_ .1 '1: 1' 44.4.1... . ‘1‘, 1.
"' W 11,111.14? géflu... :5“; 14‘1", 13-”; .

L31; W‘ffi ' 15'5";

    
   

   

.'1; an" ' . '1; .
111 ":11 h?” .- 1 1'1 ., .. f‘ .1111 X11311.r F61”? ”no. 7,17:
l a": . -41 "ml" 1.14:1.11'1 .. "11.11141310111'111'a1‘
. 111 .1 ._1.. ... -44: . 4 '1. - : 1.:
.111 1r -;

     

  

 
 

1 1 111"”, J11“) “'1‘.
.- I 111’.‘ I ‘ "I _ {17134 .2:14
1n. '4 157. .1" 44 1}? 1111,1111, 11111" ' 1 1‘
.113 1 \1' A '11» 1, w" $101.1 .1
‘R‘1‘1’J1‘P‘1112-1 “fl-R1115» '

      
  
 

    

    

     
   
  
       
     
       

 
  

-« a
11-111
d313,:

VII! .111145)-.1

41" 1“. 1fl1-1 1

'I
1"
.
1',
.
.
4
1.
‘1
c“

 

c— 41114444....

311-311711141-

Ja‘a;m1.:k$r‘:H

pl
.0-

~36“.

Q1331; 1 CJ v.-

1‘-

 

 

     

.1 1 1111 ll. ‘U‘—1' ()15' I 111 “I
' 1 1 . . 4, 1'1. 1 41 ..
(111” 11‘ 4| ' . . 1 , 1‘ ~1- ' 1. 1 .111; 11 1.11111 4 L'I ‘ 1.1‘1111 'i‘ ‘1 1 1, 1
‘ 4 ‘ '-4 11‘ $1.441. 13114-1 4114.11: ,1 1. 1.1 1
1 I, 111‘ ' .11 .' 1 ‘1'. l1. [4:111 ..' _ "'10. '1 1‘ 1 1 11 ‘ '1 1 "1:01" 1” '1' .1 ’
-. ~ '~ 1‘ -' .411,» .11. 11f?“ ' ”114‘.- * .11»; 1 1 4 914
1 ‘ 1“ - 1 4'. .1 1,...” 1, I '1
1; 11:. '1' ' ' 1" ,44-4114111111411'1 {‘11‘311‘141111’ . '. "1 '1:
1 ‘.1,,I .1 1 I: ‘ .1, 1:11 11 1| ".0 #:111,‘I 1:11 (11:11 1‘" {#14 1_ ,, “$1. 1" . kn“ .-
1 ' "-1 ' 1‘ ' .
| | I ' 111.111 ’11. 11.1 ‘1 1'. '1 I {IVE-fri- H" 1.11 1‘; ~;."1«'.L:’:_ wk‘# 4. “J. J&( J d'fi“. 1".“

THESIS

LIB KAI! 1’
Michigan State
l University

 

This is to certify that the

dissertation entitled

PROTOPLAST DEVELOPMENT, REGENERATION AND
FUSION IN Brevibacterium lactofermentum

 

presented by

Susanne E. Keller

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in EQQd Science

 

JW ‘1pr

Major professor

 

James J. Pestka
Date 7'22”“:-

 

MS U i: an Affirmatiw Action/Equal Opportunity Institution 0- 12771

 

 

MSU

LIBRARIES
m

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

 

 

 

 

PROTOPLAST DEVELOPMENT, REGENERATION AND
FUSION IN Brevibacterium lactofermentum

 

By

Susanne E. Keller

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Food Science and Human Nutrition

I985

ACKNOWLEDGMENTS

Thanks to my major professor, James Pestka, for his
continued support throughout this experience. My sincere
appreciation to Harold Sadoff without whose advice and
help I surely would have failed. Thanks also to the rest
of my committee members, Robert Brubaker, J.R. Brunner, and
Mark Uebersax, who have put up with me all this time. Last
but not least, thanks to H. Momose whose initial guidance

led me to this research.

ii

TABLE OF CONTENTS

LIST OF TABLES.
LIST OF FIGURES
INTRODUCTION.
LITERATURE REVIEW

Uses and Economic Importance of Brevibacterium.

 

Classification and Taxonomy of Brevibacterium
Protoplast Development and Fusion

 

MATERIALS AND METHODS
Strains . .
Media
Protoplast Formation. .
Microsc0pic Observation of. Prot0plasts.
Protoplast Fusion . .
Radiolabeling of Cell Wall.

RESULTS
Prot0plast Formation. . . .
Regeneration of Prot0plasts
Variation in Mutants.
Determination of Remaining Cell Wall and.
Relationships to Regeneration
Protoplast Fusion

DISCUSSION.
CONCLUSION.

APPENDIX. . .
Introduction and Literature Review.
Methods
Results
Discussion.
Conclusion.

BIBLIOGRAPHY.

Page
iv

vi

LIST OF TABLES

Table

I

ID

Development of osmotic sensitivity in Brevi-
bacterium lactofermentum and appearance of

 

protoplasts.

Effect of media on the development and regenera-
tion of protoplasts. . .

Effect of sucrose concentration of protoplast
development and regeneration . . . . .

Effect of bovine serum albumin on the regenera-
tion of prot0plasts.

Distribution of counts in each peak of
hydrolyzed cell chromatograms.

Comparison of two methods for protoplast fusion.

Effect of temperature on fusion and recombina-
tion frequency

Effect of different molecular weight PEG on the
fusion and recombination frequency . . . .

Effect of length of PEG treatment on fusion and
recombination frequency. . . . . . .

Effect of protoplast development on fusion and
recombination frequency.

Appendix Table

l

Effect of NTG concentration on mutation
frequency and cell viability in the membrane
filter rotation method

iv

Page

29

34

35

41

49
52

53

55

56

57

75

Appendix Table

2 Induction of his+ reverse mutation in his'l6l

using membrane filters and NTG containing plates.

3 Inactivation of residual NTG by placing membrane
filters on pH 9 plate after NTG treatment

4 Effect of pH 9 treatment on cell viability.

Page

76

78
8O

LIST OF FIGURES

Figure

I
2

ID

Development of protoplasts.

Appearance of osmotic sensitivity during lyso-
zyme treatment. . . . . . . . .

Regeneration of protoplasts

Regeneration of wild type and his’ mutant
protoplasts vs depth of agar.

% Osmotically insensitive cells remaining vs
length of lysozyme treatment.

% 3H in protoplasts vs % regeneration of
protoplasts . . . . . . . .

% 3H in protoplasts vs % whole cells remaining.

Chromatograms of (3H) DAP heated with and
without killed cells. . . . . . . .

Chromatograms of hydrolysed cells labelled with
(3H)DAP....................

Changes occurring over time during development
of protoplast from (3H) DAP labelled cells.

Appendix Figure

1
2

Effect of NAL on the growth of strain 2256.

Increase in viable cells of strain 2256 after
NAL treatment

Effect of NTG concentration on mutation

frequency and cell viability in the membrane
filter rotation method. . .

vi

Page
3l

33
36

39

42

44
45

46

48

50

73

74

77

Appendix Figure

Page
4 Variation in frequency of streptomycin
resistant (A) and histidine reversion (8)
mutations. . 81
5 Replication map of StrR and his+ loci in
hisl6l.................... 82

vii

INTRODUCTION

Although microorganisms are currently known to produce
hundreds of substances of commercial value, relatively few
are produced for sale (Eucheigh, 198l). This is due
primarily to the lower cost and higher yield of chemical
synthesis as compared to biological fermentation. In
microorganisms, biological pathways to the production of
various industrial chemicals are often subject to strict
control mechanisms which prevent their overproduction in
the wild type. Production, therefore, depends upon the
ability to overcome these controls either by often costly
chemical manipulations or the development of such mutants
lacking such control mechanisms. Traditionally, the
devel0pment of mutants was dependent of the screening of
enormous numbers of artificially created mutants, a tech-
nique that was at best slow and laborious. With the advent
of new genetic methods such as recombinant-DNA techniques,
direct manipulation of biological pathways within micro-
organisms becomes a possibility. The creation of such new
microorganisms makes fermentation of various economically

important substances a very attractive proposition.

However, the creation of new industrially useful
strains of microorganisms requires that basic research
provide us with the knowledge necessary to take advantage of
new genetic methods. For example, in both E. coll and
g. subtilis, various methods of genetic transfer are well
established. A genetic map and various methods of mapping
of the genomes of these organisms is also readily available.
This information is vital to the success of newer methods of
genetic engineering. Unfortunately, this is not the case
with all industrially important microorganisms. One such
genus of industrial importance for which these methods have

not been well established is Brevibacterium. It was the

 

purpose of this work to establish a reliable method of

genetic transfer and a genetic map of Brevibacterium, using

 

g. lactofermentum as the model strain.

 

The uses and economic importance of the genus Brevi-

bacterium are wide and varied. Expansion of its use and

 

improvement of strains for commercial use would be of
considerable economic interest. As mentioned such develop-
ment currently depends on traditional development and
screening of enormous numbers of mutants. Reliable methods
for genetic transfer and genetic mapping of this species
would enhance genetic improvement of these industrially

important strains of Brevibacterium.

 

LITERATURE REVIEW

Uses and Economic Importance of Brevibacterium

 

 

One bacterial genus of considerable industrial impor-

tance is Brevibacterium. For example, D. flavum is one

 

organism used commercially to produce glutamic acid or
monosodium glutamate (MSG) for use as a flavoring agent.
Approximately 300,000 tons of MSG are produced per year.
This represents sales of l,O8O million dollars per year
(Eucleigh, l98l). A number of other species of Brevi-

bacterium are also considered glutamic acid producers.

 

These species include a. divaricatum, D. aminogenes, g.

 

 

lactofermentum, D. saccharolyticum, D. roseum, D. ammonia-

 

 

gens, D. alamicum, and D. thiogentitalis (Abe and Takayama,

 

1972). In addition to the more traditional approach used
to produce MSG by g. flavum, a novel method using temperature-

sensitive mutants of D. lactofermentum was developed by

 

Momose and Takagi (1978). Normally, glutamate production

by Brevibacterium requires that the media be biotin

 

deficient or that penicillin or surface active agents such
as Tween 60 are added to the media. The mutant developed

by Momose and Takagi (I978) was found to produce glutamic

acid in biotin-rich media without the presence of peni-
cillin or surface active agents when the temperature was
raised from permissive (30°) to non-permissive (37°).
Along with glutamic acid, 8. flavum can be manipulated
to produce lysine, an essential amino acid. Currently 8 %
of lysine is produced by fermentation (Eucleigh, I981).
Other amino acids which have been reported to be produced

by various members of the genus Brevibacterium include

 

alanine, valine, homoserine, o-ethylhomoserine, o-propyl-
homoserine, phenylalanine, threonine, proline, tyrosine,
aspartate, histidine, serine, isoleucine, and Ieucine (Abe
and Takayama, I972; Ikeda et al., I976; Nakamori and Isamu,
I970; Shiio et al., I982; Tsuchidu et al., l975a; Tsuchidu
et al., I975b; Yoshinaga, I969). Although not all these
are economically viable fermentations, the variety serves
to illustrate the point that, with new genetic techniques,
the industrial fermentation of any amino acid by Brevi-

bacterium may be possible in the near future.

 

In addition to the production of amino acids, this
genus is important in the production of nucleotides and
nucleotide-related products such as 5'-inosinic acid
(Nara et al., I969; Sato and Furuya, I977; Teshiba and
Furuya, I982). Inosinic acid is an important flavor-
enhancing agent of foods and of considerable economic

significance. The species used in its production is

8. ammoniagenes. Coenzyme A is also produced via biological

 

fermentation utilizing 8. ammoniagenes (Asada et al.,

 

I982; Shimizu et al., I979).

Another important use of Brevibacterium in the food

 

industry is the production of various cheeses (Rose, l98l).

8. linens in mixed cultures usually with Streptococcus

 

lactis and S. cremoris is involved in the production of
limburger, brick, muenster, trappist, port and duSalut
cheeses (Rose, l98l; Schmidt et al., I976). Its primary
role appears to be in flavor deveIOpment.

Uses of Brevibacterium discussed to this point have

 

centered on those already of considerable economic impor-

tance. A survey of literature involving Brevibacterium

 

indicates other possible roles for this genus, including
biodegradation of herbicides (Horvath, l97l; Rott et al.,
I979). Some aid the metabolism of hydrocarbons (Atlas and
Bartha, I972; Pirnik et al., I979). This latter ability
may make them important in biodegradation of hazardous

pollutants. Another possible use of Brevibacterium, the

 

production of feed protein from mesquite wood, has been
investigated by Fu and Thayer (I975). Mesquite wood itself
is not useful as feed and its conversion to protein would
render those areas where only mesquite wood grows as

economically productive. One final use for Brevibacterium

 

may be in the isolation and use of novel endonucleases for
use in characterization of DNA or in recombinant DNA work.

A few such nucleases have already been characterized

(Basnak'yan et al., I98I; Gerlinas et al., I977).

Classification and Taxonomy of Brevibacterium

 

 

Brevibacterium as a group is rather complex and hetero-

 

genous. In the 8th edition of Bergey's Manual it is placed
within the coryneform group of bacteria Genera incertae
sedis. Attempts to clarify the position of Brevibacterium
have been made by numerous methods using different criteria
(Bergey & Manual, 8th Ed., I974; Collins et al., I979;
Crombach, I97I; Jones, I975; Keddie and Cure, I979; Komagata
et al., I969; Stackebrandt and Fiedler, I975; Seiler, I983;
Yamada et al., I976; Yamada and Komagatu, I970) and with
variable success. No attempt will be made here to reclassify

or clarify the taxonomy of Brevibacterium. However, certain

 

characteristics examined by researchers and unique to this
group will be discussed.

