Review Adhesion

J. Dairy Sci.
84:1294–1309
 American Dairy Science Association, 2001.
Invited Review: Adhesion Mechanisms of Rumen Cellulolytic Bacteria

J. Miron,* D. Ben-Ghedalia,* and M. Morrison†
*Metabolic Unit, Agricultural Research Organization, The Volcani Center,
P.O.Box 6, Bet-Dagan, 50250 Israel and
†Dept of Animal Science, The Ohio State University, Columbus, OH, 43210
ABSTRACT sesses at least two mechanisms for specific adhesion to

cellulose: a cellulosomal-like mechanism, and a CbpC
We divided the adhesion process of the predominant
(Pil)-protein mechanism that probably involves the pro-
cellulolytic rumen bacteria Fibrobacter succinogenes,
duction of fimbrial-like structures. Indirect and direct
Ruminococcus flavefaciens, and Ruminococcus albus
studies suggested that carbohydrate epitopes of CBPs
into four phases: 1) transport of the nonmotile bacteria
and CBD epitope of cellulases may also be involved
to the substrate; 2) initial nonspecific adhesion of bacte-
mostly in the nonspecific phase of adhesion of R. albus.
ria to unprotected sites of the substrate that is domi-
(Key words: Fibrobacter succinogenes, Ruminococcus
nated by constitutive elements of bacterial glycocalyx;
albus, Ruminococcus flavefaciens, specific and nonspe-
3) specific adhesion via adhesins or ligands formation
cific adhesion)
with the substrate, which can be dominated by several
bacterial organelles including cellulosome complexes, Abbreviation key: CBD = cellulose-binding domain,
fimbriae connections, glycosylated epitopes of cellulose- CBP = cellulose-binding proteins, CMC = carboxymeth-
binding protein (CBP) or glycocalyx, and cellulose-bind- ylcellulose, EG = endoglucanase, MC = methylcellulose,
ing domain (CBD) of enzymes; 4) proliferation of the PAA = phenylacetic acid, PPA = phenylpropanoic acid.
attached bacteria on potentially digestible tissues of
the substrate. Each of the phases and its significance
in the adhesion process are described. Factors affecting INTRODUCTION
bacterial adhesion are described including: 1) factors The rumen microbial ecosystem comprises at least
related to bacterial age, glycocalyx condition, and mi-
30 predominant bacterial species at a total concentra-
crobial competition; 2) factors related to the nature of
tion of 1010 to 1011/ml of rumen fluid, some 40 species
substrate including, cuticle protection, surface area, hy-
of protozoa (105 to 107/ml), and five species of fungi (<
dration, and ionic charge; and 3) environmental factors
105/ml), (Orpin and Joblin, 1997; Stewart et al., 1997;
including pH, temperature, and presence of cations and
Williams and Coleman, 1997). Bacterial species of the
soluble carbohydrate. Based on the information avail-
rumen are considered more important than protozoa
able from the literature, it appears that each of the
and fungi in determining the extent and rate of feed
predominant rumen bacteria—F. succinogenes, R. fla-
degradation and utilization for the production of micro-
vefaciens, and R. albus—has a specific mechanism of
bial protein and VFA (Stewart et al., 1997). Subse-
adhesion to cellulose. In F. succinogenes, both the glyco-
quently, the host ruminant animal absorbs VFA (mostly
sidic residues of the outer membrane CBP and espe-
cially of the 180-kDa CBP, and the distinct CBD of EG2 through the rumen wall) and digests proteins, lipids,
EGF and Cl-stimulated cellobiosidase, may play a role and carbohydrate constituents of microbes and feed res-
in the adhesion to cellulose. No direct evidence, except idues entering the small intestine to supply its mainte-
scanning electron microscopy observations, yet sup- nance needs and for the production of meat and milk.
ports the existence of either cellulosome complex or Bacteria inhabiting the rumen have been classified
fimbriae structures involved in the adhesion mecha- into five groups dependent on their environmental exis-
nism of F. succinogenes. At least two mechanisms, cellu- tence: 1) free-living bacteria associated with rumen liq-
losome-like complexes and carbohydrate epitopes of the uid phase; 2) bacteria loosely associated with feed parti-
glycocalyx layer are involved in the specific adhesion cles; 3) bacteria firmly adhered to feed particles; 4) bac-
of R. flavefaciens to cellulose. Ruminococcus albus pos- teria associated with rumen epithelium; and 5) bacteria
attached to the surface of protozoa or fungal sporangia
(Cheng and Costerton, 1980; McAllister et al., 1994).
Under ordinary feeding conditions, bacterial popula-
Received October 2, 2000.
Accepted December 12, 2000. tions associated with feed particles (groups 2 and 3)
Corresponding author: J. Miron; e-mail: jmiron@actcom.co.il. are numerically predominant and occupy up to 75%
1294
INVITED REVIEW: ADHESION OF RUMEN CELLULOLYTIC BACTERIA 1295
of the total microbial population and microbial ATP quent fiber degradation; factors affecting bacterial ad-
production in the rumen (Craig et al., 1987; Forsberg hesion to fiber; and mechanism of the adhesion process
and Lam, 1977; Minato et al., 1993). Microbial popula- in each of the rumen cellulolytic species.
tions associated with feed particles are estimated to be
responsible for 88 to 91% of ruminal endoglucanase and PHASES OF BACTERIAL ADHESION
xylanase activity, 70% of the amylase activity, and 75% TO PLANT TISSUES
of the protease activity in the rumen (Minato et al.,
1993; Williams and Strachan, 1984). These percentages When the adhesion process is described, a distinction
suggest that microbial populations associated with feed between specific and nonspecific adhesion should be
particles are pivotal for feed digestion in the rumen. emphasized (Pell and Schofield, 1993). Thus, the adhe-
Accordingly, microbes associated with the liquid phase sion process occurring in the rumen can better be de-
(20 to 30% of the total microbes), including free-living scribed by dividing the process into four phases: 1)
bacteria and bacteria detached from solid substrate, transport of the bacteria to the fibrous substrate; 2)
have little direct involvement in the digestion of insolu- initial nonspecific adhesion of bacteria to proper sites of
ble feed particles. The bacterial populations associated the substrate; 3) specific adhesion between the attached
with rumen epithelium and those attached to the sur- bacteria and the digestible tissue via the production of
face of protozoa and fungi (~1% of total rumen popula- more extensive linkages and adhesins (Figure 1A, B,
tion) have a minor role in the process of feed digestion C, and F); and 4) proliferation of attached bacteria to
in the rumen. form colonies on specific sites of the plant tissue (Figure
The importance of adhesion on subsequent cell wall 1C, D, and E). Each of these phases depends on the
degradation was demonstrated in studies employing successful completion of the previous one. The multi-
cellobiose or glucose grown mutant cells lacking the step model summarized here is similar to the general
adhesion ability of Fibrobacter succinogenes S85, Fibro- mode for microbial adhesion to solid substrate that ulti-
bacter intestinalis DR7, Ruminococcus albus SY3 and mately results in the formation of biofilm (Busscher
8, and Ruminococcus flavefaciens 007. The mutant cells and Weerkamp, 1987; Costerton et al., 1981; McAllister
were characterized by smoother appearance of surface et al., 1994; Pell and Schofield, 1993).
topology compared to the wild type cells, lacked adhes-
ins or ligands formation with the substrate, and lost Phase I, Transport of Bacteria to the Substrate
most of their cellulolytic capability (Gong and Forsberg,
1989; Miron et al., 1998; Miron and Forsberg, 1998 and Transport of rumen bacteria to the particulate sub-
1999; Reddy and Morrison, 1998; Stewart et al., 1990). strate is problematic, because the predominant rumen
Additional support to the relationship between adhe- cellulolytic bacteria lack any flagella or cilia and are
sion ability and subsequent cellulose degradation was therefore nonmotile; also, mixing in the rumen is poor
given by Morris and Cole (1987), who demonstrated (Weimer, 1996). Thus, the first contact between cellulo-
that isolates of R. albus strains that adhered to cellu- lytic bacteria and substrate is dependent on the size of
lose-degraded cellulose better than strains lacking ad- the free-suspended cellulolytic population, to bind new
hesion ability. The necessity of adhesion for subsequent particles suspended in the rumen. Several studies
cellulose digestion by rumen bacteria was further dem- (Gong and Forsberg, 1989; Miron et al., 1989, 1990,
onstrated by the observations that a low concentration 1998; Miron and Forsberg, 1998; Morris and Cole, 1987)
of methylcellulose, which mediated detachment of cel- using various assays of adhesion measurement, show-
lulolytic rumen bacteria to cellulose without affecting ing that a state in which 100% of the cellulolytic rumen
enzymes activity also blocked cellulose degradation bacterial cells are attached to cellulose does not exist;
(Kudo et al., 1987; Pell and Schofield, 1993). This review there is always a free-suspended, unattached popula-
will focus mainly on the predominant rumen cellulolytic tion of cellulolytic bacterial cells. Wells et al. (1995)
bacteria: F. succinogenes, R. flavefaciens, and R. albus, presented evidence to demonstrate that F. succinogenes
whose attachment mechanism to fibrous plant particles released cellodexterins during growth on glucose, cello-
has been extensively studied over the last decade. These biose, or cellulose. Therefore, nonadherent cells and
species are considered as firmly attached bacteria that daughter cells of either F. succinogenes or ruminococci
can adhere to cellulose but are incapable of attaching may have a source of nutrients for growth and survival
to insoluble starch (McAllister et al., 1994; Minato et in the rumen liquid phase (Russell, 1985) and possibly
al., 1993; Miron et al., 1989; Pell and Schofield, 1993). are poised for adhesion to new cell wall particles enter-
In this paper, we describe the understanding ob- ing the rumen. Thus, incoming forages would be colo-
tained so far on: phases of adhesion of rumen cellulolytic nized primarily by nonadherent cells and by daughter
bacteria to plant tissues and their importance for subse- cells released from cell division processes on the colo-
Journal of Dairy Science Vol. 84, No. 6, 2001

1296 MIRON ET AL.
Figure 1. Scanning electron micrographs of cationized-ferritin-pretreated bacterial cells of the strains (A) Ruminococcus albus 7, (B)
Ruminococcus flavefaciens FD1, (C) Fibrobacter succinogenes S85, (D) Co-culture of F. succinogenes BL2 + Butyrivibrio fibrisolvens D1, (E)
Coculture of F. succinogenes S85 + B. fibrisolvens D1, and (F) F. succinogenes BL2. Bacterial strains were grown on cell walls of alfalfa (A,
B and C) or ozone-treated cotton stalks (E and F) or SO2-treated wheat straw (D) as the sole added carbohydrate substrate, and attached
to cell wall particles. Note: 1. The presence of protuberant structures on bacterial surfaces and adhesins formation with the substrate (A,
B, C, F); 2. The dense layer of firmly attached bacterial cells covering wrapped straw stem and broken edges of vascular cotton stalks tissue,
compared to the absence of bacterial colonization on nearby unwrapped and protected tissues (D, E and F); 3. The mode of complimentary
living of attached F. succinogenes cells with B. fibrisolvens (D and E). These micrographs were previously published in (Miron et al. 1989;
Miron and Ben-Ghedalia, 1992, 1993a).

