Cellulase production by Rhizobium

MISSING IMAGE

Material Information

Title:
Cellulase production by Rhizobium
Physical Description:
x, 79 leaves : ill. ; 28 cm.
Language:
English
Creator:
Morales, Victor Manuel, 1949-
Publication Date:

Subjects

Subjects / Keywords:
Rhizobium   ( lcsh )
Cellulase   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1981.
Bibliography:
Includes bibliographical references (leaves 72-78).
Statement of Responsibility:
by Victor Manuel Morales.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000296530
notis - ABS2895
oclc - 08143828
System ID:
AA00003456:00001

Full Text












CELLULASE PRODUCTION BY Rhizobium


BY


VICTOR MANUEL MORALES























A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY



UNIVERSITY OF FLORIDA


1981


.~Cdi-- I -. a


































I dedicate this work

to my children

Susana and Victor.


c---Clc~ _I















ACKNOWLEDGEMENTS


The author would like to express his deep appreciation

to the chairman of his committee, Dr. David H. Hubbell for

his constant support, understanding and guidance. He would

like to thank his committee members Dr. Chesley B. Hall, Dr.

James F. Preston, Dr. Bob G. Volk and Dr. Edward M. Hoffmann

for their assistance and suggestions.

He would also like to thank Mr. Manuel Mesa and Mrs.

Louise Munro for their technical assistance and Garnet Jex,

Frederick Schipul, Anne Barkdoll, Ana Moliner and Karen

Kanudtsen for their friendship.

The author thanks the scientists listed in Table 1 who

provided the Rhizobium cultures used in this study.



















iii


h~~-l















TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS . .. .. ... iii

LIST OF TABLES. . ... ...... vi

LIST OF FIGURES . . vii

ABSTRACT . . ix

INTRODUCTION. . ... 1

LITERATURE REVIEW .. .. 2

MATERIALS AND METHODS . ... 10

Cellulase Activity. . .. 10
Screening for Culture Conditions Suitable for
Cellulase Production ... 13
Effect of Different Concentrations of Myo-inositol
or Potassium Gluconate on Cellulase Activity. 13
Induction of Cellulase Production by Different
Carbohydrates ... . 14
Effect of Nitrogen Source on Cellulase Production. 15
Effect of Plant Root Extracts on Cellulase
Production ... . 15
Effect of Mannitol on Cellulase Production ..... .16
Effect of the Culture Age on Cellulase Production. 17
Purification of Enzyme Activity..... 17
Effect of pH of Substrate on Cellulase Activity. 18
Effect of Calcium on Cellulase Activity. .. 19
Effect of Magnesium on Cellulase Activity. .19
Cellulase Production by Different Strains of
Rhizobium . . 19

RESULTS AND DISCUSSION. . . 21

Screening for Cultural Conditions for Cellulase
Production. . . 21
Effect of the Concentration of Myo-inositol or
Potassium Gluconate on Cellulase Production 21
Cellulase Activity Induced by Different
Carbohydrates . . 25


__I_ ~ 1_










Effect of the Nitrogen Source on Cellulase
Production . . 27
Effect of Root Extracts on Cellulase Production. 32
Effect of Mannitol on Cellulase Production ... 35
Effect of Age of Culture on Cellulase Production 35
Enzyme Purification. . . 42
Effect of pH on the Substrate on Cellulase
Activity. . . 46
Effect of Calcium on Cellulase Activity. .. 46
Effect of Magnesium on Cellulase Activity .. 52
Enzyme Characterization .. 52
Production of Cellulase by Different Rhizobium
trifolii Strains. . .. .56

CONCLUSIONS . . ... 60

APPENDIX I: VISCOSIMETRIC ASSAY FOR CELLULASE
ACTIVITY .. ... 62

APPENDIX II: GROWTH OF Rhizobium trifolii IN INO
MEDIUM . . 68

BIBLIOGRAPHY. . 72

BIOGRAPHICAL SKETCH . .. .. .. 79


C_~__ ~_















LIST OF TABLES

Table Page


1 Sources and characteristics of Rhizobium
strains used in the research .. .. 11

2 Cellulase production by Rhizobium growing
in different media. . ... 22

3 Growth of Rhizobium in two basal media with
different C sources . 23

4 Rhizobium trifolii strain BAL production of
cellulase with different carbohydrates and
N sources in the culture medium ... 28

5 Rhizobium sp. cowpeaa group" strain CIAT 79
production of cellulase with different car-
bohydrates and N sources in the culture
medium. . .. 29

6 Effect of the N source of the culture medium
on cellulase. production by Rhizobium trifolii
strains BAL and BART A. . ... 31

7 Effect of root extracts on cellulase produc-
tion by Rhizobium trifolii strains BAL and
BART A. . . .. 33

8 Effect of root extracts on cellulase produc-
tion by Rhizobium trifolii strains BAL and
BART A. . . .. 34

9 Effect of the concentration of mannitol on
the production of cellulase by Rhizobium
trifolii strain BAL . ... .36


I _















LIST OF FIGURES


Figure Page

1 Effect of the concentration of myo-inositol
in the culture medium on cellulase production
by Rhizobium. . ... 24

2 Effect of the concentration of potassium glu-
conate on cellulase production by Rhizobium 26

3 Cellulase activity of enzyme extracts of dif-
ferent Rhizobium strains. . ... 30

4 Cellulase activity per dry weight of bacteria
of cultures of Rhizobium trifolii strains BAL
and BART A at different times of culture incu-
bation. . ... ... 38

5 Cellulase activity per volume of enzyme extract
of cultures of Rhizobium trifolii strains BAL
and BART A at different times of culture incu-
bation. . .... .40

6 Elution pattern of Rhizobium trifolii strain
BAL cellulase after DEAE cellulose column
chromatography. . ... 45

7 Elution pattern of Rhizobium trifolii strain
BAL cellulase after desalting through a column
of BioGel P2. . ... 48

8 Effect of the pH of the substrate on enzymatic
activity of Rhizobium trifolii enzyme extract 50

9 Effect of CaC12 on cellulase activity of Rhizo-
bium trifolii enzyme extract. ... 51

10 Effect of MgSO4 on cellulase activity of Rhizo-
bium trifolii enzyme extract . 53

11 Comparison of plots of reduction of viscosity
and release of reducing groups by Rhizobium
trifolii enzyme extract . .... 55


vii











12 Production of cellulase by different strains
of Rhizobium trifolii . 58

13 Plot of the relationship between cellulase
amount and the slope of the reciprocal of
the viscosity vs time of reaction .. .65

14 Relationship between enzyme amount and time
to achieve a 50% reduction of viscosity of
a CMC solution. . ... 67

15 Growth and dry matter accumulation in Rhizo-
bium trifolii strains BAL and BART A cultures
in INO medium . ... 71



































viii















Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


CELLULASE PRODUCTION BY Rhizobium

By

Victor Manuel Morales

December 1981

Chairman: Dr. David H. Hubbell
Major Department: Soil Science

The production of cellulase by Rhizobium species was

studied. Rhizobium trifolii cellulase is induced by a variety

of polysaccharides, including celluloses and hemicelluloses.

Cellobiose and myo-inositol also induce enzyme production but

mannitol represses it at concentrations higher than 0.25%.

Cellulase production by "slow-growing, cowpea type"

rhizobia is more stringently regulated. Few substances induced

the production of cellulase in the strains studied of this

group of Rhizobium. Effective inducers were carboxymethyl-

cellulose, gulconate and myo-inositol.

Plant root substances were shown to stimulate cellulase

production by Rhizobium. Root substances soluble in phosphate

citric acid buffer stimulated cellulase production as measured

by the viscosimetric assay, while root solids increased the

enzymatic activity .as determined by the release of reducing










groups. This is an indication of a differential stimulation

of enzyme production by the plant root.

Cellulase production was very low under all conditions

tested. Most of the enzyme is loosely bound to the capsular

material and/or the cell surface. The enzyme in fast-growers

is an 1,4-8-D-glucan-4-glucanohydrolase (endo-glucanase EC

3.2.1.4) with specificity for high molecular weight oligosac-

charides. In slow-growers enzyme extracts show small decreases

of viscosity of carboxymethylcellulose solution with a rela-

tively large release of reducing groups.

Rhizobium cellulase loses activity under low ionic strength

conditions. Carbohydrate also seems to help stabilize the en-

zyme in raw extracts. The proteins, after separation from the

carbohydrates, are difficult to redissolve. Solubilization

is achieved when glycine, urea and small amounts of calcium

and magnesium are present in the solution.

There was no correlation between infectiveness of Rhizo-

bium trifolii strains and cellulase production. One strain,

which lacks the nodulation plasmid, proved to be capable of pro-

ducing cellulase at the same rate as its parental infective

strain.


i














INTRODUCTION


The infection of legume roots by Rhizobium is a complex

process which shows a high degree of specificity. The expres-

sion of this specificity occurs prior to the initiation of the

infection thread (Li and Hubbell, 1969). The binding of the

bacterium to the root hair surface through a lectin bridge has

been reported to play a role in the specificity of infection

(Dazzo et al., 1976). However, the biochemical mechanism by

which the infection thread is initiated is unknown. Cell wall-

degrading enzymes are likely to be involved since they are

important in cell wall-growth processes. Rhizobium produces

several polysaccharide degrading enzymes, notably pectinases

(Hubbell et al., 1978), hemicellulases and cellulases (Martinez-

Molina et al., 1979) when growing in artificial media. The

purpose of this research was to define the conditions which

induce the production of cellulases by Rhizobium, to purify

and characterize the enzymes, and to evaluate their possible

involvement in the infection process.


-1-















LITERATURE REVIEW


The Rhizobium-legume association is the symbiosis with

highest importance for agriculture. The bacteria, Rhizobium,

induce nodule formation in the roots of leguminous plants.

Inside these nodules the microbe fixes atmospheric nitrogen

(N2) to ammonia which is then used by the plant. Agronomic

exploitation of this plant-bacteria symbiosis is of great

importance for agriculture. It has been calculated that the

use of leguminous plants in agriculture provides an annual

input of fixed N approximately equal to that fixed industri-

ally for fertilizer (Delwiche, 1978).

Rhizobia are true bacteria belonging to the family

Rhizobiaceae; they are gram negative rods which can live sap-

rophytically in the soil and the rhizosphere of leguminous and

nonleguminous plants (Vincent, 1974). They are divided into

two main groups based on their growth rate on yeast extract

mannitol agar medium (YEMA). The fast-growers generally nod-

ulate temperate legumes and the slow-growers usually nodulate

tropical legumes. The difference in growth rate reflects true

biochemical and genetic differences (Graham, 1975; Martinez-de

Drets and Arias, 1972).

One important characteristic of the establishment of the

symbiosis is its specificity. Certain strains of rhizobia can
-2-










nodulate only certain species of leguminous plants. Based on

this, rhizobia have been classified in groups according to

which plants they can nodulate. This classification, although

useful for practical work, has been criticized because of its

lack of a biochemical and genetical basis and because of pro-

miscuous behavior of some strains (Graham, 1975).

