TEMPORAL AND SPATIAL COMPONENTS OF THE YEAST FLORA OF THE RED IMPORTED FIRE ANT (Hymenoptera: Formi ci dae) by AMADOU SIDIKI BA, B.S., M.Ed.

A DISSERTATION IN AGRONOMY Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY

Approved

_' 'CK'ctt I I

;

lll~V'kir'- - . ...- R:H , > '

Accepted

I Deat of th~- Graduate-'§chool

• December, 1994 ' ACKNOWLECX^EMENTS

I am first and, foremost, grateful to Dr. Sherman A. Phillips, Jr., committee chairperson, for his support, assistance, guidance, and the preparation of this dissertation. I would like also to thank the other members of my committee. Dr. Caryl E. Heintz, Dr. Harlan G. Thorvilson, Dr. John C. Zak, and Dr. Richard E Zartman for comments and guidance during distinct phases of my research. Also many thanks go to Dr. David Nes and his research team for their help in the sterols extraction and identification. I am grateful to the Department of Plant and Soil Science for this opportunity. I would like to express my thanks to Camille E. Landry and Gene White for their help in collecting colonies and the diversions they provided through the graduate student life. Special thanks to Mary Peek and Mary Benton. I would like to thank Usha S. Kallemuchikkal for her help. I express my sincere appreciation to Bob, Rene, and Seth McCleary for their friendship and for the many holidays we spent together. This dissertation is dedicated to my parents, "yerohammu" Sidiki C. Ba, and Marlam D. Bal for their unconditional love and support In my endeavors. I would like to express my love and admiration to my eighty-five-year-old grandmother Dede.

11

^ TABLE OF CONTENTS

ACKNOWLEDGEMENTS i I ABSTRACT v LISTOFTABLES vii LIST OF FIGURES i x CHAPTER I. LITERATURE REVIEW 1 References cited 2 0 II. IDENTIFICATION AND ABUNDANCE OF YEASTS ISOLATED FROM RED IMPORTED FIRE ANT HAEMOLYMPH 31 Introduction 31 Materials and Methods 32 Results 44 Discussion 5 3 References Cited 5 9 III. NUTRIENT AND YEAST FLORAL CHANGES IN SOIL INDUCED BY THE RED IMPORTED FIRE ANT 9 6 Introduction 96 Material and Methods 99 Results 102 Discussion 107 References Cited 109

111

^ IV. STEROL PROFILES OF THE RED IMPORTED FIRE ANT AND ASSOCIATED YEASTS 11 8 Introduction 118 Material and Methods 126 Results 123 Discussion 125 References Cited 127

APPENDICES A. SIGNIFICANCE LEVELS FOR ANOVAS AND MANOVA EVALUATING VARIOUS EFFECTS OF YEASTS ON THE RED IMPORTED FIRE ANT (WEIGHT, SITES, SEASONS, AND MOUND SOIL AND CONTROLS) 137 B. MASS SPECTRUM OF MAJOR STEROLS ISOLATED FROM THE RED IMPORTED FIRE ANT AND ITS ASSOCIATED YEASTS 143

IV ABSTRACT

Few investigations have examined the relationship between Solenopsis complex and its internal microflora. Most investigations on the red imported fire ant are either control oriented or emphasize the species' negative impact on ecosystems. Therefore, the objective of this research were (1) to isolate and identify yeasts from the red imported fire ant (Solenopsis invictaV (2) to determine the impact of associated yeasts on red imported fire ant colony vigor, and (3) to clarify the significance of the association by obtaining and comparing sterol profiles of the yeast flora and the red imported fire ant. Colonies were collected from five locations in Texas: Beaumont, Huntsvllle, Waco, Stephenville, and Abilene. These sites represent a transect along a decreasing moisture gradient from the southeastern to the northwestern region of the state. Colonies were randomly collected during the winter of 1992, and the spring, summer, and fall months of 1993. Physiological and morphological tests showed that the 327 yeast isolates from Solenopsis invicta haemolymph represent five yeast species: Candida parapsilosis (168 Isolates), Candida lipolytica (123 isolates), Candida guillermondii (18 isolates), Candida rugosa (9 Isolates), and Debaryomyces var. hansenii (9).

Using GLM statistical procedures, the biomass of red imported fire ant colonies with yeasts was significantly greater (P. < 0.005) than ant colonies without yeasts during winter and spring months. A significantly greater (E < 0.005) colony biomass was observed for fire ant colonies collected only from the Abilene site. This response was site specific. Moreover, 80.3% of the fourth instar larvae collected from Beaumont, Huntsville, Waco, and Abilene collected during the fall months, harbored yeasts. These data indicate that the presence of yeast does not have a negative impact on colony vigor. A high prevalence of yeasts in larvae suggests a nutritional role. The fact that the fungal sterol, ergosterol, was detected in worker larvae, adult workers, and eggs suggests that these yeast symbionts may contribute to the sterol pool of the red imported fire ant. These data suggest that this association is not random, and a highly selective process is taking place between the yeasts and the fourth instar larva. Consequently, this association may yield novel approaches in red imported fire ant control if behavioral parameters of the ant are considered in the development and application of microbial control agents. In addition, means of yeast colony forming units per gram of soil between mound soil and soil adjacent to mounds were significantly different (P. < 0.001). Also, nutrient concentration such as phosphorus in soil from brood chamber was significantly greater than control samples (P < 0.05).

VI LIST OF TABLES

2.1 Colony of red imported fire ant and yeast species associated with the red Imported fire ant obtained from sites located along a decreasing moisture gradient in Texas 64

2.2 Morphological and biochemical characteristics of the yeast flora of the red imported fire ants 76

2.3 Abundance of yeast species associated with the red imported fire ant adult haemolymph (from Texas) 8 0

2.4 Yeast frequency of occurrence In red imported fire ant workers, colonies, and fourth instar larvae obtained from colonies in Texas 80

2.5 Yeast species frequencies of occurrence in fourth instar larvae of the red imported fire ant from colonies in Texas 81

2.6 Mean of yeast colony forming units per red imported fire ant larva 81

3.1 Fatty acid profiles of yeasts isolated from mound soil and non-mound soil 111

3.2 Relative frequencies of yeast species occurrence in red imported fire ant brood chambers and soil adjacent to mounds collected from five sites along a moisture gradient in Texas 112

3.3 Descriptive statistics (mean, standard error) for number of yeast colony forming units per gram of soil among seasons 113

3.4 Descriptive statistics (mean, standard error) for number of yeast colony forming units per gram of soil among sites 11 0

Vll 4.1 Sterol composition of various developmental stages and life forms of the red imported fire ant 130

4.2 Chromatographic and spectral properties of sterols Isolated from the red imported fire ant 132

4.3 Sterol profile of the dominant yeast, Candida parapsilosis. associated with the red imported fire ant 134

4.4 Sterol profile of the second most prevalent yeast, Candida lipolytica. associated with the red imported fire ant 135

A. 1. ANOVA (GLM) evaluating the effects of yeasts on colony weight 1 38

A. 2. ANOVA (GLM) evaluating yeast effects on weights of colonies collected during winter 1992 138

A. 3. ANOVA (GLM) evaluating yeast effects on weights of colonies collected during spring 1993 138

A. 4. ANOVA (GLM) evaluating yeast effects on weights of colonies collected during summer 1993 139

A. 5. ANOVA (GLM) evaluating yeast effects on weights of colonies collected during fall 1993 139

A. 6. ANOVA (GLM) evaluating yeast effects on weights of colonies collected from Abilene 139

A. 7. ANOVA (GLM) evaluating yeast effects on weights of colonies collected from Stephenville 140

A. 8. ANOVA (GLM) evaluating yeast the effects of yeasts on colony weightsin the Waco site 140

Vlll A. 9. ANOVA (GLM) evaluating the effects of yeasts on colony weight in the Huntsville site 140

A. 10. ANOVA (GLM) evaluating the effects ofyeasts on colony weights in the Beaumont site 141

A. 11. ANOVA showing effects of dates, sites, and locations (mound soil adjacent to mound) on yeast colony forming units 141

A. 12. Two-way ANOVA evaluating the effects of dates and sites, dates and locations, and sites and locations (mound soil soil or soil adjacent to mound) on yeast colony forming units 141

A. 13. Three-way ANOVA evaluating the effects of dates, sites, and locations (mound soil or soil adjacent to mound) on yeast colony forming units 142

IX LIST OF FIGURES

2.1 Collection sites along a decreasing moisture gradient from Beaumont to Abilene, Texas 82

2.2 Comparison of mean weights of red imported fire ant colonies among locations in Texas 83

2.3 Comparison of mean weights of all red imported fire ant colonies among locations in Texas (n=200) with and without yeasts 84

2.4 Comparison of mean weights of red imported fire ant colonies with and without yeasts among locations in Texas 85

2.5 Comparison of mean weights of red imported fire ant colonies with and without yeasts collected from sites in Texas during 1992 and 1993 86

2.6 Number of yeast isolates from red imported fire ant workers obtained from sites in Texas 87

2.7 Number of yeast Isolates from red imported fire ant workers obtained during 1992 and 1993 from sites in Texas 88

2.8 Number of yeast Isolates from fourth Instar red imported fire ant larvae obtained from sites along a moisture gradient 89

2.9 Dominant yeasts growing on Dalmau plate (corn meal agar) associated with the red imported fire ant. A: C. parapsilosis: B: C. lipolytica 90 2.10 Sagital section of fourth instar larva. A: meconia; B-D: yeast cells lodged near muscle fibers, and attached to membrane-bound structures (fat bodies) 92

2.11 Gut of fourth instar larva exhibiting yeast cells. A-B: yeasts in alimentary tract showing cell division 94

3.1 Comparisons of mean nutrient concentrations (nitrogen, potassium, phosphorus, and sodium) between mound soil and soil adjacent to mounds (paired t-test) 114

3.2 Comparison of mean nutrient concentrations (magnesium, salinity, and calcium) between mound soil and soil adjacent to mounds (paired t-test) 11 5

3.3 Comparison of mean organic matter concentrations between mound soil and soil adjacent to mounds (paired t-test) 116

3.4 Comparison of mean pH values between mound soil and soil adjacent to mounds (paired t-test) 117

4.1 Structures (side chain and nucleus) of sterols isolated from the red imported fire ant 136

B. 1. Mass spectrum of the fragmentation pattern of sitosterol (M+ and other sterol diagnostic) 144

B. 2. Mass spectrum of the fragmentation pattern of campesterol (M+ and other sterol diagnostic) 145

B. 3. Mass spectrum of the fragmentation pattern of stigmasterol (M+ and other sterol diagnostic) 146

B. 4. Mass spectrum of the fragmentation pattern of cholesterol (M+ and other sterol diagnostic) 147.

XI B. 5. Mass spectrum of the fragmentation pattern of ergosterol (M+ and other sterol diagnostic) 148

B. 6. Mass spectrum of the fragmentation pattern of methylenecholesterol (M+ and other sterol diagnostic)... 149

Xll CHAPTER I LITERATURE REVIEW

All levels of associations or relationships, ranging from mutualistic through antagonistic, exist between microorganisms and ants (Steinhaus and Marsh 1967; Buchner 1965). However, the association of the red imported fire ant, Solenopsis invicta Buren, and naturally occurring microorganisms is not well understood. The red imported fire ant was introduced into the United States between 1930 and 1940 through the port of Mobile, Alabama (Green 1967). This species is a medical and agricultural pest that infests Alabama, Florida, Louisiana, and parts of North Carolina, Georgia, Mississippi, Texas, Arkansas, Oklahoma, and Tennessee, and its range is increasing. In addition, the polygyne form is becoming more prevalent and more difficult to control with chemicals (Glancey et al. 1987). This invasive pest is native to Brazil, Uruguay, and Argentina where it is beset by pathogens, parasites, and predators (Jouvenaz 1983). However, in the United States few symbiotic organisms were found in association with the red imported fire ant (Jouvenaz 1991). Approximately 10% of known insect species contain non­ parasitic microorganisms that may be intracellular or extracellular, usually in the lumen of the digestive tract (Douglas 1989). The simplest associations involve microorganisms found in the gut lumen (Koch 1960). Simple mechanical associations between insects and yeasts were described (Phaff and Starmer 1987), as

