LIBRARY
Michigan State
University

 

 

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINE return on or before date due.
MAY BE RECALLED with earlier due date if requested.

DQTE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

we m.“

IDENTIFICATION AND CHARACTERIZATION OF COLD SHOCK
LOCI IN SINORHIZOBI UM MELILOT I 1021

By

Ann M. Gustafson

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Microbiology

1999

 

11.

ab

 

IUJ
ho

at“:

El

Se

ABSTRACT

IDENTIFICATION AND CHARACTERIZATION OF COLD SHOCK
LOCI IN SINORHIZOBIUM MELILOT I 1021

By

Ann M. Gustafson

In order to survive and persist in the soil, Sinorhizobium meliloti must be
able to adapt to a wide range of temperatures. To investigate S. meliloti's response
to low temperature, transposon mutagenesis was performed on S. meliloti strain
1021 using the reporter transposon, Tn5-1062. Tn5-1062 carries the promoterless
luxAB genes of Vibrio harveyi which encode bacterial luciferase. Luciferase
activity can be easily assayed by exposing the cell culture to aldehyde and
measuring the luminescence of the sample. Luminescence of the transposon
recipients was assayed at 30°C and following a temperature downshift to 15°C.
Seven transposon insertions were isolated that exhibited increased luminescence at
15°C. Five of these insertions were in either a 168 or 238 rm gene. Further
investigation revealed that luxAB insertions in all three of the rrn operons in S.
meliloti 1021 showed increased luminescence following a temperature downshifi.
To characterize their apparent temperature regulation, the rrn promoter regions
were sequenced and the transcription start sites mapped. Each 168 rRNA gene is
preceded by a 354 bp leader region that is conserved between the three operons. A

single promoter having significant similarity to the rm P1 promoters of

10;

deg

lb

r.
315

do;

Tn.

0R

 

(“1'1

and

gen

 

Escherichia coli was located upstream of each rrn operon. The rrnA promoter
was fused to the luxAB genes at +1 and +172 and recombined into the inositol
locus of S. meliloti 1021. Study of the two promoter-luxAB fusions revealed that
the 5’ end of the rRNA transcript was required for the maximum temperature-
dependent increase in luminescence. Increased luminescence in the +172 fusion
resulted from the combined effects of increased translation, increased message
stability, and increased promoter activity at 15°C. A rrnP-luxAB fusion to +187
also showed significantly increased luminescence following a temperature
downshift from 30°C to 15°C. To investigate the factors influencing this increase,
transposon mutagenesis of S. meliloti 1021 was performed using the transposon,
Tn5-233. Several transposon insertions were isolated that increased the
luminescence of the rrnP—luxAB fusion and the original 168 rmzzTn5-1062
insertion mutants at 30°C. Two of these Tn5-233 insertions were located in an
ORF whose deduced product showed significant similarity to Ribonuclease E of E.
coli. Ribonuclease E is an essential endoribonuclease involved in the processing
and degradation of mRNA and rRNA. In S. meliloti 1021, the expression of this

gene increased transiently following a temperature downshift from 30°C to 15°C.

m"

 

"t X.._;.-.

ACKNOWLEDGMENTS

First, I want to thank my advisor, Mike Thomashow, for his encouragement
and guidance during my graduate study and in the completion of this dissertation. I
would also like to thank the members of my guidance committee, Frans de Bruijn,
Lee Kroos, Rich Lenski and Tom Schmidt, for taking the time to read this
dissertation and for their helpfirl comments over the years. Several people worked
with me on different aspects of this project. Deane Lehmann was the first to
isolate cold-induced transposon insertion mutants and helped me get started, and
keep going, as a graduate student. Kevin O’Connell contributed greatly to this
project through his own research efforts and sage advice. Patty Spagnuolo and
Beth Seymour provided technical assistance in screening for “down” mutants and
in mutant analysis, respectively. Peter Wolk and his laboratory kindly provided
use of their photonic system and the reporter transposon, Tn5-1062. I would also
like to thank Sylvia Rossbach and Anne Milcamps for their helpful advice. A
special thanks goes to all the members (past and present) of Mike Thomashow’s
and Rebecca Grumet’s labs. It has really been a privilege. Finally, I would like to
thank the Center for Microbial Ecology and the Department of Microbiology for

funding this work and providing this opportunity for graduate study.

iv

TABLE OF CONTENTS

 

 

 

LIST OF TABLES ................................................................... vii

LIST OF FIGURES .................................................................. viii

CHAPTER 1

Literature Review ..................................................................... 1
Introduction .................................................................... 1
Cold shock proteins in Escherichia coli ................................... 3
Cold shock proteins in other bacteria ...................................... 9
Regulation of the cold shock response in Escherichia coli ............. 15
The cold shock response in Sinorhizobium meliloti ..................... 24
Literature Cited ............................................................... 25

CHAPTER 2

Identification of low temperature induced genes in Sinorhizobium

meliloti 1021 ........................................................................... 31
Summary ....................................................................... 31
Introduction ................................................................... 33
Results ......................................................................... 38
Discussion ..................................................................... 62
Materials and Methods ...................................................... 69
Literature Cited ............................................................... 76

CHAPTER 3

Characterization of the Sinorhizobium meliloti 1021 rrn promoters ........... 81
Summary ....................................................................... 81
Introduction .................................................................... 83
Results .......................................................................... 89
Discussion ...................................................................... 120
Materials and Methods ....................................................... 129
Literature Cited ................................................................ 137

91d

ll
.ill

\L

"1

 

 

CHAPTER 4
A Sinorhizobium meliloti 1021 gene similar to the Escherichia

 

 

 

coli me gene is induced following a temperature downshift ................... 140
Summary ....................................................................... 140
Introduction .................................................................... 142
Results .......................................................................... 146
Discussion ..................................................................... 169
Materials and Methods ....................................................... 174
Literature Cited ................................................................ 183

APPENDD( A

Auxotrophic mutants cause increased luminescence of rrn-luxAB

fusions at 30°C ......................................................................... 188
Introduction .................................................................... 188
Results and Discussion ....................................................... 188
Materials and Methods ....................................................... 202
Literature Cited ................................................................ 209

APPENDD( B

Preliminary characterization of Tn5-233 insertions causing decreased

luminescence of rrn-luxAB fusion at 15°C ......................................... 210
Introduction .................................................................... 210
Results and Discussion ....................................................... 210
Materials and Methods ....................................................... 218
Literature Cited ................................................................ 223

APPENDD( C

Nucleotide sequence of the promoter and upstream regions of the three

rm operons in Sinorhizobium meliloti 1021 ....................................... 224

 

 

 

 

LIST OF TABLES

CHAPTER 2
Table 2.1 Identification of transposon insertion sites

in Tn5-1062 mutants .................................................. 49
Table 2.2 Bacterial strains ....................................................... 60
Table 2.3 Plasmids ................................................................ 61
CHAPTER 3
Table 3.1 Bacterial strains ....................................................... 116
Table 3.2 Plasmids ................................................................ 117
Table 3.3 Oligonucleotides ...................................................... 119
CHAPTER 4
Table 4.1 Bacterial strains ....................................................... 167
Table 4.2 Plasmids ............................................................... 168
APPENDICES
Table A.1 List of auxotrophic Tn5-233 insertional mutants ................. 189

vii

 

 

 

LIST OF FIGURES

CHAPTER 1
Figure 1.1 Regulatory elements controlling the expression of the

cold shock protein, CspA ........................................... 20
CHAPTER 2
Figure 2.1 Structure of pRL1062a and Tn5-1062 inserted in genome. . . .. 39
Figure 2.2 Screen for Tn5-1062 inserts induced by a temperature

downshift .............................................................. 40
Figure 2.3 Set of seven Tn5-1062 inserts induced by a temperature

downshift .............................................................. 41
Figure 2.4 Sequence alignment of the RM3166 transposon insertion.

site and the 16S rm of Rhizobium meliloti LMG 613 ........... 45
Figure 2.5 Sequence alignment of the RM73 transposon insertion site

and the 238 rrn of A grobacterium vitis ........................... 47
Figure 2.6 Mapping the transposon inserts to the rrn operons .............. 51
Figure 2.7 Effect of a temperature downshift on the luminescence

of the luxAB transcriptional fusion, pAGl.3 ...................... 53
Figure 2.8 Construction of pAGl.3 and pLUX ................................ 55
Figure 2.9 Levels of luxAB transcript following a temperature

downshift from 30°C to 15°C ....................................... 59
CHAPTER 3
Figure 3.1 Sequence comparison of the three Sinorhizobium

meliloti 1021 rrn promoter regions ................................ 90
Figure 3.2 Primer extension assay of the rrn transcripts using

MT179 ................................................................. 93
Figure 3.3 RNase protection assay of the rrn promoter regions ............ 94
Figure 3.4 Mapping the 5' ends of the rm transcripts by primer

extension using MT261 .............................................. 98
Figure 3.5 Construction of rrn promoter-luxAB fusions ..................... 99
Figure 3.6 Luminescence of rrn promoter-luxAB firsion strains ............ 102
Figure 3.7 Induction of luminescence and luxAB mRNA in rm

promoter-luxAB fusions after a temperature downshift ......... 106
Figure 3.8 Comparison of luxAB mRNA levels ............................... 110
Figure 3.9 Induction of the rrn-uidA fusions following a

temperature downshift ............................................... 112
Figure 3.10 The promoter region of the S. meliloti 1021 rrnA operon ...... 115

viii

 

1"3

u - n l

 

 

 

 

 

 

 

CHAPTER 4

 

. —-u—'

 

 

Figure 4] Structures of Tn5-233 and reporter plasmid, pAGl.3 ........... 147
Figure 4.2 The Tn5-233 inserts in RM26b1 and RM483 cause

increased luminescence at 30°C when transduced

into RM603 ........................................................... 149
Figure 4.3 Luminescence of RM603(26b1) and RM603(483)

at 30°C and 15°C ..................................................... 150
Figure 4.4 DNA sequence analysis of the insertion sites in RM483

and RM26b1 .......................................................... 151
Figure 4.5 Southern blot analysis of RM26bl and RM483 ................. 154
Figure 4.6 Alignment of the E. coli and S. meliloti RNase E

protein sequences .................................................... 158
Figure 4.7 Comparison of domains in E. coli and S. meliloti

RN ase E proteins ..................................................... 160
Figure 4.8 The rne transcript accumulates transiently following a

temperature downshift from 30°C to 15°C ....................... 162
Figure 4.9 Effect of me inserts on lux transcript levels at 30°C

when transduced into RM603 ....................................... 164
Figure 4.10 Effect of me inserts on cspA levels before and after a

temperature downshift ............................................... 166
APPENDD( A
Figure A] Arginine and pyrimidine biosynthesis in Escherichia coli. . . .. 190
Figure A2 The Tn5-233 insert in RM75 is located in a gene similar

to the pyrB gene of Synechocystis PCC6803 ..................... 193
Figure A.3 Set of Tn5-233 inserts that caused auxotrophy and

increased luminescence when transduced into RM603 ......... 195
Figure A.4 Luminescence of auxotrophic double mutants increases

with culture density .................................................. 197
Figure A.5 Effect of added uracil on luminescence in RM603(75) ......... 199
Figure A.6 Effect of pyrB::Tn5-233 insert on luxAB transcript

levels at 30°C when transduced into RM603 ..................... 201
APPENDIX E
Figure B.1 Set of Tn5-233 inserts that decreased luminescence at

15°C when transduced into RM603 ................................ 212
Figure B.2 Effect of RM42 insert on luminescence of 16S rrn::Tn5-

1062 in RM603 ....................................................... 214
Figure 8.3 Effect of Tn5-233 inserts in RMQ9 and RMT77 on

luminescence of 16S rrn::Tn5-1062 in RM603 .................. 216

ix

CHAPTER 1

Literature Review

Introduction

Temperature is an important physical factor that has a great impact on
bacterial growth and survival. Although bacteria have been isolated from
extremely hot and extremely cold environments, individual bacterial species are
limited to a narrower range (30-40°C) of grth temperatures (Russell, 1990).
Organisms adapted to relatively high temperatures are classed as thermophiles,
while those capable of growth at or near 0°C are defined as psychrotrophs or
psychrophiles. Psychrotrophs have optimum growth temperatures above 15°C
while psychrophiles have maximum growth temperatures of less than 20°C.
Organisms like Escherichia coli, Bacillus subtilis, and Sinorhizobium meliloti are
described as mesophiles showing optimum growth at 30 to 37°C.

At low temperature, several cellular structures and processes can be
negatively affected. Enzymatic reactions slow down and may become limiting as
temperature decreases. Protein conformation can also be affected by low
temperature resulting in the inactivation or improper regulation of enzymes.

Membrane fluidity decreases with temperature, potentially affecting the activity of

 

 

   

mi

 

any membrane-bound enzyme. Specifically, a negative effect of low temperature
on substrate uptake has been demonstrated (N edwell and Rutter, 1994). The
conformation of nucleic acids can also be negatively affected by temperature
through the potential stabilization of unfavorable secondary structures at low
temperatures. Changes in nucleic acid conformation could impact many cellular
functions including: transcription, translation, DNA replication, ribosome
synthesis and regulation of gene expression.

In considering adaptation to low temperature or the response to a “cold
shock”, it is important to note that “cold” is a relative term. To a psychrophilic
bacterium, 8°C may be an optimum growth temperature. Yet, to the mesophilic
bacterium E. coli, 8°C constitutes an extreme low temperature stress. Low
temperature stress may then be considered to be exposure to temperatures below
those to which the organism is adapted. It is interesting to note that a cold shock
response has been observed in both psychrophilic and mesophilic bacteria albeit at
different temperatures. While comparative studies of psychrophilic and
mesophilic bacteria can be informative in examining different adaptations to low
temperature, they are necessarily limited as any observed differences may result
fiom species variation as opposed to being adaptative changes. Studying the
response of a single organism to different temperatures allows a more
straightforward comparison. This type of study is particularly relevant to soil

microorganisms like Sinorhizobium meliloti that must adapt to seasonal, diurnal

 

 

PK

 

Co

and spatial temperature variation in order to persist in the environment.

This review will focus on the cold shock response in Escherichia coli as
most of the proteins that are induced in response to low temperature have been
identified in this bacterium. Characterizing the functions of these proteins and
their importance to low temperature growth will aid in the understanding of low
temperature stress and how it affects cellular physiology. Preliminary studies of
the cold shock response have also been performed in very diverse bacteria. The
cold shock proteins identified in these studies will be compared to those
characterized in E. coli. Comparison of the different responses will be helpfirl in
determining the degree to which the cold shock response is conserved among
difl‘erent bacteria. Finally, the regulation of the cold shock response will be
discussed. The cellular mechanisms for sensing and responding to low
temperature stress have been examined most thoroughly in E. coli. The ribosome
has been proposed to be the sensor of low temperature stress and its adaptation to
low temperature to be the main objective of the cold shock response. This model
will be discussed along with the regulation of CspA, the major E. coli cold shock

protein, and its potential role as a regulator in the cold shock response.

Cold shock proteins in Escherichia coli

The cold shock response in E. coli is characterized by the induction of at

least 16 cold shock proteins, the repression of heat shock protein synthesis and the

 

 

 

If.

CC]
\8
MI

sh.

PP:

hefl

 

continued expression of ribosomal proteins despite a transient inhibition of overall
protein synthesis. Three cold shock proteins, ribosome binding factor A (RbfA),
cold-shock DEAD-box protein A (CsdA) and translation initiation factor 2 (IF 2),
are associated directly with the ribosome. RbfA is a 15 kDa protein with a p1 of
6.05 that is associated with the free 30S ribosomal subunit and is thought to be
involved late in ribosome maturation or translation initiation (Jones and Inouye,
1996). Following a cold shock from 37°C to 15°C, wild type cells show a
transient change in their ribosome sedimentation profile indicating a temporary
block in the initiation of translation. Immediately after the cold shock, there is an
increase in 708 ribosomes and the 50S and 303 ribosomal subunits and a
corresponding decrease in polysomes. A similar change in the ribosome
sedimentation profile occured in a rbfA knockout mutant indicating that RbfA was
required for efficient translation initiation. Significantly, the cold shock response
was found to be constitutively induced in the rbfj4 knockout mutant following a
cold shock from 42°C to 15°C. Several cold shock proteins, including NusA, H-
NS, CsdA, CspA and CspB, were expressed at levels 2-5 times higher than in the
wild type strain. Conversely, overexpression of RbfA on a multicopy plasmid
shortened the growth lag normally observed following a cold shock by 50%.

A second cold shock protein associated with the ribosome is CsdA, a 70kDa
protein with a pI of ~9 (Jones et al., 1996). CsdA was predicted to be an RNA

helicase because of its similarity to other DEAD-box proteins. Jones et a1. (1996)

have confirmed that CsdA does have helix-destabilizing activity and propose that
this activity may be important in removing secondary structures in mRNAs at low
temperature, since CsdA is almost exclusively associated with the 708 ribosomes
at 15°C. While a csdA knockout mutant shows no difference in growth from its
parent strain at 37°C, it has an increased generation time of 16 hours at 15°C
compared to the 8 hour generation time of the wild type strain. When the cold
shock response was examined in the csdA mutant, no difference in protein
expression was observed until 3 hours following the temperature shift from 37°C
to 15°C. The expression of the heat shock proteins, DnaK and GroEL, remained
low in the csdA mutant although they are normally derepressed within three hours.
This effect was specific to low temperature as DnaK and GroEL were expressed
normally following a typical heat shock from 28°C to 42°C.

CspA, CspB, CspG and Cspl are the most highly expressed cold shock
proteins (Goldstein et al., 1990; Lee et al., 1994; Nakashima et al., 1996; Wang
et al., 1999). These four proteins are members of the E. coli CspA family that
shares significant similarity with the cold shock domain of the eukaryotic Y-box
protein family (W istow, 1990). The nine proteins in this family range in size fiom
69 to 74 amino acids, have pl’s of 5.5 to 10.7, and are from 46 to 91% identical at
the amino acid level (Yamanaka et al., 1998). The structure of CspA has been
determined and found to sonsist of five B-strands forming a B-barrel (Newkirk er

al., 1994; Schindelin, etal., 1994). Each member of the CspA family contains

 

C3.

)5:

dc

pr

 

two RNA binding sites, RNPl and RNP2, that are perfectly conserved among the
cold-induced members (Yamanaka et al., 1998). Because CspA was the first cold-
induced member identified, most functional studies have focused on this protein.
CspA is strongly induced following a cold shock and constitutes more than 10% of
protein synthesis within an hour of the temperature shift (Goldstein et al. , 1990).
Since CspA is so quickly induced, it was initially investigated as a transcriptional
activator. In fact, when CspA was overexpressed on a multicopy plasmid, the
expression of several cold shock proteins was increased following a temperature
downshift (Jones eta1., 1992b). In vitro experiments with the hns promoter region
showed that while CspA enhanced activity of this promoter, the purified protein
bound only weakly to a DNA fragment containing the promoter in gel shift and
DNA footprinting experiments (La Teana er al., 1991; Brandi etal., 1994).

CspA, however, appeared to enhance the binding of RNA polymerase to the
promoter region. The authors suggested that CspA might be stabilizing the RNA
polymerase open complex by binding the DNA strands separated by the
polymerase (Brandi et al., 1994). CspA has also been ascribed a role in translation
as a possible RNA chaperone (J iang et al., 1997). In contrast with earlier results,
Jiang et al. (1997) showed that CspA was capable of binding only ssDNA and
ssRNA and not dsDNA. Additionally, CspA bound cooperatively to RNA and
destabilized intermolecular and intrarnolecular structures. Jones et al. (1996) have

proposed that secondary structures formed in mRN As at low temperature could

inhibit their translation. Additionally, it was suggested that CsdA and CspA could
aid in translation through the unwinding of secondary structures by CsdA followed
by the stabilization of the single-stranded conformation by CspA. Further, the
moderate affinity of CspA to RNA would allow the ribosome to easily displace
bound CspA proteins as it translated the message.

Two additional cold shock proteins, trigger factor and Hsc66, are thought to
function as molecular chaperones. The identification of molecular chaperones as
cold shock proteins is quite interesting since molecular chaperones are usually
associated with heat stress. Trigger factor is thought to promote folding of nascent
polypeptides as it is associated with the 508 ribosomal subunit and has been
crosslinked to nascent peptides (Kandror and Goldberg, 1997). Trigger factor has
also been shown to have peptidyl prolyl isomerase activity. Interestingly, this step
in the folding of proteins is often rate-limiting and may be particularly sensitive to
low temperature. Although trigger factor has been shown to associate with the
heat shock protein GroEL and to enhance its activity, these two proteins have
opposite responses to temperature. In fact, overexpression of trigger factor at high
temperature actually decreases cellular viability. Following incubation for one
week at 4°C, a non-permissive grth temperature for E. coli, overexpression of
trigger factor increased survival from 15% to 40% while the reduction of trigger
factor expression led to a decrease in survival to only 1%.

The second molecular chaperone, Hsc66, has 42% similarity to the heat

shock protein DnaK and can prevent aggregation of proteins in vitro (Silberg et a1. ,
1998). Hsc66 is encoded by the gene hscA which is cotranscribed with hscB, a
gene encoding a co-chaperone, Hsc20. Although the transcript containing both
genes is induced at low temperature, only the Hsc66 protein levels have been
examined (Lelivelt and Kawula, 1995). An insertion mutant of hscA was not
greatly altered in growth following a cold shock although the expression of 5
proteins was altered at low temperature. The characterization of trigger factor and
Hsc66 as cold shock proteins suggests that multiple steps in protein synthesis are
negatively affected by low temperature in E. coli.

The remaining cold shock proteins in E. coli are involved in cellular
functions not directly related to protein synthesis (Jones el al., 1987). Two
enzymes of the pyruvate dehydrogenase complex are cold-induced,
dihydrolipoamide acetyltransferase and pyruvate dehydrogenase (lipoamide). The
cold shock protein, NusA, associates with RNA core polymerase and is involved
in transcriptional temrination, antiterrnination and pausing. Polynucleotide
phosphorylase is a 3’ to 5’ exonuclease that interacts with Ribonulease E to
degrade mRNA. RecA regulates the SOS response for DNA repair and is a key
enzyme in homologous recombination. DNA gyrase introduces negative
supercoils into the chromosome. Finally, H-NS shows increased expression
during the cold shock response (La Teana c1al., 1991). H-NS is an abundant

DNA-binding protein that negatively affects the expression of several unrelated

 

 

 

It

 

it:

 

genes including the rrn operons (reviewed in Atlung and Irrgrner, 1997). Binding
of H-NS to the rrn promoter region specifically counteracts activation by FIS
protein (Afflerbach el al., 1998). In an hns insertional mutant expressing no

detectable H-NS protein, growth was significantly inhibited at 12°C and 25°C but

only slightly affected at 37°C (Dersch et al., 1994).

Most of the cold shock proteins in E. coli are not expressed exclusively at
low temperature but are also important at optimum growth temperatures.
Presumably, the function of these proteins increases in importance at lower
temperatures. One example is the cold-sensitive grth of an H-NS null mutant.
As described above, the cold shock proteins are involved in a variety of celluluar
functions indicating that low temperature probably affects many aspects of cell
function. Protein synthesis appears to be particularly affected as 9 of the 16 cold
shock proteins are involved in either translation, RNA conformation, or protein
folding. The remaining cold shock proteins are involved in transcription, gene

regulation, metabolism, RNA degradation and DNA conformation and repair.

Cold shock proteins in other bacteria

The cold shock response has been investigated in many different bacteria.
Most bacteria examined induced the synthesis of several proteins following a
temperature downshift. CspA, in particular, appears to be highly conserved as it

has been detected in hypertherrnophiles, thermophiles, mesophiles and

psychrotrophs. In several species, multigene families have been identified that
contain from 3 to 9 CspA homologs. In each case, at least one of these proteins is
increased in expression at low temperature (Graumann et al., 1996; Mayr et al.,
1996; Michel et al., 1997; Wouters et al., 1998). In the psychrotrophic bacteria,
Pseudomonasfragi and Bacillus cereus, four and five Csp homologs have been
identified, respectively (Michel et al., 1997; Mayr et al., 1996). While all four
proteins are induced by low temperature in P. fragi, the expression of two proteins
termed cold acclimation proteins (Caps), remained high while the other two
proteins were transiently induced. Surprisingly, the transiently induced proteins
were also induced by heat shock and are,therefore, referred to as temperature
adaptation proteins (Taps). In B. cereus, two of the five CspA-like proteins were
constitutively expressed at a higher level at 7°C as compared to 30°C. Having
multiple copies of CspA-like proteins is not necessarily a characteristic of
psychrotrophic bacteria as only one CspA homolog has been identified in the
psychrotroph Arthrobacter globiformis (Berger et al. , 1997). This protein, CapA,
is expressed constitutively at a higher level at 4°C as compared to 25°C.
Multigene families of CspA homologs were also found in mesophilic strains.
Lactococcus lactis MG1363 contains five CspA homologs that are differentially
regulated in response to low temperature (Wouters el al., 1998). The L. lactis
CspE protein is constitutively expressed while the remaining CspA homologs are

induced following a shift from 30°C to 10°C. All four proteins responded

10

 

it

(”I

L' t,

H 9:1

transiently to the cold shock although CspB and CspD were induced at least 30-
fold while CspA and CspC were induced less than 10-fold. Interestingly, CspA
and CspC have a pI of ~9 which is unusually high for the cold-induced CspA
homologs. In B. subtilis, three CspA homologs have been characterized
(Graumann, et al., 1996). While all three are induced following a cold shock,
CspB and CspC are induced during stationary phase as well (Graumann and
Marahiel, 1999). Unlike most bacterial strains, no CspA homologs were found in
the cyanobacteriaAnabaena variabilis M3 (Sato, 1995). Interestingly, four of the
five RNA-binding proteins identified in A. variabilis were induced by a
temperature downshift from 38°C to 22°C. While these RN A-binding proteins are
not homologous to CspA, they may serve a similar function at low temperature.
Although the proteins induced by the cold shock response have been
enumerated using 2D-PAGE, most of the individual proteins have not been
identified. Sixteen of the 37 proteins observed to increase in Bacillus subtilis
following a temperature shift from 37°C to 15°C have been identified (Graumann,
et al., 1996). Several of these proteins were similar to cold shock proteins in E.
coli. As described above, three homologs of the E. coli cold shock protein, CspA,
were identified in addition to the ribosomal proteins S6 and L7/L12. While
individual ribosomal proteins have not been identified as being cold-induced in E.
coli, the continued synthesis or induction of groups of ribosomal proteins has been

noted (Jones et al., 1987; Jones et al., 1996). The ribosomal protein S21 has been

11

characterized as being cold-inducible in both Anabaena variabilis (Sato, 1994) and
in Sinorhizobium meliloti 1021 (O’Connell and Thomashow, 1999). Cyclophilin,
a peptidylprolyl isomerase encoded by ppiB, has been identified as a cold shock
protein in B. subtilis. The E. coli cold shock protein, trigger factor, also has
peptidylprolyl isomerase activity. B. subtilis also induces several enzymes
involved in amino acid synthesis and glycolysis: CysK, Iva, Gap and TIM.

An additional E. coli cold shock protein, polynucleotide phosphorylase,
while not identified as a cold shock protein in B. subtilis, was characterized as
such in Yersinia enterocolitica. In fact, synthesis of polynucleotide phosphorylase
(PNP) was required for growth at 5°C in this psychrotroph (Goverde er al., 1998).
Strains carrying transposon insertions in this gene were unable to form colonies at
5°C yet were indistinguishable from the wild type strain at 30°C. While cold
shock experiments were not performed on this strain, amounts of PNP were
compared in cultures grown constantly at 30°C or 5°C. Levels of PNP protein and
pnp mRNA were 1.6-fold higher at the lower temperature.

Two RNA helicases were identified in Anabaena sp. strain PCC 7120 that
showed increased expression following a temperature downshift of 10°C (Charnot
et al., 1999). Both CrhC and Cth belong to the DEAD-box family of RNA
helicases as does the E. coli cold shock gene, csdA. While cth expression was
induced under several conditions beside cold stress, crhC was induced only at low

temperature. Although CrhC is a DEAD-box RNA helicase like CsdA, it shows

12

C__r .

{W

El

 

t5,

 

more similarity to another E. coli RNA helicase, RhlE.

Although no fatty acid desaturase has been identified as a cold shock
protein in E. coli, cold induced desaturases have been identified in B. subtilis and
at least two cyanobacteria. In B. subtilis, a des-lacZ fusion was induced 10 to 15-
fold following a temperature shift from 37°C to 20°C (Aguilar etal., 1998).
Levels of unsaturated fatty acids increased at 20°C but were not required for
viability as a des null mutant showed growth similar to that of the parent strain at
both 37°C and 20°C. Of the four desaturases identified in the cyanobacterium
Synechocystsis sp. PCC 6803, three are clearly induced following a temperature
downshifi albeit at different rates (Los el al., 1997). In Synechococcus sp. strain
PCC 7002, three desaturases, DesA, DesB, and DesC, have been identified
(Sakamoto and Bryant, 1997). The expression of all three genes was induced
within five minutes of a shift from 38°C to 22°C. The three genes were also
expressed more highly in cells grown continuously at 22°C. Insertion mutants of
desA and desB were unable to grow at 15°C although the strains showed different
effects at higher temperatures (Sakamoto et al., 1998). While both mutant strains
were able to grow normally at 38°C, the desA mutant showed a decreased growth
rate at 22°C. Interestingly, the mutants were not affected in photosynthesis but
were limited in their ability to take up nutrients from the growth medium.

Unlike E. coli, B. subtilis induced several proteins in response to both heat

and cold shocks. Of the six TIPs or temperature-inducible proteins, only CheY

l3

 

Cl

 

was identified. CheY is involved in regulating rotation of the flagella. As
mentioned earlier, a set of four temperature-inducible proteins were identified in
the psychrotroph Pseudomonasfragi as well (Michel et al. , 1997). Since these
proteins were induced following both cold shock and heat shock they were
designated temperature adaptation proteins (Taps). Surprisingly, two of these
proteins, TapA and TapB were identified as homologs of the E. coli major cold
shock protein, CspA. While not all CspA homologs are induced by cold shock,
none had been identified as heat shock proteins previously.

Relatively few cold shock proteins besides the CspA homologs have been
identified in bacteria other than E. coli. The widespread occurrence of proteins
similar to CspA does suggest, however, that the cold shock response is a well
conserved response. Comparison of the best characterized species, E. coli and B.
subtilis, demonstrates that parallels do exist in the types of proteins induced.
Further similarities were found in diverse strains such as the cyanobacteria
Anabaena and the psychrotroph, Y. enterocolitica. Characterizing the specific
functions of the CspA proteins of E. coli and the RNA-binding proteins of
cyanobacteria may reveal interesting parallels as well. Differences between the
cold shock responses were also found. Unlike E. coli, both B. subtilis and P. fragi
synthesized a set of proteins that responded to both heat shock and cold shock.
Considering the opposing nature of the two stresses, characterizing the function of

these general temperature proteins should prove very interesting.

14

Regulation of the cold shock response in Escherichia coli

As stated earlier, the cold shock response in E. coli is characterized by the
overall inhibition of protein synthesis with the exception of the increased synthesis
of the cold shock proteins. The cold shock proteins in E. coli can be divided into
two classes (Thieringer et al., 1998). Class 1 includes those proteins induced
greater than lO-fold following a temperature downshift while class 11 contains
those proteins that are induced from 2 to 5-fold. The cold-induced members of the
CspA family and CsdA are class I proteins and are induced within an hour
following a temperature downshift (Wang et al., 1999). This response is transient
and the levels of synthesis decline within three hours. The remaining cold shock
proteins are class H proteins which are also induced quickly, reaching maximum
synthesis by three hours (Jones et al., 1987).

In E. coli, depending on the extent of the temperature downshift, overall
cellular protein synthesis is inhibited for 2-4 hours following the temperature shift.
In fact, this inhibition of ribosome activity by low temperature has been proposed
to act as the signal stimulating the cold shock response. The evidence for this
connection arose in part from a study that linked ribosome activity to the
regulation of both heat shock and cold shock (VanBogelen and Neidhart, 1990).
When cells were exposed to non-lethal concentrations of antibiotics affecting
different aspects of ribosome ftmction, the pattern of protein synthesis resembled

either a cold shock response or a heat shock response depending on the antibiotic

15

used. The antibiotics that induced a response similar to a cold shock appeared to
slow or block activity of the ribosome. The effects of amino acid starvation and
ppGpp concentration on the cold shock response were investigated as well (Jones
et al., 1992a). Amino acid starvation or overproduction of the regulatory
nucleotide, ppGpp, prior to a temperature downshift resulted in the delay or
prevention of the cold shock response. Conversely, a rel/1 spoT mutant that
produced no ppGpp showed no growth lag when shifted to 10°C as compared to

the 2 hour growth lag displayed by the wild type strain. Thus, the alteration of

 

ribosome function either through the action of certain antibiotics or through amino
acid starvation influenced the regulation of the cold shock response.

The “Cold-Shock Ribosome Adaptation” model proposed by Jones and
Inouye (1996) takes into account the negative effect of low temperature on
ribosome function and the potential regulatory interaction between ribosome
activity and the cold shock response. According to this model, translation of most
cellular transcripts is inhibited by a shift to low temperature signalling a cold
shock response. In contrast to most cellular transcripts, the transcripts encoding
cold shock proteins and ribosomal proteins can be translated by “non-adapted”
ribosomes due to some unique property of these mRNAs. The newly synthesized
cold shock proteins interact with the ribosome to adapt it to function at low
temperature thus allowing overall protein synthesis to resume. If this model is

correct, inhibiting the ribosome by some other mechanism besides low temperature

16

should induce a full cold shock. Ten of the fourteen cold shock proteins assayed
increased following the addition of the “cold shock” antibiotic, tetracycline
(VanBogelen and Neidhart, 1990). This partial cold shock response indicates that
additional signals may be required.

In studying the regulatory mechanisms controlling the expression of the
cold shock genes, attention has focused on CspA as a potential regulator.
Overexpression of CspA, such that CspA levels were increased five-fold above
normal levels following a cold shock, resulted in the increased induction of at least
five cold shock proteins (Jones er al., 1992b). One of the cold shock proteins
affected by overexpression of CspA was DNA gyrase. C spA was further
implicated in the regulation of DNA gyrase as it was shown to bind the gyrA
promoter (Jones et al., 1992b). CspA enhances the expression of the H-NS protein
as well (La Teana et al., 1991). While CspA did not bind strongly to the hns
promoter, it did appear to enhance the binding and activity of RNA polymerase
(La Teana et al., 1991; Brandi et al., 1994). A cspA null mutant was not greatly
affected in grth or protein expression at either 37°C or 15°C with the exception
of the increased expression of CspB, CspG, and CspE (Bae et al., 1997).
Presumably, the increased expression of these related proteins was able to
compensate for the lack of CspA. Strains containing multiple deletions of CspA
family genes will be necessary to assess the possible regulatory functions of these

proteins.

17

 

The regulation of CspA, itself, has also been studied extensively. In most
of these studies, cultures of E. coli were grown to mid-log phase at 37°C prior to a
shifi to 10 or 15°C. In these studies, the level of cspA mRNA was extremely low
at 37°C and CspA protein synthesis was not detectable. These observations led to
the conclusion that CspA was specifically synthesized for low temperature
conditions. A recent study contradicts this conclusion as CspA was observed to be
highly expressed at 37°C when stationary phase cells were first introduced into a
fresh medium (Brandi et al., 1999). Although expression of CspA declined as cell
density increased, it remained detectable by Western blot. Previous studies had
quantified the synthesis of CspA without regard to the pool of CspA protein
already present in the cell. Measuring CspA levels by Western blot, Brandi er al.
(1999) did find that CspA levels increased following a cold shock although the
degree of induction ranged from 3 to 30-fold depending on the age of the initial
culture. Regardless of the amount of CspA present in the cells at 37°C, the same
level of CspA protein was attained following the cold shock. The finding that
CspA is highly expressed at 37°C following the transfer of stationary phase cells
to fresh medium may relate to the suggestion by Jones et al. (1992a) that the effect
of a cold shock is similar to that of a nutritional upshift. In both cases, the
capacity of the cell for protein synthesis is insufficient relative to the concentration
of charged tRNAs present.

Following a cold shock, cspA mRNA levels were induced within 30

18

rrrinutes of the temperature downshift. This change in message level did not result
from a burst in promoter activity but rather from increased stability of the cspA
message. Three nucleotide substitutions introduced just upstream of the cspA
Shine-Dalgarno sequence significantly stabilized the cspA mRNA at 37°C,
increasing the half-life from 12 seconds to 30 minutes CspA protein was highly
expressed in this strain at 37°C indicating that the cspA promoter was active at
both temperatures (Fang et al., 1997). Cloning the [pp promoter in front of the
cspA gene further illustrated the role of message stability in the induction of CspA
synthesis (Fang etal., 1997). The cspA message transcribed from the lppp-cspA
fusion was identical to the wild type cspA mRN A. When the lppp-cspA fusion was
integrated into the chromosome of a AcspA strain, no CspA protein was detected at
37°C despite the high constitutive activity of the [pp promoter. CspA protein was
' detected only after a cold shock to 15°C. Further characterization of cspA mRNA
stability following a cold shock from 37°C to 15°C demonstrated that cspA mRN A
was not permanently stabilized at low temperature. The cspA message reached a
maximum half-life of over 60 minutes within the first half hour at 15°C. Message
stability then decreased to a 10 minute half-life after an additional hour at 15°C
(Goldenberg et al. , 1996).