The specificity of nodulation implies the existence of a

mechanism by which a mutual recognition of host and symbiont

occurs (Dazzo and Hubbell, 1981). In the case of a homologous

(nodulating) combination of plant and microbe, a complex in-

fection process occurs. Knowledge of this process of infec-

tion of legume roots has been recently reviewed (Bauer, 1981;

Bhuvaneswari, 1981; Dart, 1974, 1975, 1977; Dazzo and Hubbell,

1981; Solheim and Paxton, 1981).

Fast-growing Rhizobium, when proliferating in the rhizo-

sphere of a compatible host, inducesthe deformation of root

hair. "Branched," "moderately curled" and "markedly curled"

deformations have been defined as distinct categories.

"Markedly curled" root hairs have a characteristically twisted

tip known as the shepherd's crook. "Markedly curled" root

hairs appear only in homologous combinations (Yao and Vincent,

1976).

Rhizobia enclosed in the "shepherd's crook" trigger the

development of a tubular structure of plant origin called the

infection thread. The infection thread enveloping the bacteria

grows toward the base of the root hair and eventually, passing

from cell to cell, penetrates the root cortex. Rhizobia are







-4-


released from the infection thread into the cytoplasm of cer-

tain cortical cells. The bacteria are enveloped by plant

plasmamembrane, forming what is called "bacteriodal vesicles."

In these vesicles, rhizobia are transformed into bacteroids

which are responsible for N2-fixation in the nodule (Basset

et al., 1977; Paau et al., 1978; Robertson et al., 1978).

The recognition between host and microsymbiont occurs

before the initiation of the infection thread (Li and Hubbell,

1969). The most promising current hypothesis to explain the

specificity of infection is that host plant lectins interact

specifically with microbial cell surface carbohydrates attach-

ing the bacteria to the root hair surface (Dazzo et al., 1976).

In a more complete model, proposed by Graham (1981), the attach-

ment is one of the steps involved in recognition but not the

only one. Following attachment, there is a signal production

and reception sequence which induces responses by the symbionts.

and triggers the infection thread formation. The nature of

these signals is unknown, although cell surface polysaccharides

are implicated (Yao and Vincent, 1976).

Ljunggren and Fahraeus (1961) proposed that rhizobia

induce the production of polygalacturonases by the plant.

These enzymes would "soften" the root hair cell wall allowing

the Rhizobium to penetrate the plant cell and then trigger

the infection thread formation. The involvement of polygalac-

turonases in the infection process has been under discussion

with published results against (Lillich and Elkan, 1968;

McMillan and Cooke, 1969; Solheim and Raa, 1971) and favoring







-5-


(Palomares, 1975; Olivares et al., 1977; Verma and Zogbi, 1978).

Confirmation of the Ljunggren and Fahraeus hypothesis must

contend with the low levels of enzyme activity and the limits

of reliability of the assays.

It has been found, recently, that rhizobia not only pro-

duce small amounts of polygalacturonases (Hubbell et al., 1978),

but also produce cellulases and hemicellulases (Martinez-Molina

et al., 1979).

The involvement of polysaccharide-degrading enzymes in

the infection process is also suggested by electron microscope

data (Callaham, 1979; Chandler, 1978; Napoli and Hubbell, 1975).

Rhizobia enclosed in the shepherd's crook multiply and locally

disrupt the cell wall growth. A layer of new material is then

deposited at the infection site. This is apparently similar

to the defense mechanism observed at the point of entry of

microbial pathogens (Aist, 1977). Cell wall-degrading enzymes

are implicated as elicitors of defense responses in plants

(Hahn et al., 1980; Lyon and Albersheim, 1980). However, the

mechanism for infection thread formation must differ since

callose, the material usually deposited at the site of cell

wall injury, is not observed at the infection thread initia-

tion point (Callaham, 1979). The type and quantity of

polysaccharide-degrading enzymes may be this difference.

Plant polysaccharide-degrading enzymes are thought to play a

role in cell wall-growth processes (Roland and Vian, 1979)

and Rhizobium enzymes may only modify the regulation of the

cell wall-growth.







-6-


There have been many recent advances in the genetics of

Rhizobium in general and with respect to nodulation (Beringer,

1980, Beringer et al., 1980). Fast-growing Rhizobium strains

possess several large plasmids which may account for up to 10%

of the total genetic information (Casse et al., 1976; Prakash

et al., 1980). It has been established that the genes neces-

sary for nodulation reside in a plasmid (Zurkowski and Lorkie-

wicz, 1979). This nodulation plasmid has been denominated sym

plasmid. Transfer of this sym plasmid to a related microbe,

such as Agrobacterium tumefasciens, makes it capable of form-

ing root nodules in the proper host (Hooykaas et al., 1981).

Polysaccharide-degrading enzymes have been considered to

play an important role in many plant diseases (Bateman, 1964,

1976; Bateman and Millar, 1966; Byrde, 1979). Most phytopatho-

genic microorganisms are capable of producing polysaccharide-

degrading enzymes; this is a feature plant pathogens share

with members of the saprophytic flora and fauna of the soil.

It is thus apparent that the ability to produce polysaccharide-

degrading enzymes does not in itself enable a microorganism

to be a plant pathogen, but it may be an essential feature

for the pathogenic capabilities of certain pathogens (Bateman,

1976).

Cellulases are produced by bacteria, fungi, actinomycetes,

protozoa, insects and molluscs (Coughland and Folan, 1979;

Gooday, 1979). Not all cellulolytic microbes can degrade

crystalline cellulose but all of them can hydrolyse amorphous

or substituted celluloses. Reese et al. (1950) proposed two







-7-


types of cellulases, C1 and Cx. The enzymes designated C1

were supposed to be nonhydrolytic, but caused disaggregation

of the cellulose chains of the native material, thereby allow-

ing the hydrolysis of the products by the enzymes designated

Cx. The Cx enzymes hydrolyzed the chains made accessible by

the C1 cellulases. More recently, highly purified C1 material

has been isolated and demonstrated to be an exo-8-1,4 glucanase

(Nisizawa et al., 1972; Shikata and Nisizawa, 1975). Thus,

hydrolysis of native cellulose is thought to result from

synergistic action of endo- and exo-glucanases, which form

multienzyme complexes on the surface of substrate molecules

(Coughlan and Folan, 1979).

The enzyme componentsof a cellulase complex are: i) 1,4-

B-D-glucan 4-glucanohydrolases (endo-l,4-0-D-glucanases; EC

3.2.1.4), which preferentially hydrolyse internal glycosidic

linkages of cellulosic substrates; ii) 1,4-0-D-glucan cellobio-

hydrolases (exo-cellobiohydrolases; EC 3.2.1.91), which pref-

erentially cleave cellobiose units from the nonreducing end

of cellulose chains; and iii) $-glucosidases (cellobiase; EC

3.2.1.21), which cleave cellobiose units produced by the endo-

and exo-glucanases to give glucose (Coghlan and Folan, 1979).

The two most commonly used assays for Cx cellulases are

the measurement of the production of terminal reducing groups

or of glucose and the measurement of the decrease in viscosity

of a solution of carboxymethylcellulose (CMC). The viscosi-

metric method is much more sensitive for endo-glucanases,

since a random break in the cellulose chain may reduce







-8-


viscosity by 50% but causes almost no increase in reducing

groups (Ryu and Mandels, 1980).

Cellulase is an inducible enzyme system in most microbial

systems studied. Cellulose seems to be the best inducer for

the complete cellulase complex (Coughlan and Folan, 1979).

Other inducers are sophorose (Mandels et al., 1962) (2-0-0-

D-glucopyranosyl-a-D-glucose) and lactose (Ryu and Mandels,

1980). Cellulase biosynthesis is controlled by both catabolite

repression and by end-product inhibition (Montenecourt et al.,

1979; Reese, 1977).

The components of the cellulase system vary according to

the microorganism and the substrate on which it grows. The

number of enzymatic components for Cx enzymes range from 1 to

12 and for the C1 group from 1 to 3. They can be glycoproteins,

in which case the size of the oligosaccharide is another source

of variation. Endoglucanases differ in the degree of random-

ness of the splitting of glycosidic linkages and in the lim-

iting size of the oligosaccharide on which they can act. Some

have no activity on substrates smaller than 5 units (Reese,

1977). Many cellulases are able to catalyze the hydrolysis

of xylan (Capon and Thompson, 1979). Urbanek et al. (1978)

described a purified enzyme from Phoma hibernica capable of

hydrolysis of 8-1,4 glycosidic linkages formed by different

hexoses. It can degrade xylan, galactomannan, glucomannan

and galactoglucomannan.

Rhizobium trifolii strains are capable of synthesizing

cellulose (Napoli et al., 1975), small molecular weight 8-2







-9-


linked glucans (York et al., 1980; Zevenhuizen and Scholten-

Kaerselman, 1979) and acidic heteropolysaccharides containing

galactose, glucose, glucuronic acid and pyruvate (Robertson

et al., 1981). The structure of these heteropolysaccharides

are complex with branches and different types of linkages

such as a-1,4; 8-1,4; 8-1,6; 8-1,3 (Robertson et al., 1981;

Somme, 1974, 1980).

The role of microbial polysaccharides in nature is not

clear. They are thought to play a role in survival to desic-

cation, food storage, protection against viruses, clumping and

flocculation and adhesion to surfaces (Deinema and Zevenhuizen,

1971; Harris and Mitchell, 1973). The latter role seems to

be very important in pathogenesis and the establishment of

symbiotic relationships such as Rhizobium-legume root nodules

(Dazzo and Hubbell, 1981).

Rhizobium seem to be unable to hydrolyze their own extra-

cellular polysaccharide but the production of phage-induced

extracellular-polysaccharide depolymerases (Higashi and Abe,

1978).















MATERIALS AND METHODS


The strains of Rhizobium used in the study, their sources

and legume host are listed in Table 1.
The Rhizobium strains were kept on yeast extract mannitol

agar (YEMA) slants and/or in lyophilized form.
The infectivity of the strains was defined as the ability

to induce the formation of nodules in the roots of host plants

growing in Gibson's partially enclosed seedling assemblies

(Vincent, 1970).
Inocula for experiments were prepared by resuspending

the bacterial growth from petri plate cultures in sterile de-

ionized water. The bacterial population was counted using a

Petroff-Hauser chamber and the amount of the bacterial suspen-

sion to be used as inoculum calculated to obtain an initial

population of 1.5 x 107 cells/ml in the culture.

Cellulase Activity
Three methods were used for assaying enzymatic activity.

The plate-cup method as described by Dingle et al. (1953)

using carboxymethylcellulose (hereafter called CMC; type 7HF,

Hercules Inc., Wilmington, DE) in phosphate-citric acid buffer

(PCA buffer; 0.1M K2HPO4 and 0.1M citric acid mixed to pH 5.2)

as substrate and agar as solidifying agent. The plates were


10










-11-


0 C*
0- 0 m ) C0 *r4C
) N 0 V) cn 0) $ r.
U U N ) u 4 1 0 0
Sm4m u 0 $4 C --I .rq
0 0 u
A0 n Cd t4-4 0
0 m 0 -4

an a 3*
m rJr-4
0
4..
C %0


ai u



*rt 0)
Ua) I + I + + I + I + + + +
m (U f
4 1 CC

or,

$
4n P4


.0 4: CU .

