v

. V“? to m:
Lam: 4.4..” an? ’3'
.. T ~ 5-" ""f .‘ ¢ .'
o P
. , tr .a,‘1-E~$
m page emery;

‘i'
‘.

3mg 515’

 
      

This is to certify that the
thesis entitled
Characterization of a Rhizobium leguminosarum
mutant with a defect in iron uptake

presented by

Theodore Ralph John

has been accepted towards fulfillment
of the requirements for

M-S. degree in Botany and Plant
Pathology

gégflfi), W
Major professor

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

Date _La_SepJ-.emher_1983

 

 

MSU

LIBRARIES
-..—

 

 

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.

 

 

’
hem" an)“

 

 

"on m r
P‘iyi‘Zujfl‘f" ' v 7533.. .20"; 31"“ kt!

 

CHARACTERIZATION OF A RHIZOBIUM LEGUMINOSARUM MUTANT
WITH A DEFECT IN IRON UPTAKE

 

By
Theodore Ralph John

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Botany and Plant Pathology

1983

ABSTRACT

CHARACTERIZATION OF A RHIZOBIUM LEGUINOSARUM MUTANT
WITH A DEFECT IN IRON UPTAKE

 

By
Theodore Ralph John

A mutant strain (116) of Rhizobium leguminosarum which

 

accumulates porphyrins has been isolated. This study was
undertaken to characterize biochemically the lesion in 116.
The mutant induces white, ineffective nodules in peas.
In broth culture, 116 grows as well as both the parental
strain (1062) and a spontaneous revertant (strain 74-11).
116 has reduced levels of cytochromes b and c. In extracts
of cells grown in minimal medium containing high levels of
iron, activities of G-aminolevulinic acid synthease and
G-aminolevulinic acid dehydrase show no difference beteween
116 and 1062. There is also no difference in the rate of
paprhyrin formation in dense cell suyspensions and cell-free
extracts of 116, 1062, and 74-11. In media containing low
iron, 116 shows no growth while 1062 and 74-11 show normal
gorwth. The initial rate of 55Fe uptake from 116 grown
in low iron is 10 times less than the rate from 1062 and

74-11. These data suggest that 116 is defective in the

Theodore Ralph John

initial uptake of iron. The significance of rhizobial iron

uptake in the development of the Rhizobium-legume root

 

nodule is discussed.

Flower in the crannied wall,

I pluck you out of the crannies,

I hold you here, root and all, in my hand,
Little flower - but if I could understand
What you are, root and all, and all in all,

I should know what God and man is.

Alfred, Lord Tennyson

ii

ACKNOWLEDGEMENTS

I would like to thank my major professor, Dr. Kenneth
Nadler, for opening my eyes to porphyrins and iron; his
guidance through this research is most appreciated. I would
also like to thank the other members of my guidance
committee, Dr. Robert Bandurski and Harold Sadoff, whose
constructive criticisms of my work I will long value. I
also thank Dr: Robert Scheffer who referred at my thesis
defense.

The assistance of Dr. John Chisnell and Mr. Dennis
Reinecke with the HPLC is gratefully appreciated. I wish to
thank Mr. Don Versteeg for performing the atomic absorption
spectroscopy. Thanks to BMP for the PMTs. Special thanks
go to Ms. Paula Moan for typing this manuscript. I also
wish to thank Comm—Well Sales and Engineering, Inc. for the
financial assistance which they have provided over the
years.

Finally I want to thank Mr. Robert Creelman and Mr.
James Smith with whom I had many enlightening conversations

and who are ture friends.

iii

TABLE OF CONTENTS

LIST OF FIGURES O O O O O O O O O O O O O 0

LIST OF TABLES O O O O O O I O O O O O O O 0

LIST OF ABBREVIATIONS . . . . . . . . . . .

INTRODUCTION 0 I O O O O O O O O O O O O C O

The Structure, Function and Biosynthesis

of Leghemoglobin . . . . . . . . . .
The Biosynthesis of Heme . . . . . . .

Bacterial Strains with Mutations in Genes

Coding for Heme Biosynthetic Enzymes
Effects of Iron Deficiency on Heme

Biosynthesis . . . . . . . . . .
Effects of Iron Deficiency on Non-heme-

containing Enzymes . . . . . . . .
Microbial Iron Uptake . . . . . . .

MATERIALS AND METHODS . . . . . . . . . . .

Bacterial Strains . . . . . . . . . .
Media . . . . . . . . . . . . . . .
Bacterial Growth Conditions . . . .
Growth of Bacteria in Low Iron Media
Plant Growth and Nutrition . . . . .
Measurement of Nitrogen Fixation . .
Preparation of Cell-Free Extracts .
Cytochrome Difference Spectra . . .
Enzyme Assays . . . . . . . . . .
Production of Pyrroles by Dense Suspe
of Bacterial Cells . . . .

n

S

1.

Production of Tetrapyrroles by Cell- Free

Extracts . . . . . . . . . . .
Measurement of Iron Uptake by Bacterial
Dense Cell Suspensions . . . . . . .
Analytical Methods . . . . . . . . . .

RESULTS 0 O O O O O O O O O O O O O O O I 0

iv

0

S

n

Page

vi
vii

viii

Page
DISCUSSION . . . . . . . . . . . . . . . . . . . . . 58

APPENDIX C O O O O O O O O O O O O O O O O O O O O O 65

BIBLIOGRAPHY O O O O O 0 O O I O I O O O O O O O O O 68

LIST OF FIGURES

Figure Page
1. The heme biosynthetic pathway . . . . . . . . . . 4

2. Acetylene reduction by pea roots nodulated with
various Rhizobium leguminosarum strains . . . . 33

 

3. Growth of R. leguminosarum strains in minimal
med ium O O O O O O O C O O O O C O O O O O O O O 36

 

4. Difference spectra (dithionite reduced minus
ferricyanide oxidized) of cell-free extracts
of Rhizobium leguminosarum strains 1062, 16,
74-Tl . . . . . . . . . . . . . . . . . . . . . 39

 

5. HPLC tracings of porphyrin methyl esters as
isolated from log-phase cells of Rhizobium
leguminosarum strains 1062, 116, and 74-11 . . . 43

 

 

6. HPLC tracings of porphyrin methyl esters as
isolated from dense cell suspensions of log-
phase cells of Rhizobium leguminosarum strains
1062, 116, and 74-11 incubated in d-amino-
levulinic acid . . . . . . . . . . . . . . . . . 45

 

7. Growth of R. leguminosarum strains in low iron
and high iron media 0 I O I O O O O O O O O O O O 51

 

8. 55Fe uptake of Rhizobium leguminosarum strains
grown in high iron or low iron minimal media . . 54

 

vi

LIST OF TABLES

 

 

 

 

 

Table Page
1. Rhizobium leguminosarum strains . . . . . . . . . 17
2. Symbiotic properites of Rhizobium leguminosarum

strains 0 O O O O O O O O O O O O O C I O C O O 35

3. Concentrations of cytochromes in Rhizobium

leguminosarum strains . . . . . . . . . . . . . 41
4. Rate of PBG and porphyrin accumulation in

suspensions of Rhizobium leguminosarum cells

inCUbated With 0.5 mM ALA O O O O O O O O O O 48
5. Activities of enzymes of heme biosynthesis in

cell-free extracts of R. leguminosarum . . 49
6.

 

Initial and final rates of 55Fe uptake for

Rhizobium leguminosarum strains grown in high
iron andilow iron media

. . . . . . . . 57

vii

ALA

ALAS
ALAD

PBG

URO
UROgen
COPRO
COPROgen
PROTO
PROTOgen
LI

HPLC
SDS-PAGE

LIST OF ABBREVIATIONS

6-aminolevulinic acid

é-aminolevulinic acid synthase
G-aminolevulinic acid dehydrase
porphobilinogen

uroporphyrin

uroporphyrinogen

c0proprophyrin

coproporphyrinogen

protOporphyrin

protoporphyrinogen

low iron

high performance liquid chromatography
sodium dodecylsulfate polyacrilamide gel

electrophoresis

viii

INTRODUCTION

The Structure, Function, and Biosynthesis
of Leghemoglobin

 

Leghemoglobins are hemOproteins unique to nitrogen-
fixing root nodules induced by the bacterium Rhizobium in
its symbiotic legume host (3). Leghemoglobin comprises
25-30% of the total nodule soluble protein (81). Most
legumes have several chromatographically distinguishable
leghemoglobins in their root nodules (27), but they all are
similar enough to one another to classify them under the
general heading of "leghemoblobin". Leghemoglobin is
similar to vertebrate myoglobin in spectral prOperties
(65, 79) and in amino acid sequence (27). It has been
suggested that leghemoglobin is essential to nitrogen
fixation in legume root nodules because nodules which lack
leghemoglobin generally do not fix nitrogen (9, 82).

Leghemoglobins vary in molecular weight from 15,000 to
17,000 but all contain the prosthetic group protoheme IX.
The prosthetic group sits in a hydrOphobic pocket of the
ap0protein where the fifth ligand binding position of the
iron atom of the heme molecule is coordinated to the
imidazole ring of a histidine residue in the protein (4).
The sixth ligand binding position remains open to bind with
other molecules, including molecular oxygen. All

1

2

leghemoglobins have a very high affinity for oxygen relative
to other hemoglobins; e.g. leghemoglobin is half-saturated
with oxygen at a pressure of 0.05 mm Hg as compared to 4 to
14 mm Hg for mammalian hemoglobins (1, 87). The role of
leghemoglobin in the root nodule symbiosis is thought to be
related to this oxygen binding function (87).

The enzyme nitrogenase, which is responsible for the
reduction of dinitrogen to ammonia in several species of

bacteria including Rhizobium, is irreversibly inhibited by

 

molecular oxygen (18). Yet the enzyme requires large
amounts of ATP which presumably is provided through the bac-
terial electron transport chain. Leghemoglobin is thought to
act as an oxygen "buffer": it provides oxygen for rhizobial
oxidative phosphorylation, but keeps the oxygen tension low
enough to allow for rhizobial nitrogenase activity (87).

The biosynthesis of leghemoblobin adds a special qual-

ity to the Rhizobium-legume symbiosis. Indirect evidence

 

suggested that leghemoglobin apoprotein was a plant gene
product (22, 25, 80). When the same plant was nodulated

with various Rhizobium strains, the leghemoglobins produced

 

were electrophoretically indistinguishable. Similarly, when
different legume species were nodulated with the same

Rhizobium strain, then the leghemoglobins produced were

 

chromatographically and electrophoretically distinct (22,
25). It was also shown that the mRNA coding for leghemo-
globin was associated with 808 ribosomes (80). Baulcombe

and Verma (7) provided more direct evidence that the legume

 

 

3

host was the site of apoprotein synthesis by showing that
leghemoglobin cDNA hybridized to soybean DNA and not to

Rhizobium DNA. In contrast, the heme prosthetic group

 

 

appears to be bacterial in origin. It was shown in
radiotracer studies that the bacteroids make the heme moiety
(22, 33) and it was further shown that as leghemoglobin
content of soybean nodules increased, the activities of two
key heme biosynthesis enzymes increased in the bacteroids,

but not in the plant (58).

