AUMENTARY TRACT MICROBWTA 0F AQUATIC INVERTEBRMES Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSITY AMANDA KAY Man 19 7 5 ‘J J. u 1‘. 1 A 4 \ .. Mid-153m Stem Uzfivszsity JHESIS a? 7 '« £735“ am 1:, win; HUAG & SUNS" i *1 300K BINDERY ma “ LIBRARY HINDU“ _ .; snmorqutwcmsau ' ' 52v ‘ :4 ABSTRACT ALIMENTARY TRACT MICROBIOTA OF AQUATIC INVERTEBRATES By Amanda Kay Meitz Microscopical examination of twenty-six species of aquatic inver- tebrates, primarily insect larvae, revealed that a gut microbiota is widespread. Microbiota was present in the midguts and, more frequently, in the hindguts of the larvae. Morphologically diverse microbiota was observed in the gut lumen and firmly adhering to the gut wall of a number of the larvae examined. Rods were the most frequently observed bacterial morphologies, however, prosthecate and filamentous bacteria and members of an obscure class of fungi, trichomycetes, were also noted. With some exceptions, the presence or absence of a gut microbiota was found to be correlated with the food habits of the insect. Detritivorous insects were observed to possess a dense midgut or hindgut biota, and occassionally both. Invertebrates living on more nutritious substrates such as algae or insect prey possessed a sparse microbial population in their alimentary tracts . ALIMENTARY TRACT MICROBIOTA 0F AQUATIC INVERTEBRATES BY Amanda Kay Meitz A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Microbiology and Public Health 1975 ACKNOWLEDGMENTS I wish to thank my major professor, Dr. Michael Klug, for his continuing interest and patience. I express appreciation to members of my guidance comittee, Dr. Kenneth Cummins, Dr. James Tiedje, and Dr. John Breznak, for their availability and suggestions. Special thanks are extended to Dr. Darwin Buthala of Western Michigan University for many hours of instruction in electron microscopy and permission to use the electron microscope facility at Western Michigan University. Thanks also to Stuart Pankratz at Michigan State University for providing suggestions and assistance with electron microscopy techniques. To my parents and family I am indebted for years of moral support and perennial faith in me. Thanks to the graduate students and staff at the Kellogg Biological Station for numerous and productive discussions. This research was supported by National Science Foundation grant GB-36069X to Drs. Cummins and Klug at the Kellogg Biological Station. ii LIST OF TABLES . . LIST OF FIGURES . INTRODUCTION . . MATERIALS AND METHODS TABLE OF CONTENTS Collection and Maintenance of Insects . Light Microscopy and Enumeration of Biota Electron Microsc0py RESULTS . . . Gut Morphology . Gut Microbiota . DISCUSSION . . . Gut Morphology . Gut Microbiota . Feeding Categories Possible Microbe-Insect Interactions . LITERATURE CITED iii page iv l3 l3 18 41 41 43 47 54 60 LIST OF TABLES Gut morphologies and taxonomy of invertebrates . . Gut dimensions and Petroff-Hauser counts of bacteria in selected invertebrates . . . . . . . . . . Morphologies of bacteria observed in Petroff-Hauser counts. Feeding categories and gut microbiota of invertebrates iv Page 15 31 34 38 LIST OF FIGURES Figure Page 1. A simplified view of trophic relationships in a woodland stream community. Dashed arrows indicate less frequent eXChange O O O O O O O O O O O O O O O O O 3 Examples of gut morphologies observed. Left: Hydatophylax hesperus gut, characteristic of the simple gut morphology. Center: gigronia serricornis gut, a more complex gut morphology. Right: Tipula abdominalis, a complex gut with a fermentation chamber. Foreguts and hindguts are outlined with a heavy line to denote chitinization. Malpighian stubules attach at the line where the midgut and hindgut meet. g--gastric cecae; i--ileum; pr--proventriculus; py--pylorus; r--rectum; rl--rectal lobe; rs--rectal sac . . . . . . l7 Lumen bacteria from Tipula. Upper: Agar slide photomicrograph lBOOX. Lower: Shadowed electronmicrographs. Left:5280X Right: 5320K . . . . . . . . . . . . . . . 20 . Tipula rectal sac and bacteria. Upper: Epon-embedded thick sections of non-washed rectal sac and lumen contents. Left: 310x Right: 780x. Lower: Electronmicrograph of washed rectal sac. 11,550X . . . . . . . . . . 22 Bacterial filaments from Tipula. Upper left: Wet mount of sporulating filamentous bacteria and gut wall near juncture of the ileum, rectal sac and rectal lobe. Masses of shorter rods are against the wall. 800x. Upper right: Agar slide preparation of the distal ends of the filamentous bacteria with Spores and in the presporulation stage following septum formation. l300X. Lower: Electronmicrograph of the filaments and cuticle of gut wall in the region of the ileum, sac, and lobe juncture. 11,550X . . . . . . . . . 24 Prosthecate bacteria from Tipula. Upper: Prosthecates in close proximity to the gut wall. 16,800X. Lower left: 24,800X. Lower right: 42, 750x . . . . . . . 28 Trichomycete hyphae with immature spores, bacteria and detritus visible through the intact midgut wall of the blackfly Prosimulium. . . . . . . . . . . 