0 s.4 4.)
N 0 O
*H *. 0 cd U
S-4 PI tn >4.
0 rq o 7 C
44 *j *
0




Ut 0 0
4-, I 0
Hi C o N 0 Co
4 cd O 1 0
Cd 4 0 ) I -4 -4 r-4 0

EC 0 % p '0 0 0
C0 PQ 0 D n LI
4)

0



S0 0 0
** 4 0' C-
0 a r4 CU
0 >d 41 3




4) .0 VI 4c
c ,o 6o ie


H 0 0 0 4
u N N N
i 4- *= 0 *
0 4) I-4 P1-4 -(-
SC) c In
6 e e2 .^







-12-


incubated for 48 hours at 350C and developed by flooding with

a 1 w/v solution of hexadecyltrimethylammonium bromide (East-

man Kodak Co., Rochester, NY). The enzymatic activity was

proportional to the diameter of the hydrolysis zone.

Cellulase activity was also measured by the decrease in

viscosity of a solution of CMC (0.2% w/v) as described by

Martinez-Molina et al. (1979) using Cannon-Fenske viscometers.

The third method used was the measurement of the increase

in reducing power at 300C of a reaction mixture consisting of

1 ml of CMC solution (0.2% w/v) in PCA buffer and 1 ml of en-

zyme extract. Reducing sugars were assayed by the method of

Somogyi-Nelson (Somogyi, 1952).

Enzymatic activity was expressed as enzyme units (EU)

per 100 mg of dry weight of bacteria or as EU per ml of enzyme

extract. Results of the viscosimetric assay were expressed as

EU (see Appendix I) or as percentage of decrease of viscosity

after 24 hours of incubation.

One enzyme unit (EU) was defined as the amount of enzyme

that reduces the viscosity of CMC solution (0.2% w/v) by 50%

in 10 hours or has a slope (plot of the reciprocal viscosity

against reaction time, hereafter called slope 1/ns) of 1.883

x 102 hour or produces an increase in reducing power in a

CMC solution equivalent to 25 ug of glucose in 24 hours.

Enzymatic activity was corrected for microbial growth

as explained in Appendix II.

Glucose was measured by the glucose oxidase method

(Sigma Chemical, St. Louis, MO).







-13-


Screening for Culture Conditions Suitable for
Cellulase Production

To screen for a medium in which cellulase production

is induced, three strains of Rhizobium (R. trifolii BAL, R.

japonicum J61 and Rhizobium sp. CIAT 79) were grown in two

different basal media, Bergersen's (Bergersen, 1961) and RDM

(Ronson and Primrose, 1979). Six different carbon sources

were tested with each basal medium: myo-inositol (0.4% w/v),

arabinose (0.5% w/v), potassium gluconate (0.5% w/v), sodium

succinate (0.5% w/v), sodium malate (0.5% w/v) and xylose

(0.5% w/v; all obtained from Sigma Chemical, St. Louis, MO).

Media were dispensed in screw cap test tubes in 10 ml amounts

and, after inoculation, incubated in an orbital shaker at

280C and 120 rpm. After 8 days of incubation, the cultures

were sonicated for 1 min at 60W (Model W140; Heat Systems-

Ultrasonics Inc., Plainview, NJ) and centrifugated at 15,300

x g for 15 min at 40C. Cellulase activity of the supernatants

was assayed by the plate-cup method.

Effect of Different Concentrations of Myo-inositol or
Potassium Gluconate on Cellulase Activity

In order to investigate the effect of the concentration

of myo-inositol and gluconate on cellulase production, two

strains of Rhizobium (BAL and CIAT 79) were grown in Berger-

sen's broth with different concentrations of these two C

sources (0.25, 0.5, 1.0, 1.5 and 2.0% w/v). The media were

dispensed in flasks in amounts of 100 ml, and inoculated and

incubated as previously described. After 8 days of incuba-

tion, the cells were precipitated by centrifugation at 3,550


I







-14-


x g for 1 hour at 40C. The pellet from each 100 ml of medium

was resuspended in 10 ml of PCA buffer and sonicated for 1

min at 60W. The broken cells were precipitated by centrifu-

gation at 15,300 x g for 15 min at 40C. The supernatant

(hereafter called EF enzyme extract) was tested for cellulase

activity by measuring the release of reducing groups.

Induction of Cellulase Production by Different Carbohydrates

Rhizobium trifolii strain BAL and Rhizobium sp. cowpeaa

group" strain CIAT 79 were grown in Bergersen's medium with

0.1% w/v mannitol and with 5 g/l of the following substances:

CMC, xylan, locust bean gum, gum arabic and cellobiose (all

obtained from Sigma Chemical Co., St. Louis, MO, with the

exception of CMC). Three nitrogen sources were used: NH4C1

(0.1% w/v), KNO3 (0.1% w/v) and sodium glutamate (0.11% w/v).

The media were inoculated and incubated as previously described.

After 8 days the cultures were harvested and EF enzyme extracts

prepared as described except that the cells were resuspended

in 3 ml of PCA buffer instead of 10 ml. The viscosimetric

assay was used for measuring cellulase activity.

In a second part of the experiment, Rhizobium trifolii

strain BAL and the Rhizobium sp. cowpeaa group" strains CB82

and CB1650 were grown in Bergersen's medium with 0.1% mannitol

and 5 g/l of CMC. The inoculation, cultural conditions and

enzyme extractions were performed as described above. Cellu-

lase was determined by viscosimetry.







-15-


Effect of Nitrogen Source on Cellulase Production

The effect of N source on cellulase production was

studied growing Rhizobium trifolii strains BAL and BART A in

Bergersen's medium with myo-inositol (2% m/v) as C source and

with each of two inorganic N sources (NH4C1 0.5% w/v and KNO3

0.5% w/v) with or without sodium glutamate (0.11% w/v). The

control treatment was glutamate alone as N source. The media

were inoculated .and incubated as described before. After 8

days of incubation EF enzyme extracts were prepared and cellu-

lase activity measured by the viscosimetric assay.


Effect of Plant Root Extracts on Cellulase Production

To study the promotion of cellulase production by plant

root substances and agar, Rhizobium trifolii strains BAL and

BART A were grown in the presence of root extracts or agar

(0.2% w/v, Difco agar, Difco Laboratories, Detroit, MI).

For the preparation of the root extracts, seeds of Trifoliu

repens var. Louisiana Nolin were surface sterilized and ger-

minated in sterile storage dishes (80 mm height x 100 mm

diameter). These dishes had stainless steel mesh as support

and Fahraeus' minus N nutrient solution (Fahraeus, 1957).

The seeds were germinated at room temperature under fluores-

cent lights. When the seedlings were 10 days old, the roots

were cut with a razor blade and macerated in a mortar in PCA

buffer over ice. The macerate was centrifuged at 15,300 x g

for 15 min at 40C. The supernatant (hereafter called root

solubles) was sterilized by filtration using a 0.4 Um filter

(Metricel, Gelman Scientifics, Ann Arbor, MI) and stored at







-16-


40C. The pellet was resuspended in deionized water and fro-

zen, thawed and centrifugated to wash soluble compounds.

The procedure was repeated twice. After the last centrifu-

gation, the pellet was resuspended in PCA buffer, autoclaved

and stored at 40C. The basal medium used was Bergersen's

with myo-inositol (2% w/v) as C source (hereafter called INO

medium; pH was adjusted to 7.0 with a solution of NaH2PO4

20% w/v) which received root solubles, root solids suspension

or PCA buffer (5 ml in 100 ml of basal medium) or agar (0.2%

w/v). The agar was added to the medium before autoclaving.

All the others were filter sterilized and added to the medium

after autoclaving. The media were inoculated and incubated

as described previously. After 8 days of incubation, the

cells were harvested by centrifugation and an EF enzyme ex-

tract prepared and tested for cellulase activity by both

viscosimetry and release of reducing groups.

Effect of Mannitol on Cellulase Production

Mannitol is a carbohydrate commonly used in media for

Rhizobium growth. Since readily available sugars usually

repress the production of polysaccharide-degrading enzymes

in microorganisms, the effect of the presence of mannitol

in the culture medium was studied. Rhizobium trifolii strain

BAL was grown in Bergersen's medium with different concentra-

tions of mannitol (0.25, 0.50, 1.0, 1.5 and 2.0% m/v) with

or without myo-inositol (1% w/v). The media were inoculated

and incubated as described. After 8 days of incubation, an







-17-


EF enzyme extract was prepared and cellulase activity tested

by the release of reducing groups.

Effect of the Culture Age on Cellulase Production

To study the production of cellulase with respect to

time of culture incubation, Rhizobium trifolii strains BAL

and BART A were grown in INO medium with or without agar

(0.01% w/v). The medium was inoculated and incubated as des-

cribed. At every sampling time (0, 1, 2, 4, 6 and 8 days),

2 flasks of each strain were randomly selected and an EF

enzyme extract prepared from 70 ml of culture medium from

each flask. Enzymatic activity was determined by viscosimetry.

Purification of Enzyme Activity

Cells from 6.35 liters of Rhizobium trifolii strain BAL

culture (4 days) in INO medium were collected by centrifuga-

tion at 3,550 x g for 1 hour and resuspended in 120 ml of

glycine-urea buffer solution (GU buffer solution composed of

0.2M K2HPO4 in 1% w/v glycine, 1M urea and 0.1M citric acid

in 1% glycine and 1M urea mixed to give a pH of 7.5). Chlo-

ramphenicol and dithiothreitol (Sigma Chemical Co., St. Louis,

MO) were added to a concentration of 100 ug/ml and ImM, re-

spectively. The cell suspension was sonicated twice for 1

min at 60W and centrifuged at 15,300 x g for 15 min at 4*C.

The protein of the supernatant was fractionated by

(NH4)2SO4 precipitation (Schwarz/Mann, Orangeburg, NY) in

20% saturation steps up to 80% saturation. Precipitation







-18-


was carried out adding solid salt in small quantities to the

enzyme extract at 40C under constant stirring.

The precipitates from each degree of saturation were re-

dissolved and dialyzed in GU buffer solution with 0.05 ug/ml

of CaC12*2H20 for 5 hours in dialysis membranes with a molec-

ular weight cutoff of 1,000 daltons (Spectra/Por 6, Spectrum

Medical Industries, Los Angeles, CA) and at 40C. The frac-

tions which precipitated at 20%, 40% and 60% saturation,

contained cellulase activity and were pooled. The pooled

material was placed in a column (length 10.5 cm x diameter

5 cm) of DEAE cellulose (DE 52 preswollen and provided by

the manufacturer in the free base form, Whatman, Clifton, NJ)

stabilized with GU buffer solution. The column was eluted

with a linear gradient of NaCI and pH at a flow rate of

0.96 ml/min. The gradient was from OM NaCI pH 7.5 to 1M

NaCI pH 4.8. Fractions of 4.6 ml were collected and assayed

for cellulase activity by viscosimetry. The fractions which

showed activity were pooled and the protein precipitated

with (NH4)2SO4 to 80% saturation at 40C. The precipitate

(hereafter called DEAE semipurified cellulase) was resolu-

bilized in 10 ml of GU buffer solution and desalted by pass-

ing it through a column of BioGel P2 (Bio-Rad Laboratories,

Richmond, CA; 12 x 2.5 cm) stabilized with GU buffer solution.