1 2 well as beneficial associations, whereby insects derive important nutritional compounds such as sterols, amino acids, carbohydrates, vitamins, and essential fatty acids (Koch 1960; Phaff and Starmer 1987). Numerous yeast species are indeed known to play important roles in the life cycles of various insects. These yeasts are generally located in the gut or in specialized cells as intracellular symbionts (Buchner 1965). Most yeasts located in alimentary tracts of insects have been identified to species level. For example, literature on the association of yeasts with various Drosophila species is abundant (Phaff and Starmer 1987). Yeast prevalence in Drosophila diets has stimulated studies regarding the role of yeasts in most aspects of Drosophila biology (Starmer and Aberdeen 1990; Starmer and Fogleman 1986; Starmer et al. 1989). Results of these studies suggest that the two organisms are mutually dependent for survival. Drosophila spp., for example, rely on the yeasts for nutrition during larval development and egg maturation whereas the yeast relies on Drosophila for dispersal. In addition, the contribution of the yeast symbionts to the Insect host appears, in general, to be more than simply nutritional (Dowd 1989). Investigations on the taxonomy (Lachance et al. 1988), ecology (Starmer and Fogleman 1986; Starmer et al. 1976), and blogeography (Starmer et al. 1989) of yeasts and Drosophila spp. have shown mutualistic interactions between cactus yeasts and cactus- inhabiting drosophilids. Besides direct nutritional benefits of yeasts to Drosophila. these investigations revealed that some cactus-living yeasts detoxify chemical compounds present in the larval environment, thus facilitating the normal development of Drosophila mojavensis (Fogleman and Kircher 1986). A common yeast in this environment, Candida ingens. is capable of breaking down toxic compounds to their ester forms (Starmer 1982). Other yeasts, Candida sonorensis and Cryptococcus cercanus. participate in the detoxification process by metabolizing alcohol compounds, thus eliminating their toxic effects (Starmer et al. 1986). Larvae of cactus-breeding Drosophila selectively feed on yeasts found in their habitat (Fogleman and Kircher 1986). Vacek et al. (1984) studied the feeding preferences of mature, mated female Drosophila buzzatii and found that fly discrimination between yeasts developed quickly. After three days, a significant correlation was found to exist between the preferred yeasts for feeding and for oviposition, with Pichia cactophilia being the most preferred yeast species. Fogleman et al. (1981) also reported that larvae of Drosophila mojavensis preferred the yeast Pichia cactophilia compared to Candida sonorensis. In addition, laboratory tests confirmed that larvae selected patches of preferred yeasts and spent more time feeding on them. Volatiles produced by yeasts influence larval preferences. The common cactus-specific yeasts, such as Pichia cactophilia. are capable of producing relatively high levels of volatile cues (Fogleman 1982). Investigations of Drosophila buzzattii also show that yeasts are transferred vertically through the pupal stage and horizontally during courtship and mating (Starmer et al. 1988). In addition. 4 previous yeast diets can significantly affect the choice of mates. Thus, yeasts sen/e an important role in the biology and behavior of Drosophila species. Beetle associations with yeast species figures prominently in the literature of yeast and insect interactions (Phaff and Starmer 1987). For example, bark beetle associations with yeasts were reported by several investigators (Shifrine and Phaff 1956; Leufen and Nehls 1986; Whitney 1982; Phaff and Starmer 1987). Bark beetles are responsible for killing coniferous trees in North America, and various species of Dendroctonus. Ips. Scolytus. and Dryocoetes are responsible for most of the damage (Furniss and Carolin 1977; Baker 1972). Bark beetles belong to the family Scolytidae (Coleoptera), and their common characteristic of attacking living coniferous hosts probably has special significance In mutualistic associations (Whitney 1982). Microorganisms associated with bark beetles may aid the beetles in overcoming tree resistance, enhancing beetle nutrition, and producing or converting chemical substances that modify beetle behavior (Whitney 1982). The dominant yeast flora of Ips typographus Includes Hansenula holstii and Candida diddensii (Leufren and Nehls 1986). In addition, Hansenula capsulata. Pichia pinus. Candida nitratophila. and two Cryptococcus-type yeasts were also recovered in smaller numbers (Leufren and Nehls 1986). An internal survey of yeasts associated with the guts of the pine bark beetles Dentroctonus and Ixis conducted by Shifrine and Phaff (1956) found four predominant species, Pichia pinus (32 strains), Hansenula capsulata (36 strains) 5 Candida silvicola (70 strains) and Candida curvata (15 strains) In addition, lesser amounts of Candida parapsilosis (2 strains) and Candida rugosa (1 strain) were also Isolated (Shifrine and Phaff 1956). From these associations, the assumption was made that yeasts play at least a supplementary role In the nutrition of larvae In the phloem galleries (Phaff and Starmer 1987). Whitney (1982) suggests that nutritional benefits for bark beetles is more important either during the larval stage or during maturation than during reproduction of the adult beetle. In addition, yeasts were not only part of the Internal flora of every species of bark beetle, they are also present throughout all stages of beetle metamorphosis from the first-instar larva to the adult stage (Phaff and Starmer 1987). Numerous yeast species (Phaff and Starmer 1987; van der Walt et al. 1972) are also associated with ambrosia beetles (Coleoptera: Scolytidae). In a survey, van der Walt and colleagues Isolated 21 genera and 64 species of yeasts associated with this group of beetles. Also, six new species of Candida were recovered from material of various bostrichid, buprestid, cerambycid, platypid and scolytid Insect pests Infesting Indigenous trees In the Transvaal region of South Africa (van der Walt et al. 1972). However, these reports did not explain the significance of this association but simply presented a species list of yeasts. An investigation of internal yeast flora of the adult green June beetle (VIshnlac and Johnson 1990), Cotinis nitlda (L.), (Coleoptera: Scarabaeidae) from gut and feces yielded: Aureobasidium pullulans. Candida guillermondii. Candida krusei. Candida sake. Candida tropicalis. Cryptococcus albidus. var. albidus. Debaryomyces hansenii. Hanseniaspora uvarum. Hansenula californica. Rhodoturula glutinis. Rhodoturula rubra, and Trichosporon cutaneum. The major Isolate, Trichosporon cutaneum. Is likely acquired from soil by the green June beetle as it emerges (Vishlnlac and Johnson 1990). Later, this yeast becomes dominant in the gut of C.- nitlda and may play a role in green June beetle feeding and aggregation behavior (Vishlnlac and Johnson 1990). Domek and Johnson (1988) showed that aggregation behavior was triggered by C.. nitida feeding and suggested that the green June beetle was attracted to semiochemlcals released during that process. In addition, aggregation behavior was elicited in both sexes of the green June beetle when they fed on peach slices inoculated with Trichosporon cutaneum (Johnson and VIshnlac 1991). The role of microorganisms in aggregation behavior of C- nitida was investigated by Domek and Johnson (1990) using antibiotic treatment of food substrates. Their results suggested that yeasts contained in diets, and possibly in beetle digestive tracts, significantly contributed to the production of volatiles, which induced beetle aggregation. The presence of sugar In floral nectar and honey spurred Investigations regarding the Internal yeast flora of various species of bees (Phaff and Stamner 1987). Yeast species may be transferred to flower nectar by pollinating Insects (Sandhue and Waraich 1985). Batra et al. (1973) published an extensive study of the mycoflora of domesticated and wild bees (Hymenoptera: Apidae). In their study, they found 17 yeast from the midgut of the honey bee. Apis 7 mellifera. and from the midgut of the wild bees, Nomia melanderi. Megachile roundata. Hallctus rubicundus. and Anthophora occldentalis. These yeasts were also present in the floral nectar and were members of the following genera: Candida. Endomycopsls. Oidium. Hansenula. Pichia. Rhodoturula. Saccharomyces. Schizosacchasromyces. and Zygosaccharomyces. Bee and yeast associations appeared to be harmless and possibly even beneficial for bee nutrition by providing vitamins and other essential nutrients (Batra et al. 1973). However, some strains of Saccharomyces cerevisae caused bloating and premature death of Nomia larvae (Batra et al. 1973). Intestinal yeast flora of 388 adult worker honey bees was investigated by Gilliam et al. (1974). The dominant yeast species of adult worker honey bees were Torulopsis magnoliae. Candida parapsilosis. and Torulopsis glabrata. However, the significance of yeast associations with bees Is not well understood. Several authors (Khoch 1960; Brooks 1963; Buchner 1965) argued that these yeasts were ingested as a food source, and subsequently colonized the lumen of digestive tracts, and were incorporated into cells of the gut epithelium. Therefore, another aspect of microorganism association with insects is the presence of intracellular symbionts. A vast number of Insects from many different orders have symbionts which are found in cells embedded in fat bodies or in other cells of the haemocoel (Hinde 1971). However, the difficulty of culturing these intracellular microorganisms has hindered their identification. The yeast symbionts of anobiid beetles (Coleoptera: Anobiidae) were cultured 8 on standard mycological media (Pant and Frankael 1954). However, Brooks (1963) pointed out that obligate intracellular organisms would not grow on the relatively simple media used by many researchers. The difficulty of obtaining symblont-bearing tissue from insects aseptically without killing the symbionts in the process has been recognized by Brooks and Richards (1956) in their challenge to Pierre's claim that he cultured the symbionts of the cockroach Leucophae maderae (F.) (Pierre 1962). The first symbiont identified was that of the beetle Ernobius mollis L. The yeasts are globose, occurring In clusters, and the strain is weakly fermentative (Muller 1934). The yeast was named Torulopsis ernobii Lodder et Kreger-van Rij (1952). However, it was renamed Candida ernobii (Lodder et Kreger-van Rij) Yarrow and Meyer (1978). In addition, subsequent efforts to culture symbionts of anobiid beetles failed (Muller 1934). Seven strains of yeast-like symbionts, subsequently isolated from the anobiid beetle Sitodrepa panlcea. were characterized morphologically and physiologically (Bismanis 1976). The yeast was placed in the genus Torulopsis and was named T. buchnerii Kuhlwein and Jurzita (1961). This yeast obtained nitrogenous compounds and carbohydrates, such as proline and trehalose, from haemolymph of the host, and synthesized and made available to the host all essential amino acids and vitamins except biotin (Bismanis 1976). According to the hypothesis of Buchner (1965), intracellular symbionts located within specialized cells (mycetocytes) provide nutrients which are lacking in the insect diet. Insects harboring 9 mycetocytes are able to tolerate low nitrogen diets. In addition, mycetocytes provide essential amino acids, B vitamins, and sterols (Cochran 1985; Mullins and Cochran 1975; Brooks and Richards 1955; 1956; Wren and Cochran 1987). Symbionts as a sterol source for insects has been the most widely reported aspect of yeast-insect associations. Insects have no ability to synthesize sterols de novo (Svoboda et al. 1975) and therefore, a supply of exogenous sterols is required (Clayton 1964) for membrane and hormone biosynthesis. The only exceptions are those Insects in which a sterol source may be provided by gut or intracellular symbionts (Clayton 1964). Aphids raised on chemically defined artificial diets free of sterols (Dadd and Mittler 1966; Dadd and Krieger 1968; Srivastava and Auclair 1971). The symbionts serve as a source of sterols for aphids (Dadd and Mittler 1966; Srivastava and Auclair 1971; Akey and Beck 1972). Ehrhardt (1968), and Houck et al. (1976) indicated that the symbionts of the pea aphid, Acyrthosiphon pisum were capable of lipid metabolism Furthermore, Griffiths and Beck (1977a, b) suggest cholesterol biosynthesis by pea aphid symbionts. However, these studies were disputed in an Investigation of sterol biosynthesis In the aphid Schizaphis graminum (Rondani) by Campbell and Nes (1982). Their results showed that S.. graminum and its symbionts do not have the complement of enzymes In the isopentenoid pathway that are required for cholesterol synthesis (Campbell and Nes 1982). Moreover, S.. graminum accumulates sterols from sorghum plants and has the necessary enzymes to dealkylate sitosterol to cholesterol 10 (Campbell and Nes 1982). Based upon the previous discussion the following observations are generally accepted (Kircher 1982) regarding the sterol physiology and requirements of insects: (1) they require an exogenous source of sterol. It is principally supplied by the diet, but may also come from gut of intestinal microorganisms; (2) cholesterol can be universally used; (3) plant- eating Insects can also use phytosterols such as sitosterol, stigmasterol, and campesterol, and ergosterol, a sterol specific to filamentous fungi and yeasts; and (4) the ability to use various sterols and the dietary concentrations required for optimal growth vary among species. Therefore, insects, unlike vertebrates, plants, and many other organisms, are unable to synthesize the steroid nucleus, because the Isopentenoid pathway Is blocked prior to squalene formation (Clayton 1964). In addition, cholesterol is often the main sterol in insect tissue. However, plants contain little or no cholesterol. Therefore, phytophagous insects must either transform phytosterols Into cholesterol or obtain sterols from their symbionts (Karlson and Hoffmeister 1963). Thus, plant sterols (sitosterol, stigmasterol, campesterol, etc.) and the fungal sterol, ergosterol, with C28 and C29 are converted to C27 (cholesterol) by a process called dealkylation (Svoboda et al. 1975). Ergosterol was the third most effective sterol after cholesterol and sitosterol for growth of insects (Clayton 1964). The role of these yeast symbionts appears to be more than just nutritional. The enlargement of mycetomes (membrane bound mycetocytes) In insecticide-resistant aphids has fueled speculation 11 that symbionts detoxify insecticides (Amiressami and Petzold 1976; Amiressami 1980). Furthermore, symbiotic microorganisms of phytophagous insects are involved in detoxifying plant allelochemlcals (Jones 1984). The cigarette beetle, Lasloderma serricorne (F.), feeds on tobacco, straw, seed, and pepper (Metcalf et al. 1962). Dowd (1989) found that symbiotic yeasts in the cigarette beetle detoxify xenobiotlcs and thus assist In host survival. Plant- derived compounds that can be detoxified by microorganisms Include over two dozen different alkaloids (Vining 1984). Most often, however, detoxification denotes metabolism or transformation of lipophilic compounds (Dowd 1991). In obligate symbiotic relationships, the host transfers the symbionts from one generation to the next (Buchner 1965). In situations where the symbionts are located In the host's gut, symbiont migration from the midgut to the hindgut results in their incorporation in fecal pellets (Koch 1967). Subsequently, larvae eat the feces, thus assuring the transfer of symbionts to the next generation. This uptake of symbionts has been observed by Wigglesworth In the South American family Triatomidae (Koch 1967). Natural transfer can also be accomplished by smearing eggs externally with symbionts during oviposition (Brooks 1963; Koch 1967). This mechanism of symbiont transmission Is reported in anobiid, cerambycid, curculionid, chrysomelid beetles, and lygaeid bugs (Buchner 1965; Dasch et al., 1984). On the other hand, temiltes acquire flagellated protozoa by feeding on the proctodeal secretions of older insects (Grasse 1952). Symbionts can also be transferred 12 from mycetocytes to ovaries for incorporation into the oocytes (Douglas 1989). This mode of transmission is called transovarial transmission and Is reported in the following families: Blattidae, Cimicidae, Lygaeidae, Aphldidae, Bostrichidae, Silvanldae, Curculionldae, and Formicidae (Douglas 1989). Despite a wealth of descriptive information on the transfer of symbionts to insect ovaries and incorporation of symbionts by oocytes, virtually nothing is known regarding the actual mechanisms that mediate microorganism movement and their consequent transfer to the appropriate insect cells (Douglas 1989). Taylor (1983) proposed that the evolution of endosymbiosis begins from a neutral or harmful association and proceeds through the following five stages: (1) consistent and extended contact; (2) avoidance of lethal harm during contact; 3) coadaptation leading to reduced virulence and Increased tolerance; (4) further coadaptation, leading to dependence and/or interdependence; and (5) still further coadaptation, leading to facilitative Interactions that foster stability and permanence in the association. One such association which has been well studied involves aphids (superfamily Aphidoidea) (Douglas 1989). Aphids are a monophyletic group originating in the Jurassic Period, and comprise approximately 4,000 extant species (Munson et al. 1991). A phylogenetic investigation of the endosymbionts from 11 species of aphids shows that the symbionts are related and constitute a distinct lineage within the gamma-3 subdivision of the class Proteobacteria (Munson et al. 1991). Their phylogeny suggests that a successful 13 endosymblotlc association occurred in a common ancestor of the 11 aphid species and that subsequent branching of aphids and bacteria produced the present species of aphids and their endosymbionts (Munson et al. 1991). The single infection theory suggests a minimum of 80 million years for the aphid-endosymbiont association, because fossils from Cretaceous amber show that the MIndaridae, Aphldidae, and Drepanosiphidae were distinct lineages at that time (Munson et al. 1991). The eubacterial endosymbionts of whiteflies (Homoptera: Aleyrodidae) are unrelated to the endosymbionts of aphids and mealybugs (Clark et al. 1992). The whitefly species Bemisia tabaci. Siphonimus phillyreae. and Trialeuroles vaporariorum contain endosymbionts which are closely related and constitute a distinct lineage within the gamma- subdivision of the Proteobacteria (Clark et al. 1992). Mutualism is a powerful factor of evolution and could facilitate the transmutation of species by enhancing population dynamics (Nardon and Grenler 1991). In the case of weevils, two sibling species, Stophilus oryzae and S.. zeamais. are distinguishable solely by the morphology of their symbionts, yet the hybrids between the two species are sterile (Nardon and Wicker 1981). In the case of certain species from the ant tribe Camponotoni, and in particular species of Formica from the tribe Formlcini, their endosymbionts may have a common ancestor because (1) the symbiont distribution is confined to phylogenetlcally related ants; (2) symbiont morphology is identical; (3) all symbionts are mechanically transferred the same way (transovarially); and (4) one 14 ant genus, Colopsis. has intermediate characteristics of the two genera for which the tribes are named (Dash et al. 1984). These ant genera, Formica and Camponotus. were recovered from Baltic amber dated to be 35-40 million years old. Apparently the Formicine endosymbioses, if derived from a common origin, predates that age (Dash et al. 1984). The association of the fungus-growing ants (subfamily Myrmicinae, tribe Attini) is a well-known phenomenon and figures prominently In the evolution theory of mutualistic associations (Weber 1972). The mutualism between the attine ants and the fungi is probably the most elaborate of any insect-fungus association (Martin 1987; Cherrett et al. 1989) and rivals human ingenuity. The fungal symbionts of the attine produce hyphae with swollen tips, which form clusters 1-2 mm in diameter called staphylae (Martin 1992). Staphylae liquid content Is taken into the infrabuccal chamber of the worker ant and "pumped" into the crop whereby it subsequently reaches the midgut (Martin 1992). This liquid contains different enzymes which play a critical role in the attine-fungus mutualistic relationship (Martin 1992). Ingested liquid extracted from larvae and workers contain proteases, amylases, pectlnases, hemicellulase, cellulases (Martin 1992), and other compounds such as polyphenol oxidase and tannase (Cherrett et al. 1989). Attempts to identify the fungal symbionts of the attine ants have been hindered by the absence of sporogenous cells (Martin 1992). Two fungal isolates were identified, one species of Lepiota that produced fruiting bodies In the laboratory and is a well-known basidiomycete 15 genus with many free-living forms (Hervey et al. 1977); the other, Attamyces bromatiflcus. produced vegetative mycelia but did not produce sporophores (Kreisel 1972). Few investigations have examined the relationships between the Solenopsis complex and its internal microflora. Most research on the red imported fire ant are control oriented or emphasize the species negative Impact on ecosystems. Efforts were recently made to investigate the microflora of S.. invicta: however, the trend is directed particularly towards finding pathogens of the red imported fire ant. In fact, protozoa, viruses, and fungi were Isolated from the red Imported fire ant In its native lands (Allen and Buren 1974; Avery et al. 1977; Allen and Silveira-Guido 1974). Interest in using insect pathogens within integrated pest management programs has generated mission-oriented research with the objective of developing fungal pathogens as microbial insecticides (Burges 1971). Much of this research has focused on Isolation, development, and production of highly pathogenic microbes with genetic characteristics favoring their long-term storage (Ferron 1981). Several species of microsporidia and neogregarines were isolated from ants. The microsporidium Thelohania solenopsae was isolated from 12 species of Solenopsis from Uruguay and Argentina (Jouvenaz 1983). This microsporidium invades ovaries of queens and adipose tissues of workers (Knell et al. 1977). Thelohania solenopsae is transmitted transovarially but can not be transmitted Q^Q^. Infected colonies do not live as long as healthy colonies under laboratory conditions (Jouvenaz 1983). This microsporidium 16 was isolated from 67 of 865 colonies of fire ants (7.8%) collected from Matto Grosso, Brazil (Jouvenaz 1983). A second microsporidium, Burenella dimorpha. was also isolated from Solenopsis species (Jouvenaz 1983), and pupae succumbed to infection, although the fourth Instar larva was that stage which must be infected (Jouvenaz 1983). Intracolonlal infestation was usually below 5%, but some colonies occasionally reached 100% (Jouvenaz 1983). In addition, the neogregarine Mattesia geminata was isolated from the tropical fire ant. Solenopsis geminata (F.), in Florida (Jouvenaz 1983). Little information on yeasts as part of the microflora of ants exists. Yeast flora of the mound soil of Formica rufa L. was investigated by Golubev and BabeVa (1972). From this investigation, a large number of yeast cells, mainly species of the genus Debaryomyces were recovered. A total of 213 isolates of Debaryomyces was obtained, 143 strains were identified as d. cantarelli CapriottI, four strains as D.. hansenii (Zopf) Lodder et Kreger-van Rij, and 66 Isolates represented a new species, Debaryomyces formicarius Golubev and Bab'eva. However, the authors failed to describe the method of yeast isolation and mainly emphasized the characterization of the new yeast species. The significance of this association was not reported, and Internal yeast flora of the ant was not investigated further. However, higher nutrient concentration was found In mound soil of the western harvester ant, Pogonomyrmex occldentalis (Cresson) (Rogers and Lavine 1974), the Florida harvester ant, Pogonomyrmex badius 17 Latreille (Gentry and Stritz 1972), and Solenopsis invicta (Herzog et al. 1972). Higher nutrient concentration in mound soil may be explained by the fact that ants gather and store food in the mound. In addition, distribution of food via trophylaxis may contribute to nutrient accumulation. However, food accumulation alone may not explain the predominance of the genus Debaryomyces in the mound of Formica rufa.

In the search for pathogens, yeasts have been isolated from haemolymph of S.- invicta. However, little Information exists pertaining to yeasts as a normal part of the microflora of the red imported fire ant. In 1977, Jouvenaz et al. surveyed pathogens of S.. invicta In the southeastern part of the United States and found a benign, mildly pathogenic, and unidentified yeast associated with 9.3% of S.. invicta colonies. The survey showed that 34 of 54 colonies collected from Mobile Co., Alabama, harbored the yeast. In addition, it was present in six of 20 colonies collected from three counties in northern Alabama. Also, this species was isolated from 23 of 86 colonies from nine northwestern Florida counties and one from an adjacent Georgia county. In southern Florida, the yeast was present in 11 of 86 colonies and 16 of 117 colonies from northeastern Florida. Although biotic factors in different geographical areas may influence yeast prevalence in red imported fire ant colonies, the authors did not explain the possible causes of incidence variation observed at the different sites. Another point of contention is the methodology used for yeast detection. Samples of 1000-2000 mixed adult and immature ants were triturated in tissue 18 homogenizers and water (Jouvenaz 1977), and the homogenate was observed using phase-contrast microscopy. This method imposes extreme limitations in detecting yeasts because yeast cells may not be distinguished clearly from ant tissue. In addition, the origin of the yeasts may be from an external source or from the gut of the insect. A unicellular fungus was isolated from S.. Invicta haemolymph by Jouvenaz et al. (1981). The cells of the fungus were described as club-shaped and multiplied by budding. This organism proliferated as a yeast in S.. invicta haemolymph without being lethal to the ant (Jouvenaz et al. 1981). Field populations of red imported fire ants were not reduced even in areas were the yeast prevalence approached 50% (Jouvenaz 1986). Therefore, pathogenicity of the yeast occurring in S.. invicta haemolymph was not established. However, these reports also failed to describe the yeast to species. Jouvenaz (1989) reported at least three species of endoparasitic fungi occurring as budding yeasts in red imported fire ant haemolymph in the United States and South America. However, the study failed to characterize the yeasts, and pathogenicity of the association was neither observed nor was it established. In addition, two strains of a parasitic yeast, differing only In assimilation of celloblose and colonial morphology, were isolated from S.. invicta haemolymph in the United States and Argentina 19 (Jouvenaz and Kimbrough 1991). The yeast was described as a new species, Mvrmecomyces annellisae Jouvenaz, representing a new genus. However, changes in behavior In ants harboring the yeast were not observed. In many instances, mutualistic associations have been suggested between symbionts and Insect hosts. These studies suggest that the symbiotic microorganisms may provide vitamins, carbohydrates, amino acids, and lipids to the insect, regulate host reproduction, function as chemical cues responsible for insect aggregation, detoxify plant allelochemlcals, and may be a metabolic drain to the host or even pathogenic. In some Instances, the hypothesis that symbiosis may promote speciatlon has been raised, but at this point, many of the aforementioned attributes are merely speculative. Mutualism Is a complex phenomenon and should be approached with rigorous testing to assess fully the nature of the relationship. The fact that few microbiologists work with the red imported fire ant has delayed information regarding the natural microflora of this species. The objectives of this research were: (1) to isolate and identify yeasts from the red imported fire ant (Solenopsis invicta): (2) to determine the impact of associated yeasts on red imported fire ant colony vigor; and (3) to clarify the significance of the association by obtaining and comparing sterol profiles of the yeast flora and the red imported fire ant. 20 References Cited

Adams, C. T. 1986. Agricultural and medical impact of the imported fire ants, in: Fire ants and leaf-cutting ants. Biology and Management. C. S. Lofgren and R. K. Vander Meer, Eds.. Westvlew, Boulder, Colorado, pp. 48-57.

Akey, D. H., and S. D. Beck. 1972. Nutrition of the pea aphid Acyrthoslphum pisum: requirements for trace metals, sulphur, and cholesterol. J. Insect Physiol. 18:1901-1914.

Allen, G. E., and G. A. Silviera. 1974. Occurrence of microsporidia in Solenopsis richteri and Solenopsis sp. in Uruguay and Argentina. Fla. Entomol. 57:327-329.

Allen, G. E., and W. F. Buren. 1974. Microsporldian and fungal diseases of Solenopsis invicta Buren in Brazil. J. New York Entomol. Soc. 82:125-130.

Amiressami, M. 1980. Investigation of the light microscopical and ultrastructure of the demeton-S-methyl resistance aphids under consideration of the mycetome symbionts of the Phorodon humili Schrank. in: Endocytobiology, endosymbiosis and cell biology: a synthesis of recent research. W. Schwemmler and H. E. A. Schenk, Eds. Walter de Gruyter, New York. pp. 425-443.

Amiressami, M. and H. Petzold. 1976. Symbioseforeschung un Insektizidresistenz. Ein llcht-und elektronenmikroskopischer Beitrag zur Klarung der Insektizidresistenz von Aphiden unter Beruksichtigung der Mycetom-Symbioten bei Myzus persicae Suiz Z. Angnew. Entomol. 82:253-259.

Avery, S. W., D. P. Jouvenaz, W. A. Banks, and D. W. Anthony. 1977. Virus-like particle in a fire ant, Solenopsis spp. (hymenoptera: Formicidae) from Brazil. Fla. Entomol. 60:17-20.

Baker, W. L. 1972. Eastem forest insects. U.S.D.A. Forest Serv. Misc. Publ. 1175. 642 pp. 21 Batra, L. R., S. W. T. Batra, and G. E. Bohart. 1973. The mycoflora of domesticated and wild bees (Apoidae). Mycol. Appl. 49:13-44.

Brooks, M. A. 1963. Symbiosis and aposymbiosis in anthropods. In: Symbiotic associations. 13th Symp. Soc. gen. Microbiol. Nutman, P. S. and B. Mosse, Eds. Cambridge U. P., London, pp. 200-231.

Brooks, M. A., and A. G. Richards. 1956. Intracellular symbiosis In cockroaches. III. reinfection of aposymbiotic cockroaches with symbiotes. J. Exp. Zool. 132:447-465.

Brooks, M. A., and richards. 1955. Intracellular symbiosis in cockroaches: Production of aposymbiotic cockroaches. Biol. Bull. 109:22-39.

Buchner, P. 1965. Endosymbiosis of animals with plant microorganisms. Interscience, New York. p. 909.

Burges, H. D., and Thompson. 1971. Standardization and assay of microbial insecticides, in.: Microbial control of insects and mites. H. D. Burges and N. W. Hussey, Eds. Academic Press, London, pp. 591-621.

Campbell, B. C, and W. D. Nes. 1982. A reappraisal of sterol biosynthesis and metabolism in aphids. J. Insect Physiol. 29:149-156.

Cherret, J. M., R. J. Powell, and D. J. Stradling. 1989. The mutualism between leaf-cutting ants and their fungus, in: Insect fungus interaction, N. Wilding, N. M. Collins, P. M. Hammond, and J. F. Webber, Eds. Academic Press, London, pp.93-120.

Clari<, M. A., L. Bauman, M. A. Munson, P. Bauman, B. C. Campbell, J. E. Duff us, L. S. Osborne, and N. A. Moran. 1992. The eubacterial endosymbionts of whiteflies (Homoptera: Aleyrodoidae) constitute a lineage distinct from the endosymbionts of aphids and mealybugs. Current Microbiol. 25:119-123.

Clayton R. B. 1964. The utilization of sterols by insects. J. Lipids Res. 5:3-19. 22

Cochran, D. G. 1985. Nitrogen excretion in cockroaches. Annu. Rev. Entomol. 30:29-49.

Dadd R. H.. and Krieger D. L. 1968. Dietary amino acid requirements of the aphid, Myzus persicae. J. Insect Physiol. 14, 741-764.

Dadd, R. H., and T. E. Mittler. 1966. Permanent culture of an aphid on a totally synthetic diet. Experientia. 22:832-835.

Dash, G. A., E. Weiss, and K. P. Chang. 1984. Endosymbionts of insects, in: Bergey's Manual of Systematic Bacteriology, vol. I. N. R. Krieg, Ed. William and Wilkins, Baltimore, pp. 811-836.

Domek, J. M., and D. T. Johnson. 1988. Demonstration of semiochemically induced aggregation in the green June beetle, Cotinis nitida (L.) (Coleoptera: Scarabaeidae). Environ. Entomol. 17:147-149.

Domek, J. M., and D. T. Johnson. 1990. Inhbition of aggregation behavior in the green June beetle, Cotinis nitida (L.) (Coleoptera: Scarabaeidae). Environ. Entomol. 19:995-1000.

Douglas, A. E. 1989. Mycetocytes symbiosis In insects. Biol. Rev. 64:409-434.

Dowd, P. F. 1989. In situ Production of hydrolytic detoxifying enzymes by symbiotic yeasts in the cigarette beetle (Coleoptera: Anobiidae). J. Econ. Entomol. 82:396-400.

Dowd, P. F. 1991. Symbionts-medlated detoxification In insect herbivores, in: Microbial mediation of plant-herbivore interactions. P. Barbosa, V. A. Krischik, and C. G. Jones, Eds. John Wiley & Sons Inc., New York. pp. 411-440.

Ehrhardt, P. 1968. Nachweis einer durch symbiontische microorganimen bewirkten sterinsynthese in kunstlich aphiden. (Homoptera, Rhinchota, Insecta). Experientia. 24, 82-83. 23 Ferron, P. 1981. Pest control by the fungi: Beauveria and Metarhizium. in: Microbial control of pests and plant diseases. H. D. Burges, Ed. Academic Press, New York. pp. 465-482.