30 INTRODUCTION It is generally concluded that first to third order woodland stream communities depend upon inputs from surrounding terrestrial areas for the majority of their energy (Nelson and Scott, 1962; Hynes, 1963; Egglishaw, 1964; Minshall, 1967; Triska, 1970; Fisher, 1971; Hall, 1971; Fisher and Likens, 1972, 1973; Cummins g£_§l, 1972, 1973). Particulate organic matter, primarily in the form of senescent leaf material, makes up a large percentage of this input with estimates of these inputs ranging from 0.97 to 5.0 g/m2/day (Peterson and Cummins, 1974). In temperate regions the initial stages of processing of these inputs of allochthonous organic matter occurs throughout the fall and winter via several mechanisms (Hynes, 1970; Kaushik and Hynes, 1971; Cummins, 1974). Coarse particulate organic matter (CPOM) such as leaves, twigs, branches, bark, needles, nuts, and fruits undergo abiotic leaching upon entering the stream and the majority of the soluble components are released within twenty-four hours (Nykvist, 1963; Kaushik and Hynes, 1971; Cummins, 1974, Petersen and Cummins, 1974). Other abiotic losses due to mechanical and physical processing of the CPOM as a result of the rigors of lotic conditions is estimated to account for approximately 5% of the total processing of this material (Cummins g£_§l, 1973). In a scheme suggested by Cummins (1973) CPOM is material greater than 1 mm and includes leaf litter and fragments of plants and animals. Fine particulate organic matter (FPOM) is less 2 than 1 mm and includes plant and animal fragments, fecal material, free microorganisms and floculated or precipitated dissolved organic matter (DOM). Concurrent with the abiotic processes the CPOM is colonized by fungi, protozoans, and bacteria. Colonization of CPOM by fungi (aquatic hypho- mycetes) and bacteria occurs normally within two weeks in streams in the temperate zone (Suberkropp and Klug, Kellogg Biological Station, pers. comm.) This colonization leads to increases in the nitrogen content of the detritus (Kaushik and Hynes, 1971; Iverson, 1973; Suberkropp g£_al, 1975). The major processing of detrital materials occurs through the actions of the microorganisms and invertebrates as summarized in the trophic relations scheme in Figure 1 (from Cummins, 1973). The two categories of organic particles (CPOM and FPOM) are processed in the stream by different categories of microorganisms and invertebrates. Fungi are credited with a major role in CPOM decomposition (Suberkropp and Klug, 1975) while bacteria are the primary decomposers of FPOM (Klug, pers. comm.) Aquatic animals, primarily aquatic insect larvae, have been divided into four categories according to their feeding behavior (Cummins, 1973): 1)shredders--animals consuming CPOM, 2)collectors--animals utilizing FPOM, 3)grazers--ingesters of periphyton, primarily diatoms, and 4)predators which utilize members of the other three feeding groups. Animals in each of the groups have morphological and behavioral adapta- tions which equip them for their roles of grazing algae, capturing insect prey or shredding or collecting detrital material. In Figure l Pteronarcys, Tipula, and gycnopsyche are shown as typical shredders, Stenonema and Simulium as collectors, Glossosoma as a grazer and Nigronia and the fish Cottus and Salmo as predators. Litter microbes are characterized by M D'IWLVIO ”HICULA'I m U" "I clam-c IA! YE. U0" w—r— m a"’ " 'v u .v' canvas 0 /omuuc an“ \ uocc0{nou\. J nucleus ........ nil-=2 l .--.-..o.. c. Q-.-- -n-----------o--ooo _. , ,. m mono-s Into-ton L Figure l. A simplified view of trophic relationships in a woodland stream community. Dashed arrows indicate less frequent exchange. 4 fungi (aquatic hyphomycetes) and fine particle microbes by bacteria. Deciduous leaves and photosynthesis by diatoms are utilized, together with soil runoff, as representative of energy inputs to the system. An apparent nutritional dependence by some of the insects on the microbial flora associated with the ingested leaf material, rather than the leaf material itself, has been noted (Kaushik, 1969; Wallace g£_al, 1970; Kostalos, 1971; Liston, 1972; MacKay and Kalff, 1973; Iverson, 1973; Barlocher and Kendrick, 1973a, 1973b, 1975). The insects presumably ingest the detrital material only to gain access to the colonizing microbes and fulfill their dietary needs at the expense of the microbes, while deriving little nutrition from the comparatively recalcitrant detrital material. A "peanut butter and cracker" analogy has been suggested by Cummins (1974) with the "peanut butter" being the microorganisms and the "cracker" the less nutritious leaf material. Alternatively, the insects may find the partially decomposed material more palatable (less tough) than non-colonized leaves. Aside from the insects' preference for colonized vs. non-colonized leaves microbial-insect interactions are thought to occur among aquatic insect larvae comparable to those found in other insects. Microbes have been observed to occur within the insect body, either in mycetomes (specialized cells harboring bacteria or yeast) or in the alimentary tract of the insect. Microbes within the mycetomes, termed endosymbionts, have been observed in a number of insects including hemipterans, heter- opterans, coleopterans, and orthOpterans (Buchner, 1965). Buchner noted a correlation between nutritional categories of insects and endosymbionts. Insect predators which live on a complete diet do not have endosymbionts, while insects whose diets are incomplete possess endosymbionts. The American cockroach benefits from growth factors synthesized by mycetomal 5 bacteria, which are lacking from the normal cockroach diet. Aposym- biotic cockroaches (individuals in which the bacteria have been eliminated via antibiotics or gnotobiotic rearing) require a rich diet and are raised with difficulty in the laboratory (Brooks and Richards, 1955). Mittler (1971) has shown that individual omission of the ten essential amino acids from diets of aphids deprived of their endosymbionts substantially reduced the growth of these aphids compared to aphids possessing symbionts. For those aphids possessing endosymbionts the only "essential" amino acids were histidine, lysine, isoleucine and methionine. A well documented example of a mutualistic symbiosis (favorable and obligatory for both symbionts) between an insect and the microbiota of its alimentary tract occurs in termites. Protozoa in the termite paunch digest cellulose to organic acids (primarily