Scholarly article on topic 'Opportunities to improve fiber degradation in the rumen: microbiology, ecology, and genomics'

Opportunities to improve fiber degradation in the rumen: microbiology, ecology, and genomics Academic research paper on "Biological sciences"

0
0
Share paper
Academic journal
FEMS Microbiology Reviews
OECD Field of science
Keywords
{""}

Academic research paper on topic "Opportunities to improve fiber degradation in the rumen: microbiology, ecology, and genomics"

ELSEVIER

FEMS Microbiology Reviews 27 (2003) 663-693

www.fems-microbiology.org

Opportunities to improve fiber degradation in the rumen: microbiology, ecology, and genomics

Denis O. Krause ab;*, Stuart E. Denman a, Roderick I. Mackie c, Mark Morrison d, Ann L. Rae a, Graeme T. Attwood e, Christopher S. McSweeney a

a CSIRO Australia, Queensland Bioscience Precinct, St. Lucia, Qld 4067, Australia b Department of Animal Science, Faculty of Agricultural and Food Sciences, University of Manitoba, Winnipeg, MB, Canada R3T 2N2 c Department of Animal Science, University of Illinois, Champaign-Urbana, IL 61801, USA d The Ohio State University, Department of Animal Sciences, Columbus, OH 43210, USA e AgResearch, Palmerston North, New Zealand

Received 2 August 2002; received in revised form 19 June 2003; accepted 12 July 2003

First published online 30 August 2003

Abstract

The degradation of plant cell walls by ruminants is of major economic importance in the developed as well as developing world. Rumen fermentation is unique in that efficient plant cell wall degradation relies on the cooperation between microorganisms that produce fibrolytic enzymes and the host animal that provides an anaerobic fermentation chamber. Increasing the efficiency with which the rumen microbiota degrades fiber has been the subject of extensive research for at least the last 100 years. Fiber digestion in the rumen is not optimal, as is supported by the fact that fiber recovered from feces is fermentable. This view is confirmed by the knowledge that mechanical and chemical pretreatments improve fiber degradation, as well as more recent research, which has demonstrated increased fiber digestion by rumen microorganisms when plant lignin composition is modified by genetic manipulation. Rumen microbiologists have sought to improve fiber digestion by genetic and ecological manipulation of rumen fermentation. This has been difficult and a number of constraints have limited progress, including: (a) a lack of reliable transformation systems for major fibrolytic rumen bacteria, (b) a poor understanding of ecological factors that govern persistence of fibrolytic bacteria and fungi in the rumen, (c) a poor understanding of which glycolyl hydrolases need to be manipulated, and (d) a lack of knowledge of the functional genomic framework within which fiber degradation operates. In this review the major fibrolytic organisms are briefly discussed. A more extensive discussion of the enzymes involved in fiber degradation is included. We also discuss the use of plant genetic manipulation, application of free-living lignolytic fungi and the use of exogenous enzymes. Lastly, we will discuss how newer technologies such as genomic and metagenomic approaches can be used to improve our knowledge of the functional genomic framework of plant cell wall degradation in the rumen. © 2003 Published by Elsevier B.V. on behalf of the Federation of European Microbiological Societies.

Keywords: Rumen; Plant cell wall; Metagenome; Diversity; Functional genome

Contents

1. Introduction....................................................................................................................664

2. Fibrolytic ruminal microorganisms and enzymes................................................................666

2.1. Glycosyl hydrolases....................................................................................................666

2.2. Fibrobacter succinogenes............................................................................................668

2.3. Ruminococcus albus....................................................................................................668

2.4. Ruminococcus flavefaciens..........................................................................................669

2.5. Butyrivibrio/Pseudobutyrivibrio....................................................................................669

2.6. Prevotella..................................................................................................................670

2.7. Anaerobic rumen fungi..............................................................................................670

* Corresponding author. Tel.: +1 (204) 474-9383; Fax: +1 (204) 474-7628. E-mail address: denis_krause@umanitoba.ca (D.O. Krause).

0168-6445/03/$22.00 © 2003 Published by Elsevier B.V. on behalf of the Federation of European Microbiological Societies. doi: 10.1016/S0168-6445(03)00072-X

3. Adherence to cellulose and evidence for the cellulosome....................................................671

3.1. General nature of adherence to cellulose....................................................................671

3.2. F. succinogenes adherence to cellulose........................................................................672

3.3. R. flavefaciens adherence to cellulose..........................................................................673

3.4. R. albus adherence to cellulose ..................................................................................673

3.5. Fungal adherence to cellulose....................................................................................674

4. Improvement of fiber digestion by rumen inoculation........................................................674

4.1. Background ..............................................................................................................674

4.2. Genetically modified fiber-degrading bacteria..............................................................675

4.3. Non-genetically modified fiber-degrading bacteria........................................................675

5. Modification of fiber by exogenous means........................................................................676

5.1. Chemical and mechanical treatments..........................................................................676

5.2. Plants genetically modified for plant cell wall composition..........................................676

5.3. Lignolytic fungi ........................................................................................................678

5.4. Exogenous enzymes....................................................................................................679

6. (Gen)omics and metagenomics..........................................................................................681

7. Conclusions......................................................................................................................683

Acknowledgements..................................................................................................................683

References..............................................................................................................................683

1. Introduction

Ruminants make up a significant proportion of the domesticated animal species worldwide, and among farmed livestock are the best adapted to utilization of plant cell walls [1]. Almost half of the global carbon fixed annually by photosynthesis is incorporated into plant cell walls making it the most renewable carbon source on earth [2]. Lignocellulose will therefore always be important in ruminant diets, and even in intensive finishing systems is incorporated into the ration because it is both economical and necessary for normal healthy rumen function. Improvement in the ability of the rumen microbiota to degrade plant cell wall is generally highly desirable and usually leads to improved animal performance [3].

In the developing world ruminants have an important role in the sustainability of village communities, and in many cases form the major source of income. Smallholders rely on subsistence farming, with few or no inputs, but forages are usually available and generally provide the sole source of nutrition to the animal [4,5]. Even in the developed world, forage is the major source of nutrition in many situations. For example, in the major beef-producing regions of northern Australia, forage is virtually the only source of nutrition, and supplementation is impractical and expensive. This is also true in North and South America where grazing animals are common, and ligno-cellulose makes up the majority of the diet.

The symbiosis between animal and microbe in the rumen allows for a cooperative system in which both the host and animal derive a benefit [6]. The rumen is a capacious pre-gastric fermentation chamber that sustains a rich community of microorganisms that rapidly colonize and digest feed particles. Carbohydrate polymers in plants are

indigestible to most animals but can be hydrolyzed and fermented by a range of microorganisms in the rumen. The end products of this fermentation are fatty acids, which form a major metabolic fuel for the ruminant, and microbial cells that are a major source of protein and amino acids when absorbed in the lower digestive tract of the animal. One cost of this relationship is the breakdown, or sacrifice, of dietary protein by rumen microorganisms before digestion by the host enzymes [6].

Over the last 50 years, significant improvements in our understanding of the digestion of fiber in the rumen have occurred and much of this information has been translated into practical nutritional management strategies. For example, an understanding of the importance of nitrogen to the degradation of fiber by fibrolytic microorganisms has led to the inclusion of urea supplements in ruminant diets [1], and mechanical and chemical treatments of forages has improved their digestibility [7]. Probably one of the most effective means of exploiting information on rumen microbiology and digestion has been the construction of computer models that predict animal performance from the characteristics of feed ingredients [3,8,9]. These models have the ability to make significant improvements in dietary formulation, and although they do not directly manipulate fiber digestion in the rumen, they do optimize the utilization of scarce nutrients by rumen microbiota. Models are ideal examples of how incremental advances in our knowledge of rumen microbial processes have been utilized to make significant advances in rumen function.

The supposed inability of rumen microbiota to express the appropriate suit of enzymes to maximize fiber digestion is often given as a reason for research into improving the function of rumen microbiota. For example, this has led to the inoculation of bacteria into the rumen that ap-

parently have 'superior' abilities. Hungate [1] discussed early examples in which fibrolytic bacteria dosed into the rumen had little effect on fiber digestion. In fact the notion that the rumen microbiota lacks appropriate fibrolytic activities has persisted, and in recent years genetically modified rumen bacteria have been the focus of intensive research [10,11]. Unfortunately, much of the effort in producing ruminal inoculants has been unsuccessful because no improvements in rumen fiber digestion have resulted in vivo.

What has become obvious from these efforts is that our understanding of the rumen microbial ecosystem is still superficial in comparison with the complexity it encom-

passes. If one reflects on the fact that a substantial proportion of the rumen microbiota has not been cultured then these results are not surprising. Far less effort has been expended on the rumen fungi and protozoa [12], and even less work has been done on bacteriophage inhabiting the rumen [13,14]. The importance of this complexity has become apparent when inoculation studies revealed that dosed organisms usually disappear below detectable levels in the rumen and that rumen protozoa may play an important role in this decline [15,16].

The purpose of this review is to examine research that has attempted to improve fiber digestion in the rumen. We will not discuss some of the well known aspects of rumen

Table 1

GH identified in cultured rumen microorganisms and the family to which they belong

Enzymea

GH familiesb

Rumen microorganism (GH families)

Endocellulase 3.2.1.4 5, 6, 7, 8, 9, 10, 12, 26, 44, 45, 48, 51, 61, 74

Exocellulase 3.2.1.91 5, 6, 7, 9, 10, 48

L-Glucosidase 3.2.1.21 1, 3

Endoxylanase 3.2.1.8 5, 8,10,11, 16, 26, 43, 52, 62

L-Xylosidase 3.2.1.37 3, 10, 39, 43, 52, 54

a-Amylase 3.2.1.1 Licheninase 3.2.1.73

Mannanase 3.2.1.78

13, 14, 57 16

Butyrivibrio fibrisolvens (5, 9)

Fibrobacter succinogenes (5, 51, 9)

Neocallimastix frontalis (5)

Neocallimastix patriciarum (5, 6)

Orpinomyces joyonii (5)

Orpinomyces sp. (5, 6)

Piromyces equi (45, 5, 48)

Piromyces sp. E2 (9, 48)

Piromyces rhizinflata (5)

Prevotella ruminicola (26, 5, 9, Nc)

Ruminococcus albus (5, 9)

Ruminococcus flavefaciens (44, 5, 9, Nc)

Neocallimastix patriciarum (6)

Orpinomyces sp. (6)

Piromyces rhizinflata (6)

Piromyces sp. E2 (6)

Butyrivibrio fibrisolvens (3)

Orpinomyces sp. (1)

Piromyces sp. E2 (3)

Prevotella albensis (3)

Prevotella ruminicola (3)

Ruminococcus albus (3)

Butyrivibrio fibrisolvens (10)

Eubacterium ruminantium (10)

Fibrobacter succinogenes (10, 11)

Neocallimastix frontalis (11)

Neocallimastix patriciarum (10, 11)

Orpinomyces sp. (11)

Piromyces communis (11)

Piromyces sp. (11)

Prevotella bryantii B14 (10)

Prevotella ruminicola (10, 26, 5)

Pseudobutyrivibrio xylanivorans type strain: Mz5 (11)

Ruminococcus albus (11)

Ruminococcus flavefaciens (10, 11, 16)

Butyrivibrio fibrisolvens (43)

Prevotella ruminicola (43)

Butyrivibrio fibrisolvens H17c (13)

Fibrobacter succinogenes (16)

Orpinomyces sp. PC-2 (16)

Piromyces sp. (26)

Prevotella ruminicola (26)

aEnzymes are identified by their IUPAC designations.

bGH enzymes are identified by their designation in the Carbohydrate Active Enzyme server (http ://afmb.cnrs-mrs.fr/CAZY). This provides a hierarchical based web database that can easily be queried for organisms or enzymes of interest.

Table 2

Carbohydrate esterases (CE) identified in cultured rumen microorganisms and the family to which they belong

Enzymea CEb Rumen microorganism (GH families)

Acetyl xylan esterase 3.1.1.72 1, 2, 3, 4, 5, 6, 7 Fibrobacter succinogenes (6)

Neocallimastix patriciarum (3, 2, 6) Orpinomyces sp. (6) Ruminococcus albus (1, 4) Ruminococcus flavefaciens (1, 3) Ruminococcus sp. (1) Butyrivibrio fibrisolvens (Nc) Piromyces equi (1)

Feruloyl esterase 3.1.1.73 1 Orpinomyces sp. (1)

aEnzymes are identified by their IUPAC designations.

bCE enzymes are identified by their designation in the Carbohydrate Active Enzyme server (http ://afmb.cnrs-mrs.fr/CAZY). This provides a hierarchical based web database that can easily be queried for organisms or enzymes of interest.

microbiology such as taxonomy involved in fiber digestion, but rather will focus on attempts that have been made to improve fiber digestion with inoculants, fungal treatments, and exogenous enzymes, but will preface this with an update of the glycosyl hydrolases (GH) (Table 1), and other enzymes (Table 2) produced by rumen microorganisms. In addition, we will discuss the applications of genomics and metagenomics to rumen microbiology and how this new technology will affect our knowledge of rumen cell wall degradation. This review will provide a road-map of where we have been, and where this research is likely to go in the future.

2. Fibrolytic ruminal microorganisms and enzymes

A complex community of fibrolytic microorganisms catalyzes the degradation of fiber in the rumen. The taxonomy of these organisms has been extensively reviewed in recent years and the reader should refer to these publications for further information [12,17-19]. In brief, the major fibrolytic bacteria are the Gram-negative Fibrobacter succinogenes, and two species of Gram-positive bacteria, Ruminococcus albus and Ruminococcus flavefaciens. Butyrivibrio fibrisolvens are a group of highly xylanolytic Grampositive bacteria inhabiting the rumen, which have a central role in fiber digestion. Prevotella are not regarded as highly cellulolytic bacteria but do produce a range of xy-lanases. A number of less well characterized cellulolytic bacteria occur, such as Eubacterium cellulosolvens. In addition, the anaerobic rumen fungi are considered important in fiber digestion and one of the best-studied fungi is Neocallimastix sp. There is also increasing evidence that the rumen protozoa may have the capacity to digest fiber, although this is not particularly well understood [20].

2.1. Glycosyl hydrolases

Most of the enzymes involved in cellulose and hemicel-lulose degradation are GH (Table 1) that hydrolyze the glycosidic bond between carbohydrates, or between a car-

bohydrate and a non-carbohydrate molecule [21]. Hydrolysis of the glucoside results in the formation of a sugar and another compound, and the 'hydrolase' signifies that C-O, C-N, or C-C bonds can be broken during hydrolysis. We have included the carbohydrate esterases (CE) in the discussion of GH, as they hydrolyze C-O bonds, but there may be some definitional issues (Table 2). The hydrolysis step is via general acid catalysis, and requires a proton donor and a nucleophile, or base [21]. The reaction results in either a retention, or an inversion of the anome-ric carbon. Retaining GH retain the anomeric carbon configuration via a double displacement mechanism. Inverting GH invert the anomeric configuration via a single nucle-ophilic displacement.

GH and related enzymes can be classified based on their amino acid sequence similarity using hydrophobic cluster analysis, rather than simple substrate specificity [22] (Table 1). This method of classification results in all members of a family possessing a conserved catalytic mechanism, even though they may act on different substrates [23]. The CAZy (Carbohydrate Active enZyme) database (http://afmb.cnrs-mrs.fr/CAZY) maintains and updates information on GH and their classifications via amino acid sequences. Currently there are 91 structurally defined GH families, 65 glycosyltransferase families, 13 polysaccharide lyase families, 13 carbohydrate esterase families, and 32 carbohydrate-binding module families.

Efficient breakdown of cellulose in the rumen usually requires a number of GH including endoglucanases (endo-1,4-L-D-glucan hydrolase, EC 3.2.1.4), exogluca-nases (exo-1,4-L-D-glucan cellobiohydrolase, EC 3.2.1.91), and L-glucosidases (P-D-glucosidase, EC 3.2.1.21), which work synergistically to hydrolyze cellulose [24]. The model (Fig. 1) for synergism between the three types of enzymes has been proposed to be that of endoglucanase attack on amorphous regions of cellulose fibers, creating sites for cellobiohydrolases to proceed into the crystalline region of cellulose [25]. The cleaved cellobiose and short-chain cellodextrins are then converted to glucose by P-glucosi-dases to stop end product inhibition. All three types of GH have been isolated from different rumen cellulolytic

r Cellulose

Endoglucanase

• • • • •

• • •

Cellobiose Cellotriose

Cellobiohydrolase ^

Glucose

D-Glucose chains Exoplucanase 11_Exoglucanase

P-Glucosidase (cellobiase)

Fig. 1. The cellulase enzyme system consists of three major components: endo-L-glucanase (EC 3.2.1.4), exo-L-glucanase (EC 3.2.1.91), and L-glucosidase (EC 3.2.1.21). The mode of action of each of these is: (1) endo-p-glucanase, 1,4-L-D-glucan glucanohydrolase, carboxymethyl cellulase: 'random' scission of cellulose chains yielding glucose and cel-lo-oligosaccharides. (2) Exo-p-glucanase, 1,4-L-D-glucan cellobiohydrolase: exo-attack on the non-reducing end of cellulase with cellobiose as the primary structure. (3) L-Glucosidase, cellobiase: hydrolysis of cellobiose to glucose [31].

microorganisms. There are 14 GH families containing en-doglucanases, and the rumen microorganisms predominantly fall within families 5 and 9. This is hardly surprising, as these families are comprised of the most members, suggesting that many different types of organisms have found this to be the best form for an endo-acting cellulase.

Of the exoglucanases isolated from the rumen, family 6 exo-acting enzymes have only been found in anaerobic fungi (Table 1). Members of this family have an inverting mechanism that proceeds from the non-reducing end of the cellulose chain. Family 7 enzymes act with a retention mechanism and proceed from the reducing end of the cellulose chain. Currently no family 7 exo-acting enzyme has been isolated from rumen microorganisms. The absence is notable because in aerobic fungi such as Trichoderma reesei and Humicola insolens these enzymes can act in synergy with the family 6 exo-acting enzymes to attack the cellulose fibers from both ends and thus increase the degree of digestion [26].

Xylanases act on xylan, converting it to its constitutive sugars, with endo-L-1,4-xylanases (1,4-L-D-xylan xylano-

hydrolase, EC 3.2.1.8), which hydrolyze the 1,4-0-xylopy-ranosyl linkages of xylan, and L-xylosidases (1,4-L-xylan xylohydrolase EC3.2.1.37), which hydrolyzes the xylo-oli-gosaccharides produced by the endoxylanases (Table 1). These enzymes have been found in the rumen, mostly from GH families 10 and 11. A series of enzymes, which cleave side chain sugars or remove acetyl groups from the xylan backbone [27], are also involved in xylan degradation and can be found in the rumen (Table 2). Acetylxylan esterases are responsible for deacetylation of xylans and xylo-oligosaccharides. With 22-50% of xylose residues being acetylated at the O-2 and/or O-3 positions, acetyla-tion is an important factor influencing the digestibility of plant cell wall material in ruminants [28,29]. In addition arabinoxylan is one of the main hemicelluloses comprising the backbone structure of L-1,4-linked xylose to which are attached the arabinose side chains. The arabinose also has ester-linked ^-coumaric and ferulic acid, and the hemicel-lulose is linked to lignin via ferulic acid links (Fig. 2) [3032].

In this era of whole genome sequencing, tabulation of GH is in some ways redundant, because the sequence information of each organism contains the whole complement of enzymes. Traditionally, genes have been cloned and screened on various substrates that indicate a particular type of GH activity, and the inability to recover a gene of a particular class may just mean that the cloning

Fig. 2. A representation of the different kinds of lignin interactions in plant cell walls. A summary of the kinds of aromatic ester and ether cross-links between carbohydrate and lignin. Linkages are: 1 = direct ester linkage, 2 = hydroxycinnamic acid ester, 3 = hydroxycinnamic acid ether, 4 = ferulic acid bridge, 5 = direct ether linkage, 6 = dehydrodiferulic acid diester bridge, 7 = dehydrodiferulic acid diester-ether bridge. Redrawn from [30].

and screening strategy did not work. A simplistic resolution to this problem is to mine the genome and identify the genes in question. The only problem with this is that so much of the genome has no sequence homology with known genes, and as many as 30-40% of the open reading frames (ORFs) have no discernible homology/function. There is an increasing number of ORFs without known function, or ORFans, and in the case of the rumen bacteria could represent GH that have not yet been cloned [33].

The following discussion describes GH from Fibro-bacter, Ruminococcus, Butyrivibrio, Prevotella, and anaerobic rumen fungi, and is not based on discovery via genome mining techniques. Genome data for rumen bacteria are available in special purpose databases such as the 'The comprehensive microbial resource' [34]. Research on the GH of rumen organisms should always be done within the context of the genome sequence. The discussion below is based largely on functional analysis which genome sequence annotation relies upon, and these two approaches need to be pursued in parallel.

2.2. Fibrobacter succinogenes

The fibrolytic enzymes of F. succinogenes are amongst the best studied within the rumen bacteria (Tables 1 and 2). Initial enzyme purification work characterized the en-doglucanases EG1 and EG2 [35], a chloride-stimulated cellobiosidase [36], and a cellodextrinase [37]. Overall, the cloning of endoglucanases is most common [24]. An endoglucanase gene, cel-3, was one of the first to be sequenced [38] and at least seven distinct glucanase activities in F. succinogenes strain S85 have been reported [24,25]. Genomic libraries have subsequently revealed the endoglu-canase genes endB, celD, celE, celF, and celG [39-42]. A gene for a mixed linkage L-glucanase (mlg) has also been cloned and characterized [43]. More recently, a new GH family 5 endoglucanase gene, endA, from strain S85 has been reported [44]. The cellobiosidase gene (cedA), which encodes a previously purified cellodextrinase, has also been cloned and characterized [45]. Strain S85 also contains two L-glucosidase activities [46] and a cellobiase activity [47]. Where gene and derived protein sequence information is known, these genes fall into either GH family 5 or family 9, with the exception of celF that is in family 51.

Xylanases have also been purified from strain S85. The xylan debranching enzyme endoxylanase 1, and a dual function endoxylanase 2 (xylanase and endoglucanase activities) have been characterized [48]. Four xylanase genes have been cloned and fully characterized. The xylanase gene xynC encodes two catalytic domains, both with GH family 11 xylanase activity [49]. The xynB gene encodes a dual function GH family 10 domain, with both xylanase and endoglucanase activities [50]. Genomic fragments of F. succinogenes strain S85 containing GH family 10 xylan-

ase genes have also been deposited in GenBank by two groups. A 6688-bp fragment of F. succinogenes S85 DNA encoding four open reading frames showing homology to GH (celJ, cel5K, xyn10L, and xyn10M) and showing xylanase activity on RBB xylan plates (GenBank AY007248 [51]). Another overlapping fragment (7223 bp) was sequenced and three genes were identified. One had homology to a xylanase (xynD) that is identical to xyn10L, and xynE that is identical to xyn10M, previously reported as xynB (GenBank AF180368 [52]).

Glucanase genes have also been isolated from F. succinogenes strains AR1, SD35, 135 and BL2. Three gluca-nases were cloned from AR1, and one, endAFS, was characterized in detail and found to be a GH family 9 endoglucanase [53-55]. Strain SD35 contained a GH family 51 endoglucanase (endl) [56] while strain BL2 had a GH family 9 endoglucanase (egC) [57]. An endoglucanase (end3) [58] and a xylanase [247] have been described in F. succinogenes strain 135. The glucanase genes from these strains seem to be distinct from strain S85 glucanases

[24.59].