The cspA gene and the other class 1 cold shock genes, cspB, cspG, esp], and
csdA, have a long untranslated region (UTR) at the 5’ end of their message that is

145 to 226 bp in length (Figure 1.1) (Wang et al., 1999; Jiang et al., 1996).

19

 

Transcription start

 

 

 

 

 

 

 

site

+1

—- DNA
AT-rich
UP element Translation
start site

5, A 3 ,

N ”G ‘ RNA
cold box downstream box
5’ UTR ‘ ’ \
Transcriptional Ribosome
attenuation mRNA stability binding

 

 

 

 

 

 

 

 

Figure 1.1. Regulatory elements controlling the expression of the cold
shock protein, CspA. This figure is adapted fi'om Thieringer et al. (1998).

20

 

Transcriptional fusions including the full 5’ UTR of cspA showed increased
reporter gene expression following a cold shock (Goldenberg et al., 1997; Mitta et
al., 1997; Jiang et al., 1996). While a fusion to +26 showed comparable message
stability at 37°C, the full UTR sequence was required for transient stabilization at
low temperature (Mitta et al., 1997). Within the first 26 bp of the cspA message,
there is an 11 bp conserved sequence known as the cold box (Figure 1.1). i.
Overexpression of this sequence leads to the prolonged inhibition of cellular

protein synthesis and the prolonged synthesis of the class 1 cold shock proteins

 

following a temperature downshift (Jiang et al., 1996; Fang el al., 1998). The
effect on protein synthesis achieved by the overexpression of this sequence
indicates that a repressor protein may be binding this site after the cold shock. In
addition, each of the class I cold shock proteins has a cold box sequence in its 5’
UTR suggesting that these genes are coordinately regulated.

The 5’ UTR alone does not account for the full induction of the CspA
protein at low temperature. Mitta et al. (1997) constructed both transcriptional
and translational fusions in their study of cspA gene expression. Although a
transcriptional fusion to the lacZ gene showed induction of B-galactosidase
activity following a cold shock, the 5’ end of the cspA coding sequence was
required for full induction. This region contains a “downstream box” that is
complementary to the decoding region of the 168 rRN A thereby increasing the

affinity of this message for the ribosome (Figure 1.1) (Sprengart and Porter, 1997).

21

The location and sequence of the downstream box is conserved among all of the
cold-inducible cspA homologs in E. coli (Wang et al., 1999; Mitta et al., 1997).
The downstream box in the cold shock gene transcripts may explain why these
mRNAs can be translated immediately following a temperature downshift prior to
adaptation of the ribosome to low temperature.

Bae er al. (1997) have proposed that CspA negatively regulates its own
expression. In the construction of a AcspA mutant, only the cspA coding region

was removed leaving an intact cspA promoter and 5’ UTR. While the deletion of

 

the coding region did not result in a growth defect, the expression of the cspA 5’
UTR was altered. Both transcription and stability of the 5’ UTR were affected
following a cold shock. Immediately after a temperature downshift, the half-lives
of both the wild type cspA transcript and the mutant cspA 5’ UTR were over 60
minutes. The level of cspA 5’ UTR in the AcspA mutant was approximately 3-fold
higher than the level of the wild type gene. These results indicate that
immediately after a cold shock the transcription of the cspA gene in the mutant
strain has increased. Three hours after the cold shock both the cspA and cspB
transcripts in the mutant strain had almost doubled in stability relative to the wild
type strain. The authors suggest that CspA may affect its own expression in two
ways. First, as a RNA chaperone, CspA may bind its own message and enhance
its degradation. A second mechanism may be through transcriptional attenuation,

potentially mediated through the binding of CspA to its own message (Bae er al.,

22

1997)

The “Cold-Shock Ribosome Adaptation” model proposed by Jones and
Inouye (1996) describes one mechanism involved in sensing temperature and
stimulating the cold shock response, but it is likely that other mechanisms are
involved as well. The exact regulatory mechanisms leading to the induction of
cold shock protein synthesis are still not clear. While overexpression of CspA led
to increased levels of several cold shock proteins, other data suggests that CspA
acts as a RNA chaperone. The regulation of CspA, itself, involves the interaction
of several mechanisms. The extended 5’ UTR is responsible for cspA message
instability at 37°C and its transient stabilization after a cold shock. In addition, the
cold box at the 5’ end of the cspA transcript appears to be involved in negative
regulation at low temperature, possibly through the binding of CspA to its own
transcript. The downstream box may account for the extremely efficient
translation of the cspA message at low temperature. Whether the cspA promoter
plays a significant role during a cold shock is still not known as no +1 fusion has
been examined at low temperature. Finally, the growth phase dependent
regulation of CspA synthesis described by Brandi et al. (1999) may give new
insights into the regulation of the CspA family of proteins and the cold shock

response.

23

The cold shock response in Sinorhizobium meliloti

Sinorhizobium meliloti 1021, as a soil bacterium, must survive a wide range
of temperatures, including those well below its optimum grth temperature of
30°C. Considering the apparent conservation of the cold shock response among
different bacterial species, it seems likely that S. meliloti 1021 possesses a
mechanism for adapting to sub-optimal temperatures. Preliminary studies of a
related strain, Rhizobium meliloti A2, performed by Cloutier et al. (1992)
indicated that this bacterium was capable of protein synthesis at temperatures
below its minimum growth temperature of 7°C. However, the identity of the
proteins synthesized at low temperature, their relative synthesis at optimal
temperature, and their importance for the survival of R. meliloti A2 at low
temperature was not determined. In our study of S. meliloti 1021, we hoped to
identify those genes whose expression increased at low temperature, to assess their
importance to growth at low temperature, and to begin to characterize their

regulation in response to temperature.

24

Literature Cited

Afflerbach, H., O. Schriider, and R. Wagner. 1998. Effects of the Escherichia
coli DNA-binding protein H-NS on rRN A synthesis in viva. Mol. Microbiol.
28(3):641-653.

Aguilar, P. S., J. E. Cronan, JR., and D. de Mendoza. 1998. A Bacillus subtilis
gene induced by cold shock encodes a membrane phospholipid desaturase. J.
Bacteriol. 180(8):2194-2200.

Atlung, T. and H. Ingmer. 1997. H-NS: a modulator of environmentally
regulated gene expression. Mol. Microbiol. 24(1):7-17.

Bae, W., P. G. Jones, and M. Inouye. 1997. CspA, the major cold shock protein
of Escherichia coli, negatively regulates its own gene expression. J. Bacteriol.
179(22):7081-7088.

Berger, F., P. Normand, and P. Potier. 1997. capA, a cspA-like gene that
encodes a cold acclimation protein in the psychrotrophic bacterium Arthrobacter
globiformis 8155. J. Bacteriol. 179(18):5670-5676.

Brandi, A., C. L. Pon, and C. O. Gualerzi. 1994. Interaction of the main cold
shock protein CS7 .4 (CspA) of Escherichia coli with the promoter region of hns.
Biochimie 76: 1090-1098.

Brandi, A., R. Spurio, C. O. Gualerzi, and C. L. Pan. 1999. Massive presence
of the Escherichia coli ‘major cold-shock protein’ CspA under non-stress
conditions. EMBO J 18(6): 1653-1659.

Chamot, D., W. C. Magee, E. Yu, and G. W. Owttrim. 1999. A cold shock-
induced cyanobacteria] RNA helicase. J. Bacteriol. 181(6): 1728-1732.

Cloutier, J., D. Prévost, P. Nadeau, and H. Antoun. 1992. Heat and cold shock
protein synthesis in arctic and temperate strains of rhizobia. Appl. Environ.
Microbiol. 58(9):2846-2853.

Dersch, P., S. Kneip, and E. Bremer. 1994. The nucleoid-associated DNA-
binding protein H-NS is required for the efficient adaptation of Escherichia coli
K-12 to a cold environment. Mol. Gen. Genet. 245:255-259.

25

Fang, L., W. Jiang, W. Bae, and M. Inouye. 1997. Promoter-independent cold-
shock induction of cspA and its derepression at 37°C by mRN A stabilization.
Mol. Microbiol. 23(2):355-364.

Fang, L., Y. Hou, and M. Inouye. 1998. Role of the cold-box region in the 5’
untranslated region of the cspA mRN A in its transient expression at low
temperature in Escherichia coli. J. Bacteriol. 180(1):90-95.

Goldenberg, D., I. Azar, and A. B. Oppenheim. 1996. Differential mRNA
stability of the cspA gene in the cold-shock response of Escherichia coli. Mol.
Microbiol. 19(2):241-248.

Goldenberg, D., I. Azar, A. B. Oppenheim, A. Brandi, C. L. Pan, and C. O.
Gualerzi. 1997. Role of Escherichia coli cspA promoter sequences and
adaptation of translational apparatus in the cold shock response. Mol. Gen. Genet.
256:282-290.

Goldstein J., N. S. Pollitt, and M. Inouye. 1990. Major cold shock protein of
Escherichia coli. Proc. Natl. Acad. Sci. USA 87:283-287.

Goverde, R. L. J., J. H. J. Huis in’t Veld, J. G. Kusters and F. R. M001. 1998.
The psychrotrophic bacterium Yersinia enterocolitica requires expression of pnp,
the gene for polynucleotide phosphorylase, for grth at low temperature (5°C).
Mol. Microbiol. 28(3):555-569.

Graumann, P., K. Shcrtider, R. Schmid, and M. A. Marahiel. 1996. Cold
shock stress-induced proteins in Bacillus subtilis. J. Bacteriol. 178(15):46l l-
4619.

Graumann, P. L., and M. A. Marahiel. 1999. Cold shock proteins CspB and
CspC are major stationary-phase-induced proteins in Bacillus subtilis. Arch.
Microbiol. 171:135-138.

Jiang, W., P. Jones, and M. Inouye. 1993. Chloramphenicol induces the
transcription of the major cold shock gene of Escherichia coli, cspA. J. Bacteriol.
175(18):5824-5828.

Jiang, W., L. Fang, and M. Inouye. 1996. The role of the 5’-end untranslated

region of the mRNA for CspA, the major cold-shock protein of Escherichia coli,
in cold-shock adaptation. J. Bacteriol. 178(16):4919-4925.

26

Jiang, W., Y. Hon, and M. Inouye. 1997. CspA, the major cold-shock protein of
Escherichia coli, is an RNA chaperone. J. Biol. Chem. 272(1): 196-202.

Jones, P. G. and M. Inouye. 1996. RbfA, a 308 ribosomal binding factor, is a
cold-shock protein whose absence triggers the cold-shock response. Mol.
Microbiol. 21(6):1207-1218.

Jones, P. G., R. A. VanBogelen, and F. C. Neidhart. 1987. Induction of
proteins in response to low temperature in Escherichia coli. J. Bacteriol.
169(5):2092-2095.

Jones, P. G., M. Cashel, G. Glaser, and F. C. Neidhart. 1992a. Function of a
relaxed-like state following temperature downshifts in Escherichia coli. J.
Bacteriol. 174(12):3903-3914.

Jones, P. G., R. Krah, S. R. Tafuri, and A. P. Wolffe. 1992b. DNA gyrase,
CS7.4, and the cold shock response in Escherichia coli. J. Bacteriol.
174(18):5798-5802.

Jones, P. G., M. Mitta, Y. Kim, W. Jiang, and M. Inouye. 1996. Cold shock
induces a major ribosomal-associated protein that unwinds double-stranded RNA
in Escherichia coli. Proc. Natl. Acad. Sci. USA 93:76-80.

Kandror, O. and A. L. Goldberg. 1997. Trigger factor is induced upon cold
shock and enhances viability of Escherichia coli at low temperatures. Proc. Natl.
Acad. Sci. USA. 94:4978-4981.

La Teana, A., A. Brandi, M. Falconi, R. Spurio, C. L. Pan, and C. O.
Gualerzi. 1991. Identification of a cold shock transcriptional enhancer of the
Escherichia coli gene encoding nucleoid protein H-NS. Proc. Natl. Acad. Sci.
USA. 88: 10907-1091 1.

Lee, S. J., A. Xie, W. Jiang, J. Etchegaray, P. G. Jones, and M. Inouye. 1994.
Family of the major cold-shock protein, CspA (CS7.4), of Escherichia coli, whose
members show a high sequence similarity with the eukaryotic Y-box binding
proteins. Mol. Microbiol. 11(5):833-839.

Lelivelt, M. J., and T. H. Kawula. 1995. Hsc66, an Hsp70 homolog in

Escherichia coli, is induced by cold shock but not by heat shock. J. Bacteriol.
177(17):4900-4907.

27

Los, D. A., M. K. Ray, and N. Murata. 1997. Differences in the control of the
temperature-dependent expression of four genes for desaturases in Synechocystis
sp. PCC 6803. Mol. Microbiol. 25(6): 1 167-1 175.

Mayr, B., T. Kaplan, S. Lechner, and S. Scherer. 1996. Identification and
pmification of a family of dirneric major cold shock protein homologs from the
psychrotrophic Bacillus cereus WSBC 10201. J. Bacteriol. 178(10):2916-2925.

Michel, V., I. Lehoux, G. Depret, P. Anglade, J. Labadie, and M. Hebraud.
1997. The cold shock response of the psychrotrophic bacterium Pseudomonas
fragi involves four low-molecular-mass nucleic acid-binding proteins. J.
Bacteriol. 179(23):7331-7342.

Mitta, M., L. Fang, and M. Inouye. 1997. Deletion analysis of cspa of
Escherichia coli: requirement of the AT-rich UP element for cspA transcription
and the downstream box in the coding region for its cold shock induction. Mol.
Microbiol. 26(2):321-335.

Nakashima, K., K. Kanamaru, T. Mizuno, and K. Horikoshi. 1996. A novel
member of the cspA family of genes that is induced by cold shock in Escherichia
coli. J. Bacteriol. 178(10):2994-2997.

Nedwell, D. B. and M. Rutter. (1994). Influence of temperature on growth rate
and competition beween two psychrotolerant antarctic bacteria: low temperature
diminishes affinity for substrate uptake. Appl. Environ. Microbiol. 60: 1984-1992.

Newkirk, K., W. Feng, W. Jiang, R. Tejero, S. D. Emerson, M. Inouye, and G.
T. Montelione. 1994. Solution NMR structure of the major cold shock protein
(CspA) from Escherichia coli: identification of a binding epitope for DNA. Proc.
Natl. Acad. Sci. USA 91:5114-5118.

O’Connell, K. P. and M. F. Thomashow. 1999. Transcriptional organization
and regulation of a polycistronic cold shock operon in Sinorhizobium meliloti
RM1021 that includes the major cold shock gene cspA and a homolog of
ribosomal protein gene rpsU. (in preparation)

Russell, N. J. 1990. Cold adaptation of microorganisms. Phil. Trans. R. Soc.
Lond. B 326:595-611.

28

Sakamoto, T. and D. A. Bryant. 1997. Temperature-regulated mRNA
accumulation and stabilization for fatty acid desaturase genes in the
cyanobacterium Synechococcus sp. strain PCC 7002. Mol. Microbiol.
23(6):1281-1292.

Sakamoto, T., G. Shen, S. Higashi, N. Murata, D. A. Bryant. 1998. Alteration
of low-temperature susceptibility of the cyanobacterium Synechococcus sp. PCC
7002 by genetic manipulation of membrane lipid unsaturation. Arch. Microbiol.
169220-28.

Sato, N. 1994. A cold-regulated cyanobacterial gene cluster encodes RNA-
binding protein and ribosomal protein S21. Plant Mol. Biol. 24:819-823.

Sato, N. 1995. A family of cold-regulated RNA-binding protein genes in the
cyanobacterium Anabaena variabilis M3. Nucleic Acids Res. 23(12):2161-2167.

Schindelin, H., W. Jiang, M. Inouye, and U. Heinemann. 1994. Crystal
structure of CspA, the major cold shock protein of Escherichia coli. Proc. Natl.
Acad. Sci. USA 91:5119-5123.

Silberg, J. J., K. G. Hoff, and L. E. Vickery. 1998. The Hsc66-H5020
chaperone system in Escherichia coli: chaperone activity and interactions with the
DnaK-DnaJ-GrpE system. J. Bacteriol. 180(24):6617-6624.

Sprengart, M. L., and A. G. Porter. 1997. Functional importance of RNA
interactions in selection of translation initiation codons. Mol. Microbiol.

24(1):19-2s.

Thieringer, H. A., P. G. Jones, and M. Inouye. 1998. Cold shock and
adaptation. BioEssays 20:49-57.

VanBogelen, R. A., and F. C. Neidhardt. 1990. Ribosomes as sensors of heat
and cold shock in Escherichia coli. Proc. Natl. Acad. Sci. USA 87 :5589-5593.

Wang, N., K. Yamanaka, and M. Inouye. 1999. Cspl, the ninth member of the
CspA family of Escherichia coli, is induced upon cold shock. J. Bacteriol.
181(5): 1603-1609.

Wistow, G. 1990. Cold shock and DNA binding. Nature 344:823-824.

29

Wouters, J. A., J.-W. Sanders, J. Kok, W. M. de Vos, O. P. Kuipers, and T.
Abee. 1998. Clustered organization and transcriptional analysis of a family of
five csp genes of Lactococcus lactis MG1363. Microbiol. 144:2885-2893.

Yamanaka, K., L. Fang, and M. Inouye. 1998. The CspA family in Escherichia
coli: multiple gene duplication for stress adaptation. Mol. Microbiol. 27(2):247-
255.

 

 

3O

CHAPTER 2

Identification of low temperature induced genes in Sinorhizobium
meliloti 1021

Summary

Transposon mutagenesis of Sinorhizobium meliloti 1021 was conducted in
an effort to identify genes whose products might be involved in adaptation to sub-
optimal growth temperatures. The transposon Tn5-1062 containing the
promoterless luxAB genes of Vibrio harveyi was introduced into the wild type
strain S. meliloti 1021. Approximately 23,000 transposon recipients were screened
for those exhibiting increased luminescence after a temperature shift from 30°C to
15°C. Seven strains were isolated. Sequencing of the insertion sites revealed that
in five of the strains the transposon had inserted in either a 16S or a 23S rRNA
gene. Southern blot analysis indicated that the transposons had inserted into two
of the three ribosomal RNA operons present in S. meliloti 1021. Direct
mutagenesis of the three ribosomal RNA operons indicated that the luxAB genes
were affected by the temperature shift to a similar extent in each of the three
operons. The upstream region of the ribosomal RNA operon containing the

transposon in strain RM3166 was isolated and cloned in front of the luxAB genes

31

on the broad host range plasmid, pRK290. This 605 bp fragment was shown to
contain the sequences required to promote the temperature-dependent expression

of the luxAB genes.

32

Introduction

As a soil bacterium, Sinorhizobium meliloti, must adapt to a wide range of
temperatures. While the response of microorganisms to high temperature is
relatively well understood, the adaptive response of bacteria to low and freezing
temperatures is just beginning to be described. The cold shock response has been
examined in several bacterial species including several Rhizobia (Cloutier et al.,
1992), Bacillus subtilis (Graumann et al., 1996), Escherichia coli (Jones et al.,
1987), Lactococcus lactis (Panoff et al., 1994), and the psychrotrophs
Arthrobacter globiformis S155 (Berger et al., 1996), Bacillus psychrophilus
(Whyte and Inniss, 1992), and Pseudomonasfragi (Michel et al., 1997). The cold
shock response occurs following a sudden drop in growth temperature and
involves the induction of a set of proteins, termed the cold shock proteins. Overall
protein synthesis is often inhibited immediately following a temperature downshift
resulting in a temporary growth lag.

While the cold shock response has been observed in several bacterial
species and yeast, it has been studied most thoroughly in E. coli. Most of the
induced proteins, or cold shock proteins, are also synthesized at optimum
temperatures and are only induced between 2 and 10-fold (Jones et al., 1987).
These proteins include polynucleotide phosphorylase, NusA, lF20r/B, RecA,

dihydrolipoamide acetyltransferase, pyruvate dehydrogenase, DNA gyrase A, H-

33

NS (La Teana et al., 1991), RbfA (Jones and Inouye, 1996), Hsc66 (Lelivelt and
Kawula, 1995), and trigger factor (Kandror and Goldberg, 1997). These proteins
are known to have roles in important cellular processes including mRN A
degradation, transcriptional termination and antiterrnination, translation,
chromosome structure, energy metabolism and protein structure. In addition,
several ribosomal proteins continue to be synthesized and in some cases are
reported to be moderately induced despite the overall inhibition of protein
synthesis (Jones and Inouye, 1994; Jones et al., 1996).

Another class of cold shock proteins includes those which are expressed
primarily at low temperatures. Four of these proteins, CspA, CspB, CspG and
Cspl, are highly induced following a cold shock from 37°C to 15°C (Goldstein et
al., 1990; Lee et al., 1994; Nakashima et al., 1996; Wang et al., 1999). The
synthesis of CspA alone accounts for 13% of total protein synthesis within an hour
of a temperature downshift to 15°C. These proteins are members of the CspA
family of proteins that share significant similarity with the cold shock domain of
the eukaryotic Y-box protein family (Wistow, 1990). Proteins homologous to
CspA have been identified in many bacteria. In addition to being implicated in the
transcriptional regulation of two cold shock genes, hns and gyrA, CspA has also
been shown to bind cooperatively to single-stranded RNA and to destabilize
secondary structure (La Teana et al., 1991; Jones et al., 1992b; Jiang et al.,

1997). Because of its affinity for RNA, CspA has been proposed to function as a

34

RNA chaperone in preventing formation of secondary structures that could
interfere with translation of mRNA. Another protein, CsdA, is also expressed
predominantly at low temperature although it is not induced as highly as CspA and
CspB (Jones et al., 1996). CsdA contains the DEAD motif which is highly
conserved among RNA and DNA helicases and has been shown to unwind double-

stranded RNA. As CsdA is associated with both ribosomal subunits at low

 

temperature in approximately stoichiometric amounts, it has been speculated that

CsdA may aid in translation at low temperature by removing deleterious secondary

 

u.-
I ll

structures in mRN A. I
Protein synthesis has been shown to be the most cold-sensitive process in
Escherichia coli and, specifically, the initiation of translation has been shown to
be inhibited at low temperatures (Friedman et al. , 1971). Interestingly, a set of
antibiotics that inhibit ribosome function can stimulate a cold shock response at
37°C when added to cultures in non-lethal amounts (VanBogelen and Neidhart,
1990). Based on these observations and the induction of several proteins involved
in translation, a model of the cold shock response as a mechanism for adapting the
ribosome to low temperature has been developed (Jones and Inouye, 1996). The
actions of the cold shock proteins RbfA, C sdA, IF2 and CspA are thought to
enable the ribosome to function at low temperature. These proteins are either
associated with the ribosome itself or involved in destabilizing secondary

structures in RNA. CsdA and CspA are thought to interact with mRNA, removing

35

any secondary structures that may interfere with their translation.

Whether this model is accurate for the cold shock response in
Sinorhizobium meliloti 1021 is not clear. While Cloutier ei al. (1992) showed that
several proteins continued to be synthesized in Rhizobium meliloti A2 following a
cold shock to temperatures as low as 0°C, the identity and function of these
proteins is unknown. Clearly, there is much to be learned about the response of F
Sinorhizobium meliloti 1021 to low temperature.

In performing transposon mutagenesis of S. meliloti 1021, we hoped to

 

identify those genes whose expression was induced following a temperature
downshift. This method was thought to give an advantage over two-dimensional
gel electrophoresis in allowing the isolation and tagging of the induced genes. A
Tn5 derivative, Tn5-1062, was used in this study as Tn5 transposons have been
shown to work well in the mutagenesis of S. meliloti (Milcamps et al., 1998). This
reporter transposon carries the promoterless Vibrio harveyi luxAB genes encoding
bacterial luciferase (Meighen, 1991). Luciferase releases photons when catalyzing
the following reaction: FMNHZ + RCHO + Oz —> FMN + RCOOH + H20. The
luxAB genes are expressed when the transposon inserts in the correct orientation
downstream of an active promoter. In this way, Tn5-1062 can be used as an
indicator of promoter activity. Since the enzyme substrate, dodecyl aldehyde, can
penetrate the cellular membrane, luciferase activity can be assayed without

disrupting the bacterial cells and the same colonies can be assayed multiple times.

36

We present here the isolation of seven cold-inducible Tn5-1062 insertion
mutants. Five of these insertions are located within ribosomal RNA genes. These
insertions are mapped to two of the three rm operons known to exist in S. meliloti
1021 (Honeycutt et al., 1993). Sequences required for the cold-inducible
luminescence of the luxAB genes were isolated on a 605 bp BglII-Sacl fragment of

one of the rrn promoter regions.

37

 

Results

Isolation of transposon mutants induced by low temperature.
Sinorhizobium meliloti 1021 was mutagenized with the transposon Tn5-1062
carrying the promoterless luxAB gene cassette (Figure 2. 18). When inserted in the
correct orientation downstream of an active promoter, the transposon’s luxAB

genes are transcribed as part of the endogenous transcript. The transposon was

 

moved into the cells on the plasmid pRL1062 via triparental matings (Figure

 

2.1A). This plasmid is not replicated in S. meliloti 1021, so any cell gaining
kanamycin resistance must have the transposon inserted in the chromosome. After
selecting for kanamycin resistance, the resulting colonies were screened for those
that exhibited increased luminescence after a temperature downshift from 30°C to
15°C or from 30°C to 10°C.

Most transposon recipients showed a slight increase in luminescence after 4
to 6 hours at the lower temperature (Figure 2.2). This 2 to 5-fold increase was
considered to be a background increase in activity resulting from a greater stability
of either the luxAB mRNA or the luciferase protein at the lower temperature.
Three rounds of mutagenesis and screenings were performed resulting in the
isolation of seven mutant strains showing markedly increased luminescence after a
temperature downshift. The first round of screening involved a temperature shift

from 30°C to 10°C. The mutant, RM3166 (Figure 2.3A), was isolated after

38

 

B

 

 

“I‘m my Kmr T.........._m

Figure 2.1. Structure of pRL1062a and Tn5-1062 inserted in genome. The
plasmid pRL1062a was used to move Tn5-1062 into S. meliloti via conjugation.
(A) The plasmid carries the origin of transfer, oriRRKZ), and the transposon
delimited by 52 bp of the IS50L,{ , and its inverted repeat}, in rssoa The
oriV(p15A) located within the transposon supports replication of this plasmid in E.
coli but not in S. meliloti. Arrows indicate the direction of transcription of the
various genes. The hocAB genes are located within the XbaI fiagment. (B) Tn5-
1062 is shown after transposition into the bacterial genome. The inverted repeats
of IS50L, { , and 1S50R,) , delimit the transposon sequence. If the transposon
inserts in the genomic DNA downstream of an active promoter in the correct
orientation, the luxAB genes will be transcribed.

39

 

30°C

15°C

 

Figure 2.2. Screen for Tn5-1062 inserts induced by a temperature
downshift. 48 Tn5-1062 recipients were streaked on TY agar containing
kanamycin (200 [Lg/ml) and streptomycin (250 ug/ml). The plates were
incubated for 2 days at 30°C, exposed to aldehyde and the luminescence
recorded with a photonic camera (top photo). The plates were transferred to
15°C for 5 1/2 hours, re-exposed to aldehyde and observed again (bottom
photo). The arrow indicates the mutant strain RM603.

40

Figure 2.3. Set of seven Tn5-1062 inserts induced by a temperature
downshift. (A) Strains were cultured on TY agar plates at 30°C for two days
and then assayed for luminescence. The plates were then shifted to 15°C for 5
hours before being assayed again. (B) Strains were grown in TY broth to an
A600 of approximately 0.4 and assayed for luminescence before (30°C) and after
(15°C) a temperature downshift. Each column represents the mean 1 SD of
three measurements per strain. This graph is adapted fi'om O’Connell et al.
(1999)

41

'5“)
~as

£§§£§e

£16766

30°C

15°C

 

 

 

 

   

 

 

 

B
D3o°c .15°C

8

l‘

é.

m6

.5

£4

S

:32

i I I I I I I
O _ __ _

ea“ as” a‘" er" es” er” 9&‘9’66

Figure 2.3

42

examining approximately 17,000 kanamycin resistant colonies. Five additional
strains were isolated, including RM5 18, RM523, and RM603, in a second round
of mutagenesis and screening using a temperature shift from 30°C to 15°C (Figure
2.3A). Approximately 5,000 transposon recipients were screened in this round.
One thousand more mutants were analyzed in a third round of screening utilizing
the same temperature shift but a minimal growth medium, GTS, instead of the rich
medium, TY, used previously. This last screen resulted in the isolation of strain

RM73 (Figure 2.3A).

The seven Tn5-1062 recipient strains consistently showed a marked
increase in light production after a temperature downshift when colonies were
assayed for luciferase activity with a photonic camera. In order to more accurately
quantitate the change in luminescence, the mutant strains were grown in broth and
light production was quantified using a luminometer (Figure 2.38). When
analyzed in this way, luminescence could be adjusted to cell density allowing
more accurate comparisons between samples and time points. The cultures were
grown up at 30°C and assayed immediately prior to being shifted to 15°C and five
hours after the temperature shift. The mutant strains showed large increases in
luminescence after a cold shock. Luminescence in mutants RM 11 and RM509
increased over 100-fold while luminescence in the remaining mutants increased

between 26 and 76-fold.

43

Identification of the cold induced loci. In order to characterize the
transposon insertion sites, the Tn5-1062 transposons along with the neighboring
genomic DNA were cloned out of each strain. The Tn5-1062 transposon contains
an origin of replication, p15a, recognized in Escherichia coli allowing the
transposon to be isolated and maintained as a plasmid outside of S. meliloti (Figure
2.1A). Genomic DNA isolated from the strains was typically digested with an
enzyme that did not out within the transposon. In some cases an enzyme that did
out within the transposon was used but was selected such that the region encoding
kanamycin resistance and containing the origin of replication remained intact. The
resulting restriction fragments were circularized and transformed into E. coli

DHSor.

The genomic DNA flanking the transposons was sequenced and compared
to known sequences. The DNA contiguous with the transposon in RM3166 was
nearly identical to a region of a 16S rRNA gene in R. meliloti LMG 613 (Figure
2.4) while the transposon insertion site in RM73 was very similar to a 23S rRNA
gene in Agrobacterium vitis (Figure 2.5). Analysis of the remaining mutants
revealed that the transposons in 5 of the 7 mutants were inserted in rRNA genes
(Table 2.1). The transposons in mutants RM518, RM603 and RM3166 were in
168 rRNA genes and the transposons in RM73 and RM523 were inserted in 23S
rRNA genes. The two remaining insertion mutants, RMll and RM509, contained

insertions in an operon encoding homologs of the E. coli proteins, CspA and S21

44

Figure 2.4. Sequence alignment of the RM3166 transposon insertion site and
the 16S rm gene of Rhizobium meliloti LMG 613. DNA contiguous with the
transposon in RM3166 was sequenced. A BLASTN search (Altschul et al, 1997)
revealed that the sequence was similar to the 16S rm of Rhizobium meliloti LMG
613 (accession number, X67222). A = RM3166, B= R meliloti LMG 613, " I "=
identity, ":" = mismatch. Gaps were included in sequence (-) to facilitate
alignment.

45

U 0' U 3' U 3' U 3’ U 3’ U 3'

I! 3' fl! 3

U 3' U 3’

77

840

127

890

177

940

227

990

277

1040

327

1090

377

1140

426

1190

476

1240

525

1290

GGGCCGCACAAGCGGTGGAGCATGTGGTTTAATTCGAAGCAACGCGCAGA

llll|l|||||IllllllIlllllllllllllllllllllllllllllll
GGGCCGCACAAGCGGTGGAGCATGTGGTTTAATTCGAAGCAACGCGCAGA

ACCTTACCAGCCCTTGACATCCCGATCGCGGATACGAGAGATCGTATCCT

||||||||||||||||||||||||||llllllllllllllllllllllll
ACCTTACCAGCCC TTGACATCCCGATCGCGGATACGAGAGATCGTATCCT

TCAGTTCGGCTGGATCGGAGACAGGTGCTGCATGGCTGTCGTCAGCTCGT

|||||||ll||||l||||||||||||llllllllllllllllllllllll
TCAGTTCGGCTGGATCGGAGACAGGTGCTGCATGGCTGTCGTCAGCTCGT

GTCGTGAGATGTTGGGTTAAGTCCCGCAACGAGCGCAACCCTCGCCCTTA

||||||lllllIllllllllllllllllllllll||||||l|l|||||||
GTCGTGAGATGTTGGGTTAAGTCCCGCAACGAGCGCAACCCTCGCCCTTA

GTTGCCAGCATTCAGTTGGGCACTCTAAGGGGACTGCCGGTGATAAGCCG

||llllllllll|llllllllllllIllllllllllllllllllllllll
GTTGCCAGCATTCAGTTG GGGCACTCTAAGGGGACTGCCGGTGATAAGCCG

AGAGGAAGGTGGGGATGACGTCAAGTCCTCATGGCCCTTACGGGCTGGGC

llllI||l|llIll|||||||llIl||||||||||||||||||||l||||
AGAGGAAGGTGGGGATGACGTCAAGTCCTCATGGCCCTTACGGGCTGGGC

TACACACGTGCTAC-ATGGTGGTGACAGTGGGCAGCGAGACCGCGAGGTC

lllllllllllllltllllllllllll|||||||||||||||||||||||
TACACACGTGCTACAATGGTGGTGACAGTGGGCAGCGAGACCGCGAGGTC

GACTGAATCTCCAAAAGCCATCTCAGTTCGGATTGCACTCTGCAACTCGA

l|:::|llllI|Ill||l|llllllllllllllllllllllllllllIll
GAGCTAATCTCCAAAAGCCATCTCAGTTCGGATTGCACTCTGCAACTCGA

GTGCATGAAGTTGGAATCGCTAGTAATCGCAGATCAGCATGCTGC- GTGA

lllllllllllllllllllIllllllllllllllllll|||||||= llll
GTGCATGAAGTTGGAATCGCTAGTAATCGCAGATCAGCATGCTGCGGTGA

ATACGTTCCGG 535

|||||||||:|
ATACGTTCCCG 1300

Figure 2.4

46

Figure 2.5. Sequence alignment of the RM73 transposon insertion site and
the 23S rm gene of Agrobacterium vitis. The DNA bordering the transposon in
RM73 was sequenced. A BLASTN search (Altschul et a1, 1997) revealed
significant identity to the 23S rm gene in A. vitis (accession number U45329). A
= RM73, B = A. vitis, " I "= identity, ":" = mismatch. Gaps were included in
sequence (-) to facilitate alignment.

47

F

, 1! D' U

V I! 3' U 3’ U fi’ U 8’ U

3

1901

51

1949

101

1999

151

2049

201

2099

251

2149

301

2198

351

2248

401

2298

451

2348

501

2396

Figure 2.5

CTGTCTCCCAACATAGACTCAGTGAAATTTGAATTCCCCGTGAAGATGCG

llll||:|||||I||||||llllllll:||||llllllllllllllllll
CTGTCT-CCAACATAGACTCAGTGAAA-TTGAATTCCCCGTGAAGATGCG

GGGTTCCTGCGGTCAGACGGAAAGACCCCGTGCACCTTTACTATAGCTTT

llllIlllllll||ll|IIlllllllllllIlllllllllllllllllll
GGGTTCCTGCGGTCAGACGGAAAGACCCCGTGCACCTTTACTATAGCTTT

ACACTGGCATTCGTGTCGGCATGTGTAGGATAGGTGGTAGGCTTTGAAGC

llllllll|lllllllllllllllllllllll||||||||||||||||||
ACACTGGCATTCGTGTCGGCATGTGTAGGATAGGTGGTAGGCTTTGAAGC

GGGGACGCCAGTTCTCGTGGAGCCATCCTTGAAATACCACCCTTATCGTC

::|ll|l||Ill||::llll|||l|lllllllllllllllllllllllll
TTGGACGCCAGTTCGAGTGGAGCCATCCTTGAAATACCACCCTTATCGTC

ATGGATGTCTAACCGCGGTCCGTCATCCGGATCCGGGACAGTGTATGGTG

|||||lllllll||l||l|||||:||||||||||||||||l||l||ll|l
ATGGATGTCTAACCGCGGTCCGTTATCCGGATCCGGGACAGTGTATGGTG

AAATAGATCGACTGGNGTGGTCGCCTCCGAAAGAGTAACGGAGGCGCGCG

:::|||:|:|ll|l|:|:|l||||||||ll||l|l|llllllllllllll
-GGTAGTTTGACTGGGGCGGTCGCCTCCGAAAGAGTAACGGAGGCGCGCG

ATGGTGGGCTCAGAGNGGTCGGAAATCGCTCGTCGAGTGCAATGGNATAA

|||||lllll||||::llllll||||||:||||||||||||l||1:||||
ATGGTGGGCTCAGACCGGTCGGAAATCGGTCGTCGAGTGCAATGGCATAA

GCCCGTCTGACTGCGAGACTGACAAGTCGAGCAGAGATGATAGTCGGTCA

lllll:lll||||llllllllllllllllllll||||:||:|||||l|||
GCCCGCCTGACTGCGAGACTGACAAGTCGAGCAGAGACGAAAGTCGGTCA

TAGTGATCCGGTGGACCCGCGTGGAAGGGTCATCGATCAANGGATAANAG

llllllllllllll:||l|||l|l|||||:|||||:||ll:llllll:||
TAGTGATCCGGTGGTCCCGCGTGGAAGGGCCATCGCTCAACGGATAAAAG

GTANGCCGGGGCATAACAGGCTGATGACCCCCANGTAGTCCATATCGACG

|||:|||||||:||l|l||||l||llll||l|l:|:||||l||l|||||l
GTACGCCGGGG-ATAACAGGCTGATGACCCCCAAG-AGTCCATATCGACG

GTGTTNGTCTGGGTACCTNGATG 523

|:|||::::|:||:llll:||||
GGGTTG--TTTGGCACCTCGATG 2416

48

 

Hue—o uh. 33952.33 ea «3.5—58.. mama—.32. «.28 m: mini—ea» Bea—=8.

made.

