I“ I alwx .Iv. -b“‘l - ~r V um,‘ - 3m if; m :8! V- :g . N “ u't ' ‘I ifwfifizléfiim. ‘-:~§‘:: ’3 9:3? Wu 1 i n.21u... Lu’ 4 V: :‘uffi E,“ «3:522 ' 3a "£3 “5* ' h L L‘Q . . ‘ . _“. 4... .4d ' . .» ~ .J:.~ 91.9% «7 ‘ v I " l v ..'| I." '3 ” $314? :mg !‘ 'zkw out» -‘ w- 5 v: .t'. 11:3" =2 ‘ 4 '4 5. .. .. . ,~_.v. “ 'v. “1 .r‘ah " ,, ,gmrh .IVG‘IQI ‘Il‘lh. 5 I :r '2 L' “l. ,...4 11 w 51;» l .3 intuit» Milliflillliiflljiljiil 312 This is to certify that the dissertation entitled Description of a new species of Cytophaga and characterization of its xylan-degrading enzyme system presented by Sheridan Kidd Haack has been accepted towards fulfillment of the requirements for Doctor of Philosgahy degreein Microbiology Date February 3, 1992 MS U is an Affirmative Action/Equal Opportunity Institution 0- 12771 LIBRARY Michlgan State . Unlvenlty PLACE IN RETURN BOX to remove thlo checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE L_ MSU It An Affirmative Action/Equal Opportunity Institution cum nus-n1 DESCRIPTION OF A NEW SPECIES OF CYTOPHAGA AND CHARACTERIZATION OF ITS XYLAN-DEGRADING ENZYME SYSTEM BY Sheridan Kidd Haack A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Microbiology 1992 672- 3504/ ABSTRACT DESCRIPTION OF A NEW SPECIES OF CYTOPHAGA AND CHARACTERIZATION OF ITS XYLAN‘DEGRADING ENZYME SYSTEM BY Sheridan Kidd Haack Xylans are a class of cell wall polysaccharide widely distributed among both terrestrial and aquatic plants; however, our understanding of the microbiology and biochemistry of xylan degradation, especially in anaerobic habitats, is rather limited. Research described herein investigated the taxonomy and physiology of a new species of anaerobic bacterium in the genus Cytophaga and characterized the xylan-degrading enzyme system of the new isolate. Isolates of gliding bacteria that attached in masses to, and dominated the fermentation of, xylan powder in methanogenic and sulfidogenic enrichments from freshwater sediments were all Gram negative, slender rods that formed no endospores, microcysts or fruiting bodies. Representative strain XM3 was a mesophilic, aeroduric anaerobe that grew by fermentation of mono-, di-, or polysaccharides (but not cellulose) in a mineral medium producing acetate, succinate, propionate, C02, and H2. strain XM3 possessed sulphonolipids and carotenoid, but not flexirubin, pigments. Its total cellular fatty acids were dominated by C15:O anteiso (75%), n (13%), and iso (2%) isomers. Strain XM3 had 45.5 mol% G + C in its DNA, and partial sequencing of its 16S rRNA placed XM3 within the Bacteroides-Flavobacterium phylogenetic group. strain xu3 was proposed as the type strain of the new species Cytophaga xylanolytica. Similar strains were also isolated from marine sediments. The xylanase activity of Cytophaga xylanolytica strain XM3 was mainly (90-95%) cell-associated; the remainder was secreted into the culture fluid. Isoelectric focusing of Triton extracts of whole cells, followed by activity stains, suggested that xylose- and xylan- grown cells had common endoxylanaee (pI 4.3), arabinofuranosidase (pI 5.85), and xylosidase (pI 4.6) (iso)enzymee. However, xylan-grown cells produced additional xylanase activity bands. Supernatants of both xylose- or xylan—grown cultures contained a similar arabinofuranosidase, but neither supernatant contained a xylosidase. Both supernatants also exhibited xylanase activity bands not found in the Triton extracts. Xylanase was fully induced only when cells were grown on xylan; however, cells retained low, constitutive levels of xylanase when grown on monosaccharide components of xylan, or on the glucose polymer lichenan. ACKNOWLEDGEMENTS No work of the magnitude of a doctoral program takes place without the encouragement, support and patience of many people. I feel very fortunate to have been associated with an exceptional group of faculty, students, and staff who have always been willing to help and who have provided a stimulating environment in which to work and learn. I am especially indebted to the members of my Guidance Committee, Dre. R. B. Hespell, M. J. King and C. A. Reddy, for their counsel and encouragement. I am also grateful to Dr. R. Hausinger, Dr. E. R. Leadbetter, and Dr. D. stahl for technical assistance and advice at various stages of my work. I certainly want to thank Jodi Switzer Blum, Matt Kane, Jon Cooperider, Colleen Sweeney Demezas, Kristen Kennedy, Stefan Wagener, Dave Emerson, and Andreas Brune for making the lab a fun place to be, and for all the help you've each provided. I have been especially fortunate to work in the lab of Dr. John Breznak, whose wealth of knowledge has been a constant inspiration. Thank you John, for keeping me on track and for keeping the standards high. Finally, I want to thank my husband, Bob, for his patience and good humor and for being, as always, my best friend. TABLE OF CONTENTS List of Tables.....................................................vi List of Figures...................................................vii Introduction........................................................1 Chapter 1 Cytophaga xylanolytica sp. nov., a xylan-degrading, anaerobic, gliding bacterium ........36 Chapter 2 The xylan-degrading enzyme system of Cytophaga xylanolytica .................................78 Chapter 1 LIST OF FIGURES Figure 1 Phase contrast micrograph of a mass of Figure Figure Chapter 2 Figure Figure Figure Figure Figure 2 3 1 2 3 4 5 Cytophaga xylanolytica-like cells.......................47 Histogram depicting the distribution of 14C radioactivity HPLC fractions of spent culture fluid from a C. xylanolytica fermentation...............58 Absorption spectrum of an acetone/methanol (7/3, v/v) extract of glucose-grown cells of Cytophaga xylanolytica XM3........................................6O Transmission electron micrograph of xylan-grown cells of Cytophaga xylanolytica XM3.....................95 FPLC fractionation of protein and activity from cells of Cytophaga xylanolytica XM3 ....................97 Analytical isoelectric focusing of Triton extracts of xylan- or xylose-grown cells of Cytophaga xylanolytica XM3 ......................................101 Analytical isoelectric focusing of FPLC fractions of Triton extracts from xylan-grown cells of Cytophaga xylanolytica XM3.......................................104 Effect of xylose on induction of xylanase activity by xylan in cells of Cytophaga xylanolytica XM3........109 vi LIST OF TABLES Introduction Table 1 Bacterial xylan-degrading enzyme systems................9 Chapter 1 Table 1 Properties of freshwater isolates of Cytophaga xylanolytica..................................49 Table 2 Properties of marine strains of xylanolytic Cytophaga... ...... ..........................50 Table 3 Products of glucose and xylose fermentation by growing cells of Cytophaga xylanolytica XM3..........54 Table 4 Cellular fatty acids in freshwater isolates of Cytophaga xylanolytica ..............................63 Table 5 Xylanase activity of Cytophaga xylanolytica XM3.........65 Table 6 Cell-associated glycosidase activities of Cytophaga xylanolytica XM3..............................65 Chapter 2 Table 1 Distribution of xylanase in Cytophaga xylanolytica XH3 grown on 0.2% oat spelt xylan or 10 mM xylose........91 Table 2 Isoelectric point of carbohydrases of Cytophaga xylanolytica XM3.........................................99 Table 3 Depolymerase and glycosidase specific activities of Cytophaga xylanolytica XM3...........................105 INTRODUCTION Next to cellulose, hemicelluloses are the most abundant poly- saccharides in nature (Wilkie 1983). Hemicelluloses occur in the primary (growing) and secondary (mature) cell walls of all terrestrial plants, as well as in the cell walls of algae and aquatic plants (Wilkie 1983, Wong et a1. 1988). The structure of plant primary cell walls has been the subject of several recent reviews (Lamport 1986, Lamport and Catt 1981, McNeil et a1. 1984, Wilson and Fry 1986). Up to 90% of the primary cell wall is polysaccharide variously substituted with acetyl esters and methyl and feruloyl esters and ethers (McNeil et a1. 1984). Lignin is not a component of primary cell walls, but phenolic acids are common constituents of Gramineae (grasses) primary cell walls (Chesson et a1. 1982). Approximately 10% of the primary plant cell wall is protein, which is comprised of hydroxyproline-rich, arabinose-substituted extensin, arabinogalactan proteins (the biological function of these two types of protein is unknown), and other proteins (frequently glycosylated) including malate dehydrogenase, peroxidase, phosphatases, glycosyl hydrolases and transferases, and endoglucanases (McNeil et a1. 1984). In the current model, cellulose (6-1,4 linked D-glucan; 11—24% of the polysaccharide of the cell wall) fibrils are woven into an extensin matrix. The extensin helices are cross linked by isodityrosine dimers 2 resulting in an extremely insoluble protein mesh. The cellulose fibrils are "sheathed" by hemicellulose polymers. In dicots, the hemicellulose is primarily (20% of cell wall) xyloglucan, a/3-1,4- linked D-glucose with some,6-1,6-D—xylose side chains. In monocots, it is primarily (35% of cell wall) xylan, which is described below. The hemicelluloses are hydrogen bonded to the cellulose fibrils in a manner which varies with the degree of substitution of the hemicellulose backbone. Pectins (homo— and hetero- galacturonans and arabinans) form an additional meshwork around the other molecules, possibly as an amorphous matrix. Pectins comprise 34%, 10%, and 22% of the primary cell walls of dicots, monocots or softwoode (Douglas fir), respectively (Thomas et al. 1987). Much less is known about the structure of secondary plant walls due to their highly lignified nature, but xylans are abundant in woody tissue and comprise the major hemicellulose in wood from angiosperms, as well as a substantial portion (7-12% of dry weight) of the wood from gymnosperms (Timmell 1962, Whistler and Richards 1970). Xylene comprise a class of hemicellulose widely distributed among both terrestrial and aquatic plants. They are complex heteropolymers typically consisting of a,6-1,4-linked xylopyranoee backbone variously substituted with L-arabinofuranoee, 4-O-methyl-D- glucuronic acid, and O-acetyl side groups. The arabinoee substituents are typically linked to C-3 of the xylose residues while the acetic acid and methyl-glucuronic acid residues are usually linked to C-2. The nature and degree of substitution varies with the stage of plant growth as well as within different tissues of the same plant (Wong et 3 a1. 1988, Wilkie 1983). Typical chemical procedures used to isolate xylans from plant material employ alkaline hydrolysis which may alter acidic constituents or hydrolyze ester linkages, leading to a loss of characteristic in situ features of the xylan, such as the acetyl and feruloyl linkages described above (Wilkie 1983). The use of enzymes from plant-polymer degrading fungi and bacteria has been extremely useful in elucidating the structures of plant cell wall polysaccharides. The xylans of maize primary cell walls have been particularly well studied by both enzymatic and chemical methods and have been shown to exhibit several of the properties of the hypothetical xylan (Kato and Nevins 1984 a-d, 1985, Nishitani and Nevins 1988, 1989, 1990, 1991). Kato and Nevins (1984a) found that the predominant noncellulose polysaccharide of a water insoluble fraction of Zea mays shoot cell walls is an arabinoxylan. Treatment of the water insoluble fraction with an endo-(1,4)-[3- xylanase purified from a commercial preparation of Bacillus subtilis d—amylaee resulted in glucuronoarabinoxylan (GAX) fragments having a 5-1,4-D-xylopyranose backbone with 60-70% substitution at C-2 or 0-3 with arabinose, glucuronic acid or other substituents (1984d). The SAX further contained ferulic acid, later shown to be bound to the sax as O-(5-O—feruloyl-obL-arabinofuranosyl)-(1-3)—O-5-D-xylopyranoeyl-(1- 4)-D-xylopyranose (1985). The Bacillus subtilis enzyme released only 208 of the total GAx from the cell wall preparations (1984d). A second Bacillus subtilus enzyme degraded glucuronic acid substituted xylans of both bean (Vigna) and Zea mays cell walls to characteristic d—1,2-glucuronic acid substituted xylosyl oligomers (Nishitani and 4 Nevins 1991). This constitutes the first evidence of a repeating unit structure in any xylan. Non-enzymatic analyses have also been useful in elucidating the structure of xylans in the maize cell wall. Carpita (1983) extracted hemicellulose from Zea mays coleoptile cell walls by using alkali gradients and showed that hemicellulosic polymers comprised 43% of these PCW's. Three separate fractions [GAx I, GAX II and MG-GAX (a mixture of 60% mixed-linked glucans and 40% GAX)] comprised 20, 30 and 50% respectively of the total extracted hemicellulose. GAx I (extracted wth 0.01-0.045 N KOH) was highly substituted; 6 of every 7 xylose residues carried a terminal arabinose (t-ara) or an d-1,2- linked glucuronic acid. The t-ara substituente were most common although some galactose residues comprised a very small proportion of substituents. GAx II (O.45-0.8 N KOH) was substituted on 2 of every 3 xylose residues, with similar substituente to GAx I. GAx I may be a precursor of other GAx in the primary cell wall (Carpita and Whittern 1986). Carpita (1986) also examined ferulic acid in maize cell walls and found aromatic material (including aromatic amino acids) comprising only 0.3-1.2% of 2-4 day coleoptiles. Host of the radioactivity from incorporated 14C-phenylalanine or 14 C—tyrosine was incorporated into the hemicellulose fraction. Based on differential extraction Carpita (1986) proposed etherified as well as esterified aromatics as substituente of hemicelluloses in Zea mays (see also Scalbert et a1. 1985). He suggested, based on 14C-proline incorporation, that hydroxyproline-rich proteins comprise only a small 14 percentage of maize cell walls. Finally, he demonstrated that C 5 labelled glucose or xylose are metabolized to hexosee before being incorporated into the cell wall polymers, but over 80% of 14C- arabinose is incorporated directly into non-cellulosic polymers (Carpita et a1. 1982). No analysis to date has been made of the distribution of acetyl groups in maize hemicelluloses; however, Bacon et al. (1975) showed O-acetyl groups to be common in grasses. Xylan degrading enzymes. From the foregoing discussion of the nature of plant cell walls it is obvious that one important characteristic of a xylan-degrading microorganism would be possession of the requisite depolymerizing enzymes. The suite of enzymes commonly thought to be necessary for the complete degradation of a xylan include endoxylanaee (E.C. 3.2.1.8) to degrade the xylan backbone, xylobiase (E.C. 3.2.1.37) to cleave terminal xylose residues from xylooligomers produced by endoxylanase activity, and arabinofuranosidase (E.C. 3.2.1.55) to remove arabinosyl side groups. Reviews on the nature and sources of xylanases, xylosidases, and arabinofuranosidases have appeared previously (Biely 1985, Dekker and Richards 1975, Kaji 1984, Reilly 1981, Wong et al. 1988). As our concept of the structure of xylan in situ has been refined, it has become apparent that additional enzymes would be needed to remove methyl-glucuronic, as well as acetyl and feruloyl substituente. The importance of acetyl or acetyl-xylan esterases ( Biely et a1. 1986, Wood and McCrae 1986, Hespell and O'Bryan-Shah 1988),(X-4-O-Me-D-glucuronidase (Puls et a1. 1987, Johnson et a1. 1989) and ferulic acid esterase (Mackenzie and Bilous 1988, Borneman et al. 1990) has only recently been recognized. Acetyl esterases and glucuronidases have been shown to be synergistic with xylanases in the degradation of isolated xylans, but little information on the nature and mode of action of these enzymes exists (Biely et al. 1986, Puls et al. 1987). The lack of a suitable substrate for in vitro tests of enzyme activity has hampered the study of both the glucuronidases and the ferulic acid esterases. Theoretically, the enzymes described above should be sufficient to fully degrade a hypothetical xylan, and should constitute the basic components of a xylan-degrading enzyme system. However, the picture is complicated by the evidence that some xylanases exhibit a preference for, or may be specialized to degrade, xylans with specific substituente (Dekker and Richards 1975, Nishitani and Nevins 1991, Reilly 1981, Wong et al. 1988). In many (most) cases, a single microorganism possesses multiple endoxylanases which differ in substrate specificity, mode of action, or properties such as isoelectric point and molecular weight (Wong et al. 1988). It is now clear that many bacteria possess multiple xylanase genes (Flint et al. 1989, 1990, Gilbert et al. 1988, Hamamoto et al. 1987, Hazlewood et al. 1988, Honda et al. 1985, Kluepfel et al. 1990, Mackenzie et al. 1989, Sakka et al. 1989, 1990, Schwarz et al. 1990, Vats-Mehta et al. 1990, Yang et al. 1989). Wong et al. (1988) observed that many fungi and bacteria produce two forms of endoxylanaee: a high molecular weight/acidic form and a low molecular weight/basic form. These authors also reviewed evidence from their own work with Trichoderma harzianum, as well as from the work of Takenishi and Tsujisaka (1975) on Aspergillus, that multiple endoxylanases from a single 7 microorganism can cooperate in the degradation of xylan. However, in both examples, the existence and degree of cooperativity of the different endoxylanases was dependent on the nature of the test substrate (Wong et al. 1988). Finally, many endoxylanasee exhibit substrate cross-specificity (Wong et al. 1988) degrading, for example, carboxymethylcellulose or lichenan. Conversely, many endoglucanases have some activity on xylan. It is clear that xylan-degrading enzyme systems can be extremely complex; however, as pointed out by Wong et al. (1988) biotechnological applications of isolated enzymes, enzyme combinations or whole organisms will require an understanding of this complexity. Described bacterial xylan—degrading enzyme systems Xylan-degrading enzymes have been purified, or cloned (or both) from several genera of bacteria (Table 1). In no case have all the relevant enzymes been studied in a single bacterial genus. Often, research on xylan-degrading enzymes has followed naturally from previous purification or cloning of the cellulose-degrading enzymes of a particular microorganism (e.g., Clostridium thermocellum). other studies have employed combinations of denaturing and non-denaturing gel electrophoresis, isoelectric focusing, and activity stains (e.g. Bachmann and McCarthy 1991) to give an overview of the number and nature of the enzymes which comprise the xylan-degrading enzyme system of a previously unstudied microorganism. This review extends the information on bacterial xylan-degrading enzyme systems provided by Wong et al. (1988), and specifically focuses on those bacterial genera for which there is substantial information on one or more of the 8 enzymes involved in xylan degradation. In the sections below, aspects of the xylan-degrading enzyme systems of these focal genera are discussed in greater detail. mama .Hs um someones: Inuoumwox e.e oz Amv H>x NH uswwasm ammo .Hu um censuses oz oMH loozo.oozo.xozov loo HH Has lemma .o.m~mma oz and Hex lo.mv H flax .Hm um cmumcnocuo oz m.