END PRODUCT REPRESSION OF COBALAMIN SYNTHESIS LESLIE …

END PRODUCT REPRESSION OF COBALAMIN SYNTHESIS

IN SALMONELLA TYPHIMURIUM

by

LESLIE B. GARRISON, B.S.

A THESIS

IN

MICROBIOLOGY

Submitted to the Graduate Faculty of Texas Tech University in

Partial Fulfillment of the Requirements for

the Degree of

MASTER OF SCIENCE

Approved

Accepted

December, 1989

73

/3^ IS3 ACKNOWLEDGMENTS

I am deeply indebted to Dr. Randall M. Jeter for his direction of the work for this

thesis and my graduate work, and to the other members of my committee, Drs. IDoris

Lefkowitz and Llewellyn Densmore, for their assistance in the final preparation of this

thesis. I would also like to express my appreciation to my parents, grandparents and wife

for their encouragement and support during my graduate wOTk.

u

CONTENTS

ACKNOWLEDGMENTS ii

ABSTRACT iv

LIST OF TABLES v

I. INTRODUCTION 1

n. MATERL\L AND METHODS 9

in. RESULTS 22

IV. DISCUSSION 34

REFERENCES 38

m

ABSTRACT

The purpose of this research is to elucidate the mechanism that regulates the end-

product repression of the genes for cobalamin biosynthesis (cob) in Salmonella

typhimurium. Twenty-nine insertion mutants which exhibited altered repression

phenotypes were isolated. The effect of these insertion mutations on the expression of the

cob genes was quantified by using P-galactosidase assays to monitor the transcription of a

coblr.lac operon fusion. The mutants were divided into two phenotypically distinct groups

which were designated positive and negative. In the 11 negative mutants, the cobl operon

was not expressed under any conditions. All of the insertion mutations in this group were

mapped in or near the cobl operon itself. In the 18 positive mutants, the cobl operon was

less sensitive to repression by cobalamin than in the parent strain. The positive mutants

were subdivided into four groups by cotransduction and conjugational mapping. One

group of mutations is located in the cobalamin transport genes (btuB and btuCED). A

second group lies near the distal end of the ethanolamine utilization (eut) operon at 50 map

units and prevents the mutants from utilizing ethanolamine as a sole carbon source. A third

group lies between 47 and 62 map units on the S. typhimurium chromosome. The fourth

group of positive mutations remains unmapped.

The insertion mutations that are located in the eut region appear to lie downstream

from the genes that encode the enzymes and a known regulatory protein for ethanolamine

utilization. We propose that these mutations affect the synthesis of a second regulatory

protein that binds cobalamin. This protein plays a dual regulatory role by repressing

transcription of the cobl operon and activating transcription of the eut operon.

IV

UST OF TABLES

TABLE 1 Salmonella typhimurium stTddns used mitds study 10

TABLE 2 Mini-tet insertion mutants generated in this study 24

TABLE 3 Results ofcotransductionalcrossses between the positive insertion mutations and markers in or near known cobalamin transport genes 26

TABLE 4 Results ofcotransductions between ewr mini-tet insertion mutants and maikers in and near eut operon 28

TABLE 5 Cojugational mapping of mini-tet insertion mutations in strain RT887 31

TABLE 6 p-Galactosidase enzyme activity in permeabilized cells 32

CHAPTER I

INTRODUCTION

Cobalamin (B12) is the largest and most complex of the vitamins (13). It is not

synthesized de novo by plants or animals; it is produced only by certain microorganisms

(18). Therefore, animals must depend on these microorganisms for their supply of

cobalamin.

Ruminant animals obtain the cobalamin they require from the microorganisms

which grow in the rumen. Other animals, including humans, must glean their supply of

cobalamin entirely fix)m dietary sources. In humans, prolonged dietary deficiency of

cobalamin, or a defect in cobalamin absorption, may result in pernicious anemia. This is an

often fatal disease which has no animal counterpart It can be treated by intravenous

injections of the vitamin form of cobalamin (14).

The Structure of Cobalamin

The cobalamin molecule is made up of two major components, the corrin ring and

a ribonucleotide that contains the heterocyclic base 5,6-dimethylbenzimidazole (DMBI) see

Fig. 1. The corrin ring is similar to the tetrapyrrole rings found in heme and chlorophyll

(7). However, the linkages between the four pyrroles in the porphyrin ring are all methene

bridges, while the A and D pyrroles of the corrin ring are joined by a simple covalent bond

between the a carbons. An atom of cobalt is held in the center of the corrin ring by a

covalent bond to the nitrogen of the A pyrrole, and by coordinate bonds to the other three

pyrrole nitrogens (20).

The nucleotide portion of the cobalamin molecule is joined to the corrin ring by a

D-l-amino-2-propanol side chain on the D pyrrole, and by a coordinate bond between

1

NH'

NH2OCCH2CH2

hfH^CCH2

" 3 ^ H3C NH2OCCH

.CH2CONH2

CH2CH2CONH2

CH2CH2CONH2

CH2OH

Figure 1: Adenosylcobalamin.

DMBI and the cobalt atom. The nucleotide lies below the plane of the ring (13,15). This

nucleotide is unusual for two reasons: the base is linked to the C-1 of the ribose by a 7-a-

glycoside bond (in lieu of the 9-p-glycoside bond found in other nucleotides), and DMBI

itself is found only in cobalamin (7,20).

The presence of the unique base is apparently necessary for the function of

cobalamin as a coenzyme. 0)balamin analogues, in which DMBI has been replaced by

other purine bases, are less effective in vertebrates, and analogues that have pyrimidine

bases are inert as coenzymes in humans (17,20).

The sixth substituent position of the cobalt atom, which lies above the plane of the

ring, may be occupied by a number of different groups, including NO2, OH, SO3H, CNS,

or CN (20). In the commercially-produced form of cobalamin, the sixth position is most

commonly occupied by a cyano (CN) group; thus, cyanocobalamin has become

synonymous with the term vitamin B12 (10). However, cyanocobalamin is formed only

during the industrial production of cobalamin and apparendy is not produced in vivo

(10,20). In the cofactor forms, which do occur in the cell, the sixth position is filled by

either a methyl group (methylcobalamin) or by 5'-deoxyadenosine (adenosylcobalamin)

(18). Cobalamin-dependent enzymes and their cofactors are discussed in greater detail

below.

Industrial Production

Cobalamin is one of only two vitamins that are produced industrially entirely by

microbial fermentations, riboflavin being the other (7). The majority of the cobalamin is

produced in the form of cyanocobalamin, with smaller amounts of hydroxocobalamin and

adenosylcobalamin being produced as well. The bacterial species most often used to

produce cobalamin industrially is Pseudomonas denitificans, which requires the addition of

DMBI to the culture medium (7).

The originally-isolated Pseudomonas strain was able to produce 0.6 mg of

cobalamin per liter. After 12 years of strain development, the yield was increased to 60 mg

per liter (7). Strain development usually involves generating mutants with ultraviolet

radiation or chemical mutagens, indirect selection, and the screening of mutants for

increased production. Yields may also be increased by varying the culture conditions (7).

Neither of these two methods for enhancing product yield requires a detailed knowledge of

the genes involved or their regulation.

The discovery that Salmonella typhimurium synthesizes cobalamin under

anaerobic conditions (16) allows established genetic and enzymatic techniques to be applied

in an organism that has a well-understood genetic make-up. The analysis of cobalamin

biosynthesis can therefore be facilitated by studying the process in this organism.

Cobalamin-Dependent Enzvmes in Salmonella

It has been estimated that the production of cobalamin may require as many as 30

unique enzymes (11); this is a considerable investment of the cell's resources. Yet none of

the cobalamin-dependent enzymes known to exist in Salmonella seem to be vital to the ceU

under most culture conditions. Four cobalamin-dependent enzymes have thus far been

described in Salmonella. They are N^-methyltetrahydrofolate-homocysteine

methyltransferase (EC 2.1.1.13), ethanolamine ammonia-lyase (EC 4.3.1.7), propanediol

dehydratase (EC 4.2.1.28), and an unnamed enzyme that reduces epoxyqueuosine to

queuosine (12).