nized fibers or through direct physical contact between substrates (Ben-Ghedalia et al., 1993; Miron and Ben-
the incoming forage and the already colonized fibers Ghedalia, 1992, 1993a; and Figure 1D and E). The
(Weimer, 1996). Despite the poorer mixing in the ru- chewing that occurs during eating and rumination by
men, this process apparently occurs with sufficient ra- the host animal is necessary to disrupt the protective
pidity because of the relatively high volumetric concen- cutin layer to expose more digestible portions of the
tration of both fluid-phase microorganisms and fiber plant and to increase the surface area of the substrate
(Weimer, 1996). Nonspecific initial adhesion in the ru- and its hydration. This process increases the probabil-
men is dependent on various factors related to the na- ity that cellulolytic bacteria will initiate nonspecific
ture of the bacteria, the substrate, and environmental binding to fibrous sites. Several studies have shown
conditions, as described later. that although the rate of adhesion of different bacterial
species to cellulose varies, this generally occurs shortly
Phase II, Initial Nonspecific Adhesion after contact with solid substrates. Adhesion of rumino-
cocci species to damaged plant tissues or cellulose oc-
Nonspecific adhesion is initiated when the bacteria curs within 1 to 5 min after their addition into the
arrive within range (2 to 50 nm) of van der Waals medium (Latham et al., 1978a, 1978b; Morris, 1988;
forces—hydrophobic, ionic, and electrostatic interac- Morris and Cole, 1987; Roger et al., 1990). However,
tion with the solid substrate (Busscher and Weerkamp, maximum adhesion of F. succinogenes cells is not at-
1987; Pell and Schofield, 1993). Nonspecific adhesion is tained until 15 to 30 min after contact with cellulose
defined as a combination of reversible and irreversible (Gong and Forsberg, 1989; Roger et al., 1990). The ini-
processes without the involvement of specific adhesins tial nonspecific adhesion is a prerequisite for the third
or ligands between bacteria and substrate receptors phase, in which bacterial-substrate linkages and adhes-
(Pell and Schofield, 1993). In addition to the physio- ins are created.
chemical forces, conformation to substrate shape and
wedging into cavities of feed are also considered as Phase III: Specific Adhesion
nonspecific adhesion. The constitutive bacterial glyco-
calyx components appear to be involved in the initial Specific adhesion is defined as a process in which
binding process (Cheng et al., 1977, 1983; Cheng and ligands or adhesins on the bacterial cell surface recog-
Costerton, 1980). Glycocalyx was defined by Costerton nize receptors on the substrate tissue (Pell and Scho-
et al. (1981) as those polysaccharide-, glycoprotein-, or field, 1993). The surface topology of F. succinogenes, R.
protein-containing structures of bacterial origin lying albus, and R. flavefaciens cells grown on and attached
outside the outer membrane of gram-negative cells (F. to plant cell walls were characterized by extensive ag-
succinogenes) and the peptidoglycan of gram-positive gregation of protuberances on the glycocalyx layer and
cells (R. flavefaciens or R. albus). The components of formation of adhesins connections with the plant cell
bacterial glycocalyx involved in the nonspecific adhe- walls sites, that could be seen only after cationized-
sion process may include constitutive carbohydrate epi- ferritin prestaining (Figure 1A, B, C, and F). These
topes (not induced by cellulosic substrate) of the glyco- protuberance organelles created after several hours of
calyx layer and its proteins, cellulose binding proteins incubation have been suggested to be cellulosome com-
(CBP), and additional factors (Cheng et al., 1987; plexes and, more recently, to have fimbriae structure
Cheng and Costerton, 1980; Gong et al., 1996; Latham (Kim et al., 1999; Miron et al., 1989, 1990; Morrison and
et al., 1978a; Miron and Forsberg, 1998, 1999; Morrison Miron, 2000; Pegden et al., 1998). We have suggested
and Miron, 2000; Ohara et al., 2000). (Miron et al., 1989, 1990) that attached ruminal bacte-
Short- and long-term incubation studies demon- ria receive stimulating signals during initial cell wall
strated that bacteria randomly transported to sites of polysaccharide digestion for the subsequent manufac-
potential adhesion start their initial adhesion mainly ture of inducible linkages and adhesins between the
on cut or macerated surfaces of forage particles as dem- outside glycocalyx layer of the bacteria and the digest-
onstrated in Figure 1D and E and reported in several ible tissue and for the manufacture of cellulolytic en-
studies (Dinsdale et al., 1978; Latham et al., 1978a, zymes as shown in Figure 1A, B, C, and F. This sugges-
1978b; Miron and Ben-Ghedalia, 1992, 1993a, 1993b, tion has been supported by previous studies (Bera et
1993c). Forage tissues protected by cutin are resistant al., 1999; Doerner et al., 1992; Flint et al., 1999; Gong
to adhesion (Akin, 1989; Bauchop, 1980). Some forage et al., 1996; Karita et al., 1997; Malburg et al., 1997;
tissues contain 18 to 24% silica, which impedes its di- McGavin and Forsberg, 1989; Mitsumori and Minato,
gestion (McAllister et al., 1994). Physical or chemical 1995, 1997; White et al., 1997) based on incubation for
pretreatments of fibrous substrate before feeding may several hours, showing that the presence of cellulose
create more adhesion sites compared with untreated or xylan substrates induced the production of some bac-

1298 MIRON ET AL.
terial cellulolytic and xylanolytic enzymes and also pos- particles (Figure 1C, D, and E). Latham et al. (1978a,
sibly of some cellulose binding proteins. 1978b) observed that the epidermis of perennial rye-
The importance of adhesin formation in plant cell grass is colonized by R. flavefaciens within 30 min of
wall digestion has been demonstrated in several elec- incubation in rumen fluid. R. flavefaciens predominates
tron microscopy observations (Akin, 1979; Kudo et al., only on uncut surfaces of epidermis, phloem and scle-
1987; Latham 1980; Miron et al., 1989, 1990; Morris renchyma cell walls. In contrast, F. succinogenes colo-
and Cole, 1987), and adhesins always formed later in nizes more slowly on cut edges of most plant cells, ex-
the adhesion process. Structures that have been fre- cept those of xylem, and also colonizes uncut surfaces
quently proposed as adhesins in cellulolytic bacteria of mesophyll, epidermis, and phloem cell walls. Dins-
include polycellulosome complexes, fimbriae or pili, gly- dale et al. (1978) found that cell walls of grass leaves
cocalyx capsule, cellulosic fibrils, cellulose binding pro- are digested primarily by ruminococci while fibrous cell
teins, and enzyme binding domains (Morrison and walls and cotton fibers were colonized by F. succino-
Miron, 2000; Pell and Schofield, 1993). genes. Thus, bacterial species demonstrate different
Fibrobacter succinogenes, R. flavefaciens, and R. al- specificity for binding and colonization, which serves to
bus cells preadapted to grow on plant cell walls produce reduce competition. Additional documentation of adhe-
protuberant-like organelles on their cell surface, adhere sion and colonization specificity was provided by Bhat
to cellulose better via adhesins formation, and degrade et al. (1990), who found that R. flavefaciens and F.
plant cell wall faster than bacteria preadapted to grow succinogenes have separate specific adhesion sites on
on cellobiose (Kim et al., 1999; Miron et al., 1989, 1990; barley straw. After adhesion and initial colonization, a
Miron and Ben-Ghedalia, 1993a). Based on these data, film of bacterial layer were evident after several hours
we suggest that a high concentration in the rumen of
of incubation on the interior of the tissues between
viable daughter bacterial cells containing the induced
plant parenchyma and phloem cells and subsequently
glycocalyx organelles needed for specific adhesion and
begins the digestion of the plant cell walls (Cheng et
adhesins formation is responsible for reducing the lag
al., 1983; Latham et al., 1978a, 1978b; Figure 1C, D,
time of cell wall degradation in the rumen.
and E). The cell wall rich tissue is degraded by slowly
The strategy of specific and strong adhesion to the
cellulosic substrate provides several advantages for the diffused nonmotile bacteria and their enzymes from the
bacteria. First, the cellulolytic enzymes are concen- “inside-out” namely: starting from cell lumen of broken
trated on their substrate, and other microbes are ex- cells, toward S3, S2, and S1 layers of the secondary cell
cluded from the site of hydrolysis, which allows the wall, and terminates in the lignified middle lamella
attached bacteria to have first access to the products (Cheng et al., 1983; McAllister et al., 1994; Weimer,
of cellulose hydrolysis (Minato et al., 1993; Pell and 1996). The mode of degradation of plant cell walls by
Schofield, 1993; Figure 1D and E). Moreover, strong either F. succinogenes or R. flavefaciens is essentially
adhesion protects the attached bacteria from grazing through the production of well-defined pits in the colo-
by ruminal protozoa and protects their cellulolytic en- nized tissue. Cells of pure R. albus cultures are identi-
zymes themselves from ruminal proteases (Pell and fied as loosely adherent, always being at a distance
Schofield, 1993; Weimer, 1996). Finally, bacteria from the plant cell walls and surrounded by extensive
attached to food particles have a retention time in the condensed glycocalyx (Cheng et al., 1983). These obser-
rumen as much as three times longer than these free vations suggest that each of the cellulolytic bacterial
in the liquid phase, thus having better opportunity to species have different modes of adhesion to plant cell
digest plant cell walls polysaccharides (Minato et al., walls, as will be discussed later. Akin (1989) suggested
1993; Weimer, 1996). the following general pattern of ease and extent of grass
The intimate and specific linkages between the tissue digestion by rumen bacteria: mesophyll, phloem
attached bacteria and the potentially digestible sub- > epidermis, parenchymal boundle sheath > scleren-
strate results in a proliferation of new generations of chyma > lignified vascular xylem.
induced bacteria for the production of colonies, as de-
scribed in phase IV. If a bacterial cell is attached to FACTORS AFFECTING BACTERIAL ADHESION
an undigestible site of plant substrate, it would not
proliferate to produce a colony, as demonstrated in un- Several factors that affect both the nonspecific and
wrapped plant tissues (Figure 1D, E, and F). the specific phases of the adhesion process include: 1)
factors related to bacterial age, bacterial glycocalyx con-
Phase IV: Proliferation and Colonization dition and microbial competition; 2) factors related to
on Plant Tissues the substrate including, cuticle protection, surface area,
During this phase, the adhered bacteria proliferate hydration, ionic charge and cation exchange capacity;
to create colonies on potentially digestible sites of forage and 3) environmental factors including pH, tempera-