acetate) which are utilized by the termite for energy (Honigberg, 1970). Further, Breznak g£_gl (1973) and Beneman (1973) have demonstrated nitrogen fixation in termites which has been attributed to the gut microbiota. This source of fixed nitrogen is presumed to be vital to the termite since a diet of wood is deficient in combined nitrogen as evidenced by a high C/N ratio. Alimentary tract insect-microbe interactions are thought to also occur in larval stages of aquatic insects as evidenced by a dense, morphologically diverse bacteria population associated with the hindgut of larvae of the aquatic cranefly Tipula abdominalis (Klug and Kotarski, 1974). This microbiota was observed in the lumen and firmly adhering to the gut wall. The nutrient-poor characteristics of the detrital material ingested by the cranefly and other shredders and collectors suggest a possible insect-microbe gut symbiosis analogous to Buchner's observation for insects possessing mycetomes, i.e. a correlation between feeding behavior of an insect and alimentary tract microbiota similiar to the correlation between feeding behavior and presence or absence of a mycetome. The importance of the gut microbiota to the insect and to the processing of detritus in the stream may be minimal or vital depending upon the nature of the ingested food and the enzymatic capabilities of the host. A range of symbioses including mutualistic, protocooperative (favorable to both symbionts, but not obligatory), commensalistic (favorable to the commensal and the host not affected), or parasitic relationships could occur between an insect and members of the gut microbiota. For example, if the insect produced a wide range of polymer degrading enzymes, or possessed the behavioral and morphological adapta- tions so that a complete diet was available to a somewhat limited set of enzymes produced by the insect, a gut microbiota would be superfluous. Conversely, an insect consuming fiberous material low in nutrition and lacking the enzymes necessary to digest the various polymers would benefit from.a microbial population with the capacity to hydrolyze the complex dietary constitutents. Other possibilities between these extremes are that the insect gut absorbs growth factors, essential amino acids, vitamins, or other small moleucles such as fatty acids, produced by microbial symbionts. The last set of possibilities is probably more frequent among insects than the two extremes of the continuum and certainly more difficult to characterize. If the gut microbiota is superfluous or only marginally beneficial to the harboring insect the microbiota may be playing a role in the further conditioning and preparing of the detritus for the next trophic level. The biota of a shredder, for example, may not benefit the shredder, but may be a vital conditioning requirement for ingestion of shredder fecal material by the collectors; thus playing a relatively more 7 important role in processing allochthonous input to the stream than in providing nutrients to the host insect. Shredders and collectors consume 0.6% to 130% of their body dry weight/day (Cummins §£_gl, 1973). Welch (1968) reported that for detritus feeders as much as 80% of the ingested food is execreted as feces. The ingested food is presumably modified as it passes through the gut with the less resistant compounds being assimilated by the insect. The material execreted as feces is thought to be more recalcitrant and less nutritious than that ingested. However, due to the associated bacteria the feces may be considerably more palatable and nutritious to the collectors than if the bacteria were absent. The role of the bacteria associated with this FPOM may be analogous to that of the hyphomycetes associated with the CPOM, acting as the "peanut butter" to induce collectors to ingest the FPOM. Although particles in feces have attached bacteria, this possibility is not as attractive as it might seem because feces of several aquatic insects have been observed to disperse immediately upon defecation, releasing free bacteria and detrital particles. The integration of these considerations concerning the role of gut ‘microbiota into the scheme of trophic relations in a woodland stream is not possible, without more information concerning the presence or absence of a gut microbiota in aquatic insects aside from the cranefly. To provide this information a survey of aquatic insect larvae was under- taken to assess whether a gut microbiota is widespread among aquatic insect larvae and if the presence of a microbiota is correlated with the feeding behavior of the insects examined. The basic hypothesis being that a gut microbiota would frequently be observed among animals consuming detritus (shredders and collectors) and absent in animals utilizing more nutritious substrates (predators and grazers). To test this hypothesis selected members of the various feeding categories were dissected and their alimentary tracts examined with phase microscopy. Gut morphology, presence or absence of significant numbers of bacteria, their location in the guts, and whether the population was lumen or gut wall associated were noted. An estimate of the numbers present was determined and light and electron photomicrographs of typical examples I of the microbiota obtained. MATERIALS AND METHODS Collection and Maintenance of Insects Most of the insect larvae examined were collected from Augusta Creek, a small woodland stream in Kalamazoo and Barry counties, Michigan. Larvae of insects were maintained in containers of water and detritus, stones colonized with algae, or insect prey depending on the feeding category of the insect to be examined. In this way the larvae had available a fairly normal diet prior to death and dissection. The chambers were aerated via compressed air forced through an aquarium bubbling stone. Temperature was held at 5-100 C for periods of several days for most larvae to several months for the larval craneflies. Four species from streams in the Cascade Mountains were kindly provided by members of Dr. James Sedell's laboratory at Oregon State University. Selection of insects was based upon availability of information concerning their feeding habits, ease of collection and potential interest as an unusual or typical insect. A crustacean, Gammarus, and a mollusk, Goniobassis, ‘were also included in the survey because of their frequent occurrence in Augusta Creek and their similiar feeding behavior to some insects.