Fogleman, J. C. 1982. the role of volatiles in the ecology of cactophllic Drosophila. In: Ecological Genetics and Evolution: The Cactus-veast-Drosophila Model System (J. S. F. Barker, and W. T. Starmer, eds), Academic Press Australia, Sydney, pp. 191-206.

Fogleman, J. C, and H. W. Kircher. 1986. Differential effects of fatty acid chain length on the viability of two species of cactophllic Drosophila. Comp. Biochem. 83A:761-764.

Fogleman, J. C, W. T. Starmer, and W. B. Heed. 1981. Larval selectivity for yeast species by Drosophila mojavensis in natural substrates, Proc. natn. Acad. Sci. USA. 78:4435-4439.

Furniss, R. L., and V. M. Carolin. 1977. Western forest insects. U.S.D.A. For. Serv. Misc. Publ. No. 1339. 654 pp.

Gentry, B. G. and K. L. Stiritz. 1972. The role of the Florida ant, Pogonomyrmex badius. in old field mineral nutrient relationships. Environ. Entomol. 1:39-41.

Gilliam, M.,L. J. Wickerham, H. L. Morton, and R. D. Martin. 1974. Yeasts isolated from honey bees. Apis mellifera fed 2, 4- D and antibiotics. J. Invertrbr. Pathol. 24: 349-346.

Glancey, B. M., J. C. E, Nickerson, D. P., Wojclk, J. Trager, and C. T. Adams. 1987. The increasing incidence of the polygenous form of the red imported fire ant, Solenopsis invicta (Hymenoptera: Formicidae) in Florida. Florida Entomologist. 60:275-2789.

Golubev, V. I., and I. P. Bab'Eva. 1972. Debaryomyces formicarius sp. n. and Debaryomyces cantarelli associated with the ants of the group Formica rufa L. J. Gen. Appl. Microbiol. 8:249-254. 24 Grasse, P. P. 1952. Role des flagettes symbiotiques chez les blattes et les termites. TIdschr. Ent. 95:70-80.

Green, H. B. 1967. The imported fire ant in Mississippi. Miss. Agr. Expt. Stat. Bull. 737, 23 pp.

Griffiths, G. W., and S. D. Beck. 1977a. In vivo sterol biosynthesis by pea aphid symbiotes as determined by digitonin and electron microscopic autoradiography. Cell TIss. Res. 176: 179-190.

Griffiths, G. W., and S. D. Beck. 1977b. Effect of dietary cholesterol on the pattern of osmium deposition in the symbiote- containing cell of the pea aphid. Cell TIss. Res. 176: 191-203.

Hervey, A., C. T. Rogerson, and I. Leong. 1977. Studies on fungi cultivated by ants. Brittonia 29:226-236.

Herzog, D. C, T. E. Reagan, D. C. Sheppard, K. M. Hyde, S. S. Nllakhe, M. Y. B. Hussein, M. L. Mcmahan, R. C. Thomas, and L. D. Newsom. 1974. Solenopsis invicta Buren: Influence on Louisiana pasture soil chemistry. Environ. Entmol. 5:160-162.

Hinde, R. 1971. The control of the mycetome symbiotes of the aphids Brevicoryne brassicae. Myzus persicae. and Macrosiphum rosae. J. Insect Physiol. 17:1791-1800.

Houck, E. J., G. W. Griffiths, and S. D. Beck. 1976. Lipid metabolism in the symbiotes of the pea aphid, Acyrthosiphon pisum. Comp. Biochem. Physiol. 54B:427-431.

Johnson, D. T., and H. S. Vishniac. 1991. The role of Trichosporon cutaneum in eliciting aggregation behavior in Cotinis nitida (L.) (Coleoptera: Scarabeidae). Environ. Entomol. 20:15-21.

Jones, C. G. 1984. Microorganisms as mediators of plant resource exploitation by insect hertDlvores. in: A new ecology, novel approaches to interactive systems. P. W. Price, C. N. Slobodchikoff, and W. S. Gould, Eds. Wiley, New York. pp. 53-99.

Jouvenaz, D. P. 1983. Natural enemies of fire ants. Fla. Entomol. 66:111-121. 25

Jouvenaz, D. P. 1986. Diseases of fire ants: Problems and opportunities.ln: Fire ants and leaf-cutting ants: Biology and management. C. S. Lofgren, and R. K. Vander Meer (Eds), Fire ants. Westvlew Press, Boulder, Colorado, pp. 280-296.

Jouvenaz, D. P. 1989. Endoparasitic fungi of fire ants. Proceedings of the 1989 imported fire ant conference, pp. 19-20.

Jouvenaz, D. P. G. E. Allen, W. A. Banks, and D. P. Wojclk. 1977 A survey for pathogens of fire ant Solenopsis spp. in the southeastern United States, Fla. Entomol. 60:275-279.

Jouvenaz, D. P., and J. W. Kimbrough. 1991. Mvrmecomyces gnnQlliggg gen. nov., sp. nov. (Deuteromycotina: Hyphomycetes), an endoparasitic fungus of fire ants, Solenopsis spp. (Hymenoptera: Formicidae). Mycol. Res. 95. 12:1395-1401.

Jouvenaz, D. P., C. S. Lofgren and W. A. Banks. 1981. Biological control of imported fire ants: A review of current knowledge. Bull. Entomol. Soc. America, 27:203-208.

Kartson, P. and H. Hoffmeister. 1963. Biogenesis of ecdysone: conversion of cholesterol to ecdysone. Z. Physiol. Chem. 331:298-300.

Kircher, H. W. 1982. Sterol and insects, in: Cholesterol system in insects and animals. J. Dupont, Ed. CRC Press, Boca Raton, Florida, pp. 1-50.

Knell, J. D., G. E. Allen, and E. I. Hazard. 1977. Light and electron microscope study of Thelohania solenopsae n. sp. (microsporidia: Protozoa) in the red imported fire ant, Solenopsis invicta. J. Invertebr. Pathol. 29:192-200.

Koch, A. 1960. Intracellular symbiosis in insects. Annu. Rev. Microbiol. 14:121-140.

Koch, A. 1967. Insects and their endosymbionts. In: Symbiosis, vol. 2. M. Henry, Ed. Academic Press, New York. pp. 1-96. 26

Kreisel, H. 1972. Pilze aus Pilzgarten von Atta insularls in Kuba. Z.AIIg. Mikrobiol. 12:643-654.

Kuhlwein, H., and G. Jurzita. 1961. Studlen an der symbiose der anobiiden. I. Die Kultur des symbioten von Sitodrepa panlcea. Arch. Mikrobiol. 40:247-260.

Lachance, M. A., W. T. Starmer, and H. J. Phaff. 1988. Identification of yeasts found in decaying cactus tissue. Can. J. Microbiol. 14:79-86.

Leufeven, A. and L. Nels. 1986. Ouantification of different yeasts associted with the bark beetle, Ips typographus. during its attack on a a spruce tree. Microb. Ecol. 12:237-243.

Lodder, J. and Kreger-van Rij, N.J.W. 1952. The yeasts, a taxonomic study. Interscience Publishers, Inc. New York. pp. 713.

Lofgren, C. S., W. A. Banks, and B. M. Glancey. 1975. Biology and control of imported fire ants. Annu. Rev. Entomol. 20:1-30.

Martin, M. M. 1992. The evolution of insect-fungus associtions: from contact to stable symbiosis.Amer. Zool. 32:593-605.

Metcalf, C. L., W. P. Flint, and R. L. Metcalf. 1962. Household insects and pests of stored grain, seeds, and cereal products, in: Destructive and useful insects: their habits and control. C. L. Metcalf, W. P. Flint, and R. L. Metcalf, Eds. McGraw-Hill, New York. pp. 888-938.

Meyer, S. A., D. G. Ahearn, and D. Yarrow. 1984. Genus 4. Candida Berkhourt. in: The yeasts, a taxonomic study, 3rd edn. N. J. Kreger-van Rij, Ed. Elsevier Science, Amsterdam, Holland, pp. 585-844.

Mullins, D. E. and D. G. Cochran. 1975. Nitrogen metabolism in the American cokroaches. Comp. Biochem. Physiol. 50A:489-510. 27 Munson, M. A., P. Bauman, M. A. Clark, L. Baumann, N. A. Moran, D. J. Voeglin, and B. C. Campbell. 1991. Evidence for the establishment of aphid-eubacterium endosymbiosis in an ancestor of four aphids families. J. Bacterid. 173:6321- 6324.

Muller, W. 1934. Untersucungen uber die Symbiose von Tieren mit Pilzen und Bakteriem. Uber die Pilzsymbiose holzfressender Insektenlarven. Arch. Mikrobiol. 5, 84-47.

Nardon, P., and A. M., Grenler. 1991. Serial symbiosis theory and weevil evolution: the role of symbiosis, in: Symbiosis as a source of evolutionary Innovation. L. Margulis and R. Fester, Eds. MIT Press, Cambridge, Massachusetts, pp.153-169.

Nardon, P., and C. Wicker. 1981. La symbiose chez le genre Stophilus (Coleoptere Curculionldae). Principaux aspects morphologlques, physiologlques et genetiques. Ann. Biol. 20:327-373.

Pant, N. C, and G. Fraenkel. 1954. Studies on the symbiotic yeasts of two insect species, Lasloderma serricrne F. and Stegoblum paniceum L. Biol. Bull. 107:420-432.

Phaff, H. J., and W. T. Starmer. 1987. Yeasts associated with plants, insects and soil, in: The yeasts. Biology of Yeasts. A. H. Rose and J. S. Harrison, Eds. Academic Press. New York. pp. 123- 179.

Pierre, L. L. 1962. Synthesis of ascorbic acid by the normal fat body of the cockroach Leucophae maderae. and by its symbionts. Nature. 201:54-55

Rogers, L. E., and R. J. Lavigne. 1974. Environmental effects of western harvester ants on the shortgrass plains ecosystem. Environ. Entomol. 3:994-997.

Sandhu, D. K., and M. K. Waraith. 1985. Yeasts associated with pollinating bees and flower nectar. Microbio. Ecol. 11:51-58. 28 Shifrine, M., and H. J. Phaff. 1956. The association of yeasts with certain bark beetles. Mycologia. 48:41-55.

Starmer, W. T., and J. C. Fogleman. 1986. Coadaptation of Drosophila and yeasts in their natural habitat. J. Chem. Ecol. 12:1037- 1055.

Stamner, W. T., and V. Aberdeen. 1990. The nutritional Importance of pure and mixed cultures of yeasts in the development of Drosophila mulleri larvae in opuntia tissues and its relationship to host plant shifts, in: Ecological and evolutionary genetics of Drosophila. J. S. F. Barker, W. T. Starmer, and R. J. Maclntyre, Eds. Plenum Press, New York. pp. 145-160.

Starmer, W. T., F. Peris, and A. Fontdevila 1988, the transmission of yeasts by Drosophilia buzzattii during courtship and mating. Anim. Behav. 36:1691-1695.

Starmer, W. T., J. C. F. Baker, H. J. Phaff, and J. C. Fogleman. 1986. Adaptation of Drosophila and yeasts In their interactions with the volatile 2-propanol in the cactus-microorganism- drosophila model system. Aust. J. Biol. Sci. 39:69-79.

Starmer, W. T., M. A. Lachance, H. J. Phaff, and W. B. Heed. 1990. The blogeography of yeasts associated with decaying cactus tissue in North America, the Caribbean and Nothern Venezuela. Evol. Biol. 24:253-296.

Starmer, W. T., W. B. Heed, M. Miranda, M. W. Miller, and H. J. Phaff. 1976. The ecology of yeast flora associated with cactiphllic Drosophila and their host plants In the Sonoran Desert. Microbiol Ecology. 3:11-30.

Steinhaus, E. A., and Marsh, G. A. 1967. Previously unreported accesions for diagnosis and new records. J. Invertebr. Pathol. 9:436-438. 29 Svoboda, J. A., J. N. Kaplans. W. E. Robbins, and M. J. Thompson. 1975. Recent developments in insect steroid metabolism. Ann. Rev. Entomol. 20:205-220.

Taylor, F. J. R. 1983. Some eco-evolutlonary aspects of intracellular symbiosis, in: Intracellular symbiosis. In K. W. Jeon, Ed. Academic Press, New York. pp. 1-28.

Vacek, D. C, P. D. East, J. S. F. Barker, and M. H. Soliman. 1984. Feeding and oviposition preferences of Drosophila buzzatii for microbial species Isolated from its natural environment. Biological journal of the Linnean Society. 24:175-187.

Van der Walt, J. P., D. B. Scott, and W. C. Van der Klift. 1972. Six new Candida species from South African Insect sources. Mycopath. Mycol. Appl. 47:221-236.

Vining, L. C. 1984. Microbial transformations of alkaloids and nitrogenous compounds, in: Handbook of microbiology, vol. Vll, microbial transformation. A. I. Laskin and H. A., Eds. CRC Press, Boca Raton, Florida, pp. 467-477.

Vishniac, H. S., and D. T. Johnson. 1990. The development of yeast flora in the adult green June beetle, Cotinis nitida (L.) (Coleoptera: Scarabeidae). Mycologia. 82:470-478.

Weber, N. A. 1972. Gardening ants, the attine. Mem. Am. Phil. Soc. 92:1-146.

Whitney, H. S. 1982. Relatioships between bark beetles and symbiotic organisms, in: Bark beetle In North American conifers: a system for the study of evolutionary biology. Milton, J. B., and K. B. Surgeon (Eds). University of Texas Press, Austin, Texas, pp. 183-211.

Wren, H. N., and D. G. Cochran. 1987. Xanthine dehydrogenase activity in the cockroach endosymbiont Blattabacterlum cuenoti (Mercier, 1906), and in the cockroach fat body. Comp. Biochem. Physiol. 88B:1023-1026. 30 Yarrow, D., and S. A. Meyer. 1978. Proposal for amendement of the diagnosis of the genus Candida Berkhout nom. cons. Int. J. Syst. Bacterid. 28:611-615. CHAPTER II IDENTIFICATION AND ABUNDANCE OF YEASTS ISOLATED FROM RED IMPORTED FIRE ANT HAEMOLYMPH

Introduction Although the literature regarding medically and industrially important yeasts is abundant, the identification of yeast from the general environment poses pitfalls. While imposing certain limitations, identification schemes used for medical and industrial purposes are of some value in identifying organisms isolated from the general environment. Even so, a combination of different schemes is necessary for identification to species. The purpose of this portion of the investigation was to isolate and identify yeasts from the haemolymph of red imported fire ant workers. Yeasts were identified based on morphology, physiology, and biochemistry. Three groups of yeasts are recognized: the ascosporogenous (Ascomycetes), the basidiosporogenous (Basidiomycetes), and the imperfect yeasts (Deuteromycetes) (Barnett et al. 1983). The Diazonium Blue B (DBB) test stains basidiosporogenous yeasts dark red but does not induce color change in ascoporogenous yeasts. Ascosporogenous yeasts are differentiated from imperfect yeasts by staining with malachite green and counterstaining with safranin. This procedure confirms the presence of ascospores in ascoporogenous yeasts or the absence of ascospores in Imperfect yeasts. Therefore, these simple staining procedures can be used to separate the three classes of yeasts.

31 32 Materials and Methods Collection of Colonies

Collection sites. Colonies were collected from five locations in Texas: Beaumont, Huntsville, Waco, Stephenville, and Abilene. These sites represent a transect of red imported fire ant-infested regions along a decreasing moisture gradient from the southeastern to the northwestern region of the state. The amount of precipitation (ten year means obtained from the U.S. National Weather Service) is highly correlated to sites (r = 0.97; Cricket Software 1987), supporting the existence of this moisture gradient from Beaumont to Abilene (Figure 2.1). Beaumont is in Jefferson County. This county Is located in the extreme southeastern part of Texas (Crout et al. 1960) and is bordered by the Gulf of Mexico to the south. The climate has characteristics of both tropical and temperate zones, and sea breezes prevent high temperature during summer months. The county is mostly a flat, featureless plain and is dissected by very few streams (Crout et al. 1960). Red imported fire ant colonies were collected 7 km from Beaumont towards the town of Sour Lake on State Highway 105. Huntsville Is located in Walker County which is in the southeastern part of Texas. The topography is undulating to gently rolling (McClintock et al. 1979). The climate is hot during summer 33 months and cool during winter months, and the humidity often reaches 90% (McClintock et al. 1979). Colonies were collected along

Interstate 45, 7 km south of the Intersection of State Highway 30 and Interstate 45 near Huntsville.

Waco is the county seat of Mclennan County. This county is located in the Blackland Prairie of central Texas (Templln et al. 1958). Colonies were collected 17 km from Waco at the intersection of State Highways 6 and 340. Stephenville Is located In Erath County. This county is In the central part of Texas and has a warm-temperate, permanently humid climate characterized by hot summers (Wagner et al. 1973). The relative humidity is fairly uniform throughout the year, although slightly lower during the summer months (Wagner et al. 1973). Colonies were collected 400 m south of Ingraham Road. Abilene is located In Taylor County. This county is in north- central Texas and lies within the Edwards Plateau (Conner et al. 1976). This county represents the boundary between the humid climate of eastern Texas and the semiarid climate of northwestern Texas (Conner et al. 1976). Colonies were collected beginning at the intersection of County Roads 707 and 1750. Method of collection. Ten colonies were randomly (stratified) collected (166 m apart) from each site during the winter of 1992, and the spring, summer, and fall months of 1993. Colonies were collected 12 December 1992, 15 March, 17 June, and 25 September 1993. Therefore, 50 colonies were collected each season (5 sites x 10 colonies), resulting in a total of 200 colonies for the 34 Investigation (10 colonies per site x 5 sites x 4 seasons). Mounds were excavated with shovels and placed in plastic buckets. Shovels were washed in 25% Clorox solution (5.25% sodium hypochlorite) to avoid possible yeast contamination between mounds. The Inner surface of the buckets was dusted with talc (unscented baby powder; Hy-Top) to prevent escape of ants. Handling of colonies. Colonies were transported to the laboratory and separated from soil according to the methods of Banks et al. (1981). Ants were scooped from water and placed In lightweight, clear styrene boxes (111 x 111 x 39 mm; Tri-State plastics, Dixon, KY) with a paper towel lining the bottom to absorb excess water. Boxes were sealed hermetically to prevent the escape of ants. After 30 min. ants were transferred to another empty box of known weight (Mettler PL 3000 balance) and weighed again. The final weight of each colony was recorded in grams (to the nearest 0.1 g). Colonies were maintained at room temperature and fed every other day a sterile diet (dog food; Alpo brand altemated with boiled eggs). Sterile, deionlzed water was always available to the ants.

Isolation of Yeasts from Adult Workers A total of 10,000 worker ants were processed during this Investigation. Each season 2,500 ants were screened for yeasts (50 workers per colony x 10 colonies per site x 5 sites). Workers were surface sterilized by successive immersions in 70% ethyl alcohol for 5 min., 10% Clorox solution with 1 drop/200 ml of wetting agent Triton X-100 (Sigma) for 5 min. and three rinses in sterile. 35 deionlzed water (Poinar and Thomas 1978). Ants were dead following sterilization procedures. Haemolymph was extracted using standard sterile procedures. The second tergite of the gaster was removed, with care being taken not to sever the gut. Ants were held dorsal side down to allow haemolymph to flow onto plates (100 X 15 mm) containing acidified yeast extract-malt extract agar [AYM: Difco YM agar plus 0.7% (v/v), 1 M hydrochloric acid, final pH of 3.7- 4.0] to inhibit bacterial growth (Starmer et al. 1976; Phaff et al. 1978). Intact, surface-sterilized ants were used as controls for the sterilization procedure by placing 50 specimens from the same colony on AYMA plates overnight. To Induce growth, plates were incubated for approximately three weeks at 28°C which Is also presumed to be the optimal growth temperature for the red Imported fire ant. After growth, each representative colony type was subcultured for purification on AYM media.

Isolation of Yeasts from Worker Larvae Fourth instar larvae from colonies collected on 25 September 93 were also screened for the presence of yeasts. Only colonies containing at least 50 fourth Instar larvae were chosen. A total of 12 colonies, three each from Abilene, Waco, Huntsville, and Beaumont, was Investigated. Fourth instar larvae were not present in colonies collected from the Stephenville site at that time. Six hundred lan/ae were processed (3 colonies x 4 sites x 50 larvae). Meconia were removed with fine-tipped forceps from larvae, washed with sterile, deionlzed water, and placed on AYMA plates which were 36 incubated at 28oC. In addition, 10 lan/ae (lan/al weight ca. 0.50 mg) from each of the 12 colonies were individually homogenized in a tissue grinder containing 99.5 ^il of sterile, deionlzed water. The homogenate was mixed with 9.9 ml of sterile, deionlzed water and vortexed. From this mixture, 0.1 ml was plated on AYM agar and Incubated at 28^0 for three weeks. Yeasts present in the in third instar larvae were also investigated. This larval stage was not prevalent in any of the colonies reared in the laboratory. However, a total of 50 third Instar larvae was obtainable from all colonies, and each was Individually screened for yeasts by culturing the gut content on AYMA plates incubated at 28oC for three weeks. Sterilization procedures for all larvae were the same as those previously stated for adults.