2.3. Ruminococcus albus

A large range of GH have been isolated from R. albus strains (Tables 1 and 2). R. albus strain F-40 has attracted substantial attention and seven distinct endoglucanases, a cellobiosidase, and two ß-glucosidases have been reported

[24.60]. Enzyme purification and characterization studies have in addition identified two endoglucanases [61], a cel-lobiosidase [62], and a ß-glucosidase [63]. The ß-glucosi-dase was later cloned and sequenced [64]. Recently, a family 5 GH was cloned and sequenced (egV), and contained a dockerin domain thought to be involved in cellulosome assembly [65]. This was followed by the isolation of the strain F-40 cellulosome complex [66] and the demonstration of cellulase and/or xylanase activity in 11 of at least 15 proteins that comprised the complex.

The fiber-degrading enzymes and associated genes from an R. albus strain isolated in Australia, AR67, have also been studied [67,68]. This particular strain is extremely interesting as it is one of the most active strains yet cultured [69]. A gene encoding exo-1,4-ß-D-glucosidase from R. albus AR67 was expressed in Escherichia coli. The cloned enzyme was located in the cytoplasm (40%) and attached to insoluble cell components (48%). After purification to homogeneity, the enzyme was found to be specific for cellulose with a ß-D-glucopyranosyl configuration and was inactive against a-glucosides, lactosides and xylo-sides.

Strain SY3 of R. albus is another highly active degrader of plant cell walls [69]. To determine which subcellular proteins might be involved in adhesion and a possible cellulosome complex, SY3 wild-type was compared with an SY3 adhesion-deficient mutant. The adhesion-defective mutant produced significantly less (5-10-fold) overall gly-

canase activity, and the 'true cellulase activity' appeared to be entirely confined to the cell membrane fractions. Two family 4 GH, endoglucanases EGA (celA) and EGB (celB), have also been cloned from R. albus SY3 and sequenced [61,70].

Xylanase activity has been reported in R. albus strain 8 [71]. Xylanase 1 and 2 have been purified and are thought to interact synergistically with an arabinofuranosidase during degradation of alfalfa cell walls [71]. R. albus strain 7 has been found to harbor a ß-glucosidase, an endoglu-canase, and three xylanases [72-74]. The xynA gene from strain 7 encodes a protein with three domains: a GH family 11 catalytic domain, a xylanase-stabilizing domain similar to those from R. flavefaciens, and a domain with sequence similarity to deacetylases [74]. The xynB gene encodes a protein of 680 amino acid residues, including a family 11 GH catalytic domain, a repeated asparagine-tyrosine region and a C-terminal domain showing homology to the Clostridium stercorarium xynY domain E [73]. The xynC gene also show discrete domains: an N-terminal domain with homology to xylanases of Bacillus subtilis and Erwinia chrysanthemi, a region with homology to the C. stercorarium xynY thermostabilizing domain and a linker sequence containing a 20-residue repeat of aspara-gines [73]. An unidentified Ruminococcus species has also been found to contain a xylanase gene (xynl) belonging to the family 11 GH, which shows homology to the R. flavefaciens xylanase gene, xynE [75] and R. albus 7 xylanase (xynB) [74].

2.4. Ruminococcus flavefaciens

GH from R. flavefaciens strains are common (Tables 1 and 2). R. flavefaciens FD-1 has an exoglucanase activity [76], a cellodextrinase [77], and at least four glucanases [78-82]. Three of these glucanase genes (celB, celD and celE) are inducible, while one glucanase (celC) and the FD-1 cellodextrinase (celA) are expressed constitutively [76,82]. Strain 17 has an endoglucanase in GH family 5 (endA [83]), and a recently characterized GH family 44 endoglucanase gene, endB [84]. R. flavefaciens is able to degrade xylan, and four xylanase genes have been identified. The genes xynA and xynD encode xylanases with dual catalytic domains [85,86]. Although the XynA domains exhibit similar substrate specificities, they fall into different GH families; domain A is in GH family 11 while domain C belongs to GH family 10 [86]. Similarly, the two xynD domains fall in different GH families: domain A is a family 11 xylanase while domain B is a family 16 gluca-nase [85]. On the other hand the R. flavefaciens 17 xylan-ase, xynB, encodes an enzyme with a single GH family 11 catalytic domain [86].

Three enzymes carrying esterase domains have recently been identified in the rumen cellulolytic anaerobe R. flavefaciens 17 [75]. The cesA gene product includes domains for an N-terminal acetylesterase and an unidentified C-ter-

minal domain, while the previously characterized XynB enzyme (781 amino acids) includes an internal acetylester-ase domain in addition to its N-terminal xylanase catalytic domain. A third gene, xynE, is predicted to encode a multidomain enzyme of 792 amino acids including a family 11 xylanase domain and a C-terminal esterase domain. A family 3 GH gene, xylA, has been identified in an 11-kb genomic DNA fragment from R. flavefaciens 17 [75] which also carries two putative AraC-type regulators, xynD, and a putative xylose isomerase and ABC-type sugar transporter genes.

A phage library constructed from strain 186 genomic DNA produced a carboxymethylcellulose (CMC)-degrad-ing clone containing four ORFs corresponding to an endoglucanase (renA), an exoglucanase (rex), a L-glucosi-dase, and a xylanase (rxy) [87]. The predicted protein encoded by renA had an N-terminal proline-threonine-ser-ine-rich region, a central catalytic domain and a C-termi-nal binding domain. The rex-encoded exoglucanase had its catalytic site located at the N-terminus followed by a pro-line-threonine-serine-rich region and a putative binding domain at the C-terminus.

2.5. Butyrivibrio/Pseudobutyrivibrio

Ruminal strains of Butyrivibrio are genetically diverse and have recently been reclassified with the creation of a new genus, Pseudobutyrivibrio [88]. The phylogenetic diversity of the Butyrivibrio/Pseudobutyrivibrio assemblage is mirrored by the large number of strains in which GH have been detected (Tables 1 and 2) [89]. Although cellu-lolytic strains of Butyrivibrio have been isolated in the past [90], this activity is often lost upon subculture in the laboratory. The Butyrivibrio are, however, efficient utilizers of xylans [91,92].

Correspondingly, an abundance of xylanase genes have been isolated and only a few endoglucanases have been described. B. fibrisolvens strain H17c(SA) contains a GH family 5 endoglucanase (endl [93]), a GH family 9 cellodextrinase (cedl [94]), a GH family 3 L-glucosidase (bglA [95]) and three xylanase genes (xynB [96], xynE and xynF [89]. B. fibrisolvens strain A46 produces a GH family 5 glucanase (celA [97]) while GS113 produces an interesting GH family 43 enzyme which has both xylosidase and ara-binofuranosidase activity [98,99]. Rumbak et al. [100] cloned and sequenced an a-amylase gene (amyA) from strain H17c and found that it shared homology with pro-karyotic and eukaryotic amylases of GH family 13. Man-narelli et al. [101] cloned and sequenced a xylanase gene (xynA) from strain 49 and showed via hybridization studies that a similar gene was present in strains H17c and CF3. Subsequent to this, Dalrymple et al. [89] used polymerase chain reaction (PCR) primers designed to different xylanase gene families to survey the xylanase genotypes in 28 Butyrivibrio strains. They found 11 new xylanase genotypes, all of which fall into GH family 10. More recently

multiple xylanases have been reported from Pseudobuty-rivibrio xylanovorans strain Mz5 [102]. Strain Mz5 was found to have high xylanolytic activity and the cell-associated fraction showed 14 xylanase activities on sodium dodecyl sulfate-polyacrylamide gel electrophoresis zymo-grams ranging in molecular mass from 26.7 kDa up to 145 kDa. Two of these xylanases (99.8 and 77.4 kDa) showed weak CMCase activity indicating some endoglucanase activity. Xylanase expression was generally inducible by growth on oat spelt xylan, but two xylanases showed low-level expression when glucose, cellobiose, maltose and soluble starch were used as the carbon source. Subsequently, a partial xylanase sequence (xynT) was deposited in GenBank [103] and is placed within family 11 GH.

2.6. Prevotella

Prevotella are able to utilize a wide variety of polysac-charides, and are thought to be important contributors to xylan degradation in the rumen [104] and as such a range of GH have been identified (Table 1 and 2). Prevotella bryantii has been found to contain a GH family 26 endo-glucanase, which was notable due to a frame shift in its coding region [105]. Matsushita et al. [106] sequenced an endoglucanase gene (celA) encoding an enzyme with substantial homology to members of the GH family 26 cellu-lases. A membrane-associated glucosidase with cellodex-trinase and cyanoglycosidase activities has also been characterized [107]. The enzyme appears to have an exo-1,4-L-glucosidase activity as it attacks cellodextrins from the non-reducing end. It was able to cleave the cyanogenic glycosides, amygdalin and prunasin, and may play a part in cyanide toxicity in ruminants. Another study by Verco and White [108] identified a phage clone from a Bi4 ge-nomic library expressing endoglucanase and mannanase activities. Subclones containing EcoRI fragments were sequenced and six ORFs were identified: ORFs 1 and 2 had no significant homology with any database sequences, ORF 3 had regions of homology with a cellulose-binding protein (CBP) from Clostridium cellulovorans, ORFs 4 and 5 encoded two related L-1,4-endoglucanases, while ORF 6 encoded a mannanase. The gene sequences up- and downstream of the ORFs indicated that these genes were transcribed as a single unit.

A xylanase gene from Prevotella ruminicola 23 [109] with endoglucanase activity has a catalytic domain that shares homology with B. fibrisolvens, R. flavefaciens, and Clostridium thermocellum xylanases. A phage library of P. ruminicola (bryantii) B14 screened for xylanase activity revealed four chromosomal regions associated with activity [110]. One clone encoded an endoxylanase, with ^-ni-trophenyl (pNP)-xylosidase and pNP-arabinofuranosidase activities. The DNA sequence of this clone [111] consisted of two genes, xynA, which was responsible for endoxyla-nase activity, and xynB, which encoded xylosidase activity (using an exoxylanase mechanism), and a weak arabino-

furanosidase activity. Two other xylanases, xynC, from strain B14, and a strain D31d xylanase, have been characterized [112]. They were found to possess highly unusual structures in which their catalytic domains were interrupted by possibly non-coding sequences. The sequence of a cellulase gene from P. ruminicola strain 23 showing homology with GH family 5 enzymes has also been lodged (GenBank AB022865). Recently, further characterization of the P. bryantii B14 gene cluster containing xynA and xynB identified additional genes [113]: xynD which appears to be involved as a solute transporter, xynE which has homology to genes encoding acylhydrolases and ary-lesterases, xynF which shares homology with L-glucuroni-dases, and xynR, a multidomain regulatory protein which may serve to activate the xynABD gene cluster and therefore xylanase expression.

2.7. Anaerobic rumen fungi

The actual role of anaerobic fungi in the rumen is still an issue that has not been completely resolved [104] but our knowledge of their GH is increasing as evidenced by the range of enzymes identified (Tables 1 and 2). It is, however, known that anaerobic fungi colonize plant tissue and appear to degrade lignified tissue that is not degraded by other microorganisms [114]. Having said this, it is also true that the rates of growth and degradation of fungi are much slower than those of the bacteria, and their ability to persist is limited because their growth rates are much lower than the rumen dilution rate. Fungi are able to degrade up to 34% of the lignin in plant tissue, can penetrate the plant tissue as a result of their filamentous growth [115], have a broad range of highly active enzymes, and are the only known rumen organisms with exo-acting cellulase activity [24]. The importance of fungi is further supported by the synergy that occurs with rumen bacteria [116]. Methanogenic co-cultures of non-autoclaved stem fragments were degraded more extensively by Neocallimastix frontalis and Piromyces isolates than by Caecomyces isolates and N. frontalis and Piromyces isolates showed the greatest rates of stem degradation. When interactions between F. succinogenes and methanogenic co-cultures of fungi growing on ryegrass stems were investigated, N. frontalis inhibited F. succinogenes. In contrast, a Cae-comyces species grown with F. succinogenes increased stem degradation, indicating that F. succinogenes and Caecomy-ces spp. may have complementary fibrolytic activities [117].

Neocallimastix spp. are the best studied of the rumen fungi and are highly active against crystalline cellulose. Earlier work [118] reported that the activity on crystalline cellulose by RK21 was even higher than that of T. reesei C30. To unravel the cellulases a cDNA library prepared from Neocallimastix patriciarum CX was constructed [119]. Four cDNAs of note were identified: celA, celB, celC, and celD. The celA was a cellobiohydrolase that

hydrolyzed crystalline cellulose, while celB and celD had endoglucanase activities. The most interesting cDNA was celD that had multiple activities including endoglucanase, licheninase, cellobiosidase, and xylanase. Several researchers have also identified cellobiohydrolase (celA) [120], glucanase (celB) [121], glucosidase [122], L-glucosidase [123], and cellobiase [124] from rumen fungi.

Cellulase activity is also associated with anaerobic Orpinomyces joyonii SG4, which has significant cellulase activity [125]. Two cellulases, celB29 and celB2, were isolated from a cDNA library. The cloned enzymes had high activities towards barley L-glucan, lichenin and CMC, but not Avicel, laminarin, pachyman, xylan and pullulan. In addition, CelB29 and CelB2 showed activity against pNP-L-D-cellobioside and pNP-L-D-cellopentaoside but not pNP-L-D-glucopyranoside with preferential activity against pNP-L-D-cellotrioside [125]. Recently identified from a cDNA library of Orpinomyces PC2 [126] was a cDNA designated celF encoding a cellulase with a signal peptide, a carbohydrate-binding module (CBM), a linker, and a catalytic domain similar to that of celA from N. patriciarum. The catalytic domain was also homologous to CelA and CelC from the same fungus which contain N-terminal docking domains for a cellulase-hemicellulase complex.

Xylanases have also been cloned (xynA and xynB) from N. patriciarum [127]. The xynA encodes an enzyme with two catalytic domains and is highly active [127]. Interestingly, the truncated form of this enzyme has at least a fivefold higher activity than the native. The xynB encoded both xylanase and cellobiosidase activity [127]. Similar data were obtained with Piromyces sp. [128].

Forages are rich in xylan containing hemicellulose and as many as 22 ± 50% of the xylose residues are acetylated at the O-2 and/or O-3 positions and acetylation is an important factor influencing the digestibility of plant cell wall material in ruminants [129,130]. It can be demonstrated that chemical deacetylation of xylan is easily achieved using dilute alkali solutions and significantly increases the digestibility of cellulose by enzymes [131]. Several workers [132,133] have suggested that enzymic deacetylation may be a prerequisite for the breakdown of acetylxylan or may enhance the rate of its hydrolysis by other enzymes.

Acetylesterases able to remove O-acetyl groups from xylose residues in xylan and xylo-oligomers are classified as acetylxylan esterases [132,133]. Although not all esterases that hydrolyze such substrates are acetylxylan esterases, it is likely that under conditions of high levels of synthesis of fiber-degrading enzymes, a significant proportion of esterases able to hydrolyze naphthyl acetate (a-NA), or similar compounds, may be acetylxylan esterases. The limited number of cloned acetylxylan esterases isolated is partly due to the difficulties encountered in finding suitable substrates for screening of libraries for acetylxylan esterase clones. However, a number of acetyl-xylan esterases have been reported as having activity against a-NA and other artificial substrates [134].

More recently acetylesterase and cinnamoyl ester hydro-lase activities were described in N. patriciarum [135,136]. None of the enzymes had true cinnamoyl ester hydrolase activity, but two of the enzymes had acetylxylan esterase activity, and bnaA, bnaB and bnaC encode proteins with several distinct domains. Carboxy-terminal repeats in BnaA and BnaC were homologous to protein-docking domains in other enzymes from Neocallimastix species and another anaerobic fungus, a Piromyces sp. The catalytic domains of BnaB and BnaC are members of Ser/His active site hydrolases [137]. Feruloyl and ^-coumaroyl esterases have also been purified from Neocallimastix strain MC2 [138,139], and significant extracellular acetylesterase activity can be detected in the MC2 strain [140].

3. Adherence to cellulose and evidence for the cellulosome

3.1. General nature of adherence to cellulose

Fiber-degrading bacteria, and fungi, usually adhere to the surface of plant cell walls (Fig. 3) and a lack of understanding in exploiting this process may be one of the reasons for the difficulty in establishing inoculant microorganisms in the rumen [141-143]. The intimate association

Fig. 3. Adherence of mixed rumen bacteria to plant material. A: Scanning electron micrograph of adherence to plant cell wall. B: Close examination of bacterial cells reveals protuberances that are likely adherence factors that bind the cells to the plant surface [147].

with plant cell walls has evolved in conjunction with a sophisticated molecular structure, the cellulosome, which facilitates the adherence process (Fig. 4). The cellulosome is an extracellular structure, and appears in many cases to be essential for degradation of crystalline cellulose and associated plant cell wall polysaccharides [144]. The cellu-losome arrangement also promotes adherence to the plant cell walls, and provides individual microbial cells with a direct competitive advantage in the utilization of the soluble hydrolysis products [144]. It also appears that the anaerobic fungi have cellulosome-like machinery that is involved in adherence to cellulose [145,146].

Presently the adherence of bacteria to cellulose in rumen ecosystems can be divided into four phases in F. succino-genes, R. flavefaciens, and R. albus [147], but it is likely that many of the steps are similar in fungi. (1) Transport of the nonmotile bacterium to the plant substrate. (2) Nonspecific adhesion of bacteria to available sites on the plant cell wall [147]. (3) Specific adhesion via adhesions or li-gand formation with the substrate, that may be facilitated by structures such as cellulosome complexes, fimbrial connections, glycosylated epitopes of CBP or the glycocalyx, and CBM [147]. (4) Proliferation of the attached bacteria on potentially digestible plant tissues [147]. Bacterial adhesion is not straightforward and can be affected by a range of factors including: (1) bacterial age, glycocalyx condition, and competition with other microorganisms

[147], (2) the nature of the substrate, including cuticular covering, surface area, hydration, and ionic charge [147], and (3) environmental factors such as pH, temperature, and presence of cations and soluble carbohydrate [147].

3.2. F. succinogenes adherence to cellulose

F. succinogenes binds tightly to the surface of plant materials via adhesions leading to extensive plant cell wall degradation [148-151]. Three F. succinogenes enzymes, the endoglucanases EG2 and EGF and the chloride-stimulated cellobiosidase, are likely to contain a CBM, and may be involved in bacterial adhesion to cellulose [152,153]. There is strong evidence which suggests that seven CBPs, with masses of 40, 45, 50, 120, 180, 220, and 240 kDa located in the outer membrane of F. succinogenes, may be involved in adhesion. Immunogold labelling of the 180-kDa CBP demonstrated its importance in adhesion to cellulose via a common glycosidic epitope [154]. The importance of glycoproteins to the adhesion process can be demonstrated more directly by removing the carbohydrate structures on the cell surface with periodate and surface proteins with proteases as has been described for Streptococcus bovis and E. coli [155,156]. In a recent study with Fibrobacter intestinalis DR7 carbohydrate components of a glycosylated CBP isolated from the outer membrane and periplasm of F. intestinalis DR7 were shown to

Typell

dockerin

domain

Type cohesio: domain

Cellulose microfibrils

Cellulose

Type I dockerin domain

Type I

cohesion

domain

Bacterial cell surface

Fig. 4. Idealized representation of fiber and its component cellulose, microfibrils, hemicellulose, and lignin that are degraded via the cellulosome complex. The 'cellulosome' is a multienzyme complex produced by many cellulolytic rumen organisms. The cellulosomes are associated with the microbial cell surface, mediate cell attachment to the insoluble substrate, and degrade it to soluble products that are then absorbed. The multiple subunits of the cellulosome are composed of numerous functional domains that interact with each other and with the cellulosic substrate. One of these subunits, a large glycoprotein, which is called the 'scaffoldin', comprises a distinctive class of non-catalytic scaffolding polypeptide. The scaffoldin subunit selectively integrates the various cellulases and xylanase subunits into the cohesive complex, by combining its 'cohesin' domains with a 'dockerin' domain present on each of the subunit enzymes. The cellulose-binding domain attaches the cellulosome to the cellulose surface. The 'catalytic domains' constitute the various GH. Redrawn from various authors [30,147,174].

play a significant role in adhesion to cellulose. The isolated CBP included residues of glucosamine, galactosamine, glucuronic acid, and galacturonic acid that blocked adhesion to cellulose [148,157]. However, additional biochemical and genetic evidence is needed to explore the role of gly-cocalyx carbohydrates in the adhesion process for F. succinogenes. Thus, both the glycosidic residues of the outer membrane CBP and especially of the 180-kDa CBP, and the CBM of EG2, EGF, and Cl-stimulated cellobiosidase may play a role in the adhesion of F. succinogenes.

Even though a number of proteins appear to be involved in adherence, this is still indirect evidence. The only direct evidence to support the existence of either a cellulosome complex or fimbria structures involved in the adhesion mechanism of this bacterium is scanning electron microscopy observations [148,150,151,158,159]. However, in bacterial cells prestained with cationized ferritin, the presence of ultrastructural protuberances is sometimes connected to growth rate rather than to induction of cel-lulolytic systems [160].

3.3. R. flavefaciens adherence to cellulose

R. flavefaciens adheres immediately and strongly to fiber particles and degrades plant cell walls at a comparatively rapid rate [161-163]. For example, R. flavefaciens FD-1 has a maximum dilution rate on crystalline cellulose (0.1 h31), which is higher than that of other ruminal bacteria on the same substrate [164]. When R. flavefaciens is allowed to grow in the presence of plant cell walls, the cells will attach, and close examination of the areas of attachment reveals the existence of protuberances extending from the bacterium to the plant surface [148,161,165]. It has also been reported [166] that the endoglucanases of R. flavefaciens FD-1 exist in two forms: a large enzyme complex with a molecular mass of 3000 kDa called complex A, and a smaller fraction (89 kDa) designated complex B. It appears that complex A contains at least 13 different endoglucanases, whereas complex B has only five unique endoglucanases and some of the polypeptides in these complexes are glycosylated. Gene sequence analysis of three endoglu-canases, and one cellodextrinase in R. flavefaciens FD-1, and three xylanases (xynB, C, and D), one endoglucanase (endA), and an esterase (estA) in R. flavefaciens 17, demonstrates that these enzymes lack any distinct CBM ([82,166,167]. However, some R. flavefaciens 17 enzymes including XynB, XynD, EndA, and EstA contain a dock-erin-like domain, suggesting that they may in some way interact to form a cellulosome-like complex that could be involved in the adhesion mechanism [167,168].

Recently, a scaffoldin protein, a member of the cellulosome complex of R. flavefaciens 17, has been identified, and several cohesins connecting the scaffoldin with a type 1 dockerin associated with several catalytic enzymes have been identified [147]. A 30-kDa protein attached to the CBP in R. flavefaciens has been identified but its role

in adhesion is not clear [169,170]. Early work [161] suggested the possible role of glycocalyx glycoproteins in mediating adhesion of R. flavefaciens cells to cellulose, which was supported by a study of lectins which inhibited adhesion probably by blocking specific epitopes [171]. The finding that polypeptides of the two complexes identified in R. flavefaciens FD-1 are glycosylated [166] supports the possible importance of carbohydrate epitopes in adhesion of the bacterium. The adhesion of R. flavefaciens to cellulose was inhibited when methylcellulose or CMC was added to medium (0.1%) but not by the addition of cello-biose (1%), suggesting that the recognition site of cellulose-binding factors of this bacterium is larger than a repeating cellobiose moiety [169,170,172]. Thus, at least two mechanisms, cellulosome-like complexes and carbohydrate epitopes of the glycocalyx layer, are involved in the adhesion of R. flavefaciens to cellulose.

