REMOTE STORAGE .chESlS TITl E; ~2 I III III II This is to certify that the dissertation entitled Bacterial Wilt of Toronto Creeping Bentgrass presented by David Lee Roberts has been accepted towards fulfillment of the requirements for Ph. D degreein Plant Pathology / A7 Iéé/w L Major p ofessor \ .c Date Nov. 8, 1982 MSU i: an Affirmative Action/Equal Opportunity Institution IlIIlIIII L2 0-12771 REMOTE STORAGE KS I: PLACE IN RETURN BOX to remove this checkout from your record. To AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE 112317 #ARIéZQifl ' 20:: Blue 10/13 pz/ClRC/DateDueForms_2013.undd - 09.5 BACTERIAL WILT OF TORONTO CREEPIN G BENTGRASS By David Lee Roberts A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Botany and Plant Pathology 1982 ABSTRACT BACTERIAL WILT OF TORONTO CREEPING BENTGRASS BY David Lee Roberts Transmission electron microscopy disclosed the consistent association in five states of a rod—shaped bacterium with a Toronto creeping bentgrass (Agrostis palustris Huds cv. Toronto) disease previously known as the C-15 problem. Symptoms of the disease included blue—green shriveled leaf tips of randomly selected plants and suggested a vascular wilt. In electron micrographs the bacterium had rippled cell walls, was xylem limited and measured 0.3 to 0.5 pm in width by 1.0 to 1.5 um in length. A slow-growing bacterium forming pale-yellow, mucoid, circular, convex colonies on nutrient agar was isolated from locations in several midwestern states and proven by artificial inoculation to be the etiological agent of the Toronto creeping bentgrass disease. The bacterium exhibited the following characteristics: Gram-, oxidase—, catalase+, obligate aerobic, H28 production, starch hydrolysis, utilization of casein, gelatin liquefaction and growth at 36°C. These attributes resembled the bacterial genus Xanthomonas, although serology and nonmotility suggested no relatedness of the bentgrass bacterium to three xanthomonad isolates. Growth studies revealed similar temperature Optima and generation times for the bentgrass bacterium and Xanthomonas. Oxytetracycline applied in 1 g/L solutions at 2 L per m2 field plots suppressed disease development whereas streptomycin sulfate had no effect. An _i_n_ vitro bioassay showed that the bentgrass bacterium was David Lee Roberts approximately 100 times more sensitive to oxytetracycline than streptomycin sulfate. Based on bacterial etiology and symptomatology a suggested name for the disease is bacterial wilt of Toronto creeping bentgrass. ACKNOWLEDGMENTS I would like to express my sincere thanks to my major professor, Dr. Joseph M. Vargas, Jr., for his support and guidance during fulfillment of my degree. A very special appreciation is due to Dr. Karen K. Baker for her encouragement, enthusiasm and friendship. I also wish to acknowledge Drs. Dennis W. Fulbright and Paul E. Rieke for serving on my guidance committee and for their valuable criticism of my research and dissertation. A very warm thanks to my wife, Adele, for her continual love and patience during preparation of my dissertation and publications. Lastly, but certainly not least, a very sincere appreciation to my parents, Guy and Betty Roberts, and family, Jim, Phil, Sharon, Stephanie and Rebecca, for their eternal love and devotion. I also wish to acknowledge receipt of the Ernst A. Bessey Award for outstanding research and will treasure it as a token of my memorable experiences in the Department of Botany and Plant Pathology at Michigan State University. ii TABLE OF CONTENTS Page LIST or TABLES v LIST OF FIGURES vi INTRODUCTION AND LITERATURE REVIEW 1 SECTION I. ASSOCIATION OF A BACTERIUM WITH THE C-15 PROBLEM 11 Materials and Methods 11 Results and Discussion 12 SECTION II. DIAGNOSIS AND DISTRIBUTION OF THE BACTERIUM ASSOCIATED WITH THE C-15 PROBLEM 23 Materials and Methods 24 Results and Discussion 25 SECTION III. ISOLATION OF THE BACTERIUM CAUSING A WILT DISEASE OF TORONTO CREEPIN G BENTGRASS 34 Materials and Methods 35 Isolations 35 Inoculations 35 Host Range 36 Electron Microscopy 36 Results and Discussion 37 Isolations 37 Inoculations 38 Host Range 47 Electron Microscopy 47 SECTION IV. CHARACTERIZATION OF THE BACTERIUM CAUSING A VASCULAR WILT DISEASE OF TORONTO CREEPING BENTGRASS 57 Materials and Methods 58 Physiological and Morphological Tests 58 Ultrastructural Analysis 60 iii Growth Studies 60 Serology 61 Results and Discussion 62 Physiological and Morphological Tests 62 Ultrastructural Analysis 66 Growth Studies 66 Serology 70 SECTION V. SYMPTOM SUPPRESSION OF BACTERIAL WILT WITH OXYTETRACYCLINE 79 Materials and Methods 79 Results and Discussion 81 DISCUSSION AND CONCLUSIONS 92 LIST OF REFERENCES 101 iv LIST OF TABLES Table Page 1 Distribution of the bacterium associated with diseased Toronto creeping bentgrass as diagnosed by three electron microscopy techniques. 26 2 Inoculation tests utilizing various inoculation times, injured and noninjured plants, bacterial isolates from four states, and host range. 43 3 Representative plant pathogenic bacteria used as positive and negative controls in all tests for characterization of the bentgrass bacterium. 59 4 Reaction of the bentgrass bacterium and representative genera of plant pathogenic bacteria to general physiological-biochemical tests. 63 5 Reaction of the bentgrass bacterium and Xanthomonas spp. to specific physiological-biochemical and morphological tests. 65 6 Optical densities (O.D.) and numbers of colony forming units (CFU) of three isolates of the bentgrass bacterium at various dilutions in nutrient broth (NB) and 0.01 M phosphate buffer. 69 7 Approximate generation time of the bentgrass bacterium and C. thhomonas camgestris pv. pelargonii in nutrient broth at 5 20 C, 25 C and 30 C. 73 8 Serological analysis by gel double diffusion of the bentgrass bacterium and representative species of common plant pathogenic bacteria. 74 9 Effects of two antibiotics and a copper compound on a disease of Toronto creeping bentgrass. 82 10 Colony formation of the bentgrass bacterium on antibiotic amended nutrient agar. 90 Fi re 1and2 4and5 6, 7, 8, 9 10 11 and 12 13 LIST OF FIGURES Page Diseased Toronto creeping bentgrass in the field. (1) Leaf blades appeared blue-green, twisted, and shriveled, indicating a rapid wilt. (2) Random disease selection of individual plants of the stolon propagated Toronto creeping bentgrass. Diseased Toronto creeping bentgrass putting green. Poa annua (light green), other cultivars of creeping bentgra_ss (dark green) and individual Toronto creeping bentgrass plants remained unaffected. Transmission electron micrographs of ultrathin sections from diseased Toronto creeping bentgrass. (4) Leaf cross section with four xylem vessels (X) containing cross sections of bacteria (B). (5) Longitudinal section showed xylem limitation as evidenced by secondary thickenings (ST). Transmission electron micrographs of diseased (6-8) and symptomless (9) Toronto creeping bentgrass: (6) Cross section of plant in advanced stages of disease with bacterium advancing through the xylem wall (arrows). (7) Xylem at root-crown interface with longitudinal and cross sections of bacteria (B). Bacteria measured approximately 0.3 to 0.5 u m in width by 1.0 to 1.5 um in length. (8) Single bacterium with rippled cell walls (arrows). (9) Root cross section from symptomless plant. Xylem (X) contained no bacteria. Transmission electron micrographs of ultrathin sections of diseased Toronto creeping bentgrass from various locations revealed characteristic infection of xylem (X) by bacteria (B). Scanning electron microscopy (SEM) of diseased Toronto creeping bentgrass root. (11) Vascular bundles (VB) are arranged in a circle in a bentgrass root. (12) Close observation revealed bacteria (B) in the xylem of one vascular bundle. ~ Bacteria in TEM negative stains of expressed sap from diseased Toronto creeping bentgrass. Examination of individual bacteria revealed characteristic convoluted external surface (arrows). vi 14 17 19 21 28 30 32 14 and 15 16 17 18 19 20 21 and 22 23 Isolation of the bentgrass bacterium by dilution plating of macerated diseased crowns. (14) No bacterium with consistent color and growth habit were visible in high populations during 48 hours incubation. (15) By 144 hours, a bacterium with consistent color and growth characteristlg had beggme quite distinct on the higher dilution plates, 10 and 10 . The bentgrass bacterium exhibited circular, convex and mucoid4colony fgrmation with pale yellow pigmentation on the 10 and 10 nutrient agar dilution plates. Uninoculated control and inoculated Toronto creeping bentgrass plants. Several days after inoculation with the bentgrass bacterium Toronto creeping bentgrass developed characteristic blue-green, wilt symptoms (arrows) similar to those observed in the field. Transmission electron micrograph of single negatively stained bentgrass bacterium isolated from diseased 'Toronto.’ Isolated bacterium exhibited similar convoluted surface (arrows) and size to those bacteria observed in negative stain of expressed plant sap. Transmission electron microscopy of a bentgrass bacterial isolate embedded in agar. Longitudinal ultrathin section of the isolated bacterium showed similar size and morphology to ultrathin sections of bacterium observed in diseased 'Toronto.‘ Artificial inoculation of Toronto creeping bentgrass with the frequently-isolated bacterium. Transmission electron micrograph of ultrathin sections confirmed infection by bacteria (B) in the plant xylem (X) similar to natural infection in the field. Scanning electron microscopy confirmed infection by the bentgrass bacterium in artificial inoculations. (21) Artificially-inoculated plant xylem (X) with bacteria (B). (22) Uninoculated control plant xylem (X). Ultrastructural characteristics of the bentgrass bacterium. The cell wall possessed an inner membrane (1M) and outer membrane (OM), typical of Gram negative cell walls. Cytoplasm contained DNA-like strands (DNA) and electron dense spheres resembling ribosomes (R). vii 40 42 45 49 51 53 55 68 24, 25, 26 27 and 28 Figure 27: Figure 28: 29 30 and 31 32 and 33 Growth of the bentgrass bacterium and Xanthomonas cam estris pv. pelargonii as measured by optical denSIty (absorbance at 520 nm) Suring gncubationo in nutrient broth shaker flasks at 20 C, 25 C and 30 C. Neither bacterium grew at 5 C. (24) X. camEstris cpv. Elargonii and bentgrass bacterium at 20 C. (25) X. camgstris cpv. Elargonii and bentgrass C Facterium at 25 . (26) X. campestris (pv. Elargonii and bentgrass C Facterium at 30 . Serological analysis by agar double diffusion of the bentgrass bacterial isolates (numbered 7, 20, 30, 40, 50) and known plant pathogenic bacteria. Precipitate lines indicate positive reaction of antiserum (A/S) dilution 1/32 (center well) to bacterial test isolate (outer wells). # 7 = Illinois isolate # 20 = Illinois isolate # 30 = Michigan isolate # 40 = Ohio isolate # 50 = Wisconsin isolate X.p. = Xanthomonas campestris pv. Elargonii # 7 = Illinois isolate C.m. = Corynebacterium michiganense pv. michiganense A.t. = AgLobacterium radiobacter var. tumefaciens Biovar E.h. = Erwinia herbicola P.s. = Pseudomonas solanacearum X.p. = Xanthomonas campestris pv. Elargonii Oxytetracycline (OT) at 1 g/L suppressed disease symptoms after two applications whereas disease deveIOped in the streptomycin sulfate (1.5 g/L) treated plot (ST) at Location A. Toronto creeping bentgrass treated with antibiotics at location B. (30) Oxytetracycline (OT) at 1.0 g/L suppressed development of disease compared to the nontreated c eck (CK). (31) After single applications (1 g/L at 190 L/93m ) of each antibiotic to adjacent halves of a golf green, disease developed on the streptomycin-treated (ST) half, whereas no diseased developed on the oxytetracycline-treated (OT) half even after six weeks. Scanning electron microscopy of crown freeze-fractures of Toronto creeping bentgrass. (32) Xylem vessels from untreated plant containing many bacteria (B). (33) Vascular viii 72 76 84 86 tissue from plant treated (1.0 g/L) with oxytetracycline. Xylem vessels (X) contain no bacteria. 89 ix INTRODUCTION AND LITERATURE REVIEW Some of the world's most economically important plant species are in the family Gramineae. Rice, wheat, corn, oats, sugar cane and others are of unquestionable value for the well being and survival of maple and animals. Turfgrasses are also members of the Gramineae, although often not thought to be as economically important as the grain plants. Even so, billions of dollars are spent annually throughout the world for turfgrass maintenance (Beard, 1973). In the United States approximately 4.3 billion dollars were expended for turfgrass maintenance in 1965. Estimates suggest that as much as 260 million dollars were spent in 1969 for turfgrass maintenance in Michigan alone (see Hanson and Juska, 1969). Today the figure would undoubtedly be far in excess of this 1969 estimate, including such expenses as seed, fertilizer, pesticides, mowing equipment and labor. Turfgrass maintenance ranges from very low orders of maintenance such as for roadsides and parks to more highly maintained areas such as home lawns and golf courses. Management of turfgrass diseases and pests accounts for a considerable proportion of turfgrass maintenance. While herbicides, insecticides and fungicides are applied to a wide range of turfgrass culture situations, their use increases dramatically where turfgrass receives intense cultivation. Besides chemical pesticides, cultural practices and varietal differences can also be quite effective in controlling diseases and pests. Certain cultivars of grasses are quite susceptible to particular disease-causing microorganisms whereas other cultivars of grasses are immune or highly resistant to the same 2 pathogens. One remarkable example has become known as the C-15 problem or C-15 decline. In the 1930's, the United States Department of Agriculture (U.S.D.A.) and the United States Golf Association (U.S.G.A.) selected various cultivars of Agrostis palustris Huds. (creeping bentgrass) for high maintenance turf areas (Hanson and Juska, 1969). One such cultivar was designated 'Toronto' and assigned cultivar number 15 (C-15). Such characteristics as dense growth, fine texture, light color, close mowing adaptation, aggressive creeping habit, and resistance to segregation after a long time have made 'Toronto' a choice cultivar for high maintenance turf areas, particularly golf course‘putting greens. Toronto creeping bentgrass has been so widely cultivated and popular in the midwestern United States that it has even been called the 'elite' among the bentgrasses. After decades of cultivation, however, a disease of unknown etiology emerged which has destroyed the otherwise outstanding reputation of Toronto creeping bentgrass. In the late 1960's and early 1970's several universities in the Midwest devoted considerable effort to solving the disease of unknown etiology by extensive fungicide and fertility experimentation (personal communication). Neither consistent control nor isolation of a pathogenic agent was achieved. Because of this inconclusive research, very little information concerning symptomatology and conducive