”735

5,380

”Kw—mo

Ed

Emma

momeosoo mam—Kuhn

Go 3

mum Eu

mum Eu

mnu 3

85 3

wow" mono 830:

am 3.:

Swan:

3m}:

Nwm 3.:

“mm 3.:

0.. an...

BeehiSaS mu.
Nwoaw H

W SmEez. PEG a;
quuww

N 3326a. ESQ 3 u
v3.3.5

Smiegaeles ea...»
:3ch

Smaegamlei 5.5
C330

953.2 830..

mam

moo

mun

H:

A8

m «Eco—c

 

No.3

0-75

ob

ob

9.3

.35 came—1m:— RSS 25 mecca—So 8858: 3:309. mm mad: m9. 9o com» woe—goo :58: m9. one: 5830: $8.
V5.5 253.2 mooa 25 m <25 Ed mg": mm 8833 5 one: wr>mfiz 830: 3.80::— Q E. 30$. .55 m «ice

53088 ~52 =53. moneoaoom $08 :5 3838 $9.5 ca 86098 8 8876 En SEQ—8a wgmfiz 808 3. 0.558.

95% mos—8:8 2mm awesom— 8 3:83 on «95: am «W2.» moacoaoom 525mg 90 can 589

49

(O'Connell and Thomashow, 1999).

Mapping of transposon insertions to specific rm operons. Honeycutt et
al. (1993) identified three rm operons in S. meliloti 1021. To determine whether
the apparent temperature-dependent regulation was specific to a particular set of
ribosomal RNA genes, the transposon inserts were mapped to the individual rrn
operons. The enzyme Sacl does not cut within or between the 163 or 23S rRNA
genes, allowing three fragments representative of the three rrn operons to be
resolved on a Southern blot (Figure 2.6A, RM1021). Southern blot analysis of
RM3166 and RM603 showed that the transposon was in the rrn operon contained
in fragment 1 as this band was missing in these strains and another, larger band
was present (Figure 2.6). In RM518, RM73, and RM523 the transposons were
inserted in the rrn operon represented by fragment 2 as this band had been shifted
due to the presence of the transposon (Figure 2.6). The question then arose as to
whether an insert in the third operon would also be induced by low temperature.
When a vector containing a fragment of the 16S rrn gene fused to the luxAB genes
was recombined into all three of the rrn operons, the inserts appeared to be
regulated by temperature in a similar fashion (data not shown).

Fusing the 16S rm upstream region to the luxAB genes. Because all of
the transposons were inserted well within the rRNA genes, the luxAB transcript
was joined to a significant amount of either the 16S or 23S rRNA in each mutant.

The ribosomal RNA transcript contains a large amount of secondary structure,

50

 

 

<3
1»-

2—> '7 .
3—>

 

 

 

3166 603 _ lkb

 

fragment 1 23S rRNA

518 523 73

23S rRNA

    

fiagment 2

Figure 2.6. Mapping the transposon inserts to the rm operons. (A)
Genomic DNA isolated from the mutant strains and S. meliloti 1021 was
digested with SacI and electrophoresed on an agarose gel. The DNA was
tranferred to a nylon membrane and probed with 32P-labeled 23S rDNA. The
three rm operons are separated onto three fragments, see RM1021. The
presence of Tn5-1062 within an operon is shown by the increase in size of a
band relative to its position in the wild type sample, RM1021. (B) Maps of two
of the three rm operons are shown. Based on sequence comparison and the
Southern blot analysis (A), the relative positions of the different transposon
insertion sites are depicted.

51

 

undergoes significant processing and interacts with the ribosomal proteins causing
concern that even the partial rRNA transcript in each mutant strain may adversely
affect the expression of the luxAB genes. Therefore, the plasmid, pAGl.3, was
constructed so that the luxAB genes would not be preceded by any structural
rRNA. Based on the location of the transposon within the 168 rRNA gene in
RM3166 as compared to the E. coli 16S rRNA gene, the location of the mature
end of the 16S rRN A gene was predicted to be approximately 1 kb upstream of the
transposon insert. In E. coli, two tandem rm promoters are located 190 bp from
the 5’ end of the 16S rRNA gene (Srivastava and Schlessinger, 1990). After
allowing for this leader region, the BglII-Sacl fragment was selected for cloning as
the rm promoter was predicted to be within that region (Figure 2.7A). The
plasmid, pAGl.3, was constructed by cloning the BglII-Sacl fragment from the
16S rm upstream region in front of the luxAB genes on the broad host range
plasmid pRK290 (Figure 2.8). The plasrrrid, pLUX, was constructed to serve as a
negative control and is identical to pAGl.3 except for not containing the Bng-
SacI fragment (Figure 2.7A).

The two plasmids were moved into the wild type strain, RM1021, and their
luminescence compared to an original insertion mutant, RM603 (Figure 2.7B).
The transposon in RM603 is inserted in the same 16S rRNA gene as is tagged in
RM3166. Both RM1021/pAGl.3 and RM1021/pLUX emitted measurable light

although pLUX contained no exogenous promoter. The luminescence from

52

 

Figure 2.7. Effect of a temperature downshift on the luminescence of the
luxAB transcriptional fusion, pAGl.3. (A) The structures of pAGl.3,
pLUX, and RM603 are shown. A 605 bp BgIII-Sacl fragment containing a
16S rm upstream region was subcloned from RM3166 and used to make a
transcriptional fusion to luxAB (see Figure 2.8). The activity of this fusion was
compared to that of RM603 (16S rrn::Tn5-1062) and pLUX, a plasmid
identical to pAGl.3 but lacking the BgIII-Sacl fragment. (B) GTS medium
containing Km (50 ug/ml) was inoculated 1:1000 and allowed to grow to an
A600 of 0.1-0.2 at 30°C. The cultures were sampled immediately prior to
shifting the culture flasks to a 15°C water bath and 1, 2, 4, and 6 hours
afterward. The aliquots were assayed for luciferase activity as described in
Materials and Methods. Three cultures were grown for each construct and
three aliquots taken from each flask at every time point. The relative light unit
(RLU) measurements were averaged for each flask and divided by the A600 to
correct for culture density. Each bar represents the average of values fiom
three flasks :t the standard deviation.

53

 

 

BglII SacI

 

 

 

- . .-. . ‘ " .. ."/‘ ‘.".l- ""'.""'r' 5". ' .\ I’l' 1. ....-..i... "’..'4'd‘ .u.... a »,t- .q-h'rnre‘
.1; . .V. ,7. ._. A, 1 _ ., . . 7.3 . v . .‘. '_ . . :/
-.'.".l." . . I > ‘ I '
v. . .:. .333: . - : ‘ , r. / ,
hm...“ ii, i l. 'L . . Lam“ . . r’ ‘ ”

RM603 (16S rrn::Tn5-1062)

 

Bglll SacI

 

 

 

 

M ori V KmR
pAGl.3

 

 

 

pLUX Lima oriV KmR

 

 

   

 
 

 

I RM1021/pLUX
I RM1021/pAGl.3
RM603

 

    

 

RLU/Am x 104‘
U.

 

 

’/.r’////////////z7////////////////// ‘

///////////////////////// .

I './//'// - I

'/.////'V///////A .

67/ I

0 1 2 4 6
Hours at 15°C

Figure 2.7

54

Figure 2.8. Construction of pAGl.3 and pLUX. The rm promoter region
directing the transcription of the luxAB genes in strain RM3166 was cloned
using XmaI to form the plasmid, pAGlX. The BamHI fragment containing the
promoter region and a portion of the 16S rRNA gene was then isolated fiom
this plasmid and cloned into the BamHI site of pBluescript SK forming
pAG1X7. The BamHI-Sacl fragment containing the rm promoter region was
isolated and treated with T4 DNA polymerase to form blunt ends. This
fiagment was then ligated to the larger BgIII fragment of pRL1062 that had
been treated with the Klenow fragment of DNA polymerase. The resulting
intermediate plasmid was digested with BgIII and Xhol. This fragment
contained a 605 bp fiagment of genomic DNA in front of the left portion of
Tn5-1062. The plasmid pRK290 was digested with BgIII and SalI and the
largest band isolated. The BgIII-Xhol fragment was ligated to the BglII-Sall
fi'agment of pRK290 to form pAGl.3 which was then moved into S. meliloti
1021 via triparental mating. The plasmid, pLUX, was constructed to serve as a
negative control in the luciferase assays. This plasmid was made by digesting
pRL1062 with BgIII and XhoI and isolating the fragment containing the luxAB
genes. This fiagment was then ligated to the BglII-SalI fragment of pRK290
described above. This plasmid is identical to pAGl .3 but lacks the 605 bp rm
promoter fragment sub-cloned from RM3166. Restriction enzyme sites used in
a cloning step are indicated by an asterisk. The 52 bp inverted repeats at each
end of the transposon are represented by arrowheads.

55

 

 

*BglII X7101

.2
BamH I/BglII It???
Step 1
pRK290 *XhoI
.2.
9:,
*SaII
*Bg I
EcoRI BgIII
Figure 2.8

 

 

56

 

 

RM1021/pLUX is most likely the result of transcription originating from pRK290
sequences serving as a cryptic promoter. The amount of luminescence produced
by RM1021/pAGl.3 at 30°C was approximately 10-fold higher than that produced
by RM603. This may be due to the multiple copies of the plasmid-borne
promoter-luxAB fusion as compared to the single transposon insert in RM603.
While both RM102 l/pAG1.3 and RM603 increased in luminescence to
comparable levels after the temperature downshift, the calculated inductions, 28-
fold and 219-fold respectively, were quite different. The difference in induction is
primarily due to the relatively elevated luminescence of RM 1021/pAGl.3 at 30°C.
While the induction of RM1021/pAGl.3 luminescence is lower than that of
RM603, the 28-fold increase was still higher than the 4.6-fold increase in
luminescence from RM1021/pLUX. The increase in luminescence observed with
RM1021/pLUX is comparable to what was observed when screening the random
Tn5-1062 inserts for induced loci (Figure 2.2). Thus, the BglII-Sacl fragment
contained sequences required for the cold-induced expression of the luxAB genes
in RM 102 1.

Analysis of luxAB transcript levels following a temperature downshift.
The induction of luminescence in the rrn insertion mutants is quite high and varies
significantly between different growth media. For instance, the increase in
luminescence in RM603 when grown in TY broth is approximately 74-fold (Figure

2.3) as compared to the 219-fold induction measured during growth in GTS

57

 

medium (Figure 2.7). To determine whether the increase in luminescence was
paralleled by an increase in luxAB message, RNA was isolated from two rm
insertion mutants and RM1021/pAGl.3, before and after a temperature downshift.
Strains RM518 and RM73 contain Tn5-1062 insertions in a 16S or a 23S rRNA
gene respectively (Figure 2.6). Figure 2.9 shows the result of a Northern blot
probed with 32P-labeled luxAB DNA. In all three strains, levels of the luxAB
transcript increased within an hour at 15°C and were still elevated after 6 hours.
The pattern of message accumulation is different than the pattern of luminescence
at 15°C (compare to Figure 2.7B). While maximum levels of the luxAB transcript

are reached within 1-2 hours, luminescence is still increasing up to 6 hours

following the temperature shift. The Northern blot did confnm, however, that
levels of luxAB transcript increased at 15°C in the insertion mutants and in

RM1021/pAGl.3.

58

 

Hours at 15°C
0 1 2 4 6

 

<- 4.4 kb
RM1021/pAGl.3

(- 2.4 kb

(- 7.5 kl)
RM518 (- 4.4 kb

(16Srrn::Tn5-l062)

' '4 (- 7.5 kb
(- 4.4 kb

 

RM73

(23Srrn::Tn5-l062) g (. 2 4 kb

 

Figure 2.9. Levels of luxAB transcript following a temperature downshift
from 30°C to 15°C. Strains RM1021/pAGl.3, RM518 and RM73 were grown
in TY broth at 30°C until reaching an A600 of 0.2. Aliquots were taken before
shifting the flasks to a 15°C shaking water bath and 1, 2, 4, and 6 hours
following the temperature downshift. RNA was isolated from 5 ml aliquots as
described in Materials and Methods. 2 ug of each RNA sample were separated
on an agarose gel and transferred to a nylon membrane. The membranes were
probed with a 32P-labeled luxAB DNA fragment.

59

Table 2.2. Bacterial strains

 

 

Swing Description Reference or source
Sinorhizobimi meliloti
RM1021 Str“; derivative of SU47 Meade et al, 1982
RMII KmR; RM1021 containing Tn5-1062 O’Connell and
insert in ORF 2 Thomashow, 1999
RM509 KmR; RM1021 containing Tn5-1062 O’Connell and
insert between cspA and ORF 2 Thomashow, 1999
RM73 KmR; RM1021 containing Tn5-1062 This study
insert in the 23S rrnB gene
RM518 Km“; RM1021 containing Tn5-1062 This study
insert in the 16S rrnB gene
RM523 KmR; RM1021 containing Tn5-1062 This study
insert in the 23S rrnB gene
RM603 Km“; RM1021 containing Tn5-1062 This study
insert in the 16S rrnA gene
RM3166 Km“; RM1021 containing Tn5-1062 This study
insert in the 16S rrnA gene
Escheg'chia cgli
DHSa supE44 hstI 7 recAI (hi-l Hanahan, 1983
Alac U169(¢ 801acZAM15)
endAl gyrA96 reIAI

60

 

 

Table 2.3. Plasmids

Pl mis

pRL1062a

pRK2013

pAGlX

pAG1X7

pRK290

pAGl.3

pLUX

Description

NmR/Kmk; vector used in Tn5-1062
mutagenesis and construction of
pAGl .3 and pLUX

KmR; helper plasmid used to mobilize
plasmids into RM1021 strains

Km“; Xmal fragment subcloned fiom

RM3166 containing left half of
Tn5-1062 and upstream sequences

ApR; BamHI fi'agment containing the
rm promoter subcloned fi'om pAGl X
and cloned into BamHI site of
pBluescript SK

Tet“; broad host range (RK2) plasmid

used in constructing pAGl .3 and pLUX.

KmR', 605 bp BgIII/Sacl fragment
containing the rm promoter,
subcloned fi'om pAG1X7, and cloned
in fiont of the luxAB genes in pRK290

KmR; BglII/Xhol fragment of pRL1062

cloned between the SalI and BgIII sites
of pRK290

61

Reference or source

Cohen et al, 1998

Figurski et al, 1979

This study

This study

Ditta et al., 1980

This study

This study

 

 

Discussion

Seven cold-inducible Tn5-1062 insertion mutants were isolated in this
study. Five insertions are within the 16S and 23S rRNA genes of two ribosomal

RNA, rm, operons, while the other two insertions are within an operon containing

 

genes similar to cspA and rpsU of E. coli. All three of the rm operons found in S. r
meliloti 1021 appear to be coordinately regulated, as the luxAB genes recombined

into the third rrn operon also showed cold-inducible luminescence (data not "3
shown). While these results suggest that more ribosomal RNA is required by the E

cells following a cold shock, a previous study of the relationship of ribosome
synthesis and growth rate did not come to this conclusion (Ryals et al., 1982).
The identification of the rrn operons as potential cold-inducible loci in S. meliloti
was, therefore, a novel and unexpected finding.

Following a temperature downshift from 30°C to 15°C, S. meliloti 1021
slows from a generation time of 2.4 hours to 22 hours (Lehmann, 1994). An
increased demand for ribosome synthesis and, therefore, increased transcription of
the rm operons is unexpected in view of this decline in growth rate. Ryals et al.
(1982) examined the amount of stable RNA present in E. coli cells when grown at
different temperatures and found that the amount of stable RNA and, therefore, the
number of ribosomes per cell, remained the same in the temperature range from

40°C to 20°C. This observation suggests that at these temperatures the different

62

 

cellular enzymes are affected similarly such that the translational capacity of the
cell remains proportional to the demand for protein synthesis. A study of a
psychrophilic Pseudomonas demonstrated that if growth rate was held constant,
the RNA content of the cell increased at the lower temperature (Harder and
Veldkamp, 1967). In this case, it was proposed that the cells synthesized more
ribosomes in order to compensate for the decline in translation rate. Synthesizing
more ribosomes would presumably provide the translational capacity required to
maintain the same growth rate. The crucial difference between these two
experiments is most likely the manipulation of the grth rate and not the use of
difi‘erent bacteria. A possible conclusion from these experiments would be that
rrn expression can be regulated by temperature but a net increase in the number of
ribosomes is not necessary if growth rate is allowed to vary as temperature
declines.

Protein synthesis has been shown to be particularly sensitive to low
temperatures in many mesophilic bacteria (Broeze et al., 1978). In E. coli, it is the
initiation of translation that is inhibited below 8°C, resulting in the accumulation
of the ribosomal subunits and cessation of growth (Friedman et al., 1971). In fact,
Jones and Inouye (1996) have proposed the “Cold-Shock Ribosome Adaptation”
model. In this model, the cold shock response is viewed primarily as an adaptive
response to a decline in translational capacity. The cold shock proteins are

thought to adapt the ribosome for optimum function at low temperature. Three of

63

the cold shock proteins in E. coli, translation initiation factor 2, RbfA and CsdA,
are associated directly with the ribosome and its function. RbfA and CsdA have
been shown to be particularly important at low temperature. In this model of the
cold shock response, synthesis of the ribosome is not being induced but a direct
connection between the cold shock response and ribosome function is made.
Another aspect of the cold shock response in E. coli is the drop in level of
guanosine 3’, 5’-bis(pyrophosphate), or ppGpp, irmnediately following a cold
shock (Mackow and Chang, 1983). This nucleotide is involved in regulating the
stringent response (Cashel et al., 1996). Levels of ppGpp increase following a
nutrient downshift, repressing the transcription of rRNA and tRNA and inducing
the expression of biosyntlretic enzymes. In this way, higher levels of ppGpp are
associated with a sudden decline in growth rate and lower levels with a nutrient
shift up or sudden increase in growth rate. Paradoxically, this relationship is
reversed following a temperature downshift when the levels of ppGpp fall even as
growth rate declines. It has been proposed that the levels of ppGpp decrease
following a cold shock in E. coli in response to the sudden inhibition of translation
initiation (Jones et al., 1992a). This inhibition of translation is thought to lead to a
surplus of charged tRNAs. A similar condition would occur following a nutrient
upshift. While the response of ppGpp levels to a temperature downshift has not
been measured in S. meliloti 1021, the response to amino acid starvation is similar

to that observed in E. coli (Howorth and England, 1999). If ppGpp levels decrease

64

following a temperature downshift in S. meliloti as in E. coli, transcription from
the rm operons could potentially increase.

Most studies of cold shock have focused on analyzing changes in protein
synthesis using two-dimensional gel electrophoresis. These studies would not be
able to detect increased synthesis of rRN A. If, however, the increased synthesis of
rRNA were leading to the synthesis of more ribosomes, one would expect to see r
increased synthesis of the ribosomal proteins. In E. coli, several ribosomal

proteins continue to be synthesized following a cold shock and in some cases

 

appear to be induced (Jones et al., 1987; Jones et al., 1996). Although selected
ribosomal proteins have been shown to be cold-inducible in other organisms,
including S. meliloti -- S6 and L7/L12 in Bacillus subtilis (Graumann et al., 1996)
and S21 in Anabaena variabilis (Sato, 1994) and Sinorhizobium meliloti 1021
(O'Connell and Thomashow, 1999) -- the coordinate induction of all the ribosomal
proteins has not been observed. It is, however, possible that a novel mechanism
exists in Sinorhizobium meliloti and that the ribosomal proteins are cold-induced.
While the Tn5-10632 mutagenesis of S. meliloti 1021 identified only one cold-
inducible ribosomal protein, the possible existence of other induced ribosomal
proteins cannot be eliminated (O’Connell and Thomashow, 1999). Because the
regulation of ribosomal protein synthesis in E. coli occurs post-transcriptionally
(Keener and Nomura, 1996), the luminescence of the luxAB genes inserted in a

ribosomal protein gene may not accurately reflect the regulation of that ribosomal

65

protein's synthesis.

Another aspect of cold shock involves its predicted effect on RNA
secondary structure. It has been proposed that secondary structures in RNA affect
translation at low temperature but altered secondary structure may also affect
rRNA processing and consequently ribosome synthesis. Two of the cold shock
proteins in E. coli, CspA and CsdA, are thought to interact with RNA at low
temperature. A role for CspA as a RNA chaperone has been proposed because of
its ability to bind single-stranded RNA and destabilize secondary structures in
RNA (Jiang e1al., 1997). CsdA has been shown to unwind double-stranded RNA
and shows sequence similarity to the DEAD box family of RNA helicases (Jones
et al., 1996). That CspA and CsdA are induced after a cold shock suggests that
mRNA and perhaps rRNA are negatively affected at low temperatures. At low
temperature, RNA molecules may form aberrant secondary structures that could
interfere with their function. In addition, rRNA processing has been shown to be
inhibited at low temperature in the yeast, Saccharomyces cerevisiae (Kondo et al.,
1992). One of the cold shock proteins identified in this organism is NSRl, a
nucleolin-like protein that is involved in ribosomal RNA maturation (Kondo et al.,
1992). Inhibition of rRNA processing would necessarily inhibit ribosome
synthesis. This inhibition may in turn activate a feedback mechanism that could
stimulate transcription from the rm promoters. So there may be two factors

leading to a decline in translational capacity: inhibition of ribosome function and

66

the inability to synthesize more functional ribosomes.

A final consideration involves the isolation of only highly induced loci in
our mutagenesis and screen. This result was unexpected as several moderately
induced loci have been identified in other organisms including Escherichia coli
(Jones et al., 1987), Arthrobacter globiformis (Berger et al., 1996), Lactococcus
lactis (Panoff et al., 1994), and Bacillus subtilis (Graumann et al., 1996). After
extensive screening, only two classes of induced loci were isolated: the rrn
operons and the operon containing homologs of the major cold shock protein CspA
and the ribosomal protein S21. Yet, Cloutier et al. (1992) showed that a strain of
Sinorhizobium meliloti continued to synthesize several proteins following a
temperature downshift to 0°C or 5°C. Although the mutagenesis appeared to be
thorough, the sensitivity of the screen for altered luminescence may have been
limited. The luminescence of the average insertional mutant appeared to be
affected by the temperature shift. This led to a background level of variable
luminescence that made it difficult to identify moderately induced loci that
increased by only 3 to 5-fold.

Further experiments are required to assess the effect of a cold shock on
ribosomal RNA regulation in S. meliloti. The testing of direct promoter fusions is
needed to accurately measure changes in rrn promoter activity in response to
temperature shifts. While the rrn promoter was brought closer to the luxAB genes

in constructing pAGl.3, the exact location of the promoter is still unknown.

67

Intervening sequences between the start of transcription and the luxAB genes may
interfere with the measurement of rrn promoter activity. The construction and

testing of direct promoter-luxAB fusions is described in the following chapter.

68

Materials and Methods

Bacteria and growth conditions. Cultures of Escherichia coli DHSot were
grown aerobically at 37°C in LB broth (Sambrook et al., 1989) or on solid LB
medium containing 1.4% agar. Cultures of Sinorhizobium meliloti were grown
aerobically at 30°C in either tryptone-yeast extract medium, TY, (Beringer, 1974),
or GTS medium (Kiss et al., 1979). Cultures were maintained on solid TY or GTS
medium containing 1.4% agar, or frozen to -80°C in TY broth containing 15%
glycerol. S. meliloti was cultured on solid media containing the following
antibiotics when appropriate: kanamycin, Km (200 ug/ml) and streptomycin, Sm
(250 rig/ml). E. coli was grown on plates containing ampicillin, Ap (100 ug/ml),
Km (100 ug/ml), or tetracycline, Tet (10 ug/ml) when appropriate. A lower
concentration, 50 rig/ml, of Km and Sm was used in broth cultures for both
bacteria.

Transposon mutagenesis of Sinorhizobium meliloti 1021. Wild type S.
meliloti 1021 was subjected to mutagenesis by Tn5-1062 carried on the plasmid
pRL1062 (Figure 2.1) (Cohen et al., 1998). The plasmid pRL1062 was introduced
into S. meliloti 1021 via triparental matings (de Bruijn and Rossbach et al., 1994).
Cells were isolated from overnight cultures of E. coli DHSot carrying the plasmids

pRL1062 and pRK2013 and also from two-day old cultures of S. meliloti 1021.

69

 

These cells were rinsed once and then resuspended in TY broth before being
mixed and spotted on TY agar plates. The plates were incubated at 30°C for 12-16
hours. Individual mating spots were resuspended in a growth medium and
dilutions were spread on selective plates containing Km and Sm and incubated at
30°C. When single colonies had formed (3-5 days), they were transferred to new
plates, grown up at 30°C (2 days), and assayed for luciferase activity under
diflerent conditions. In some cases the colonies were not transferred to fresh
plates before being assayed. Three sets of mutagenesis and screening were
performed. In the first set which resulted in the isolation of RM3166, over 17,000
colonies were screened for increased luminescence after a temperature shift from
30°C to 10°C. A second screening of approximately 5,000 colonies resulted in the
isolation of RM518, RM523 and RM603 by looking at increased luminescence
after a temperature shift from 30°C to 15°C. In the third mutagenesis and
screening, transposon recipients were selected and screened on GTS medium. In
this case approximately 1,000 colonies from eleven matings were examined and
one mutant, RM73, showed significantly increased luminescence at 15°C.

Detection and quantification of bacterial luciferase activity. Luciferase
activity was measured differently depending on the culture conditions used to
grow the bacteria. If the bacteria were grown on a solid agar medium,

luminescence was measured with the Hamamatsu Photonic System model C1966-

20 (Photonic Microscopy, Oak Brook, IL). The Berthold LUMAT luminometer

70

LB 9501 (W allac Inc., Gaithersburg, MD) was used to measure aliquots of liquid
bacterial cultures. Bacteria grown on agar plates were exposed to the enzyme
substrate, N-decyl aldehyde, for two minutes prior to being assayed for light
production for 59 seconds. The aldehyde was spread on a glass petri dish which
was then placed over the agar plate. The Hamamatsu system converts photon
counts to an image displayed on a monitor. This system can also quantitate the
amount of light produced by each colony.

Bacteria grown in broth culture were assayed quantitatively for light
production. In this assay, the N-decyl aldehyde was dispersed in an aqueous
solution of bovine serum albumin (20 mg/ml) at a concentration of 1 Ill/ml. A 5-
10 ul aliquot of the cell culture was added to 50—100 pl of the aldehyde solution,
vortexed for 30 seconds, then assayed for 60 seconds. The measured
luminescence was reported in arbitrary units, RLU, and adjusted for cell number
by dividing this number by the A600 of the culture. All assays were performed at
room temperature.

DNA isolation, cloning and sequencing. Genomic DNA was isolated from
3-5 ml cultures of S. meliloti according to a standard CTAB extraction protocol
(Ausubel et al., 1987) with one modification. Prior to extraction with CTAB, the
bacterial lysate was passed several times through an 18-gauge needle using a
sterile syringe in order to partially shear the DNA. This step reduced sample

viscosity and facilitated extraction of the samples with phenol and chloroform.

71

The Tn5-1062 transposons and their contiguous genomic DNA segments
were cloned out of the transposon recipients and transformed into E. coli DHSa.
Genomic DNA isolated from the mutants was digested with the appropriate
restriction enzyme and circularized. The transposons contain an origin of
replication, p15A, that is recognized by E. coli DHSoi allowing the ligation
mixture to be electroporated directly into E. coli DH50t. The circularized
transposons were selected by spreading on plates containing Km.

Plasmid DNA was isolated using a Qiagen maxi-prep column (Qiagen, Inc.,
Valencia, CA) or a standard alkaline lysis method (Sambrook et al., 1989). DNA
sub-cloned from strain RM3166 was sequenced manually using the T7 Sequenase
v.2 kit according to the recommended protocol (Amersharn Life Science, Inc.,
Cleveland, OH) and examined using standard gel electrophoresis and
autoradiography techniques. DNA subcloned from all other mutants was
sequenced using an automatic fluorescent sequencing method. The clones were
sequenced at the MSU-DOE-PRL DNA Sequencing facility using the ABI Catalyst
800 for Taq cycle sequencing and the 373A Sequencer for the analysis of the
products. Sequences to the left of the transposons were obtained using the primer
MT47 5’-CTAAGCTGCCTCCATCC-3’ which binds within the luxAB gene
cassette. Sequences to the right of the transposon were obtained using the primer
MT84 5’-CACATGGAATATCAG-3’ which is specific for sequences within the

1S5 OR of the transposon. Both primers were synthesized by the Macromolecular

72

Structure Facility located in the Department of Biochemistry at Michigan State
University.

Southern blotting. DNA fragments were separated on an agarose gel and
transferred to a nylon membrane in 20X SSC via capillary action according to the
standard protocol (Sambrook et al., 1989) with two modifications. Instead of the
acid depurination step, the gel was first exposed to UV light for 10 rrrinutes in the
UV Stratalinker 2400 (Stratagene, La J olla, CA). After transfer, the DNA was
crosslinked to the nylon membrane by exposure to 20 mjoules of light energy in
the UV Stratalinker. The blot was hybridized to a 32P-labeled probe following a
protocol based on the one described in Sarnbrook et al., 1989. The pre-
hybridization and hybridization solutions contained 6X SSC, 0.5% SDS, and
0.25% nonfat dry milk. Hybridization was performed at 60-65°C in a Robbins
Hybridization Incubator Model 400 (Robbins Scientific Corp., Sunnyvale, CA).
The probe was prepared from a band isolated fragment by a random priming
method (F einberg and Vogelstein, 1983). The blots were washed four times in a
solution of 0. 1X SSC and 0.1% SDS at 60-65°C. The blot was then exposed to
film.

DNA sequence analysis. The DNA contiguous with the transposons was
sequenced by priming out of the ends of the transposons. The resulting sequences
were joined at the nine base pair direct repeat created during transposition and

compared to the non-redundant database including GenBank, EMBL, DDBJ, and

73

01

qu

tie

PDB sequences using the BLASTN algorithm (Altschul et al., 1997).

Isolation of RNA. RNA was isolated from S. meliloti strains according to a
modified protocol derived from that found in Ausubel et al. (1987) for the
isolation of RNA from gram negative bacteria. Bacteria were harvested from a 3-
20 ml aliquot of cell culture to which rifarnpicin had been added to a final
concentration of 0.2 mg/ml. The cell pellets were resuspended in 1.4 ml of
protoplasting buffer and 80 ul of 50 mg/ml of lysozyme added. The tubes were
incubated on ice 5-10 minutes. The protoplasts were collected by centrifugation
for 2 nrinutes at 7,000 rpm. The pellets were resuspended in 0.5 ml of gram
negative lysing buffer, mixed with 15 pl of diethylpyrocarbonate, DEPC, and

incubated for 5 minutes at 37°C. Tubes were chilled briefly on ice, mixed with

250 pl of 5M NaCl, incubated 10 minutes on ice and centrifuged for 10 nrinutes.
The supernatant was removed, added to a tube containing 1 ml ethanol and stored
overnight at -20°C. The RNA was collected by centrifuging for 15 minutes at 4°C,
dried under vacuum and resuspended in DEPC-treated water. Samples were
quantified using a spectrophotometer and measuring the A260 and A230.

Northern blots. RNA samples were treated with formaldehyde and
electrophoresed on formaldehyde/agarose gels according to standard protocols
(Sambrook et al., 1989). RNA was transferred from the gel to a nylon membrane
in 10X or 20X SSC via capillary action according to the standard protocol

(Sambrook et al., 1989) with one modification. After transfer, the RNA was

74

crosslinked to the membrane by exposure to 20 mjoules of light energy in the UV
Stratalinker 2400 (Stratagene, La J olla, CA). The membranes were pre-hybridized
at 42°C in the following solution: 50% forrnamide, 5X SSC, 50mM potassium
phosphate, 5X Denhardt’s solution, 0.5% sodium dodecyl sulfate, and lOOug/ml
denatured DNA. After a minimum of 2 hours of pre-hybridization treatment, 3‘ZP-
labeled probe was added to the solution and incubated overnight at 42°C in a
Robbins Hybridization Incubator Model 400 (Robbins Scientific Corp., Sunnyvale,
CA). The probe was prepared from a band isolated fragment by a random priming
method (F einberg and Vogelstein, 1983). The membranes were treated first at
room temperature with three washes of 2X SSC and 0.5% SDS. After one wash in
0.1X SSC and 0.5% SDS, the temperature was raised to 50°C for two more washes
in the same solution. Washes were continued until the spent wash solution was no

longer detectably radioactive.

75

Literature Cited

Altschul, S. F., T. L. Madden, A. A. Schiiffer, J. Zhang, Z. Zhang, W. Miller,
and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of
protein database search programs. Nucleic Acids Res. 25: 3389-3402.

Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A.
Smith, and K. Struhl (ed.). 1987. Current protocols in molecular biology. John
Wiley and Sons, NY.

Berger, F., N. Morellet, F. Menu, and P. Potier. 1996. Cold shock and cold
acclimation proteins in the psychrotrophic bacterium Arthrobacter globiformis
8155. J. Bacteriol. 178(11):2999-3007.

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

Broeze, R. J., C. J. Solomon, and D. H. Pope. 1978. Effects of low temperature
on in vivo and in vitro protein synthesis in Escherichia coli and Pseudomonas

fluorescens. J. Bacteriol. 134(3):861-874.

Cashel, M., D. R. Gentry, V. J. Hernandez, and D. Vinella. 1996. The
stringent response, p. 1458-1496. In F. C. Neidhart, R. Curtiss 111, J. L. Ingraharn,
E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznifltoff, M. Riley, M.
Schaechter and H. E. Umbarger (eds), Escherichia coli and Salmonella: cellular
and molecular biology, 2nd edition. ASM Press, Washington, D. C.

Cloutier, J., D. Prévost, P. Nadeau, and H. Antoun. 1992. Heat and cold shock
protein synthesis in arctic and temperate strains of rhizobia. Appl. Environ.
Microbiol. 58(9):2846-2853.

Cohen, M. F ., J. C. Meeks, Y. A. Cal, and C. P. Wolk. 1998. Transposon
mutagenesis of heterocyst-forming filamentous cyanobacteria. Meth. Enzymol.
297 :3-17.

de Bruijn, F. J. and S. Rossbach. 1994. Transposon mutagenesis, p. 387-405.
In P. Gerhardt, R. G. e. Murray, W. A. Wood and N. R. Krieg (ed.), Methods for
general and molecular bacteriology. American Society for Microbiology,
Washington, DC.

76

Ditta, G., S. Stanfield, D. Corbin, and D. R. Helinski. 1980. Broad host range
DNA cloning system for Gram-negative bacteria: construction of a gene bank of
Rhizobium meliloti. J. Bacteriol. 77(12):7347-7351.

F einberg, A. P., and B. Vogelstein. 1983. A technique for radiolabeling DNA
restriction endonuclease fragments to high specific activity. Anal. Biochem.
132(1):6-13.

Figurski, D., and D. Helinski. 1979. Replication of an origin containing
derivative of plasmid RK2 dependent on a plasmid fimction provided in trans.
Proc. Natl. Acad. Sci. USA. 76: 1648-1652.

Friedman, H., P. Lu, and A. Rich. 1971. Temperature control of initiation of
protein synthesis in Escherichia coli. J. Mol. Biol. 61: 105-121.

Goldstein, J., N. S. Pollitt, and M. Inouye. 1990. Major cold shock protein of
Escherichia coli. Proc. Natl. Acad. Sci. USA. 87:283-287.

Graumann, P., K. Schriider, R. Schmid, and M. A. Marahiel. 1996. Cold
shock stress-induced proteins in Bacillus subtilis. J. Bacteriol. 178(15):4611-
4619.

Hanahan, D. 1983. Studies on the transformation of Escherichia coli with
plasmids. J. Mol. Biol. 66:557-580.

Harder, W. and H. Veldkamp. 1967. A continuous culture study of an
obligately psychrophilic Pseudomonas species. Achiv. fur Mikro. 59:123-130.

Honeycutt, R. J., M. McClelland, and B. W. S. Sobral. 1993. Physical map of
the genome of Rhizobium meliloti 1021. J. Bacteriol. 175:6945-6952.

Howorth, S. M. and R. R. England. 1999. Accumulation of ppGpp in symbiotic
and free-living nitrogen-fixing bacteria following amino acid starvation. Arch.
Microbiol. 171(2): 131-134.

Jiang, W., Y. Hon, and M. Inouye. 1997 . CspA, the major cold-shock protein of
Escherichia coli, is an RNA chaperone. J. Biol. Chem. 272(1): 196-202.

Jones, P. G. and M. Inouye. 1994. The cold-shock response-a hot topic. Mol.
Microbiol. 11(5):811-818.

77

Jones, P. G. and M. Inouye. 1996. RbfA, a 30S ribosomal binding factor, is a
cold-shock protein whose absence triggers the cold-shock response. Mol.
Microbiol. 21(6):1207-1218.

Jones, P. G., R. A. VanBogelen, and F. C. Neidhart. 1987. Induction of
proteins in response to low temperature in Escherichia coli. J. Bacteriol.
169(5):2092-2095.

Jones, P. G., M. Cashel, G. Glaser, and F. C. Neidhart. 1992a. Function of a
relaxed-like state following temperature downshifts in Escherichia coli. J.
Bacteriol. 174(12):3903-3914.

