h $.39- svltffliul
{a ‘Li 3 .
III)!!!"
I! as?

. v
. u. a. . a
vxwmufib. sun}

4....

. w 1..
F5”. ....

9...... 500:.
, «Pd .2...
”F‘Kw
a

113:0: ..

xr. .2
. iii

\1,
.ifiix :

nu:

.. £413....
. . . iii!
..
, . .x
ELY 1:,
:1?
32:3
1 . .i .u... .

z
. T aft

63...! it... . L. .17
5.4. .. :...zx:.)u‘.

ti: {I 3;. 1)
£11.17.

.
b
‘1

5... .
1.5.7.3..

' Adi

 

... nil 11.
. v

 

 

 

|ES

\\\\ii\'ii\ii“\i\’\"\ \~. \\\ .\ Wit

 

 

LIBRARY -“
Michigan State
University

 

 

 

This is to certify that the
dissertation entitled
Genetic Analysis of Wax Ester and Tri-

acylglycerol Biosynthesis in Acinetobacter
calcoaceticus Strain BD413

 

 

presented by

Steven E. Reiser

has been accepted towards fulfillment
of the requirements for

Ph.D- degree in Biochemistry

 

Major pro essor

Date—344mg—

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

r

PLACE ll RETURN BOX to remove We checkout from your record.
TO AVOID FINES return on or before dete due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Ie An Afflrmetive Action/Emu Opportunity lnetltwon
Wm:

——-_._..__

GENETIC ANALYSIS OF WAX ESTER AND TRIACYLGLYCEROL
BIOSYNTHESIS IN ACHVETOBACTER CALCOACETIC US STRAIN BD413

By

Steven Edward Reiser

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

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1996

ABSTRACT

GENETIC ANALYSIS OF WAX ESTER AND TRIACYLGLYCEROL
BIOSYNTHESIS IN ACINETOBACTER CALCOACETICUS STRAIN BD413

By

Steven Edward Reiser

The phenomena I have investigated is the accumulation of wax esters and
triacylglycerol in the form of intracellular inclusions in the gram negative aerobic
bacterium Acinetobacter calcoaceticus strain BD413. This strain of A. calcoaceticus
accumulates both wax esters and triacylglycerol as a means of carbon storage when it
undergoes nutrient starvation.

Mutants of strain BD413 were induced using both chemical and transposon
mutagenesis methods. By screening colonies for wax accumulation by staining with
the lipophilic dye, Sudan black B, followed by TLC, a total of 8 transposon mutants
and 21 chemically induced mutants were isolated. These mutants were separated into
3 general categories: wax'tag”, wax+tag’ and wax'tag'.

Nutritional supplementation experiments on the wax’tag“ mutant, WowlS,
indicated that this mutants genetic lesion was an inability to catalyze the conversion of
acyl-CoA to fatty aldehyde. Complementation of this mutant with a cosmid from a
cosmid genomic library allowed the identification of an open reading frame encoding
the wowlS gene. The open reading frame shared considerable similarity to an open
reading frame described in Mycobacterium tuberculosis called ORF 2. It is believed
that ORF 2 acts as a B-keto reductase involved in mycolic acid biosynthesis based on

its similarity to a known B-keto reductase.

By expressing the identified open reading frame from A. calcoaceticus in E.
coli, and assaying for the encoded protein's activity, the genes enzymatic activity was
identified. The gene product was observed to catalyze the formation of fatty alcohol
from acyl-CoA via a fatty aldehyde intermediate. The isolated gene has been named
acrl for acyl-CoA reductase.

A second aspect of the research described herein concerns mutants isolated
following transposon mutagenesis with mini-TnlOPtth. Flanking sequence
surrounding a transposon insertion in strain ll-C7, a wax"tag' mutant, was cloned by
inverse PCR (IPCR). DNA sequencing of this DNA allowed the identification of an
open reading frame that shares considerable homology to gInE, glutamate-ammonia-
ligase adenylyltransferase, from E. coli. This gene is involved in the regulation of
glutamine synthetase (GS), an important enzyme involved in amino acid biosynthesis.
Examination of the open reading frames found near the disrupted gInE gene indicated
that one of these had significant homology to a branched-chain—amino-acid
transaminase and a third open reading frame that did not share homology to anything
in GenBank, release 92.0. It seems likely that the putative mutation in this line
disrupts the ability of the mutant to respond properly to nitrogen concentrations

surrounding it.

Copyright by
Steven Edward Reiser
1996

I dedicate this work to my wife, Sharon. What is that elusive "something" that allows
us to love each other forever? You are my effervescent soda bubble, my risotto, in
short, my Boopie. This is also for my parents, I love you both. I also dedicate this to
rockin' music from the likes of Pearl Jam and the Violent Femmes. Sometimes the

situation calls for music with an attitude.

ACKNOWLEDGMENTS

I would like to thank Dr. Chris Somerville for his tutelage and for providing an
environment for independent thought and growth. I really enjoyed our conversations
about things like TIGR, biotechnology and the like. Looking at the future through
your eyes has been very exciting. I would also like to thank Dr. Christoph Benning,
whose guidance as a mentor helped this project really get off the ground. You are a
good friend. I also owe thanks to Dr. Jan Jaworski who helped me put some
Biochemistry in my Biochemistry degree. Thank you, Amy Stroven who endeavored
to screen through thousands of bacterial colonies looking for mutants, and then went
back for more to see if they were complemented. I would like to thank Dr. Joe Ogas
for providing the pileup which was used to clone a partial acyltransferase. Additional
thanks goes to Dr. Michele Nikoloff who took time out to review this manuscript.
Special thanks to my committee members Dr. Green, Dr. Kaguni, Dr. McIntosh, Dr.
Ohlroggee and Dr. Zeikus for providing guidance and their time. This work was
supported by the United States Department of Energy grant DE-FG02-94-ER-20133.