‘ Light Microscopyiand Enumeration of Biota Larvae were killed by immersion in boiling water for approximately five seconds or by decapitation and examined using a variety of micro— scopic techniques. After dissection in 0.1 M phosphate buffer, pH 7, 9 10 gut morphology was traced using a camera lucida attachment on the dissecting microscope. Wet mount preparations of lumen contents and washed gut wall were made by excising portions of the gut (foregut, midgut, pylorus, and rectum) and slicing the resulting cylindrical tissue longitudinally to obtain a flat preparation. The lumen contents removed during this process were collected in a drop of buffer on a microscope slide. The gut wall was vortexed vigorously for 10-20 seconds in 0.1 M phosphate buffer, placed on a slide, and the surrounding muscle teased away. The resulting tissue was examined for adhering microorganisms with phase microscopy. Numbers of bacteria in the wet mounts of lumen contents were deter- mined subjectively. More than fifty bacteria/field at 1000X was categorized as +2, 1-50 bacteria/field as +1, and less than one bacterium/ field as 0. Subsequently the Petroff-Hauser counter was used on one- half of the larvae initially examined to provide further documentation of the subjective observations. For enumeration of gut microflora larvae were dissected and midgut and hindgut dimensions were recorded. Midguts and hindguts were macerated in a tissue grinder in phosphate buffered 4% formalin solution. With the larger larvae it was possible to macerate individual midguts and hindguts. For analysis of the smaller larvae several midguts and hindguts were pooled. Pieces of gut wall and detrital particles in the gut were observed to settle within a minute after maceration and were not disturbed when samples of suspension were placed in the Petroff-Hauser chamber and counted at 640x. The suspension was agitated and large particles allowed to resettle between successive samples. Three to six samples of the suspension were counted, each count including three to twenty-five fields (twenty-five fields equals volume of Petroff-Hauser counter) depending upon bacterial density. The gut 11 was presumed to be a cylinder and volume of the gut was calculated from the gut measurements. Average numbers of bacteria per midgut and hindgut were calculated from the Petroff-Hauser counts and numbers/ml of gut were calculated. The wet mount procedure was used for all invertebrates examined (Tables 1 and 4) and Petroff-Hauser counts on invertebrates listed in Tables 2 and 3. Since bacteria populations were, to some extent, similiar among insects possessing biota and due to time constraints phase, light, and electron micrographs presented in the "Results" are of microbiota from fourth instar larvae of the cranefly Tipula. This was chosen for photography since it was large and easier to manipulate than smaller larvae, possessed greater morphological diversity of bacteria than some of the smaller larvae, and was readily available. For phase photography of lumen populations contents from Tipula gut were suspended in phosphate buffer or 0.75% sodium chloride solution and drops of the suspension placed on agar slides. Agar slides were made by dripping a sterile 1.5% agar solution on microscope slides, allowing it to spread out, wiping off excess agar, and allowing the remainder to solidify on the slide. Sections 111m thick were cut on an LKB microtome from epon-embedded alimentary tract tissue (described below) and placed in drops of 10% ethanol on microscope slides. Excess ethanol was blotted off and the remainder evaporated over a hot plate causing the sections to adhere to the slide. Several drops of 1% toluidine blue and 1% sodium borate (1:8) were added with the slide still on the hotplate. The hotplate was allowed to heat until the edges of the staining solution were dry. The staining solution was washed off with distilled water, excess water blotted and 12 remaining water near the sections allowed to air dry. The thick sections were photographed using light microscopy. All photographs were taken with Kodak Panatomic or Plus-X-Pan film on a Zeiss Universal Microscope equipped with a 35 mm camera. Electron Microscopy Preparations of lumen bacteria for electron microscopy were made by dissecting larvae of the cranefly Tipula in 0.75% saline. Portions of the hindgut were placed in test tubes, sliced longitudinally, and vortexed in 1% glutaraldehyde in 0.1 M phosphate buffer for thirty minutes at room temperature. The cells were washed in distilled water to remove the glutaraldehyde and placed on 200 mesh formvar coated grids. The grids were shadowed with platinumrpalladium (80:20) in a Kinney Vacuum Evaporator prior to viewing in the electron microscope. Larvae for embedding in epon were decapitated and dissected in 3% cold glutaraldehyde in 0.1 M phosphate buffer. Guts were perfused with 6% purified buffered glutaraldehyde (Electron Microsc0py Sciences) to insure immediate contact with the fixative. The tissue was fixed overnight at 40 C, washed in phosphate buffer, post-fixed in 1% osmium tetroxide