Jones, P. G., R. Krah, S. R. Tafuri, and A. P. Wolffe. 1992b. DNA gyrase,
CS7.4, and the cold shock response in Escherichia coli. J. Bacteriol.
174(18):5798-5802.

Jones, P. G., M. Mitta, Y. Kim, W. Jiang, and M. Inouye. 1996. Cold shock
induces a major ribosomal-associated protein that unwinds double-stranded RNA
in Escherichia coli. Proc. Natl. Acad. Sci. USA 93:76-80.

Kandror, O. and A. L. Goldberg. 1997. Trigger factor is induced upon cold
shock and enhances viability of Escherichia coli at low temperatures. Proc. Natl.
Acad. Sci. USA. 94:4978-4981.

Keener, J. and M. Nomura. 1996. Regulation of ribosome synthesis, p. 1417-
1431. In F. C. Neidhart, R. Curtiss 111, J. L. lngraharn, E. C. C. Lin, K. B. Low, B.
Magasanik, W. S. Reznifltoff, M. Riley, M. Schaechter and H. E. Umbarger (eds.),
Escherichia coli and Salmonella: cellular and molecular biology, 2nd edition.
ASM Press, Washington, D. C.

Kiss, G. B., E. Vincze, Z. Kalman,T. Forrai, and A. Kondorosi. 1979. Genetic
and biochemical analysis of mutants affected in nitrate reduction in Rhizobium
meliloti nodule bacteria, nitrogen fixation. J. Gen. Microbiol. 113: 105-1 18.

Kondo, K., L. R. Z. Kowalski, and M. Inouye. 1992. Cold shock induction of
yeast NSRl protein and its role in pre-rRNA processing. J. Biol. Chem.
267(23): 16259-16265.

La Teana, A., A. Brandi, M. F alconi, R. Spurio, C. L. Pon, and C. O.
Gualerzi. 1991. Identification of a cold shock transcriptional enhancer of the
Escherichia coli gene encoding nucleoid protein H-NS. Proc. Natl. Acad. Sci.
USA. 88:10907-10911.

78

Lee, S. J., A. Xie, W. Jiang, J. Etchegaray, P. G. Jones, and M. Inouye. 1994.
F arnily of the major cold-shock protein, CspA (CS7.4), of Escherichia coli, whose
members show a high sequence similarity with the eukaryotic Y-box binding
proteins. Mol. Microbiol. 11(5):833-839.

Lehmann, M. D. 1994. Studies on cold-regulated gene expression in Rhizobium
meliloti. Master's thesis, Michigan State University.

Lelivelt, M. J., and T. H. Kawula. 1995. Hsc66, an Hsp70 homolog in
Escherichia coli, is induced by cold shock but not by heat shock. J. Bacteriol.
177(17):4900-4907.

Mackow, E. R. and F. N. Chang. 1983. Correlation between RNA synthesis and
ppGpp content in Escherichia coli during temperature shifts. Mol. Gen. Genet.
19225-9.

Meade, H. M., S. R. Long, G. B. Ruvkun, S. E. Brown, and F. M. Ausubel.
1982. Physical and genetic characterization of symbiotic and auxotrophic mutants
of Rhizobium meliloti induced by transposon Tn5 mutagenesis. J. Bacteriol.

149: 114-122.

Meighen, E. A. 1991. Molecular biology of bacterial bioluminescence.
Microbiol. Rev. 55(1): 123-142.

Michel, V., I. Lehoux, G. Depret, P. Anglade, J. Labadie, and M. Hebraud.
1997. The cold shock response of the psychrotrophic bacterium Pseudomonas
fragi involves four low-molecular-mass nucleic acid-binding proteins. J.
Bacteriol. 179(23):7331-7342.

Milcamps, A., D. M. Ragatz, P. Lim, K. A. Berger, and F. J. de Bruijn. 1998.
Isolation of carbon- and nitrogen-deprivation-induced loci of Sinorhizobium
meliloti 1021 by Tn5-luxAB mutagenesis. Microbiol. 144:3205-3218.

Nakashima, K., K. Kanamaru, T. Mizuno, and K. Horikoshi. 1996. A novel
member of the cspA family of genes that is induced by cold shock in Escherichia
coli. J. Bacteriol. 178(10):2994-2997.

O’Connell, K. P. and M. F. Thomashow. 1999. Transcriptional organization
and regulation of a polycistronic cold shock operon in Sinorhizobium meliloti
RM1021 that includes the major cold shock gene cspA and a homolog of
ribosomal protein gene rps U. (in preparation)

79

Olfm
1999.
report

Panof
shocl

RyalsL
synihl
Sambi
laboral
Spfing
I
Sato,’I
bindin

SfiVas
- I
coh,p
Sdfles'
I
Cwfluh

ltnlh
and Col

“dngl
Chitl
181(5;

Pfiflthi

llistot

 

O'Connell, K. P., A. M. Gustafson, M. D. Lehmann, and M. F. Thomashow.
1999. Identification of cold shock loci in Sinorhizobium meliloti using a luxAB
reporter transposon. (in preparation)

Panofl', J., S. Legrand, B. Thammavongs, and P. Boutibonnes. 1994. The cold
shock response in Lactococcus lactis Subsp. lactis. Cmr. Microbiol. 29:213-216.

Ryals, J., R. Little, and H. Bremer. 1982. Temperature dependence of RNA
synthesis parameters in Escherichia coli. J. Bacteriol. 151(2):879-887.

Sarnbrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a
laboratory manual, 2nd edition. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, NY.

Sato, N. 1994. A cold-regulated cyanobacterial gene cluster encodes RNA-
binding protein and ribosomal protein S21. Plant Mol. Biol. 24:819-823.

Srivastava, A. K. and D. Schlessinger. 1990. rRNA processing in Escherichia
coli, p. 426-434. In W. E. Hill, A. Dahlberg, R. A. Garrett, P. B. Moore, D.
Schlessinger and J. R. Warner (ed.), The ribosome: structure, function, and
evolution. American Society for Microbiology, Washington, DC.

VanBogelen, R. A. and F. C. Neidhardt. 1990. Ribosomes as sensors of heat
and cold shock in Escherichia coli. Proc. Natl. Acad. Sci. USA. 87 :5589-5593.

Wang, N., K. Yamanaka, and M. Inouye. 1999. Cspl, the ninth member of the
CspA family of Escherichia coli, is induced upon cold shock. J. Bacteriol.

181(5): 1603-1609.

Whyte, L. G., and W. E. lnniss. 1992. Cold shock proteins and cold acclimation
proteins in a psychrotrophic bacterium. Can. J. Microbiol. 38: 1281-1285.

Wistow, G. 1990. Cold shock and DNA binding. Nature 344:823-824.

80

CHAPTER 3

Characterization of the Sinorhizobium meliloti 1021 rrn promoters

Summary

The promoter regions of the three rrn operons of S. meliloti 1021 were
isolated from a genomic clone bank and sequenced. The three promoter regions
share a large section of identity beginning with the 16S structural gene and
continuing 408 bp 5’ of the predicted mature end of the 16S rRN A. The start of
transcription was determined by primer extension and ribonuclease protection
assays. A single promoter was located 354 bp upstream of the predicted fully
processed ends of the three 16S rRNA transcripts. These promoters are very‘
similar in structure to the rm Pl promoters in Escherichia coli. Two promoter
fragments were fused to the luxAB genes and recombined into the S. meliloti
genome in order to study the regulation of the rrn promoters. Fragment A began
with the +1 nucleotide and contained only the promoter and upstream regions.
Fragment B contained 172 bp of the 5’ end of the rrn transcript in addition to the
rrn promoter. The 172 bp region included in fragment B was primarily

responsible for the increased luminescence after a cold shock. The +1 fusion

81

showed only a 6-fold increase in luminescence compared to the 42-fold increase
observed with the +172 fusion. Examination of the luxAB mRNA levels at 30°C
and 15°C revealed that the greater increase of luminescence by fragment B

resulted from the apparent inhibition of translation at 30°C and the increased
stability of the luxAB transcript in this rrn-luxAB fusion. The rrn promoter
fragments were also cloned in front of the uidA reporter gene. The rrn-uidA
fusions showed a transient accumulation of reporter gene transcript following a
temperature downshift. In addition, the 5’ end of the rrn transcript included in

fragment B was demonstrated to have antiterrnination activity.

82

Introduction

The rate of protein synthesis in bacteria is primarily determined by the
number of ribosomes per cell. Since ribosome synthesis requires the coordinate
synthesis and assembly of over fifty proteins and three ribosomal RNA molecules,
it represents a significant investment of cellular resources. Consequently, many
overlapping mechanisms appear to be in place such that ribosome number is
tightly correlated with cellular growth rate under most conditions. The cellular
growth rate is in turn primarily influenced by the nutrient content and temperature
of the growth medium.

The regulation of ribosome synthesis has been extensively studied in
Escherichia coli (for reviews Keener and Nomura, 1996; Gourse et al. , 1996).
The expression of the rm operons has been identified as the primary target for
regulation. Each operon normally produces a single transcript containing the 16S,
23S and 5S rRNAs which is then processed into the individual rRNAs. The
competition for binding of the ribosomal proteins by the newly transcribed rRNAs
and the mRNAs encoding the ribosomal proteins coordinates the synthesis of
rRNA with the synthesis of the ribosomal proteins. If no rRNA is present to bind
to the ribosomal proteins, then the ribosomal proteins bind to their own mRN As
and prevent their translation. While the expression of the rrn operons involves the

initiation of transcription, transcriptional antitermination, and the processing of the

83

 

 

 

rRNA transcript, the initiation of transcription is thought to be the key regulatory
target.

As stated above, the synthesis of ribosomes and consequently the activity of
the rrn promoters is positively correlated with the growth rate of the cell. Studies
of the relationship of growth rate to ribosomal synthesis have focused on
variations in the nutritional content of the growth medium. Three different

regulatory mechanisms have been proposed to interact with the rrn promoters

 

under different situations including FIS-dependent activation, grth rate-
dependent regulation and stringent control. Upstream of the rrn promoters in E. E
coli, there are three binding sites for the FIS protein (Ross et al., 1990). The

binding of FIS to the rrn promoters contributes significantly to the activation of

these promoters. The significance of FIS binding has been noted particularly in

relation to the different growth phases of a bacterial culture. The most rapid

growth occurs during exponential phase and then declines as the medium is

exhausted and the culture enters stationary phase. This cycle continues if the

stationary phase cells are introduced into fresh medium and resume rapid growth.

FIS protein is implicated in the positive regulation of the rrn promoters because its

concentration is highest during the stages of rapid growth and then declines during

slow growth. Appleman et al. (1998) have shown that the concentration of FIS in

the cell is directly related to its binding of the rrn promoters so that only at high

cellular concentrations are the FIS binding sites fully occupied. The H-NS protein

84

has been shown to counteract the FIS-dependent activation of the rrn P1 promoters
in in vitro and in vivo experiments (Tippner et al., 1994; Afflerbach et al., 1998).
The concentration of the H-NS protein was shown to increase by approximately 5-
fold as the culture slowed from exponential growth and entered stationary phase
supporting the idea that H-NS is important in reducing expression of the rrn P1
promoter during the transition into stationary phase (Afflerbach et al. , 1998).
While an earlier study found no effect of H-NS on rrn expression during this stage
of growth (Aviv er al., 1996), Afflerbach ei al. (1998) suggested that the different
strains were not maintained at the same growth rate thus preventing an accurate
assessment of the effect of H-NS.

Growth rate-dependent control involves the regulation of the rrn promoters
in response to different growth rates achieved in media of different nutrient
contents. This type of control requires only the core rrn promoter (-41 to +1) and
is thought to involve the cellular ATP and GTP concentrations (Gaal et al., 1997).
The levels of ATP and GTP are positively correlated with the cellular growth rate
and either ATP or GTP is the first nucleotide incorporated into all of the rrn
transcripts. The concentration of the initiating nucleotide greatly affects the
stability of the RNA polymerase-rm promoter complex so that stability is
enhanced at higher concentrations resulting in a higher rate of transcription.

Levels of the nucleotide guanosine 3’, 5’-bis(pyrophosphate), or ppGpp, also vary

with growth rate and were initially thought to mediate growth rate control.

85

However, an E. coli strain completely lacking ppGpp still exhibited growth rate-
dependent control (Gaal and Gourse, 1990).

The stringent response refers to the cellular response to starvation for anrino
acids (see review Cashel et al., 1996). This response is characterized by a
dramatic increase in ppGpp levels and the simultaneous activation or repression of
various genes. The levels of ppGpp also respond to variations in other nutrients
but ppGpp has not always been shown to be the key regulator in these responses.
In response to starvation for anrino acids, ppGpp acts a negative regulator of rRNA
and tRNA transcription and a positive regulator of certain biosynthetic genes.

The regulation of rrn operon expression in response to temperature is not
fully understood. Ryals et al. (1982) found that stable RNA content and synthesis
remained constant when cultures of E. coli were grown at temperatures between
20°C and 40°C. The levels of ppGpp, however, were found to vary inversely with
temperature. Mackow and Chang (1983) also found that ppGpp levels decreased
in E. coli following temperature downshifts from 40° to 23°C; however, they
observed that the RNA synthesis rates changed as well. RNA synthesis fell
initially after the temperature downshift only to increase again as the ppGpp levels
continued to decrease. The inverse relationship between ppGpp and RNA
synthesis was supported in this study while the absolute amounts of ppGpp
required to affect the rRNA regulation changed with temperature. Another

interesting finding concerns the relationship between the stringent response and

86

 

cold shock response in E. coli. Jones et al. (1992) found that the two responses
were antagonistic and that mutants lacking ppGpp were able to more quickly adapt
to low temperature. While this study did not examine rrn synthesis directly, many
of the proteins up-regulated at low temperature are involved in transcription and
translation and may be regulated similarly to the rrn operons. Thus, the nucleotide
ppGpp may be important to the temperature regulation of the rm operons although
more precise studies must be conducted to determine the contributions of other
regulatory systems.

As described in Chapter 2, we performed transposon mutagenesis of S.
meliloti 1021 using the reporter transposon Tn5-1062 which carries the
promoterless luxAB genes encoding Vibrio harveyi luciferase. Seven cold-induced
insertion mutants were isolated in the study, including five mutants with insertions
in ribosomal RNA genes. While these insertions were distributed between only
two of the three rrn operons present in S. meliloti 1021, recombination of the
luxAB genes into the third 16S rRNA gene revealed that the third rm operon
showed temperature-dependent expression of the luxAB genes as well.

Preliminary characterization of the rrnA promoter indicated that a 605 bp BglII-
SacI fragment cloned in fiont of the luxAB genes led to temperature-dependent
luminescence and accumulation of the lwcAB transcript.

We report here the further characterization and sequencing of the rrn

promoters and the role of temperature in their regulation. The three rrn promoters

87

 

 

present in S. meliloti 1021 were isolated from a cosnrid clone bank and sequenced.
The transcriptional start sites were mapped indicating the presence of a single
promoter controlling the expression of each rrn operon. Two rrn promoter
fragments were fused to the luxAB and uidA reporter genes, recombined into the
inositol locus of S. meliloti 1021, and their regulation in response to a temperature
downshift examined. The first 172 bp of the rRNA transcript appeared to be
primarily responsible for the increase in luminescence observed in the original
rrn::Tn5-1062 insertions. Luminescence in the rrn promoter +1 fusion to luxAB
increased only 4 to 6-fold following a temperature downshift from 30°C to 15°C.
The greater increase in luminescence by the rrn::Tn5-1062 insertions and the rrn
promoter +172 fusion to luxAB resulted from the apparent inhibition of translation
in these strains at 30°C. The 5’ end of the rrn transcript also increased stability of
the luxAB transcript following the temperature downshift. The rrn promoter
fusions to the uidA gene showed a transient increase in uidA transcript following a
temperature downshift. Further, the rrn promoter +172 fusion to the uidA gene

indicated that the 5’ end of the rrn transcript has antitermination activity.

88

 

Results

DNA sequence analysis of the three rrn promoter regions in S. meliloti
1021. The promoter regions of the three rrn operons of S. meliloti 1021 were
isolated in order to study their regulation. A S. meliloti 1021 cosmid clone bank
was probed with MT119, an oligomer complimentary to a central region of the
16S rRNA gene. The three rrn promoter regions were isolated on three separate
cosrrrids. The sequence 5' to that encoding the mature l6S rRNA was determined
for each of the three operons (Figure 3.1). Two operons, rrnB and rrnC, were
identical over the first 441 bp of this “leader” region while the sequence of rrnA
diverged after 392 bp. The designations rrnA, rrnB, and rrnC, refer to those
operons isolated on fragments l, 2, and 3, respectively (Figure 2.6). Because the
luxAB genes behave similarly in response to a temperature downshift when
recombined into each of the three operons, it seems likely that the regulatory
elements required for this response are present within the region shared by all
three rm operons.

Transcriptional mapping of the rrn promoters. In order to precisely
locate the rrn promoters, the 5' ends of the rrn transcripts were identified using
primer extension analysis. The primer MT179 (Figure 3.1) was chosen for this
assay because it is complementary to sequences downstream of the BglII-Sacl

fragment cloned into pAGl.3 (Figure 2.7). This fragment was able to support

89

Figure 3.1. Sequence comparison of the three Sinorhizobium meliloti 1021 rm
promoter regions. The three rm promoters were isolated fiom a clone bank and
sequenced. Dashes indicate sequence identity to rrnB. The promoter regions of
rrn A, B, and C, refer to the rm operons isolated on fi'agments 1, 2, and 3,
respectively (Figure 2.6). The putative -35 and -10 hexamers are indicated and the
lower case letters represent bases that differ from the E. coli rm Pl consensus
sequences (Keener and Nomura, 1996): (-35) TTGTc/gA and (-10) TATAAT.
The +1 identifies the 5' end of the pre-16S rRNA as determined by primer
extension. Sequences complementary to the primers used in the primer extension
assays, MT179 and MT261, are underlined. The 5' end of the mature 16S rRNA
was predicted based on sequence identity to R sphaeroides (Dryden and Kaplan,
1993)

90

I ‘9 3.: nah]

 

Huuuw
hunk”
Huuur

wanna
Hunk”
uuuur

huuuw
Huuxu
unuur

wwwnmoommoooaaamacromooomwmomaonom>>>>mm>amoomemoomoowmoaeeoomoeoomooaaaoomeamo»
wmmaoaowaooooemwemomnnowewonnomowwwmomo>oooma>aamma>mo>momaamoromanaaaaomomaooaa
weennameom>eamaowemwwmamwmmmomwwwaawamawaonwnwmmoamwmmmwmnmmnenmwwwewmwwanemoowo
rum
em>>H>wwo>ooaaaw>>>>n>man>oaa>omwaaaaoemoamwaaeaaamwww>aomaamaaesmnemaammwomooam
ammwweemoo tttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttt
oomowww>nooeooeaaaoommmoaaaoaowmwww>oamao>ea ttttttttt mtwmoo ttttttttttttttttttttt
Lo +—
mamaomwmwwmnoomwaownameno>mmmomoommomnamonomomwoowemaooaaomnaoaemmmaaeooewmweewm

2.33

meow

7%Huqo

m. 25 on :8 38:8 3m :95
953Genoano»:0E3agnoaomfiaegooogeoeéamgaanag

Ema—d u.—

91

temperature-dependent expression of the luxAB genes in S. meliloti 1021 (Figure
2.7). The primer extension assay of the rm transcripts isolated from RM1021 at
30°C gave two major products of 100 bp and 272 bp (Figure 3.2A, lane 1). These
products were also present when RNA isolated from RM1021 at 15°C was used
(data not shown). These results indicated that there might be two promoters
driving the transcription of the rm genes (Figure 3.28).

Pausing of reverse transcriptase sometimes produces prematurely
terminated extension products that can be mistaken for transcriptional start sites.
Consequently, ribonuclease protection assays were performed to determine
whether either of the two extension products resulted from premature termination.
The ribonuclease protection assay identifies transcriptional start sites without the
use of reverse transcriptase. A strain carrying the rrnA promoter region fused to
the luxAB genes was constructed so that a single rm promoter could be analyzed.
The rrn-luxAB fusion produces a transcript that can be distinguished from those
produced by the three endogenous rrn promoters. The resulting strain, RM311,
(Figure 3.3A) contains a fragment of the rrnA promoter region beginning with the
+1 nucleotide of P2 and extending to the BglII site used in the construction of
pAGl.3. This fragment was cloned in front of the luxAB genes in the plasmid
pMW193 (Bosworth et al., 1994) and recombined into the inositol locus of

RM1021 (Figure 3.5) such that the fusion was present in single copy.

92

 

*

meowoa»

 

T*

t
1
63636)

B
P1 P2

I I IA Mature 16S rRNA

Figure 3.2. Primer extension assay of the rm transcripts using MT179. (A)
RNA was isolated from RM1021 grown at 30°C. 20 pg of RNA was added to
500 fmoles of MT179 (see Figure 3.1 for primer sequence) and annealed
overnight at 65°C as described in Materials and Methods. The primer extension
products (lane 1) were compared to a sequence ladder generated fi'om MT179
priming off of pAGlX7. The 5’ ends of the rRNA transcripts are shown in bold
with an asterisk. (B) Schematic of the rm promoter based on the primer
extension products generated with MT 179.

 

93

 

[E

Figure 3.3. RNase protection assay of the rm promoter regions. (A) A
schematic of the rm promoter-luxAB fusion in RM311 is shown (1). The
transcripts that would be produced by the putative promoters, P1 (2) and P2 (3),
are depicted and the dotted lines show the portion that would be complementary
to the probe. A map of the RNA probe used in the experiment (4) is included
as well. (B) RM311 was grown up in TY broth at 30°C then shifted to 15°C for
six hours before RNA was isolated. lug of RNA was hybridized to 4 x 106
CPM of radiolabeled probe overnight at 55°C. The samples were then treated
with RNase as described in Materials and Methods. In lane 1 are the
radiolabeled RNA standards with sizes in base pairs. Lane 2 contains
radioactive probe hybridized to 1 pg tRNA and treated in parallel with the
experimental sample. Lane 3 contains the RNA probe fragments protected by
the complementary RM311 RNA .

94

A

9“ P1 P2 $36
[ I: —->
(1) RM311 , ll_leB

 

 

 

 

 

 

 

 

 

172 bp
96“
(2) P1 transcript —> 5’ MAE I 3’
(3) P2 transcript —> 5W ....... ' 3’
(411 bp)
\9 a P
9% f" 1”
(4) Probe 3’ l J 5’
rrnP (“JAB
B
(bp) 1 2 3
780
‘ 4... ~ 583 b
530 ; p
400 ‘_ ‘— ~ 411 bp
280
4— ~ 172 bp
155 "
Figure 3.3

95

The single-stranded RNA probe used in this assay to detect transcripts
originating from the rrn promoter-luxAB gene fusion in RM311 was synthesized
from the plasmid, pAG31 1-1 (Table 3.2). This probe (Figure 3.3A) is
complementary to 399 bp of the luxAB transcript and the entire rrn upstream
region in RM311. The probe was hybridized to RNA isolated from RM311 and
then digested with ribonucleases T1 and A. These ribonucleases do not degrade
double-stranded RNA; therefore, any transcript produced that is complementary
to a portion of the 32P-labeled, single-stranded probe will protect the probe from
degradation. The transcript originating from P1 should protect 583 bp of the probe
while the transcript originating from P2 should protect a 411 bp fragment (Figure
3.3A). The transcripts synthesized by the endogenous rm promoters will also
protect a portion of the probe. The 5’ end of the transcripts originating from the
endogenous P1 promoters are complementary to 172 bp of the probe but the
transcripts originating from the endogenous P2 promoters have no region of
complementarity to the probe.

RNA was isolated from RM311 six hours after a temperature shift from
30°C to 15°C. After the RNA was hybridized to the probe and digested with the
ribonucleases, two major bands were detected on the gel (Figure 3.38). The
smallest band of approximately 172 bp represents the region of the probe protected
by the endogenous P1 transcript. The band of approximately 583 bp is the portion

of the probe protected by the P l-luxAB transcript. The slightly larger band is most

96

likely the product of a partial digestion of the P l-luxAB transcript. The probe has
70 bp at its 5’ end that do not hybridize with the transcript. This short segment
may be removed with less efficiency by the ribonucleases. Of primary importance
is the absence of a band in the vicinity of 411 bp indicating the lack of any
transcript originating from P2. These results suggest that P2 is not an active
promoter under the conditions tested and that the 100 bp primer extension product
was most likely the product of pausing and premature termination by reverse
transcriptase.

A second primer extension assay was performed using the primer MT261
(Figure 3.1). The purpose of this experiment was to confirm the +1 nucleotide
determined for the Pl transcript in the earlier experiments. A single major band
was produced in this primer extension assay when RNA isolated from RM1021 at
both 30°C and 15°C was used (Figure 3.4A). This assay confrrmed that adenine
was the first base in the rm transcript with no additional promoters present further
upstream (Figure 348). Based on the above data, we propose that each of the rm
operons in S. meliloti 1021 is controlled by a single promoter.

Expression of the rm promoter-luxAB fusions. To begin to characterize
the regulatory mechanisms acting on the rrn operons in response to low
temperature, two fragments of the promoter region were amplified and cloned in
front of the luxAB genes (Figure 3.5). Fragment A extends from -419 to the +1

nucleotide, thus containing only the promoter and upstream regions. Fragment B

97

T C?"

    

O>QHOT§OH>QO

B
-35 -10 +1
TGTTGanTGTTGGAGGGCTGGGGTCTATAAgCCCGATCACTGA

Figure 3.4. Mapping the 5' ends of the rrn transcripts by primer
extension using MT261. (A) RNA was isolated from RM1021 grown at
30°C. 100 finoles of MT261 (see Figure 3.1 for description of primer) were
added to IOug of RNA and annealed overnight at 60°C as described in
Materials and Methods. The primer extension products were compared to a
sequence ladder primed from MT261 using pAGlX7b as a template. The 5’
end of the rRNA transcript is shown in bold with an asterisk. (B) Sequence of
the rm promoter region. The putative -35 and -10 hexamers are underlined.
Nucleotides not in agreement with the E. coli rm P1 consensus sequences are
in lowercase letters. The E. coli rm Pl consensus: (-35) TTGTc/gA (~10)
TATAAT.

98

Figure 3.5. Construction of rm promoter-11MB fusions. (A) Two
fragments were amplified from the S. meliloti 1021 rm promoter region using
three different primers: MT215, MT213, and MT214. The left primer,
MT215, incorporates the SacI and KpnI restriction sites 5’ to position -419.
The two right primers contain the NotI site. The 3’ end of fragment A begins
with the first transcribed nucleotide and continues 5’ to -419. Fragment B
begins at the 3’end with base +172 of the rm transcript and continues to -419.
The amplified fragments were cloned into pBluescript utilizing the SacI and
NotI sites. The luxAB genes were cloned into the XbaI site downstream of the
fragments. (B) The promoter-luxAB fusions were cloned into the KpnI site of
pMW l 93 between the 0 fragment containing a spectinomycin resistance gene
flanked by the T4 translational and transcriptional terminators and two
tandemly arranged rm terminators (T1T2) (Bosworth et al, 1994). The
resulting plasmids were introduced into RM1021 along with an incompatible
plasmid, pJB251. Single recombinants were obtained by selecting for SpR and
GmR colonies. Double recombinants were isolated by screening for Ino-, SpR
and TetS colonies. RM] 11 and RM311 contain fragment A and fragment B,
respectively, directing transcription of the luxAB genes.

99

A +1 +354

 

 

 

-419 +1
r)
luxAB fragment A
-419 +1 +172

 

l

B

 

 

ii-Efi-Efi-Egi-3;?-513-53-Eli-fi-Efi-EZE-Efi-EZE-Efl (2 I LILIZI
inositol DNA

 

 

 

Pm-quAB

 

 

‘RMlll (-4l9to +1 fusedto luxAB)
‘RM311(-419to +172 fusedto IwcAB)

Figure 3.5

100

extends from -419 to +172 so that it includes the 5’ end of the rm transcript but no
structural 16S rRNA. This fragment encompasses all but 12 bp at the 3’ end of the
rrnA promoter fragment subcloned into plasmid pAGl.3 (Figure 2.7). The
promoter-luxAB fusions were cloned into pMW193 between the omega fragment
and two tandemly arranged rm terminators (Bosworth et al., 1994). The omega
fragment confers spectinomycin resistance and contains T4 transcriptional and
translational terminators. The vector also contains portions of the inositol locus of
S. meliloti allowing the promoter-luxAB fusions to be recombined into the wild
type strain and maintained in single copy in the genomic DNA. The fusions
constructed from fragments A and B were recombined into the inositol locus

forming strains RM1 11 and RM311, respectively.

The luminescence of the fusion strains was compared to that of the original
168 rrnA::Tn5-1062 insertion mutant, RM603, at both 30°C and 15°C (Figure
3.6A). The luminescence of RM3 11 (-419 to +172) closely resembled that of
RM603 at both 30°C and 15°C. The luminescence of RM1 11 (-419 to +1),
however, was significantly greater than that of either RM311 or RM603 at 30°C.
After incubation at 15°C, the difference in luminescence between the three strains
was no longer evident. To more accurately quantitate luminescence, cultures were
grown in TY broth at 30°C and then subjected to a temperature downshift to 15°C

(Figure 3.68). At 30°C, RM1 11 produced approximately 7.8-fold more light than

101

Figure 3.6. Luminescence of rm promoter-MB fusion strains. (A) The
two fusion strains, RM311 and RM1 1 1, were streaked onto TY agar along
with RM603 and grown for 2 days at 30°C. The plates were exposed to
aldehyde and the luminescence monitored with the photonic camera. The
plates were then incubated for approximately 5 1/2 hours at 15°C and re-
exposed to aldehyde. The images recorded by the photonic camera are shown.
(B) The two fusion strains and RM603 were cultured in TY broth at 30°C until
reaching an A600 between 0.4 and 0.5. At this point 5ul aliquots were
measured for luminescence as described in Materials and Methods. The
cultures were then shifted to a 15°C shaking water bath. The measurement
reported for 15°C was taken 3 hours after the temperature shifi. Luminescence
is reported in relative light units or RLU divided by the A600 of the culture to
correct for culture density. Each bar represents the average luminescence fiom
two experiments with a total of four samples per strain. The error bars
represent : the standard deviation. (C) The two fusion strains and RM603
were cultured in TY broth at 30°C until reaching an A600 between 0.4 and 0.5.
At this point, samples were taken and luminescence measured as described in
Materials and Methods. The cultures were then shifted to a 15°C shaking
water bath and luminescence measured 1, 3, 5, and 7 hours after the
temperature shift. Each bar represents the average fold induction determined
in two experiments with a total of five flasks measured for each strain type.
The stande deviation is shown with error bars.

102

 

 

 

 

 

60000
“ 50000
E 40000 I RM603
g 1:: RM311
1: 30000 RMlll
E 20000
10000
0
30°C 15°C
E
E
g I RM603
3 E31 RM311
3 RM111
é
o 1 3 5 7
Hours at 15°C
Figure 3.6

103

RM311 and 54-fold more than RM603. After three hours at 15°C, the difl‘erence
in the luminescence of the strains had decreased to less than 2-fold with RM603
being brighter than RM] 1 1. In Figure 3.6 C, luminescence is expressed in terms
of fold induction after a temperature downshift from 30°C to 15°C. Luminescence
in RM311 and RM603 increased approximately 40-fold as compared to the 6-fold
increase by RM1 11. Since RM311 and RM603 achieve levels of luminescence

comparable to RM 1 11 at 15°C, the difference in induction is primarily due to their

relatively reduced luminescence at 30°C (Figure 3.6 B).

RNA analysis of the rm promoter-luxAB fusions. The reduction in
luminescence at 30°C observed in RM311 and RM603 relative to RM 1 11 may
have resulted from the destabilization of the luxAB mRNA by the 5’ end of the rm
transcript. A similar mechanism has been observed in the regulation of the S.
meliloti 1021 cspA gene (O’Connell and Thomashow, 1999). To determine
whether the luxAB genes were in fact being destabilized in RM311 and RM603 at
30°C, the luxAB transcript and luminescence levels were assayed. RNA was
isolated from RM] 11, RM311, and RM603 grown in TY broth at 30°C and three
hours after being shifted to 15 0C. RNA samples were probed with 32P-labeled
luxAB DNA. Luminescence and luxAB transcript levels at 30°C are compared in
Figure 3.7A. The values measured for each strain are reported relative to those in
RM] 1 1. Surprisingly, the level of luxAB transcript in RM311 and RM603 was not

reduced relative to RM1 11 as the low level of luminescence suggested. While

104

RM1 11 contained two-thirds less luxAB mRNA than RM603, RM1 11 produced 5-
fold more light. As in RM603, the amount of light produced by strain RM311 was
less than expected considering the amount of luxAB mRNA present. The lack of
correlation between the levels of luxAB transcript and luminescence in RM311 and
RM603 suggests that translation of the luxAB genes is inhibited in these strains at
30°C.

As seen in Figure 368, RM] 1 1, RM311 and RM603 reached comparable
levels of luminescence following a temperature downshift. Since the low level of
luminescence at 30°C appeared to result from inhibition of translation, the relief of
this inhibition at 15°C may have accounted for the observed increase in
luminescence. Alternatively, the luxAB transcripts produced by RM311 and
RM603, while continuing to be translated less efficiently at 15°C, may have been
more stable and accumulated to higher levels than those produced by RM] 1 1. In
order to distinguish between these possibilities, the levels of luxAB transcript and
the corresponding levels of luminescence at 15°C were examined. Figure 3.7B
compares the luminescence and luxAB transcript levels present in strains RM1 11,
RM311 and RM603 three hours after a shift to 15°C. All quantities are expressed
relative to strain RM1 11 at 30°C. It can be seen that the luminescence of RM] 11

increased approximately 4-fold at 15°C and that the luminescence of RM311 and

RM603 reached levels comparable to RM] 11 at 15°C (Figure 3.7B, light bars).

105

Figure 3.7. Induction of luminescence and 11MB mRNA in rm promoter-
luxAB fusions after a temperature downshift. The two fusion strains and
RM603 were cultured in TY broth at 30°C until reaching an A600 between 0.4
and 0.5. At this point, samples were taken for RNA isolation and luminescence
measured as described in Materials and Methods. The cultures were then
shifted to a 15°C shaking water bath and sampled 3 hours after the temperature
shift. Panel A is a graph of the measurements taken at 30°C and panel B is a
graph of the measurements taken after 3 hours of incubation at 15°C. The light
bars represent the average luminescence determined in two experiments with a
total of four flasks measured for each strain type. The values expressed are
normalized to the luminescence of RM1 11 at 30°C. Each solid bar represents
the average levels of luxAB mRNA normalized to the levels in RM] 11 at 30°C.
The luxAB mRNA levels were measured from Northern blots using a
phosphoirnager. The error bars show the standard deviations.

106

{:1

Fl

 

 

 

 

 

 

a Luminescence
I mRNA

 

Relative to RM] 11

O

28
24
20
16
12

Relative to RM1 1 l at 30°C

Figure 3.7

below-hut

 

 

 

o-boo

 

RMlll RM311 RM603

 

 

 

 

 

a Luminescence
I mRNA

 

 

 

 

 

 

RMlll RM311 RM603

107

 

The levels of luxAB mRNA and luminescence in RM] 11 remained
consistent following the temperature downshift. As message levels increased
approximately 4-fold at 15°C, luminescence also increased approximately 4-fold.
If the increase in luminescence at 15°C in RM311 and RM603 was solely due to
an increase in luxAB mRNA stability, the ratio of luminescence and transcript
level should have remained constant at each time point. This was not the case,
however, as the level of light production began to approach the level of luxAB
transcript at 15°C in RM311 and RM603. While the ratio of transcript to
luminescence in RM311 at 30°C was approximately 7.3, the ratio was reduced to
1.6 after three hours at 15°C. The same pattern was seen in RM603 as the ratio of
transcript to luminescence was reduced from 16.4 to 2.5 after the shift to 15°C.
The translation of the lwcAB genes in RM311 and RM603 appeared to be
preferentially inhibited at 30°C with the efficiency of translation increasing at
15°C. However, the increased stability of the luxAB message at 15°C in strains
RM311 and RM603 may also have contributed to the increased luminescence in
these strains as the luxAB transcript did accumulate more at 15°C in these two
strains than in RM1 1 1. Evidently, the increased luminescence observed in the
original rrn: :Tn5-1062 strains resulted from the combined effect of multiple
factors including the altered stability and translation of the luxAB transcript
(Figure 3.7; RM603). Since the rrn promoter-luxAB fusion strain RM311 behaves
similarly to RM603, a significant amount of the increased luminescence may be

108

 

caused by the 172 bp sequence located at the 5’ end of the rm transcript.

The induction of luxAB transcript levels was examined more thoroughly in
RM1 11 and RM311 following a temperature downshift from 30°C to 15°C (Figure
3.8). The results confirmed the earlier observation that the presence of the 5’ end
of the rm transcript in RM311 resulted in the greater accumulation of quAB
transcript at 15°C. Although the induction of luxAB mRNA was higher in both
RM311 and RM] 11 in this experiment, the level of luxAB mRNA in RM311 was

approximately 2-fold greater than that in RM] 11 after 3 hours at 15°C. This is the

same ratio seen in Figure 3.78. After 4 hours at 15°C, the levels of luxAB
transcript began to drop in both RM] 11 and RM311 (Figure 3.8), indicating that
the increased accrunulation of luxAB transcript following a temperature downshift
may be transient. Increased promoter activity appeared to contribute to the
increase in luxAB expression in RM311 as luxAB transcript increased 6-fold in
RM1 11 following the temperature downshift. The precise increase in promoter
activity cannot be determined from these experiments, however, as both luxAB
message stability and rrn promoter activity contributed to luxAB expression in
RM] 1 1.