Effect of pH of Substrate on Cellulase Activity

Solutions of CMC with different pH values (pH 5, 6, 7

and 8) were prepared by addition of 5N NaOH. The viscosimetric







-19-


method was used to measure cellulase activity adding 0.5 ml

of a 1:5 dilution of the DEAE semipurified cellulase to 5 ml

of the CMC solution.

The experiment was also performed using Rhizobium

trifolii strain BAL EF enzyme extract.


Effect of Calcium on Cellulase Activity

The effect of calcium was studied adding 0.1 ml of solu-

tion of CaC12*2H20 at different concentrations (the final

concentrations were 0, 0.46, 2.3, 4.5, 9.1 and 18.1 x 10-4M)

or of solution of ethylenediaminetetraacetic acid (EDTA) to

0.9 ml of EF enzyme extract. The cellulase activity was

measured by the viscosimetric assay.


Effect of Magnesium on Cellulase Activity

The effect of magnesium on the Rhizobium cellulase activ-

ity was studied as previously described for calcium. Solu-;

tions with different concentrations of MgSO4*7H20 were used.

The final concentrations of MgSO4 were 0, 1.25, 2.5, 5, 10

and 20 ug/ml. The cellulase activity was measured by the

viscosimetric assay.


Cellulase Production by Different Strains of Rhizobium

Several strains of Rhizobium trifolii (BAL, BART A, 0403,

0435, 0435-2, 521 and 521nod-8) with different infectiveness

on Trifolium repens were grown in INO medium. Inoculation

and incubation were performed as described previously. After

8 days of incubation, the cells were harvested by centrifugation








-20-


and an EF enzyme extract prepared as described. Cellulase

activity was measured by viscosimetry.















RESULTS AND DISCUSSION


Screening for Cultural Conditions for Cellulase Production

Table 2 shows the results of cellulase production by

Rhizobium in different cultural conditions. The fast-growing

Rhizobium trifolii strain BAL was more fastidious with re-

spect to the C source for growth (Table 3). Bergersen's me-

dium with myo-inositol was the best medium for cellulase

production for this strain. The slow-growing cowpeaa type"

strain CIAT 79 grew well on most carbohydrates but showed

cellulase activity only with xylose and gluconate, while

Rhizobium japonicum strain J61 did not produce detectable

cellulase activity in any of the media used. It is apparent

that cellulase production is triggered by different inducers

in different Rhizobium species.


Effect of the Concentration of Myo-inositol or Potassium
Gluconate on Cellulase Production

Figure 1 shows cellulase levels in cells grown at dif-

ferent concentrations of myo-inositol. The cellulolytic

activity increases dramatically at concentrations higher

than 0.25% but there is not too much difference above 0.5%.

Enzymatic activity was higher at a concentration of 2% than

at 1%, albeit nonsignificantly, in spite of the fact that


-21-







-22-


TABLE 2: Cellulase production by Rhizobium growing in
different media.


Basal medium
Bergersen's RDM
C Source Strains Strains
CIAT CIAT
BAL 79 J61 BAL 79 J61

Gluconate ++ -

Xylose +++ +++

Succinate -

Arabinose + +

Myo-inositol +++ -

Malate -


Enzymatic activity from the cup-plate assay:
no enzymatic activity
+ traces of activity
++ light enzymatic activity
+++ clear enzymatic activity







-23-


TABLE 3: Growth of Rhizobium in two basal media with
different C sources.


Basal medium
Bergersen's RDM
C Source Strains Strains
CIAT CIAT
BAL 79 J61 BAL 79 J61

Gluconate 3 4 3 1 4 4

Xylose 0 5 1 0 4 0

Succinate 0 2 2 0 3 3

Arabinose 1 5 4 0 4 1

Myo-inositol 5 4 2 5 1 1

Malate 0 4 3 0 4 3


Scale of microbial-growth:


0 no growth to
5 prolific growth.







-24-


CO



b1.0




IUR
L R
r-
















FIGURE 1:


Strains:

BAL

CIAT79 '~


0.25 0.5 1.0 1.5 2.0
Myo-INOSITOL CONCENTRATION (/o)


Effect of the concentration of myo-inositol in
the culture medium on cellulase production by
Rhizobium. Enzymatic activity was assayed by
the increase of reducing groups in a CMC solution.
For enzyme unit definition see Materials and
Methods section. Vertical bars represent stan-
dard deviations.







-25-


Rhizobium trifolii grows faster and has a larger number of

viable rhizobia at 1% myo-inositol than at 2% at harvest time.

The reason may be that at 2% concentration of cyclitol there

is a higher rate of death and lysis when the culture reaches

the stationary phase, releasing the enzyme from the cells.

"Cowpea type" Rhizobium CIAT 79, which did not show de-

tectable activity with myo-inositol by the cup-plate assay,

had a higher activity than BAL when measured by the release

of reducing groups (Figure 1). Enzyme extracts of this

strain usually give small decreases in viscosity of CMC solu-

tion (Table 5). This might indicate the presence of an exo-

glucanase in the preparation. Unfortunately, CIAT 79 cultures

have high levels of reducing sugars which are probably the

cause of the high variation observed in the results of the

Somogyi-Nelson assay.

Gluconate does not induce cellulase production in Rhizo-

bium trifolii strain BAL (Figure 2) but it does with cowpeaa

type" rhizobia CIAT 79. Cellulase activity in CIAT 79 cul-

tures with gluconate was more than twice that found in cul-

tures with myo-inositol of the same strain as measured by

the reducing groups assay. It seems, therefore, that cellu-

lase is induced by different substances in fast- and slow-

growing Rhizobium.


Cellulase Activity Induced by Different Carbohydrates

Rhizobium trifolii strain BAL produced cellulase equally

.with all the carbohydrates tested with the exception of xylan






-26-


0.25 0.5 1.0 1.5 2.0
CONCENTRATION OF GLUCONATE (%)


Strain: BAL
BAL


FIGURE 2:


Effect of the concentration of potassium gluconate
on cellulase production by Rhizobium. Enzymatic
activity was determined by the increase in reduc-
ing groups in a CMC solution. For enzyme units
definition see Materials and Methods section.
Vertical bars represent standard deviations.


5
5.





LJ



I-
-3






LU

5=1


CIAT79







-27-


when NH4C1 was the N source (Table 4). There was no differ-

ence between inducers in enzymatic activity or in growth.

Rhizobium sp. cowpeaa type" CIAT 79 did not show signif-

icant cellulase production with most inducers tested using

the viscosimetric assay (Table 5). Cellulase was detected in

enzyme extracts of this strain only when growing with cello-

biose and KNO3 as N source. However, this is not a general

feature of all cowpeaa type" strains. Cultures of Rhizobium

sp. strain CB82 showed considerable cellulase activity when

growing with CMC and low levels of mannitol, while CB1650 is

similar to CIAT 79 with respect to viscosity reduction (Fig-

ure 3). It seems that slow-growing Rhizobium strains differ

in ability to produce cellulase. This is not surprising

since cowpeaa group" is very heterogeneous with differences

in range of host plants, growth on artificial media and en-

vironmental conditions where they are found.


Effect of the Nitrogen Source on Cellulase Production

Cellulase production by Rhizobium trifolii strains BAL

and BART A growing with different N sources is shown in Table

6. BAL showed the highest enzyme production per 100 mg of

dry bacteria with NH4C1 regardless of the presence of glutam-

ate, but the yield of the enzyme per ml of enzyme extract was

20 times lower than with glutamate alone. This probably was

an effect of the poor growth displayed by BAL with mineral

salts (data not shown). BART A, by contrast, grew more slow-

ly with mineral N than with glutamate alone but produced the







-28-


TABLE 4: Rhizobium trifolii strain BAL production of
cellulase with different carbohydrates and N
sources in the culture medium.


Carbohydrate N source
NH4Cl KNO3 Na glutamate

CMC 82 86 73

Xylan 0 76 63

Locust bean gum 73 78 62

Gum arabic 70 69 69

Cellobiose 78 79 47


NOTE: The control treatment consisted of the mixture
of 1 ml PCA buffer and 5 ml of CMC solution,
and incubated as described. Control showed
a 2% decrease in viscosity after 24 hours.

Percentage of viscosity decrease after 24 hours of
incubation. Average of two replications.







-29-


TABLE 5: Rhizobium sp. cowpeaa group" strain CIAT 79
production of cellulase with different carbo-
hydrates and N sources in the culture medium.


Carbohydrate N source
NH4CI KNO3 Na glutamate

CMC 11 4 4

Xylan 3 7 6

Locust bean gum 9 3 4

Gum arabic 5 4 5

Cellobiose 2 20 8


NOTE: For the control treatment 1 ml of PCA buffer
was mixed with 5 ml of CMC solution and incu-
bated as described. Control showed a 3% de-
crease in viscosity in 24 hours.

Percentage of viscosity decrease after 24 hours of
incubation. Average of two replications.







-30-


SoI




S20 STRAINS:
Ca
44 R.trifolii BAL -
-- *
" R.sp. CB82 o -0 -..--
. R.sp. CB1650 A
S\ CONTROL

S40 %


CZ


=

60 "' ----...




0 5 10 15 20
TIME (hours)




FIGURE 3: Cellulase activity of enzyme extracts of different
Rhizobium strains. The strains were grown for 8
days, in Bergersen's basal medium with 5 g/l of CMC
and 0.1% w/v myo-inositol. Each point is the average
of 2 replications. Boiled BAL enzyme extract was
used as control treatment.







-31-


TABLE 6: Effect of the N source of the culture medium
on cellulase production by Rhizobium trifolii
strains BAL and BART A.


Strain
BAL BART A
N Source Enzyme Units
Per Per** Per Per
volume mass volume mass

NH4Cl with glutamate 0.06 47 2.00 190

NH4C1 without glutamate 0.07 160 0.00 0


KNO3 with glutamate 0.00 0 1.84 131

KNO3 without glutamate 0.00 0 0.00 0


Glutamate only 1.22 28 1.54 118


NOTE: All cultures were processed to obtain the same
volume of enzyme extract. Cellulase was assayed
by the viscosimetric method. Each value is the
average of two replications. For definition of
enzyme unit see the Materials and Methods section.

Enzymatic activity expressed as EU/ml of enzyme extract.
Enzymatic activity expressed as EU/100 mg of dry bacteria.







-32-


same amount of enzyme when growing with glutamate with or

without mineral N salts. In this strain, then, enzyme pro-

duction was directly proportional to the growth.


Effect of Root Extracts on Cellulase Production

The root solids solution showed a protein content of

32 ug/ml (by the Lowry assay) and a total carbohydrate con-

tent of 71 Ug/ml (by the phenolsulfuric acid method).

Root extracts seem to stimulate the production of cellu-

lase by strain BAL (Table 7). Root solids showed a striking

stimulation of enzyme production as measured by the release

of reducing groups. However, the same stimulation was not

observed in.the results of the viscosimetric assay (Table 8).