environmental conditions was published during this period. Even though the disease was well known and recognized, no publications other than popular articles were published which even recognized the problem. In the mid and late 1970's, research on the C-15 problem increased. In 1975, researchers at the University of Illinois found that red leaf spot, caused by Helminthosporium erythrospilum (= Drechslera erythrospila) was an important 3 disease on Toronto creeping bentgrass in the Chicago, Illinois area (Meyer and Turgeon, 1975). They reported that fungicides previously known to be effective for diseases caused by Helminthosporium spp. were not effective for the C-15 disease. Red leaf spot was reported to be controlled by the fungicide chlorothalonil when used in conjunction with early spring fertilization. In 1979 and 1981, researchers at The Ohio State University reported leaf blight and crown rot as a new and important disease of Toronto creeping bentgrass. Leaf blight and crown rot was controlled by application of the fungicide iprodione (Larsen, 1979; Larsen et al., 1981). The reports of these two fungal diseases did not appear to be presented with adequate descriptions of symptoms. The 1975 report (Meyer and Turgeon) suggested that symptoms were similar to the disease described much earlier (Drechsler, 1935). Although the other two studies (Larsen, 1979; Larsen et al., 1981) reported initial symptoms as being red leaf lesions and "leaf tip dieback," plants eventually "blighted" to the crown. Couch (1973) also reported symptoms of red leaf spot as a "withering" of the leaves and that an overall view of a diseased stand of 'Toronto' may give a "drought-stricken appearance." Meyer and Turgeon (1975) noted that the "red leaf spot" epidemic was affecting Toronto creeping bentgrass locations that had obtained their stolons from one nursery. Likewise, in 1980, the C-15 problem reached epiphytotic pr0portions in the midwest on many locations that had cultivated 'Toronto' stolons from that particular nursery location. This indicated that a specific strain of 'Toronto' was being affected by the disease. One such location that gained national recognition was the Butler National Golf Course in Oak Brook, Illinois. The C-15 problem devastated 'Toronto' greens of Butler National several weeks before the 1980 Professional Golfers Association (PGA) Western Open. Because of the unpredictable nature of the disease, turfgrass specialists from 4 across the nation consulted on the C-15 problem at Butler National. The dynamic existence of the disease in various stages of development on many 'Toronto' putting greens at Butler permitted extensive experimentation and sampling. Symptoms and spread of the disease suggested a pathogenic agent, however, applications of many common fungicides as advised by turf specialists did not inhibit the progress of the disease. In spite of the lack of disease control with iprodione and chlorothalonil, the disease was tentatively diagnosed as red leaf spot or leaf blight and crown rot, or labeled the "mystery disease" by scientists, turf specialists and the news media. Most turfgrass diseases are caused by fungi, however, one mchplasma disease and one virus disease are also known (Vargas, 1981; Couch, 1973). Because several broad spectrum fungicides were not effective at Butler National and other midwestern locations, fungal etiology of this disease was highly unlikely. It was suspected that microorganisms other than fungi were responsible for the poorly-understood C-15 problem. After fungi, bacteria are the second largest group of microorganisms that cause plant diseases. Bacteria are classified’in the Kingdom Procaryotae (see Buchanan and Gibbons, 1974). Thousands of different species of bacteria have been characterized (see Buchanan and Gibbons, 1974) but only a few genera are known to cause plant diseases. Corynebacterium, Agrobacterium, Erwinia, Pseudomonas and Xanthomonas are the five most prominent disease-causing genera of bacteria currently recognized by plant pathologists (see Schaad, 1980; Agrios, 1978). Another group of bacteria that have in recent years been shown to cause plant diseases are the so-called rickettsia—like bacteria (RLB) (Goheen et al., 1973; Hopkins et al., 1973a; Hopkins, 1977; Kitajima et al., 1975), also known as bacteria of uncertain affiliation (see Schaad, 1980). These bacteria of uncertain 5 affiliation were called rickettsia-like bacteria due to their similarities in morphology, size and supposed obligate parasitism, to the true rickettsias that cause diseases in mammals (see Nester et al., 1973). The major groups of plant pathogenic bacteria cause a wide variety of diseases, including leaf spots, wilts, cankers, rots and blights (Agrios, 1978). Vascular wilt diseases are caused by many different phytopathogenic bacteria. Bacterial canker of tomato (Corynebacterium michiganense pv. michiganense), southern bacterial wilt of solanaceous cr0ps (Pseudomonas solanacearum), bacterial wilt of cucurbits (Erwinia tracheiphila) and black rot of cabbage (Xanthomonas campestris pv. campestris) are a few representative vascular diseases caused by members of several prominent genera of phytopathogenic bacteria (see Schaad, 1980). Mechanisms of vascular wilt induction by bacteria is controversial (Dimond, 1970; Buddenhagen and Kelman, 1964). Symptom development in vascular diseases is generally attributed to physical blockage of xylem vessels by the bacteria (Nelson and Dickey, 1966), to secretion of extracellular polysaccharide (slime) by bacteria (Husain and Kelman, 1958) which inhibits the movement of moisture, to accumulated growth substances (Sequiera and Kelman, 1962), or to toxic glchpeptides (Hess and Strobel, 1969). Bacterial diseases are managed by a variety of methods. Sanitation, cultural practices, and host plant resistance are among the most common methork of disease control. Chemical control has been far less successful for control of bacteria than for fungal disease control. Of the chemicals in use, copper compounds have been most effective for control of bacterial leaf spots and blights (Agrios, 1978). Antibiotics have been used in recent years with some success. Tetracyclines and streptomycin are very broad spectrum antibiotics, inhibiting 6 many genera of both gram negative and gram positive bacteria (Dekker, 1963; Nester et al., 1973). Both antibiotics are known to be systemic in plants (Dekker, 1963; Frederick et al., 1971; Sinha and Peterson, 1972) which probably explains their suppressive effects on vascular diseases. Hopkins and Mortensen (1971) and Hopkins et al. (1973b) found good symptom suppression of Pierce's disease of grapevines with tetracycline antibiotics. Streptomycin has been proven effective against fireblight of apple (Goodman, 1954) caused by Erwinia amylovora and walnut blight (Miller, 1959) caused by Xanthomonas juglandis. Many reports with less success of antibiotic disease control are well documented (Dekker, 1963; Ark and Alcorn, 1956). Expense, ineffective control and antibiotic administration in human medicine have made antibiotics undesirable for management of plant disease. However, preliminary use of antibiotics have an important benefit of demonstrating suppression of plant diseases caused by procaryotic microorganisms (Hopkins et al., 1973b; Hopkins and Mortensen, 1971; Maramorosch, 1976). Tetracycline antibiotics have been particularly useful for inhibiting and demonstrating the etiology of plant diseases caused by mycoplasmas (Nienhaus and Sikora, 197 9; Maramorosch, 1976). Plant pathogenic bacteria have been identified or characterized on the bases of morphological, physiological and pathogical analyses (see Buchanan and Gibbons, 1974; Schaad, 1980). Morphological criteria include size, shape, motility, and type of flagellation. Physiological or biochemical analyses typically test for respiration or utilization of certain carbon compounds and relate the presence of various enzyme systems. Pathological tests delimit the various plant host(s) on which a bacterium is capable of inciting a disease. Bergey's Manual (see Buchanan and Gibbons, 1974) currently lists plant pathogenic bacteria in designated parts based on cell morphology, Gram reaction, oxygen requirement and mode of energy obtainment. Many of the plant 7 pathogens are recorded as separate species even though they differ only on the host attacked and are biochemically indistinguishable. For example, five species of Xanthomonas are accepted yet several hundred other species are listed separately and can only be distinguished from Xanthomonas campestris on the basis of the disease host, for which the bacteria are also given their species name. Various bacteriologists have recognized the problems with bacterial taxonomy (Young et al., 1978; Dye, 1962; Dye, 1974; Stolp et al., 1965). In a study involving 209 phytopathogenic Xanthomonas cultures comprising 57 recognized species, Dye (1962) could not distinguish any of the various species from one another on the basis of standard laboratory tests. Dye (1962) suggested that because the Xanthomonas cultures varied only in pathogenicity tests, they should be regarded as one species. Recently Young et a1. (1978) recognized the inherent problems in classifying many species of bacteria that can only be differentiated on the basis of its host. Many similar bacterial species that do not have claims to species rank were classified as pathovars (Young et al., 1978). Thus far, the classification of the RLB is uncertain (see Schaad, 1980; see Davis et al., 1981b). Several of the RLB have been cultured on specific media (Davis et al., 1978; Davis et al., 1981a; Wells et al., 1981) and have been shown in some instances to be serologically related to one another (see Davis et al., 1981b), but the relationship to other plant pathogenic bacteria has yet to be resolved. In recent years DNA base composition and hybridization have also been regarded as extremely important criteria for bacteria systematics. Bergey's Manual (see Buchanan and Gibbons, 1974) generally lists guanine to cytosine (GC) ratios for most bacteria. Usually bacteria that have similar GC ratios and a high DNA hybridization index are regarded as more closely related than those bacteria that do not. De Ley (1968) noted the percent GC was similar for both 8 Pseudomonas and Xanthomonas and concluded that both genera should be united under the name Pseudomonas campestris. This is a typical example of the problems encountered in taxonomy when only one characteristic is considered. Xanthomonas and Pseudomonas are similar in many ways. Both genera are Gram negative, motile, rod shaped, similar in size, and contain species that produce similar types of diseases (see Schaad, 1980). Even though these similarities may be important, the two genera are far different in other ways (see Schaad, 1980). Many pseudomonads produce a fluorescent pigment and are oxidase negative. Other pseudomonads do not produce the fluorescent pigment and are oxidase positive. Xanthomonas species are known for the production of a polysaccharide called "xanthan" (Smith et al., 1952). Another unique characteristic of the genus is the production of a yellow pigment. This pigment was named "xanthomonadin" and is far different from the carotenoid-type pigments found in many other bacteria. The xanthomonadin pigments consist of brominated, aryl-polyene esters with absorption maxima at 418, 437 and 463 nm in petroleum ether (Andrews et al., 1973; Starr and Stephens, 1964). Starr et a1. (1977) suggested that the pigment was of taxonomic significance since it could be found only in the genus Xanthomonas. Other characteristics of the xathomonads usually include starch hydrolysis, gelatin liquefaction, utilization of casein, and hydrogen sulfide production (see Schaad, 1980). Serology has developed into an important tool in plant pathology for (1) detecting plant pathogenic bacteria and viruses and for (2) characterizing or classifying viruses and bacteria into distinct sero-groups. Many serology techniques have been develOped and are used routinely by plant virologists (see Ball, 1974). Dilution endpoint, agar diffusion, immunoelectrOphoresis, and fluorescent antibody are several broad categories of immunological techniques. 9 Plant bacteriologists are becoming increasingly aware of the importance of serological techniques. Numerous reports abound concerning detection and identification of plant pathogenic bacteria. Schaad (1978) used direct and indirect immunofluorescence for rapid identification of X. campestris. All strains of X. campestris and X. vesicatoria cross reacted with X. campestris. Serology has been used extensively to detect the RLB and has shown that these fastidious bacteria of uncertain affiliation form an immunological distinct group (see Davis et al., 1981b). The RLB do not exhibit any serological relationship with the common genera of phytopathogenic bacteria. Agar diffusion was used by Jenkins et a1. (1966) for distinguishing peanut wilt caused by P. solanacearum from other diseases with similar symptoms. In addition to distinguishing unrelated bacteria from one another, Elrod and Braun (1947) could differentiate members of the genus Xanthomonas into several immunological groups. Thus serology may differentiate between members on the generic level or on the subspecies level, depending upon the specificity of the tests. Like serology, electron microscopy (EM) has become a very important tool for many plant pathologists. Typically, EM has been used chiefly as an instrument of high resolution for examination of fine ultrastructural detail (Meek, 1976). Ultrastructural examination is most effectively achieved with standard transmission electron microscopy (TEM) although generally considered quite laborious. Standard TEM usually involves the following procedures: fixation in glutaraldehyde and osmium tetroxide, dehydration in a graded series of ethanol, resin embedding, ultrathin sectioning and positive staining (Hooper et al., 1979). Scanning electron microscopy (SEM) requires less preparation than standard TEM and involves: fixation in glutaraldehyde and osmium tetroxide, dehydration in a graded series of ethanol, critical point drying, and gold coating (Hooper et al., 1979). Although not capable of as high resolution as standard 10 TEM, SEM is quite useful for studying surfaces of biological specimens or nonbiological objects. Although EM is not ordinarily used in plant disease diagnosis, virologists have used leaf dips to detect viruses in diseased tissue and to ascertain the structure, size and shape of virus particles (Gibbs and Harrison, 1976). Negative stain TEM, the method commonly employed for leaf dips by virologists, is a very rapid procedure, involving little time and preparation (Hooper et al., 1979). Besides diagnosis, EM has potentially important implications in bacterial taxonomy. The ultrastructure of the cell wall, cell t0pography, flagellation, cytoplasmic components and other significant characteristics of bacteria can easily be examined with EM. The size, shape and t0pography of cell walls and cell enveIOpes have been viewed extensively with TEM (Joklik and Willet, 1976). The C-15 problem has remained a mystery for at least a decade. Although conventional isolation and chemical application techniques have been very important for discerning most turfgrass diseases (Vargas, 1981), the C-15 problem has neither been effectively controlled nor accurately diagnosed by these methods (Meyer and Turgeon, 1975; Larsen et al., 1981). The purpose of this dissertation research was to reevaluate and solve the C-15 problem by incorporation of methods not commonly employed by turfgrass pathologists. 