The homocysteine methyltransf erase, which uses methylcobalamin, (11) transfers

a methyl group from N^-methyltetrahydrofolate to homocysteine to form methionine. This

enzyme is the product of the metH gene (29). The final step of methionine synthesis may

also be carried out by a cobalamin-independent methyltransferase, which is the product of

the metE gene (16). However, the metH enzyme is much more efficient than the metE

enzyme, which may constitute as much as 5 % of the total soluble protein in cells which

are using it to supply their need for methionine (32). The production rates of these two

enzymes are regulated by the availability of cobalamin. In the presence of cobalamin, the

expression of metH is increased while the activity of the metE gene is suppressed (25).

The suppression of metE by cobalamin requires the products of the metH and metF (5-

methylenetetrahydrofolate reductase) genes, which suggests that 5-methyltetrahydrofolate

must bind to the metH holoenzyme in order for metE to be repressed (22).

Ethanolamine ammonia-lyase is encoded within the eut operon, which lies at 50

map units on the Salmonella chromosome between the cysA and purC genes (24). The

coenzyme for ethanolamine ammonia-lyase is adenosylcobalamin (1,4). This enzyme

cleaves ethanolamine to yield acetaldehyde and ammonia (2,27). The acetaldehyde is then

converted into acetyl coenzyme A by acetaldehyde dehydrogenase, which is also encoded

within the eut operon (24).

In addition to the two enzymes, the eut operon also encodes a regulatory protein

that enhances the transcription of its own operon. This positive regulatory protein appears

to be transcribed finom the main promoter for the operon and also fh>m a weak secondary

promoter just in front of the regulatory gene (24). The main promoter of the eut operon is

induced by the simultaneous presence of cobalamin and ethanolamine in the culture medium

(3,24).

Propanediol dehydratase is another enzyme which uses adenosylcobalamin as a

coenzyme (1). This enzyme transforms 1,2-propanediol into propionaldehyde (19). This

is the first step in the utilization of propanediol as a carbon and energy source by

Salmonella (24).

The last of the four cobalamin-dependent enzymes reduces the 2,3-epoxy-4,5-

dihydroxycyclopentane ring of epoxyqueuosine to the 4,5-dihydroxycyclo-2-pentene ring

found in queuosine. C^euosine is a nucleoside which replaces guanosine in position 34 of

the anticodon of certain tRNA molecules in eubacteria and many eucaryotes. When

cobalamin is not available, all of the tRNAs will contain guanosine. If cobalamin is

available, then some of these tRNAs will contain queuosine at position 34 (12).

Transport of Cobalamin

Cells may obtain cobalamin for use as a cofactor either by transporting external

cobalamin into the cell or by internal synthesis. In Escherichia coli, the transport of

cobalamin into the cell requires the products of genes btuB^ btuC^ btuD, btuE and tonB

(5). The btuB gene product is an outer-membrane protein which binds cobalamin. The

tonB gene product is thought to be an energy-transducing protein that is responsible for the

active transport of cobalamin across the cell wall (23). The transport of cobalamin fix)m the

periplasmic space across the cytoplasmic membrane and into the cell is perfonned by the

btuC, btuD, and btuE proteins. Mutants that have lost one or more of these proteins are

only slightiy impaired in their ability to transport cobalamin as long as they have an intact

btuB-tonB system (5).

The Genes Involved in Cobalamin Synthesis

The biosynthesis of cobalamin begins in the porphyrin pathway, which also

produces heme and chlorophyll. The last intemoediate that the two pathways have in

common is uroporphyrinogen HI (7,33). The specifics are not fiilly understood, but the

general outline of the cobalamin pathway has been elucidated in nonenteric bacteria and

several of the intermediates are known (7,33).

While litUe is known about the enzymatic pathway that produces cobalamin, the

biosynthetic genes and their regulation in Salmonella have been described. The genes

involved in the production of cobalamin (cob) are located at 42 map units (between his and

supD). The cob genes appear to be organized into three discrete operons that arc

designated cob I, cob n, and cob HI. The order of the operons (in the counterclockwise

direction along the S. typhimurium genetic map) is cobl-cobUl-cobll. The contribution of

each of these operons to the production of cobalamin has been determined by nutritional

studies, cobl mutants cannot synthesize the corrin ring, and therefore require an

exogenous supply of the biosynthetic interatiediate cobinamide in order to produce

cobalamin. cobU mutants are unable to produce DMBI, and so an exogenous supply of

DMBI is required in order to produce cobalamin. cobUl mutants cannot join the corrin ring

and the DMBI portions of the molecule together, so that these mutants cannot produce

cobalamin even if cobinamide and DMBI are both provided (16).

Regulation of the cofe Genes

The expression of the cob genes is affected by a number of factors. These genes

are maximally expressed under conditions of anaerobic respiration with fumarate as the

electron acceptor. The presence of oxygen represses the transcription of the cob operons in

general and cobl in particular. The expression of cobl may be reduced 40-fold from its

maximal level in the absence of oxygen. The cob genes are also regulated by adenosine

3',5'-cyclic monophosphate (cAMP) (11). An increase in cAMP levels within the cell

causes the cAMP-catabolite regulatory protein complex to bind to and enhance the

expression of the cob genes. Conversely, a drop in the cAMP level allows the complex to

be released from the cob genes, which reduces their rate of transcription (28). This type of

regulation by cAMP is typical of genes involved in the utilization of carbon sources. The

cob genes are also subject to end-product repression by cobinamide (the ring portion of the

cobalamin molecule), DMBI, and cobalamin itself. Individually, DMBI and cobinamide

have only a slight effect on cobl, but together they decrease the operon's level of

transcription by half Cobalamin has the strongest repressive effea of all, decreasing the

expression of the cob I genes by up to 80 % (11).

8

I propose that the end-product lepression of the cob operon is effected by a

legulatofy protein ^ch binds cobalamin, changes conformation, then blocks transription

by binding to an operator sit on the operon. In this study we will use genetic and

enzymatic techniques to map the location of this regulatory gene and to quantify its effect

on thecob genes in S. typhimurium.

CHAPTER n

MATERIAL AND METHODS

Bacteria. Culture Media, and Growth Conditions

The bacterial strains used in this study were derivatives of Salmonella

typhimurium LT2; the genotypes are listed in Table 1. The bacteriophage used to perform

transductions was P22 HT 105/1 mr-201(10). The complex media used routinely in this

study were nutrient agar and nutrient broth (Difco Laboratories), and the minimal medium

was the E medium of Vogel and Bonner (31). NCE (No carbon E) minimal medium (8)

was used in experiments involving the utilization of specific carbon sources. The final

concentrations of various carbon sources added to NCE medium were ethanolamine,

14mM; fumarate, 40mM; glucose, llmM; glycerol, 22mM; lactose, 7mM; and 1,2-

propanediol, lOmM. Other supplements were spread onto the minimal medium plates as

required to support the growth of auxotrophic mutants (0.1 ml of these supplements were

added to each plate). The concentrations of the stock solutions were as follows:

cyanocobalamin, 4mg/l; cobinamide, 4mg/l; 5,6-dimethylbenzimidazole (DMBI), 50 mg/1;

and L-methionine, 9g/l. The concentrations of the antibiotics used in the complex media

were 30 mg/1 of ampicillin, 50mg/l of kanamycin, and 30 mg/1 of tetracycline. The

antibiotic concentrations used in the minimal media were 125 mg/1 of kanamycin, 1(X)0

mg/1 of streptomycin, and lOmg/1 of tetracycline.