In BAL cultures with root solids, cells tended to adhere to

the root pieces and did not grow as well as in other treat-

ments. No. effect was shown by the addition of root extracts

on BART A growth or cellulase production as measured by vis-

cosimetry, but the release of reducing groups was strongly

stimulated by root solids and, to a lesser extent, by agar.

The difference in results obtained with these two assays may

indicate that a different set of enzymes is produced by dif-

ferent treatments. Enzyme extracts from cultures with root

solids and agar showed delayed or no pigment production (see

Appendix II for discussion on pigment production) and also

low content of reducing sugars.


I







-33-


TABLE 7: Effect of root extracts on cellulase production
by Rhizobium trifolii strains BAL and BART A.


Strain
BAL BART A
Culture amendment Enzyme units
Per Per** Per Per
volume mass volume mass

Root solubles 1.9 29.6 1.8 85.9

Root solids 2.9 161.8 12.0 513.2

Agar 0.7 12.2 4.7 227.1

No amendment 1.0 16.0 2.3 111.3


NOTE: Enzymatic activity was measured by the release
of reducing groups assay. Each value is the
average of 4 replications. For definition of
enzyme unit see the Materials and Methods section.

Enzymatic activity expressed as EU/ml of enzyme extract.
Enzymatic activity expressed as EU/100 mg of dry
bacteria.








-34-


TABLE 8: Effect of root extracts on cellulase production
by Rhizobium trifolii strains BAL and BART A.


Strain
BAL BART A
Culture amendment Enzyme units
Per Per** Per Per
volume mass volume mass

Root solubles 1.1 16.4 1.7 79.7

Root solids 0.3 15.4 1.2 52.9

Agar 0.4 6.6 1.2 58.1

No amendment 0.4 6.0 1.4 68.6


NOTE: Enzymatic activity was assayed by the visco-
simetric method. Each value is the average
of 2 replications. For definition of enzyme
unit see the Materials and Methods section.

Enzymatic activity expressed as EU/ml of enzyme extract.
Enzymatic activity expressed as EU/100 mg of dry
bacteria.


_1 _~







-35-


Effect of Mannitol on Cellulase Production

Mannitol in the culture medium inhibits the production

of cellulase (Table 9). In the presence of myo-inositol

there was more cellulase produced at a concentration of man-

nitol of 0.25% than at higher concentrations. When mannitol

is the only C source, cellulase production drops steadily as

mannitol concentration increases. Above 1.5% mannitol there

is only a very small enzymatic activity. The high enzyme con-

centration when bacteria were grown with mannitol as the only

C source is a result of the correction for bacterial growth;

actually the enzyme concentration per ml of enzyme extract

in these treatments is low. The results indicate that cellu-

lase production is controlled. However, no reason seems ap-

parent for myo-inositol to be an inducer or mannitol to be a

repressor; both substances are readily metabolized by the

microbe. It may be that Rhizobium once possessed the capa-

bility not only of synthesizing but of degrading and utiliz-

ing cellulose for growth, and they partially lost the capa-

bility for degradation. Myo-inositol, being commonly present

in plant tissues (Loewus and Loewus, 1980), could act as an

inducer of production of polysaccharide-degrading enzymes.

But, it would be uncommon that a microorganism, in the highly

competitive soil environment, would retain genes which do

not contribute to its ability to survive.


Effect of Age of Culture on Cellulase Production

Figures 4 and 5 show cellulase production with respect

to age of culture. The enzymatic activity is expressed in


- -







-36-


TABLE 9:


Effect of the concentration of mannitol on the
production of cellulase by Rhizobium trifolii
strain BAL.


Mannitol Myo-inositol Enzyme units (EU)
concen- concen- Per Standard Per Standard
traction traction volume deviation mass deviation
-%- -I-

0.25 1 2.9 0.5 36.2 6.2
0.50 1 1.2 0.8 15.0 10.0
1.00 1 1.2 0.3 15.0 3.7
1.50 1 1.1 0.6 14.1 7.7
2.00 1 0.8 0.3 10.2 3.8
0.25 0 1.0 0.3 40.0 12.0
0.50 0 0.8 0.3 25.8 9.7
1.00 0 0.6 0.3 16.0 8.0
1.50 0 0.0 0.2 0.0 6.8
2.00 0 0.3 0.3 2.7 7.7


NOTE: Enzymatic activity was measured by
of reducing groups. Each value is
of 4 replications. For definition
unit see the Materials and Methods


the release
the average
of enzyme
section.


Enzymatic activity expressed as EU/ml of enzyme extract.
Enzymatic activity expressed as EU/100 mg of dry
bacteria.






























.H.i< U n

a) $ .0 CO
41 0
4-* S90 1r73 t rt
cd *iH 4
cd -4 cd
14. 4-) r-4 v) > 4
U a) 4-)
4*(-r4 n cd
IC *-I 4-H
Mr- 0 Mo


*-4 R 1 04


4- 0 H 04 O


0. c u -*H
u I .,.o
N 0 0 .-




SU *N,4 *
(,CM rtC


1 < cd N +-






r :4 *w H 4)



U 0 C 4-


__




















E
I


~\\\\\\\\`\\P~\~\\\\\\~\~\\~\\\\\

I


R\\\\\\\\m\\~\~\~\~P~\~\~\\\\\~

I


-38-


LUiK
=.
_ w 2


bO





I-
'5=


's





a c


d0







-2
d


I


S0 0 u 0 0

VIU313VG AVG 10 ZWO0I /SiINO 3WAZN3


~\\\~\\~\\\\~\~\\~\~\\\\\\~\\\\n\\~n\\\\


mnmmamml\\\\W\\\\\\\\ W



































S0 r4
> H 4-)C

N -i-4 (t
,-4 4-I CdU



0o 0
*H "0 M
0)0 4-1 e -r




0 L *,
0, 4-C3 U




> 1.c U 0





UU 4 ..4-
'4-< *0
n0 C r-
f0 E-0 0










41 z4 U)r-i
S


aC> +0 cVl )


P41 O
U 0 tn 04








-40-


E \\~\\\\~\\\,\~\\\\\,\~\\\~\~\\\\\\\\\\\





IS\\~\\\`\\\\\\ \\\\\\\\\\\\\ \\\'
I_
________ Pr^.^^^^ <%^
c^^^ ^m ^^^


.m i
--
F3-


::u
(U: (U C


...







cc
ba
cc


-J


I 0

I lZ

13V1VX3 3NIAZN3JO lu/SlINn 3NAZN3


(U
0.





C.3


c.t





CI'
---
&a










LL-
L,

-J


-ie

I,-


\\~P~n~. \\7\\7\\\\\ \


r-


I\\\\\\\\\\\\\\\\\\\\\\







-41-


Figure 4 on the basis of dry weight of bacteria while Figure

5 shows the same results expressed per ml of enzyme extract.

The two strains differ with respect to cellulase production.

BART A produces cellulase earlier than BAL. BART A cultures

showed traces of activity after 1 day of incubation while

BAL cultures did so after 2 days of incubation.

The activity expressed on a basis of dry weight of bac-

teria may be misleading, especially with young cultures. A

trace of activity in cultures with very small dry weight of

bacteria per ml will give a high activity per mg of dry bac-

teria.

The peak of production of enzyme as expressed in Figure

4 occurs after 2 days of incubation for BART A and after 6

days for BAL. At these ages the cultures are in early sta-

tionary phase (see Appendix II). The highest rate of extract-

able enzyme occurs after 4 days of incubation with the excep-

tion of BART A growing with agar (Figure 5). Agar seems to

depress the production of cellulase, but this could be due

to the higher viscosity of the cultures with agar which could

make sonication less efficient in releasing the enzyme, or the

enzyme may bind to the agar and not be extracted. All Rhizo-

bium trifolii strains tested release a yellow pigment when

growing in INO medium after reaching early stationary phase

(unpublished results). Cultures of BART A with agar start

producing the pigment one day before cultures without agar.

In contrast, pigment production is delayed in BAL cultures

with agar. BAL cultures in INO medium show clumping. The







-42-


clumps are very numerous, of uniform size and shape. They

are not affected by divalent cations, EDTA, detergents, trypsin,

pectinase or cellulase. Only sonication disruptsthem. BAL

cultures with agar do not produce this clumping.

Cellulase production increases with time of incubation,

reaches a peak and declines, with the exception of BART A in

medium with agar, which showed a continuous upward trend.


Enzyme Purification

Most of the cellulase produced by Rhizobium is loosely

bound to the.cell surface and/or the capsular material. There

is also a small detectable activity in the culture supernatant

(ranging from 0.05 to 0.2 enzyme units/ml). This low enzyme

concentration in such a large volume accounts for a good pro-

portion of the total enzyme produced. But, it is very diffi-

cult to separate the enzyme from the extracellular polysac-

charide. Of the total amount of enzyme (as measured by the

viscosimetric assay) that remains with the cells after centrif-

ugation, 40% is released into the buffer solution after soni-

cation, 20% can be released after repetition of the extraction

procedure and 40% remains attached to the cells.

The type of extracting solution used influences the amount

of enzyme released. The highest recovery was obtained with

deionized water. PCA and GU buffers extract around 60% of

the amount extracted with water. Tris HC1 buffer (pH 8.0)

releases only 20% of this amount. Enzyme extracts made up

with water maintain the enzymatic activity for several weeks







-43-


when stored at 40C or for 2 or 3 months when sterilized by

filtration (polycarbonate filters, 0.4 um, Gelman Industries,

Ann Arbor,MI) and stored at 41C.

When ammonium sulfate precipitation is performed slowly

(allowing the solution to stand for at least 5 hours between

steps), the fractions with 20%, 40% and 60% saturation show

cellulase activity after dialysis. When a quick precipita-

tion is performed, the fraction with 80% saturation also

shows activity. GU buffer solution was essential for main-

taining enzymatic activity. Without glycine and urea the

precipitate did not dissolve completely and usually part of

it reprecipitated after a few hours. A small amount of CaC12

2H20 (0.05-0.1 g/ml) helped to solubilize the protein. Dial-

ysis time should be short (4-5 hours; 10 hours maximum).

Extensive dialysis resulted in complete loss of cellulase

activity. This may be due to binding of the enzyme to the

cellulosic dialysis membranes or denaturation. Several pro-

cedures were tested to recover enzyme from the surface of the

dialysis membranes without success. The affinity of the en-

zyme for cellulosic material seems to increase as the ionic

strength of the solution decreases.

Figure 6 shows the elution pattern of proteins from a

DEAE cellulose column. Cellulase activity appeared in frac-

tions eluting immediately before the main protein peak. Elu-

tion should be performed with a gradient of salt and pH. A

salt gradient alone does not release the protein from the

column.































r_
*Mr1 I
( U



*H 0 *
OIU >n -
oo 4




1-4 u I r-

0 4) 4.J
H l U *






6 r-1 (0 0
0 r 4t

0N q.-4 ( 0
OCp y






-4 C U 0

4- 3 0O t04


-4 PO
<' r-l *

r-4 < X r-I
W PQ U 0 4)-




I-W
*p*
0




en*
o4







-45-


(AI!JBIeu) NOIIVUIN33NO3 13ON

U- I5


U -S-j~g~g-


i
I
I i


i i
I
T i



cl

C-3

O uc:



r6 C2


lUU 08Z e 33NV8aHOSaV







-46-


Desalting after ammonium sulfate precipitation using

a BioGel P2 column gave two peaks (Figure 7). The first peak

showed low cellulase activity. This enzymatic activity dis-

appeared in less than 24 hours, indicating that the enzyme

is denatured under these conditions. Mixing the proteins

from the two peaks did not restore cellulase activity.