The Biosynthesis of Heme

The biosynthesis of heme in Rhizobium is similar to

 

that in other microorganisms and animals and occurs via the
pathway shown in Figure 1. The first step in the pathway is
the reaction of glycine and succinyl-CoA to 6-aminolevulinic
acid (ALA). This reaction is catalyzed by the enzyme
G-aminolevulinic acid synthase (ALAS, EC 2.3.1.37). Two
molecules of ALA are condensed by the action of 6-amino-
levulinic acid dehydrase (ALAD, EC 4.2.1.24) to form the
dicarboxylic monopyrrole porphobilinogen (PBG). The enzyme
PBG-deaminase (EC 4.3.1.8) catalyzes the sequential head to
tail condensation of four PBG molecules into the linear
tetrapyrrole hydroxymethylbilane (also known as preuropor-
phyrinogen), with a loss of four molecules of ammonia.
Hydroxymethylbilane is an unstable intermediate; it will
nonenzymatically cyclize into uroporphyrinogen 1. However,

in the presence of the next enzyme of the heme biosynthesis

GLYCINE + SUCCINYL CoA

 

co2
S-AMINOLEVULINIC ACID
(ALA)
1 +ALA
PBG
+3PBG

4~H3 — — - -)URO I
UROgen III — — -) UROIII

“oat/1

COPROgen III- — -) copnom:

49' + zco2 4

PROTOgon IX

6e- (/1

PROTO II
1 +Fe’2

m

 

Figure 1. The Heme Biosynthetic Pathway (Pigmented

compounds are underlined)

Abbreviations PBG = porphobilinogen
URO = urOporphyrin
UROGEN = uroporphyrinogen
COPRO = coprOporphyrin
COPROgen = coprOporphyrinogen
PROTO = protoporphyrin
PROTOgen = protoporphyrinogen

5

pathway, uroporphyrinogen cosynthetase, hydroxymethylbilane
is converted into uroporphyrinogen III (UROgen III). The
four acetyl carboxyl groups of UROgen III are removed by
uroporphyrinogen III decarboxylase (EC 4.1.1.37) to give
coproporphyrinogen III (COPROgen III), a tetracarboxyl
compound. COPROgen III is oxidatively decarboxylated to
form protoporphyrinogen IX (PROTOgen IX) by the enzyme
COPROgen oxidative decarboxylase (EC 1.3.3.3). PROTOgen IX
is then oxidized to protoporphyrin IX (PROTO IX) by PROTOgen
oxidase (EC 1.3.3.4). The final step of heme biosyntehsis
is mediated by the enzyme ferrochelatase (EC 4.99.1.1),
which inserts ferrous iron into PROTO IX, with heme as the
product.

The rate-limiting step of tetrapyrrole formation in
bacteria, plants, and animals is at ALAS (34); in fungi,

ALAD activity is the rate-limiting step. In Rhizobium, the

 

rate-limiting step also appears to be at ALAS (28, 43, 58).

The regulation of heme biosynthesis in Rhizobium probably

 

involves the control of ALAS activity. In Rhodopseudomonas

 

spheroides, for example, it has been shown that ALAS is

 

subject to feedback inhibition (16) and enzyme repression
(48) by heme, activation by organic sulfides (63, 64), and

oxygen (46). In Micrococcus denitrificans, iron induces

 

formation of ALAS and heme inhibits ALAS activity (76). In

Rhizobium japonicum, it has been shown that restricted

 

 

aeration increases ALAS activity (58), while low iron

decreases ALAS activity (70). There is a recent report (50)

6

which describes the cloning of the gene for ALAS from R.
meliloti. This clone will be useful as a probe for the
studies of transcription of the ALAS gene and should provide
more information about the regulation of this enzyme under

different growth conditions.

Bacterial Strains with Mutations in Genes Coding
for Heme Biosynthetic Enzymes

 

 

A hemin-deficient bacterial mutant strain was first
isolated in 1953 by Jensen and Thofern (39). This

Staphylococcus aureus mutant appeared to be "leaky" (it did

 

not have a strict auxotrophic requirement for heme) in that
the mutant grew as "small colony variants" in the absence of
heme. Since then, similar heme mutant strains have been
reported in a variety of organisms. In most cases, mutagen-
esis was carried out with N-methyl-N'-nitro-N'nitrosoguan-
idine. Heme mutants were determined by a variety of methods
including: (i) response of mutant growth to added hemin or
precursors; (ii) loss of cytochromes or respiratory activ-
ity; (iii) loss of catalase activity; and (iv) accumulation
of porphyrins in mutant cultures as seen by the appearance
of a reddish color in the culture or by seeing a pinkish-red
fluorescence when illuminated with ultraviolet light.
Various mutant strains have been isolated with defects
in almost every step of the heme biosynthetic pathway. The
most frequent type appearing in the literature are mutants
deficient in ALAS activity (35, 49, 50, 73, 88). When these

mutants are grown in the presence of ALA, they show wild

type levels of cytochromes and catalase activity. The
reason that so many mutants of this type have been isolated
may be related to the regulatory role that ALAS plays in
heme biosynthesis: a mutation occuring either on the ALAS
gene itself or on one of the regulatory gene sequences for
ALAS will result in an ALAS-defective mutant. Also, ALA
readily enters cells to overcome a mutant deficiency,
whereas other heme precursors generally do not. McConville
and Charles (53) reported the isolation of a variety of heme
synthesis secondary mutants which were obtained from
hemin-permeable mutants of E. 2911. Their group 1 and group
2 mutants were purported to be defective in ALAD based on
the fact that these mutants were unable to convert ALA to
ur0porphyrin. However, they did not assay for ALAD activity
in their mutants; it is possible that these mutants were not
defective in ALAD as reported, but in PBG deaminase. The
same authors did report the isolation of three E. ggli K 12
mutants which accumulated PBG (52). In this case they
showed that the mutants had wild type activities of ALAD,
but only one-tenth the wild type activity of PBG deaminase.
Uroporphyrin- (53), coproporphyrin- (20, 56), and
protoporphyrin-accumulating mutants (20, 32, 53) have also
been reported. The lesions in these organisms were
determined by feeding ALA to whole cells and determining the
porphyrins accumulated by thin layer chromatography.

There have been few reports on heme biosynthesis

mutants in Rhizobium. Leong, 35 31. (50) reported on the

 

8

isolation of two 3. meliloti mutants, A-34 and A-36 which
showed a strong dependence on ALA for growth. When assaying
ALAS activities in the mutants, they found the expected low
levels of synthase activity in mutant A-34; however, mutant
A-36 had wild type levels of ALAS activity in vitro. They
were unable to explain this observation but suggested that
the ALAS enzyme from A-36 may have an abnormally high Km

for one of its substrates or a very low K1 for an

inhibitor such as heme. Noel, 25 al. (66) isolated a number

of nitrogen fixation mutants from B. japonicum. Several of

 

them were described as strains which form white nodules on
soybeans, the white color being due to a presumptive lack of
leghemoglobin. They went on to show that nodules induced by
these mutants lacked heme when detected by the pyridine
hemochromogen assay. This assay is the least sensitive
method of quantitating heme, its limit of sensitivity being
about 30 mM heme. More sensitive methods are (i) measuring
the pyridine hemochromogen as the reduced minus oxidized
spectrum, and (ii) fluorometric determination of heme (74).
Thus, the mutants of Noel, 35 al. (66) may indeed have had
low levels of heme. Also, the heme deficiency that they
noted may have been only an indirect manifestation of the

true lesion causing the fix' phenotype.

Effects of Iron Deficiency on Heme Biosynthesis

 

The insertion of Fe+2 into protoporphyrin IX,

mediated by the enzyme ferrochelatase, is the final step in

9

heme biosynthesis. The lack of available iron thus has
profound effects on heme biosynthesis in particular and cell
metabolism in general.

One of the first observations on the effects of iron
deficiency was the loss of activity of enzymes which contain
heme as a prosthetic group. During their classic studies on
iron deficiency, Waring and Werkman (85) noted a loss of

catalase and peroxidase activity in Aerobacter indologenes

 

when grown under low iron. They were also unable to detect

any cytochromes in low iron-grown A. indolqgenes. A loss of

 

catalase activity was also demonstrated in Arthobacter JG-9

 

(17) and Nocardia gpaca (86) when grown in low iron medium.

 

Kauppinen also noted the loss of hemoproteins under low iron

growth in the yeast Candida guilliermondii (42). In addi-

 

tion to a loss of cytochromes and peroxidase/catalase activ-
ities, he found a decrease of succinate dehydrogenase activ-
ity and almost complete loss of aconitase activity. Further-
more, he noted that upon loss of activity for the two TCA
cycle enzymes just mentioned, under low iron there was an
increase in alcohol dehyrogenase activity which corresponded
to an increase in fermentation by these organisms. The
yeast would of course be required to shift its metabolism to
a fermentative pathway due to the loss of its cytochromes
and TCA cycle. Finally, in a recent study, Roessler and

Nadler noted that Rhizobium japonicum grown under low iron

 

also had a lower cytochrome content and decreased ALAS and

ALAD activities over that of high iron-grown cells (70).

10

Another common phenomenon of iron deficient growth in
microorganisms is the accumulation and excretion of
porphyrins (47). Middleton and Gunner (55) reported that,

in low iron-grown cultures, Arthrobacter globiformis showed

 

a general increase in the total porphyrins excreted; these
investigators did not determine the identity of the
porphyrins. Increased porphyrin excretion under iron
deficiency has been attributed to the release of feedback
inhibition by heme on ALAS (47). It was also suggested that
iron plays a role in the conversion of COPROgen to
protoporphyrin IX because, in many cases of iron deficiency,
coprOprOphyrin was determined to be the excreted porphyrin
(47, 77, and references therein). However, coproporphyrin
is not always the porphyrin which is accumulated. In both

R. japonicum (70) and Achromobacter metalcaligenes (26),

 

 

 

protoporphyrin was shown to be the porphyrin excreted under
iron-deficient growth. Although it appears that iron is
necessary for the conversion of COPROgen to PROTO in some
organisms, e.g. E. 2911, this is not a general requirement

for all microorganisms.

Effects of Iron Deficiency on Non-heme-
containing Enzymes

 

 

Iron has been shown to have an effect on other enzymes
not containing heme. An obvious class of proteins affected
by iron are the non-heme iron proteins. These proteins
contain iron and sulfur in their "active sites." Three

examples of these proteins are NADH dehydrogenase, succinate

11

dehydrogenase, and ferredoxin. The enzyme NADH dehydro-
genase is responsible for NADH oxidation in the terminal
electron transport system of aerobes and has several
paramagnetic iron-sulfur centers (68). In iron-deficient
batch cultures of Micrococcus denitrificans, Imai, gt a},
(37) could detect little or no EPR signal corresponding to
NADH dehydrogenase. As mentioned above, Kauppinen (42)
noted a decrease in succinate dehydrogenase activity from

Candida guilliermondii with decreasing iron in the growth

 

medium. Finally, Knight and Hardy (44) reported that iron

deficiency in Clostridium pasteurianum resulted in the loss

 

of synthesis of ferredoxin.

In contrast to enzymes which contain iron as an inte-
gral component, there are other enzymes and enzyme activi-
ties that are altered under iron-deficient conditions. As
noted above, Kauppinen (42) reported a decrease in the TCA
cycle enzyme aconitase, but increased activities of alcohol

dehydrogenase in iron deficient Candida guilliermondii.

 

This observation was attributed to a change in cellular
metabolism from dependence on the TCA cycle and electron
transport to fermentation. Finally, and not surprisingly,
iron limitation increases the activity of a number of
enzymes involved with synthesizing and degrading iron
chelating agents (siderophores) which function in trans-
porting iron into a cell (see below). Ito and Neilands (38)
isolated the siderophore 2,3-dihydroxybenzoylg1ycine from

iron-deficient cultures of Bacillus subtilis; this

 

12

compound is not found in iron-sufficient cultures. In

Escherichia coli, iron has been shown to repress the enzymes

 

which synthesize the siderophore enterochelin, and also the
enzyme enterochelin esterase, which is responsible for the

hydrolysis of enterochelin and release of iron to the cell

(67).