~m + .o.ov sax ooH newness o.m mm + ADV m a»: 0.0 m.om + on N sax mama .Hu so was» o.m mm + loo H can «mom omz moswsouwu s.¢ mm sax Add a»: «man H.m ma «x u «x Alex loo m can .Hm um cmnmuum m.¢ mm «x u ax Alex lay a a»: mains ucueaouwo ammo .Hm um oz we HzxmcmuuamxA.nxAucx is. z ass a. ouosusmz in.MmmmH .Ha no meson oz me HzxmemuuamxA.NxAuox Ao.ov a a»: mnauo .on usauwoun mwoa .Hs oz mm + loo m :zx um osmoxro .mmoH oz an + loo 2 a»: moms .Hu um onus oz was «xx .Nx Alex lo.mv o aux «an .o: .ou Ham." . Am 30 “04> N . m NN Hixmflmhum NXA . NNAIGN Amy Chan le3 Oflflbflb QQROEOHOQ oosououo: Hm :8 huwouuod elhusn qussmuc .usouuhn oshnco undocumoolssahx aseuouosm .u eunsa 10 Ham” .Hu um uuo oz ow uuua\ozx .0. o Hon moons moosaomwuoeu Homo seasons one coo oz no Aoozo .cxv ADV m chx Uhflm msoswomwhoflw coma .Ha um Haaousccsz o.m m.N¢ + ADV d can we msoeuouwunflh ONHQRSHAuom coma Hooomoz use soundness oz 00 mx Alex ADV aux swoowswfish momma Hooommz oz on uuum\oox loo «mum\ozx com ummomuooz oz oz + loo can mauuso MOEWOHOHOQN mama anoomuoooa sou anouuonoz m.m mo «and is. onus Hans museuonm Homo moo>mz can oomuoomoz oz so ”noose .oxaoooxuoxcooo lo. a»: ammo mca>mz ocs ounx oz m.m + Am. chx mowflunsn moms .Hs no modem no.mmoma a.m on + .o. as: .ou um monsoon oz an nx .Nx .anox loo as: maaooo mawwuoum maeawuuo ousouoHOI no 3: heavens: slhusn ludsdmuo A.U.uooov .H CHAIR ll Huuo>on oz + AU. m cox mama Huuo>on mm + luv u oz: .Hn um descends: Heum>mm oz + ADV H aux ammo .Hm um Hoax oz om .ox .ozov lo. a doc o.mmmma .Ha no venomous oz om ioozo .ooz .oxv Ao.ov u can suwaooosuoou oz om .omo loozo .ozo .oxe loo x H00 oz me loozo .ozo .oxv lo. 3 Hon oz on onus zoo m was oz me maum\ozx loo o one oz mo ouuu\ozx loo o Hzxo cams oz oz a»: loo 4 Hzxn .Hm um nuuzoom oz os.~m.~v + loo a ozx mesoo oHoz anucuouuouu mm.v no me .mx Alex loo 0 czx mama m¢.¢ ms nu .mx Aiox zoo o ozx .am no Moosouom mm.¢ ow mx .Nx Alex Am. t sax Eswhuhouuoou omma .Hu um moons mw.m an” HxAuoxnmx .xozo lo. Hex «moo ammo announce new moo om.o on ma .Nx Alex .o.ov o ozx .smmo .o« no moo me.v mo oz I «x Alex lo. a a»: «No oooz soowazusoouoou eswowhquNU ousouom08 no 3! huwkuot elhusn laws-mac ..o.uooo. .H OHA‘H 12 ammo mumomuoo commo.HA om loses cz.mx.ux.xAuox¢ lo. a ozx sou muss: oommo.HA s.mm onus ooaex.mz.¢Auoxo .ov a ozx ammo .Hm no unoom oz oz + loo ozx mmm nosouosfloosm Monounsaowm oz can ozx couseooumslgoo oz on zooo as m oz ow zone ox a oz me zooo ox n oz he zooo as n oz mo zooo ox H HsEOmoHoaaooisoz oz om + «on oz mm loo: .ox. Assess". mm oz oma zoo: .oxv mm oz em ooze .oxv mom oz so loo: .ozo oo\am oz so loo: .ox. em ammo oz oao .ozo .ox. an Inn #0 UMHOZ HdEOmOHDHHOU ESNHOOOEOC” sownwuuuouo ousououem no 38 husbuuni Essa Indulge ..o.uooo. . .— 0."an 1. l ammo o.m mm m: .: Auo: loo Ho oz: .oa um coonoms u.¢ em a: .: Auo: zoo H oz: omnooo mooouwow>osuoou Homo moouuooo pom o:uomuo m.m m.m~ mx .m: .oAuoxo lo. oz: nonwoouowumwomoh omoo .Hm no snow: nmum> .ommH .Hm um Hooooaoz .oomo .Hm um ooooo: .omma e.o on m: n m: Auo: .o.oo o oz: .Ha no ooomouoz ~.m no m:.:AuomoHo o: .o.o. o oz: mo mooowswa moozsoumohom. oz omnom mxA.m: Auo: loo o oz: Homo .Hm no soaoz oz om oozo.:ozoa~:.:xuo: loo o oz: Huoo moowuouosuau momma .Hm U0 0>0u0 O.¢ moml MMHM Nflhfl neomo m.¢ com + m oz: .Hm um m>muo w.m our + H of: m 93:. MDOOOUOEWEDN mama .oo um uumoooo .ommo oz om onus ooiokuozo 0 oz: .Hm um uuooomz oz on m: .zAuozo o oz: ammo .oa um Homo m.q om + c oz: «moaaawoo .ouoom mcoomohoswm mesofiopsoom doseuomox no 3: husk-80¢ Olhuflfl lad—=90 ..o.uooo. .o sands Iazconmouuwolo .m: omuoooooz: .w: .xozo "nouos mmouw>wuom Hmooeuaoom .A «csazx azumom .cxlod acmazxosousosam .cxo unwoo oauuooaoone «Hm omcHEumuoo poo .oz uzox. unwed: umaooosoe "3: .endowmoenoHHoolazuouzaaunEstnuoEIv .oox nouMHsHHooncvoszonuso .ozo aomsoamowooHamelazcocmouueclm .oozo aonsoamoooamlazcoomouuocan .uozo «omsoHuonx .omcoHo .Hoo mommowooosuswocenmumlol v «oompwmonxmonuu.azxocsuu “onceuuonx «osazxocanmus .c:¢ mosazx .ox "hudbduofl «omwuwuso .on «womuouoo qumos .mo¢ ammmcsosamooso .wmum «mmmoemonxl .Hz: «omscmazxooco .czx ”useduesmduoo Olhusn mama .Hu um ammouoo oz on + .0. oz: :z woman ammo zouumooz ocm ocmsnomo zm.¢ moH Amy Hz: oz om oxnoo ouom zoo zoo oz mm loo noun ooooua .HA on + oz: w.m mm + aha Homo zouuuooz ~.m mm + oz: com oomsoomm m.z «m + oz: Hmoo woman shomnososofihouh ousououox Hm I! hvw>wuot Olhusfl lswssuuo ..o.uooo. .H OHACB 15 Acre-ones. The genus Aeromonas includes Gram negative, facultatively anaerobic bacteria related to the vibrios. The species in Table 1 were isolated in one case on the basis of alkalophilic growth as well as xylan hydrolysis (sp. no. 212), and in the other case simply on the ability to degrade xylan. All the studied enzymes were produced extracellularly but there is no indication that cell-associated enzymes, if present, were investigated. The xylanase of A. caviae was induced by xylan and not by xylo-oligosaccharides or cellulose. This enzyme possessed transxylosidase activity, which has been noted in several other xylanases (Table 1, Reilly 1981, Biely 1985, Wong et al. 1988). The xylanase L gene, cloned from sp. no. 212, hybridized with only one genomic fragment from sp. no. 212, suggesting that xylanases S and M from this species were encoded by separate genes. Bacillus. Several species in the Gram positive, endospore-forming genus Bacillus have been investigated with respect to xylan- degradation. All the investigated xylanases have been produced extracellularly, and as was the case for Aeromonas, there is no indication of whether these bacteria produce cell-associated xylanases. The xylosidase of B. circulans WL-12 was produced extracellularly while those of B. pumilus 12 and IPO were located intracellularly. This is the only genus in which two xylosidase genes have been identified (B. pumilus IPO, Table 1). In B. pumilus IPO, the xylanase gene and xylosidase I gene were encoded on the same genomic restriction fragment (Panbangred et al. 1983a,b). A B. subtilus PAP115 22 kD xylanase exhibited 50% amino acid homology with the B. pumilus IPO xylanase, but no identity with a 32 kD xylanase 16 previously cloned from the same organism (Paice et al. 1986). While several Bacillus species are recognized to produce a number of polysaccharide-degrading enzymes other than xylanases (Esteban et al. 1982, Nishitani and Nevins 1991), there have been no comprehensive studies, in one species, of the circumstances under which these multiple enzyme activities are expressed. Rules bacteria. The rumen bacterial genera Bacteroides, Butyri- vibrio, Fibrobacter and Ruminococcus have been studied in some detail to determine the conditions under which their polysaccharide-degrading enzymes are expressed (Williams and Withers 1982a,b, 1985; Martin and Akin 1988, Hespell and O'Bryan-Shah 1988, Sewell et a1. 1988, Pettipher and Latham 1979, Greve et al. 1984c). Butyrvibrio is considered to be the major xylan/hemicellulose degrading bacterium in the rumen (Hespell et al. 1987). Although Bacteroides, Fibrobacter and Ruminococcus produce xylan-degrading enzymes (Table 1), these genera grow poorly on xylan (Hespell et al. 1987) and are generally thought to utilize primarily cellulose in the rumen. Williams and Withers (1982a,b) comprehensively examined the effect of carbohydrate growth substrate on glycosidase (14 enzymes), depolymerase (6 enzymes) and Lolium multiflorum cell wall degrading activities of Ruminococcus, Fibrobacter, Bacteroides, and Butyrivibrio isolates after growth on several monomeric or polymeric carbohydrates. Interestingly, although not pointed out by the authors, each of these genera best degraded L. multiflorum cell walls after growth on the carbohydrate substrate (different for each genus) which best induced cellulase activity (Williams and Withers 1982a). Only in Fibrobacter succinogenes $85 17 were the highest levels of xylanase activity found after growth on xylan. In the other genera the highest levels of xylanase activty were noted after growth on xylose (Ruminococcus albus RUMS), arabinan (Ruminococcus flavefaciens 123), maltose (Bacteroides ruminicola strains), or glucose (Butyrivibrio fibrisolvens HlOb). All of these isolates also produced xylanase (although at lower levels) after growth on xylan. Bacteroides ovatus, included in Table 1, is not a rumen isolate, but occupies the human colon. Both Bacteroides and Butyrivibrio have yielded enzymes which on good evidence appear to be bifunctional xylosidase/ arabinofurano- sidases (Table 1). In Bacteroides ovatus the gene for this bifunctional enzyme appears on a 3.8 kb genomic fragment which also contains a xylanase gene (Whitehead and Hespell 1990). However, the xylanase gene was regulated independently of the bifunctional gene, and there was some evidence for a second arabinofuranosidase gene. The bifunctional xylosidase/ arabinofuranosidase of Butyrivibrio fibrisolvens 68113 did not exhibit amino acid homology with any known xylan-degrading enzymes or glycosidases. The xynA gene product from Butyrivibrio fibrisolvens 49 exhibited 37% amino acid homology with the Bacillus sp. C-125 xynA protein (Mannarelli et al. 1990), but hybridized to chromosomal sequences in only two of five closely related Butyrivibrio strains. The B. fibrisolvens H17c xynB gene product exhibited 32%, 38%, and 40% amino acid homology with the cellobiohydrolase/endocellulase of Clostridium saccharolyticum, the xylanases of Bacillus sp. C-125, and the xylanase of B. fibrisolvens 49, respectively (Lin and Thomson 1991). 18 Fibrobacter and Ruminococcus each produce more than one xylanase protein (Table 1). There was clear evidence for at least two genes (possibly more) in R. flavefaciens FD—l, as well as indication of multiple copies of both the xynA and xynB genes in the R. flavefaciens FD-l genome (Flint et al. 1991). There was also evidence of increased mRNA production from both xynA and xynB in R. flavefaciens FD-l grown on xylan as compared to cellobioee (Flint et a1. 1991). The endoxylanases purified from the non—sedimentable culture fluid of Fibrobacter succinogenes 585 exhibited single bands on SDS-PAGE gels but multiple bands (4 each) after isoelectric focusing (Matte and Forsberg 1992). This phenomenon was also observed by Bachmann and McCarthy (1991) in their study of the xylanases of Thermomonospora fusca, as well as by MacKenzie et al. 1989, in work on the xylanases of Clostridium thermocellum. The causes and significance of this heterogeneity remain unknown. Endoxylanase 1 from F. succinogenes 885 exhibited the interesting property of removing arabinose residues from arabinoxylan without accompanying activity on aryl-OFL-arabinofurano— sides, arabinan, or arabinogalactan (Matte and Forsberg 1992). Similar activity has been shown for Streptomyces roseiscleroticus (Table 1) as well as for a xylanase from the fungus Trichoderma koningii (Wong et al. 1988). Clostridiun. This genus includes Gram positive, anaerobic, endospore formers widely recognized for diverse substrate utilization. Many of the species studied with regard to xylan-degradation were originally isolated for their ability to degrade cellulose, and in fact the original C. acetobutylicum ATCC 84 strain, as well as C. thermocellum 19 grow very poorly on xylan. During growth on xylan as sole substrate, xylose oligomers and xylose accumulated in the C. thermocellum culture fluid, indicating a preference by this species for high molecular weight xylan fragments (Weigel et al. 1985). other bacterial genera which grew rapidly on xylan utilized xylose first, and then xylo-oligomers. C. thermocellum excretes a high molecular weight complex comprised of 26 subunits separable by SDs-PAGE (Bayer et al. 1983, Kohring et al. 1990), which binds tightly to cellulose, and has been designated the "cellulosome." Xylanolytic activity in the cellulosome has only recently been investigated. A similar, secreted high molecular weight cellulolytic and xylanolytic complex has also been examined in the meeophilic Clostridium sp. C7 (Cavedon et al. 1990). Non-cellulosome associated cellulases and xylanases have been found in the culture fluid of both species (Morag et al. 1990). A total of 18 genes cloned from C. thermocellum were found to encode 15 endoglucanases, 2 xylanases, and one,6-glucosidase (Hazlewood et al. 1988). of the endoglucanases, 6 were observed to have xylanase activity as well. Kohring et a1. (1990) found that 15 of 26 bands appearing after SOS-PAGE exhibited endoglucanase activity, and 13 had xylanase activity. Both activities were found in 8 of the 26 bands. Many (14) of these proteins were glycosylated, and debranching activities were also detected. Grepinet et al.(1988a,b) determined the nucleotide sequence of the xynz gene, and purified and studied the encoded protein (Table 1). Mackenzie et al. (1989) obtained an additional three cloned fragments encoding xylanase 20 grow very poorly on xylan. During growth on xylan as sole substrate, xylose oligomers and xylose accumulated in the C. thermocellum culture fluid, indicating a preference by this species for high molecular weight xylan fragments (Weigel et al. 1985). other bacterial genera which grew rapidly on xylan utilized xylose first, and then xylo-oligomers. C. thermocellum excretes a high molecular weight complex comprised of 26 subunits separable by SDS-PAGE (Bayer et al. 1983, kohring et a1. 1990), which binds tightly to cellulose, and has been designated the "cellulosome." Xylanolytic activity in the cellulosome has only recently been investigated. A similar, secreted high molecular weight cellulolytic and xylanolytic complex has also been examined in the meeophilic Clostridium sp. C7 (Cavedon et al. 1990). Non-cellulosome associated cellulases and xylanases have been found in the culture fluid of both species (Morag et al. 1990). A total of 18 genes cloned from C. thermocellum were found to encode 15 endoglucanases, 2 xylanases, and one B-glucosidase (Hazlewood et al. 1988). Of the endoglucanases, 6 were observed to have xylanase activity as well. kohring et a1. (1990) found that 15 of 26 bands appearing after SOs-PAGE exhibited endoglucanase activity, and 13 had xylanase activity. Both activities were found in 8 of the 26 bands. Many (14) of these proteins were glycosylated, and debranching activities were also detected. Grepinet et al.(1988a,b) determined the nucleotide sequence of the xynz gene, and purified and studied the encoded protein (Table 1). Mackenzie et al. (1989) obtained an additional three cloned fragments encoding xylanase 21 activity, none of which appeared to be identical to the xynz gene. The xynz gene encoded a polypeptide with a centrally located 60 amino acid sequence, containing duplicated 24-amino acid segments, which was similar to conserved domains found at the carboxy-terminus of C. thermocellum endoglucanases (Grepinet et al. 1988b). The conserved domain was not required for catalysis. The xynB gene product from C. acetobutylicum P262 exhibited 65% homology with the Bacillus pumilus IPO xylanase, and 40% homology with the B. subtilus PAP115 xylanase (Zappe et al. 1990). Actinomycetes. The actinomycetes have been found to be excellent producers of extracellular xylanases, as well as of other depolymerases (Ball and McCarthy 1989, Johnson et al. 1988, Zimmermann et al. 1988). Advances in the molecular biology and genetics of these bacteria have allowed cloning between species, avoiding problems encountered with expression and excretion when foreign genes are cloned into Escherichia coli (e.g., Vats-Mehta et al. 1990). Two xylanase-encoding genes have been cloned from Streptomyces lividans 66 (Mondou et al. 1986, Vats-Mehta et a1. 1990) into a xylanase/endo- xylanaee double mutant of S. lividans. The genes appear to encode two proteins previously purified from s. lividans 66, which are not immunologically related (kluepfel et al. 1990, Morosoli et a1. 1986). There appears to be little xylanase associated with S. lividans cells, but the production of extracellular enzyme was increased by the addition of Tween 80 or olive oil to cultures during growth (Bertrand et al. 1989). The gene for one of four xylanases found in the culture medium of Thermomonospora fusca was cloned into S. lividans Tk24 22 (Ghangas et al. 1989). The xylanaee encoded by this gene bound to and hydrolyzed insoluble xylan. The partial amino acid sequence for xylanase I from s. thermoviolaceus 0PC-520 showed 47% homology with the Cellulomonas fimi exoglucanase, while the xylanase II protein exhibited 46% homology with the Bacillus pumilus IPO xylanase (Teujibo et al. 1992). Thermomonospora fusca BD21 produced extracellular and intracellular xylanase, acetyl esterase, and arabinofuranosidase proteins; however, the,6-xylosidase was located intracellularly (Bachmann and McCarthy 1989, 1991). In this species, acetyl esterase was responsible for all observed acetyl-xylan esterase activity. Addition of purified,6-xylosidase to T. fusca endoxylanaee increased the hydrolysis of xylan, and similar addition of purifieth-L- arabinofuranosidase resulted in increased saccharification of wheat straw (Bachmann and McCarthy 1991). Mackenzie et al. (1987) examined the induction of extracellular cellulase and xylanase enzymes from S. flavogriseus and S. olivochromogenes after growth on cellulose, oat spelt xylan or wheat bran. Xylan-containing substrates induced cellulase, xylanase, and debranching enzyme activities (acetyl—xylan esterase, d-L-arabinofuranosidase, and(X-L-O-Me-glucuronidase) in both species, as well as ferulic acid esterase in S. olivochromogenes. Pseudo-ones. Two genomic fragments encoding three xylan-degrading enzymes have been cloned from Pseudomonas fluorescens subsp. cellulosa (Gilbert et al. 1988). The xynA gene was encoded on the same genomic fragment as an endoglucanase gene with which it exhibited substantial 23 homology due to the possession by both of highly conserved domains (see below; Hall et al. 1989). The gene for xynB appeared on a different fragment along with the xync gene. The product of the xync gene was shown to release only arabinose from arabinoxylan (i.e., there appeared to be no endoxylanaee activity) and this protein exhibited no activity on aryl-er-arabinofuranosides (Gilbert et al. 1988). Both xynB and C possessed conserved sequences which conferred on the protein products the ability to bind to cellulose (but not xylan), and which were not required for catalysis (Kellett et al. 1990). These results elicit intriguing questions with regard to the origin of polymer-degrading enzymes of similar function, a topic recently reviewed for the,8—1,4-glycanases by Gilkes et al. (1991). These authors suggest that cellulases and xylanases of both procaryotic and eucaryotic origin can be grouped into families based on conserved amino acid sequences in their putative catalytic domains. Many of these enzymes exhibit additional sequences, again frequently conserved across species, genus and procaryotic-eucaryotic boundaries, which are similar to sequences known to confer the ability of some enzymes to bind to a substrate (usually cellulose), or which encode for protein segments rich in proline or threonine residues which may be glycosylation sites (Gilkes et al. 1991). They propose that fusion or shuffling of these different domains may account for the somewhat bewildering array of enzymes which hydrolyze the,6-1,4-glycosidic bond, with varying degrees of specificity, in an equally diverse array of polysaccharides. 