Maintenance of Yeast Cultures Pure cultures were obtained from single, well-Isolated colonies which were subcultured twice. Yeasts were maintained on glucose peptone yeast agar (GPYA; Difco, Detroit, Ml) to reduce potential mutations In their fermentation and assimilation abilities (Arx and Shipper 1978). Cultures were stored at 5^0 and subcultured at 3- to 4-month intervals. 37 Identification of Yeasts

Yeast isolates were physiologically characterized according to the methods of Wickerham (1951) and van der Walt and Yarrow (1984). Cellular morphology and mode of reproduction of the strains were determined using malt extract agar. Formation of pseudomycelia and true mycelia were assessed by the Dalmau plate technique (yeast suspensions growing under sterile coversllps) of Wickerham (1951). Assimilation and fermentation tests of carbohydrates were performed following the procedures of Wikerham (1951) and van der Walt and Yarrow (1984). In addition, API 20C System was used for yeast identification. Yeast isolates were also sent to Centraalbureau voor Schimmelcultures, Delft, Netheriands, for confirmation. Preparation of inoculum. Prior to conducting the assimilation and fermentation tests, cultures were grown on starvation media consisting of Bacto yeast nitrogen base (Difco) with 1.5% (w/v) agar (Difco) and 0.1% glucose (Sigma, St. Louis, MO) as suggested by Dalton et al. (1984) and Incubated at 28oC for four days. The inoculum was standardized (Wickerham 1951) by diluting with sterile deionlzed water until black lines drawn on white cardboard could only be seen as diffuse bands, equivalent to +2 growth (approximately 10^ cells ml-i).

Quality of the media for yeast identification was assessed using known pure yeast cultures received from C. P. Kurtzman of the Nothern Regional Research Laboratory, Peoria, Illinois. The following pure stock Isolates were obtained: NRRL Y-173 (Candida 38 famata). NRRL Y-17301 FSporidiobolus (Sporidiobolomyces) salmonicolorl. NRRL Y-1774 (Candida zeylanoides^. NRRL Y-11996 (Candida cariosilignicolaV NRRL Y-1498 (Candida tenuis^. NRRL Y- 10942 (Candida naeodendraV NRRL Y-1589 (Rhodoturula mlnuta^. NRRL Y-2332 (Candida boidinin. and NRRL Y-8281 (Kluveromvces marxianus). Macromorphology. Pigmentation, texture, elevation, margin, and shape of colonies were recorded after three weeks Incubation at 280C on 5% malt extract agar (Difco) (van der Walt and Yarrow 1984). The presence of pellicle, ring, and sediment formation was obsen/ed In yeast morphology broth (YM, Difco) cultures after 2, 7 and 21 days at 28oC. Micromorphology. The formation of pseudohyphae, true mycelia, and arthrospores was determined by microscopic examination of Dalmau plates on corn meal agar (Difco) after 10 days incubation at 280C (Wickerham 1951; Van der Walt and Yarrow 1984). Vegetative reproduction, cell size and shape were examined in YM broth after two days incubation at 28oC and on 5% malt extract agar after three days Incubation at 28^0 (Van der Walt and Yarrow 1984). After 21 and 28 days incubation at 28oC, water suspensions of yeast cultures were made from com meal agar, 5% (w/v) malt extract agar (Difco) (Wickerham 1951), and from slants of potassium acetate agar [1% (w/v) potassium acetate, 0.1% (w/v) glucose, 0.25% (w/v) yeast extract, 1.5% (w/v) agar] for microscopic examination of ascospores (100X). Heat-fixed yeast cells were stained with 5% (w/v) aqueous malachite green and counterstalned with 5% (w/v) safranin (Van der 39 Walt and Yarrow 1984), to confirm the presence or absence of ascospores. Water mounts were used to establish the detailed shape of ascospores. API 2QC system. The API 20C system Is used by medical laboratories for yeast identification and consists of plastic strips containing dehydrated reagents for 19 assimilation tests and a control. This system was employed first in the Identification scheme of yeasts in the study herein, because it Is efficient, reliable, and quickly limits the choices for Identification. The assimilation of compounds was tested In the following order: glucose, glycerol, 2-keto-gluconate, arabinose, xylose, adonitol, xylltol, galactose. Inositol, sorbose, methyl-alpha-D-glucoside, N- acetyl-glucosamine, cellobiose, lactose, maltose, saccharose, trehalose, melzitose, and raffinose. The test strips innoculated with yeasts were Incubated at 30°C, and results were recorded after 24, 48, and 72 hours. Identifications were made by referring to the API Analytical Profile Index. Some numbers did not correspond to species listed in the index. Therefore, the API Computer Service Center was contacted to obtain the corresponding species. Fermentation of carbohydrates. Fermentation of carbohydrates by yeast species assesses their ability to use specific sugars anaerobically by observing the formation of gas (CO2). One part (0.5 ml) of filter-sterilized (Whatman filters, 0.2^m), 6.7% (w/v) Bacto yeast nitrogen base (Difco), and the carbon source [2% (w/v) of either glucose, celloblose, galactose, maltose, sucrose or 4% (w/v) raffinose; Sigma] was added to 9 parts (4.5 ml) sterile, deionlzed 40 water In test tubes (16 x 150 mm) containing Durham tubes (6 x 25 mm) (Wickerham 1951; Van der Walt and Yarrow 1984). Tests tubes were inoculated with 0.1 ml of yeast suspensions and Incubated at 280C. Observations for gas production as an Indication of fermentation were made at 2, 7, 14, 21 and 28 days. Assimilation of organic compounds. Assimilation tests were used to assess the ability of different yeasts to use organic compounds for aerobic growth. Yeast species differ In their abilities to utilize certain organic compounds. For carbon compound assimilation tests, a 10-fold concentrated medium was prepared from 6.7 g of Bacto yeast nitrogen base (Difco) and 5 g of the carbohydrate In 100 ml sterile, deionlzed water (Wickerham 1951; Van der Walt and Yarrow 1984). The solution was sterilized by filtration (Whatman filters, 0.2 jam), and aliquots of 0.5 ml of the carbohydrate tested were added to test tubes containing 4.5 ml of sterile deionlzed water. The following compounds were tested: L- arabinitol, D-arabinose, L-arabinose, cellobiose, citrate, erythritol, ethanol, galacitol, D-galactose, D-glucitol, D-gluconate, D- glucosamlne, D-glucuronate, glycerol, myo-inositol, inulin, DL- lactate, lactose, maltose, D-mannitol, melizitose, mellbiose, 2- methyl D-glucoside, raffinose, L-rhamnose, ribitol, D-ribose, salicin, soluble starch, L-sorbose, succinate, sucrose, trehalose, xylitol, D-xylose. All compounds were analytical grade reagents (Sigma, St. Louis, MO). A positive control (with glucose) and a negative control (with no carbon source) were inoculated for each set of yeast isolates tested. 41 Assimilation of nitrogen cQmpniinH.*^ The medium for testing the ability of yeasts to utilize various nitrogen sources was prepared in a 10-fold concentration by dissolving 11.7 g of Bacto Yeast Carbon Base (Difco) together with the requisite amount of the nitrogen source in 100 ml deionlzed water. For nitrate assimilation testing, 0.78 g of potassium nitrate was used; whereas, for nitrite assimilation, 0.26 g of sodium nitrite was used. In the case of amino nitrogen, 0.64 g ethylamine hydrochloride, 0.68 g of cadaverine dehydrochloride, and 0.56 g of L-lysine were used (Wickerham 1946, 1951). The final medium was prepared by aseptically transferring 0.5 ml aliquots of the solution sterilized by filtration (Whatman filters, 2|im) Into 4.5 ml of sterile deionlzed water. Growth at different temperatures. Yeast inocula were transferred to 2% malt extract agar (2% w/v) malt extract, 1.5% (w/v) agar. Duplicate plates were incubated at 25°, 30°, 37°, and 420C for 10 days. Growth on vitamin-free media. The 10-fold concentrated medium was prepared by dissolving 16.7 g Bacto vitamin-free yeast base (Difco) in 100 ml deionlzed sterile water. One part (0.5 ml) filter-stenlized solution containing vitamin-free yeast base (Difco) was added to 9 parts (4.5 ml) sterile deionlzed water in test tubes. Yeast suspensions (0.1 ml) were added to test tubes, incubated at 280C, and examined for growth at 7, 14, and 21 days (van der Walt 42 and Yarrow 1984). Inocula were taken from tubes with growth and placed into a second Bacto vitamin-free yeast base broth (Difco). Growth was recorded as positive only If yeast growth was observed in the second tube (van der Walt and Yarrow 1984). Production of urease. Slants of Christensen's urea agar (Difco) were inoculated and Incubated at 2B°C. Cultures were observed up to five days. Development of a pink color (due to phenol red indicator) after 24-hour incubation at 28oC was recorded as positive for urea hydrolysis. Diazonium Blue B color test. Hagler and Ahearn (1981) modified the DBB test previously described by van der Walt and Hopsu-Havu (1976). Results obtained with the modified test reconfirmed that the color reaction was restricted to basidiomycete yeasts. Ten day- old cultures grown on malt-yeast-glucose-peptone agar at room temperature were held at 55^0 for two hours in a hot-air oven and subsequently flooded with cold DBB reagent. If cultures became dark red within 2 min. at room temperature, the results were recorded as positive (Hagler and Ahearn 1981). Cycloheximide. Yeasts vary in their sensitivity to the antibiotic actldione (cycloheximide) (Whiffen 1948). This test was conducted in liquid yeast nitrogen base (Difco) with D-glucose (Sigma), and filter-stenlized cycloheximide (Sigma) was added to give concentrations of 0.1% or 0.01% (v/v). Growth on 50% glucose. Actively growing cultures were streaked on 50% (w/w) glucose agar tubes. Tubes were incubated at 28oC and examined for growth after five days. 43 Electron Microscopy

Yeast cells obtained from the Dalmau plates and fourth Instar larvae were prefixed separately at room temperature in 2% glutaraldehyde, 0.1 M sodium cacodylate buffer, (pH 7.0). Lan/ae were fixed for two hours, whereas yeast cells were fixed for one hour. Samples were subsequently washed in 0.1 M cacodylate buffer (pH 7.0) and post-fixed In 2% with osmium tetroxide for one hour. Yeasts and larval specimens were washed separately in 1 M cacodylate buffer (pH 7.0) and dehydrated through a graded ethanol series. After critical point drying, larval specimens were sagitally sectioned for internal observation of yeast cells. Subsequently, yeast and larval specimens were sputter-coated with argon (Ploron SEM) and viewed with a scanning electron microscope (Hitachi S- 500).

Light Microscopv Light microscopy was used to observe yeasts in the gut of adult workers. Ants were dehydrated by Immersion in 95% ethyl alcohol for 3 min. to pelletize the gut content. Ants were dissected with sterilized forceps and the guts harvested and spread on microscope slides, stained with methylene blue, and covered with sterile coversllps. Subsequently, the gut was observed for the presence of yeast cells. 44 Statistical Analyses

Comparisons were made between mean weights of colonies with yeasts (51 colonies) and without yeasts (149 colonies) to determine if they were significantly different (E < 0.05). In addition, comparisons of mean weights of colonies with and without yeasts were made among locations and seasons (E < 0.05). Data were analyzed using (ANOVAs) the general linear model procedure of the Statistical Analysis System (SAS Institute 1985, Cary, NC). Mean separation analysis was performed using Duncan's (1951) multiple range test.

Results Yeasts Isolated from Adult Haemolymph Yeast isolates were obtained from the haemolymph of 327 adult workers. This number represents 3.27% of all workers dissected (10,000). Table 2.1 shows the weight of each colony, number of ants harboring yeasts, and the yeast species occuring In that colony. Determination of pure or mixed cultures based on morphogical examination of well-isolated colonies on malt extract agar showed that each ant harbored a single yeast strain. In addition, yeast Isolates from workers of the same colony represented the same yeast strain based on morphological and physiological characteristics. According to Barnett et al. (1983), just over half of all yeast species identified are D-glucose fermenters. Of the yeast Isolates obtained from the study herein (327), approximately 60% of them 45 (195) fermented D-glucose. All isolates in this Investigation reacted negatively with DBB, which stains cell walls of basidlomycetous yeasts but not ascomycetous yeasts. The majority of the yeast isolates obtained In this study did not form ascospores. Therefore, most of the yeast flora of the red Imported fire ant belongs to the yeast Imperfecti (Deuteromycotina, Blastomycetes, Cryptococcacea). However, one yeast Isolated from Huntsville and Waco was identified as an ascomycetous yeast, Debaryomyces hansenii (Ascomycotina, Endomycetales, Saccharomycetaceae).

Glucose Fermenters Imperfect yeasts. The majority of yeasts isolated from haemolymph of the red imported fire ant were Identified as Q_. parapsilosis (Figure 2.9 A and Table 2.2). After three weeks' incubation at 28^0 on malt extract agar, cultures of this isolate were cream-colored, glistening, soft, smooth margined, and flat to umbonate. Colony diameter was 2 mm after three days and 5-8 mm after seven days. Sediment and ring were observed after three weeks on malt extract agar. After two days at 28oC in yeast morphology broth, cells were globose to oval (2.0-3.5) x (3.0-4.5 |im). Also, long pseudomycelial cells were observed on Dalmau plate cultures. The profile number recorded from the API 20C System was C-6656171 which was a perfect match for Q_. parapsilosis in the API Analytical Profile Index. Furthermore, these isolates were Identified as £.. parapsilosis using the key of Meyer et al. (1984). This species fermented glucose strongly but did not ferment other 46 carbohydrates tested. Few isolates of this species obtained from the red imported fire ant assimilated D-ribose. All strains of £L parapsilosis did not assimilate cellobiose, which demonstrates that this yeast is not common in the soil because neariy all regular soil yeasts can assimilate celloblose produced from microbial degradation of cellulose (Phaff and Starmer 1987). Of the 327 yeast isolates, 168 were C parapsilosis. which represents 51.4% of the total yeast isolates obtained from the haemolymph of the red imported fire ant (Table 2.3). A second species of Candida, guillermondii. was isolated from 5.5% of red Imported fire ant (Table 2.2). After three weeks' Incubation at 28oC on malt extract agar, cultures of this yeast were yellowish to cream colored, soft, smooth and shiny, and developed a disagreeable odor. Colony diameter measured 2 mm after three days and 3-6 mm after seven days. A sediment formed and a ring was observed after three weeks on malt extract agar. After two days at 280C in yeast morphology broth, cells were long and oval (2-4) x (2.5-6) ^m. In addition, pseudomycelium development on Dalmau plate cultures on corn meal agar was abundant. The profile number recorded from the API 20C System was C-6676373. These yeast isolates were also identified as C.. guillermondii using the key of Meyer et al. (1984). This species fermented glucose, D-galactose, raffinose, and sucrose. None of the £.. guillermondii Isolates differed in their pattern of assimilation. Based on the aforementioned characteristics, they were considered to be the same strain. However, cellobiose was successfully assimilated by 47 this yeast. Therefore, C auillermondii may be common In soil. Among the 327 yeast isolates were 18 C. guillermondii. which represents 5.5% of total yeasts isolated from S.. invicta haemolymph (Table 2.3).

Perfect §t9te. The only ascosporogenous yeast Isolated from ant haemolymph was DebarvomyceR hansenii (Zopf) Lodder et Kreger van-Rlj (Table 2.2). After three weeks' Incubation at 28oC on malt extract agar, cultures were grayish white, soft and shiny, and slightly wrinkled. Colony diameter measured 2 mm after three days and 3-6 mm after seven days. A sediment was formed at three weeks on malt extract agar. After two days at 28oC in yeast morphology broth, cells were globose (2-4) x (3-6) |im. However, pseudohyphae did not develop on Dalmau plate cultures on corn meal agar. The profile number recorded from the API 20C System was C- 6776273. Although this number was not cited in the API Analytical Profile Index, It was acceptable for Q_. hansenii after consulting the API data base. Furthermore, these isolates were confirmed as fi. hansenii by using the key of Kreger van Rij (1984). This species fermented glucose but did not ferment other carbohydrates. Few isolates of this species were recovered from the red imported fire ant haemolymph. This strain of fi. hansenii assimilated cellobiose, suggesting that it is also a common soil yeast. Among the 327 yeast isolates, nine were EL. hansenii. which represents 2.75% of those isolated from S.. invicta haemolymph (Table 2.3). 48 Glucose non-fermenters Imperfect state. The teleomorph of Candida lipolytica (Harrison) Diddens et Lodder is named Yarrowla lipolytica (Wickertiam et al.) van der Walt and von Arx. This species was the second most prevalent yeast isolated from red imported fire ant adult haemolymph (Table 2.2). After three weeks' incubation at 28^0 on malt extract agar, cultures were cream-colored, moist, soft, delicately wrinkled, and distinctly elevated approximately 8 mm. Colony diameter measured 3 mm after three days and 5-10 mm after seven days. A sediment was formed and a thin pellicle was observed at three weeks on malt extract agar. After two days at 28oC In yeast morphology broth, cells were long and oval. Moreover, short, oval cells were abundant. Cells measured (2-4) x (3-20) fxm. Also pseudomycelial cells were well developed on Dalmau plate cultures on corn meal agar. In addition, long branched filaments with septa were present. The profile number recorded from the API 20C System was C-6000200 which was a good match for £.. lipolytica in the API Analytical Profile Index. Furthermore, these Isolates were identified as £.. lipolytica (Figure 2.9 B) using the key of Myer et al. (1984). This species did not ferment glucose. Two strains, based on their ability to assimilate sorbose, were detected. Neither grew at 370C; however, they did grow at 30oC. None of the isolates assimilated cellobiose, which again demonstrates that this yeast is not a common resident in soil. Among the 327 yeast isolates, 123 were Q.. lipolytica. which represents 37.6% of all yeast isolates from S.. invicta haemolymph (Table 2.3). 49 Candida rugosa (Anderson) Diddens et Lodder has not been found. This species was not commonly Isolated from the red imported fire ant haemolymph (Table 2.2). After three weeks' incubation at 28^0 on malt extract agar, cultures were cream-to-grey, moist and soft, and partly wrinkled. Colony diameter measured 2.5 mm after three days and 4-6 mm after seven days. A sediment was formed and a ring was observed at three weeks. After two days at 28oC In yeast morphology broth, cells were long and oval (2-3) x (3.5-10) |im. Also, pseudomycelial cells were well developed on Dalmau plate cultures on corn meal agar. The profile number recorded from the API 20C System was C-2452062, which was a very good match for Q_. rugosa In the API Analytical Profile Index. Furthermore, these isolates were identified as Q_. rugosa using the key of Myer et al. (1984) based on physiological characteristics. Because no differences in their assimilation patterns were observed, this species represented a single strain. This species does not ferment glucose. Isolates of Q_. rugosa do not assimilate cellobiose, which excludes this yeast as a regular soil resident. Among the 327 yeast Isolates, nine were £.. rugosa. which represents 2.75% of the total isolated from S.. invicta haemolymph (Table 2.3).

Yeast Prevalence Physiological and morphological tests showed that the 327 yeast isolates from S.. invicta haemolymph represented five yeast species (Table 2.3): Candida parapsilosis (168 Isolates), Candida lipolytica (123 isolates), Candida guillermondii (18 isolates), Candida rugosa 50 (9 isolates), and Debaryomynp.*^ h^ngonii war h^r^^^mj (9 Isolates). The species Q_. par^psilQsi.s and C lipolytica represented almost 90% of all yeast isolates detected (Table 2.3). The dominant yeast species, £.. parapsilc)>«?l,s. was isolated from S.. Invicta at all sites. In addition, this species was isolated during all four seasons, but not from all sites during all seasons. The second most prevalent yeast, C lipolytica. was recovered during all seasons and from four of the five sites (not Abilene).

Statistical Analyses

The colonies collected from Beaumont, regardless of whether yeasts were present or not, displayed a significantly greater weight (E < 0.05) in comparison to the other four sites (Figure 2.2). ANOVA tables are presented In Appendix A. Comparison of mean weights of the 51 colonies which were harboring yeasts and the 149 colonies which were not (Figure 2.3) showed that biomass of colonies with yeasts was significantly heavier (E < 0.005) than colonies without yeasts. The mean weight of colonies with and without yeasts was 35.8 g and 23.6 g, respectively (df = 5, 194). More specifically, after mean separation analysis was performed using Duncan's (1951) multiple range test, a significantly greater weight (E < 0.005) was observed only in colonies collected from the Abilene site. The mean weight of colonies from Abilene with and without yeast (df = 1, 38) was 34.8 g and 20.9 g, respectively (Figure 2.4). However, no significant differences were detected in weights of colonies with and without yeast from the other four sites. 51 Colonies with yeasts were significantly greater (E < 0.0001) in weight (Figure 2.5) than colonies without yeasts collected during winter and during spring. Mean weight of winter-collected colonies with and without yeasts was 42.2 g and 23.3 g, respectively (df = 5, 44). Mean weight of spring-collected colonies with and without yeasts was 48.9 g and 28.6 g, respectively (df = 5, 44). However, no significant differences were detected among colonies with and without yeasts collected during the summer and the fall. ANOVA tables are presented in Appendix A. Figures 2.6 and 2.7 show the number of yeast isolates from the red imported fire ant for each site and each season, respectively. Yeast Isolates were detected in 3.3% of the 10,000 adult workers and 25.5% of the 200 colonies (Table 2.4). In general, the biomass of colonies harboring yeasts appears to be of greater weight than colonies without yeasts among both locations and seasons.

Urvag A total of 600 fourth instar larvae were screened for yeasts This investigation showed that 80.3% of the fourth Instar larvae were harboring yeasts (Table 2.4). Yeast prevalence In the 12 colonies screened was 100%. In Abilene, 123 of 150 lan/ae harbored yeasts (Figure 2.8). In addition, yeasts were present in 131 of 150 larvae from colonies collected In Waco. In Huntsville, yeasts were present in 117 of 150 lan/ae, and 111 larvae of 150 contained yeasts from colonies collected in Beaumont. This investigation showed (Table 2.5 ) that among the 482 yeasts detected from fourth 52 mstar larvae were Candida parapsilosis (319 Isolates), Candida ''P^^lyt'gg (76 isolates), Oandida auillermondii (44 Isolates) and Candida rggpga (43 isolates). Although D.. hansenii was isolated from the haemolymph of adult workers, this species was not isolated from the lan/al gut. In addition, the mean colony forming units of yeast cells In a fourth instar lan/ae was 8.8 x 102 (Table 2.6). Moreover, C. parapsilosis. representing 51.4% of the yeast flora of adult workers and 66.2% of the yeast flora of fourth Instar larvae, is the dominant species associated with the red imported fire ant Third instar lan/ae, half the size of fourth instar larvae, were not prevalent in colonies collected during the fall of 1993. Individual cultures of gut contents on AYMA plates from 50 third instar larvae failed to yield yeasts.