in S-collidine buffer and embedded in epon according to the procedures of Luft (1961). Silver or gold sections were cut with glass knives and placed on grids. Sections were stained with aqueous 5% uranyl acetate and lead citrate according to Reynolds (1963). A Siemens Elmiskop 1a at 80 RV was used to examine the grids. RESULTS Gut Morphology The alimentary canal in insects is divided into three main regions: the foregut, derived from the embryonic ectoderm, the midgut which is endodermal in origin and the ectodermally-derived hindgut. Depending upon the morphological complexity of the insect these regions may be further divided: the foregut into the oesophagus, crop and proventriculus; the midgut is comprised of the gastric cecae and ventriculus; and the hindgut includes the pylorus, ileum, and rectum (Chapman, 1971; Wigglesworth, 1974). In the insects examined boundaries of the foregut, midgut, and hindgut were usually fairly distinct; the subdivisions of these regions were often less discernible or absent. A tissue change and often a color change was observable where the foregut and midgut connected. The hindgut was separated from the midgut at the point where the Malpighian tubules intersect the alimentary tract. The majority of the insect guts examined exhibited morphological similarity as shown in Table l. The simple gut possessed no gastric cecae, proventriculus, convolutions or enlarged fermentation chamber. Small variations of the simple gut occurred and correlated with the phylogeny of the insect; for example the three mayflies were more similar to each other than to any of the others in this group. The simple gut shown (Figure 2) is a tracing of Hydatophylax hesperus and is characteristic of the limnephilids and filipalpians. The remaining 13 14 trichopterans and elmid beetle had a longer, narrower, and slightly curved ileum. Ephemeropterans possessed a wider ileum and the rectum was surrounded by a prominent layer of muscles. A few insects exhibited more complex morphology including a proventriculus in the foregut, gastric cecae in the midgut, or a convolu- tion or fermentation chamber in the hindgut (Table 1). The diagram of one of the more complex guts is that of the megalopteran Nigronia. Two other guts exhibited an intermediate complexity between the simple gut and the more complex megalopteran gut shown. The pylorus and ileum of the blackfly Prosimulium*were not well defined, forming a 100p that lead to an enlarged rectum. The caddisfly Hydropsyche possessed a straight tube gut with the addition of a proventriculus. The cranefly Tipula was the only insect with a truly enlarged fermentation chamber in the hindgut (Figure 2). In the majority of insects the midgut was the most voluminous and prominent of the three regions, frequently larger than the hindgut by a factor of ten. In the cranefly and megalopterans these proportions were reversed, though in Tipula the hindgut was only two to three times larger than the midgut (Table 2). Hindguts of all of the insects examined were fairly durable tissues and could be manipulated for observation as described previously. The midgut of the majority of insects was fragile and could not be handled in this manner. It was possible to make squash mounts of midgut and observe fragments of wall, but not to make a clean preparation of wall. The four exceptions-~the chironomids and simulid, possessed a midgut wall that consisted of a tough, transparent membrane (Figure 7). Due to the small size of these guts slicing the cylindrical midgut longitudinally to aquire a flat tissue was not achieved. Instead, the contents could 15 Table l. Gut morphologies and taxonomy of invertebrates SIMPLE GUT Hexagenia limbata (Ephemeroptera, Ephemeridae) Stenonema spp. (Ephemeroptera, Heptageneidae) Leptophlebia nebulosa (Ephemeroptera, Leptophlebiidae) Pteronarcysgpictetii (Plecoptera, Filipalpia, Pteronarcidae) Pteronarcys sp. (Plecoptera, Filipalpia, Pteronarcidae)* Taeniopteryxgparvula (Plecoptera, Filipalpia, Taeniopterygidae) Lara sp. (Coleoptera, Elmidae)* Brillia flavifrons (Diptera, Chironomidae, Orthocladiinae) Brillia sp. (Diptera, Chironomidae, Orthocladiinae) Stictochironomus annulicrus (Diptera, Chironomidae, Chironominae) Platycentropus radiata (Trichoptera, Limnephilidae) Pycnopsyche guttifer (Trichoptera, Limnephilidae) Hydatophylax argus (Trichoptera, Limnephilidae) fiydatophylax hesperus (Trichoptera, Limnephilidae)* Lepidostoma costalis (Trichoptera, Lepidostomatidae) Heteroplectron sp. (Trichoptera, Calamoceratidae)* Brachycentrus occidentalis (Trichoptera, Brachycentridae) Glossosoma nigrior (Trichoptera, Glossosomatidae) CONVOLUTION Prosimulium sp. (Diptera, Simuliidae) PROVENTRICULUS Hydropsyche bronta (Trichoptera, Hydropsychiidae) PROVENTRICULUS + GASTRIC CECAE + CONVOLUTION Paragnetina sp. (Plecoptera, Setipalpia, Perlidae) Corydalus Sp, (Megaloptera, Corydalidae) Nigronia serricornis (Megaloptera, Corydalidae) GASTRIC CECAE + FERMENTATION CHAMBER Tipula abdominalis (Diptera, Tipulidae) NON-DIFFERENTIATED CUTS Goniobassis sp. (Mollusca, Gastropoda) Gammarus pseudolimnaeus (Crustacea, Amphipoda) * species from Oregon 16 0mm Hmuomu--mu mood Hmuoouuuau Bauumuuuu mononduuha moasowuuco>ouaunua afloaauum mmooo uauummwuum .uooa uswvcas van uswvwa mSu ouog3 mega man um nomuum moasnsu amflswmaamz .coflumuwcfiuaso ouocov cu mafia m>mon m saws voawauso mum musmvaan can muswouom .uonamno coHumucoEumw m saws usw memEOo w .mwamcfieovnm manage "unwwm .zononduoa usw xoaaaoo ouoa m .uaw mwcuoowuuom mwcousz "umucou .xwoaosauoa usw oamEam map mo oaumwuouomumno .usw monommoa xmamzm0uwvmm "umoq vo>ummpo mowwo~onduoa usw mo moaaamxm .N unawam 17 \_\—’/A\' \, 0‘ ‘\\\VI ..,....nuuu\\\ ‘, VT“ 1 \ ’7 ‘ ' / 'C ‘1 \ . / >~ Q A \\\\\ ‘ I- L -. . ..,,.....uu \\\\\\\‘ (y \l l! ‘ MA Figure 2. 