3.4. R. albus adherence to cellulose

Electron microscopic and antibody studies provided some evidence for the existence of cellulosome-like structures in R. albus [148,165,173,174]. It has been reported [175] that the cellulase activity of cellobiose-grown R. albus SY3 was cell-associated, had a high molecular mass, but formed an unstable complex (1.5 MDa) that could be easily disassociated into proteins of low molecular mass. Later it was found that phenylpropanoic acid (PPA) or phenylacetic acid (PAA) stabilized this complex [176]. Miron et al. [148] reported the isolation and separation of the glycocalyx capsule, inner membranes, and peptido-glycan cell walls of cellobiose-grown R. albus SY3, from the extracellular fluid and the cytoplasm [147]. They found that most of the cellulases, xylanases, and cellulose-degrading activities of R. albus SY3 were associated with the capsule and cell walls. These findings are consistent with electron microscopic observations that adherent R. albus cells are surrounded by a glycocalyx capsule and may not actually touch the cellulose surface [177, 178]. Genetic evidence is provided by sequence analysis of endoglucanases celA and celB from R. albus SY3 and endoglucanases I, II, III, and IV from R. albus F-40, as well as several xylanases. These analyses indicate that these enzymes lack an epitope for either a CBM or a dockerin-like domain, suggesting that they are not integrated as part of a cellulosome complex [61,66,166].

Even though most of the R. albus SY3 endoglucanases are not integrated into cellulosome-like organelles, high-molecular-mass complexes have been isolated that contain mainly xylanases and some endoglucanase activity. Most of this activity is associated with cellulose [165]. Thus it is suggested that the cellulosome complex of R. albus may contain a CBM or enzymes employing CBM [147]. In parallel, a cellulosome complex was isolated from the culture supernatant of R. albus F-40 grown on cellulose, and its components were identified as three previously se-

quenced endoglucanases (egV, egVI, and egVII). In addition five endoglucanases, three xylanases, and four non-enzymatic proteins were described that had not previously been identified [65,66].

3.5. Fungal adherence to cellulose

Molecular evidence is accumulating that enzymes are associated with a fungal cellulosomes in the genera Neocallimastix, Orpinomyces, and Piromyces, and are modular [145,179]. In addition to a catalytic domain, they contain several copies of a conserved 40-amino acid cysteine-rich, non-catalytic docking domain (NCDD), which do not show sequence homology to bacterial dockerins [125,127, 180,181]. NCDDs are interspaced by short linkers and are separated from the catalytic domain(s) by a serine-threo-nine-rich linker(s). Given these molecular signatures the NCDD subunits can be regarded as members of fungal cellulosomes.

Recently, it was shown that a glutathione S-transferase (GST) reporter protein fused to one, two, or three NCDDs from Piromyces equi was able to specifically recognize a 97-kDa protein present in a cellulosome preparation purified by a cellulose affinity procedure [182]. These results strongly indicate the presence of a scaffoldin analog in the fungal cellulosome and show that, in contrast to bacterial dockerins, a single NCDD may serve as the interacting unit. Thus far, estA from P. equi [182] and celB2 from O. joyonii [125] are the only examples of genes encoding a cellulosome component containing only one NCDD. The majority of the genes encoding components contain two NCDDs.

Significant progress has been made in unravelling the nature of the cellulosome in Piromyces spp. The cellulo-some produced by Piromyces sp. strain E2 during growth on filter paper was purified [183]. Three dominant proteins were identified in the cellulosome preparation, with molecular masses of 55, 80 and 90 kDa. To investigate the major 90-kDa cellulosome protein further, the corresponding gene, cel9A, was found to be an endoglucanase. Cel9A includes a 445-residue GH family 9 catalytic domain, and is the first fungal representative of this large family. The catalytic domain was succeeded by a putative L-sheet module of 160 amino acids with unknown function, followed by a threonine-rich linker and three fungal docking domains. Homology modelling of the Cel9A dockerins suggested that the cysteine residues present are all involved in disulfide bridges. Further investigations of cDNAs in P. equi and Piromyces sp. strain E2 revealed that they both encoded a GH 48 cellulase, containing two C-termi-nal fungal dockerin domains [184]. Immunoscreening with anti-cellulosome antibodies was used to isolate L-gluca-nase activity in the strain E2 cellulosome. The C-terminal end of the encoded Cel3A (GH family 3) protein consisted of an auxiliary domain and three fungal dockerins, typical for cellulosome components. Cel3A catalytic domain was

specific for L-glucosidic bonds and functioned as an exo-glucohydrolase on soluble substrates as well as cellulose [179].

There has also been significant progress elucidating the molecular determinants of substrate recognition in Piro-myces. Two CBM were described [146], which were highly discriminatory for gluco- and manno-configured ligands and accompanying data suggested that these sugars were likely to play a pivotal role in the efficient degradation of the plant cell wall by the Piromyces cellulosome. This unusual ligand specificity presents an excellent model system for studying protein-carbohydrate recognition. A model for CBM ligand recognition in Piromyces based on a crystal structure has recently been elucidated [185]. The major molecular determinant in recognition is the orientation of the aromatic residues in the binding site which complement the conformation of gluco- an manno-configured li-gands.

4. Improvement of fiber digestion by rumen inoculation

4.1. Background

The idea that the microbiota of the digestive tract does not always comprise an optimum balance was noted in the early 1800s. Brugnone (referenced by Hungate [1]) stated that "..as one encounters it [the bolus] in ruminants, a bolus which is removed very easily from the mouth, and which some veterinarians give as a sure remedy to induce rumination in animals, in which this function is suspended due to illness''. Metchnikoff [186] discussed a similar concept in relation to human gut health [187]. He believed that the human colon was a reservoir of potentially pathogenic bacteria and the toxic substances produced by these bacteria were harmful to the host. The observation that milk fermented by lactic acid bacteria did not support the growth of potentially pathogenic bacteria allowed him to arrive at this theory.

Inoculation clearly is of benefit to rumen function. Hun-gate [1] describes a number of studies in which 'supposedly' fibrolytic ruminal inoculants were added to the rumen, but there was no evidence suggesting that ruminal fiber degradation was enhanced. He did, however, cite studies that showed quite convincingly that inoculation modifies rumen function, other than fiber digestion. For example, rumen inoculation improves performance during transition from a forage ration to a high-grain diet. If very large amounts of rumen contents are transferred from grain-adapted animals to unadapted animals the transition to the high-grain diet can be made quite quickly without the ensuing problems of ruminal acidosis [188]. Similarly, adaptation of calves to a roughage ration can be accelerated if normal adult microbiota is dosed into calves [189]. More recent studies have confirmed that dosing of newborn dairy calves with rumen fluid from adult cattle causes

them to gain more weight and have less diarrhea than untreated controls [190].

Forage utilization can also clearly be improved with inoculation. The forage legume Leucaena leucocephala produces a goitrogenic compound (3-hydroxy-4[1H]pyri-done) that causes toxicity in ruminants. Synergistes jonesii, is able to break down this compound, and was originally found in Hawaiian goats but not in cattle in Australia. This organism when inoculated into susceptible ruminants grazing L. leucocephala conferred resistance to the toxin [191-193]. Similarly, monofluoroacetate is found in various pasture plants at levels of up to 5 g kg31 [194] and has an LD50 in ruminants of approximately 0.3 mg kg31 of body weight. Gregg et al. [143] inoculated four strains of recombinant B. fibrisolvens, transformed with a gene encoding fluoroacetate dehalogenase into sheep. The inoculated sheep showed a significant reduction in toxicological symptoms after fluoroacetate poisoning when behavioral, physiological, and histological data were compared with those of five uninoculated sheep. It also appeared as if the inoculated strains persisted at approximately 106-107 ml31.

4.2. Genetically modified fiber-degrading bacteria

With regard to fiber degradation in the rumen, much effort has been expended in developing genetically modified bacteria that would have superior fiber-degrading abilities. The construction of genetically modified bacteria has proceeded under the assumption that the rumen mi-crobiota does not produce the correct mixture of enzymes to maximize plant cell wall degradation. For example, Ruminococcus and Fibrobacter do not produce exocellulases that are active against crystalline cellulose, so that adding this activity would make them more potent. Transformation of Ruminococcus and Fibrobacter (the most fibrolytic rumen bacteria) [11] has not been successful, but it has been possible with B. fibrisolvens [195-197], S. bovis [198], and Prevotella spp. [199]. Bu. fibrisolvens is primarily hemicellulolytic [200,201], and is considered ecologically robust making it a good choice as a host for recombinant plant cell wall-degrading enzymes. Published studies [202204] have demonstrated that modification of B. fibrisolvens with GH has been successful, and in vitro digestibility of fiber can be improved, but unfortunately this still does not allow them to compete with the far more fibrolytic Fibro-bacter and Ruminococcus species.

Initially B. fibrisolvens was modified with a xylanase (family 10 GH) from N. patriciarum, a GH family different from the family 11 typically present in B. fibrisolvens [195,205]. GH 10 and GH 11 xylanases differ in their catalytic properties [206] and it was reasoned that the introduction of a GH 11 xylanase would increase its ability to digest fiber [195]. Fiber digestibility could be improved in vitro by more than 28% in comparison with the native untransformed strain [195]. Subsequent studies demon-

strated that these recombinants were unable to compete with highly cellulolytic Ruminococcus strains in vitro and did not persist in the rumen beyond 10-15 days [207].

Cotta et al. [204] transformed the human colonic bacterium Bacteroides thetaiotaomicron strain BTX with a xy-lanase. In a continuous fed-batch culture system with mixed rumen bacteria this strain only persisted when chondroitin sulfate (a mucopolysaccharide used by B. the-taiotaomicron) was added to the medium. In a subsequent study [208], B. thetaiotaomicron was inoculated into dual flow continuous culture fermenters and persisted at approximately 1% of the total population for at least 144 h. Some increase in fiber digestion could be observed. It should be noted that the fermenters used did not support a full complement of rumen microorganisms, and in particular, no protozoa were present [209].

An alternative approach is to create recombinant bacteria that can degrade fiber at low pHs. It is well known that fiber digestion declines in animals on high-grain diets because the pH of the rumen drops below 6.5 [210] and Ruminococcus and F. succinogenes are sensitive to even mildly acidic pH [211]. Russell and Wilson [212] proposed that the addition of fibrolytic activity to an acid-resistant species such as Prevotella would create an organism which would be far more competitive because it would be filling an 'acidic niche' which autochthonous cellulolytic bacteria are unable to fill. Subsequent studies have constructed the appropriate organisms but no in vivo studies have been conducted [213,214].

4.3. Non-genetically modified fiber-degrading bacteria

The establishment of non-cellulolytic bacteria in the gnotobiotic rumen is possible [215,216], but persistence of fibrolytic strains is more difficult and there are no cases of Ruminococcus becoming established [217]. F. succino-genes S85 has been successfully established in gnotobiotic lambs but introduction as a member of a mixed microbial community is a prerequisite [216,218]. Fonty et al. [216] tried to define which combination of 182 rumen cultures was essential to colonization but could only conclude that increased complexity increased colonizing ability. Alternative approaches to establishing cellulolytic strains have included repeated dosing of lambs [15] and adult sheep [219], and completely replacing the contents of the rumen with single strains of cellulolytic bacteria [220] (D.O. Krause and C.S. McSweeney, unpublished). All these attempts have been unsuccessful.

Given that measurement of fiber digestion in vivo would require the inoculant to be at reasonably high levels for at least 2 weeks, and that dosed strains disappear quickly, attempts have been made to measure digestibility by artificially elevating cellulolytic bacteria in the rumen. Deho-rity and Tirabasso [221] increased the numbers of fibro-lytic bacteria in the rumen by feeding a high-cellulose diet composed of purified wood cellulose. There was a 10-fold

increase in the number of cellulolytic bacteria but they could not demonstrate a significant increase in the digestion of alfalfa (lucerne) cellulose using in situ nylon bag digestibility. Similar results were obtained if the number of Ruminococcus species was elevated by continuous inoculation [222]. Collectively, these results would indicate that microbial enzyme activity is not the limiting factor in ru-minal fiber digestion.

Several groups of workers have dosed fibrolytic bacteria into the rumen but have found that the inoculant usually disappears from the rumen [223-226]. One of the reasons given for this is that we do not yet understand how fibro-lytic bacteria exist and reproduce at the fiber surface. What we do know is that organisms do not exist in ecosystems on their own, but reproduce and persist as members of complex microbial communities. Ultimately bacteria persist, or survive, because they reproduce within the physiological and ecological limits of the ecosystem (community-level reproductive strategies) [227]. These strategies derive from the evolution of cooperative networks among microorganisms in which some members cleave specific bonds, others utilize particular substrates, and still others produce inhibitors [227]. A ruminal example is the production of cellodextrins by cellulolytic bacteria [228], which are utilized by non-structural carbohydrate-fermenting bacteria (NSC). The NSC in turn produce ammonia and branched-chain volatile fatty acids that are consumed by cellulolytic bacteria [229,230]. In the case of fibrolytic organisms, it is likely that our understanding of nutrient requirements and other interactions between organisms at the fiber surface is inadequate.

In the continuous dosing studies of both lambs [15] and adult sheep [219] 16S rRNA-based observations indicated that there was a significant increase in the eukaryotic population and this appeared to be primarily the result of an increase in protozoa [15]. Protozoa have a considerable capacity to ingest rumen bacteria [231-233] and in vitro studies with sheep rumen fluid have shown that lysis of Methanobrevibacter, and Selenomonas ruminantium decreased significantly when protozoa were absent from the rumen fluid [234]. Sharp et al. [235] have also demonstrated that an inoculated Lactobacillus plantarum strain disappeared from the rumen largely because of protozoal predation. There were also suggestions of protozoal predation of F. succinogenes S85 in the gnotobiotic study of Fonty et al. [216]. In contrast, there did not appear to be a suppression of F. succinogenes S85 when dosed to protozoa-free gnotobiotic lambs in the presence of a complex mixture of axenic rumen bacteria [236].

Bacteriocin production by certain cellulolytic bacteria has only recently become a subject of research and may be an essential component in the formation of cooperative microbial networks. R. albus strains can produce bacter-iocin-like substances that inhibit the growth of R. flavefa-ciens but not F. succinogenes [237,238]. There also appears to be an unusually high incidence of bacteriocin-like ac-

tivity among Butyrivibrio isolates and butyrivibriocin has been isolated from B. fibrisolvens AR10 [239]. How the ability to produce bacteriocins, or resistance to bacterio-cins, is involved in the establishment and persistence of dosed ruminal bacteria is difficult to assess with current knowledge, but these compounds may have important ecological consequences.

When cellulolytic bacteria are grown together in diculture, cellulose degradation is often below that of the pure culture [201,240], which is probably the result of competitive and non-competitive interactions between cellulolytic bacteria. Shi [241] demonstrated that cell numbers of individual species were approximately equal in cellulose excess dicultures of R albus plus R flavefaciens, R. albus plus F. succinogenes, and R. flavefaciens plus F. succinogenes. However, when cellulose was limiting R. flavefaciens > R. albus, R. flavefaciens > F. succinogenes, and F. succinogenes > R. albus. These competitive outcomes were likely the result of the superior ability of R. flavefaciens to adhere to cellulose [242]. It is interesting to note that R. albus survived under cellulose-limited conditions. This was probably a combination of its ability to utilize glucose (R. flavefaciens does not) [243], grow at low concentrations of cellobiose [244], and its capacity to produce bacteriocins [238].

5. Modification of fiber by exogenous means

5.1. Chemical and mechanical treatments

There are a number of well-established technologies that can reliably be used to increase the digestibility of fiber in the rumen based on mechanical and chemical treatment of plant material before it is consumed. In this review we will not discuss this technology, which has been previously considered [7,245-247]. We will only concentrate on developments of some newer technologies.

5.2. Plants genetically modified for plant cell wall composition

5.2.1. Lignin synthesis

The relationship between lignin and carbohydrates in the plant cell wall is complex, constituting a range of compounds (Fig. 5), various lignin iterations with plant cell walls (Fig. 2), within the context of a multifaceted lignin biosynthetic pathway (Fig. 6). Changes to the components of lignin result in improvements in digestibility. Natural or chemically induced 'brown midrib' (bm) mutations in several forage species result in altered lignin contents and consequently higher digestibilities [248]. Two of the bm mutations in maize occur in genes that encode enzymes in the lignin biosynthetic pathway (Fig. 6). The bm1 mutation affects the gene encoding cinnamyl alcohol dehydrogenase [249] and the bm3 mutation results in changes to

the gene encoding caffeic acid O-methyltransferase (COMT) [250]. These mutants clearly indicate the potential for manipulation of the lignin biosynthetic pathway to improve digestibility of forages.

Recently, molecular strategies for manipulating the enzymes of lignin biosynthesis have contributed greatly to our understanding of the process of lignin biosynthesis and the effects of altered lignin content and composition on digestibility [251]. The enzymes that catalyze lignin formation are encoded by multi-gene families and may carry out a number of related reactions with variable specificity. These factors generate an enormous plasticity in lignin structure that allows adaptation during development and environmental stress [251]. Consequently, the effects on lignin structure of suppressing a single enzymic reaction are not entirely predictable.

Modification of lignin biosynthesis has been achieved by sense and antisense suppression of plant gene expression. These experiments have produced somewhat conflicting results showing that changes both to lignin content and to composition appear to alter cell wall digestibility. Transgenic tobacco plants with reduced expression of cin-namaldehyde dehydrogenase (CAD) produced lignin with a decreased S/G ratio and increased degradability even though there was no change to the total amount of lignin [252]. Similarly, down-regulation of CAD activity in alfalfa altered the composition of the lignin and led to improved digestibility in the rumen [253]. The activity of COMT has also been reduced by sense or antisense suppression in a number of species. The digestibility of cell walls from tobacco increased following suppression of COMT activity, which caused a reduction in the S/G ratio without altering total lignin content [254]. Similar improvements to digestibility were seen in Stylosanthes hu-milis, a tropical forage legume, when the composition of

Tyrosine

p-coumaryl alcohol coniferyl alcohol slnapyl alcohol

Fig. 5. Major monolignol constituents of lignin in their hydrocinnamic acid. Redrawn from [30].

CafleicCOMj; Ferulic_F5H

p-Coumaraldehyde

acid UCL ► Feruoyl-CoA I CCR Coniferaldehyde

-►5-Hydroxy C0MV Sinapic ferulic acid

5-Hydroxyferuloyl acid I CCR

acid |4CL

Sinapoyl-IA

5-Hydroxy coniferaldehyde

CCR Sinapaldehyde

p-Coumaryl alcohol

Coniferyl alcohol

I AD Sinapyl alcohol

LIGNIN

Fig. 6. The lignin biosynthetic pathway. The lignin biosynthetic enzymes are: PAL = phenylalanine ammonia-lyase, TAL = tyrosine ammonia-lyase, C4H = cinnamate 4-hydroxylase, C3H = 4-hydroxycinnamate 3-hy-droxylase, COMT = caffeic acid 3-O-methyltransferase, F5H = ferulate 5-hydroxylase, 4CL = 4-coumarate:CoA ligase, CCoA-3H = coumaroyl-coenzyme A 3-hydroxylase, CCoA-OMT = caffeoyl-coenzyme A O-meth-yltransferase, CCR = cinnamoyl-CoA reductase, and CAD = cinnamyl alcohol dehydrogenase [352].

lignin was altered by antisense suppression of the COMT gene [255]. Although these results suggest that lignin composition is an important factor in digestibility, contradictory results were obtained from a study of induced lignin polymerization in a model system. The digestibility of cell walls from maize suspension-cultured cells was investigated following induced polymerization of monolignols added in various proportions.

While lignification substantially reduced the degradabil-ity of the cell walls compared to controls, varying the ratios of monolignols (Fig. 5) had no effect [256]. Some experiments with transgenic plants concur with these results. Sewalt et al. [257] showed that the increased digestibility of tobacco stems with suppressed COMT activity was associated with reduced lignin content, not with changes in composition. An increase in digestibility was also achieved in transgenic tobacco plants by down-regulation of caffeoyl coenzyme A 3-O-methyltransferase, which caused a small change in lignin content but no change in S/G ratios [258]. Furthermore, major changes to lignin composition have been shown to have no effect on digestibility. The fahl mutants of Arabidopsis thaliana are deficient in ferulate-5-hydroxylase and produce lignin with no syringyl units. There was no difference in cell wall degradability of these mutants compared to wild-type plants [259].

It seems clear that a reduction in lignin content can improve digestibility. In all cases where a reduction in lignin content was achieved in transgenic plants, digestibility increased. Presumably, this increase is due to improved access of rumen microorganisms and secreted enzymes to plant cell wall polysaccharides. Lignin appears to play a major role in cross-linking and anchoring cell wall

polysaccharides. In an A. thaliana mutant in which lignin was reduced by 50% by a defect in the gene encoding cinnamoyl-CoA reductase, the cell wall architecture was altered to the extent that xylem vessels collapsed [260]. Detrimental changes to plant growth and metabolism have been described in other transgenic plants with reduced lignin contents [261].

Notwithstanding the effects of reduced lignin content, there are many documented examples of improvement to digestibility achieved through altered lignin composition. These improvements may be due to associated changes in cell wall composition and structure. Alterations of single enzymes can affect other products of the phenylpropanoid pathway, which are known to have multiple roles in metabolism. Furthermore, altering one enzyme may interrupt the very fine control mechanisms that maintain the heterogeneity of lignin across different cell types and developmental stages [251]. A recent study confirmed that different alterations to lignin biosynthetic enzymes could cause different changes to spatial organization of lignin and cell wall polysaccharides at the cellular and subcellular levels [262]. The interactions between lignin and cell wall poly-saccharides are likely to be a very important determinant of digestibility.

An improved understanding of the types and quantity of lignin required by plant cells for normal development will be important in genetic manipulation of plant cell walls for improved digestibility. Large decreases in lignin concentration appear to cause defects in plant growth and metabolism [259,260]. More subtle changes to lignin concentration and cell wall ultrastructure may be as effective as reducing the overall lignin content. Molecular tools for manipulating plant cell wall architecture are becoming available, and offer new approaches for understanding and improving forage digestibility.

5.2.2. Cellulose synthesis

Some progress has also been made recently in our understanding of the regulation of cell wall polysaccharide synthesis and structure. The model plant species A. thaliana has proved particularly valuable in these studies because of the ease of screening large numbers of plants and the availability of a complete genome sequence. Mutations have been found in genes encoding cellulose synthesis that act in primary cell wall synthesis [263] and in secondary cell wall synthesis [264]. The mutant plants had no changes in concentrations of lignin or non-cellulose poly-saccharides, but major changes to cell wall ultrastructure. There is some evidence for coordinated regulation of lig-nin and cellulose deposition in tree species. In transgenic Populus trees with reduced activity of 4-coumarate:CoA ligase, the reduction in the amount of lignin was counteracted by an increase in the amount of cellulose, so that the total lignin-cellulose mass was unchanged [265]. Genes encoding other enzymes involved in cell wall synthesis have also been isolated from Arabidopsis recently, includ-

ing a family of genes related to xyloglucan fucosyltrans-ferases [266]. Analysis of cell walls from plants altered in expression of these genes will help assemble models of cell wall structure.