Unless otherwise stated, all bacterial cultures were incubated at 37°C. Broth

cultures were grown aerobically at 200 rpm in a rotary incubator shaker (Series 25, New

Brunswick Scientific C!o.). Plate cultures were incubated in standard incubators (Model

330, National Appliance Co.). Anaerobic incubations were performed in an anaerobic

chamber (Model 1025, Forma Scientific Co.) under an atmosphere of 93% N2, 5% CO2,

9

10

TABLE I, Salmonella typhimurium stndns used in this study.

Strains

RT787 RT789 RT801 RT869 RT870

RT933 RT934 RT935 TR5654 TR5655

TR5656 TR5657 TR5658 TR5660 TR5662

TR5663 TR5666 TR5667 TR5668 TR5669

TR5671 TR5686 TR5688 TR6583 TT627

Tr628 TT10271 TT10423 TT10426 TT10427

TT10723 TT10852 Tni580 TT13775 TT13893

Genotype

metE205 ara-9 cob-21:: Mu dA aroDSSrpsU argfl87 purC7 cysA533

leu^7 cob-24::MudJ Aeut-237 cob-24::Mu dJ thrA9rpsLl leu^SSrpsLl

proA36 rpsLl purES rpsLl pyrCJrpsLl pyrF146 rpsLl his01242 hiS'2236 rpsLl

purFMSrpsLl serA13 rpsLl cysG439rpsLl cysE396 rpsLl ilV'508rpsLl

pyrB64 rpsLl aroDMOrpsU purAlSS rpsLl metE205 ara-9 purC7 rpsLI (Ftsl 14 lac^zrf'20::TnlO)

pyrC7 rpsLl (Ftsl 14 lac+zzf-21::TnlO ) eut-I8::Mu dJ proAB -^7(F128pro+ toc+ zzf'1831::TnlO Ll6M7TtX^ proAB47iFl2S pro+ lac+ zzf'1834::TnlO Ai6 AiTKan pNK972

metE205 ara-9 metH2355:'Mu dA metE205 ara-9 cdb-24:Ma dJ t^m-237 btu-2vM\x dJ em-20S:\lTdO

11

and 2% H2. Centrifugations were performed in a table-top clinical centrifuge (International

Equipment Co.). The 30°C incubations for the p-galactosidase assay were performed

using a Dri-Bath (Type 16500, Thermolyne Corp.), and the absorbance readings were

made with a Spectronic 20 spectrophotometer (Milton-Roy Co.).

The organic chemicals and antibiotics used in this study were obtained from Sigma

Chemical Co., with the exception of the ethanolamine and 1,2-propanediol, which were

purchased fiom Aldrich Chemical O).

Lysate Production and Transformations

Lysates were produced by adding a 0.3 ml sample of an overnight culture of S.

typhimurium (various mutant strains) to a tube containing 5 ml of nutrient broth and 0.03

ml of a stock P22 lysate (1 x 10 -1 x 10 ^ plaque-forming units/ml) which had been

grown on LT2. The cells were grown overnight, and then the culture was centrifuged at

2,500 ipm for 20 min. The supernatant containing the phage was poured into a sterile

tube, and 0.5 ml of chloroform was added. The supernatant and chloroform were mixed

using a vortex mixer, and the chloroform was allowed to settie out of the aqueous layer

ovemight at room temperature.

Transductions were perforated by adding 0.03 ml of a lysate (of phage grown on

the desired donor bacterial strain) to an ovemight culture of recipient cells. The culture was

incubated for 30 min to allow for phenotypic expression of the transduced marker.

Samples (0.1 ml) of the culture were then spread onto plates which would only allow the

cells that had inherited the desired trait (an antibiotic maricer or the repair of some

auxotropic mutation) to grow.

12

Gene Pool Construction

A TnlO Al6 A17Tet^ (mini-tet) gene pool was produced by using Salmonella

strains TT10423 and TT10427. The strain TT10423 contains an F plasmid which has a

mini-tet element (a TnlO with the gene for transposase removed) inserted into it (8,10).

TT10427 contains the plasmid pNK972 which carries only the transposase gene.

A P22 lysate grown on TT10423 was used to transduce an ovemight culture of

TT10427, and 0.1 ml aliquots of the cells were spread onto nutrient agar + tetracycline

plates. The plates were incubated overnight, and the antibiotic-resistant colonies (-100

colonies^late) were washed from the plates with 0.1 M phosphate buffer with 10 mM

ethyleneglycol-bis-(p-aminoethyl ether) N,N,N',N'-tetraacetic acid (EGTA) to bind free

calcium ions in the solution and thereby prevent P22 phage fix)m attaching to the uninfected

cells (8). The cells were collected in a 15 ml tube, centrifiiged at 2,500 rpm for 20 min,

and washed twice with phosphate buffer with EGTA. The cells were resuspended in

phosphate buffer without EGTA, and a lysate was grown on the pooled cells. This lysate

was the mini-tet gene pool which was used in later experiments. A similar procedure was

also used to prepare a TnlO A16A17 Kan^ (mini-kan) gene pool with TT10426 as the

donor strain.

Isolation of Mutants Using Mini-Tet Insertions

S. typhimurium strain TT10852 has a Mu dJ element inserted into the cobl

operon. Mu dJ contains a partial lac operon which has no promoter of its own, so that it

must depend on the promoter of the operon into which it is inserted for its transcription.

LT2, which is the parent strain of all the S. typhimurium strains that were used or

generated in this study, does not produce P-galactosidase. The only lac operon that is

contained by TT10852 is the one within the Mu dJ element. Thus, the expjression of the

cobl operon can be monitored by assaying for P-galactosidase enzyme activity.

13

Strain Tri0852 was grown ovemight in nutrient broth. The broth culture was

transduced with a mini-tet gene pool, and 0.1 ml aliquots were spread onto nutrient agar

plates with tetracycline to select for cells which had inherited the mini-tet elements by

homologous recombination into the chromosome. The plates were incubated overnight,

then replica-plated onto lactose MacConkey agar plates with tetracycline and with or

without a supplement of cyanocobalamin. If the colonies produced P-galactosidase, they

appeared red on the MacConkey agar plates; if they did not produce p-galactosidase, they

appeared white.

Two classes of regulatory mutants were detected by this selection. In the first

class, termed positive mutants, the cob genes were rendered insensitive to end-product

repression, and the cells were Lac"*" in both the presence and absence of cobalamin. In the

second class, termed negative mutants, the cob genes were not transcribed whether

cobalamin was present or absent, and these cells were Lac" under both sets of conditions.

All mutants of both classes were isolated, and their phenotypes were confirmed by

restreaking them onto the same type of indicator medium that had been used in the original

screening. The mutants were also streaked onto E glucose minimal medium plates and

incubated ovemight to determine if the production of cobalamin in these cells was still

repressed by oxygen. This test was performed in a metE (TR6583) background.

Determining the Proximity of the Putative Regulatory Mutations to the cobl Operon

The ability of the mini-tet insertions in each mutant to cotransduce with the

cobli'Mu dJ of TT10852 was determined. P22 lysates were grown on each the putative

regulatory mutants. Ovemight cultures of TT10852 were transduced with these lysates,

and 0.1 ml samples were spread onto nutrient agar plates with tetracycline. The plates were

incubated ovemight and the colonies were scored for kanamycin sensitivity, which

14

indicates that inheritance of the mini-tet results in simultaneous loss of the Mu dJ, and that

the two elements are cotransducible.

Mutants that carried cotransducible mini-tet insertions were tested to determine if

the insertions were located in the cob genes. All of the mutants, along with LT2 and

TT10852 controls, were streaked from nutrient agar stock plates onto four different sets of

E glucose minimal medium plates. The control set of plates was spread with cobalamin,

while the three test sets contained either no supplanent, cobinamide, or DMBI. All four

sets of plates were incubated anaerobically for 48 hr. Mutants in the cobl operon are

unable to produce cobinamide, and they grow only on the plates where cobinamide has

been supplied. Mutants in cobU are unable to produce the DMBI portion of the molecule,

and they require a supplement of DMBI in the medium for growth. Mutants which do not

grow on all three types of test medium are considered to have cobUl insertions (16).