24 Factors affecting the complexity of xylan-degrading enzyme systems It is obvious that we have learned a great deal in recent years about the number and nature of xylan-degrading enzymes from several genera of bacteria. Use of denaturing and non-denaturing gel electro- phoresis, isolelectric focusing and activity stains, has revealed information about the xylan-degrading enzyme systems of genera for which genetic systems are not well-established. Concurrently, molecular techniques have been developed for several well-studied genera and protein isolation methods have become more routine. The combination of these diverse methods has begun to reveal a previously unsuspected complexity in bacterial xylan-degrading enzyme systems. It is now clear that most bacteria possess multiple xylanase genes encoding proteins which exhibit further heterogeneity in physical characteristics, substrate specificity, or mode of action. In turn, many xylanase genes may be part of a larger gene complex encoding polysaccharide degrading activities in a multitude of procaryotic and eucaryotic organisms. Although this diversity will enhance the probablity of finding very specific enzymes for biotechnological applications, it will encumber the search for pattern and definition. It seems clear that xylan—degrading enzyme systems will not be as neatly packaged as, for example, bacterial cellulase systems, where three types of enzymes (endoglucanase, exoglucanase, and/g-glucosidase) essentially perform the same functions in all bacteria, and are sufficient to degrade cellulose wherever it may be found. The heterogeneity of bacterial 25 xylan-degrading enzyme systems is probably most directly a function of the heterogeneity of xylan, and thereafter a function of the heterogeneous environment of the plant cell-wall. While enzymatic tests for xylanase activity employ isolated and soluble substrates, bacteria degrading xylan in nature will encounter xylan inextricably linked to a multitude of other polysaccharides. Most bacteria probably do not need to fully degrade xylan in order to survive. Instead, multiple enzyme activities which partially degrade xylan, as well as other plant cell wall polysaccharides, or which clip away residues from a variety of parent molecules, may be more advantageous than a "complete" xylan—degrading enzyme system. In this respect, it seems logical that many xylanases have some activity on other polysaccharide substrates, and vice versa. It also seems logical that, as for the rumen bacteria discussed above, xylan-degrading enzymes would be induced by mono- or polysaccharides other than xylan. It is also important to recognize that our understanding of the complexity of bacterial xylan-degrading enzyme systems is restricted by the types of questions we have asked. In some cases the primary research objective has been to obtain enzymes with unique substrate specificities or tolerances, and no effort has been made to examine other components of the xylan-degrading enzyme system, or to determine the circumstances under which the producing organism utilizes xylan in nature. In some cases, no effort has been made to determine whether a given xylanase has activity on other substrates, or whether the conditions under which the enzyme was studied represent those under 26 which it would typically, or optimally, be expressed. As more xylan- degrading enzymes are isolated we will have a better concept of the criteria to be met by specific classes of xylanases. Finally, much of our knowledge comes from research with bacteria which do not grow on xylan. Should we expect the components of the xylan-degrading system of Clostridium thermocellum (which, as noted above, does not grow well on xylan) to be similar to those of, for example, Butyrivibrio? The evolutionary constraints on these two systems will be quite different. The C. thermocellum system does not need to produce xylose, or even apparently, xylo-oligomers for the cells (Wiegel et al. 1985) and may exist only to free cellulose for further cellulaee attack. It may be that some competition for plant cell wall polymers is based on the ability of a cell to "carve out" of the cell wall, and utilize, the unique products of its own polymer- degrading enzyme system. 27 References Bachmann SL, McCarthy AJ (1989) Purification and characterization of a thermostable B-xylosidase from Thermomonospora fusca. J Gen Microbiol 135:293-299 Bachmann SL, McCarthy AJ (1991) Purification and cooperative activity of enzymes constituting the xylan-degrading system of Thermomonospora fusca. Appl Environ Microbiol 57:2121—2130 Bacon JSD, Gordon AH, Morris DE (1975) Acetyl groups in cell-wall preparations from higher plants. Biochem J 149: 485-487 Ball AS, McCarthy AJ (1989) Production and properties of xylanases from actinomycetes. J Appl Microbiol 66:439-444 Bayer EA, kenig R, Lamed R (1983) Adherence of Clostridium thermocellum to cellulose. J Bacteriol 156: 818-827 Berenger J-F, Frixon C, Bigliardi J, Creuzet N (1985) Production, purification, and properties of thermostable xylanase from Clostridium thermocellum. Can J Microbiol 31:635-643 Bernier R Jr, Desrochers M, Jurasek L, Paice M G (1983) Isolation and characterization of a xylanase from Bacillus subtilis. Appl Environ Microbiol 46: 511-514 Bernier R Jr., Driguez H, Desrochers M (1983) Molecular cloning of a Bacillus subtilus xylanase gene in Escherichia coli. Gene 6:59-65 Bertrand J-L, Morosoli R, Shareck F, kluepfel D (1988) Expression of the xylanase gene of Streptomyces lividans and the production of the enzyme on natural substrates. Biotech Bioeng 33:791-794 Biely P (1985) Microbial xylanolytic systems. Trends Biotechnol 3: 286-290 Biely P, Mackenzie CR, Puls J, Schneider H (1986) Cooperativity of esterases and xylanases in the enzymatic degradation of acetyl xylan. Bio/Technology 4: 731-733 Biely P, Markovic o, Mislovicova D (1985) Sensitive detection of endo-l,4-beta-glucanases and endo-1,4-beta-xylanases in gels. Anal Biochem 144: 147-151 Biely P, Puls J, Schneider H (1985) Acetyl xylan esterases in fungal cellulolytic systems. FEBS Lett 186:80-84 Bornemann WS, Hartley RD, Morrison WH, Akin DE, Ljungdahl, LG (1990) Feruloyl and p-coumaroyl esterase from anaerobic fungi in relation to plant cell wall degradation. Appl Microbiol Biotechnol 33:345- 351 28 Carpita NC (1983) Hemicellulosic polymers of cell walls of Zea coleoptiles. Plant Physiol 72: 515-521 Carpita NC (1984) Cell wall development in maize coleoptiles. Plant Physiol 76: 205-212 Carpita NC (1986) Incorporation of proline and aromatic amino acids into cell walls of maize coleoptiles. Plant Physiol 80: 660-666 Carpita NC, Whittern D (1986) A highly substituted glucuronoarabino- xylan from developing maize coleoptiles. Carbohydr Res 146: 129-140 Carpita NC, Brown RA, Weller, RA (1982) Uptake and metabolic fate of glucose, arabinose, and xylose by Zea mays coleoptiles in relation to cell wall synthesis. Plant Physiol 69: 1173-1180 Cavedon k, Leschine SB, Canale-Parola E (1990) Cellulase system of a free-living meeophilic Clostridium (strain C7). J Bacteriol 172:4222—4230 Chesson A, Stewart CS, Wallace RJ (1982) Influence of plant phenolic acids on growth and cellulolytic activity of rumen bacteria. Appl Environ Microbiol 44: 597-603 Dekker RFH, Richards GN (1975) Hemicellulases: Their occurrence, purification, properties, and mode of action. Adv Carbohydr Chem Biochem 32:277-352 Esteban R, Villanueva JR, Villa TG (1982) Beta-D-xylanases of Bacillus circulans WL-12. Can J Microbiol 28: 733-739 Flint HJ, McPherson CA,, Bisset J (1989) Molecular cloning of genes from Ruminococcus flavefaciens encoding xylanase and/3(1-3,1-4)- glucanase activties. Appl Environ Microbiol 55:1230-1233 Flint HJ, McPherson CA, Martin J (1991) Expression of two xylanase genes from the rumen bacterium Ruminococcus flavefaciens 17 cloned in pUC13. J Gen Microbiol 137:123-129 Forsberg CW, Beveridge TJ, Hellstrom A (1981) Cellulase and xylanase release from Bacteroides succinogenes and its importance in the rumen environment. Appl Environ Microbiol 42: 886—896 Ghangas GS, Hu Y-J, Wilson DB (1989) Cloning of the Thermomonospora fusca xylanase gene and its expression in Escherichia coli and Streptomyces lividans. J Bacteriol 171:963-2969 Gilbert HJ, Sullivan DA, Jenkins G, kellett LE, Minton NP, Hall J (1988) Molecular cloning of multiple xylanase genes from Pseudomonas fluorescens subsp. cellulosa. J Gen Microbiol 134:3239-3247 29 Gilkes NR, Henrissat B, kilburn DG, Miller RCJr, Warren RAJ (1991) Domains in microbial[3-1,4-glycanases: sequence conservation, function, and enzyme families. Microbiol Rev 55:303-315 Gilkes NR, Langsford ML, kilburn DG, Miller RCJr, Warren RAJ (1984) Mode of action and substrate specificities of cellulases from cloned bacterial genes. J Biol Chem 259: 10455-10459 Grabeki AC, Jeffries TW (1991) Production, purification, and characterization of,6-(l,4)-endoxylanase of Streptomyces roseiscleroticus. Appl Environ Microbiol 57:987-992 Grepinet O, Chebrou M-C, Beguin P (1988a) Purification of Clostridium thermocellum xylanase Z expressed in Escherichia coli and identification of the corresponding product in the culture medium of C. thermocellum. J Bacteriol 170:4576—4581 Grepinet O, Chebrou M-C, Beguin P (1988b) Nucleotide sequence and deletion analysis of the xylanase gene (xynZ) of Clostridum thermocellum. J Bacteriol 170:4582-4588 Greve LC, Labavitch JM, Hungate RE (1984a)CX-L-arabinofuranosidase from Ruminococcus albus 8: Purification and possible role in hydrolysis of alfalfa cell wall. Appl Environ Microbiol 47:1135- 1140 Greve LC, Labavitch JM, Hungate RE (1984b). Xylanase action on alfalfa cell walls. In: Dugger WM and Bartnicki-Garcia S (eds) Structure, Function, and Biosynthesis of Plant Cell Walls. Proc. 7th Symposium in Botany, Univ Calif, Riverside, pp 150-166 Greve LC, Labavitch JM, Stack RJ, Hungate RE (1984c) Muralytic activities of Ruminococcus albus 8. Appl Environ Microbiol 47:1141-1145 Grohmann R, Mitchell DJ, Himmel ME, Dale BE, Schroeder HA (1989) The role of ester groups in resistance of plant cell wall polysaccharides to enzymatic hydrolysis. Appl Biochem Biotechnol 20/21:45-61 Hall J, Hazlewood GP, Barker PJ, Gilbert HJ (1988) Conserved reiterated domains in Clostridium thermocellum endoglucanases are not essential for catalytic activity. Gene 69:29-38 Hall J, Hazlewood GP, Huskieson NS, Durrant AJ, Gilbert HJ (1989) Conserved serine-rich sequences in xylanase and cellulase from Pseudomonas fluorescens subsp. cellulosa: internal signal sequence and unusual protein processing. Mol Microbiol 3:1211-1219 Hamamoto T, Honda H, kudo T, Horikoshi K (1987) Nucleotide sequence of the xylanase A gene of alkalophilic Bacillus sp. strain C-125. Agric Biol Chem 51:953-955 A 30 Hazlewood GP, Romaniec MPM, Davidson k, Grepinet O, Beguin P, Millet J, Raynaud O, Aubert J-P (1988) A catalogue of Clostridium thermocellum endoglucanase,[3-glucosidase and xylanase genes cloned in Escherichia coli. FEMS Microbiol Lett 51:231-236 Hespell RB, O'Bryan-Shah PJ (1988) Esterase activities in Butyrivibrio fibrisolvens strains. Appl Environ Microbiol 54: 1917-1922 Hespell RB, Wolf R, Bothast RJ (1987) Fermentation of xylans by Butyrivibrio fibrisolvens and other ruminal bacterial species. Appl Environ Microbiol 53: 2849-2853 Honda H, kudo T, Ikura Y, Horikoshi R (1985) Two types of xylanases of alkalophilic Bacillus sp. no. C-125. Can J Microbiol 31: 538— 542 Honda H, Kudo T, Horikoshi R (1985) Molecular cloning and expression of the xylanase gene of alkalophilic Bacillus sp. strain C—125 in Escherichia coli. J Bacteriol 161: 784-785 Johnson kG, Harrison BA, Schneider H, Mackenzie CR, Fontana JD (1988) Xylem-hydrolyzing enzymes from Streptomyces spp. Enzyme Microb Technol 10: 403-409 Johnson kG, Silva MC, Mackenzie CR, Schneider H, Fontana JD. (1989) Microbial degradation of hemicellulosic materials. App Biochem Biotech 20/21: 245-258 kaji A (1984) L-arabinosidases. Adv Carbohydr Chem Biochem 42: 383-394 kato Y, Nevins DJ (1984) Enzymic dissociation of Zea shoot cell wall polysaccharides. I. Preliminary characterization of the water-insoluble fraction of Zea shoot cell walls. Plant Physiol 75: 740—744 kato Y, Nevins DJ (1984) Enzymic dissociation of Zea shoot cell wall polysaccharides. II. Dissociation of (1->3), (1->4)-beta-D- glucan 4-glucanhydrolase from Bacillus subtilis. Plant Physiol 75: 745-752 kato Y, Nevins DJ (1984) Enzymic dissociation of Zea shoot cell wall polysaccharides. III. Purification and partial characterization of an endo-(1—>4)-beta—xylanase from a Bacillus subtilis enzyme preparation. Plant Physiol 75: 753-758 kato Y, Nevins, DJ (1984) Enzymic dissociation of Zea shoot cell wall polysaccharides. IV. Dissociation of glucuronoarabinoxylan by purified endo-(1->4)-beta-xylanase from Bacillus subtilis. Plant Physiol 75: 759-765 31 Kato Y, Nevins DJ (1985) Isolation and identification of o-(S-O- feruloy1-alpha-L-arabinofuranosyl)-(1->3)-O-beta-D-xylopyranosyl- (1->4)-D-xylopyranose as a component of Zea shoot cell—walls. Carbohydr Res 137:139—150 Kellett LE, Poole DM, Perreira LMA, Durrant AJ, Hazlewood GP, Gilbert HJ (1990) Xylanase B and an arabinofuranosidase from Pseudomonas fluorescens subsp. cellulosa contain identical cellulose-binding domains and are encoded by adjacent genes. Biochem J 272:369-376 Kersters-Hilderson H, Loontiens FG, Claeyssens M, De Bruyne CK (1969) Partial purification and properties of an induced beta-D-xylosidase of Bacillus pumilus 12. Eur J Biochem 7: 434-441 Kluepfel D, Vats-Mehta S, Aumont F, Shareck F, Morosoli R (1990) Purification and characterization of a new xylanase (xylanase B) produced by Streptomyces lividans 66. Biochem J 267:45-50 Kohring s, Wiegel J, Mayer F (1990) Subunit composition and glycosidic activities of the cellulase complex from Clostridium thermocellum JW20. Appl Environ Microbiol 56:3798-3804 Kudo T, Ohkoshi A, Horikoshi K (1985) Molecular cloning and expression of a xylanase gene of alkalophilic Aeromonas sp. no. 212 in Escherichia coli. J Gen Microbiol 131:2825-2830 Lamport DTA (1986) The primary cell wall: A new model. In Young RA and Rowell RM (eds.), Cellulose: Structure, Modification, and Hydrolysis. Wiley, New York pp 77-90 Lamport DTA, Catt Jw (1981) Glycoproteins and enzymes of the cell wall. Encyl Plant Physiol 138: 133-165 Lee SF, Forsberg CW, Gibbins LN (1985) Xylanolytic activity of Clostridium acetobutylicum. Appl Environ Microbiol 50: 1068-1076 Lee SE, Forsberg CW, Rattray JB (1987) Purification and characterization of two endoxylanases from Clostridium acetobutylicum ATCC 824. Appl Environ Microbiol 53: 644-650 Lee SP, Forsberg CW (1987) Isolation and some properties of a beta-D—xylosidase from Clostridium acetobutylicum ATCC 824. Appl Environ Microbiol 53: 651-654 Lin L-L, Thomson JA (1991) Cloning, sequencing and expression of a gene encoding a 73 kDa xylanase enzyme from the rumen anaerobe Butyrvibrio fibrisolvens H17c. Mol Gen Genet 228:55—61 MacKenzie CR, Bilous D (1988) Ferulic acid esterase activity from Schizophyllum commune. Appl Environ Microbiol 54:1170-1173 32 Mackenzie CR, Bilous D, Schneider H, Johnson KG (1987) Induction of cellulolytic and xylanolytic enzyme systems in Streptomyces spp. Appl Environ Microbiol 53:2835-2839 MacKenzie CR, Yang RCA, Patel GB, Bilous D, Narang S (1989) Identification of three distinct Clostridium thermocellum xylanase genes by molecular cloning. Arch Microbiol 152:377-381 Mannarelli BM, Evans S, Lee D (1990) Cloning, sequencing, and expression of a xylanase gene from the anaerobic ruminal bacterium Butyrivibrio fibrisolvens. J Bacteriol 172:4247-4254 Martin SA, Akin DE (1988) Effect of phenolic monomers on the growth and B-glucosidase activity of Bacteroides ruminicola and on the carboxymethylcellulase, B-glucosidase, and xylanase activities of Bacteroides succinogenes. Appl Environ Microbiol 54:3019-3022 Matte A, Forsberg CW (1992) Purification, characterization, and mode of action of endoxylanases 1 and 2 from Fibrobacter succinogenes $85. Appl Environ Microbiol 58:157-168 McNeil M, Darvill AG, Fry SC, Albersheim P (1984) structure and function of the primary cell walls of plants. Annu Rev Biochem 53: 625-663 Mondou F, Shareck F, Morosoli R, Kluepfel D (1986) Cloning of the xylanase gene of Streptomyces lividans. Gene 49:323-329 Morag E, Bayer EA, Lamed R (1990) Relationship of cellulosomal and noncellulosomal xylanases of Clostridium thermocellum to cellulose- degrading enzymes. J Bacteriol 172:6098-6105 Moriyama H, Hata Y, Yamaguchi H, Sato M, Shinmyo A, Tanaka N, Okada H, Katsube Y (1986) Crystallization and preliminary X-ray studies of Bacillus pumilus IPO xylanase. J Mol Biol 193: 237-238 Morosoli R, Bertrand J-L, Mondou F, Shareck F, Kluepfel D (1986) Purification and properties of a xylanase from Streptomyces lividans. Biochem J 239:587-592 Nishitani K, Nevins DJ (1991) Glucuronoxylan xylanohydrolase A unique xylanase with the requrement for appendant glucuronosyl units. J Biol Chem 266:6539-6543 ohkoshi A, Rude T, Mase T, Horikoshi K (1985) Purification of three types of xylanases from an alkalophilic Aeromonas sp. Agric Biol Chem 49:3037-3038 Paice MG, Bourbonnais R, Desrochers M, Jurasek L, Yaguchi M (1986) A xylanase gene from Bacillus subtilus: nucleotide sequence and comparison with B. pumilus gene. Arch Microbiol 144:201-206 33 Panbangred W, Shinmyo A, Kinoshita S, Okada H. 1983a. Purification and properties of endoxylanaee produced by Bacillus pumilus. Agric. Biol. Chem. 47: 957-963. Panbangred W, Rondo T, Negoro S, Shinmyo A, okada H. 1983b. Molecular cloning of the genes for xylan degradation of Bacillus pumilus and their expression in Escherichia coli. Mol. Gen. Genet. 