5.3. Lignolytic fungi

In nature, lignin is degraded by several groups of organisms, among which white-rot fungi (WRF) belonging to the basidiomycetes are one of the most active groups [267,268]. Two families of ligninolytic enzymes are widely considered to play a key role in this process: phenol oxidase (laccase) and peroxidases (lignin peroxidase (LiP), manganese peroxidase (MnP)) (see reviews [32,269]). A number of genera produce these enzymes, which include Pleurotus, Phanerochaete, Coprinus, and Pycnoporus spp. [270].

Initially fungal cells colonize plant material and utilize soluble carbohydrates (although this is not true for all strains) before degrading lignin. Initial work by Agosin et al. [271] indicated that in vitro digestibility of wheat straw increased from 38 to 68% when treated with strains of Dichomitus squalens and Cyathus stercoreus. Agosin et al. [272] did detailed analyses of structural components of wheat straw to assess the effects of WRF on forage chemistry. There was a rapid degradation of 14C-labelled lignin until a plateau was reached, but further weight loss was a result of carbohydrate degradation.

Gamble et al. [273] used nuclear magnetic resonance to investigate the structural changes that occurred when Ceri-poriopsis subvermispora and Cyathus stercoreus delignified Bermuda grass (Cynodon dactylon) stems. The in vitro digestibility of Bermuda grass after 6 weeks of incubation was improved from 29 to 32% by C. stercoreus, and from 63 to 77% by C. subvermispora, while dry matter losses were about 20%. C. subvermispora removed 23% and C. stercoreus 41% of total aromatics. Removal of aromatic compounds, including ester-linked p-coumaric and ferulic acid, occurred before that of carbohydrates. However, even highly specialized lignolytic fungi that preferentially degrade lignin will start consuming the structural carbohydrates once the tissue has been delignified [274-276], making it important to stop the process if animal feeding is contemplated.

One of the problems with this process at an agricultural scale is that putrefying bacteria on the forage will overgrow causing the forage to rot. To solve this problem the material can be pasteurized during solid-state fermentation. This is the most successful delignification process available because temperature, oxygen, aeration etc., are tightly controlled. The process is technically viable at an industrial scale but not economically feasible for farmers because of the energy inputs. For example, Zadrazil et al. [277] incubated 1500 kg of wheat straw with Pleurotis spp. and demonstrated improved in vitro digestibility (40% to 53.7% dry matter disappearance).

Huttermann et al. [278] described a novel process for the recycling of agricultural wastes by Pleurotus spp. of fungi. This included pasteurization of wet straw by solar heat, treatment with detergents and amendment of straw with wastes from the food industry, such as potato pulp and tomato pomace. Careful analysis found this methodology to be suitable for on-farm application where low energy requirements are a prerequisite. In vitro and in vivo experiments have clearly shown that biological treatment with Pleurotus spp. improves the availability of poor roughage for animals, owing to its effect on cellulose and lignin. In feeding experiments, rams fed fungus-treated straw exhibited increased body weight [278].

Alternative means of pasteurizing the material are the use of chemicals such as NaOH, H2SO4, and NH3 that inhibit putrefying bacteria before treatment with WRF [104,113,118]. However, WRF are typically sensitive to high-pH environments and this can restrict the use of chemicals. Coprinus spp. of fungi are alkaliphilic and can grow at pHs as high as 10.0 [279]. This particular property of Coprinus fungi suggests a solution to two of the practical requirements of forage conservation, prevention of putrefactive microorganisms and increasing the protein content of the feed. Urea on forage rapidly converts to ammonia due to the activity of autochthonous microorganism on the forage; the pH consequently increases resulting in inhibition of many putrefying organisms. If Co-prinus fungi are used delignification can proceed in the absence of putrefying organisms while at the same time increasing the NPN content of the forage [279-282].

Another mechanism by which putrefying bacteria are inhibited and the degradation of cellulose retarded is by ensiling plant material after initial treatment with WRF. The anaerobic conditions will prevent activity by fungi due to the lack of oxygen, and the low pH produced by the acidic fermentation in the silage will retard the growth of putrefactive bacteria. Yang et al. [283] used a two-stage process that combined solid-state fermentation and ensiling with corn straw. The solid-state fermentation increased the level of protein in the feed from 6.7% to 14.7% but decreased the cellulose by 38.0% and hemicellulose by 21.2%. A high-protein roughage with increased digestibility resulted after ensilage of the material.

5.4. Exogenous enzymes

5.4.1. Historical perspective

During the 1960s evidence that the direct application of enzymes to forage before consumption can improve cattle performance became available [284]. In comparison to controls, cattle gained 6.8-24.0% more weight and converted feed 6.0-21.2% more efficiently when ground maize, oat silage, maize silage, or lucerne hay was treated with an enzyme cocktail containing amylolytic, proteolytic, and cellulolytic enzymes. Four different enzyme preparations given in combination with the growth stimulant diethyl-

stilbestrol allowed cattle fed a maize-lucerne hay diet to gain 14% more weight than controls [285]. Further studies confirmed that enzyme supplements could improve the performance of cattle on silage diets [286], but other studies have not necessarily supported this work and have even demonstrated negative results. Leatherwood [287] added a fungal enzyme extract to a grain-supplemented lucerne diet fed to young calves and found no improvement in gain or feed utilization. Unfortunately enzyme cocktails containing enzymes have also resulted in quite significant decreases in animal performance when applied to forage-based diets fed to ruminants [288-290].

These earlier studies have provided valuable information on the potential benefits of enzymes for beef cattle production. More recent studies have been designed to address issues related to the inconsistency of results by investigating the effects of feed type [291], application levels [292], enzyme products [293,294], and enzyme applications [292,295]. Other factors, such as whether dry forage or silage is used [291,293], the enzyme is infused directly into the rumen or not [295,296], have been investigated. Given this variation in responses it is obvious that a mechanistic understanding of how enzymes interact with forages and how synergies between exogenous and autochthonous microbiotas in the rumen occur are still not well understood.

5.4.2. Mode of action

An early criticism of enzyme inoculation into the rumen was that the enzymes were likely to undergo inactivation by the proteolytic activity of rumen fluid. It has subsequently been demonstrated that CMCase and xylanase activities contributed from endogenous sources can remain active in the rumen [297] and all exogenous enzyme activity is not necessarily inactivated once introduced into the rumen [297]. The survival of exogenous enzymes in the rumen raises the prospect that enzymes may improve digestion through the direct hydrolysis of ingested feed. Several researchers have demonstrated that exogenous enzymes may enhance fiber digestion by ruminal microorganisms in vitro [298,299] as well as in situ but these observations do not confirm work by other researchers [298,300].

Release of reducing sugars by exogenous enzymes is probably an important mechanism by which exogenous enzymes operate [292]. The degree of sugar release is dependent on the feed type as well as the type of enzyme. For example, only two of 11 enzymes tested released significant amounts of reducing sugars from barley silage [299]. In addition, the enzymes most effective at releasing reducing sugars from lucerne hay were not those that released the most reducing sugars from barley silage. The release of sugars from feeds is at least partially the result of solubilization of NDF (neutral detergent fiber) and ADF (acid detergent fiber) [301,302] and is consistent with the observed increases in the soluble fraction and

rate of in situ digestion [295,301]. However, most studies have not found that exogenous enzymes improve the extent of in situ or in vitro dry matter digestion [301, 303,304]. These studies would suggest that exogenous enzymes only digest substrates that would normally be digested by enzymes produced by the autochthonous micro-biota in the rumen. Additionally, although exogenous enzymes affect the release of soluble carbohydrates, the amount liberated represents only a minute portion of the total carbohydrate present in the diet. It would be difficult to attribute any production responses solely to the increased availability of reducing sugars given that comparable increases in yield were not seen when up to 9% of total dietary dry matter was supplied as molasses [305].

Although adding exogenous enzymes may increase the activity of xylanases and cellulases in ruminal fluid, enzyme activity in the rumen fluid usually represents less than 30% of the total enzyme activity in the rumen, the remainder being associated with the feed particles [306308]. For example, application of fibrolytic enzymes to a grass hay diet fed to sheep prior to consumption increased endoglucanase activity and xylanase activity in ruminal fluid, but this activity accounted for only 0.5% of the total endoglucanase activity in the rumen [298]. Given that exogenous enzymes represent only a small fraction of the ruminal enzyme activity, and that the ruminal microbiota is inherently capable of digesting fiber it is difficult to envision how exogenous enzymes would enhance ruminal fiber digestion through direct hydrolysis [298].

A more likely explanation for the mode of action of exogenous enzymes is that they work synergistically with the rumen microbiota. This would only be true if the exogenous enzyme inoculant contained enzymes not produced by rumen microbes and was therefore contributing unique activity. Presently, it does not seem likely that the addition of exogenous enzymes has added unique activity to the rumen as only the rate, and not the extent, of cell wall digestion has been improved [306,309]. Of course, as discussed previously, fibrolytic enzymes normally act as members of a cellulosome complex, presumably because there are a range of glycosidic bonds that need to be hydrolyzed synergistically. A logical extension of this reasoning would therefore be to add enzymes that are not produced by rumen microorganisms and attack structures in forage that result in an increased extent of digestion. Lignolytic enzymes clearly fit within these requirements.

5.4.3. Lignolytic enzymes

As discussed, lignin is clearly one of the major constraints to plant cell wall degradation (Figs. 2 and 5). Most of our understanding of the enzymology of lignin biodegradation comes from studies of a single WRF species, Phanerochaete chrysosporium. The principal enzymes implicated in this process are LiP and MnP [310-312]. Despite the extensive literature on LiP and Mn-dependent peroxidases, there is relatively little information on the

depolymerization of synthetic lignin in vitro [310,313315], and it has not been possible to demonstrate extensive lignin depolymerization using isolated LiP or MnP to date. Furthermore, many WRF, including a number of aggressive lignin degraders, seem to operate without expression of LiP activity [316]. Conversely, the WRF Len-tinula edodes achieves an almost imperceptible rate of lig-nin biodegradation despite producing a greater specific LiP activity than optimally expressed P. chrysosporium [251]. Similarly, a LiP-overproducing mutant of this fungus did not show an increased rate of lignin mineralization [317].

Laccases (benzenediol:oxygen oxidoreductase, EC 1.10.3.2) are produced by WRF in combination with LiP and MnP [318,319]. The role played by laccases in lignin degradation has remained obscure since the low redox potential of this enzyme appeared to make it incapable of oxidizing non-phenolic lignin constituents. However, the identification of efficient lignin-degrading WRF that lack LiP and MnP has stimulated research on the role played by laccases in this process. The redox potential of laccases is lower than that of LiP and horseradish peroxi-dase [320] and it was thought that this low redox potential precluded laccases from playing a significant role in the oxidation of non-phenolic polymeric lignin. However, recent studies show that, in the presence of appropriate low-molecular-mass 'mediators' (such as 2,2'-azino-bis-(3-eth-ylthiazoline-6-sulfonate (ABTS) or 3-hydroxyanthranilic acid (3-HAA)) laccase is able to oxidize a wide range of other aromatic lignin compounds [321-323]. These recent studies have indicated that laccases may play a far more important role in lignin biodegradation than was previously thought.

The WRF Pycnoporus cinnabarinus was recently demonstrated to degrade lignin in the absence of both LiP and MnP [323-325]. Under ligninolytic conditions, laccase was the predominant phenol oxidase secreted and neither LiP nor MnP was produced [322]. Despite the absence of these two LiPs, lignin appeared to be degraded just as rapidly, and to the same extent, as P. chrysosporium. The mechanism by which laccases can assume some of the function of the LiP was described by [251,326-328]. The identification of 3-HAA as a naturally occurring redox mediator for laccase is the first evidence to support this ligninolytic system as having equivalent potential to ligninolytic systems based on LiP or MnP [251,316,322-324].

Synthetic compounds that have been used as redox mediators in combination with fungal laccases include 3-HAA, ABTS and 1-hydroxybenzotriazole which has been used in a process-scale pulp bio-bleaching application [329]. The laccase/3-HAA system was significantly more effective in degrading 14C-labelled lignin (polymerized guaiacyl dehydroperoxidase) to fragmentation products (with molecular masses ranging from 4000 to monomeric units) than was the ABTS/laccase couple [322]. The compound 3-HAA is produced as an intermediate in the bio-

synthesis of a major fungal orange pigment (cinnabarinic acid) by the WRF P. cinnabarinus.

Given that the enzymes involved in lignin oxidation are too large to penetrate the unaltered wood cell wall, the use of low-molecular-mass, diffusible compounds to oxidize the polymer is a logical solution. Although the exact reactive species that mediate the laccase-catalyzed oxidation remain unidentified, it is clear that diffusible and reactive low-molecular-mass compounds are responsible for the degradative attack on the lignin polymer [251]. It is likely that each laccase has a preferred low-molecular-mass 'mediator' substrate, which may represent a major secreted metabolite. It has been proposed that LiP uses the veratryl cation radical as mediator [329]. Finding the physiological laccase substrate and matching it to the laccase secreted by the fungus has recently been applied to an industrial process. It appears that a laccase and mediators with NO, NOH, or HRNOH groups can be combined in a lac-case-mediator system (Lignozyme® process) and are effective in delignifying wood in a pilot pulp and paper process [329]. Given the success with this process it is likely that it could be extended to forage.

6. (Gen)omics and metagenomics

High-throughput DNA sequencing and the advent of the '-omics' disciplines now offer the potential to obtain a complete blueprint for the lifestyle of a specific microbe, and to assess its genetic potential in a comparative and functional manner. Although early initiatives in microbial genome sequencing focused on microbes with either small genomes (e.g. mycoplasmas and some archaebacteria) and (or) pathogens, similar approaches with agriculturally and environmentally relevant microbes have been conducted. The genomes of the rumen microbes with relevance to fiber degradation for which genome sequence is available are F. succinogenes, R. albus, and P. ruminicola strain 23.

The F. succinogenes genome has yielded significant insights, and some of those related to GH are detailed in this publication [330]. From the gene list at least 24 genes encoding endoglucanases and cellodextrinases were identified, far exceeding the six genes previously characterized by recombinant DNA strategies employing E. coli as a cloning host. A relatively large number of these genes (13/24) appeared to encode GH belonging to enzyme family 5, but family 7 and one family 45 GH were also identified. Three genes, not previously described, appeared to encode CelF homologs, which possess multiple catalytic domains. Interestingly, no family 6 or family 48 GH were identified, up until now thought to be the source of exo-acting cellulases and processive endoglucanases in eu-bacteria. An additional 23 genes were identified that were presumed to encode xylanases and other enzymes that hydrolyze non-cellulosic polysaccharides. The F. succino-genes genome encodes type I and type II secretion systems,

in addition to PilT homologs that coordinate twitching motility in other Gram-negative bacteria. The genome sequence data have already revealed the limited scope of our knowledge of the genetic blueprint of polysaccharide hydrolysis in F. succinogenes, the most intensively studied aspect of this bacterium's physiology. The vast majority of this genome sequence carries sequence information that cannot confidently be assigned to any specific function.

The R. albus project has revealed a number of ORFs containing type I dockerins, supportive of recent biochemical and genetic evidence that R. albus produces a cellulo-some-like complex. It has also long been recognized that efficient cellulose hydrolysis by R. albus is conditional on the provision of micromolar amounts of PAA/PPA, which elicit substantial changes to cell surface ultrastructure. Cel-lulase activity is also retained within cell-associated, high-molecular-mass protein complexes, thought to be cellulo-somes. A combination of several '-omics' approaches was recently used to define this response [331,332]. Two-dimensional polyacrylamide gel electrophoresis revealed two proteins, 'PpaA' (108 kDa, pi 5.6) and 'PpaB' (94 kDa, pi 5.4), which increase dramatically in extracts prepared from cells cultured in the presence of PAA/PPA. These polypeptides were subjected to N-terminal sequence analysis by Edman degradation and a peptide mass fingerprint from tryptic digests of each protein was generated by matrix-assisted laser desorption/ionization time of flight analysis. These proteomic data were then used to query the R. albus genome sequence, and the two ORFs encoding these proteins were identified. Based on these findings, 'PpaA' is a family 9 GH (Cel9B) and 'PpaB' is a family 48 GH (Cel48A), and homology searches suggest these proteins provide processive endoglucanase (Cel9B) and exoglucanase (Cel48A) activities. These results show for the first time some of the enzymes rate-limiting to cellulose hydrolysis by R. albus. PAA/PPA apparently does not have the same effect when xylan is the substrate [333].

Despite the unequivocal value associated with whole genome sequencing, it is still cost-prohibitive for many investigators to obtain a similar degree of sequence data for multiple strains of the same microbe, or for related species. Subtractive hybridization (SH) can be used as a means to recover 'unique' genomic information from other strains of Ruminococcus spp. and F. succinogenes. Using these methods Antonopoulos and White [334] have generated a set of 288 clones that are unique to R. flavefaciens FD-1 with respect to R. flavefaciens JM1. This clone library includes sequences with homology to a non-ribo-somal peptide synthetase from Streptomyces avermitilis, and the restriction endonuclease SaH. These clones were also used to query the R. albus genome, and clones with significant sequence identity to genes present in R. albus include a putative acetyl-CoA synthetase/carbon monoxide dehydrogenase (78% identical) and a nitrogenase subunit (60% identical). These results substantiate how SH can broaden the scope of functional and comparative ge-

nomics in a cost-effective manner, and will facilitate the examination of gene diversity and genome plasticity among related ruminal microbes. These methods might also help elucidate which gene(s), as well as other ecological or physiological process(es), are rate-limiting to fiber degradation. By doing so, some of the major 'informational' constraints to improving ruminal fiber degradation may ultimately be alleviated, and hypothesis-driven rather than empirical experimental designs will be employed.

Microbial genome sequencing is a useful tool that will allow us to extend our understanding of individual fibro-lytic species of rumen bacteria. However, we should keep these genome sequencing efforts in context. It is by no means the end game given that the majority of microbial species cannot be, or have not been, cultured. As discussed, microorganisms live in community and analyses of individual organisms do not comprise the complex ecological networks that optimize ecosystem function found in most microbial systems, including those for fiber digestion. These complex systems are not well understood and it is by studying fiber digestion at a community genomic level (the metagenome) that the new clues to manipulating fiber digestion will emerge.

The analysis of the rumen metagenome is complicated by the fact that the majority of microorganisms have not been cultured (estimates range from 85 to 95% [335-337]), and probably comprise upwards of 1000 individual species of bacteria, fungi, and protozoa [338]. This situation is not unique to the rumen and the vast majority of microbes in the biosphere, often thousands of species in a single environmental niche [339]), cannot be grown in the laboratory. It is estimated that, on average, less than 1% have ever been identified [340]. Traditional methods for culturing microorganisms fail to represent the scope of microbial diversity in nature and they limit analysis to those that grow under laboratory conditions [341,342].

The use of PCR in conjunction with phylogenetically stable molecular markers like 16S rDNA as species-specific identifiers has provided us with the ability to detect single cells in microbial ecosystems [337,343], irrespective of their culturability, and thereby widened our view of biodiversity [343]. The recent surge of research in molecular microbial ecology provides compelling evidence for the existence of many novel types of microorganisms in high numbers and provides an entirely different approach for tapping into the potentially limitless resource of uncultured bacteria. The rumen is no different, and clone libraries based on 16S rRNA clearly suggest that a vast reservoir of physiologies is present in the rumen [335-337]. Handelsman et al. [344] first coined the term 'metagenome' to describe genomic and associated functional analysis at the community level. One of the keys to metagenomic studies is the ability to clone large fragments of community DNA (> 100 kb) into bacterial artificial chromosomes (BAC). BAC vectors are controlled for copy number and insert stability, allowing single genes and even operons to be

cloned. An extremely useful phylogenetic spin-off is that because of the insert size many clones will contain 16S genes, which together with an analysis of associated functional genes provides a phylogenetically informed view of uncultured diversity [344,345].

This technology is still in its infancy and issues such as representative cloning, quantitative lysis, and expression still need to be addressed [346]. In particular, quantitative lysis is difficult to overcome. It is imperative that large fragments be cloned, and to do this techniques based on pulsed field gel electrophoresis must be employed [345]. To prevent shearing of DNA in aqueous solutions cells are embedded in agarose plugs and lysed in situ. This immediately presents the problem of quantitative lysis of the microbiota. We have developed technology in our laboratory that allows us to obtain quantitative lysis and at the same time obtain clones, on average, of above 150 kb (D.O. Krause et al., unpublished). The importance of clone size should not be underestimated. If insert size averages 100 kb, a library of 50000 clones will have a low probability of covering the genome of an organism that is less than 0.5% of the population. If, on the other hand, insert size is 200 kb a library of 50 000 clones will contain organisms of as little as 0.2% of the population. Libraries of less than 100 kb require clone libraries of impossible dimensions. Handelsman et al. [344] estimated that in the order of 106 BAC clones with an average insert size of 100 kb will be necessary for complete representation of the soil metagenome. This seems an impossible task, and with the resources available to most institutions, it is. However, if one accepts that major microbial species are the target of these activities, and that the metagenomes include incredible molecular diversity, 50 000 clones may be adequate.

The choice of appropriate clone recipients will be important as cloning of metagenomic DNA will be in heter-ologous hosts, and primarily in E. coli, simply because high clone numbers are necessary to get anywhere near a representation of the microbial diversity present in the metagenome [346]. Similarly, downstream cloning in Bacillus species or Streptomyces lividans for the sake of optimizing gene expression will be the method of choice [346]. Although activity screening based on the functional expression of enzymes is attractive because of unique physiologies that will be discovered, problems associated with heterologous gene expression would naturally limit its success. However, it appears that diversity in metagenomic DNA is so great that even inherently flawed cloning strategies represent so much molecular novelty that at present optimization of these strategies seems irrelevant in comparison to the constraints in downstream processing of this 'mega-diversity'.

Already metagenome research has provided some unique insights into microbial ecosystems. Analysis of large clones derived from oceanic libraries has revealed the importance of proteorhodopsin obtained from BAC inserts of uncultured bacterioplankton [347]. It was dem-

onstrated that photoactive proteorhodopsin is present in oceanic surface waters and that there may be an extensive family of globally distributed proteorhodopsin variants. The protein pigments comprising this rhodopsin family were spectrally tuned to different habitats and absorb light at different wavelengths in accordance with light available in the environment. These authors suggest that proteorho-dopsin-based phototrophy is a globally significant oceanic microbial process.

In another study [348] on oceanic microbial assemblages, large-scale genome sequence analysis was conducted on marine archaea to better describe the population genetics, genome content, and biological properties of naturally occurring, uncultivated pelagic crenarchaeotes. Sequencing and analysis of the entire DNA insert from one Antarctic marine archaeon revealed differences in genome structure and content between Antarctic surface water and temperate deep water archaea. Analyses of the predicted gene products revealed many typical archaeal proteins but also several proteins that so far have not been detected in archaea.

Metagenome analysis is not simply limited to discovery of more and novel diversity, but has practical use in finding fresh enzymes for industrial and agricultural use. This is a logical assumption when one realizes that because of the poor ability to culture environmental microorganisms, most industrial enzymes will necessarily have originated from the small percentage that have been cultured. Using BAC-based technologies, direct DNA extraction and cloning strategies from environmental DNA have already been demonstrated and revealed novel enzyme activities that are of industrial importance [344,349-351]. The list of reported enzyme activities discovered this way (lipase, esterase, amylase, nuclease, chitinase, xylanase) is still rather small, but will undoubtedly grow rapidly. This approach has significant scope for mining the rumen microbiota for novel enzyme activities, or lack of, so that more effective exogenous enzyme additives can be designed.