General characteristics of Brevibacterium are similar

 

to those of Corynebacterium and Arthrobacter. They have

 

 

considerable morphological diversity but typically appear
as short, aerobic, non-sporulating, gram-positive rods
(Bergey's Manual, 8th Ed., I974). Yamada and Komagata

(I970)attempted to characterize Brevibacterium as well as

 

the other related groups of coryneform bacteria on the
basis of type of cell division, cell wall amino acid content
and DNA base composition. None of the factors discussed

could be utilized to distinguish Brevibacterium from other

 

coryneform bacteria. GC content was found to vary con-
siderably (46.6-70.5). This is supported by latter studies
done by Pitcher (I983). Cell wall content varied but

remained consistent with variation found in Corynebacterium

 

and Arthrobacter. It is interesting to note that glutamic

 

acid producers, regardless of genus, had a narrow range of
GC content, similar cell wall amino acid content, and a
single mode of division. Examinations by Keddie and Cure
(I979) of cell wall composition of coryneform bacteria also

found wide diversity in Brevibacterium in general. However,

 

glutamic acid producers, regardless of genus, appeared
similar. Such strains all contained DL-DAP along with the
sugars, arabinose and galactose. Thus these glutamic acid
producing strains among different genera would seem, there-
fore, to be more related to each other than individual

strains within the grouping Brevibacterium.

 

Other researchers have attempted to use the unusual
menaquinones found in coryneform bacteria as a classification
method (Collins et al., I979; Yamada et al., I976). Cory-
neform bacteria were found to contain the same type of
menaquinone, referred to as MK-9 (H2). Because this is the

same type found in Corynebacterium bovis and also on the

 

basis of similar mycolic acids within glutamic acid producers
and true corynebacteria, Collins et aI. (I979) recommended
the transfer of all glutamic acid-producing strains they

tested to Corynebacterium-sensu stricto. These researchers

 

also point out that their data support the hypothesis that
glutamic acid producers should in fact be reduced to

synonymy with C. glutamicum. This hypothesis is further

 

SUPported by the numerical taxonomic studies of Seiler
(I983). One method to resolve the difficulty in classifi-

cation of Brevibacterium within Corynebacterium may be found

 

 

in DNA-DNA or DNA-RNA homology studies. Such studies

could provide more conclusive evidence on the relationship
of the various genus and species to each other. Stackebrant
and Fiedler (I975) have done one such study. Since, the
number of organisms used within corynebacteria in general

and specifically Brevibacterium are very small and of little

 

use in classification, additional work in this area would be

beneficial.

Protoplast Development and Fusion

 

Transformation, transduction, or conjugation have never

been reported for Brevibacterium. However, protoplast

 

fusion has been reported for 8. flavus, 8. lactofermentum

 

and g. glutamicum, and therefore, holds promise as a method

 

of genetic exchange (Ajinomoto Co., I983a, l9836). Optimal
conditions have not yet been well established.

Protoplast fusion has been used for genetic studies in
a large number of organisms (Ferenezy, I98l; Hopwood, l98l).
With the advent of the use of PEG as a fusogen by Koa and

co-workers with plant protoplasts, various procedures for

fusion have been deveIOped for both procaryotes and
eucaryotes (Baltz, I978; Fodor and Alfold, I979; Hales,
I977; Kao and Michayluk, I969; Kao et al., I974; Ponte-
corvo, I975; Schaeffer et al., I976). Proper use of
prot0plast fusion as a genetic technique depends upon the
ability to develop and revert protoplasts. The method used
depends upon the type of organism in question. Within the
gram-positive procaryotes such as 8. subtilus, 8. lichen-

formis, and 8. megaterium, which have been extensively

 

studied, a simple lysozyme treatment was sufficient to
produce protoplasts (Bugaichuk et al., I98l; Fodor and
Alfoldi, I976; Schaeffer et al., I976; Wyrick and Rogers,
I973). In the case of 8. brevis, achromopeptidase was used
successfully to produce protoplasts. Resistence to lysozyme
of this 8. brevis strain was attributed to a thick protein
layer on the cell surface (Nimi et al., I983). Other

gram-positive organisms such as Mycobacterium, Brevibacterium,

 

 

Streptgmyces and Streptococcus are also resistant to treat-

 

ment with lysozyme. In the case of Mycobacterium and

 

Streptomyces, protoplasts have been obtained by first

 

treating the cells with glycine in growth medium (Baltz,
I978; Hopwood, I98l; Udou et al., I982). The glycine is
believed to replace alanine in the peptidoglycan, thereby
interfering with crosslinking. Cells treated in this manner
become more sensitive to lysozyme. Lysozyme and a amylase

were used together to obtain protoplasts of Streptococcus

 

IO

lactis (Okamoto et al., I983). Prot0plasts have been

obtained in lysozyme-resistant Brevibacterium and Coryne-

 

bacterium by treating with penicillin in combination with

 

lysozyme either in a sequential manner or simultaneously
(Kaneko and Sakeguchi, I979; Shtannikov et al., I98l;

Rytir et al., I982). Most gram-negative bacteria peptido-
glycan is sensitive to lysozyme (Schaitman, l98l). However,
the peptidoglycan is surrounded by a lysozyme-impermeable
outer membrane, which must first be destroyed before the
lysozyme can act. Digestion of the peptidoglycan does not
result in the removal of this outer membrane. Therefore,
such osmoticalIy-sensitive structures are more appropriately
termed spheroplasts. Removal of the outer membrane to
produce true protoplasts is an important and probably
essential step if fusion is to occur. In gram-negative
bacteria such as 8. £911, a procedure has been developed
involving lysozyme treatment and EDTA that will produce

true protoplasts (Weiss, I976). Spheroplasts with about

l5% of exposed cytoplasmic membrane of Providence alcali-

 

faciens were produced using a combined glycine and lysozyme
EDTA treatment (Cortzec et al., I979). These were success-
fully utilized in subsequent fusion experiments.

Conditions forregeneration of prot0plasts appear to be
far more complex than for their development and may vary
even in closely related species (Hopwood, l98l). Results

of Fodor and Alfoldi (I979) emphasize that various

II

physiological factors can greatly affect regeneration rates.
They claim complete distortion of genetic information

derived from their fused phenotypes due to variable regenera-
tion of the parental mutant types. Gabor and Hotchkiss
(I982) also noted physiological factors such as crowding

on plates played an important role in regeneration of fused
protoplasts, thus supporting observations of Fodor and
Alfoldi. In an earlier study by Landman et al. (I968)
crowding was found to delay reversion markedly. In studies

with Streptomyces by Baltz and Mutsushima (l98l), both the

 

temperature of cell growth prior to protoplast development
and the temperature during regeneration were found to effect
regeneration rates. The optimum growth temperatures
required for good regeneration varied from species to species
and was not necessarily the same as that temperature optimum
required during regeneration. In addition dehydration of
regeneration plates and plating by using a soft agar overlay
was said to result in a more rapid, synchronous and efficient
regeneration. As a result auto-inhibition exhibited by
some species was overcome. Using optimal conditions for
strains tested nearly IOO% regeneration was achieved.

Careful examination of parameters for the genera-

tion of Bacillus subtilus have resulted in improvements

 

to the established hypertonic media that can obtain up
to IOO% regeneration rates (Gabor and Hotchkiss, I979).

The primary factor responsible for this high rate was the

I2

addition of I% bovine serum albumin (BSA) to the growth
and dilution media. Unfortunately regeneration did
drop to ID to 75% after treatment with PEG. BSA was

also reported to enhance regeneration of Staphylococci.

 

However, the use of BSA in the regeneration media precludes

the direct selection of recombinants. Akamatsu and
Sekiguchi (I984) developed a method of regeneration of
Bacillus species protoplasts using a preincubation with
3% polyvinylpyrollidone. No appreciable growth of
auxotrophs was obtained whereas regeneration frequencies
increased SOD-fold in some cases. In addition, a 20
fold increase in regeneration rates was obtained when

an agar overlay method was used as opposed to spread
plates. Lower agar concentrations also resulted in
increased regeneration frequencies.

Another interesting observation made by Akamatsu and
Sekiguchi (I984) was that regenerating protoplasts
increased in size first then underwent a non-oriented
division. In regenerant colonies which had been plated
using an overlay method, cells were in the bacillary form
in 3 or 4 days. When the protoplasts were plated using a
spread plate method the majority stopped growing after a
few divisions. Akamatsu and Sekiguchi (I984) hypothesized
that the newly synthesized cell wall components at high
concentrations around the protoplasts might be an important

factor for their regeneration.

I3

Past studies also indicate a possible role of cell wall
components in the regeneration of protoplasts. Differences
in the envelope structure of three different L-forms of

Proteus mirabilis were examined by Demonty et al. (I973).

 

An incomplete cell wall about 8 mm thick outside the plasma
membrane was observed in two unstable L-forms whereas the
stable L-form was enveloped only by the plasma membrane.
The remaining cell wall of the unstable L-forms was
believed to be made up largely of Iipopolysaccharides.
Peptidoglycan content did not correlate to the unstability
of the various L-forms.

Reversion of Bacillus subtilis was stimulated by

 

the presence of cell wall extractions and a variety
of autoclaved intact microorganisms (Landman and Forman,

2+ + . . .
and K 1ons were requ1red for revers1on.

I969). Mg
Prot0plasts were said to enlarge but not divide in liquid
medium and reversion was stimulated with gelatin or hard
agar. Reversion was studied via the use of various
antibiotics blocking DNA, RNA, protein, and cell wall
synthesis in order to determine if and when these events
were required. Reversion was broken down into three

steps. It was speculated that the first step involved an
alteration of the membrane, the second step, the depression
of DAP and mucopeptide wall synthesis, and the last step,

teichoic acid synthesis. The physical immobilization of

excreted cell products at the protOpIast surface early in

I4

step two was also believed to be important.

Conditions for regeneration of Brevibacterium have

 

been examined. Best reversion rates when succinate is
used as to osmotic stabilizer (as opposed to sucrose)
and when the protoplasts are imbedded in the top of a
semisoft 0.8% agar layer (Shtannikov et al., l98l).
Enrichment with casein hydrolysate or amino acids and
nitrogen bases also increased reversion with reported
reversion rates being 20 to 70%. The results correspond
fairly well to those in an earlier work by Kaneko and
Sakaguchi (I979) where a 30% regeneration rate was
reported under similar conditions of regeneration. No
other variables that might affect regeneration of

Brevibacterium protoplasts were investigated.

 

As mentioned earlier, fusion of protoplasts have been
accomplished with both procaryotes and eucaryotes. Fusion
has not been limited to fusion of protoplasts of the same
species, but can be made to occur with different species
and genus (Cocking et al., I98l; Ferenzy, l98l; Hales,
I977; Hopwood, I98l; Shepard et al., I983). Interspecific
protoplast fusion has been reported more often for eucary-
otes than for procaryotes. Presumably, this does not stem
from the inability to achieve interspecific protoplast
fusion but rather simply that few have tried it. Hopwood
(I98l) reports attempts at interspecific fusion with

Streptomyces and concludes that although such recombinants

 

 

[rm—r-.-sx

I5

occur, they occur at very low frequency and are not very
stable. Shtannikov et al. (l98l) report that fusion does
occur between Corynebacterium glutamicum and Brevibacterium
flavum. However, it should be noted that these two genus
are so closely related that some researchers indicate they
are in fact the same. Interspecific fusion has also been
reported for Bacillus species (Akamatsu and Sekiguchi,
I983). Numerous examples of intraspecies fusion occur in
the literature and have been reviewed in some detail
(Ferenezy, l98l; Hopwood, l98l; Pebery, I980).

Fusion itself is thought to occur in distinct phases

LIT”? fiffimic -..-.- at- .|.=. E

(Ahkong et al., I975; Frehel et al., I979; Gumpert, I980;
Knutton and Pasternak, I979). The model proposed by
Gumpert (I980), which also encompassed those proposed by
others, specifies five steps. The first is the formation
of a contact zone. Presumably, this is promoted by fusogens
such as polyethylene glycol (PEG). PEG is said to promote
aggregation of the protoplasts (Tilcock and Fisher, I982).
The second step is the establishment of molecular contacts
and alteration of membrane structures. Changes in the
bilayer phase of membrane lipids have been reported by
Cullis and Hope (I978 and l98l) using 3IP-NMR techniques.
This alteration of the membrane includes increased fluidity
of the phospholipids and lateral diffusion of intrinsic
proteins which may be promoted by the action of both PEG
and Ca2+ (Tilcock and Fisher, I979). PEG has been shown to

I6

cause changes in phospholipid hydration and polarity which
may account for these alterations (Arnold et al., I983). In
particular, it may be the removal of water associated with
the bilayer that is the principle destabilizing effect
which results in fusion (Gibson and Strauss, I984). The
third step is the formation of a fusion membrane and the
fourth the formation of a separation layer by membrane
material. The final step is the migration of lipid
vesicles and intermixing of cell contents. Sucess of
fusion may depend in part on the partial removal of

the PEG (Ferenezy, I98l; Wojcieszyn et EH .,

I983).

Some controversy does exist on the ability of PEG by
itself to cause fusion. Honda et al. (l98la) reported that
purified PEG mw 6000 was unable to cause cell fusion. They
concluded that the commercial grade PEG contained at least
two components, one which caused aggregation and one which
resulted in the perturbation of the phospholipid bilayer.
Honda et al. (l98lb) later identified the components removed
during PEG purification as antioxidants like a-toc0pherol
or other phenolic derivatives which are often added to
commercial PEG. Fusion activity could be restored to the
purified PEG when such components were added. These results
were supported by the work of Wojcieszyn et al. (I983). In
contrast, Smith et al. (I982) reported that four out of five

commercial preparations of PEG retained fusogenic activity

I7

upon purification. Although the addition of small quan-
tities of compounds of the type utilized by Honda et al.
(l98lb) was observed to enhance cell fusion up to 5 %.

PEG is now used almost exclusively to facilitate fusion.
Factors investigated included pH, concentration of Ca2+,
molecular weight of the PEG, time and temperature of treat-
ment. Generally speaking, more concentrated solutions of
50% PEG (wt/vol) are more effective than dilute solutions.
For fusion of mammalian cells, PEG mw 6000 have been used
(Pontecorvo, I975). In later procedures, DMSO was added to

enhance fusion by 4l.7% (wt/vol) PEGat mw HEM (Hales, I977).