Assay for Transport Mutant Phenotypes

The ability of the mutants to transport cobalamin across the cell envelope was

tested by the mutants' ability to grow with limited cobalamin (5). The positive mini-tet

insertion mutations were tested in a metE205 (TR6583) genetic background. Strain LT2

(mef^ btu^) and TT13775 (mef^ btu-2::Mu dJ) controls were streaked onto E glucose and

NCE ethanolamine minimal medium plates which had been spread with 0.1 ml of a 0.04

mg/1 solution of cyanocobalamin. This is 1/lOOth of the standard amount of cobalamin per

plate. These plates were incubated aerobically for 5 days.

Test for Cotransduction of the Positive Mutations with Markers In or Near Known Transport Genes

Three genetic loci have been identified that are involved in the uptake of cobalamin

in Salmonella: btuB, btuC, and tonB. S. typhimurium strain TT13775 has a kanamycin-

15

resistance marker at the btuC locus (5). This strain was transduced with positive mutant

lysates and inoculated onto nutrient agar plates with tetracycline. After ovemight

incubation, 100 colonies were patched to the same medium. The patch plates were

incubated ovemight and then replica-plated onto nutrient agar plates with tetracycline and

with or without kanamycin. These plates were incubated ovemight. The patches were then

scored for the loss of kanamycin resistance.

The ability of the remaining unmapped positive mutations to cotransduce with the

other transport genes was determined by using S. typhimurium strains RT801 and RT797,

which have auxotrophic markers near btuB and tonB, respectively. These strains were

transduced with the positive mutant lysates, inoculated onto nutrient agar plates with

tetracycline, and incubated ovemight Patch plates were prepared and replica-plated onto E

glucose minimal medium with tetracycline. The replica plates were incubated ovemight and

the patches were scored for repair of the auxotrophic mutations.

Test for Cotransduction of the Positive Mutations with a Marker in metH

S. typhimurium strain TT10723 contains an ampicillin-resistance marker in the metH

gene, which is responsible for the production of the methylcobalamin-dependent

homocysteine methyltransferase. Ovemight broth cultures of TT10723 were transduced

with the lysates of the remaining unmapped positive mutants. A 0.1 ml sample of each

transduced culture was spread onto nutrient agar plates with tetracycline, and these plates

were incubated ovemight. Patch plates were prepared and then replica-plated onto nutrient

agar plates with tetracycline, with or without ampicillin. These plates were incubated

ovemight and the patches were scored for loss of ampicillin resistance.

16

Assay for Conversion of the Vitamin Form of Cobalamin into the Cofactor Forms

Cyanocobalamin, the vitamin form of cobalamin, is modified to make the different

cofactors that are used by the cobalamin-dependent enzymes. Some of the mutations which

conferred a transport-deficient phenotype did not cotransduce with maikers near the three

loci known to be involved in cobalamin transport. These mutations were placed into a

TR6583 (metE) background by transduction. The mutants were then incubated in liquid E

glucose minimal medium with methionine to minimize the carryover of nutrients. The cells

were streaked onto three different sets of minimal medium plates with either glucose,

propanediol, or ethanolamine as a sole carbon source. All three types of media had been

spread with 0.1 ml of a 400 mg/1 solution of cyanocobalamin (1(X)X the usual

concentration) to allow transport mutants to grow on the plates. The plates were incubated

for 5 days and screened for the presence or absence of growth.

Test for Proximitv of the Positive Mutations to the Ethanolamine Genes

While performing the above experiment, two mutants were characterized which

grew on both the glucose and propanediol plates, but did not grow on ethanolamine as a

sole carbon source. Transductions were performed to determine if these mutations were in

the region of the ethanolamine utilization (eut) operon. S. typhimurium strain TT10271,

which has a Mu dJ inserted in the ethanolamine genes, was transduced with P22 lysates

grown on the two ethanolamine-negative mutants. The transductants were plated on

nutrient agar + tetracycline and these plates were incubated ovemight. Patch plates were

prepared and replica-plated onto nutrient agar plates with tetracycline with or without

kanamycin. These plates were incubated ovemight and the patches were then scored for

loss of kanamycin resistance.

17

Two other transductions were performed with the mutant strains RT869 and

RT870, which have auxotrophic mutations in the pwrC and cysA genes, respectively (Table

1). These two genes lie on either side of the ethanolamine genes (24). The procedure for

both cotransduction experiments was the same. Ovemight broth cultures of recipient cells

were transduced with the mini-tet insertion mutant lysates and then inoculated onto nutrient

agar plates with tetracycline. These plates were incubated ovemight Patch plates were

prepared and replica-plated onto E glucose minimal medium plates with tetracycline.

Repair of the auxotrophic mutation indicated that the mini-tet insertion and the auxotrophic

mutation are cotransducible.

To determine if the mini-tet insertions were in a control region that is known to

exist at the distal end of the ethanolamine operon (24), the phenotype of a previously-

isolated mutant in the control region was compared to the phenotype of the positive mutants

with insertions that were found to map near the ethanolamine genes. S. typhimurium strain

TT13893, with a TnlO insertion in this control region, was transduced to kanamycin

resistance with a P22 lysate grown on TT 10852 (coW::Mu dJ). The double insertion

mutant was streaked onto MacConkey plates with and without cyanocobalamin, and was

then checked for the positive phenotype (Lac"*" on both plates).

The ability of the two eut insertion mutants to repair a deletion of the eut operon in

strain TTl 1580 was also used as a method of determining if the mutants were inserted in

the eut control region. Two sets of ethanolamine minimal medium plates were prepared.

One set was spread with an ovemight culture of TTl 1580, while the control set was spead

with an ovemight culture of RT933 (leu-447). One drop of each P22 lysate grown on cells

containing the two eut insertion mutations was placed onto each plate. One drop of a lysate

grown on LT2 was also placed on each plate as a control. The plates were incubated for 2

days and then scored for transductants that inherited the ability to use ethanolamine as a

sole carbon source.

18

Arranging the Unmapped Mutations into Groups by Cotransduction

All of the mutants which were not successfully mapped by cotransduction were

arranged into groups by replacing the mini-tet insertions in a few of the mutants with a

linked mini-kan marker, and then transducing the cells containing the new mini-kan marker

with each of the remaining mini-tet insertions. If the mini-kan marker is at the same

chromosomal location as the mini-tet insertion with which it is being transduced,

inheritance of the mini-tet will cause the mini-kan to be lost from the chromosome at a

detectable frequency. If all of the colonies resulting fix)m a transduction are both

kanamycin- and tetracycline-resistant, then the two mutations are not P22-cotransducible

and they must be inserted into different parts of the chromosome. If, however, some of the

transductants are kanamycin-sensitive, then the mutations are P22-cotransducible and they

are located near to each other on the chromosome.

Certain mutants, chosen at random, were transduced with a mini-kan gene pool

and inoculated onto nutrient agar plates with kanamycin. These plates were incubated

ovemight Patch plates were prepared and replica-plated onto nutrient agar plates with

kanamycin and with or without tetracycline. Patches that were kanamycin-resistant but

tetracycline-sensitive were purified, grown ovemight in broth culture, and used as

recipients for transductions in which the donor lysates were grown on the other mini-tet

insertion mutants. If the inheritance of mini-tet caused the simultaneous loss of the mini-

kan marker, then that mini-tet mutant was placed in the same group as the mini-tet mutant

from which the min-kan had been derived originally. In this way, unmapped mutations

were grouped at separate chromosomal locations.