Thank you DOE!

vi

RESULTS AND DISSCUSION .............................. 70

CONCLUSIONS ........................................ 106
REFERENCES ......................................... 108
CHAPTER 4: ISOLATION AND CHARACTERIZATION OF TRANSPOSON
MUTANTS FROM A. CALCOACETICUS STRAIN BD413 .......... Ill
ABSTRACT ........................................... 111
INTRODUCTION ....................................... 1 12
MATERIALS AND METHODS ............................. 114
RESULTS AND DISCUSSION .............................. 126
CONCLUSION ......................................... 141
REFERENCES ......................................... 145
CHAPTER 5: CONCLUSIONS AND PERSPECTIVES .................. 146

APPENDIX A: DNA SEQUENCE INFORMATION AND CONTIG MAPS . . . . 155

APPENDIX B: CONSTRUCTION OF A TRANSCRIPTIONAL EXPRESSION

VECTOR FOR A. CALCOACETICUS AND E. COLI ............... 174
REFERENCES ......................................... 188
APPENDIX C: CLONING OF A PARTIAL DNA FRAGMENT OF SN-l
ACYLTRANSFERASE FROM A. CALCOACETICUS . . . ........... 189
REFERENCES ......................................... 194

viii

LIST OF TABLES

Table 1-1: A comparison of some naturally occuring wax esters ............. 3
Table 1-2: Relative concentrations of neutral lipids in hexadecane inclusions. . . . . 9
Table 1-3: Lipid composition of intracellular inclusions isolated from

hexadecanol and hexadecane-grown Acinetobacter sp. strain HOl-N ...... 9
Table 1-4: Quantitation of Acinetobacter sp. HOl-N cellular lipids .......... 12
Table 1-5: Quantitation of Acinetobacter sp. HOI-N extracellular lipids ....... 12
Table 1-6: Energy yields in moles of ATP per mole of energy source ........ 14
Table 1-7: Relative changes in the amounts of aliphatic compounds comprising

the epicuticular waxes of 20 eceriferum (cer) mutants compared to wild

type levels ............................................ 15
Table 2-1: Bacterial strains fi'om Chapter 2 .......................... 25
Table 2-2: Effect of inclusion of hexadecane (0.3% w/v) and hexadecanol

(0.3% w/v) on neutral lipid composition of various mutants .......... 41
Table 3-1: Bacterial strains used in Chapter 3 ......................... 53
Table 3-2: Plasmid sources and derivations for Chapter 3 ................ 55
Table 3-3: List of synthetic oligonucleotides used in Chapter 3 ............. 63
Table 3-4: Cosmids that share homology to 1A-3F ..................... 74
Table 4-1: Bacterial strains used in Chapter 4 ......................... 115
Table 4-2: Plasmid sources and derivations for Chapter 4 ................ 117
Table 4-3: List of synthetic oligonucleotides used in this Chapter 4 .......... 123

ix

Table B-1: Plasmid sources and derivations for Appendix B ............... 175

Table B-2: B-Galactasidase content of cells transformed with pSER120
constructs ............................................. 181

Table C-l: Synthetic degenerate oligonucleotides used in Appendix C ........ 191

LIST OF FIGURES

Figure 1-1: Structure of a generic wax ester ........................... 2

Figure 1-2: Proposed pathway for wax ester biosynthesis in Acinetobacter
calcoaceticus strain BD413 .................................. 6

Figure 2-1: A comparison of wax ester accumulation from four strains of A.
calcoaceticus ........................................... 3 1

Figure 2-2: Effect of duration of NTG treatment on cell viability over time
following NTG mutagenesis and induced number of Rif colonies ...... 33

Figure 2-3: Example of staining for mutants affected in neutral lipid
accumulation using the lipophilic dye Sudan black B ............... 35

Figure 2-4: TLC plate illustrating the three general classes of mutants ........ 36

Figure 2-5: Qualitative TLC analysis of the mutants isolated following NTG
mutagenesis ........................................... 37

Figure 2-6: Proposed pathway for wax ester biosynthesis in Acinetobacter
calcoaceticus strain BD413 ................................. 39

Figure 2-7: Chemical complementation of class I and class III mutants grown
under nitrogen starvation (wax inducing) conditions in the presence of

0.3% hexadecanol ....................................... 40
Figure 2-8: Chemical complementation of the WowlS mutant .............. 43
Figure 3-1: Detailed restriction map of pLA2917 (taken from Allen, 1985) . . . . 71
Figure 3-2: Complementation of Wowl ............................. 73
Figure 3-3: Restriction map of cosmid 4A-55 and locations of the transposon

insertions ............................................. 76
Figure 3-4: Restriction map of cosmid lA-3F ......................... 77

xi

Figure 3-5: EcoRV digestions of 4A-55 and transposon containing derivatives . . 80

Figure 3-6: Southern analysis of transposon mutagenized cosmids ........... 81
Figure 3-7: Map showing sequence ID#1-9 in respect to one another ......... 82
Figure 3-8: DNA and protein sequence of the region containing acrl ........ 85
Figure 3-9: Optimized FASTA alignment between acrland ORF2 ........... 87

Figure 3-10: Three-way alignment of acrl (ACAR1.AMI), ORF2 from
Mycobacterium tuberculosis (ORF2.AMI) and ActIII from Streptomyces

cinnamonensis (ACT3.AMI) ................................ 89
Figure 3-11: Optimized FASTA alignment between acrl and ActIII ......... 90
Figure 3-12: Pr0posed pathway for wax ester biosynthesis in Acinetobacter

calcoaceticus strain BD413 ................................. 91
Figure 3-13: Complementation of the mutant WowIS with pSER2zacr1 ....... 93
Figure 3-14: Kyte and Doolittle plots of the protein sequence from acrl ...... 94
Figure 3-15: SDS-PAGE gel showing protein induction .................. 96
Figure 3-16: In vitro acyl-CoA reductase assay ........................ 97
Figure 3-17: Demonstration of an in vitro assay showing reductase activity of

acrl with different radiolabelled substrates ...................... 98
Figure 3-18: Cofactor dependence of acyl-CoA reductase ................. 99
Figure 3-19: In vitro acyl-CoA reductase timecourse .................... 101

Figure 3-20: In vitro enzyme assay testing for reductase activity using 1-"C-
palmitoyl aldehyde ....................................... 102

Figure 3-21: Secondary structure prediction for acrland conserved amino acid
residues specific to the family of short chain alcohol dehydrogenases . . . . 103

Figure 3-22: In vitro acyl-CoA reductase assay carried out in the presence of
unlabelled cis-l l-hexadecenal ............................... 105

Figure 4-1: Positional mapping of the BglII fragment in the cosmid 5A-Fl . . . . 125

xii

Figure 4-2:

Southern blot analysis of 9 random colonies following mutagenesis

with mini-TnIOPtth transposon ............................. 127

Figure 4-3:

Southern analysis of isolated mutants carrying mini-TnlOPtth . . . . 129

Figure 4-4: Southern analysis of genomic DNA from 30-F10 and 11-C7 ...... 131
Figure 4-5: Extended IPCR of 11-C7 and 30-F10 ...................... 132
Figure 4-6: Southern analysis of 11-C7 with pSERIO ................... 133
Figure 4-7: Southern analysis of 5A-F1 cosmid DNA ................... 135
Figure 4-8: Southern analysis of 11-C7 with pSRll .................... 137
Figure 4—9: Map showing sequence ID#10-14 in respect to one another ....... 138
Figure 5-1: DNA Sequence of Arabidopsis EST T21872 ................. 149
Figure 5-2: Optimized FASTA alignment between Acrl and the Arabidopsis