For fusion of various Streptomyces, PEG from IOOO to

 

6000 has been used with and without the addition of DMSO.
PEG was generally in the range of 36 to 50% (Akamatsu and
Sekiguchi, I983; Hopwood and Wright, I978; Hopwood and
Wright, l98l; Ochi et al., I979; Rose, l98l). In fusion
experiments with Bacillus species, PEG 6000 is most often
used (Fodor and Alfoldi, I976; Fodor and Alfoldi, I979;
Frehel et al., I979; Horvath, l97l; Schaeffer et al.,
I976). With Brevibacterium, PEG 6000 was also used at 33%

 

by Kaneko and Sakaguchi (I979) and at 40% by Shtannikov

et al. (l98l). Fusion by Shtannikov et al. (l98l) was

done at room temperature, whereas Kaneko and Sakaguchi (I979)
used 30°C. The possible effect of different temperatures was
not investigated by either group. Kaneko and Sakaguchi

(l98l) report that pH did not much effect the fusion

I8

mixture. The range that they tested is not given.
Shtannikov et al. (I98l) did not vary pH conditions.
Kaneko and Sakaguchi (I979) did test exposure time to PEG
and report no effect on fusion frequency over a l to 5
minute period. Neither group investigated the possible

2+

influence of Ca on the fusion of Brevibacterium. Controls

 

utilizing DNase to assure that genetic transfer was indeed
due to protoplast fusion as opposed to DNA uptake are not

mentioned. A requirement for Ca2+

has been noted by groups
working with fungi, but it has not been well investigated
in fusion with bacteria (Ferenezy, I98l; Peberdy, I980).

Ca2+

has been shown to be involved with the fusion of
artificial membranes in a number of studies (Ingolia and
Koshland, I978; Ito and Ohnishi, I974; Papahadiopoulos
et al., I974; Papahadiopoulos et al., I976; Sun et al.,
I979). The mechanism of action of the Ca2+ is thought to
be by causing phase separation or disturbances in the lipid
molecules. This could result in the formation of small
bridges when cells come into contact, which then enlarge
for fusion.

It should be pointed out at this time that fusion does
not by itself ensure genetic exchange. Recombination is a
separate event. Frehel et al. (I979) placed emphasis on an
optimal post-PEG incubation period. The method of selection

of recombinants may exert a strong influence on recombinant

genotypes recovered. Generally, both direct and indirect

I9

screening methods have been used. Fodor and Alfoldi (I979)
used a direct method, screening on the basis of nutritional
requirements. As mentioned previously, this led to con-
siderable bias in the recovery of the recombinants. Under
the circumstances, an indirect method might produce better
results.

With Brevibacterium, direct selection for antibiotic

 

markers has been utilized (Kaneko and Sakaguchi, I979;
Shtannikou et al., l98l). Recombinants are selected on
the basis of rifampsin and streptomycin-resistance markers
obtained from each parental type, then classified on the
basis of a non-selected auxotrophic marker. Presumably,
this could avoid physiological bias introduced by variable
nutrient content in the regeneration media as utilized by
Fodor and Alfoldi (I979).

In the literature, there are a few notable novel
approaches to selection of recombinants. These involve the
selection of one or both the parental protoplasts so that
only the resulting recombinant progeny will grow. One
method by Hopwood and Wright (I98l) involves the use of
ultraviolet irradiated protoplasts. The ultraviolet
radiation presumably leaves the protoplast intact but
nonviable unless fused with another protoplast not carrying
the same defect. This is the same reasoning used by Wright
(I978). Here cells were given lethal doses of a specific

reagent or inhibitor, which selectively damages a specific

 

20

molecule. The two parental types are then fused and only
those recombinants receiving a full set of undamaged mole-
cules can survive, thus eliminating the need for selectable
markers.

The fusion of nonviable parental organisms gives rise
to another method of genetic transfer involving the fusion
of protoplasts with phospholipid vesicles containing genetic
information. Such a method has been developed by Fraley 2

et al. (I979). In this study liposomes containing pBR322

 

DNA were fused with 8. coli. Such a procedure could be

adapted for use with other microorganisms.

In addition to fusion of whole protoplasts and of
liposomes with protoplasts, a great deal of success has been
achieved with the transformation of protoplasts with DNA
(Akamatsu and Sekiguchi, I982; McDonald and Burke, I984;
Kondo and McKay, I982). The procedures used are generally
similar to those used for protoplast fusions.

Protoplast fusion and transformation have become
increasingly important as a means to improve strains of
organisms for which no other genetic exchange system exists.
There are presently numerous examples in the literature
where these methods are now being applied to improve or
enhance various properties of organisms with no other
exchange system. A second purpose for the development of
such systems is to develop a functional map of the bacterial

chromosome. One notable effort has been made in this

21

direction by Stahl and Pattee (I983a, I983b). Location of
various markers was first done using protoplast fusion.
Calculation of frequencies of recombinants and possible
linkages of markers was done directly by entering results
of 9 and ID factor crosses into a computer. Results were
then confirmed by transformation of competent cells of

Staphylococcus aureus.

 

In summary, protoplast fusion has been utilized for
the exchange of genes and the mapping of the chromosome in
a variety of organisms. Because other avenues of genetic

exchange do not appear to exist for Brevibacterium lacto-

 

fermentum, protoplast fusion could fill this void. In

 

addition, protoplast fusion can provide a method of mapping

for Brevibacterium lactofermentum. As such protoplast

 

fusion could provide a convenient method for strain improve-

ment via the genetic manipulation of this microorganism.

MATERIALS AND METHODS

Strains

Brevibacterium lactofermentum strain 2256 was kindly

 

_..€-n.iI

supplied by Dr. H. Momose, Central Research Laboratories,

Ajinomoto Co., Japan. Mutants derived from this strain

 

were obtained by N—methyl-N-nitro-N-nitrosoquanidine (NTG)
R

fgfialfi‘fl

mutagenesis. Str strains were resistant to lOO pg/ml
streptomycin. RifR strains were resistant to IO ug/ml

rifampsin.
Media

Complex media (CM) was that used by Momose and Takagi
(I978) and had the following composition per liter: yeast
extract, l0.0 9; glucose, 5.0 9; NaCl, 5.0 g; and poly-
peptone, I0.0 g at pH 7.0. Minimal media (MM) per liter
composition was: glucose, 20.0 g; (NH4)ZSO4, I0.0 g; urea,
2.5 g; KH2P04, l.0 g; MgSO4°7H20, 0.4 g; biotin, 50.0 pg;
thiamin-HCI, 200 pg; NaCl, 50.0 mg; FeSO4-7H20, l0.0 mg;
at pH 7.0. Plates of CM or MM media contained 2% agar.
Amino acids were added to a final concentration of 0.0I%
as required by the particular aurotrophic strain. Selective

plates for selection of recombinants were CM with l00 ug/ml

22

23

streptomycin and ID pg/ml rifampsin or MM with I00 pg/ml
streptomycin and I0 ug/ml rifampsin and appropriate amino
acid(s) at a final concentration of 0.0l%. Buffer for the
dilution of protoplasts was a modification of sucrose-
maleate (SMM) buffer used by Wyrick and Rogers (I973). It
contained 0.4l M sucrose, 20 mM maleate and 0.2M M9504

at pH 6.5. All other dilutions were carried out using 0.I M
potassium phosphate buffer at pH 7.0. Lysozyme treatment
solutions were either MM with ID mg/ml lysozyme, 0.2 u/ml
penicillin and 0.4I M sucrose or a modification of that
used by Chang and Cohen (I979) which consisted of SMM plus
35 g/l Difco pennassay media (SMMP) with ID mg/ml lysozyme,
0.3 u/ml penicillin, and 0.4l M sucrose. Filter sterilized
BSA, O.I% (w/v) final concentration was added to stabilize
protoplasts. Lysozyme solutions were made fresh prior to
each use and filter sterilized. Two types of regeneration
media were used. Minimal regeneration media was MM with
l35 g/l Na succinate. Complex regeneration media (CR) was
a modification of that utilized by Wyrick and Rogers (I973)
with a per liter composition as follows: Tris, l2.0 g;

KCl, 0.5 9; glucose, I0.0 g; MgCI2°6H20, 8.I g; CaCl2-2H20,
2.2 9; (group A), Naz succinate, I35 9; (group B), peptone,
4.0 g; yeast extract 4.0 g; casamino acid, l.O g; K2PO4,
0.2 9; (group C). Agar, 0.45% final concentration, was
added to group A. Each group was brought to pH 7.0 and

autoclaved separately, group A and B at ll5°C and group C

24

at l200C. Once again, BSA at 0.I% final concentration was

added to stabilize protoplasts.

Protgplast Formation

 

Development of protonlast procedure was modified from
those utilized by Kaneko and Sakaguchi (I979) and Shtannikov
et al. (I98l). An overnight culture grown in either MM or
CM was inoculated into the same type media to an initial
00(575) of 0.150 - 0.200 and incubated on a shaker at 3200
until an 00(575) of approximately 0.600 was reached. This
00 corresponds to about IO8 cells/ml. Fresh filter sterilized
penicillin was added to give a final concentration of
0.3 u/ml. Cells were cultivated for one more generation
period (l.5 - 2.0 hrs), then washed one time with SMM
buffer, resuspended in one of the lysozyme solutions and
incubated without shaking at 320C overnight. Rate of
protoplast development was determined by taking samples
from lysozyme solution at various intervals and diluting
with potassium phosphate buffer to lyse all osmotically
sensitive cells. Samples were then plated on CM to deter-

mine the number of osmotically insensitive cells remaining.

Reversion of Prot0plasts

 

To revert protoplasts, lysozyme treated cells were
washed one time with SMM plus BSA and resuspended in a volume
of SMM equal to the starting volume. The sample was then

diluted appropriately in SMM plus BSA and plated in ID mls

25

of regeneration media plus BSA. Plates were incubated
upright at 320C for up to 2 weeks to ensure all prot0plasts
capable of reversion had grown. This number represented
regenerated protoplasts plus any remaining osmotically
insensitive cells. To obtain % regeneration, this total
was corrected for the number of cells remaining osmotically
insensitive as was the initial number prior to lysozyme

treatment.

 

% regeneration = (ggg) x IOO
where “
a = p0pulation from regeneration plates {A
b = population from "whole cell“ plates "'
c = initial population

The number of reverted protoplasts was then divided by the
corrected initial number, and multiplied by lOO to give the

% regeneration.

Microsc0pic Observation of Prot0plasts

 

Protoplast development was monitored on a phase micro-
scope. Direct cell and protoplast counts were done with a
Petrof—Hauser counting chamber. Protoplast reversion
was observed by taking samples at various time intervals
from the soft agar regeneration media plates, placing on

a slide and covering with a cover slip.

Protoplast Fusion

 

Prot0plasts of two strains, Met'rifR and Arg’strR,

R

or Thr'rifR and Arg'str were mixed (I ml each type) and

26

centrifuged. They were then resuspended in one-tenth the
volume of SMMP at 37°C. Two mls of polyethylene glycol
(PEG mw 6000, MCB, 33% w/v in SMM without MgCl) and filter
sterilized) prewarmed to 37°C was added and rapidly mixed
with the protoplasts. The solution was then incubated

30 mins at 370C. After incubation, 8 mls of SMMP was added
to the PEG - protoplast suspension. Prot0plasts were then
centrifuged and resuspended in 2 mls of SMMP. The fused
protoplasts were then plated for nonselective regeneration
by plating l ml per plate of IO mls regeneration media.
These plates were incubated 3 days at 32°C. After 3 days,
IO mls phosphate buffer (0.I M, pH 7.0) was added to each
plate and homogenized with the agar by quickly drawing the
suspension up and down a l0 ml pipet. Once homogenized,
the sample was diluted appropriately and spread plated on CM
plates to determine the population of regenerated cells and
on selective media to determine fusion and recombination.
Controls were run in exactly the same manner except 2 mls of
only one type mutant was used per fusion. To determine if
protoplasts were transformed as well as fused, DNase I
(Sigma) was added to the prot0plast mixture at a final
concentration of IO mg/ml prior to and during PEG treatment

in some experiments.

Radiolabeling of Cell Wall

 

Cells were grown overnight in MM plus 0.0l% lysine and

transferred to fresh MM plus 0.0l% lysine plus IOO pI of

27

of 3H—DL-mesodiaminopimelic acid (Research Products Inter-
national Corp., Mt. Prospect, Illinois, 2 pCi/ul 32 pCi/mMol)
at a starting 00(575) of .2. Cells were then cultivated and
protoplasted as before. One ml samples were taken prior to
lysozyme treatment and at various time intervals during
lysozyme treatment. Controls were treated in exactly the
same manner except lysozyme was left out of the lysozyme
solution. Samples were centrifuged and washed 2 times with
SMM and resuspended in l N HCl. 0.5 mls of each sample
(duplicates) were placed in Iypholyzation vials, flushed
with N2, and sealed. The sealed vials were heated at IIOOC
overnight. Ten pl of each hydrolysed sample was counted
directly in a 20 ml glass vial with IS mls of aqueous
scintillation cocktail (RPI Safety solve). Fifty pl of
each hydrolysed sample was chromatogramed on thin layer
cellulose plates. The solvent system used was methanol:
H20:pyridine:l2NHCl (l00:35:20:5). Each chromatogram was
cut into 0.5 x l.0 inch fractions. The cellulose was
scraped off each fraction and placed into a 20 ml glass vial
with IS mls of nonaqueous scintillation cocktail (RPI 3a20).
These samples were left overnight in the dark at room
temperature prior to counting. Amino acid standards were
run with the same solvent system and sprayed with %
ninhydrin in acetone and heated briefly to visualize their
locations. DAP was identified by its characteristic green

spot.