11 SECTION I ASSOCIATION OF A BACTERIUM WITH THE C-15 PROBLEM Toronto (C-15) creeping bentgrass has been a popular turfgrass for golf course putting greens ever since its selection in the 1930's. During the past 15 years a disease of unknown etiology, subsequently called the "C-15 problem" or "C-15 decline," has destroyed the elite turfgrass in many 'Toronto' growing areas. The C—15 problem reached epiphytotic proportions in the Summers of 1979 and 1980 in many locations in the Midwest. However, the disease gained national recognition when the 'Toronto' greens at the Butler National Golf Course in Oak Brook, Illinois were devastated several weeks in advance of the 1980 Professional Golfers Association (PGA) Western Open tournament. Previous reports of two fungal diseases (Meyer and Turgeon, 1975; Larsen, 1979; Larsen et al., 1981) had not provided final solutions to the C-15 problem because fungal pathogens could not be isolated and recommended fungicide control programs were not effective in 1980 at Butler National. Symptomatology of the C-15 problem had not been sufficiently described previous to this study. Because of the disease symptoms and the nature of its spread, the purposes of this part of the research were to describe the symptomatology of the disease and investigate the possibility of an infectious, disease-causing agent. MATERIALS AND METHODS Segments measuring 5-10 mm of root, crown, and leaf tissue from diseased plants were surface sterilized in 0.596 sodium hypochlorite or 0.196 mercuric 12 chloride or were dipped in 95% ethanol and flamed. After three washes in sterile distilled water, the plant segments were plated on potato dextrose agar (PDA) and 296 water agar for isolation of fungi, on nutrient agar (NA) and sucrose (0.596) nutrient agar (SNA) for isolation of bacteria, and on J D-3 medium (Davis et al., 1978) for isolation of certain fastidious bacteria. Plates were incubated at 25 C and observed daily for two weeks for growth of microorganisms. Isolation attempts were repeated throughout the summer. Symptomless plants were taken through identical sterilization and isolation procedures. Electron microscopy was performed on samples collected in June and August, 1980. Several 1 mm pieces of leaves, crowns, and roots from each of 10 diseased plants exhibiting wilting of the leaf tips were fixed in cold 4% glutaraldehyde in 0.1 M phosphate buffer, pH 7.2, for two hours; postfixed overnight in cold 1% buffered osmium tetroxide, and dehydrated in a graded series of ethanol. Specimens were embedded in a 1:1 mixture of epon-araldite and ERL epoxy resin (Hooper et al., 197 9). Ultrathin sections were stained with 2% uranyl acetate and subsequently in lead citrate (Reynolds, 1963). Root and crown segments from five symptomless plants were prepared in the same manner. Sections were examined in a Philips 300 or 201 transmission electron microscope. RESULTS AND DISCUSSION Initially affected Toronto creeping bentgrass plants at Butler National possessed blue-green leaf tips indicating a rapid wilt. The wilt progressed from the leaf tip to the crown, and eventually the entire leaf blade appeared blue- green, twisted and shriveled (Figure 1). Roots and crowns initially appeared white and in good general health but later turned brown after shoot tissue was severely wilted. Entire decomposition of the plant usually followed. In some instances disease development stopped and new shoots emerged from the crown. 13 Figures 1 and 2. Diseased Toronto creeping bentgrass in the field. (1) Leaf blades appeared blue-green, twisted, and shriveled, indicating a rapid wilt. (2) Random disease selection of individual plants of the stolon prepagated Toronto creeping bentgrass. 14 15 The disease advanced randomly through the putting green killing individual plants of the stolon propagated C-15 (Figure 2). Some individual C-15 plants as well as 22a annua L. (annual bluegrass) and other cultivars of Agrostis palustris remained unaffected (Figure 3). Disease development appeared to be favored by periods of high moisture (precipitation and relative humidity) followed by bright, warm (25°C or above) days. Large areas of the putting green were destroyed in a few days. No fungal pathogens commonly reported from turfgrass nor any other fungi were observed or isolated regularly from diseased plants obtained from Butler National Golf Course. Several bacteria, differing in colony color and morphology, were frequently isolated from diseased plants on PDA, NA and SNA, and J D—3. Pathogenicity tests by either the immersion of injured roots or crown injection, both using pure bacterial broth cultures, were inconclusive after several attempts. Bacteria were not isolated consistently from symptomless plants. Transmission electron miroscopy was used to ascertain the presence of microorganisms in affected tissues because the symptomatology suggested a vascular disorder. Examination of ultrathin sections from all 10 diseased creeping bentgrass plants revealed large numbers of rod-shaped bacteria in the“ roots, crowns, and leaves. The bacteria were limited exclusively to the xylem vessels (Figures 4 and 5) except for advanced stages of the disease when bacteria moved into adjacent areas (Figure 6). The bacteria measured approximately 0.5 u m in diameter and 1-1.5 u m in length (Figure 7). In longitudinal section, the bacterial cell wall appeared rippled (Figure 8). These bacteria were similar in size, morphology, and location within the plant to the rickettsia-like bacteria associated with plum leaf scald (Kitajima et al., 1975), phony disease of peach (Hopkins et al., 1973a), Pierce's disease of grapevines (Goheen et al., 1973), 16 Figure 3. Diseased Toronto creeping bentgrass putting green. Poa annua (light green), other cultivars of creeping bentgrass (dark green) and individual Toronto creeping bentgrass plants remained unaffected. Figures 4 and 5. 18 Transmission electron micrographs of ultrathin sections from diseased Toronto creeping bentgrass. (4) Leaf cross section with four xylem vessels (X) containing cross sections of bacteria (B). (5) Longitudinal section showed xylem limitation as evidenced by secondary thickenings (ST). 19 Figures 6-9. 20 Transmission electron micrographs of diseased (6-8) and symptomless (9) Toronto creeping bentgrass: (6) Cross section of plant in advanced stages of disease with bacterium advancing through the xylem wall (arrows). (7) Xylem at root-crown interface with longitudinal and cross sections of bacteria (B). Bacteria measured approximately 0.3 to 0.5 u m in width by 1.0 to 1.5 um in length. (8) Single bacterium with rippled cell walls (arrows). (9) Root cross section from symptomless plant. Xylem (X) contained no bacteria. 21 22 alfalfa dwarf (Goheen et al., 1973), and in symptomless johnsongrass (Sorghum halepense), which has been implicated as a possible reservoir of rickettsialike bacteria for phony peach disease (Weaver et al., 1980). No bacteria were found in any ultrathin sections from symptomless plants (Figure 9). The discovery of bacteria within the xylem vessels of wilted plants suggests a possible causal agent for the disease that destroyed the Toronto greens at the Butler National Golf Course. In addition, the association of bacteria with the disease may represent a significant breakthrough in the unsolved C-15 problem of Toronto creeping bentgrass, for which no effective fungicidal control has been found. Bacterial diseases of turfgrasses have not been reported previously. Although bacterial diseases are common for other genera of the Gramineae (see Schaad, 1980), the bacterium association with Toronto creeping bentgrass represents the first bacterial disease encountered on a turfgrass species. SECTION II DIAGNOSIS AND DISTRIBUTION OF THE BACTERIUM ASSOCIATED WITH THE C-15 PROBLEM Since the discovery of two fungal diseases (Meyer and Turgeon, 1975; Larsen, 1979; Larsen et al., 1981) did not appear to provide adequate solutions to the C-15 problem, the discovery of bacteria in the xylem vessels of diseased Toronto creeping bentgrass at the Butler National Golf Course in 1980 was an important and timely alternative. A turfgrass disease caused by a bacterium had not previously been described and represented a plausible explanation to the symptoms associated with the poorly understood C-15 problem. The bacterium association was first discovered by routine examination of ultrathin sections of diseased Toronto creeping bentgrass with transmission electron microscopy (TEM). Standard TEM ultrathin sectioning has been used in basic research primarily for studying fine structural detail (Meek, 1976) and has not often been employed for diagnosis of plant diseases. TEM preparation is an expensive, tedious and time-consuming process involving approximately 1 to 2 weeks for completion and subsequent observation in an electron microscope (H00per et al., 197 9). Other electron microscopy methods require less time and expense. Scanning electron microscopy (SEM) can provide fine detail of plant surfaces or other objects and can be completed after 1 to 2 days of preparation (Hooper et al., 1979). In addition, negative stain TEM, used quite commonly to observe structure and size of viruses or other very small particles, is a very rapid technique requiring only several minutes for completion. 23 24 Thus, the possible employment of SEM and/or TEM negative staining as alternate means of determining the distribution of the association of the bacterium with diseased Toronto creeping bentgrass seemed to provide a good test system for methods of disease diagnosis. The association of the bacterium with diseased 'Toronto' at locations other than Butler National would present more evidence for the presumption that a bacterium is the cause of the C-15 problem. The purposes of this segment of the research were to determine the distribution of the bacterium associated with diseased 'Toronto' and to evaluate three electron microscopy techniques for use in rapid disease diagnosis. MATERIALS AND METHODS Golf and country clubs reporting the C-15 problem on their 'Toronto' greens were sampled from 1980 to 1982. Ten cm diameter cores were obtained from affected golf greens exhibiting the previously described dark green, shriveled leaf tips and characteristic random disease spread. Two symptomless and at least four plants exhibiting the wilted leaf tips in the early stages of disease development were selected from each sample and washed for at least two hours in running tap water in preparation for electron microscopy. For standard TEM ultrathin sectioning, leaves, crowns, and roots were processed as described previously (see previous section). Plants were prepared for SEM in a similar manner as for standard TEM. After dehydration in a graded series of ethanol, plant pieces were freeze-fractured in liquid nitrogen and critical point dried (Hooper et al., 1979). Samples were then mounted on aluminum stubs and gold-coated for 3.5 minutes using a Film Vac. Inc. sputter coater. For the rapid TEM negative stain technique, crowns were dissected from selected plants and squeezed or macerated in a drop of distilled water on a glass slide. Drops of the crushed crown-water slurry were placed on carbon- and 25 parlodion-coated copper mesh grids. Grids were dried after 1 minute by blotting with filter paper and stained with 2% aqueous uranyl acetate for 45 seconds and blotted dry. Both diseased and symptomless 'Toronto' plants from various locations in five states were prepared for standard TEM, TEM negative staining and/or SEM. Samples prepared for SEM were observed in a JEOL JSM 35C scanning electron microscope while samples for standard TEM and TEM negative stain were observed in either a Philips 300 or Philips 201 transmission electron microscope. RESULTS AND DISCUSSION The rod-shaped bacterium was associated with diseased Toronto creeping bentgrass at many locations where the previously described symptomatology was observed (Table 1). As expected, TEM ultrathin sections showed that the bacterium was limited to xylem vessels, possessed rippled cell walls and measured approximately 0.3 to 0.5 um in width by 1.0 to 1.5 um in length (Figure 10) and hence was very similar to the bacterium observed in diseased 'Toronto' from Butler National. SEM proved to be an outstanding diagnostic technique. Numerous rod- shaped bacteria were observed in the xylem vessels of vascular cylinders of diseased Toronto creeping bentgrass (Figure 11 and Figure 12). Bacteria were not found in any SEM or TEM samples from symptomless Toronto creeping bentgrass plants. The rapid TEM negative stain technique also proved quite useful. Macerated crowns of diseased Toronto creeping bentgrass plants from five locations correlated 100% with standard TEM and SEM for the presence of the bacterium (Table 1). Observation of large numbers of uniform bacteria on the parlodion-coated grid was recorded as positive (Figure 13). Close examination of the bacteria revealed the characteristic rippled external surfaces (Figure 13). 26 .wmwemucon mcfiooeo 3:959 firs E32323 2:. mo 533083 83065 =+=w u + 330 F5550 xoohnzocwosc oEomm Ems—8&3 - + 6535 223 59:22 - + $595 Baum 220 59:2: - + + 88585 :2st 596% - + See 585sz 9:2 59:22 - + ems? Boats—ls 535 2:0 - + 3:33 .3: «as: .E ages - + + SE moov 3:53: Roam: - + SE. BEE 289:: scam: I + AEESBBSEB 878:0 .5 £95: - + + + 3:8 32:50 3:20 .59 8:20 .a mesa n + 30a; voozomumv man—ram 32:3 £95: - + + 3:: 82:3 .56 58 Roan: - + + 358 2C 2602 mesa - a+ 22332 base :85 do s95: Ems Ems 8:83 93m 53m 558m 65882 saw 552:: mmBanmEzm commofla 833503 30898:: 5.588 8%: t3 38:me ma magmas—on @5398 3:28. 8336 at? @3308? 832323 2: no 533235 A 033. 27 Figure 10. Transmission electron micrographs of ultrathin sections of diseased Toronto creeping bentgrass from various locations revealed characteristic infection of xylem (X) by bacteria (B). Figures 11 and 12. 29 Scanning electron microscopy (SEM) of diseased Toronto creeping bentgrass root. (11) Vascular bundles (VB) are arranged in a circle in a bentgrass root. (12) Close observation revealed bacteria (B) in the xylem of one vascular bundle. 30 31 Figure 13. Bacteria in TEM negative stains of expressed sap from diseased Toronto creeping bentgrass. Examination of individual bacteria revealed characteristic convoluted external surface (arrows). 