7. Conclusions

This review has discussed the major problems that face scientists trying to improve plant cell wall digestion in the rumen. A semi-historical survey covered the various attempts and strategies that have been employed to improve fiber digestion in the rumen. It seems clear from this review that several major challenges are presented and that improvement of fiber digestion is possible providing that it is based on a rational scientific basis, something that until now has not always been obvious.

In retrospect, many of the strategies that have been undertaken to improve rumen digestion have floundered because of our lack of understanding of a very complex system. First among these is the fact that we have based almost all our knowledge about rumen microbiology on

just a few species of bacteria, and in fact just a few strains. For example, it is not clear that we are even working with the major cellulolytic rumen bacteria in pure culture. In addition, attempts to improve rumen function by addition of exogenous enzymes have been based largely on availability of enzymes and little attention has been paid to actual enzyme 'requirements' of the rumen. This of course has been because of the fact that we have essentially been ignorant of the functional genomic framework within which the rumen operates; the cellulosome is an example of this. This situation can be likened to a ship's captain who can only see the tip of an iceberg, the mass of which is vast in comparison to what sticks up above the water. To say the least it makes navigation difficult.

(Gen)omics technologies, including metagenomes, now provide rumen microbiologists with their best opportunity, to date, to see the iceberg in its entirety, rather than just its tip, in both a comparative and functional system. This is both an exciting and daunting prospect. Clearly, (gen)-omics will provide a massive increase in the rate of information acquisition, and both novel and conventional methodologies and techniques must flourish for the potential of (gen)omics to be fully realized. For instance, there will be a need for novel in silico methods to mine and extract relevant information from seemingly disparate systems. Renewed interest in microbial physiology and in the isolation of 'unculturable' or 'not-yet-cultured' microbes is also necessary, if we are to fully exploit the opportunities provided by (gen)omics.

Acknowledgements

The authors would like to thank the Australian, New Zealand, and US funding agencies who have supported various aspects of the authors' work.

References

[1] Hungate, R.E. (1966) The Rumen and its Microbes. Academic Press, New York.

[2] Enquist, B.J., Economo, E.P., Huxman, T.E., Allen, A.P., Ignace, D.D. and Gillooly, J.F. (2003) Scaling metabolism from organisms to ecosystems. Nature 423, 639-642.

[3] Fox, D.G., Barry, M.C., Pitt, R.E., Roseler, D.K. and Stone, W.C. (1995) Application of the Cornell net carbohydrate and protein model for cattle consuming forages. J. Anim. Sci. 73, 267-277.

[4] McDermott, J.J., Randolph, T.F. and Staal, S.J. (1999) The economics of optimal health and productivity in smallholder livestock systems in developing countries. Rev. Sci. Technol. 18, 399-424.

[5] Kelley, J. and Paterson, R. (1997) Crop residues as a resourse. The use of fungi to upgrade lignocellulosic wastes. Biol. Int. 26, 16-20.

[6] Hungate, R.E. (1984) Microbes of nutritional importance in the alimentary tract. Proc. Nutr. Soc. 43, 1-11.

[7] Wilkins, R.J. and Minson, D.J. (1970) The effects of grinding, supplementation and incubation period on cellulose digestibility in vitro and its relationship with cellulose and organic matter digestibility in vivo. J. Agric. Sci. 74, 445-451.

[8] NRC (1996) Nutrient Requirements for Beef Cattle. National Academy Press, Washington, DC.

[9] Reichl, J.R. and Baldwin, R.L. (1975) Rumen modeling: Rumen input-output balance models. J. Dairy Sci. 58, 879-890.

[10] Vercoe, P.E. and White, B.A. (1997) Genetics of ruminal anaerobic bacteria. In: Gastrointestinal Microbiology (Mackie, R.I., White,

B.A. and Isaacson, R.E., Eds.), pp. 321-372. Chapman and Hall, New York.

[11] Teather, R.M., Hefford, M.A. and Forster, R.J. (1997) Genetics of rumen bacteria. In: The Rumen Microbial Ecosystem (Hobson, P.N. and Stewart, C.S., Eds.), pp. 427-466. Blackie, Melbourne.

[12] Hespell, R.B., Akin, D.E. and Dehority, B.A. (1997) Bacteria, fungi, and protozoa of the rumen. In: Gastrointestinal Microbiology (Mackie, R.I., White, B.A. and Isaacson, R.E., Eds.), pp. 59-141. Chapman and Hall, New York.

[13] Klieve, A.V. and Swain, R.A. (1993) Estimation of ruminal bacterio-phage numbers by pulsed-field gel electrophoresis and laser densitom-etry. Appl. Environ. Microbiol. 59, 2299-2303.

[14] Klieve, A.V. and Bauchop, T. (1988) Morphological diversity of ruminal bacteriophages from sheep and cattle. Appl. Environ. Microbiol. 54, 1637-1641.

[15] Krause, D.O., Smith, W.J.M., Ryan, F.M.E., Mackie, R.I. and McSweeney, C.S. (2000) Use of 16S-rRNA based techniques to investigate the ecological succession of microbial populations in the immature lamb rumen: tracking of a specific strain of inoculated Ruminococcus and interactions with other microbial populations in vivo. Microbiol. Ecol. 38, 365-376.

[16] Krause, D.O., Bunch, R.J., Conlan, L.L., Kennedy, P.M., Smith, W.J., Mackie, R.I. and McSweeney, C.S. (2001) Repeated ruminal dosing of Ruminococcus spp. does not result in persistence, but changes in other microbial populations occur that can be measured with quantitative 16S-rRNA-based probes. Microbiology 147, 17191729.

[17] Stewart, C.S., Flint, H.J. and Bryant, M.P. (1997) The rumen bacteria. In: The Rumen Microbial Ecosystem (Hobson, P.N. and Stewart, C.S., Eds.), pp. 10-72. Blackie, Melbourne.

[18] Williams, A.G. and Coleman, G.S. (1997) The rumen protozoa. In: The Rumen Microbial Ecosystem (Hobson, P.N. and Stewart, C.S., Eds.), pp. 73-139. Blacke, Melbourne.

[19] Orpin, C.G. and Joblin, K.N. (1997) The rumen anaerobic fungi. In: The Rumen Microbial Ecosystem (Hobson, P.N. and Stewart, C.S., Eds.), pp. 140-195. Blackie, Melbourne.

[20] Devillard, E., Bera-Maillet, C., Flint, H.J., Scott, K.P., Newbold,

C.J., Wallace, R.J., Jouany, J.P. and Forano, E. (2003) Characterization of XYN10B, a modular xylanase from the ruminal protozoan Polyplastron multivesiculatum, with a family 22 carbohydrate-binding module that binds to cellulose. Biochem. J. 373, 495-503.

[21] Henrissat, B. and Bairoch, A. (1993) New families in the classification of glycosyl hydrolases based on amino acid sequence similarities. Biochem. J. 293, 781-788.

[22] Henrissat, B. (1991) A classification of glycosyl hydrolases based on amino acid sequence similarities. Biochem. J. 280, 309-316.

[23] Davies, G.J. and Henrissat, B. (2002) Structural enzymology of carbohydrate-active enzymes: implications for the post-genomic era. Biochem. Soc. Trans. 30, 291-297.

[24] Forsberg, C.W., Cheng, K.J. and White, B.A. (1997) Polysaccharide degradation in the rumen and large intestine. In: Gastrointestinal Microbiology (Mackie, R.I. and White, B.A., Eds.), pp. 319-379. Chapman and Hall, New York.

[25] Lynd, L.R., Weimer, P.J., van Zyl, W.H. and Pretorius, I.S. (2002) Microbial cellulose utilization: fundamentals and biotechnology. Microbiol. Mol. Biol. Rev. 66, 506-577.

[26] Boisset, C., Petrequin, C., Chanzy, H., Henrissat, B. and Schulein, M. (2001) Optimized mixtures of recombinant Humicola insolens cel-lulases for the biodegradation of crystalline cellulose. Biotechnol. Bioeng. 72, 339-345.

[27] Poutanen, K., Tenkanen, M., Korte, H. and Puls, J. (1991) Accessory

enzymes involved in the hydrolysis of xylans. In: Enzymes in Biomass Conversion (Leatham, G.F., Ed.), pp. 426-436. American Chemical Society, Washington, DC.

[28] Hespell, R.B. and Cotta, M.A. (1995) Degradation and utilization by Butyrivibrio fibrisolvens H17c of xylans with different chemical and physical properties. Appl. Environ. Microbiol. 61, 3042-3050.

[29] Hespell, R.B. and Whitehead, T.R. (1990) Physiology and genetics of xylan degradation by gastrointestinal tract bacteria. J. Dairy Sci. 73, 3013-3022.

[30] Carpita, N. and McCann, M. (2000) The cell wall. In: Biochemistry and Molecular Biology of Plants (Buchanan, B., Gruissmen, W. and Jones, R., Eds.), pp. 52-108. John Wiley and Sons, Somerset, NJ.

[31] Bhat, M.K., Hazlewood, G.P., Bedford, M.R. and Partridge, G.G. (2000) Enzymology and other characteristics of cellulases and xylan-ases. In: Enzymes in Farm Animal Nutrition (Bedford, M. and Partridge, G., Eds.), pp. 11-60. CAB International, Wallingford.

[32] Sun, Y. and Cheng, J. (2002) Hydrolysis of lignocellulosic materials for ethanol production: a review. Bioresour. Technol. 83, 1-11.

[33] Siew, N. and Fischer, D. (2003) Twenty thousand ORFan microbial protein families for the biologist? Structure 11, 7-9.

[34] Peterson, J.D., Umayam, L.A., Dickinson, T., Hickey, E.K. and White, O. (2001) The comprehensive microbial resource. Nucleic Acids Res. 29, 123-125.

[35] McGavin, M. and Forsberg, C.W. (1988) Isolation and characterization of endoglucanases 1 and 2 from Bacteroides succinogenes S85. J. Bacteriol. 170, 2914-2922.

[36] Huang, L., McGavin, M., Forsberg, C.W., Lam, J.S. and Cheng, K.J. (1990) Antigenic nature of the chloride-stimulated cellobiosidase and other cellulases of Fibrobacter succinogenes subsp. succinogenes S85 and related fresh isolates. Appl. Environ. Microbiol. 56, 12291234.

[37] Huang, L. and Forsberg, C.W. (1988) Purification and comparison of the periplasmic and extracellular forms of the cellodextrinase from Bacteroides succinogenes. Appl. Environ. Microbiol. 54, 1488-1493.

[38] McGavin, M.J., Forsberg, C.W., Crosby, B., Bell, A.W., Dignard, D. and Thomas, D.Y. (1989) Structure of the cel-3 gene from Fibrobacter succinogenes S85 and characteristics of the encoded gene product, endoglucanase 3. J. Bacteriol. 171, 5587-5595.

[39] Forano, E., Broussolle, V., Gaudet, G. and Bryant, J.A. (1994) Molecular cloning, expression, and characterization of a new endogluca-nase gene from Fibrobacter succinogenes S85. Curr. Microbiol. 28, 7-14.

[40] Malburg Jr., L.M., lyo, A.H. and Forsberg, C.W. (1996) A novel family 9 endoglucanase gene (celD), whose product cleaves substrates mainly to glucose, and its adjacent upstream homolog (celE) from Fibrobacter succinogenes S85. Appl. Environ. Microbiol. 62, 898-906.

[41] Malburg, S.R., Malburg Jr., L.M., Liu, T., Iyo, A.H. and Forsberg, C.W. (1997) Catalytic properties of the cellulose-binding endogluca-nase F from Fibrobacter succinogenes S85. Appl. Environ. Microbiol. 63, 2449-2453.

[42] Iyo, A.H. and Forsberg, C.W. (1994) Features of the cellodextrinase gene from Fibrobacter succinogenes S85. Can. J. Microbiol. 40, 592596.

[43] Teather, R.M. and Erfle, J.D. (1990) DNA sequence of a Fibrobacter succinogenes mixed-linkage beta-glucanase (1,3-1,4-beta-D-glucan 4-glucanohydrolase) gene. J. Bacteriol. 172, 3837-3841.

[44] Cho, K.K., Kim, S.C., Woo, J.H., Bok, J.D. and Choi, Y.J. (2000) Molecular cloning and expression of a novel family A endoglucanase gene from Fibrobacter succinogenes S85 in Escherichia coli. Enzyme Microb. Technol. 27, 475-481.

[45] Gong, J., Lo, R.Y.C. and Forsberg, C.W. (1989) Molecular cloning and expression in Escherichia coli of a cellodextrinase gene from Bacteroides succinogenes S85. Appl. Environ. Microbiol. 55, 132-136.

[46] Buchanan, C.J. and Mitchell, W.J. (1992) Two beta-glucosidase activities in Fibrobacter succinogenes S85. J. Appl. Bacteriol. 73, 243250.

[47] Gong, J. and Forsberg, C.W. (1993) Separation of outer and cyto-

plasmic membranes of Fibrobacter succinogenes and membrane and glycogen granule locations of glycanases and cellobiase. J. Bacteriol. 175, 6810-6821.

[48] Matte, A. and Forsberg, C.W. (1992) Purification, characterization, and mode of action of endoxylanases 1 and 2 from Fibrobacter succinogenes S85. Appl. Environ. Microbiol. 58, 157-168.

[49] Paradis, F.W., Zhu, H., Krell, P.J., Phillips, J.P. and Forsberg, C.W. (1993) The xynC gene from Fibrobacter succinogenes S85 codes for a xylanase with two similar catalytic domains. J. Bacteriol. 175, 76667672.

[50] Cheng, K.J., Selinger, L.B., McAllister, T.A., Yanke, L.J., Bae, H.D., Shin, H.T,, Goto, M., Takenaka, A., Forsberg, C.W., Shelford, J.A., Onodera, R., Itabashi, H., Ushida, K., Yano, H. and Sasaki, Y. (1997) Exploitation of rumen microbial enzymes to benefit ruminant and non-ruminant animal production. In: Rumen Microbes and Digestive Physiology in Ruminants (Itabashi, H., Onocera, R., Sasaki, Y., Ushida, K. and Yano, H., Eds.), pp. 25-34. Karger Landes Systems, Basel.

[51] Shaw, N.D., Irwin, D.C. and Wilson, D.B. (2000) Cloning and se-qeencing of a 6.7 kb frangment of Fibrobacter succinogens S85 DNA fragment encoding four open reading frames homologous to glycosyl hydrolases and showing xylanase activity on RBB xylan plates. GenBank accession number AY007248.

[52] Ha, J.K., Malburg, L.M.Jr., Park, S.-Y. and Forsberg, C.W. (1999) Fibrobacter succinogenes S85 family 10 glycosyl hydrolase XynD (xynD), family 10 glycosyl hydrolase XynE (xynE), and family 10 glycosyl hydrolase XynB (xynB) genes, complete cds. GenBank accession number AF180368.

[53] Cavicchioli, R., East, P.D. and Watson, K. (1991) endAFS, a novel family E endoglucanase gene from Fibrobacter succinogenes AR1. J. Bacteriol. 173, 3265-3268.

[54] Cavicchioli, R. and Watson, K. (1991) Molecular cloning, expression, and characterization of endoglucanase genes from Fibrobacter succi-nogenes AR1. Appl. Environ. Microbiol. 57, 359-365.

[55] Cavicchioli, R. and Watson, K. (1991) The involvement of transcrip-tional read-through from internal promoters in the expression of a novel endoglucanase gene FSendA, from Fibrobacter succinogenes AR1. Nucleic Acids Res. 19, 1661-1669.

[56] Ozcan, N., Cunningham, C. and Harris, W.J. (1996) Cloning of a cellulase gene from the rumen anaerobe Fibrobacter succinogenes SD35 and partial characterization of the gene product. Lett. Appl. Microbiol. 22, 85-89.

[57] Bera, C., Broussolle, V., Forano, E. and Gaudet, G. (1996) Gene sequence analysis and properties of EGC, a family E (9) endoglucanase from Fibrobacter succinogenes BL2. FEMS Microbiol. Lett. 136, 79-84.

[58] Lin, C. and Stahl, D.A. (1995) Comparative analyses reveal a highly conserved endoglucanase in the cellulolytic genus Fibrobacter. J. Bacteriol. 177, 2543-2549.

[59] Flint, H.J., McPherson, C.A., Avgustin, G. and Stewart, C.S. (1990) Use of a cellulase-encoding gene probe to reveal restriction fragment length polymorphisms among ruminal strains of Bacteroides succinogenes. Curr. Microbiol. 20, 63-67.

[60] Miyagi, T., Javorsky, P., Pristas, P., Karita, S., Sakka, K. and Oh-miya, K. (1998) Partial purification and characterization of RalF40I, a class II restriction endonuclease from Ruminococcus albus F-40, which recognizes and cleaves 5'-/GATC-3'. FEMS Microbiol. Lett. 164, 215-218.

[61] Poole, D.M., Hazlewood, G.P., Laurie, J.I., Barker, P.J. and Gilbert, H.J. (1990) Nucleotide sequence of the Ruminococcus albus SY3 endoglucanase genes celA and celB. Mol. Gen. Genet. 223, 217-223.

[62] Ohmiya, K., Shimizu, M., Taya, M. and Shimizu, S. (1982) Purification and properties of cellobiosidase from Ruminococcus albus. J. Bacteriol. 150, 407-409.

[63] Ohmiya, K., Shirai, M., Kurachi, Y. and Shimizu, S. (1985) Isolation and properties of beta-glucosidase from Ruminococcus albus. J. Bacteriol. 161, 432-434.

[64] Ohmiya, K., Takano, M. and Shimizu, S. (1990) DNA sequence of a beta-glucosidase from Ruminococcus albus. Nucleic Acids Res. 18, 671-678.

[65] Ohara, H., Noguchi, J., Karita, S., Kimura, T., Sakka, K. and Ohmiya, K. (2000) Sequence of egV and properties of EgV, a Ruminococcus albus endoglucanase containing a dockerin domain. Biosci. Biotechnol. Biochem. 64, 80-88.

[66] Ohara, H., Karita, S., Kimura, T., Sakka, K. and Ohmiya, K. (2000) Characterization of the cellulolytic complex (cellulosome) from Ruminococcus albus. Biosci. Biotechnol. Biochem. 64, 254-260.

[67] Ohara, H., Miyagi, T., Kaneichi, K., Karita, S., Kobayashi, Y., Ki-mura, T., Sakka, K. and Ohmiya, K. (1998) Structural analysis of a new cryptic plasmid pAR67 isolated from Ruminococcus albus AR67. Plasmid 39, 84-88.

[68] Ware, C.E., Lachke, A.H. and Gregg, K. (1990) Mode of action and substrate specificity of a purified exo-1,4-beta-D-glucosidase cloned from the cellulolytic bacterium Ruminococcus albus AR67. Biochem. Biophys. Res. Commun. 171, 777-786.

[69] Krause, D.O., Bunch, R.J., Smith, J.M. and McSweeney, C.S. (1999) Diversity of Ruminococcus strains: a survey of genetic polymorphisms and plant digesting ability. J. Appl. Bacteriol. 86, 487-495.

[70] Miron, J., Jacobovitch, J., Bayer, E.A., Lamed, R., Morrison, M. and Ben-Ghedalia, D. (2001) Subcellular distribution of glycanases and related components in Ruminococcus albus SY3 and their role in cell adhesion to cellulose. J. Appl. Microbiol. 91, 677-685.

[71] Greve, L.C., Labavitch, J.M. and Hungate, R.E. (1984) a-L-Arabi-nofuranosidase from Ruminococcus albus 8: Purification and possible role in hydrolysis of alfalfa cell wall. Appl. Environ. Microbiol. 47, 1135-1140.

[72] Woo, J.H., Cho, K.K., Min, H.K. and Choi, Y.J. (1995) Cloning of gene for beta glucosidase from Ruminococcus albus 7. Mol. Cell 5, 448-451.

[73] Nagamine, T., Aminov, R.I., Ogata, K., Sugiura, M., Tajima, K. and Benno, Y. (1997) Cloning of xylanase genes from Ruminococcus albus and chromosome mapping of Fibrobacter succinogenes. In: Rumen Microbes and Digestive Physiology in Ruminants (Itabashi, H., Ono-cera, R., Sasaki, Y., Ushida, K. and Yano, H., Eds.), pp. 59-67. Japan Scientific Societies Press, Tokyo.

[74] Nakamura, M., Nagamine, T., Takenaka, A., Aminov, R.I., Ogata, K., Tajima, K., Matsui, H., Benno, Y. and Itabashi, H. (2002) Molecular cloning, nucleotide sequence and characteristics of a xylanase gene (xynA) from Ruminococcus albus 7. J. Anim. Sci. 73, 347-352.

[75] Aurilia, V., Martin, J.C., McCrae, S.I., Scott, K.P., Rincon, M.T. and Flint, H.J. (2000) Three multidomain esterases from the cellulolytic rumen anaerobe Ruminococcus flavefaciens 17 that carry divergent dockerin sequences. Microbiology 146, 1391-1397.

[76] Gardner, R.M., Doerner, K.C. and White, B.A. (1987) Purification and characterization of an exo-beta-1,4-glucanase from Ruminococcus flavefaciens FD-1. J. Bacteriol. 169, 4581-4588.

[77] Wang, W.Y. and Thomson, J.A. (1990) Nucleotide sequence of the celA gene encoding a cellodextrinase of Ruminococcus flavefaciens FD-1. Mol. Gen. Genet. 222, 265-269.

[78] Howard, G.T. and White, B.A. (1988) Molecular cloning and expression of cellulae genes from Ruminococcus albus in Escherichia coli bacteriophage lambda. Appl. Environ. Microbiol. 54, 1752-1755.

[79] Vercoe, P.E., Finks, J.L. and White, B.A. (1995) DNA sequence and transcriptional characterization of a ß-glucanase gene (celB) from Ruminococcus flavefaciens FD-1. Can. J. Microbiol. 41, 869-876.

[80] Vercoe, P.E., Spight, D.H. and White, B.A. (1995) Nucleotide sequence and transcriptional analysis of the celD ß-glucanase gene from Ruminococcus flavefaciens FD-1. Can. J. Microbiol. 41, 27-34.

[81] Wang, W., Reid, S.J. and Thomson, J.A. (1993) Transcriptional regulation of an endoglucanase and a cellodextrinase gene in Ruminococcus flavefaciens FD-1. J. Gen. Microbiol. 139, 1219-1226.

[82] Doerner, K.C. and White, B.A. (1990) Assessment of the endo-1,4-beta-glucanase components of Ruminococcus flavefaciens FD-1. Appl. Environ. Microbiol. 56, 1844-1850.

[83] Cunningham, C., McPherson, C.A., Martin, J., Harris, W.J. and Flint, H.J. (1991) Sequence of a cellulase gene from the rumen anaerobe Ruminococcus flavefaciens 17. Mol. Gen. Genet. 228, 320323.

[84] Rincon, M.T., McCrae, S.I., Kirby, J., Scott, K.P. and Flint, H.J. (2001) EndB, a multidomain family 44 cellulase from Ruminococcus flavefaciens 17, binds to cellulose via a novel cellulose-binding module and to another R. flavefaciens protein via a dockerin domain. Appl. Environ. Microbiol. 67, 4426-4431.