ture, and the presence of O2, cations (Na+, Ca+2, and et al., 1989; Roger et al., 1990). These studies suggest
Mg+2) and soluble carbohydrate. Much of the work on that both proteins and carbohydrate components of
the effects of bacteria, substrate, and environmental these bacteria were involved in their adhesion to cellu-
conditions on the adhesion process, as described below, lose, and adhesion also occurred by dead bacteria if
has been done in short-term incubations before de novo their protein structures were not modified or denatur-
glycocalyx synthesis. Therefore, the implications of ated. Features of these protein and carbohydrate struc-
these studies are more relevant to the initial nonspecific tures involved in adhesion mechanism are discussed
adhesion phase than to the specific phase. below.
Different types of assays have been used to measure Microbial competition. In vitro studies have dem-
adhesion: 1) a turbidimetric assay that measure adhe- onstrated that adhesion of F. succinogenes is inhibited
sion based on change in optical density after cellulose to a limited quantity of cellulose when the ruminococci
has been added to a culture and allowed to sediment species are added simultaneously (Bhat et al., 1990;
(Gong and Forsberg, 1989; Minato et al., 1993; Roger Mosoni et al., 1997). This competition was explained
et al., 1990); 2) an assay using bacteria radiolabeled by the faster adhesion of Ruminococci species (within
with 14C or 3H (Morris, 1988; Mosoni et al., 1997; Pell 5 min) compared with the slower adhesion of F. succino-
and Schofield, 1993; Rasmussen et al., 1989); and 3) genes (Mosoni et al., 1997). When R. flavefaciens and
an assay based on protein determination of bacteria F. succinogenes are already adherent, R. albus 20 adhe-
adhered to cellulosic substrate (Bhat et al., 1990; sion occurred without inhibition but involved R. flave
Weimer and Schmidt, 1989). The assay used may faciens detachment (Mosoni et al., 1997). These data
greatly influence the outcome of these studies (McAllis- suggest that the two Ruminococcus species have the
ter et al., 1994; Minato et al., 1993; Pell and Schofield, same adhesion site or physically hinder each other dur-
1993). However, the results summarized below are nec- ing their adhesion. However, the Ruminococcus species
essary to better understand the adhesion mechanism and F. succinogenes probably have different adhesion
and can serve to identify areas for further research sites (Mosoni et al., 1997). The rumen xylanolytic bacte-
aimed at improving the adhesion process in the rumen. ria Butyrivibrio fibrisolvens and Prevotella ruminicola
do not interfere with adhesion to and digestion of plant
Factors Related to Bacteria cell walls by ruminococci and F. succinogenes and act
with the cellulolytic species in a complementary man-
Age. Maximal adhesion of F. succinogenes and R. ner (Miron and Ben-Ghedalia, 1992, 1993a, 1993b,
flavefaciens was obtained when bacterial cultures were 1993c; Miron et al., 1994; Figure 1D and E). Some evi-
in their midexponential stage of growth (Bhat et al., dence suggests that rumen fungi can inhibit bacterial
1990; Weimer and Schmidt, 1989). fibrolytic activity (Joblin, 1997); however, interaction
Bacterial envelope condition. Several studies for adhesion to cellulose between protozoa or fungi and
demonstrated the effects of various treatments of bacte- the rumen cellulolytic bacteria have not been identified.
rial cultures on their adhesion ability. In F. succino-
genes, treatments of bacterial cells with several en- Factors Related to the Substrate
zymes (including trypsin, pronase, proteinase K, ther-
molysin, protease, and lipase), carbohydrate removal We have already noted the need to remove parts of
by periodate treatment, and protein fixation with glu- the cutin wrapping of the plant tissue before consequent
taraldhyde (0.1%) significantly decreased bacterial ad- rumen bacterial adhesion.
hesion to cellulose. Killing the bacterium by sodium Using a variety of cellulosic substrates, Weimer and
azide or formalin without protein denaturation had no Schmidt (1989) demonstrated that F. succinogenes cells
effect on adhesion, whereas killing the bacteria by heat adhered better to cationic cellulose ethers than to neu-
treatment that also cause complete protein denatur- tral crystalline cellulose, whereas anionic cellulose-
ation, strongly inhibited the adhesion (Gong and Fors ether substrate reduced adhesion of this bacterium. In-
berg, 1989; Minato et al., 1993; Pell and Schofield, 1993; creasing the surface area of the cellulosic substrate by
Roger et al., 1990). Similarly, in R. albus protease treat- fine grinding resulted in increasing adhesion of F. succi-
ments with trypsin and pronase, and carbohydrate re- nogenes.
moval by dextranase or periodate oxidation, decreased
bacterial adhesion to cellulose (Pell and Schofield, Environmental Factors
1993). In R. flavefaciens, fixation of bacterial proteins
with formaldhyde strongly inhibited adhesion, whereas Temperature. The adhesion to cellulose of the three
killing the bacterium with sodium arsenate or heat cellulolytic species was completely inhibited at temper-
treatment reduced adhesion only by 15% (Rasmussen atures below 4°C, and in R. albus and F. succinogenes

1300 MIRON ET AL.
adhesion also decreased in temperature above 50°C and genes specifically therefore recognize a cellobiose moi-
achieved maximal values at 30 to 38°C (Gong and Fors- ety site of cellulose and therefore addition of excess
berg, 1989; Minato et al., 1993; Morris and Cole, 1987; cellobiose blocks the bacterial binding factors (Minato
Pell and Schofield, 1993; Roger et al., 1990). Heat dena- et al., 1993).
turation (100°C) completely reduced F. succinogenes ad- The adhesion of the three rumen cellulolytic species
hesion, but was less effective on R. flavefaciens (Roger to cellulose is strongly inhibited by soluble derivatives
et al., 1990). Some differences between studies may be of cellulose including sodium-carboxymethylcellulose
ascribed to variations in the technique used for measur- (CMC) and methylcellulose (MC) added at concentra-
ing adhesion. tions of 0.1% (Bhat et al., 1990; Kudo et al., 1987; Mi-
pH. The effect of pH on adhesion of cellulolytic bacte- nato et al., 1993; Morris, 1988; Rasmussen et al., 1989).
ria to cellulose varied according to bacteria. Roger et However, these results are in conflict with the data of
al. (1990) showed that the adhesion of F. succinogenes Roger et al. (1990), reporting that the adhesion of F.
to cellulose increased as pH was increased from 4.5 to 6, succinogenes and R. flavefaciens is not inhibited by the
remained stable between pH 6 and 7, and fell abruptly addition of CMC. These findings suggest that the recog-
above pH 7.5. Notwithstanding, Gong and Forsberg nition site of cellulose binding factors of R. albus and R.
(1989) reported that the adhesion of this bacterium did flavefaciens is larger than a cellobiose unit or repeating
not change over a pH range of 5.3 to 6.8. Roger et al. cellobiose moiety, and therefore adhesion of these bacte-
(1990) also showed that the adhesion of R. flavefaciens ria to cellulose is blocked when the bacterial cells are
to cellulose was stable at pH values between 3.3 and coated with high molecular weight MC or CMC.
7.5, and decreased at pH 8, whereas Rasmussen et al. These data show the effect of several factors on the
(1989) reported that the adhesion of the bacterium was nonspecific adhesion phase, since adhesion measure-
not affected by changes in pH between 6 and 8. The ments are for short incubation periods (~1 h). However,
adhesion of R. albus was not affected by changes in pH some environmental factors that affect the viability of
between 5.5 and 8 (Morris, 1988). These differences bacteria (e.g., pH, Na+ depletion, and temperature) or
between studies may be ascribed to variations in the inhibition of bacterial growth due to bacteriocin secre-
technique and bacterial strains used for measuring ad- tion by coculture bacterial species (Chan and Dehority,
hesion. 1999) may also inhibit the specific adhesion phase that
Presence of cations. Gong and Forsberg (1989) re- occurred after several hours of fermentation.
ported that the adhesion of F. succinogenes to cellulose
was enhanced by the presence of either Ca+2 or Mg+2 MECHANISMS OF BACTERIAL ADHESION
and Na+, whereas Roger et al. (1990) reported that the TO CELLULOSE
adhesion of this bacterium was insensitive to the pres-
ence of divalent cations, although Na+ was essential Investigators have focused on four structures be-
for adhesion. The adhesion of R. flavefaciens was not lieved to be important in specific adhesion to cellulose
affected by deprivation of either Ca+2, Mg+2, or Na+, but of the rumen cellulolytic bacteria: 1) large multicompo-
was significantly reduced by the deprivation of both nent complexes called cellulosomes (Flint et al., 1999;
divalent cations. Therefore, Roger et al. (1990) sug- Karita et al., 1997; Lamed et al., 1987; Miron et al.,
gested that interaction between the bacterial cell and 1989, 1990; Morrison and Miron, 2000; Ohara et al.,
the divalent cations Ca+2 and Mg+2 is the main mecha- 2000); 2) fimbriae or pili adhesins (Morrison and Miron,
nism involved in R. flavefaciens adhesion, although hy- 2000; Pegden et al., 1998); 3) carbohydrate epitopes of
drophobic interactions and enzymes may also be in- bacterial glycocalyx layer (Cheng and Costerton, 1980;
volved. Gong et al., 1996; Miron and Forsberg, 1998, 1999; Pell
Soluble carbohydrates. The effect of soluble carbo- and Schofield, 1993); and 4) enzyme binding domains
hydrate on adhesion of cellulolytic rumen bacteria to (Gong et al., 1996; Karita et al., 1997; McGavin and
cellulose-solid substrate has been studied in several Forsberg, 1989; Mitsumori and Minato 1995, 1997). Ev-
works. The adhesion of F. succinogenes, R. albus, and idence for the occurrence of these structures in each of
R. flavefaciens to cellulose is not inhibited by glucose, the rumen bacterial species will be discussed below.
mannose, xylose, maltose, cellodextrins, and soluble
starch added at concentration of 1% (Minato et al., Adhesion via Cellulosome-like Complexes
1993). However, the adhesion of F. succinogenes to cel-
lulose is strongly inhibited by 1% cellobiose, whereas Cellulosomes are large, stable, multienzyme com-
that of ruminococci cells is only slightly affected by plexes specialized in the adhesion to and degradation
cellobiose (Minato et al., 1993; Morris, 1988; Rasmus- of cellulose that reside within protuberances visible on
sen et al., 1989). Cellulose-binding factors of F. succino- the cell surface (Bayer et al., 1998, 1999; Beguin et