RESULTS

Protoplast Formation

 

Logarithmically growing wild type cells of 8. lacto-

fermentum were subjected to variable concentrations of

 

penicillin pretreatment and lysozyme-penicillin treatment
in order to determine optimal conditions for the deveIOp—
ment of protoplasts (Table I). When treated with lysozyme
only, osmotic sensitivity increased with increasing amounts
of lysozyme. However, microscopic examination indicated
poor (<25%) protoplast development at all concentrations of
lysozyme used. When penicillin was added to the procedure,
the proportion of protoplasts increased with increasing
amount of penicillin used. Penicillin at concentrations
$0.3 units/IO8 cells did not inhibit growth. Attempts to
replace penicillin treatment with 2% glycine gave the same
results as treatment with lysozyme only, regardless of the
length of preincubation in media with 2% glycine. Attempts
to use stationary phase cells grown overnight in penicillin
resulted in greater than 99% of the population sensitive
to osmotic shock but only approximately 50% appeared to be
free protoplasts under microscopic observation.

Utilizing 0.3 units/ml penicillin and ID mg/ml lysozyme

at a starting population of IO8 cells/ml, protoplast

28

29

.3mkx n +++ .xmm-mm u ++
.3mmv u + “xmx .cowpm>cmmno owaoumocuwe >3 cmcwsempmv mm3 “cmanpm>mv ammFgouoeau

.ZU co mewmeQ new ewmmzn mpmcqmo;a >3 umcwecmpmu mm3 newsmemL mFqu epocza
o— uzonm mm3 uemsummcw mexNomxp mo pcmam mcu pm cowumngoa Frmu .pcmsummcu wexNOmzp

w uzocozocgu um::_peoo use ucmsgmmcp mEANOmxp op Lowcn meson N gamma mm3 gemEpmmcp
cwppwuwcma .meso; om cow 822m aw mexNomzr spw3 umummcu ecu :2 :3 :3ocm wcm3 mppmum

._E\

 

 

 

 

+++ o_o.o ++ 030.0 ++ 030.0 + m—o.o or

+++ F_o.o ++ «30.0 ++ mmo.o + ow.o m

++ mm.o + No.0 + F.N + mm F

acme mew pews mew upcme mew ugcws mew
-mo_m>we -cmemc Imopw>mc -evesmc Iaopw>mn -cwmemc -QoPm>mu -cmemc

ummpu mppmu “more mprmu pmmpa mPqu ummpa mFqu A \m V
.0 egg m o 3 a .0 can 6 o 3 o .0 038 m o 3 w .0 one u o 3 a PE E

u a F c \ u a P g a p a P g \ p n F g \ msx~omxp a

m.o N.O ~.O O cowumchmucou

 

APE\mpw::V
:TFFeuwcma mo cowumcucmucoo

 

m.mumm_a0poca mo mocmcmmaqm
use EzpcmEcmmopump Ezwcmuumnw>mem cw 3u_>wuwmcmm uwuoEmo mo pewsao~m>mo .F mpnee

 

30

development appeared nearly complete in 3-4 hours (Figure l).
Osmotic sensitivity showed greatest increases in the first
few hours and at 4 hours only approximately 0.I% of the
starting population remained osmotically insensitive
(Figure 2). A decrease in 00(575) paralleled development
of protoplasts (Figure 2 inset).
The effects of growth media (CM vs MM) and lysozyme
treatment medium on the protoplast development were
examined (Table 2). MM resulted in substantially fewer
cells (l0 times) remaining osmotically insensitive than did
CM. However, the media used during lysozyme treatment did
not appear to have a marked effect on protoplast devel0pment.
Sucrose concentration has been reported to have an
effect on protoplast development (Kaneko and Sakaguchi,
I979). Two different concentrations were used in this
study to determine the optimal one (Table 3). No difference
in the development of protoplasts was detected regardless
of the media used to cultivate the cells. However, a
slight advantage was detected in the regeneration when

0.4l M sucrose was used as opposed to .5 M sucrose.

Regeneration of Prot0plasts

 

Prot0plasts were plated in CR and examined microsc0pi-
cally after 48 hours of incubation. Two types of patterns
were detected (Figure 3). The first was a large swollen
protoplast-like form with few protrusions that appeared to

be regenerated cells. The second was smaller more numerous

3I

Figure l. Development of prot0plasts. A. Normal cells of
Brevibacterium lactofermentum; B. After 3 hours
in lysozyme solution; C. After l6 hours in
lysozyme solution. The bar represents l0 um.

 

33

 

 

 

 

 

Q.

 

".52

Colony Forming Units/ ml

 

 

 

Hours

Figure 2. Appearance of osmotic sensitivity during lysozyme
treatment. Population of Whole cells remaining
was determined by dilution in phosphate buffer
and plating on CM. Inset: change in 00(570)
during lysozyme treatment.

34

Table 2. Effect of media on the development and regeneration
of prot0plastsa

 

 

Growth Lysozyme Regeneration % Whole % Regeneration
Media Treatment Media Cells
Media Remaining
CM SMP CR 1.5x10-I 1.2
CM SMP MM 1.5x10"1 O.6I
CM MM CR 3.3x10'1 0.6l
CM MM MM 3.3xio‘1 0
MM SMP CR 5.2x10'2 12
MM SMP MM 5.2x10'2 13
MM MM CR 4.0x10-2 7.5
MM MM MM 4.0x10‘2 9.1

 

aCells grown and penicillin treated in growth media indicated.
Lysozyme treatment and regeneration were in media indicated.

Abbreviations as

indicated previously. % whole cells
remaining and % regeneration determined as before.

35

 

 

Table 3. Effect of sucrose concentration on protoplast
development and regeneration.
Media % whole cellsc % regenerationC
remaining

Experiment la

.5 M sucrose 3.0xlO’2 0.4

.4I M sucrose 4.0xl0'2 I.5
Experiment 2a 2

.5 M sucrose l.6xl0' I.l

.41 M sucrose 5.9x10-3 1.2
Experiment 3b _3

.5 M sucrose 6.3xl0 3 3.2

.4l M sucrose 9.4xl0’ 5.7

 

aCells were grown and penicillin treated in CM and trans-
ferred to lysozyme in SMP for development of protOpIasts.

b

Cells were grown and penicillin treated in MM and trans-

ferred to lysozyme in SMP for deveI0pment of protoplasts.

CWhole cells remaining was determined by dilution in
phosphate buffer and plating on CM.
determined by dilution in SM and plating on CR.

% regeneration was

Figure 3.

36

Regeneration process of protoplasts. Prot0plasts
were plated in CR. After 48 hours of incubation
(32°C) samples were withdrawn and inspected.

(A) Large type prot0plast that often developed.
(B) Small protoplast with larger amounts of
regeneration. The bar represents lO pm.

37

   

38

protoplast-like forms with many pleomorphic cells in Close

proximity in addition to what appeared to be regenerated
cells. Colonies on CR plates usually appeared in 3 to 5
days. However, plates were kept incubated two weeks to
ensure that any reverting protoplasts had sufficient time
to develop. Microscopic examination of colonies on CR
plates revealed a variety of pleomorphic forms in addition
to normal rods and some remaining coccoid shapes.

Agar depth appeared to effect regeneration rates of the
protoplasts (Figure 4). Maximal regeneration rates appeared
to be at l0 mls of agar per petri dish. Rates dropped
sharply with increasing agar depth after l0 mls. The
addition of catalase (.02% final concentration) or peni-
cillinase (l unit/ml final concentration) to CR had no
effect on regeneration.

Effect of growth media on the development of proto-
plasts was previously pointed out as quite dramatic
(Table 3). There was also a noticable effect of this
parameter on regeneration. Regeneration rates ranged from
l3-7.5% when MM was used as the growth media as opposed to
1.2-0% when CM was used as growth media. In addition,
although lysozyme treatment, as stated previously, did not
appear to effect protoplast development, it did appear to
affect regeneration. Regeneration rates were approximately
2 fold higher when SMMP was used for lysozyme treatment

than when MM was used.

39

 

3.8

3.6

9”
n:

Wild Typo
Colony Forming Units x107

.09
CD

2.6 '

 

 

 

 

Figure 4.

~6.4
O
E
16.0 °
3 a;
a1
3"
<5.6 3§
5'8
19 a
.52 E
S
x
d413 E;
m
.44
«14.0
s 10 1‘5 20
ml agar

Regeneration of wild type and His-mutant proto-
plasts vs depth of agar: Prot0plasts were pour
plated with various agar amounts into petri
plates. Wild type H , His-mutant

 

40

Regeneration media (CR vs MM) itself had a slight
effect on regeneration rates. CR appeared to have a slight
advantage when CM was used as the growth media. When MM was
used as the growth media the tendency was reversed.

Finally, the effect of bovine serum albumin (BSA) on

the regeneration of protoplasts was tested (Table 4). In

II. x _
.’
h
'.
p

each of three separate experiments where BSA was added at a
final concentration of 0.I% w/v to both the lysozyme treat- 3
ment and regeneration media, approximately a 3 fold increase .

in regeneration rates was observed. 6

 

Variation in Mutants

 

All investigations of conditions for the development
and regeneration of protoplasts were done using the wild
type strain as a model. Optimized conditions were then
applied to mutants developed for fusion. Figure 5 shows
the rate of protoplast deveI0pment measured as % osmotically
insensitive cells remaining vs time for 3 different mutants
plus the wild type. For each different mutant a different
rate was observed. Regeneration rates of mutants appeared
to be related to the completeness of prot0plast development.
This effect was further investigated by labeling studies

with the wild type strain.

Determination of Remaining Cell Wall and Relationships to

 

Regeneration

 

The relationship between fraction of cell wall remaining

and regeneration was examined. Prot0plasts were developed

4I

Table 4. Effect of bovine serum albumin on the regeneration
of prot0plastsa

 

% whole cell % regeneration
remaining

 

Experiment I
-2

 

 

w BSA 7.2xl0 2 3.9
w/o BSA 4.8xl0' 1.2 ,3?
Experiment 2 1
w BSA 1.3x10" 1.2
w/o BSA 3.0xl0'2 .39
Experiment 3 1
w BSA 2.1x10'1 11 y
w/o BSA 2.2x10' 3,4 .ee

 

aCells were cultivated and penicillin treated in CM,
lysozyme treated in SMP and regenerated on CR. BSA was
added to lysozyme solution and CR at a final concentration
of 0.I% w/v. % whole cells remaining and % regeneration
was determined as before.

42

 

%Whole Cells Remaining

20

Figure 5.

 

 

 

 

b A
I
b
b ‘ t
F a
. A A j I ._A
2 4 t I

% Osmotically insensitive cells remaining vs
length of lysozyme treatment. Cells were
treated for the deveI0pment of protoplasts
as described in the materials and met ods.
“Wild ty e. HLys'. O-OHis-Str.
D—Dmet'Rif .

 

43

as stated in Materials and Methods. Concentrations of
lysozyme and penicillin were varied in order to obtain
intervals in which protoplast formation was still incomplete.
Initial cell concentration was always I08/ml.

In Figure 6 the results of 5 separate experiments with
variable treatments as stated in the legend are plotted.
It can be seen from these results that as total 3H decreases, Im“
the % regeneration also decreases. The correlation

coefficient was 0.9 (P<.005). Figure 7 illustrates

the relationship between osmotic sensitivity and 3H

 

remaining. Osmotic sensitivity appears to drop sharply 3!

3

after more than 4 % of the H has been removed.

To determine that the total counts in samples were due
to (3H) DAP, hydrolyzed samples from experiments using

5 mg/ml lysozyme with 0.2 units/ml penicillin were chromato-
gramed and counted. To determine if any (3H) DAP was lost
during the hydrolysis procedure and to determine location

of (3H) DAP two controls were run as described in the legion
of Figure 8. Both controls showed a Single major peak at
fractions 4-5 corresponding to an Rf of 0.24. DAP standard
(unlabeled) were sprayed with ninhydrin and gave an Rf of
0.24. The Rf for lysine was found to be 0.45. The Rf'S of
other amino acids were not determined Since it has been

previously reported that 8. lactofermentum will only convert

 

DAP to lysine and will not break down lysine (Tosaka and

Takinami, I978). In addition to the single major peak, a

44

 

%’H

 

Figure 6.

 

 

A L A g A J A

20 40 5| I0

p
p

% Regeneration

% 3H in protoplasts vs % regeneration of proto-
plasts. Data was obtained from 5 separate
experiments; 2 experiments where I mg/ml lyso-
zyme and 0.2 units/ml penicillin was utilized,C];
2 experiments where 5 mg/ml lysozyme and 0.2
units/ml penicillin was utilized,.; and l
experiment where IO mg/ml lysozyme and 0.3 units/
ml penicillin was utilizedfi) Initial cell
concentration was always I08/ml. Procedure for
development and regeneration of protoplasts as
described in materials and methods. Slope of

the line calculated to be 0.8l using linear
regression. Calculated correlation was 0.9.

%"H

45

 

 

 

 

‘m_n MIL-“'45:"! agl-E'
‘ 7

 

A A 1 n l A

L
b

 

Figure 7.

20 40 3| ll llfl

%Whole Cells Remaining

% 3H in protoplasts vs % whole cells remaining.
Data was obtained from 4 separate experiments.
I experiment utilized l mg/ml lysozyme and 0.2
units/ml penicillin,.; 2 experiments utilized
5 mg/ml lysozyme and 0.2 units/ml penicillin,

l experiment utilized IO mg/ml lysozyme and
0.3 units/ml penicillinKl Initial cell concentra—
tion was always l0 /ml. Procedure for devel0pment
and regeneration of protoplasts as described in
materials and methods.

46

 

 

.FUI
z e :3 eeeeeee>o ee333¥ .38: z e e?
.eeee see; ooope we emeeee eemeeseso uocfip em eeeeee
aqc Azmv we Emcooumsoccu .mw weaned a<oflzmv mo Emcaoumsosgu .<m mssmwa
awn—2:2 20:05.“.
= = . a . N 2 2 . a . N

. L~
. 3
. ..

(.00!) mo

L
A
N
—

 

 

 

 

l
’
—

 

 

 

 

 

47

minor contaminant was detected at fractions lO-ll. This
contaminant represented l.3 and l.7% of the total counts
respectively.