[85] Flint, H.J., Martin, J., McPherson, C.A., Daniel, A.S. and Zhang, J.X. (1993) A bifunctional enzyme, with separate xylanase and beta (1,3-1,4)-glucanase domains, encoded by the xynD gene of Ruminococcus flavefaciens. J. Bacteriol. 175, 2943-2951.

[86] Zhang, J.X., Martin, J. and Flint, H.J. (1994) Identification of non-catalytic conserved regions in xylanases encoded by the xynB and xynD genes of the cellulolytic rumen anaerobe Ruminococcus flavefa-ciens. Mol. Gen. Genet. 245, 260-264.

[87] Asmundson, R.V., Huang, C.M., Kelly, W.J., Yu, P.-L. and Curry, M.M. (1990) The cellulase of Ruminococcus flavefaciens strain 186: Characterization, cloning and use in ruminant nutrition. In: Micro-bial and Plant Opportunities to Improve Lignocellulose Utilization by Ruminants (Akin, D.E., Ljungdahl, L.G., Wilson, J.R. and Harris, P.J., Eds.), pp. 401-409. Elsevier, New York.

[88] Kopecny, J., Logar, R.M. and Kobayashi, Y. (2001) Phenotypic and genetic data supporting reclassification of Butyrivibrio fibrisolvens isolates. Folia Microbiol. 46, 45-48.

[89] Dalrymple, B.P., Swadling, Y., Layton, I., Gobius, K.S. and Xue, G.P. (1999) Distribution and evolution of the xylanase genes xynA and xynB and their homologues in strains of Butyrivibrio fibrisolvens. Appl. Environ. Microbiol. 65, 3660-3667.

[90] Shane, B.S., Gouws, L. and Kistner, A. (1969) Cellulolytic bacteria occurring in the rumen of sheep conditioned to low protein Teff hay. J. Gen. Microbiol. 55, 445-447.

[91] Hespell, R.B., Wolf, R. and Bothast, R.J. (1987) Fermentation of xylans by Butyrivibrio fibrisolvens and other ruminal bacteria. Appl. Environ. Microbiol. 53, 2849-2853.

[92] Sewell, G.W., Aldrich, H.C., Williams, D., Mannarelli, B., Wilke, A., Hespell, R.B., Smith, P H. and Ingram, L.O. (1987) Isolation and character-ization of xylan-degrading strains of Butyrivibrio fibri-solvens from grass-fed anaerobic digester. Appl. Environ. Microbiol. 54, 1085-1090.

[93] Berger, E., Jones, W.A., Jones, D.T. and Woods, D R. (1989) Cloning and sequencing of an endoglucanase (end!) gene from Butyrivibrio fibrisolvens H17c. Mol. Gen. Genet. 219, 193-198.

[94] Berger, E., Jones, W.A., Jones, D.T. and Woods, D R. (1990) Sequencing and expression of a cellodextrinase (ced!) gene from Butyrivibrio fibrisolvens H17c cloned in Escherichia coli. Mol. Gen. Genet. 223, 310-318.

[95] Lin, L.L. and Thomson, J.A. (1991) Cloning, sequencing and expression of a gene encoding a 73 kDa xylanase enzyme from the rumen anaerobe Butyrivibrio fibrisolvens H17c. Mol. Gen. Genet. 228, 55-61.

[96] Lin, L.L. and Thomson, J.A. (1991) An analysis of the extracellular xylanases and cellulases of Butyrivibrio fibrisolvens H17c. FEMS Microbiol. Lett. 68, 197-203.

[97] Hazlewood, G.P., Davidson, K., Laurie, J.I., Romaniec, M.P. and Gilbert, H.J. (1990) Cloning and sequencing of the celA gene encoding endoglucanase A of Butyrivibrio fibrisolvens strain A46. J. Gen. Microbiol. 136, 2089-2097.

[98] Sewell, G.W., Utt, E.A., Hespell, R.B., Mackenzie, K.F. and Ingram, L.O. (1989) Identification of the Butyrivibrio fibrisolvens xylosidase gene (xylB) coding region and its expression in Escherichia coli. Appl. Environ. Microbiol. 55, 306-311.

[99] Utt, E.A., Eddy, C.K., Keshav, K.F. and Ingram, L.O. (1991) Sequencing and expression of the Butyrivibrio fibrisolvens xylB gene encoding a novel bifunctional protein with ß-D-xylosidase and a-L-arabinofuranosidase activities. Appl. Environ. Microbiol. 57, 1227-1234.

[100] Rumbak, E., Rawlings, D.E., Lindsey, G.G. and Woods, D R. (1991) Cloning, nucleotide sequence, and enzymatic characterization of an alpha-amylase from the ruminal bacterium Butyrivibrio fibrisolvens H17c. J. Bacteriol. 173, 4203-4211.

[101] Mannarelli, B.M., Evans, S. and Lee, D. (1990) Cloning, sequencing, and expression of a xylanase gene from the anaerobic ruminal bacterium Butyrivibrio fibrisolvens. J. Bacteriol. 172, 4247-4254.

[102] Zorec, M., Cepeljnik, T., Nekrep, F.V. and Logar, R.M. (2001) Multiple endoxylanases of Butyrivibrio sp. Folia Microbiol. 46, 94-96.

[103] Cepeljnik, T., Grebenc, T., Krizaj, I. and Marinsek-Logar, R. (2003) Endo-1,4-xylanase XynT from the rumen bacterium Pseudobutyrivi-brio xylanivorans : isolation and characterisation. GenBank Accession number AJ543424.

[104] Hespell, R.B., Akin, D.E. and Dehority, B.A. (1997) Bacteria, fungi, and protozoa of the rumen. In: Gastrointestinal Microbiology (Mackie, R.I., White, B.A. and Isaacson, R.E., Eds.), pp. 59-141. Chapman and Hall, New York.

[105] Vercoe, P.E. and Gregg, K. (1992) DNA sequence and transcription of an endoglucanase gene from Prevotella (Bacteroides) ruminicola AR20. Mol. Gen. Genet. 233, 284-292.

[106] Matsushita, O., Russell, J.B. and Wilson, D.B. (1991) A Bacteroides ruminicola 1,4-beta-D-endoglucanase is encoded in two reading frames. J. Bacteriol. 173, 6919-6926.

[107] Wulff-Strobel, C.R. and Wilson, D.B. (1995) Cloning, sequencing, and characterization of a membrane-associated Prevotella ruminicola B14 beta-glucosidase with cellodextrinase and cyanoglycosidase activities. J. Bacteriol. 177, 5884-5890.

[108] Vercoe, P.E. and White, B.A. (1997) Genetics of ruminal anaerobic bacteria. In: Gastrointestinal Microbiology (Mackie, R.I., White, B.A. and Isaacson, R.E., Eds.), pp. 321-372. Chapman and Hall, New York.

[109] Whitehead, T.R. (1993) Analyses of the gene and amino acid sequence of the Prevotella (Bacteroides) ruminicola 23 xylanase reveals unexpected homology with endoglucanases from other genera of bacteria. Curr. Microbiol. 27, 27-33.

[110] Gasparic, A., Marinsek-Logar, R., Martin, J., Wallace, R.J., Nek-rep, F.V. and Flint, H.J. (1995) Isolation of genes encoding beta-D-xylanase, beta-D-xylosidase and alpha-L-arabinofuranosidase activities from the rumen bacterium Prevotella ruminicola B1. FEMS Microbiol. Lett. 125 (4), 135-141.

[111] Gasparic, A., Martin, J., Daniel, A.S. and Flint, H.J. (1995) A xylan hydrolase gene cluster in Prevotella ruminicola B14: sequence relationships, synergistic interactions, and oxygen sensitivity of a novel enzyme with exoxylanase and beta-(1,4)-xylosidase activities. Appl. Environ. Microbiol. 61, 2958-2964.

[112] Flint, H.J., Whitehead, T.R., Martin, J.C. and Gasparic, A. (1997) Interrupted catalytic domain structures in xylanases from two distantly related strains of Prevotella ruminicola. Biochim. Biophys. Acta 1337, 161-165.

[113] Miyazaki, K., Miyamoto, H., Mercer, D.K., Hirase, T., Martin, J.C., Kojima, Y. and Flint, H.J. (2003) Involvement of the multidomain regulatory protein XynR in positive control of xylanase gene expression in the ruminal anaerobe Prevotella bryantii B14. J. Bacteriol. 185, 2219-2226.

[114] Lowe, S.E., Theodorou, M.K. and Trinci, A.P. (1987) Cellulases and xylanase of an anaerobic rumen fungus grown on wheat straw, wheat straw holocellulose, cellulose, and xylan. Appl. Environ. Mi-crobiol. 53, 1216-1223.

[115] McSweeney, C.S., Dulieu, A., Katayama, Y. and Lowry, J.B. (1994) Solubilisation of lignin by the ruminal anaerobic fungus Neocallimastix patriciarum. Appl. Environ. Microbiol. 60, 2985-2989.

[116] Dehority, B.A. and Tirabasso, P.A. (2000) Antibiosis between ru-minal bacteria and ruminal fungi. Appl. Environ. Microbiol. 66, 2921-2927.

[117] Joblin, K.N., Matsui, H., Naylor, G.E. and Ushida, K. (2002) Degradation of fresh ryegrass by methanogenic co-cultures of ruminal

fungi grown in the presence or absence of Fibrobacter succinogenes. Curr. Microbiol. 45, 46-53.

[118] Wood, T.M., Wilson, C.A., McCrae, S.I. and Joblin, K.N. (1986) The highly active extracellular cellulase from the anaerobic ruminal fungus Neiocallimastix frontalis. FEMS Microbiol. Lett. 34, 37-40.

[119] Xue, G.P., Orpin, C.G., Gobius, K.S., Aylward, J.H. and Simpson, G.D. (1992) Cloning and expression of multiple cellulase cDNAs from the anaerobic rumen fungus Neocallimastix patriciarum in Escherichia coli. J. Gen. Microbiol. 138, 1413-1420.

[120] Denman, S., Xue, G.P. and Patel, B. (1996) Characterization of a Neocallimastix patriciarum cellulase cDNA (celA) homologous to Trichoderma reesei cellobiohydrolase II. Appl. Environ. Microbiol. 62, 1889-1896.

[121] Zhou, L., Xue, G.P., Orpin, C.G., Black, G.W., Gilbert, H.J. and Hazlewood, G.P. (1994) Intronless celB from the anaerobic fungus Neocallimastix patriciarum encodes a modular family A endogluca-nase. Biochem. J. 297, 359-364.

[122] Hebraud, M. and Fevre, M. (1990) Purification and characterisation of an aspecific glycoside hydrolase from the anaerobic ruminal fungus Neocallimastix frontalis. Appl. Environ. Microbiol. 56, 31643169.

[123] Wilson, C.A., McCrae, S.I. and Wood, T.M. (1994) Characterization of a beta-D-glucosidase from the anaerobic rumen fungus Neo-callimastix frontalis with particular reference to attack on cello-oli-gosaccharides. J. Biotechnol. 37, 217-227.

[124] Li, X.L. and Calza, R.E. (1991) Fractionation of cellulases from the ruminal fungus Neocallimastix frontalis EB188. Appl. Environ. Microbiol. 57, 3331-3336.

[125] Qiu, X., Selinger, B., Yanke, L. and Cheng, K. (2000) Isolation and analysis of two cellulase cDNAs from Orpinomyces joyonii. Gene 245, 119-126.

[126] Chen, H., Li, X.L., Blum, D., Ximenes, E. and Ljungdahl, L. (2003) CelF of Orpinomyces PC-2 has an intron and encodes a cellulase (CelF) containing a carbohydrate-binding module. Appl. Biochem. Biotechnol. 108, 775-786.

[127] Gilbert, H.J., Hazlewood, G.P., Laurie, J.I., Orpin, C.G. and Xue,

G.P. (1992) Homologous catalytic domains in a rumen fungal xy-lanase: evidence for gene duplication and prokaryotic origin. Mol. Microbiol. 6, 2065-2072.

[128] Fanutti, C., Ponyi, T., Black, G.W., Hazlewood, G.P. and Gilbert,

H.J. (1995) The conserved noncatalytic 40-residue sequence in cel-lulases and hemicellulases from anaerobic fungi functions as a protein docking domain. J. Biol. Chem. 270, 29314-29322.

[129] Fry, S.C., Smith, R.C., Renwick, K.F., Martin, D.J., Hodge, S.K. and Matthews, K.J. (1992) Xyloglucan endotransglycosylase, a new wall-loosening enzyme activity from plants. Biochem. J. 282 (Pt 3), 821-828.

[130] York, W.S., Kumar Kolli, V.S., Orlando, R., Albersheim, P. and Darvill, A.G. (1996) The structures of arabinoxyloglucans produced by solanaceous plants. Carbohydr. Res. 285, 99-128.

[131] Teixeira, L.C., Linden, J.C. and Schroeder, H.A. (2000) Simultaneous saccharification and cofermentation of peracetic acid-pre-treated biomass. Appl. Biochem. Biotechnol. 86, 111-127.

[132] Biely, P., Mislovicova, D. and Toman, R. (1985) Soluble chromo-genic substrates for the assay of endo-1,4-beta-xylanases and endo-

I.4-beta-glucanases. Anal. Biochem. 144, 142-146.

[133] Biely, P., Markovic, O. and Mislovicova, D. (1985) Sensitive detection of endo-1,4-beta-glucanases and endo-1,4-beta-xylanases in gels. Anal. Biochem. 144, 147-151.

[134] McDermid, K.P., Forsberg, C.W. and MacKenzie, C.R. (1990) Purification and properties of an acetylxylan esterase from Fibrobacter succinogenes S85. Appl. Environ. Microbiol. 56, 3805-3810.

[135] Cybinski, D.H., Layton, I., Lowry, J.B. and Dalrymple, B P. (1999) An acetylxylan esterase and a xylanase expressed from genes cloned from the ruminal fungus Neocallimastix patriciarum act synergisti-cally to degrade acetylated xylans. Appl. Microbiol. Biotechnol. 52, 221-225.

[136] Dalrymple, B.P., Cybinski, D.H., Layton, I., McSweeney, C.S., Xue, G.P., Swadling, Y.J. and Lowry, J.B. (1997) Three Neocallimastix patriciarum esterases associated with the degradation of complex polysaccharides are members of a new family of hydrolases. Microbiology 143, 2605-2614.

[137] Upton, C. and Buckley, J.T. (1995) A new family of lipolytic enzymes? Trends Biochem. Sci. 20, 178-179.

[138] Borneman, W.S., Ljungdahl, L.G., Hartley, R.D. and Akin, D.E. (1992) Purification and partial characterization of two feruloyl esterases from the anaerobic fungus Neocallimastix strain MC-2. Appl. Environ. Microbiol. 58, 3762-3766.

[139] Borneman, W.S., Ljungdahl, L.G., Hartley, R.D. and Akin, D.E. (1991) Isolation and characterization of p-coumaroyl esterase from the anaerobic fungus Neocallimastix strain MC-2. Appl. Environ. Microbiol. 57, 2337-2344.

[140] Borneman, W.S., Hartley, R.D., Himmelsbach, D.S. and Ljungdahl, L.G. (1990) Assay for trans-p-coumaroyl esterase using a specific substrate from plant cell walls. Anal. Biochem. 190, 129-133.

[141] Fonty, G., Gouet, P.H., Jouany, J.P. and Senaud, J. (1983) Ecological factors determining establishment of cellulolytic bacteria and protozoa in the rumens of meroxenic lambs. J. Gen. Microbiol. 129, 213-223.

[142] Fonty, G., Gouet, P., Ratefiarivelo, H. and Jouany, J.P. (1988) Establishment of Bacteroides succinogenes and measurement of the main digestive parameters in the rumen of gnotobiotic lambs. Can. J. Microbiol. 34, 938-946.

[143] Gregg, K., Hamdorf, B., Henderson, K., Kopecny, J. and Wong, C. (1998) Genetically modified ruminal bacteria protect sheep from fluoroacetate poisoning. Appl. Environ. Microbiol. 64, 3496-3498.

[144] Schwarz, W.H. (2001) The cellulosome and cellulose degradation by anaerobic bacteria. Appl. Microbiol. Biotechnol. 56, 634-649.

[145] Steenbakkers, P.J., Li, X.L., Ximenes, E.A., Arts, J.G., Chen, H., Ljungdahl, L.G. and Op Den Camp, H.J. (2001) Noncatalytic docking domains of cellulosomes of anaerobic fungi. J. Bacteriol. 183, 5325-5333.

[146] Freelove, A.C., Bolam, D.N., White, P., Hazlewood, G.P. and Gilbert, H.J. (2001) A novel carbohydrate-binding protein is a component of the plant cell wall-degrading complex of Piromyces equi. J. Biol. Chem. 276, 43010-43017.

[147] Miron, J., Ben-Ghedalia, D. and Morrison, M. (2001) Invited review: adhesion mechanisms of rumen cellulolytic bacteria. J. Dairy Sci. 84, 1294-1309.

[148] Miron, J., Yokoyama, M.T. and Lamed, R. (1989) Bacterial cell surfacestructures involved in lucerne cell wall degradation by pure cultures of cellulolytic rumen bacteria. Appl. Microbiol. Biotechnol. 32, 218-222.

[149] Mironov, A.S. and Sukhodolets, V.V. (1979) Promoter-like mutants with increased expression of the Escherichia coli uridine phosphor-ylase structural gene. J. Bacteriol. 137, 802-810.

[150] Miron, J. and Ben-Ghedalia, D. (1993) Digestion of cell-wall mono-saccharides of ryegrass and alfalfa hays by the ruminal bacteria Fibrobacter succinogenes and Butyrivibrio fibrisolvens. Can. J. Micro-biol. 39, 780-786.

[151] Miron, J. and Ben-Ghedalia, D. (1993) Digestion of structural poly-saccharides of Panicum and vetch hays by the rumen bacterial strains Fibrobacter succinogens BL2 and Butyrivibrio fibrisolvens D1. Appl. Microbiol. Biotechnol. 39, 756-759.

[152] Huang, L., Forsberg, C.W. and Thomas, D.Y. (1988) Purification and characterizationof a chloride-stimulated cellobiosidase from Bacteroides succinogenes S85. J. Bacteriol. 170, 2923-2932.

[153] McGavin, M. and Forsberg, C.W. (1989) Catalytic and substrate-binding domains of endoglucanase 2 from Bacteroides succinogenes. J. Bacteriol. 171, 3310-3315.

[154] Gong, J., Egosimba, E.E. and Forsberg, C.W. (1996) Cellulose binding proteins of Fibrobacter succinogenes and the possible role of a 180-kDa cellulose binding glycoprotein in adhesion to cellulose. Can. J. Microbiol. 42, 453-460.

[155] Galfi, P., Neogrady, S., Semjen, G., Bardocz, S. and Pusztai, A. (1998) Attachment of different Escherichia coli strains to cultured rumen epithelial cells. Vet. Microbiol. 61, 191-197.

[156] Styriak, I., Galfi, P. and Kmet, V. (1994) The adherence of three Streptococcus bovis strains to cells of rumen epithelium primoculture under various conditions. Arch. Tierernahr. 46, 357-365.

[157] Miron, J. and Forsberg, C.W. (1999) Characterisation of cellulose-binding proteins that are involved in the adhesion mechanism of Fibrobacter intestinalis DR7. Appl. Microbiol. Biotechnol. 51, 491-497.

[158] Gong, J. and Forsberg, C.W. (1989) Factors affecting adhesion of Fibrobacter succinogenes subsp. succinogenes S85 and adherence-defective mutants to cellulose. Appl. Environ. Microbiol. 55, 30393044.

[159] Miron, J. and Ben-Ghedalia, D. (1987) Digestibility by sheep of total and cell wall monosaccharides of wheat straw treated chemically or chemically plus enzymatically. J. Dairy Sci. 70, 1876-1884.

[160] Blair, B.G. and Anderson, K.L. (1998) Comparison of staining techniques for scanning electron microscopic detection of ultrastructural protuberances on cellulolytic bacteria. Biotech. Histochem. 73, 107113.

[161] Latham, M.J., Brooker, B.E., Pettipher, G.L. and Harris, P.J. (1978) Adhesion of Bacteroides succinogenes in pure culture and in the presence of Ruminococcus flavefaciens to cell walls in leaves of perennial ryegrass (Lolium perenne). Appl. Environ. Microbiol. 35, 1166-1173.

[162] Latham, M.J., Brooker, B.E., Pettipher, G.L. and Harris, P.J. (1978) Ruminococcus flavefaciens cell coat and adhesion to cotton cellulose and to cell walls in leaves of perennial ryegrass (Lolium perenne). Appl. Environ. Microbiol. 35, 156-165.

[163] Miron, J., Duncan, S.H. and Stewart, C.S. (1994) Interactions between rumen bacterial strains during the degradation and utilization of the monosaccharides of barley straw cell-walls. J. Appl. Bacteriol. 76, 282-287.

[164] Shi, Y. and Weimer, P.J. (1992) Response surface analysis of the effects of pH and dilution rate on Ruminococcus flavefaciens FD-1 in cellulose-fed continuous culture. Appl. Environ. Microbiol. 58, 2583-2591.

[165] Morrison, M. and Miron, J. (2000) Adhesion to cellulose by Ruminococcus albus : a combination of cellulosomes and Pil-proteins ? FEMS Microbiol. Lett. 185, 109-115.

[166] White, B.A., Cann, I.K.O., Mackie, R.I. and Morrison, M. (1997) Cellulase and xylanase genes from ruminal bacteria: domain analysis suggest a non-cellulosome-like model for organization of the cellulase complex. In: Rumen Microbes and Digestive Physiology in Ruminants (Onodera, R., Itabashi, H., Ushida, K., Yano, H. and Sasaki, Y., Eds.), pp. 69-80. Japan Scientific Societies, Tokyo.

[167] Flint, H.J., Aurilia, V., Kirby, J., Miyazaki, K., Ricon-Torres, M.T., McCrae, S.I. and Martin, J.C. (1999) Organization of plant cell wall degrading enzymes in the ruminal anaerobic bacteia Ruminococccus flavefaciens and Prevotella bryantii. In: Genetics, Biochemistry and Ecology of Cellulose Degradation (Onodera, R., Itabashi, H., Ushi-da, K., Yano, H. and Sasaki, Y., Eds.), pp. 511-519. Uni Publishers, Tokyo.

[168] Kirby, J., Martin, J.C., Daniel, A.S. and Flint, H.J. (1997) Dock-erin-like sequences in cellulases and xylanases from the rumen cel-lulolytic bacterium Ruminococcus flavefaciens. FEMS Microbiol. Lett. 149, 213-219.

[169] Mitsumori, M. and Minato, H. (1995) Distribution of cellulose-binding proteins among the representative strains of rumen bacteria. J. Gen. Appl. Microbiol. 41, 297-306.

[170] Mitsumori, M. and Minato, H. (1997) Cellulose-binding proteins from rumen microorganisms. In: Rumen Microbes and Digestive Physiology in Ruminants (Onodera, R., Itabashi, H., Ushida, K., Yano, H. and Sasaki, Y., Eds.), pp. 47-57. Japan Scientific Societies, Tokyo.

[171] Baintner, K., Duncan, S.H., Stewart, C.S. and Pusztai, A. (1993)

Binding and degradation of lectins by components of rumen liquor. J. Appl. Bacteriol. 74, 29-35.

[172] Rasmussen, M.A., White, B.A. and Hespell, R.B. (1989) Improved assay for quantitating adherence of ruminal bacteria to cellulose. Appl. Environ. Microbiol. 55, 2089-2091.