EST, T21872 ........................................... 150
Figure A-l: Sequence ID#I ..................................... 156
Figure A-2: Sequence ID#2 ..................................... 157
Figure A-3: Sequence ID#3 ..................................... 158
Figure A-4: Sequence ID#4 ..................................... 159
Figure A-5: Sequence ID#5 ..................................... 159
Figure A-6: Sequence ID#6 ..................................... 160
Figure A-7: Sequence ID#7 ..................................... 161
Figure A-8: Sequence ID#8 ..................................... 162
Figure A-9: Sequence ID#9 ..................................... 163
Figure A-10: Sequence ID#10 ................................... 164
Figure A-11: Sequence ID#ll ................................... 166
Figure A-12: Sequence ID#12 ................................... 167

Figure A-13: Sequence ID#13 ................................... 168
Figure A-14: Sequence ID#14 ................................... 169
Figure A-15: Sequencing contigs of pSR6 ........................... 170
Figure A-16: Sequencing contigs of pSR2 ........................... 171
Figure A-17: Sequencing contigs of pSRll .......................... 172
Figure A-18: Sequencing contigs of pSR12 .......................... 173
Figure B-l: Restriction maps of pBR322, pWH1274, pSERI and pSER2 ...... 176

Figure B-2: Illustration of the steps taken to constuct pSER200-1 and
pSER200-4 transcriptional expression vectors for use in A. calcoaceticus . 178

Figure B-3: Sequence ID#15 .................................... 183

Figure B-4: Sequence ID#16 .................................... 183

Figure B-5: Best known restriction map of A. calcoaceticus/E. coli
transcriptional expression vector pSER200-1 ..................... 184

Figure B-6: Results of BLASTX alignment using sequence ID#17 as a query

sequence .............................................. 186
Figure B-7: Sequence ID#17 .................................... 186
Figure B-8: Best known restriction map of A. calcoaceticus/E. coli

transcriptional expression vector pSER200-4 ..................... 187
Figure C-l: Pileup of known and putative acyltransferases ................ 190
Figure C-2: Sequence ID#18 .................................... 193

Figure C-3: BLASTX alignment between a putative acyltransferase from A.
calcoaceticus and Sn-l acyltransferase from E. coli ................ 193

xiv

LIST OF ABBREVIATIONS

ATP Adenosine triphosphate

cfu Colony forming unit

EDTA (Ethylenedinitilo)tetraacetic acid
IPCR Inverse polymerase chain reaction
NADH Nicotinamide adenine dinucleotide

(reduced form)

NADPH Nicotinamide adenine dinucleotide
phosphate (reduced form)

NTG Nitrosoguanidine
PCR Polymerase chain reaction
TLC Thin layer chromatography

XV

CHAPTER 1

INTRODUCTION

Wax esters are found in a number of very different organisms. Over a century
ago, Melville wrote about wax esters as a liquid gold hunted by Whalers in his
eighteenth century novel, Moby Dick (Melville, 1851). Referred to as spermaceti oil,
these wax esters are stored in the heads of the sperm whale and have the general
chemical structure that is shared by all wax esters (Figure l-l). The North American
shrub, jojoba, is a new source of wax esters that is cultivated in the southwestern
portion of the United States. In jojoba (Simmondria chinensis), wax esters are stored
in the seeds of the plant where they serve as a means of energy and carbon storage for
developing seedlings. Wax esters have also been found in a number of microbial
organisms such as Acinetobacter caIcoaceticus strain BD413, a gram negative aerobic
bacteria that was first isolated by Juni et. a1. (Juni & Janik, 1969) as an unencapsulated
form of A. calcoaceticus strain BD4. A. calcoaceticus strain BD413 accumulates wax
esters when grown under nitrogen limited conditions (F ixter et al., 1986). Although
these organisms are very diverse, examination of the chemical structures of the waxes
they produce reveals that these waxes are very similar to each other structurally, in
terms of their fatty acid and fatty alcohol components, carbon chain lengths and degree

of saturation (Table 1-1)(Ervin et al., 1984).

l—_ Fatty Acid Segment _.| |._. Fatty Alcohol Segment ——|

‘u’
e.

 

 

EstcrL'mkage

Figure 1-1: Structure of a generic wax ester. The fatty acid, or acyl segment
and the fatty alcohol, or alkoxy segment are shown.

3

Table 1-1: A comparison of some naturally occuring wax esters'

 

 

Sperm Microbial Jojoba
whale oil wax esters oil
Carbon number of intact 28-40 32-40” 36-44
wax esters
Fatty acid segments:
Carbon number 14-22 l6-20" 16-24
Number of unsaturations 0 or 1 0 or 1° 1
Fatty alcohol segments:
Carbon number 16-20 16-20" 18-24
Number of unsaturations 0 or 1 0 or 1‘ I
Predominant sites of 037, m9 and 011 1 m7 and (1)9 (09

unsaturation‘I
'Adopted from Ervin et al., 1984
"Carbon numbers dependent on n-alkane used as substrate for growth.
6Degree of unsaturation is dependent on growth temperature.
dPosition of double bond is indicated relative to the end carbon position.

 

The characterization of wax esters that occur in Acinetobacter species has
previously been described by several different researchers (Gallagher, 1971; Fixter,
1986). It was observed that when cells of different strains of A. calcoaceticus were
grown with carbon in abundance but some other nutrient limiting, wax esters
accumulated both intracellularly and extracellularly (Makula, 1975; Scott, 1976). Wax
esters were also observed when cells were grown under nutrient rich conditions, but to
a lesser extent. Cells can be starved for nitrogen, sulphur or phosphorous, and if
carbon is in abundance, wax esters will accumulate (F ixter, 1991). The amount of
wax ester accumulation was found to be highly variable depending on the strain

examined. Wax contents ranged from 0.6 mg to 141.1 mg per gram of bacteria dry

weight (Fixter, 1986).

Wax composition of A. calcoaceticus species has been extensively studied.
Thirty two carbon length waxes make up the majority (roughly 50%) of waxes
produced by A. calcoaceticus species (F ixter et al., 1986), while 34 carbon length
waxes make up about 40% of the total wax ester content. The remaining 10% can be
found in the form of 30 and 36 carbon waxes. These waxes are typically greater than
80% saturated. Examination of the isomeric composition of wax ester species by gas
liquid chromatography/mass spectrometry in terms of fatty acid-fatty alcohol
combinations was observed to be, what would be expected if one assumed random
combination of available fatty acids with fatty alcohols (F ixter & Sherwani, 1991).
The degree of desaturation of wax esters can be altered by changing growth -
temperature. Ervin et. al. report that when Acinetobacter species were grown at 30°C,
80% of the wax esters produced were saturated with 18% being monounsaturated and
2% being diunsaturated (Ervin et a1, 1984). When the growth temperature was
dropped to 24°C, 61% of the waxes were saturated, but the amount of monosaturated
waxes climbed to 31% and diunsaturated waxes composed 8% of the total. Finally,
when cultures were incubated at 17°C, the amount of saturated wax esters fell to 50%,
monounsaturated waxes hovered near 35% and diunsaturated wax esters jumped to
15%. In all of these measurements, diunsaturated compounds were the result of one
unsaturation in the acyl segment, and the other unsaturation in the alkoxy segment.
Monounsaturated wax esters were the result of an unsaturation in either segment.
Finally, unsaturations occurred at the (07 and (09 sites (when numbering from the

terminal methyl carbon of the carbon chain) on the acyl segments, and at the (07 site

only in the alkoxy segment.