Expression of rm promoter-uidA fusions. Promoter fragments A and B
were also fused to the reporter gene, uidA, to confirm the results obtained with the
rrn promoter-luxAB fusions. The fusions were cloned into the le93 plasmid

and recombined in RM1021 in the same way as the luxAB fusions with one

109

 

18
16 i l
14 1
12 Rm3ll

 

 

 

 

 

 

 

 

lell

 

 

 

 

Fold induction

 

 

 

 

 

 

 

 

 

ere-nae:

 

0 2 3
Hours at 15°C

Figure 3.8. Comparison of MB mRNA levels. The rm promoter-luxAB
fusion strains, RM311 and RM] 1 1, were grown in TY broth supplemented
with Sp. RNA was isolated from aliquots taken at 30°C and 1, 2, 3 and 4
hours following a temperature downshift to 15°C. Northern blots were
performed and probed with a 32P-labeled XbaI fragment containing the luxAB
genes. The Northern blots were quantitated using a phosphoimager and the
fold induction calculated. Each bar represents the average fold induction of 3-
4 flasks from two different experiments. The standard deviation of the fold
induction is indicated by the error bars.

110

 

exception (Figure 3.9A). The promoter-uidA fusions were cloned into the Kpnl
site of pMW193 such that transcription initiating from the rrn promoter proceeds
toward the omega fragment. Fragment A (-419 to +1) is cloned in front of uidA in
RM112 and fragment B (-419 to +172) is cloned in front of uicM in RM312.
Levels of the uidA transcript were examined before and 1, 2, 3, 4, and 8 hours

following a temperature downshift from 30°C to 15°C (Figure 3.98).

The rrn promoter appeared to be transiently induced at 15°C as the levels of
uidA message increased within one hour and then declined by four hours in
RM112. This general pattern was also observed in RM312 although the 5’ end of
the rRNA transcript (+1 to +172) appeared to either stabilize or increase the
expression of the uidA transcript at both 30°C and 15°C. While the induction of
uidA appears to be slightly prolonged in RM312, the effect was still transient.

Interestingly, two fragments with complementarity to uidA were transcribed
in RM312 while only the smaller fragment was produced in RM] 12. Since the
only difference between RM312 and RM112 is the presence of the 5’ end of the
rrn transcript, the larger transcript must result from the transcription of both the
uicM gene and the omega cassette. The omega cassette consists of the
spectinomycin resistance gene flanked by an inverted repeat containing a T4
transcriptional terminator. Apparently, sequences within the first 172 bp of the rrn
transcript allow some transcription to proceed through the first terminator.

Therefore, it appears that the first 172 bp of the rrn transcript has antitermination

lll

 

Figure 3.9. Induction of the rrn-uidA fusions following a temperature
downshift. (A) Schematic of the rrn promoter-uidA fusions is shown.
Promoter fiagments A and B were cloned in front of the uidA gene within
pMW193 in the opposite orientation as the luxAB fusions (see Figure 3.5).
Transcription from the rm promoter proceeds toward the omega cassette. The
inverted repeats containing the T4 transcriptional terminator are depicted by the
shaded boxes. (B) RM112 (-419 to +1) and RM312 (-419 to +172) were grown
in TY broth at 30°C then shifted to 15°C. RNA was isolated from each culture
prior to shifting the cultures to 15°C and then 1, 2, 3, 4, and 8 hours after the
temperature shift. 4pg RNA fi'om each sample was electrophoresed and
transferred to a nylon membrane. The membrane was probed with 3:‘P-labeled
uidA DNA. Two exposures of the RM112 portion of the Northern blot are
shown: 2.5 hours and 30 hours. The blot was subsequently stripped of
radioactive probe and hybridized with 32P-labeled 23S rDNA to check loading.

112

A lkb

{11'er M“ i Q 1

 

 

 

    

 

 

 

 

 

B RM112 RM312
('419 to +1) (419 to +172)
Hours at 15°C Hours at 15°C
0 1 2 3 4 8 0 1 2 3 4 8
2.5 hr - 4,... ,,
””9“" .. 4 4.1 kb
1% <23 kb
0 1 2 3 4 8
30hr
exposure
Figure 3.9

113

activity. In fact, this region contains a putative boxA element and associated
structures similar to those involved with the transcriptional antitermination of the

rm operons in E. coli (Figure 3.10).

114

cw mwmamnn 1mm “Hm uvv 1H0 +H
wwwWOHmanfiaaaaaaaemfimeWGnnamaambnhamHammwmmmoemOQQHOHVH)?mOOOm>annHQWOmwmomnmmommomoamoa

*** * *****

 

111111111111 V
mmomfiomfifimaooH800006062000866002>026a>momOPHQOHQPHHOHO00020HmmanOHHHOWOQHHHHQOHOHHQHHOQ>§H

A 11111111111 wox w vcnmnw<m Unoommmwso mnmpx
manobnmmmwaamaaaoanmanHmaaeaaanvnvmaaowbayawmvhvvvnfinrvvnomammnmmnnmbmnHonnnmmvnnao>>m>mwa
** 4 $** *** ****$**********t ***** **4*** * **

Hamomfianammwb>mwmbhfiammoQQHO>OOHHHHO>02>Q>O>OabbnfionfimfiaaanaommaanmGOOHHOOHHOQOOOHO>HHO>FO

m. man om amncmm Hmm szw
2H0moanemfimaananmanOPHHOEHQOEOmamwaeagmnogamwaamgaenagflnfigfivgggg

ENE.» v.3. .25 $3532. game: en :5 h. 3556a. ~3— §> 35.6... .25 82538 e». :5 2.05054 5min o». 95
12588». 22> £53: 33> 5 302: £55. .25 2.57. 38859 m. «a on :5 3m ~22.» 5 £52: 5 .55 28:9... .25
5» 25:. e». $5 55:8 5253 Beams” 8 won: :5 Recommim m8? e». n5 3m 22> mm 392: 5 ~55 5.55. .25 .551me
39.85 :5 35m ’59 32:5. 8 59553 w. e». :5 3w ~22? .25 u. 3a 2. mamas: w 225m 58 27.22 25
27.2 5 a macs: 23:: n5 uncoommim a»? won—5:8 565138 85 5 28:8. .25 093 massage 5 uses: 5 ~55 «<3.
8535 59325 $53 «SE 5035. 8 En m. new. 583 moo—~38. Dem—5a 382m 5985 595.53 3855a 8 wen:
u man. .eon :3qu cm 583. .25 385552 .3 Ba -_ 0 «£553 25 £52: 5335 o». :5 magnum 255355 5
~55 2:: 851me 59535 $033. 2:: :5 m. 8:. 3.: 3 3.255.. 8:535. .25 .5835 Cd 05:53 5 .0592: 8

25:.

+Nn

+HHA

+moa

+mmn

+wma

115

Table 3.1. Bacterial strains
m Description

Sinorhizobium m_eIiIoti
RM1021 StrR; derivative of SU47

RM603 KmR; RM1021 containing Tn5-1062
insert in the 16S rrnA gene

RM] 11 Ino',SpR; RM1021 containing rm
promoter-luxAB fusion (-419 to +1)
recombined into inositol locus

RM311 Ino',SpR; RM1021 containing rm
promoter-luxAB fusion (-419 to +172)
recombined into inositol locus

RM112 Ino',SpR; RM1021 containing rm
promoter-uidA fusion (-419 to +1)
recombined into inositol locus

RM312 Ino',SpR; RM1021 containing rm
promoter-uidA fusion (-419 to +172)
recombined into inositol locus

 

floherichiggoli
DHSa supE44 hstI 7 recA I thi-I
AlacUl 69(4) SOIacZAMI 5)

endAI gyrA96 relAI

116

Reference or source

Meade et al, 1982

This study

This study :-

This study

 

This study

This study

Hanahan, 1983

Table 3.2. Plasmids

Plasmids

6H7

pAG1X7

pAGlX7b

pAGl.3

pKOl 1

pK012

pAGl 11

pAGl l lrec

pAG112

Description

TetR; cosmid carrying the promoter
region of the rrnA operon

ApR; BamHI fragment containing
the rm promoter subcloned fi'om
pAGIX and cloned into BamHI site

of pBluescript SK-

ApR; BgIII/Xbal fragment subcloned
from pAG1X7 and cloned into
pBluescript SK-

KmR; 605 bp BgIII/Sacl fi'agment
containing the rm promoter, subcloned
from pAGlX7, and cloned in front

of the luxAB genes in pRK290

ApR; 2.5 kb XbaI fragment carrying
the luxAB genes subcloned from
pRLl 0623 cloned into the XbaI site
of pBluescript KS

ApR; 2.5 kb XbaI fi'agment carrying
the uidA cloned into the EcoRI site
of pBluescript KS

ApR; fiagment A (419 to +1) of the
rrnA promoter cloned into the SacI/Notl
sites of pKOll

TetR,SpR; Kpnl fiagment containing the
rm promoter-luxAB fusion (-419 to +1)
cloned into pMW193

ApR; fragment A (419 to +1) ofthe

rrnA promoter cloned into the SacI/NotI
sites of pK012

117

Reference or source

This study

This study

This study

This study

This study

This study

This study

This study

This study

 

Table 3.2. Plasmids (cont'd)

Elfimids

pAGl 12rec

pAG3] 1

pAG3 1 lrec

pAG311-1

pAG3 12

pAG3 12rec

pMW193

pRK2013

pJB25 1

Description

TetR,SpR; KpnI fragment containing the
rm promoter-uidA fusion (-419 to +1)
cloned into pMW193

ApR; fragment B (-419 to +172) of
the rrnA promoter cloned into the
SacI/Notl sites of pKOII

TetR,SpR; KpnI fragment containing the
rm promoter-luxAB fusion (-419 to
+172) cloned into pMW193

ApR; a derivative of pAG3] 1,
following the digestion of pAG311
with PstI, this plasmid contains
rm promoter fragment B fused to
the 399 bp XbaI-PstI fragment

of the luxAB genes

ApR; fragment B (419 to +172) of
the rrnA promoter cloned into the
SacI/Notl sites of pK012

TetR,SpR; KpnI fiagment containing the
rm promoter-uidA fusion (-419 to
+172) cloned into pMW193

SpR,TetR; pRK290-based vector
designed to recombine promoter-
reporter gene fusions into the ino-
sitol locus of RM1021

KmR; helper plasmid used to mobilize
plasmids into RM1021 strains

GmR; this plasmid incompatible with
RK2 vectors

118

Reference or source

This study

This study

This study

 

This study ‘v

This study

This study

Bosworth et al., 1994

Figurski et al., 1979

Bosworth et al., 1994

 

.285 uh. CE::§.3::8

 

Damage—8:5 8182.8 350:0: 8:: 8:8..8
72.— 5 m. >>HOE>QOOQHOQQQO>QH u. 83.55855. 8 8:5: 5m8: om 8m 3.: mafia
:8: 8 :88 M. 35:8: 88:: 8:5.
72.30 m. OO>Q__._._.H.OA.OQQ.:.OQ u. 88208853. 8 :5 3.: 58:81:" :65 +8
8 +_ 3w F8: 8.. 2.85.. 3838:
22M _ w m. OHOQOQQOOQO>OQ§OHOEOHO 1%.: 2.55.. 8.. 88:—58:0: em 8865:: w o». :5
.H>H>:O>>:QHO u. 3.: 32:05.. Emma: 258823 8 +3~w 808:8
:5 2o: :5 8.. 08:8m
22.2: m. OHOQOQQOOOOHO>HO§O§>H> 1%: 2:80.. 8.. 83258.8: o». 2.8853 > 0». :5
b>OOOO>QOOOHOO>> w. 3.: 32:05.. 5m8: 88:83 2:: +_w 808:8
:5 2o: :5 8a o_o:::m
72:; m. OHOQ>OOHOQQH>OOQ>flOfiOQOQO 5: “5.5.. 8~ 88:—88:8: o». 8885 > 8:: w
-OQQHOOQQHQ w. o». :5 3.: 32:29 5m8: 8888.5 8" .30 8::
828:5 Man: 8:: 8:: 88 8.. 08:5m
72.8. m. 8 2.: ammenjmjmmqufiomc» w. 8.853853. 8 :5 3.: 8:81.: :65 +8: 8

+3950: 8.. 25.5.. 9888:

119

Discussion

In a previous study it was shown that Sinorhizobium meliloti 1021
contained three operons encoding rm genes (Honeycutt et al., 1993). Here we
report the sequence and regulation of the promoters controlling the expression of
these operons (Figure 3.10). The sequence upstream of the predicted 5’ end of the
mature 16S rRNA transcript is essentially identical in each operon suggesting that

the three operons are coordinately regulated (Figure 3.1). Two methods were used

 

to map the transcriptional start sites of the three operons. While the primer
extension analysis suggested that there might be two transcription start sites and,
therefore, two promoters (Figure 3.2), the ribonuclease protection assay ruled out
the existence of one of the start sites (Figure 3.3). Although neither assay can
distinguish between the original 5’ end of a transcript and a processed end, the
primer extension assay can produce artefactual bands due to the premature
termination of reverse transcriptase. Thus, we conclude that a single promoter
controls the expression of each of the three operons. The banding patterns
obtained from samples taken before and after a temperature downshift were the
same, indicating that the same promoter sequences were used at both temperatures
(data not shown).

The predicted -10 and -35 hexamers in the S. meliloti 1021 promoter

regions are very similar to those of the E. coli Pl rm promoters (Figures 3.10).

120

There is however, a larger separation between the two hexamers in S. meliloti
1021, 18 bp as opposed to the 16 bp found in E. coli. The rRNA transcript in all
three S. meliloti 1021 rrn operons begins with adenine as do most of the rRNA
transcripts in E. coli. This similarity suggests that S. meliloti and E. coli may
share a similar mechanism for growth rate-dependent control. In E. coli, the
efficiency of transcriptional initiation at the rrn promoters is sensitive to the :-
availability of the initiating nucleotide, usually ATP (Gaal et al., 1997). The level
of ATP has been positively correlated with increasing growth rates. The

efiiciency of transcription initiation, therefore, increases with growth rate leading

 

to higher expression of the rrn genes.

The UP element also contributes to rm promoter activity in E. coli (Ross et
al., 1993). This element located between -60 and -40 is rich in adenines and
thymines and is thought to interact with the or subunit of RNA polymerase. The
region between -67 and -40 in the S. meliloti rrn promoter regions also contains a
large number of adenines and thymines suggesting that this may be an activation
sequence as well (Figure 3.10). It is interesting to note that while the sequence of
rrnA begins to diverge in this region, the high frequency of adenines and thymines
is maintained (Figure 3.1).

The leader regions of rRNA transcripts typically contain conserved
elements important in the processing of the 16S rRNA and transcriptional

antitermination. In E. coli, inverted repeats on either side of the mature 168 and

121

23S rRN A sequences anneal to form processing stalks which are then cleaved by
RNase III to form the pre-l6S and pre-23S rRNAs (see review Srivastava and
Schlessinger, 1990). In S. meliloti 1021, inverted repeat sequences were also
found on either side of the 168 rRNA sequence (Figure 3.10). It is important to
note that the second transcriptional start site predicted by the first primer extension
assay lies within the upstream repeat, (+158 to +194), further supporting the
existence of only one rrn promoter. Consequently, the 5’ end of the repeat is
contained within promoter fragment B (-419 to +172).

A second conserved element, boxA, was located within the leader region
and also contained in promoter fragment B cloned into RM311 and RM312. This
element is important for the expression of the rRNA genes through a
transcriptional antitermination mechanism. The boxA element in E. coli consists
of the 12 bp sequence TGCTCTTTAACA. Point mutations made in this sequence
showed that nucleotides 2 to 8 (GCTCTTT) were functionally important for
antitermination (Berg e1al., 1989). This element appears to be highly conserved
among many diverse bacteria including Caulobacter crescentus, Pseudomonas
aeruginosa, Bacillus subtilis, Mycoplasma capricolum, Mycoplasma
hypopneumoniae, T hermus thermophilus, T hermatoga maratima, and
Halobacterium halobium (Berg et al., 1989). These sequences were flanked by a
stem loop and a short sequence high in adenines. An element with 75% identity to

the 12 bp E. coli boxA Was found in the rrn leader in S. meliloti 1021 (Figure

122

 

3.10). Five of the seven key nucleotides were identical to the E. coli sequence.
This was the same criteria used in locating boxA in the bacteria listed above. The
sequence immediately following the boxA sequence was reasonably high in
adenines and a putative stem loop was present in the upstream sequence. A further
characteristic found in the bacteria examined by Berg et al. (1989) was the
placement of the boxA element adjacent to or within the processing stalk of the If
16$ rRNA. The boxA element identified in the S. meliloti 1021 rm promoter is

only 11 bp 5’ to the inverted repeat predicted to form the processing stern of the

 

l6S rRNA (Figure 3.10).

Previously, Tn5-1062 insertions in the S. meliloti 1021 168 and 238 rm
genes were observed to increase in luminescence following a temperature shift
from 30°C to 15°C (Chapter 2). The level of lwcAB transcript was induced as
well, suggesting that the rm promoters might be activated by the shift to low
temperature (Figure 2.9). To determine whether the rrn promoters were, in fact,
activated by a temperature downshift, two promoter fragments were fused to the
luxAB reporter genes and integrated into the S. meliloti genome. Fragment A (-419
to + 1) contained only the upstream and promoter regions thus reflecting promoter
activity. Since fragment B (-419 to +172) contained a ponion of the rRNA leader
region, this promoter fragment fused to the luxAB genes would reflect any

additional promoter-independent regulation.

123

The pattern of luminescence in the luxAB fusion strain, RM311 (-419 to
+172), while similar to that observed in the Tn5-1062 insertion mutant, RM603,
was quite different to that of RM1 11 (-419 to +1) (Figure 3.6). Luminescence in
both RM311 and RM603 increased approximately 40-fold while luminescence in
RM] 11 increased by only 6-fold (Figure 3.6C). Comparing levels of
luminescence to levels of luxAB transcript revealed that while RM 1 l 1, RM311 and
RM603 contained comparable amounts of luxAB transcript, RM1 11 produced 5 to
8-fold more light than RM603 and RM311 at 30°C (Figure 3.7A). These data
suggest that the luxAB transcript is translated less efficiently in RM603 and
RM311. Since the luxAB genes are co-transcribed with a portion of the rrn
transcript in RM603 and RM311, secondary structures formed in the rRNA
portion of the transcript may interfere with the translation of the luxAB genes.
Alternatively, proteins interacting with the rRN A transcript may also contribute to
the inhibition of translation. When comparing RM311 and RM603, it is important
to note that adding only 172 bp of the rm transcript to the luxAB genes created a
result comparable to the presence of over 1 kb of rrn transcript. While it is not
clear what sequences at the 5’ end of the rrn transcript caused the reduction in
luminescence, it interesting to note that these sequences precede genes which are
not translated.

While it is reasonable that a particular transcript may be translated less

efficiently than another, it is unexpected that this effect would be lessened

124

following a temperature downshifi. Although the ratio of luxAB transcript to
luminescence in RM] 11 is comparable at both 30°C and 15°C, this ratio changes
markedly in RM311 and RM603 (Figure 3.7B). At 15°C, it appears that the luxAB
transcript is translated more efficiently in RM311 and RM603 than at 30°C. The
mechanism causing this change is not clear although a possible explanation
involves the nature of the cold shock response. In E. coli, adaptation of the r

ribosome to low temperature is thought to be central to the cold shock response

(Jones and Inouye, 1996). Further, the cold shock proteins, CspA and CsdA, have

 

been proposed to work together to remove secondary structure from mRNA thus
improving translation initiation. If the inhibition of translation observed at 30°C in
RM311 and RM603 is resulting from secondary structures in the rrn portion of the
luxAB transcript, translation of the luxAB transcript may be improved following
adaptation of the ribosome. Alternatively, the shift to low temperature may
directly disrupt whatever inhibitory structures or interactions were affecting the
rrn-luxAB transcript at 30°C.

The rm promoter fused to the luxAB genes at +1 showed a modest
induction in activity within an hour of the temperature downshift (Figures 3.6-3.8).
In Figure 3.8, the levels of luxAB transcript began to decline after three hours at
15°C suggesting that the induction is temporary. A transient induction of promoter
activity following an abrupt temperature downshift may be understood within the

context of a sudden inhibition of translation following a shift to low temperature.

125

Since rm promoter activity is modulated by the demand for protein synthesis,
inhibition of protein synthesis may signal the need for more ribosomes. Jones et
a1. (1992) found that the stringent response and the cold shock response were
opposing activities in E. coli. The stringent response occurs when the
concentration of uncharged tRNAs increases relative to charged tRNAs resulting
in an increase in ppGpp levels. Jones et al. (1992) found that triggering the W
stringent response prior to a cold shock delayed the cold shock response. In cells
lacking the ability to synthesize ppGpp, several cold shock proteins were induced

at higher temperatures. As one aspect of the stringent response involves the

 

repression of rRNA transcription, the equivalent of a nutrient upshift may lead to
the activation of rm expression. Previous studies indicated that more ribosomes
were not needed at low temperature as long as growth rate was allowed to
decrease. Other regulatory mechanisms could then predominate resulting in the
subsequent repression of rRNA transcription (Ryals et al., 1982). Thus, there may
be a transient increase in rm expression immediately following a temperature
downshift that is quickly moderated as the cell adapts to the new temperature. In
fact, when stable RNA was measured in cells of S. meliloti 1021 grown at 30°C
and 15°C, no significant difference was observed between cultures grown at the
different temperatures (T. Schmidt, personal communication). This data indicates

that there is no net increase in the number of ribosomes in cells grown at 15°C

relative to those grown at 30°C.

126

Fusing the rm promoter fragments to the uicM gene in RM112 and RM312
confnmed the transient induction of rrn promoter activity following a temperature
downshift. The transient nature of the promoter activation can be seen more
clearly in RM112 and RM312 as the uidA transcript appears to be less stable than
the luxAB transcript. The peak of induction was between 1-2 hours in the
experiment shown in Figure 3.9 as opposed to the peak at 3 hours shown in Figure
3.8. This difference is not necessarily characteristic of a particular reporter gene
as some variation was seen within each set of uidA and luxAB fusions (data not
shown). An interesting difference between the expression of the uidA and luxAB
genes is the effect of fragment B on the level of gene expression. The level of
uidA transcript is much higher at both temperatures in RM312 than in RM112
while the level of luxAB transcript differs by no more than two-fold between
RM311 and RM 1 l l. The increased expression of the uidA transcript in RM312 is
most likely due to antitermination activity originating from the 172 bp of rm
leader region. Apparently, transcription of the uidA gene in RM112 is sensitive to
premature termination. Since the levels of luxAB transcript are comparable in
RM311 and RM] 1 1, the transcription of the luxAB genes may not be affected by
premature termination.

Taken together, our results indicate that each rrn operon in S. meliloti 1021
is controlled by a single promoter. As these promoters are very similar to the rm

Pl promoters of E. coli, they may be regulated by similar mechanisms. Contrary

127

 

to our previous studies of rrn::Tn5-1062 insertions, the rrnA promoter fused at +1
to the luxAB genes showed only a modest 4 to 6-fold induction in luminescence
following a temperature downshift. However, fusing the luxAB genes to the rrnA
promoter at +172 essentially recreated the induction of luminescence initially
observed in the transposon insertion mutants. This induction appeared to be due
primarily to the inhibition of luxAB translation at 30°C. Construction of rm !'
promoter fusions to uidA confirmed the transient induction of the rm promoters

and indicated the presence of antitermination activity in the 5’ end of the rm

 

transcript.

128

Materials and Methods

Bacteria and growth conditions. Cultures of Escherichia coli DHSa were
grown aerobically at 37°C in LB broth (Sambrook et al., 1989) or on solid LB
medium containing 1.4% agar. Cultures of Sinorhizobium meliloti were grown
aerobically at 30°C in either tryptone-yeast extract medium, TY (Beringer, 1974),
GTS medium (Kiss et al., 1979), BSM medium (Bergersen, 1961), or BSI
medium. BSl medium is modified BSM containing 1% inositol instead of 1%
mannitol as the carbon source. Cultures were maintained on solid medium
containing 1.4% agar, or frozen to -80°C in TY broth containing 15% glycerol. S.
meliloti was cultured on solid media containing the following antibiotics when
appropriate: gentamycin, Gm (50 ug/ml), kanamycin, Km (200 rig/ml),
spectinomycin, Sp (100 ng/ml), streptomycin, Sm (250 rig/ml), and tetracycline,
Tet (10 ug/ml). E. coli was grown on plates containing ampicillin, Ap (100
ug/ml), Gm (50 pg/rnl), Km (100 pg/ml), Sp (100 rig/ml), or Tet (10 ug/ml) when
appropriate. A lower concentration, 50 rig/ml, of Km and Sm was used in broth
cultures. When used together in broth cultures, Gm and Sp were present at 20

rig/ml and 50 rig/ml, respectively.

129

 

DNA sequencing and analysis. Cosmid DNA was prepared for sequencing
from E. coli bacteria using Qiagen columns (Qiagen Inc., Valencia, CA). The
cosmids carrying the rm promoter regions were sequenced at the Biotechnology
Resource Laboratory, Yale University. Sequence alignments were performed
using the Pileup program in the GCG sequence analysis package (Wisconsin
Package Version 9.1, Genetics Computer Group, Madison, WI).

Isolation of RNA. RNA was isolated from S. meliloti strains according to a
modified protocol derived from that found in Ausubel et al., 1987, for the isolation
of RNA from gram negative bacteria. Bacteria were harvested from a 3-20 ml
aliquot of cell culture to which rifarnpicin had been added to a final concentration
of 0.2 mg/ml. The cell pellets were resuspended in 1.4 ml of protoplasting buffer
and 80 pl of 50 mg/ml of lysozyme added. The tubes were incubated on ice 5-10
minutes. The protoplasts were collected by centrifugation for 2 minutes at 7,000
rpm. The pellets were resuspended in 0.5 ml of gram negative lysing bufler,
mixed with 15 pl of diethylpyrocarbonate, DEPC, and incubated for 5 minutes at
37°C. Tubes were chilled briefly on ice, mixed with 250 pl of 5M NaCl,
incubated 10 minutes on ice and centrifuged for 10 minutes. The supernatant was
removed, added to a tube containing 1 ml ethanol and stored overnight at -20°C.
The RNA was collected by centrifuging for 15 minutes at 4°C, dried under
vacuum and resuspended in DEPC-treated water. Samples were quantified using a

spectrophotometer and measuring A260 and A230. Alternatively, the cell pellets

130

 

were frozen quickly in liquid nitrogen and the samples stored at -80°C. RNA was
then isolated using Rneasy Mini Kits (Qiagen Inc., Valencia, CA).

Primer extension analysis. Primer extension assays were performed
according to the manufacturer’s recommendations using the AMV Reverse
Transcriptase Primer Extension System (Promega, Madison, WI). The primers
used in this experiment were synthesized by the Macromolecular Structure Facility F
located in the Department of Biochemistry at Michigan State University. The
annealing temperatures used in the assay were specific to the primer, 65°C for

MT179 and 60°C for MT261. Depending on the experiment, 100 or 500 finoles of

 

end-labeled primer were used in a reaction with 10 or 20 pg of S. meliloti RNA.
The primer extension reaction products were compared to a sequencing ladder
synthesized with the same primer used in the primer extension assay. Standard
protocols were used as described in the T7 Sequenase v. 2 DNA Polymerase DNA
sequencing kit (Amersham Life Science, Inc., Cleveland, OH). The plasmid
template used with MT179 to produce a sequencing ladder was pAGlX7 (BamHI
fragment of pAGlX cloned into the Xbal site of pBluescript SK-). The template
used with MT261 was a smaller version of the above plasmid, pAGlX7b (BglII-
Xbal fragment cloned into pBluescript SK-). Primer extension products and
sequencing ladders were analyzed using standard gel electrophoresis and

autoradiography techniques.

131

RNase protection assay. RN ase protection assays were performed
according to standard protocols (Ausubel et al., 1987). The RNA probe was
synthesized from the plasmid pAG3 l l-1. The plasmid was digested with Sacl and
the 3 ’ overhang removed by incubation with T4 DNA polymerase. The cut
plasmid was then treated with proteinase K to remove any contaminating
ribonucleases. The template DNA was extracted once with
phenol:chloroform:isoamyl alcohol, ethanol precipitated, and resuspended in water
to a concentration of 500 ng/ pl. The RNA probe was then synthesized from the
template plasmid using the T3 DNA polymerase from Boehringer Mannheim. The
Promega Riboprobe system components (Promega, Madison, WI) were used
according to the manufacturer’s suggested protocol to synthesize a 32P-labeled
RNA probe complementary to a 399 bp fi'agment of the luxAB gene cassette fused
to a 500 bp fragment of the S. meliloti rrn promoter region. The RNeasy
Miniprep columns (Qiagen, Inc., Valencia, CA) were used to separate the RNA
probe from the unincorporated nucleotides according to the kit’s RNA cleanup
protocol. The radioactivity of the RNA probe was quantified using the Packard

Tri-Carb® liquid scintillation analyzer model 1500 (Packard Instrument Company,
Inc., Downers Grove, IL). 10 pg aliquots of S. meliloti RNA were briefly dried
under vacuum and resuspended in 30p1 of hybridization buffer containing 5 x 105
cpm of probe. Samples containing 10pg of tRN A were included with each assay

as a control. After 5 minutes at 85°C samples were hybridized at 55°C overnight.

132

After ribonuclease digestion and treatment according to the standard protocol, the
samples were separated on a 4% acrylamide/urea gel and visualized by
autoradiography. The GibcoBRL 0.16-1.77 RNA ladder was end-labeled with [y-
32P]ATP to act as a size standard for gel electrophoresis.

PCR cloning and integration of the resulting promoter-luxAB fusions.
Two regions of the rrnA promoter region were amplified from the cosmid 6H7
(Figure 3.1) which carries the upstream region of the 16S rm isolated on fragment
1 (Figure 2.6) and corresponds to the region cloned into pAGl.3 (Figures 2.7 and
2.8). The left primer used for both fragments was MT215 5'-CTCGAGCTCGGT
ACCGATCTCGCGCCGGTCCGGTG-3' and includes the Sac] and Kpnl sites for
cloning. The right primer used for the larger fi'agment B (Figure 3.5) was MT213
5'-CTCGCGGCCGCACGTTTCTCTTTCTCTATATTCAATTGTC-3'
incorporating the Not] site. The right primer used for the smaller fragment A was
MT214 5'-CTCGCGG CCGCTGATCGGGCTTATAGACCCCAGCCCTCCAA-3'
with the Not] site included as well. VentR® DNA polymerase (New England
Biolabs Inc., Beverly, MA) was used to amplify the fragments from 0.05 ng of
6H7 DNA with 20 pmoles of each primer. The PCR conditions were as follows:
94°C (5 minutes) then 30 cycles at 94°C (1 minute)/ 50°C (1 minute)/ 72°C (1
minute) followed by 5 minutes at 72°C. A second round of amplification was
necessary to obtain enough of the fragments for cloning. The conditions of the

second round were the same as the first with two exceptions: only 26 cycles were

133

 

used and the first PCR product was used as template. The PCR products were
then digested with Sac] and Not] and band isolated. These fragments were cloned
into the SacI and Not] sites in the multiple cloning site of pKOl 1. The plasmid
pKOll is a derivative of pBluescript KS containing the luxAB genes cloned into
the Xbal site. The resulting plasmids, pAG] 11 or pAG311, were isolated and
digested with the enzyme Kpnl. The enzyme cut at the Kpnl site engineered into
the left primer and the Kpnl site in the multiple cloning site from pKOI 1. The
band isolated contained the promoter fragment in the correct orientation to control
transcription of the luxAB genes. This fragment was then cloned into pMW193
forming the plasmids, pAGl 1 lrec and pAG311rec. The plasmids were moved into
the wild type strain RM 1021 via triparental matings using the plasmid pRK2013 as
a helper plasmid. Transformed colonies were selected for on BSM plates
containing Tet or Sp. The incompatible plasmid pJB251 (GmR) was moved into
the RM102] strains carrying pAG] 1 lrec or pAG311rec using the triparental
mating method and pRK2013 (KmR) as a helper plasmid. Mating spots were
streaked on BSM plates containing Gm and Sp. The resulting colonies were
grown up in TY broth containing Sp and Gm until cultures were dense and then
diluted and spread on BSM with Sp and Gm. These colonies were then screened
for the inability to grow on inositol and the loss of tetracycline resistance. Three
replicate plates were made: BSI with Sp and Gm, TY with Sp, Gm, and Tet, and

TY with Sp and Gm. Single recombinants should grow on both sets of TY plates

134

 

while double recombinants should grow only on TY with Sp and Gm. In the
construction of these plasmids the second recombination event happened at a high
enough frequency such that no single recombinants were isolated, only double
recombinants.

Detection and quantification of bacterial luciferase activity. Luciferase
activity was measured differently depending on the culture conditions used to
grow the bacteria. If the bacteria were grown on a solid agar medium,
luminescence was measured with the Hamamatsu Photonic System model C1966-
20 (Photonic microscopy, Oak Brook, IL). The Berthold LUMAT luminometer
LB 9501 (W allac Inc., Gaithersburg, MD) was used to measure aliquots of liquid
bacterial cultures. Bacteria grown on agar plates were exposed to the enzyme
substrate, N-decyl aldehyde, for two minutes prior to being assayed for light
production for 59 seconds. The aldehyde was spread on a glass petri dish which
was then placed over the agar plate. The Hamamatsu system converts photon
counts to a visual image displayed on a monitor. This system can also quantitate
the amount of light produced by each colony.

Bacteria grown in broth culture were assayed quantitatively for light

production. In this assay, the N-decyl aldehyde was dispersed in an aqueous

solution of bovine serum albumin (20 mg/ml) at a concentration of ] pl/ml. A 5-

10 p1 aliquot of the cell culture was added to 50-100 pl of the aldehyde solution,

vortexed for 30 seconds, then assayed for 60 seconds. The measured

135

 

luminescence was reported in arbitrary units or RLU and adjusted for cell number
by dividing this number by the A600 of the culture. All assays were performed at
room temperature.

Northern blots. RNA samples were treated with formaldehyde and
electrophoresed on formaldehyde/agarose gels according to standard protocols
(Sambrook et al., 1989). RNA was transferred from the gel to a nylon membrane
in 10X or 20X SSC via capillary action according to the standard protocol
(Sambrook et al., 1989) with one modification. After transfer, the RNA was
crosslinked to the membrane by exposure to 20 mjoules of light energy in the UV
Stratalinker 2400 (Stratagene, La Jolla, CA). The membranes were pre-hybridized
at 42°C in the following solution: 50% formarnide, 5X SSC, 50mM potassium
phosphate, 5X Denhardt’s solution, 0.5% sodium dodecyl sulfate, and 100pg/ml
denatured DNA. After a minimum of 2 hours of pre-hybridization treatment, 32P-
labeled probe was added to the solution and incubated overnight at 42°C in a
Robbins Hybridization Incubator Model 400 (Robbins Scientific Corp., Sunnyvale,
CA). The probe was prepared from a band isolated fragment by a random priming
method (Feinberg and Vogelstein, 1983). The membranes were treated first at
room temperature with three washes of 2X SSC and 0.5% SDS. After one wash in
0.1X SSC and 0.5% SDS, the temperature was raised to 50°C for two more washes
in the same solution. Washes were continued until the spent wash solution was no

longer detectably radioactive.

136

 

Literature Cited

Afflerbach, H., O. Schrr’ider, and R. Wagner. 1998. Effects of the Escherichia
coli DNA-binding protein H-NS on rRNA synthesis in viva. Mol. Microbiol.
28(3):641-653.

Appleman, J. A., W. Ross, J. Salomon, and R. L. Gourse. 1998. Activation of
Escherichia coli rRNA transcription by FIS during a growth cycle. J. Bacteriol.
180(6): 1525-1532. FE

Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A.
Smith, and K. Struhl (ed.). 1987. Current protocols in molecular biology. John
Wiley and Sons, NY.

 

Aviv, M., H. Giladi, A. B. Oppenhiem, and G. Glaser. 1996. Analysis of the
shut-off of ribosomal RNA promoters in Escherichia coli upon entering the
stationary phase of growth. FEMS Microbiology Letters 140271-76.

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

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

Bosworth, A. H., M. K. Williams, K. A. Albrecht, R. Kwiatkowski, J. Beynon,
T. R. Hankinson, C. W. Ronson, F. Cannon, T. J. Wacek, and E. W. Triplett.
1994. Alfalfa yield response to inoculation with recombinant strains of Rhizobium
meliloti with an extra copy of dctABD and/or modified nifA expression. Appl.
Environ. Microbiol. 60(10):3815-3832.

Cashel, M., D. R. Gentry, V. J. Hernandez, and D. Vinella. 1996. The
stringent response, p. 1458-1496. In F. C. Neidhart, R. Curtiss [11, J. L. Ingraham,
E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznifl<off, M. Riley, M.
Schaechter and H. E. Umbarger (eds.), Escherichia coli and Salmonella: cellular
and molecular biology, 2nd edition. ASM Press, Washington, D. C.

Dryden, S. C. and S. Kaplan. 1993. Identification of cis-acting regulatory

regions upstream of the rRN A operons of Rhodobacter sphaeroides. J. Bacteriol.
175(20):63 92-6402.

137

Fang, L., W. Jiang, W. Bae, and M. Inouye. 1997. Promoter-independent cold-
shock induction of cspA and its derepression at 37°C by mRNA stabilization.
Mol. Microbiol. 23(2):355-364.