Microscopy Because of the high prevalence of yeasts in larvae, scanning electron microscopy was used to observe yeasts In the gut of the fourth instar larva (Figures 2.10 A-D and 2.11 A-B). Scanning electron microscopy showed that yeast cells were not only present, but were also abundant in the gut of the fourth instar lan/ae (Figures 2.10 and 2.11). In addition, electron micrographs showed some cytological differences in the location of yeast cells in the larval gut. Yeast cells could be obsen/ed mixed with the gut content of the larvae. Cell division was taking place at this location (Figure 2.11). Yeasts were also lodged close to muscle fibers, and some yeast cells were attached to large membrane-bound structures (possibly fat 53 body; Figure 2.10 B, C, and D). In addition, light microscopy showed that yeast cells were present in the gut of adult ants (workers and queens). Furthermore, fecal material was also obsen/ed to contain yeast cells.

Discussion Obviously, yeasts are part of the Internal flora of S.. invicta from fourth instar larva to the adult. The gut of viable fourth Instar larvae was packed with yeast cells. This association suggests that a highly specific process of selection is taking place between the yeasts and the fourth instar larva which appears to be the first stage of yeast colonization. These data indicate that the presence of yeast does not have a negative impact on colony weight. Because, colony weight is an indication of vigor, yeast association with the red Imported fire ant does not appear to be pathogenic. In addition, a high prevalence of yeasts in lan/ae suggests a nutritional role. However, the origin of the specification is not available. Therefore, this yeast and S.- invicta may be locked in a mutualistic association. Consequently, the association of yeasts with the larval stage may yield novel approaches in red imported fire ant control if behavioral parameters of the ants are considered In the development and application of microbial control agents. The presence of an infrabuccal filter capable of preventing the Ingestion of microparticles as small as 0.9 |im was reported by Glancey et al. (1981). This report has probably slowed investigations on the administration of microbial organisms per os 54 to control the red Imported fire ant. However, conidia of Beauveria l?a§Siana measuring from 1|j,m to 2^im in diameter were found In the gut of S.. invicta workers (Siebeneicher et al. 1992). Unlike solid microparticles and rigid cell walls of fungi, yeasts are more malleable in shape, and the cells of budding yeasts are approximately l^im. Thus, the uptake of yeast cells by adult worker is a strong possibility. On the other hand, yeasts may be acquired during the larval stage and maintained, thus becoming permanent members of the microbial flora of the red imported fire ant. This investigation is the first quantitative study of the yeast flora of the red imported fire ant. Yeasts that have been Isolated In this study have previously been isolated from other Insects. However, the significance of the interaction has seldom been addressed. The study of Shifrine and Phaff (1956) mentioned £.. parapsilosis and £.. rugosa as part of the yeast flora associated with the bark beetles, i^s and Dendroctonus. but did not elucidate the nature of the Interaction between these organisms. In addition, the occurrence of £.. guillermondii and Q_. hansenii in the gut of the adult green June beetle, Cotinis nitida (L.) (Coleoptera: Scarabaeidae), was reported by Vishlnlac and Johnson (1990). In a series of investigations on the yeast flora of the green June beetle, only one yeast, Trichosporon cutaneum. was studied further. This yeast species was shown to elicit aggregation behavior in £.. nitida (Johnson and Vishniac 1991). The association of C. lipolytica with Insects was first reported for bees (Gilliam et al. 1974). Gilliam et al. (1974) found that the yeast flora of healthy bees included £. 55 parapsilosis. Q,. guillermondii. £.. rugosa. and D.- hansennii. Therefore, all yeasts isolated in the current study from the red Imported fire ant were previously isolated from other social insects. Therefore, the yeast may be collected with the pollen and carried by the bees in their pollen basket. The question which arises is, does S.. invicta offer the yeasts the same incentives? The five yeast species associated with the red imported fire ant present some similarities in their assimilation patterns. While all yeasts assimilated glycerol and the amino nitrogens ethylamine, L- lysine, and cadaverine, all isolates failed to assimilate nitrate. Phaff et al. (1952) Isolated some of the same yeast species from marine environments, and found them to be tolerant to high pH. Yeasts exposed to a high pH regime maintain a lower intracellular pH by accumulating glycerol (Gustafsson and Norkrans 1976; Brown 1978; Reed et al. 1987; Andre et al. 1988; Jovall et al. 1990). Yeasts subjected to osmotic stress respond by increasing their Internal osmolarity by accumulating low molecular weight organic solutes (LeRudulier et al. 1984; HIggins et al. 1987; Jovall et al. 1990). Therefore, the yeast flora of the red imported fire ant may be able to withstand ant alkaloids sprayed during gaster flagging. Under high pH regimes, these yeasts produce glycerol (polyol) which may sen/e to balance the osmotic pressure resulting from venom alkaloids. Consequently, yeasts associated with £. jnvjcta may be able to exclude alkaloids and achieve their osmoregulation by accumulating selective solutes such as glycerol which is compatible with cell enzymes. Using in yivfi and in vjlm ^^C NMR spectroscopy, 56 Jovall et al. (1990) confirmed that glycerol production was induced in Q,. hansenii by high pH. In addition, a study was conducted by Fogleman and Kircher (1983) on the effects of senita cactus alkaloids on yeasts associated with Drosophila. Their investigation showed that yeast growth and sterol biosynthesis was not inhibited by alkaloids. On the other hand, the inhibition of the red imported fire ant venom alkaloids on the in vitro germination and development of certain entomogenous fungi was reported by Storey et al. (1990). However, venom alkaloid concentrations in mounds were not quantified for field condition simulation In the laboratory. In addition, unicellular organisms like yeasts may have a higher tolerance for alkaloids than filamentous fungi by intracellular adjustments to osmotic shock. The presence of yeasts from the larval to the adult stage of S.. invicta supports the Idea that these yeasts are gut symbionts. The hypothesis that these yeasts may be pathogenic to the ant should be discounted because they did not cause any observable debilitating effects. Numerous yeast cells occupy the larval gut. However, a lesser yeast density Is exhibited in adult workers. Their differences suggests that yeast contribution to red imported fire ant nutrition is more critical during Immature stages than during the adult stage. Among the non-fermenting yeasts, £.. lipolytica llpolyzes fats (Barnett et al. 1983), and C. rugosa. hydrolyze lipids via lipase (Veeragavan and Gibbs 1989). Consequently they may be involved in the degradation of proteins, in general, and residual yolk in the larval gut in particular. 57 The dominant yeast species associated with S.. invicta have different vitamin requirements. The most prevalent yeasts, £.. parapsilosis and Q_. lipolytica. require biotin and thiamin, respectively, for growth. The case could be made that these two yeasts complement each other in helping to meet the vitamin requirement of S.. invicta. The carbohydrate most abundant in insect haemolymph, trehalose, is assimilated by £. parapsilosis but Is not used by £.. lipolytica. The provision of vitamins, carbohydrates, amino acids, and lipids by yeasts to insect hosts is an area of interest in symbiotic research and merits investigation. Moreover, preoccupation with a simple working hypothesis may be counterproductive (Graham 1967). Weight differences in colonies harboring yeasts and colonies without yeasts were observed in the Abilene site only. This location represents the most recent site of red imported fire ant infestation along a moisture and an invasion gradient. Therefore, yeasts may be most important during the establishment phase. During winter and spring months, biomass was greater in colonies harboring yeasts compared to colonies without yeasts. Their differences may be explained by the fact that during cool seasons, colonies are less active and mounds tend to become more stable. However, in warmer seasons, colonies are more dynamic. In part, as a result of reproductive processes. Therefore, several small mounds may bud from larger mounds (Vargo and Porter 1989), and still exhibit the same yeast flora as the mother mound. Therefore, the variable, 58 colony weight, may not have as significant an impact in warmer seasons as it has in cooler seasons. On the other hand, some colonies weighing more than 100 g did not exhibit yeasts and some colonies weighing just 2 g did. The difference in yeast prevalence between the haemolymph of adult workers and the gut of worker larvae may be explained by the rupture of the peritrophic membrane during meconium expulsion. The peritrophic membrane passes from the gut largely intact as a wrapping which surrounds the faecal pellet, but sometimes the membrane Is broken in the hindgut by muscular contractions or spinous projections (Bignell 1985). Therefore, disruption of the peritrophic membrane may be of importance in allowing ingested organisms access to colonization sites, and also permiting some organisms to reach the mesenteric epithelium by fon/vard migration along the ectoperitrophic space (Le Berre 1967). 59 References Cited

Andre, L., A. Nilsson, and L. Adier. 1988. The role of glycerol of osmotlerance of the yeast Debaryomyces hansenii. J. Gen. Microbiol. 134:669-677

Arx, J. A. von, and M. A. A. Schipper. 1978. The CBS fungus collection. In: Fungus collection, in: Advances in applied microbiology, vol. 24. D. Periman, Ed. Academic Press, New York. pp. 215-236.

Banks W. A., C. S. Lofgren, D. P. Jouvenaz, C. E. Stringer, P. M. Bishop, D. F. Williams, D. P. Wojcik, and B. M. Glancey. 1981. Techniques for collecting, rearing and handling imported fire ants. U.S. Dep. Agr. Sci. Ed. Admin., AAT-S-21:1-9.

Bamett, J. A., R. W. Payne, and D. Yarrow. 1983. Yeasts: characteristics and Identification. Cambridge Univ. Press, Cambridge. 811 p.

Bignell, D. E. 1984. The arthropod gut as an environment for microorganisms, in; Anderson, J.M., A.D.M. Rayner, and D.W.H. Walton [eds]. Invertebrate-microbial interaction, Brithish Mycological Society and the British ecological Society. 205- 227.

Brown, A. D. 1978. Compatible solutes and extreme water stress in eucaryotic microrganisms. Adv. Microb. Physiol. 17:181-242.

Conner, N. R., C. L. GIrdner, and L. Watson. 1976. Soil sun/ey of Taylor county, Texas. United states Soil Conservation Service in cooperation with Texas Agricultural Experiment Station. 66 pp.

Cricket Software. 1987. 40 Valley Stream Parkway, Malvern, P.A.

Crout, J. D., D. G. Symmank, and G. A. Peterson. 1965. Soil survey, Jefferson county, Texas. U.S. Dept. of Agriculture, Soil Conservation Service, Washington. 73 pp. 60 Dalton, H. K., R. G. Board, and R. R. Davenport. 1984. The yeasts of British fresh sausage and minced beef. Antonie van Leeuwenhoek. 50:227-248.

Duncan, D. B. 1951. A significance test for differences between ranked treatments in an analysis of variance. Va. J. Sci. 15:697-706.

Fogleman, J. C, and H. W. Kircher. 1983. Senita alkaloids: no inhibition of sterol biosynthesis in yeasts or cacti. J. Nat. Prod. 46:279-282.

Gilliam, M.,L. J. Wickerham, H. L. Morton, and R. D. Martin. 1974. Yeasts isolated from honey bees. Apis mellifera fed 2, 4- D and antibiotics. J. Invertrbr. Pathol. 24: 349-346.

Glancey, B. M., Vander Meer, R. K., Glover, A., Lofgren, C. S., and Vinson, S. B. 1981. Filtration of microparticles from liquids ingested by the red imported fire ant Solenopsis invicta Buren. Insectes Soc. 28:395-401.

Graham, K. 1967. Fungal-insect mutualism in trees and timber. Ann. Rev. Entomol. 12:105-126.

Gustafsson, L., and B. Norkrans. 1976. On the mechanism of salt- tolerance. Production of glycerol and heat during growth of Debaryomvces hansenii. Arch. Microbiol. 110:118-183.

Hagler, A. N., and Ahearn, D. G. 1981. A rapid DBB test to detect basidlomycetous affinity of yeasts. Int. J. Syst. Bacterid. 31, 204-208.

HIggins, C. F., J. Cairney, D. A. Stiriing, L. Sutheriand, and I. R. Booth. 1987. Osmotic regulation of gene expression: ionic strength as an intracellular signal. Trends Biochem. Sci. 12:339-344.

Johnson, D. T., and H. S. Vishniac. 1991. The role of Trichosporon cutaneum in eliciting aggregation behavior in Cotinis nitlda (L.) (Coleoptera: Scarabeidae). Environ. Entomol. 20(1): 15-21. 61 Joval, P.,I. Tunblad-Johanson, and L. AdIer. 1990. 3C NMR analysis of production and accumulation of osmoregulatory metabolites in the salt-tolerant yeast Debaryomyces hansenii. Arch. Microbiol. 154:209-214.

Kreger-van Rij, N. J. W. 1984. Debarvomyces Lodder et Kreger-van Rij nom. cons, in: The yeasts, a taxonomic study. N. J. W. Kreger-van Rij, Ed. Elsevier Science Publishers, New York, pp.130-145.

Le Berre, R. 1967. Les membranes peritrophique chez les arthropodes. Leur role dans la digestion et leur intervention dans revolution d'organismes parasitaire. Cahler ORTSM (Series D'entomologies Medicine et Parasitologies), 5, 147- 204.

LeRudulier, D., A. R. Strom, A. M. Dandekar, L.T. Smith, and R. O. Valentine. 1984. Molecular biology of osmoregulation. Science. 224:1064-1068.

McClintock, W. R. Jr., J. J. Castille, and M. Stewart. 1979. Soil of Walker County, Texas. Dept. of Agriculture, Soil conservation Service in cooperation with Texas Agricultural Experiment Station. 124 p.

Meyer, S. A., and H. J. Phaff. 1972. DNA base composition and DNA- DNA homology studies as tools in yeast systematlcs. in: Yeast models In science and technics. A. Kockova-Kratochvilova and E. MIranik, Eds. Proc. Spec. Symp. Yeasts, Smolenice. Publishing House of the Slovak Academy of Sciences, Bratislava, pp. 375-387.

Meyer, S. A., D. G. Aheam, and D. Yan-ow. 1984. Candida Beri

Phaff, H. J., and W. T. Starmer. 1987. Yeasts associated with plants. Insects and soil, in: The yeasts. Biology of Yeasts. A. H. Rose and J. S. Harrison, Eds. Academic Press. New York. pp. 123- 179. 62

Phaff, H. J., E. M. Mrak, and O. B. Williams. 1952. Yeasts Isolated from shrimp. Mycologia. 44:431-451.

Phaff, H. J., M. W. Miller, and E. M. Mrak. 1978. The life of yeasts, 2nd ed. Han/ard University Press, Cambridge, Massachusetts. 341 pp.

Poinar, G. O., and G. M. Thomas. 1978. Diagnostic manual for the identification of insect pathogens. Plenum Press, New Yori<. 218 pp.

Reed, R. H., J. A. Chudek, R. Foster, and G. M. Gadd. 1987. Osmotic significance of glycerol accumulation in exponentially growing yeasts. Appl. Environ. Microbiol. 53:2119-2123.

SAS Institute. 1985. SAS User's guide: basics, 5th ed. SAS Institute, Cary, North Carolina.

Shifrine, M., and H. J. Phaff. 1956. The association of yeasts with certain bark beetles. Mycologia. 48:41-55.

Siebeneicher, S. R., S. B. Vinson, and C. M. Keneriey. 1992. Infection of the red imported fire ant by Beauveria basslana through various routes of exposure. J. Invertebr. Pathol. 59:280-285.

Starmer, W. T., W. B. Heed, M. Miranda, M.W. Miller, and H.J. Phaff. 1976. The ecology of yeast flora associated with cactiphllic Drosophila and their host plants In the Sonoran Desert. Microbial Ecology. 3: 11-30.

Storey, G. K., R. K. Vander Meer, D. G. Boucias, and C. W. McCoy. 1991. Effect of fire ant (Solenopsis Invicta) venom alkaloids on the In vitro germination and development of selected entomogenous fungi. J. Invertebr. Pathol. 58:88-95.

Templln, E. H., A. L. Nabors, T. R. Atkins, A. W. Grain, I. C. Mowery, D. T. Horton, J. S. Williams, and R. M. Voigtel. 1958. Soil survey Mclennan County, Texas. United States Soil Conservation Service In cooperation with Texas Agricultural Experiment Station. 124 p. 63

Van Der Walt, J. P., and D. Yarrow. 1984. Methods for the isolation, maintenance, classification, and identification of yeasts pp. 45-104. in: N. J. W. Kreger-Van Rij [ed]. The yeasts, a Taxonomic Study, 3rd ed. Elsevier, Amsterdam.

Van der Walt, J. P., and V. K. Hopsu-havu. (1976). A colour reaction for the differentiation of ascomycetous and hemibasidlomycetous yeasts. Antonie van Leuwenhoek. 39:635-647.

Vargo, E. L., and S. D. Porter. 1989. Colony reproduction by budding in the polygyne Form of Solenopsis invicta (Hymenoptera: Formicidae). Ann. Entomol. Soc. Am. 3:307-313.

Veeraragavan, K. and B. F. Gibbs. 1989. Detection and partial purification of two lipases. Biotechnology Letters. 5:345-348.

Vishniac, H. S., and D. T. Johnson. 1990. The development of yeast flora in the adult green June beetle, Cotinis nitida (L.) (Coleoptera: Scarabeidae). Mycologia. 82:470-478.

Wagner, B. J., J. R. Thomas, E. R. Harris, E. Deleon, C. G. Ford, and J. D. Kelly. 1973. Soil survey of Erath County, Texas. United States Soil Conservation Service in cooperation with Texas Agricultural Experiment Station. 82 pp.

Whiffen, A. J. 1948. The production, assay and antibiotic activity of Actldione, an antibiotic from Streptomyces griseus. J. Bacterid. 56:283-291.

Wickeriiam, L. J. 1946. A critical evaluation of the nitrogen assimilation tests commonly used in the classification of yeasts. J. Bacterid. 52:293-301.

Wickerham, L. J. 1951. Taxonomy of yeasts. U.S. Department of Agriculture Tech. Bull. 1029, 56p. 64

Table 2.1. Colony weight d red imported fire ant and yeast species associated with the red imported fire ant obtained from sites located along a decreasing moisture gradient in Texas

Mound No. Colony No. Ants Yeast and Location wt(a) w/Yeast Species ABL#1-I 33.2 6 Candida parapsilosis ABL#2-I 10.1 0 none ABL#3-I 20.1 0 none ABL#4-I 41.9 0 none ABL#5-I 25.3 0 none ABL#6-I 22.2 0 none ABL#7-I 19.2 0 none ABL#8-I 7.0 0 none ABL#9-I 12.4 0 none ABL#10-I 40.0 0 none STV#1-I 9.7 0 none STV#2-I 0.4 0 none STV#3-I 6.5 0 none STV#4-I 6.4 0 none STV#5-I 12.3 0 none STV#6-I 11.0 0 none

ABL = Abilene; STV = Stephenville; WCO = Waco; HTV = Huntsville; BMT = Beaumont; I = Winter 1992; II = Spring 1993; III = Summer 1993; IV = Fall 1993. 65

Table 2.1 Continued

Mound No. Colony No. Ants Yeast and Location Wt(Q) w/Yeast Species STV#7-I 4.6 0 none STV#8-I 0.5 0 none STV#9-I 10.6 0 none STV#10-I 19.5 0 none WC0#1-I 4.1 0 none WC0#2-I 11.2 0 none WC0#3-I 44.3 0 none WC0#4-I 34.9 1 1 Candida lipolytica WCO#5-l 33.8 8 Candida lipolytica WC0#6-I 7.1 0 none WCO#7-l 13.5 0 none WC0#8-I 8.5 0 none WC0#9-I 11.8 0 none WCO#10-I 7.4 0 none HTV#1-I 12.7 0 none HTV#2-I 13.9 4 Debaryomyces hansenii HTV#3-I 15.3 0 none HTV#4-I 19.4 0 none 66

Table 2.1 Continued

Mound No. Colony No. Ants Yeast and Location Wt (g) w/Yeast Species HTV#5-I 13.3 0 none HTV#6-I 40.2 7 Candida lipolytica HTV#7-I 31.5 0 none HTV#8-I 10.6 5 Candida lipolytica HTV#9-I 23.5 6 Candida lipolytica HTV#10-I 34.8 0 none BMT#1-I 60.5 0 none BMT#2-I 54.1 0 none BMT#3-I 19.5 0 none BMT#4-I 111.8 0 none BMT#5-I 8.7 4 Candida lipolytica BMT#6-I 96.4 6 Candida parapsilosis BMT#7-I 136.2 0 none BMT#8-I 9.7 0 none BMT#9-I 98.2 9 Candida lipolytica BMT#10-I 32.3 0 none ABL#1-II 42.3 7 Candida parapsilosis ABL#2-II 15.7 0 none ABL#3-II 52.1 0 none 67

Table 2.1 continued

Mound No. Colony No. Ants Yeast and Location Wt(g) w/Yeast Species ABL#4-II 21.9 0 none ABL#5-II 32.7 0 none ABL#6-II 18.7 0 none ABL#7-II 9.4 0 none ABL#8-II 29.6 0 none ABL#9-II 23.7 0 none ABL#10-II 26.8 0 none STV#1-II 18.1 0 none STV#2-II 23.1 0 none STV#3-II 3.2 0 none STV#4-II 20.3 4 Candida lipolytica STV#5-II 9.4 0 none STV#6-II 21.6 0 none STV#7-II 6.1 5 Candida lipolytica STV#8-II 12.8 0 none STV#9-II 15.7 0 none STV#10-II 19.3 0 none WC0#1-II 19.4 0 none WCO#2-ll 12.4 0 none WC0#3-II 3.9 0 none 68