18 be extruded from the tubular gut by applying pressure to the exterior of the gut with tweezers, thus obtaining preparations of contents and gut wall (Figure 7). Neither the amphipod Gammarus or the snail Goniobassis possessed the foregut, midgut, and hindgut divisions of the alimentary tract that is characteristic of insects. Alimentary tracts of both were undiffer- entiated, consisting of fairly fragile membranes surrounding the ingested food. Gut Microbiota Two populations of organisms were distinguished in the insects examined-~lumen and wall-associated organisms. Lumen organisms were defined to be those organisms removed when the gut wall was sliced longitudinally and vortexed in buffer. (The majority of the lumen organisms were removed when the contents were collected by longitudinally slicing the gut wall and allowing the contents to fall in a drop of buffer on a glass slide.) The wall organisms remained attached to the tissue in spite of the vigorous washing. For the Petroff-Hauser counts the lumen population was assumed to consist of those organisms that were suspended as a result of the tissue grinding procedure. Only rarely were fragments of typical wall-attached organisms observed in the lumen preparations. The predominent lumen bacteria in all guts were flagellated rods ranging in length from 1 pm to 30 pm with the majority 5 pm long or smaller. In regions of guts with 105-107 cells/ml only small rods, approximately 1-3 pm long were observed. Guts with greater numbers possessed greater diversity including larger sporulating and non- sporulating rods, spiral-shaped organisms, and more unusual morphologies. 19 Morphological diversity of the lumen biota varied slightly from insect species to species and between individual members of the same species &; shown in the percentages of bacterial morphological types in Table 3. Morphological diversity and flagellation among members of the lumen population of the cranefly Tipula are shown in Figure 3. Similar or somewhat less diversity was observed in wet mount preparations of hindgut lumen contents of the other obligate shredders, collectors, and predators that possessed a dense microbiota. Thick sections of non—washed epon-embedded gut wall from the cranefly Tipula (Figure 4) revealed dense populations close to the wall and some- what less dense populations and detrital particles in the lumen. The gut is surrounded by two layers of muscles. A total of three nuclei of gut wall cells can be seen in the two photomicrographs. Thick sections of the caddisflies Pycnopsyche and Hydropsyche revealed a similiar situation to that shown in the cranefly. Thin sections of washed cranefly gut wall revealed bacteria surrounded by amorphous material in close proximity to the wall (Figure 4). The wall consists of a cuticle adjacent to the bacteria. Parallel infoldings of the cell membrane with mitochondria between occupy approximately one-half of the cell. The nucleus and nuclear membrane are prominent above an electron-transparent area and the basement membrane. Preliminary electron microscopy of thin sections of the caddisfly Pycnopsyche hindgut revealed a similiar association. The most distinctive members of the wall populations were filamentous bacteria. Phase and electron microscopy of these organisms (Figure 5) revealed an end-on attachment to the wall and extension several hundred um into the lumen. At the proximal end of the filaments few septa were observed. At the distal end of the filaments numerous septa and t I o s‘: 1 . , f»; ‘ "‘4'3. Figure 3. Lumen bacteria from Tipula. Upper: Agar slide photomicrograph. 1300X Lower: Shadowed electronmicrographs. Left: 5280K Right: 5320X 21 Figure 4. Tipula rectal sac and bacteria. Upper: Epon-embedded thick sections of non-washed rectal rectal sac and lumen contents. Left: 310x Right: 780x Lower: Electronmicrograph of washed rectal sac. 11,550X 7,2 we. .. .5 Figure 4. Figure 5. 23 Bacterial filaments from Tipula. Upper left: Upper right: Lower: Wet mount of sporulating filamentous bacteria and gut wall near juncture of the ileum, rectal sac, and rectal lobe. Masses of shorter rods are against the wall. 800x Agar slide preparation of the distal ends of the filamentous bacteria with spores and in the presporulation stage following septum formation. 1300X Electronmicrograph of the filaments and cuticle of gut wall in the region of the ileum, sac, and.lobe juncture. 11,550X Figure 5. 25 inclusion bodies, presumably spores, occur. During the sporulation process the cells appear to shorten somewhat as evidenced in the agar slide preparation (Figure 5) where cells without spores are slightly longer and not as wide as those with spores. Distal portions of filamen- tous bacteria in the lumen can also be observed in Figure 4 with the attachment site presumably being outside the sectioned area. In the cranefly Tipula the bacterial filaments were most dense at the juncture of the ileum with the rectal sac and rectal lobe. However, this localization was not observed in every individual. In some indi- viduals these organisms were observed to colonize the rectal sac in densities similiar to that at the ileum-sac-lobe juncture. In the trichopterans the filamentous bacteria were localized in the posterior half of the pylorus or the anterior portion of the rectum, Relative densities of filaments in these two regions varied from individual to individual. Location of the mass of filaments could frequently be observed immediately after opening the body cavity and exposing the alimentary tract. The mass of filaments would impart a yellowish tinge while the rest of the gut was brown due to the detrital food material. Under the dissecting microscope (120x or 250x) the filaments appeared as fine wispy strands protruding from.the gut