Microbial Iron Uptake

 

Iron is one of the most abundant elements in the
earth's crust, its abundance is surpassed only by aluminum,
silicon, and oxygen (60). The iron content of soils range
from 0.5 to 5% (13). However, only a minute part is
available to living organisms. Both Fe(II) and Fe(III) have
a high affinity for hydroxide ions which results in
extremely low solubilities of the iron hydroxides. The
KSp of Fe(OH)3 at 25°C has been estimated to be 10‘”-7
(10) and the equilibrium concentration of ferric ion at pH 7
is about 10'18 M (69). The major way in which
microorganisms solubilize and sequester this highly
unobtainable iron is through the use of siderophores.

Siderophores are specific chelators of iron (III).

They are synthesized and used by a variety of microorganisms
including bacteria (both gram positive and gram negative)
and fungi (45). Compounds with the characteristics of
siderophores have not been isolated from plants. However,
citrate, which is common in plants, is thought to play an

important role in iron metabolism (15). As Neilands has

13

pointed out (59), practically all microorganisms for which
the presence of siderOphores has been critically examined
has resulted in a positive test. The only possible
exception to the ubiquitous presence of siderophores in
microorganisms may be in the strict anaerobes; in their
reducing environment, iron is present as the more easily
attainable iron(II). Whether or not siderophores are

present in Rhizobium remains to be seen.

 

Under conditions of high iron (greater than ca. 1 uM),
siderophore production is repressed and the organism, e.g.
E. 2213, assimilates iron by the so-called "low affinity"
system. There is not much known about this method of iron
uptake. It is not thought that any specific chelators are
required for iron(III) uptake. Apparently under high iron
conditions, the organism is able to obtain iron from the
Fe(III) oxy-hydroxide polymers which exist under these
conditions (75). Nitrilotriacetate, a chelator of iron, has
been used to block the low affinity system (29). Mutants
which are unable to produce or assimilate siderophores are
still able to grow unimpaired provided that there is a
sufficient concentration of iron in their growth medium (30,
83).

Under low iron conditions (less than ca. 1 uM), sider-
ophore production is derepressed. Siderophores are defined
as relatively low-molecular weight (500-1000) compounds
which are virtually specific toward binding iron(III).

Structurally, siderophores fall into two categories:

14

hydroxamates and phenolates-catecholates (61). These
ligands have an extremely high affinity for iron, the
formation constants lie in the range of 10+20 to 10+50
(69). The molecular shape of the siderophore is designed
such that the ligand coordinates with all six bonds of the
iron octahedron.

Ferricsiderophore complexes are too large to freely
diffuse across a membrane into a cell. For example, the
permeability barrier of the outer membrane of two gram

negative organisms (E. coli and Salmonella typhimurium) have

 

been determined to be near 550-650 daltons (24). This means
that along with synthesizing a siderophore, a cell must also
synthesize a membrane-bound siderophore receptor protein.

The presence of these receptors have been demonstrated with

both mutants in E. coli and Salmonella typhimurium (59, 72)

 

and by direct visualization of the receptor proteins with
SDS-PAGE (14). Iron deficiency causes a derepression of the
synthesis of these receptor proteins. Once inside the cell,
the method of release of Fe(III) from the coordinating
ligand is thought to be through reduction of the iron (31,
51, 69) or by hydrolysis of the ferricsiderophore. The
redox potentials of some ferric siderophores have been
determined (19) and those for the hydroxamate-type have been
found to be within the range of physiological reductants.
After releasing its bound iron, some siderophores are
recycled back out of the cell to assimilate more iron; but

other, e.g. enterobactin, a catecholate-type, must be

15

metabolically degraded. This may have to do with the low
redox potentials found with this type of siderophore:
metabolic degradation is perhaps the only way to release the
iron from this siderophore (19, 69).

In a root nodule, Rhizobium is entirely dependent on

 

the plant host for iron. The endosymbiotic bacteroids
require iron for many important symbiotic functions. The
enzymes nitrogenase, ALAS, and ferrochelatase all require
iron; the latter two enzymes are necessary for providing the
heme prosthetic group of leghemoglobin. How the bacteroids
assimilate iron is not known. This investigation provides a
beginning toward understanding iron assimilation in

Rhizobium.

 

MATERIALS AND METHODS

Bacterial Strains

 

Rhizobium leguminosarum strain 1062 was derived from

 

strain 300 (Table 1, 41); it exhibits a pop+ (non
porphyrin-accumulating) phenotype and induces dinitrogen-

fixing root nodules on host peas (Pisum sativum L.). Strain

 

116 is a porphyrin-accumulating mutant of strain 1062
isolated after NTG mutagenesis (57) and exhibits a nod+
fix' pop' phenotype. Strain 74-11 is an apparently
spontaneous fix+ revertant of 116 (with a nod+ fix+

pop+ phenotype) isolated from the one reddish, effective

nodule, obtained upon inoculation of peas with 116.

Media

Rhizobium minimal medium (Y) and complete medium (TY)

 

were prepared as described by Beringer (8). (See appendix
for media recipes.)

Low iron (LI) minimal medium was Y medium except that
no iron was added and with the following modifications. All
chemicals used were chosen from commercially available
sources for having low amounts of iron contamination. The
chemicals were used without further purification except for
KZHP04, which, when made into a stock solution, was

16

17

 

opp mo
ucmuuo>ou msoocmucoam

xuoz mwnu

Awumpv omouqmo .omm
.musumz .Hm um .COumGLOh

+me +aoa +voc

uxwm uaoa +voc

+xwm +aom +vo:

+P mom Hum o. app «P on:

_ mom yum o— gnu a. mu:

mum op mum ep mus

_pu¢~
op—

moo.

 

mocoumwom

mmhuocmsm

omwuocmw ucm>maom

umnazz

:Hmuum

 

 

mcflwuum Enummocfieswoa ESHLONHLM

._ magma

18

extracted with 8-hydroxyquinoline by the procedure of Waring
and Werkman (84). All water used in LI medium was doubly
distilled, deionized, and passed over a Chelex 100 column
(H+ form). To remove any contaminating iron, all
glassware used in low iron studies was washed as described
by Waring and Werkman (84): 1) soap and water, 2) rinse
with distilled water, 3) soak in 95% ethanol saturated with
KOH, 4) rinse with distilled water, 5) soak in aqua regia
(18% conc. HN03 and 82% conc. HCl, v/v), 6) rinse with
distilled water, and 7) fill glassware to the brim with
distilled, deionized water and autoclave for 20 minutes.
The concentration of iron in L1 medium was determined
to be 3.3 parts per billion using a Hitachi 180-80 Atomic
Absorption Spectrophotometer. This corresponds to a
concentration of 0.059 umol per liter. The iron concen-

tration of normal Y medium was 122 umol per liter.

Bacterial Growth Conditions

 

Bacterial strains were maintained in either 20% or 40%
glycerol (v/v) at -20°C. Outgrowth was accomplished by
inoculating a TY slant with about 40 ul of frozen stock and
incubating 3 or 4 days at 28°C. Liquid cultures were inocu-
lated and grown as follows. Cells were washed from the
slant with 5 m1 sterile distilled, deionized water and used
to inoculate 500 ml medium (in 1 liter Erlenmeyer flasks)
with 1 m1 bacterial suspension from the slant. Shake

cultures at 130 rpm at 28°C on an orbital shaker; cultures

19

were harvested after 24 hours, when they had reached mid-log
phase.

Bacterial growth was measured either i) Turbidometri-
cally by using a Klett calorimeter equipped with a standard
red filter, or ii) by enumeration of colony forming units by

plating appr0priate dilutions of bacteria on TY solid media.

Growth of Bacteria in Low Iron Media

Growth of bacteria in low iron media was accomplished
as follows. Outgrowth of bacteria from frozen stocks on TY
slants was done as described above. Cells were washed off
the slant with 5 ml sterile distilled, deionized water and
transferred to a sterile disposable screw cap centrifuge
tube. The cells were pelleted at high speed in a clinical
centrifuge, the supernatant fluid was discarded, and the
pellet was suspended with 5 ml sterile Chelex-treated water
(see above). One-half ml of this cell suspension was used
to inoculate 70 ml of LI medium; the culture was then shaken
at 130 rpm at 28°C on an orbital shaker to deplete cellular
iron pools.

Cells used for iron uptake experiments were harvested
(as subsequently described) after 25 hours of growth, when
they had reached mid-log phase.

To further deplete endogenous iron pools and to carry
out low iron growth experiments, 0.5 ml of low iron culture
was transferred to 70 m1 of fresh LI medium after the first

culture reached mid-log phase. This served as 'time zero'

20

for the growth experiments (Figure 7). Bacterial growth was

measured as described above.

Plant Growth and Nutrition

 

Pea seeds (Pisum sativum L. var. Alaska) were surface

 

disinfected first for 5 min. in ethanol, then for 30 min in
5% (w/v) sodium hypochlorite, and finally for 3 x 10 min in
sterile distilled, deionized water. Surface disinfected
seeds were then germinated on water agar plates (0.25% [w/v]
Sigma agar in distilled water) for 3 days at room
temperature.

Seeds were planted in 10 x 15 cm paper cups containing
equal amounts of perlite and vermiculite. First the cup was
filled with perlite/vermiculite to 1/2 full, then 20 ml of

appropriate Rhizobium strain was added (wash bacteria off

 

slant with 5 ml distilled, deionized water and then add to
500 ml sterile distilled, deionized water). After this
about 10 pea seedlings were placed in the cup and finally,
the cup was filled to about 2/3 full with perlite/
vermiculite.

Plants were grown in a greenhouse at 23°C with natural
lighting. They were irrigated with a nitrogen-free nutrient
solution (40) for two days, followed by tap water every
third day. Nodules matured 28 to 32 days after inoculation.

(See appendix for nitrogen-free nutrient solution recipe.)

21

Measurement of Nitrogen Fixation
Nitrogen fixation was determined by the acetylene
reduction assay (36). Roots from pea plants inoculated with

the apprOpriate E. leguminosarum strain were weighed, placed

 

in 55 ml test tubes, and capped with a serum stopper. Ten
percent of the air in the tube was replaced with an equal
volume of acetylene (generated from calcium carbide).
Ethylene formed was measured on a Varian 3700 gas chromato-
graph with flame ionization detection: column, 0.02 (I.D.)
x 70 cm Porapak R; injector temperature, 100°C, column

temperature, 30°C; FID temperature, 100°C.

Preparation of Cell-Free Extracts

 

Cells were harvested by centrifugation in 250 ml
bottles at 13,000 x g for 20 minutes. The pellet was
resuspended in ice-cold sonication buffer (described below)
and the washed cells were recentrifuged at 17,000 x g for 10
min. The pellet was then resuspended in 5 ml of ice-cold
sonication buffer and the cells were disrupted by sonication
with six 20 sec bursts at power setting 4 on a Branson model
S-125 sonifier. Each burst was followed by a 10 to 15 sec
cooling period. The sonically treated material was centri-
fuged at 27,000 x g and the resulting supernatant fluid was
used for the enzyme assay. For all assays, final protein
concentration of the cell-free extract was adjusted with

appropriate buffer, described below, to a value between 0.8

22

and 1.5 mg per ml. Approximately 25 mg protein was obtained
from 500 ml of a mid-log culture.

The sonication buffer for ALAS assays contained (in
mmol per liter): HEPES buffer (pH 8.0), 100; MgClz, 1;
and 2-mercaptoethanol, 2.5. The sonication buffer for ALAD
assays contained (in mmol per liter): HEPES buffer (pH
7.5), 30; MgSO4, 10; and 2-mercaptoethanol, 2.5. The
sonication buffer for cell-free PBG incubations contained

(in mmol per liter): KZHP04 buffer (pH 7.5), 5.