Table 2.1 Continued Mound No. Colony No. Ants Yeast and Location Wt (g) w/Yeast Species WC0#4-II 33.7 0 none WC0#5-II 48.2 6 Candida lipolytica WC0#6-II 16.9 0 none WC0#7-II 6.1 0 none WC0#8-II 44.9 0 none WCO#9-ll 22.8 0 none WCO#10-II 29.3 0 none HTV#1-II 32.1 0 none HTV#2-II 23.3 0 none HTV#3-II 36.8 0 none HTV#4-II 5.3 0 none HTV#5-II 12.1 0 none HTV#6-II 25.8 0 none HTV#7-II 29.3 0 none HTV#8-II 48.9 0 none HTV#9-II 46.3 4 Candida lipolytica HTV#10-II 44.6 8 Candida lipolytica 69

Table 2.1 Continued Mound No. Colony No. Ants Yeast and Location wt(g) w/Yeast Species BMT#1-II 22.2 0 none BMT#2-II 112.3 0 none BMT#3-II 79.6 0 none BMT#4-II 11.2 0 none BMT#5-II 98.4 7 Candida lipolytica BMT#6-II 84.9 5 Candida lipolytica BMT#7-II 36.7 0 none BMT#8-II 71.3 0 none BMT#9-II 69.8 0 none BMT#10-II 42.5 0 none ABL#1-MI 54.6 6 Candida parapsilosis ABL#2--III 7.3 0 none ABL#3-III 6.8 0 none ABL#4-III 13.1 7 Candida parapsilosis ABL#5-III 40.5 8 Candida parapsilosis ABL#6-III 3.95 0 none ABL#7-III 38.1 0 none ABL#8-III 58.3 9 Candida guillermondii ABL#9-III 5.3 0 none 70

Table 2.1 Continued

Mound No. Colony No. Ants Yeast and Location Wt (g) w/Yeast Species ABL#10-III 17.2 0 none STV#1-III 1.2 7 Candida parapsilosis STV#2-III 5.8 0 none STV#3-III 9.9 0 none STV#4-III 28.3 0 none STV#5-III 23.6 0 none STV#6-III 3.4 0 none STV#7-III 12.6 7 Candida parapsilosis STV#8-III 7.0 0 none STV#9-III 2.4 4 Candida parapsilosis STV#10-III 112.8 8 Candida parapsilosis WC0#1-III 24.0 5 Candida parapsilosis WC0#2-III 14.9 0 none WC0#3-III 31.2 0 none WC0#4-III 38.5 0 none WC0#5-III 36.0 8 Candida parapsilosis WC0#6-III 62.3 1 1 Candida parapsilosis 71

Table 2.1 Continued Mound No. Colony No. Ants Yeast ar>d Location Wt (g) w/Yeast Species WC0#7-lll 15.7 4 Candida parapsilosis WC0#8-III 24.4 0 none WC0#9-III 2.0 6 Candida parapsilosis WCO#10-I II 17.6 0 none HTV#1-III 44.2 0 none HTV#2-III 28.9 6 Candida parapsilosis HTV#3-III 9.8 0 none HTV#4-III 53.4 0 none HTV#5-III 4.7 0 none HTV#6-III 18.9 8 Candida lipolytica HTV#7-HI 4.7 0 none HTV#8-III 21.4 5 Candida parapsilosis HTV#9-III 4.4 0 none HTV#10-I II 4.1 0 none BMT#1-II 70.4 9 Candida guillermondii BMT#2-III 12.4 0 none BMT#3-II 4.4 6 Candida parapsilosis BMT#4-II 1 30.2 4 Candida parapsilosis 72

Table 2.1 Continued

Mound No. Colony No. Ants Yeast and Location Wt (g) w/Yeast Species BMT#5-III 30.3 0 none BMT#6-III 27.4 0 none BMT#7-III 27.8 0 none BMT#8-III 67.2 0 none BMT#9-III 17.6 0 none BMT#10-III 20.3 7 Candida parapsilosis ABL#1-IV 17.3 0 none ABL#2-IV 15.3 0 none ABL#3-IV 23.6 8 Candida parapsilosis ABL#4-IV 24.3 0 none ABL#5-IV 22.1 5 Candida parapsilosis ABL#6-IV 24.9 0 none ABL#7-IV 20.2 0 none ABL#8-IV 19.2 0 none ABL#9-IV 26.0 6 Candida parapsilosis ABL#10-IV 19.1 0 none STV#1-IV 62.0 7 Candida parapsilosis 73

Table 2.1 Continued Mound No. Colony No. Ants Yeast and Location wt(g) w/Yeast Species STV#2-IV 7.8 4 Candida parapsilosis STV#3-IV 9.1 0 none STV#4-IV 9.2 0 none STV#5-IV 13.0 0 none STV#6-IV 7.1 4 Candida lipolytica STV#7-IV 10.2 0 none STV#8-IV 16.4 0 none STV#9-IV 18.3 0 none STV#10-IV 7.3 0 none WCO#1-IV 24.2 0 none WC0#2-IV 45.0 0 none WC0#3-IV 17.3 7 Candida parapsilosis WC0#4-IV 16.8 5 Debaryomyces.hansennii WC0#5-IV 8.3 0 none WC0#6-IV 15.0 0 none WC0#7-IV 110.0 0 none WC0#8-IV 16.9 0 none 74 Table 2.1 Continued Mound No. Colony No. Ants Yeast and Location Wt (g) w/Yeast Species WC0#9-IV 70.0 8 Candida lipolvtica WCO#10-IV 6.3 0 none HTV#1-IV 4.6 0 none HTV#2-IV 27.6 6 Candida parapsilosis HTV#3-IV 6.8 0 none HTV#4-IV 9.0 5 Candida lipolytica. HTV#5-IV 18.3 5 Candida lipolytica. HTV#6-IV 7.1 0 none HTV#7-IV 11.2 0 none HTV#8-IV 11.6 0 none HTV#9-IV 72.0 0 none HTV#10-IV 7.0 0 none BMT#1-IV 13.4 0 none BMT#2-IV 47.6 0 none BMT#3-IV 39.3 0 none BMT#4-IV 6.8 0 none BMT#5-IV 110.0 9 Candida rugosa. BMT#6-IV 13.1 0 none 75 Tade 2.1 Continued

Mound No. Colony No. Ants Yeast and Location Wt(g) w/Yeast Species BMT#7-IV 8.1 0 none BMT#8-IV 61.9 0 none BMT#9-IV 9.3 0 none BMT#10-IV 16.7 0 none Table 2.2. Morphological and biochemical characteristics of the yeast flora of the red imported fire ant.

Yeast Species

Characteristic O^ Yl Cg Cr Dh Morphology Pellicle - + - - + Pseudo/ + + + + . true hyphae Chlamydospores - - - . Ascospores - - - - + Capsule - - - _ . Fermentation D-glucose + - + - + celloblose - - _ . . D-galactose - - + _ . maltose - - . . . raffinose - - + - . sucrose - . + . .

Cp = Candida parapsilosis. Yl = Yarrowla lipditica. Cg = Candida guillermondii. Cr = Candida rugosa. and Dh = Deban/omyces hansenii. + = positive response, - = negative response, w = weak response, D = delayed response, v = variable response obtained, meaning different strains of that species were isolated. 77 Table 2.2. Continued.

Yeast Species

Characteristic cp Yl Cg Cr Dh Assimilatinn!

L-arablnltol - - + +

D-arablnose - - + •

L-arablnose + - + + celloblose - - + + citrate + + + + erythritol - + - + ethanol + + + + + galactitol - - - +

D-galactose + - + + + D-glucitol + + + + gluconate + + <• ^ +

D-glucoseamine + - + + glucuronate - - - - glycerol + + + + +

/77yo-lnositol + - - +

Inulin - - + -

DL-lactate - ^ - - lactose - - - - maltose + - + + D-manltol + + + + 78 Table 2.2. Continued.

Yeast Species

Characteristic Cp Yl Cg Cr Dh melzitose + + +

mellbiose - + + methyl-D- + + + glucoside

raffinose - ^ ^ ^ -

L-rhamnose - _ - - - ribitol + + + D-ribose V + -

salicin - + +

soluble - — •• ~ W starch L-sorbose + D + + succinate + + + sucrose + + + trehalose + + + xylitol + D + D-xylose + + + DL-lactate V + D

• 79 Table 2.2. Continued.

Yeast Species

Characteristic ^ Yl Cg Cr Dh

Nitrate nitrite — " " • + ethylamine + + + + + L-lyslne + + + + + cadaverine + + + + + Growth

w/o vitamins •• - - - - at 250C + + + + + at 30OC + + + + + at 370c + - + + - at 420C V - - - - with 50% glucose + + + D D with 0.01% + + - - D cyclohexmide

with 0.1% - + - - - cyclohexmide

urea hydrolysis - + - ~ • Diazonium - - - - - Blue B 80

Table 2.3. Abundance of yeast species^ associated with red imported fire ant adult haemolymph (from Texas). .§Eecles Percent Candida parapsilosis 51.37 Candida lipolytica 37.61 Candida auillermondii 5 50 Candida rugosa 2.75 Debaryomyces hansenii 2.75

a Percent of 327 isolates

Table 2.4. Yeast frequency of occurrence in red imported fire ant workers, colonies, and fourth instar larvae obtained from colonies In Texas. Categories Percent Workers 3.27 Colonies 25.5 Lan/ae 80.3 81

Table 2.5. Yeast species frequencies of occurrence in fourth instar larvae of the red imported fire ant from colonies In Texas. ^ Species Percent^ Candida parapsilosis 66.20 Candida lipolytica 15.80 Candida guillermondii 910 Candida rugosa 8.90 a n = 600

Table 2.6. Mean of yeast colony forming units per red imported fire ant larva. Sites CFU/larva Abilene 9.43 x 102 Waco 8.03 X 102 Huntsville 8.50 x 102 Beaumont 9.00 x 102 82

160 BMT ' 1 • E o b.y 140-

o Sf 120- ^HTV co 100- 1

- A 80-- WCO CO 0) ABL A y ^STV 1 ' 1 ' 1 ' 1 60 1 ' 160 270 430 590 Kilonneters

Figure 2.1 Collection sites along a decreasing moisture gradient from Beaumont to Abilene, Texas (ABL = Abilene, STV Stephenville, WCO = Waco, HTV = Huntsville, BMT = Beaumont). 83

O

o o O c CO 0)

Figure 2.2. Comparison of mean weights of red imported fire ant colonies among locations in Texas (ABL = Abilene, STV Stephenville, WCO = Waco, HTV = Huntsville, BMT = Beaumont; ANOVA: GLM). Means with same letters are not significantly different (£ < 0.05) as detected by Duncan's multiple range test (n = 40 colonies/ site). 84

40

M WITH YEASTS

O) WITHOUT YEASTS 2 30-1

20- O o O c CO 10- 0)

W/WO YEASTS

Figure 2.3. Comparison of mean weights of all red imported fire ant colonies (n = 200) with and without yeasts. Means with different letters are significantly different (P. < 0.05) as indicated by Duncan's multiple range test (with yeasts = 51 colonies; without yeasts = 149). 85

80

u> n WITH YEASTS

O) WITHOUT YEASTS 0) 5

o o o

Figure 2.4. Comparison of mean weights of red imported fire ant colonies with and without yeasts among locations in Texas (ABL = Abilene, STV = Stephenville, WCO = Waco, HTV = Huntsville, BMT = Beaumont). Means within sites followed by an asterisk are significantly different (P. < 0.05) using Duncan's multiple range test. 86

U>

0)

o o o

CO 0)

WINTER SPRING SUMMER FALL SEASONS

Figure 2.5. Comparison of mean weights of red imported fire ant colonies with and without yeasts collected from sites n Texas during 1992 and 1993. Means within seasons followed by an asterisk are significantly different (P. < 0.05) as determined by Duncan's multiple range test. 87

0) i5 o (0

(0 (0 0) >-

E 3

ABL STV WCO HTV BMT Sites

Figure 2.6. Number of yeast isolates from red imported fire ant workers obtained from sites in Texas. (ABL = Abilene; STV = Stephenville; WCO = Waco; HTV = Huntsville; BMT Beaumont; n = 2,000 for each site). 88

200

(0 i5 o (0

(0 CO 0) 100-

0) E 3

WINTER SPRING SUMMER FALL SEASONS

Figure 2.7. Number of yeast isolates from red imported fire ant workers obtained during 1992 and 1993 from sites in Texas. 89

200 (0 •^ (0 (0 0) >

0 (0 > 100- (0

0) E

ABL WCO HTV BMT SITES

Figure 2.8. Number of yeast isolates from fourth instar red imported fire ant larvae obtained from sites along a moisture gradient in Texas (ABL = Abilene, WCO = Waco, HTV = Huntsville, BMT = Beaumont; n =150 for each site). 90 91 92 93 94 95 CHAPTER III NUTRIENT AND YEAST FLORAL CHANGES IN SOIL INDUCED BY THE RED IMPORTED FIRE ANT

Introduction Yeast flora of the terrestrial environment was thoroughly reviewed by Carmo-Sousa (1969). Species were isolated from soils of differing texture, chemical composition, humidity, and pH (Carmo-Sousa 1969), and in general, were found to occur in lower numbers than bacteria and molds (Phaff and Starmer 1987). Therefore, soil simply serves as reservoir for yeasts from plants and animals and not specifically as a unique habitat for their growth and reproduction (Phaff et al. 1978; Carmo-Sousa 1969). Most of the early studies regarding yeast isolation from soil employed enrichment techniques in liquid media, and efforts were devoted to determine the species present in a particular type of ecosystem (Carmo-Sousa 1969). Therefore, these qualitative investigations were not designed to estimate colony forming units per gram of soil nor relative frequencies of yeast species (Carmo- Sousa 1969). Quantitative estimates of yeast flora in soil were reported by Di Menna (1960, 1966). Both quantitative and qualitative studies of yeasts from soils of New Zealand showed that yeast species varied from location to location with soil type and vegetation. However variations resulting from seasonal changes

96 97 were not observed (Di Menna 1957). Di Menna (1957) observed that Candida Curv^t9 was the dominant species in forest soils, whereas the genus Crvotococcus dominated grassland soils. Yeasts isolated from soil have a wide range of metabolic abilities, but are not good decomposers of plant products. They depend upon bacteria and filamentous fungi to assimilate plant compounds (Phaff and Starmer 1987). For example, cellulose is degraded to cellobiose by other microbial agents in the soil and subsequently assimilated by the yeasts (Phaff and Starmer 1987). Phaff et al. (1978) have indicated that three factors are important in determining the composition of soil yeast community: (1) ability of a particular yeast species to withstand competition from other soil microorganisms; (2) chemical composition of soil, soil type, season, climate and temperature, and soil moisture; and (3) soil flora and fauna, in general. Subterranean ants may play an important role in the accumulation of nutrients in soils. Ant mounds are dynamic microhabitats where food is stored and shared among colony members, and where waste products are deposited (i.e., ant graveyard, feces, seed covers, exoskeleton of dead insects, etc.). Therefore, ants may not only Influence soil nutrients, they may actually have a positive impact on nutrient accumulation in soils. The Impact of the western harvester ant, Pogonomyrmex nn^irientalis (Cresson), on soil nutrient was investigated by Rogers and Lavigne (1974). Soil samples from beneath mounds, discs, and areas beyond colonies showed significant differences in nitrogen and 98 phosphorous content. Soluble nitrogen and phosphorous concentrations were significantly higher beneath mounds compared to areas adjacent to mounds (Rogers and Lavigne 1974). This higher nutrient concentration may be explained by the fact that ants gather and store food In their mounds. In addition, trophylaxic distribution of food may contribute to nutrient accumulation. Gentry and Stirltz (1972) showed that the Florida harvester ant Pogonomyrmex badius Latreille contributes to greater phosphorous and potassium accumulation on the surface of the mound than observed In adjacent soils. The effects of the red Imported fire ant, Solenopsis Invicta Buren, on pasture soil initially assessed by Herzog et al. In 1972. In this Investigation soil from mounds contained significantly higher concentrations of phosphorous and potassium and had a less acidic pH than from off mound soils. Moreover, chemical analyses of grass growing on S.. Invicta mounds exhibited significantly higher protein (14.2 vs. 12.2%), carotene (127.2 vs. 8.62 mg/kg), and phosphorous (0.30 vs 0.28 ppm) than grasses growing on adjacent areas. The objectives of this portion of the investigation were (1) to determine If S.. invicta activities within mounds modified specific soil nutrients; and (2) to determine If the species composition of the yeast community differs between soil mound and non-mound soil. 99 Materials and Methods Soil Collection Soil samples were collected from colony brood chambers, and approximately two meters from mounds (controls). The control samples consisted of soil cores approximately 20 cm deep by 5 cm in diameter. Yeasts are known to be concentrated In the upper layers of soil between 2 and 10 cm and rarely exceed 30 cm (Phaff et al. 1978). Soil samples were collected In the same randomly chosen mounds previously described for yeast isolation. A total of 200 soil samples from brood chambers and 200 soil samples from areas adjacent to the mounds was collected (10 mounds x 5 sites x 4 seasons). Shovels were used for collecting soil from brood chambers; whereas, a soil core sampler was used for obtaining controls. These tools were washed in approximately 25% Clorox (5.25% hypochlorite) to avoid possible yeast contamination between mounds. Samples were sealed In plastic bags, placed in an Ice chest for transportation to the laboratory, and stored at -5oC.

Yeast Isolation and Enumeration Ants and brood remaining In samples were removed by hand. Ten grams of soil were added to 90 ml of sterile deionlzed water and blended for 30 sec. Serial dilutions of 10-2, 10-3 and 10-* were made for each sample and 0.1 ml of each dilution spread on 0.1 ml on acidified yeast exract-malt extract agar [AYM: Difco YM agar plus 0.7% (v/v), 1 M hydrochloric acid, final pH of 3.7-4.0]. This acidic medium Inhibits bacterial growth and provides selectivity for yeast 100 Isolation (Starmer et al. 1976; Phaff et al. 1978). Propionic acid (0.1%) was also added to AYMA to decrease the growth of filamentous fungi (Buhagiar and Barnett 1971). Plates were Incubated at 28^0 for 10 days and obsen/ed daily for yeasts growth. Subsequently, a quantitative estimate of total yeasts from each mound and from each area adjacent to mounds was made from counts of viable CFU/g of soil. In addition, relative frequency of each yeast species was determined by randomly selecting 25 colonies from one plate within each set of dilutions. However, In certain instances, plates contained less than 25 colonies. Consequently, less than 25 colonies were available for selection. In other instances, plates were covered with filamentous fungi and none could be selected. Yeasts were identified, and the number of isolates for each species was expressed as a percent of the total number of isolates.

Yeast Identification Morphologv. The standard identification method always included morphological obsen/ations. The formation of pseudohyphae, true mycelium, and arthrospores was determined by microscopic examination of Dalmau plates of yeasts cultures growing on corn meal agar (yeast suspensions growing under sterile coversllps) after 10 days incubation at 28^0 (Wickerham 1951). /VPl PQC System. The API 20C System was performed on yeasts from soil as previously described for identification of yeasts isolated from the adult worker and worker lan/ae. When Identification by the API 20C System matched the morphological 101 profile of yeast isolates, the identification was usually considered final. In specific Instances, fatty acid profiles were employed for additional confirmation. In addition, fatty acid profiles were always used when the API 20C Identification System proved ambiguous.

Soil Analyses Fifteen soil samples from brood chambers and 15 soil samples from adjacent areas (5 sites x 3 samples) were analyzed by the Texas Agricultural Extension Service, Lubbock, Texas, for organic matter, pH, nitrogen, phosphorus, potassium, calcium, magnesium, salinity, and sodium.

Statistical Analvses Three-way analyses of variance (ANOVAs) were used to determine significant effects of seasons, sites, and number of yeast forming colony units per gram of soil collected from brood chambers and non-mound soil (Sokal and Rohlf 1981). Bartletfs test was used to determine homogeneity of variance (SSPS Inc. 1988, Chicago, IL) Extractable nutrients (nitrogen, phosphorus, potassium, sodium, calcium, and magnesium), pH, salinity, and percent of organic matter in mound soil and non-mound soil were analyzed using a paired t- test (Steel and Torrie 1960). 102 Results Crvotococcus terreus DI Menna was Isolated from the soil adjacent to mounds and from brood chambers with a relative frequency of 33.4% and 14.7%, respectively. Yeast cells from YM broth were globose (3.0-6.5) x (4.0-7) ^im. Pseudomycelial cells were not present on Dalmau plate cultures on corn meal agar. The profile numbers from the API 20C System were C-2543740, C- 2543700 and C-2552350 which were very reliable for £i. terreus In the API Analytical Profile Index. This identification was confirmed by fatty acid analysis (Table 3.1). Geotrichum candidum Link was isolated from the soil adjacent to mounds and from brood chambers with a relative frequency of 19.6% and 9.9%, respectively. Yeast cells from YM broth were globose (3.0- 6.5) X (4.0-7) |xm. True mycelia and arthrospores were present on AYMA plates. The profile number from the API 20C System was C- 6700121 which was a very good match for Q.. candidum In the API Analytical Profile Index. This Identification was also confirmed by fatty acid analysis (Table 3.1). Trichosporon cutaneum (De Beurm. et Vaucher) Ota, also known as Trichosporon begelll (Carmo-Sousa 1970) was isolated from the soil adjacent to mounds and from brood chambers with a relative frequency of 12.1% and 11.2%, respectively. Yeast cells from YM broth were long (3.0-6.5) x (3.0-12) ^im. True mycelia and arthrospores were observed on AYMA plates. The profile numbers from the API 20C System were C-2557333, C-2557372, and C- 25545770 which were very good matches for Jr. cutaneum in the API 103 Analytical Profile Index. Although the fatty acid analysis Indicated that this yeast was Crvptococcu.*; neoformans. the presence of arthrospores definitely shows that this yeast belongs to the genus Trichosporon.