F einberg, A. P., and B. Vogelstein. 1983. A technique for radiolabeling DNA
restriction endonuclease fragments to high specific activity. Anal. Biochem.
132(1):6-13.

Figurski, D., and D. Helinski. 1979. Replication of an origin containing
derivative of plasmid RK2 dependent on a plasmid function provided in trans.
Proc. Natl. Acad. Sci. USA. 76:1648-1652.

Gaal, T., and R. L. Gourse. 1990. Guanosine 3’-diphosphate 5’-diphosphate is
not required for growth rate-dependent control of rRNA synthesis in Escherichia
coli. Proc. Natl. Acad. Sci. USA. 87:5533-5537.

Gaal, T., M. S. Bartlett, W. Ross, C. L. Turnbough Jr., and R. L. Gourse.
1997. Transcription regulation by initiating NTP concentration: rRNA synthesis
in bacteria. Science 278:2092-2097.

Gourse, R. L., T. Gaal, M. S. Bartlett, J. A. Appleman, and W. Ross. 1996.
rRNA transcription and growth rate-dependent regulation of ribosome synthesis in
Escherichia coli. Annu. Rev. Micriobiol. 50:645-677.

Hanahan, D. 1983. Studies on the transformation of Escherichia coli with
plasmids. J. Mol. Biol. 66:557-580.

Jones, P. G., M. Cashel, G. Glaser, and F. C. Neidhardt. 1992. Function of a
relaxed-like state following temperature downshifts in Escherichia coli. J.
Bacteriol. 174(12):3903-3914.

Keener, J. and M. Nomura. 1996. Regulation of ribosome synthesis, p. 1417-
1431. In F. C. Neidhart, R. Curtiss H], J. L. Ingraham, E. C. C. Lin, K. B. Low, B.
Magasanik, W. S. Rezniflmff, M. Riley, M. Schaechter and H. E. Umbarger (eds),
Escherichia coli and Salmonella: cellular and molecular biology, 2nd edition.
ASM Press, Washington, D. C.

Kiss, G. B., E. Vincze, Z. Kalman, T. Forrai, and A. Kondorosi. 1979. Genetic

and biochemical analysis of mutants affected in nitrate reduction in Rhizobium
meliloti nodule bacteria, nitrogen fixation. J. Gen. Microbiol. 113:105-118.

138

Mackow, E. R. and F. N. Chang. 1983. Correlation between RNA synthesis and
ppGpp content in Escherichia coli during temperature shifts. Mol. Gen. Genet.
192:5-9.

Meade, H. M., S. R. Long, G. B. Ruvkun, S. E. Brown, and F. M. Ausubel.
1982. Physical and genetic characterization of symbiotic and auxotrophic mutants
of Rhizobium meliloti induced by transposon Tn5 mutagenesis. J. Bacteriol.
149:114—122.

Ross, W., J. F. Thompson, J. T. Newlands, and R. L. Gourse. 1990. E. coli Fis
protein activates ribosomal RNA transcription in vitro and in vivo. EMBO J
9(11):3733-3742.

Ryals, J., R. Little, and H. Bremer. 1982. Temperature dependence of RNA
synthesis parameters in Escherichia coli. J. Bacteriol. 151(2):879-887.

Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a
laboratory manual, 2nd edition. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, NY.

Tippner, D., H. Afllerbach, C. Bradaczek, and R. Wagner. 1994. Evidence for
a regulatory function of the histone-like Escherichia coli protein H-NS in
ribosomal RNA synthesis. Mol. Microbiol. 11(3):589-604.

139

 

CHAPTER 4

A Sinorhizobium meliloti 1021 gene similar to the Escherichia coli me
gene is induced following a temperature downshift

Summary

Sinorhizobium meliloti 1021 was mutagenized with the transposon, Tn5-

 

233, to identify genes whose products affected the expression of the cold-inducible
rrnP-luxAB fusion, pAGl .3. The transposon recipients were assayed for their
ability to affect the light production of pAGl.3 which contains the luxAB gene
cassette fused to the rrnA promoter at +187. Approximately 28,000 colonies were
screened for increased luminescence at 30°C and 14,000 colonies for decreased
luminescence at 15°C. Thirteen transposon inserts were found that increased
luminescence from pAGl.3 at 30°C. Two of these inserts, RM26b1 and RM483
increased light production approximately 5-fold at 30°C when they were
transduced into strain RM603 (16S rrnAzzTn5-1062). Both inserts were located
within an ORF whose deduced protein product showed significant similarity to
RNase E, an essential endoribonuclease involved in the processing and

degradation of mRNA and rRNA. The transposon insertions are located near the

140

carboxy terminus of this protein. The insertion mutants, therefore, should produce
a truncated form of RNase E. The inserts were shown to affect the accumulation
and processing of the luxAB mRNA at 30°C in the double mutants RM603(26b1)
and RM603(483). They also influenced the processing and accumulation of the
cspA transcript following a temperature downshift. When expression of me, the
gene encoding RN ase E, was examined before and after a temperature downshift,

the levels of me transcript were found to be transiently induced.

141

Introduction

While it is known that many bacteria induce a cold shock response
following a temperature downshift, the mechanisms by which the change in
temperature is sensed and the protein expression regulated are not fully
understood. The cold shock response has been most thoroughly studied in E. coli.
Following a shift in temperature from 37°C to 15°C, E. coli induces a set of 16
proteins despite a transient inhibition of overall protein synthesis (Thieringer,
1998). These cold shock proteins can be divided into two classes. The first class
includes proteins that are primarily expressed at low temperature and the second
comprises proteins that are expressed at the higher temperature and only
moderately induced following a cold shock (Mitta el al., 1997). While these
proteins have a variety of cellular functions, several proteins including CsdA,
RbfA and IF 2 are specifically involved with the ribosome (Jones et al., 1996;
Jones and Inouye, 1996; Jones et al., 1987). Because of this and the transient
inhibition of protein synthesis that occurs in E. coli following a cold shock, the
cold shock response has been viewed in terms of adapting the translational
machinery to function at low temperature (Thieringer er al., 1998). One aspect of
this adaptation involves the action of proteins like CsdA and CspA in destabilizing
RNA secondary structure (Jones et al., 1996; Jiang et al., 1997). Secondary

structures formed in mRN A at low temperature are thought to inhibit translation.

142

 

'1

According to the “Cold-Shock Ribosome Adaptation” model proposed by Jones
and Inouye (1996), CsdA, a protein with RNA helix unwinding activity, and

C spA, a putative RNA chaperone, are thought to interact to remove and prevent
formation of secondary structures in mRNA.

Interestingly, the addition of low levels of antibiotics like chloramphenicol
that inhibit ribosome function can induce the cold shock response at 37°C
(VanBogelen and Neidhart, 1990). This observation and the effect of low
temperature in inhibiting translation in E. coli led to the proposal of the ribosome
as a sensor of temperature change. The relationship of the cold shock response
and the stringent response was then investigated. The stringent response involves
the changes in gene expression following arrrino acid starvation. While treatment
with chloramphenicol, low temperature and amino acid starvation all result in the
inhibition of protein synthesis, the immediate effect on the ribosome differs.
Treatment with chloramphenicol and low temperature blocks the ribosome in such
a way that charged tRNAs are in excess. Amino acid starvation results in the
opposite situation. Indeed, the stringent response was found to work in opposition
to the cold shock response (Jones et al., 1992). The overexpression of ppGpp, the
signal molecule for the stringent response, inhibited the cold shock response in E.
coli.

The regulation of the major cold shock protein, CspA, has been studied in

detail and has been found to occur at many levels. Promoter activity is only a

143

minor component in this regulation which affects message stability, translation and
potentially transcription attenuation. The cspA transcript contains a 159 bp
untranslated region at its 5’ end that causes the transcript to be extremely unstable
at 37°C (Brandi et al., 1996). The message is then transiently stabilized at 15°C
allowing the rapid induction of CspA protein. Further investigation of this region
by Mitta et al. (1997) suggests that while the first 26 bp are sufficient to F-

destabilize the transcript at 37°C, the remainder of the UTR is required to inhibit

expression of the cspA gene. Contrary to its inhibitory function at 37°C, the larger

 

fragment of the UTR was also found to be required to achieve the full level of E
mRNA stability at 15°C. Mitta et al. (1997) observed that the downstream box,

located at the 5’ end of the coding region, increased translation at 15°C. This
downstream box was required for the full induction of the CspA protein and is
thought to increase translation by interacting with the ribosome. Four other highly
induced cold shock proteins have UTR sequences at their 5’ ends and appear to be
derepressed in a cspA mutant although the full interaction of these regulatory
elements is not yet clear (Bae et al. , 1997).

In studying the cold shock response in Sinorhizobium meliloti 1021, we
isolated several reporter transposon insertions that exhibited cold-inducible gene
expression (Chapter 1). While two of these insertions were within an operon
containing genes similar to cspA and rpsU (O’Connell and Thomashow, 1999), the

five other insertions were in ribosomal RNA genes. The rrnA promoter was

144

isolated on a 605 bp DNA fragment and fused to the luxAB genes at +187 in the
plasmid, pAGl.3. This construct showed temperature-dependent luminescence
when propagated in S. meliloti 1021. In order to identify potential regulatory
factors acting on the rrnP-luxAB fusion, S. meliloti 1021 was mutagenized and the
transposon mutants screened for their ability to alter the luminescence of the
luxAB fusion. Two insertional mutants were isolated that caused increased 1‘
luminescence of the rrnP-luxAB fusion and a 16S rrn Tn5-1062 insertion at 30°C.
These two insertions were located within an ORF encoding a protein with

significant similarity to Ribonuclease E. The temperature-dependent expression of ,-

 

this locus is described as well as the effect of the insertional mutants on the 16S

rm Tn5-1062 insertion and on cspA expression.

145

Results

Transposon mutagenesis and screen for transposon inserts affecting the
luminescence of pAGl.3. Sinorhizobium meliloti 1021 was mutagenized with the
spectinomycin resistant Tn5-derivative, Tn5-233 (De Vos et al., 1986), (Figure
4.1A). The rrnP-lwcAB fusion, pAGl.3, was mobilized into groups of transposon
recipients (Figure 4.1B). Approximately 28,000 of the resulting colonies were
screened for increased luminescence at 30°C and 14,000 colonies were screened
for decreased luminescence at 15°C. No transposon inserts were found that
affected luminescence at 15°C, but thirteen colonies showed increased
luminescence at 30°C.

Characterization of the putative regulatory mutants. Southern blot
analysis of the mutant strains showed that each contained a single Tn5-233
insertion (data not shown). Each of these insertions was reconstructed in strain
RM603 using the transducing phage CID-M12 (Finan et al., 1984). RM603 contains
a Tn5-1062 transposon inserted in the 16S rrnA gene that show cold-inducible
luminescence. By transducing the Tn5-233 insertions into RM603, we hoped to
correlate increased luminescence at 30°C with the presence of the transposon
insertion. In addition, the number of luxAB genes present in the cell was reduced
to a single chromosomal insertion, thus decreasing background luminescence at
30°C. All thirteen of the inserts affected the luminescence of RM603 at 30°C.

146

 

A Tn5-233

 

 

 

 

 

 

 

 

 

Kpnl EcoRI

[ ]

IS50L } 1S5 0R
GmR/KmR SpR/SmR
B pAGl.3
BglII ac] X7101 /SalI
rrnP f |
i luxAB oriV KmR
pRK290

 

Figure 4.]. Structures of Tn5-233 and reporter plasmid, pAGl.3. (A) The
structure of Tn5-233 (De Vos et al, 1986) is shown. Tn5-233 is a derivative of
Tn5 from which the central BglII fiagment has been removed and replaced by
genes encoding resistance to Gm, Km, Sp, and Sm. The two restriction enzyme
sites used in cloning out the ends of the transposon are shown. (B) The
structure of the rmP-lwcAB plasmid, pAG] .3, is shown. The plasmid, pAGl .3,

is described in detail in Figure 2.8.

147

 

Two of these insertions, RM26bl and RM483, are described below.
Characterizing the luminescence of the double mutants RM603(26b1)
and RM603(483). The Tn5-233 inserts, RM26b1 and RM483, were transduced
into RM603 forming the double mutants, RM603(26b1) and RM603(483). When
cultured on TY agar plates, the double mutants exhibited increased luminescence
at 30°C relative to the original strain RM603 but showed little difference in
luminescence at 15°C (Figure 4.2). The double mutants were then grown in TY

broth with the original insertion mutant RM603 to better quantitate the effect of

 

..--.- __

the Tn5-233 insertions on the luminescence of the Tn5-1062 insertion. At 30°C,
RM603(26b1) and RM603(483) both produced approximately 5-fold more light
than RM603 (Figure 4.3). However, within an hour of a temperature downshift to
15°C, the difference in lunrinescence between the double mutants and RM603 had
declined significantly. After three hours at 15°C, the transduced strains produced
only 17 to 20% more light than RM603. While the Tn5-233 insertions did
increase luminescence at 30°C, luminescence was not fully induced until after the
temperature downshift. In fact, luminescence in the double mutants still increased
approximately 5-fold after the temperature downshift.

Identification of the insertion sites in RM26b1 and RM483. The
insertion sites in RM26b1 and RM483 were identified by cloning and sequencing
the DNA contiguous with either one or both ends of the transposon in each of the

strains (Figure 4.4). Comparison of the sequences indicated that the two

148

3...; I I

RM603 RM603 RM603
(26bl) (483)

 

15°C

 

Figure 4.2. The Tn5-233 inserts in RM26bl and RM483 cause increased
luminescence at 30°C when transduced into RM603. RM603 (16S rrn::Tn5-
1062) was compared to two double mutants RM603(26b1) and RM603(483).
Each double mutant is a derivative of RM603 containing a Tn5-233 insert.
Cultures were streaked on TY agar plates and incubated at 30°C for two days
before being assayed for luminescence. After being assayed for luminescence
at 30°C , the plates were moved to 15°C for 5-6 hours then assayed again.

149

 

700

 

 

I RM603
RM603(26b1)
r3 RM603(483)

 

 

RLU/A600 x 10‘5

 

 

 

 

 

 

 

c
H
N
u

Hours at 15°C

Figure 4.3. Luminescence of RM603(26b1) and RM603(483) at 30°C and
15°C. Luminescence of the two double mutants RM603(26b1) and
Rm603(483) was compared to that of RM603 (16S rrn::Tn5-1062) during
- growth at 30°C and after a shift to 15°C. The double mutants are derivatives of
RM603 containing Tn5-233 inserts. The strains were cultured in TY broth with
Km (50 pg/ml). Cultures were grown at 30°C until they reached an A600 of
approximately 0.3. At that point, samples were taken and the luminescence
measured. The cultures were then shifted to 15°C and samples taken 1, 2, and 3
hours after the temperature downshift. Luminescence is measured in relative
light units, RLU, and adjusted to culture density by dividing by the A600 of the
culture. Each bar represents the average of measurements of three flasks and
the error bars give the standard deviation of these measurements.

150

 

GAATTCGATT
CCTGGCGAAG
ATTACGGCGG
GACTACTACC
AGCCGAAGAG
AACAGAAATT
GCCGAAGCCG
GCCGAAGGCC
CCGGCGAAGA
GGCGACAAGA
GGATATCGAC
ACGACGGCGA
GAAGAGGTGG
CCAGGAAGTG
AAGAGCGCGG
GGCCGCTACT
GCGCAAGATC
GCGGTCTCGA
GCCAATCGCA
CCTGTGGGAG
TCGTCTACGA
ACCAAGGACA
AGCCAAGGGC
AGCCCTATCG
CAGCTCGATC
CATCATCATC
GTCGTTCGAC
CTCGAGGCAG
CGGCCTGATC
GCTCCGTCGA
ATCCAGGTCG
GCGCATCCGC
GCAACGGCAC
CTGCGCGGCA
CGTGCGGACG
GTTCGATCAC
GCGGACGCCC
GGTCGAAAAT
CGGAAGACGA
GTGGAAGAGA
CGAACAAGGC

GTGCGGGTCA

GAGGACGAGA
CGACGCTACG

Figure 4.4

TCGAATCGGA
GTAACCAGAG
CAATCGCCAC
AGATTCCCCT
GCTCGCCGCG
CGCGCCGGCT
CCGAGGCAGC
AAGGCCAAGC
GGCCGTGGCG
ACGAAATGGC
GCGCGCCGTC
AAAGGAAATC
CGGATCGTCA
ATCAAGCGCC
CAACAAGGGT
CCGTTCTCAT
ACCAACCTGC
CGTGCCGCAG
CGAAAGTCGA
AACGTCCGCA
GGAGGGCAGC
TCAGCGAGAT
TTCATGAAGA
CGACGTGCAT
GCATGCTGCA
AACCAGACCG
GCGCGAGCAC
CCGAGGAAGT
GTCATCGACT
GAAGAAGCTA
GCCGCATCTC
GCCTCGGTGC
GGGCCACATT
TCGAGGAACA
ACACCCGATA
GGACTACGAG
ATGTCGGTGC
CCGGTGAGGA
AGAAGAGGTC
TTGCTGCCGA
GGCCGCAAGC
RMI83
GGGGGCGGAC
CCGGCGACGA
GATGCCTTGA

ACATAAGAAG
TGGAGCCGTC
GGGTTTCTGG
CGCCGATCGT
ACGATGACGA
GGCGATACGG
CGGAGAAGCC
CGAAGCGTAC
CCTGAAAGCG
CGCGGTGGTC
AGCGTGACGA
ATCGAATCCG
TATCCGCAAG
GGCAGATCCT
GCGGCACTCA
GCCGAACACG
AGGACCGCAA
GGGATGGGTG
GATCAAGCGC
CGCTGACGCT
CTTATCAAGC
CGTCGTTTCG
TGCTGATGCC
CCGATCTTCT
GCCGCAGGTG
AAGCTCTGGT
TCGATCGAGG
AGCGCGGCAA
TCATCGACAT
AAGGATTGCC
GCATTTCGGC
TGGAAAGCAC
CGCTCGCAAT
TCTGCTCAAG
TCGCGCTCTA
CAGCGCTTCG
GCAGCACTTC
TCGACCAGAT
CTCATCGAGG
GCGTCAGGAC
GCAAGCGTCG

GAGGCCTTCG

AGCCGCCGTG
ATGGCGATAG

151

CAGATCCGCG
GCTCCAGGCG
CCTTCGCCGA
CAGGCCCTGT
GCCGAACGAC
TTGCCGAAGT
GCTGAGGCCG
CCGCCGCGCG
GCCCGTCGGA
GATACCGACT
CGACGATGAC
TCGGCGCCGA
CCGCGCAAGC
GCTGGTGCAG
CGACCTATCT
GCGCGCGGCG
GCGTCTGAAG
TGATCCTGCG
GACTTCGAAT
GAACTCGACT
GATCGATCCG
GGCGAGGAAG
GAGCCACGCA
CCCGCTCCGG
ACGCTGAAGT
TTCCATCGAT
ATACGGCTCT
TTGCGCCTTC
GGAAGAGAAG
TGAAGAACGA
CTGCTCGAAA
GATGCAGACC
CCTCCGTTGC
AACACCACGC
CCTGCTCAAT
GTGTCTCCAT
GCGATCGATC
CCTGCAGTTC
AGGAGTCCGA
CCGAAAAAGG
TCGCCGCCGG

CTGACGCCGC
GACGATGCCG
CGAGCAGAAG

GCAATATCTA
GCCTTCGTCG
AATCCATCCC
TGAAGGCGGA
ACCGTGGCGG
CGTGGCACAG
CTGCGGAGAA
AAGCCGAAGG
AGACGGCGAA
CGATCTCCGA
GACGACAATC
AGATGCCATG
AATACCGCAT
GTCGCAAAGG
TTCGCTCGCC
GCGGTATTTC
GAGATCGCCC
CACCGCCGGA
ATCTCATGCG
GCCCCCTGCC
CGACCTCTAC
GCTACAAGGA
AAGGTCGTGC
CATCGAAGCA
CGGGCGGCTA
GTGAACTCCG
CCAGACGAAT
GCGACCTTGC
CGGAACAACC
TCGTGCCCGT
TGTCTCGCCA
TGCCCGCACT
CCTGCATGTC
ACGACATCAG
CAGAAGCGCA
CTTCATCGAA
GCGGTGAACC
GAGCCGGAGC
CGAAGAGGAA
GGCAGACCGA
CGCGGCAAGG

CGACGCGGCT
AGGATGAAGT
CGCAAGCGTC

50
100
150
200
250
300
350
400
450
500
550
600
650
700
750
800
850
900
950

1000
1050
1100
1150
1200
1250
1300
1350
1400
1450
1500
1550
1600
1650
1700
1750
1800
1850
1900
1950
2000
2050

2100
2150
2200

GCCGCCGCGG CAAGCGTGGC GGCCGCCGCA GCCGTCAGGA CGAACTGGCC 2250
GAAGCCGAGG GCGAGACCGG CATGGCTGCA GAAGCCGGCG AAGGAGCCGG 2300
CGACGACGCA GAGGCGGAAG TCGCCCCAGC CGCGGCAGCC GAGGAGACGT 2350
RM26b1
CACCGGATAT CGCGGCAGCC GCTGCCGTCG AGGCAGCAGC GGAGGAGCCG 2400
AAACCGGCAA.AACCCCGGCG CAGCCGCAAG AAAGCAGTCA.AGGCCGAGGA 2450
GCCGGCCGAG AGTGTCGATG AAGTAGCCGA GAAGACCGTC GAGGCCCAAC 2500
CCGAAGCAAC CGAAGCCGCC CCCGAGGCCG TGGAGGAAGC TGTTGCCGAT 2550
CTGGAAGGCG CAAAGCCCGC GCGCGCAAAC CGGGATCTGT CGGCGATCGC 2600
TACCGAGCCG GTCGTGACGT CCAGCAAGGG CGAGGGCGAA GAGGAGCCGA 2650
CAAAACCCAA AAAGGGCGGC TGGTGGCAGC GCCGCGGTTT CTTCTAAgaa 2700
>>>>>>>>>>>>>> <<<<<<<<<<<<<<

gaataggaaa aagcccggag cgatccgggc tttttcatgg ctggggccgc 2750
gtggcggctc agttgtgacc tgcctctttg aaaactccgc actttccgga 2800
cggaaagcgg cacacgcttt tccttatccc gctctatact ttcccgcagg 2850
gaggacgagg ccATGCGCAA GTTCACAGTC ATCGTTACGG AAGAGTTCGA 2900
AGCCGACACT GCCGAGGAGG CCGCCTTGCT CATGTATCAC CAACTGACCA 2950
ACGGGCCGGC ACCTCTGCAT TATTCGGTCA CGGATGAAAC GAAGATTGCA 3000
ACCAGCCTGA TACTGGACCG GAAAAAGGCA.GATGAATTTG CCAGTGTCGA 3050
TCACACTGCA GACCCCGGCA ATTGGTAAgc cacgcgtaac cggccgatgt 3100
gcggataccc atgtctggag cagagcac 3128

Figure 4.4 (cont'd). DNA sequence analysis of the insertion sites in RM483
and RM26b1. The DNA contiguous with the transposons in RM483 and
RM26bl was cloned and sequenced. Putative coding sequences are capitalized.
Non-coding sequence is typed in lowercase. Predicted terminator sequences are
indicated by arrowheads above the relevant sequence. The insertion sites for each
strain are labeled with the strain name and the relevant bases printed in bold.

152

 

transposon insertion sites were separated by only 310 bp. Two open-reading
frames (ORFs) were predicted for this region using the codon preference program,
CodonUse 3.1, developed by Conrad Halling (University of Chicago). A
schematic of the region showing the predicted ORFs and the location of the Tn5-
233 insertions in the two mutants is shown (Figure 4.5A). Both inserts are located
within ORF l which encodes a peptide 898 anrino acids in size. This first ORF is
followed by a putative rho-independent terminator and a second, smaller ORF
encoding a 71 arrrino acid peptide (Figure 4.4). The presence of the terminator
suggests that ORF 1 may be the last ORF in an operon or is transcribed alone.
Alternatively, ORF2 may be co-transcribed with ORF]. The insertions, then, may
be exerting a polar effect on the transcription of this second ORF in addition to
truncating the product of ORF].

Southern blot analysis of strains RM26b1 and RM483. Since the
transposons from RM26bl and RM483 were isolated as two different clones, there
existed the formal possibility that a gene similar to ORF] existed and two different
loci had been mutagenized. A Southern blot was performed to confirm that the
transposons were, in fact, inserted in the same locus (Figure 4.5). RM1021,
RM483, and RM26b1 were digested with BamHI, EcoRI, and Sal] and then probed
with a 32P-labeled DNA fragment complementary to ORF] upstream of the
RM483 insertion site (Figure 4.5A). As the enzyme BamHI does not out within

the transposon, the BamHI fragments in the mutant strains are larger than the

153

Figure 4.5. Southern -blot analysis of RM26b] and RM483. (A) A
schematic of the predicted ORF’s is shown with the insertion sites of the two
transposons indicated by arrows. Pertinent restriction sites are noted and the
size of the sequence shown in base pairs. The predicted rho-independent
terminator is represented by a stem loop. The region of the sequence
complementary to the double-stranded DNA probe used in the Southern blot
analysis is shown by the dotted line. (B) DNA isolated from the reference
strain, RM1021, and the mutant strains, RM483 and RM26bl, was digested
with BamHI (B), EcoRI (E), and Sal] (S). The DNA fragments were
separated by electrophoresis through an agarose gel and blotted to a nylon
membrane. The membrane was probed with a 32P-labeled XhoI-Notl
fragment corresponding to a section of ORF].

154

Sal]
.EcoRI RM483 RM26bl
q, 4, ORF 2

o... PLE-

lbp 0.00;;OObOe... 3].an

 

 

B RM10217RM483 RM26B1
B E SIB E s B E s

 

 

     

 

 

 

 

 

 

Figure 4.5

155

corresponding fragment in RM1021 because of the transposon insertion. The
restriction enzyme, E coRI, cuts within Tn5-233 and produces a slightly larger
fragment in RM26bl as would be expected if that strain’s insertion site were 3 ’ to
the insertion site in RM483. The restriction enzyme, Sal], cuts between the
insertion sites of RM483 and RM26bl. Because the DNA fragment used to probe
the blot was complementary to sequences 5’ of the transposon in RM483 and the E“
enzyme cuts 3’ to the transposon, (Figure 4.5A), the Sal] fragment in RM483 .

should be larger than the corresponding fragment in RM 102 1. The Sal] fragment

 

in RM26bl is the same size as the fragment in RM1021 as the enzyme cuts L,
between the insertion site of RM483 and RM26b1. Since the probe is '
complementary to sequences 5’ to the transposon in RM483, the Sal] fragment it
anneals to in RM26bl does not contain the transposon. Southern blot analysis,
therefore, confirms that the transposon insertions are in the same locus.
To determine whether ORF] is a member of a gene family, DNA isolated
from RM 1021 was digested with BamHI, EcoRI, and Sal], and probed in 6X SSC
with a 32P-labeled DNA fragment complementary to ORF]. The blots were
washed in solutions containing 0.1X to 5X SSC. The lowest stringency wash, 5X
SSC, (data not shown), gave the same banding pattern as the highest stringency
wash, 0.1X SSC, indicating that there is most likely a single copy of ORF] in S.

meliloti 1021 (Figure 4.53, RM1021).

156

ORF] encodes a protein similar to RNase E. The ORF containing the
two insertion sites of RM26b1 and RM483 encodes a peptide with significant
similarity to the RNase E proteins of E. coli and H. influenzae. In E. coli, RNase
E has been shown to be an essential endoribonuclease involved in the processing
and degradation of mRNA and rRNA (see review Cohen and McDowall, 1997;
Besseraub et al., 1998). Figure 4.6 shows an alignment of the S. meliloti RNase E r
protein sequence with the E. coli RNase E sequence. The S. meliloti protein is
most similar to the E. coli RNase E protein in the region between residues 25 and

525 of the E. coli protein. This region is the most highly conserved portion of the

 

protein, (Kaberdin et al. 1998), and has been shown to contain the enzyme’s
catalytic domain (McDowall and Cohen, 1996). Within this region, an Sl-like
RNA-binding domain was identified between residues 4] and 122 (Bycroft et al.,
1997). While the S. meliloti protein shows significant similarity to this domain,
this region of sequence homology is interrupted by a non-homologous stretch of
163 residues (Figure 4.6). A second region of E. coli RNase E is required for site-
specific cleavage of several RNA species (McDowall and Cohen, 1996). This
region is located between residues 321 and 498 and is contained within the region
of greatest similarity between the E. coli and S. meliloti RNase E proteins (Figures
4.6 and 4.7).

The carboxy terminal end of the E. coli protein was initially thought to be

required for cell viability (Wang and Cohen, 1994), but has since been shown to

157

 

so ------------------------------------------------------------

96 VIPISGPQHIEIGDKII

 

so --------------------—---—-----------------------------------

156 mmsrnmnnnnnmnnmmurm!

90
216

 

148 : ..:i :L' PESMG OLIVRTAG
276 :.~ :1 ' IOHHuVILRIXG

208
336

261
388

321
448

381
508

441 IA YLLNngng‘
568 - IAHYLLN.7"

500
627

555
661

615
672

675
711

711
771

 

741 _ ..
831 mmmmmnsnfl ‘vax sst-------------

774 AQEIPIQQIINIIDNRDNGGIPRRSBBSPRBLRVBGQRRRRIRDIRIPEQBFIPLTVICA

878 ---------------------------------—-—-—--——-—_-----_--_------

Ifigurelk6

158

 

I. 894 GVVIAPVQVIIPQPIVVITTBPIVIAAIVTIQPQVITISDVINIQIVIIOIIPVVIPQII
8. 889 ------------------------------------------------------------

a . 954 mrmmmsmmwmunm
s. ass ------------------------------------------------------------

I'll”
r’ 9171

    
   

    

I. 1014 RAPIIVPIAPRnir DR
8. 889 ----------- f‘ E‘JllF—«I

AGKGAAGGHIAIHHABAAEARPQPVI

Figure 4.6 (cont'd). Alignment of the E. coli and S. meliloti RNase E
protein sequences. The predicted peptide encoded by ORF 1 was compared
to the database and found to be similar to E. coli RNase E (PZlSl3). An
alignment of the two proteins is shown above, E. = E. coli RNase E and S. = S
meliloti RNase E. Identical regions are shaded in black and similar regions are
shaded in gray. The insertion sites of the transposons in RM483 and RM26b1
are labeled and the affected residue is underlined.

159

 

Oman—<30 moan”:

%////% 5

N8o ace moo 50¢

/a« <

mama

 

 

 

 

 
 

 

 

‘—
u

 

 

 

m5

 

 

 

 

 

 

Ems—.0 ah. 0:55:18: 2. 325:3. :— m. 8:. 2:. h. 3838:. 52»: a 13855.. .56 522.6 m @3858 :63 m. 8:.
9.885 M. 38.8: Ava man @8389 Wammoam 0». 9o M. 38.8: 3385 $5. 33:913. 8 m... 8:. 8.0 Sana—58 23.. m
8:: :50 .58 9-58 52>-Eaa5m ammo: m: m. 8:. mm man—woman 3. «5595 I 23. concaveafism «3:303 5 m.
3850:. .55 ammo: 5 m. 8:. 895.8 m2. So anaoagacofimma 83.5. 8.5 Eo nonomgsaim 89.288 5 m. 33.8:
wan 59889 3. mimosa 8.68 a . .56 flan—50.3.2. 52>-Eafiam ammo: 5 m. 8:. a 5:898; 3. 90 9.8m-
ESE—5m @ . >~m§5¢ior ammosm 5 h. 33:82. v.8 :28 3. manna—nu ». .55 Ema: 25 £38 Spam—om 52:88 So
39:25.. a». 50 c.8588: 533.25.. 5 523mm 25 5 5252: . 888:3?

160

be dispensable. Kido el al. (1996) found mutants in me expressing RNase E
proteins truncated at residue 592 that were fully viable. The carboxy-terminal half
is not well conserved and is thought to serve as a platform for protein-protein
interactions in the assembly of the ribonucleolytic complex termed the
degradosome (Kaberdin et al., 1998). An arginine-rich RNA-binding site is
located within the carboxy-terminal half of the protein between residues 597 and
684 (McDowall and Cohen, 1996). While there is not a corresponding domain in
the S. meliloti RN ase E protein, there are two arginine-rich regions, residues 670-
683 and residues 730-745. The two Tn5-233 transposons in mutants RM483 and
RM26b1 are located at residues 688 and 791, respectively.

ORF2 encodes a novel polypeptide. The second predicted ORF encodes a
small hydrophilic protein 71 amino acids in length. This peptide showed no
significant similarity to any other protein in the database. The pl of the deduced
polypeptide is 4.51.

The rne transcript accumulates transiently after a temperature
downshift. We examined the expression of the me transcript before and after a
temperature downshift. RNA was isolated from RM212, a derivative of RM1021,
at 30°C and 1, 2, 3, 4, and 8 hours after the culture was shifted to 15°C (Figure
4.8). Samples were probed with a 32P-labeled fragment of the me gene. The
largest band on the gel is approximately 3.5 kb in length. The sequence data

predicted a transcript of approximately 2.7 kb if ORFI is transcribed alone or 3.1

161

 

 

Hours at 15°C
2 ,3 4 8

    

 

23S probe

 

 

 

 

Figure 4.8. The me transcript accumulates transiently following a
temperature downshift from 30°C to 15°C. RNA was isolated from
RM212, a derivative of RM1021 containing a promoterless-uidA fusion
recombined into the inositol locus. The strain was grown in TY broth at
30°C then shifted to 15°C. Samples were taken immediately before the
temperature shifi and 1, 2, 3, 4, and 8 hours afterward. A XhoI-NotI DNA
fragment corresponding to a portion of the me gene was labeled with [on-32
P]CTP and used to probe the Northern blot. Subsequently, the blot was
stripped of the original probe and hybridized to 32P-labeled 23$ rDNA to
check sample loading.

162

kb if ORF2 is included in the same transcript. Alternatively, it is possible that
ORFl extends further upstream of the sequenced region. It is not clear from these
data whether ORF] and ORF2 are co-transcribed. While relatively low levels of
the transcript were present at 30°C, the me transcript levels increased dramatically
within 1 hour of the temperature downshift. Levels remained high for 3 hours at
15°C and then decreased significantly at 4 hours. Transcript was barely detectable E—
8 hours after the temperature downshift indicating that the me gene is expressed at g
a lower constitutive level at 15°C than at 30°C.

Effect of Tn5-233 insertions on tax mRNA. Since the Tn5-233 insertions

 

caused increased luminescence at 30°C when transduced into RM603, the level of
luxAB transcript was quantified in these strains. RM603, RM603(483),and
RM603(26b1) were grown in TY broth and the luminescence and the luxAB
mRNA levels measured for each time point. Samples were taken at an A600 of 0.5
and 0.9 during growth at 30°C. As expected, the double mutants, RM603(483)
and RM603(26bl), produced 5 to 6-fold or 7 to 9-fold more light than RM603 at
each time point. The level of luxAB transcript appeared to be elevated at each time
point in RM603(483) and RM603(26bl) (Figure 4.9). Interestingly, the
processing or degradation of the 16S rrn-luxAB message appeared to be inhibited
by the Tn5-233 insertions. At both sampling times, more signal was present in the
higher molecular weight bands in RM603(483) and RM603(26b1) than in RM603.

The increase in luminescence in RM603(26b1) and RM603(483) appears to be due

163

 

RM603 RM603(483)th603(26b1)l
A... 0.5 0.9 0.5 0.9 0.5 0.9|

 

 

 

 

 

~3.5 kb ‘>
~2.5 kb "

 

23S probe

 

Figure 4.9. Effect of me inserts on luxAB transcript levels at 30°C when
transduced into RM603. The two double mutants, RM603(26b1) and
RM603(483), carrying a Tn5-233 insert in me, and RM603 (16S rrn::Tn5-
1062) were grown in TY broth at 30°C. RNA was isolated fiom samples taken
at the following A600: 0.5 and 0.9. 2 pg of each sample was electrophoresed on
a formaldehyde/agarose gel and blotted to a nylon membrane. The samples
were probed with a 32P-labeled Xbal-Pstl, luxAB DNA fragment. The blot was
then stripped and reprobed with 32P-labeled 23S rDNA to check sample
loading.

in part to the increased stability of the rrn-luxAB transcript.

Effect of me mutants on the cspA mRNA. RNase E has been implicated
in the regulation of CspA, one of the major cold shock proteins in E. coli (Fang et
al., 1997). The cspA transcript contains a 5’ untranslated region, UTR, containing
a RNase E target sequence. The presence of this site is thought to contribute to the
extreme instability of this message at 37°C. The homologous transcript in S.
meliloti 1021 also contains a 5’ UTR with a putative RNase E cleavage site
(O’Connell and Thomashow, 1999). In order to determine whether RN ase E is
involved in the regulation of CspA in S. meliloti, the levels of cspA transcript in
RM1021, RM26bl and RM483 were examined during a temperature downshift
from 30°C to 15°C (Figure 4.10). The cspA gene is the first ORF in an operon
containing two additional genes. The lower molecular weight band seen in all
samples represents the cspA gene while the larger molecular weight band seen
immediately following the temperature downshift represents the entire operon
(O'Connell and Thomashow, 1999). Interestingly, the me inserts appeared to have
no effect on the cspA message at 30°C. However, in RM483 and RM26bl at 2-4
hours after the downshift, the cspA transcripts appeared to accumulate to a greater

extent than in the wild type strain.