32 33 A few bacteria of various sizes and morphologies were occasionally observed in very low numbers on grids of macerated healthy plants and were presumed to be natural populations of saprOphytic bacteria. SEM and ultrathin sectioning TEM are both quite valuable for different reasons. Freeze-fractured plant segments could easily be observed in the SEM for both the presence and location of bacteria. With TEM, fine detail of plant (xylem) and bacterial (cell wall) ultrastructure was readily obtained. However, preparation for SEM was much more easily accomplished, making SEM the technique of choice for rapid diagnostic evaluation. Although TEM negative stains of expressed plant material were quite useful and the most rapid technique, its use did not permit the observation of bacteria i_n s_it_u (in xylem vessels). For diagnosis, SEM and TEM negative stain techniques appeared to be suitable substitutes for the laborious standard TEM. The results of the distribution study showed that a bacterium was consistently associated with the C-15 problem. The bacterium was associated with diseased 'Toronto' from 15 locations in Illinois, Indiana, Ohio, Michigan and Wisconsin (Table 1). This widespread distribution of the bacterium associated with diseased 'Toronto' and not with apparently healthy plants in the Midwest proved that the original finding at Butler National was not an isolated phenomenon. This study presents additional, convincing evidence that a bacterium is the likely incitant of the C-15 problem. SECTION III ISOLATION OF THE BACTERIUM CAUSING A WILT DISEASE OF TORONTO CREEPIN G BENTGRASS The previous studies presented evidence for the consistent association of a bacterium with the poorly understood Toronto creeping bentgrass disease known as the C-15 problem. Electron microscOpy has shown the bacterium to be associated with disease symptoms in at least 15 locations in Illinois, Indiana, Ohio, Michigan and Wisconsin, and that the disease is highly specific for Toronto creeping bentgrass. Electron microscopy also showed that the bacterium possessed a rippled cell wall, measured approximately 0.3 - 0.5 by 1.0 - 1.5 u m, and was limited to xylem vessels. These characteristics resembled the so-called rickettsia-like bacteria (RLB) or bacteria of uncertain affiliation, which cause diseases in other crop plants (Goheen et al., 1973; Hopkins et al., 1973a; Kitajima et al., 1975). Since 1978, several of the fastidious RLB in other crop diseases have been isolated into pure culture using a JD—3 or other specialized media (Davis et al., 1978; Wells et al., 1981). The objectives of this portion of the research were to isolate the bacterium observed with electron microscopy and to prove by ultrastructural comparisons and by inoculation with subsequent symptom development that the bacterium is the etiological agent of the C-15 problem. 34 35 MATERIALS AND METHODS Isolations Toronto creeping bentgrass plants from at least one location in each of four states, Ohio, Michigan, Illinois and Wisconsin, were washed thoroughly in running tap water with two drops of Tween 20 (Atlas Chemical Ind., Inc.) for two hours. In preparation for three separate isolation techniques, plants were surface sterilized in 0.525% sodium hypochlorite for two minutes, and washed in three changes of 0.01 M phosphate buffer. In a direct plating method, leaves, crowns, and roots were plated onto agar plates. In the second method, plant sap was expressed into a drop of sterile 0.01 M phosphate buffer and streaked onto an agar plate with a sterile transfer loop. In the third method, plant crowns were macerated in 1 ml sterile 0.01 M phosphate buffer, diluted with 9 ml 1 to 10-5. One-tenth ml of each phosphate buffer and serially diluted from 10- dilution was spread with a sterile L-shaped glass rod onto agar plates and incubated at 23°C for one week. The standard 15 x 100 mm petri plates contained 15 m1 of either nutrient agar (NA-Difco) or JD-3 medium (Davis, 1978). Two diseased plants and two symptomless plants from each of the four locations were prepared for each of the three isolation techniques. Pure cultures of isolated bacteria were obtained by streaking onto J D-3 or NA and stored on yeast salts broth (See Schaad, 1980). Inoculations Individual Toronto creeping bentgrass stolons were rooted in sand for one week. Simultaneously, pure cultures of isolated bacteria from each of the four locations were grown on NA for four days, transferred to 50 ml of nutrient broth (NB) and placed on an Eberbach shaker (180 cycles/minute) for 72 hours. Bacterial suspensions were centrifuged at 2500 x g for 10 minutes, resuspended in 0.01 M phosphate buffer and adjusted to 105 colony-forming units (CFU) per 36 ml by methods described in Section IV. Rooted stolons were injured by clipping off the leaf tip and then immersed in the bacterial suspension on the shaker (50 cycles/minute) for 15 minutes. Plants were removed, planted in 500 m1 polystyrene cups containing a 50:50 mix of sand and sterile soil, and incubated at 28°C and 95 - 100% relative humidity for 48 hours. At that time, plants were placed in the greenhouse or under artificial lights (28 - 30°C) and observed for symptoms. Twenty individual plants were tested, five plants for a representative isolate from each state. Five uninoculated control plants were treated following the same procedure using buffer without bacteria. Experiments were repeated three times. Preliminary inoculation tests incorporated slight modifications of the above procedure. In one experiment, ten plants were injured and ten plants remained uninjured before immersion for 15 minutes in bacterial suspensions. In another test, plants were immersed for 5, 15, 30, 60 or 120 minutes (five plants/treatment time) in bacterial suspensions to determine any minimum time of plant exposure to bacteria for successful inoculation. Host Range Fifteen seedlings each of A. palustris Huds. cv. Penncross, A. palustris cv. Penneagle, and _Po_a annua L. were propagated from seed and subjected to the same inoculation and incubation procedures as described above. Ten plants of each cultivar were inoculated with a mixed suspension of two bacterial isolates known to be pathogenic on 'Toronto' and five served as controls. Electron Microscopy Electron microscopy was used for comparison of isolated bacteria with bacteria observed in diseased plants from the field and for verification of infection by artificial inoculation. Samples for scanning electron microscopy (SEM) and for ultrathin section transmission electron microscOpy (TEM) were 37 prepared as described previously (See sections I and II). For TEM ultrathin sections of isolated bacteria, single colonies on NA were mixed with molten agar (55°C) and prepared in a similar manner to that described previously for plant preparation (Hooper et al., 1979). For TEM negative staining, bacteria were obtained from NA or NB cultures, or from expressed plant sap and suspended in 0.5 m1 dHZO. A dr0p of the suspension was placed on a parlodion-carbon-coated copper mesh grid and drained off after one minute. A drop of aqueous, saturated uranyl acetate was placed on the grid for 30 seconds, then drawn off. Grids were examined in a Philips 300 TEM operated at 80 kV. RESULTS AND DISCUSSION Isolations With the direct plating method, many bacterial colonies of varying size, morphology and color developed from plant pieces within 48 hours. Similar colony types were observed with plant species from both diseased and symptomless plants. Similar results were also obtained with streaking of expressed sap. Bacterial colonies of varying color, size and morphology developed from expressed sap of diseased and symptomless plants. No differences were observed between NA and the JD-3 medium for either the direct plating method or the expressed sap method. With the dilution plate method, numerous colonies of varying size, color 1 2 and morphology appeared on the 10- and 10- dilution plates from both diseased and symptomless plants within 48 hours. No colonies were evident on the 10'.4 to -5 10 dilution plates by 48 hours (Figure 14). Continued incubation at room temperature for an additional 24 hours revealed numerous uniformly minute 4 (< 0.1 mm) bacterial colonies on the 10_ to 10"5 dilution plates from diseased C-15 plants. After 144 hours the numerous, uniform bacterial colonies on the 38 4 10" to 10‘5 dilution plates increased to approximately 2 mm diameter and appeared yellow, circular and convex (Figure 15 and Figure 16). There were no additional colonies of differing morphology on these higher dilution plates. Colony formation of the yellow bacterium was induced equally on either NA or the JD-3 medium. The yellow bacterium was not observed on the higher dilution plates (10'4 to 10-5) from symptomless plants. At least three isolates of the yellow bacterium were obtained from each of the four sampled states. Of the three isolation techniques, the dilution plating technique proved to be the only effective method for consistent isolation of high populations of the uniform (yellow) bacterium. Because contaminating epiphytic bacteria grew quite rapidly on all isolation media, techniques entailing plating of plant pieces and streaking of expressed sap were not conducive for isolation of the yellow bacterial pathogen which formed colonies very slowly on agar media. Colony formation by the yellow bacterium could easily be overlooked if petri plates are not incubated for a sufficient length of time. At least 72 hours of incubation at room temperature was necessary before small colonies of less than 0.1 mm diameter are observed and after 96 hours of incubation, colonies were approximately 1 mm in diameter and yellow non-diffusible pigmentation was evident. Inoculations All 20 Toronto creeping bentgrass plants inoculated with a representative isolate of the yellow bacterium obtained by dilution plating from each of the four sampled states developed symptoms approximately five to seven days after inoculation (Table 2). Plants became stunted as growing ceased while affected leaves wilted from the tip to the crown and appeared bluegreen and shriveled (Figure 17). Root and crown regions initially appeared white and in good general health but eventually discolored to brown as they decomposed. Reisolation and Figures 14 and 15. 39 Isolation of the bentgrass bacterium by dilution plating of macerated diseased crowns. (14) No bacterium with consistent color and growth habit were visible in high populations during 48 hours incubation. (15) By 144 hours, a bacterium with consistent color and growth characteristic had becomesquite distinct on the higher dilution plates, 10_4 and 10- . 40 Toronto Cree in: — m mm / .I‘ i It? . 41 Figure 16. The bentgrass bacterium exhibited circular, convex and mycoid colgny formation with pale yellow pigmentation on the 10 and 10 nutrient agar dilution plates. 42 43 Table 2. Inoculation tests utilizing various inoculation times, injured and noninjured plants, bacterial isolates from four states, and host range. Test Host 96 Symptom Developmentsl Inoculation Timeb 'Toronto' (5 plants/time) 5 minutes 100 15 minutes 100 30 minutes 100 60 minutes 100 120 minutes 100 Noninoculated control 0 Leaf Injury 'Toronto' 10 injured plants 100 10 noninjured plants 40 10 injured (noninoculated) 0 10 noninjured (noninoculated) 0 Bacterial Pathogenicity 'Toronto' (5 plants/isolate) Illinois isolate 100 Michigan isolate 100 Ohio isolate 100 Wisconsin isolate 100 Noninoculated control 0 Host Range 10 seedlings 'Penncross' 0 10 seedlings 'Penneagle' 0 10 seedlings Egg annua 0 10 seedlings 'Toronto' 100 Noninoculated control (5 seedlings/cultivar) 0 aBluegreen leaf wilt after five to ten days. Time of exposure to bacterial suspension in shaked flask. Figure 17. 44 Uninoculated control and inoculated Toronto creeping bentgrass plants. Several days after inoculation with the bentgrass bacterium Toronto creeping bentgrass developed characteristic blue-green, wilt symptoms (arrows) similar to those observed in the field. 45 .g\\ ,¢}// ("7 ( CONTROL", INOCULATED G 46 inoculation using similar procedures resulted in the isolation of the same bacterium and the same symptom development. Noninoculated control plants remained symptomless through all repetitions of these experiments and the yellow bacterium was not isolated from any of these control plants. Varying inoculation times resulted in symptom development for all Toronto creeping bentgrass regardless of inoculation period (Table 2). Four of the 10 noninjured plants became diseased while all of the injured plants developed symptoms (Table 2) indicating that injury is apparently not necessary. Perhaps plants were injured by shaking or large numbers of bacteria entered through stomates. Isolates of several commonly-isolated bacteria other than the frequently- isolated yellow bacterium were obtained by the direct plating or expressed sap isolation methods and did not result in symptom production when the previously- described inoculation procedures were used. Hygrothermograph environmental data which coincided with disease development in the field at two locations had been recorded previously and were used in this study as realistic environmental parameters for artificial inoculation. In the field, disease development is apparently initiated by periods of high relative humidity, particularly after precipitation, followed by periods of lower humidity and warm temperatures. As with field observations, artificial inoculations have shown similar symptom development when periods of high relative humidity (> 9596) were followed by periods of lower relative humidity (5096) and high temperature (28-30°C) in the greenhouse environment. Leaf blade injury was also viewed as realistic for artificial inoculations because 'Toronto' putting greens are mowed daily through the summer months. Early morning mowing probably encourages the spread, survival and efficient infection of the turfgrass by the bacterium under periods of high humidity. 47 Host Range A. palustris cv. Penneagle, A. palustris cv. Penncross and 1393 321313 did not develop symptoms even after two weeks (Table 2). Toronto creeping bentgrass, used as a control, became diseased after inoculation with the yellow bacterium. All non-inoculated control plants remained symptomless. Hence, host range studies with two other common cultivars of creeping bentgrass, and the turfgrass invader, 3% w, resulted in similar conclusions to those observed in field situations (see Section I). The bacterial disease appears specific for 'Toronto' and does not appear to affect other turfgrasses. The specificity of this bacterium is important because many locations which grow Toronto creeping bentgrass have been invaded by annual bluegrass or are being converted to other cultivars of creeping bentgrass. Unless the bacterium changes in its specificity, I would not expect these other turfgrass species or cultivars to be affected by this disease. Electron Microscopy Negative stains showed that the isolated bacterium was identical in size (0.3-0.5 x 1-1.15 u m) and morphology (rippled cell wall) to the bacterium from expressed plant sap (Figure 18). TEM ultrathin sections of the isolated bacterium embedded in agar (Figure 19) also showed similar morphology (rippled cell wall) and size to the bacterium observed in diseased plants from the field (see Section I). TEM and SEM of inoculated Toronto creeping bentgrass confirmed that with artifically—inoculated turf stolons, bacteria were limited to xylem vessels in diseased plants (Figures 20 and 21) and resembled those observed in plants from the field. Wilt characteristics are presumably due in part to the physical presence of bacteria or the excretion of polysaccharide and/or toxic substances 48 Figure 18. Transmission electron micrograph of single negatively stained bentgrass bacterium isolated from diseased 'Toronto.