Wax composition can also be altered by varying the composition of the
medium (Dewitt & Ervin, 1982). When cultures are grown in minimal mineral
medium with succinate or acetate as a carbon source, wax compositions are observed
to be what was previously described, i.e. 50% of the waxes are 32 carbons in length
and another 40% being 34 carbons in length. By growing the bacteria in minimal
mineral medium with hexadecane (a 16 carbon alkane) as a carbon source 100% of the
waxes were 32 carbons in length. Incubation of cultures in longer chain alkanes was
found to give rise to wax ester compositions of C2,, C2“,2 and C2!” (Dewitt & Ervin,
1982). The alcohol segment of the wax esters derived from alkanes was always the
same length as the substrate, indicating that the variation observed was due to the
length of the fatty acid.

Very little research has been aimed at understanding the biochemistry of wax
ester production. Most ideas are based on observed enzymatic activities involved in
hydrocarbon oxidation, with the leap in reasoning that what is observed for the
degradation of alkanes might be running backwards and be responsible for wax
biosynthesis (F ixter et al., 1991). This has given rise to the proposed pathway for wax
ester biosynthesis in A. calcoaceticus that is illustrated in Figure 1-2. In the
production of wax esters de novo, it is believed that the starting substrate is either
acyl-ACP or acyl-CoA. Acyl-ACP is generally regarded as being the substrate of
choice. This is because in many bacteria, particularly E. coli, fatty acid biosynthesis
typically involves acyl-ACPs rather than CoA derivatives (Rock & Cronan, 1985). In

the first committed step of wax ester biosynthesis, acyl-ACP is thought to be reduced

0
ll

VWW/c\s/COA

Acyl-CoA
Reductase

V\/\/\/\/\/\/\/\H

Fatty Aldehyde
Reductase

MCI-1201'!

Fatty Alcohol \
Fatty Acyl-CoA Transferase

 

 

Figure 1-2: Proposed pathway for wax ester biosynthesis in Acinetobacter
calcoaceticus strain BD413.

7

to the corresponding aldehyde. An aldehyde intermediate is proposed to occur based
of the observation of a constitutive NAD-dependent long chain alkanal dehydrogenase
and an inducible (induced in the presence of alkanes) NADP-dependent alkanal
dehydrogenase in the Acinetobacter strain HOl-N (Fox et al., 1992; Singer & F innerty,
19850). These activities were detected when strain HOl-N was investigated in relation
to its ability to oxidize hexadecane and hexadecanol as carbon sources. With the
observed activity of these two enzymes, the idea was put forth that one, or both, of the
enzymes might be catalyzing the reverse reaction, reducing acyl-ACP to the
corresponding aldehyde. There was no experimental evidence put forth to support this
idea. The second step in wax ester formation involves the reduction of the fatty
aldehyde to its corresponding fatty alcohol. Here again, the same'logic was proposed.
Two independent reports describe cofactor dependent and independent fatty alcohol
dehydrogenases, and again it is believed that these enzymes may play a role in wax
ester biosynthesis (Fox et al., 1992; Singer & F innerty, 1985b). In this instance also,
no experimental evidence was put forth to support the model. No experimental
evidence has been presented on the third and final step proposed in wax ester
biosynthesis, the formation of wax esters from either an acyl-ACP, or acyl-CoA, and
fatty alcohol by acyl-CoA (or ACP):alcohol transferase.

In contrast to A. calcoaceticus, the enzymes involved in wax ester biosynthesis
of jojoba have been extensively studied and characterized. In this organism it is
known that there are two enzymes that directly catalyze the formation of wax esters
(Pollard & Metz, 1995; Metz et al., 1995). The first enzyme that was characterized

was a fatty acyl-CoA reductase. This enzyme is known to be highly substrate specific

8

for tetracosenoyl-CoA (a 24 carbon acyl-CoA), and is known to catalyze the formation
of a long chain alcohol directly from this substrate via an aldehyde intermediate
(Pollard & Metz, 1995). The second enzyme, an acyl-CoA-fatty alcohol transferase
catalyzes the formation of an ester linkage between acyl-CoA and a fatty alcohol to
yield a wax ester. Assays on this enzyme, found it to be acyl-CoA specific, preferring
Czo-monounsaturated acyl-CoA’s and Cl, and C13 mono- and di- unsaturated fatty
alcohols (Metz et al., 1995).

It has been observed for A. calcoaceticus that when cellular concentrations of
wax esters reach concentrations of 20 mg/g dry weight, inclusion bodies are formed
(Fixter et al., 1991). Characterization of these inclusion bodies has never been
performed on cells grown in plain, unsupplemented minimal mineral medium. Rather,
our knowledge of the composition of inclusion bodies comes fi'om examination of cells
grown in minimal mineral medium, supplemented with hexadecane. Table 1-2
recapitulates the neutral lipid profile of inclusion bodies isolated from cells grown
under such conditions (Scott & Finnerty, 1976). In a different report put out by the
same group 9 years later, the lipid composition of inclusion bodies grown on
hexadecane and hexadecanol had the compositions shown in Table 1-3 (Singer &

F innerty, 1985a). These data appear to be contradictory, but represent the only data
available. It is also important to note that these studies, along with many others, use
A. calcoaceticus strain HOl-N. This strain may be very different from other A.
calcoaceticus species. It was first isolated as Micrococcus cerificans by enrichment on
hexadecane (F innerty et al., 1962). It was later reclassified as an A. calcoaceticus

species. During growth of this culture on alkanes, it was observed that it could

Table 1-2: Relative concentrations of neutral lipids in hexadecane inclusions.‘

 

Neutral Lipid Relative Percentage (%)

 

Wax ester 50.5

Free fatty acid 5.8
Free fatty alcohol 17.6

Triglyceride 5.7

Diglyceride 6.9

Monoglyceride 1 .3

Hexadecane 12

'Adapted from Scott & Finnerty, 1976

 

Table 1-3: Lipid composition of intracellular inclusions isolated from hexadecanol and
hexadecane-grown Acinetobacter sp. strain HOl-N.‘

 

 

Lipid component Wax ester Hydrocarbon
inclusions (%) inclusions ("/o)b
Phospholipid 9.6 4.7
Wax ester 85.6 1.3
Fatty alcohol 4.8 0.4
Hexadecane 0.0 93.3
Other lipidsc 0.0 0.3

 

1'Taken from Singer et al., 1985a.

l’This data is reported to be derived from the same data presented in Table 1-2.
°Fatty acid and mono-, di- and triglycerides.