165

 

 

 

 

15°C
30°C 1 hr 2 hr 4 hr
8"" 6°» .385 e") @‘ gs’” a“ 6&5? 6‘} 6°~ ,3?)
$4396“ «flair? as” «3“ «398’s?

 

 

 

 

 

 

 

Figure 4.10. Effect of me inserts on cspA levels before and after a
temperature downshift. The three strains were grown at 30°C in TY broth
supplemented with either Sm or Sp at SOug/ml. At an A600 of 0.4-0.5, RNA
was isolated and the cultures were shifted to 15°C. Samples were taken 1, 2,
and 4 hours afier the temperature shift. (A) Zug of each RNA sample was run
on a formaldehyde/MOPS/agarose gel and blotted to a nylon membrane. The
membrane was probed with 32P-labeled antisense cspA DNA and exposed to
film. (B) Photograph of agarose gel prior to blotting. Equal amounts of
ethidium bromide were added to each sample prior to loading on the gel with
one exception. The sample indicated by the ** received excess ethidium
bromide.

166

Table 4.1. Bacterial strains

Strains

Description

Sinorhrjgobiym meliloti

RM1021

RM603

RM26b1

RM483

RM603(26b1)

RM603(483)

RM212

Egcherichig coli
DHSor

Sufi; derivative of SU47

KmR; RM1021 containing
Tn5-1062 insert in the 16S
rrnA gene

SpR, GmR/KmR ; RM1021
containing Tn5-233 inserted
in the me gene

SpR,GmR/KmR; RM1021
containing Tn5-233 inserted
in the me gene

SpR,GmR/KmR/NmR; Rm1021
containing 168 rrnA::Tn5-1062
and mezzTn5-233

SpR,GmR/KmR/NmR; Rm1021
containing 168 rrnAzzTn5-1062
and mezzTn5-233

SpR; RM1021 containing the

+1 to +172 fragment of the rrnA
promoter region fiised to the
110MB genes and recombined
into the inositol locus

supE44 hst17 recAI thi-I

Alac U1 69(4) 801acZAMI5)
endA 1 gyrA 96 reIAI

167

Meade et al, 1982

This study

This study

Thisstudy

This study

This study

This study

Hanahan, 1983

Reference or source

 

Table 4.2. Plasmids

Plasmids

pRK607

pAGl .3

pRK2073

Description

SpR,GmR/KmR; plasmid used
to perform Tn5-233 mutagenesis

KmR; 605 bp BgIII/Sacl fragment
containing the rrn promoter,
subcloned from pAGlX7, and
cloned in front of the luxAB genes
in pRK290

SpR; helper plasmid used to

mobilize pAGl .3 into RM1021
strains

168

Reference or source

De Vos et al, 1986

This study

Better et al, 1983

Discussion

In this study, we have identified a Sinorhizobium meliloti 1021 cold-
inducible transcript that encodes a protein with significant similarity to
Ribonuclease E (RNase E) (Figures 4.4 and 4.6). RNase E is encoded by the gene,
me, and has been identified in Escherichia coli, Haemophilus influenzae Rd,
Mycobacterium tuberculosis, Synechocystis sp., and Porphyra purpurea (Kaberdin
et al., 1998). The enzyme has been studied most thoroughly in E. coli and is
thought to be a key enzyme in the degradation and processing of mRN A and rRN A
(see review Cohen and McDowall, 1997; Besseraub et al., 1998). RNase E has
been isolated as a member of a multiprotein complex known as the degradosome
which includes polynucleotide phosphorylase (PNP), DnaK, rhIB RNA helicase,
and enolase (Miczak et al. , 1996). Interestingly, PNP is a cold-inducible protein
in the psychrotroph Yersinia enterocolitica and is required for growth at 5°C
(Goverde et al., 1998). PNP has also been identified as a cold shock protein in E.
coli (Jones et al., 1987). The rhlB RNA helicase is a member of the DEAD-box
family of RNA helicases and has been shown to assist in RNA degradation
through the removal of secondary structures (Py etal., 1996). Another member of
the DEAD-box family of RNA helicases, CsdA, is an E. coli cold shock protein
although this enzyme is associated with the ribosome (Jones et al. , 1996). In S.

meliloti 1021, we have shown that the me transcript levels transiently increase

169

 

 

 

following a temperature downshift from 30°C to 15°C (Figure 4.8). This is an
interesting observation considering the cold-inducible expression of related
proteins in other organisms.

Although the me transcript levels are cold-inducible, the level of RNase E
protein must be examined directly. In E. coli, the translation of mRN A has been
shown to be differentially affected by cold shock. The induction of three cold
shock proteins, CspA, CsdA and IF2, was compared to that of the reporter gene
cat encoding the protein chloramphenicol acetyltransferase (CAT), (Goldenberg er
al. , 1997). When cat was fused to the cspA promoter and UTR, the transcript
levels were induced maximally at 1 hour following a temperature downshift,
similarly to the cspA transcript. The synthesis of the CAT protein, however, was
delayed and did not reflect the levels of cat transcript in the cell. In contrast, the
three cold shock proteins showed peak induction within three hours of the cold
shock. To accurately assess the role that RNase E may play at low temperature,
the level of the protein must be monitored following a temperature downshifi.

The identification of RN ase E as a cold shock protein would emphasize
another facet of low temperature’s effect on cellular physiology and the cold shock
response. While much attention has been paid to the effect low temperature may
have on the secondary structure of mRNA and its effect on translation, less
attention has been paid to its potential efi’ect on RNA processing and degradation.

In E. coli, the understanding of the cold shock response has centered around the

170

adaptation of the ribosome to low temperature (Jones and Inouye, 1996).
Interestingly, the major cold shock proteins, CspA and CsdA, have been shown to
interact with RNA and can destabilize secondary structures (Jiang et al., 1997;
Jones et al., 1996). While destabilizing secondary structure couldimpact
translation of mRNA, it would also influence the degradation of mRNA. The
identification of PNP as not only cold-inducible but essential for growth at low
temperature in Y. enterocolitica emphasizes the importance of message
degradation to cell viability and adaptation to low temperature (Goverde et al.,
1998). The mechanisms required for RNA degradation are also related to
ribosomal RNA processing. Ribosomal RNA processing is inhibited by a cold
shock in yeast and deletion of the cold shock protein, NSRl, has been shown to
further inhibit processing, especially at low temperature (Kondo et al., 1992).
The two rne::Tn5-233 insertion mutants, RM26bl and RM483, were
identified in a screen of Tn5-233 mutagenized S. meliloti 1021. The two inserts
caused increased luminescence from the rrn-quAB fusion, pAGl.3 and the 16S
rrn::Tn5-1062 strain, RM603, at 30°C (Figures 4.2 and 4.3 ). The transposon
insertions are located downstream of the putative S l-like RN A-binding and
endoribonucleolytic domains of RNase E and should lead to the production of
truncated proteins (Figure 4.7 ). Although RNase E is an essential enzyme in E.
coli , Kido et al. (1996) found that RNase E proteins truncated at residue 592,

though exhibiting reduced activity, were still able to support cell growth. A

171

similar situation likely exists in RM483 and RM26b1. Southern blots indicated
that while there is only one me gene in S. meliloti, the insertions in RM483 and
RM26b1 are not lethal. In RM603(483) and in RM603(26bl), the 16S rrn-luxAB
transcript appears to be processed more slowly and to be generally more stable
(Figure 4.9). These data suggest that the truncated RN Ase E proteins, while
retaining essential nucleolytic activities, are functioning at a reduced level.
RNase E has been implicated in the regulation of the cspA transcript in E.
coli (Fang et al., 1997). At the normal incubation temperature of 37°C, the cspA

transcript is very unstable. Mutation of a putative RNase E site in the untranslated

 

leader region or UTR of the cspA gene resulted in the dramatic stabilization of the
message at 37°C. The cspA transcript was also somewhat stabilized in a
temperature-sensitive RNase E mutant at the non-permissive temperature of 44°C.

A similar mechanism may contribute to the regulation of cspA in S. meliloti 1021.

Fusing the cspA UTR to the luxAB genes lowered expression of the luxAB genes at
30°C relative to a direct cspA promoter-luxAB fusion (O’Connell and Thomashow,

1999). As the S. meliloti cspA UTR contains a putative RNase E site, the levels of

cspA were examined in the me insertion mutants. Interestingly, an effect was
detectable at 15°C but not at 30°C (Figure 4.10); the stabilization of the cspA

message appeared to be prolonged in the me insertion mutants. It must be noted

 

that the insertions only partially affect the function of RNase E and the lack of an

effect at 30°C does not mean that full length RNase B does not interact with the

172

cspA mRN A at that temperature.

In sum, we have identified a S. meliloti 1021 homolog of RNase E, an
enzyme involved in mRN A degradation and processing. The rne transcript
accumulates significantly following a temperature downshift suggesting that
RNase E is a cold shock protein in S. meliloti 1021. Although the me gene is
present as a single copy, two transposon insertions 3’ to residue 688 were not
lethal. The truncated RN ase E protein appeared to be affected in its processing of E—
the cspA transcript immediately following a temperature downshift while having 3

no effect on its expression at 30°C. Finally, the identification of RNase E as a

 

cold shock protein suggests that cold shock may have an important influence on

mRNA degradation and processing.

 

173

Materials and Methods

Bacteria and growth conditions. Cultures of Escherichia coli DHSa were
grown aerobically at 37°C in LB broth (Sambrook et al., 1989) or on solid LB
medium containing 1.4% agar. Cultures of Sinorhizobium meliloti were grown
aerobically at 30°C in either tryptone-yeast extract medium, TY (Beringer, 1974), E'

GTS medium (Kiss etal., 1979), BSM medium (Bergersen, 1961), or M9 medium

 

(Sambrook et al., 1989). Cultures were maintained on solid medium containing
1.4% agar, or frozen to -80°C in TY broth containing 15% glycerol. S. meliloti i
was cultured on solid media containing the following antibiotics when appropriate:
kanamycin, Km (200 jig/ml), neomycin, Nm (200 rig/ml), spectinomycin, Sp (100
ng/ml), streptomycin, Sm (250 rig/ml), and tetracycline, Tet (10 uglrnl). E. coli
was grown on plates containing ampicillin, Ap (100 ug/ml), Km (100 rig/ml), Sp
(100 rig/ml), or Tet (10 rig/ml) when appropriate. A lower concentration, 50
jig/ml, of Km and Sm was used in broth cultures.
Transposon mutagenesis and screen. Sinorhizobium meliloti 1021 was
mutagenized with the transposon Tn5-233 via biparental conjugation. The
transposon was carried on the plasmid pRK607. DHSa containing the plasmid

pRK607 was spotted with S. meliloti 1021 on TY agar plates. After 12-24 hours,

these spots were resuspended in 0.5 ml of TY broth, diluted 1 to 10 and 100141

174

spread onto TY agar plates containing Sp (100 rig/ml) and Sm (250 jig/ml). After
three days of incubation at 30°C, 2 mls of 0.85% NaCl was applied to each plate

and the colonies resuspended. The cells were harvested by a brief centrifugation

and resuspended in 1.0 ml of TY broth. 0.5 ml of this suspension was used to
inoculate 2.5 mls of TY broth containing Sp and Sm at 50 ng/ml. After 3-5 hours
of incubation at 30°C, 1.5 mls of the culture was harvested and rinsed with 1 ml of
TY broth. The cells were finally resuspended in only 150 pl of TY broth. 20-25
pl of this suspension was spotted onto a TY agar plate with an equal volume of
Escherichia coli DHSa containing both the reporter plasmid, pAGl .3, and the
helper plasmid, pRK2073. After incubating overnight at 30°C, the spots were
each resuspended in 1.0 ml of TY broth, diluted to various concentrations and
spread on selective TY agar plates containing Nm (200 jig/ml) and Sm (250
jig/ml). These plates were incubated for 3-5 days until colonies were large enough
to be assayed for luciferase activity. Plates were either exposed to aldehyde
immediately and screened for those showing increased luminescence at 30°C or
they were incubated at 15°C for 4-6 hours before screening for colonies showing
decreased luminescence.

Measurement of luciferase activity. Luminescence was measured with the

Hamamatsu Photonic System model Cl966-20. Bacteria grown on agar plates

were exposed to the enzyme substrate, N-decyl aldehyde, for two minutes prior to

175

 

being assayed for light production for 59 seconds. The aldehyde was spread on a
glass petri dish which was placed over the agar plate. The Hamamatsu system
converts the photon counts to a visual image on a monitor which can be used to
quantitate the amount of light produced by each colony.

Bacteria grown in broth culture were assayed quantitatively for light
production. In this assay, the N-decyl aldehyde was dispersed in an aqueous

solution of bovine serum albumin (20 mg/ml) at a concentration of l ul/ml. A 5-

10 ul aliquot of the cell culture was added to 50—100 111 of the aldehyde solution,
vortexed for 30 seconds, then assayed for 60 seconds. The measured
luminescence was reported in arbitrary units, RLU, and adjusted for cell number
by dividing this number by the A600 of the culture. All assays were performed at
room temperature.

(ID-M12 transduction. The Tn5-233 transposons were introduced into
strains already carrying the Tn5-1062 transposon by phage transduction using the
phage CID-M12 (Finan el al., 1984). Preparations of CID-M12 were prepared from
dense cultures of the donor strain. Cultures of the donor strains were grown to late
log phase in LB broth supplemented with 2.5 mM of MgSO4 and CaClz. The
cultures were then inoculated with CID-M 12 to a final concentration of 108 PFU/ml.
After 6-8 hours, the cultures were checked for evidence of lysis and centrifuged to
remove remaining intact cells and debris. The supernatant was collected and 0.05

ml chloroform added per 1.0 ml lysate. The mixtures were vortexed thoroughly

176

and stored overnight at 4°C. The tubes were spun again to pellet the chloroform

and remaining debris. The supernatant was removed and stored at 4°C.

In order to determine the titer of the phage lysates, the lysate was diluted
and mixed 1:10 with a 1-2 day old culture of S. meliloti cultured in LB broth
supplemented with 2.5 mM MgSO4 and CaClz. After incubating for 20 minutes at
30°C, the phage/bacteria mixtures were added to 3 rrrls of top agar [Nutrient Broth
(Difco Laboratories, Detroit, MI) containing 0.65% agar with 2.5 mM MgSO4 and
CaClz] and spread on LB agar plates. Plaques were easily visible after incubating
the plates overnight at 30°C.

The phage lysates prepared from the donor strains were used to transduce
various strains containing Tn5-1062. The recipient strain was cultured in LB broth
supplemented with 2.5 mM MgSO4 and CaClz. After 1-2 days of growth, the
CFU/ml were estimated by measuring the optical density of the culture and an
aliquot mixed with phage lysate such that a ratio of 1 PFU/ 2 CFU was achieved.
0.5 ml of cells was added to a volume of 0.5 ml of phage lysate. Supplemented
LB broth was used to adjust the volume of the phage lysate to achieve the correct
ratio. After incubation at 30°C for 20 minutes, 0.1 ml of 0.2M sodium citrate was
added to each tube in order to chelate the Mg+ and CaH ions and, therefore,
interrupt the infection of the bacterial cells. The cells were then rinsed with a
sterile solution consisting of 0.85% NaCl and 0.01 M sodium citrate. The cells

were resuspended in the saline solution and plated on selective plates, LB agar

177

with Sp (100 ug/ml) and Nm (200 ug/ml).

Cloning, DNA sequencing, and sequence analysis. Genomic DNA was
isolated from 3-5 ml cultures of S. meliloti according to a standard CTAB
extraction protocol (Ausubel et al. , 1987) with one modification. Prior to
extraction with CTAB, the bacterial lysate was passed several times through an
18-gauge needle using a sterile syringe in order to partially shear the DNA. This
step reduced sample viscosity and facilitated extraction of the samples with phenol
and chloroform.

Genomic DNA was digested with either EcoRI or Xmal and Kpnl. EcoRI
and Kpnl cut within the transposon allowing the isolation of either the left end or
the right end of the transposon. Since Tn5-233 has identical ends, this facilitated
sequencing directly out of the transposon. Digested genomic DNA was cloned
into pBluescript SK- and transformed into E. coli DHSa. Colonies resistant to
either kanamycin or spectinomycin were selected depending on which end of the
transposon was cloned.

The resulting plasmids were isolated using a Qiagen column (Qiagen, Inc.,
Valencia, CA) or a standard alkaline lysis method (Sambrook et al., 1989). The
plasmids were sequenced manually using the T7 Sequenase v.2 kit according to
the recommended protocol (Amersham Life Science, Inc., Cleveland, OH) or at the
MSU-DOE-PRL DNA Sequencing facility using the ABI Catalyst 800 for Taq

cycle sequencing and the 373A Sequencer for the analysis of the products. Initial

178

 

sequences were obtained using primers complementary to the end of Tn5. Either
the primer, "Tn5 right", 5'-CACA TGGAATATCAG-3' or the primer, MT93, 5'-
CACGATGAAGAGCAGAA GTTAT-3' were used for the first round of
sequencing. Subsequent primers were chosen using PrimerSelect 3.11
(DNASTAR Inc., Madison, WI). Primers were synthesized by the
Macromolecular Structure Facility located in the Department of Biochemistry at
Michigan State University. Sequence data alignments were carried out using
SeqMan II 3.61 (DNASTAR Inc., Madison, WI).

Potential open reading frames were identified using a codon preference
program, CodonUse 3.1, developed by Conrad Halling (University of Chicago).
Sequences encoding proteins similar to the predicted peptides were identified
using the BLAST algorithms (Altschul et al., 1997). Possible terminator
sequences were located using the Terminator program, (Brendel and Trifonov,
1984), in the GCG sequence analysis package (Wisconsin Package Version 9.1,
Genetics Computer Group, Madison, WI). Protein sequence alignments were
created by using the Match-Box Web Server 1.2 at
http://www.fundp.ac.be/sciences/biologie/bms/matchbox_submit.htrnl

Southern blotting. DNA fragments were separated on an agarose gel and
transferred to a nylon membrane in 20X SSC via capillary action according to the
standard protocol (Sambrook et al. , 1989) with two modifications. Instead of the

acid depurination step, the gel was first exposed to UV light for 10 minutes in the

179

UV Stratalinker 2400 (Stratagene, La J olla, CA). After transfer, the DNA was
crosslinked to the nylon membrane by exposure to 20 mjoules of light energy in
the UV Stratalinker. The blot was hybridized to a 32P-lobeletl probe following a
protocol based on the one described in Sarnbrook etal., 1989. The pre-
hybridization and hybridization solutions contained 6X SSC, 0.5% SDS, and
0.25% nonfat dry milk. Hybridization was performed at 60-65°C in 3 Robbins
Hybridization Incubator Model 400 (Robbins Scientific Corp., Sunnyvale, CA).
The probe was prepared from a band isolated fragment by a random priming
method (Feinberg and Vogelstein, 1983). The blots were washed under standard
conditions four times in a solution of 0. 1X SSC and 0.1% SDS at 60-65°C. The
blot was then exposed to film. For the low stringency washes, the concentration of
SDS was held constant and SSC added to a final concentration of 0.5x, 2.5X or
5.0x. The washes were performed at 60-65°C.

Isolation of RNA. RNA was isolated from S. meliloti strains according to a
modified protocol derived from that found in Ausubel et a1. (1987) for the
isolation of RNA from gram negative bacteria. Bacteria were harvested from a 3-
20 ml aliquot of cell culture. In all but one experiment, the assay of cspA message
levels, rifarnpicin was added to culture samples to a final concentration of 0.2
mg/ml just after harvesting. The cell pellets were resuspended in 1.4 ml of
protoplasting buffer and 80 ul of 50 mg/ml of lysozyme added. The tubes were

incubated on ice 5-10 minutes. The protoplasts were collected by centrifugation

180

for 2 minutes at 7,000 rpm. The pellets were resuspended in 0.5 ml of gram
negative lysing buffer, mixed with 15 ul of diethylpyrocarbonate, DEPC, and
incubated for 5 minutes at 37°C. Tubes were chilled briefly on ice, mixed with

250 pl of 5M NaCl, incubated 10 minutes on ice and centrifuged for 10 nrinutes.
The supernatant was removed, added to a tube containing 1 ml ethanol and stored
overnight at -20°C. The RNA was collected by centrifugation for 15 minutes at
4°C, dried under a vacuum and resuspended in DEPC-treated water. Samples
were quantified using a spectrophotometer and measuring the A260 and A230.
Alternatively, the cell pellets were frozen quickly in liquid nitrogen and the
samples stored at -80°C. RNA was then isolated using RNeasy Mini Kits (Qiagen
Inc., Valencia, CA).

Northern blots. RNA samples were treated with formaldehyde and
electrophoresed on formaldehyde/agarose gels according to standard protocols
(Sambrook et al., 1989). RNA was transferred from the gel to a nylon membrane
in 10X or 20X SSC via capillary action according to the standard protocol
(Sambrook et al., 1989) with one modification. After transfer, the RNA was
crosslinked to the membrane by exposure to 20 mjoules of light energy in the UV
Stratalinker 2400 (Stratagene, La Jolla, CA). The membranes were pre-hybridized
at 42°C in the following solution: 50% forrnamide, 5X SSC, 50mM potassium
phosphate, 5X Denhardt’s solution, 0.5% sodium dodecyl sulfate, and lOOug/rnl

denatured DNA. After a rrrinimum of 2 hours of pre-hybridization treatment, 32P-

181

labeled probe was added to the solution and incubated overnight at 42°C in a
Robbins Hybridization Incubator Model 400 (Robbins Scientific Corp., Sunnyvale,
CA). The membranes were treated first at room temperature with three washes of
2X SSC and 0.5% SDS. After one wash in 0.1X SSC and 0.5% SDS, the
temperature was raised to 50°C for two more washes in the same solution.
Washes were continued until the spent wash solution was no longer detectably
radioactive. The probe was prepared from a band isolated fragment by a random
priming method with one exception.(Feinberg and Vogelstein, 1983). The cspA
antisense probe was prepared from a PCR-amplified fragment using one of the
primers, MT171, 5’-CCGGAATTCAAACGCTCCCTGCCAGTA CATCCG-3’.
This primer is complementary to the 3’ end of cspA and therefore will amplify the
non-coding strand of the cspA gene. The protocol for random labeling was used
with one modification. The randomized oligo was replaced by 11.4 pmoles of the

primer MT171.

182

 

 

Literature Cited

Altschul, S. F., T. L. Madden, A. A. Schfiffer, J. Zhang, Z. Zhang, W. Miller,
and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of
protein database search programs. Nucleic Acids Res. 25: 3389-3402.

Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A.
Smith, and K. Struhl (ed.). 1987. Current protocols in molecular biology. John
Wiley and Sons, NY.

Bae, W., P. G. Jones, and M. Inouye. 1997. CspA, the major cold shock protein
of Escherichia coli, negatively regulates its own gene expression. J. Bacteriol.
179(22):7081-7088.

Barlow, T., M. Berkmen, d. Georgellis, L. Bayr, S. Arvidson, and A. von
Gabain. 1998. RNase E, the major player in mRNA degradation, is down-
regulated in Escherichia coli during a transient growth retardation (diauxic lag).
Biol. Chem. 379:33-38.

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

Bessarab, D. A., V. R. Kaberdin, C. Wei, G. Lion, and S. Lin-Chao. 1998.
RNA components of Escherichia coli degradosome: Evidence for rRN A decay.
Proc. Natl. Acad. Sci. USA 95:3157-3161.

Better, M., and D. R. Helinski. 1983. Isolation and characterization of the recA
gene of Rhizobium meliloti. J. Bacteriol. 155(1):311-316.

Brandi, A., P. Pietroni, C. O. Gualerzi, and C. L. Pon. 1996. Post-
transcriptional regulation of CspA expression in Escherichia coli. Mol.
Micriobiol. 19(2):231-240.

Brendel, V. and E. N. Trifonov. 1984. A computer algorithm for testing
prokaryotic terminators. Nucl. Acids Res. 12:4411-4427.

Bycroft, M., T. J. P. Hubbard, M. Proctor, S. M. V. Freund, and A. G.

Murzin. 1997. The solution structure of the 81 RNA binding domain: a member
of an ancient nucleic acid-binding fold. Cell. 88:235-242.

183

Cohen, S. N., and K. J. McDowall. 1997. RNase E: still a wonderfully
mysterious enzyme. Mol. Microbiol. 23(6): 1099-1 106.

De Vos, G. F., G. C. Walker, and E. R. Signer. 1986. Genetic manipulations in
Rhizobium meliloti utilizing two new transposon Tn5 derivatives. Mol. Gen.
Genet. 204:485-491.

Fang, L., W. Jiang, W. Bae, and M. Inouye. 1997. Promoter-independent cold-
shock induction of cspA and its derepression at 37°C by mRN A stabilization.
Mol. Microbiol. 23:355-364.

Feinberg, A. P., and B. Vogelstein. 1983. A technique for radiolabeling DNA
restriction endonuclease fragments to high specific activity. Anal. Biochem.
132(1):6-13.

Finan, T. M., E. Haartwieg, K. Lemieux, K. Bergman, G. Walker and E.
Signer. 1984. General transduction in Rhizobium meliloti. J. Bacteriol. 1592120-
124.

Goldenberg, D., I. Azar, A. B. Oppenheim, A. Brandi, C. L. Pon, and C. 0.
Gualerzi. 1997. Role of Escherichia coli cspA promoter sequences and
adaptation of translational apparatus in the cold shock response. Mol. Gen. Genet.
256:282-290.

Goverde, R. L. J., J. H. J. Huis in’t Veld, J. G. Kusters and F. R. Mooi. 1998.
The psychrotrophic bacterium Yersinia enterocolitica requires expression of pnp,

the gene for polynucleotide phosphorylase, for growth at low temperature (5°C).
Mol. Microbiol. 28(3):555-569.

Hanahan, D. 1983. Studies on the transformation of Escherichia coli with
plasmids. J. Mol. Biol. 66:557-580.

Jain, C. and J. G. Belasco. 1995. RNase E autoregulates its synthesis by
controlling the degradation rate of its own mRN A in Escherichia coli: unusual
sensitivity of the me transcript to RNase B activity. Genes & Dev. 9:84-96.

Jiang, W., Y. Hou, and M. Inouye. 1997. CspA, the major cold-shock protein of
Escherichia coli, is an RNA chaperone. J. Biol. Chem. 272(1):196-202.

Jones, P. G. and M. Inouye. 1996. RbfA, a 308 ribosomal binding factor, is a
cold-shock protein whose absence triggers the cold-shock response. Mol.
Microbiol. 21(6): 1207-1218.

184

Jones, P. G., R. A. VanBogelen, and F. C. Neidhart. 1987. Induction of
proteins in response to low temperature in Escherichia coli. J. Bacteriol.
169(5):2092-2095.

Jones, P. G., M. Cashel, G. Glaser, and F. C. Neidhart. 1992. Function of a
relaxed—like state following temperature downshifts in Escherichia coli. J.
Bacteriol. 174(12):3903-3914.

Jones, P. G., M. Mitta, Y. Kim, W. Jiang, and M. Inouye. 1996. Cold shock
induces a major ribosomal-associated protein that unwinds double-stranded RNA
in Escherichia coli. Proc. Natl. Acad. Sci. USA 93276-80.

Kaberdin, V. R., A. Miczak, J. S. Jakobsen, S. Lin-Chao, K. J. McDowall, and
A. von Gabain. 1998. The endoribonucleolytic N-terrninal half of Escherichia
coli RN ase E is evolutionarily conserved in Synechocystis sp. and other bacteria
but not the C-terminal half, which is sufficient for degradosome assembly. Proc.
Natl. Acad. Sci USA 95:11637-11642.

Kido, M., K. Yamanaka, T. Mitani, H. Niki, T. Ogura, and S. Hiraga. 1996.
RNase E polypeptides lacking a carboxyl-terminal half suppress a mukB mutation
in Escherichia coli. J. Bacteriol. 178:3917-3925.

Kondo, K., L. R. Z. Kowalski, and M. Inouye. 1992. Cold shock induction of
yeast NSRl protein and its role in pre-rRNA processing. J. Biol. Chem.
267(23): 16259- 16265.

McDowall, K. J. and S. N. Cohen. 1996. The N-terminal domain of the me gene
product has RN ase B activity and is non-overlapping with the arginine-rich RNA-
binding site. J. Mol. Biol. 255:349-355.

Miczak, A., V. R. Kaberdin, C. Wei, and S. Lin-Chao. 1996. Proteins
associated with RN ase E in a multicomponent ribonucleolytic complex. Proc.
Natl. Acad. Sci. USA 93:3865-3869.

Mitta, M., L. Fang and M. Inouye. 1997. Deletion analysis of cspA of
Escherichia coli: requirement of the AT-rich UP element for cspA transcription
and the downstream box in the coding region for its cold shock induction. Mol.
Microbiol. 26(2):321-335.

185

O’Connell, K. P. and M. F. Thomashow. 1999. Transcriptional organization
and regulation of a polycistronic cold shock operon in Sinorhizobium meliloti
RM1021 that includes the major cold shock gene cspA and a homolog of
ribosomal protein gene rpsU. (in preparation)

Py, B., C. F. Higgins, H. M. Krisch, and A. J. Carpousis. 1996. A DEAD-box
RNA helicase in the Escherichia coli RNA degradosome. Nature 381:169-172.

Sarnbrook, J., E. F. F ritsch, and T. Maniatis. 1989. Molecular cloning: a
laboratory manual, 2nd edition. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, NY.

Thieringer, H. A., P. G. Jones, and M. Inouye. 1998. Cold shock and
adaptation. BioEssays 20:49-57.

Van Bogelen, R. A. and F. C. Neidhardt. 1990. Ribosomes as sensors of heat
and cold shock in Escherichia coli. Proc. Natl. Acad. Sci. USA. 87:5589-5593.

Wang, M. and S. N. Cohen. 1994. ard-I: a human gene that reverses the effects
of temperature-sensitive and deletion mutations in the Escherichia coli rne gene
and encodes an activity producing RN ase E-like cleavages. Proc. Natl. Acad. Sci.
USA 91:10591-10595.

186

 

APPENDICES

187

Appendix A

Auxotrophic mutants cause increased luminescence of rrn-luxAB fusions
at 30°C

Introduction

Tn5-233 insertion mutants were tested for their ability to alter the
luminescence of plasmid, pAGl.3, at 30°C and 15°C. The plasmid, pAGl.3,
contains the luxAB genes of Vibrio harveyi fused to the rrnA promoter at +187 and
exhibits increased luminescence following a temperature downshift from 30°C to
15°C. Thirteen Tn5-233 insertion mutants that caused increased luminescence at
30°C were isolated. Initial characterization of the mutants revealed that nine

mutants were unable to grow on minimal medium. Further analysis of the nine

auxotrophic mutants is described below.

Results and Discussion

The nine auxotrophic mutants causing increased luminescence from
pAGl.3 required addition of either uracil or arginine, or both, for growth on
minimal medium (Table A. 1). The biosynthetic pathways for uracil and arginine
are connected through the use of carbamoyl phosphate as a substrate (Figure A. 1).

Carbarnoyl phosphate is synthesized for each pathway by a single enzyme,

188

Table A.l. List of auxotrophic Tn5 -233 insertional mutants.

 

 

 

 

 

Mutant AuxotrOphy Location of Tn5 -233
strains insertion
RM75 uracil pyrB
RM6Sa uracil/arginine carAB
RM945 uracil/arginine carAB
RM] 5a uracil --
RM35 uracil --
RM835 uracil --
RM26a uracil -—
RM522 uracil -—
RM65b arginine ---

 

189

 

Figure A.1. Arginine and pyrimidine biosynthesis in Escherichia coli. The
synthesis of L-arginine from L-glutamate is shown on the left of the figure.
Synthesis of the pyrimidine nucleotides from carbamoyl phosphate and
aspartate is shown on the right. The reaction of carbamoyl phosphate
synthetase is shown at the top of the figure as carbamoyl phosphate is used in
both pathways. The genes affected by Tn5-233 inserts are shown in italics at
the step in biosynthesis performed by the enymes they encode. PRPP = 5-
phosphoribosyl-l-pyrophosphate. This figure is adapted from G. Gottschalk,
fictenfiMemmlism , 2nd Edition, Springer-Verlag, New York, 1986.

190

The

and
rare
1 in
s at

L-glutamine + HCO3' + 2ATP

 

carAB
L- glutamate *carbamoyl phosphate +
aspartate
acetyl-CoA
C CoA pyrB r P‘
carbamoyl aspartate
N-acetyl glutamate

l

 

P

4,5-dihydroorotate

 

N—acetylglutamate—
y—semialdehyde

glutamate 'c NAD+
C a-oxoglutarate NADH+H+

N-acetylornithine orotate

orotidine monophosphate

 

 
  

 

L-omithine
*carbamoyl phosphate
C co.
Pi
L-citrulline uridine monophosphate
aspartate + ATP
Ami-PP, 2ATP ZADP-l-Pi
L-argininosuccinate uridine triphosphate
NH3+ ~ ATP
fumarate ' C
ADP + P,
L-arginine cytidine triphosphate

Figure A.l

191

carbamoyl phosphate synthetase. The transposon insertion site in the uracil
auxotroph, RM75, was identified by cloning out one end of the transposon and the
neighboring genomic sequences. Sequence analysis indicated that the transposon
was located within an ORF encoding a protein showing significant similarity to
aspartate carbamoyltransferase in Synechocystis PCC6803 (Figure A.2). In E. coli,
this protein is the second enzyme in the pyrimidine biosynthetic pathway and is
encoded by the pyrB gene (Figure A. 1). Two other mutants, RM65a and RM945,
required both arginine and uracil for growth on a minimal medium. In S. meliloti,
auxotrophs requiring both arginine and uracil for growth have been complemented
by either the carA or the carB loci (Kerppola and Kahn, 1988). These loci encode
the two subunits of the enzyme carbarnoylphosphate synthetase (Figure A.2).

In order to correlate the presence of the Tn5-233 transposon with the
observed increase in luminescence, each insert was transduced into strain RM603
(16S rrnAzzTn5-1062) using the phage <D-M12. Luminescence was increased in
all of the double mutants relative to RM603 when the strains were cultured on TY
agar plates (Figure A.3). When cultured in TY broth, the relative increase in
luminescence by the double mutants was not constant over the growth curve but
became more pronounced as the cultures approached stationary phase. Figure A.4
shows the relative luminescence of six double mutants during growth in TY broth
at 30°C. Five of the strains showed a greater than 8-fold increase in luminescence

relative to RM603 at the final time point as opposed to a less than 4-fold increase

192

Figure A.2. The Tn5-233 insert in RM75 is located in a gene similar to the
pyrB gene of synechocystis PCC6803. (A) The DNA continuous with one
end of the transposon in RM75 was cloned and sequenced. The DNA
sequence is shown and the one letter amino acid codes of the predicted peptide
are shown in italics above. (B) An alignment of the S. meliloti peptide, Sm,
and the Synechocystis PCC6803 aspartate transcarbamoylase (P74163), Sy, is
shown. Identical residues are shaded in black and similar residues in gray.

193

«mu
.dr one
DH
8W8
eflm
SJ

A

1 57
T Q I N L F F E A S T R T Q S S F E L
ACGCAGATCAATCTGTTCTTTGAAGCCTCGACCCGTACCCAGTCCTCATTCGAGCTC
58 114
A G K R L G A D V' Mr N' M’ S V G .N S S V
GCCGGAAAGCGGCTCGGCGCCGACGTCATGAACATGTCCGTCGGCAATTCCTCGGTG
115 171
X’ K G E T L I D T A .M T L N' A M’ X P D
AANAAAGGCGAGACGCTGATCGATACGGCGATGACGCTGAACGCCATGCNCCCGGAC
172 228
V L V V .R H S S A G A A S L L A X'.K V
GTGCTGGTCGTTCGCCACTCGTCGGCGGGGGCTGCCTCGCTGCTTGCCCANAAGGTC
229 285
S C S V V .N .A G D X’ X .X E H P T Q A L
TCCTGCTCGGTCGTCAACGCCGGCGACNGCCANCNCGAGCACCCGACNCAGGCGCTG
286 339
L X .X X D X’ P A R P X' A S S X’ R D H
CTCNACNCCNCTGACNATCCGGCGCGCCCTANGGCNAGCTCTCNGCGGGATCAT

B

Sy l MEMVISITRNBILDLTDIRGIELDIVLQEAITFQQVLBGQTKKVEALQGQ

   
   

’STRTWSSFELAEKRL:

i _ DVMN
~STRTqSSFELA$KRLe-IVMN.

  
     

Sy 51
Sm 1

    

Sy 101
Sm 50

 

Sy 151 I EAPDNRAIQCLQGKKIAVVGDILHSRVIRSNBIBLTIAGAD
Sm 90 AR? RDH

S y 2 O 1 WPMMLMGSMCMIQPMGADW
S y 2 5 1 MWSIMMITHDRLKVCQWPGWISSILK

Sy 301 nomrsuonqv'rsmrsrmmmcmn

Figure A.2

194

Figure A.3. Set of Tn5-233 inserts that caused auxotrophy and increased
luminescence when transduced into RM603. RM603 (16S rrn::Tn5-1062)
was compared to a collection of double mutants. Each double mutant is a
derivative of RM603 containing a Tn5-233 insert. Cultures were streaked on
TY agar plates and incubated at 30°C for two days before being assayed for
luminescence. After being assayed for luminescence at 30°C , the plates were
moved to 15°C for 5-6 hours then assayed again.