Effect of pH of the Substrate on Cellulase Activity

Rhizobium cellulase increases activity with an increase

in pH (Figure 8). The highest activity was found at pH 8.0,

which was the highest pH tested. However, the increase was

mainly on the initial rate of activity and was relatively

small.

Assays with DEAE semipurified enzyme did not show any

effect of pH on enzymatic activity. This was probably due

to the presence of urease in the preparation. Urease activ-

ity, although small, increased the pH during incubation.


Effect of Calcium on Cellulase Activity

Calcium did not show effect on the activity of Rhizobium

cellulase. Although the enzymatic activity was higher when

calcium was present, the differences were not significant

(Figure 9). There was no difference either when EDTA was

added to the CMC solution or when the CaC12*2H20 concentra-

tion was increased to 4.5 x 10-2 M.
tion was increased to 4.5 x i0 M.































ca UO
4- u W
cCd
(-4


0 0


4J >



4-I C$U
0 t 0*




-) cP H
O *1
cS 3 Ot
$-4 r-4 0
O4) V- 0

Scd >



M 3 r0 .4
4->) CO U
*i- U -l
*P PQ U






-48-


(%) eseoJ3ep AI!So3S!A


/~,~


CS

0
0l ^
3
-L


wu o08 ie 33NV8HOS8V


La

-J




I-

m ~I




























U

U 1U




0 4-


N

4 0)




IH
. eU
.0 0 0





4) 41,
Srt










o 1
3 0)
3e 441










4) 4 *1




0 tn u
W *P
MMU Ut

C O








0:4







-50-


/ li

a I 0i -
C=


'* I /I
- I ca
o I -i

jS y / l -
I-1-
U) 3
i/ //i _T
I I / / I



,I / -I
I i I / =






/ // /



oa
C Io




( 04 -/ *I-



V- CI C4
IS03SIA SV333








(%) A11SO3SIA NI 3SV3li330








-51-


I-


1.1
UIJ
S
crs


LU
*j 0.9




LU
z
= 0.1
w
w
=E

LU9


cm


CaCI2 CONCENTRATION ( x10-4M)














FIGURE 9: Effect of CaCI2 on cellulase activity of Rhizo-
bium trifolii enzyme extract. Enzymatic activity
was assayed by the viscosimetric method. Each
value is the average of 2 replications.







-52-


Effect of Magnesium on Cellulase Activity

Figure 10 shows the cellulase activity of Rhizobium

trifolii strain BAL enzyme extract. There is not a clear

trend in the changes of cellulase activities with different

concentrations of magnesium. Magnesium did not affect enzy-

matic activity either alone or mixed with calcium (data not

shown).


Enzyme Characterization

Figure 11 shows the plot of decrease in viscosity of a

solution of CMC and the release of reducing groups caused by

a Rhizobium trifolii strain BAL enzyme extract. There was a

slow release of reducing groups. Eight hours of incubation

were required to liberate the equivalent of 0.11 mmoles of

glucose during which the viscosity of the solution decreased

by 29%. This indicates that the enzyme is an 1,4-B-D-glucan

4-glucanohydrolase (endo-l,4-B-D-glucanase). CMC is degraded

to oligosaccharides rather than to glucose or cellobiose.

Glucose production was not detected by the glucose oxidase

assay upon incubation with enzyme extract. Rhizobium trifolii

is able to grow on cellobiose as sole C source and, therefore,

must be able to produce cellobiase (B-glucosidase); however,

it is unable to grow on CMC, indicating that no cellobiose

was formed.

The relatively low activities observed and the difficulty

of increasing it by purification may indicate that CMC is not

the natural substrate for the enzyme. Enzymatic activities







-53-


0.9


~ ~ a nr -c
CMSO4 CONCENTRATION ( /m)
M9S04 CONCENTRATION (,ug/ml)


FIGURE 10: Effect of MgSO4 on cellulase activity of Rhizo-
bium trifolii enzyme extract. Enzymatic activity
was assayed by the viscosimetric method. Each
value is the average of 2 replications.



























h fo
4-> I0 0
0 V-O CO


U >sU (0
W 0 +"H P4
-rl N -r4 : 3

0t 0 0
0 M

.0 t> 0
0 44 -l
*- i .0 0 U
+j04au 01 0
u 3 Cd a)'
30 ~M Cb4 0
) 60 4- cd4
004) > 0 U
0 *4l *0 od-
U M 0 cn p,





O RO OO
Sr-i 4a)
c )4) *d '-

0 No0
a c!4-4 a
0 9o$4 (d o bo
0 to > 10 w
T-I *H




0 (= -i 3 ct 43






rl.








-55-


(%) 3SV3N330
CD C*


AlIS03SIA
Co


CO -



LS

Um

Ik--


(oeljx au AzuS IU/813') SdflOHS 9WN13l3OM JO 3JSV313H







-56-


on xylan, locust bean gum and gum arabic have been detected

but at an even lower level than cellulase.

Rhizobium cellulase is resistant to heat. Overnight

incubation at 30*C did not affect enzymatic activity; further-

more, increases in enzymatic activity have been observed after

this treatment. Heating at 600C for 30 min destroyed 70% of

the activity while 15 min of boiling destroyed 95% of the

cellulase.


Production of Cellulase by Different Rhizobium trifolli
Strains

Figure 12 shows the level of cellulase production by

strains of Rhizobium trifolii with different infectiveness.

There were differences between pairs of infective and non-

infective strains. BART A, a noninfective derivative, pro-

duced 6 times more cellulase than the infective parental

strain 0403 or BAL. All of these three strains possess

three plasmids (Taylor, 1981), but BART A has a deletion in

the sym plasmid. As a contrast, 0435, which is infective,

produced almost 4 times the amount of cellulase produced

by 0435-2, its noninfective derivative.

The strain 521nod-8 is a derivative of 521, cured of

the sym plasmid (190 x 106 daltons) (Zurkowski and Lorkiewicz,

1979). The two strains did not show any difference in cellu-

lase production in the cultural conditions used.

The results obtained indicate that the genes) coding

for the Rhizobium endoglucanase is not in the sym plasmid.

Hooykaas et al. (1981) showed that the transfer of the sym































0 v 0
*V >, r

9i *4r *i-

1-* 4 q
4J >% 4

+j th lu
41 U M r-
g c0 Wir-i

.UU4

*u )
N 4q HO



,-l 0 )

to tn
t *M + P4(

S-4H 0 0
*a -He 4
v-I 0

SI 4o


o1
4-4 .w-I 4

(0 0 i >i




0 0 Mu




r-o







-58-


STRAIN CHARACTERISTICS:
Infective Non-infective
BAL BART A
0403 0435-2
0435 521nod"8
521


C.2



Lb


E
C=







N
Lu







-59-


plasmid of Rhizobium trifolii strain LPR 5001 (180 x 106 dal-

tons) to a Ti plasmid cured derivative of Agrobacterium

tumefaciens makes the transcojugants to induce nodule for-

mation in clover roots. The nodules were white, nonefficient

and appeared after a long period of incubation.

The results do not rule out the possibility of a role

of these endoglucanases in other stages of nodule development

such as bacteroid differentiation or the release of the bac-

teria from the infection thread. They could also be important

in the infection process of tropical legume roots which are

nodulated through a different pathway of infection (Chandler,

1978).















CONCLUSIONS


Cellulase production by Rhizobium strains is inducible.

Cellulase was produced by Rhizobium trifolii strains when

growing in Bergersen's medium with myo-inositol or cellobiose

as C sources; or when one of a variety of polysaccharides was

present in the medium. Enzyme production was repressed by

mannitol at concentrations higher than 0.25% w/v. Slow-

growing cowpeaa type" Rhizobium strains produced cellulase

(in all conditions tested), but the highest production was

found when potassium glutamate was the C source.

Rhizobium trifolii strains produce an endo-B-glucanase

which degrades CMC to oligosaccharides but not to cellobiose.

There are indications that cowpeaa type" rhizobia also pro-

duce an exo-a-glucanase.

Cellulase activity in raw enzyme extracts is stable,

but decreases when carbohydrates and ions are removed. The

presence of glycine, urea, calcium and magnesium helps to

maintain the activity during the purification process, but

fails to do so in the latter stages of purification.

Root substances stimulate the production of cellulase

by Rhizobium trifolii strains. Strains growing in the pres-

ence of root solids showed higher cellulase activity, as


-60-







-61-


assayed by the release of reducing groups, than do cultures

growing with root solubles or without amendment. However,

cultures growing with root solubles had the highest activity

when assayed by the viscosimetric method. This indicates

that a different set of enzymes is induced by each root

fraction.

Cellulase activity of enzyme extracts increases as pH

of the substrate increases up to pH 8, which was the highest

level tested. No effect on activity by calcium or magnesium

was observed.

There was no correlation between ability to produce

endo-8-glucanase by Rhizobium trifolii strains and infectiv-

ity of white clover. One of the strains lacking the nodula-

tion plasmid proved capable of producing cellulase at the

same rate as its infective parental strain.















APPENDIX I
VISCOSIMETRIC ASSAY FOR CELLULASE ACTIVITY


Amongst the various assays used for cellulase activity

determination, the viscosimetric method has the advantages

of being rapid and extremely sensitive, especially for endo-

glucanases. Its main disadvantage is that enzymatic activ-

ity is measured not in absolute terms but in relation to a

physical change, a decrease in viscosity. The viscosity of

a polymer solution is a function of its average molecular

weight. Viscosimetric assays can indeed, after certain fac-

tors are controlled, measure enzymatic activity in absolute

terms (Almin and Eriksson, 1967). Most workers, however,

express the results in relative terms, either as percentage

of decrease of viscosity per unit of time, time required for

attaining a 50% decrease in viscosity, or the slopes of the

straight lines resulting from plotting reciprocal specific

viscosities (1/ns) against reaction time (Rexova-Benkova and

Markovic, 1976). The fact that Rhizobium cellulase activ-

ities are very low necessitated testing the reliability of

the viscosimetric assay.

Figure 13 shows the relation between enzyme amount and

the slope of the line of the plot of reciprocal specific vis-

cosities against reaction time (hereafter called slope 1/ns).


-62-







-63-


The different amounts of enzyme were obtained by dissolving

a Rhizobium trifolii strain BAL enzyme extract with water.

An enzyme unit was defined as the amount of cellulase that

reduces the viscosity of a CMC solution (0.2% w/v) by 50%

in 10 hours or has a slope l/ns of 1.883 x 10-2 hours ,

which is approximately equivalent to the liberation of re-

ducing groups equal to 25 ug of glucose in 24 hours.