When hydrolysed samples were chromatogramed, three
radioactive peaks were apparent (Figure 9). Identical

results were obtained in duplicate experiments. The first

 

peak can be identified as DAP, the second as lysine, the :33
third peak is unknown, although its location corresponds i
to that of the contaminant in the (3H) DAP standard. The E
intensity of all three peaks decreases over time in the I
lysozyme treated cells. The DAP peak appeared to decrease E3?

faster than the other two. Controls also showed some loss
of label, however not as much as lysozyme treated cells.
When distribution of counts in each of the three peaks was
examined, the proportion of counts in the DAP peak in the
lysozyme treated cells was seen to decrease with respect
to time whereas it remains high in the controls (Table 5).
Labeled lysine amounts varied between experiments but did
not Show a decrease over time.

When all the various parameters are examined together,

3

loss of H (total, 3H (DAP), and % regeneration seem to

follow similar curves (Figure I0). Osmotic sensitivity

as measured by cell lysis seems to develop faster than

tritium loss. In the controls, some drop in (3H) DAP, and

3H total is also observed. This drop however is not of the

same magnitude as lysozyme treated cells.

48

 

up A

II’

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LJ

 

 

 

l1

FRACTION POWDER FRACTION "UNDER

Figure 9.

Cgromatograms of hydrolysed cells labelled with
( H) DAP. Log phase cells were labeled with
(3H) DAP then treated with 5 mg/ml lysozyme and
0.2 units/ml penicillin. A. Hydrolysed cells
prior to lysozyme treatment. B. Hydrolysed
cells after 3 hours Iysoz me treatment. , and
3 hour untreated control . C. Hydrolyzed
cells after 6 hours lysozyme treatment 0 , and
6 hour untreated control I. D. Hydrolyzed
cells after I8 hours lysozyme treatment 0, and
I8 hour control, I .

 

49

 

.::L 0303 mucme3ewqxo 0300330=o .mxmma

00353 330 :3 mpczoo 30003 30 mcwcw>3n 0:0 3000 0:0 :3 magnoo mewcum >0 00:3530300 003

3000 £000 :3

030200

.353wo3 003 coppmsucmucou 3300 30333:3
30330 2333303300 353033;: 3.0 new 05330033 ~E\mE m 5333 vapmmeu m3m3

.000 30

0 0033 00330003
03300 among 003

 

 

 

3.03 0.03 330. 30.3 3.03 0.03 .0e0000 e; 03
3.03 3.03 330. 33.0 3.03 0.03 30e0000 e0 0
0.03 3.33 330. 03.0 3.03 0.33 30e0000 e; 3
0.00 0.00 330. 00.0 3.0 0.03 0E3~0m33 e0 03
0.33 0.03 000. 30.3 0.33 3.33 02330033 e; 0
0.03 3.33 030. 30.0 0.03 3.33 0E3~0m33ee 3
0.03 0.00 030. 33.0 3.33 0.33 3030303
axe ucm axm 303 0x0 0:3 0x0 amp qu vcm axm amp
030:302 033 aqo

 

.mEmsmoumsocgu 3300 003330303; 30 300a 3000 :3 030300 mo cowpsnwspmwo .m w3003

50

0 .000. $00 34% .3208 330: :3 30.0338 03003 30.0 420003003

we: 3... wumfaouoen mo :ozmcmcmmmc a: 33 Army 3414 .23:on 30.30.:

:m a 0.0 3:03.300: 3.3303805 3:332:03 03:8 30:3 30.0 500.5005

oEANomapu<op meamva .muozpme use 030330302 :3 cmavcummu mm 003 mcacmuosq

.0E330033 000003; 0:20 03 0000000000e 03300 00330003 000 3:30 000:

:03 ago: mmpasmm Focucou .cwppwuwewa Fs\033:: ~.o 0:0 05330033 Pe\oe m

003: eeeees» use: 0<0 MI 0 0033 00330003 m_300 00000 000 .03300 00330003

0<o Azmv soc» mummpaouoc mo acmsaopm>mc newest we?» em>o chLesuuo mmmcazu
30:0: 00:0:

2 2 2 2 a o 0 ~ 2 2 2 2 . o o

I 1‘ l R 4

 

‘ R I

I ‘ R R 4 I
Q

 

 

 

 

 

.op mcsmwm

°/o

 

 

SI

Protoplast Fusion

 

The procedure for fusion of 8. lactofermentum was as

 

stated in Materials and Methods. A nonselective regenera-
tion method was utilized because direct selection with
antibiotic markers resulted in high background rates.
Frequency was determined by dividing the number of colonies
that appeared on selective plates by the population added If
to each plate. Population was determined by plating various I
dilutions on CM plates.

Initially, two different fusion procedures were compared.

 

The first was based on that described by Kaneko and L:
Sakaguchi (I979) and the second was based on that presented
in a patent by Ajinomoto Co., Japan (I983) (Table 6). For
the first procedure, 33% PEG at an average m.w. of 6000
plus 5 mM EDTA at pH 7.0 was used. For the second proce-
dure, 33% PEG at an average mw of 6000 plus l0 mM CaClZ at
pH 8.0 was used. The data indicated a much higher frequency
of recombination for the first procedure than for the
second. In fact, the frequency of recombination in the
crosses for the second procedure are not substantially
different from the control values.

The effect of temperature on fusion and recombination
was determined in three separate experiments (Table 7).
From these three experiments a trend toward higher frequency
rates can be detected with increasing temperature. The

highest frequency was obtained at 37°C in Experiment 2.

52

Table 6. Comparison of two methods for prot0plast fusion.

 

Frequency of StrRRifR

 

 

PEG+EDTA PEG+CaCl2
Arg'StrR x Arg'StrR 4.lxl0'7 I.3xlO-8
Met-RifR x Met'RifR <I.6xl0'° <2.1x10'8

in cm nib-1.3:} .' . I

Arg‘StrR x Met-RifR 2.8xlO"6 l.6xl0-8

 

Table 7.

53

Effect of temperature on fusion
frequency.

and recombination

 

Experiment I

RxArg’StrR

R

Arg'Str
Met'RifoMet'Rif

R R

Arg’Str xMet'Rif

Experiment 2
Arg'StrRxArg’StrR
Met-RifoMet'RifR
Arg-StrRxMet'RifR

Experiment 3

Arg-StrRxArg'StrR
Met'RifoMet'RifR
Arg-StrRxMet'RifR

25°C
7.3xl0-
<2.3xlO—
l.6xl0-

-8
8

<I.4xl0
2.3xIO-

370

C
l.7xlO'
<1.9x10’

1.7x10'

8
8
7

7
8
6

32°C
5.8xl0
1.1x10'
1.8x10'

<l

<l

.8xl0-
.2xl0-
.4xl0'

.6xl0'
.9xl0-

37°C
7
8

7

37°C
9.4x10'8

<1.7x10‘8

2.8x10'6

45°C
8
8

.leO'6

 

3

u ‘f‘ -AMM"‘.2 A: 1
a

pm;
i

54

Frequency of recombination did vary from experiment to
experiment even when the same conditions were maintained.
However, frequency of crosses were consistently higher than
frequency of controls.

Table 8 gives the effect of different average molecular
weights of PEG on fusion and recombination frequencies.
The highest frequency was obtained when PEG mw 6000 was
used. Fusion with PEG mw IOOO did not result in a
frequency different from the controls. Length of PEG

treatment was also investigated (Table 9). For this

 

experiment PEG mw 6000 was used. No substantial differences L2
were detected over the time range tested. All frequencies
were in the same order of magnitude.
Another factor investigated was the effect of protoplast
development on fusion and recombination frequency (Table IO).
There appears to be a slight increase in frequency after
5 hours lysozyme/penicillin treatment as compared to 2
hours or l6 hours.
When all fusion experiments are examined as a whole, the
R)

frequencies of crosses (Arg'StrRxMet'R4 were in the range

of IO'°. The best frequencies ever obtained were about

l0'5. These frequencies can be compared to control fre-

quencies of ID"8 to IO’7. Frequencies of appearance of the
StrRRifR phenotype was l.9xl0'6 when an alternate mutant

Thr'RifR was used. This is comparable to the frequency

obtained with the Met'RifR mutant. Addition of DNase to

55

Table 8. Effect of different molecular weight PEG on the
fusion and recombination frequency.

 

Average molecular weight PEG

 

 

6000 3000 IOOO
Arg’StrRxArg'StrR 6.9xl0"8 8.8xl0-8 8.3xl0-8
Met-RifoMet'RifR <2.5xI0-8 <2.8le8 <2.6le8
FT?
Arg-StrRxMet'RifR 3.0x10'5 7.3x10'7 <3.Bx10'8

 

 

56

Table 9. Effect of length of PEG treatment on fusion and
recombination frequency.

 

Length of Treatment Frequency of StrRRifR
(33% PEG mw 6000)

 

30 min 3.0 x 10'5
IS min l.2 x l0"5
5 ill
5 min 2.5 x l0-

1 min 1.2 x 10'5

 

 

57

Table l0. Effect of protoplast devel0pment on fusion and
recombination frequency.

 

Frequency of StrRRifR

 

 

Length of - R - . R - R
Lysozyme/ Arg Str Met R1f Arg Str
Penicillin x R x , R - , R
Treatment Arg Str Met R1f Met R1f
2 hours l.2xl0-8 <I.3xl0'8 6.9xl0'7
5 hours 3.9xio‘8 <1.3x10'8 1.2x10'6
- - -7
15 hours 5.2xl0 8 <I.le0 8 4.2xlO

 

 

58

the fusion procedure had no effect on the frequency of

recombination.

 

DISCUSSION

8. lactofermentum cells are resistent to treatment with

 

lysozyme. Some method of pretreatment is required in order
to obtain good protoplast development. Glycine has been .
used by others with some success for the development of
protoplasts (Baltz, I978; Hopwood, l98l; Udou et al.,

I982). Glycine is thought to replace alanine in the t

 

peptidoglycan structure therefore preventing crosslinking ;~
and thus resulting in cell walls more susceptible to
lysozyme treatment. In these experiments with 8. lacto-

fermentum, glycine did not result in good development of

 

prot0plasts as determined by microsc0pic inspection.
Therefore penicillin was utilized for development of

protoplasts from 8. lactofermentum in a procedure Similar

 

to that used by Kaneko and Sakaguchi (I979). Care was

taken that penicillin at concentrations used for the
devel0pment of protoplasts was not inhibitory to growth of

the organism. An attempt was made to utilize an overnight
culture of stationary cells grown in media containing low
levels of penicillin. It was hoped that such cells might

be more resistant to harsh conditions of treatment and
therefore result in higher regeneration rates. Unfortunately,

cells treated in this manner also had poor devel0pment of

59

6O

protoplasts as determined by microscopic inspection.
Several different combinations of penicillin treatment
concentrations and lysozyme treatment concentrations were
examined. The best treatment was the most severe attempted
which was 3 u/ml penicillin and ID mg/ml lysozyme. Higher
concentrations were not utilized due to possible inhibitory
affects to cell growth. Once penicillin and lysozyme 3‘
concentrations were established, the effect of media was
examined. Growth conditions have been reported to effect

not only the development of protoplasts but there regenera-

 

tion aS well (Baltz, I978). In this study, the growth media
and media used to lysozyme treat cells did not display much
of an effect on the development of protoplasts. In each
case far less than I% of the cells remained osmotically
insensitive. However, the regeneration of protoplasts seemed
dramatically affected. Growth in a defined media resulted in
regeneration that was I0 fold higher than with growth in a
complex media. Regeneration media, by contrast, did not have as
dramatic an effect unless prot0plasts first grown in complex
media were plated for regeneration in a defined media. Such
protoplasts exhibited very poor regeneration rates.
Previously published accounts of prot0plast development

and regeneration in Brevibacterium species indicate that a

 

slight reduction in sucrose concentration resulted in
improved development of protoplasts (Kaneko and Sakaguchi,

I979; Shtannikov et al., l98l). However no such effect was

6l

observed in this study. Protoplast development was virtually
the same under both concentrations utilized. A slight
advantage was detected in regeneration rates when the lower
concentration was used, therefore the reduced sucrose
concentration was utilized in later procedures.

Initial attempts at protoplast regeneration often

resulted in erratic rates even between duplicate samples.

‘1'!

It was then discovered that the depth of regeneration agar
influenced the rate of regeneration. Maximal regeneration

seemed to occur when IO mls of agar per petri plate was

 

used. It could be hypothesized that less than this amount
probably resulted in too much drying at the surface and

more may have caused limitation of oxygen to the regenera-
ting protoplasts. It was also hypothesized that accumulation
of H202 might cause the decrease in regeneration rates.
Accumulation of H202 has been Shown to affect enumeration

of stressed Staphylococcus aureus cells (Flowers et al.,

 

I979). This hypothesis was eliminated when addition of
catalase had no effect on regeneration rates.

In the current literature, a number of studies have
concluded that the use of BSA can greatly enhance the
regeneration of protoplasts (Gabor and Hotchkiss, I979;
Gotz et al., I98l; Minton and Morris, I983). In this study
a similar effect was noted. BSA increased regeneration
rates approximately 3 fold in every instance. A possible

explanation for this affect is that BSA may act to protect

62

the proteins at the protoplast surface.

When regenerating protoplasts were examined micro-
sc0pically, two predominant patterns seemed to emerge.
One was a rather large protoplast many times the size of a
normal cell. These on occasion had smaller more rod shaped

forms in close association with them. More common, however,

were smaller protoplasts that seemed to elongate or "bud"
with more rod forms adjacent to them. One possible

explanation is that the larger forms may be protoplasts that

r
M
i
I
III
1

have lost the capacity to revert or divide, therefore merely

accumulate more mass.

 

As noted previously, all optimization studies were run
with the wild type strain. When mutants were used,
variable results were obtained. Much of the variability
in the development of protoplasts could be attributed to
growth rates in MM and to somewhat variable penicillin
sensitivity as a possible function of growth rate. Two
solutions present themselves as methods to eliminate any
detrimental effects of this variability. One is to optimize
the procedure for each mutant used. The other is to choose
mutants with as Similar a response as possible to the
response of the wild type under optimal conditions developed
for the wild type.