Candida famgitg (Harrison) Meyer et Yarrow was isolated from the soil adjacent to mounds and from brood chambers with a relative frequency of 6.4% and 7.2%, respectively. Yeast cells from YM broth were globose (2.0-4.0) x (4.0-7) ^im. Pseudomycelial cells were not present on Dalmau plate cultures on corn meal agar. The profile numbers from the API 20C System were not In the data base and thus could not be Identified by this method. However, identification was achieved by fatty add analysis (Table 3.1). Rhodoturula rubra (Deme) Lodder was isolated from the soil adjacent to mounds with a relative frequency of 5.2% . However, this yeast was not recovered from brood chambers. Yeast cells from YM broth were long and ovoid (2.-5.0) x (2.5-6) ^m, and colonies exhibited pink pigmentation. Pseudomycelial cells were not present on Dalmau plate cultures on corn meal agar. The profile numbers from the API 20C System were C-6670072 and C-2550073 which were good matches for RJi. rubra in the API Analytical Profile Index. This identification was confirmed by fatty acid analysis (Table 3.1). Rhodoturula minuta (Salto) Harrison was isolated from the soil adjacent to mounds with a relative frequency of 3.5%. However, this yeast was not recovered from brood chambers. Yeast cells from YM broth were ovoid to globose (2.0-6.0) x (2.5-6.5) ^im, and colonies exhibited pink pigmentation. Pseudomycelial cells were not present 104 on Dalmau plate cultures on corn meal agar. The profile numbers from the API 20C System were C-6700121 and C-2270073 which were good matches for Rh.. minuta in the API Analytical Profile Index. This Identification was confirmed by fatty acid analysis (Table 3.1). Candida ymi (Desmazieres Ex Lodder) van Uden et Buckley was isolated from the soil adjacent to mounds with a relative frequency of 3.4%. This yeast was not recovered from brood chambers. Yeast cells from YM broth were ovoid and long (2.0-4.2) x (5.0-10) |im. Pseudomycelial cells were observed on Dalmau plate cultures on corn meal agar. The profile numbers from the API 20C System were not In the API Analytical data base; however, the yeast was identified as £.. vini based on fatty acid analysis (Table 3.1^.Candida tenuis Diddens et Lodder was isolated from the soil adjacent to mounds and from brood chambers with a relative frequency of 12.3% and 1.2%, respectively. Yeast cells from YM broth were ovoid and long (1.0-3.0) x (4-10) |im. Pseudomycelial cells were abundant on Dalmau plate cultures on corn meal agar. The profile numbers from the API 20C System were ambiguous. Subsequently, this yeast was Identified as g. tenuis using fatty acid analysis (Table 3.1). In addition to the aforementioned yeast species found in the soil, other species previously isolated from worker larvae and adult workers were also recovered from soil. These species include £.. parapsilosis. C.. lipolvtica. £.. guillermondii. C rugpga, and C hansenii which were Identified using the API 20C System. The relative frequency of £.. parapsilosis in brood chambers and soil adjacent to 105 mounds was 16.6% and 2.1%, respectively. The yeast, C. lipolytica. exhibited a relative frequency of 17.7% in the mound and 3.2% in controls (soil adjacent to mounds). The relative frequency of £. guillermondii in the brood chambers and the soil adjacent to mounds was 8.5% and 4.4%, respectively. The relative frequency of £. rugosa was 5.1% In the brood chambers and 1.3% in areas adjacent to mounds. The species, D.. hansenii. exhibited a relative frequency of 8.1% In the mound and 3.2% in controls. Ten yeast species were isolated from mound soil, whereas 13 species were Isolated from soil adjacent to the mounds (Table 3.2). Common soil yeast residents, £r. terreus. Q,. candidum. and Tr. cutaneum were the dominant species in soil adjacent to mounds. These yeasts represented 65% of the total isolated from soil adjacent to mounds and represent almost 36% of yeasts present in brood chambers (Table 3.2). However, yeast flora of the red imported fire ant, previously described from fourth instar larvae and adult workers, represented 56% of yeast species found in brood chambers (Table 3.2). A total of 400 samples was processed; however, 143 samples were lost because of the abundant growth of filamentous fungi making the yeast population count impossible. Mean numbers of yeast colony forming units were not significantly different among seasons (df = 3, E > 005) and sites (df = 4, E > 0.05). However, mean numbers of yeast colony forming units per gram of soil between 106 mound soil and soil adjacent to mounds (Tables 3.3 and 3.4) were significantly different (df = 1, E < 0.001). The three-way

Interaction of season, site and location was not significant. ANOVAS are shown In Appendix A.

The mean concentration of nitrogen in mound soil and In soil adjacent to mounds (Figure 3.1) was 46.27 ppm and 18.27 ppm, respectively. Nitrogen In mound soil was not significantly different from nitrogen concentration in control samples (t = 1.78, df = 14, E < 0.05). However, concentration of phosphorous In soil from brood chambers with a mean of 33.73 ppm was significantly greater (t = 3.53, df=14, E < 0.05) than control samples which contained 17.81 ppm of phosphorous. Amounts of potassium (Figure 3.2) In mound soil and non-mound soil (314.2 vs 145.3 ppm, respectively) were not significantly different (t = 2.09, df = 14, E < 0.05). Amounts of sodium In mounds and in soil adjacent to mounds (76. 60 vs 113.33 ppm, respectively) were not significantly different (t = 2.09, df = 14, E < 0.05). The red imported fire ant increased significantly the salinity (Figure 3.2) of mound soil (t = 2.33, df = 14, E < 0.05) in comparison to control samples (523.73 vs. 331.07 ppm, respectively). The mean percents of organic matter in mound soil and adjacent areas were 1.96% and 1.67%, respectively (Figure 3.3). However the amount of organic matter in brood chambers was not

significantly different from controls (t = 1.19, df = 14, E < 0.05). Soil pH was not significantly different between treatments and

controls (Figure 3.4). The mean pH of mound soil was 7.21; whereas

the mean pH for controls was 7.34 (t = 0.53, df = 14, E < 0.05). 107 Calcium concentration In mound soil and controls (5,529.13 vs. 5,417.9 ppm, respectively) was not significantly different (t = 0.164, df = 14, E < 0.05). Amount of magnesium In mound soil and controls (246.6 vs. 311.9 ppm, respectively) did not significantly differ (t = 1.12, df = 14, E < 0.05).

Discussion Colony founding by the red imported fire ant results in soil disturbance, and disturbance is defined by White and Pickett (1985) "as any relatively discrete event In time that disrupts ecosystem, community, or population structure and changes resources, substrate availability, or the physical environment" (p. 7). The abundance of those yeasts previously isolated from the red imported fire ant is greater In mound soil (56%) than in non-mound soil (14%). Therefore, a microenvlronment suitable to these yeasts may be mediated by the red imported fire ant. Moreover, salinity and phosphorus were significantly greater in mound soil than In soil adjacent to mounds. Therefore, this nutrient concentration reinforces the contention that changes are caused by the red imported fire ant In mound soil regarding substrate accumulation. In addition, nitrogen and organic matter accumulated In mound soil; however, significant differences were not observed between mound soil and non-mound soil. Disturbance may act as a buffer against species elimination through competition (Zak 1992). Even so, nutrient patches are obviously created by the presence of red Imported fire ant colonies, and these patches may control the 108 distribution of microorganisms such as yeasts. Environmental patchwork exerts a powerful Influence on the distribution of organisms, their interactions, and their adaptations (WIens 1976). The questions that arise are: does nutrient accumulation have any biological significance?, and does It really "matter" to the yeasts? 109 References Cited

Buhagiar, R. W. M. and J. A. Barnett. 1971. The yeasts of strawberries. J. Appl. Bact. 34:727-739.

DI Menna, M. E. 1957 The Isolation of yeasts from soil. J. Gen. Microbiol. 17:678-688.

Di Menna, M. E. 1960 Yeasts from Antarctica. J. Gen. Microbiol. 23:295-300.

Di Menna, M. E. 1966. Yeasts In Antarctic soils. Antonie van Leeuwenhoek. 32:29-38.

Carmo-Sousa, L. do 1969. Distribution of yeasts In nature. In: Biology and activity of yeasts. A. H. Rose and J. S. Harrison, Eds. Academic Press, New York, pp. 79-105.

Do Carmo-Sousa, L. do 1970. The genus Trichosporon Behrend. in: The yeasts, a taxonomic study. J. Lodder, Ed. North-Holland Publ. Co., Amsterdam, pp. 441 447.

Gentry, B. G. and K. L. Stiritz. 1972. The role of the Florida ant, Pogonomyrmex badius. in old field mineral nutrient relationships. Environ. Entomol. 1:39-41.

Herzog, D. C, T. E. Reagan, D. C. Sheppard, K. M. Hyde, S. S. Nllakhe, M. Y. B. Hussein, M. L. Mcmahan, R. C. Thomas, and L. D. Newsom. 1974. Solenopsis invicta Buren: Influence on Louisiana pasture soil chemistry. Environ. Entmol. 5:160-162.

Phaff, H. J., and W. T. Starmer. 1987. Yeasts associated with plants, Insects and soil. In: The yeasts. Biology of Yeasts. A. H. Rose and J. S. Harrison, Eds. Academic Press. New Yori<. pp. 123- 179.

Phaff,H. J., M. W. Miller, and E. M. Mrak. 1978. The life of yeasts. Harvard University Press, Cambridge, Massachusetts. no Rogers, L. E. and R. J. Lavigne. 1974. Environmental effects of western harvester ants on the shortgrass plains ecosystem. Environ. Entomol. 3:994-997.

Starmer, W. T., W. B. Heed, M. Miranda, M. W. Miller, and H. J. Phaff. 1976. The ecology of yeast flora associated with cactiphllic Drosophila and their host plants In the Sonoran Desert. Microbiol Ecology. 3:11-30.

Steel, R. G., and J. H. Torrie. 1960. Principles and procedures of statistics. McGraw-Hill Book Co. Inc., New York. 481 pp.

Van der Walt, J. P. and D. Yarrow. 1984. Methods for the isolation, maintenance, classification and identification of yeasts. In.: The yeasts, a taxonomic study, 3rd ed. J. W. Kreger-van Rij, Ed. Elsevier, Amsterdam, pp. 45-104.

White, P. S., and Pickett, S. T. A. 1985. Natural disturbances and patch dynamics: an introduction. In: The ecology of natural disturbances and patch dynamics, S. T. A. Picket and P. S. White, Eds. Academic Press, New York, pp. 3-13.

WIens, J. A. 1976. Population responses to patchy environments. Ann. Rev. Ecol. Syst. 7:81-120.

Wikerham, L. J. 1951. Taxonomy of yeasts. U.S. Department of Agriculture. Tech. Bull. 1029, 56p.

Zak, J. C. 1992. Response of soil fungal communities to disturbance, la: The fungal community: Its organization In the ecosystems. G. C. Carroll and Donald T. WIcklow, Eds. Marcel Dekker, Inc., New York. pp. 403-425. Ill

Table 3.1. Fatty acid profiles of yeasts Isolated from mound soil and non-mound soil.

Yeast Soecles Fatty adds Cr. t. G.c. Tr. c. C. f. Rh. m . Rh. r. c. t. C.v

14:0 t r t r t r t r t r t r t r

15:0 - t r - - t r - - 1

16:1 - t r t r 1 t r 2 9 10 16:0 15 13 17 15 17 15 16 12

17:1 - t r t r t r t r t r t r 3

17:0 - t r 1 t r 1 t r - -

16:0 2 OH - t r - - t r - - t r 18:2 58 58 35 26 47 24 31 39 18:1 23 21 37 53 27 54 40 33 18:0 6.0 2 7 3 5 t r 2 1 Total 270592 221528 area 253800 176848 176800 269432 448840 259432 %named 9 9 100 100 100 100 100 100 99

Cr. t. = pryptOCOCCUS terreus. G. c . = Geotrichum candidum. Tr. c. = •^cnr^r^n nutaneijm . C. f.= Car idida famata. Rh. m. = Rhodoturula minuta. Rh. r = Rhodoturula ruiira. C. t. = gaiMda IsmiiS. C.v. = Candida vini. tr refers to trace less than 1%; - refers to not detected. 112

Table 3.2. Relative frequencies of yeast species occurrence in red imported fire ant brood chambers and soil adjacent to mounds collected from five sites along a moisture gradient In Texas. Species Mound (%) Control (%)

Cryptocococcus terreus 14.7 33.4 Geotrichum candidum 9.9 19.6 Trichosporon cutaneum 11.2 12.1 Candida famata 7.2 6.4 Rhodoturula rubra 0 5.2 Rhodoturula minuta 0 3.5 Candida vini 0 3.4 Candida tenuis 1.2 2.3 Candida parapsilosis 16.5 2.1 Candida lipolytica 17.7 3.2 Candida guillermondii 8.5 4.4 Candida rugosa 5.1 1.3 Debaryomy^f^fi hansenii 8.1 3.2 113

Table 3.3. Descriptive statistics (mean, standard error) for number of yeast colony forming units per gram of soil among seasons. Values in parentheses are standard errors. Soil origin

Season Mound soil Non-mound soi

Winter 2,350.0 (183.4) a 1,123.0 (177.7) b Spring 2,359.0(181.4)3 959.4(60.9) b Summer 2,097.1 (116.0) a 821.2(46.4) b Fall 2,145.2 (149.6) a 957.6 (67.6) b ^'b Means with different superscript within season differ (P < 0.0001).

Table 3.4. Descriptive statistics (mean, standard error) for number of yeast colony forming units per gram of soil among sites. Values in parentheses are standard errors. Soil origin

Site Mound soil Non-mound soil

Abilene 2.248.2 (121.7) a 923.1 (121.7) b

Stephenville ' 2,275.0 (224.3) a 1,089.5 (224.3) b Waco 2,380.8 (198.0) a 981.8(71.1) b

Huntsville 1,704.2(96.7) a 796.4(65.4) b

Beaumont 2614.3 (242.1)8 a 1,073.1(89.4) b a,b Means with different superscript within site differ (P < 0.0001). 114

400

E

Q. 300-

V) Xc 200- 3

tcr 100- 3 o E < NITROGEN POTASSIUM PHOSPHORUS SODIUM NUTRIENTS

Figure 3.1. Comparisons of mean nutrient concentrations between mound soil and soil adjacent to mounds (paired t-test). Means for the same nutrient with an asterisk are significantly different (P. < 0.05). 115

E Q. Q.

(0 C 0)

o E < MAGNESIUM SALINITY CALCIUM

NUTRIENTS

Figure 3.2. Comparisons of mean nutrient concentrations between mound soil and soil adjacent to mounds (paired t-test). Means for the same nutrient with an asterisk are significantly different (E. < 0.05). 116

y\

MOUND CONTROL

2- *-» •- CO

o "c (0 1 -

MOUND CONTROL

Figure 3.3. Comparison of mean organic matter concentrations between mound soil and soil adjacent to mounds (paired t-test). Means with the same letter are significantly different (P. < 0.05). 117

MOUND

CONTROL

0) _3 (0 >

MOUND CONTROL

Figure 3.4. Comparison of mean pH values between mound soil and soil adjacent to mounds (paired t-test). Means with the same letter are not significantly different (E < 0.05). CHAPTER IV STEROL PROFILES OF THE RED IMPORTED FIRE ANT AND ASSOCIATED YEASTS

Introduction The early nutritional studies of sterols as growth factors in various Insect species showed that ergosterol was the third most effective sterol after cholesterol and sitosterol (Clayton 1964). Ergosterol is the dominant sterol associated with fungi. Cholesterol provided good growth for most of the insects in a study by Clayton (1964). However, it was not as effective as some phytosterols for certain phytophagous species (Clayton 1964). The plant sterol, sitosterol, was utilized more efficiently than cholesterol by the silkworm, Bombvx mori. and aquatic larvae of the mosquito, Aedes egvpti. The corn borer, Pvrausta nubilasis. however, utilized ergosterol more effectively than cholesterol (Clayton 1964). In addition, a group of eight omnivorous and phytophagous species grew equally well with ergosterol as with cholesterol. The early nutritional studies of sterols as growth factors in various insect species showed that ergosterol was the third most effective sterol after cholesterol and sitosterol (Clayton 1964). One of the first reports of ergosterol conversion to dehydrocholesterol was that of Clark and Bloch (1959a). In their investigation, nymphs of the German cockroach, Blatella g^rmanica., were raised successfully on synthetic diets containing ergosterol as the sole sterol source. Clark and Bloch (1959a) showed the

118 119 conversion of ergosterol to dehydrocholesterol by feeding Blatella nymphs radiolabeled ergosterol. Therefore, dehydrocholesterol In Bl9tell9 nymphs feeding on ergosterol progressively became the major sterol. The ability of the German cockroach to grow and to develop normally without cholesterol would Imply the existence of mechanisms for converting dehydrocholesterol directly to the necessary metabolic products, by-passing cholesterol (Clark and Bloch 1959a). Therefore, ergosterol may serve as a "sparing" sterol in natural systems (Clark and Bloch 1959a). Other Investigations have shown the ability of insects to grow and develop normally on ergosterol as a sole source of sterol. Fraenkel and Blewitt (1943) Investigated the sterol requirements of seven insect species belonging to the genera Tribolium. Lasloderma. Sllvanus. Ptinus. Sitodrepa. Ephestica. and Dermestes. Their study showed that ergosterol satisfied the growth requirements of six of the seven. The only genus which failed to grow on yeast sterols was Dermestes. This failure may be explained by the fact that species of Dermestes are carnivorous and mostly rely on cholesterol to satisfy their dietary requirements. Dermestes vuloinus lan/ae fed phytosterols or yeast sterols failed to pupate or showed abnormal development (Clark and Bloch 1959b). On the other hand, ergosterol satisfied the sterol requirement of the angoumous grain moth, g^itotrooa nerealella (Chippendale 1971). Fifteen adults developed from ergosterol-fed lan/ae, while 19 adults developed from cholesterol-fed larvae. A series of sterols were also tested on the khapra beetle, Trngoderma granarium (AganA/all 1970). Of those 120 tested for growth and development, cholesterol, cholestanol, stigmasterol, B-sitosterol, 7-dehydrocholesterol, and ergosterol were satisfactory (AganA^all 1970). Drosophila melanogaster lan/ae were tested for sterol utilization (Cooke and Sang 1969), and ergosterol was also found to provide faster development than cholesterol for H. melanogaster larvae. However, the tea totrix, Homona coffearia. showed different percents of emergence depending on diet (Sivapalan and Gnanapragasama 1978). Ergosterol as the sole sterol provided 67% moth emergence, while a mixture of sucrose, multivitamins, and ergosterol provided 87% emergence. No moth emergence was recorded when cholesterol was included in the diet as the only sterol (Sivapalan and Gnanapragasama 1978). Sterol requirements for pupation of the beetle, Xyleborus ferrugineus. was investigated by Chu et al. (1970) who found that aposymbiotic Individuals of the species failed to pupate on a diet supplemented with cholesterol or lanosterol. However, aposymbiotic individuals produced several broods or generations of progeny that pupated and formed normal adults when ergosterol or 7-dehydrocholesterol was the sole source of sterol (Chu et al. 1970). Therefore, these various investigations Indicate that ergosterol can satisfy the sterol requirement of different Insect species. The objective for this portion of the study was to determine if the dominant yeast species produce sterols that could be utilized by the red imported fire ant. 121 Materials and Methods Colonies were collected 25 September1993 and 15 April 1994 from Abilene, Texas. Those colonies collected In September were kept in the laboratory for four months and fed a mixture of boiled chicken eggs and soybean oil (ratio of one egg to 2 ml soybean oil) and water every other day. From these colonies, 26.72 g of workers (approximately 15,089 individuals) were used for sterol extraction. Those colonies collected In April were immediately separated from soil and frozen for sterol extraction. From these colonies, 0.55 g of eggs (approximately 80,000), 0.75 g of fourth Instar larvae (approximately 1,400), 10.29 g of workers (approximately 5,800 ), and 3.74 g of queens (approximately 420) were used for sterol analysis. The dominant yeasts of the red Imported fire ant were found In this study to be Candida parapsilosis and Candida lipolytica. Each species was cultured In four cotton-plugged flasks with 250 ml yeast extract malt extract broth on a shaker at room temperature for two weeks. Total yield was 35.2 g and 25.6 g of C parapsilosis and C. lipolytica. respectively.

.qtftrni Extraction Ant and yeast specimens were homogenized in liquid nitrogen with a mortar and pestel and subsequently saponified in 10% (w/v) potassium hydroxide In 90% methanol under reflux for 30 min. After the extract was filtered, an equal volume of glass-filtered water was added. The mixture was partioned with an equal volume of 122 diethyl ether three times. The ether phase was dried in a rotary evaporator under vacuum. Subsequently, the extract was dried further with benzene by rotary evaporation under vacuum to yield the non-saponifiable fractions (NSF). Furthennore, the NSF was completely dried at 60^0 with nitrogen. The non-saponifiable lipids were developed using preparative thin layer chromatography (TLC, Analtech, Silica Gel) with benzene-ether (85:15). Cholesterol and lanosterol standards were used to locate 4-desmethyl, 4- monomethyl, and 4,4-dimethyl sterol bands. Subsequently, the three areas on TLC plates, corresponding to the three classes of aforementioned sterols, were scraped separately and extracted with acetone.

Sterol Analvsis Gas liquid chromatography (GLC) was performed on a Hewlett- Packard Model 5890 gas chromatograph. The following operating parameters were set: column temperature, 245^0; detector temperature, 300oC; injector temperature, 275^0; and helium carrier gas flow-rate, 20 ml/mln. Quantitative analysis of sterols by GLC was relative to cholesterol (RRTc). Reversed-phase high- performance liquid chromatography (HPLC) was performed on a Hewlett-Packard Model 1090. The HPLC was equipped with a Zorbax ODS CIS-reversed phase column connected to an ISCO variable wavelength detector set at 205 nm. The column was eluted with methanol at room temperature at a flow-rate of 1 ml/mln. In addition, sterols were analyzed using a gas chromatography-mass 123 spectrometer system (Hewlett-Packard 5870 MSD). A capillary

column was used for GC-MS (15 M DB-5, 0.25 mm i.d, 0.25 |im film).