Cytochrome Difference Spectra

 

Cytochromes were detected in cell-free extracts by
difference spectroscopy (2). Cell-free extracts were ob-
tained as described above by washing and resuspending cells
in 0.1 M potassium phosphate buffer (pH 7.0) and sonicating.
The sonically treated material was centrifuged at 30,000 x g
for 10 minutes and the protein in the supernatant fluid was
adjusted to 3 to 8 mg protein per ml. One 2 ml sample was
reduced with a few grains of sodium dithionite, and another
sample was oxidized wth 50 pl of 3 mM K3Fe(CN)6. A
reduced minus oxidized difference spectrum was obtained on a
Cary 15 recording spectrophotometer equipped with a 0.1
slidewire. Cytochromes were quantitated according to the
method of Appleby (2) by measuring AA between a wavelength
pair corresponding to an absorption maximum and trough for
each cytochrome. Thus, for cytochrome c, AA.mM = 23.2 (550-
536 nm) and for cytochrome b, AA mM = 17.9 (559-580 nm).

23

Enzyme Assays

 

G-aminolevulinic acid synthease (ALAS, EC 2.3.1.37)

ALA production in cell-free extracts was measured as
described (70). The reaction mixture contained (in mmol per
liter): HEPES buffer (pH 8.0), 100; MgClz, 32; Naz-
succinate, 200; glycine, 200; ATP, 14; coenzyme A, 0.1;
pyridoxal phosphate, 0.6. The reaction was started by
adding 0.5 ml of cell-free extract to a test tube containing
0.5 ml of reaction mixture. After incubating 2 hours in a
water bath at 30°C, the reaction was stepped by placing the
tubes on ice and immediately adding 0.2 ml of 33% ice-cold
(w/v) trichloroacetic acid. The tubes were mixed well and
allowed to stand overnight. Precipitated protein was
removed by centrifugation at 800 x g for 10 min. One ml of
the supernatant fluid from each tube was combined with 0.25
ml of 0.75 m Na3P04 and 0.1 ml of ethylacetoacetate.

The contents of the tubes were mixed, capped with marbles
and placed in a boiling water bath for 15 min. After
cooling, 1.35 ml of modified Ehrlich's reagent (see
appendix) was added to each tube and the absorbance of the
resulting color complex was measured in a spectr0photometer
(Gilson model 240) at a wavelength of 553 nm. The molar
extinction coefficient of the G-ALA-ethylacetoacetate-

Ehrlich's complex at 553 nm was assumed to be 6.2 x

1o+4.

24

a—aminolevulinic acid dehydrase (ALAD, EC 4.2.1.24)

 

Porphobilinogen (PBG) production in cell-free extracts
was measured as described (58, 70). The reaction mixture
contained (in mmol per liter): HEPES buffer (pH 7.5), 30;
MgSO4 10; 2-mercaptoethanol, 25; ALA, 5. The reaction was
started by adding 0.5 ml of cell—free extract to a test tube
containing 0.5 ml reaction mixture. After proceeding 2
hours in a water bath at 30°C, the reaction was stopped by
placing the tubes on ice and immediately adding 0.25 ml 20%
(w/v) trichloroacetic acid which was saturated with HgClz.
The tubes were mixed and allowed to stand on ice overnight.
Precipitated protein was removed by centrifugation at 800 x
g for 10 min. One ml of the supernatant fluid from each
tube was combined with one ml modified Ehrlich's reagent.
After 15 min, the absorbance of the resulting color complex
was read in a spectrophotometer set at 555 nm. The molar
extinction coefficient of the PBG-Ehrlich's complex was

assumed to be 6.1 x 10+4.

Production of Pyrroles by Dense Suspensions
of Bacterial Cells

 

 

Porphyrin production was determined by incubating dense
cell suspensions with ALA. Bacterial cells were harvested
as described previously, washed once in MOPS, and re-
suspended in 50 mM MOPS buffer (pH 6.9) to a protein con-
centration of 0.8 to 1.5 mg per ml. The reaction mixture

contained (in mmol per 10 ml final volume): MOPS buffer (pH

25

6.9), 0.5 and Nag-succinate, 0.41. In addition to the
above, the ALA-containing samples also had 50 1 of 0.1 M
G-aminolevulinic acid stock (final concentration = 5 mol
per 10 ml reaction mixture). From the time = 0 samples, one
ml cell suspension was removed for PBG determination,
hematoporphyrin Naz-salt (112 mnols as a 0.5 m1 aliquot in
ethanol) was added as a standard to estimate porphyrin
recovery, and the remainder was immediately frozen for
subsequent analysis of porphyrins (see below). Samples
containing ALA were incubated in 50 ml Erlenmeyer flasks
with shaking in a 30°C water bath. After 20 hours one ml of
cell suspension was removed for PBG determination, 0.5 ml of
the same hematoporphyrin stock was added, the samples were
frozen, and PBG and porphyrin production was determined as

described below.

Production of Tetrapyrroles by Cell-Free Extracts

 

Uro- and Coproporphyrin production was determined in
cell-free extracts by incubating with PBG. Bacterial cells
were harvested and lysed as described previously and the
cell-free supernatant fluid was suspended to 8.0 to 11.5 mg
cell protein per ml with 5 mM potassium phosphate buffer (pH
7.5). The 2.1 ml reaction mixture, made up in 12 x 75 mm
culture tubes, contained 0.7 ml of cell-free extract plus
(in umol): TRIS buffer (pH 7.5) 55; 2-mercaptoethanol,
13.75; and PBG, 0.30. The reaction mixture was prepared

while the culture tubes were in an ice/water bath. Once

26

prepared (and still in the ice bath), the tubes were
vortexed, capped with a serum stopper, and sparged with
nitrogen for 10 minutes. Sparging was accomplished by
inserting two hypodermic needles into the serum stopper
(providing a nitrogen inlet and outlet). At time = 0 hours,
the tubes were transferred to a 30°C water bath and
incubated under nitrogen for 3 hours. At the end of the
incubation, 22.4 nmol of hematoporphyrin Nag-salt, as a

100 pl aliquot in ethanol, was added to each sample and the

tubes were frozen for subsequent analysis of porphyrins.

Measurement of Iron Uptake by Bacterial
Dense Cell Suspensions

 

 

The rate of iron uptake by dense suspensions of whole
cells was determined with 55Fe using the method of
Rosenberg (71). Cells were harvested as described
previously: cultures were first centrifuged at 13,000 x g
for 20 minutes and the pellet was washed twice with ice-cold
phosphate buffer (pH 6.9), with centrifugation at 17,000 x g
for 10 minutes. The phosphate buffer contained (in mmol per
liter): NaHzPOz, 30; KH2PO4, 10; (NH4)2804, 10; MgClz,
1; and CaClz, 0.04. This mixture, at pH 4.5, was auto-
claved and allowed to stand at room temperature for several
days to allow any precipitates to form. The solution was
then passed through a membrane filter (Millipore, 0.45 pm)
and the pH was adjusted to 6.9 with 3:1 (v/v) 5 M NaOH/KOH.

This solution was then diluted with an equal amount of

27

distilled, deionized water before use. For 55Fe uptake
studies, bacteria were suspended to 1.0 to 1.5 mg soluble
protein per ml with the phosphate buffer, above, and ten m1
of cell suspension was placed in a 100 ml polypropylene
beaker and preincubated for 10 minutes in a 30°C shaker
water bath.

Radioactive iron was added to the cell suspension as
55FeCl3 (specific activity = 26.37 Ci per gram) in a
solution of 2 mM sodium nitrilotriacetate (NTA). NTA was
used to prevent non-specific binding of 55Fe+3 to
bacterial cell surfaces. The 55Fe/NTA solution was
prepared in an Eppendorf microfuge tube by adding 15 ul
55FeC13 (5.0 pCi, in 0.1 N HC1) stock to 0.1 m1 of 2
mM NTA. This solution was incubated for 30 minutes at 30°C,
115 pl of 55Fe/NTA solution was added to the cell
suspension and a 200 pl aliquot was immediately removed and
filtered with suction through a membrane filter (Millipore,
0.45 um). The membrane filters were previously soaked
overnight in 40 mM Fe EDTA. The cells trapped on the filter
were immediately washed twice with 2 ml of ice-cold 0.9%
NaCl (w/v). Aliquots were removed from the cell suspension
at subsequent times and filtered as described. The cell
suspension was shaken at 30°C except for the brief periods
during sampling. All samples were taken in triplicate. To
account for non-specific adsorption of 55Fe, the
radioactivity of the t = 0 filter was subtracted from the

subsequent time points.

28

The filters were then placed in scintillation vials, 10
ml of scintillation cocktail (Safety-Solve, Research
Products International) was added, and the samples were
counted in the 3H channel of a Beckman LS 7000
scintillation counter. The program used provided an H#
which measures counting efficiency; pmol 55Fe taken up
was calculated using the counting efficiency and specific

activity.

Analytical Methods

 

PBG Determination

 

One ml of cell suspension was added to a 10 ml culture
tube containing 50 pl of 100% (w/v) trichloracetic acid.
The tubes were mixed and then centrifuged for 10 min at high
speed in a clinical centrifuge. The supernatant fluid was
decanted into another 10 ml test tube and one m1 modified
Ehrlich's reagent was added. Samples were mixed and after
15 min absorbance at 555 nm was determined using a Cary 15
recording spectrophotometer. The molar extinction
coefficient of the PBG-Ehrlich's complex was determined to

be 6.1 x 10+4 (78).

Porphyrin separation and quanititation

 

After removal of the 1 ml aliquot for PBG determin-
ation, the remainder of the cell suspension was quantita-

tively transferred to a 50 m1 round bottom flask. This

29

remainder was then frozen in a dry-ice/acetone bath and
lyOphilized to dryness. The lyophilate was then suspended
in either 5 ml methanol : sulfuric acid (19:1, v/v) for
whole cell incubations or 2 ml MeOH/HZSO4 for cell free
incubations and incubated for 24 hours at -20°C followed by

the porphyrins (see K. Smith, Porphyrins and

 

Meta110porphyrins p. 835-835).

 

For chromatography, the porphyrin methyl esters were
quantitatively transferred into chloroform. First, a few
drops of methanolic iodine (0.005%, w/v) was added to the
methanol : sulfuric acid suspension to oxidize porphyrino-
gens present up to porphyrins. The suspension was centri-
fuged at 27,000 x g for 9 min. The supernatant fluid was
decanted and the pellet was re-extracted with 2 ml
methanol : sulfuric acid. The combined supernates were
neutralized with 3 volumes of 5% (w/v) sodium bicarbonate
and the porphyrin methyl esters were quantitatively trans-
ferred to diethyl ether. The aqueous fraction was extracted
with ether until no red fluorescence (as detected with long
wavelength ultraviolet illumination) could be seen in the
ether fraction. The combined ether layers were twice back
extracted with 1/2 volume 7% (w/v) NaCl. The ether fraction
was taken to dryness with a rotary evaporator. The porphy-
rin methyl esters were immediately taken up in 8 ml chloro-
form and this was quantitatively transferred to a 15 ml

conical centrifuge tube. Before injection onto the HPLC,

30

the volume of the sample was reduced under vacuum and
samples were resuspended in 50 pl of chloroform.