Another revelation of the variability in mutant response
was that regeneration might correlate to incomplete protoplast

devel0pment. To further study this relationship the cell

63

3H) DAP. These cells

walls of wild type were labeled with (
were then rendered protoplasts under various conditions

and their ability to regenerate evaluated. Although a

clear relationship appears to exist between remaining cell

wall and regeneration capacity, it is not possible to deter-

mine an exact amount of cell wall remaining required to

promote regeneration. This was due in part to a large r:
unidentified peak which appeared in Chromatograms of cell

fractions thus making exact analytical calculation impossible.

This unidentified peak corresponded exactly to a very small

 

contaminating peak found in the commercially obtained
(3H) DAP. The peak did not appear to be a breakdown
product as further harsh treatment of (3H) DAP (incubation
at IIO°C in 6 N HCI) did not result in any change in the
proportion of counts found in the contaminating peak. A
possible explanation for its increased occurrence in the
wild type cell hydrolysates may be that the cells preferen-
tially take up the labeled contaminating compound. Never-
theless, even with the contaminating fraction, an analysis
of the % of counts in each peak of chromatogramed samples
indicates that DAP is preferentially lost during lysozyme
treatment. Some overall label count is also lost. This
may be due to some cell lysis or some leaking of inter-
cellular components due to loss of integrity of the cell

wall. The total % of labeled lysine also varies between

the two experiments Shown in Table 5. Although the exact

64

cause of this variability is unknown, the amount of lysine
appears to be unrelated to penicillin/lysozyme treatment.

For fusion of bacterial protoplasts a sequence of
events has been hypothesized as stated previously (Ahkong
et al., I975; Frehel et al., I979; Gumpert, I980; Knutton
and Pasternak, I979). The first step requires that the
protoplasts are brought into close contact with each other.
Generally, such close contact is brought about by the use
of PEG. The next step is the formation of molecular
contacts and alteration of membrane structures. This step
may also be enhanced by the action of PEG and may also be
affected in some procedures by the addition of divalent
cations such as Ca2+ (Tilcock and Fisher, 1979). The
final phases of fusion involve the formation of a fusion
membrane and an intermixing of cellular contents. The
success of this final phase may depend on removal of PEG
(Ferency, l98l; Wojcieszyn et al., I983).

Ultimately, in order to determine the success or failure
of any particular fusion procedure some method must exist
to determine if fusion has occurred. In the fusion of
eucaryotic cells this method is often visual, making
determination of fusion relatively simple. However, for
procaryotes, a visual method cannot be utilized due to
the much smaller Size of the cells involved. As a result,
determination of the success of fusion often relies on the

appearance of markers, such as antibiotic resistance in the

65

recombinants. This method of determination is then not a
simple measure of fusion, but also of recombination and of
the expression of any particular markes. Therefore, fusion
expressed as the frequency of appearance of phenotypic
markers may actually reflect poorly on the actual frequency
of fusion within a p0pulation. Nonetheless, it remains the
best method available and as such has been used to determine ET
the relative rates of fusion in these studies.

For the fusion of Brevibacterium Species protoplasts

 

and protoplasts of closely related organisms, two basic

 

Fl . Ml; -3'4‘~ ..n '. w’u'a .‘n' I- '~. ...'
i

procedures have been described. The first involves the use
of 33% PEG plus I0 mM CaClZ at a pH of 8.0. In this study
these procedures were compared. The procedure using high
concentrations of CaCI2 resulted in rates of recombination
no higher than that expressed by controls leading to the
conclusion that fusion had not occurred under these condi—
tions. This procedure was therefore discarded and future
fusions were run using the first procedure.

AS mentioned previously, mutant types did express
somewhat different behavior than did the wild type when
protoplasted and regenerated using the method optimized for
the wild type. In addition, it was noted that although
conditions were standardized as much as possible, the
success or completeness of protoplast development did vary
somewhat from experiment to experiment. This variation

made it difficult to compare experiments run with different

66

sets of protoplasts. Therefore, comparisons of frequencies
under differing conditions could only be made when the
protoplasts used were all of the same lot.

When the effect of temperature on fusion and recombina-
tion frequency was examined, three separate experiments
were run. Background frequencies varied but never exceeded
3.0xl0'7. Recombination frequencies achieved a rate of
2.8xl0'6 under best conditions. This seemed to occur at
370C. When the recombination frequency at 37°C was compared
to frequencies at other temperatures, the frequency at 37°C
was always the highest within any one particular experiment.
However, variation of recombination frequency at 37°C
between the three experiments was quite large and recombina-
tion frequencies at different temperatures (both higher and
lower) using a different lot of protoplasts did exceed
frequencies at 37°C. This indicates the possible relation-
ship between fusion frequencies and state of development of
prot0plasts.

The effect of molecular weight of PEG and length of PEG
treatment on fusion and recombination frequency was also
examined. No clear cut difference could be established
between the different time ranges tested. Once again,
variability in recombination frequency was high. Molecular
weight did appear to affect the frequencies somewhat.
However, it should be noted that the various molecular

weight ranges of PEG were obtained from different

67

manufacturers and may have contained variable amounts of
impurities. Some researchers have claimed that PEG by
itself cannot cause fusion (Honda et al., l98la). As
such, the results obtained here may reflect a difference
in the amount of impurity present rather than an actual
effect of molecular weight. In addition, it should also
be noted that the % PEG at each molecular weight was held :-
constant. One explanation hypothesized for the success of I
fusion with PEG treatment is that PEG destabilizes membranes

by the removal of water (Gibson and Strauss, I984). This g

 

effect may be more pronounced with higher molecular weight
PEG when compared to lower molecular weight PEG when the
concentration of the lower molecular weight PEG is not
increased. Lower molecular weight PEG is more commonly
used in eucaryotic cell fusion due to its decreased
viscosity which increases the ease of handling of PEG.
Concentrations used are generally higher than 33%.
Finally, an attempt was made to determine if the age or
developmental state of the product could affect the
frequency of recombination. A small effect was observed.
Frequencies were highest after 5 hours of lysozyme treat-
ment. This would correspond to a point in time where
protoplast development would be maximal. Rates may drop
after this period of time due to decreased viability of
the protoplasts over time. Prior to 5 hours, there may be

too much cell wall remaining, thus interfering with fusion.

CONCLUSION

Prot0plasts can be developed from logarithmically

growing cells of 8. lactofermentum using a combination of

 

penicillin and lysozyme treatment. The concentrations of
penicillin and lysozyme required for good devel0pment of
protoplasts are 0.3 units and ID mg per l08 logarithmically
growing cells respectively. The protoplasts developed in
this manner can be effectively regenerated in a complex
semisoft agar media using a pour plate method. Regenera-
tion of protoplasts is aided by the addition of BSA to the
regeneration media.

Optimal regeneration of 8. lactofermentum protoplasts

 

appears to be related to the degree of protoplast
devel0pment. A correlation was found between cell wall
remaining and regeneration rates, the less cell wall, the
lower the regeneration rates. This could indicate that
some fraction of ce-l wall is required for prot0plasts to
regenerate effectively.

Prot0plasts can be fused effectively using 33% PEG
(mw 6000) at 37°C. Temperature does play a role in fusion
probably by increasing the fluidity of the membrane.
Recombination frequency is, however, highly variable and

may depend on the state of development of the protoplasts.

68

69

Although fusion can be concluded to have occurred since
recombination frequencies under the best Circumstances were
approximately l0 fold higher than controls, the frequencies
obtained were still quite low. AS a result, it would not
be possible to use this method as developed here for the
mapping of genetic markers based on recombination frequencies.
However, it does remain a viable method for the exchange of
genetic information between various strains and as such
will be a useful procedure for strain improvement of

industrially useful Brevibacterium strains.

 

APPENDIX

APPENDIX

Introduction and Literature Review

 

In conjunction with establishing a possible method to

map the chromosome of 8. lactofermentum by protoplast

 

fusion, an attempt was made to develop a second alternative
method. The method developed was that of nitrosoquanidine
(NTG) sequential mutagenesis which is based on work by

Ward et al. (I970). NTG sequential mutagenesis first
requires the development of a synchronously growing cell
culture. This has been accomplished in a variety of ways.
Ward et al. (I970) uses nalidixic acid, a gyrase inhibitor,
to produce a synchronous culture of 8. £211: Others have
used (I) Size of cell, (2) the tendency of a stationary
culture to begin growing again in a synchronous manner
(Cutler and Evans, I966) and (3) nutrient limitation (Kepes
and Kepes, l98l; Shehata and Marr, I970). Once a synchronous
culture is achieved, successive samples are treated with

NTG (Cerda-Olmedo et al., I968). NTG is believed to cause
mutation at the replicating point of the chromosome.
Therefore, the type of mutant obtained should reflect the
area of chromosome exposed at any specific time. This method

has been used in the past to develop a map of 8. coli

70

7I

(Cerdo-Olmedo et al., I968). The inherent defect of this
method is that only a crude map can be established, due

to the speed at which the chromosome replicates. For
organisms such as 8. £9119 much better methods such as
conjugation are currently available. NTG treatment would
be best suited to a slow-growing organism or one in which
no other method has been established. For example, it has

been used to map just such an organism, Mycobacterium

 

tuberculosis (Woodley et al., l98l). If this method were

 

developed for use in 8. lactofermentum it could provide a

 

means of confirming any results obtained by protoplast
fusion. It could also provide an alternative method of
mapping Should protoplast fusion prove to be unsuitable for

establishing a map of the chromosome.

Methods

8. lactofermentum was cultivated in CM media or CM agar

 

plates (See Materials and Methods). Overnight cultures

were transferred to 5 mls of fresh CM media. After approxi-
mately 2 hours, Nalidixic acid was added to a final concen-
tration of 250 pg/ml to synchronize the culture. Cells

were removed by filtration and washed with fresh media,

then resuspended in fresh media. Fifty ul of culture was then
placed on sterile milipore filters (0.45 um pore Size) on

CM plates. For NTG treatment, these filters were placed

on plates containing NTG, 0.05 m Tris-maleate or O.l M

72

potassium phosphate buffer (pH 5.5) and 0.45% Difco agar.
After appropriate time treatment periods, filters were
placed on plates with 0.I M potassium phosphate (pH 9.0)
and 0.45% Difco agar for 5 minutes to destroy residual NTG.
Filters with cells were then transferred to CM agar

plates for l hour to allow expression of mutation, then
finally transferred to selective media plates. Selective
media plates were MM or CM plus streptomycin or rifampcin

(See Materials and Methods).

Results
The optimal concentration nalidixic acid (Nal) was
determined as that which completely inhibited growth of 8.

lactofermentum (Figure I). For 8. lactofermentum this was

 

 

determined to be 250 ug/ml. This concentration was found’
to successfully result in synchronous growth as Shown by
viable cell counts over time (Figure 2).

The amount of NTG required and the length of treatment
required to induce reverse mutation of a his' mutant of

8. lactofermentum by a filter rotation method was determined

 

as shown in Tables I and 2 and in Figure 3. A concentration
of 50 pg/ml NTG in the agar for 2 minutes or less seemed to
give best results. It was necessary to destroy residual

NTG on the filters by brief exposure to agar at pH 9.0.

The length of this treatment was determined as shown in

Table 3. The effect of this treatment on cell viability

73

 

 

 

 

 

 

 

 

1.5 $
5 D
l.2 "
E u r .
D
C) r a
0.4 r 9
C
QJI'I’ #7:
I ’9’“ I
o 4 I 12
Hours

Appendix Figure I.

Effect of NAL on the growth of strain
2256. Cultivation was carried out in
CM supplemented with various concentra-
tions of NAL at 30°C with Shaking.”
250 ug/ml NAL. A—A l25 ug/ml NAL.

CHI] 75 ug/ml NAL. 0050 ug/ml NAL.
H 0 ug/ml NAL.

74

 

 

 

 

 

 

 

Appendix Figure 2.

O C
«It:
0
.0 an
O.
O
W
h
3
464°
3:
O O
or
.0 d—
O
O
5 E
TIC-1'3

Increase in viable cells of strain 2256
after NAL treatment. Cultivation was
as in Figure l. After treatment of
cells with 250 ug/ml NAL for 4 hours,
NAL was removed by centrifugation and
washing of cells with MM devoid of
nitrogen. Cultivation was reinitiated
at a cell combination of l.3xl08/ml.

75

Appendix Table I. Effect of NTG concentration on mutation
frequency and cell viability in the
membrane filter rotation method.

 

 

Concentration No. of his+ No. of surviving Frequency
of NTG (ug/ml) per filtera cells pEr of his+
filteE reverse
(x l0 ) mutation
(a/105 b)
-7
O 9 i 4 275 i 6 3.3 X IO
25 72 a 7 235 a 15 3.1 x 10'6
50 234 a 11 121 + 9 1.9 x 10'5
100 12 a 3 3 e 2 4 x 10'5

 

Experimental procedure: see text.

a,b: Mean (in duplicate) i standard error.

76

Appendix Table 2. Induction of his+ reverse mutation in
his-I6l using membrane filters and NTG
containing plates (IOO ug/ml)

 

Placement time of membrane
filter on NTG plate (min)

Number of his+ colonies
per membrane filter

 

Non-NTG treatment (control)

53

27

 

Experimental procedure: see text.

77

 

 

 

 

10"r . 10" :1
- 3
.‘.>“. g
E a
a, .2
u. D
o Q
3 3
0 <
C 0
8 E.
g 10"I . 10“ 8
U.

0 540 100
NTG lug/m1)

Appendix Figure 3. Effect of NTG concentration on mutation
frequency and cell viability in the
membrane filter rotation method.

78

Appendix Table 3. Inactivation of residual NTG by placing
membrane filters on pH 9 plate after NTG

treatment.