wall. Localization of bacteria in a portion of the gut was particularly evident in the wood boring midge Brillia. A very prominent clump of sporulating rods (approximately 8 X 1.1m) occurred just anterior to the juncture of the Malpighian tubes with the alimentary tract. This mass of rods was observable as a light brown clump through the body of the insect at 120x. Individuals were observed to defecate and the hindgut and posterior of the midgut were emptied without the removal of the 26 mass. Phase microscopy of the feces revealed woody particles and less than one bacterium/field. Apparently, the mass of rods inscribes the midgut wall and is firmly attached. Clumping of the rods was observed and difficulties encountered when attempts were made to spread the bacteria for observation on agar slides. That the bacteria were not removed with the feces together with the clumping implies that the rods form a "donut" inside the midgut and adhere while allowing the feces to pass through the ”hole" of the "donut." Not only were specific members of the flora localized (the filaments of some shredders and collectors and the mass of rods in the wood borer), but frequently the entire flora was localized in the anterior portion of the rectum or in the pylorus. Due to the size of the washed Tipula gut wall the most notable example of localization of the entire gut flora to a region of the hindgut was observed by Klug and Kotarski (1974). They described a "line of demarcation" just anterior to the rectum, observable macroscopically on a washed gut wall. The wall had a fuzzy white appear- ance in colonized areas and was clear and transparent at the posterior end. The shredding caddisflies also possessed a distinct line between colonized and non-colonized regions. Prosthecate bacteria, shown in Figure 6, possess appendages (prosthe- cae) approximately 0.2-0.7 um X 0.05-0.12 pm and the longest dimension of the cell (excluding prosthecae) is approximately 0.6-1.31rm across. Prosthecae of the cells are surrounded by the bacterial cell wall (lower electronmicrographs, Figure 6). Some morphological variability occurs among the prosthecate bacteria and many of them contain vesicles. They are embedded in a somewhat more electron-dense material than the other bacteria in the gut. Prosthecates were observed in thin sections of gut wall with bacteria in the rectal lobe, rectal sac, or ileum-sac- 27 lobe juncture in six different cranefly larvae. Occurrence was fairly rare, but when they were observed they were in groups very often within ten um.of the gut wall. Prosthecates have also been noted by Klug and Kotarski (1974) in scanning electron micrographs of Tipula gut wall. Trichomycetes, a poorly understood class of fungi were observed in the midguts of three dipterans and the hindgut of a mayfly (Table 4). After removal of portions of detritus and bacteria from.the midguts of the dipterans the trichomycetes could be observed through the intact transparent midgut wall (Figure 7). Observation of trichomycetes in the mayfly hindgut was by the usual method of slicing open the gut. Coloni- zation of larvae by trichomycetes varied with time. The mayfly, for example, was well colonized in some collections and poorly colonized later in the spring. Degree of colonization also varied between individ- uals collected on the same day. Asexual reproductive structures, trichospores, were characteristic of Stachylina or Harpella, members of the Harpellales, the only order of trichomycetes known to conjugate. occassionally conjugation of hyphae was observed in the blackfly Prosimulium and the midge Stictochironomos. The subjectively determined categories of +2, +1, and 0 were found to correspond to 109-1010, 108, and 105-107 bacteria/ml of gut respectively in the Petroff-Hauser counting procedure for the insects that were counted. Half of the insects were examined by both methods and half received subjective examination only. Due to the correlation noted above between the subjective and Petroff-Hauser examinations it is reasonable to extrapolate from the insects that were counted in the Petroff-Hauser chamber to those not counted as was done in Table 4. Tables 2 and 3 list Figure 6. 28 Prosthecate bacteria from Tipula. Upper: Prosthecates in close proximity to the gut wall. 16,800X Lower left: 24,800X Lower right: 42,750X Figure 6. '- . .‘ "a“? mi“ '0' s 9 ..". . {h I: . I» d -‘l‘ I 4, (as? v , . r. . _ .u w, " l‘i‘ . ‘ ’1 . u'35,%;n_! ,. A o f I '3" w‘ a" 30 Figure 7. Trichomycete hyphae with immature spores, bacteria and detritus visible through the intact midgut wall of the blackfly Prosimulium. 31 :oflumcwauoumv vmaooa onu mo omnuo>m ago mu usw\muona:a .mucoawuammoa oumumdmm mo ammuo>m onu a“ oasao> mam wouasoo can coaooa who? musw uofiawam .haamsva>wvafi voussoo mums musm ammumfi « Bea x w.~ sod x o.s -OA x e.a m “amass 50H x w.s moH x m.H m cs x N.m n Sumac“: maamumou «soumoeaama ¢NI “OH x o.H OH x o.a ~-oH x ~.m a Roi x A.N moi x m.h ~-o~ x m.m a “amuse Ones x o.a as x N.H ~-oH x N.H A sea x ~.m mes x o.m N-Oa x H.s H Oz x o.n ca x o.m N-0a x o.a a sea x A.“ NOS x ~.a -OA x ~.a a owes x o.a Mes x o.m M OH x ~.m a “amass; ammauusmxmeoswmocomm oH x ~.m cm x m.s a-oa x s.a a mop x s.a MOS x ~.H p-0a x ~.H a “amass Nos x s.N mes x n.a -oH x m.o H on x A.m ace x m.m m-0a x s.m a mstoasa oH x w.~ mes x N.H H-0H x s.m a owes x s.a mes x ~.s A-oH x m.m a mos x s.n OH x o.a H oH x m.a a oHoH x o.~ MOS x o.w AHoH x o.s H Ausweaae mes x s.a OH x s.H H-0a x o.a a scauoumoaV ace x N.~ mes x N.H «-0s x m.m a mass as asses masseuseenm magmas muCMUCOU u—JO H0 H5 USU AHEV ®§HO> *m.