192: 335—341. Panbangred W, Kawaguchi O, Tamita T, Shinmyo A, Okada H. 1984. Isolation of two beta-xylosidase genes of Bacillus pumilus and comparison of their gene products. Eur. J. Biochem. 138: 267- 273. Pettipher GL, Latham MJ. 1979. Characteristics of enzymes produced by Ruminacoccus flavefaciens which degrade plant cell walls. J. Gen. Microbiol. 110: 21—27. Puls J, Schmidt 0, Granzow C. 1987. d-glucuranidase in two microbial xylanolytic systems. Enzyme Microb. Technol. 9: 83-88. Reilly PJ. 1981. Xylanases: Structure and function. Pages 111-129 in A. Hollaender (Ed.), Trends in the Biology of Fermentation for Fuels and Chemicals. Plenum, New York. Sakka K, Furuse S, Shmada K (1989) Cloning and expression in Escherichia coli of thermophilic Clostridium sp. F1 genes related to cellulose hydrolysis. Agric Biol Chem 53:905-910 Sakka K, Kojima Y,, Yoshikawa K, Shimada K (1990) Cloning and expression in Escherichia coli of Clostridium stercorarium strain F-9 genes related to xylan hydrolysis. Agric Biol Chem 54:337-342 Scalbert, A., B. Monties, J.—Y. Lallemand, E. Guittet, and C. Rolando. 1985. Ether linkage between phenolic acids and lignin fractions from wheat straw. Phytochemistry. 24: 1359-1362. Schwarz WH, Adelsberger H, Jauris S, Hertel C, Funk B, Staudenbauer WL (1990) Xylan degradation by the thermophile Clostridium stercorarium: cloning and expression of xylanase, B-D-xylasidase, and d-L— arabinofuranosidase genes in Escherichia coli. Biochem Biophys Res Commun 170:368-374 Sewell GW, Utt EA, Hespell RB, Mackenzie RF, Ingram L0 (1989) Identification of the Butyrivibrio fibrisolvens xylosidase gene (xle) coding region and its expression in Escherichia coli. Appl Environ Microbiol 55:306-311 Sipat, A., K. A. Taylor, R. Y. C. Lo., C. W. Forsberg, and P. J. Rrell. 1987. Molecular cloning of a xylanase gene from Bacteroides succinogenes and its expression in Escherichia coli. Appl. Environ. Microbial. 53:477-481. 34 Takenishi S, Tsujisaka Y (1975) On the modes of action of three xylanases produced by a strain of Aspergillus niger van Tieghem. Agric Biol Chem 39:2315-2323 Teather RM, Wood PJ. 1982. Use of Congo red-polysaccharide interactions in enumeration and characterization of cellulolytic bacteria from the bovine rumen. Appl. Environ. Microbiol. 43: 777-780. Thomas JR, McNeil M, Darvill AG, Albersheim P. 1987. Structure of plant cell walls. XIX. Isolation and characterization of wall polysaccharides from suspension cultured Douglas fir cells. Plant Physiol. 83: 659-671. Timmel TE (1962) Enzymatic hydrolysis of a 4-O-methylglucuronoxy1an from the wood of white birch (Betula papyrifera Marsh.) Sven Papperstidn 65:435-447 Tsujibo H, Miyamoto K, Kuda T, Minami K, Sakamato T, Hasegawa T, Inamori Y (1992) Purification, properties, and partial amino acid sequences of thermostable xylanases from Streptomyces thermaviolaceus OPC-520. Appl Environ Microbiol 58:371-375 Utt EA, Eddy CK, Keshav KF, Ingram Lo (1991) Sequencing and expression of the Butyrivibrio fibrisolvens xle gene encoding a novel bifunctional protein withfl3-D-xylasidase andCX-L-arabinofurano- sidase activties. Appl Environ Microbial 57:1227-1234 Vats—Mehta S, Bouvrette P, Shareck F, Morosoli R, Kluepfel D (1990) Cloning of a second xylanase-encoding gene of Streptomyces lividans 66. Gene 86:119-122 Weinstein L, Albersheim P (1979) Structure of plant cell walls. Ix. Purification and partial characterization of a wall-degrading enda-arabanase and an arabinofuranosidase from Bacillus subtilus. Plant Physiol 63: 425-432 West CA, Elzanowski A, Yeh L-s, Barker WC (1989) Homologues of catalytic domains of Cellulamanas glucanases found in fungal and Bacillus glycosidases. FEMS Microbiol Lett 59:167-172 Whitehead TR, Hespell RB (1990) Heterologous expression of the Bacteraides ruminicola xylanase gene in Bacteroides fragilis and Bacteroides uniformis. FEMS Microbiol Lett 66:61-66 Whitehead TR, Hespell RB (1990) The genes for three xylan-degrading activities from Bacteroides ovatus are clusterred in a 3.8-kilabase region. J Bacterial 172.2403-2412 Whistler RL, Rchards EL (1970) Hemicellulases. In Pigman W, Horton D (eds.) The carbohydrates-Chemistry and biochemistry, 2nd edition, Vol Ila, Academic Press, New York, pp 447-469 35 Wiegel J, Mothershed CP, Puls J (1985) Differences in xylan degradation by various noncellulolytic thermophilic anaerobes and Clostridium thermocellum. Appl Environ Microbiol 4:656-659 Wilkie, KCB (1983) Hemicellulase. Chemtech 13:306-319 Williams AG, Withers SE. 1982a. The production of plant cell wall polysaccharide-degrading enzymes by hemicellulalytic rumen bacterial isolates grown on a range of carbohydrate substrates. J. Appl. Bacterial. 52: 377-387. Williams AG, Withers SE. 1982b. The effect of the carbohydrate growth substrate on the glycosidase activity of hemicellulose- degrading rumen bacterial isolates. J. Appl. Bacterial. 52: 389- 401. Williams AG, Withers, SE. 1985. Formation of polysaccharide depolymerase and glycoside hydrolase enzymes by Bacteraides ruminicola subsp. ruminicola grown in batch and continuous culture. 12: 79-84. Wilson LG, Fry JC. 1986. Extensin -- a major cell wall glycoprotein. Plant Cell Environ. 9: 239-260. Wong KKY, Tan LUL, Saddler, JN (1988) Multiplicity of/3-1,4—xylanase in microorganisms:functions and applications. Microbiol Rev 52:305- 317 Wood TM, McCrae SI (1986) Studies of two law-malecular—weight endo— (1,4)-B—D—xylanases constitutively synthesized by the cellulolytic fungus Trichoderma koningii. Carbohydr Res 148:321-330 Yang RCA, MacKenzie CR, Bilous D, Narang S (1989) Identification of two distinct Bacillus circulans xylanases by molecular cloning of the genes and expression in Escherichia coli. Appl Environ Microbiol 55:568-572 Vats-Mehta S, Bouvrette P, Shareck F, Morosoli R, Kluepfel D (1990) Cloning of a second xylanase-encoding gene of Streptomyces livdans 66. Gene 86:119-122 Viet DN, Kamia Y, Abe N, Kaneko J, Izaki K (1991) Purification and properties of/3-1,4-xylanase from aeromonas caviae W-61. Appl Environ Microbiol 57:445-449 Zappe H, Jones WA, Woods DR (1990) Nucleotide sequence of a Clastridium acetobutylicum P262 xylanase gene (xynB). Nuc Acids Res 18:2179 Zimmermann W, Winter B, Brada P (1988) Xylanolytic enzyme activities praduceed by meeophilic and thermophilic actinomycetes grown on graminaaceous xylan and lignocellulose. FEMS Microbiol Lett 55:181- 186 Chapter One Cytophaga xylanolytica sp. nov., a xylan-degrading, anaerobic gliding bacterium 36 37 Abstract. Gliding bacteria attached in masses to, and dominated the fermentation of, xylan powder in methanogenic and sulfidogenic enrichments from various freshwater sediments. Isolates of such bacteria were all Gram negative, slender rods (0.4 x 4-24 um) that formed no endospores, microcysts or fruiting bodies. Representative strain XM3 was a mesophilic, aeroduric anaerobe that grew by fermentation of mono-, di-, and polysaccharides (but not cellulose) in a mineral medium containing up to 3% NaCl. However, COZ/HCO3- was required in media for consistent initiation of growth. Fermentation products included acetate, propionate, succinate, C02, and H2. Xylan- grown cells had xylanase and various glycosidase activities that were mainly or almost entirely cell-associated, respectively. Strain XM3 was weakly catalase positive, but oxidase negative; it possessed sulphonolipids and carotenoid, but not flexirubin, pigments; and its total cellular fatty acids were dominated by 615:0 anteiso (751), n (131) and iso (21) isomers. Strain XM3 had 45.5 molZ G + C in its DNA, and partial sequencing of its 168 rRNA placed XM3 within the Bacteroides-Flavobacterium phylogenetic group. Similar strains have recently been isolated from marine sediments. Strain XM3 is herewith proposed as the type strain of the new species, CytoPhaga xylanolytica. Results, which are discussed in terms of our current concept of the genus CytoPhaga, suggest that the importance of C. xylanolytica in anaerobic biopolymer decomposition has not been fully appreciated. 38 Xylans are a class of cell wall hemicelluloses widely distributed among both terrestrial and aquatic plants. They are heteropolysaccharides consisting of a B-l,4-linked xylopyranose backbone, often with a-l,3-arabinofuranoside and a-1,2- methylglucuronic acid side groups. In addition, C2 and C3 of some of the xylopyranose residues may be acetylated (Biely 1985). Next to cellulose, xylans are probably the most abundant polysaccharides in nature; and it has been estimated that nearly 1010 metric tons are turned over annually (Wilkie 1983). Unfortunately, however, our understanding of the microbiology and biochemistry of xylan decomposition is fairly limited (Biely 1985). This is particularly true for anaerobic decomposition of xylans by free-living microbial communities. Accordingly, the present study was initiated. This paper reports on the isolation and characterization of freshwater and marine cytophagas capable of decomposing xylan and other polysaccharides under strictly anaerobic conditions. Freshwater isolates are typified by strain XM3 which is proposed as the type strain of the new species, CytoPhaga xylanolytica. The properties of C. xylanolytica reported herein extend our current concept of the genus CytOphaga and suggest that the importance of such organisms in anaerobic decomposition of biopolymers has not been fully appreciated. 39 Materials and Methods Enrichment and isolation of bacteria Approximately 1 ml samples of aquatic sediment slurry were inoculated into 50 m1 of NaZS-reduced, COZ/HC03--buffered, defined mineral media containing, as sole fermentable substrate, 53 mg of xylan (larchwood or oat spelt; pre- extracted with 702 ethanol; Sigma Chem. Co.) in powdered (i.e. non- heat-sterilized) farm. Media contained 0 to 20 mH $04-2 (for methanogenic or sulfidogenic enrichments, respectively) and a salt concentration compatible with the source of inoculum (freshwater or marine; see below) and were held in rubber-stoppered serum bottles under Nz/CO2 (80/20, v/v). Incubation was at 30°C. From robust enrichments, pure cultures of xylanolytic bacteria were isolated by using 11 agar shake tubes of homologous medium containing 0.1-0.22 heat-sterilized (solubilized) xylan. Isolated strains were maintained in liquid medium or on 1.5% agar bottle plates (Hermann et a1. 1986). Media and growth studies Defined "freshwater" mineral medium #1 contained (per liter): KH2P04, 0.2 g; NH4 CaClz.2H20, 0.15 g; NaCl, 1.0 g; MgClZ.6H20, 0.62 g; NaHCO3 (84 g/l; Widdel and Pfennig 1984), 30 ml; SL-lO trace element solution Cl, 0.25 g; xc1, 0.5 g; solution (Widdel et al. 1983), 1 m1; W/Se solution (0.1 mM each of Na28e03 and Na NO in 20 mM NaOH), 1 ml; vitamin B solution (50 mg/l), 1 ml; 2 4 12 mixed vitamin solution (below), 1 m1; and 1 M NaZS (prepared and stored under N2), 1 ml. The mixed vitamin solution contained (per liter): 4-aminobenzoic acid, 40 mg; D-(+)'biotin, 10 mg; nicotinic acid, 100 mg; Ca-D(+)-pantothenate, 50 mg; pyridoxamine 40 dihydrochloride, 100 mg; and thiamine dihydrochloride, 100 mg. All solutions were heat-sterilized, except for the mixed vitamin solution which was filter-sterilized. Energy sources (eg. xylan, xylose, etc.) were sterilized separately and included at a final concentration of 0.1-0.22 or 5-10 mM. Freshwater mineral medium #2 was identical to #1, except that MgC12.6H20 was replaced by Mg804.7H20 (0.74 g/l). Defined "marine" medium was identical to freshwater medium #2, except that the following salts were increased to (g/l): NaCl, 25; CaCl .2H 0, 1; and 2 2 KCl, 1; and MgSO4.7H20, 5. "Brackish" water medium was prepared by mixing equal volumes of marine medium and freshwater medium #2. Broth and agar media were usually prepared and incubated under NZ/CO2 (80/20, v/v), which resulted in a final pH of 7.2 - 7.4. Adjustment of the pH, if necessary, was made by adding sterile 1M Na CD3 or HCl, or by varying the COZ/NaHCO ratio (Costilow 1981). The 2 3 requirement by cells for COZ/HCO3- was examined by omitting NaHCO3 from the medium, but including 10 mM 3-(N-morpholino)propanesulfonic acid (MOPS buffer; adjusted to pH 7.3 and heat-sterilized separately) and incubating cells under a gas phase of 1002 N To test aerobic 2. growth, cells were incubated under air in sulfide-free, bicarbonate- free freshwater medium #2 which was buffered with Tris/TRIZMA base (pH 7.3; 20 mM final concentration). Turbidimetric measurement of cell growth, cell dry mass determinations, and assessment of utilization of organic and inorganic substrates were done as previously described (Breznak et al. 1988). 41 Metabolic studies Material balances for the fermentation of uniformly- labeled 14C-glucose and 14C-xylose were done with cells growing in freshwater medium #2 containing these substrates. Products were analyzed as previously described (Patrikus and Breznak 1977), but analyses also included separation and quantification of soluble products by high performance liquid chromatography (HPLC), trapping of 14 C02, and determinations of specific 14C radioactivity all as described by Breznak and Switzer (1986). Tests for aerobic respiration of xylose were done by using a Warburg type respirameter and conventional manometric techniques (Umbreit et al. 1957). Reaction mixtures (3.0 ml) were held in double sidearm respirameter vessels and contained: 20 mM Tris buffer and 10 mM K-phosphate buffer, both at pH 7.3; NaCl, 0.17 H; xylose, 20 mH; dithiothreitol (DTT; for anoxic incubations only), 3 mM; and cells [grown anaerobically on xylose; harvested by centrifugation and resuspended in Tris-phosphate-NaCl (above)], 2.54 mg dry cell mass equiv. Reactions were initiated by introducing xylose from one of the $0 from 2 4 the other sidearm. Incubation was at 30°C under air or 1001 N2, and sidearms, and they were terminated by adding 0.2 ml of 5N H any gas exchange was corrected for that occuring in xylose-free (endogenous metabolism) reaction mixtures. Enzyme assays Late exponential to early stationary phase xylan-grown cells were harvested at 5°C by centrifugation at 10,000 x g for 10 min to obtain extracellular (supernatant fluid) and cell-associated (pellet) enzyme fractions. The extracellular fraction was either used 42 without further treatment or was dialyzed (12-14 kD mol. wt. cutoff) for 3 h at 4°C against 3 changes of ca. 50 volumes each of 0.01 M N- [2-hydroxyethyllpiperazine-N'-[2-ethanesulfonic acid] buffer (pH 6.0) containing 1 mM CaCl2 (= HFPES/CaClz). Dialyzed preparations were then concentrated by ultrafiltration through an Amicon PM10 membrane under 20 psi N . Pellets were resuspended in HEPES/CaCl2 (pH 6.8), and 2 cells were disrupted at 4°C by using a Branson sonicator operating for 1.5 min at 602 duty cycle and an output setting of 5. Xylanase and carboxymethylcellulase (CMCase) were determined by measuring the release of reducing sugar from xylan or CMC, respectively. Reaction mixtures contained 2.5 m1 of 0.12 (w/v) heat- solubilized xylan (oat spelt, lot no. 116F-0240; Sigma Chemical Co.) or CMC (Na salt; Sigma Chem. Co.), 1.5 ml of 0.1 M HEPES/CaCl2 (pH 6.8), and 1 ml enzyme solution (added to start the reaction) and were incubated at 30°C. Reducing sugar liberated was quantified by colorimetric assay (Nelson 1944) with xylose as standard. Glycosidase and acetylesterase activities were determined by measuring the release of p-nitrophenol from p-nitrophenyl substrates (Greve et al. 1984). Reaction mixtures contained 1 ml of substrate (10 mM in 0.1 M HEPES, pH 6.8) and 0.2-1.0 m1 enzyme solution. Incubation was at 30°C. Chemical assays Glucose was determined by HPLC (Breznak and Switzer 1986) and by assay with glucose oxidase (Sigma Chem. Co.). Xylose was determined by HPLC (above) and by the orcinol assay (Herbert et al. 1971). Xylan was determined by using the orcinol or phenol-H 2804 assay (Herbert et a1. 1971). Pyruvate was quantified and qualitatively 43 identified as the 2,4-dinitrophenylhydrazone derivative (Freidmann and Haugen 1943), as well as by HPLC of the underivatized acid (above). Protein was determined by using the Bradford reagents (Bio-Rad Chem. Co.). Sulfide was assayed as described by Cord-Ruwisch (1985), and production of methane and volatile fatty acids (VFA’s) in enrichment cultures was determined by gas chromatography (Schink and Pfennig, 1982). Other procedures For sequencing of 16S rRNA, total nucleic acids were extracted from cells with hot phenol and were precipitated from the aqueous phase with ethanol (Lane et al. 1985). Nucleotide sequences were determined by using the dideoxynucleotide method with: reverse transcriptase; 16S rRNA as template; and three universal oligonucleotide primers (Sanger et al. 1977; Lane et al. 1985) provided by Dr. D. Stahl (Univ. Illinois, Urbana, USA). Total cellular fatty acids were analyzed by using the MIDI system (Microbial ID, Inc., Newark, DE, USA) according to procedures described in the technical manual which accompanied the MIDI equipment. Briefly, this involved: 1) saponification of cellular lipids with methanolic NaOH; 2) esterification of fatty acids with acidic methanol; and 3) extraction of fatty acid methyl esters (FAMES) into methyl tert-butyl ether/hexane, then into NaCl-saturated 0.3 N NaOH, prior to analysis by gas chromatography on a 25 m fused silica capillary column. 44 Spot tests for carotenoid pigments were performed as described by Skerman (1967). Carotenoids were also extracted from cell pellets of strain XM3 with acetone/ methanol (7/3, v/v) for determination of absorption spectra. Tests for flexirubin pigments were done on cell pellets as described by Reichenbach et al. (1974). Sulphonolipid analysis was kindly performed by Drs. W. Godchaux and E. R. Leadbetter (Univ. of Connecticut, Storrs, USA) as described by these same workers (1983). DNA base composition of strain XM3 was kindly determined by Dr. J. Flossdorf (GBF, Braunschweig, FRG) by using a buoyant density method (Flossdorf 1983). Previously described methods were also used for electron microscopy (Breznak and Pankratz 1977) and for determination of nitrate reduction (Smibert and Krieg 1981). Phase contrast photomicrographs were made by using the procedure of Pfennig and Wagener (1986). 