cohesin domain, which is present in a surface protein

incorporated into the bacterium’s S-layer (Bayer et al.,
1998 and 1999).
Adhesion of bacterial cellulosome to cellulose is medi-
ated by cellulose-binding domain (CBD) of the scaf-
foldin plus CBD of enzymes connected to the scaffoldin.
Although six distinct families of CBD have been identi-
fied in cellulolytic microorganisms, the CBD of C. ther-
mocellum provides a good example of the structure and
function of scaffoldin’s CBD. Scaffoldin’s CBD is com-
posed of nine-stranded β-sandwich of jellyroll topology,
that form two antiparallel β sheets. The planar face of
the CBD molecule interacts with three successive
chains on the cellulose surface, by harboring a planar
strip of aromatic residues of amino acids aligned pre-
cisely along one of the cellulose chains. The stacking
interactions performed between the planar strip resi-
Figure 2. A schematic model of the Clostridium thermocellum
dues and the glucose rings along the cellulose chain are
cellulosome. Provided by Ed Bayer, Dept. of Biological Chemistry, considered the major cause of the specificity and strong
The Weizmann Institute of Science, Rehovot, Israel. binding of the CBD to crystalline cellulose substrate
(Bayer et al., 1998). Some CBD also appear to catalyze
the disruption of the hydrogen bonds interactions be-
al., 1996; Lamed et al., 1983). Electron microscopic, tween the chains of the crystalline cellulose (Din et
biochemical, and immunochemical methods were used al., 1991).
to establish the existence of “cellulosomes” in a phyloge- Our knowledge of “cellulase” enzyme structure and
netically diverse range of anaerobic bacteria, including function has progressed quite rapidly, and molecular
rumen cellulolytic species (Lamed et al., 1983, 1987; details describing the structure and function of various
Miron et al., 1989, 1990). Much of our knowledge of scaffoldins are also now beginning to emerge (Bayer et
cellulosomes has been derived from the study of Clos- al., 1999). The discovery and characterization of the
tridium spp., and a model of the Clostridium thermocel- cellulosomes from a number of bacterial species has
lum cellulosome is reproduced in Figure 2. An integral helped to explain why many anaerobic cellulolytic bac-
feature of the cellulosome is the “scaffoldin” or “cellulo- teria adhere tightly to plant surfaces, and why their
some-integrating protein.” It is a large, glycosylated “cellulase” activity is principally located on the bacte-
protein that possesses a series of functional domains rial cell surface.
shown to facilitate either enzyme adhesion (via type I Based on electron microscopy observations, Miron et
cohesin domains), cellulose binding or, in the case of al. (1989, 1990) suggested that the predominant rumen
the C. thermocellum scaffoldin, anchoring to the bacte- cellulolytic bacteria contain cellulosome-like com-
rial cell surface (via a type II dockerin domain). The plexes. Recently, genetic evidence was provided to this
catalytic components of the cellulosome include en- suggestion, showing that some of the enzymes of R.
zymes involved with the hydrolysis of plant cellulose flavefaciens and R. albus are integrated to form cellulo-
and heteroxylan. These proteins are also characterized some-like complexes (Flint et al., 1999; Karita et al.,
by their modular architecture, with discrete domains 1997; Kirby et al., 1997; Morrison and Miron, 2000;
coordinating either cleavage of polysaccharide bonds, Ohara et al., 2000; Reddy and Morrison, 1998).
cellulose binding or enzyme activity, and adhesion to
the scaffoldin (via a type I dockerin-type I cohesin inter- Adhesion via Fimbriae or Pili
action) (Bayer et al., 1998, 1999). A recent study of two
Clostridium species indicated that the type I dockerin Fimbriae or pili, which have been implicated in bacte-
of the catalytic proteins interact selectively and in a rial adhesion, are surface appendages that are 5 to 7
species-specific manner with the type I cohesins of the nm in width and 100 to 200 nm in length in gram-
scaffoldin (Pages et al., 1997). The anchoring of the C. negative bacteria (Pell and Schofield, 1993). Initially,
thermocellum cellulosome to the bacterial surface also fimbriae were identified on gram-negative bacteria, but
appears to be mediated by a cohesin-dockerin type of they also are involved in adhesion of gram-positive bac-
interaction. The type II dockerin of the C. thermocellum teria (Fives-Taylor and Thompson, 1985). The fimbriae
scaffoldin forms a stable interaction with a type II of gram-positive bacteria are aggregated proteins

1302 MIRON ET AL.
rather than the highly ordered structures observed in lyx carbohydrate in the adhesion process of the rumen
gram-negative cells (Doyle and Sonnenfeld, 1989). As cellulolytic bacteria.
more has been learned about the role of fimbriae in
adhesion, it has become clear that structural subunits Adhesion via Cellulose-Binding Domains
of fimbriae are the actual adhesins. Some subunits in of Cellulolytic Enzymes
the gram-positive bacteria Actinomyces viscosus (Yeung
and Cisar, 1990) and S. sanguis (Fenno et al., 1989) Examination of cellulase structure in some organ-
associated with the fimbriae have been identified as isms has revealed two functional domains: the active
adhesins. In E. coli, the carbohydrate-binding sites of catalytic domain that is responsible for the hydrolytic
three types of fimbriae are in small (28 to 35 kDa) cleavage of the glycosidic bonds, and the binding do-
repeated subunits, most of which are in the tips of the main that binds the bacterial enzymes to its substrate.
fimbriae with a few additional sites along their length In many of these cases, the CBD is linked to the cata-
(Lindberg et al., 1987). lytic core by linkers rich in hydroxy amino acids. Most
Recently, the research team of Morrison has identi- of the CBD identified in nonruminal bacteria share con-
fied in R. albus 8 a novel form of cellulose-binding pro- siderable homology in the presence of four conserved
tein (cbpC, 17.7 kDa) that belongs to the Pil protein tryptophan and additional two cysteine residues, and
family, being most similar to the type 4 fimbrial pro- the CBD do not participate in catalysis (Tomme et al.,
teins of gram-negative, pathogenic bacteria (Larson et 1995). Some CBD may participate in disruption of hy-
al., 1999; Pegden et al., 1998). Thus, R. albus appears drogen bonds between cellulosic chains (Din et al.,
to produce a fimbriae mechanism involved in its adhe- 1991). Because of the conserved aromatic residues, it
sion to cellulose, which is consistent with the morphol- is thought that CBD attached to cellulose either by
ogy of the cell glycocalyx observed by electron micro hydrogen bonding or hydrophobic interactions (Tomme
scopy (Costerton et al., 1981; Morrison and Miron, 2000; et al., 1995). Previous experiments have identified CBD
Patterson et al., 1975; Stack and Hungate, 1984). and shown that bacteria lacking these domains were
less adherent, and in some cases, less able to digest
Adhesion via Carbohydrate Epitopes crystalline cellulose (McGavin and Forsberg, 1989;
of Bacterial Glycocalyx Tomme et al., 1995).
Distinct binding domains have been identified in F.
Most of the evidence that implicated polysaccharides succinogenes, including the CBD of endoglucanases 2
in adhesion is from electron microscopy observations (EG2) and EGF, which are induced by cellulose, and
(Cheng and Costerton, 1980; Cheng et al., 1983; Dins- the chloride stimulated cellobiosidase (Clcbsase) (Fors
dale et al., 1978; Latham et al., 1978a, 1978b). Several berg et al., 1993; Gong et al., 1996; Huang et al., 1988;
studies reported that the slime layer surrounding R. Malburg et al., 1997; McGavin and Forsberg, 1989; Mit-
albus and R. flavefaciens was composed of glycoproteins sumori and Minato, 1995, 1997).
that their carbohydrate residues were involved in adhe- Recently, Karita et al. (1997) cloned a gene egVI en-
sion of the bacteria (Cheng and Costerton, 1980; La- coding a family 9 cellulase from R. albus F-40 and found
tham et al., 1978a). Treatment with protease (trypsin that the enzyme contained a distinct CBD.
and pronase) and dextranase and removal of glycocalyx No genetic evidence is available to demonstrate
carbohydrate by periodate oxidation significantly de- whether R. flavefaciens contains a distinct CBD in its
creased adhesion of R. albus and F. succinogenes to enzymes; however, a noncatalytic 30-kDa CBP has been
cellulose (Pell and Schofield, 1993). Using specific lec- identified in this bacterium (Mitsumori and Minato,
tins, Baintner et al. (1993) demonstrated that R. fla- 1997).
vefaciens, R. albus, and F. succinogenes react with lec-
tins that can specifically bind to glucose or mannose of ADHESION MECHANISMS IN
bacterial envelope, and F. succinogenes also reacts with FIBROBACTER SUCCINOGENES
lectins specific to galactose. These indirect studies sug-
gest that both protein and carbohydrate are involved Fibrobacter succinogenes binds tightly to the surface
in adhesion mechanism of cellulolytic bacteria. of plant materials via adhesins leading to extensive
More direct evidence for the role of carbohydrate in plant cell wall degradation (Forsberg et al., 1993; Miron
adhesion was given recently in Fibrobacter species et al., 1989; Miron and Ben-Ghedalia, 1993a, 1993b).
(Gong et al., 1996; Miron and Forsberg, 1998, 1999) At least nine different glucanase genes and one
and in R. albus SY3 (Miron, 2001, unpublished data), cellodextrinase gene plus four xylanase genes have
as described later. However, additional biochemical and been cloned from F. succinogenes S85 (Bera et al., 1999;
genetic evidence is needed to explore the role of glycoca- Forsberg et al., 1993; Iyo and Forsberg, 1996; Malburg