The graph (Figure 13) indicates that the relationship

is linear above 0.2 EU. A regression calculated with the

data obtained with the enzyme amounts ranging from 0.3 to

1.5 EU produced the following equation:
EU = -0.93 + 1.51 log(slope l/ns x 1,000)

Correlation = 0.9996

Figure 14 shows the plot of enzyme amounts against time

for achieving 50% of viscosity reduction. The relationship

is that of a straight line above 0.3 EU. For enzyme quanti-

ties smaller than 0.6 EU, the time to achieve 50% viscosity

reduction is more than 20 hours. In practice,these times

should be calculated by extrapolation from a plot of 1/ns vs

time of reaction because incubations longer than 24 hours

are not reliable. Therefore, the best way to calculate low

cellulase activities is through calculation of the slopes

1/ns.
































SOA
0
Cfl 4-4


Cd 0 *
CU O

rt 004
00 -H

S-*4 0 (U




.0 0 -4

0, O4
-4 4-4 0 .
= 0 0
9) 0

H o +Co
4 rl r --

r >l >


0Q) 4)-r4

*4-' : 0



r-43 0
CL4ct 0







-65-


I-












0-




cq



*d


9 ( o QO O 1O 3 d O






























d)

*- 0 0
4 ) +-)



0 r- .i -l
+-)U C 4P
r 4 tO
O $ *. *


0
0 ) ,l .
t t0 '-4







0 *0
4) U):) 4-






0 t Cn t 0
>N oo a
. -H 4.4







+ d -cd1 P4
4.up =p )
mocoC
OCU
ce 0t o Q,



CO O :
MC ca o

*l- UO*i-
to p-











p4-







-67-


Nli


























CD II-



n ci







(sinoq) AlIS03SIA NI 3SV3H33d %OS 3A31H3V 01 3W11















APPENDIX II
GROWTH OF Rhizobium trifolii IN INO MEDIUM



Rhizobium trifolii cultures growing in INO medium produce

a dark yellow pigment which strongly absorbs at wavelengths

shorter than 400 nm and gives a positive reaction with the

Lowry assay for proteins (Lowry et al., 1951). This makes

it difficult to quantify protein or bacterial population by

UV absorption or optical density. For this reason,bacterial

growth was measured by dry weight of bacteria per unit of

volume of culture.

Clumping often occurs in Rhizobium trifolii cultures

in INO medium. For this reason, bacterial population was

determined by direct counting using a Petroff-Hauser chamber

after 30 sec of sonication at 50W (Heat systems-Ultrasonics

Inc., model W140). The dry weight of bacteria was estimated

filtering a sample of culture through a Nucleopore filter

(Polycarbonate, 0.4 pm, Nucleopore Corp., Pleasanton, CA).

The filtrate was washed twice with filtered deionized water

and dried overnight at 700C. The weight of the bacteria was

calculated as the difference in weight before and after fil-

tration.

Figure 15 depicts the growth of Rhizobium trifolii

strains BAL and BART A in INO medium. The dry weights of


-68-







-69-


bacteria reach a peak toward the end of the log phase of

growth and then drop. The reason may be cell lysis and

indeed the viable cell count drops during stationary phase.

The optical density of the culture at 620 nm continues in-

creasing, however, due probably to accumulation of dead

cells, extracellular polysaccharide and pigment. The yellow

pigment is produced by Rhizobium trifolii growing in a de-

fined medium with myo-inositol as C source. Yeast extract

inhibits pigment production.
























50 O

o* 4 th 4
.e 0 *
; 0 9: b :
*3 C 3 n 3 e%

4 V) ul o r4 r U
0 0 H4-4 0 O
00 p4r-4
cd 4 4-4


:3 04 W 0*
0f u *0 0 rq
"-4 4-4 P4-
4-).94 $ e t
rl 4) r-4lU


q < (n0 9 0




4 Qarr 4 4
,o a to r *
^ &0 >C > a .





4J c0 cc
3 A (4- 12 0




0 x 4-) 4.)
>%-M 0 .-4) 34-4


(A EU-. 4)1-4 bOw-4

*I 4) a cd) 0 cd


O*'U 0*0l
I- *x 0 t4 4P (n
c4 0 3 4 4 *l
(3 -r4 C> 4J p.4
F 0 -r- tI (d C (0
~C? 6<-ii a 4- a








-71-


Strains:
BAL BART A
Population: o ---

Dry weight: -* ----
0 0


o0 0


1x100


N
E





C3


Ca
o



C-
LO


a-


-A 1x108

re
QC
W
mc
W


1x107'


1.0 "




18S
Do-

m



mc
=.

oA


1 2 3 4 5 6 7 8

TIME OF INCUBATION (days)















BIBLIOGRAPHY


Aist, J. R. 1977. Mechanically induced wall appositions of
plant cells can prevent penetration by a parasitic fungus.
Science 197:568-571.

Almin, K. E., and K. Eriksson. 1967. Enzymic degradation of
polymers. I- Viscosimetric method for the determination
of enzymic activity. Biochim. Biophys. Acta 139:238-247.

Basset, B., R. N. Goodman and A. Novacky. 1977. Ultrastruc-
ture of soybean nodules. I- Release of rhizobia from the
infection thread. Can. J. Microbiol. 23:573-582.

Bateman, D. F. 1964. Cellulase and the Rhizoctonia disease
of beans. Phytopathol. 54:1372-1377.

Bateman, D. F. -1976. Plant cell wall hydrolysis by pathogens.
p. 79-103. In: J. Friend and D. R. Threlfall (ed.).
Biochemical aspects of plant-parasitic relationships.
Academic Press, London.

Bateman, D. F., and H. G. Basham. 1976. Degradation of plant
cell walls and membranes by microbial enzymes. p. 316-
355. In: R. Heitefuss and P. H. Williams (ed.). Phys-
iologial plant pathology. Springer-Verlag, Berlin.

Bateman, D. F., and R. L. Millar. 1966. Pectic enzymes and
tissue degradation. Ann. Rev. Phytopathol. 4:119-146.

Bauer, W. D. 1981. Infection of legumes by rhizobia. Ann.
Rev. Plant Physiol. 32:407-449.

Bergersen, F. J. 1961. The growth of Rhizobium in synthetic
media. Aust. J. Biol. Sci. 14:349-360.

Beringer, J. E. 1980. The development of Rhizobium genetics.
J. Gen. Microbiol. 116:1-7.

Beringer, J. E., N. J. Brewin and A. W. B. Johnston. 1980.
The genetic analysis of Rhizobium in relation to symbio-
tic nitrogen fixation. Heredity 45:161-186.

Bhuvaneswari, T. V. 1981. Recognition mechanisms and infec-
tion process in legumes. Economic Bot. 35:204-223.


-72-







-73-


Byrde, R. J. W. 1979. Role of polysaccharide degrading enzymes
in microbial pathogenicity. p. 417-436. In: R. C. W.
Berkeley, G. W. Gooday and D. C. Ellwood (e-.). Microbial
polysaccharides and polysaccharases. Academic Press,
London.

Callaham, D. A. 1979. A structural basis for infection of
root hairs of Trifolium repens by Rhizobium trifolii.
M.S. thesis, University of Massachusetts, Amherst, 41 pp.

Capon, B., and J. W. Thompson. 1979. Studies on cellulolytic
enzymes. I- The use of 3,4-dinitrophenyl glycosides as
substrates. Bioorganic Chem. 8:147-173.

Casse, F., C. Boucher, J. S. Julliot, M. Michel and J. Denarie.
1976. Identification and characterization of large plas-
mids in Rhizobium meliloti using agarose gel electrophore-
sis. J. Gen. Microbiol. 113:229-242.

Chandler, M. R. 1978. Some observations of infection of
Arachis hypogaea L. by Rhizobium. J. Exp. Bot. 29:749-755.

Coughlan, M. P., and M. A. Folan. 1979. Cellulose and cellu-
lases: Food for thought, food for the future. Int. J.
Biochem. 10:103-108.

Dart, P. J. 1974. The infection process. p. 381-429. In:
A. Quispel (ed.). The biology of nitrogen fixation.
North Holland Publ., Amsterdam.

Dart, P. J. 1975. Legume root nodule initiation and develop-
ment. p. 467-506. In: J. G. Torrey and D. T. Clarkson
(ed.). The development and function of roots. Academic
Press, London.

Dart, P. J. 1977. Infection and development of leguminous
nodules. p. 367-473. In: R. W. F. Hardy and W. S.
Silver (ed.). A treatise on dinitrogen fixation. Sec-
tion III. John Wiley, New York.

Davies, E., and 0. Ozbay. 1980. A potential role for cellu-
lase in hormone-controlled elongation. Z. Pflanzenphysiol.
99:461-469.

Dazzo, F. B., and D. H. Hubbell. 1981. Control of root hair
infection. p. 274-309. In: W. J. Boughton (ed.).
Ecology of nitrogen fixation. Vol. 2. Oxford Univ. Press,
Oxford.

Dazzo, F. B., C. A. Napoli and D. H. Hubbell. 1976. Adsorption
of bacteria to roots as related to host specificity in the
Rhizobium-clover symbiosis. Appl. Environ. Microbiol.
32:166-171.







-74-


Deinema, M. H., and L. P. T. M. Zevenhuizen. 1971. Formation
of cellulose fibrils by gram-negative bacteria and their
role in bacterial floculation. Arch. Mikrobiol. 78:42-57.

Dekker, R. F. H., and G. N. Richards. 1976. Hemicellulases--
their occurrence, purification, properties and mode of
action. Adv. Carbohydrate Chem. 32:277-352.

Delwiche, C. C. 1978. Legume--past, present and future.
Biosci. 28:565-570.

Dingle, J., W. W. Reid and G. L. Solomons. 1953. The enzyme
degradation of pectin and other polysaccharides. II-
Application of the "cup-plate" assay to the estimation
of enzymes. J. Sci. Food Agric. 3-4:149-155.

Fahraeus, G. 1957. The infection of clover root hairs by
nodule bacteria studied by a simple glass slide tech-
nique. J. Gen. Microbiol. 16:374-381.

Gooday, G. W. 1979. A survey of polysaccharise production.
A search for phylogenetic implications. In: R. C. W.
Berkeley, G. W. Gooday and D. C. Ellwod (e-d.). Microbial
polysaccharides and polysaccharases. Academic Press,
London.

Gourret, J. P., and H. Fernandez-Arias. 1974. Etude ultra-
structrale et cytochimique de la differenciation des
bacteroides de Rhizobium trifolii Dangeard dans les
nodules de Trifolium reopens L. Can. J. Microbiol. 20:
1169-1181.

Graham, P. H. 1975. Identification and classification of
root nodule bacteria. p. 99-112. In: P. S. Nutman
(ed.). Symbiotic nitrogen fixation in plants. Vol. 7.
Cambridge Univ. Press, Cambridge.

Graham, T. L. 1982. Recognition in the Rhizobium-legume
symbiosis. Int. Rev. Cytol. (in press).

Hahn, M. G., A. G. Darvill and P. Albersheim. 1980. Poly-
saccharide fragments from the walls of soybean cells
elicit phytoalexin (antiobiotic) accumulation in soybean
cells. Plant Physiol. 65:136.

Harris, R. H., and R. Mitchell. 1973. The role of polymers
in microbial aggregation. Ann. Rev. Microbiol. 27:27-50.