45 Results Enrichment and isolation of bacteria Various freshwater sediments collected in the vicinity of Konstanz, FRG, all gave rise to methanogenic and sulfidogenic enrichments when inoculated into mineral medium containing xylan powder. During the first 2 weeks of incubation, most enrichments became turbid, and the otherwise readily- dispersible xylan powder became gummy and dispersible only with vigorous swirling. Phase contrast microscopy of xylan particles at this time revealed large numbers of slender, rod-shaped bacteria attached to, and gliding over, the particle surface (Fig. 1a). In methanogenic enrichments, 4 mM acetate and 3 mM propionate appeared as the dominant (but transient) VFA’s in culture fluid, and after 8 weeks approximately 94% of the original xylan carbon could be accounted for as CH4 and C02. In sulfidogenic enrichments, acetate (4-5 mM) was the single dominant VFA produced, but xylan decomposition did not proceed to completion, presumably because sulfide accumulated to inhibitory levels (4-5 m”). From such enrichments, 20 strains of xylanolytic gliding bacteria were isolated, representing the numerically dominant colonies which developed in xylan shake tubes. All strains were virtually identical in size and in cell and colony morphology, and they all resembled the gliding bacteria seen on the surface of xylan particles in enrichments, so one of these (strain XM3) was selected for detailed study as reported below. Subsequently, similar gliding bacteria were enriched and isolated from freshwater muds (including a swine waste 46 Figure l.a. Phase contrast micrograph of a mass of CytoPhaga xylanolytica-like cells attached to the surface of a xylan particle in a methanogenic, freshwater enrichment culture. Individual cells can be discerned on some portions of the particle, along its edges, and after having glided out on the agar-covered slide (arrows). b.Phase contrast micrograph of C. xylanolytica XM3 cells in pure culture. c. Trans- mission electron micrograph of a thin section of xylan-grown C.xylano- lytica XM3. Note the vesicular evagination of the outer membrane (arrow). Marker bars = 5 um (a & b); 0.4 pm (c). 47 48 lagoon sediment) collected in East Lansing, Michigan (Table 1) and from marine sediments collected in Woods Hole, Massachusetts USA (Table 2). On first transfer into broth media, XM3 and many other strains grew as blotches attached to the inside wall of culture tubes or as macroscopically visible clumps in the liquid. However, such a phenotype was usually lost after several transfers, with most strains eventually growing as uniformly suspended cells. Isolates grew as swarms, not discrete colonies, on the surface of bottle plates. Moreover, when the Michigan and Woods Hole enrichments were streaked directly onto xylan bottle plates, thin, spreading swarms of gliding bacteria developed among, and extending out from, the more typical colonies of non-gliding bacteria. Selective transfer of the leading edge of such swarms to fresh plates resulted in pure populations of gliding bacteria, from which isolated colonies could ultimately be obtained by using agar shake tubes. In shake tubes solidified with Z 1% agar, isolated colonies of XM3 were about 2 mm in diameter, dense, lenticular or round in shape, and usually pigmented (see below). In less solid agar medium, XM3 colonies were spherical in shape, less dense, and increased in diameter with time (up to 1 cm). Gliding bacteria were never isolated from enrichments by streaking conventional Petri plates of xylan medium which lacked NaHCO3 and NaZS [but which contained 10 mM MOPS buffer (7.3), 1 mM DTT, and 3 mH 804.2], and which were incubated in an anaerobe chamber under N2/H2 (92/8, v/v). This was true even if the gas phase was enriched with 20% (v/v) CO Furthermore, if enrichments were established by using heat- 2. 49 Table 1. Properties of freshwater isolates of Cytophaga xylanolytica. x143 9.1.1b MA3 mu Cell size (um) 0.4 x 5-10 0.4 x 4-24 0.4 x 4-24 0.4 x 3-15 Substrates fermentedc xylan starch pectin inulin laminarin lichenin pullulan cellobiose lactose maltose sucrose D-glucose D-fructose D-galactose D-mannose D-xylose L-arabinose D-glucuronic acid D-galacturonic acid I + + I++I++I' I + + + + + + + I + + I + + + I + + + + + + I + I + + + + + + I + + + + + + + + + I + I + + + + + + I + + + + + + + + + I + I aAll strains were orange pigmented (subsurface colonies), catalase +, and oxidase -, and all formed acetate, propionate and succinate as major acidic products of xylan fermentation. bIsolated from a swine waste lagoon sediment. cNone of the strains fermented: cellulose (Sigmacell 20 microcrystalline cellulose; or ball-milled Whatman #1 filter paper), CMC, arabinogalactan, mannan, chitin, N-acetylglucosamine, L-sorbose, D-ribose, or L-rhamnose. Strain XM3 did not ferment: chitosan, sorbitol, glycerol, malate, citrate, formate, benzoate, gallate, syringate, caffeate, phenylacetate, xanthine, uracil, methanol, ethanol, phenylalanine, Brij 58, or Tween 20. 50 Table 2. Properties of marine strains of xylanolytic Cytophaga. OPZB OPZF EPA EPB PR2L BPFW Cell size (um) width 0.4-0.6 0.4-0.5 0.4-0.6 0.4-0.6 0.6-0.7 0.4-0.5 length 2-8 3-12 4-9 4-34 2-7 4-28 Colora orange white gold gold salmon white Carotenoid + - + + - - Flexirubin + — + + - - Catalase + + + + + + b Oxidase - + - - + ND Growth in (medium) brackish water + + + + + + freshwater #2 - - + + - + Acidic fermenta- tion productsc A,P A,P A,P A,P,S,B ND ND Agarolytic + + - - + - a . Subsurface colonies or agar surface swarms. bND, not determined c . . . During growth on xylan. A, acetate; P, propionate; S, succ1nate; B, butyrate. 51 sterilized (solubilized) xylan instead of xylan powder, CytoPhaga-like bacteria (although microscopically visible) were far less abundant. Morphology Cells of strain XM3 were Gram negative rods, possessing both an inner (cytoplasmic) and outer membrane and measuring 0.4 x 5- 10 um (Fig. lb,c). In thin sections, the outer membrane appeared relatively smooth, but possessed evaginations at various points to form cushion-like knobs or elongated vesicles. The latter were often 0.2-0.4 um long (Fig. 1c) and resembled those observed on certain cytophagas (Follett and Webley 1965; Oyaizu et al. 1982). Cells occured singly or in pairs, less frequently as chains or asterisk- shaped aggregates, and they lacked flagella. However, cells of XM3 and other isolates in liquid media could bend or flex slightly, translocate parallel to their long axis when attached to surfaces, transport minute particles of debris over the cell surface, and pivot on one cellular pole as described for typical cytophagas (Reichenbach 1989c). No endospores, microcysts, or fruiting bodies were formed, although ghost-like spherical bodies (both free and attached to cells) were formed in old cultures. The size range of all strains is shown in Tables 1 & 2. Based on their general phenotypic properties (described more fully below), isolates were concluded to be members of the genus CytOphaga (Reichenbach 1989c) and will be referred to by this epithet for the remainder of this paper. Nutrition and growth characteristics Cytophaga XM3 (as well as other freshwater isolates) grew in defined mineral media with various mono-, 52 di-, and polysaccharides as sole fermentable energy source (Table 1). Three of the 6 marine isolates were also agarolytic (Table 2). XM3 grew within a pH (initial) range of 6.1-8.7 (optimum 7.2-8.2) and temperature range of 19-37OC (optimum 30-320C). No growth occurred at 4°C or 45°C. When grown at 30°C with 10 mM glucose as energy source, cells had a doubling time of 2.6 h and attained an OD nm of 1.0-1.2 600 (equiv. 0.40 mg dry cell mass/ml). Cytophaga XM3 could use NH4+, glutamine (but not glutamate, N03-, or N2), or peptone as sole N source, and did not require vitamins for growth. However, cells did not respire anaerobically with N03- or 804-2 as terminal electron acceptor. Surprisingly, strain XM3 and other freshwater isolates grew equally well in media containing up to 3% NaCl. Conversely, all marine isolates grew in brackish water medium; 2 strains also grew in freshwater medium (Table 2). Luxurious growth of all freshwater isolates required the presence of COZ/NaHCO3 in the medium. Cells grew poorly or not at all in COZ/NaHCO -free, MOPS-buffered media under 1001 N Likewise Na S, 3 2' 2 used as the reducing agent in anoxic media, was the best S source. Strain XM3 grew poorly and inconsistently in NaZS-free medium containing 1 mM 804 2, cysteine, or glutathione (with or without 1 mM DTT included as reducing agent). No growth at all was observed in similar media containing methionine or thiosulfate. None of the other freshwater isolates obtained so far will use 304.2 as 8 source (with or without DTT as reducing agent). Interestingly, however, increasing the NaCl concentration in freshwater media to 10-20 g/l permitted maintenance of strain XM3 on marginal S sources such as 804 2 and S3 glutathione. CytOphaga XM3 grew in the presence of air in non-reduced, Tris/TRIZMA~buffered medium with glucose as energy source, but only if the following conditions were met: [NaCl] 2 12; [CD2 + Hco3‘] _>_ [energy source]; and the surface/volume ratio of the culture broth was small (i.e. held in a vertical culture tube). Under these conditions, 804 2 was provided as S source. However, cells of XM3 (and other freshwater isolates) grew poorly or not at all if such cultures were shaken, or if streaked on the surface of homologous agar medium exposed to air. Moreover, cell suspensions of strain XM3 did not exhibit xylose-dependent 02 uptake in manometric experiments. Based on these characteristics, XM3 was concluded to be an aeroduric anaerobe. The recently-isolated marine strains have not been examined in sufficient detail to state their precise relationship to oxygen. Fermentation products During anaerobic growth in COZ/NaHCO3-buffered medium, Cytophaga XM3 fermented 14C-UL-glucose and 14C-UL-xylose to acetate, propionate, succinate, and CO as major products (Table 3). 2 The same, or similar, major organic acid profiles were displayed by XM3 (and other isolates examined) when grown on xylan, with the exception that marine strain EPB also produced butyrate (Table 2). H2 was also produced by XM3, as was (occasionally) pyruvate (Table 3). However, although 98% of the substrate 14C was recovered in both experiments, only about 702 was associated with specifically identified products + cells: the remainder was present in HPLC fractions (Fl-F4) that elicited little or no response from the 54 Table 3. Products of xylose and glucose fermentation by growing cells of CytOphaga xylanolytica XM3. Products Pro- a pio- Ace- Succi- Pyru- b Substrate nate tate nate C02 H2 vate Unk Cells UL-14C-Xylose Recovery of 14C (I) 11.2 9.2 9.5 11.5 - 12.2 27.5 17.5 mmol per 100 mmol xylose fermentedc d 32.9 36.4 21.0 57.5 7.2 19.0 - - dpm per mmol (x 107) 2.2 1.7 3.0 1.3 - 2.0 - - mmol exogenous CO2 fixed per mmol product e 1.3 0.7 1.7 - - 1 5 ND ND per 100 mmol xylose 42.7 26.9 36.5 - - 27.7 ND ND UL-lac-Glucose Recovery of 14C (X) 14.6 10.0 5.7 14.4 - 0 32.5 21.0 mmol per 100 mmol glucose fermentedc d 53.1 41.6 9.8 86.2 3.2 0 - - dpm per mmol (x 107) 1.3 1.1 2.8 0.8 - 0 - - mmol exogenous CO2 fixed per mmol product 1.4 0.6 0.5 - - 0 ND ND per 100 mmol glucose 71.7 23.9 4.7 - - 0 ND ND 55 Table 3 Cont'd. alReCific activities were (107 dpm/mmol): UL'14C’glucose, 4.78; UL- C-xylose, 6.54. bUnknown soluble material (see text). Fermentation products tested for, but not detected in significant amounts, included: glycerol, sorbitol, dulcitol, mannitol, inositol, propanol, butanol, ethanol, 2,3-butanediol, acetoin, diacetyl, lactate, formate, butyrate, isobutyrate, valerate, isovalerate, crotonate, fumarate, citrate, oxalate, oxaloacetate, a-ketoglutarate, a-ketogluconate, fructose, mannose, galactose, ribulose, and glucuronic acid. cBased on direct chemical analyses, except for CO2 which was based on radioactivity determination. d . . . Determined in separate experiments. e . ND, not determined. 56 refractive index detector (Table 3; Fig. 2). The material in such fractions remains to be identified, but it was not residual glucose or xylose, nor was it any of 28 other common fermentation products including tricarboxylic acid cycle intermediates and sugar alcohols (Table 3, footnote d). Based on its retention time during HPLC, it may be a heterogeneous mixture of di- or oligosaccharides and/or derivatives thereof. Comparison of the specific radioactivity of identified products with that of the initial substrate indicated that COZ/HCO3 fixation (or exchange) was involved in sugar fermentation by strain XM3. From 0.5 to 1.7 mmol exogenous C02/HCO3- was incorporated per mmol organic acid formed (Table 3). Overall, however, about 1 mmol COZ/HC03- was incorporated into soluble products per mmol glucose or xylose dissimilated. Additional phenotypic and genotypic characteristics CytOphaga XM3 produced a cell-associated pigment which resulted in orange subsurface colonies in xylan shake tubes, orange to salmon colored swarms on xylan plates, and light pink cell pellets from glucose broth cultures. Spot tests suggested that the pigment was carotenoid, and acetone/methanol extracts of glucose-grown cells displayed an absorption spectrum typical of that of carotenoids (Zechmeister 1962) with major peaks at 458 (shoulder), 484 and 518 nm, and minor peaks at 357, 373, and 673 nm (Fig. 3). Qualitatively and quantitatively similar pigmentation was observed whether cells were grown in the dark or with incandescent illumination. No evidence was obtained for the 57 Figure 2. Histogram depicting the distribution of 14C radioactivity in HPLC fractions of spent culture fluid from a C. xylanolytica fermentation of 20 mM UL-14C-xylose and 10 mM UL-IAC-glucose. Py, S, A, and P are fractions containing pyruvate, succinate, acetate, and propionate, respectively, as indicated by the peaks in detector response. F1-F4 are arbitrary fractions. 58 A IIIV mszammm mOkOMFwD m>_._.<._wm E E a e m .u. m L ) - W A . A a . .5F_ tr M 1m _ my! 4_ nlu 2% N 3_ i . N .4 t F . m .1 41 it. l E 1L 1 r|5R e. M nw 5 O aw mu cw O aw 0 5 2 1 1 3 2 2 1 1 =2n_D I_<._.O._. “.0 e5 :UV 0.3....— ._.Z<._. n > iso. Among the hydroxy fatty acids, 3-0H-a-C15.0 was a major and consistently present species (Table 4). Of 1542 nucleotides typically present in 16S rRNA, 927 of those present in XM3 were sequenced, spanning Escherichia coli position numbers (Woese 1987) 218-510, 554-906, and 1093-1380. These were aligned against the E. coli, Bacteroides fragilis, and Flavobacterium [= CytOphaga; Christensen (1980)] heparinum complete sequences as reported by Weisburg et al. (1985). At position numbers 290 (G), 310 62 (C), 501 (G), 569 (U), 881 (A), 570 (U), 680 (G), 710 (C), 724 (U), 866 (A), and 1340 (U), strain XM3 bore the sequence signature of the Bacteroides-Flavobacterium subgroup (Weisburg et al. 1985). At position numbers 450 (G), 483 (C), and 484 (C), it resembled that of other eubacteria. Seven additional nucleotide positions useful in distinguishing the bacteroides and flavobacteria were not determined in the present study. Of 12 nucleotide positions found by Woese (1987) to be useful in distinguishing bacteroides from flavobacteria: 4 would place XM3 with the bacteroides; 4 would place it with the flavobacteria; and 1 is more typical of other eubacteria. Three others remain to be determined. The partial sequence for strain XM3, based on three internal nonoverlapping fragments, has been deposited with GenBank (accession numbers M80585, M80586, and M80587), or it can be obtained by writing the authors directly. Cells of CytOphaga XM3 were catalase and oxidase negative when first isolated, but became weakly catalase positive upon prolonged laboratory subculture. Other isolates were also weakly catalase positive, and all but 2 marine strains were oxidase negative (Table 2). The DNA base composition of strain XM3 was 45.5 molZ G + C. Xylan-degrading enzymes Oat spelt xylan-grown CytOphaga XM3 possessed xylanase activity which was largely cell-associated (912) and which exhibited pH and temperature optima of 6.8 and 45°C, respectively (Table 5). The extracellular activity did not appear to be an artifact of cell lysis, inasmuch as 7 glycosidase activities assayed in similar experiments were almost exclusively cell-associated (Table 6). 63 Table 4. Cellular fatty acids in freshwater isolates of Cytophaga xylanolytica . X of Total Fatty acidsb XM3 SL1 MA3 EW Straight chain c n 14:0 ND ND 3 7 ND n 15:0 19.9 1 3 4 12.4 27 4 42.7 n 16:0 4.1 1 1 8 1.4 1 1 ND Branched 1 14:0 ND ND 3.7 ND i 15:0 3.7 1 0.8 3.7 10.2 4.0 a 15:0 57.2 1 8.7 71.0 49.3 45.1 a 17:0 1.8 1 0.3 2.6 ND ND 3-Hydroxy n 15:0 5.5 1 2.0 1.7 4.3 8.1 a 17:0 3.3 1 0.6 5.2 ND ND Total 94.1 1 1.6 97.9 99.7 100 8Values for strain XM3 are the mean 1 SEM of 3 independent determinations. Values for other strains are the results from a single analysis. bC atomszdouble bonds. n, normal; 1, iso-branched; a, anteiso- branched. cND, not detected. 64 Table 5. xylanase activity of CytOphaga xylanolytica XM3 Cell-associated Extracellular activity activity pH optimum 6.8 6.8 50% activity 5.5 and 8.0 5.5 and 8.0 Temperature (0C) optimum 45 45 50% activity 30 and 49 26 and 56 Specific activity8 1.35 4.14 2 of total culture activity 90.3 9.7 a . . . . umol reducing sugar liberated per min per mg protein. Substrate was 0.5% (final conc.) oat spelt xylan (Lot #116F-0240; Sigma Chem. Co.). 65 Table 6. Cell-assgciated glycosidase activities of CytOphaga xylanolytica XM3 . Specific Activityb a-D-glucopyranosidase 0.344 B-D-glucopyranosidase 0.671 a-D-galactopyranosidase 0.703 B-D-galactopyranosidase 0.384 B-D-xylopyranosidase 5.156 B-D-fucopyranosidase 0.184 a-L-arabinofuranosidase 4.628 a . . . . a-D- and B-D-mannopyran051dase and acetylesterase act1v1t1es were not detected. Except for a trace of a-L-arabinofuranosidase, no activity of any of the enzymes tested was found in extracellular culture fluid. umol p-nitrophenol released per min per mg protein. 66 Xylanase-containing cell extracts (9.7-11.2 pg protein/ml) typically hydrolyzed about 501 of added oat spelt xylan (500 ug/ml) after 1 h at 40°C and pH 6.8. A maximum of 71.9% hydrolysis was observed after 48 h. Addition to reaction mixtures of fresh xylan resulted in an abrupt increase in the rate of release of reducing sugar, but addition of fresh enzyme did not. This observation suggested that some glycosidic linkages present in the xylan substrate were resistant to attack by XM3 xylanase enzyme(s). Strain XM3 possessed no a-D- or B-D- mannopyranosidase or acetylesterase activity (Table 6); and CMCase specific activity was typically 100- to 1000-fold lower than that of xylanase (data not shown). The latter observation was consistent with the inability of XM3 to grow on various cellulases or CMC (Table l). 