—.QEH§< COfiuGUOQ flmBHC< \guouomm .oz \mmumuomm .oz 35 mo .02 moumupouuo>cH wouooamw CH mwumuomn mo muasoo panamznmmouuom was mfimecoEflv uau .N ofinme 32 coaumcwaumuov poaooa ecu mo owwuo>w ecu ma u:w\muonasa .muaoamHSmmoa ouwumaom mo omnuo>w osu ma oabao> usw mounsoo can voaooa mum? muaw umaamam .maamsvfi>«vafi voucsoo mum? musm ummuma % ca x o.a sea x m.H s-oH x m.s A Wee x m.s mod x m.~ s as x m.~ ca “amass ca x N.~ oH x m.a n-0a x m.o A manuaasacu owes x m.H was x m.~ n-0a x m.H ca usmucae moaoaouae00uoaum as x o.~ as x m.m OH x ~.m m unwaaa maamuamaaooo Mod x o.m Mes x m.s “was x H.n m “omega: mauucmoseomum sea x m.a OH x m.m -oA x a.~ s unwoaa sea x w.a mes x ~.H w-o~ x o.m s “amends mumaaaa swammem: Nod x s.a OH x m.~ m-oH x m.a 0 spaces egos x A.m Mes x m.H s as x w.a a Suwanee mucosa meosmmauese sou x ~.A oa x o.m s-oH x m.~ a uawaaa saucy as x s.m mmoa x s.~ s-oH x 0.5 a “canvass--u=meaa mes x ~.m mes x «.5 ca x m.m a magma m n: mfiuouomnunusmvaa mod x m.m sea x s.a m-oH x n.m a “swans; .am maaaaam neg x w.m qoa x w.m muoH x o.~ m uomuu usm Hauou mammaawaovsomm.msumaamu cod x o.m mes x m.H cm x a.s a unweaa Nos x m.o mes x ~.m MHOA x ~.m a “sweep; auumuoaa msoamaoumum mucoucou usu mo Ha uso AHEVoEDHo> . *mamaa:< cowumooq amawc< \mwuouomm .oz \mfiuouomm .02 use mo .02 Au.uaoov N manna N" cowumcmauoumv cmaooa msu mo owmuo>m on» ma u=w\muonasa .muamamHSmmoE mumumaom mo ammuo>m on» ma oasio> usm wouasoo was coaoom ouo3 musw Headmam .maamsvw>av:« voucsoo mums muam nowuma a sea x 0.0 mes x o n as x 0.x a ace x m.s mes x N m wHoH x a.“ a “amuse; .Am massesuou OH x H.N cm x m A OD x H.~ H unmeas owes x H.H mes x m H mHoH x o.s a Suwanee macuouauuwm «Heatwaz oH x s.e sea x o H .OH x s.N A “amass .u mes x o.a Goa x s a N oH x s.a A somecu; Hoamwa: «sesammoau muGOUSOU USU HO H5 USU QESHO> *mHmEHS< GOHquOA Hmefi< \mfiuouomm .oz \mwuouumm .02 use «0 .oz Au.uaoov N magma 34 voucdoo mam voHooa one? munw uoHHmEm .mHHmsvH>chH voucsoo oum3 musw HmemH « ooH H uawvHa HHumuuHm ooH H uswvcHn mhdumcououm 00H m unwvHa mHHmumoo 00H m uswvaHz «BeumodemH ooH H ooH H uswvHa 0.0 v.0 m.~ m.om H m.o 5.0 ¢.H o.wm H 0.0 N q.N mm H q.o o.mm H ooH H unwvaHn HoMHuunm.o;ommdoco>m ooH H ooH H HaweHa H mm H ooH H msuona N oH mm H m a so H ¢.m o.H m ow H N.H m.H m.w mm H AuswvcHg m.H m.H a.m om H uoHHoumoav m m 0H an H macs ou asoHH mHHmcHEownm mHSQHH mEchmeo mucoEmHHm mouoam mvom N *mHmEHc< :oHumOOH HmEHc< amsmam N HHHz Ho .oz -HmHHam x mace x mucsoo unnammummouuom cH vo>uomno «Huouomn mo monoHosmuoz .m oHnme 35 N.N N.0 0.00 H a 0.0 H.00 H uswvcHs .am musvhuoo 00H H unwuHB mHauooHunm m N 00 H usmccHs chousz 00H m uswvHE uoHuch 00H N usmvaHs mBOmOmmoHo N m 00 N N N 0» 0H ustuHa H 00 N msuoHHsacm m N0 0H usmvcHs msaocouH£UOuoHum 00H 0 unwvHB mHHmucmvHooo 00H m uswvcH: mHuucoomnomHm 00H 0 uswvHE mumAEHH 00H 0 uswvcH: wHCowwxm: 00H 0 uawvHa mucoun m.0 0.H m.0 N0 0 uswvaHn onommmouvmm 50 *23 0.3 m ”3038 H33 m mm as m uswvHa HoHuoucm 0.00 0.0m m dasHo mHHouomnuuuswvHE 00H m uswvcHL .mm mHHHHHm «somcaHHovaomm 00H m uomuu usw Hmuou nonmaamw mBchmwuo mucmEmHHm monomm mvom N mHmch< coHumUOH HweHc< moamnm N nqu mo .02 -HmuHam N avom N Hu.uc000 m mHHmH 36 the insects which were examined with the Petroff-Hauser counting proce- dure, the numbers of bacteria computed, and percentages of different morphologies observed. A summary of the feeding categories of the invertebrates examined and the gut microbiota observed is presented in Table 4. The shredder feeding category was subdivided on the basis of the observed or presumed feeding habits of the invertebrates (as explained in the "Feeding cate- gories" section). The obligate shredders possessed an abundant microbiota in the lumen (109-1010/m1 of hindgut lumen) and associated with the hindgut wall, however lacked a significant population in the midgut (105-107/ml of midgut lumen). Facultative shredders and grazers had sparse populations of bacteria in both the midgut and hindgut (105- 107/m1). More diversity in localization and types of associated microbiota was exhibited in the collector category. Some members (Hydropsyche, Stictochironomos, Hexagenia, and Stenonema) possessed a hindgut population similiar to the hindgut biota in the obligate shredders. The caddisfly Brachycentris and the blackfly Prosimulium possessed a 8 7 less dense (10 /ml) and scant (105-10 /ml) hindgut population respectively. The blackfly and a midge, Stictochironomos, possessed a midgut biota that included spiral-shaped organisms and trichomycetes in addition to the rod-shaped bacteria commonly observed in the hindguts of the obligate shredders and collectors. The mayfly Leptophlebia possessed the usual rod-shaped and filamentous bacteria and was the only insect observed to harbor a hindgut population of trichomycetes. Of the two wood borers examined, the wood tunneling midge Brillia possessed a dense midgut population including spiral-shaped organisms and the localized rods 10 8 (109-10 /ml) and a less dense hindgut population (10 /ml). The 37 remaining wood borer, Lara, contained a meager population in both midgut and hindgut. Of the predators, the megalopterans possessed a dense hindgut lumen population (109-1010/ml) and a scarce midgut population (105-107/m1). The stonefly Paragnetina contained sparce populations in 5 7 both midgut and hindgut (10 -10 /ml). 38 Euflw mo HE\mHu0uown I OH. oH 0H s usw mo Ha\wHuouoan 0H uswvaH: was uawvHavoochov voxumH mouoo%EOSUHHH machwwuo vodmsmuHmHHam itll-IUJE-‘m 0 mucoamHHm HmHuouumm I as» Ho 3:238; B3....2 303 3532 I I meHMH>mHM MHHHHHm I l 93>qu iuouhoHcmmH l I .90. mhlouchumuh I l HHumuuHm mNuumcoumuh mamanmmmm M>HHcH mo muOHnouoHa usw can moHHowoumo wsHuooh .0 oHan 39 [-i (1) E4 (I) cu l l | m U) WI H.hu¥ amoHsnoc mHnoHsmouhoH huh .mmm mascocoum mus oumnaHH mHmwwwxom h-« mauoHHsccm masoc0uH£oouUHum .mm BSHHsaHmoum mHHmucovHuoo msuucoohnomum huh mucoun onohmmouvhm mmoeomgqoo * .mm «HHHHHm .mm mumH 00550 9003 msochHHovammh.msumEEmu Ame couuoonououom Hu.ucoov mamaommmm m>HH