19

Mapping the Location of the Mutations bv Conjugation

Hfr donor strains were constructed using representative mutants taken from each

of the genetic groups defined in the above experiments. The Hfr strains were constmcted

using F' plasmids which had Tn70s inserted into them. The TnlOs in the plasmid created

an area of homology between the plasmid and the mini-tet insertions so that the plasmid

integrated into the chromosome at the point of the mini-tet insertion (6). Two plasmid

donor strains (TT627 and TT628) were used for the constructions. These two strains

contain Flac plasmids with Tn70 inserted in opposite orientations (6). Thus, two Hfr

donors could be constmaed for each mutant, one with a clockwise and the other with a

counterclockwise gradient of chromosomal transfer. The two Flac plasmid donors were

grown ovemight in NCE glucose minimal broth at 3(y*C. The donor and recipient cells

were then cross-streaked onto NCE lactose minimal plates with tetracycline. The colonies

that grew on the plates were composed of tetracycline-resistant recipient cells that had

inherited the Flac plasmid. A healthy colony was taken from each plate and grown

ovemight at 30°C in NCE lactose minimal broth. A 0.1 ml sample of this culture was

added to 2.9 ml of the same medium and incubated ovemight at 42°C. Because the

plasmid's replication is temperature-sensitive, only cells in which the plasmid has

integrated into the host chromosome will grow at 42°C (6). These cells were then streaked

onto NCE lactose minimal plates with tetracycline. The plates were incubated for 2 days at

42°C. Healthy colonies were taken from the plates, inoculated into NCE lactose minimal

broth, and were incubated ovemight at 42°C. These cells were the Hfr donors. Seventeen

different streptomycin-resistant auxotrophic mutants were used as recipients in the

conjugational matings (6). Aliquots (0.1 ml) of each Hfr stain were mixed with 0.1 ml of

each auxotropic recipient in 1 ml of nutrient broth. The matings were allowed to proceed

for 3 hrs. Samples (0.1 ml) were then taken from each tube and spread onto NCE glucose

20

minimal medium plates with streptomycin. The plates were incubated for 2 days, and the

numbers of recombinants on each plate were counted.

Determining the Effect of the Mutations on the Transcription of the cob Genes

p-galactosidase assays were performed on all the mutants that were not located in

the cobl stmctural genes. Four replicates of the assay were done for each mutant The

mini-tet insertions were placed into an LT2 background by transduction. The cells carrying

the mini-tets were transduced to kanamycin-resistance with a P22 lysate that had been

grown on strain TT10852 (coblvMn dJ) lysate. In these transductants, the Mu dJ element

was inserted into the cobl operon and the expression of the lac genes was direcdy related to

the transcription of the cobl operon. Strains LT2 (wild-type) and TTl 1580 (Aeut-237)

were also transduced to kanamycin resistance and used as controls. A wild-type

Escherichia coli K-12 strain (RT504) that was grown with and without an inducer for the

lac genes (isopropyl-P-D-thiogalactopyranoside; IPTG) was also used as a positive control

for p-galactosidase activity. For these assays, cells were grown in NCE glycerol-fumarate

minimal medium under anaerobic conditions so that the cob genes would be maximally

expressed (11).

One ml volumes of glycerol-fumarate medium were placed into test tubes and

incubated ovemight under anaerobic conditions to remove dissolved oxygen. The tubes

were inoculated with 0.1 ml of ovemight cultures of the 18 mini-tet strains, which had been

pre-grown in nutrient broth under aerobic conditions. The tubes were incubated

anaerobically for 6 hrs., then centrifuged at 2,500 rpm for 20 min. The cells were

resuspended and diluted 1:10 in sterile physiological saline (0.85%), and placed on ice for

20 min to prevent further growth. They were then diluted 1:10 with complete Z buffer

(21). Chloroform and 0.1% sodium dodecyl sulfate (SDS) were added to the tubes, and

21

the tubes were mixed with a vortex mixer for 10 sec to penneabilize the cells. The tubes

were equilibrated at 30°C for 2 min, and then o-nitrophenyl-P-D-galaaopyranoside

(ONPG) was added to each tube. The tubes were mixed again and the reaction was

allowed to proceed at 30°C. When a faint yellow color had developed, the reaction was

stopped by adding 1 M Na2C03 to the tubes. At this point, the tubes were placed in a

30°C shaking incubator for 5 min. Absorbance readings of the reaction mixture at 420 nm

and 550 nm were taken with a spectrophotometer. A reading of the diluted cells was also

taken at 650 nm. The reaction time and the three readings were used to calculate the

enzyme activity in nanomoles of ONPG produced/min/units of absorbance at 650 nm (11).

CHAPTER m

RESULTS

Major Classes of Mutants

Twenty-nine insertion mutants with altered phenotypes for the end-product

repression of cobalamin biosynthesis were isolated in S. typhimurium during this study.

All of the mutants were generated using the mini-tet insertion element This insertion

element was used for two reasons: the antibiotic resistance marker (Tet^ provides a means

for isolating the mutants by direct selection, and insertion mutations produce dismptions in

the coding sequence of a target gene, which usually results in complete loss of the gene's

activity.

There are no direct assays presendy available for any of the enzymes involved in

the biosynthesis of cobalamin. Therefore, the transcription of the cob genes in the mutant

cells was monitored by assaying for the activity of p-galactosidase in a cobl::lac gene

fusion strain. On MacConkey indicator medium, lactose-fermenting (Lac"*") colonies

appear deep red, and lactose-nonfermenting (Lac") colonies appear white. The fusion

strain is Lac" in the presence of cyanocobalamin and Lac"*" in its absence. Thus, the

production of p-galactosidase by the fusion is normally repressed when cyanocobalamin is

added to the culture medium. Following mutagenesis, the mini-tet insertion strains were

screened for changes in the mutant cells' ability to ferment lactose on MacConkey plates.

Based on the color of the colonies, it was possible to arrange the mini-tet insertion mutants

into two different classes. The first group of 18 mutants was Lac"*" on MacConkey agar

regardless of whether cyanocobalamin was present or absent in the medium. They

appeared to be constitutive for cob gene transcription. The second class of 11 mutants was

Lac" regardless of whether cyanocobalamin was present or absent in the medium. These

22

23

were designated negative mutants. All of the mutants produced in this study are listed in

Table 2.

Characterization of the Negative Mutants

The coblv.lac gene fusion was originally generated by the insertion of the defective

bacteriophage Mu dJ into the cobl operon. This genetic element carries a kanamycin

resistance maricer (Kan') in addition to the lac genes. All of the mini-tet insertions were

tested to determine whether or not they were cotransducible with Mu dJ by using

bacteriophage P22. Each mini-tet was transduced into the Mu dJ-containing fusion strain

(TTl0852), and the tetracycline-resistant transductant colonies were tested for simultaneous

loss of kanamycin resistance (indicating that the two markers are cotransducible). All of

the negative mutants were found to be cotransducible with the Mu dJ in the cobl operon.

These results indicated that the mini-tet insertions might also be somewhere in the cob

genes. A nutritional study was performed to determine if the mini-tets were also interfering

with the transcription of the cobU and cobTH operons. Each of the mini-tet mutants was

tested for growth under conditions that required cobalamin synthesis from the biosynthetic

intermediates cobinamide and DMBI. All of the negative mutants could make their own

DMBI and join it with cobinamide to form cobalamin; however, they were not able to

produce their own cobinamide. Thus, this result indicated that all the negative mutants

interfere only with the transcription of the cobl operon; transcription of the cobVi and coWII

operons is not disrupted.

Mapping the Positive Mutants

None of the 18 positive mutants were cotransducible with the Mu dJ insertion in the

cobl operon. These mutants were tested for the ability to produce cobalamin aerobically, in

order to ascertain if the repression of cob by oxygen had also been affected by these

24

TABLE 2. Mini-tet insertion mutants generated in this study.