67 Discussion Taxonomy of xylanolytic isolates The generic epithet "Cytophaga" currently embraces a heterogeneous assembly of chemoorganotrophic, free-living, rod-shaped gliding bacteria that have the abiltity to degrade a variety of biopolymers, but which form neither resting cells nor fruiting bodies (Reichenbach 1989c). It is a genus that is likely to undergo extensive revision once the natural relationships amongst its members become more precisely defined (Reichenbach 1989c, personal communication). However, until such revision dictates otherwise it seems reasonable to assign the isolates described herein (both freshwater and marine strains) to CytOphaga, inasmuch as most of their phenotypic properties are consistent with the genus description. In particular we note their gliding motility, uniform cellular morphology, and free-living habitat which, taken together, are inconsistent with the decriptions of other genera within the CytOphagaceae (Reichenbach 1989b) or CytOphagales (Reichenbach 1989a). Our decision is supported by the apparent presence in strain XM3 of sulfonolipids, a class of lipids widely distributed among members of the CytOphaga supergroup (Godchaux and Leadbetter 1983). Furthermore, the major soluble fermentation products of freshwater and marine isolates (i.e. acetate and propionate 1 succinate; Tables 1-3) are the same as those formed by facultatively anaerobic Cytophaga species during fermentative growth (Reichenbach 1989c). Nucleotide sequence analysis of the 16S rRNA of strain XM3 was 68 not especially helpful in generic assignment, because such data are almost nonexistant for cytophagas. Nevertheless, results of such analysis placed strain XM3 within the Bacteroides - Flavobacterium subgroup of the eubacteria. This is the same subgroup to which CytOphaga johnsonae and Flavobacterium heparinum [= CytOphaga heparina, Christensen (1980)] have been assigned (Weisburg et a1. 1985). The freshwater strains described herein possess several properties which make them distinct from existing CytOphaga species, however. Most notable among these is their relationship to oxygen. Whereas all known CytOphaga species are strict or (rarely) facultative aerobes, the present freshwater isolates are strict anaerobes which, although aeroduric, will not grow under conventional oxic conditions nor will use oxygen as terminal electron acceptor for metabolism (strain XM3). Fatty acid analyses (Table 4) revealed additional contrasts. Cellular fatty acid profiles of known CytOphaga species are dominated by C15:0 branched isomers of the iso-type, whereas in the present isolates these were anteiso. Where anteiso-fatty acids have been found to occur in CytOphaga as major constituents, as in C. hutchinsonii, they have been C17:0 (Reichenbach 1989c, and references therein). By contrast, little or no anteiso C17:0 fatty acids were detected in the present isolates. Likewise hydroxy fatty acids, which occur in relatively large amounts in cytophagas (IS-552 of total), were present in low amount in the current isolates (4-92) and were identified as 3-OH-n-C and 3-OH-a-Cl7zo, two forms poorly 15:0 represented among known CytOphaga species (Reichenbach 1989c). In 69 addition, the salt tolerance of the present freshwater isolates (up to 3% NaCl) was much greater than that of other freshwater or terrestrial cytophagas, which rarely grow above 1 2 NaCl (Reichenbach 1989c). Finally, the mall G+C in the DNA (45.5; strain XM3) is nearly the highest of any CytOphaga, with the possible exception of C. (Flavobacterium) heparinum (40-46) and C. arvensicola (43-47) which are both strict aerobes (Reichenbach 1989c). We believe that the above-mentioned differences are significant enough to warrant creation of a new species, CytOphaga xylanolytica, to accommodate the xylan-degrading freshwater isolates described herein, with strain XM3 as the type strain. A description of C. xylanolytica is presented below. Regarding the recently-isolated marine cytophagas, it seems wisest to postpone taxonomy of these strains until they have been studied in greater detail. Ecological significance The repeated, ready enrichment and isolation of C. xylanolytica and xylanolytic marine cytophagas from various sites in the United States and southern Germany suggest that they are important to the decomposition of xylan in anoxic freshwater and marine sediments. As described above, insoluble (i.e. non-heat- sterilized) xylan particles in primary enrichments quickly became colonized by an attached mass of C. xylanolytica-like cells (Fig. 1a) which glided over, and grew to occupy, virtually every square um of surface area on the particles and which dominated their hydrolysis and initial fermentation. Not surprisingly, strain XM3 was found to possess most of the enzyme activities necessary for effective xylan 7O degradation (Biely 1985). Inasmuch as over 902 of the xylanase activity of C. xylanolytica XM3 is cell-associated (Table 5), we suspect that both attachment to, and gliding over, the surface of xylan particles were important in bringing the bulk of the cells’ xylanase activity into contact with its substrate and in allowing such enzymes to continually gain access to new, perhaps more labile, microsites for hydrolysis. That attachment and motility are competitive traits for xylan degradation by C. xylanolytica is supported by the observation that such cytophagas appeared to be far less abundant in enrichments established with heat-sterilized (hence solubilized) xylan, wherein other, non-gliding, bacteria came to the fore. Presumably, in vitro enrichment cultures with powdered xylan mimic natural habitats, wherein xylan-containing plant materials are normally presented to degradative microbial communities in insoluble, particulate form. Such enrichments allow for selection of xylan- degrading bacteria in which phenotypic properties other than xylan hydrolysis per as are equally (and perhaps more) relevant to their competitive success. However, xylan is never found in pure form in nature. It is normally complexed with, and secondary in abundance to, cellulose in plant cell walls (Biely 1985, Wilkie 1983). Therefore, effective decomposition of xylan by C. xylanolytica in nature probably occurs within microbial consortia containing cellulolytic bacteria, because C. xylanolytica strains could not grow on cellulose and possessed only low levels of CMCase activity (strain XM3). 71 Their attachment to, and conspicuous enrichment with, particulate xylan as opposed to heat-solubilized xylan, as well as their anaerobicity, probably both contribute to the obscurity of C. xylanolytica until now. Moreover, their inability to form dense, discrete, "typical" bacterial colonies on the surface of agar media, instead of film-like swarms, would decrease the chances that they would be noticed by the unsuspecting eye, a problem compounded by the tendency of cells to attach together to particles and be quickly diluted out if source material were diluted prior to direct plating (Reichenbach 1988). In retrospect, we might well have overlooked them ourselves if original anoxic enrichments were not prepared with powdered xylan and followed by isolation with agar shake tubes. These considerations suggest that the importance of C. xylanolytica to xylan degradation in anoxic habitats may not be fully appreciated. Description of CytOphaga xylanolytica Sp. nov. CytOphaga xylanolytica sp. nov. [xy.lan.o.ly'ti.ca. xylan xylan, a xylose-containing heteropolysaccharide in plant cell walls; Gr. v. lyein to loosen, dissolve; M.L. fem. adj. lytica; M.L. fem. adj. xylanolytica xylan- dissolving]. Slender, flexible, Gram negative rods with slightly tapering ends, 0.4 x 3-24 pm. Move by gliding motility. No endospores, cysts, or fruiting bodies formed. Cells orange (subsurface colonies in xylan agar medium) or orange to salmon (surface swarms on xylan agar) due to the presence of carotenoid pigment(s). Flexirubins absent. Subsurface colony morphology (lenticular or spherical) and size (2 mm or spreading up to 72 1 cm) in xylan agar both depend on agar concentration, the latter characteristics being prevalent in media solidified with <12 agar. Best growth occurs in NaZS-reduced, HCO3-/CO -buffered, anoxic 2 mineral media with mono-, di-, or polysaccharides as fermentable energy sources. Vitamins not required (strain XM3). Sulfide used as sole 8 source. Poor and inconsistent growth with sulfate, cysteine, or glutathione as sole S source. Methionine and thiosulfate not used as S 3 or N2), glutamine (but not glutamate), and peptone used as sole N sources. HCO3-/CO2 required or greatly 4. sources. NH“ (but not NO stimulatory for growth. All strains fermented xylan, laminarin, lichenin, cellobiose, lactose, glucose, galactose, mannose, xylose, and arabinose as energy sources for growth. Individual strains fermented starch, pectin, inulin, pullulan, maltose, sucrose, fructose, glucuronic acid, and galacturonic acid. None fermented cellulose, carboxymethylcellulose, arabinogalactan, mannan, chitin, N- acetylglucosamine, sorbose, ribose or rhamnose. Growth occurs within a pH range of 6.1-8.7 (optimum 7.2-8.2), a temperature range of 19-37OC (optimum 30-320C), and in media containing up to 3% NaCl. Aeroduric anaerobe; metabolism fermentative. Weakly catalase positive (may be negative when first isolated); oxidase negative. Acetate, propionate, and succinate are major products of xylan, xylose, and glucose fermentation. Small amounts of CO2 and H2 also detected in detailed analyses of fermentations by strain XM3. N03 and -2 . . . SO4 not used as terminal electron acceptors during anaerobic growth. No growth under conventional aerobic conditions, with or without CO2 enrichment of the atmosphere. 02 not used as terminal electron acceptor (only strain XM3 tested). 73 Cellular fatty acids dominated by C isomers, with anteiso > n 15:0 > iso forms. Hydroxy fatty acids include 3-OH-n-C and 3-OH-a-C 15:0 17:0 types. Preliminary analyses suggest the presence in cells of a sulfonolipid(s) of a type slightly less polar than N-acylcapnine (strain XM3). Partial nucleotide sequencing of the 16S rRNA of the type strain places it in the Bacteroides - Flavobacterium subgroup of the eubacteria. The molZ G+C in the DNA of the type strain is 45.5 (Bd). Source: Freshwater sediments and swine waste lagoon sediment. Type strain: Strain XM3; deposited with the Deutsche Sammlung von Mikroorganismen, Braunschweig (DSM No. 6779). 74 Acknowledgements. This research was supported in part by National Science Foundation grants DIR88-09640 to the Center for Microbial Ecology and DCB86-14756 to J.A.B.; and by a Barnett Rosenberg Fellowship from the College of Natural Science of Michigan State University to S.K.H. This work was initiated while J.A.B. was on sabbatical leave in the laboratory of Prof. Norbert Pfennig, Univ. of Konstanz, as a Senior US Scientist Awardee of the Alexander von Humboldt-Stiftung, FRG. J.A.B. wishes to thank the Humboldt-Stiftung for its support and Prof. Pfennig and his research group for advice, encouragement and hospitality during his stay. We are also grateful to Dr. J. Flossdorf (GBF, Braunschweig) for analysis of the DNA base composition of strain XM3; to Drs. E. R. Leadbetter and W. Godchaux for sulfonolipid analysis; to Dr. D. Stahl for help and advice on 16S rRNA sequencing; and to H.S. Pankratz for help with electron microscopy. 75 References Biely P (1985) Microbial xylanolytic systems. Trends Biotechnol 3:286- 290 Breznak JA, Pankratz HS (1977) In situ morphology of the gut microbiota of wood-eating termites [Reticulitermes flavipes (Kollar) and Captotermes formosanus Shiraki]. Appl Environ Microbiol 33:406-426 Breznak JA, Switzer JM (1986) Acetate synthesis from H2 plus CO2 by termite gut microbes. Appl Environ Microbiol 52:623-630 Breznak JA, Switzer JM, Seitz H-J (1988) Sporomusa termitida sp. nov., an H /CO -utilizing acetogen isolated from termites. Arch Microbiol 150: 82- 88 Christensen P (1980) Description and taxonomic status of CytOphaga heparina (Pazya and Korn) comb. nov. (basionym: Flavobacterium heparinum Pazya and Korn 1956). Int J Syst Bacteriol 30:473-475 Cord-Ruwisch R (1985) A quick method for the determination of dissolved and precipitated sulfides in cultures of sulfate-reducing bacteria. J Microbiol Meth 4:33-36 Costilow R (1981) Biophysical factors in growth. In: Gerhardt P (ed) Manual of methods for general bacteriology, chapter 6. American Society for Microbiology, Washington, D.C. Flossdorf J (1983) A rapid method for the determination of the base composition of bacterial DNA. J Microbiol Meth 1:305-311 Follett EAC, Webley DM (1965) An electron microscope study of the cell surface of CytOphaga johnsonii and some observations on related organisms. Antonie van Leeuwenhoek J Microbiol Serol 31:361-382 Friedemann TE, Haugen GE (1943) Pyruvic acid II. The determination of keto acids in blood and urine. J Biol Chem 147:415-442 Godchaux W III, Leadbetter ER (1983) Unusual sulfonolipids are characteristic of the Cytophaga-Flexibacter group. J Bacteriol 153:1238-1246 Greve LC, Labavitch JM, Stack RA, Hungate RE (1984) Muralytic activities of Ruminococcus albus 8. Appl Environ Microbiol 47:1141- 1145 Herbert D, Phipps PJ, Strange RE (1971) Chemical analysis of microbial cells. In: Norris JR, Ribbons DW (eds) Methods in microbiology, vol. 5B. Academic Press Inc., New York, pp 209-344 76 Hermann M, Noll KM, Wolfe RS (1986) Improved agar bottle plate for isolation of methanogens or other anaerobes in a defined gas atmosphere. Appl Environ Microbiol 51:1124-1126 Lane DJ, Pace B, Olsen GJ, Stahl DA, Sogin ML, Pace NR (1985) Rapid determination of 16S ribosomal RNA sequences for phylogenetic analyses. Proc Natl Acad Sci USA 82:6955-6959 Nelson N (1944) A photometric adaptation of the Somogyi method for the determination of glucose. J Biol Chem 153:375-380 Oyaizu H, Komagata K, Amemura A, Harada T (1982) A succinoglycan- decomposing bacterium, CytOphaga arvensicola sp. nov. J Gen Appl Microbiol 28:369-388 Pfennig N, Wagener S (1986) An improved method of preparing wet mounts for photomicrographs of microorganisms. J Microbiol Meth 4:303-306 Potrikus CJ, Breznak JA (1977) Nitrogen-fixing Enterobacter agglomerans isolated from guts of wood-eating termites. Appl Environ Microbiol 33:392-399 Reichenbach H (1988) Gliding bacteria in biotechnology. In: Rehm H-J, Reed G (eds) Biotechnology, vol. 68. VCH Verlagsgesellschaft, Weinheim pp 673-696 Reichenbach H (1989a) Order I. Cytophagales Leadbetter 1974, 99AL. In: Staley JT, Bryant MP, Pfennig N, Holt JG (eds) Bergey's manual of systematic bacteriology, vol. 3. Williams & Wilkins, Baltimore pp 2011-2013 Reichenbach H (1989b) Family I. CytOphagaceae Stanier 1940, 630,AL emend. In: Staley JT, Bryant MP, Pfennig N, Holt JG (eds) Bergey’s manual of systematic bacteriology, vol. 3. Williams & Wilkins, Baltimore pp 2013-2015 Reichenbach H (1989c) Genus I. CytOphaga Winogradsky 1929, 577, AL emend. In: Staley JT, Bryant MP, Pfennig N, Holt JG (eds) Bergey’s manual of systematic bacteriology, vol. 3. Williams & Wilkins, Baltimore pp 2015-2050 Reichenbach H, Kleinig H, Achenbach H (1974) The pigments of Flexibacter elegans: novel and chemosystematically useful compounds. Arch Microbiol 101:131-144 Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain- terminating inhibitors. Proc Natl Acad Sci USA 74:5463-5467 Schink B, Pfennig N (1982) Fermentation of trihydroxybenzenes by Pelobacter acidigallici gen. nov. sp. nov., a new strictly anaerobic, non-sporeforming bacterium. Arch Microbiol 133:195-201 77 Skerman VBD (1967) A guide to the identification of the genera of bacteria. Williams & Wilkins, Baltimore Smibert RM, Krieg NR (1981) General characterization. In: Gerhardt P (ed) Manual of methods for general bacteriology, chapter 20. American Society for Microbiology, Washington, DC Umbreit WW, Burris RH, Stauffer JF (1957) Manometric techniques. Burgess Pub. Co., Minneapolis Weisburg WG, Oyaizu Y, Oyaizu H, Woese CR (1985) Natural relationship between bacteroides and flavobacteria. J Bacteriol 164:230-236 Widdel F, Kohring G-W, Mayer F (1983) Studies on dissimilatory sulfate-reducing bacteria that decompose fatty acids. III. Characterization of the filamentous gliding Desulfonema limicola gen. nov. sp. nov., and Desulfonema magnum sp. nov. Arch Microbiol 134:286-294 Widdel F, Pfennig N (1984) Dissimilatory sulfate- or sulfur-reducing bacteria. In: Krieg NR, Holt JG (eds) Bergey’s manual of systematic bacteriology, vol. 1. Williams & Wilkins, Baltimore, pp 663-679 Wilkie KCB (1983) Hemicellulase. Chemtech 13:306-319 Woese CR (1987) Bacterial evolution. Microbiol Rev 51:221-271 Zechmeister L (1962) Cis-trans isomeric carotenoids, vitamins A, and arylpolyenes. Springer-Verlag, Wien Chapter Two The xylan-degrading enzyme system of Cytophaga xylanolytica 78 79 Abstract. xylanase activity of Cytophaga xylanolytica strain XM3 was mainly (90-95t) cell-associated, the remainder was secreted into the culture fluid. The pH and temperature optima (6.8 and 45 0C, respectively) were similar for both the cell-associated and extracellular activities. The extracellular activity was not sedimentable by ultracentrifugation, however approximately half of the cell-associated activity was associated with cell wall fragments. Treatment of whole cells with 0.2% Triton X-100 released 74% of the xylanase activity without lysing the cells, suggesting that xylanase was associated with the outer membrane and/or with the periplasmic space. This was supported by transmission electron microscopy of Triton-treated cells. Isoelectric focusing of Triton extracts of cells, followed by activity stains, suggested that xylose- and xylan- grown cells had common endoxylanaee (pI 4.3), arabinofuranosidase (pI 5.85), and xylosidase (pI 4.6) (iso)enzymes. However, xylan-grown cells produced additional xylanase activity-bands (pI 4.5 - 5.1 and > 5.5) as well as several lichenanases. An arabinofuranosidase with the same pI as the cell associated protein was found in supernatants of both xylan- and xylose-grown cultures, but neither supernatant contained a fi-xylosidase. Both supernatants also exhibited endoxylanaee activity bands not found in the Triton extracts. Xylanase and lichenanase were fully induced only when cells were grown on the homologous polymer; however, cells retained low, constitutive levels of xylanase and lichenanase when grown on monosaccharide components of the respective polymer, and even when grown on other polymers. Glycosidase activities followed similar patterns. 