and Forsberg, 1993; Malburg et al., 1997). Cellulolytic to cellulose. The only evidence to support the existence
enzymes of F. succinogenes S85 that have been purified, of either cellulosome complex or fimbriae structures
cloned, and characterized in these studies include en- involved in the adhesion mechanism of this bacterium
doglucanases (EG) EG1, EG2, EG3, EGB, EGD, EGE, is scanning electron microscopy observations (Gong and
EGF and EGG, a chloride stimulated cellobiosidase Forsberg, 1989; Miron et al., 1989; Miron and Ben-
(Clcbsase); a cellodextrinase (CEDA); and the endoxyla- Ghedalia, 1992, 1993a, 1993b, 1993c; Figure 1 C to F).
nases EX1, EX2, and XynC. In addition, a number of However, in bacterial cells prestained with cationized
heteroxylan-debranching enzymes, including acetyl-xy- ferritin, the presence of ultrastructural protuberances
lan esterase, arabinofuranidase, α-glucoronidase, and is sometimes connected to growth rate rather than to
lichenase were also identified in this bacterium. Parts induction of cellulolytic systems (Blair and Anderson,
of those enzymes were cell associated with the outer 1998).
layers of the bacterium, and production of almost all
enzymes was quantitatively induced by cellulosic sub- ADHESION MECHANISMS IN
strate (Bera et al., 1999). However, only three of the RUMINOCOCCUS FLAVEFACIENS
F. succinogenes enzymes including the endoglucanases
EG2 and EGF and the chloride-stimulated cellobiosi- The rumen cellulolytic bacterium Ruminococcus fla-
dase (Clcbsase) probably contained a distinct CBD. vefaciens adhere immediately and firmly to fibrous
Thus, at least those enzymes of F. succinogenes that plant particles and degrade grass and straw cell-wall
contain a CBD may be involved in bacterial adhesion polysaccharides faster than the other ruminal cellulo-
to cellulose (Forsberg et al., 1993; Huang et al., 1988; lytic species (Latham et al., 1978a, 1978b; Miron et al.,
McGavin and Forsberg, 1989). 1994). For example, R. flavefaciens FD-1 has a maxi-
In addition, evidence suggests that seven CBP of 40, mum dilution rate on crystalline cellulose (0.1/h) that
45, 50, 120, 180, 220, and 240 kDa from outer mem- is higher than that of other ruminal bacteria (Shi and
brane of F. succinogenes, which possess a common car- Weimer, 1992). Ultrastructural observations of several
bohydrate epitope, are involved in the adhesion mecha- strains of R. flavefaciens cells grown on and attached
nism. Immunogold labeling of the 180-kDa CBP demon- to plant cell walls demonstrated that the bacterium
strated its importance in adhesion to cellulose via the contained protuberances on its surface and forms ad-
common glycosidic epitope (Gong et al., 1996). Periodate hesins with the cellulosic substrate (Latham et al.,
oxidation of carbohydrate and protease treatments of 1978a; Miron et al., 1989, 1990; Stewart et al. 1990).
bacterial cells prior to adhesion demonstrate the possi- White et al. (1997) reported that the endoglucanases
ble importance of these glycoproteins in mediating the of R. flavefaciens FD1 exist in two forms: a large enzyme
adhesion to cellulose of F. succinogenes (Pell and Scho- complex of molecular mass greater than 3000 kDa
field, 1993). termed as complex A and a smaller fraction of enzyme
In a recent study employing another member of the activity (89 kDa) designated as complex B. They sug-
Fibrobacter family (the lower gut bacterium F. intesti- gested that complex A contains at least 13 different
nalis DR7), the role of carbohydrate in Fibrobacter ad- endoglucanase activities, whereas complex B has five
hesion was further supported. Using immunochemical unique endoglucanase activities. Furthermore, some of
methods, we have demonstrated that the carbohydrate the polypeptides in these complexes were glycosylated.
components of glycosylated CBP isolated from the outer Gene sequence analysis of the three endoglucanases
membrane and periplasm of Fibrobacter intestinalis and one cellodextrinase identified in R. flavefaciens
DR7 (lower gut bacteria) play a significant role in the FD1, and of the three xylanases (Xyn B, C, and D),
adhesion of this bacterium to cellulose. These isolated endoglucanase (endA), and esterase (EstA) identified
CBP included residues of glucosamine, galactosamine, in R. flavefaciens 17, demonstrate that these enzymes
glucuronic acid, and galacturonic acid that blocked ad- lack any distinct CBD (Doerner et al., 1992; Flint et
hesion to cellulose when premixed as neutral monosac- al., 1999; White et al., 1997). However, some of R. fla-
charides with cellulose in the growth medium before vefaciens 17 enzymes including Xyn B, Xyn D, EndA,
bacterial addition (Miron and Forsberg, 1998, 1999). and EstA contain a dockerin-like domain, suggesting
However, additional biochemical and genetic evidence that they are integrated to form a cellulosome-like com-
is needed to explore the role of glycocalyx carbohydrate plex that may be involved in adhesion mechanism (Flint
in the adhesion process F. succinogenes. et al., 1999; Kirby et al., 1997). Recently, in a collabora-
Thus, both the glycosidic residues of the outer mem- tive study between the research groups of Flint (Scot-
brane CBP and especially of the 180-kDa CBP, and the land) and the Israeli group of Lamed and Bayer, the
distinct CBD of EG2, EGF, and Cl-stimulated cellobio- scaffoldin protein of the cellulosome complex of R. fla-
sidase may play a role in the adhesion of F. succinogenes vefaciens 17 has been identified and sequenced, and

1304 MIRON ET AL.
several cohesins connecting the scaffoldin with type 1 cellulose-degrading activities of R. albus SY3 were in-
dockerin of the catalytic enzymes have been identified. deed associated with outer bacterial layers (capsule and
However, it is not clear yet whether R. flavefaciens cell walls) and not secreted into the extracellular me-
scaffoldin contains a distinct CBD or not (Lamed and dium. However, only small portion of the cellulolytic
Bayer, 2000, personal communication). and xylanolytic activities were associated with a cellulo-
Mitsumori and Minato (1995 and 1997) identified a some-like enzymatic complexes of molecular mass >400
firmly attached CBP of 30 kDa in some R. flavefaciens kDa (Miron, 2000, unpublished data). Our findings are
strains; however, its role in the adhesion process is not consistent with electron microscopy observations that
clear yet. the adherent R. albus cells are embedded in a glycocalyx
The early study of Latham et al. (1978a) regarding capsule and their cell membranes do not directly con-
the possible role of glycocalyx glycoproteins in mediat- tact cellulose surface (Cheng et al., 1983; Dinsdale et
ing adhesion of R. flavefaciens cells to cellulose, was al., 1978). Genetic support for our quantitative activity
further supported by a study with lectins responding data is provided by sequence analysis of endogluca-
with bacterial glycocalyx (Baitner et al., 1993). The
nases celA and celB of R. albus SY3 and endoglucanases
finding that polypeptides of the two complexes identi-
I, II, III, and IV of R. albus F-40, and several xylanases,
fied in R. flavefaciens FD1 are glycosylated (White et al.,
showing that these enzymes lack any epitope of either
1997) supports the possible importance of carbohydrate
a CBD or a dockerin-like domains, suggesting that they
epitopes in adhesion of the bacterium. The adhesion of
R. flavefaciens to cellulose was inhibited by MC or CMC are not integrated as part of a cellulosome complex
added to medium (0.1%) but not by the addition of cello- (Karita et al., 1997; Nagamine et al., 1997; Ohara et
biose (1%), suggesting that the recognition site of cellu- al., 2000; Poole et al., 1990; White et al., 1997). Even
lose binding factors of this bacterium is larger than a though most of the R. albus SY3 endoglucanases are
repeating cellobiose moiety (Bhat et al., 1990; Minato not integrated into cellulosome-like organelles, we have
et al., 1993; Rasmussen et al., 1989). isolated high molecular weight complexes that contain
Thus, at least two mechanisms, cellulosome-like com- mainly xylanases and some endoglucanase activity, and
plexes and carbohydrate epitopes of the glycocalyx most of the activity was attached to cellulose. Thus it
layer, are involved in the adhesion of R. flavefaciens is suggested that the cellulosome complex of R. albus
to cellulose. may contain a CBD or enzymes employing CBD (Miron,
2001, unpublished data).
ADHESION MECHANISMS IN In parallel, a cellulosome complex was isolated from
RUMINOCOCCUS ALBUS the culture supernatant of R. albus F-40 grown on cellu-
lose, and its components were identified as three pre-
The Cellulosome Paradigm
viously sequenced endoglucanase (egV, egVI, and eg-
Indirect evidence for the presence of cellulosome-com- VII) plus additional unidentified five endoglucanases,
plex organelles in R. albus was obtained by a combina- three xylanases, and four nonenzymatic proteins (Kar-
tion of electron microscopic and immunohistochemical ita et al., 1997; Ohara et al., 2000). Genetic evidence
methods (Kim et al., 1999; Lamed et al., 1987; Miron for the presence of cellulosome-like organelles in R.
et al., 1989, 1990, 1998). Wood et al. (1982) reported albus 8 were also provided by Reddy and Morrison
that the cellulases activity of cellobiose grown R. albus (1998), who isolated mutants incapable of adhesion and
SY3 was cell associated and of high molecular mass found that mutant strains lacked a 115- (CbpD) and
unstable complex (1.5 MDa) and could be disrupted by 95-kDa (CbpE) CBP. Several independent clones that
dissociating agents into proteins of low molecular mass.
hybridized to the CpbD probe have been isolated and
Later it was found that rumen fluid factors identified
partially sequenced. Sequences highly similar to the
as phenylpropanoic acid (PPA) or phenylacetic acid
type I dockerins of various Clostridium endoglucanases
(PAA), stabilized R. albus 8 cellulases and prevent dis-
sociation of its surface organelles (Pegden et al., 1998; and xylanases, as well as the dockerin domain recently
Stack and Hungate, 1984). In contrast to the findings reported by Karita et al. (1997), were identified in these
of Wood et al. (1982), we have isolated and separated clones (Morrison and Miron, 2000; Reddy and Mor-
the glycocalyx capsule, inner membranes, and peptido- rison, 1998).
glycan cell walls of cellobiose grown R. albus SY3, from The questions whether the cellulosome-like complex
the extracellular fluid and the cytoplasm, by using a of R. albus contains a CBD on its scaffoldin skeleton
combination of different buffers extraction, centrifuga- is still open, and the structure and sequence of the
tion, ultracentrifugation, and enzymatic solubilization. scaffoldin, dockerins, cohesions, and CBD elements are
We found that most of the cellulases, xylanases, and still unknown.