Mutant Strains » Genotype or Description

RT871 RT872 RT873 RT874 RT875

RT876 RT877 RT878 RT879 RT880

RT881 RT882 RT883 RT884 RT885

RT886 RT887 RT888 RT889 RT890

RT891 RT892 RT893 RT894 RT895

RT896 RT897 RT898 RT899

btu-151 btu-152 btu-153 btu-154 btu-155

btu-156 loss of end-product repression for cob loss of end-product repression for cob loss of end-product repression for cob loss of end-product repression for cob

loss of end-product repression for cob loss of end-product repression for cob loss of end-product repression for cob loss of end-product repression for cob eut-lOOl

eut-1002 loss of end-product repression for cob loss of end-product repression for cob cob-2001 cob-2002

cob-2003 cob-2004 cob-2005 cob-2006 cob-2007

cob-2008 cob-2009 cob-2010 cob-2011

^ The mutations in strains RT877 through RT884 belong to a single group but have not been mapped by conjugation.

^ The mutations in strains RT887 and RT888 have been mapped by conjugation and are located between 47 and 62 map units on the 5. typhimurium chromosome.

25

mutations. This test was performed in a metE (TR6583) genetic background. None of the

positive mutants produced enough cobalamin to grow aerobically on glucose minimal

medium plates. This result indicated that Uie cob genes arc still sensitive to oxygen

repression in this group of mutants.

The ability of tiiese mutants to grow on limited (100-fold dilution) cobalamin was

also tested to determine if the appearance of the repression-insensitive phenotype in these

cells was being caused by a loss of cobalamin transport into the cell. The only positive

mutant which did not grow aerobically with limited cobalamin was RT871. However, this

is not conclusive evidence that the other mutants are not located in cobalamin transport

genes, since a mutation in the btuC locus might still allow the cells to grow with limited

cobalamin (5). Further experiments were performed to detemine if any of the other positive

mutants were cotransducible with markers near the three known loci for cobalamin

transport The results of these cotransduction experiments indicate that RT871 and RT874

are btuB mutants (they have lost the outer membrane-binding protein for cobalamin). Four

additional strains (RT872, RT873, RT875, and RT876) appear to be btuC mutants (they

have lost one or more of the cytoplasmic membrane proteins for cobalamin transport).

None of the strains appear to be tonB mutants (loss of an energy-transfer protein involved

in cobalamin transport). The results of these experiments are listed in Table 3.

The metE gene, which encodes the cobalamin-independent homocysteine

transferase, is repressed by cyanocobalamin (25). Mutations in or near the metH gene,

which encodes the cobalamin-requiring homocysteine methyltransferase, block this

repression (26). Thus, the ability of the positive insertion mutations to be cotransduced

with metH was also tested. None of the mutations was cotransducible with metH.

26 TABLE 3. Results of cotransductional crosses between the positive insertion mutations

and markers in or near known cobalamin transport genes.^

Mutant Strains

RT871 RT872 RT873 RT874 RT875

RT876 RT877 RT878 RT880 RT881

RT882 RT883 RT884 RT885 RT886

RT887 RT888

TT13775

(btu-2:Mu dJ)^

0 99 90 0

100

93 0 0 0 0

0 0 0 0 0

0 0

RT801 (argH87) c

48 ND ND 96 ND

ND 0 0 0 0

0 0 0 0 0

0 0

RT797 (trpA3)^

ND ND ND ND ND

ND 0 0 0 0

0 0 0 0 0

0 0

ND, not determined.

a The data are expressed as frequency of cotransduction [(number of colonies that inherit the unselected maricer/100 colonies that inherit the selected maricer) x 100].

b Inserted at the frmC locus.

c Cotransduccs with ^mB.

d Cotransduces with tonB.

27

Ability of Mutants tn Cnnvert Cobalamin into the r/>fantor^

The form of cobalamin (cyanocobalamin, metiiylcobalamin, or adenosycobalamin)

that mediates the end-product repression of the cob genes is unknown. Thus, a mutation

that affects the cells' ability to convert cyanocobalamin (used to supplement the plates) into

one of the cofactors could result in the positive phenotype. For the mutants to grow on

propanediol or ethanolamine as a sole cartx)n source, the cells would have to be able to

produce adenosylcobalamin; to grow on glucose in the absence of methionine, they must

produce methylcobalamin. The ability of the mutants to grow on minimal medium plates

with either glucose, propanediol, or ethanolamine as a sole carbon source was tested. The

plates were supplemented with a 100-fold concentrated solution of cobalamin to overcome

any transport deficiencies. All of the mutants were able to grow on the propanediol and the

glucose plates. However, two of the mutants were unable to grow on the ethanolamine

plates. This result indicated that these two mutants produce both methylcobalamin and

adenosylcobalamin, but that their ability to utilize ethanolamine as a carbon source is

impaired in some other way.

Mapping the Ethanolamine Mutants

The positive mutants that were unable to utilize ethanolamine as a sole cartx)n

source were characterized further. The eut operon encodes the enzymes for ethanolamine

utilization and requires the simultaneous presence of cobalamin and ethanolamine to induce

transcription (3,24). The operon also contains the gene (eutR) for a regulatory protein that

is required in order for the operon to be expressed.

Cotransdution experiment were performed to determine if the two ethanolamine-

deficient mutations were located within the eut operon. The data for this experiment are

shown in Table 4. These cotransduction experiments indicate that these mutations lie at the

28 TABLE 4. Results of cotransductions between Euf mini-tet insertion

mutants and markers in and near the eut operon a.

Mutants

TT10271

(eM/-18::MudJ)

RT869

(purCJ)^

RT870

(cysA533) ^

RT885 24 0 23

RT886 34 20

a The data are expressed as finequency of cotransduction (as defined in Table 3).

^ The purC gene is upstream from the proximal end and the cysA gene is downstream from the distal end of the eut operon.

29

distal end of the eut operon. The eut R regulatory gene has also been mapped at the distal

end of the operon (24). Therefore, the two mini-tet insertions may occur near or within the

eutR gene itself.

To determine if these two mutants were within the eutR gene, they were

transduced into S. typhimurium strain TTl 1580. This strain contains a deletion in the eut

operon which runs from die promoter through the eutR gene. If the mutations occuired

within the deleted area, then it would not be possible for the transduction to produce a Eut"*"

phenotype. However, both of the eut insertion mutations were able to repair the eut

operon. Thus, both of the mutations appear to lie outside of the deletion and downstream

from the eutR gene.

Mapping the Remaining Mutants

The 10 positive mutants which did not appear to be blocked in cobalamin

transport or ethanolamine catabolism were arranged into two additional groups. Mini-kan

insertions were isolated that were linked to several randomly-chosen mini-tet insertions.

Linkage to other unmapped mini-tet insertions was then tested by cotransduction. One

group contains two mutants (RT887 and RT888), while the other group contains the

remaining eight mutants (RT877-RT884). The mutations in the first group were mapped

by conjugation, they lie between 47 and 62 map units on the 5. typhimurium chromosome

(Table 5). The mutations in the second and larger group were not mapped.

p-Galactosidase Assavs

p-galactosidase assays were performed on all of the positive mutants in order to

quantify the effect that these mini-tet insertions have on the end-product repression of the

cob genes. Control assays were also performed on the following three control strains

30

TABLE 5. Cbnjugational mapping of mini-tet insertion mutations in strain RT887.