[173] Sakka, K., Kimura, T., Karita, S. and Ohmiya, K. (1999) [Structure and function of cellulolytic complex 'cellulosome']. Tanpakushitsu Kakusan Koso 44, 1487-1496.

[174] Bayer, E.A., Chanzy, H., Lamed, R. and Shoham, Y. (1998) Cellulose, cellulases and cellulosomes. Curr. Opin. Struct. Biol. 8, 548557.

[175] Wood, T.M., Wilson, C.A. and Stewart, C.S. (1982) Preparation of the cellulase from the cellulolytic anaerobic rumen bacterium Ruminococcus albus and its release from the bacterial cell wall. Biochem. J. 205, 129-137.

[176] Stack, R.J. and Hungate, R.E. (1984) Effect of 3 phenylpropanoic acid on capsule and cellulases of Ruminococcus albus 8. Appl. Environ. Microbiol. 48, 218-223.

[177] Cheng, K.J., Stewart, C.S., Dinsdale, D. and Costerton, J.W. (1983) Electronmicroscopy of bacteria involved in the digestion of plant cell walls. Anim. Feed Sci. Technol. 10, 93-120.

[178] Dinsdale, D., Morris, E.J. and Bacon, J.S.D. (1978) Electron microscopy of the microbial populations present and their modes of attack on various cellulosic substrates undergoing digestion in the sheep rumen. Appl. Environ. Microbiol. 36, 160-168.

[179] Steenbakkers, P.J., Harhangi, H.R., Bosscher, M.W., van der Hooft, M.M., Keltjens, J.T., van der Drift, C., Vogels, G.D. and op den Camp, H.J. (2003) ß-Glucosidase in cellulosome of the anaerobic fungus Piromyces sp. strain E2 is a family 3 glycoside hydro-lase. Biochem. J. 370, 963-970.

[180] Fujino, Y., Ogata, K., Nagamine, T. and Ushida, K. (1998) Cloning, sequencing, and expression of an endoglucanase gene from the rumen anaerobic fungus Neocallimastix frontalis MCH3. Biosci. Biotechnol. Biochem. 62, 1795-1798.

[181] Millward-Sadler, S.J., Davidson, K., Hazlewood, G.P., Black, G.W., Gilbert, H.J. and Clarke, J.H. (1995) Novel cellulose-binding domains, NodB homologues and conserved modular architecture in xylanases from the aerobic soil bacteria Pseudomonas fluo-rescens subsp. cellulosa and Cellvibrio mixtus. Biochem. J. 312, 39-48.

[182] Fillingham, I.J., Kroon, P.A., Williamson, G., Gilbert, H.J. and Hazlewood, G.P. (1999) A modular cinnamoyl ester hydrolase from the anaerobic fungus Piromyces equi acts synergistically with xylanase and is part of a multiprotein cellulose-binding cellulase-hemicellulase complex. Biochem. J. 343, 215-224.

[183] Steenbakkers, P.J., Ubhayasekera, W., Goossen, H.J., van Lierop, E.M., van der Drift, C., Vogels, G.D., Mowbray, S.L. and Op den Camp, H.J. (2002) An intron-containing glycoside hydrolase family 9 cellulase gene encodes the dominant 90 kDa component of the cellulosome of the anaerobic fungus Piromyces sp. strain E2. Biochem. J. 365, 193-204.

[184] Harhangi, H.R., Steenbakkers, P.J., Akhmanova, A., Jetten, M.S., van der Drift, C. and Op den Camp, H.J. (2002) A highly expressed family 1 ß-glucosidase with transglycosylation capacity from the anaerobic fungus Piromyces sp. E2. Biochim. Biophys. Acta 1574, 293-303.

[185] Charnock, S.J., Bolam, D.N., Nurizzo, D., Szabo, L., McKie, V.A., Gilbert, H.J. and Davies, G.J. (2002) Promiscuity in ligand-binding : The three-dimensional structure of a Piromyces carbohydrate-binding module, CBM29-2, in complex with cello- and mannohexaose. Proc. Natl. Acad. Sci. USA 99, 14077-14082.

[186] Metchnikoff, E. (1908) The Nature of Man. Studies in Optimistic Philosophy. William Heinemann, London.

[187] Metchnikoff, E. (1907) The Prolongation of Life. Optimistic studies. William Heinemann, London.

[188] Allison, M.J., Bucklin, J.A. and Dougherty, R.W. (1964) Ruminal changes after overfeeding with wheat and the effect of intraruminal

inoculation on adaptation to a ration containing wheat. J. Anim. Sci. 23, 1164—1171.

[189] Conrad, H.R. and Hibbs, J.W. (1953) A high roughage system for raising dariy calves based on the early development of rumen function. III. Effect of rumen inoculations and ther ratio of hay to grain on digestion and nitrogen retention. J. Dairy Res. 36, 1326— 1334.

[190] Muscato, T.V., Tedeschi, L.O. and Russell, J.B. (2002) The effect of ruminal fluid preprations on the growht and health of newborn, milk-fed dairy calves. J. Dairy Sci. 85, 648—656.

[191] Allison, M.J., Cook, H.M. and Jones, R.J. (1983) Detoxification of 3-hydroxy-4(1H)-pyridone, the goitrogenic metabolite of mimosine, by rumen bacteria from Hawaiian goats. XVII Conference on Rumen Function, Chicago, IL.

[192] Allison, M.J., Mayberry, W.R., Mcsweeney, C.S. and Stahl, D.A. (1992) Synergistes jonesii, gen. nov., sp. nov. : A rumen bacterium that degrades toxic pyridinediols. Syst. Appl. Microbiol. 15, 522— 529.

[193] McSweeny, C.S., Allison, M.J. and Mackie, R.I. (1993) Amino acid utilization by the ruminal bacterium Synergistes jonesii strain 78-1. Arch. Microbiol. 159, 131—135.

[194] McSweeney, C.S. and Mackie, R.I. (1997) Gastrointestinal detoxification and digestive disorders in ruminant animals. In: Gastrointestinal Microbiology (Mackie, R.I. and White, B.A., Eds.), pp. 583— 634. Chapman and Hall, New York.

[195] Gobius, K.S., Xue, G.P., Aylward, J.H., Dalrymple, B.P., Swadling, Y.J., McSweeney, C.S. and Krause, D.O. (2002) Transfromation and expression of an anaerobic fungal xylanase in several strains of the rumen bacterium Butyrivibrio fibrisolvens. J. Appl. Microbiol. 93, 122—133.

[196] Beard, C.E., Hefford, M.A., Forster, R.J., Sontakke, S., Teather, R.M. and Gregg, K. (1995) A stable and efficient transformation system for Butyrivibrio fibrisolvens OB156. Curr. Microbiol. 30, 105— 109.

[197] Clark, R.G., Cheng, K.J., Selinger, L.B. and Hynes, M.F. (1994) A conjugative transfer system for the rumen bacterium, Butyrivibrio fibrisolvens, based on Tn916-mediated transfer of the Staphylococcus aureus plasmid pUB110. Plasmid 32, 295—305.

[198] Whitehead, T.R. (1992) Genetic transformation of the ruminal bacterium Butyrivibrio fibrisolvens and Streptococcus bovis by electro-poration. Lett. Appl. Microbiol. 15, 186—189.

[199] Shoemaker, N.B., Anderson, K.L., Smithson, S.L., Wang, G.-R. and Salyers, A.A. (1991) Conjugal transfer of a shuttle vector from the human colonic anaerobe Bacteroides uniformis to the ruminal anaerobe Prevotella (Bacteroides) ruminicola B[4. Appl. Environ. Microbiol. 57, 2114—2120.

[200] Dehority, B.A. (1965) Degradation and utilization of isolated hemi-cellulose by pure cultures of cellulolytic rumen bacteria. J. Bacteriol. 89, 1515—1520.

[201] Dehority, B.A. and Scott, H.W. (1967) Extent of cellulose and hemi-cellulose digestion in various forages by pure cultures of cellulolytic rumen bacteria. J. Dairy Sci. 50, 1136—1141.

[202] Gobius, K.S., Xue, G.P., Aylward, J.H., Dalrymple, B.P., Swadling, Y.J., McSweeney, C.S. and Krause, D.O. (2002) Transfromation and expression of an anaerobic fungal xylanase in several strains of the rumen bacterium Butyrivibrio fibrisolvens. J. Appl. Microbiol. 93, 122—133.

[203] Krause, D.O., Bunch, R.J., Dalrymple, B.D., Gobius, K.S., Smith, W.J.M., Xue, G.P. and McSweeney, C.S. (2001) Expression of a modified Neocallimastix patriciarum xylanase in Butyrivibrio fibrisolvens digests more fibre but cannot effectively compete with highly fibrolytic bacteria in the rumen. J. Appl. Microbiol. 90, 388—396.

[204] Cotta, M.A., Whitehead, T.R. and Rasmussen, M.A. (1997) Survival of the recombinant Bacteroides thetaiotaomicron strain BTX in in vitro rumen incubations. J. Appl. Microbiol. 82, 743—750.

[205] Robertson, J.M., McKenzie, N.H., Duncan, M., Allen-Vercoe, E., Woodward, M.J., Flint, H.J. and Grant, G. (2003) Lack of flagella

disadvantages Salmonella enterica serovar Enteritidis during the early stages of infection in the rat. J. Med. Microbiol. 52, 91—99.

[206] Baldwin, R.L. and Palmquist, D.L. (1965) Effect of diet on the activity of several enzymes in extracts of rumen microorganisms. Appl. Microbiol. 13, 194—200.

[207] Krause, D.O., Bunch, R.J., Dalrymple, B.D., Gobius, K.S., Smith, W.J.M., Xue, G.P. and McSweeney, C.S. (2001) Expression of a modified Neocallimastix patriciarum xylanase in Butyrivibrio fibrisolvens digests more fibre but cannot effectively compete with highly fibrolytic bacteria in the rumen. J. Appl. Microbiol. 90, 388—396.

[208] Ziemer, C.J., Sharp, R., Stern, M.D., Cotta, M.A., Whitehead, T.R. and Stahl, D.A. (2002) Persistence and functional impact of a micro-bial inoculant on native microbial community structure, nutrient digestion and fermentation characteristics in a rumen model. Syst. Appl. Microbiol. 25, 416—422.

[209] Ziemer, C.J., Sharp, R., Stern, M.D., Cotta, M.A., Whitehead, T.R. and Stahl, D.A. (2000) Comparison of microbial populations in model and natural rumens using 16S ribosomal RNA-targeted probes. Environ. Microbiol. 2, 632—643.

[210] Goad, D.W., Goad, C.L. and Nagaraja, T.G. (1998) Ruminal mi-crobial and fermentative changes associated with experimentally induced subacute acidosis in steers. J. Anim. Sci. 76, 234—241.

[211] Russell, J.B. and Dombrowshi, D.B. (1980) Effect of pH on the efficiency of growth by pure cultures of rumen bacteria in continuous culture. Appl. Environ. Microbiol. 39, 604—610.

[212] Russell, J.B. and Wilson, D.B. (1996) Why are ruminal cellulolytic bacteria unable to digest cellulose at low pH? J. Dairy Sci. 79, 1503— 1509.

[213] Gardner, R.G., Wells, J.E., Russell, J.B. and Wilson, D.B. (1995) The effect of carbohydrates on the expression of the Prevotella ruminicola 1,4-ß-D-endoglucanase. FEMS Microbiol. Lett. 125, 305— 310.

[214] Gardner, R.G., Wells, J.E., Russell, J.B. and Wilson, D.B. (1995) The cellular location of Prevotella ruminicola ß-1,4-D-endoglucanase and its occurrence in other strains of ruminal bacteria. Appl. Environ. Microbiol. 61, 3288—3292.

[215] Fonty, G., Gouet, P. and Nebout, J.M. (1989) Development of the cellulolytic microflora in the rumen of lambs transferred into sterile isolators a few days after birth. Can. J. Microbiol. 35, 416—422.

[216] Fonty, G., Jouany, J.P., Thivend, P., Gouet, P. and Senaud, J. (1983) A descriptive study of rumen digestion in meroxenic lambs according to the nature and complexity of the microflora. Reprod. Nutr. Dev. 23, 857—873.

[217] Mann, S.O. and Stewart, C.S. (1974) Establishment of a limited rumen flora in gnotobiotic lambs fed on a roughage diet. J. Gen. Microbiol. 84, 379—382.

[218] Fonty, G., Gouet, P.H., Jouany, J.P. and Senaud, J. (1983) Ecological factors determining establishment of cellulolytic bacteria and protozoa in the rumens of meroxenic lambs. J. Gen. Microbiol. 129, 213—223.

[219] Krause, D.O., Smith, W., Dalrymple, B.P., Mackie, R.I. and McSweeney, C.S. (1999) 16S rDNA sequencing of Ruminococcus albus and Ruminococcus flavefaciens : design of a signature probe and its application in adult sheep. Microbiology 145, 1797—1807.

[220] Varel, V.H., Yen, J.T. and Kreikemeier, K.K. (1995) Addition of cellulolytic clostridia to the bovine rumen and pig intestinal tract. App. Environ. Microbiol. 61, 1116—1119.

[221] Dehority, B.A. and Tirabasso, P.A. (1998) Effect of ruminal cellu-lolytic bacterial concentrations on in situ digestion of forage cellulose. J. Anim. Sci. 76, 2905—2911.

[222] Krause, D.O., Bunch, R.J., Conlan, L.L., Kennedy, P.M., Smith, W.J., Mackie, R.I. and McSweeney, C.S. (2001) Repeated ruminal dosing of Ruminococcus spp. does not result in persistence, but changes in other microbial populations occur that can be measured with quantitative 16S-rRNA-based probes. Microbiology 147, 1719— 1729.

[223] Wallace, R.J. and Walker, N.D. (1993) Isolation and attempted

introduction of sugar alcohol-utilizing bacteria in the sheep rumen. J. Appl. Bacteriol. 74, 353-359.

[224] Attwood, G.T., Lockington, R.A., Xue, G.P. and Brooker, G.P. (1988) Use of a unique gene sequence as a probe to enumerate a strain of Bacteroides ruminicola introduced into the rumen. Appl. Environ. Microbiol. 54, 534-539.

[225] Flint, H.J., Bisset, J. and Web, J. (1989) Use of antibiotic resistance mutations to track strains of obligately anaerobic bacteria introduced into the rumen of sheep. J. Appl. Bacteriol. 67, 177-183.

[226] Miyagi, T., Kaneichi, K., Aminov, R.I., Kobayashi, Y., Sakka, K., Hoshino, S. and Ohmiya, K. (1995) Enumeration of transconju-gated Ruminococcus albus and its survival in the goat rumen ecosystem. Appl. Environ. Microbiol. 61, 2030-2032.

[227] Caldwell, D.E., Wolfaardt, G.M., Korber, D R. and Lawrence, J.R. (1997) In: Do Bacterial Communities Transcend Darwinism? (Jones, J.W., Ed.), pp. 105-191. Plenum Press, New York.

[228] Russell, J.B. (1985) Fermentation of cellodextrins by cellulolytic and non-cellulolytic rumen bacteria. Appl. Environ. Microbiol. 49, 572576.

[229] Miller, T.L. and Wolin, M.J. (1979) Fermentations by saccharolytic intestinal bacteria. Am. J. Clin. Nutr. 32, 164-172.

[230] Wolin, M.J. and Miller, T.L. (1988) In: Microbe-Microbe Interactions (Hobson, P.N., Ed.), pp. 343-359. Elsevier, New York.

[231] Coleman, G.S. and Hall, F.J. (1984) The uptake and utilization of Entodinium caudatum, bacteria, free amino acids, and glucose by the rumen ciliate Entodinium bursa. J. Appl. Bacteriol. 56, 283-294.

[232] Coleman, G.S. and Sandford, D.C. (1979) The uptake and utilization of bacteria, amino acids, and nucleic acid components by the rumenciliate Eudiplodinium maggi. J. Appl. Bacteriol. 47, 409-419.

[233] Coleman, G.S. (1964) The metabolism of Escherichia coli and other bacteria by Entodinium caudatum. J. Gen. Microbiol. 37, 209-223.

[234] Newbold, C.J., Ushida, K., Morvan, B., Fonty, G. and Jouany, J.P. (1996) The role of ciliate protozoa in the lysis of methanogenic ar-chaea in rumen fluid. Lett. Appl. Microbiol. 23, 421-425.

[235] Sharp, R., Hazlewood, G.P., Gilbert, H.J. and O'Donnell, A.G. (1994) Unmodified and recombinant strains of Lactobacillus plan-tarum are rapidly lost from the rumen by protozoal predation. J. Appl. Bacteriol. 76, 110-117.

[236] Fonty, G., Gouet, P., Ratefiarivelo, H. and Jouany, J.P. (1988) Establishment of Bacteroides succinogenes and measurement of the main digestive parameters in the rumen of gnotobiotic lambs. Can. J. Microbiol. 34, 938-946.

[237] Odenyo, A.A., Mackie, R.I., Stahl, D.A. and White, B.A. (1994) The use of 16S rRNA-targeted oligonucleotide probes to study competition between ruminal fibrolytic bacteria: pure-culture studies with cellulose and alkaline peroxide-treated wheat straw. Appl. Environ. Microbiol. 60, 3697-3703.

[238] Odenyo, A.A., Mackie, R.I., Stahl, D.A. and White, B.A. (1994) The use of 16S rRNA-targeted oligonucleotide probes to study competition between ruminal fibrolytic bacteria: development of probes for Ruminococcus species and evidence for bacteriocin production. Appl. Environ. Microbiol. 60, 3688-3696.

[239] Kalmokoff, M.L. and Teather, R.M. (1997) Isolation and characterization of a bacteriocin (Butyrivibriocin AR10) from the ruminal anaerobe Butyrivibrio fibrisolvens AR10: evidence in support of the widespread occurrence of bacteriocin-like activity among rumi-nal isolates of B. fibrisolvens. Appl. Environ. Micorbiol. 63, 394402.

[240] Dehority, B.A. (1973) Hemicellulose degradation by rumen bacteria. Fed. Proc. 32, 1819-1824.

[241] Shi, Y., Odt, C.L. and Weimer, P.J. (1997) Competition for cellulose among three predominant ruminal cellulolytic bacteria. Appl. Environ. Microbiol. 63, 734-742.

[242] Shi, Y. and Weimer, P.J. (1996) Utilization of individual cellodextrins by three predominant ruminal cellulotyitc bacteria. Appl. Environ. Microbiol. 62, 1084-1088.

[243] Helaszek, C.T. and White, B.A. (1991) Cellobiose uptake and me-

tabolism by Ruminococcus flavefaciens. Appl. Environ. Microbiol. 57, 64-68.

[244] Shi, Y. and Weimer, P.J. (1997) Competition for cellobiose among three predominant ruminal cellulolytic bacteria under substrate-excess and substrate-limited conditions. Appl. Environ. Microbiol. 63, 743-748.

[245] Berger, L.L., Fahey, G.C. Jr., Bourquin, L.D. and Titgemeyer, E.C. (1994) Modification of forage quality after harvest. In: Forage Cell Wall Structure and Digestibility (Fahey, J.G.C., Ed.), pp. 922-966. American Society of Agronomy, Madison, WI.

[246] Laredo, M.A. and Minson, D.J. (1975) The effect of pelleting on the voluntary intake and digestibility of leaf and stem fractions of three grasses. Br. J. Nutr. 33, 159-170.

[247] Hu, Y.J., Smith, D.C., Cheng, K.J. and Foresberg, C.W. (1991) Cloning of a xylanase gene from Fibrobacter succinogenes 135 and its expression in Escherichia coli. Can. J. Microbiol. 37, 554-561.

[248] Cherney, J.H., Cherney, D.J.R., Akin, D.E. and Axtell, J.D. (1991) Potential of brown-rib low-lignin mutants to improve forage quality. Adv. Agron. 46, 157-198.

[249] Halpin, C., Holt, K., Chojecki, J., Oliver, D., Chabbert, B., Monties, B., Edwards, K., Barakate, A. and Foxon, G.A. (1998) Brown-midrib maize (bml) : a mutation affecting the cinnamyl alcohol dehydrogenase gene. Plant J. 14, 545-553.

[250] Vignols, F., Rigau, J., Torres, M.A., Capellades, M. and Puigdome-nech, P. (1995) The brown midrib3 (mb3) mutation in maize occurs in the gene encoding caffeic acid O-methyl transferase. Plant Cell 7, 407-416.

[251] Whetten, R.W., MacKay, J.J. and Sederoff, R.R. (1998) Recent advances in understanding of lignin biosynthesis. Annu. Rev. Plant Physiol. 49, 585-609.

[252] Bernard, L., Chaise, J.P., Delval, E. and Poncet, C. (1998) Validation of the main modeling methods for the estimation of marker mean retention times in the different compartments of the gastrointestinal tract in sheep. J. Anim. Sci. 76, 2485-2495.

[253] Baucher, M., Bernard-Vailhe, M.A., Chabbert, B., Besle, J.M., Opsomer, C., Van Montague, M. and Botterman, J. (1999) Down-regulation of cinnamyl alcohol dehydrogenase in transgenic alfalfa (Medicago sativa L.) and the effect on lignin composition and digestibility. Plant Mol. Biol. 39, 437-447.

[254] Bernard-Vaihe, M.A., Migne, C., Cornu, A., Maillot, M.P., Grenet, E., Besle, J.M., Atanossova, R., Martz, F. and Legrand, M. (1996) Effect of modification of the O-methyl transferase activity on cell wall composition, ultrastructure and degradability of transgenic tobacco. J. Sci. Food Agric. 72, 385-391.

[255] Rae, A.L., Manners, J.M., Jones, R.L., McIntyre, C.L. and Lu, D.-Y. (2001) Antisense suppression of the lignin bisynthetic enzyme, caffeate O-methyltransferase, improves in vitro digestibility of the tropical pasture legume, Sylosanthes humilis. Aust. J. Plant Physiol. 28, 289-297.

[256] Grabber, J.H., Hatfield, R.D. and Quideau, S. (1997) p-Hydroxy-phenyl, guaiacyl and syringly lignins have similar inhibitory effects on cell wall degradability. J. Agric. Food Chem. 45, 2530-2532.

[257] Sewalt, V.J.H., Beauchemin, K.A., Rode, L.M., Acharya, S. and Baron, V.S. (1997) Lignin impact on fiber degradation. IV. Enzymatic saccharification and in vitro digestibility of alfalfa and grasses following selective solvent delignification. Biores. Technol. 61, 199206.

[258] Guo, D., Chen, F., Wheeler, J., Winder, J., Selman, S., Peterson, M. and Dixon, R.A. (2001) Improvement of in-rumen digestibility of alfalfa forage by genetic manipulation of lignin O-methyltransfer-ases. Transgenic Res. 10, 457-464.

[259] Jung, H.J., Ni, W., Chapple, C.C.S. and Meyer, K. (1999) Impact of lignin composition on cell wall degradability in an Arabidopsis mutant. J. Sci. Food Agric. 79, 922-928.

[260] Jones, L., Ennos, A.R. and Turner, S.R. (2001) Cloning and characterisation in irregular xylem 4 (irs4) a severrely lignin deficient mutant of Arabidopsis. Plant J. 26, 205-216.

[261] Ralph, J., Hatfield, R.D., Piquemal, J., Yahiaoui, N., Pean, M., Lapierre, C. and Boudet, A.M. (1998) NMR characterisation of altered lignins extracted from tobacco plant down-regulated for lignification enzymes cinnamyl-alcohol dehydrogenase and cinnamoyl-CoA reductase. Proc. Natl. Acad. Sci. USA 95, 12803-12808.