The porphyrin methyl esters were separated and quanti-
tated by high performance liquid chromatography on a Varian
model 5000 HPLC: column, 20 x 0.46 cm Partisil-lO ODS
(Whatman 10 pm) with a 7.5 x 0.21 cm Co:Pell ODS (Whatman)
precolumn; fluorescence detecter, Kratos FS 950 Fluoromat,
Aex = 405 nm, Aem > 595 nm with output to an
Omniscribe linear recorder. Solvent program was as follows,
time = 0 min 60% (v/v) ethanol (40% water), t = 7 min 100%
ethanol; flow rate, 2 ml per min. Porphyins were quanti-
tated by determining the weight of sample peaks and compar-
ing with the weight of peaks resulting from injection of
porphyrin methyl ester standards. The hematoporphyrin
standard added at the end of whole cell and cell-free incu-
bations was used to determine and convert for incomplete
recovery of porphyrins from the samples. Results presented
in the tables are the mean : S.E. of three essays done in

triplicate.

Protein Determination

 

The concentration of protein in cell-free extracts was
determined using the Bradford method (12), using bovine
gamma globulins as a standard. This technique involves the
quantitative binding of protein to the dye Coomassie blue.
0.1 ml of cell-free extract was added to 5 ml of the dye

solution and the absorbance of the resulting color complex

31

was measured in a spectrophotometer at 595 nm. (See

appendix for reagent recipe.)

RESULTS

Nodulation Characteristics

 

Rhizobium leguminosarum mutant strain 116 forms white,

 

ineffective nodules. Host pea plants nodulated with this
strain show the typical nitrogen deficiency symptoms, being
shorter and yellower than plants nodulated with the wild
type or revertant strains (data not shown). When nitrogen-
ase activity is estimated by the acetylene reduction assay,
strains 1062 and 74-11 reduce acetylene while 116 does not
(Figure 2, Table 2). The mutant strain also forms signifi-
cantly more nodules per host plant root than either the wild
type or revertant strains (Table 2). This is true whether
these values are expressed as a per weight or per number
basis. Hypernodulation is characteristic of ineffective

Rhizobium strains.

 

Growth Characteristics

 

In order to characterize the biochemical defect in the
mutant strain, growth curves were determined for the three
strains. Figure 3 shows growth as a function of time for
strains 1062, 116, and 74-11. When grown in defined,

minimal (Y) medium, the mutant strain shows no apparent

32

Figure 2.

33

Acetylene reduction by pea roots nodulated with

various Rhizobium leguminosarum strains.

At time = 0 min, C2H2 was added to root plus
nodule samples in stoppered culture tubes.
C2H2 was added to a final concentration of

10% (v/v). Ethylene was determined by FID gas

chromatography.

Symbols: 0, 1062;A, 116; I, 74-11

nmol 02H4 lgrom root

34

 

 

 

 

so 100

time (min)

 

35

Osman know ummma um

nwuooaom m aouw

.cgmuuw mumwumoummm onu nuw3 noumHDoocH muoou

mo .m.m H cams msu acmmmuaou noncomoum mumn uoou unwam you wasnoz

.ucmawumaxm

numb w>wuwucommuamu mum aofiuunnmu mamamumom pom nmucommnm muadwmm

 

 

 

 

 

 

 

 

: u z. m. H E a H om. 35 :1:

nu no. mum: wumf cod 0:

a H 8 2 H 3 2 H m2 «he 88
uoou ucmHa uoou amuw uoou Osman u; . uoou Emuw Camuum
mmaano: we mmasno: we mmasco: mo uwnan: qu0 ~08:

:oHuosnou
ocoazuoom
mcfimuum abummoafisswma ESwDONALM mo mmwuummoua ofiuoHnamm .N manna

Figure 3.

36

Growth of E. leguminosarum strains in minimal

 

medium.

Colony forming units were determined by plating
appropriate dilutions of broth cultures onto TY

plates and incubating for 3 days at 28°C.

Symbols: 0 , 1062;A, 116;., 74-11

number of cells / ml

10“9

 

I Illlll

0.
lll'llll I

I

0:.

 

 

 

lllllll

IJIIILLI

 

 

L— a

E 3
IO.6 "5

.. J
Io*5 1 1 1 1 1

0 IO 20 3O 4O 50 60

time (hours)

38

additional requirements for growth over that of the wild
type or revertant.

Although the mutant grows as well as the wild type and
revertant, it has a lower cytochrome content (Figure 4).
Cytochromes were determined in cell-free extracts of broth
cells by means of reduced minus oxidized difference spectra.
It is possible to both qualitatively and quantitatively
estimate the presence of cytochromes by this technique
because the different cytochromes show different absorption
maxima in their difference spectra. Figure 4 shows the
dithionite-reduced minus ferricyanide oxidized difference

spectra from E. leguminosarum strains 1062, 116, and 74-11.

 

The absorption maxima for the a- and 8- peaks for cytochrome
c are, respectively, at 555 nm and 530 nm. The maxima for
the a- and 8- peaks for cytochrome b are, respectively, at
563 nm and 530 nm. Although less well defined in this
figure, the absorption maximum for cytochrome oxidase (a-
peak) is around 600 nm. Concentrations of cytochromes c and
b in the three strains are presented in Table 3. Figure 4
and Table 3 show that the mutant has no detectable
cytochrome c and less cytochrome b than either the wild type
or revertant. The limit of sensitivity for this assay is 10
pmol cytochrome per mg protein. These data suggest that 116
is defective in heme biosynthesis; however, the mutant
strain apparently makes enough cytochromes to grow at wild

type rates in Y medium (see Figure 3).

39

Figure 4. Difference spectra (dithionite reduced minus
ferricyanide oxidized) of cell-free extracts of
Rhizobium leguminosarum strains 1062, 116, and

74-11.

Log-phase cells were harbested, washed in 0.1 M
K-phosphate buffer (pH 7.0), lysed by sonication,
and centrifuged at 27,000 x g for 10 min.

a) Strain 1062, protein concentration 8.0 mg ml‘l
b) Strain 116, protein concentration 4.9 mg ml"1
c) Strain 74-11, protein concentration 6.8 mg ml'1

d) baseline

40

 

 

 

Absorbonce

 

 

 

560 550 660 650
Wavelength (nm)

41

Table 3. Concentrations of cytochromes in Rhizobium

 

leguminosarum strains

 

 

pmol gytochrome i S.E.

 

mg protein

 

Strain cyt c cyt b
1062 59 i 13 76 i 25
116 _ a) 31 i 3
74-11 44 + 11 58 i 19

 

a) trace, below limit of detection (10 pmol/mg protein)

Results presented are the mean : S.E. of three independent

determinations.

42

High Performance Liquid Chromatography of Porphyrins

Mutant strain 116 accumulated porphyrins (precursors to
heme). In order to identify the lesion in this strain,
whole cell suspensions and cell-free extracts were incubated
in the presence of ALA and intermediate of heme biosynthesis
were estimated; presumably one would see an accumulatioin of
porphyrins at the step in the pathway where the lesion
occurs. In order to quantitate the porphyrins formed during
feeding experiments, it was necessary to deveIOp an HPLC
method for separating porphyrins.

The HPLC method which was developed involved using a
linear gradient of ethanol and water; from zero to seven
minutes the gradient ran from 60% to 100% ethanol. The
stationary phase was a C-18 (reversed phase) column which
gave very repeatable results.

Representative HPLC tracings of porphyrin methyl ester
standards and porphyrin methyl esters as extracted from the

three E. leguminosarum strains are shown in Figures 5 and 6.

 

Figure 5 shows the tracings for porphyrin methyl esters as
isolated from log phase cells of the three strains. Figure
6 shows the tracings resulting from incubating dense cell
suspensions (which were harvested during log phase) with
ALA. It can be seen that incubating cell suspensions with
ALA results in a large increase in the porphyrins

recovered.

 

Figure 5.

43

HPLC tracings of porphyrin methyl esters as

isolated from log-phase cells of Rhizohbium

 

leguminosarum strains 1062, 116, and 74-11.

 

Porphyrins were isolated as their methyl esters
by extracting washed, lyophilyzed cells with

methanol-sulphuric acid.

For all tracings, the fluorescence detector range

was set at 0.2 (most sensitive scale).

a) Strain 1062
b) Strain 116
c) Strain 74-11

d) Porphyrin methyl ester standards:

F - solvent front

1 - UROoctamethyl ester

2 - COPROtetramethyl ester
3 - PROTOdimethyl ester

44

 

 

 

 

 

 

3:33.62“. 952mm

IO

6
Time (min)

4

Figure 6.

45

HPLC tracings of porphyrin methyl esters as

isolated from log-phase cells of Rhizohbium

 

leguminosarum strains 1062, 116, and 74-11

 

incubated in G-aminolevulinic acid.

Cells suspensions were incubated aerobically with
0.5 mM ALA for 20 hours at 30°C and were then
frozen. Porphyrins were isolated as their

methyl esters by extracting lyOphilized cells

with methanol-sulphuric acid.

The fluorescence detector range was set at 1.0.
In tracings a,b, and c the range was changed to

0.2 at the point indicated by the arrow.

a) Strain 1062

b) Strain 116

c) Strain 74-11

d) Porphyrin methyl ester standards:

F - solvent front

1

UROoctamethyl ester

2

COPROtetramethyl ester

U.)
I

PROTOdimethyl ester

46

 

I

I

I

 

L. l.

 

 

 

A

 

 

 

 

 

8:33.03“. Sigma

6 8 IO

Time (min)

4

47

Qpantitation of Selected Heme Biosynthesis Activities

 

The results of incubating dense suspensions of cells
with ALA are shown in Table 4. It can be seen that there is
no significant difference in the rate of either PBG or
porphyrin formation between the wild type, mutant, or
revertant. These data suggest that 116 is not impaired in
protoporphrin biosynthesis.

The activities of several enzymes of the heme biosyn-
thesis pathway in cell-free extracts of the three bacterial
strains are presented in Table 5. The mutant and wild type
strains show no signficant difference in activities of the
first two heme synthesis enzymes: ALAS and ALAD. Cell-free
extracts were incubated with PBG under a N2 atmosphere and
porphyrins formed were determined by HPLC. As can be seen
in Table 5, the three strains also show no difference in
uro- or coproporphyrin forming ability. Under these experi-
mental conditions, only traces (less than 1 pmol PROTO per
mg protein per hour) of protoporphyrin were detected.

Because the mutant strain showed no obvious accumula-
tion of heme synthesis intermediates, the results shown in
Tables 4 and 5 do not indicate that the mutant is defective

in porphyrin biosynthesis.

Experiments on Cells Grown in Low Iron

 

Since strain 116 shows no apparent defects in porphyrin
and biosynthesis, and indeed would convert ALA to PROTO

(Table 4), it was decided to look at iron metabolism. If

48

.numvcmum HmcumOCw

swumnmuomoumEmn m mo zum>oomu unmoumm mnu no women emuumuuoo mumB newumnauoa mo mum>ouom

.oumoHHaHuu CH econ mummmm manna mo .m.m H came onu mum newcommum muflamom

 

 

 

 

So H e to H m I H 8 mod H 35 :1:
«we wfli 2H3 taufié o:
a H m a H 2 e H S 85 H 35 88
080mm ommoo om: samuum
I “so: . Camuoua we l_ueoml. :wmuoua we
.m.m + awumnumom Hoeml .m.m + vwm Hoe:

 

.<A< ze m.o sues noumnsocw mHHmo

 

enummocweswoa abanONwsm mo mconcmamSm cw sewumHsesvom Gauznauom new 0mm mo oumm .q manna

.cumvcmum Hmcuoucw

cHuhcmuoaoqumn m mo wum>oomu unwouma wsu co comma vmuomuuoo wum3 wcHumnauoa mo hum>oomm

.wuwowamwuu CH meow mzwmmm wounu mo .m.m H cmme one mum nmucwmoum muHDmmm

cocHEHoumn uo: Am

 

49

 

 

 

 

 

 

 

o.o.H A _.o.H A As Am _.-en
s H o w H s. as._ H mm.m Nw.o H mm.l op.
o H m a H m_ mm.~ H mm.m mm.o H.nm.P Hoop
. . lune; . :Hmuoum we . . I use: . chuoum we . . 1 H505 . :Hmuoumlwe
w m + :Humnmuom Hoem. m w + mum Hoec m m + <A< Hoe:
ommoo om: Hue>euua a<q< eue>euum m<e< :Hmoum
enummocHemme

.M mo muomnuxo omumnaamo CH meoSucmmoHn oEmn mo moEkuco mo wwHuH>Huo< .m mHQMH

50

the mutant strain was defective in iron metabolism, this
would result in porphyrin accumulation and reduced
cytochrome levels because reduced iron is required to form
heme. Similarly, a rhizobial iron metabolism mutant would
result in a non-nitrogen fixing phenotype because
nitrogenase in a root nodule is an iron protein and is
inactive in the absence of leghemoglobin. To test the
hypothesis that strain 116 is defective in iron metabolism,

growth curves for the three E. leguminosarum strains grown

 

in low iron medium were determined.