The Fimbriae Paradigm al., 1985). These motifs are thought to have an im-
portant role in the recognition and adhesion of Rickett-
Microscopic examinations of several strains of R. al- sia spp. to cell surfaces. These similarities in structure
bus revealed structures other than cellulosome-like pro- and function add credibility to the proposed role for
tuberances that appear to mediate adhesion to cellu-
CbpC in adhesion to plant surfaces.
lose. Patterson et al. (1975) described the presence of
Further studies with the R. albus 8 wild type and
“extensive amounts of fibrillar, extracellular material,”
mutant strains have further established CbpC role in
which projected as much as 600 nm from the cell sur-
the adhesion process. Morrison’s group showed (Larson
face, and were believed to be primarily responsible for
and Morrison, 1999; Larson et al., 1999; Pegden et al.,
adhesion to cellulose. Stack and Hungate (1984) later
1998) that R. albus adhesion in cellulose-binding assays
described the presence of “fimbrial-like structures”
and cbpC transcript abundance were significantly in-
when R. albus 8 was provided with PAA or PPA, and
creased by the inclusion of either ruminal fluid or micro-
it was shown that R. albus adhesion to cellulose is
molar concentrations of both PAA or PPA in the growth
increased by the inclusion of PAA orPPA in the growth
medium, that probably induce the production and stabi-
medium (Pegden et al., 1998). Electron microscopic ex-
lization of fimbriae structures (Morrison and Miron,
amination of R. albus strain F-40 also illustrated simi-
2000). A series of cellulose-binding experiments have
lar structures mediating adhesion to cellulose (Kim et
also been conducted with preparations of the CbpC pro-
al., 1999). With this background, Morrison and cowork-
tein in the presence or absence of CMC (Pegden et al.,
ers decided to take a functional proteomics approach to
1998). Results suggest that CMC affects the association
identify and isolate cellulose-binding proteins (Larson
constant (Ka) of cbpC binding to cellulose, but not its
and Morrison, 1999; Larson et al., 1999; Pegden et al.,
maximal binding (Vmax), and, therefore, CMC serves
1998; Reddy and Morrison, 1998). Several proteins of
as a competitive inhibitor of cbpC binding to cellulose.
relatively small molecular mass (16 to 25 kDa) were
Southern blot analysis has confirmed that a number of
identified, and initial characterization of one of these
other R. albus strains possess cbpC gene homologue(s).
proteins, hereafter referred to as cellulose-binding pro-
Additionally, Western immunoblot analysis identified
tein type C (cpbC) was completed (Larson and Morrison,
1999; Larson et al., 1999; Pegden et al., 1998). The cbpC a protein of V25-kDa present in R. albus SY3 that was
gene was isolated by a combination of reverse genetics cross-reactive with anti-CbpC antibodies, and in subse-
and genomic walking procedures, and shown to encode quent experiments the same protein was shown to be
a protein of 169 amino acids with a calculated molecular a cellulose-binding protein.
mass of 17,655 Da. Although the CbpC protein pos- Based on these findings, it seems reasonable to pro-
sesses no sequence similarity with existing CBD fami- pose that the CbpC protein that is the repeated building
lies, motifs characteristic of other, relatively well-char- block for fimbriae creation, is a newly identified strat-
acterized proteins are present. Notably, the amino-ter- egy for the adhesion of gram-positive bacteria to cellu-
minal third of the CbpC protein possesses a leader lose (Morrison and Miron, 2000; Pegden et al., 1998).
peptide, cleavage site, and motif characteristic of the
Pil-family of proteins, especially the type 4 fimbrial CBD of Enzymes
subunit proteins from the gram-negative species: Di-
chelobacter nodosus, Moraxella bovis, Neisseria gonorr- Recently, Karita et al. (1997) cloned a gene egVI en-
hoeae, and Pseudomonas aeruginosa. The type 4 fim- coding a family 9 cellulase from R. albus F-40. The
brial subunits are relatively low molecular mass poly- sequence analysis revealed that this enzyme consists
peptides (20 to 25 kDa) and the resulting fimbriae are of three domains: a family 9 catalytic domain, a family
located at the polar ends of the bacterial cell (Hobbs IIIb CBD, and a dockerin-like reiterated sequence, simi-
and Mattick, 1993). The CbpC and the fimbrial proteins lar to clostridial type dockerin. This is the first genetic
all possess a fairly long stretch of relatively hydrophilic evidence available for the presence of a cellulase CBD
residues at their carboxy-terminus, while other mem- among the seven endoglucanases and three xylanase
bers of the Pil-family do not (Pegden et al., 1998). The that have been cloned and sequenced from R. albus F-
similarities among CbpC and type 4 fimbrial proteins 40 (Karita et al., 1997; Nagamine et al., 1997; Ohara
suggest that cbpC binding to cellulose is most likely et al., 2000). Although egVI contained dockerin type 1
mediated via the carboxy two-thirds of the protein. Sig- element and thus is a part of the cellulosome complex
nificant sequence identity exists between this part of of R. albus F-40, the CBD of the enzyme provides addi-
the CbpC protein and 72- and 75-amino acid motifs that tional mechanism for bacterial adhesion to cellulose,
are tandemly repeated 13 times within the 190-kDa apart from the CBD of the scaffoldin (whose existence
surface antigen protein of Rickettsia spp. (Anacker et has not yet been proven).

1306 MIRON ET AL.
Table 1. Summary of adhesion mechanisms in rumen cellulolytic bacteria.1
Fibrobacter Ruminococcus Ruminococcus

Adhesion mechanism succinogenes flavefaciens albus
Surface protuberances
(electron microscopy) + (1) + (1) + (1,2,3)
Adhesins connections
(electron microscopy) + (1,4) + (1,15) + (1,3)
Dockerin, (5)
Cellulosome organelles Scaffoldin, (6) Dockerin,
(genetic evidence) ? Cohesins (6) (7,8,11)
Fimbriae or pili
(genetic evidence) ? ? CbpC (9)
EG2, EGF
CBD of enzymes CI-Cbsase
(biochemical evidence) (10,16) ? EGVI (11)
Carbohydrate epitope of
identified CBP + (12) ? + (17)
Exopolysaccharides of
glycocalyx layer + (12,13) + (15) + (2,13,14)
1
References: 1 = Miron et al. (1989), 2 = Lamed et al. (1987); 3 = Kim et al. (1999); 4 = Miron et al. (1993a);
5 = Flint et al. (1999); 6 = Lamed and Bayer (2000, unpublished); 7 = Ohara et al. (2000); 8 = Reddy and
Morrison (1998); 9 = Pegden et al. (1998); 10 = Forsberg et al. (1993); 11 = Karita et al. (1997); 12 = Gong
et al. (1996); 13 = Pell and Schofield (1995); 14 = Cheng et al. (1977); 15 = Latham et al. (1978a); 16 =
Malburg et al. (1997); 17 = Miron. (2001, unpublished data). CBD = cellulose-binding domain, CBP = cellulose-
binding proteins, EG = endoglucanase.
Glycocalyx Exopolysaccharides tion of fimbrial-like structures, whose interaction with

cellulose can be competitively inhibited by CMC. Carbo-
Cheng et al. (1977) reported that the slime layer sur-
hydrate epitopes of some CBP and distinct CBD of cellu-
rounding R. albus was composed of glycoproteins and
lases are probably involved mostly in the nonspecific
formed a layer of approximately 100 nm at the cell
phase of the adhesion process.
surface. They suggested that the slime layer involved
in adhesion of these bacteria. Treatment with protease
(trypsin and pronase) and dextranase, and removal of CONCLUSIONS
glycocalyx carbohydrate by periodate oxidation, sig-
nificantly decreased adhesion of R. albus to cellulose Based on the literature, it appears that each of the
(Pell and Schofield, 1993). R. albus contained several predominant rumen bacteria F. succinogenes, R. fla-
monosaccharide residues on its glycocalyx capsule, in- vefaciens and R. albus has a specific mechanism of adhe-
cluding glucose or mannose and galactose that reacted sion to cellulose as summarized in Table 1.
with several specific lectins (Baitner et al., 1993) and In F. succinogenes, both the glycosidic residues of the
in particular with a lectin that recognized specific glyco- outer membrane CBP and especially of the 180-kDa
sylated epitope of C. thermocellum cellulosome (Lamed CBP, and the distinct CBD of EG2 EGF and Cl-stimu-
et al., 1987). These indirect studies suggest that carbo- lated cellobiosidase, may play a role in the adhesion to
hydrates are also involved in adhesion mechanism of cellulose. There is no direct evidence yet, except scan-
R. albus, although their role has not yet been clarified. ning electron microscopy observations, to support the
We have recently demonstrated that part of the CBP existence of either cellulosome complex or fimbriae
of R. albus SY3 contain glycosidic epitopes that are structures involved in the adhesion mechanism of F.
specifically immunoreactive with antibodies specific to succinogenes.
adhesion, suggesting their possible involvement in the At least two mechanisms, cellulosome-like complexes
adhesion process. These organelles were found in cells and carbohydrate epitopes of the glycocalyx layer are
grown either on cellulose or on cellobiose + glucose as involved in the specific adhesion of R. flavefaciens to cel-
the sole carbohydrate substrate, suggesting constitu- lulose.
tive nature, which is not induced by the substrate Ruminococcus albus possesses at least two mecha-
(Miron, 2001, unpublished data). nisms for specific adhesion to cellulose: a cellulosomal-
Based on these results, we propose that R. albus pos- like mechanism, and a cbpC (Pil)-protein mechanism
sesses at least two mechanisms for specific adhesion to probably involving the production of fimbrial-like struc-
cellulose: a cellulosomal-like mechanism, and a CbpC tures. Indirect and direct studies suggested that carbo-
(Pil)-protein mechanism probably involving the produc- hydrate epitopes of CBP and CBD epitope of cellulases