80 Next to cellulose, hemicelluloses are the most abundant polysaccharides in nature (Wilkie 1983), yet our understanding of the decomposition of these heteropolysaccharides by microbes is limited (Biely 1985, Wong et al. 1987). Xylans are a major class of cell-wall hemicellulose widely distributed among both terrestrial and aquatic plants. They are complex polymers consisting of al3-1,4-linked xylopyranoside backbone variously substituted withci-1,3- arabinofuranoside, 0.9 M NaCl. Neither NaCl in concentrations up to 1.5 M nor a reduction of the pH to 4.0 would remove the yellow material from the DEAE—Sephadex A50 column. Therefore, a batch procedure was employed as follows: DEAE-Sephadex A50 previously equilibrated with 0.01 M potassium phosphate buffer (pH 7) was added to the crude extracellular fraction at a ratio of 1:6 (v/v). The mixture was stirred gently at 4°C for 30 min, then filtered through a 0.45 um filter (Millipore GSWP) under vacuum, being careful not to allow the gel to become dry. The filtrate was designated Fraction 1 (F1). The gel was washed from the filter with 10 ml of 0.1M potassium phosphate buffer (pH 7) containing 1.2 M NaCl, stirred at 4 °C for 60 min, then filtered again, this time allowing the gel to become dry; this filtrate was designated Fraction 2 (F2). These fractions were each concentrated (Amicon, PM10) to 1-2 mg protein/ml. Fractionation of cell-associated xylanase activity Since xylan was believed to be too large a molecule to enter the cells, it was assumed that the cell-associated activity must be located in or on the cell wall; therefore, various treatments were employed in an effort to extract the cell-associated activity without disrupting the cells. Unsonicated cell pellets were suspended to 1/10 the original culture volume with each of the following solutions: medium containing no added substrate; distilled water; 50 mM EDTA; 0.2 or 1 M NaCl; 50 mM potassium phosphate buffer (pH 6.8); 0.1M 87 HEPES/CaCl2 (pH 6.8); or 0.01% or 0.1% Triton x-1oo (w/v; dissolved in either distilled water or 50 mM potassium phosphate buffer pH 6.8). The suspensions were mixed gently for 30 min at 4 0C. In a separate experiment, cells were resuspended in 1/50 culture volume (1.5 m1) of phospholipase C (Celesk and London 1987; Sigma Chem. Co., 62.5 U) in 0.15 M triethanalamine (pH 7.6) containing 0.03 M CaC12. Cells were kept at 37 °C for 3 h with periodic gentle mixing before being brought to 1/10 culture volume with 0.01 M HEPES/CaClz. The extracts (supernatants) and cells (sonicated as previously described) obtained from each of the above treatments after centrifugation at 10,000 x g (20 min, 10 °C), were assayed for distribution of protein and xylanase activity. In an additional experiment, cells were subjected to osmotic shock essentially as described in Huang and Forsberg 1987. Cells previously washed with 0.1 M HEPES/CaCl2 were sequentially suspended in sucrose/EDTA, then in ice cold distilled water. After centrifugation at 30,000 x g (20 min, 10 0C), the resulting supernatant was collected as the periplasmic fraction. The cell pellet, resuspended in 0.01 M HEPES/CaClz, was sonicated and centrifuged again to obtain membrane (pellet) and cytoplasmic (supernatant) fractions. All supernatants, as well as designated fractions, were assayed for distribution of protein and xylanase activity. Extraction with 0.1% Triton x—1oo (whether dissolved in water or phosphate buffer) removed the most protein and activity from the 88 cells. For all subsequent experiments, 0.2% (chosen to optimize the detergent to extractable protein ratio) Triton x-1oo (w/v in distilled water) was used to extract xylanase activity from the cells. The 0.2% Triton extract from xylan-grown cells was further subjected to fast protein liquid chromatography (FPLC) Triton extract (4.1 ml containing ca. 3.4 mg protein) was applied to a MONO-Q column equilibrated with 50 mM HEPES pH 7.1. Fractions were eluted with a linear gradient of 0-2 M NaCl. The flow rate was 1 m1/min, and 1 ml fractions eluted from the column were each analyzed for protein (A280 and Lowry) and xylanase activity. Fractions exhibiting xylanase activity were combined (4-7 ml), concentrated to ca. 0.5 ml (Amicon, PM10). Triton x-1oo was then added to a final concentration of 0.2% and the solution was further concentrated (Centricon 10) to 1-4 mg protein/ml. Isoelectric focusing and activity stains Analytical thin layer IEF was performed by using the Bio-Rad Bio- Phoresis Horizontal Electrophoresis Cell and application protocols supplied by the manufacturer. Typical procedures employed 1% agarose gels (0.8 mm) with Bio/Lyte ampholytes and both Bio-Rad and Pharmacia (low pH range) IEF standards. Gels were stained for protein with Coomassie R-250 (Bio-Rad). After IEF the separation gel was overlayered with a 0.4 mm thick activity gel of 1% agarose (prepared in 0.1M HEPES/CaClz) cast on agarose support film (Bio-Rad) as described in Lee and Forsberg (1987) and Biely and Markovic (1988). Activity gels contained either 0.5-1% Remazol Brilliant Blue-dyed xylan (RBB-xylan, Biely 1985), 1% oat spelt xylan (Lot # 116F-0240), 89 1% lichenan, or 0.1-0.2% 4-methylumbelliferyl-glycosides (MU- glycosides). All substrates were from Sigma Chem. Co. Activity gels were prewarmed to 40 oC and the sandwich was kept at 40 °C for 1-2 h (RBB-xylan, oat spelt xylan or lichenan) or 10 min-1 h (MU- glycosides). MU-glycoside hydrolysis was visualized by exposure of the gels to UV light (Biely and Markovic 1988). RBB-xylan hydrolysis was observed as a clear zone after 30 min- 1 h rinse with 0.05 M acetate buffer (pH 5.4) in 95% ethanol (1:2, v/v, Biely et al. 1985). Cat spelt xylan and lichenan hydrolysis were observed as clear zones after staining the activity gel with Congo Red (1 mg/ml for lichenan; 10 mg/ml for xylan) for 15 min with subsequent rinsing in 1 M NaCl (Teather and Wood 1982). Since freshwater strains of Cytophaga have been reported to hydrolyze agar (Agbo and Moss 1979), a control activity gel containing no added substrate was tested on one occasion. Other procedures Previously described methods were used for electron microscopy (Breznak and Pankratz 1977) and high performance liquid chromatography (HPLC; Breznak and Switzer 1986). To test for a requirement by the cells for contact with xylan for growth, a method similar to that of Kauri and Kushner (1985) was used in which xylase- or xylan-grown cells were inoculated onto the surface of 0.05 um filters (Millipore) held on the surface of agar medium plates containing either 0.4% xylan or no added substrate. 90 Results Characteristics of crude enzyme fractions General characteristics of the crude extracellular and cell-associated fractions were previously reported (Haack and Breznak 1992). Both exhibited similar pH and temperature optima (6.8 and 45 0C, respectively). Both fractions exhibited 50% or greater activity from pH 5.5 to 8.0, and from 30 to 50 oC. The crude cell-associated fraction maintained stable levels of activity for 3 h at 30 °C in the pH range 5.5-8.3. The crude cell-associated fraction typically hydrolyzed 40-50% of added (0.5%) oat spelt xylan in 1 h at 30 °C and pH 6.8; a maximum hydrolysis of 70% was obtained in one assay. The decrease in activity after 1 h was not due to enzyme instability, because addition of fresh xylan at 3 h caused an immediate increase in activity, but addition of fresh enzyme did not. In addition to xylanase activity, the crude cell-associated fraction exhibited 6 mo. when stored at 4 oC. Triton treatment did not appear to cause marked disruption of cells, which retained their rod—shaped morphology and phase-dark Character when observed by phase contrast microscopy. However, 93 electron microscopy revealed that the Triton treatment appeared to draw the outer membrane Closer to the cell wall, thereby collapsing somewhat the rather thick periplasmic space typical of untreated cells (Fig. 1). Moreover, vesicles found as outer membrane blebs or free in the intracellular space around untreated cells were not as apparent in the Triton-treated preparations (Fig. 1). Similarly, after osmotic shock, 38.6% of the cell-associated activity was cytoplasmic, 47.8% was found in the membrane fraction, and 13.5% was periplasmic. In this same experiment, virtually alllg-D-galactosidase activity was associated with the cytoplasmic fraction, while 16.3%, 71.0%, and 12.8% of thefl3-D-xylosidase activity was membrane associated, cytoplasmic, or periplasmic, respectively. Together, the results from Triton treatment as well as from osmotic shock suggested a cell wall (both membrane and periplasmic space) location for the majority of the xylanase activity of C. xylanolytica XM3, Fractionation of xylanase activity Treatment of the culture supernatant of xylan-grown cells with DEAE- Sephadex A50 removed (presumed) residual carbohydrate and yielded two fractions with combined protein and activity representing approximately 50% of that found in the crude supernatant fraction (Table 1). This treatment was not required for the supernatants of xylose-grown cells. FPLC, used to fractionate xylanase activity present in the Triton extracts of xylan-grown cells, yielded four fractions with xylanase activity (FA-FD) and one non-active fraction (FX) (Fig.2). These 94 FIGURE 1. Transmission electron micrographs of thin sections of xylan-grown cells of Cytophaga xylanolytica strain XM3. 1A, 18; normal, untreated cells. 2A, 23; cells treated with 0.2% Triton x-100 for 1 hr at 4 °C. Bars: 1A, 2A, 0.5 um; 1B, 28, 0.2 um. 95 96 FIGURE 2. FPLC fractionation of protein and xylanase activity from the Triton extract of cells of Cytophaga xylanolytica XM3. MONO-Q column; linear NaCl concentration gradient in 0.05 M HEPES pH 7.1 FA: Fractions 1- 7; FX: Fractions 8-13; FB: Fractions 14-17; FC: Fractions 18-21; FD: Fractions 22-26. (L 10 97 Ail/\IiO‘v' BSVNV'IAX 'IViOi w x pesee|aJ Jsfins BugonpeJ '00:”) Q ‘0 V N _u 0 IO T I l l; " N “x o /09 ‘ N LU >. 33 I: Z < Z _l I- ). 0 X ‘1 l0 1- ‘-“ - I E 11.1 0 l- . O 1) I 0. if 4| 9 I II l0 0 e 0 l l l l o - «2 <0. '5. 0a o C C C C LJW X NlalOHd 9W J0 Ailth'lOW IOBN FRACTION 98 fractions together accounted for 100% of the protein but only 19% of the total activity applied to the column. The activity of individual fractions, when summed, accounted for only 8.5% of the total culture activity (Table 1). However, when these five fractions were recombined (maintaining the same relative proportions of total protein per fraction) the total activity was increased to ca. 14% of the total culture activty (Table 1), suggesting a synergistic interaction of enzyme activity between the fractions. Although FPLC fractionation apparently resulted in significant loss of enzyme activity, it allowed increased resolution of the pI's of endoxylanases in the 4.5-5.1 range (see below). Isoelectric focusing Analytical isoelectric focusing indicated that the Triton extracts of both xylan- and xylose-grown cells contained several proteins, all with pI < 6.5 (Fig. 3). Activity stains suggested that xylan- and xylose-grown cells possess common xylanase, arabinofuranosidase, and B-xylosidase (iso)enzymes (Fig. 3, Table 2); however, xylan-grown cells exhibited additional xylanase activity bands representing proteins with pI's ranging from 4.5 - 5.9, as well as several bands with lichenanase activity, none of which were easily separated by isoelectric focusing of the complete Triton extract (Fig. 3). Xylan- grown cells exhibited extracellular xylanase activity at five distinct pI's, whereas xylose-grown cells exhibited only three (Table 2). An arabinofuranosidase activity band with the same pI as the cellular protein was found in the supernatant of both xylan- and xylose-grown cells, but neither supernatant displayed a xylosidase activity band 99 TABLE 2. Isoelectric point (pI) of carbohydrases of C. xylanolytica XM3. XMAS! XYLOSIDASE ARABINOSIDABB LIME! llhbflzflfieflfl TRITON EXTRACT OI" CELL PELLET CULTURE SUPERNATANT 5 . 85 IXLQ§§:§BQ!! TRITON EXTRACT 4.3 4.6 5.85 OF CELL PELLET CULTURE SUPERNATAN T 100 FIGURE 3. Analytical isoelectric focusing of Triton extracts of xylan- or xylose-grown cells of Cytophaga xylanolytica strain XM3. A: Coomassie protein stain. B: activity stain; Bl, xylan hydrolysis (Remazol Brilliant Blue dyed xylan); 82, lichenan hydrolysis (Congo Red); 83, Methylumbelliferylxyloside hydrolysis (Photographed under UV illumination); 84, Methylumbelliferylarabinofuranoside hydrolysis. LANES: 1 and 4, p1 Standards (positions of selected pI markers are indicated); 2, Triton extract of xylan-grown cells; 3, Triton extract of xylose-grown cells. A _..v amé sod _a vamp vamp QMNP emww 102 (Table 2). This pattern suggested that the extracellular arabinofuranosidase activity was not the result of cell lysis. FPLC fraction FB, which contained the greatest total xylanase activity (Fig. 2), exhibited xylanase activity bands at pI 4.5, 4.7, 4.8, 5.0, and 5.4, as well as the arabinofuranosidase and the xylosidase activity bands (Table 2, Fig. 4). Three of the xylanase activity bands in FPLC fraction FB appeared to also have lichenanase activity; however, this fraction also contained an additional lichenanase activity band at pI 4.9 (Table 2). Some xylanase activity bands present in fraction FB also had analogues in fractions FC (pI 4.5) and FA (4.7 and 4.8); however, fractions FA and FC both contained a xylanase activity band at p! 4.3 which was not seen in fraction FB. Fraction FA also displayed a band at pI 5.9 (Table 2, Fig. 4). FPLC fraction FD exhibited no discrete activity band on IEF. No activity bands corresponding to)6-glucosidase or cellobiohydrolase were detected in FPLC fractions FA, F8, or FC after 1 h or 24 h incubation. Induction of enzyme activity In C. xylanolytica XM3, the enzymes for degrading xylan or lichenan were fully induced only when cells were grown on the appropriate polymer (Table 3); however, cells retained low constitutive levels of xylanase and lichenanase when grown on monosaccharide or disaccharide constituents of the appropriate polymer, and even when grown on other polymers (Table 3). Glycosidase activities followed similar trends (Table 3). C. xylanolytica could not grow on cellulose (ball-milled filter paper, or Sigmacell 20) or carboxymethylcellulose (Haack and Breznak 1992). Thus, it was not surprising that CMCase specific 103 FIGURE 4. Analytical isoelectric focusing of FPLC fractions of Triton extracts of xylan-grown cells of C. xylanolytica strain XM3. A: Coomassie protein stain. B: activity stain; clear zones after staining with Congo Red indicate xylan hydrolysis. Lanes: STDS, pI standards (positions of selected pI markers are indicated); FA, FB, FC, FPLC fractions (see Fig. 2). 104 s FCFBFA s FCFBFA 5 pl 6.0 > 5.1 p 4.5 F 105 0«0H.0 H000.0 va0.0 0000.0 02 h000.0 02 0000.0 0000.0 0««.0 «hm.a h0«.0 0«0.0 H00.0 000.0 0H0.0 0«0.0 H«0.0 0H0.0 «50.0 0H0.0 0«0.0 000.0 0am.0 0H0.0 000.0 000.0 0Hh.0 vo«.m 000.0 mvfl.n 00h.H 000.H 00«.« H50.0 000.0 0000.0 Hfi00.0 0000.0 «000.0 «000.0 0000.0 0000.0 0000.0 «000.0 HOH.0 mvm.0 050.0 ««0.0 0H0.0 0«0.0 Oz 02 m0«.0 000.0 0«0.0 000.0 000.0 0«0.0 H00.0 050.0 Hv0.0 ZGZMxOHA 000.0 02‘ Z¢Aux 000.0 Z¢ZNZOHA 0H.0 flmOHQOAQmO SE 0 mmOODAU :8 0a Mmoflhx :8 0 02¢ ZH0IH>COCQOMDACIQ no oa~08>aom se>a0 o Baum concedes Hoconmowu0CIw no woman meanness H09: 0 A.o.ucoo. m 04009 107 activity was 100- to 1000-fold lower than that of xylanase activties. Starvation for 48 hr after growth on xylose also did not result in induction of xylanase activity (data not shown). When grown in a medium in which xylose or lichenan was included with oat spelt xylan, late exponential phase cultures exhibited an appreciable level of xylanase (and, in the latter case, also lichenanase) activity, but which was still less than that observed when grown on xylan alone (Table 3). The induction of xylanase activity during cell growth on 5 mM xylose plus 0.1% xylan was investigated at several points in the growth curve (Fig. 5). Xylose steadily disappeared from the culture medium and was completely utilized by the time culture protein values had reached 100 ug/ml data not shown). Xylanase specific activity increased steadily during growth, and was approximately 5-fold greater than uninduced levels even at early time points in the growth curve when as much as 3 mM xylose remained in the medium; the greatest increases in specific activty occurred after xylose was depleted (ca. 150 ug/ml cell protein; Fig. 5). Cells grown on xylan alone exhibit fully induced xylanase activity at much lower cell densities (Fig. 5). Such cells do not appear to release xylose into the medium during growth on xylan (detection limit 0.1 mM). Other studies In experiments with C. xylanolytica XM3 in which cells were grown with combinations of sugars (5 mM xylose + 5 mM arabinose, or 5 mM cellobiose + 5 mM glucose), both substrates disappeared from the medium concurrently, as detected by HPLC (data not shown). 108 FIGURE 5. Effect of xylose on induction of xylanase activity by xylan in Cytophaga xylanolytica XM3. Xylose-grown cells were inoculated into medium containing only 0.1% oat spelt xylan (Sigma; Lot # 116F-0240) or 0.1% xylan + 4.2 mM xylose. 109 .8330 ...E x £065 :00 03 P I._.>>Om_0 4.50 00« 00.. 00F 00 .\r . _ . L! I‘ mwO..>x + z<._>x 20 _>_._.0< 0.0.0000 mw<2<4>x e 33: 20 9 5.2.52 Orzomem mmx I‘ e ‘\ ...0 «.0 0.0 #0 0.0 (bugsimd 5w x I__u!u: x Jefins 5U!Onp8.l lawn) Ail/\liOV OlleEdS 110 In a test similar to that of Kauri and Kushner (1985), it was found that C. xylanolytica (previously grown on either xylose or xylan) could grow on 0.05 um membrane filters which were placed on the surface of agar plates containing 0.4% xylan as sole carbon source. No growth occurred on filters placed on 1.5% agar alone, and Clearing of the underlying xylan could be seen when the filter was removed, although all growth was confined to the filter surface. 111 Conclusions The majority of the xylan-degrading enzyme system of C. xylanolytica XM3 is cell-associated. The effectiveness of Triton treatment in removing cell-associated activity, taken together with the intact appearance of the cells after Triton treatment, and the association of substantial activity with insoluble material after ultracentrifugation of sonicated cells strongly suggest that the xylanase activity of C. xylanolytica is largely associated with the outer membrane of the cell wall. However, the release of some xylanase by osmotic shock suggests that some of the xylanase activty may reside in the periplasmic space. An outer membrane or cell-surface location was originally proposed for Cytophaga polymer-hydrolyzing enzymes by Stanier in 1942, and this has been confirmed for several enzymes or enzyme complexes of cytophagas including agarase (Duckworth and Turvey 1969), protease(s) (Christison and Martin 1971), carboxymethycellulase (Chang and Thayer 1977), amylase (McKay 1991), and a dextranase (Janson 1975). In the closely-related Sporocytophaga myxococcoides, from 10% to 100% of the cellulase activity was reported to be located in the cell wall, and both groups found the activity to be extractable with Triton X-100 (Charpentier and Rabic 1974, Osmundsvag and Goksoyr 1975). Additionally, in Bacteroides a genus related to Cytophaga on the basis of 16S rRNA homology (Reichenbach 1989), an outer membrane mannanase (B. ovatus; Gherardini and Salyers 1987), an inner membrane associated polygalacturonic acid lyase (B. thetaiotaomicron; McCarthy et a1. 