Auxotrophic recipient

TR5654 (thrA9)

TR5655 (leu-485)

TR5656 (proA36)

TR5657 (purE8)

TR5658 (pyrCT)

TR5660 (pyrF146)

TR5686 (aroDMO)

TR5662 (his01242 his-2236)

TR5663 (purFI45)

TR5666 (serA13)

TR5667 (cysG439)

TR5668 (cysE396)

TR5669 (ilv-508)

TR5671 (pyrB64)

TR5688 (PMM755)

Map units

0

2

6

11

23

33

36

42

47

62

72

79

83

98

96

Numbers of prototrophi ic colonies in matings with Hfr donor derived from

TT627

597

164

255

270

340

552

430

612

980

41

57

146

219

152

16

TT628

ND

ND

ND

ND

ND

ND

ND

76

31

752

623

ND

ND

ND

ND

ND, not determined.

a The Hfr donors were derived from strains containing F plasmids that were originally inherited from TT627 and TT628. The plasmids insert into the chromosome in opposite orientations.

31

RT504, RT934, and RT935. The RT934 strain contains a cobl.Mu dJ insertion but has an

otherwise wild-type (LT2) genetic background. This strain was used to establish the

normal level of cobl transcription in 5. typhimurium. In addition to the cobl:Mu dJ

insertion, the RT935 strain also contains a deletion which removes the entire eut operon.

This strain was used to determine what effect the eut operon has on the expression of the

cobl operon. RT504 is a wild-type E. coli K-12 strain which was grown in the presence

and absence of isopropyl-p-D-thiogalactoside (IPTG) as an inducer for the lac genes.

These control assays were used to establish the upper and lower limits for the entire set of

assays. The assays on all of the S. typhimurium strains were performed both in the

presence and absence of cyanocobalamin (Table 6).

32

TABLE 6. p-Galactosidase enzyme activity in permeabilized cells.a b

Strains ^

Parent strain

RT934

Transport mutants

RT871 RT872 RT873 RT874 RT875 RT876

Unmapped mutants

RT877 RT878 RT880 RT881 RT882 RT883 RT884

Eut" mutants

RT885 RT886 RT935

Cells grown with cobalamin

1396 ±87.55

1400 -1- 49.65 1835+92.16 1659+ 0.00 1708 + 99.02 1689 ±90.07 1658 ± 20.00

2030+ 79.00 2017+ 27.05 2103+ 57.88 2314+ 84.92 1993+ 90.29 1087 + 105.90 1619 ± 52.28

2121+ 22.91 1774+139.30 1707+ 56.60

Mutants mapped by conjugation

RT887 RT888

1202 + 59.25 914 ±32.65

Cells grown without cobalamin

235 ± 27.00

1450 + 72.22 1400 + 66.37 1861 +21.72 905 + 22.73

1084+ 0.00 1485 ±68.13

878 + 33.49 1422 + 87.53 1350 + 38.58 2105 + 26.86 1645 + 34.42 928 + 45.10

1070 ± 32.00

2062 + 21.48 1532 ± 4.30 1008 ± 56.60

854 + 26.51 781 ± 5.32

%

change in activity

83.17

-3.45 23.71 10.85 47.01 35.82 10.43

56.75 29.50 35.81 9.03

17.46 14.63 33.91

4.95 13.64 40.95

28.95 14.55

^ The specific activity is given in units of nmol of ONPG/min/units of absorbance at 650 nm. The numbers in this table represent the average of four replications + the standard deviation.

"v- .

33

TABLE 6, continued.

^ Two £. coli controls that were grown with and without IPTG as the inducer were also assayed. Four replications of each control were perfonned. The induced cells produced 9899 ± 331.20 units of activity, while die uninduced control produced 162.0 ± 19.04 units.

^ The cob'24v^yx dJ insertion was transduced from strain TT10852 into each of the strains listed in this table in order to assay for p-galactosidase activity.

CHAPTER IV

DISCUSSION

Mini-tet insertions which alter end-product repression of the cob genes were

isolated in this study. The effect of these mutations on the expression of die cob genes was

determined by performing growth studies, and by assaying for the p-galactosidase activity

of coblr.lac fusions in die mutant cells. The 29 different mutants were divided into two

distinct groups, based on the Lac phenotype that they exhibited on lactose MacConkey

indicator plates in the presence and absence of cyanocobalamin. These two groups were

designated positive and negative mutants.

The 11 negative mutants were shown by both cotransduction and nutritional

experiments to lie within or near the cobl operon. The mini-tet insertions that confer the

negative phenotype could be regulatory mutations that disrupt an operator site that is

involved in the end-product repression of the cobl operon. Alternatively, they might

interfere with the transcription of the operon by being inserted in the promoter or between

the promoter and the cobv.lac fusion. The exact location of these mini-tet insertions can be

determined more precisely by deletion mapping. This has not been done at this time for

lack of a detailed genetic map of the cobl genes.

The importance of this group of mutants lies in the fact that all of the mutations

that inactivate transcription of the cobl operon are inserted in or near the c^ron itself.

These mutations appear to produce the negative phenotype in the cells because they

physically block the transcription of the operon. If this is true, it suggests that the cob

genes are not regulated by an activation mechanism.

The 18 positive mutants can be subdivided into four different groups based on

their map locations, and the types of genes in which they are found. Six of these mutants

34

35

belong to the first group and are blocked in cobalamin transport. Two of die insertions

occur in die btuB gene,while the otfier four are in the btuCED genes. These mutations give

die cells a repression-insensitive phenotype because diey decrease die ability of die cells to

transport cobalamin across die cell wall. The remaining groups of positive mutants are

probably involved more direcdy in die regulation of cob. The insertions in die second

group of mutants have been mapped by cotransduction, and diey lie near die distal end of

die eut operon at 50 map units on die chromosome (Table 4). The insertions in die diird

group of mutants have been mapped more crudely by conjugation. They lie between 47

and 62 map units on the chromosome, but they are not cotransducible with eut genetic

markers. The insertions in the fourth and largest group of mutants have not yet been

mapped. Because of the larger number of mutants belonging to this group, the operon in

which these insertions lie may be larger than the operons in which the other insertions are

located.

The mutations in the eut operon render the ceU incapable of utilizing ethanolamine

as a sole carbon source. They are located near the region of the operon that codes for a

regulatory protein (EutR) which must be produced in order for the operon to be expressed.

This part of the eut operon has been designated region in (24). The induction of the eut

operon is dependent on the simultaneous presence of cobalamin and ethanolamine; cither

compound by itself does not stimulate transcription. I propose a regulatory model in which

two proteins are involved, one that binds to ethanolamine, and a second which binds to

cobalamin. After these proteins bind to their effectors, they change conformation and bind

to separate regulatory sites at the eut operon. This would allow the operon to be expressed.

In this model, the protein which binds to cobalamin would also bind to a regulatory site at

the cob genes. When this protein binds to the cob operator site, it blocks the transcription

of the cob operon. This would only occur when cobalamin was present in excess of the

cell's needs.

35

In order for this model to work, the cobalamin-binding protein would have to be

expressed by a weak promoter which is regulated independendy of die eut operon's main

promoter, so that a low level of this protein would always be present in the cell. A weak

promoter of this type is known to exist in the eut operon just upstream from the gene for

the EutR regulatory protein (24). However, the coblr.lac fusions are still partially

repressed in a strain (RT935) that also carries a deletion of the eut operon including the

eutR gene, and in strains (derivatives of RT885 and RT886) that carry the two eut insertion

mutations (Table 6). This in^lies diat the regulatory protein which is affecting the

transcription of both the eut and cob operons is not EutR but is produced by a region of the

eut operon that is downstream from the eutR gene. Further evidence for this conclusion is

the fact that P22 lysates that were grown on both strains containing the eut insertion

mutations could repair the eut deletion.

The effect of both the deletion and the insertion mutations on the end-product

repression of die cob genes seems to be diat they reduce die regulatory protein's rate of

syndiesis. They do not disrupt the regulatory gene itself, since die cob genes are still

partially repressed by cobalamin in all of diese strains. In die model stated above, diis

would mean that bodi of die regulatory proteins would have to have dieir own weak

promoters, and that the EutR protein is die ethanolamine-binding regulatory protein.