may also be involved mostly in the nonspecific phase Cheng, K. J., C. S. Stewart, D. Dinsdale, and J. W. Costerton. 1983.
Electron microscopy of bacteria involved in the digestion of plant
of adhesion of R. albus. cell walls. Anim. Feed Sci. Technol. 10:93–120.
Costerton, J. W., R. T. Irvin, and K. J. Cheng. 1981. The bacterial
glycocalyx in nature and disease. Annu. Rev. Microbiol.
ACKNOWLEDGMENTS 35:299–324.
Craig, W. M., G. A. Broderick, and D. B. Ricker, 1987. Quantitation of
The research conducted in J. Miron’s laboratory (Is- microorganisms associated with the particulate phase of ruminal
rael) and in M. Morrison’s laboratory (USA) has been ingesta. J. Nutr. 117:56–64.
Din, N., N. R. Gilkes, B. Tekant, R. C. Miller, R. Anthony, J. Warren,
partially supported by the US-Israel USDA-BARD Proj- and D. G. Kilburn. 1991. Non-hydrolytic disruption of cellulose
ect 2783-96. We also thank Ed Bayer for helpful discus- fibres by the binding domain of a bacterial cellulase. Biotechnol-
sions and the provision of Figure 2. ogy 9:1096–1099.
Dinsdale, D., E. J. Morris, and J.S.D. Bacon. 1978. Electron micro-
scopy of the microbial populations present and their modes of
REFERENCES attack on various cellulosic substrates undergoing digestion in
the sheep rumen. Appl. Environ. Microbiol. 36:160–168.
Akin, D. E. 1979. Microscopic evaluation of forage digestion by rumen Doerner, K. C., G. T. Howard, R. I. Mackie, and B. A. White. 1992. β-
microorganisms—A review. J. Anim. Sci. 48:701–710. Glucanase expression by Ruminococcus flavefaciens FD-1. FEMS
Akin, D. E. 1989. Histological and physical factors affecting digestibil- Microbiol. Lett. 93:147–154.
ity of forages. Agron. J. 81:17–23. Doyle, R. J., and E. M. Sonnenfeld. 1989. Properties of the cell surfaces
Anacker, R. L., R. H. List, R. E. Mann, S. F. Hayes, and L. A. Thomas. of pathogenic bacteria. Int. Rev. Cytol. 118:33–92.
1985. Characterisation of monoclonal antibodies protecting mice Fenno, J. C., D. J. LeBlanc, and P. Fives-Taylor. 1989. Nucleotide
against Rickettsia rickettsii. J. Infect. Dis. 151:1052–1060. sequence analysis of a type I fimbrial gene of Streptococcus san-
Baintner, K., S. H. Duncan, C. S. Stewart, and A. Pusztai. 1993. guis FW213. Infec. Immun. 57:3527–3533.
Binding and degradation of lectins by components of rumen li- Fives-Taylor, P. M., and D. W. Thompson. 1985. Surface properties
quor. J. Appl. Bacteriol. 74:29–35. of Streptococcus sanguis FW213 mutants nonadherent to saliva-
Bauchop, T. 1980. Scanning electron microscopy in the study of micro- coated hydroxyapatite. Infect. Immun. 47:752–759.
bial digestion of plant fragments in the gut. Pages 101–110 in Flint, H. J., V. Aurilia, J. Kirby, K. Miyazaki, M. T. Rincon-Torres,
Contemporary Microbial Ecology. D. C. Elwood, J. N. Hedger, M. S. I. McCrae, and J. C. Martin. 1999. Organization of plant cell
J. Latham, J. M. Lynchand, and J. H. Slater, eds. Academic Press, wall degrading enzymes in the ruminal anaerobic bacteria Rumi-
New York, NY. nococcus flavefaciens and Prevotella bryantii. Pages 511–519 in
Bayer, E. A., L.J.W. Shimon, R. Lamed, and Y. Shoham. 1998. Cellulo- Genetics, Biochemistry and Ecology of Cellulose Degradation. K.
somes: structure and ultrastructure. J. Struct. Biol. 124:221–234. Ohmiya, K. Hayashi, K. Sakka, Y. Kobayashi, S. Karita, and T.
Bayer, E. A., S. Y. Ding, Y. Shoham, and R. Lamed. 1999. New Kimura, eds. Uni Publishers, Tokyo, Japan.
perspectives in the structure of cellulosome-related domains from Forsberg, C. W., J. Gong, L.M.J. Malburg, H. Zhu, A. Iyo, K. J. Cheng,
different species. Pages 428–436 in Genetics, Biochemistry and P. J. Krell, and J. P. Phillips. 1993. Cellulases and hemicellulases
Ecology of Cellulose Degradation. K. Ohmiya, K. Hayashi, K. of Fibrobacter succinogenes and their roles in fibre digestion.
Sakka, Y. Kobayashi, S. Karita, and T. Kimura, Eds., Uni Publish- Pages 125–136 in Genetics, Biochemistry and Ecology of Lignocel-
ers, Tokyo, Japan. lulose Degradation. K. Shimada, S. Hoshino, K. Ohmiya, K.
Beguin, P., and M. Lemaire. 1996. The cellulosome: an exocellular, Sakka, Y. Kobayashiand, and S. Karita, eds. Uni Publishers Co.,
multiprotein complex specialized in cellulose degradation. Crit. Ltd, Tokyo, Japan.
Rev. Biochem. Mol. Biol. 31:201–236. Forsberg, C. W., and R. Lam. 1977. Use of adenosine-5-triphosphate
Ben-Ghedalia, D., J. Miron, and R. Solomon. 1993. The degradation as an indicator of the microbial biomass in rumen contents. Appl.
and utilization of structural polysaccharides of sorghum straw Environ. Microbiol. 33:528–534.
by defined ruminal bacteria. Anim. Feed Sci. Technol. 42:283–295. Gong, J., E. E. Egosimba, and C. W. Forsberg. 1996. Cellulose binding
Bera, C., G. Gaudet, and E. Forano. 1999. Regulation of glycosyl- proteins of Fibrobacter succinogenes and the possible role of a
hydrolase genes expression in Fibrobacter succinogenes S85. 180-kDa cellulose binding glycoprotein in adhesion to cellulose.
Pages 541–544 in Genetics, Biochemistry and Ecology of Cellulose Can. J. Microbiol. 42:453–460.
Degradation. K. Ohmiya, K. Hayashi, K. Sakka, Y. Kobayashi, Gong, J., and C. W. Forsberg. 1989. Factors affecting adhesion of
S. Karita, and T. Kimura, Eds., Uni Publishers, Tokyo, Japan. Fibrobacter succinogenes S85 and adherence defective mutants
Bhat, S., R. J. Wallce, and E. R. Orskov. 1990. Adhesion of cellulolytic to cellulose. Appl. Environ. Microbiol. 55:3039–3044.
ruminal bacteria to barley straw. Appl. Environ. Microbiol. Hobbs, M., and J. S. Mattick. 1993. Common components in the
56:2698–2703. assembly of type 4 fimbriae, DNA transfer systems, filamentous
Blair, B. G., and K. L. Anderson. 1998. Comparison of staining tech- phage and protein-secretion apparatus: a general system for the
niques for scanning electron microscopic detection of ultrastruc- formation of surface-associated protein complexes. Mol. Micro-
tural protuberances on cellulolytic bacteria. Biotech. Histochem. biol. 10:233–243.
73:107–113. Huang, L., C. W. Forsberg, and D. Y. Thomas. 1988. Purification
Busscher, H. J., and A. H. Weerkamp. 1987. Specific and non-specific and characterization of a chloride-stimulated cellobiosidase from
interactions in bacterial adhesion to solid substrate. FEMS Micro- Bacteroides succinogenes S85. J. Bacteriol. 170:2923–2932.
biol. Rev. 46:165–174. Iyo, A. H., and C. W. Forsberg. 1996. Endoglucanase G from Fibro-
Chan, W. W., and B. A. Dehority. 1999. Production of Ruminococcus bacter succinogenes S85 belongs to a class of enzymes character-
flavefaciens growth inhibitors by Ruminococcus albus. Anim. ized by a basic C-terminal domain. Can. J. Microbiol. 42:934–943.
Feed Sci. Technol. 77:61–71. Joblin, K. N. 1997. Interactions between ruminal fibrolytic bacteria
Cheng, K. H., D. E. Akin, and J. W. Costerton. 1977. Rumen bacteria: and fungi. Pages 3–10 in Rumen Microbes and Digestive Physiol-
Interaction with particulate dietary components and response to ogy in Ruminants. R. Onodera, H. Itabashi, K. Ushida, H. Yano,
dietary variation. Fed. Proc. 36:193–203. and Y. Sasaki, eds. Japan Scientific Societies Press, Tokyo, Japan.
Cheng, K. J., and J. W. Costerton. 1980. Adhesive bacteria—Their Karita, S., K. Sakka, and K. Ohmiya. 1997. Cellulosomes and cellu-
role in the digestion of plant material, urea and ephithelial cells. lase complexes of anaerobic microbes: their structure, models and
Pages 225–250 in Digestive Physiology and Metabolism in Rumi- function Pages 47–57 in Rumen Microbes and Digestive Physiol-
nants. Y. Ruckebusch and P. Thivend, Eds, MTP press Ltd., Lan- ogy in Ruminants. R. Onodera, H. Itabashi, K. Ushida, H. Yano,
caster, England. and Y. Sasaki, eds. Japan Scientific Societies Press, Tokyo.

1308 MIRON ET AL.
Kim, Y. S., S. G. Wi, and K. H. Myung. 1999. Ultrastructural studies Miron, J., and D. Ben-Ghedalia. 1993c. Digestion of cell wall monosac-
of a Ruminococcus albus surface structures involved in lignocellu- charides of ryegrass and alfalfa hays by the rumen bacteria Fibro-
lose degradation. Pages 531–540 in Genetics, Biochemistry and bacter succinogenes and Butyrivibrio fibrisolvens. Can. J. Micro-
Ecology of Cellulose Degradation. K. Ohmiya, K. Hayashi, K. biol. 39:780–786.
Sakka, Y. Kobayashi, S. Karita, and T. Kimura, eds. Uni Publish- Miron, J., D. Ben-Ghedalia, M. T. Yokoyama, and R. Lamed. 1990.
ers, Tokyo, Japan. Some aspects of cellobiose effect on cell surface structures in-
Kirby, J., J. Martin, A. Daniel, and H. J. Flint. 1997. Dockerin-like volved in lucerne cell walls utilization by fresh isolates of rumen
sequences in cellulases and xylanases from the rumen cellulolytic bacteria. Anim. Feed Sci. Technol. 30:107–120.
bacterium Ruminococcus flavefaciens. FEMS Microbiol. Lett. Miron, J., S. H. Duncan, and C. S. Stewart. 1994. Interactions between
149:213–219. rumen bacterial strains during the degradation and utilization of
Kudo, H., K. J. Cheng,, and J. W. Costerton. 1987. Electron micro- the monosaccharides of barley straw cell walls. J. Appl. Bacteriol.
scopy study of the methyl-cellulose mediated detachment of cellu- 76:282–287.
lolytic rumen bacteria from cellulose fibers. Can. J. Microbiol. Miron J., and C. I. Forsberg. 1998. Features of Fibrobacter intestinalis
33:267–272. DR7 mutant which is impaired with its ability to adhere to cellu-
Lamed, R., E. Setter, R. Kenig, and E. A. Bayer. 1983. The cellulosome lose. Anaerobe 4:35–43.
a discrete cell surface organelle of Clostridium thermocellum Miron J., and C. I. Forsberg. 1999. Characterisation of cellulose bind-
which exhibits separate antigenic, cellulose-binding, and various ing proteins which are involved in adhesion mechanism of Fibro-
cellulolytic activities. Biotechnol. Bioeng. 13:163–181. bacter intestinalis DR7. Appl. Microbiol. Biotechnol. 51:491–497.
Lamed, R., J. Naimark, E. Morgenstern, and E. A. Bayer. 1987. Miron J., E. Morag, E. A. Bayer, R. Lamed, and D. Ben-Ghedalia.
Specialized cell surface structures in cellulolytic bacteria. J. Bac- 1998. An adhesion defective mutant of Ruminococcus albus SY3 is
teriol. 169:3792–3800. impaired in its capability to degrade cellulose. J. Appl. Microbiol.
Larson, M. A., N.C.K. Heng, and M. Morrison. 1999. Identification 84:249–254.
of phenyl-substituted acid responsive operons in Ruminococcus Miron, J., M. Yokoyama, and R. Lamed. 1989. Bacterial cell surface
albus using differential display RT-PCR. Pages 1–7 in Proc. 99th structures involved in lucerne cell wall degradation by pure cul-
Annu. General Mtg. Am. Soc. Microbiol., Washington, DC. tures of cellulolytic rumen bacteria. Appl. Microbiol. Biotechnol.
Larson, M. A., and M. Morrison. 1999. Application of the differeren- 32:218–222.
tial-display RT-PCR technique to examine conditional gene ex- Mitsumori, M., and H. Minato. 1995. Distribution of cellulose-binding
pression in Ruminococcus albus. In Proc. 8th Int. Symp. Microbial proteins among the representative strains of rumen bacteria. J.
Ecol. (in press) Gen Appl. Microbiol. 41:297–306.
Latham, M. J. 1980. Adhesion of rumen bacteria to plant cell walls. Mitsumori, M., and H. Minato. 1997. Cellulose-binding proteins from
Pages 339–350 in Microbial Adhesion to Surfaces. R.C.W. Berke- rumen microorganisms. Page 47–57 in Rumen Microbes and Di-
ley, et al., eds. Ellis Horwood, Ltd, Chichester. gestive Physiology in Ruminants. R. Onodera, H. Itabashi, K.
Latham, M. J., B. E. Brooker, G. L. Petipher, and P. J. Harris. 1978a. Ushida, H. Yano, and Y. Sasaki, eds. Japan Scientific Societies
Ruminococcus flavefaciens cell coat and adhesion to cotton cellu- Press, Tokyo, Japan.
lose and cell walls in leaves of perennial ryegrass. Appl. Environ. Morris, E. J. 1988. Characteristics of the adhesion of Ruminococcus
Microbiol. 35:156–165. albus to cellulose. FEMS Microbiol. Lett. 51:113–118.
Latham, M. J., B. E. Brooker, G. L. Petipher, and P. J. Harris. 1978b. Morris, E. J., and O. J. Cole. 1987. Relationship between cellulolytic
Adhesion of Bacteroides succinogenes in pure cultures and in the activity and adhesion to cellulose in Ruminococcus albus. J. Gen.
presence of Ruminococcus flavefaciens to cell walls in leaves of Microbiol. 133:1023–1032.
perennial ryegrass. Appl. Environ. Microbiol. 35:1166–1173. Morrison, M., and J. Miron. 2000. Adhesion to cellulose by Ruminococ-
Lindberg, F., B. Lund, L. Johansson, and S. Normark. 1987. Localiza- cus albus: a combination of cellulosomes and Pil-proteins? FEMS
tion of the receptor-binding protein adhesin at the tip of the Microbiol. Letters, 185:109–115.
bacterial pilus. Nature 328:84–87.
Mosoni, P., G. Fonty, and P. Gouet. 1997. Competition between rumi-
Malburg, L. M., and C. W. Forsberg. 1993. Fibrobacter succinogenes
nal cellulolytic bacteria for adhesion to cellulose. Curr. Microbiol.
possesses at least nine distinct glucanase genes. Can. J. Microbiol.
35:44–47.
39:882–891.
Nagamine, T., R. I. Aminov, K. Ogata, M. Sugiura, K. Tajima, and
Malburg, S.R.C., L. M. Malburg, T. Liu, A. H. Iyo, and C. W. Forsberg.
Y. Benno. 1997. Cloning of xylanase genes from Ruminococcus
1997. Catalytic properties of the cellulose-binding endoglucanase
albus and chromosome mapping of Fibrobacter succinogenes.
F from Fibrobacter succinogenes S85. Appl. Environ. Microbiol.
Pages 59–67 in Rumen Microbes and Digestive Physiology in
63:2449–2453.
Ruminants. R. Onodera, H. Itabashi, K. Ushida, H. Yano, and Y.
McAllister, T. A., H. D. Bae, G. A. Jones, and K. J. Cheng. 1994.
Sasaki, eds. Japan Scientific Societies Press, Tokyo, Japan.
Microbial attachment and feed digestion in the rumen. J. Anim.
Sci. 72:3004–3018. Ohara, H., S. Karita, T. Kimura, K. Sakka, and K. Ohmiya. 2000.
McGavin, M., and C. W. Forsberg. 1989. Catalytic and substrate Characterization of cellulolytic complex (cellulosome) from Rumi-
binding domains of endoglucanase 2 from Bacteroides succino- nococcus albus. Biosci. Biotechnol. Biochem. 64:254–260.
genes. J. Bacteriol. 171:3310–3315. Orpin, C. G., and K. N. Joblin. 1997. The rumen anaerobic fungi.
Minato, H., M. Misumori, and K. J. Cheng. 1993. Attachment of Pages 140–195 in The Rumen Microbial Ecosystem. P. N. Hobson
microorganisms to solid substrates in the rumen. Pages 139–145 and C. S. Stewart, Eds. Blackie Academic and Professonal, Pub-
in Proc. MIE Bioforum on Genetic, Biochemistry and Ecology of lishers, London.
Lignocellulose Degradation. Institut Pasteur, Paris, France. Patterson, H. A., R. Irvin, J. W. Costerton, and K. J. Cheng. 1975.
Miron, J., and D. Ben-Ghedalia. 1992. The degradation and utilization Ultrastructure and adhesion properties of Ruminococcus albus.
of wheat straw cell wall monosaccharide components by defined J. Bacteriol. 122:278–287.
ruminal cellulolytic bacteria. Appl. Microbiol. Biotechnol. Pages, S., A. Belaich, J. P. Belaich, E. Morag, R. Lamed, Y. Shoham,
38:432–437. and E. A. Bayer. 1997. Species-specificity of the cohesin-dockerin
Miron, J., and D. Ben-Ghedalia. 1993a. Untreated and delignified interaction between Clostridium thermocellum and Clostridium
cotton stalks as model substrates for degradation and utilization cellulolyticum: prediction of specificity determinants of the dock-
of cell wall monosaccharide components by defined ruminal cellu- erin domain. Proteins. 29:517–527.
lolytic bacteria. Bioresource Technol. 43:241–247. Pegden, R. S., M. A. Larson, R. J. Grant, and M. Morrison. 1998.
Miron, J., and D. Ben-Ghedalia. 1993b. Digestion of structural poly- Adherence of the gram-positive bacterium Ruminococcus albus
saccharides of panicum and vetch hays by the rumen bacterial to cellulose and identification of a novel form of cellulose-binding
strains Fibrobacter succinogenes BL2 and Butyrivibrio fibrisol- protein which belongs to the Pil family of proteins. J. Bacteriol.
vens D1. Appl. Microbiol. Biotechnol. 39:756–759. 180:5921–5927.