195

mud
- 1 062)
11 M
ed 011
ed for
s were

Figure A.3

30°C

\
\
x
\
x
\
\
\
\

 

RM603

RM603(65A)

RM603(945)

RM603(75)

RM603(15A)

RM603(35)

RM603(522)

RM603(835)

RM603(26A)

RM603(65B)

196

15°C

//

 

\
x
‘5
x
\
\
\
\

    

0 14 I RM603

E 12 RM603(26A)

§ 1: Er RM603(15A)

E, 6 RM603(65A)

E 4 l RM603(65B)

i 2 a RM603(75)
0 3 I RM603(945)

0.1-0.2 0.4-0.6 0.8-1.0

A600

Figure A.4. Luminescence of auxotrophic double mutants increases with
culture density. The luminescence of RM603 (16S rrn::Tn5-1062) was
compared to that of six double mutants. These double mutants are derivatives
of RM603 containing an additional Tn5-233 insert that disrupts the arginine
and/or pyrimidine biosynthetic pathways. All strains were cultured at 30°C in
TY broth. Samples were taken when the cultures reached an A600 of 0.1-0.2,
0.4-0.6, and 0.8-1.0 and assayed for luminescence. Luminescence was reported
as RLU or relative light units and adjusted for cell density by dividing the RLU
by the A600 of the sample. Relative luminescence was calculated by dividing
the RLU/A600 of each double mutant by the RLU/A600 of RM603.

197

at the earlier time point. RM603 (26A) showed a smaller effect on luminescence
overall; however, luminescence of this strain relative to RM603 increased from
1.3 to 3.9-fold at an A600 of approximately 1.0 (Figure A.4).

Because the difference in luminescence between the double mutants and
RM603 increased as the cultures approached saturation, it was thought that uracil
may have become limiting in the growth medium at the higher cell density and that
this limitation might be required for an effect on luminescence to be observed. If
this were true, supplementing the TY medium with uracil should delay the effect
on luminescence in the double mutants. To test this idea the pyrB mutant double
mutant, RM603(75), and RM603 were grown in TY broth with and without added
uracil at 30°C (Figure A.5). When the cultures not supplemented with uracil were
sampled at an A600 of 0.1-0.2, RM603(75) produced 2.7 times more light than
RM603. At an A600 of 08-09, the difference in luminescence had increased to
82-fold. In cultures supplemented with uracil, RM603(75) produced only 2.7
times more light than strain RM603 at the higher culture density. These results
indicate that the amount of uracil in the growth medium influences luminescence
in pyrB mutant, RM603(75).

The results obtained with this mutagenesis are quite interesting in relation
to the model for growth rate-dependent regulation of the rm promoters proposed
by Gaal et al. (1997). In Escherichia coli, the open complexes formed by RNA

polymerase at the rrn Pl promoters are unusually sensitive to the concentration of

198

 

I I A6000.1-0.2 A6000.8-0.9

 

12000

 

E

 

 

 

 

 

 

 

RLU/Am x 103
8
8

 

 

 

 

 

RM603 RM603(75) RM603 RM603(75)
+ uracil + uracil

Figure A.5. Effect of added uracil on luminescence in RM603(75). The
luminescence of RM603 (16S rrn::Tn5-1062) was compared to that of
RM603(75), a derivative of RM603 containing a Tn5-233 insert in pyrB. The
strains were grown at 30°C in either TY broth or TY supplemented with uracil
(20 rig/ml). Samples were taken at A600 of 0.1-0.2 and again at A600 of 0.8-0.9
and luminescence measured. Luminescence is reported as relative light units,
RLU, and is adjusted to cell density by dividing RLU by the A600 of the culture.
Each bar represents the average of three flasks and error bars show the standard
deviation.

199

the initiating nucleotide. In E. coli all but one of the rrn promoters initiates
transcription with ATP. Because the concentration of ATP varies depending on
nutrient conditions and its rate of consumption within the cell, it could function as
a signal molecule. Gaal et al. (1997) demonstrated in vivo and in vitro that rrn P1
promoter activity varies positively with ATP concentration. Further, the rrn
promoters were shown to respond specifically to ATP concentration through the
use of mutants defective in pyrimidine biosynthesis. When these mutants were
grown under partial pyrimidine limiting conditions, ATP concentration was no
longer correlated with growth rate. P1 rrn promoter activity increased with higher
levels of ATP although growth rate was decreasing. While sensitivity of the S.
meliloti 1021 rrn promoters to the initiating nucleotide has not been described, it is
interesting that all three rrn transcripts begin with adenine. Thus, the S. meliloti
1021 pyrimidine auxotrophs might be expected to increase the activity of the rrn-
luxAB fusions when uracil becomes limiting in the growth medium.

In order to further investigate the apparent increase in rm expression in a
pyrimidine auxotroph, RM603(75) and RM603 were grown in TY broth at 30°C
and RNA isolated at A600 of 0.5 and 0.9. Surprisingly, the luxAB transcript
accumulated only slightly more in RM603(75) than in RM603 (Figure A.6).

While this result could indicate that the pyrimidine auxotrophy simply has a
general effect on luciferase activity, previous observations of the pyrB insertion do

not support this conclusion. The Tn5-233 insertion in RM75 was transduced into

200

 

RM603 RM603(75)
0.5 0.9 0.5 0.9

 

~2.5 kb 9

 

238 probe

 

Figure A.6 Effect of pyrB::Tn5-233 insert on luxAB transcript levels at
30°C when transduced into RM603. The double mutant, RM603(75), and
RM603 were grown in TY broth at 30C. RNA was isolated from samples taken
at an A600 of 0.5 and 0.9. 2 pg of each sample was electrophoresed on a
formaldehyde/agarose gel and blotted to a nylon membrane. The samples were
probed with a 32P-labeled Xbal-Psi], luxAB DNA fi'agment. The blot was then
stripped and reprobed with 32P-labeled 23S rDNA to check for sample loading.

201

 

three different Tn5-1062 insertion mutants to test the specificity of its effect on
luminescence (data not shown). Although the insertion did increase the
luminescence in these strains, the increase in luminescence was 2-3 fold as
opposed to the 12-fold increase observed in RM603(75) in the same experiment.
Apparently, the pyrB insertion is affecting some post-transcriptional step in the
expression of the RM603 Tn5-1062 insertion.

The experiments presented here do not directly address the growth rate-
dependent regulation of the rm promoters in S. meliloti 1021. Growth rate-

dependent regulation of rrn expression in E. coli was studied using only the core

 

promoter sequences of the rrn P1 promoter (-41 to +1) while the rrn-quAB fusions
in this study included the full promoter and over 1 kb of 16S rm. While the
initiation of rrn promoter activity was shown to be influenced by the ATP
concentration alone, other factors could negate this activity resulting in no net
increase in rRNA. If this regulation is to be examined in S. meliloti, rm promoter

fusions at +1 must be used instead of the Tn5-1062 insertions used in this study.

Materials and Methods

Bacteria and growth conditions. Cultures of Escherichia coli DHSa were
grown aerobically at 37°C in LB broth (Sambrook et al., 1989) or on solid LB
medium containing 1.4% agar. Cultures of Sinorhizobium meliloti were grown

aerobically at 30°C in either tryptone-yeast extract medium, TY (Beringer, 1974),

202

 

GTS medium (Kiss etal., 1979), BSM medium (Bergersen, 1961), or M9 medium
(Sambrook et al., 1989). Cultures were maintained on solid medium containing
1.4% agar, or frozen to -80°C in TY broth containing 15% glycerol. S. meliloti
was cultured on solid media containing the following antibiotics when appropriate:
kanamycin, Km (200 ug/ml), neomycin, Nm (200 ug/ml), spectinomycin, Sp (100
rig/ml), streptomycin, Sm (250 ug/ml), and tetracycline, Tet (10 rig/ml). E. coli
was grown on plates containing ampicillin, Ap (100 ug/ml), Km (100 rig/ml), Sp
(100 ug/ml), or Tet (10 ug/ml) when appropriate. A lower concentration, 50
rig/ml, of Km and Sm was used in broth cultures.

Determination of auxotrophy. Each mutant strain was initially tested for
auxotrophy by streaking on a minimal medium agar plate containing Sp
(100ug/ml). Two different minimal media were used, M9 or BSM. If the mutant
strain was not capable of growth on a minimal medium, the strain was
complemented with various combinations of anrino acids, purines, pyrimidines and
vitamins. Ten different pools of supplements were added to agar plates of the
minimal medium, M9. Each component was present in two of the pools in
different combinations so that if growth occurred in the presence of two pools,
those pools should have only one component in common. The concentrations of
amino acids used to prepare the media were as recommended, (Cutting and
Youngman, 1994), with two exceptions: cysteine, lOOug/ml, and aspartic acid,
100 ug/ml. The pyrimidines and purines were added as recommended with two

203

 

exceptions: uracil, 40 ug/ml, and guanine was not tested. Thiamin was added to a
final concentration of Zug/ml. The mutants, 75 and 15A, were tested only with
cytosine and uracil. Because these compounds were able to support growth of
these strains in a minimal medium, no other factors were tested. The mutant
strains were tested either as the original mutant strain or as RM603 transductants.
Because RM603 is not an auxotroph, any auxotrophy of the transduced strain
could be attributed to the Tn5-233 insert.

41M 1 2 transduction. The Tn5-233 transposons were introduced into
strains already carrying the Tn5-1062 transposon by phage transduction using the
phage CD-M12 (Finan et al., 1984). Preparations of (ID-M12 were prepared from
dense cultures of the donor strain. Cultures of the donor strains were grown to late
log phase in LB broth supplemented with 2.5 mM of MgSO4 and CaClz. The
cultures were then inoculated with (D-M 12 to a final concentration of 108 PFU/ml.
After 6-8 hours, the cultures were checked for evidence of lysis and centrifuged to
remove remaining intact cells and debris. The supernatant was collected and 0.05
ml chloroform added per 1.0 ml lysate. The mixtures were vortexed thoroughly
and stored overnight at 4°C. The tubes were spun again to pellet the chloroform
and remaining debris. The supernatant was removed and stored at 4°C.

In order to determine the titer of the phage lysates, the lysate was diluted
and mixed 1:10 with a 1-2 day old culture of S. meliloti cultured in LB broth

supplemented with 2.5 mM MgSO4 and CaClz. After incubating for 20 minutes at

204

 

 

30°C, the phage/bacteria mixtures were added to 3 rrrls of top agar [Nutrient Broth
(Difco Laboratories, Detroit, MI) containing 0.65% agar with 2.5 mM MgSO., and
CaClz] and spread on LB agar plates. Plaques were easily visible after incubating
the plates overnight at 30°C.

The phage lysates prepared from the donor strains were used to transduce
various strains containing Tn5-1062. The recipient strain was cultured in LB broth
supplemented with 2.5 mM MgSO.; and CaClz. After 1-2 days of growth, the
CFU/ml were estimated by measruing the optical density of the culture and an
aliquot mixed with phage lysate such that a ratio of 1 PFU/ 2 CFU was achieved.
0.5 ml of cells was added to a volume of 0.5 ml of phage lysate. Supplemented
LB broth was used to adjust the volume of the phage lysate to achieve the correct
ratio. After incubation at 30°C for 20 minutes, 0.1 ml of 0.2M sodium citrate was
added to each tube in order to chelate the Mg+ and CaH ions and, therefore,
interrupt the infection of the bacterial cells. The cells were then rinsed with a
sterile solution consisting of 0.85% NaCl and 0.01 M sodium citrate. The cells
were resuspended in the saline solution and plated on selective plates, LB agar
with Sp (100 rig/ml) and Nm (200 pig/ml).

Measurement of luciferase activity. Luminescence was measured with the
Hamamatsu Photonic System model C 1966-20. Bacteria grown on agar plates
were exposed to the enzyme substrate, N-decyl aldehyde, for two minutes prior to

being assayed for light production for 59 seconds. The aldehyde was spread on a

205

 

glass petri dish which was placed over the agar plate. The Hamamatsu system
converts the photon counts to a visual image on a monitor which can be used to
quantitate the amount of light produced by each colony.

Bacteria grown in broth culture were assayed quantitatively for light
production. In this assay, the N-decyl aldehyde was dispersed in an aqueous

solution of bovine serum albumin (20 mg/ml) at a concentration of l Ill/ml. A 5-

10 ul aliquot of the cell culture was added to 50-100 u] of the aldehyde solution,
vortexed for 30 seconds, then assayed for 60 seconds. The measured
luminescence was reported in arbitrary units, RLU, and adjusted for cell number
by dividing this number by the A600 of the culture. All assays were performed at
room temperature.

Cloning, DNA sequencing, and sequence analysis. Genomic DNA was
isolated from 3-5 ml cultures of S. meliloti according to a standard CTAB
extraction protocol (Ausubel et al., 1987) with one modification. Prior to
extraction with CTAB, the bacterial lysate was passed several times through an
18-gauge needle using a sterile syringe in order to partially shear the DNA. This
step reduced sample viscosity and facilitated extraction of the samples with phenol
and chloroform.

Genomic DNA was digested with either E coRI or Xmal and Kpnl. E coRI
and Kpnl cut within the transposon allowing the isolation of either the left end or

the right end of the transposon. Since Tn5-233 has identical ends, this facilitated

206

 

sequencing directly out of the transposon. Digested genomic DNA was cloned
into pBluescript SK— and transformed into E. coli DHSa. Colonies resistant to
either kanamycin or spectinomycin were selected depending on which end of the
transposon was cloned.

The resulting plasmid was isolated using a Qiagen column (Qiagen, Inc.,
Valencia, CA) or a standard alkaline lysis method (Sambrook et al. , 1989). The
plasmids were sequenced manually using the T7 Sequenase v.2 kit according to
the recommended protocol (Amersham Life Science, Inc., Cleveland, OH) or at the
MSU-DOE-PRL DNA Sequencing facility using the ABI Catalyst 800 for Taq
cycle sequencing and the 373A Sequencer for the analysis of the products. Initial
sequences were obtained using primers complementary to the end of Tn5. Either
the primer, "Tn5 right", 5'-CACA TGGAATATCAG-B' or the primer, MT93, 5'-
CACGATGAAGAGCAGAA GTTAT-3' were used for the first round of
sequencing. Subsequent primers were chosen using PrimerSelect 3.11
(DNASTAR Inc., Madison, WI). Primers were synthesized by the
Macromolecular Structure Facility located in the Department of Biocherrristry at
Michigan State University. Sequence data alignments were carried out using
SeqMan H 3.61 (DNASTAR Inc., Madison, WI).

Potential open reading frames were identified using a codon preference
program, CodonUse 3.1, developed by Conrad Halling (University of Chicago).

Sequences encoding proteins similar to the predicted peptides were identified

207

 

using the BLAST algorithms (Altschul et al., 1997). Protein sequence alignments
were created by using the Match-Box Web Server 1.2 at
http://www.fundp.ac.be/sciences/biologie/bms/matchbox_submit.htrnl

Uracil complementation experiment. RM603(75) contains a Tn5-233 insert
in a gene homologous to pyrB and is auxotrophic for pyrimidine synthesis. This
auxotrophy can be complemented by supplementing the growth medium with
uracil. In order to examine the influence of uracil limitation on luminescence in

RM603(75), RM603 and RM603(75) were grown at 30°C in TY broth with and

without added uracil. TY broth containing streptomycin at 250 ug/ml was
inoculated with a single colony of either RM603 or RM603(75). After two days
of incubation at 30°C, each culture was used to inoculate six flasks of TY broth.
Three of the six flasks were supplemented with uracil at a concentration of 20
ug/ml. The cultures were grown with shaking at 30°C. Luciferase activity was
assayed when the cultures reached an A600 of approximately 0.1 and again at an
A600 of 0.8 to 0.9. Three measurements were taken at each time point from each
flask. These values were averaged and adjusted for culture density by dividing by
the A600. The measurements from each set of flasks were then averaged and

represented on the graph.

208

Literature Cited

Altschul, S. F., T. L. Madden, A. A. Schfiffer, J. Zhang, Z. Zhang, W. Miller,
and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of
protein database search programs. Nucleic Acids Res. 25: 3389-3402.

Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A.
Smith, and K. Struhl (ed.). 1987. Current protocols in molecular biology. John
Wiley and Sons, NY.

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

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

Cutting, S. M. and P. Youngman. 1994. Gene transfer in gram-positive bacteria,
p. 348-364. In P. Gerhardt, R. G. e. Murray, W. A. Wood and N. R. Krieg (ed),
Methods for general and molecular bacteriology. American Society for
Microbiology, Washington, DC.

Finan, T. M., E. Haartwieg, K. Lemieux, K. Bergman, G. Walker and E.
Signer. 1984. General transduction in Rhizobium meliloti. J. Bacteriol. 159:120—
124.

Gaal, T., M. S. Bartlett, W. Ross, C. L. Turnbough Jr., and R. L. Gourse.
1997. Transcription regulation by initiating NTP concentration: rRNA synthesis
in bacteria. Science. 278:2092-2097.

Kerppola, T. K. and M. L. Kahn. 1988. Genetic analysis of
carbarnoylphosphate synthesis in Rhizobium meliloti 104A14. J Gen. Microbiol.
134:921-929.

Kiss, G. B., E. Vincze, Z. Kalman, T. Forrai, and A. Kondorosi. 1979. Genetic
and biochemical analysis of mutants affected in nitrate reduction in Rhizobium
meliloti nodule bacteria, nitrogen fixation. J. Gen. Microbiol. 113:105-118.

Sarnbrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a

laboratory manual, 2nd edition. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, NY.

209

 

 

Appendix B

Preliminary characterization of Tn5-233 insertions causing decreased
luminescence of a rrn-luxAB fusion at 15°C

Introduction

Sinorhizobium meliloti 1021 was mutagenized with the transposon, Tn5-
233, in order to identify factors involved with the low temperature regulation of
cold shock loci. Transposon insertion mutants were tested for their ability to alter
the luminescence of one of two luxAB reporter plasmids, pAGl.3 and pKOC1.l, at
30°C and 15°C. The plasmid, pAGl.3, contains the luxAB genes of Vibrio harveyi
fused to the rrnA promoter at +187 and shows increased luminescence following a
temperature downshift. The plasmid, pKOC1.1, contains the luxAB genes inserted
downstream of the S. meliloti cold-induced gene, cspA (O’Connell and
Thomashow, 1999). The preliminary characterization of the six Tn5-233 insertion
mutants that decreased luminescence from either of the reporter plasmids, pAGl .3

or pKOC1.1, is described below.

Results and Discussion
Six Tn5-233 insertion mutants that decreased the luminescence from

pAGl.3 or pKOC1.1 at 15°C were isolated. To correlate the Tn5-233 insertions

210

 

with the observed decrease in luminescence and to examine the effect of all the
inserts on a rrn::Tn5-1062 insertion, each Tn5-233 insertion was transduced into
strain RM603 (16S rrnAzzTn5-1062) using phage (D-MIZ (Figure B. 1). While
luminescence was less in each of the double mutants relative to RM603 at 15°C,
two of the Tn5-233 insertions, RMDll and RMAE5, also caused decreased
luminescence at 30°C (Figure B. 1). Interestingly, the insertion in RM603(AE5)
showed a temperature-specific effect on growth rate. The growth of RMAE5 was
compared to RM1021 at 30°C and 15°C in TY broth. While RMAE5 grew only

4% slower than RM1021 at 30°C, the difference in growth rates increased to 43%

at 15°C. This result suggests that the transposon insertion has interfered with a
cellular process that becomes more important at low temperature.

Although the inserts in RM42 and RM546 cause decreased luminescence at
15°C (Figure B. l), the mutants were originally isolated because they increased the
luminescence from pAGl.3 at 30°C. These Tn5-233 inserts also increased
luminescence when transduced into RM603 and grown on TY agar (data not
shown). However, this relative increase in light production at 30°C was only
observed under the original assay conditions and not when the colonies were
exposed continuously to aldehyde as in the assay shown in Figure 8.1. When
RM603(42) and RM603(546), were grown in TY broth at 30°C, no significant
increase in light production was observed relative to the original strain RM603. In

fact, when RM603(42) was compared to RM603 during growth at 30°C in TY

211

 

 

 

RM603 RM603 RM603 RM603 RM603 RM603 RM603
(42) (546) (Q9) (T77) (AE5) (D11)

 

Figure B.1. Set of Tn5-233 inserts that decreased luminescence at 15°C
when transduced into RM603. RM603 (16S rrn::Tn5-1062) was compared to
a collection of double mutants. Each double mutant is a derivative of RM603
containing an additional Tn5-233 insert. Cultures were streaked on TY agar
plates and incubated at 30°C for two days before being assayed for
luminescence. After being assayed for luminescence at 30°C, the plates were
moved to 15°C for 5-6 hours then assayed again.

212

 

broth, luxAB mRNA levels were reduced in the double mutant at the three culture
densities measured, A600 of 0.2, 0.5 and 0.9 (Figure 8.2).

The Tn5-233 insertion mutants, RMQ-9 and RMT-77, were isolated
because of their effect on the luminescence of the plasmid pKOCl.1 at 15°C.
When transduced into RM603, these inserts also reduced luminescence of the 16S
rrnA::Tn5-1062 insert at 15°C (Figure B. 1). Interestingly, RM603(Q9) produced
light at levels comparable to that produced by RM603(42) and RM603(546) at

both 30°C and 15°C (Figure B. l). The double mutant RM603(T77) also showed

decreased luminescence at 15°C but not to the same extent as the other three
strains. When RM603(Q9) and RM603(T77) were grown in TY broth culture and
exposed to a temperature downshift, the Tn5-233 inserts’ effect on luminescence
was different than that observed during growth on agar plates (Figure 8.1 and
8.3). In RM603(Q9), the Tn5-233 insert had no effect on luminescence at either
temperature. The Tn5-233 insert in RM603(T77) appeared to decrease
luminescence at 30°C while luminescence gradually increased after the
temperature downshift approaching the levels observed in RM603 after 3 hours.
The effect of this insert on luminescence during growth on plates was essentially
opposite to that observed in broth culture. Although RM42, RM546, RMQ9, and
RMT77, showed consistent effects on luminescence when transduced into RM603
and assayed on plates, further study of these mutants would be difficult without

reproducing these effects in broth cultures.

213

Figure B.2. Effect of RM42 insert on luminescence of 16S rrn::Tn5-1062
in RM603. (A) RM603 and RM603(42) were grown in TY broth at 30°C and
sampled at A600 of 0.2, 0.5, and 0.9. RNA was isolated, separated on a
denaturing formaldehyde/agarose gel, transferred to a nylon membrane and
probed with a radiolabeled DNA fragment complementary to the luxAB
message. (B) Luminescence was measured at the same time as the above
samples were taken. Relative luminescence was calculated by dividing the
RLU/A600 of RM603(42) by the RLU/A600 of RM603. Each bar represents the
average relative luminescence calculated from two experiments.

214

 

RM603 RM603(42)
A600 0.2 0.5 0.9 0.2 0.5 0.9

 

 

 

 

 

 

 

 

 

 

B I l RM603 RM603(42) |
8
fl
8
a
.5
a
a
3
'5
a
0
a:

0.1-0.2 0.5 0.9

A
Figure B.2 coo

215

Figure B.3. Effect of Tn5-233 inserts in RMQ9 and RMT77 on
luminescence of 16S rrn::Tn5-1062 in RM603. (A) RM603 and
RM603(Q9) were grown in TY broth at 30°C then shifted to 15°C.
Luminescence was measured immediately before the shift and 1, 2, and 3 hours
afterward. Relative luminescence was calculated by dividing the RLU/A600 of
RM603(Q9) by the RLU/A600 of RM603. Each bar represents the average
relative luminescence from two experiments. (B) RM603 and RM603(T77)
were grown in TY broth at 30°C then shifted to 15°C. Luminescence was
measured immediately before the shift and 1, 2, and 3 hours afterward.
Relative luminescence was calculated by dividing the RLU/A600 of
RM603(T77) by the RLU/A600 of RM603.

216

 

 

 

I l RM603 RM603(Q9) I

 

 

 

 

Hours at 15C

 

[ I RM603 RM603(T77)]

 

 

 

 

 

 

 

 

Hours at 15°C

Figure B.3

217

Materials and Methods

Bacteria and growth conditions. Cultures of Escherichia coli DHSa were
grown aerobically at 37°C in LB broth (Sambrook et al., 1989) or on solid LB
medium containing 1.4% agar. Cultures of Sinorhizobium meliloti were grown
aerobically at 30°C in tryptone-yeast extract medium, TY (Beringer, 1974).
Cultures were maintained on solid medium containing 1.4% agar, or frozen to -
80°C in TY broth containing 15% glycerol. S. meliloti was cultured on solid
media containing the following antibiotics when appropriate: kanamycin, Km (200
rig/ml), neomycin, Nm (200 ug/ml), spectinomycin, Sp (100 ug/ml), streptomycin,
Sm (250 ug/ml), and tetracycline, Tet (10 ug/ml). E. coli was grown on plates
containing ampicillin, Ap (100 ug/ml), Km (100 ug/ml), Sp (100 ug/ml), or Tet
(10 rig/ml) when appropriate. A lower concentration, 50 rig/ml, of Km and Sm
was used in broth cultures.

(Ii-M12 transduction. The Tn5-233 transposons were introduced into
strains already carrying the Tn5-1062 transposon by phage transduction using the
phage (ID-M12 (F inan et al., 1984). Preparations of CD-M12 were prepared from
dense cultures of the donor strain. Cultures of the donor strains were grown to late
log phase in LB broth supplemented with 2.5 mM of MgSO4 and CaClz. The
cultures were then inoculated with CD-M12 to a final concentration of 108 PFU/ml.

After 6-8 hours, the cultures were checked for evidence of lysis and centrifuged to

218

 

remove remaining intact cells and debris. The supernatant was collected and 0.05
ml chloroform added per 1.0 ml lysate. The mixtures were vortexed thoroughly

and stored overnight at 4°C. The tubes were spun again to pellet the chloroform

and remaining debris. The supernatant was removed and stored at 4°C.

In order to determine the titer of the phage lysates, the lysate was diluted
and mixed 1:10 with a 1-2 day old culture of S. meliloti cultured in LB broth
supplemented with 2.5 mM MgSO4 and CaClz. After incubating for 20 minutes at
30°C, the phage/bacteria mixtures were added to 3 mls of top agar [Nutrient Broth
(Difco Laboratories, Detroit, MI) containing 0.65% agar with 2.5 mM MgSO4 and
CaClz] and spread on LB agar plates. Plaques were easily visible after incubating
the plates overnight at 30°C.

The phage lysates prepared from the donor strains were used to transduce
various strains containing Tn5-1062. The recipient strain was cultured in LB broth
supplemented with 2.5 mM MgSO4 and CaClz. After 1-2 days of growth, the
CFU/ml were estimated by measming the optical density of the culture and an
aliquot mixed with phage lysate such that a ratio of 1 PFU/ 2 CFU was achieved.
0.5 ml of cells was added to a volume of 0.5 ml of phage lysate. Supplemented
LB broth was used to adjust the volume of the phage lysate to achieve the correct
ratio. After incubation at 30°C for 20 minutes, 0.1 ml of 0.2M sodium citrate was
added to each tube in order to chelate the Mg+ and CaH ions and, therefore,

interrupt the infection of the bacterial cells. The cells were then rinsed with a

219

sterile solution consisting of 0.85% NaCl and 0.01 M sodium citrate. The cells
were resuspended in the saline solution and plated on selective plates, LB agar
with Sp (100 ug/ml) and Nm (200 ug/ml).

Measurement of luciferase activity. Luminescence was measured with the
Hamamatsu Photonic System model Cl966-20. Bacteria grown on agar plates
were exposed to the enzyme substrate, N-decyl aldehyde, for two minutes prior to
being assayed for light production for 59 seconds. The aldehyde was spread on a

glass petri dish which was placed over the agar plate. The Hamamatsu system

 

converts the photon counts to a visual image on a monitor which can be used to
quantitate the amount of light produced by each colony.

Bacteria grown in broth culture were assayed quantitatively for light
production. In this assay, the N-decyl aldehyde was dispersed in an aqueous

solution of bovine serum albumin (20 mg/ml) at a concentration of 1 ul/ml. A 5-

10 u] aliquot of the cell culture was added to 50-100 111 of the aldehyde solution,
vortexed for 30 seconds, then assayed for 60 seconds. The measured
luminescence was reported in arbitrary units, RLU, and adjusted for cell number
by dividing this number by the A600 of the culture. All assays were performed at
room temperature.

Cloning, DNA sequencing, and sequence analysis. Genomic DNA was
isolated from 3-5 ml cultures of S. meliloti according to a standard CTAB

extraction protocol (Ausubel et al. , 1987) with one modification. Prior to

220

extraction with CTAB, the bacterial lysate was passed several times through an
18-gauge needle using a sterile syringe in order to partially shear the DNA. This
step reduced sample viscosity and facilitated extraction of the samples with phenol
and chloroform.

Genomic DNA was digested with Xmal and Kpnl. Kpnl cuts within the
transposon allowing the isolation of the right end of the transposon. Since Tn5-
233 has identical ends, this facilitated sequencing directly out of the transposon.
Digested genomic DNA was cloned into pBluescript SK- and transformed into E.
coli DHSor. Colonies resistant to spectinomycin were selected.

The resulting plasmid was isolated using a Qiagen column (Qiagen, Inc.,
Valencia, CA) or a standard alkaline lysis method (Sambrook et al., 1989). The
plasmids were sequenced manually using the T7 Sequenase v.2 kit according to
the recommended protocol (Amersham Life Science, Inc., Cleveland, OH) or at the
MSU-DOE-PRL DNA Sequencing facility using the ABI Catalyst 800 for Taq
cycle sequencing and the 373A Sequencer for the analysis of the products. Initial
sequences were obtained using primers complementary to the end of Tn5. Either
the primer, "Tn5 right", 5'-CACA TGGAATATCAG-3' or the primer, MT93, 5’-
CACGATGAAGAGCAGAA GTTAT-3' were used for the first round of
sequencing. Subsequent primers were chosen using PrimerSelect 3.11
(DNASTAR Inc., Madison, WI). Primers were synthesized by the

Macromolecular Structure Facility located in the Department of Biochemistry at

221

 

Michigan State University. Sequence data alignments were carried out using
SeqMan II 3.61 (DNASTAR Inc., Madison, WI).

Potential open reading frames were identified using a codon preference
program, CodonUse 3.1, developed by Conrad Halling (University of Chicago).
Sequences encoding proteins similar to the predicted peptides were identified
using the BLAST algorithms (Altschul et al., 1997). Protein sequence alignments
were created by using the Match-Box Web Server 1.2 at

http://www.fundp.ac.be/sciences/biologie/bms/matchbox_subrnit.htrnl

222

 

Literature Cited

Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller,
and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of
protein database search programs. Nucleic Acids Res. 25: 3389-3402.

Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A.
Smith, and K. Struhl (ed.). 1987. Current protocols in molecular biology. John
Wiley and Sons, NY.

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

Finan, T. M., E. Haartwieg, K. Lemieux, K. Bergman, G. Walker and E.
Signer. 1984. General transduction in Rhizobium meliloti. J. Bacteriol. 159:120-
124.

O’Connell, K. P. and M. F. Thomashow. 1999. Transcriptional organization
and regulation of a polycistronic cold shock operon in Sinorhizobium meliloti
RM 1021 that includes the major cold shock gene cspA and a homolog of
ribosomal protein gene rpsU. (in preparation)

Sarnbrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a

laboratory manual, 2nd edition. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, NY.

223

 

Appendix C

Nucleotide sequence of the promoter and upstream regions of the three
rm operons in Sinrhizobium meliloti 1021

Sequence of rrnA upstream region obtained from cosmid 6H7.

A putative ORF is indicated by the capitalized sequences. Translation is predicted
to proceed from nucleotide 390 toward nucleotide 1. The deduced polypeptide
shows no significant similarity to any protein in the database. The -10 and -35 rrn
promoter sequences are shown in bold type.

GGCAAGCGCN
TCTCGCCGTC
GGCTTGGCGA
GCCGACCGGC
CCAGTTGCTG
AGACTTGAGG
GATCTCGCGC
GGTCGCCAAG
ggtcccctaa
gtgtgcatct
gacgccgcgc
attgtgaaga
ctcgaaaaag
agaaaactgt
gggtctataa

.ATCATGACAT
ANGGATCATG
CGATGTTGGG
GCCGACGGAT
CCGGCCGTCG
TGATGCCGGT
CGGTCCGGTG
CGAACGAAGC
atgctctgcc
tccgtatacg
ctccggaacg
agtgagggcg
aatctgcgac
catttttttt
gcccgatca

GCACGAAGGG
GTCGAGACTT
GGAAAGAACG
CGAGCGCGGC
AGCGCCGCAA
CAAGAACGTG
CGCGACGGCC
ATGTTCTTAT
gtatgcttat
gaagtcaaat
aagcggcgat
aaattatgta
ccgcaaaacc
gaagaagcct

224

CTGGCGGCGC
CCATCACGGT
CGATCTCCCC
AAAGGCTTCC
AGAGCGCGAC
CCCGAGAGCC
GTCCGCTAGG
CGCTGCTCAT
cgtcgatacg
cgcagcgcag
gatctgggtt
tccacaggct
ctgcttttcc
gttgacgtgt

ACCACCTCCT
CTTGTCGGCT
GTTTGGCGGC
GCCGGAATCG
GCCCATCAGA
AGCGCATCGA
ATCGGCGGGT
gcaggctgga
ttgcgcgaag
ggcaggcccc
agatcggtcg
gagggagcgg
gggctttgtc
tggagggctq

50
100
150
200
250
300
350
400
450
500
550
600
650
700
719

 

Sequence of rrnB upstream region obtained from cosmid 8D2.

A putative ORF is indicated by the capitalized sequences. Translation is predicted
to proceed from nucleotide 224 to nucleotide 577. The deduced polypeptide

sequence shows limited similarity to several transcriptional regulators. The -10
and -35 rm promoter sequences are shown in bold type.

tctcgcttga
gaggcggata
tgccgaagtg
tctcgccggt
gcttcgtcgg
CTCGGCGAGA
CCTCACCCAG
CGCACTTCAT
GAAATCGAGC
CGCGATACCC
TCGAAAACCG
CCTCGGCGGG
aaacgccggg
gccaggtttc
cagtcactta

tggagggctg

gctcgtggcg
tcgcggtagc
ctgacgccct
cagccgccga
ttatatcgac
TACAGCCGGG
TTGAAGGCGG
GGCGAAGGAC
TGAAACGGAC
AGGGTGAGGG
CGGCTGCTTC
TTCTTGAACG
cctttccgag
cggtccggct
cgattttctg
gggtctataa

atccaacaga
gctcgccgca
atacgctgca
tccgcacacg
gacATGATCT
GCTGCGCGCC
TGCGCCAGGG
GAACCCAACC
CTATTGGCTG
CCGTACGGGA
CTGCGTGAGA
TCAATAGggc
gcgcgagcgt
ttcggatgca
ctgatttttt
gcccgatca

225

tcatgtcttt
ccgaaagccg
tgtctatggg
cagcgatctc
TCACGCCGGG
CATTTTCAAA
ATTAGGGCTT
TCGAGATAGT
ATCTGCCACC
TTTCCTGGTG
CGCCGTTCGC
gttaaccata
cccgaaaagg
agaataaaca
gaaaatcgtt

gtcgcgcaag
gagcttacca
cacgcagcta
ggctcccacc
GCTCGATTAT
GCTCGAGCAT
TGCGTGCTCC
GCTGCCGGAT
GCGATCTGAT
GATGCCGCCG
TTTGACCGGT
taggcgaggc
atgccgagcc
cctttaaaaa
gttgacgtgt

50
100
150
200
250
300
350
400
450
500
550
600
650
700
750
779

 

Sequence of rrnC upstream region obtained from cosmid 5C4.

A putative ORF is indicated by the capitalized sequences. Translation is predicted
to proceed from nucleotide 372 toward nucleotide 1. The deduced polypeptide
sequence shows 61% similarity over 97 amino acids to a predicted ORF in
Escherichia coli (AE000270). The -10 and -35 rm promoter sequences are shown

in bold type.

GCCGGGCGNG
CGTTGATCCA
TTTTCCGCGA
ATGGACGACG
TGATATAGGC
GTCTGGCCGC
GGCGGACGCC
TCTTCCCTGC
ttgccgccga
ctgctgccgt
gagcggcaat
gaaaggcgga
aagcccgcaa
gtgctttgga
ttttttgaaa
gatca

TAAAGCAGCC
CACGAGATCG
GCACGGTCGC
GTGACGCCGT
ATTGACGTTC
TCGCCTCCGC
CTGCCGGCAA
CCGCCTGATC
gggtaagaca
cacgtttcgg
gggcttccag
actgcgttaa
agcggacccg
aatggccctt
atcgttgttg

CTNGGCGTTT
ACCGATCCGC
GGTCGCCTTA
AGCGCGCCTT
TCGGAGCCGC
GGCGACCGCC
GGCCGAGCGC
ATgacgctcc
agggcggtct
tgaaacctgc
agctaggaat
ccatataggt
tattggtagc
taaaaacagt
acgtgttgga

226

NATCGCGGCG
CCTCGTCCTT
GCGGTGTCGT
CATTTCATCG
CCCAGGCGTT
TCCCAGTTCC
CAGCAGCGCT
tctacccccg
gcgtcaatgc
tctggtcgcc
tcgcgtcacg
ctcatcccca
agcgtcggac
cacttacgat
gggctggggt

AAGTTCTCGC
TCCTGCGGAT
CGAGCTTCAC
CCGGCCCATC
GAAATAGACC
CAGCGTCCGT
GCGGCGATGA
ccacatgccg
acgcggccgc
tctttcgcgg
cagggccctt
gaagcgccga
gtcttttcgc
tttctgctga
ctataagccc

50
100
150
200
250
300
350
400
450
500
550
600
650
700
750
755