‘ Isolated bacterium exhibited similar convoluted surface (arrows) and size to those bacteria observed in negative stain of expressed plant sap. 49 50 Figure 19. Transmission electron microscopy of a bentgrass bacterial isolate embedded in agar. Longitudinal ultrathin section of the isolated bacterium showed similar size and morphology to ultrathin sections of bacterium observed in diseased 'Toronto.‘ 51 Figure 20. 52 Artificial inoculation of Toronto creeping bentgrass with the frequently-isolated bacterium. Transmission electron micrograph of ultrathin sections confirmed infection by bacteria (B) in the plant xylem (X) similar to natural infection in the field. 53 54 Figure 21 and 22. Scanning electron microscopy confirmed infection by the bentgrass bacterium in artificial inoculations. (21) Artificially-inoculated plant xylem (X) with bacteria (B). (22) Uninoculated control plant xylem (X). 55 56 (Dimond, 1970). Bacteria were not observed in noninoculated control plants (Figure 22). The consistent isolation of high populations of a yellow bacterium from diseased Toronto creeping bentgrass from several locations in Ohio, Illinois, Michigan and Wisconsin, and subsequent inoculation studies showed that the bacterium caused symptoms similar to those observed in the field, proving that the yellow bacterium is the cause of the disease. Growth characteristics and biochemical tests (Section IV) provided assurance that the yellow bentgrass bacterium was consistent throughout all isolations and artificial inoculations. The rippled cell wall, small size, xylem limitation and slow growth of the yellow bacterium were suggestive of a RLB (Davis, 1981b). However, growth on NA strongly indicates that the yellow bacterium is not a fastidious RLB. Taxonomic characterization of the yellow bacterium is discussed in Section IV. Exhaustive evidence derived from consistent association, electron microscopy comparisons, host range, consistent isolations and pathogenesis has proven the bacterial etiology of the C-15 problem. This is the first report of a turfgrass disease caused by a bacterium. A suggested name for the disease is "bacterial wilt" of Toronto creeping bentgrass. SECTION IV CHARACTERIZATION OF THE BACTERIUM CAUSING A VASCULAR WILT DISEASE OF TORONTO CREEPIN G BENTGRASS Bacteria are identified and classified according to various morphological, physiological and pathological analyses (see Schaad, 1980; Buchanan and Gibbons, 1974). Gram reaction, pigment production, motility, flagellation, size, morphology, and oxygen requirements are some of the most common criteria for classifying bacteria into broad groups. Utilization of various nutritional substrates for growth and pathogenicity on different hosts are more specific tests for differentiating related bacteria. Serological methods have been used for rapid and fairly accurate identification and classification of bacteria in recent years (Schaad, 1978; Jenkins et al., 1966; Davis, et al., 1981b; Elrod and Braun, 1947). Serology is of limited use, however, due to the unavailability of antisera. Other characteristics also aid in classifying bacteria. Growth at various temperatures and generation time can also be important in identifying or classifying a particular bacterium (see Buchanan and Gibbons, 1974). Ultrastructural examination of the bacterial cells and/or infected plants cells can also be of benefit. The bacterium causing the C-15 problem (bacterial wilt of Toronto creepiong bentgrass) as determined in previous sections of this research showed similar size, morphology and xylem limitation as observed with RLB diseases (Goheen et al., 1973; Hopkins et al., 1973a; Kitajima et al., 1975). 57 58 However, the non-fastidious nature as indicated by growth on a common microbiological medium (NA) suggested the bacterium was not an RLB. The purpose of this portion of the research was to determine the taxonomic affiliation of the bentgrass bacterium in relation to other common plant pathogenic bacteria. MATERIALS AND METHODS Physiological and Morphological Tests Unless otherwise stated, all tests were performed according to methods described in the Laboratory Guide for Identification of Plant Pathogenic Bacteria (see Schaad, 1980). The following tests were conducted for general delineation of several isolates of the bentgrass bacterium: Gram reactions, oxidase, catalase, oxygen requirement, pigment production, and motility. More specific tests included: hydrogen sulfide production, starch hydrolysis, casein utilization, gelatin liquefaction, growth on SX and YDC agars, phenol red dextrose, growth at 36° and 41°C, and hypersensitive reaction on tobacco (see Schaad, 1980). All tests were conducted using identified bacteria as positive and negative controls, supplied by Dr. Dennis W. Fulbright, Dr. Alfred Saettler and Carol Ishimaru, Department of Botany and Plant Pathology, Michigan State University. These known bacteria are listed in Table 3. Motility tests were accomplished by (1) light microscopy of bacteria suspended (gently) in 0.01 M phosphate buffer and (2) TEM negative stains (H00per et al., 1979) of bacteria from liquid (NB) and solid (NA) media. Pigment and extracellular polysaccharide production were determined on NA and yeast extract-dextrose calcium carbonate agar (YDC - see Schaad, 1980). Oxygen requirement was. determined by use of a thioglycolate broth (Difco), and the method of Hugh and Leifson (1953). 59 1 Table 3. Representative plant pathogenic bacteria used as positive and negative controls in all tests for characterization of the bentgrass bacterium. Corynebacterium michiganense pv. michiganense Agrobacterium radiobacter var. tumefaciens Biovar I Pseudomonas marginalis Pseudo monas syringae pv. sgingae Pseudomonas solanacearum Erwinia amylovora Erwinia chrysanthemi Erwinia carotovora subsp. atroseptica Erwinia carotovora subsp. carotovora Erwinia herbicola (1124) Erwinia herbicola (C9—1) Xanthomonas campestris pv. lar onii Xanthomonas campestris pv. haseoli (612) Xanthomonas campestris pv. phaseoli (var. fuscans 19124) t Isolates provided by Dr. Dennis W. Fulbright, Dr. Alfred Saettler and Carol Ishimaru. 60 Ultrastructural Analysis For ultrastructural analysis, diseased plant samples were prepared for TEM ultrathin sectioning as described by Hooper et a1. (1979). After preparation and ultrathin sectioning, samples were observed with high magnification on either a Philips 300 or a Philips 201 transmission electron microscope. Structural details of the bacterial cell wall and cyt0plasmic contents were examined. Growth Studies Three isolates of the bentgrass bacterium were used in all growth studies. Side arm flasks (300 ml) containing 50 ml of NB were steam sterilized (15 minutes at 30 psi) and inoculated with pure colonies of bacteria from NA. Flasks were incubated on a Eberbach shaker (180 cycles/minute) at 5°C, 20°C, 25°C or 30°C. At various times throughout the growth cycle, optical densities (O.D.) were recorded using a Bausch and Lomb Spectronic 20 Spectrophotometer operated at a wavelength of 520 nm. Results were recorded and graphed. Other studies were conducted for standardization of bacterial populations. Nutrient broth was inoculated with cultures of the bentgrass bacterium and allowed to grow to an optical density of approximately 0.8, ending the exponential phase of growth. This concentration was diluted to 3/4, 1/2, 1/4, 1/8, 1/10, 1/50, 1/100 using NB. Absorbance at 520 nm was recorded for each dilution, and the nondiluted bacterial suspension and the 1/10 and 1/100 dilutions were each serial diluted in 0.01 M phosphate buffer to 10-10. One-tenth ml of each dilution was plated on NA plates and incubated at 25°C for six days at which time colony forming units (CFU) were recorded. Another standardization test was conducted in 0.01 M phosphate buffer (pH 7.2). Bacterial suspensions in NB (O.D. = 0.8) were pelleted at 2000 g for 10 minutes. The pellet was resuspended in 0.01 M phosphate buffer and diluted to 1/2, 1/4, 1/10, 1/50 and 1/100. Optical densities at 520 nm were recorded for 61 each suspension and the undiluted, 1:10 and 1:100 were each diluted from 10"1 to --10 10 and plated in a similar method to the NB studies. In all growth studies, an isolate of Xanthomonas camestris pv. Elargonii was used as a comparative control. Serology In preparation for serology, bacterial cultures # 7 and # 8 (both isolated from St. Charles, Illinois) were streaked from storage onto NA plates. Flasks containing 50 ml of NB were inoculated with each of the two cultures and shaken on an Eberbach shaker (160 cycles/minute) for 72 hours or until cloudy. Bacterial cells were pelleted for 10 minutes at 4000 xg, and the supernatant was removed. Bacterial cells were resuspended in 0.85% saline and adjusted to at least 108 CFU/ml as determined by optical density. A New Zealand white female rabbit weighing approximately 7.0 pounds was injected intramuscularly with 0.5, 1, 1.5, 2 and 3 ml on days 1, 7, 14, 21 and 28 respectively with an equal volume of the whole cell bacterial suspension and Freud's incomplete adjuvant. Normal serum was collected prior to injection of bacteria. Serum (50 ml aliquots) was collected three times from the rabbit on days 35, 42 and 49. After incubation at 37°C for two hours and storage in a refrigerator (5°C) for 24 hours, antiserum was removed from the blood clot with the aid of a pipette and frozen without preservatives until needed. In prepraration for serology all known bacteria (Table 3) and isolates of the bentgrass bacteria from several states were grown in NB shake flasks until cloudy, pelleted at 4000 xg for 15 minutes, and resuspended in 0.85% saline with 0.025% sodium azide to an O.D. of approximately 1.0 at 520 nm. All suspensions of bacteria were sonicated for two minutes in a Heathkit sonicator (Model GD- 1151) at 41 kHz. Ouchterlony double gel diffusion was employed as the serology test procedure (see Jenkins, 1966). Agar plates (10 ml) 62 were prepared with 0.5% Colab Ion Agar ZS and 0.025% sodium azide. After solidification 6 mm wells were cut in the agar plates using a Feinberg cutter that left six wells surrounding and 3 mm distant from a center well. Sonicated test samples were added to the outer wells and a 1:32 dilution of the antiserum was added to the center well. Precipitation lines between the peripheral and center wells showed a positive reaction and were visible within 48 hours. All tests were repeated at least three times. Bentgrass bacterial isolates were also fixed on glass slides with ethanol and sent to Dr. Michael J. Davis, University of Florida, for indirect immunofluorescence comparison with RLB antiserum (Davis et al., 1981% RESULTS AND DISCUSSION Physiological and Morphological Tests Generalized tests of the bentgrass bacterium are compared to the five commonly encountered genera of plant pathogenic bacterium in Table 4. Negative Gram reaction and oxidase tests suggest that the bentgrass bacterium is not a member of the genus Corynebacterium or most Agrobacterium spp. Oxygen requirement was determined by both the thioglycolate and Hugh-Leifson (1953) methods and suggested that the bentgrass bacterium was not related to the genus Erwinia which is composed of facultative anaerobes. The bentgrass bacterium produces a nondiffusible pale-yellow pigment on nutrient agar and thus was differentiated from the pseudomonads and agrobacteria which are primarily oxidase positive and usually produce no yellow pigment (see Schaad, 1980 and Buchanan and Gibbons, 1974). Those Pseudomonas species that are oxidase negative usually produce a fluorescent pigment (see Schaad, 1980). The bentgrass bacterium was catalase positive and hence similar to most plant pathogenic bacteria. Of the five genera of plant pathogenic bacteria, the bentgrass bacterium appeared to be most closely related to the genus 6235 88: 82695:: 93 52m 338525. .3 8553er .32 .228 meme + + + + + @33qu 2263365: oEmmstEcoc Bozo» 1 3:88.83.“ 2:98 Bone» :oflwucoEmE 30.598 0223 onosoa “333303 30.85 5:869:2qu 5905 t u .+ I + .. $8me 1 .. n u + .. :ozowom EEC dam dam dam dam dam E32325 «mob $25805an mwcoEocsomm SEEM 832303056. Estouownofifoo 8295i «.383 Essencesleeaseifi Epocow 3 «2325 Ecowofiwa «5.3 no 205w ozuwucomoaaou 05. 532325 $29.55 05 no :ofloaom é 2an 64 Xanthomonas (Table 4). More specific tests also suggest similarity to Xanthomonas (Table 5). Hydrogen sulfide production, starch hydrolysis, casein utilization, and gelatin liquefaction are important similarities of the bentgrass bacterium and the xanthomonads. Growth on a dextrose medium (YDC) causes bright yellow colony formation with large amounts of slime (presumably extracellular polysaccharide) production and is characteristic of both microorganisms. The bentgrass bacterium also grows at 36°C but not at 41°C, and does not grow on SX agar (Table 5). These same characteristics are variable among the xanthomonads (see Schaad, 1980). Stabs of phenol red dextrose changed the tubes to deep red at the top with no growth in the stab for both the bentgrass bacterium and Xanthomonas indicating that neither were fermenters. Many tests for comparison of the bentgrass bacterium with known bacterial species were difficult to discern due to the slow growth of the bentgrass bacterium. Generally the reactions of the bentgrass bacterium on various media were 48 to 36 hours after reactions of the other known bacteria used as positive and negative controls. One characteristic of the bentgrass bacterium which differed from the xanthomonads was motility (Table 5). Xanthomonas is characterized as being motile by a single polar flagellum (see Schaad, 1980 and Buchanan and Gibbons, 1974). The bentgrass bacterium was not motile as ascertained by light microscopy of the bentgrass bacterium in 0.01 M phosphate buffer from either solid or liquid media. Brownian motion was easily detected but no directional movement of bacterial cells was observed. Also, no flagella or remnants of flagella were observed with negative stain TEM (see electron micrograph - section III). Perhaps the bentgrass bacterium is nonmotile as a result of mutation or motility may exist for a brief period in the life cycle of the bacterium. 65 Table 5. Reaction of the bentgrass bacterium and Xanthomonaas spp. to specific physiological-biochemical and morphological tests. Test Bentgrass Bacterium Xanthomonas spp. HZS Production + + Starch Hydrolysis + + Casein Utilization + + Gelatin Liquefaction slow + Growth on YDC yellow, slimy yellow, slimy Phenol Red Dextrose red at top red at top Growth on SX Agar - +, - Growth at 36° + + - Growth at 41° - - Motilityb non motile motile , single flagellum HSRc .. _ _ aSee Schaad, 1980. Determined by TEM negative stain and light microscopy. cDetermined on tobacco. 