10

produce up to 4.3 g of wax esters per liter of culture, and most of these waxes were
extracellular (Fixter et al., 1991). This is different in respect to other A. calcoaceticus
species where wax ester accumulation is observed to be intracellular. Mutants of A.
calcoaceticus strain HOl-N were isolated that were unable to grow on cetyl palmitate
(a 32 carbon wax ester) as a sole carbon source. Examination of the amount of wax
esters produced by these mutants showed an increase in extracellular wax
accumulation. This has led to the speculation that wax ester content in Acinetobacter
is dependant on the rate of wax ester degradation, rather than the rate of wax ester
synthesis (Geigert et al., 1984; Fixter et al., 1991).

Wax ester inclusions of A. calcoaceticus strain HOl-N have been examined by
electron microscopy where it was observed that they appeared to be surrounded by a
monolayer phospholipid membrane. Analysis of this monolayer found it to be
composed of phospholipid and proteins. At least five major polypeptide bands were
observed when proteins from the membranes were separated on an SDS-PAGE gel
(Scott & Finnerty, 1976). Unfortunately, size standards were not included with this
data so it is not possible to tell what sizes the proteins are. The lipid and protein
composition of the membranes surrounding the vesicles was compared with other
membrane fractions of the cell. Their different phospholipid composition, and the
banding pattern of proteins isolated from the inclusion membranes led the author to
suggest that the membranes arise de novo, rather than being derivatives of the
cytoplasmic or outer membrane (F ixter et al., 1991). When cells were grown in the
presence of hexadecane and hexadecanol, some cells exhibited intracytoplasmic

membranes which the author states appeared as lamellar sheaths extending through the

ll

cytoplasm (Scott & Finnerty, 1976; Singer et al., 1985a). Intracellular membranes
were also observed in cells grown in the presence of hexadecanol. Examination of
succinate grown cultures by electron microscopy never showed the presence of such
structures, suggesting that their formation may have something to do with growing the
cultures in hydrocarbons.

As indicated in Tables 1-2 and 1-3, in addition to the formation of wax esters,
A. calcoaceticus also accumulates triacylglycerol (TAG) and other neutral lipids when
grown under nitrogen deficient conditions. There has been no investigation of TAG
biosynthesis in Acinetobacter species other than to note its presence when measuring
neutral lipid profiles of cultures grown to induce wax ester biosynthesis. Intracellular
and extracellular lipid composition have been measured for A. calcoaceticus strain
HOl-N grown in nutrient rich NBYE medium versus minimal mineral medium
containing hexadecane as a carbon source. The results of these experiments by
Makula, et. al. (Makula et al., 1975), are shown in Tables 1-4 and 1-5. Based on these
observations it might be assumed that other Acinetobacter strains also accumulate
triacylglycerol. TAG may be serving as another form of energy storage, in addition to
the accumulation of wax esters. Although no biochemical characterization of
triacylglycerol biosynthesis has been presented, it could be proposed that, under
nitrogen limited conditions, a diacylglycerol-acyltransferase is expressed, leading to an
increase in the conversion of diacylglycerol to triacylglycerol. Alternatively, the
accumulation of TAG may be the result of a reduction in the conversion of TAG to
free fatty acids and glycerol. Based on the data presented in Table 1-2 from Scott et.

a1. (Scott & F innerty, 1976) which describes the neutral lipid content of inclusion

12

Table 1-4: Quantitation of Acinetobacter sp. HOl-N cellular lipids.‘

 

Medium

Lipid class NBYE" Minimal plus
Hexadecane

 

nmol/g of cell dry weight

Phospholipids 46.0 129.0
Triglyceride 1 .8 2.5
Mono- and diglyceride 0.4 6.8
Free fatty acid 7.5 8.2
Free fatty alcohol Trace 2.6
Wax ester 11.5 18.0

Hexadecane Not detected 360.0

 

‘Adapted from Makula et al., 1975
t’Nutrient broth-yeast extract medium. Composition of medium not stated by author.

Table 1-5: Quantitation of Acinetobacter sp. HOl-N extracellular lipids.‘

 

 

 

Medium
Lipid class NBYE" Minimal plus
Hexadecane
pmol/L
Phospholipids Not detected Not detected
Triglyceride 2.4 25.6
Mono- and diglyceride 0 410.0
Free fatty acid 4.0 60.0
Free fatty alcohol 0 0.5
Wax ester 0 280.0
Hexadecane 0 Not determined

 

‘Adapted fiom Makula et al., 1975
”Nutrient broth-yeast extract medium. Composition of medium not stated by author.

13

bodies formed in A. calcoaceticus grown under low nitrogen conditions in the presence
of hexadecane, it can be assumed that triacylglycerol is being accumulated together
with wax esters, and other neutral lipids in the form of inclusion bodies.

Why do organisms synthesize wax esters? Wax esters are composed of carbon
in its most reduced form, which yields the maximum amount of energy when oxidized.
Additionally, wax esters are extremely hydrophobic and are completely insoluble in
water. Therefore, when they are stored intracellularly as inclusion bodies, there is no
effect on the osmotic balance of the cell. Together, these two properties make wax
esters an excellent means of energy storage. A comparison can be made between the
energy yields of glucose, triacylglycerol and wax esters based on the information
presented by Mathews and van Holde (Mathews & van Holde, 1990). They calculated
the energy yields for glucose and a fatty acyl-CoA (palmitoyl-CoA) in moles of ATP
synthesized from ADP. Assuming that the energy yield from one mole of
triacylglycerol is equal to the oxidation of three fatty acyl-CoA chains (palmitoyl-
CoA), and the oxidation of a wax ester is essentially equivalent to the oxidation of two
fatty acyl-CoA chains (i.e. palmitoyl-CoA), Table 1-6 can be derived. It is clear that
triacylglycerol and wax esters yield more energy than the oxidation of glucose. This
clearly illustrates that both of these compounds can serve as excellent means of energy
storage.

Wax esters play many important biological roles other than providing a means
of carbon storage. Waxes are critical in the plant kingdom where they provide
protection from DNA damaging ultra-violet light, a line of defense against insects,

fungus and bacterial pathogens, and probably most importantly, prevention of

14

Table 1-6: Energy yields in moles of ATP per mole of energy source.