The normal defined (Y) medium which was used in the
preceeding experiments contains 122 pM iron. This iron
concentration is far greater than that which is needed for
optimal bacterial growth. Y medium was modified to L1
medium to more rigourously reduce the available iron. The
iron concentration in L1 medium was determined to be 0.059
UM by atomic absorption spectroscopy.

The growth of the three E. leguminosarum strains in L1

 

medium is shown in Figure 7a. It can be seen that while the
wild type and revertant show normal growth, the mutant grows
poorly if at all. As a control, the three strains were
grown in the same medium containing 122 uM iron (Figure 7b).
In high iron medium, the mutant grows as well as the wild
type and revertant. From this it was concluded that mutant
strain 116 is defective in iron metabolism.

It should be noted that prior to the growth experiments

depicted in Figure 7, the cells were grown to mid-log phase

Figure 7.

51

Growth of E. leguminosarum strains in low iron

 

and high iron media.

Colony forming units were determined by plating
appropriate dilutions of broth cultures onto TY

plates and incubating for 3 days at 28°C.

A) growth in low iron medium [Fe] 0.059 “M

B) growh in high iron medium [Fe] 122 UM

Symbols: 0,., 1062;A,A, 116;D,l, 74-11

52

 

IIIIIII

 
 

 

IIIIIIIT I

[IIIIII I

I

IIIIIII I

I

30 4O 50 60

20

IO

 

 

 

[111/ sues go aequmu

 

 

 

II7IIII I I

IIIIITr I

IIIIII I

d
4
1111111 1 1 1111111 1 111411

I

so 4o so so

"me (hours)

IO 20

 

O

 

KY“

'3'

79
nit/sues To uqumu

0
O

O
l 5

time (hour!)

53

in L1 medium to use up any endogenous iron pools. At the
beginning of the growth experiment depicted in Figure 7, the
cells were transferred to fresh LI medium or fresh high iron
medium. These LI growth experiments suggest that strain 116
has used up all of its internal iron pools by the time of
transfer to fresh medium because upon the transfer to L1,
this strain shows no further growth. For this reason, when
assaying iron uptake by cells of strain 116 (described
below), the mutant cell cultures grown in L1 were harvested
at the first mid-log phase.

To further define the nature of the iron metabolism
defect in strain 116, the uptake of radioactive iron by the
three strains was determined. Iron uptake was measured by
incubating dense suspensions of whole cells in a solution
containing 55FeCl3. Working with inorganic iron in
biological systems presents two problems. First, at
physiological pH, Fe+3 will form Fe(OH)3, which is
extremely insoluble (KSp is less than 10'38) (10).

Second, because a bacterial cell wall has a net negative
charge, Fe+3 will presumably adhere by electrostatic
interactions to the cell wall. For these reasons, when
measuring 55Fe uptake, the radioactive iron was first
chelated to nitrilotriacetate and it was this chelate which
was presented to the bacterial cell suspensions.

A time course for the uptake of 55Fe by the wild
type, mutant, and revertant grown in L1 and high iron medium

is shown in Figure 8. The initial and final 55Fe uptake

Figure 8.

54

55Fe uptake of Rhizobium leguminosarum

 

strains grown in high iron or low iron minimal

media.

Cells were harvested, washed, and resuspended in
a buffered solution to which 5 uCi 55Fe/NTA

was added. Samples were removed at various times
and cells were collected on membrane filters and
washed with 9% NaCl. An initial time point was
taken and this value was subtracted from
subsequent time points to account for
non-specific adsorption. The data presented
represent the mean of three independent
experiments where the samples were taken in
triplicate.

Open symbols: cells grown in low iron medium

Closed symbols: cells grown in high iron medium

Symbols: 0,0, 1062;A ,A, 116;D,', 74-11

55

 

 

O
7

//.../ /

.././. ..,./
L . p ./.n.u/..P/-//l

6 5 4 3 2 l
23.93 =3 oExEES 9.2a: sumo

 

0

20

.5

10
time (min)

0

56

rates are given in Table 6. For cells grown in high iron
medium, there is no significant difference in either the
initial or final 55Fe uptake rates between the wild

type, mutant, and revertant. However, in low iron-grown
cells, the mutant initially has a lower rate of uptake than
the wild type and revertant; but the final rate of uptake of
the three strains approach equal values. These results
suggest that mutant strain 116 is defective in the initial

uptake of iron.

57

.mumoHHmHuu CH :mxmu wuw3

moaaemm onu wumna mucoewummxo ucovcoamvcw mousu mo .m.m H came oSu mum noucommua muaamom

 

 

 

No.o_H on.o P_.o H NH._ s_.o.H mo._ m_.o H No.e .P-sn
oo.o H no.0 o~.o H.s_.o ms.o H mo.~ Hm.o H mm.“ QPF
mo.o H.mm.o mm.o H_mn._ Nw.o H No.~ No._ H oa.m moo,
macaw Hansen“ Hanan Haauacu
couH 30H couH an:

 

. . l awe . chuoum we
w m + mmmm Hoem

 

 

.mwwme couH 30H cam couH nwws CH eBOHw

 

mchuum enummOCHeewwH eeHnoNHSm How oxmum: mmmn mo moumu Hmewm van HmHuHcH .o waan

DISCUSSION

To test the hypothesis that rhizobia produce the heme
of leghemoglobin, Nadler has isolated a mutant of E.

leguminosarum which was presumed to be defective in heme

 

synthesis (57). The characterization of this mutant strain
(116) presented in this investigation was undertaken to
determine the biochemical basis of this mutation and thereby
to gain possible insight into the nature of the

Rhizobium-legume symbiosis. The results presented suggest

 

that 116 is defective in the initial uptake of iron.

Strain 116 does not grow in low iron medium (Figure 7)
and is defective in the initial uptake of iron (Figure 8 and
Table 6). Under "normal" 55Fe uptake conditions, i.e.
the kinetics found in strains 1062 and 74-11 grown in low
iron, there is first a rapid initial phase of uptake follow-
ed by a slower steady state rate (Figure 8). In contrast to
this, 116 grown in low iron shows a slow initial rate of
55Fe uptake and a more rapid final rate; the final rate
of uptake in 116 is more or less equivalent to the steady
state rate found in 1062 or 74-11. The biphasic kinetics of
55Fe uptake in the three bacterial strains suggest the
presence of an 55Fe-saturable space in free-living
Rhizobium leguminosaum; this iron-saturable space may be the

58

 

59

periplasmic space. Thus 116 may be unable to concentrate
iron in its periplasmic space, as shown by the initial slow
rate of uptake. However, once iron is in the periplasm,
there is no problem for 116 to transport it into the
cytosol; this is seen in the faster steady state rate of
iron uptake in Figure 8.

The results of the iron uptake experiments suggest two
possibilities as to the exact nature of the mutation in 116:
i) 116 is not able to synthesize a functional siderophore or
ii) 116 is unable to synthesize a functional siderophore re-
ceptor protein. There are relatively simple assays which
test for the presence of phenolate/catechol (5) and hydroxa-
mate (21) siderophores. A positive test for siderophores in
116 would imply that the mutant is not defective in sidero-
phore production and would also imply that the defect in 116
is in the siderophore receptor. Under iron stress, certain
outer membrane proteins are overproduced by various bacteria
(14, 62) as revealed by SDS-PAGE; these have been suggested
to serve as siderophore receptor proteins. The presence or
absence of these proteins in strain 116 should be determined
to see if this strain is somehow defective in their
production.

Another conclusion which can be drawn from this in-
vestigation is that rhizobia produce the heme of leghemo-
globin. Strain 116 induces white, non-nitrogen fixing root
nodules on host peas. Lack of nitrogenase activity may be

due to a lack of leghemoglobin in the mutant-induced

6O

nodules. When measured by a radioimmunoassay, the mutant
nodules contain about 3% of the leghemoglobin found in the
wild type (11). On the other hand, the wild type and re-
vertant strains induce pink, nitrogen-fixing nodules (Table
2). One difference between the mutant strain and the wild
type and revertant strains is that the mutant is unable to
synthesize enough heme (due to its impaired iron uptake) to
meet the nodule's demand for leghemoglobin. If the plant
genome was responsible for producing the heme of leghemo-
globin, then the nodules on that plant induced by a mutant

Rhizobium which is impaired in heme biosynthesis would

 

result in pink nitrogen fixing nodules. Strain 116, along
with E. meliloti mutants A-34 and A-36 of Leong, gg‘gl.
(50), provides the first genetic evidence that rhizobia
produce the heme of leghemoglobin.

If 116 is indeed an iron uptake mutant, then this also
implies that iron is not freely available to bacteroids in a
nodule; if iron was available, then nodules of strain 116
would not show its non-nitrogen fixing phenotype. Presum-
ably, there are other metabolic consequences occuring in 116
nodules because of lack of iron to the bacteroids. Perhaps
116 bacteroids have a lower cytochrome content than 1062 or
74-11 bacteroids. If this is the case, it would be useful
to determine the respiratory activity of these bacteroids to
see if there is any difference in the three strains.

Another consequence of impaired iron uptake is that 116

bacteroids are probably accumulating porphyrins. It is

61

tempting to speculate that if the iron concentration in 116
nodules was increased, then the mutant might be able to
overcome the heme synthesis defect and allow for sufficient
leghemoglobin synthesis that these nodules would then become
effective nitrogen fixers.

Strain 116 makes less cytochromes than strains 1062 and
74-11 (see Figure 3 and Table 3), but in minimal medium the
mutant grows as well as the wild type and revertant (Figure
2). These discoveries resulted in the conclusion that
strain 116 is a "leaky" mutant: it is not completely
blocked in heme biosynthesis and under the proper condi-
tions, the mutant is able to produce enough heme for
"normal" free-living growth. This point will be further
discussed below.

Incubating cell-free extracts or whole cell suspensions
respectively with PBG or ALA resulted in no significant
differences in the rate of porphyrin formation between the
mutant and wild type or revertant. Since 116 is defective
in heme synthesis, it would be expected to find more heme
intermediates in this strain than in 1062 or 74-11, but this
was not the case. This can possibly be explained in two
ways. First, the true manifestation of the pOp' mutation
is seen in root nodules, where the bacteria have changed
into bacteroids. Under nodule conditions, where the bac-
teroid metabolism differs from the metabolism of lab-
cultured bacteria, the pep' mutation would be more fully

expressed due to the higher demand for heme. Second, under

62

free-living growth, pOp' is not fully expressed because of
the relatively high iron concentrations present in Y medium
when growing the cells up for an assay.’ This second point
will be further discussed below.