66 Ultrastructural Analysis TEM ultrathin sections provided an opportunity for fine structure detail. Close observation of the internal composition of the bentgrass bacterium revealed cytOplasmic strands which were characteristic of DNA (Figure 23). The small electron-dense spheres were typical of ribosomes. Close examination of the cell wall of the bacterium revealed an inner and outer membrane which is characteristic of Gram negative bacteria, thus confirming the Gram stain results. These studies confirm that the bentgrass bacterium is very similar in ultrastructure to Gram negative bacteria (see Joklik and Willett, 1976). Growth Studies Standardization of bacterial populations is crucial for any experimental study. Table 6 reviews the optical densities and number of colony forming units in relation to various dilutions of bacteria in susupensions of either NB or 0.01 M phosphate buffer. NB suspensions with Optical densities of 0.014, 0.139 and 0.880 7 suggest approximately 8.5 x 10 , 7.4 x 108 and 5.0 x 109 CFU per ml, respectively. In buffer, O.D. of 0.04, 0.35 and 1.75 relates to approximately 2.1 x 108, 2.2 x 109 9 and 9.0 x 10 CFU/ml. Dilutions, O.D. and CPU exhibit a linear relationship from O.D. of 0.01 to approximately an O.D. of 0.5. In other words, for a ten-fold decrease in dilution there is an approximately ten-fold increase in O.D. and CFU. This information may be used to easily obtain 105 or less CFU/ml by simple dilution. The number of CFU is important for determining approximate bacterial populations prior to such studies as pathogenicity and serology. Figure 23. 67 Ultrastructural characteristics of the bentgrass bacterium. The cell wall possessed an inner membrane (IM) and outer membrane (OM), typical of Gram negative cell walls. Cytoplasm contained DNA-like strands (DNA) and electron dense spheres resembling ribosomes (R). 68 69 Table 6. Optical densities (O.D.) and numbers of colony forming units (CFU) of three isolates of the bentgrass bacterium at various dilutions in nutrient broth (NB) and 0.01 M phosphate buffer. Dilution 0.1). at 520 nma CFUblml NB 0.000 7 1/100 0.014 8.5 x 10 1/50 0.033 8 1/10 0.139 7.4 x 10 1/8 0.164 1/4 0.309 1/2 0.560 3/4 0.750 9 Undiluted 0.880 5.0 x 10 Buffer 0.000 8 1/100 0.040 2.1 x 10 1/50 0.078 9 1/10 0.350 2.2 x 10 1/4 0.750 1/2 1.250 9 Undiluted 1.750 9.0 x 10 aAbsorbance at 520 nm on a Bausch _ Plating of dilutions serially from 10 arid Lom pectronic 20. to 10b1° 70 Figures 24 through 26 review the growth of the bentgrass bacterium compared to a Xanthomonas campestris pv. pelargonii isolate at temperatures of 5°, 20°, 25° and 30°C in nutrient broth shaker flasks (160 cycles/ minute). Both bacteria grew at 30°, 25° and 20°C but neither grew at 5°C. The bentgrass bacterium and the Xanthomonas isolate grew equally well at 20° and 25°C (Figures 24 and 25) but the bentgrass bacterium grew more rapidly at 30°C as indicated by a separation of the plots in Figure 26. Because a linear relationship exists with dilutions, absorbance and CFU (Table 6) at absorbance readings between 0.1 and 0.5, the growth curves in Figures 24 to 26 can be used to determine approximate generation times (times for doubling of the population). Approximate generation times as measured by the time for absorbance to change from 0.2 to 0.4 (in the exponential phase of growth) are presented in Table 7. Generation times for the bentgrass bacterium and the Xanthomonas isolate are each 8.5 hours and 4.5 hours at 20°C and 25°, respectively. At 30°C, however, the bentgrass bacterium appears to grow more rapidly with a generation time of 5.0 hours versus 7.5 hours for the xanthomonad. Bergey's Manual (see Buchanan and Gibbons, 1974) reports that all xanthomonads grow at 30°C, some at 36°C, and none at 5°C, with optimum growth at 25 - 27°C. Growth of the bentgrass bacterium is very similar to the Xanthomonas isolate except near 30°C which is apparently closer to the optimum growth temperature of the bentgrass bacterium. Serology None of the commonly-encountered plant pathogenic bacteria reacted with the antiserum produced to bentgrass bacterial isolates # 7 and # 8 (Table 8). Isolate # 9 from Illinois location A and isolates # 20, # 21 and #22 from Illinois location B reacted very strongly with the antiserum. In addition, isolates from each of the other three states also reacted equally well to the antiserum Figures 24 - 26. 71 Growth of the bentgrass bacterium and Xanthomonas cam estris pv. elar onii as measured by Optical density absorbance at 520 nm during incubation in nutrient broth shaker flasks at 20°C, 25°C and 30°C. Neither bacterium grew at 5°C. (24) X. campestris pv. pelargonii and bentgrass bacterium at 20 C. (25) X. campestris pv. pelargonii and bentgrass B—acterium at 25 C. (26) X. campestris pv. pelargonii and bentgrass Bacterium at 30 C. RBSORBRNCE '- [520 abssaepwi NM) 72 20°C E] I. CRHPESIIIS PV. PELRIOONII ‘ BENIORRSS ORCTEIIUH 24 —o ‘T—F T T T I I T I T‘ T I O 4 8 12 16 20 24 28 32 36 4O 44 48 HOURS ,) 25°C 1.4 D X. CRHPESYRIS PV. PELRROONII / ‘ BENTORRSS BRCIERIUH ,..4 (3 (‘1 CI‘ Ls lg, ax _ __r—“T ’"T I- “FT-f I T I l I T 4 8 12 16 2O 24 28 32 36 4O 44 48 OW _L_¢._. L ._L.._-L————L——_L___J———4L~—._J—— \ RX \ 25 run-au- HOURS 1 1,_ 30°C ‘— m X. CRHPCSTRIS PV. PELRROONII ‘ 8041051935 BRCIERIUH 1.0“ J 1/ ,/ {a / 4-4 /’ .1 fl/ 1/ / ,/ x« .1 // g,// 1/ 4V ‘ .c' qy‘fT ' I T T ‘V T T T T T T T T U 4 8 12 16 20 24 28 32 36 4O 44 48 HOURS 26 73 Table 7. Approximate generation time of the bentgrass bacterium 3nd Xanthomonas campestris pv. pelargonii in nutrient broth at 5 C, 20°C, 25°C and 30°C. Generation Tim_ea - Hours Temperature Bentgrass Bacterium _X. campestris pv. pelargonii 5°C no growth no growth 20°C 8.5 8.5 25°C 4. 5 4. 5 30°C 5.0 7.5 aTime for population to double as measured by change in absorbance from 0.2 to 0.4 at 520 nm wavelength on a Bausch and Lomb Spectronic 20. 74 6.82:5: ma wouacwmmoo 333v: 9:530 .3 93 S» 32:02 83:30:: 889:3 3 83:93 EEoflE< =.+= m: 08.83.. 33360.5 no :ofiafiuom o : a + 5282: Se .34 a: + 2.6 a; .34 a; + 59:22 3* £2 a: + N .8383 £055 a: J: a: + fl 5383 $25: 2 .3, .2 o=Eatouoam 829:3 GNHS 2:33 .33 moons—E <3 £58956 ma:oEo:u:aN 3H8 momma—E Sm 2.58““an m::oEofi::N a: .5 Sn $.58“:an 3:050:23" 3&9 283.8: 2:255 35 3853; SEE: 23328 .amozw 23828 £55m mozmomobw dwnsm 29,328 £55m wEozecsflEo SEEM 206%,: 3535 Ezemooafiflom m::oEoc:omm oawmmmwm 5a owwctwm mgoEonsomm £85»..me mm:oEocsomm H .885 288: 2:3 33> $83062 Es_amzémfiosmeV 3:98 ESE .3 0228922.: 822888580 a £53335. 8 :oflowom .5383 Eztououm “woe 6:823 2:omofiam «53 5558 Ho 83on 33553.59. was 62:30:: annua—8n 93 .«o commit: 03:0: 8» .3 mini—3 308223 .m 233. 75 Figures 27 and 28. Serological analysis by agar double diffusion of the bentgrass bacterial isolates (numbered 7, 20, 30, 40, 50) and known plant pathogenic bacteria. Precipitate lines indicate positive reaction of antiserum (A/S) dilution 1/32 (center well) to bacterial test isolate (outer wells). Figure 27: # 7 = Illinois isolate # 20 = Illinois isolate # 30 2 Michigan isolate # 40 = Ohio isolate # 50 = Wisconsin isolate X.p. = Xanthomonas campestris pv. pelargonii Figure 28: # 7 = Illinois isolate C.m. = Corynebacterium michiganense pv. michiganense A.t. = Agrobacterium radiobacter var. tumefaciens Biovar I E.h. = Erwinia herbicola P.s. = Pseudomonas solanacearum X.p. = Xanthomonas campestris pv. pelargonii 77 produced to culures # 7 and # 8 (Table 8). Precipitate lines were clearly evident with each bentgrass bacterial isolate within 48 hours (Figure 27), whereas no precipitate lines developed with any of the common plant pathogenic bacteria even after five days (Figure 28). The antiserum dilution 1:32 was used in all tests because this concentration provided narrow and precise precipitate reactions between wells. Preliminary tests with higher concentrations of antisera resulted in very wide, vague bands that surrounded the antigen (outer) wells. No reaction occurred with the normal antiserum that was collected prior to injection of the rabbit with the antigens. Even though the physiological-biochemical analyses suggested that the bentgrass bacterium was similar to the xanthomonads, these serology results suggest that the bentgrass bacterium is not related antigenically to three Xanthomonas isolates nor any of the other genera of plant pathogenic bacteria used in this test. The double gel diffusion technique used in this study was a good test as evidenced by the strong reactions of the bentgrass pathogens from other states to antiserum produced to two Illinois isolates. The lack of reactions with other known bacteria may indicate unrelatedness or may suggest that the antiserum is very specific. Antisera are frequently produced which differentiate among species of a bacterial genus (Elrod and Braun, 1947). Several bentgrass bacterial isolates that were fixed on slides and sent to Dr. M. J. Davis, University of Florida, did not show any relationship to RLB as determined by indirect immunofluorescence (Davis et al., 1981b). Various physiology and morphological tests suggest that the bentgrass bacterium is similar to the genus Xanthomonas in most criteria except motility. Ultrastructural analysis was characteristic of many Gram negative bacteria. Growth at various temperatures and generation time are also similar to 78 Xanthomonas spp. (see Buchanan and Gibbons, 1974). Serology implies an unrelatedness, however, the possibilities of exceptionally specific antiserum or other xanthomonad serotypes should not be overlooked. Further analysis regarding DNA hybridization and GC content (DeLay 1968), and pigmentation (Starr et al., 1977) would greatly enhance the identification of the xanthomonad- like bentgrass bacterium. 79 SECTION V SYMPTOM SUPPRESSION OF BACTERIAL WILT WITH OXYTETRACYCLINE In recent years, red leaf spot caused by Helminthosporium erythrospilum (Meyer and Turgeon, 1975), and leaf blight and crown rot caused by Drechslera (= Helminthosporium) catenaria (Larsen, 1979; Larsen et al., 1981) provided solutions to the C-15 problem. Both diseases were reported to be managed by common fungicide application, however, the recommended fungicides were not effective at the Butler National Golf Course and other locations in 1980 and 1981. The initial phases of the present research showed that a bacterium was the cause of the C—15 problem and that the disease was widely distributed. Other bacterial diseases of plants have been effectively suppressed by chemical treatment (Hopkins and Mortensen, 1971 and HOpkins et al., 1973b). Thus, the purposes of the final portion of this research were to ascertain the sensitivity of bacterial wilt to several chemical treatments and to develop effective control measures to manage the disease. MATERIALS AND METHODS Field trials were established at two Illinois locations that had experienced the C-15 problem the previous year - the St. Charles Country Club in St. Charles, Illinois (location A) and Village Links Golf Course in Glen Ellyn, Illinois (location B). The test chemicals were oxytetracycline (Mycoshield: 1796 WP, Pfizer Chemicals Division), streptomycin sulfate (Agrimycin 17: 21.296 WP, 80 Pfizer Chemicals Division), and cupric hydroxide (Kocide 101: 7796 WP, Kocide Chemical Corporation). At location A, 1.0 and 1.5 g/L solutions of each chemical were applied as drench treatments with a sprinkling container at the rate of 2 L/m2 (five 2 gallons/100 feet2) to 0.93-m plots of turf. At location B, 1.0 g/L solutions of the three test chemicals were applied at the rate of 2 L/m2 (five gallons/100 feetz) to 1.86-m2 plots. A randomized complete block design (including untreated check plots) with three replicates was used at each location. At location A, two applications were made one week apart starting 20 May 1981, as symptom development was becoming apparent. At location B, applications were made weekly for six weeks starting 5 May 1981, before symptoms were evident. In an experiment at location B that was not replicated, one application of 1 g/L solutions of oxytetracycline and streptomycin were each applied 23 April 1981 to adjacent halves of a golf green at the rate of 2 L/m2 (five gallons/100 feetz). Disease severity was assessed 25 June 1981 after a severe outbreak of disease. Estimates of the percentage of disease on a scale of 1 (0-1096) to 10 (91-10096) were made of all treatments. The data were statistically analyzed by analysis of variance and Duncan's multiple range tests. For scanning electron microscopy (SEM) studies, three Toronto creeping bentgrass plants were selected from each treatment and washed in running tap water for two hours. Crowns were removed, fixed in 4% glutaraldehyde, postfixed in 1% osmium tetroxide, dehydrated in a graded series of ethanol, and freeze-fractured in liquid nitrogen (Hooper et al., 1979). Following critical point drying and sputter-coating, crown pieces were observed in a JEOL JSM-35C seaming electron microscope. 81 The E! 2129. sensitivity of the bacterium to oxytetracycline and streptomycin was tested using a laboratory bioassay. Each of the antibiotics was added to molten (55°C) nutrient agar (NA) to obtain final concentrations of 0.001, 0.01, 0.1, 1.0, 10.0, 100.0 and 1000.0 ug antibiotic per ml of agar. Twenty ml of antibiotic amended agar were added to 9 cm standard culture dishes. Three bacterial isolates were adjusted in 0.01 M phosphate buffer (ph 7.2) to obtain 10 to 100 colony-forming units (CFU) per non-amended agar control dish. One-tenth m1 of each bacterial suspension was placed on each of the antibiotic- amended agar concentrations and spread with a sterile L—shaped glass rod. Growth as evidenced by colony formation after six days of incubation at 25°C was recorded as positive. RESULTS AND DISCUSSION Excellent symptom suppression of bacterial wilt of Toronto creeping bentgrass was achieved with oxytetracycline. At location A, two applications of oxytetracycline at 1.0 and 1.5 g/L significantly (P = 0.01) reduced the severity of the C-15 problem when compared with the untreated check, streptomycin, and cupric hydroxide treatments (Table 9). Small areas of decaying turf and many individual wilted plants with characteristic shriveled, blue green leaf tips were noted in the check, streptomycin, and cupric hydroxide plots, whereas all plants in the oxytetracycline plots remained healthy with no symptom expression (Figure 29). Streptomycin and cupric hydroxide test plots were not significantly different from check plots, and no differences were noted between the 1.0 and 1.5 g/L levels of any treatment. A marked reduction in disease severity was also obtained at location B (Figure 30) after six applications of oxytetracycline (Table 9). Streptomycin sulfate and cupric hydroxide did not reduce disease severity. In the single application experiment at location B, disease deveIOped on the streptomycin- 82 Table 9. Effects of two antibiotics and a copper compound on a disease of Toronto creeping bentgrass. Disease ratingz Location Location A B Treatmenty 1.0 g/L 1.5g/L 1.0 g/L Oxytetracycline 1.0 a 1.0 a 1.0 a Streptomycin 6.0 b 5.6 b 3.6 b Cupric hydroxide 5.6 b 5.3 b 4.3 b Check 5.6 b 4.6 b yTwo treatments at location A and Sig at location B were applied as 1.0 g/L and 1.5 g/L solutions at the rate of 2 L/m . zOn a scale of 1 (0-10%) to 10 (91-10096). Each number represents the mean of three replicates. Means followed by different letters are statistically significant according to Duncan's multiple range test at P = 0.01. Locations A and B were analyzed separately. 