 

ATP Yield per Carbon
Oxidized to CO,
(moles of ATP)

Oxidative Yield

Energy Source (moles of ATP)

 

Glucose 38 6.3
Wax Ester (32 carbons) 262 8.2
Triglyceride (1e. 3, 16 393 8.2

carbon fatty acids)

 

desiccation. The steps believed to take place in the biosynthesis of wax esters in
plants is similar to the pathway illustrated in Figure 1-2 for A. calcoaceticus, with the
exception that the starting substrate is a very long chain fatty acid that is derived by
elongating stearoyl-CoA (18C) to eicosanoyl-CoA (20C). These long chain fatty acids
are believed to be reduced to their corresponding fatty aldehydes, which in turn are
further reduced to fatty alcohols. There is biochemical evidence that each of these
steps is carried out separately by a fatty acid reductase and a fatty aldehyde reductase,
respectively. Demonstration of such enzymatic activities in crude extracts has been
shown by Kolattukudy et. al. in Brassica oleracea (Kolattukudy, 1971). Wax esters
are thought to be formed by linking the long chain fatty alcohols to the long chain
fatty aldehydes by an acyl-CoAzfatty alcohol transferase, as has been observed in
jojoba. Genetic evidence for the complexity of this pathway in plants can be found in
Arabidopsis thaliana where at least 21 different loci are thought to be involved in wax
biosynthesis based upon the identification of mutants (cer for ecerzferum) (Lemieux et
al., 1994). The cer mutants of Arabidopsis are easily identifiable based on their glossy

phenotype. Typically, the stems of wild type plants are coated with waxes giving

15

them a muted green color. Cer mutants, on the other hand, are glossy green in color,
presumably because the reduction in the amount of waxes covering the stem leads to a
decrease in the amount of difiacted light that is reflected.

Plant waxes are actually composed of many different materials; wax esters are
only one component. Besides wax esters, plants accumulate long chain aldehydes,
fatty acids, primary alcohols, hydrocarbons, secondary aldehydes and ketones
(Lemieux et al., 1994). Thus, with all of these components being synthesized and
secreted, it is not surprising to find so many different mutants. The problem then, in
dealing solely with wax ester biosynthesis, is to identify mutants affected in only wax
esters and their intermediates. To characterize the cer mutants of Arabidopsis,
Lemieux et. a1. (Lemieux et al., 1994) measured relative changes in neutral lipid
compounds that make up the epicuticular waxes. This allowed them to assign possible
defects to 5 of the 21 mutants. Two of these mutants, cer4 and cer8 appear to be
blocked in steps directly related to the synthesis of wax esters. The data for these 2
mutants is summarized in Table 1-7 (Lemieux et al., 1994).

Table 1-7: Relative changes in the amounts of aliphatic compounds comprising the
epicuticular waxes of 20 eceriferum (cer) mutants compared to wild type levels.”

 

Line Fatty Primary Alde- Alkanes Second- Ketones

 

acids alcohols hydes ary
alcohols
cer4 70.4 4.5 127 1 12 177 107
cer8 434 63.5 4.3 31.0 8.1 8.2

 

‘Amounts expressed as percent of wild type levels.
hFrom Lemieux et al., 1994.

16

Because of the reduction of primary alcohols observed in cer4, this mutant is
thought to have a lesion in its ability to convert fatty aldehydes to fatty alcohols. The
other locus, cer8, is thought to regulate the reduction of fatty acids to fatty aldehydes,
because of the dramatic accumulation of fatty acids and reduction in aldehyde content
(Lemieux et al., 1994). Thus, based on genetic and biochemical evidence it would
appear that wax ester biosynthesis in Arabidopsis is carried out in a similar manner to
what is hypothesized to occur in Acinetobacter.

As stated previously, very little is known about the enzymes involved in wax
ester biosynthesis in A. calcoaceticus species. By understanding the biosynthesis of
wax esters, we can gain an insight into an alternative means of energy storage, which,
on the surface, seems as energy rich as triacylglycerol (in terms of ATP yield per
carbon atom). Additionally, an understanding of wax ester biosynthesis in
Acinetobacter could lead to insights about similar pathways in other organisms, such
as Arabidopsis.

Because of its ability to synthesize wax esters and triacylglycerol, A.
calcoaceticus seemed like an excellent organism to use for the study of wax ester
biosynthesis. Although there is direct biochemical evidence of enzymes involved in
wax biosynthesis in plants, and genetic support of the proposed pathway in the form of
the cer mutants of Arabidopsis, there were many advantages to using A. calcoaceticus
instead of Arabidopsis for such a study. Both facile genetics and fast growth rate
make A. calcoaceticus a more attractive model organism for wax ester biosynthesis
than Arabidopsis. As previously mentioned wax ester and triacylglycerol accumulation

are both induced under nitrogen limited conditions in A. calcoaceticus. This makes an

17

excellent premise for a genetic screen. Both of these compounds are presumably not
necessary for the vitality of the organism under laboratory conditions. Thus, mutations
in either pathway could be suffered by the organism with very little or no adverse
affects. This is in contrast to some of the cer mutants of Arabidopsis which are male-
sterile, or must be grown under very humid conditions to avoid death by desiccation
(Lemieux et al., 1994). Waxes and triacylglycerol are both neutral lipids and can be
readily separated and observed by thin layer chromatography (Kates, 1986). This
presents an easy, although somewhat tedious, way of screening for mutants in either
one of these pathways at the same time. The ability of A. calcoaceticus to be grown
on hydrocarbons, fatty aldehydes (this work) and fatty alcohols as sole carbon sources
presents a unique and simple means of characterizing mutants involved in wax ester
biosynthesis. Presumably mutations in the wax ester pathway can be bypassed simply
by adding a precursor of wax ester metabolism such as hexadecane, or intermediates
such as hexadecanal or hexadecanol. The ability of these compounds to complement
the mutants might signal where in the pathway a mutant has been effected.

Finally, an important characteristic of certain A. calcoaceticus species is that
certain strains of A. calcoaceticus are naturally competent for DNA transformation
(Juni & Janik, 1969; Palmen et al., 1993). It has been observed as early as 1969, that
when actively growing cultures of A. calcoaceticus were presented with DNA they are
able to up take the molecule and alter their genotype by utilizing genes from the
exogenously acquired DNA. This property could be used to facilitate genetic
complementation of mutants.

In this work I present experiments in which I have taken a genetic approach to

18

understanding wax ester biosynthesis in A. calcoaceticus. Chapter 2 details the
conditions used to mutagenize and screen for mutants affected in wax ester and
triacylglycerol accumulation. Also, the isolated mutants were characterized by
nutritional supplementation experiments involving wax ester intermediates. Chapter 3
describes the genetic complementation of a mutant affected in wax ester biosynthesis.
The cloning of the gene acrl is described along with in vitro enzyme assays
demonstrating the function of the cloned gene. Chapter 4 deals with new mutants that
were isolated by transposon mutagenesis using mini-Tn10Ptth. Flanking sequence
surrounding the transposon in one of the triacylglycerol mutants is subcloned,
sequenced and described. Finally, Chapter 5 summarizes the knowledge gained from
the previous chapters and the three appendixes at the end of this text. Recommended
directions of research are presented in context to the tools and discoveries highlighted

in the previous chapters.