There are two remaining groups of mutants which have not been characterized at

this time. The role that these genes which have been inactivated by insertion mutations

normally play in die regulation of cob is not known, but diey may be involved in modifying

cobalamin into the chemical form diat is used as an effector for die regulatory protein. For

example, the regulatory protein may bind only adenosylcobalamin, but not

methylcobalamin or cyanocobalamin.

Research into die mechanism involved in die end-product repression of cob has just

begun. There are a number of questions left unanswered, and a number of additional

37

experiments that should be perfOTmed. The location of the other two groups of mutants

that affect the repression of cob should be mapped more precisely. A more detailed map of

the eut operon should be produced to determine whether the mutations affecting the

regulation of cob are in region HI. The regulatory protein itself should be characterized to

confirm that it does bind cobalamin, and to determine its binding characteristics.

REFERENCES

1. Abeles, R. H., and H. A. Lee, Jr. 1961. An intramolecular oxidation-reduction requiring a cobamide coenzyme. J. Biol. Chem. 236: 2347-2350

2. Babior,B. M. 1982. Ethanolamine ammonia-lyase, p. 263-288. In: D. Dolphin (ed.), B.12, YQl. 2. John Wiley & Sons, New York.

3. Bassford, P. J., Jr., and R. J. Kadner. Genetic analysis of components involved in vitamin Bn uptake in Escherichia coli. J. Bacteriol. 132: 795-805.

4. Blackwell, C. M., and J. M. Turner. 1978. Microbial metabolism of amino alcohols: formation of coenzyme Bn-dependent ethanolamine ammonia-lyase and its concerted induction in Escherichia coli. Biochem. J. 176: 751-757.

5. Bradbeer, C. 1965. The clostridial fermentation of choline and ethanolamine. J. Biol. Chem. 240: 4675-4681.

6. Chumley, F. G., R. Menzel, and J. R. Rodi. 1979. Hfr formation directed by Tn70. Genetics 91: 639-655.

7. Crueger, W., and A. Crueger. 1982. Biotechnologv: textbook of industrial microbiolopv. Science Tech Pubhshers, Madison, Wis.

8. Davis, R. W., D. Botstein, and J. R. Rodi. 1980. Advanced bacterial generics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

9. Ellenbogen, L. 1975. Absorption and transport of cobalamin: intrinsic factor and die transcobalamins, p. 215-287. In: B. M. Babior (ed.), Cobalamin: biochemistrv and pathophvsiolo^. John Wiley & Sons, New York.

10. Elliott, T., and J. R. Rodi. 1988. CharacterizationofTn70rf-Cam: a transposition-defective TnlO specifiying chloramphenicol resistance. Mol. Gen. Genet. 213: 332-338.

11. Escalante-Semerena, J. C, and J. R. Rodi. 1987. Regulation of cobalamin biosynthetic operons in Salmonella typhimurium. J. Bacteriol. 169: 2251-2258.

38

39

12. Frey, B. J. McCloskey, W. Kersten, and H. Kersten. 1988. New function of vitamm B^: cobamide-dependent reduction of epoxyqueuosine to queuosine in dlNAs of Escherichia coli and Salmonella typhimurium. J. Bacteriol. 170: 2078-2082.

13. FolUs , R. H., Jr. 1958. Deficiencv disea ce, Charles C. Thomas Publishers, Spnngfield, 111.

14. Friedmann, H. C. 1975. Biosyndiesis of corrinoids, p. 75-109. In: B. M. Babior (ed.), CQbalamin: biochemistry and pathophvsiology. John Wiley & Sons, New York.

15. Hogenkamp, H. P. C. 1975. The chemistry of cobalamins and related compounds, p. 21-93. In: B.M. Babior (ed.). Cobalamin: biochemistry and pathophvsiology. John Wiley & Sons, New Yoric.

16. Jeter, R. M., and J. R. Roth. 1987. Cobalamin (vitamin B12) biosynthetic genes of Salmonella typhimurium. J. Bacteriol. 169: 3189-3198.

17. Knivett, A. 1960. The microbiological production ofvitamin B12 and sulphide from sewage. Prog. Ind. Microbiol. 2: 29-45.

18. Lehninger, A. L. 1982. Principles of Biochemistry, edition III. Worth Publishers, New York.

19. McGee, D. E., and J. H. Richards. 1981. Purification and subunit characterization of propanediol dehydratase, a membrane-associated enzyme. Biochemistry 20: 4293-4398.

20. Mervyn, L., and E. L. Smith. 1964. The biochemistry ofvitamin B12 fennentation. Prog. Ind. Microbiol. 5: 151-201.

21. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

22. Mulligan, J. T., W. Margolin, J. Heilig Krueger, and G. C. Walker. 1982. Mutations affecting regulation of methionine biosynthetic genes isolated by use of met-lac fusions. J. Bacteriol. 151: 609-619.

40 23. Reynolds, P. R., G. P. Mottur. and C. Bradbeer. 1980. Transport of vitamin Bn in

tscherichia coli. Some observations on die role of die die gene products of btuC and / onB. J. Biol Chem. 255:4313-4319.

24. Roof, D. M., and J. R. Rodi. 1988. Edianolamine utilization in Salmonella t typhimurium. J. Bacteriol. 170: 3855-3863.

25. Rowbury, R.J. 1983. Mediionine biosyndiesis and its regulation, p. 191-211. In: K. M. Hermann and R. L. Somerville (ed.), Amino acids: biosynthesis and generic regularion. Addison-Wesley Publishing Co., Reading, Mass.

26. Saint-Girons, I., N. Duchange, G. N. Cohen, and M. M. Zakin. 1984. Structure and autoregulation of die metl regulatory gene in Escherichia coli. J. Biol. Chem. 259: 14282-14285.

27. Scarlett, F. A., and J. M. Turner. 1976. Microbial metabolism of amino alcohols. Ethanolamine catabolism mediated by coenzyme Bi2-dependent edianolamine ammonia-lyase in Escherichia coli and Klebsiella aerogenes. J. Gen. Microbiol. 95: 173-176.

28. Stryer, L. 1981. Biochemistry. 2nd edition. W. H. Freeman and O)., New York.

29. Taylor, R. T., and H. Weissbach. 1973. N^-Mediyl-tetrahydrofolate-homocysteine methyltransferases, p. 121-161. In: P.D. Boyer (ed.). The Enzymes. 3rd edition. Academic Press, New York.

30. Urbanowski, M. L., L. T. Stauffer, L. S. Plamann, and G. V. Stauffer. 1987. A new methionine locus, metR, that encodes a trans-acting protein required for activation of metE and metH in Escherichia coli and Salmonella typhimurium. J. Bacteriol. 169: 1391-1397.

31. Vogel, H. J., and D. M. Bonner. 1956. Acetylomithinase of £5cAenc/ua c<?//: partial purification, and some properties. J. Biol. Chem 218: 97-106.

32. Whitfield, C. D., E. J. Steers, Jr., and M. Weissbach. 1970. Purification and properties of 5-methyltetrahydropteroyltriglutamate-homocysteine methyltransferase. J. Biol. Chem. 245: 390-401.

33. Wolf, J. B., and R. N. Brey. 1986. Isolation and genetic characterization of BanV/wj megaterium cobalamin biosynthesis-deficient mutants. J. Bacteriol. 166: 51-58.

PERMISSION TO COPY

In presenting this thesis in partial fulfillment of the

requirements for a master's degree at Texas Tech University, I agree

that the Library and my i&ajor department shall make it freely avail­

able for research purposes. Permission to copy this thesis for

scholarly purposes may be granted by the Director of the Library or

my major professor. It is understood that any copying or publication

of this thesis for financial gain shall not be allowed without my

further %n:itten permission and that any user may be liable for copy­

right infringement.

Disagree (Permission not granted) Agree (Permission granted)

Student's signature ^^^•^r 12^ J^

Student's signature Ain^mM*^

Date Date ii-fn-m?