83 Figure 29. Oxytetracycline (OT) at 1 g/L suppressed disease symptoms after two applications whereas disease developed in the streptomycin sulfate (1.5 g/L) treated plot (ST) at Location A. 84 Figures 30 and 31. 85 Toronto creeping bentgrass treated with antibiotics at location B. (30) Oxytetracycline (OT) at 1.0 g/L suppressed development of disease compared to the nontreated check CK). (31) After single applications (1 g/L at 190 L/93m ) of each antibiotic to adjacent halves of a golf green, disease developed on the streptomycin- treated (ST) half, whereas no diseased developed on the oxytetracycline-treated (OT) half even after six weeks. 86 87 treated half of the green six weeks after the single application; however, complete symptom suppression was observed on the oxytetracycline half of the green (Figure 31). Because antibiotic applications were made after symptom development at location A but before symptom development at location B, both curative and preventive control appeared to be possible with oxytetracycline. The drench method was used because previous 1980 field tests with light spray applications (0.6 g/L at 0.2 L/mz) proved ineffective. In the present study, oxytetracycline at 1.0 g/L was as effective as at 1.5 g/L. Although these high rates have given excellent control, lower rates may be sufficient for control of this disease. Streptomycin and cupric hydroxide had no effect on the disease at either rate. Some yellowing of turf in oxytetracycline-treated areas was noted when the chemical was applied during the high light-intensity portion of the day when temperatures exceeded 23°C. The yellowing was not evident on plots treated in ' the early morning or late afternoon. Examination of bentgrass plants by EM revealed many bacteria in xylem vessels of all plants from untreated check plots (Figure 32). Similarly, bacteria were also found in xylem vessels in plants from the plots treated with cupric hydroxide and streptomycin sulfate. No bacteria were found in any plants from the oxytetracycline-treated plots (Figure 33), and no differences were discerned between any of the treatment levels. The I absence of bacteria in oxytetracycline-treated plants, as ascertained by SEM, presumably resulted from antibiotic effects of oxytetracycline upon the bacteria. The laboratory bioassay showed greater sensitivity of the bentgrass bacterium to oxytetracycline than to streptomycin. Growth of three bentgrass bacterial isolates, as indicated by colony formation, was inhibited by 0.1 ug/ml oxytetracycline and by 10 u g/ml streptomycin sulfate (Table 10). These results Figures 32 and 33. 88 Scanning electron microscopy of crown freeze-fractures of Toronto creeping bentgrass. (32) Xylem vessels from untreated plant containing many bacteria (B). (33) Vascular tissue from plant treated (1.0 g/L) with oxytetracycline. Xylem vessels (X) contain no bacteria. 89 90 Table 10. Colony formation of the bentgrass bacterium on antibiotic amended nutrient agar. Antibiotic conc . u g/ ml Streptomycin Oxytetracycline Check 0 . 001 + + + 0 . 01 + + + 0 . 1 + - + 1 . 0 + - + 10 . 0 - - + 100 . 0 - - + 1000 . 0 - - + Appearance of colonies by six days recorded as "+." 91 suggest that the bentgrass bacterium is inherently 100 times more sensitive to oxytetracycline and may explain why the equally broad-spectrum streptomycin sulfate was not effective in the field experiments. Our findings at location A indicated that no more than two applications of oxytetracycline were necessary for disease control for a four-week period. The single application experiment at location B indicated that oxytetracycline appeared to persist and be effective for an extended period of time. Further testing with regard to timing and application frequency may be necessary before practical management schemes are feasible. Because tetracycline antibiotics are frequently used in treating human diseases, precautionary measures must be seriously considered before registration as a turfgrass pesticide is recommended. DISCUSSION AND CONCLUSIONS This research presents conclusive evidence for the bacterial etiology of the C-15 problem which has destroyed Toronto creeping bentgrass for many years. The initial discovery of the association of a bacterium with diseased 'Toronto' at the Butler National Golf Course in Oak Brook, Illinois was important and timely, and suggested a probable solution to the poorly-understood C-15 problem. Subsequent distribution studies showed that the bacterium associated with diseased Toronto creeping bentgrass was not an exclusive association with Butler National but was widely distributed among at least 15 locations in five midwestern states: Illinois, Indiana, Michigan, Ohio and Wisconsin. The wide distribution was crucial because transmission electron micrographs showed similar plant location (xylem), size (approximately 0.3 - 0.5 by 1.0 - 1.5 u m) and morphology (rippled cell wall) to the RLB which are fastidious and hence very difficult to culture. All early reports of diseases caused by RLB (Goheen et al., . 1973; Hopkins et al., 1973a; Kitajima et al., 1975) were based entirely upon association since the first RLB was only recently isolated into pure culture by Davis et al. (1978). Hence the consistent association of a bacterium with the C- 15 problem produced additional evidence for bacterial etiology of the C-15 problem even though the bacterium was later found to be unrelated to the RLB. A bacterium forming circular, mucoid, convex, pale yellow colonies was consistently isolated after at least five days incubation on NA at room temperature. The common turfgrass isolation method of plating plant pieces and 92 93 streaked expressed plant sap were not conducive for isolation of the pale yellow bentgrass bacterium. Using either of these two methods resulted in the rapid growth of saprophytic bacteria and fungi on isolation plates. Only the dilution plating method allowed consistent isolation of high populations of the slow- growing pale yellow bentgrass bacterium from many locations in different states. The specialized JD—3 medium for isolation of RLB (Davis et al., 1978) was not necessary because the pale yellow bentgrass bacterium grew equally well on NA. Proof of pathogenicity by inoculation was initially difficult due to the lack of understanding of pathogenesis of the C-15 problem. Environmental parameter data were collected by recording hygrothermographs in the field. Symptom development in the field occurred during'warm (80° or above), bright periods immediately following periods of high relative humidity (95-10096) or high moisture. Artificial inoculations were quite successful in the laboratory if these environmental parameters were followed. Symptomatology of shriveled blue- green leaf blade wilt, followed by decomposition was typical of those symptoms observed in the field and laboratory. Leaf blade wounding was conducive to disease development and was viewed as realistic since greens are mowed daily during the summer months. The successful inoculation and consistent reisolation (Koch's Postulates) proved that the bentgrass bacterium was the cause of the C-15 problem. Other supporting evidence included morphological comparisons and infection confirmation with the aid of electron microscopy. Symptom suppression of the disease in the field by oxytetracycline also suggested bacterial etiology because antibiotics are widely reported as inhibiting bacterial growth (Goodman, 1954; Hopkins and Mortensen, 1971; Miller, 1959; Nester et al., 1973). Because of symptoms and bacterial etiology, a suggested name for the C-15 disease is "bacterial wilt of Toronto creeping bentgrass." 94 That the C-15 problem remained unresolved for so long was undoubtedly a result of misunderstanding. Virtually all previous reports of turfgrass diseases were of fungal etiology (Vargas, 1981; Couch, 1973). Because no bacterial diseases of turfgrass have previously been reported, the lack of awareness of bacterial incitants was understandable among turfgrass pathogists. The disease was not only atypical in turfgrass pathology, but in some ways did not resemble vascular diseases caused by bacteria in other crops (Agrios, 1978). Generally, there was no "bacterial ooze," and vascular discoloration was difficult if impossible to observe in such a small plant as Toronto creeping bentgrass. Previous reports attempted to provide answers for the C—15 problem by furnishing fungal diseases as solutions (Meyer and Turgeon, 1975; Larsen, 1979; Larsen et al., 1981). These fungal diseases, red leaf spot (Drechslera egthrosfla), and leaf blight and crown rot (Drechslera catenaria), were undoubtedly legitimate, but whether they were of primary importance in the C- 15 problem is impossible to determine. In the first report (Meyer and Turgeon, 1975), no adequate description of the disease was presented. And because consistent, absolute control of the disease was not attained with only the fungicide chlorothalonil, the bacterium presumably was complicating their understanding of the C—15 disease. In the second report (Larsen et al., 1981), symptomatology was described as "red leaf lesions and leaf tip dieback" where plants were eventually "blighted to the crown." Such leaf tip dieback and blighting could be confused with bacterial wilt symptoms. Perhaps the bacterium was also affecting the results in this latter report since absolute (100%) control of the disease was not achieved with iprodione. If red leaf spot (Drechsler, 1935), and leaf blight and crown rot (Larsen, 1979; Larsen et al., 1981) were found to frequently occur with bacterial wilt, then the C-15 problem should be regarded as a "complex" or "syndrome" of diseases. This, however, was 95 not the case. Obviously, for accuracy in reporting new diseases, meticulous isolation and symptom discription are necessary. Symptomatology was thoroughly observed in this study and was helpful in determining the cause of the C-15 problem. Symptom development progressed in a similar manner to many vascular wilt diseases already studied (Dimond, 1970; Buddenhagen and Kelman, 1964). Wilt inducement by the bentgrass bacterium occurred very quickly as evidenced by the appearance of blue-green shriveled leaf tips on a bright afternoon during the summer. The succulent bentgrass plant was observed to recover very quickly in some instances. If environmental conditions were favorable, all leaves may wilt to the crown with eventual destruction of the entire plant. The mechanism of wilting has not been studied in this research. Electron microscopy revealed extraordinarily high populations of bacteria in xylem vessels. Presumably these large populations of bacteria would account for some inhibition of water movement as proposed by Nelson and Dickey (1966) regarding carnation wilt. The present study also found that the bentgrass bacterium was capable of copious production of extracellular polysaccharides (on agar media) which may support the theory proposed by Husain and Kelman (1958). Possibilities of accumulated growth substances (Sequiera and Kelman, 1962) and toxic glycopeptides (Hess and Strobel, 1969) cannot be postulated at this time. The taxonomic affiliation of the bentgrass bacterium is uncertain. Being Gram negative and an obligate aerobe, this bentgrass bacterium certainly belongs in Part 7 of Bergey's Manual (see Buchanan and Gibbons, 1974). Further, physiological-biochemical and morphological examinations revealed that the bentgrass bacterium may possess many xanthomonad-like qualities. Cell size of 0.5 u m by 1-1.5 u m, yellow pigmentation, slow mucoid growth, hydrogen sulfide production, utilization of starch, casein and gelatin are outstanding 96 characteristics of the genus Xanthomonas (see Schaad, 1980). Production of an extracellular polysaccharide on dextrose-containing media is a unique characteristic of both the bentgrass bacterium and Xanthomonas spp. Both microorganisms are also capable of causing similar types of vascular diseases even though different plants are affected. In nutrient broth shake cultures, X. campestris pv. pelargonii and the bentgrass bacterium had similar growth characteristics at 25°C. Generation times were approximately equal at all temperatures except 30°C, where the bentgrass bacterium reproduces more rapidly than the xanthomonad test isolate. Growth on solid media (NA) is far different than in liquid media. Even though Xanthomonas spp. are considered slow growers among the plant pathogenic bacterium, the bentgrass bacterium formed colonies at least 24 to 48 hours after the Xanthomonas isolates used in the physiology-biochemical studies. Because growth in liquid cultures are approximately equal, the slower colony formations of the bentgrass bacterium on agar media is undoubtedly due to non-motility. Motility is of primary importance in bacterial systematics. In every report Xanthomonas spp. are described as being motile by a single, polar flagellum (see Schaad, 1980; Buchanan and Gibbons, 1974). Besides serology, motility was the only other characteristic where Xanthomonas spp. and the bentgrass bacterium differed to any appreciable extent. Flagellation was not observed in TEM negative stains and motility was not observed with light microscopy. Perhaps the bentgrass bacterium (1) requires stringent nutrition for flagella inducement, (2) no longer develops flagella as a result of mutation or some other genetical phenomenon, (3) has flagella for a short time in its life cycle, or (4) is inherently nonmotile. Serology is becoming important as an identification tool in bacteriology. Its value is of unquestionable importance in virology (Ball, 1974). In the present 97 study, antiserum was produced to the bentgrass bacterium and used in gel double diffusion tests for determining possible relatedness to other bacteria. No precipitation occurred with any of the known common plant pathogenic bacteria whereas strong precipitate reactions occurred with all isolates of the bentgrass bacterium. This strong serological reaction, even with bentgrass bacterium isolates from other states, suggests a uniform sero—group of the bentgrass pathogen. Perhaps isolates of Xanthomonas or other plant pathogenic bacteria other than the ones used in study may react with the bentgrass bacterium antiserum. Elrod and Braun (1947) noted several immunological groups in the genus Xanthomonas, suggesting not only serological differences but also varying levels of specificity. Schaad (1978) found that all isolates of _X. campestris and X. vesicatoria reacted to a X. campestris antiserum. Fixed smears of the bentgrass bacterium sent to Dr. M. J. Davis, University of Florida, did not react to RLB antiserum by an indirect immunofluorescent technique (Schaad, 1978; Davis et al., 1981b). This, as well as the non-fastidious nature of the bentgrass bacterium, suggests that it is not related to the RLB. Most known RLB are serologically related (Davis et al., 1981b).. Besides