The operating temperatures were programmed in the GC-MS as

follows: 170^0 for 1 min, increasing to 270^0 at a rate of

20oC/min.; Isothermal at 270^0 for 3 min.. Increasing to 280oC at a rate of 2oC/min.; and Isothermal at 280oC for six minutes. Nuclear Magnetic Resonance ("IH-NMR) was performed on a AF-200 Spectrometer.

Results A total of 26 sterols was isolated from the red Imported fire ant (Table 4.1). Queens showed the highest diversity in their sterol pool. A total of 24 sterols was Isolated from queens, and the sterols were Identified based on their chromatographic and spectral properties (Table 4.2). The plant sterol, sitosterol, was the most abundant sterol In queens followed by cholesterol (Appendix B). The position of the double bond configuration of sitosterol and cholesterol was confirmed using "^H-NMR after elutlon of these sterols fom HPLC. In addition, traces of other 24-ethyl sterols were present in queens. Eggs, larvae, and workers exhibited similar sterol profiles (Table 4 1) Cholesterol was the most abundant sterol In these life stages. Moreover, workers from colonies maintained for four months on a cholesterol-rich diet showed a similar sterol profile to the non-fed 124 workers which were saponified shortly after collection. However, 70% of total sterol from those ants on a cholesterol-rich diet was cholesterol. The non-fed workers contained only 55% of their sterols In the form of cholesterol.

In general, cholesterol and the plant sterols, sitosterol, campesterol, and stigmasterol, were the dominant sterols associated with the red imported fire ant. From the sterol profile of eggs, worker lan/ae, adult workers, and queens, the dealkylation of the phytosterols to cholesterol seems to occur. The structure of sterols extracted from the red imported fire ant are shown In Table 4.3. The chromatographic and spectral properties of these sterols are presented in Table 4.2.

The fungal sterol, ergosterol, was detected In worker larvae, adult workers, and eggs. In addition, brasslcasterol, which may also be found in brassica plants (Nes and Mckean 1977), was Isolated from all developmental stages of the red imported fire ant. These sterols possess A22-24-methyl groups, and this side chain Is not normally found in phytosterols. The principal sterols extracted from Candida, parapsilosis were identified (Table 4.3) as ergosterol (88%), A5,23-methylenecholesterol (6.5%), and zymosterol (5%). However, lesser amounts of ergosta-5,7,9(11)22(23)-tetraenol, ergosta-5,23-dlenol, obtusifoliol, A8-24-methylenelophenol, sitosterol, citrostadienol, and traces of unusual sterols were detected in this yeast species. On the other hand, Candida lipolytica 125 however, showed less diversity In Its sterol components (Table 4.4).

Subsequently, the three predominant sterols in C. lipolytica were ergosterol (52%), zymosterol (39%), ergosta-5,7,9(11)22(23)- tetraenol (8%), and traces of unidentified sterols.

Discussion Noda et al. (1979) found that the smaller brown planthopper, Laodelphax striatellus Fallen, has two sterol sources, one from the rice plant sap and the other from the symbiont, which appears to provide 24-methylene cholesterol. In addition, the role of intracellular symbionts as sterol donors In the planthoppers, Nilaparvata lugens Stal and Laodelphax striatellus Fallen, was investigated by Eya et al. (1989). They Identified the yeast-like symbiont Isolated from the host's eggs as a member of the genus Candida, and this symbiont produces ergosterol when cultured. Eya et al. (1989) concluded that ergosterol produced by the symbiont is supplied to the host and possibly transformed in the host insects to cholesterol via methylenecholesterol (Eya et al. 1989). Diversity in sterol production by symbiotic microorganisms from planthoppers was also shown by Wetzel et al. (1992). Yeasts associated with the red Imported fire ant show some similarities In their sterol profiles to those symbionts found In planthoppers. Planthopper symbionts produce 24- methylenecholesterol for the host Insect (Wetzel et al. 1992). Therefore, the production of 24-methylenecholesterol In Q_. g^r^gsWosiS. may be Important to the ant and may contribute to its 126 selection as the dominant yeast species associated with S.. invicta. Traces of methylenecholesterol in fourth Instar larvae and red imported fire ant eggs may be of fungal origin. The presence of most of these sterols In S.. Invicta suggests that the yeast symbionts may contribute to the sterol pool of the red Imported fire ant. In addition, the red imported fire ant may reduce the A7-bond of ergosterol. However, definitive conclusions could only be made if pathways of radiolabelled compounds are followed. Nevertheless, ergosterol provided by £.. parapsilosis and Q_. lipolytica may contribute to the sterol pool of S.- invicta. The fact that ergosterol did not accumulate suggests that this sterol may be used by the insect. Adult workers feed readily on ergosterol (95% purity). Moreover, Vinson (1972) reported that the red Imported fire ant was feeding on Paspalum seeds (Dallas grass), but the seeds Infested with ergot fungi were the only ones selected. Ergosterol obtained from the yeast may act only as a "sparing" sterol as it does In RiatPlla germanica (Clark and Bloch 1959a), because the red imported fire ant has a diverse source of sterols available from both plants and animals. 127

References Cited

Agarwall, H. C. 1970. Sterol requirements of the beetle Trogoderma. J. Insect Physiol. 16:2023-2026.

Akey, D. H., and S. D. Beck. 1972. Nutrition of the pea aphid Acyrthoslphum pisum: requirements for trace metals, sulphur, and cholesterol. J. Insect Physiol. 18:1901-1914.

Campbell, B. C, and W. D. Nes. 1982. A reappraisal of sterol biosynthesis and metabolism in aphids. J. Insect Physiol. 29:149-156.

Chippendale, G. M. 1971. Lipid requirement of the angoumois grain moth, Sitotroga cerealella. J. Insect Physiol. 17:2169-2177.

Chu, H., D. M. Norrls, and L. T. Kok. 1969. Pupation requirement of the beetle, Xyleborus ferrugineus: sterols other than cholesterol. J. Insect. Physiol. 16:1379-1387.

Clark, A. J., and K. Bloch. 1959a. Function of sterols in Dermestes vulplnus. J. Biol. Chem. 234:2583-2588.

Clark, A. J., and K. Block. 1959b. Conversion of ergosterol to 22- dehydrocholesterol in Blattella germanica. J. Biol. Chem. 234:2589-2594.

Clayton R. B. 1964. The utilization of sterols by Insects. J. Lipids Res. 5:3-19.

Cooke, J., and J. H. Sang. 1969. Utilization of sterols by lan/ae of Drosophila Melanooaster. J. Insect Physiol. 16:801-812.

Dadd R. H., and Krieger D. L. 1968. Dietary amino acid requirements of the aphid, Myzus persicae. J. Insect Physiol. 14, 741-764.

Dadd, R. H., and T. E. Mittler. 1966. Permanent culture of an aphid on a totally synthetic diet. Experientia. 22:832-835. 128 Ehrhardt, P. 1968. Nachweis einer durch symbiontische microorganimen bewirkten sterinsynthese in kunstlich aphiden. (Homoptera, Rhinchota, Insecta). Experientia. 24, 82-83.

Eya, B. K., P. T. M. Kenny, S. Y. Tamura, M. Ohnishi, Y. Naya, K. Nakanishi, and M. Suglura. 1987. Chemical association In symbiosis, sterol donors in planthoppers symbiosis. 1989. J. Chem. Ecol. 15:373-380.

Fraenkel, G., and M. Blewitt. 1943. The sterol requirements of several insects. Biochem. J. 37:692-695.

Griffiths, G. W., and S. D. Beck. 1977a. In vivo sterol biosynthesis by pea aphid symbiotes as determined by digitonin and electron microscopic autoradiography. Cell TIss. Res. 176: 179-190.

Griffiths, G. W., and S. D. Beck. 1977b. Effect of dietary cholesterol on the pattern of osmium deposition in the symbiote- contalnlng cell of the pea aphid. Cell TIss. Res. 176: 191-203.

Houck, E. J., G. W. Griffiths, and S. D. Beck. 1976. Lipid metabolism In the symbiotes of the pea aphid, Acyrthosiphon pisum. Comp. Biochem. Physiol. 54B:427-431.

Karlson, P., and H. Hoffmeister. 1963. Biogenesis of ecdysone: conversion of cholesterol to ecdysone. Z. Physiol. Chem. 331:298-300.

Noda, H., W. Kojiro, and T. Salto. 1979. Sterols in Laodelphax striatellus with special reference to the Intracellular yeastlike symblootes as a sterol source. J. Insect Physiol. 25:443-447.

Nes, W. R. and M. L. Mckean. 1977. Biochemistry of steroids and other Isopentenolds. University Park Press, Baltimore, pp. 412-518.

RItter, S. R., B. A. Weiss, and A. L. Norrt)om. 1981. Identification of A5.7.24 methylene- and methylsterols In the brain and whole body of Atta cephalotes isthmlcola. Comp. Biochem. Physiol. 71:345-349. 129

Sivapalan, P., and N. C. Gnanapragasam. 1978. The influence of llnoieic acid and linolenic acid on adult moth emergence of Homona coffearia from merldic diets In vitro. J. Insect physlol. 25:393-398.

Svoboda, J. A.,and J. N. Kaplans, W. E. Robbins, and M. J. Thompson. 1975. Recent developments In insect steroid metabolism. Ann. Rev. Entomol. 20:205-220.

Vinson, S. B. 1972. Imported fire ant feeding on Paspalum seed. Ann. Entomol. Soc. Amer. 65:988. 130

Table 4.1. Sterol composition of various developmental stages and life forms of the red imported fire ^. sterols structure b egg larva worker queen

Cholesta- 5,22dlenol 1A 0.8 N.D. 0.7 N.D. Cholesterol MB 44.4 50.1 55.0 21.7 Cholest-7 -enol 2B 4.0 9.3 1.0 7.4 Cholestanol 3B N.D. N.D. N.D. 2.8 Ergosterol 4D 0.8 1.2 0.5 N.D. Brasslcasterol ID 12.5 9.1 11.2 3.6 Ergosta-5, 23-dlenol IE 0.7 3.2 N.D. 2.3 Campesterol 1G 12.5 8.2 10.6 5.6 Campest-7-enol 2G 0.6 1.0 N.D. 2.8 Isofucosterol 1 1 3.4 1.2 4.3 t r Avenasterol 21 0.5 1.3 N.D. 2.4 Stigmasterol 1 H 0.8 1.4 3.4 3.2 Stigmast-7, 22 dienol 2H N.D. N.D. N.D. t r Stigmast- 4^ 22-enol 3H N.D. N.D. N.D. t r Sitosterol U 17.4 10.1 12.8 32.8 Stigmast-7-enol 2J 1.1 3.1 0.2 11.9 Sitostanol 3J N.D. N.D. N.D. 2.7 Obtusifoliol 5F N.D. N.D. N.D. t r

24-Methylene & 1 ^^ A ^ lophenol 6F N.D. N.D. N.D. t r 3-24-Methylene N.D. t r lophenol 7F N.D. N.D. N.D. 0.6 N.D. t r Citrastadlenol 61 ^ 1 p^ A. .. Cycloartenol 8B N.D. N.D. N.D. t r ^ 1 r*\ Cvcloartanol 8C N.D. N.D. N.D. t r 131

Table 4.1. Continued sterols structure egg larva worker queen

24-Methylene lanosterol 9F N.D. N.D. N.D. t r 24-Methylene parked 10F N.D. N.D. N.D. t r 24-Methylene cycloartanol 8F N.D. N.D. N.D. t r Total sterol (|ig/part) 0.02 2.44 3.18 24.17 Number of specimens used 8000 1400. 5800 420 a As a percent total sterol; tr = trace; N.D. = not detected, b Structures of sterols are shown In Figure 4.1. 132

Table 4.2. Chromatographic and spectral properties of sterols isolated from the red imported fire ant. TLCa GLCb HPLCC MS (M+ and other sterol (Rf) (RRTc) (ac) diagnostic ions) d Cholesta-5, 22-dienol 0.18 0.92 0.82 384 369 366 351 300 255 Cholesterol 0.18 1.00 1.00 386 371 368 353 301 255 Cholest-7-enol 0.16 1.10 1.09 386 371 368 353 273 255 Cholestanol 0.18 1.03 1.11 388 373 370 355 311 262 Ergosterol 0.18 1.21 0.76 396 381 378 363 337 271 Brasslcasterol 0.18 1.12 0.85 398 383 380 300 271 255 Ergosta-5, 23-dienol 0.18 1.26 0.95 398 383 365 339 314 271 Campesterol 0.18 1.29 1.13 400 385 382 367 315 289 Ergost-7-enol 0.16 1.42 1.23 400 385 379 314 273 255 Isofucosterol 0.18 1.65 1.00 412 397 369 314 299 281 Avenasterol 0.16 1.72 1.03 412 397 368 314 299 271 Stigmasterol 0.18 1.40 1.10 412 397 394 351 300 271 Stlgmasta-7, 22-dlenol 0.16 1.54 1.20 412 397 379 300 271 255 Stlgmast-22 -enol 0.18 1.44 1.22 414 399 396 353 300 273 Sitosterol 0.18 1.60 1.18 414 399 396 354 329 303 Stigmast- 7-enol 0.16 1.76 1.29 414 399 381 314 273 255 Stigmastanol 0.18 1.64 1.30 416 401 359 314 299 255 Obtusifoliol 0.25 1.48 0.92 426 411 383 327 285 245 24-Methylene 328 313 285 lophenol 0.25 1.64 0.97 412 397 379 133

Table 4.2. Continued. TLCa GLCb HPLCC MS (M+ and other sterol (Rf) (RRTc) (ac) diagnostic lons)d

4a-Methyl24- methylene- cholest- 8-enol 0.25 1.49 0.97 412 397 383 327 313 285 Citrastadlenol 0.25 2.15 1.13 426 411 368 328 313 285 Cycloartenol 0.29 1.84 1.04 426 411 408 393 365 339 Cycloartnol 0.29 1.88 1.14 428 413 410 393 367 341 24-Methylene lanosterol 0.29 1.88 1.07 440 425 422 411 379 300 24-Methylene parked 0.29 2.03 1.07 440 425 407 397 341 313 24-Methylene cycloartanol 0.29 2.10 1.12 440 425 422 407 379 353

^TLC was developed In benzene/ether (85:15). ^GLC was performed on 3% SE-30 glass packed column at 245^0 isothermally. CHPLC was operated on Zorbax ODS column eluted with methanol at I.OOml/mln. at room temperature. The UV detector was set at 205 nm. dMS fragmentation pattern of sterols; only significant ions (< 10% abundance) are given between m/z 255 and 440 amu. 134

Table 4.3. Sterol profile of the dominant yeast, Candida parapsilosis, associated with the red Imported fire ant.

Sterols Percent ergosterol 88.0 methylene cholesterol 6.5 zymosterol 4.5 ergosta-5,7,9(11)22(23)tetraenol trace ergosta-5,23-dlenol trace obtusifoliol trace A8-24-methylene lophenol trace sitosterol trace citrostadienol trace 135

Table 4.4. Sterol profile of the second most prevalent yeast, Candida lipolytica. associated with the red Imported fire ant. sterols percent ergosterol 52.0 zymosterol 39.0 ergosta-5,7,9(11)22(23)tetraenol trace 136

28 17N fl B D

H

Figure. 4.1. Structures (sides chain and nucleus) of sterols isolated from the red imported fire ant. APPENDIX A: SIGNIFICANCE LEVELS FOR ANOVAS AND MANOVA ASSESSING DIFFERENT VARIABLES (COLONY WEIGHT SITE, SEASON, SOIL MOUND AND NON-MOUND SOIL) OF THE YEAST ASSOCIATION WITH THE RED IMPORTED FIRE ANT

137 138

Table A. 1. ANOVA (GLM) evaluating the effects of yeasts on colony weight.

Source of Sum of variation DF Square F P

Model 5 29310.701 11.81 0.0001 Error 194 96263.331 Corrected 199 125574.033

Table A. 2. ANOVA (GLM) evaluating yeast effects on weights of colonies collected during winter 1992.

Source of Sum of variation DF Squares

Model 5 18328.891 7.15 0.0001 Error 44 22556.850 Corrected Total 49 40885.741

Table A. 3. ANOVA (GLM) evaluating yeast effects on weights of colonies collected during spring 1993. Source of Sum of variation DF Squares F Value P

Model 5 15678.610 10.20 0.0001 Error 44 13525.094 norrected Total 49 139

Table A. 4. ANOVA (GLM) evaluating yeast effects on weights of colonies collected during summer 1993. Source of Sum of variation DF Squares F Value

I < <• Model 5 2370.180 0.94 0.4665 Error 44 22259.290 Corrected Total 49 24629.471

Table A. 5. ANOVA (GLM) evaluating yeast effects on weights of colonies collected during fall 1993. Source of Sum of variation DF Squares F Value

Model 5 4076.833 1.44 0.2276 Error 44 24843.934 Corrected Total 49 28920.768

Table A. 6. ANOVA (GLM) evaluating yeast effects on weights of colonies collected from Abilene. Source Sum of of variation DF Squares

Model 1 1354.668 8.91 0.0049 Error 38 5778.130 Corrected Total 39 7132.799 140

Table A. 7. ANOVA (GLM) evaluating yeast effects on weights of colonies collected from Stephenville. Source of Sum of variation DF Squares

Model 1 1123.632 3.28 0.0782 Error 38 13031.812 Corrected Total 39 114155.444

Table A. 8. ANOVA (GLM) evaluating the effects of yeasts on colony weight In the Waco site. Source Sum of variation DF Squares

Model 1 877.145 1.99 0.166 Error 38 16743.378 Corrected Total 39 17620.523

Table A. 9. ANOVA (GLM) evaluating the effects of yeasts on colony weight in the Huntsville site. Source of Sum of variation DF Squares F P

Model 1 250.712 0.94 0.3392 Error 38 10169.918 Corrected Total 39 10420.631 141

Table A. 10. ANOVA (GLM) evaluating the effects of yeasts on colony weight in the Beaumont site. Source of Sum of variation DF Squares F P

Model 1 2972.065 2.27 0.1400 Error 38 49703.970 Corrected Total 39 52676.036

Table A. 11. ANOVA showing effects of dates, sites, and locations (mound soil or soil adjacent to mound) on yeast colony forming units.

Source of Sum of variation Squares DF

Main Effects 56162608.0 8 9.686 0.000 dates 672285.0 3 0.769 0.512 sites 3897710.0 4 1.344 0.254 locations 50734842.0 1 69.999 0.000

Table A. 12. Two-way ANOVA evaluating the effects of dates and sites, dates and locations, and sites and locations (mound soil or soil adjacent to mound) on yeast colony forming units. Sum of Source of Variation Squares DF F P

2-Way Interactions 19486573.0 19 1.415 0.121 Date by Sites 14662377.0 12 1.686 0.071 Dates by Locations 1442540.0 3 0.663 0.575 Sites by Locations 3832731.0 4 1.322 0.263 142

Table A. 13. Three-way ANOVA evaluating the effects of dates, sites, and locations (mound soil or soil adjacent to mound) on yeast colony forming units. Sum of Source of variation Squares DF F P

3-Way Interactions 10416064.0 ll 1.198 0.286 Dates by Sites by Loc. 10416064.0 12 1.198 0.286 APPENDIX B: MASS SPECTRUM OF MAJOR STEROLS ISOLATED FROM THE RED IMPORTED FIRE ANT AND ASSOCIATED YEASTS

143 144

55

a, 8« HO C Sitosterol ro = 60 81 105 c 14 41 40- 119 21 329 3fl3 23 1 25,5 396 173 273 20

aun 4M ^ 1I Ui JLIMJUU - II ^ 300 "2^5^ ' ' 4d0' M/z 58e 10*0 50 m

Figure B. 1. Mass spectrum of the fragmentation pattern of sitosterol (M+ and other sterol diagnostic). 145

5b 80

1^0 8 ^5 Campesterol 119 145 |40^ 400 161 289 213 31 a: 255 382 20^ 199 273 340 0 Jt J 11 mil II JJ 1 Ll. iJilli^ In \ .1 M/Z 50 100 150 200 5?0 300 350 400

Figure B. 2. Mass spectrum of the fragmentation pattern of campesterol (M+ and other sterol diagnostic ions). 146

55 80 u c ro 60 81 T3 Stigmasterol C ^ 40 105 41 CE 15i3y9 ^^"30271 0 20 ill III ^"'^^l^ ^^^ ^''^ 0 U IJ M/z 50 100 150 200 250 300 350 400 450 500

Figure B. 3. Mass spectrum of the fragmentation pattern of stigmasterol (M+ and other sterol diagnostic ions). 147

55 386 80-1 HO u 105 Cholesterol S 60 30 1 ro 145 27 368 = 40 213 cc 20^ 199 326

0 U T'"T'I'Vi'1T y-i~a^ i r i-i-i—r-T—i—i—i—i—i—i—i—i—X i i—i—r—i y—< M/Z 50 100 150 200! 250 300 350 40f0 450 500

Figure B. 4. Mass spectrum of the fragmentation pattern of cholesterol (M+ and other sterol diagnostic ions). 148

36 80 u 55 396 I 60 HO

£40 143 81 337 20 105 311 471 0 Mh t|ilJii]ii,^t,l,J I I I 11)''' M/z 5i0 100 1530 200 250 300 350 400 450 500 Figure B. 5. Mass spectrum of the fragmentation pattern of ergosterol (MH- and other sterol diagnostic ions). 149

271

Methylenecholesterol 314 36,5 398 lil^i . 19,7„^27 299 4i ju,

Figure B. 6. Mass spectrum of the fragmentation pattern of 24- methylenecholestero (M+ and other sterol diagnostic ions).