[262] Chabannes, M., Ruel, K., Yoshinaga, A., Chabbert, B., Jauneau, A., Joseleau, J.P. and Boudet, A.M. (2001) In situ analysis of lignins in transgenic tobacco reveals a different impact of individual transformation on the spatial patterns of lignin deposition at the cellular and subcellular levels. Plant J. 28, 271-282.

[263] Arioli, T., Peng, L., Betzner, A.S., Burn, J., Wittke, W., Herth, W., Camilleri, C., Hofte, H., Palzinski, J., Birch, R., Cork, A., Glover, J., Redmond, J. and Williamson, R.E. (1998) Molecular analysis of cellulose biosynthesis in Arabidopsis. Science 279, 717-720.

[264] Turner, S.R., Taylor, N. and Jones, L. (2001) Mutations of the secondary cell wall. Plant Mol. Biol. 47, 209-219.

[265] Hu, W.J., Harding, S.A., Lung, J., Popko, J., Ralph, J., Stokke, D.D., Tsai, C.J. and Chiang, V.L. (1999) Repression of lignin biosynthesis promotes cellulose accumulation and growth in transgenic trees. Nat. Biotechnol. 17, 808-812.

[266] Sarria, R., Wagner, T.A., O'Neill, M.A., Faik, A., Wilkerson, C.G., Keegstra, K. and Raikhei, N.V. (2001) Characterisation of a faimily of Arabidopsis genes related to xyloglucan fucosyltransferase 1. Plant Physiol. 127, 1595-1606.

[267] Akin, D.E., Rigsby, L.L., Sethuraman, A., Morrison III, W.H., Gamble, G.R. and Eriksson, K.E. (1995) Alterations in structure, chemistry, and biodegradability of grass lignocellulose treated with the white rot fungi Ceriporiopsis subvermispora and Cyathus sterco-reus. Appl. Environ. Microbiol. 61, 1591-1598.

[268] Akin, D.E., Rigsby, L.L., Sethuraman, A., Morrison III, W.H., Gamble, G.R. and Eriksson, K.-E.L. (1995) Alterations in structure, chemistry, and biodegradation of grass lignocellulose treated with the white rot fungi Ceriporiopsis subvermispora and Cyathus sterco-reus. Appl. Environ. Microbiol. 61, 1591-1598.

[269] Mayer, A.M. and Staples, R.C. (2002) Laccase: new functions for an old enzyme. Phytochemistry 60, 551-565.

[270] Zadrazil, F. (1980) Conversion of different plant waste into feed by basidiomycetes. Eur. J. Appl. Microbiol. 9, 243-248.

[271] Agosin, E. and Odier, E. (1985) Solid-state fermentation, lignin degradation and resulting digestibility of wheat straw fermented by selected white-rot fungi. Appl. Microbiol. Biotechnol. 21, 397403.

[272] Agosin, E., Monties, B. and Odier, E. (1985) Structural changes in wheat straw components during decay by lignin-degrading white-rot fungi in relation to improvement of digestibility for ruminants. J. Sci. Food Agric. 36, 925-935.

[273] Gamble, G.R., Sethuraman, A., Akin, D.E. and Eriksson, K.E. (1994) Biodegradation of lignocellulose in Bermuda grass by white rot fungi analyzed by solid-state 13C nuclear magnetic resonance. Appl. Environ. Microbiol. 60, 3138-3144.

[274] ten Have, R. and Teunissen, P.J. (2001) Oxidative mechanisms involved in lignin degradation by white-rot fungi. Chem. Rev. 101, 3397-3413.

[275] Pointing, S.B. (2001) Feasibility of bioremediation by white-rot fungi. Appl. Microbiol. Biotechnol. 57, 20-33.

[276] Aust, S.D. (1995) Mechanisms of degradation by white rot fungi. Environ. Health Perspect. 103 (Suppl. 5), 59-61.

[277] Zadrazil, F., Puniya, A.K. and Singh, K. (1995) Pilot scale reactor for biological treatment of lignocellulosics for animal feed production. Indian J. Dairy Sci. 48, 110-117.

[278] Hüttermann, A., Hamza, A.S., Chet, I., Majcherczyk, A., Fouad, T., Badr, A., Cohen, R., Persky, L. and Hadar, Y. (2000) Recycling of agricultural wastes by white-rot fungi for the production of fodder for ruminants. Agro-Food Ind. Hi-Tech 6, 29-32.

[279] Tripathi, P. and Yadev, J.S. (1991) Comparative lignolytic and poly-saccharolytic potentials of and alkaliphilic basidiomycete on native lignocellulose. Int. Biodeter. 27, 49-59.

[280] Tripathi, J.P. and Yadav, J.S. (1989) Selection of pre-treatment for an alkaliphilic Coprinus fermentation of wheat straw in a two-stage process. Int. J. Anim. Sci. 4, 128-133.

[281] Rai, S.N., Walli, T.K. and Gupta, B.N. (1989) The chemical composition and nutritive value of rice straw after treatment with urea or Coprinus fimetarius in a solid state fermentation system. Anim. Feed Sci. Technol. 26, 81-92.

[282] Somasundaram, R., Shashreka, M.N., Bano, Z. and Rajarathnam, S. (1992) Biopotentialities of the basidiomacromycetes. Adv. Appl. Microbiol. 37, 233-361.

[283] Yang, X., Chen, H., Gao, H. and Li, Z. (2001) Bioconversion of corn straw by coupling ensiling and solid-state fermentation. Bio-resour. Technol. 78, 277-280.

[284] Burroughs, W., Woods, W., Ewing, S.A., Greig, T. and Theurer, B. (1960) Enzyme additions to fattening cattle rations. J. Anim. Sci. 19, 458-464.

[285] Nelson, F. and Damon, V. (1960) Comparison of different supplemental enzymes with and without diethylstilbestrol for fattening steers. J. Anim. Sci. 19, 1279-1289.

[286] Rovics, J.J. and Ely, C.M. (1962) Response of beef cattle to enzyme supplement. J. Anim. Sci. 21, 1012-1022.

[287] Letherwood, J.M., Mochrie, R.D. and Thomas, W.E. (1960) Some effects of a supplementary cellulase preparation of feed utilization by ruminants. J. Dairy Sci. 43, 1460-1464.

[288] Clark, J.D., Dyer, I.A. and Templeton, J.A. (1961) Some nutritional and physiological effects of enzymes for fattening cattle. J. Anim. Sci. 20, 928-938.

[289] Perry, T.W., Cope, D.D. and Beeson, W.M. (1960) Low vs high moisture shelled corn with and without enzymes and stilbestrol for fattening steers. J. Anim. Sci. 19, 1284-1294.

[290] Perry, T.W., Purkhiser, E.D. and Beeson, W.M. (1966) Effects of supplemental enzymes on nitrogen balance, digestibility, or energy and nutrients, and on growth and feed efficiency of cattle. J. Anim. Sci. 25, 760-764.

[291] Beauchemin, K.A., Rode, L.M. and Sewalt, V.J.H. (1995) Fibrolytic enzymes increase fiber digestibility and growth rate of steers fed dry forages. Can. J. Anim. Sci. 75, 641-644.

[292] Beauchemin, K.A. and Rode, L.M. (1996) Use of feed enzymes in ruminant nutrition. In: Animal Science Research and Development - Meeting Future Challenges (Rode, L.M., Ed.), pp. 103-131. Minister of Supply and Services Canada, Ottawa, ON.

[293] Feng, P., Hunt, C.W., Pritchard, G.T. and Julien, W.E. (1996) Effect of enzyme preparations on in situ and in vitro degradation and in vivo digestive characteristics of mature cool-season grass forage in beef steers. J. Anim. Sci. 74, 1349-1357.

[294] Lewis, G.E., Sanchez, W.K., Hunt, C.W., Guy, M.A., Pritchard, G.T., Swanson, B.I. and Treacher, R.J. (1999) Effect of direct-fed fibrolytic enzymes on the lactational performance of dairy cows. J. Dairy Sci. 82, 611-617.

[295] McAllister, T.A., Oosting, S.J., Popp, J.D., Mir, Z., Yanke, L.J., Hristov, A.N., Treacher, R.J. and Cheng, K.J. (1999) Effect of exogenous enzymes on digestibility of barley silage and growth performance of feedlot cattle. Can. J. Anim. Sci. 79, 353-360.

[296] Krause, M., Beauchemin, K.A., Rode, L.M., Farr, B.I. and Nor-gaard, P. (1998) Fibrolytic enzyme treatment of barley grain and source of forage in high-grain diets fed to growing cattle. J. Anim. Sci. 76, 2912-2920.

[297] Hristov, A.N., McAllister, T.A. and Cheng, K.J. (1998) Stability of exogenous polysaccharide-degrading enzymes in the rumen. Anim. Feed Sci. Technol. 76, 161-168.

[298] McAllister, T.A., Hristov, A.N., Beauchemin, K.A., Rode, L.M., Cheng, K.J., Bedford, M.R., and Partridge, G.G. (2000) Enzymes in ruminant diets. In: Enzymes in Farm Animal Nutrition (Bedford, M. and Partridge, G., Eds.), pp. 273-298. CAB International, Wallingford.

[299] Hristov, A.N., Rode, L.M., Beauchemin, K.A. and Wuerfel, R.L. (1996) Effect of a commercial enzyme preparation on barley silage in

vitro and in sacco dry matter degradability. Proc. West. Sec. Am. Soc. Anim. Sci. Rapid City, South Dakota 47, 282-284.

[300] Nsereko, V.L., Morgavi, D.P., Rode, L.M., Beauchemin, K.A. and McAllister, T.A. (2000) Effects of fungal enzyme preparations on hydrolysis and subsequent degradation of alfalfa hay fiber by mixed rumen microorganisms in vitro. Anim. Feed Sci. Technol. 88, 153170.

[301] Hristov, A.N., McAllister, T.A. and Cheng, K.J. (2000) Intrarumi-nal supplementation with increasing levels of exogenous polysaccha-ride-degrading enzymes : effects on nutrient digestion in cattle fed a barley grain diet. J. Anim. Sci. 78, 477-487.

[302] McAllister, T.A., Stanford, K., Bae, H.D., Treacher, R.J., Hristov, A.N., Baah, J., Shelford, J.A. and Cheng, K.J. (2000) Effect of a surfactant and exogenous enzymes on digestibility of feed and on growth performance and carcass traits of lambs. Can. J. Anim. Sci. 80, 35-44.

[303] Wang, Y., McAllister, T.A., Rode, L.M., Beauchemin, K.A., Morgavi, D P., Nsereko, V.L., Iwaasa, A.D. and Yang, W. (2001) Effects of an exogenous enzyme preparation on microbial protein synthesis, enzyme activity and attachment to feed in the Rumen Simulation Technique (Rusitec). Br. J. Nutr. 85, 325-332.

[304] Dong, Y., Bae, H.D., McAllister, T.A., Mathison, G.W. and Cheng, K.J. (1999) Effects of exogenous fibrolytic enzymes, alpha-bromo-ethanesulfonate and monensin on fermentation in a rumen simulation (RUSITEC) system. Can. J. Anim. Sci. 79, 491-498.

[305] Wing, J.M., Van Horn, H.H., Sklare, S.D. and Harris Jr., B. (1988) Effects of citus molasses, distillers solubles, and molasses on rumen parameters and lactation. J. Dairy Sci. 71, 414-420.

[306] Wallace, R.J., Wallace, S.J., McKain, N., Nsereko, V.L. and Hart-nell, G.F. (2001) Influence of supplementary fibrolytic enzymes on the fermentation of corn and grass silages by mixed ruminal microorganisms in vitro. J. Anim. Sci. 79, 1905-1916.

[307] Morgavi, D P., Newbold, C.J., Beever, D.E. and Wallace, R.J.

(2000) Stability and stabilization of potential feed additive enzymes in rumen fluid. Enzyme Microb. Technol. 26, 171-177.

[308] Graham, H., Balnave, D., Wallace, R.J. and Chesson, A. (1995) Dietary enzymes for increasing energy availability. In: Biotechnology in Animal Feeds and Animal Feeding (Wallace, R.J. and Cehs-son, A., Eds.), pp. 295-309. VCH Press, Weinheim.

[309] Lewis, G.E., Hunt, C.W., Sanchez, W.K., Treacher, R., Pritchard, G.T. and Feng, P. (1996) Effect of direct-fed fibrolytic enzymes on the digestive characteristics os a forage-based diet fed to beef steers. J. Anim. Sci. 74, 3020-3028.

[310] Hofrichter, M., Lundell, T. and Hatakka, A. (2001) Conversion of milled pine wood by manganese peroxidase from Phlebia radiata. Appl. Environ. Microbiol. 67, 4588-4593.

[311] Gomez-Toribio, V., Martinez, A.T., Martinez, M.J. and Guillen, F.

(2001) Oxidation of hydroquinones by the versatile ligninolytic per-oxidase from Pleurotus eryngii. H2O2 generation and the influence of Mn2+. Eur. J. Biochem. 268, 4787-4793.

[312] Glenn, J.K. and Gold, M.H. (1985) Purification and characterization of an extracellular Mn(II)-dependent peroxidase from the lig-nin-degrading basidiomycete, Phanerochaete chrysosporium. Arch. Biochem. Biophys. 242, 329-341.

[313] Tekere, M., Zvauya, R. and Read, J.S. (2001) Ligninolytic enzyme production in selected sub-tropical white rot fungi under different culture conditions. J. Basic Microbiol. 41, 115-129.

[314] Eichlerova, I., Ruel, K., Homolka, L., Joseleau, J.P. and Nerud, F. (2000) Ligninolytic characteristics of Pleurotus ostreatus strain F6 and its monokaryotic protoplast derivative P19. Can. J. Microbiol. 46, 1153-1158.

[315] Hirano, T., Honda, Y., Watanabe, T. and Kuwahara, M. (2000) Degradation of bisphenol A by the lignin-degrading enzyme, manganese peroxidase, produced by the white-rot basidiomycete, Pleuro-tus ostreatus. Biosci. Biotechnol. Biochem. 64, 1958-1962.

[316] Eggert, C., Temp, U. and Eriksson, K.E. (1996) The ligninolytic system of the white rot fungus Pycnoporus cinnabarinus : purification

and characterization of the laccase. Appl. Environ. Microbiol. 62, 1151-1158.

[317] Tien, M. and Myer, S.B. (1990) Selection and characterisation of mutants of Phanerochaete chrysosporium exhibiting ligninolytic activity under nitrogen-rich conditions. Appl. Environ. Microbiol. 56, 2540-2544.

[318] Antorini, M., Herpoel-Gimbert, I., Choinowski, T., Sigoillot, J.C., Asther, M., Winterhalter, K. and Piontek, K. (2002) Purification, crystallisation and X-ray diffraction study of fully functional lac-cases from two ligninolytic fungi. Biochim. Biophys. Acta 1594, 109-114.

[319] Baldrian, P. and Gabriel, J. (2002) Copper and cadmium increase laccase activity in Pleurotus ostreatus. FEMS Microbiol. Lett. 206, 69-74.

[320] Kersten, P.J., Kalanaraman, B., Hammel, H. and Reinhammar, B. (1990) Comparison of lignin peroxidase, horseradish peroxidase and laccase in the oxidation of methoxybenzenes. Biochem. J. 268, 475480.

[321] Johannes, C. and Majcherczyk, A. (2000) Natural mediators in the oxidation of polycyclic aromatic hydrocarbons by laccase mediator systems. Appl. Environ. Microbiol. 66, 524-528.

[322] Eggert, C., Temp, U., Dean, J.F. and Eriksson, K.E. (1996) A fungal metabolite mediates degradation of non-phenolic lignin structures and synthetic lignin by laccase. FEBS Lett. 391, 144-148.

[323] Eggert, C., Temp, U. and Eriksson, K.E. (1997) Laccase is essential for lignin degradation by the white-rot fungus Pycnoporus cinnaba-rinus. FEBS Lett. 407, 89-92.

[324] Eggert, C., LaFayette, P.R., Temp, U., Eriksson, K.E. and Dean, J.F. (1998) Molecular analysis of a laccase gene from the white rot fungus Pycnoporus cinnabarinus. Appl. Environ. Microbiol. 64, 1766-1772.

[325] Temp, U., Zierold, U. and Eggert, C. (1999) Cloning and characterization of a second laccase gene from the lignin-degrading basidio-mycete Pycnoporus cinnabarinus. Gene 236, 169-177.

[326] Bourbonnais, R., Paice, M.G., Reid, I.D., Lanthier, P. and Yaguchi, M. (1995) Lignin oxidation by laccase isozymes from Trametes versicolor and role of the mediator 2,2'-azinobis(3-ethylbenzthiazoline-6-sulfonate) in kraft lignin depolymerization. Appl. Environ. Micro-biol. 61, 1876-1880.

[327] Min, K.L., Kim, Y.H., Kim, Y.W., Jung, H.S. and Hah, Y.C. (2001) Characterization of a novel laccase produced by the wood-rotting fungus Phellinus ribis. Arch. Biochem. Biophys. 392, 279286.

[328] Ohga, S. and Royse, D.J. (2001) Transcriptional regulation of lac-case and cellulase genes during growth and fruiting of Lentinula edodes on supplemented sawdust. FEMS Microbiol. Lett. 201, 111-115.

[329] Call, H.P. and Mücke, I. (1997) History, overview and applications of mediated lignolytic systems, especially laccase-mediator-systems (Lignozym0-process). J. Biotechnol. 53, 163-202.

[330] Nelson, K., Aminov, R., Forsberg, C., Mackie, R.I., Russell, J.B., White, B.A., Wilson, D.B., Mulligan, S., Tran, K., Carty, H., Khouri, H., Nelson, W., Daugherty, S., Fraser, C. and Morrison, M. (2002) The Fibrobacter succinogenes strain S85 genome sequencing project. Beyond antimicrobials - the future of gut microbiology. RRI-INRA, O-19.

[331] Morrison, M., Devillard, E. and Goodheart, D. (2002) The effects of phenyl-substituted fatty acids and carbon source on the cellulose-binding sub-proteome of Ruminococcus albus strain 8. 102nd Gen. Meet. Am. Soc. Microbiol., Salt Lake City, UT, J-20.

[332] Devillard, E., Goodheart, D. and Morrison, M. (2002) Proteomics-based analysis of Ruminococcus albus 8 adhesion-defective mutants. Beyond antimicrobials - the future of gut microbiology. RRI-INRA, O-18.

[333] Reveneau, C. and Morrison, M. (2002) The effects from phenylace-tic and phenylpropionic acids on xylan degradation, and growth of Ruminococcus albus. 102nd Gen. Meet. Am. Soc. Microbiol., Salt Lake City, UT, K-75.

[334] Antonopoulos, D.A. and White, B.A. (2002) Identification of functional gene subsets from the cellulolytic organism Ruminococcus flavefaciens using a comparative genomics approach. Beyond antimicrobials - the future of gut microbiology. RRI-INRA, O-20.

[335] Whitford, M.F., Teather, R.M. and Forster, R.J. (2001) Phyloge-netic analysis of methanogens from the bovine rumen. BMC Micro-biol. 1, 5.

[336] Forster, R.J., Whitford, M.F., Teather, R.M. and Krause, D.O. (1998) Investigations into rumen microbial diversity using molecular cloning and probing techniques. In: Genetics, Biochemistry, and Ecology of Cellulose Degradation (Onodera, R., Itabashi, H., Ushi-da, K., Yano, H. and Sasaki, Y., Eds.), pp. 16-24. Sukuka, Japan.

[337] Tajima, K., Aminov, R.I., Nagamine, T., Ogata, K., Nakamura, M., Matsui, H. and Benno, Y. (1999) Rumen bacterial diversity as determined by sequence analysis of 16S rDNA libraries. FEMS Microbiol. Ecol. 29, 159-169.

[338] Krause, D.O. and Russell, J.B. (1996) How many ruminal bacteria are there? J. Dairy Sci. 79, 1467-1475.

[339] Torsvik, V., Goksoyr, J. and Daae, F.L. (1990) High diversity in DNA of soil bacteria. Appl. Environ. Microbiol. 56, 782-787.

[340] Amann, R.I., Ludwig, W. and Schleifer, K.-H. (1995) Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59, 143-169.

[341] Hugenholtz, P., Pitulle, C., Hershberger, K.L. and Pace, N.R. (1998) Novel division level bacterial diversity in a Yellowstone hot spring. J. Bacteriol. 180, 366-376.

[342] Rondon, M.R., August, P.R., Bettermann, A.D., Brady, S.F., Grossman, T.H., Liles, M.R., Loiacono, K.A., Lynch, B.A., MacNeil, I.A., Minor, C., Lai Tiong, C., Gilman, M., Osburne, M.S., Clardy, J., Handelsman, J. and Goodman, R.M. (2000) Cloning the soil metagenome: a strategy for accessing the genetic and funtional diversity of uncultured microorganisms. Appl. Environ. Microbiol. 66, 2541-2547.

[343] Pace, N.R. (1997) A molecular view of microbial diversity and the biosphere. Science 276, 734-740.

[344] Handelsman, J., Rondon, M.R., Brady, S.F., Clardy, J. and Goodman, R.M. (1998) Molecular biological access to the chemistry of unknown soil microbes: a new frontier for natural products. Chem. Biol. 5, R245-R249.

[345] Beja, O., Suzuki, M.T., Koonin, E.V., Aravind, L., Hadd, A., Nguyen, L.P., Villacorta, R., Amjadi, M., Garrigues, C., Jovanovich, S.B., Feldman, R.A. and DeLong, E.F. (2000) Construction and analysis of bacterial artificial chromosome libraries from a marine microbial assemblage. Environ. Microbiol. 2, 516-529.

[346] Gupta, R., Beg, Q.K. and Lorenz, P. (2002) Bacterial alkaline proteases : molecular approaches and industrial applications. Appl. Mi-crobiol. Biotechnol. 59, 15-32.

[347] Beja, O., Suzuki, M.T., Heidelberg, J.F., Nelson, W.C., Preston, C.M., Hamada, T., Eisen, J.A., Fraser, C.M. and DeLong, E.F. (2002) Unsuspected diversity among marine aerobic anoxygenic phototrophs. Nature 415, 630-633.

[348] Giovannoni, S.J., Britschgi, T.B., Moyer, C.L., Field, KG. and DeLong, E. (1990) Genetic diversity in Sargasso Sea bacterioplank-ton: Archael means and extremes. Nature 345, 60-63.

[349] Cottrell, M.T., Moore, J.A. and Kirchman, D.L. (1999) Chitinases from uncultured marine microorganisms. Appl. Environ. Microbiol. 65, 2553-2557.

[350] Henne, A., Daniel, R., Schmitz, R.A. and Gottschalk, G. (1999) Construction of environmental DNA libraries in Escherichia coli and screening for the presence of genes conferring utilization of 4-hydroxybutyrate. Appl. Environ. Microbiol. 65, 3901-3907.

[351] Henne, A., Schmitz, R.A., Bomeke, M., Gottschalk, G. and Daniel, R. (2000) Screening of environmental DNA libraries for the presence of genes conferring lipolytic activity on Escherichia coli. Appl. Environ. Microbiol. 66, 3113-3116.

[352] Campbell, M.M. and Sederoff, R.R. (1996) Variation in lignin content and composition (mechanisms of control and implications for the genetic improvement of plants). Plant Physiol. 110, 3-13.