Pell, A. N., and P. Schofield. 1993. Microbial adhesion and degrada- son and C. S. Stewart, eds. Blackie Academic and Professional
tion of plant cell walls Pages 397–423 in Forage Cell Wall Struc- Publishers, London.
ture and Digestibility. R. D. Hatfield, H. G. Jung, J. Ralph, D. Tomme, P., R.A.J. Warren, and N. R. Gilkes. 1995. Cellulose hydroly-
R. Buxton, D. R. Mertens, and P. J. Weimer, eds. ASA-CSSA- sis by bacteria and fungi. Advances in Microbial Physiology,
SSSA, Madison, WI. 37:1–81.
Poole, D. M., G. P. Hazlewood, J. I. Laurie, P. J. Barker, and H. J. Weimer, P. J. 1996. Why don’t ruminal bacteria digest cellulose
Gilbert. 1990. Nucleotide Sequence of the Ruminococcus albus faster? J. Dairy Sci. 79:1496–1502.
SY3 endoglucanase genes celA and celB. Mol. Gen. Genet. Weimer, P. J., and J. K. Schmidt. 1989. Attachment of Fibrobacter
223:217–223. succinogenes to cellulose and cellulose derivatives. Pages 64–65
Rasmussen, M. A., B. A. White, and R. B. Hespell. 1989. Improved in U.S. Dairy Forage Research Center, 1989 research summaries,
assay for quantitating adherence of ruminal bacteria to cellulose. USDA, Madison, WI.
Appl. Environ. Microbiol. 55:2089–2091. Wells, J. E., J. B. Russell, Y. Shi, and P. J. Weimer. 1995. Cellodextrin
Reddy, S.K.K., and M. Morrison. 1998. Biochemical and molecular efflux by the cellulolytic ruminal bacterium Fibrobacter succino-
characterization of adherence-defective mutants of Ruminococcus genes and its potential role in the growth of nonadherent bacteria.
albus strain 8. Page 132 in Proc. MIE Bioforum on Cellulose Appl. Environ. Microbiol. 61:1757–1762.
Degradation, Institut Pasteur, Paris, France. White, B. A., I.K.O. Cann, R. I. Mackie, and M. Morrison. 1997.
Roger, V., G. Fonty, S. Komisarczuk-Bondy, and P. Gouet. 1990. Cellulase and xylanase genes from ruminal bacteria: domain
Effects of physiochemical factors on the adhesion to cellulose analysis suggest a non-cellulosome-like model for organization
avicel of the ruminal bacteria Ruminococcus flavefaciens and Fi- of the cellulase complex. Pages 69–80 in Rumen Microbes and
brobacter succinogenes. Appl. Environ Microbiol. 56:3081–3087. Digestive Physiology in Ruminants. R. Onodera, H. Itabashi, K.
Ushida, H. Yano, and Y. Sasaki, eds. Japan Scientific Societies
Russell, J. B. 1985. Fermentation of cellodextrins by cellulolytic and
Press, Tokyo, Japan.
non-cellulolytic rumen bacteria. Appl. Environ. Microbiol.
Williams, A. G., and G. S. Coleman. 1997. The rumen protozoa. Pages
49:572–576.
73–139 in The Rumen Microbial Ecosystem. P. N. Hobson and
Stack, R. J., and R. E. Hungate. 1984. Effect of 3-phenylpropanoic
C. S. Stewart, Eds. Blackie Academic and Professional Publish-
acid on capsule and cellulases of Ruminococcus albus 8. Appl. ers, London.
Environ. Microbiol. 48:218–223. Williams, A. G., and N. H. Strachan. 1984. Polysaccharide degrading
Shi, Y., and P. J. Weimer. 1992. Response surface analysis of the enzymes in microbial populations from the liquid and solid frac-
effects of pH and dilution rayte on Ruminococcus flavefaciens FD- tions of bovine rumen digesta. Can. J. Anim. Sci. 64:58–59.
1 in cellulose-fed continuous culture. Appl. Environ. Microbiol. Wood, T. M., C. A. Wilson, and C. S. Stewart. 1982. Preparation of
58:2583–2591. cellulase from the cellulolytic anaerobic rumen bacterium Rumi-
Stewart, C. S., S. H. Duncan, and H. J. Flint. 1990. The properties nococcus albus and its release from the bacterial cell wall. Bio-
of forms of Ruminococcus flavefaciens which differ in their ability chem. J. 205:129–137.
to degrade cotton cellulose. FEMS Microbiol. Lett. 72:47–50. Yeung, M., and J. O. Cisar. 1990. Sequence homology between the
Stewart, C. S., H. J. Flint, and M. P. Bryant. 1997. The rumen bacte- subunits of two immunologically and functionally distinct types
ria. Pages 10–72 in The Rumen Microbial Ecosystem. P. N. Hob- of fimbriae of Actinomyces spp. J. Bacteriol. 172:2462–2468.

Review Adhesion

Uploaded by

Copyright:

Available Formats

Review Adhesion

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Review Adhesion

Uploaded by

Copyright:

Available Formats

J. Dairy Sci.

Invited Review: Adhesion Mechanisms of Rumen Cellulolytic Bacteria

ABSTRACT sesses at least two mechanisms for specific adhesion to

Journal of Dairy Science Vol. 84, No. 6, 2001

Journal of Dairy Science Vol. 84, No. 6, 2001

Journal of Dairy Science Vol. 84, No. 6, 2001

Journal of Dairy Science Vol. 84, No. 6, 2001

Journal of Dairy Science Vol. 84, No. 6, 2001

Journal of Dairy Science Vol. 84, No. 6, 2001

cohesin domain, which is present in a surface protein

Journal of Dairy Science Vol. 84, No. 6, 2001

Journal of Dairy Science Vol. 84, No. 6, 2001

Journal of Dairy Science Vol. 84, No. 6, 2001

Journal of Dairy Science Vol. 84, No. 6, 2001

Journal of Dairy Science Vol. 84, No. 6, 2001

Table 1. Summary of adhesion mechanisms in rumen cellulolytic bacteria.1

Fibrobacter Ruminococcus Ruminococcus

Glycocalyx Exopolysaccharides tion of fimbrial-like structures, whose interaction with

Journal of Dairy Science Vol. 84, No. 6, 2001

Journal of Dairy Science Vol. 84, No. 6, 2001

Journal of Dairy Science Vol. 84, No. 6, 2001

Journal of Dairy Science Vol. 84, No. 6, 2001

You might also like