Similarly, under heme stress, one would expect to see
an increase in ALAS activities because this is the rate

limiting enzyme of heme biosyntehsis in Rhizobium; heme is

 

known to regulate this enzyme in other organisms (16, 54).
116 does not show increased ALAS activities over that of
1062 are probably the same as why the mutant does not form
more porphyrins than the wild type and revertant: under the
assay conditions employed, pep’ was not fully expressed.

As shown in Figure 6, strain 116 clearly appears to be
a mutant defective in some aspect of iron metabolism. The
results of this experiment also explain why the afore-
mentioned results in some of the previous experiments were
unexpected. Under the high iron conditions present in Y
medium, 116 does not express pop’; this expression does
not occur until the mutant is placed under iron stress.
Under high iron conditions, the medium is rate-saturating
with respect to iron and any defect in the cell's ability to
utilize iron is overcome by diffusion of iron into the
mutant cells. It is possible that if strain 116 were grown
in low iron medium and the various heme biosynthetic
activities were assayed, the phenotype would be that of a

heme synthesis mutant. The mutant would have higher ALAS

63

and porphyrin-forming activities than either strains 1062 or
74-11.

In order to analyze the porphyrins formed by whole cell
suspensions and cell-free extracts, I develoPed my own
method of poprphyrin separation by high performance liquid
chromatography since published methods were not
reproducible. I wish to leave these recommendations for
anyone attempting porphyrin HPLC in the future: i) of the
two methods of detection, fluorescence is a better method
than absorbance at 400 nm. It is more sensitive than
absorbance detection and, by using a red cutoff emission
filter, the only peaks seen on the recorder tracing are due
to porphyrins. ii) Porphyrins should be run in columns as
methyl esters rather than free acids. Although it involves
added time to esterify porphyrin samples, I feel that this
extra time is worth the effort for two reasons: it is
easier to extract porphyrin methyl esters from biological
samples than porphyrin free acids. Free acids are insoluble
and precipitate out of solution at pH approaching
neutrality, and thus any chromatographic system which
separates porphyrin free acids must run in a solvent system
buffered below pH 4.0 in order to fully protonate the
porphyrin samples. In the long run, I feel that running a
solvent system like this can be harmful to both the HPLC
solvent pump and to the column.

In summary, this investigation has shown that mutant

strain 116 is defective in the initial uptake of iron.

64

Since this defect is not seen in high iron-grown cells, then
this implies that 116 has high and low affinity iron meta-
bolism systems. This study also gave further support to the
hypothesis that the heme of leghemoglobin is of bacteroid
origin. Further experiments which need to be done are (i)

examine the heme biosynthetic activities of E. leguminosarum

 

strains grown in low iron (Tables 4 and 5), (ii) further

define the iron uptake requirements for the three Rhizobium

 

strains, e.g. look for the presence of siderophores and the
presence of outer-membrane proteins which act as siderophore
receptors, and (iii) continue these experiments on bac-
teroids. Studies such as these should provide more insight
into the role of iron uptake and its relationship to the

Rhizobium-legume symbiosis.

 

APPENDIX

APPENDIX

Y minimal medium (Johnston, EE gl., 1978)

 

for growth of E. leguminosarum strains 1062, 116, and

 

 

74-11
final concentration
g1;
MgSO4-7H20 0.1
CaC12-6H20 0.22
KZHP04 0.22
FeCl3 (stock solution in 0.1 N HCl) 0.02
Nag-glutamate 1.1
D-biotin 5.6 mg/l
thiamine 5.6 mg/l
DL-pantothenic acid (Ca salt) 5.6 mg/l
Nag-succinate 1.35
uracil 2.25 mg/l
L-tryptophan 11.25 mg/l
TY complete medium (Johnston, gg‘gl., 1978)
gLE
Difco Bacto-Tryptone 5
Difco Bacto-Yeast Extract 3
CaClz-2H20 1.5

65

66

Minimal and complete solid media were made as described
above plus the addition of Difco Bacto-Agar to a

concentration of 1.5% (w/v).

Nitrogen-Free Medium (Johnson, gE 31., 1966)

 

for pea growth

final concentration

311
MgSO4-7H20 0.49
K2504 0.02
KH2PO4 0.02
K2HPO4 0.01
CaSO4-2H20 1.03
CaC12-2H20 1.8
trace elements BELL
FeEDTA 1
H3BO3 0.25
MnSO4-H20 0.25
ZnSO4-7H20 0.05
CuSO4'5H20 0.02
NazMoO4-2H20 0.01

CoC12-6H20 0.05

67

Modified Ehrlich's Reagent (Urata and Granick, 1963)

 

Glacial acetic acid 43 ml
70% perchloric acid 10 ml
p-dimethylaminobenzaldehyde 1.0 g
HgClz 0.175 g

Bradford Reagent (Bradford, 1976)

 

Coomassie Brillant Blue G 50 mg
95% ethanol 25 ml
85% phosphoric acid 50 ml
Distilled, deionized water 425 ml

Dissolve brillant blue in ethanol, add phosphoric acid, and
dilute with water. Filter through one layer of Whatman

Number 5 filter paper.

BIBLIOGRAPHY

10.

BIBLIOGRAPHY

Appleby, C.A. 1962. The oxygen equilibrium of
leghemoglobin. Biochim. BiOphys. Acta 62:226-235.

Appleby, C.A. 1969. Electron transport systems if
Rhizobium japonicum. I. Haemoprotein P-450, other
CO-reacfive pigment, cytochromes and oxidases in
bacterioids from Nz-fixing root nodules. Biochim.
BiOphys. Acta 172:71-87.

Appleby, C.A. 1974. Leghemoglobin. p. 521-524 In
The Biology of Nitrogen Fixation, A. Quispel (ed.).
North Holland Press. Amsterdam.

Appleby, C.A., W.E. Blumberg, J. Peisach, B.A.
Wittenberg, and J.A. Wittenberg. 1976. Leghemoglobin:
an electronic paramagnetic resonance and optical

spectral study of the free protein and its complexes

with nicotinate and acetate. J. Biol. Chem.
251:6090-6096.

Arnow, L.E. 1937. Colorimetric determination of the
components of 3,4-dihydroxyphenylalanine-tyrosine
mixtures. J. Biol. Chem. 118:531-537.

Avissar, Y.J. and K.D. Nadler. 1978. Activation of
tetrapyrrole formation in Rhizobium 'a onicum by
restricted aeration. J. Bacteriol. : -789.

 

Baulcombe, D. and D.P.S. Verma. 1978. Preparation of
a complementary DNA for leghemoglobin and direct
demonstration that leghaemoglobin is encoded by the
soybean genome. Nuc. Acids Res. 5:4141-4153.

Beringer, J.E. 1974. R factor transfer in Rhizobium
leguminosarum. J. Gen. Microbiol. 84:188-198.

 

 

Beringer, J.E., N. Brewin, A.W.R. Johnston, H.M.
Schulman, and D.A. Hopwood. 1979. The Rhizobium-
legume symbiosis. Proc. R. Soc. Lon. 204:219-233.

Biedermann, G. and P. Schindler. 1957. On the

solubility product of precipitated iron(III) hydroxide.
Acta Chem. Scand. 11:731-740.

68

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

69

Bisseling, T., C. Been, J. Klugkist, A. van Kammen and
K. Nadler. 1983. Nodule specific host protein in
effective and ineffective root nodules of Pisum
sativum. EMBO J. 2:961-966.

Bradford, M.M. 1976. A rapid and sensitive method for
the quantitation of microgram quantities of protein
utilizing the principle of protein-dye binding. Anal.
Biochem. 72:248-254.

Brady, N.C. 1974. The Nature and Pr0perties of
Soils,, 8th Ed. MacMillan, New York. p. 23.

 

Braun, V., Hancock, R.E.W., Hantke, K., and Hartman, A.
1976. Functional organization of the outer membrane of
Escherichia coli: Phage and colicin receptors as

 

components of iron uptake systems. J. Supramol. Struc.
5:37-58.

Brown, J.C. 1979. In Iron, pp. 57-78. Baltimore:
Univ. Park Press. (Comm. Med. Biol. Effects Environ.
Pollut., Subcomm. Iron, Natl. Res. Conc.).

Burnham, B.F. and J. Lascelles. 1963. Control of
porphyrin biosynthesis through a negative-feedback
mechanism. Studies with preparations of 6-aminolevuli-
nate synthetase and 5-aminolevulinate dehydratase from
Rhodopseudomonasm spheroides. Biochem. J. 87:462-472.

 

Burnham, B.F. and J.B. Neilands. 1961. Studies on the
metabolic function of the ferrichrome compounds. J.
Biol. Chem. 236:554-559.

Burris, R.H. and W.H. Orme-Johnson. 1974. Nitrogenase
and its EPR properties. In Microbial Iron Metabolism,
J.B. Neilands, (ed.). Academic Press, New York.

 

C00per, S.R., J.V. McArdle, and K.N. Raymond. 1978.
SiderOphore electrochemistry: relation to
intracellular iron release mechanism. Proc. Natl.

Acad. Sci. USA 75:3551-3554.

Cox, R. and H.P. Charles. 1973. Porphyrin-accumu-
lating mutants of Escherichia coli. J. Bacteriol.
113:122-132.

 

Cséky, T.Z. 1948. On the estimation of bound
hydroxylamine in biological materials. Acta Chem.
Scand. 2:450-454.

Cutting, J.A. and H.M. Schulman. 1969. The site of
heme synthesis in soybean root nodules. Biochim.
Biophys. Acta 192:486-493.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

70

Cutting, J.A. and H.M. Schulman. 1971. The biogenesis
of leghemoglobin. The determinant in the Rhizobium-
legume symbiosis far leghemoglobin specificity.
Biochim. Biaphys. Acta 229:58-62.

 

Decad, G.M. and H. Nikaido. 1976. Outer membrane of
gram negative bacteria. XII. Molecular-sieving
function of cell wall. J. Bacterial. 128:325-336.

Dilwarth, M.J. 1969. The plant as the genetic
determinant of leghaemoglobin production in the legume
root nodule. Biochim. Biophys. Acta 184:432-441.

Doss, M. and W.K. Philipp-Darmstan. 1971. The
influence of iron on the biosyntehsis of porphyrins and
the prosthetic group of cytochrome b in Achromobacter
megaicaligenes. Happe-Seyler's Z. Physial. Chem.

35 : 3-51.

Ellfalk, N. 1972. Leghaemaglabin, a plant
haemoglobin. Endeavour 31:139-142.

 

 

Falk, J.E., C.A. Appleby, and R.J. Parra. 1959.
Nature, function, and biosynthesis of the haem
compounds and porphyrins of legume root nodules. Symp.
Soc. Exp. Biol. XIII:73-86.

Frost, G.E. and H. Rosenberg. 1973. The inducible
citrate-dependent iron transport system in Escherichia
coli 1912. Biochim. Biophys. Acta 330:90-10T.

 

Frost, G.E. and H. Rosenberg. 1975. Relationship
between the tonB locus and iron transport in
Escherichia coli. J. Bacterial. 124:704-712.

 

Gaines, C.G., J.S. Lodge, J.E.L. Arceneaux, and B.R.
Byers. 1981. Ferrisideraphare reductase activity
associated with an aromatic biosynthetic anzyme complex
in Bacillus subtilis. J. Bacterial. 148:527-533.

 

Geargapapadakau, N.H., J. Petrilla, and A.I. Scott.
1976. Tetrapyrrale biosynthesis: N-methyl-N'-nitra-
saquanidine-induced mutants of Prppian:bacterium
shermanii. Genet. Res. Camb. 28}93-100.