1985), and a protease located on the surface of the outer membrane (B. gingivalis; Grenier and McBride 1989) could all be released from 112 isolated membranes by Triton x-100. In no case is the exact nature of the interaction of these enzymes with the cell wall understood, and in almost all the cytophaga enzyme studies Cited above, some activity could also be found in the culture supernatant. Both Duckworth and Turvey (1969) and Christinson and Martin (1971) proposed an interaction of cell-associated enzymes with extracellular polysaccharide (slime) which is produced by many Cytophaga species (Reichenbach 1989). Bacon et al. (1970) found yeast-cell-wall-degrading enzymes of C. johnsonae to be primarily extracellular, although some activities were cell-associated, and Agbo and Moss (1979) noted both extracellular and cell-associated enzyme complexes after growth of several cytophaga isolates on agar. Vance et al. (1980) reported the cellulase of S. myxococcoides to be extracellular. However, in that study the cells were grown in a stirred fermentor with a surfactant added to control foaming, and the cellulase could have been stripped from the cells by this procedure. Inasmuch as most of the xylanase of C. xylanolytica is cell- associated, it would seem that the most efficient dissociation of xylan by this bacterium would occur when cells are in direct contact with insoluble xylan---a situation entirely consistent with the lifestyle of the gliding bacterium. However, C. xylanolytica does exhibit extracellular xylanase activity, and such activity does not appear to be an artifact of cell lysis, as the supernatants and Triton extracts of either xylan- or xylose-grown cells exhibited different patterns of xylanase and)8-xylosidase activity hands after 113 electrophoresis. Since C. xylanolytica could grow on microporous membrane filters placed on the surface of 0.4% oat spelt xylan plates, and in so doing hydrolyze the xylan beneath the filter, as Kauri and Kushner found for cellulolytic Cytophaga spp., C. xylanolytica apparently does not absolutely require direct contact with a polymeric growth substrate in order to utilize it. This further suggests that the extracellular activity of this bacterium may specifically function to attack polymeric substrates at some distance from the cell. The fact that even xylose-grown cells, though not fully induced for xylanase activity, still displayed extracellular xylanase activity bands on IEF gels, may indicate that this extracellular activity serves an important function in initiating degradation of polymeric substrates to produce fragments which in turn serve to induce full xylanase activity. C. xylanolytica appears to have a complex enzyme system which is induced by xylan and is relatively specific for xylan degradation. Clearly, growth on xylan increases the xylanolytic activity of the cells and results in the appearance of a multitude of new xylanase activity hands after isoelectric focusing. Multiple xylanase activity bands have been observed by others, but the significance of multiple IEF activity bands is not yet Clear. Multiple xylanase genes have been clearly identified in several bacterial species (Bacillus: Yang et al. 1989, Hamamoto et al. 1987, Honda et al. 1985, Sakka et al. 1989 and 1990; Clostridium: Schwarz et al. 1990, Mackenzie et a1. 1989, Hazlewood et al. 1988; Streptomyces: Kluepfel et al. 1990, Vats- Mehta et al. 1990; Ruminococcus: Flint et al. 1989 and 1990; 114 Pseudomonas: Gilbert et al. 1988). In addition, two xylosidase genes have been identified in Bacillus (Panbangred 1984). However, a 2.8 kb genomic fragment from Clostridium thermocellum Cloned into E. coli encoded a single molecular weight band (25,000 kD) on SDS-PAGE gels but multiple (at least 4) bands on IEF gels (MacKenzie et al. 1989). Studies with proteins isolated from the host species have produced similar results. For example, in Thermomonospora fusca, Bachmann and McCarthy (1991) demonstrated that a single 32-kD endoxylanaee produced three activity bands at pI 7.9, 8.2 and 8.6. This 32-kD protein, itself active on denaturing PAGE gels developed as zymograms, was proposed to be a subunit of a yet larger protein which was active on native PAGE gels. A second proposed subunit of 24 kD was presumed to account for three additional IEF activity bands at lower pI's. Three bands of acetyl-esterase activity revealed by IEF, were correlated with an 80-kDa intracellular protein secreted to yield 40- kD active subunits. Similarly, purified endoxylanaee 1 from Fibrobacter succinogenes (Matte and Forsberg 1992) produced a single band of 53.7 kDa by SDS-PAGE analysis, but four activity bands at pI 7.3, 8.1, 8.3 and 8.9 after IEF. This heterogeneity may be due to a variety of postranslational events, including subunit activity as noted above, glycosylation (Kluepfel et a1. 1990), or proteolytic processing (Biely 1985). In preparations containing several proteins, multiple activity bands may also be due to the possession of multiple activities by a single protein, such as the apparent lichenanase activity of three of the endoxylanaee activity bands in our study and also observed in several of the purified or Cloned enzymes noted above. Finally, heterogeneous activities of purified proteins or 115 Cloned gene products toward the commonly used test substrates have been described (Kluepfel et al. 1990, Gilkes et al. 1984, Sakka et al. 1989). For example, MacKenzie et al. (1989) observed that one gene product from C1. thermocellum formed Clear halos in RBB-xylan plates, clearings on Congo Red-stained xylanase activity gels, and substantial quantities of reducing sugar in typical xylanase assay procedures. However, two additional Clones also formed Clear halos on RBB-xylan, and on Congo Red-stained aspen or larchwood xylan activity gels, but not on oat spelt xylan activity gels, and both converted less than 1% of xylan substrates to reducing sugar. Determination of the exact nature and significance of the multiple IEF activity bands in enzyme preparations from C. xylanolytica XM3 must await further analysis. C. xylanolytica XM3 grows on few compounds other than sugars or sugar polymers, but can grow on a diverse array of both (Haack and Breznak 1992), and it appears to be able to utilize more than one sugar substrate concurrently. We believe that cytophagas grow primarily on polymers in nature and we assume that the natural habitat of C. xylanolytica and other cytophagas is a complex matrix of biopolymers, varying in degree of substitution and linkage, over and through which these bacteria glide or swarm. Potentially, an individual cell will encounter a multitude of sugar residues, in a variety of linkages, within relatively short spatial or temporal intervals. In this environment, the ability to completely degrade a given polymer may not be as important to survival as the ability to utilize many of the sugar residues, regardless of linkage or parent molecule association. Hofle (1982) demonstrated that C. johnsonae 116 possessed two different glucose uptake systems, both of which were unspecific in that mannose and D-glucosamine or N-acetyl-D-glucosamine also appeared to be transported by these uptake systems. He proposed that this would be advantageous in environments where monomer components of the degradation of cellulose, starch, mannan and Chitin would be available concurrently. Agbo and Moss (1979) showed that agar-degrading enzyme complexes in freshwater cytophaga species were induced by agar, but also (and occasionally at higher levels), by galactans and other plant cell wall polysaccharides. These complexes also degraded a variety of polysaccharides. Similarly, whether grown on xylose, glucose, cellobiose or lichenan, C. xylanolytica retained the ability to respond to xylan through constitutive expression of xylanase, xylobiase and arabinofuranosidase enzymes. Xylan-grawn cells produced additional endoxylanases, and retained lichenanase activity, while cells grown on xylan plus lichenan expressed induced levels of both depolymerase activities. Although C. xylanolytica does not release xylose or arabinose into the medium during growth on xylan, the presence of xylose in the medium with xylan (potentially encountered in real environments through the degradative activities of other microbes) did not repress xylanase activity. Apparently, C. xylanolytica and other cytophagas are well adapted, through the location and regulation of their polymer degrading activities, for growth in a complex insoluble matrix of multiple potential growth substrates. However, C. xylanolytica cannot grow on or degrade cellulose, nor can it effectively degrade more complex plant cell wall derived substrates containing xylan, yet it 117 possesses a complex enzyme system which is inducible by xylan. This suggests that in the anoxic environments where C. xylanolytica exists, xylan residues are sufficiently exposed by the activities of other polymer-degrading microorganisms to warrant the retention of this enzyme system as one of several depolymerizing systems ecologically advantageous to these cells. It is anticipated that continued study of these ubiquitous biopolymer degrading cytophagas will greatly increase our understanding of the biochemistry, and microbial ecology of polymer degradation in natural environments. 118 References Agbo JA, Moss MO (1979) The isolation and characterization of agarolytic bacteria from a lowland river. J Gen Microbiol 115:355-368. Bachmann SL, McCarthy AJ (1991) Purification and cooperative activity of enzymes constituting the xylan—degrading system of Thermomonospora fusca. Appl Environ Microbiol 57:2121-2130 Bacon JSD, Gordon AH, Jones D, Taylor IF, Webley DM (1970) The separation of,6-glucanases produced by Cytophaga johnsonii and their role in the lysis of yeast cell walls. Biochem J 120:67-78 Biely P (1985) Microbial xylanolytic systems. Trends Biotechnol 3: 286-290 Biely P, Markovic O (1988) Resolution of the glycanases of Trichoderma reesei with respect to cellulose and xylan degradation. Biotech Appl Biochem 10:99-106 Biely P, MacKenzie CR, Puls J, Schneider H (1986) Cooperativity of esterases and xylanases in the enzymatic degradation of acetyl xylan. Bio/Technology 4: 731-733 Biely P, Markovic O, Mislovicova D (1985) Sensitive detection of endo-l,4-beta-glucanases and endo-l,4-beta-xylanases in gels. Anal Biochem 144: 147-151 Biely P, Puls J, Schneider H (1985) Acetyl xylan esterases in fungal cellulolytic systems. FEBS Lett 186:80-84 Bornemann WS,, Hartley RD, Morrison WH, Akin DE, Ljungdahl, LG (1990) Feruloyl and p-coumaroyl esterase from anaerobiic fungi in relation to plant cell wall degradation. Appl Microbiol Biotechnol 33:345- 351 Breznak JA, Pankratz HS (1977) In situ morphology of the gut microbiota of wood-eating termites [Reticulitermes flavipes (Kollar) and Coptotermes formosanus Shiraki]. Appl Environ Microbiol 33: 406-426 Breznak JA, Switzer JM (1986) Acetate synthesis from H2 plus CO2 by termite gut microbes. Appl Environ Microbiol 52: 623-630 Celesk RA, London J (1980) Attachment of oral Cytophaga species to hydroxyapatite containing surfaces. Infect Immun 29:768-777 Chang WTH, Thayer DW (1977) The cellulase system of a Cytophaga species. Can J Microbiol 23:1285-1292 119 Charpentier M, Rabic D (1974) Degradation de la cellulose par un micro-organisme du sol: Sporocytophaga myxococcoides: Characterisation d'une exoglucanase. C R Acad Sc Paris, Serie D 279:863-866 Christison J, Martin SM (1971) Isolation and preliminary Characterization of an extracellular protease of Cytophaga sp. Can J Microbiol 17:1207-1216 Duckworth M, Turvey JR (1969) An extracellular agarase from a Cytophaga species. Biochem J 113:139-142 Ferguson WJ, Good NE (1968) Anal Biochem 104:300 Flint HJ, McPherson CA, Bisset J (1989) Molecular Cloning of genes from Ruminococcus flavefaciens encoding xylanase and,6(1-3,1- 4) glucanase activties. Appl Environ Microbiol 55:1230-1233 Flint HJ, McPherson CA, Martin J (1991) Expression of two xylanase genes from the rumen bacterium Ruminococcus flavefaciens 17 Cloned in pUC13. J Gen Microbiol 137:123-129 Gherardini FC, Salyers AA (1987) Characterization of an outer membrane mannanase from Bacteroides ovatus. J Bacteriol 169:2031- 2037 Gilbert HJ, Sullivan DA, Jenkins G, Kellett LE, Minton NP, Hall J (1988) Molecular cloning of multiple xylanase genes from Pseudomonas fluorescens subsp. cellulosa. J Gen Microbiol 134:3239-3247 Gilkes NR, Langsford ML, Kilburn DG, Miller RCJr, Warren RAJ (1984) Made of action and substrate specificities of cellulases from Cloned bacterial genes. J Biol Chem 259: 10455-10459 Grenier D, McBride 8 (1989) Surface location of a Bacteroides gingvalis glycylprolyl protease. Infec Immun 57:3265-3269 Greve LC, Labavitch JM, Stack RJ, Hungate RE (1984) Muralytic activities of Ruminococcus albus 8. Appl Environ Microbiol 47:1141- 1145 Haack SR, Breznak JA (1992) Cytophaga xylanolytica sp. nov., a xylan- degrading anaerobic gliding bacterium. Arch Microbiol (accepted) Hamamoto T, Honda H, Kudo T, Horikoshi K (1987) Nucleotide sequence of the xylanase A gene of alkalophilic Bacillus sp. strain C-125. Agric Biol Chem 51:953-955 Hazlewood GP, Romaniec MPM, Davidson K, Grepinet O, Beguin P, Millet J, Raynaud O, Aubert J-P (1988) A catalogue of Clostridium thermocellum endoglucanase, -glucosidase and xylanase genes Cloned in Escherichia coli. FEMS Microbiol Lett 51:231-236 120 Hespell RB, O'Bryan-Shah PJ (1988) Esterase activities in Butyrivibrio fibrisolvens strains. Appl Environ Microbiol 54:1917-1922 Hofle, MG (1982) Glucose uptake of Cytophaga johnsonae studied in batch and Chemostat culture. Arch Microbiol 133:289-294 Honda H, Rudo T, Horikoshi K (1985) Molecular Cloning and expression of the xylanase gene of alkalophilic Bacillus sp. strain C-125 in Escherichia coli. J Bacteriol 161:784-785 Huang L, Forsberg CW (1987) Isolation of a cellodextrinase from Bacteroides succinogenes. Appl Environ Microbiol 53:1034-1041 Johnson KG, Silva MC, MacKenzie CR, Schneider H, Fontana JD. (1989) Microbial degradation of hemicellulosic materials. App Biochem Biotech 20/21:245-258 Kaji A (1984) L-arabinosidases. Adv Carbohydr Chem Biochem 42: 383-394 Kauri T, Kushner DJ (1985) Role of contact in bacterial degradation of cellulose. FEMS Miicrobiol Lett 31:301-306 Kluepfel D, Vats-Mehta S, Aumont F, Shareck F, Morosoli R (1990) Purification and Characterization of a new xylanase (xylanase 8) produced by Streptomyces lividans 66. Biochem J 267:45-50 Lee SF, Forsberg CW, Rattray JB (1987) Purification and Characterization of two endoxylanases from Clostridium acetobutylicum ATCC 824. Appl Environ Microbiol 53: 644-650 MacKenzie CR, Bilous D (1988) Ferulic acid esterase activity from Schizophyllum commune. Appl Environ Microbiol 54:1170-1173 MacKenzie CR, Yang RCA, Patel GB, Bilous D, Narang S (1989) Identification of three distinct Clostridium thermocellum xylanase genes by molecular cloning. Arch Microbiol 152:377-381 Matte A, Forsberg CW (1992) Purification, Characterization, and mode of action of endoxylanases 1 and 2 from Fibrobacter succinogenes $85. Appl Environ Microbiol 58:157-168 McCarthy RE, Kotarski SF, Salyers AA (1985) Location and Characteristics of enzymes involved in the breakdown of polygalacturonic acid by Bacteroides thetaiotaomicron. J Bacteriol 161:493-499 ' McKay AM (1991) Cell-associated amylase activity in Cytophaga johnsonae. Lett Appl Microbiol 12:36-38 Nelson N (1944) A photometric adaptation of the Somogyi method for the determination of glucose. J Biol Chem 153: 375-380 121 Osmundsvag K, Goksoyr J (1975) Cellulases from Sporocytophaga myxococcoides. Eur J Biochem 57:405-409 Peterson GL (1977) Anal Biochem 70:241 Puls J, Schmidt 0, Granzow C. 1987.CJ-glucuronidase in two microbial xylanolytic systems. Enzyme Microb. Technol. 9: 83-88. Reichenbach, H (1988) Gliding bacteria in biotechnology. In: Rehm H-J, Reed G (eds.) Biotechnology, vol. 6B. VCH Verlagsgesellschaft, Weinheim, pp 674-696 Reichenbach, H (1989) Order I. Cytophagales Leadbetter 1974, 99. In: Staley JT, Bryant MP, Pfennig N, Holt JG (eds) Bergey's manual of systematic bacteriology, vol. 3. Williams and Wilkins, Baltimore London, pp 2011-2081 Reilly PJ (1981) Xylanases: Structure and function. Pages 111-129 in Hollaender A (Ed.), Trends in the Biology of Fermentation for Fuels and Chemicals. Plenum, New York. Sakka K, Furuse S, Shmada K (1989) Cloning and expression in Escherichia coli of thermophilic Clostridium sp. F1 genes related to cellulose hydrolysis. Agric Biol Chem 53:905-910 Sakka K, Kojima Y,, Yoshikawa K, Shimada K (1990) Cloning and expression in Escherichia coli of Clostridium stercorarium strain F-9 genes related to xylan hydrolysis. Agric Biol Chem 54:337-342 Schwarz WH, Adelsberger H, Jauris S, Hertel C, Funk 8, Staudenbauer WL (1990) Xylan degradation by the thermophile Clostridium stercorarium: cloning and expression of xylanase,lS-D-xylosidase, andCX-L- arabinofuranosidase genes in Escherichia coli. Biochem Biophys Res Commun 170:368-374 Somogyi M (1952) Notes on sugar determination. J Biol Chem 195:19-23 Stanier RY (1942) The cytophaga group: a contribution to the biology of myxobacteria. Bacterial Rev 6:143-196 Teather RM, Wood PJ (1982) Use of Congo red-polysaccharide interactions in enumeration and Characterization of cellulolytic bacteria from the bovine rumen. Appl Environ Microbiol 43: 777-780. Vance I, Topham CM, Blayden SL, Tampion J (1980) Extracellular cellulase production by Sporocytophaga myxococcoides NCIB 8639. J Gen Microbiol 117:235-241 Vats-Mehta S, Bouvrette P, Shareck F, Morosoli R, Kluepfel D (1990) Cloning of a second xylanase-encoding gene of Streptomyces lividans 66. Gene 86:119-122 122 Wilkie, KCB (1983) Hemicellulase. Chemtech 13:306-319 Wong KKY, Tan LUL, Saddler, JN (1988) Multiplicity of,6-1,4-xylanase in microorganisms:functions and applications. Microbiol Rev 52:305- 317 Wood TM, McCrae SI (1986) Studies of two low-molecular-weight endo- (l,4)16-D-xylanases constitutively synthesized by the cellulolytic fungus Trichodenma koningii. Carbohydr Res 148:321-330 Yang RCA, MacKenzie CR, Bilous D, Narang S (1989) Identification of two distinct Bacillus circulans xylanases by molecular cloning of the genes and expression in Escherichia coli. Appl Environ Microbiol 55:568-572 MICHIGAN STATE UNIV. LIBRARIES mllWWHIWINIHNIWMlWWW 31293008824017