REFERENCES

Dewitt, S., Ervin, J. L., Howes-Orchison, D., Dalietos, D., and Neidleman, S. L..
1982. Saturated and Unsaturated Wax Esters Produced by Acinetobacter sp. H01-N
Grown on C.,-C20 n-Alkanes. JAOCS 59(2): 69-74.

Ervin, J. L., Geigert, J ., Neidleman, S. L., and Wadsworth, J .. Substrate-Dependent
and Growth Temperature-Dependent Changes in the Wax Ester Compositions Produced

by Acinetobacter sp. HOl-N. In Biotechnology for the Oils and Fats Indusg,
American Oil Chemists Society, 1984. Eds. Ratledge, C., Dawson, P. and Rattray, L..

F innerty, W. R., Hawtrey, E. and Kallio, R. E.. 1962. Alkane-Oxidizing Micrococci.
Zeeitschrift fiir Allg. Mikrobiologie 2(3): 169-177.

Fixter, L. M., Nagi, M. N., McCormack, J. G. and Fewson, C. A.. 1986. Structure,
Distribution and Function of Wax Esters in Acinetobacter calcoaceticus. J. Gen.
Micro. 132: 3147-3157.

l9

Fixter, L. M. and Sherwani, M. K.. Energy Reserves in Acinetobacter. In T_h§
Biology of Acinetobacter: Taxonomy, Clinical Importance. Molecular Biology,
Physiology, Industrial Relevance. Plenum Press, New York, New York. 1991. Eds.
Towner, K. J., Bergogne-Bérézin, E., and C. A. Fewson.

Fox, M. G. A., Dickinson, M. and Ratledge, C.. 1992. Long-Chain Alcohol and
Aldehyde Dehydrogenase Activities in Acinetobacter calcoaceticus strain HOl-N. J.
Gen. Micro. 138:1963-1972.

Gallagher, 1. H. C.. 1971. Occurrence of Waxes in Acinetobacter. J. Gen. Micro.
68:245-247.

Geigert, J, Neidleman, S. L. and DeWitt, S. K.. 1984. Further Aspects of Wax Ester
Biosynthesis by Acinetobacter sp. HOl-N. JAOCS, 61(11): 1747-1750.

Juni, E. and Janik, A.. 1969. Transformation of Acinetobacter calco-aceticus
(Bacterium anitratum). J. Bact. 98(1): 281-288.

Kates, M.. 1986. Techniques of Lipidology: Isolation, Analysis and Identification of
Lipids. Elsevier Science Pub. Co., New York, New York.

Kolattukudy, P. E.. 1971. Enzymatic Synthesis of Fatty Alcohols in Brassica
oleracea. Arch. Biochem. Biophys. 142: 701-709.

Lemieux, B., Koomneef, M. and Feldmann, K. A.. Epicuticular Wax and eceriferurn
Mutants. In Arabidopsis. Cold Spring Harbor Laboratory Press, New York, New
York. 1994. Eds. Meyerowitz, E. M. and Somerville, C. R..

Makula, R. A., Lockwood, P. J. and Finnerty, W. R.. 1975. Comparative Analysis of
Lipids of Acinetobacter Species Grown on Hexadecane. J. Bact. 121(1): 250-258.

Mathews, C. K. and van Holde, K. E.. 1990. Biochemisgy. Benjamin/Cummings
Pub. Co., Redwood City, California.

Melville, H. 1851. Moby Dick.

Metz, J. G., Lardizabal, K. D., and Lassner, M. W.. 1995. Calgene Inc.. United
States Patent #5,445,947. Jojoba Wax Biosynthesis Gene. Filed May 20, 1993,
awarded Aug. 29, 1995.

Ohlrogge, J. B., Pollard, M. R. and Stumpf, P. K.. 1978. Studies on Biosynthesis of
Waxes by Developing Jojoba Seed Tissue. Lipids 13(3): 203-210.

Palmen, R., Vosman, B., Buijsman, P., Breek, C. K. D., and K. J. Hellingwerf. 1993.
Physiological Characterization of Natural Transformation in Acinetobacter

20
calcoaceticus. J. Gen. Micro. 139:295-305.

Pollard, M. R. and Metz, J. G.. 1995. Calgene, Inc.. United States Patent
#5,411,879. Fatty Acyl Reductases. Filed Nov. 8, 1993, awarded May 2, 1995.

Rock, C. O. and Cronan, J. E.. 1985. Lipid Metabolism in Prokaryotes. In
Biochemistgy of Lipids and Membranes. Eds. Vance, D. E. and Vance, J. E..
Benjamin/Cummings, Menlo Park, California.

Scott, C. C. L. and Finnerty, W. R.. 1976. Characterization of Intracytoplasmic
Hydrocarbon Inclusions from the Hydrocarbon-Oxidizing Acinetobacter Species H01-
N. J. Bact. 127(1): 481-489.

Singer, M. E., Tyler, S. M. and F innerty, W. R.. 1985a. Growth of Acinetobacter sp.
Strain HOl-N on n-Hexadecanol: Physiological and Ultrastructural Characteristics.
Mol. Gen. Genetics 162: 162-169.

Singer, M. E. and F innerty, W. R.. 1985b. Alcohol Dehydrogenases in Acinetobacter
sp. Strain HOl-N: Role in Hexadecane and Hexadecanol Metabolism. J. Bact.
164(3): 1017-1024.

Singer, M. E. and Finnerty, W. R.. 1985c. Fatty Aldehyde Dehydrogenases in
Acinetobacter sp. Strain HOl-N: Role in Hexadecane and Hexadecanol Metabolism.
J. Bact. 164(3): 1011-1016.

CHAPTER 2

A. CALCOACETICUS STRAIN BD413 AS A MODEL ORGANISM FOR THE
BIOSYNTHESIS OF WAX ESTERS AND TRIACYLGLYCEROL

ABSTRACT

Mutants of A. calcoaceticus that are deficient in wax ester and triacylglycerol
accumulation were induced using nitrosoguanidine. Colonies with reduced wax or
triacylglycerol content were identified using the lipophilic dye Sudan Black B to stain
replicas of colonies. Lighter stained colonies were selected and subsequently analyzed
by TLC for wax and triacylglycerol content. In this manner over 6400 colonies were
examined. A total of 25 mutants of three different phenotypic classes were isolated.
Class I mutants are affected in wax ester accumulation, but have normal levels of
triacylglycerol. Class II mutants have decreased levels of triacylglycerol, but normal
amounts of wax esters, and class III mutants are affected in both wax ester and
triacylglycerol accumulation.

Characterization of the mutants was carried out using chemical
complementation experiments involving growing the cultures in the presence of 0.3%
hexadecane or 0.3% hexadecanol under low nitrogen conditions. None of the mutants
had any phenotypic change when cultured in the presence of hexadecane. However,
all of the class I mutants were able to synthesize wax esters when incubated in the

presence of hexadecanol. Hexadecanol had no affect on any of the class II or class III

21

22

mutants.