 

 

Godfrey, C.A. and MAJ. Dilwarth. 1971. Haem
biosynthesis from 1 C 6-aminalevulinic acid in

laboratory grown and root nodule Rhizobium lupini. J.
Gen. Microbial. 69:385-390.

 

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

71

Granick, S. and S. Sassa. 1971. ALA synthetase and
the control of heme and chlorophyll synthesis. In
Metabolic Regulation V, H. Vagel (ed.). Academic
Press, New York.

 

Haddock, B.A. and H.U. Schairer. 1973. Electron-
transpart chains of Escherichia coli. Reconstitution of

respiration in a S-aminalevulinic acid-requiring
mutant. Eur. J. Biochem. 35:34-45.

 

Hardy, R.W.F., R.D. Holsten, E.K. Jackson, and R.C.
Burns. 1968. The acetylene-ethylene assay for

Nz-fixatian: laboratory and field evaluation. Plant
Physiol. 43:1185-1207.

Imai, K., A. Asana, and R. Sata. 1968. Oxidative

phaspharylatian in Micrococcus denitrificans. J.
BiOChemo 63:219-225.

 

Ito, T. and J.B. Neilands. 1958. Products of
"law-iron" fermentation with Bacillus subtilis:
isolation, characterization, and synthesis of
2,3—dihydraxybenzaylglycine. J. Am. Chem. Soc.
80:4645-4647.

 

Jensan, V.J. and E. Thafern. 1953. Chlarhamin
(ferriparphyrinchlarid) als Bakterienwuchstaff. I. Z.
Naturfarsch. 86:599-603.

Johnson, G.V., H.J. Evans, and T. Ching. 1966.
Enzymes of the glyaxylate cycle in rhizobia and nodules
of legumes. Plant Physiol. 41:1330-1336.

Johnston, A.W.B., J.L. Beynan, A.V. Buchanan-Wallastan,
S.M. Setchell, P.R. Hirshc, J.E. Beringer. 1978. High
frequency transfer of nadulating ability between
strains and species of Rhizobium. Nature 276:634-636.

 

Kauppinen, V. 1968. Iron deficiency and glucose
metabolism of Candida guilliermundii. Saum.
Kemistilehti B41:47:50:

 

Kliewer, M. and J.J. Evans. 1963. Cabamide coenzyme
contents of soybean nodules and Nz-fixing bacteria in
relation to physiological conditions. Plant Physiol.
28:99-1040

Knight, E. and R.W.F. Hardy. 1966. Isolation and
characteristics of flavadaxin from nitrogen-fixing
Clastridium pasteurianum. J. Biol. Chem.

- S60

 

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

72

Lankfard, C.E. 1973. Bacterial assimilation of iron.
CRC Critical Rev. Microbial. 2:273-330.

Lascelles, J. 1960. Synthesis of the enzymes
concerned with bactertiachlaraphyll formation in
growing cultures of Rhodopseudomonas ppheraides. J.
Gen. Microbial. 23:487-498.

 

Lascelles, J. 1964. Tetrapyrral biosyntehsis and its
regulation. W.A. Benjamin Ca., New York.

Lascelles, J. and T.P. Hatch. 1969. Bacteria-
chlaraphyll and heme synthesis in Rhodopseudomonas
ppheraides: possible role of heme in regulation of the
branched biosynthetic pathway. J. Bacterial.
98:712-720.

 

 

Lascelles, J., B. Rittenberg, and G.D. Clark-Walker.
1969. Growth and cytochrome synthesis in a hemin-
requiring mutant of Spirillum itersanii. J. Bacterial.

97:455-456.

 

Leong, S.A., G.S. Ditta, and D.R. Helinski. 1982.

Hema biosynthesis in Rhizobium. Identification of a
cloned gene coding for 5-aminalevulinic acid synthetase
from Rhizobium meliloti. J. Biol. Chem.

257:8727-8730.

Lodge, J.S., C.G. Gains, J.E.L. Arceneaux, and B.R.
Byers. 1982. Ferrisideraphare reductase activity in

Agrabacterium tumefaciens. J. Bacteriol.
9: - 40

McConville, M.C. and H.P. Charles. 1979. Mutants of
Escherichia coli K 12 accumulating porphobilinogen: a
new locus, hemC. J. Gen. Microbial. 111:193-200.

 

 

 

McConville, M.L. and H.P. Charles. 1979. Isolation of
haemin-requiring mutants of Escherichia coli K 12. J.
Gen. Microbial. 113:115-164.

Meyers, R.B. and K.D. Nadler. 1978. Regulation of
heme biosyntehsis in Rhizobium japonicum. Abstract,
Proc. IIIrd Int'l. Symposium Nitrogen Fixation, Univ.
Wisconsin, Madison, WI.

 

 

Middleton, K.S. and H.B. Gunner. 1968. Porphyrin
production by Arthrobacter glabifarmis. Can. J.
MICIObIOl. IAIIUgS‘Ilogo

 

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

73

Miyake, S. and T. Sugimura. 1968. Capraparphyrina-
genase in a respiration-deficient mutant of yeast
lacking all cytochromes and accumulating caprapar-
phyrin. J. Bacterial. 96:1997-2003.

Nadler, K.D. 1981. In A.H. Gibson and W.E. Newton
(eds.). Proceedings of the 4th International Symposium
on Nitrogen Fixation, Australian Academy of Sciences,
Canberra, p. 143.

Nadler, K.D. and Y.J. Avissar. 1977. Heme
biosynthesis in soybean raat nodules. I. On the role
of bacterial 5-aminalevulinic acid synthase and
6-aminalevulinic acid dehydrase in the synthesis of the
heme of leghemoglobin. Plant Physiol. 60:433-436.

Neilands, J.B. 1977. Siderophores: biochemical
ecology and mechanism of iron transport in enterobac-
teria. In Biainarganic Chemistry II (Adv. in Chem.
162). ACS Press, Washington D.C.

 

Neilands, J.B. 1981. Iran absorption and transport in
microorganisms. Ann. Rev. Nutr. 1:27-46.

Neilands, J.B. 1981. Microbial iron compounds. Ann.
Rev. Biochem. 50:715-731.

Neilands, J.B. 1982. Microbial envelope proteins
related to iron. Ann. Rev. Microbial. 36:285-309.

Neuberger, A., J.D. Sandy, and G.H. Tait. 1973.
Control of 5-aminalevulinate synthetase activity in
Rhodopseudomonas spheroides. The involvement of
sulphur metabolism. Biochem. J. 136:477-490.

 

Neuberger, A., S.D. Sandy, and G.H. Tait. 1973.
Control of 5-aminolevulinate synthase activity in
Rhadapseudamanas spheroides. The purification and
properties of an endogenous activator of the enzyme.

Biochem. J. 136:491-499.

 

Nicola, N.A., E. Minasian, C.A. Appleby, and S.J.
Leach. 1975. Circular dichroism studies of myoglobin
and leghemoglobin. Biochemistry 14:5141-5149.

Noel, K.D., G. Stacey, S.R. Tandon, L.E. Silver, and
W.J. Brill. 1982. Rhizobium japonicum mutants

defective in symbiotic nitrogen fixation. J.
Bacterial. 152:485-494.

 

O'Brien, I.G., G.B. Cox, and F. Gibson, 1971. Enter-
achelin hydrolysis and iron metabolism in Escherichia
coli. Biochim. Biophys. Acta 237:537-549.

 

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

74

Orme-Jahnsan, N.R., W.H. Orme-Jahnsan, R.E. Hansen, H.
Beinert, and Y. Hatefi. 1971. EPR detectable electron
acceptors in submitachandrial particles from beef heart
with special reference nta the iron sulfur components
of DPNH-ubiquinane reductase. Biochem. Biophys. Res.
Commun. 44:446-452.

Raymond, K.M. and C.J. Carrana. 1979. Coordination
chemistry and microbial iron transport. Acc. Chem.
Res. 12:183-190.

Roessler, P.G., and K.D. Nadler. 1982. Effects of
iron deficiency on heme biosynthesis in Rhizobium
japonicum. J. Bacterial. 149:1021-1026.

 

 

Rosenberg, H. 1979. Transport of iron into bacterial
cells. In Methods in Enzymalagy LVI, 388-394. S.
Fleischer and L. Packer, editors. Academic Press, New
York.

 

Rosenberg, H. and I.G. Young. 1974. Iran transport in
the enteric bacteria. In Microbial Iran Metabolism.
J.B. Neilands (ed.). Academic Press, New York.

 

Sasarman, A., M. Surdeanu, G. Szégli, T. Haradniceanu,
V. Greceanu, and A. Dumitrescu. 1968. Hemin-deficient
mutants of Escherichia coli K 12 J. Bacterial.
96:570-572.

 

Sassa, S. and A. Kappas. 1977. Induction of
5-aminalevulinate synthase and porphyrins in cultural

liver cells maintained in chemically defined medium.
J. Biol. Chem. 252:2428-2436.

Spiro, T.G. and P. Saltman. 1969. Palynuclear
complexes of iron and their biological implications.
Struc. Banding 6:116-156.

Tait, G.H. 1973. Aminalevulinate synthetase of
Micrococcus denitrificanis. Biochem. J. 131:389-403.

 

Tawnsley, P.M. and J.B. Neilands. 1957. The iron and
porphyrin metabolism of Micrococcus lysadeikticus. J.

 

 

Urata, G. and S. Granick. 1963. Biosynthesis of
G-aminaketanes and the metabolism of aminaacetane. J.
Biol. Chem. 238:811-820.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

75

Vainshtein, B.K., E.H. Harutyunyan, I.P. Kuranav, V.V.
Barisav, N.J. Sasfenav, A.G. Pavlovsky, A.J. Grebenha,
and N.V. Kanareva. 1975. Structure of leghemoglobin
from lupin root nodules at 5A resolution. Nature
254:163-164.

Verma, D.P.S. and A.K. Bal. 1976. Intracellular site
of synthesis and localization of leghemoglobin in root
nodules. Proc. Natl. Acad. Sci. USA 73:3843-3847.

Verma, D.P.S., D.T. Nash, and H.M. Schulman. 1974.
Isolation and 12 vitra translation of soybean
leghaemoglobin mRNA. Nature 251:74-77.

Virtanen, A.I., J. Erkama, and H. Linkala. 1947. On
the relation between nitrogen fixation and
leghemoglobin content of leguminaus root nodules. Acta
Chem. Scand. 1:861-870.

Wang, C.C. and A. Newton. 1969. Iran transport in
Escherichia coli: roles of energy-dependent uptake and
2,3-dihydraxybenzaylserine. J. Bacterial.
98:1142-1150.

 

Waring, W.S. and C.H. Werkman. 1942. Growth of
bacteria in an iron-free medium. Arch. Biochem.
1:303-310.

Waring, W.S. and C.H. Werkman. 1944. Iron deficiency
in bacterial metabolism. Arch. Biochem. 4:75-87.

Webley, D.M., R.B. Duff, and G. Anderson. 1962. The
metabolism of iron-, zinc-, and manganese-deficient
Nocardia apaca. J. Gen. Microbial. 29:179-187.

 

Wittenberg, J.B., F.J. Bergersan, C.A. Appleby, and
G.L. Turner. 1974. Facilitated oxygen diffusion. The
role of leghemoglobin in nitrogen fixation by
bacteriods isolated from soybean root nodules. J.
Biol. Chem. E32, 4057-4066.

Wulff, D.L. 1967. G-Aminalevulinic acid-requiring
mutant from Escherichia coli. J. Bacteriol.
93:1473-1474.