serology, guanine to cytosine (GC) ratios and DNA hydridization are also becoming quite routine in bacterial systematics (DeLey, 1968). Although not performed in the present research, determination of GC ratio and DNA hydridization are of considerable importance in determining the relation of the bentgrass bacterium to Xanthomonas spp. and other bacteria. Analysis of the yellow pigment of the bentgrass bacterium is another important test in taxonomy and nomenclature. Xanthomonas spp. produce a class of pigments called xanthomonadin (Andrews et al., 1973) which is far different from the carotenoid-type pigments produced by other bacteria. The taxonomic significance of xanthomonadin as reviewed by Starr et a1. (1977) 98 is considerably important since only bacteria that have been classified as Xanthomonas are capable of producing the pigment (Starr and Stephens, 1964). The problem of bacterial nomenclature and taxonomy is well recognized (Dye, 1962; Dye, 1974; Stolp et al., 1965). With further deveIOpment and refinement of new techniques, taxonomy and nomenclature will continue to evolve. The position of the bentgrass bacterium in relation to other bacteria cannot be ascertained at this time. The present research may conclude that the bentgrass bacterium is a xanthomonad-er organism but differs in several criteria from the latter. Extensive research, requiring great effort and time, is necessary before the bentgrass bacterium can be named and placed in its prOper taxonomic position. Like serology and physiological tests, antibiotic inhibition was also an important assay for determining characteristics of the bentgrass bacterium and the C-15 problem. Prior to this study, broad spectrum antibiotics have not been used for control of turfgrass diseases. The suppression of disease symptoms of the C-15 problem was easily attained with oxytetracycline. Streptomycin sulfate, an equally broad spectrum bacteriostat (Nester et al., 1973) which has been effective against several bacterial diseases (Goodman, 1954; Miller, 1959), was not effective in suppressing the symptoms of bacterial wilt of Toronto creeping bentgrass. The inefficacy of streptomycin sulfate could possibly be attributed to: (1) differential uptake by the plant, (2) rapid decomposition of streptomycin in the field, or (3) inherent sensitivity difference of the bentgrass bacterium. Both oxytetracycline and streptomycin have been shown to be systemic in other plants (Dekker, 1963; Frederick et al., 1971; Sinha and Peterson, 1972) but differential uptake may be possible in Toronto creeping bentgrass. Likewise, differences in residual prOperties were not determined in the present study. In a laboratory bioassay, however, differences in sensitivity 99 of the bentgrass bacterium were demonstrated. The bentgrass bacterium was approximately 100 times more sensitive to oxytetracycline than to streptomycin sulfate. Because of the expense of the antibiotic, oxytetracycline may not be a practical method of control of bacterial wilt of Toronto creeping bentgrass. Perhaps slightly lower rates may provide good management of the disease while making application techniques more easily adaptable to the 'Toronto' grower. Impact on environment should be seriously considered before widespread application since tetracyclines are frequently administered in treating human diseases (Nester et al., 1973). Electron microscopy was an extremely important tool in this research. Without the original discovery of the bentgrass bacterium association with the C-15 problem, the disease may have been unresolved for yet another decade. Electron microscopy techniques also aided in ascertaining the widespread distribution of the bacterium in five states, providing convincing evidence that the association of a bacterium with the C-15 problem was a viable solution. TEM, while tedious and time consuming, did provide valuable information concerning bacterial and plant ultrastructure. TEM is used primarily for fine structural detail (Meek, 1976; Joklik and Willet, 1976) but may become an important problem-solving technique for such poorly understood diseases as the C-15 problem. TEM negative stain was a very rapid technique for determining the association of large numbers of bacteria with diseased plants. Though not as useful as standard TEM and SEM for observing location and infection of bacteria in the plant, TEM negative stain was of benefit in providing morphological characteristics (convoluted cell walls and non-flagellation) of the bentgrass bacterium. TEM negative stain is used commonly in virology (Ball, 1974) and has been employed for observing the topographical nature of bacteria (see Joklik and 100 Willet, 1976). SEM was a beneficial technique because of its rapid preparation and the valuable information obtained in diagnosis and distribution. Though not as rapid as TEM negative stain, SEM allowed easy access to the xylem vessels which harbored the bacteria. Electron microscopy was also very valuable for comparisons of the isolated bacterium with the bacterium observed in diseased plants. Equally important was electron microscopy's contribution to confirming infection by the isolated bacterium in artificially-inoculated plants. TEM and SEM showed that the isolated bentgrass bacterium was observed in high populations in artificially-inoculated plants but were confined to the xylem vessels, in a similar manner to naturally-infected plants from the field. Lastly, SEM supported the suspicions that oxytetracycline prohibited reproduction of the bentgrass bacterium in the plant, whereas bacterial multiplications continued unchecked in streptomycin-treated plants. The C-15 problem of Toronto creeping bentgrass has now been resolved in the present research. The disease remained a devastating problem for at least a decade as a result of poor understanding of etiology, symptomatology and control. Incorporation of new ideas and techniques has proven that the C-15 problem is of bacterial etiology. Bacterial wilt, the new descriptive name for the C—15 problem, is the first disease of bacterial origin to be reported on turfgrass. This study provides innovative and valuable information regarding diagnosis, etiology, bacterial systematics and disease management for other bacterial diseases which undoubtedly exist. LIST OF REFERENCES Agrios, G. N. 1978. Plant Pathology. 2nd ed. Academic Press, New York. 703 PP- Andrews, A. G., S. Hertzberg, S. Liaaen-Jensen and M. P. Starr. 1973. The Xanthomonas "carotenoids"-non-carotenoid, brominated, aryl-polyene esters. Acta Chem. Scand. 27:2383-2395. Ark, P. A. and S. M. Alcorn. 1956. Antibiotics as bactericides and fungicides against diseases of plants. Plant Disease Reporter 40:85-92. Ball, E. M. 1974. Serological tests for the identification of plant viruses. The American PhytOpathological Society, Inc., St. Paul, MN. 32 pp. Beard, J. B. 1973. Turfgrass: Science and Culture. Prentice-Hall, Inc., Englewood Cliffs, NJ. 658 pp. Buchanan, R. E. and N. E. Gibbons (Eds.). 1974. "Bergey's Manual of Determinative Bacteriology," 8th edition. Williams and Wilkins Co., Baltimore. 1268 p. Buddenhagen, I. and A. Kelman. 1964. Biological and physiological aspects of bacterial wilt caused by Pseudomonas solanacerum. Annual Review of Phytopathology 2:203-230. Couch, H. B. 1973. Diseases of Turfgrasses. Robert E. Krieger Publishing Co., Huntington, NY. 348 pp. Davis, M. J., W. J. French and N. W. Schaad. 1981a. Axenic Culture of the Bacteria Associated with Phony Disease of Peach and Plum Leaf Scald. Current Microbiology 6:309-314. Davis, M. J., A. H. Purcell and S. V. Thomson. 1978. Pierce's disease of grapevines: Isolation of the causal bacterium. Science 199:75-77. Davis, M. J., R. F. Whitcomb and A. G. Gillaspie, Jr. 1981b. Fastidious bacteria of plant vascular tissue and invertebrates (Including so-called rickettsia- like bacteria) In The Prokaryotes: A Handbook on Habitats, Isolation and Identification 61' Bacteria. M. P. Starr, H. Stolp, H. G. Truper, A. Balows and H. G. Schlegel (eds.). Springer-Verlag, Berlin. p. 2164-2188. Dekker, J. 1963. Antibiotics in the control of plant diseases. Annual Review Microbiology 17:243-262. De Ley, J. 1968. DNA base composition and hydridization in the taxonomy of phytOpathogenic bacteria. Annual Review Phytopathology 6:63-90. 101 102 Dimond, A. E. 1970. Biophysics and biochemistry of the vascular wilt syndrome. Annual Review Phytopathology 8:301-322. Drechsler, C. 1935. A leaf spot of bentgrasses caused by Helminthosporium erythrospilum n. sp. Phytopathology 25:344-361. Dye, D. W. 1962. The inadequacy of the usual determinative tests for the identification of Xanthomonas spp. New Zealand Journal of Science 5:393- 416. Dye, D. W. 1974. The problem of nomenclature of the plant pathogenic pseudomonads. Review of Plant Pathology 53:953-962. Elrod, R. P. and A. C. Braun. 1947. Serological studies of the genus Xanthomonas: l. Cross-agglutination relationships. Journal of Bacteriology 53:509-518. Frederick, R. J., M. Klein and K. Maramorosch. 1971. Acquisition and retention of tetracycline hydrochloride by plants. Plant Disease Reporter 55:223- 226. Gibbs, A. and B. Harrison. 1976. Plant Virology, The Principles. Edward Arnold Ltd., London. 292 pp. Goheen, A. C., G. Nyland and S. K. Lowe. 1973. Association of a rickettsia-like organism with Pierce's disease of grapevines and alfalfa dwarf and heat therapy of the disease in grapevines. PhytOpathology 63: 341-345. Goodman, R. N. 1954. Fireblight control with sprays of Agri-mycin, a streptomycin-Terramycin combination. Plant Disease Reporter 38:874— 878. Hanson, A. A. and F. V. Juska. 1969. Turfgrass Science. American Society of Agronomy, Inc., Madison, WI. 715 p. Hess, W. H. and G. A. Strobel. 1969. Ultrastructural investigations of tomato stems treated with the toxin glchpeptide of Corynebacterium sepedonicum. Phytopathology 59:12. H00per, G. R., K. K. Baker and S. L. Flegler. 1979. Exercises in Electron Microscopy: A Laboratory Manual for Biological and Medical Sciences. Michigan State University, E. Lansing, MI. 144 pp. Hopkins, D. L. 1977. Diseases caused by leafhopper—borne rickettsia-like bacteria. Annual Review of Phytopathology 17:272. HOpkins, D. L., H. H. Mollenhauer and W. J. French. 1973a. Occurrence of a rickettsia-like bacterium in the xylem of peach trees with phony disease. PhytOpathology 63:1422—1423. Hopkins, D. L. and J. A. Mortensen. 1971. Suppression of Pierce's disease symptoms by tetracycline antibiotics. Plant Disease Reporter 55:610-612. 103 Hopkins, D. L., J. A. Mortensen and W. C. Adlerz. 1973b. Protection of grapevines from Pierce's disease with tetracycline antibiotics. (Abstr.) Phytopathology 63:443. Hugh, R. and E. Leifson. 1953. The taxonomic significance of fermentative versus oxidative metabolism of carbohydrates by various Gram-bacteria. Journal of Bacteriology 66:24-26. Husain, A. and A. Kelman. 1958. Relation of slime production to mechanism of wilting and pathogenicity of Pseudomonas solanacearum. Phytopathology 48:155-165. Jenkins, S. F., Jr., D. J. Morton and P. D. Dukes. 1966. Distinguishing Pseudomonas solanacearum infections from other peanut wilt diseases by the use of serological techniques. Plant Disease Reporter 50:836-838. Joklik, W. K. and H. P. Willett. 1976. Zinsser Microbiology, 16th ed. Appleton- Century-Crofts, NY. 1223 pp. Kitajima, E. W., M. Bakarcic and M. V. Fernandez-Valiela. 1975. Association of rickettsia-like bacteria with plum leaf scald disease. PhytOpathology 65:476-479. Larsen, P. O. 1979. Leaf blight and crown rot of 'Toronto' creeping bentgrass. Greens Sect. Rec. 17:5-7. Larsen, P. O., A. K. Hagan, B. G. Joyner and D. A. Spilker. 1981. Leaf blight and crown rot on creeping bentgrass, a new disease caused by Drechslera catenaria. Plant Disease Reporter 65:79-81. Maramorosch, K. 1976. Plant mycoplasma diseases. Pages 150-171 in: Encyclopedia of Plant Physiology. Vol. 4. Physiological Plant Pathology. R. Heitefuss and P. H. Williams, eds. Springer-Verlag, New York. 890 pp. Meek, G. A. 1976. Practical Electron Microscopy for Biologists. 2nd ed. John Wiley and Sons, New York. 528 pp. Meyer, W. A. and A. J. Turgeon. 1975. Control of red leaf spot on 'Toronto' creeping bentgrass. Plant Disease Reporter 59:642-645. Miller, P. W. 1959. A preliminary report on the comparative efficacy of copper- lime and Agri-mycin dust mixtures for control of walnut blight in Oregon. Plant Disease Reporter 43:401-402. Nelson, P. E. and R. S. Dickey. 1966. Pseudomonas caLyophylli in carnation. II. Histological studies of infected plants. Phytopathology 56:153-154. Nelson, P. E. and R. S. Dickey. 1970. Histopathology of plants infected with vascular bacterial pathogens. Annual Review of PhytOpathology 8:259-280. Nester, E. W., C. E. Roberts, B. J. McCarthy and N. N. Pearsall. 1973. Microbiology. Holt, Rinehart and Winston, Inc., Chicago. 719 pp. 104 Nienhaus, F. and R. A. Sikora. 1979. Mycoplasmas, spiroplasmas and rickettsia- like organisms as plant pathogens. Annual Review of Phytopathology 17:37-58. Reynolds, E. S. 1963. The use of lead citrate at high pH as an electron opaque stain in electron microscopy. Journal of Cell Biology 17:208. Schaad, N. W. 1978. Use of direct and indirect immunofluorescence for identification of Xanthomonas campestris. PhytOpathology 68:249-252. Schaad, N. W. (ed.). 1980. Laboratory Guide for Identification of Plant Pathogenic Bacteria. The American Phytopathological Society, St. Paul. 72 p. Sequiera, L. and A. Kelman. 1962. The accumulation of growth substances in plants infected by Pseudomonas solanacearum. Phytopathology 52:439-448. Sinha, R. C. and E. A. Peterson. 1972. Uptake and persistence of oxytetracycline in aster plants and vector leafhoppers in relation to inhibition of clover phyllody agent. Phytopathology 62:377-383. Smith, N. R., R. E. Gordon and F. E. Clark. 1952. Aerobic spore-forming bacteria. U. S. Department of Agriculture Monograph 16, 148 pp. Starr, M. P., C. L. Jenkins, L. B. Bussey and A. G. Andrews. 1977. Chemotaxonomic significance of the xanthomonadins, novel brominated aryl-polyene pigments produced by bacteria of the genus Xanthomonas. Archives of Microbiology 113:1-9. Starr, M. P. and W. L. Stephens. 1964. Pigmentation and taxonomy of the genus Xanthomonas. Journal of Bacteriology 87:293-302. Stolp, H., M. P. Starr and N. L. Baigent. 1965. Problems in speciation of phytopathogenic pseudomonads and xanthomonads. Annual Review of PhytOpathology 3:231-264. Vargas, J. M., Jr. 1981. Management of Turfgrass Diseases. Burgess Publishing Co., Minneapolis. 204 pp. Weaver, D. J., B. C. Raju, J. M. Wells and S. K. Lowe. 1980. Occurrence in johnsongrass of rickettsia-like bacteria related to the phony peach disease organisms. Plant Disease 64:485-487. Wells, J. M., B. C. Raju, G. Nyland and S. K. Lowe. 1981. Medium for isolation and growth of bacteria associated with plum leaf scald and phony peach diseases. Applied and Environmental Microbiology 42:357-363. Young, J. M., D. W. Dye, J. F. Bradbury, C. G. Panagopoulos and C. F. Robbs. 1978. A proposed nomenclature and classification for plant pathogenic bacteria. New Zealand Journal of Agricultural Research 21:153-177. "‘iiiiiiiiiiiiii“