Higashi, S., and M. Abe. 1978. Phage induced depolymerase
for exopolysaccharide of Rhizobiaceae. J. Gen. Appl.
Microbiol. 24:143-153.







-75-


Hooykaas, P. J. J., A. A. N. van Brussel, H. den Dulk-Ras,
G. M. S. van Slogteren and R. A. Schilperoort. 1981.
Sym plasmid of Rhizobium trifolii expressed in different
rhizobial species and Agrobacterium tumefasciens. Nature
291:351-353.

Hubbell, D. H., V. M. Morales and M. Umali-Garcia. 1978.
Pectolytic enzymes in Rhizobium. Appl. Environ. Micro-
biol. 35:210-213.

Kelman, A., and E. B. Cowling. 1965. Cellulase of Pseudomonas
solanacearum in relation to pathogenesis. Phytopathol.
55:148-155.

Li, D., and D. H. Hubbell. 1969. Infection thread formation
as a basis of nodulation specificity in Rhizobium-
strawberry clover associations. Can. J. Microbiol.
15:1133-1136.

Lillich, T. T., and G. H. Elkan. 1968. Evidence countering
the role of polygalacturonase in invasion of root hairs
of leguminous plants by Rhizobium spp. Can. J. Micro-
biol. 14:617-625.

Ljunggren, H., and G. Fahraeus. 1961. The role of polygalac-
turonase in root-hair invasion by nodule bacteria. J.
Gen. Microbiol. 26:521-528.

Loewus, F. A., and M. W. Loewus. 1980. Myo-inositol biosyn-
thesis and metabolism. p. 43-76. In: J. Preiss (ed.).
Biochemistry of plants. Vol. 3. Academic Press, New
York.

Lowry, 0. H., N. J. Rosebrough, A. Farr and R. J. Randall.
1951. Protein measurements with the Folin phenol reagent.
J. Biol. Chem. 193:265-275.

Lyon, G., and P. Albersheim. 1980. The nature of the phyto-
alexin elicitor of Erwinia carotovora. Plant Physiol.
65:137.

Mandels, M., F. W. Parrish and E. T. Reese. 1962. Sophorose
as an inducer of cellulase in Trichoderma viride. J.
Bacteriol. 83:400-408.

Martinez-de Drets, G., and A. Arias. 1972. Enzymatic basis
for differentiation of Rhizobium into fast- and slow-
growing groups. J. Bacteriol. 109:467-470.

Martinez-Molina, E., V. M. Morales and D. H. Hubbell. 1979.
Hydrolytic enzyme production by Rhizobium. Appl. Environ.
Microbiol. 38:1186-1188.







-76-


McMillan, J. D., and R. C. Cooke. 1969. Evidence against
involvement of pectic enzymes in the invasion of root
hairs by Rhizobium trifolii. Can. J. Microbiol. 15:
643-645.

Montenecourt, B. S., D. H. J. Schamhart and D. E. Eveleigh.
1979. Mechanism controlling the synthesis of the Tri-
choderma reesei cellulase system. p. 327-337. In:
R. C. W. Berkeley, G. W. Gooday and D. C. Ellwoo~-(ed.).
Microbial polysaccharides and polysaccharases. Academic
Press, London.

Napoli, C. A., F. B. Dazzo and D. H. Hubbell. 1975. Produc-
tion of cellulose microfibrils by Rhizobium. Appl.
Environ. Microbiol. 30:123-131.

Napoli, C. A., and D. H. Hubbell. 1975. Ultraestructure of
Rhizobium-induced infection threads in clover root hairs.
Appl. Microbiol. 30:1003-1009.

Nisizawa, K., Y. Tomita, T. Kauda, H. Sukuki and K. Wakabayashi.
1972. Substrate specificity of C1 and C cellulase com-
ponents from fungi. Fermen. Technol. Today. 719-725.

Olivares, J., E. Montoya and A. Palomares. 1977. Some effects
derived from the presence of extrachromosomal DNA in
Rhizobium meliloti. p. 375-385. In: W. Newton, J. R.
Postgate and C. Rodriguez-Barrueco-Ted.). Recent develop-
ments in nitrogen fixation. Academic.Press, New York.

Paau, A. S., J. R. Cowles and D. Raveed. 1978. Development
of bacteroids in alfalfa (Medicago sativa) nodules.
Plant Physiol. 62:526-530.

Palomares, A. J. 1975. Production de poligalacturonasas en
la asociacion Rhizobium-leguminosa. Ph.D. thesis, Uni-
versidad de Granada. Granada, Spain. 153 pp.

Prakash, R. K., P. J. J. Hooykaas, A. M. Ledebaer, J. W. Kyne,
R. A. Schilperoort, M. P. Nuti, A. A. Lepidi, F. Casse,
C. Boucher, J. S. Julliot and J. Denarie. 1980. Detec-
tion isolation and characterization of large plasmids in
Rhizobium. p. 139-163. In: E. Newton and W. H. Orme-
Johnson (ed.). Nitrogen iTxation. Vol. 2. University
Park Press, Baltimore.

Reese, E. T. 1977. Degradation of polymeric carbohydrates
by microbial enzymes. Recent Adv. Phytochem. 11:311-367.

Reese, E. T., R. G. H. Siu and H. S. Levinson. 1950. The
biological degradation of soluble cellulose derivatives
and its relationship to the mechanism of cellulose hy-
drolysis. J. Bacteriol. 59:485-497.







-77-


Rexova-Benkova, L., and 0. Markovic. 1976. Pectic enzymes.
Adv. Carbohydrate Chem. Biochem. 33:323-385.

Robertson, B. K., P. Aman, A. G. Darvill, M. McNeil and P.
Albersheim. 1981. Host-symbiont interactions. V-
The structure of acidic extracellular polysaccharides
secreted by Rhizobium leguminosarum and R. trifolii.
Plant Physiol. 67:389-40.

Robertson, J. G., M. P. Warburton, A. M. Fordyce, P. Lyttleton
and S. Bullivant. 1978. Membrane in lupine root nodules.
II- Preparation and properties of peribacteroidal mem-
branes and bacteroid envelope inner membranes from
developing lupin nodules. J. Cell Sci. 30:151-174.

Roland, J. C., and B. Vian. 1979. The wall of the growing
plant cell: Its three dimensional organization. Int.
Rev. Cytol. 61:129-166.

Ronson, C. W., and S. B. Primrose. 1979. Carbohydrate metab-
olism in Rhizobium trifolii: Identification and symbiotic
properties of mutants. J. Gen. Microbiol. 112:77-88.

Ryu, D. D. Y., and M. Mandels. 1980. Cellulases: Biosynthe-
sis and applications. Enzyme Microbial Technol. 2:91-102.

Shikata, S., and K. Nisizawa. 1975. Purification and proper-
ties of an exo-cellulase component of novel type from
Trichoderma viride. J. Biochem. 78:499-512.

Solheim, B., and J. Paxton. 1981. Recognition in Rhizobium-
legume systems. p. 71-83. In: R. C. Staples Ced.).
Plant disease control. John~Wiley, New York.

Solheim, B., and J. Raa. 1971. Evidence countering the theory
of specific induction of pectin-degrading enzymes as the
basis for specificity in Rhizobium leguminous associations.
Plant Soil 35:275-280.

Somme, R. 1974. Chemical analysis of exocellulas, acid poly-
saccharide from seven Rhizobium strains. Carbohydrate
Res. 33:89-96.

Somme, R. 1980. Neutral oligosaccharides from Rhizobium
trifolii BART A. Carbohydrate Res. 85:156-159.

Somogyi, M. 1952. Notes on sugar determination. J. Biol.
Chem. 195:19-24.

Swenson, H. A. 1963. Intrinsic viscosity and its conversion
to molecular weight. Methods in Carbohydrate Chem. 3:
84-91.







-78-


Taylor, R. 1981. Plasmid screening of Rhizobium trifolii
by agarose gel electrophoresis. M.S. thesis, Univer-
sity of Florida, Gainesville, 48 p.

Urbanek, H., J. Zalewska-Sobczak and A. Borowinska. 1978.
Isolation and properties of extracellular cellulase-
hemicellulase complex of Phoma hibernica. Arch. Micro-
biol. 118:265-269.

Verma, D. P. S., and V. Zogbi. 1978. A cooperative action
of plant and Rhizobium to dissolve the host cell wall
during development of root nodule symbiosis. Plant Sci.
Letters 13:137-142.

Vincent, J. M. 1970. A manual for the practical study of
root-nodule bacteria. Blackwell Sci. Publ., Oxford.

Vincent, J. M. 1974. Root-nodule symbiosis with Rhizobium.
p. 265-341. In: A. Quispel (ed.). The biology of
nitrogen fixation. North Holland Publ., Amsterdam.

Wood, T. M., and W. McCrae. 1978. The cellulase of Trichoderma
koningii. Biochem. J. 171:61-72.

Yao, P. Y., and J. M. Vincent. 1976. Factors responsible for
the curling and branching of clover root hairs by Rhizo-
bium. Plant Soil 45:1-16.

York, W. S., M. McNeil, A. G. Darvill and P. Albersheim. 1980.
Beta-2-linked glucans secreted by fast-growing species
of Rhizobium. J. Bacteriol. 142:243-248.

Zevenhuizen, L. P. T. M., and H. J. Scholten-Kaerselman. 1979.
Surface carbohydrates of Rhizobium. I- 8-1,2 glycans.
J. Microbiol. Antonie van Leeuwenhoek 45:165-175.

Zurkowski, W., and Z. Lorkiewicz. 1979. Plasmid-mediated
control of nodulation in Rhizobium trifolii. Arch.
Microbiol. 123:195-201.















BIOGRAPHICAL SKETCH


Victor Manuel Morales was born March 3, 1949, in

Itsmina, Colombia. He graduated from Colegio de. Santa

Librada in Cali, Colombia, on July 1965. In December,

1971, he received his degree of Ingeniero Agronomo from

the Universidad Nacional de Colombia in Palmira.

He held the position of Research Assistant.in CIAT

(Centro Internacional de Agricultura Tropical) from Jan-

uary 1972 to March 1975.

Upon being awarded a scholarship from CIAT, he came

to the United States for graduate studies. He received

his Master of Science degree from the University of Florida

in August 1977.

He is a member of the American Society for Microbiology

and the Florida Soil and Crop Science Society.

He is currently a candidate for the Ph.D. degree in

the Soil Science Department of the University of Florida.


-79-




.. 1 -1


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.



Davyd- H. Iubbell,' raY
Professor of Soil Science



I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.



Edward M. Ho ma /
Professor of MiCrobi gy and
Cell Science


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy..



Chesley B. Hall
Professor of Horticultural Science



I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.


Associate Professor of Soil Science


1










I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.



es F. Preston
sociate Professor of Microbiology
and Cell Science



This dissertation was submitted to the Graduate Faculty
of the College of Agriculture and to the Graduate Council,
and was accepted as partial fulfillment of the requirements
for the degree of Doctor of Philosophy.


December 1981 'j, I*
Dean, bl1llege of Agr culture



Dean for Graduate Studies and
Research















































UNIVERSITY OF FLORIDA
122IIIII 055II3III
3 1262 08553 9608


_ I_