 

 

NTG treatmenta Condition for NTG Number of
inactivation surviving
pH Period (min) cells per
membraBe
filters
- - - 257 1 l3
+ 7.2 3 25 i 6
+ 9.0 2 I99 i II
+ 9.0 3 228 i 6
+ 9.0 5 209 i 9

 

aMembrane filters were placed for l min at 25°C on a
NTG-soft agar plate containing 200 ug/ml NTG, 0.05 M
Tris-maleate buffer (pH 5.5) and 0.45% Difco agar, then
transferred on to pH 9 or pH 7.4 soft agar plate
containing 0.I M phosphate buffer and 0.45% agar.

bMean (in duplicate) 1 standard error.

79

iS shown in Table 4.

After determination of optimal conditions of treatment,
a few Short experiments were run to attempt mapping with
this method. The results are Shown in Figure 4. The time
required for replication of the chromosome appears to be
approximately 90 minutes. A possible replication map from

these experiments is shown in Figure 5.

Discussion

 

Nalidixic acid provided a convenient method to produce

synchronously growing cultures of 8. lactofermentum.

 

Figure l demonstrates the effect of concentration of Nal on

the growth of 8. lactofermentum. Because Nal treatment is

 

to be used for a relatively short period of time only to
provide synchrony, it was decided that 250 ug/ml Nal would
be better than less repressive doses. Figure 2 illustrates
the synchrony achieved when growing cells were treated for
4 hours at this concentration of nalidixic acid.
Traditional methods of NTG treatment are somewhat
cumbersome and slow. Therefore, a new method was deveI0ped
to facilitate more rapid and accurate NTG treatment of
synchronous cultures. This method involved placing the
synchronous cells on filter disks, treating cells with NTG
by placing these disks on agar containing NTG and allowing
the NTG to diffuse up into the disks. Residual NTG was then

denatured by placing the filters plus treated cells on agar

80

Appendix Table 4. Effect of pH 9 treatment on cell
viability.

 

 

Period of treatment Number of viable cells
(min) after treatment
0 2ll
3 223
5 214
IS 235
90 2l8
I35 2l0

 

50 ul of his I6I cell suspension was applied to each
membrane filter, which were then placed on pH 9 plates.
At intervals, filters were transferred onto CM-ZG plates,
followed by incubating them at 30°C for viable cell
counting.

8I

 

 

A
l9>
3'
$1111»
U
a»
. 1 # A A A
211» B
ISM

 

 

 

Appendix Figure 4. Variation in frequency of streptomycin
resistant (A) and histidine reversion
(B) mutations.

82

 

hisiGi-i str' Iflsisi-Z

I 30

0 . ‘1‘ 3'
l I l I I I I fl
. 21 11 11

Appendix Figure 5.

Replication map of strR

in his I6I.

" ' m (°/o)

and his+ loci

83

at a very high pH for a short period of time. The effective
NTG treatment parameters were determined as Shown in Tables I,
2 and 4. Higher concentration of NTG for shorter time

periods seemed most effective. The optimal NTG concentra-
tion is described using the parameters as shown in Figure 3.
This provides the highest amount of reversion for a particular
surviving fraction. In this case the most appropriate i
concentration is approximately 60 ug/ml NTG. The final A
treatment for inactivation of NTG was determined as described

in Table 2. pH 9.0 for at least 3 minutes was required to

 

‘3.

denature the NTG as shown by the number of surviving cells.
Treatment at this pH was found not to effect cell viability
for up to l35 minutes (Table 3).

Figure 4 gives the results from one experiment using
NTG treatment on synchronously growing cells. The mutant

used was a his” mutant of 8. lactofermentum. The generation

 

time under the conditions used was estimated to be 90-l00
minutes. Within this generation period, one peak was
obtained for streptomycin resistence and 2 peaks were found
for reversion to the WT his+. A tentative map of these
markers is shown in Figure 5. The second his+ peak may
represent a suppression mutation. Unfortunately, all data
using this method was not as straightforward as that
depicted. For the most part, the number of colonies
obtained for any particular peak was low and difficult to

distinguish from background counts. Initial cell counts

84

were hard to standardize and may provide a source of error.
In addition, standardization of NTG treatment is essential
to reproducing results from experiment to experiment and

was difficult with the membrane filters.

Conclusion

 

The method developed here for NTG treatment of synchronous
cultures is faster and more efficient than those described
previously. Treatment of membrane filters on agar at pH
9.0 for a period of time to denature the residual NTG is
essential to obtaining consistent results. Preliminary data
has shown some inconsistency and some problems with low
frequency of mutation over background. However, the results
are promising. A tentative placement of some markers has
been proposed. At this point in time the method may be as

useful as protoplast fusion for chromosome mapping.

a"?

 

T

BIBLIOGRAPHY

 

BIBLIOGRAPHY

Abe, S. and Takayama, K. I972. Amino acid-producing
microorganisms, variety and classification. In 188
Microbial Production of Amino Acids. K. Yamada, S.
Kinoshitu, T. Tounodu and K. Aida, eds. Halsted Press
Publishers, 605 Third Ave., N.Y., N.Y. lOOl6.

 

73913:”
'7

Ahkong, Q.F., Fisher, 0., Tampion, W. and Lucy, J.A.
I975. Mechanism of cell fusion. Nature 253:l94-I95.

Ajinomoto Co., Inc. l983a. Method for fusion of microbial *
protoplast. Jpn. Kokai Tokkyo Koho JP 58 l58, l8l
83,l58,l8l) (Original not seen. Cited from Chem. ,
Abstracts 100 352 (1984)). Er

 

Ajinomoto Co., Inc. l983b. Fusion of microbial protoplasts.
Jpn. Kokai Tokkyo Koho JP 58, l58, I86). (Ori inal
not seen. Cited from Chem. Abstracts l00:350 Il984)).

Akamatsu, T., and Sekiguchi, J. I982. Transformation of
Bacillus protoplasts by plasmid p TP4 DNA. Agric.
Biol. Chem. 46:l6l7-l62l.

Akamatsu, T., and Sekiguchi, J. I983. Selection methods in
bacilli for recombinants and transformants of intro- and

interspecific fused protoplasts. Arch. Microbiol. l34:
303-308.

Akamatsu, T., and Sekiguchi, J. I984. An improved method
of protoplast regeneration for Bacillus species and its
application to protoplast fusion and transformation.
Agric. Biol. Chem. 48:65l-655.

Arnold, K., Pratsch, L., and Gawrisch, K. I983. Effect of
poly(ethylene glycol) on phospholipid hydration and
polarity of the external phase. Biochim. Biophys. Acta
728:l2l-l28.

Asada, M., Nakanishi, K., Matsuno, R. and Kamikubo, T.
I982. Continuous CoA production with immobilized
Brevibacterium ammoniagenes cells. Agric. Biol. Chem.

 

85

86

Atlas, R.M. and Bartha, R. I972. Degradation and minerali-
zation of petroleum by two bacteria isolated from coastal
waters. Biotech. Bioeng. l4:297-307.

Baltz, R.H. I978. Genetic recombination in Streptomyces
fradiae by prot0plast fusion and cell regeneration. J.
Gen. Microb. l07:93-l02.

 

Baltz, R.H. and Matsushima, P. l98l. Protoplast fusion in
Streptomyces: Conditions for efficient genetic recombina-
tion and cell regeneration. J. Gen. Microbiol. l27:
l37-I46.

 

Basnak'yan, I., Votrin, I. and Bebow, S.S. I98l. Study of T}
the peculiarities of the action of Brevibacterium i'
ammoniagenes endonucleases on DNA. Biochemistry (Engl.
trans. Biokimiya) 46:l054-l06l.

 

 

Bergey's Manual of Determinative Bacteriology, 8th Ed.
Williams & Wilkins Publ. Co., Buchanan, R.E. et al., f
eds. I974. Part l7, pg. 559, Actinomycetes and Related 8;
Organisms.

 

Bugaichuk, Y.D., Zuonigorodski, and Zhdanov, V.G. l98l.
Conditions for the formation of and reversion to

bacillary form of Bacillus lichenoformis protoplasts.
Microbiol. 50:494-497.

 

Cerda-Olmedo, E., Hanawalt, P.C. and Guerola, N. I968.
Mutagenesis of the replication point by nitrosoquanidine:
Map and pattern of replication of the Escherichia coli
chromosome. J. Mol. Biol. 33:705-7l9.

 

Chang, S. and Cohen, S. I979. High frequency transforma-
tion of Bacillus subtilis protoplasts by plasmid DNA.
Molec. Gen. Genet. l68:lll-ll5.

 

Cocking, E.C., Davey, M.R., Pental, D. and Power, J.B.
l98l. Aspects of plant genetic manipulation. Nature
293:265-270.

Cortzee, J.N., Sirgel, F.A. and Lecatsas, G. I979. Genetic
recombination in fused spheroplasts of Providence
alcalifaciens. J. Gen. Microbiol. ll4:3l3-322.

 

 

Collins, M.D., Goodfellow, M. and Minnikin, D.E. I979.
Isoprenoid quinone in the classification of Corneform
and related bacteria. J. Gen. Microbiol. IIO:l27-l36.

 

87

Crombach, W.H.J. l97l. DNA base composition and DNA
homology of Coryneform bacteria isolated from soil,
Cheese and sea fish. J. of Gen. Microbiol. 69:xii.

Cullis, P.R., and Hope, M.J. I978. Effects of fusogenic
agent on membrane structure of erythrocyte ghosts and
the mechanism of membrane fusion. Nature 27l:672-674.

Cutler, R.G., and Evans, J.E. I966. Synchronization of
bacteria by a stationary-phase method. J. Bacteriol. 9l:
469-476.

Demain, A.L. and Soloman, N.A. l98l. Industrial micro-
biology. Scientific American 245:67-75.

Demonty, J.A., Robaye, B., and Calberg—Bacq, C.M. I973.
Envelope structure in three different L-forms of
Proteus mirabilis. Antonie van Leeuwenhoek 39:2l7-228.

 

Eucleigh, D.E. l98l. The microbial production of industrial
chemicals. Scientific American 245:l55-l78.

Ferenczy, L. l98l. Microbial prot0plast fusion, in Genetics
as a Tool in Microbiology. S.W. Glover and D.A. Hopwood,
eds. pp.7l-34, Cambridge University Press.

 

Flowers, R.S., Martin, S.E., Brewer, 0.0. and Ordal, Z.J.
I979. Catalase and enumeration of stressed Staphylo-
coccus aureus cells. Appl. Environ. Micro. 33:lll2-
lll7.

 

 

Fodor, K. and Alfoldi, L. I976. Fusion of protoplasts of
Bacillus megaterium. Proc. Nat. Acad. Sci. 73:2l47-2l50.

 

Fodor, K. and Alfoldi, L. I979. Polyethylene glycol-
induced fusion of bacterial protoplasts: direct
selection of recombinants. Molec. Genet. l68:55-59.

Fraley, R.T., Fornairi, C., and Kaplan, S. I979. Entrapment
of a bacterial plasmid in phospholipid vesicles:
potential for gene transfer. Proc. Nat. Acad. Sci. 76:
3348-3352.

Frehel, C., Lheritier, A., Sanchez-Rivas, C., and Schaeffer,
P. I979. Electron microscope study of Bacillus
subtilis. Protoplast Fusion I37zl354-l36l.

Fu, T.T. and Thayer, D.W. I975. Comparison of batch and
semicontinuous cultures for production of protein from
mesquite wood by Brevibacterium. SPJM 98A Biotechnol.
Bioeng. l7:l749-l760.

 

 

88

Gabor, M.H. and Hotchkiss, R.D. l979. Parameters governing
bacterial regeneration and genetic recombination after
fusion of Bacillus subtilus protoplasts. J. Bacteriol.
l37:l346-l353.

 

Gabor, M.H. and Hotchkiss, R.D. I982. Analysis of randomly
picked genetic recombinants from Bacillus subtilus.
Protoplast Fusion. In Genetic and CeTlular Technology I.
U.N. Streips, S.H. Goodgal, W.R. Guild'and G.A. WiTSon,
eds. Marcel Dekker Inc., Publ.

 

 

Gerlinas, R.E., Meyers, D.A., Weiss, G.A., Roberts, R.J. and
Murray, K. I977. A specific endonuclease from
Brevibacterium albidum. J. Mol. Biol. ll4:433-440.

 

Gibson, S.M., and Strauss, G. I984. Reaction characteristics
and mechanisms of lipid bilayer vesicle fusion. Biochim.
Biophys. Acta 769:53l-542.

Gotz, F., Ahrne, S. and Lindberg, M. l98l. Plasmid transfer
and genetic recombination by protoplast fusion by
Staphylococci. J. Bacteriol. l45:74-8l.

 

Guerda, N., Ingraham, J.L. and Cenda-Olmeda, E. l97l.
Induction of closely linked multiple mutation by
nitrosoquanidine. Nature New Biol. 230:l22-l25.

Gumpert, J. I980. Electron microscopy analysis of proto-
plast fusion in Streptomyces hydroscgpicus and considera-
tion of structural alteration in fusing membranes. Arch.
Microbiol. I26:263-269.

 

Hales, A. I977. A procedure for the fusion of cells in
suspension by means of polyethylene glycol. Somatic
Cell Genetics 3:227-230.

Honda, K., Maeda, Y., Sasakawa, S., Ohno, H., and Tsuchida,
E. I98la. The components contained in polyethylene
glycol of commercial grade (PEG-6,000) as cell fusogen.
Biochem. Biophys. Research Commun. l0l:l65-I7l.

Honda, K., Maeda, Y., Sasakawa, S., Ohno, H., and Tsuchida,
E. l981b. Activities of cell fusion and lysis of the
hybrid type of chemical fusogens (1) Structure and
function of the promoter of cell fusion. Biochem.
Biophys. Research Commun. l00:442-448.

Hopwood, D.A. l98l. Genetic studies with bacterial proto-
plasts. Ann. Rev. Microbiol. 35:237-272.

'1 I" Io':‘s

 

w:

 

89

Hopwood, D.A. and Wright, H.M. 1978. Bacterial protoplast
fusion: recombination in fused protoplasts of Strepto-
myces caelicolor. Molec. Gen. Genet. l62:307-3l7.

 

Hopwood, D.A. and Wright, H.M. 1981. Protoplast fusion in
Streptomyces: fusions involving ultraviolet irradiated
protoplasts. J. Gen. Microbiol. 126:21-27.

 

Horvath, R.S. l97l. Cometabolism of the herbicide
2,3,6-trichlorobenzoate. J. Agric. Food Chem. l9:29l.

I'keda, 1., Fujita, I. and Yoshinaga, F. 1976. Screening
of L-isoleucine producers among ethionine-resistant
mutants of L-threonine-producing bacteria. Agric.
Biol. Chem. 40:511-516.

Ingolia, T.D. and Koshland Jr., D.E. 1978. The role of
calcium in fusion of artificial vesicles. J. Biol.
Chem. 253:3821-3829.

Ito, T. and Ohnishi, S. 1974. Ca2+-induced lateral phase
separations in phosphatidic acid-phosphatidyl choline
membranes. Biochem. Biophys. Acta. 352:29-37.

Jones, D. 1975. A numerical taxonomic study of Coryneform
and related bacteria. J. Gen. Microbiol. 87:52-96.

Kaneko, H. and Sakaguchi, K. 1979. Fusion of protoplasts
and genetic recombination of Brevibacterium flavum.
Agric. Biol. Chem. 43:1007-1013.

 

Kao, K.N. and Michayluk, M.R. 1969. A method for high-
frequency intergenetic fusion of plant protoplasts.
Planta. 115:355-367.

Kao, K.N., Constabel, F., Michayluk, M.R. and Gamborg, 0.C.
1974. Plant protoplast fusion and growth of inter-
genetic hybrid cells. Planta 120:215-227.

Keddie, R.M. and Cure, G.L. 1979. The cell wall compo-
sition of distribution of free mycolic acid in named
strains of Coryneform bacteria and in isolated from

various natural sources. J. Appl. Bacteriol. 42:229-
252.

Kepes, F. and Kepes, A. 1981. Long-lasting synchrony of
the division of enteric bacteria. Biochem. Biophsy.
Research Rommun. 99:761-767.

 

9O

Knutton, S. and Pasternak, C.A. 1979. The mechanism of
cell-cell fusion. Trends Biochem. Science 220-223.

Komagata, K., Yamada, K., and Ogawa, H. 1969. Taxonomic
studies on Coryneform bacteria l: Division of bacteria
cells. J. Gen. Appl. Microbiol. 15:243-259.

Kondo, J.K., and McKay, L.L. 1982. Transformation of
Streptococcus lactis protoplasts by plasmid DNA. Appl.
Environ. Microbiol. 43:1213-1215.

 

Landman, O.E., and Forman, A. 1969. Gelatin-induced
reversion of protoplasts of Bacillus subtilis to the
bacillary form: Biosynthesis of macromolecules and wall
during successive steps. J. Bacteriol. 99:576-589.

 

m I; 1——‘—‘- C?

W5

Landman, O.E., Ryter, A., and Frehel, C. 1968. Gelatin-
induced reversion of protoplasts of Bacillus subtilis
to the bacillary form: Electron-microscopic and physical
study. J. Bacteriol. 96:2154-2170.

um .14.." - Tali

 

irhaib“

McDonald, K.0., and Burke, W.F. Jr. 1984. Plasmid trans-
formation of Bacillus sphaericus 1593. J. Gen. Microbiol.
130:203-208.

 

Minton, N.P., and Morris, J.G. 1983. Regeneration of
Clostridium pasteurianuin ATCC 6013. J. Bacteriol.
155:432-434.

 

Momose, H. and Takagi, K. 1978. Glutamic acid production
in biotin-rich media by temperature-sensitive mutants
of Brevibacterium lactofermentum, a novel fermentation
process. Agric. Biol. Chem. 42:1911-1917.

 

Nakamori, S. and Isamu, S. 1970. Microbial production of
L-threonine. Part 11. Production by alpha-amino-beta-
hydroxyvaleric acid-resistant mutants of glutamate—
producing bacteria. Agric. Biol. Chem. 34:448-456.

Nara, T., Misawa, M., Komuro, T. and Kinoshita, S. 1969.
Production of nucleic-acid related substances by fermen-
tative processes. Part 21: Biosynthetic mechanisms
involved in 5 purine ribonucleotide production by
Brevibacterium ammoniagenes. Agric. Biol. Chem. 33:
358-369.

 

Nimi, 0., Kubo, M., and Sugiyama, M. 1983. Protoplast
formation and the regeneration of Bacillus brevis ATCC
9999 and its mutants. Biotech. Lett. 5:709-714.

91

Ochi, K., Hitchcock, M.J.M. and Katz, E. 1979. High-
frequency fusion of Streptomyces parvulus or Strepto-
myces antibioticus protoplasts induced by polyethylene
glycol. J. Bacteriol. 139:984-992.

 

 

 

Okamoto, T., Fujita, Y., and Irie, R. 1983. Protoplast
formation and regeneration of Streptococcus lactis
cells. Agric. Biol. Chem. 47:259-263.

 

Papahadjopoulos, D., Poste, G., Schaeffer, B.E. and Vail,
W.J. 1974. Membrane fusion and molecular segregation
in phospholipid vesicles. Biochem. Biophys. Acta 352:
10-28. F:

PapahadjOpoulos, 0., Vail, W.J., Pangborn, W.A. and Poste,
G. 1976. Studies of membrane fusion 11. Induction of
fusion in pure phospholipid membranes by calcium ions
and other divalent metals. Biochem. Biophys. Acta 448:
265-283.

 

F12

Peberdy, J.R. 1980. Protoplast fusion - a tool for genetic
manipulation and breeding in industrial microorganisms.
Enzyme Microbiol. Technol. 2:23-27.

Pirnik, M.P., Atlas, R.M. and Bartha, R. 1979. Hydrocarbon
metabolism by Brevibacterium erythrogenes: normal and
branched alkanes. J. Bacteriol. 119:868-878.

 

Pitcher, 0.6. 1983. Deoxyribonucleic acid base composition
of Corynebacterium dipheriae and other corynebacteria
with cell wall type IV FEMS Microbiol. Lett. 16:291-295.

 

Pontocorvo, G. 1975. Production of mammalian somatic cell
hybrids by means of polyethylene glycol treatment.
Somatic Cell Genetics 1:397-400.

Rose, A.H. 1981. The microbiological production of food
and drink. Scientific American 245(3), pp. 127-138.

Rott, B., Nitz, S. and Freidhelm, K. 1979. Microbial
decomposition of sodium pentachlorophenolate. J. Agric.
Food Chem. 27:306.

Rytir, V., Caslavska, J., Konickova-Radochova, M., and
Konicek, J. 1982. Modification of the cell wall in
Brevibacterium sp. M27. Folia Microbiol. 27:267-268.

 

Ryu, D.D.Y., Kim, K.S., Cho, N.Y. and Pai, H.S. 1983.
Genetic recombination in Micromonospora rosaria by
protoplast fusion. Appl. Environ. Microbiol. 45:
1854-1858.

 

 

92

Sato, A., and Furuya, A. 1977. Accumulation of guanosine-
poly-phosphates by Brevibacterium ammoniagenes; conditions
for the accumulation. Agric. Biol. Chem. 41:641-646.

 

Schaeffer, P., Cami, B. and Hotchkiss, R.D. 1976. Fusion
of bacterial protoplasts. Proc. Nat. Acad. Sci. 73:
2151-2155.

Schaitman, C.H. 1981. Cell fractionation. In Manual of
Methods for General Bacteriology. pp. 52-61, P. Gerhardt,
R.G.E. Murray, R.N. Costilow, E.W. Nester, N.A. Wood,

N.R. Krieg and G.B. Phillips, eds. American Society
for Microb., pub. Washington, 0.C.

 

 

Schmidt, R.H., Morris, H.A., Castberg, H.B. and McKay, L.L.
1976. Hydrolysis of milk proteins by bacteria used in
cheese making. J. Agric. Food Chem. 24:1106-1113.

Seiler, H. 1983. Identification key for coryneform
bacteria derived by numerical taxonomic studies.
J. Gen. Microb. 129:1433-1471.

Shehata, T.E. and Marr, A.G. 1970. Synchronous growth of
enteric bacteria. J. Bacteriol. 103:789-792.

Shepard, J.F., Bidney, D., Barsby, T. and Kemble, R. 1983.
Genetic transfer in plants through interspecific
protoplast fusion. Science 219:683-688.

Shiio, I., Sugimoto, S., and Kawamura, K. 1982. Production
of L-tryptophan by azaserine-resistant mutants of
Brevibacterium flavum. Agric. Biol. Chem. 46:1849-1854.

 

Shimizu, S., Tani, Y., and Ogata, K. 1979. Synthesis of
CoA and its biosynthetic intermediates by microbial
processes. In Methods in Enzymology. McCormic, D.B.,
and Wright, L.D. eds. Academic Press, N.Y. 62:236-245.

 

Shtannikov, A.V., Livshits, V.A. and Zhdanova, N.I. 1981.
Fusion of protoplasts and genetic recombination in
Coryneform bacteria and in Brevibacterium flavum.

Sov. Genet. (Engl. transl. Genetika) 17:924-931.

 

Smith, C.L., Ahkong, Q.F., Fisher, 0. and Lucy, J.A. 1982.
Is purified polyethylene glycol able to induce cell
fusion? Biochim. Biophys. Acta 692:109-114.

Stackebrandt, E. and Fiedler, F. 1975. DNA-DNA homology
studies among strains of Arthrobacterium and Brevi-
bacterium. Arch. Microbiol. 120:289-295.

 

 

93

Stahl, M.L. and Pattee, P.A. 1983a. Computer-assisted
chromosome mapping by protoplast fusion in Staphylo-
coccus aureus. J. Bacteriol. 154:395-405.

 

 

Stahl, M.L. and Pattee, P.A. 1983b. Confirmation of
protoplast fusion - derived linkages in Staphylococcus
aureus by transformation with protoplast DNA. J.
Bacteriol. 154:406-412.

 

Sun, S.T., Hsand, C.C., Day, E.P. and Ho, H.T. 1979.
Fusion of phosphatidylserine and mixed phosphatidyl-
serine-phosphotadylcholine vesicles: Dependence on
calcium concentration and temperature. Biochem.
Biophys. Acta. 557:45-52.

Teshiba, S., and Furuya, A. 1982. Mechanism of 5'-Inosinic
acid accumulation by permeability mutants of Brevibacterium
ammoniagenes. I. Genetical improvement of 5'IMP
productivity of a permeability mutant of B. ammonia-
genes. Agric. Biol. Chem. 46:2257-2263.

 

 

 

Tilcock, C.P.S. and Fisher, D. 1979. Interaction of
phospholipid membranes with poly(ethylene glycol)s.
Biochem. Biophys. Acta. 557:53-61.

Tilcock, C.P.S., and Fisher, 0. 1982. The integration of
phospholipid membranes with poly(ethylene glycol)
vesicle aggregation and lipid exchange. Biochim.
Biophys. Acta 688:645-652.

Tosaka, 0., and Takinami, K. 1978. Pathway and regulation
of lysine biosynthesis in Brevibacterium lactofermentum.
Agric. Biol. Chem. 42:95-100.

 

 

Tsuchida, T., Yoshinaga, F., Kubota, K. and Momose, H.
1975a. Production of L-valine by 2-thiazolea1anine-
resistant mutants derived from glutamic acid-producing
bacteria. Agric. Biol. Chem. 39:1319-1322.

Tsuchida, T., Yoshinaga, F., Kubota, K., Momose, H. and
Okumura, S. 1975b. Cultural conditions for L-leucine
production by strain no. 218, a mutant of Brevibacterium
lactofermentum 2256. Agric. Biol. Chem. 39:1149-1153.

 

 

Udou, T., Ogawa, M. and Mizuguchi, Y. 1982. Spheroplast
formation of Mycobacterium smegmatis and morphological
aspects of their reversion to the bacillary forms. J.
Bacteriol. 151:1035-1039.

 

94

Ward, G.B., Hang, M.W. and Glazer, D.A. 197D. Synchronous
reinitiation of chromosome replication in E. coli B/r

after Nalidixic acid treatment. Proc. Nat. Acad. Sci.
66:365-369.

Weiss, R.L. 1976. Protoplast formation in E. coli. J.
Bacteriol. 128:668-670.

 

Wojcieszyn, J.W., Schlegal, R.A., Lumley-Sapanski, K., and
Jacobson, K.A. 1983. Studies on the mechanism of
polyethylele glycol-mediated cell fusion using fluores-
cent membrane and cytoplasmic probes. J. Cell Biol. 96:
151-159.

Woodley, C.L., Baldwin, J.N. and Greenberg, J. 1981.
Nitrosoguanidine sequential mutagenesis mapping of

Mycobacterium tuberculosis genes. J. Bacteriol. 197:
176-180.

 

Wright, W.E. 1978. The isolation of heterkaryous and
hybrids by a selective system using irreversible
biochemical inhibitors. Exper. Cell Research 112:395-
406.

Wyrick, P.B. and Rogers, H.J. 1973. Isolation and
characterization of cell wall defective variants of
Bacillus subtilus and Bacillus lichenformis. J.
Bacteriol. 116:456-465.

 

 

Yamada, Y., Inouye, G., Tahara, Y. and Kondo, K. 1976.
The menaquinone system in the classification of
Coryneform and Nocardioform bacteria and related
organisms. J. Gen. Appl. Microbiol. 22:203-214.

Yamada, K. and Komagatu, K. 1970a. Taxonomic studies
on Coryneform bacteria, II. Principal amino acids in
the cell wall and their taxonic significance. J. Gen.
Appl. Microbiol. 16:103-113.

Yamada, K. and Komagatu, K. 1970b. Taxonomic studies
on Coryneform bacteria, III. DNA base composition
of Coryneform bacteria. J. Gen. Appl. Microbiol. 16:
215-224.

Yoshinaga, F. 1969. Studies on the fermentation production
of L-proline. IV. Mechanism of L-proline production by
Brevibacterium flavum. 2247 no. 14-5. J. Gen. Appl.
Microbiol. 15:387-398.

 

M 11111111111111 “1111111111111 1111111111 E
3 1293 03083 2699