Characterization of the Wow15 mutant (wax') was further conducted with
additional nutritional supplementation experiments using 0.3% Tween-40 or 0.3% cis-
ll-hexadecenal. Incubation of this mutant in the presence of Tween-40 had no affect
on the mutant phenotype, while grth on cis-l l-hexadecenal as a carbon source
restored the ability of the mutant to produce wax esters. This result indicates that the
Wow15 mutant has been affected in its ability to convert acyl-CoA (or acyl-ACP) to

the corresponding aldehyde.
INTRODUCTION

Acinetobacter are known to accumulate both wax esters and triacylglycerol as
intracellular inclusions under nutrient starvation conditions (Makula et al., 1975; Scott
& Finnerty, 1976). It has been reported that the inclusions are synthesized under low
nitrogen or low phosphorous conditions, presumably as a means of carbon storage
(Fixter et al., 1991). The inclusions are primarily composed of wax esters which are
predominantly 32 carbons in length (Fixter et al., 1986). Waxes are composed of an
alkoxy segment donated by hexadecanol, a 16 carbon fatty alcohol, and a 16 carbon
fatty acid segment presumably originating from either palmitoyl-CoA or palmitoyl-
ACP. Wax esters composed of 34 carbon atoms are also common and are derived
from 16 carbon alcohols and 18 carbon fatty acids (Fixter et al., 1986). Smaller
amounts of different chain length wax esters are also observed, being derived from
different chain length substrates (ie. 30, 31, 33, 35 and 36 carbon wax esters). It has

been reported that temperature plays a key role in the saturation of the wax ester

23

compounds (Ervin et al., 1984). Mono and disaturated compounds become more
prevalent as incubation temperatures decrease. No characterization of the
triacylglycerol found in these inclusions has taken place, other than to note its presence
(Makula et al., 1975; Scott & Finnerty, 1976; Singer et al., 1985).

Wax composition can also be altered by varying the carbon source in the
medium (Dewitt & Ervin, 1982). A. calcoaceticus is able metabolize fatty alcohols
and alkanes. Different wax ester products can be predicted when the cells are cultured
in their presence. For example, addition of an 18 carbon alcohol to the medium results
in wax ester inclusions that are predominantly 34 carbons in length. This is the result
of the 18 carbon alcohol being linked to palmitoyl-CoA. Thus, by growing A.
calcoaceticus in the presence of fatty alcohols or alkanes, it is possible to directly
provide both an intermediate or a precursor, respectively, to wax ester biosynthesis.
This allows the formulation of experiments that would allow the biochemical analysis
of wax ester production through the addition of these intermediates to the medium.

A. calcoaceticus can be transformed by incubating the bacteria in the presence
of DNA (Juni & Janik, 1969; Palmen et al., 1993). This facilitates the application of
recombinant DNA techniques where such a property makes working with A.
calcoaceticus amenable to such methods as targeted gene replacement and
complementation by introduction of exogenous DNA. As an interesting side note, A.
calcoaceticus is so competent that it has become an increasing problem in hospitals
where immune compromised patients fall prey to nosocomial (infections obtained when
visiting a hospital, usually in an immune compromised state) infections from A.

calcoaceticus. Although normally not a problem, recently A. calcoaceticus strains

24

have been identified as harboring antibiotic resistance genes which they have acquired
from other organisms. It is thought that microbes like A. calcoaceticus may be sharing
their genetic information with other more lethal pathogens, thus attributing to the
increase in deaths in this country by contagious disease (Bergogne-Bérézin & Joly-
Guillou, 1991).

In this chapter several different strains of A. calcoaceticus are compared in
terms of wax ester and triacylglycerol accumulation. A particular strain, BD413, was
selected for further study. Mutagenesis of this strain was carried out using
nitrosoguanidine to generate mutants affected in wax ester and triacylglycerol
biosynthesis. In order to characterize the mutants, I took advantage of the fact that A.
caIcoaceticus is able to utilize exogenous fatty alcohols and alkanes. By growing the
mutants in the presence of these substrates, it was possible to get some understanding

of the biochemical nature of the lesions in some of the mutants.

MATERIALS AND METHODS

Bacterial stains and plasmids. The bacterial strains used in the experiments
described in this chapter are shown in Table 2-1.

Growth and culture conditions. Low nitrogen minimal medium (per liter, 2.0
g KHZPO” 1.18 g succinic acid, 0.1 g NH,SO,, pH adjusted to 7.0 with solid KOH;
afier autoclaving add 20 ml of sterilized 2% MgSO,) was used in experiments with
Acinetobacter calcoaceticus strain BD413 to induce wax ester formation. High

nitrogen minimal medium, which was the same as the above except for the addition of

25

Table 2-1: Bacterial strains from Chapter 2.

 

Bacterial Strains

Relevant
Characteristics

Source or Reference

 

A. calcoaceticus
ACB #14987
ATCC #23055
ATCC #19606

ATCC #33305, strain
BD413

Wowl

Wow2

Wow3

Wow4

Wow5

Wow6

Wow7

Wow9

Wow10

Wowll

Wow12

wild type
wild type
wild type

wild type used during
these studies,
unencapsulated mutant
of A. calcoaceticus strain
BD4

wax' null mutant of
BD413

wax‘ null mutant of
BD413

wax‘tag’ null mutant of
BD413

wax'tag‘ leaky mutant of
BD413

wax'tag' leaky mutant of
BD413

tag' leaky mutant of
BD413

tag' null mutant of
BD413

wax’tag' leaky mutant of
BD413

wax‘tag' leaky mutant of
BD413

tag' leaky mutant of
BD413

wax'tag' leaky mutant of
BD413

ATCC stock center
ATCC stock center
ATCC stock center
ATCC stock center

This study, Chapter 2

This study, Chapter 2

This study, Chapter 2

This study, Chapter 2

This study, Chapter 2

This study, Chapter 2

This study, Chapter 2

This study, Chapter 2

This study, Chapter 2

This study, Chapter 2

This study, Chapter 2

26

Table 2-1 (cont'd)

Wowl3 wax' null mutant of This study, Chapter 2
BD413

Wow14 wax' null mutant of This study, Chapter 2
BD413

Wow15 wax“ null mutant of This study, Chapter 2
BD413

Wow16 wax'tag' leaky mutant of This study, Chapter 2
BD413

Wow17 tag' leaky mutant of This study, Chapter 2
BD413

Wow18 tag' null mutant of This study, Chapter 2
BD413

Wow20 tag‘ leaky mutant of This study, Chapter 2
BD413

Wow22 tag' leaky mutant of This study, Chapter 2
BD413

Wow23 wax’tag' leaky mutant of This study, Chapter 2
BD413

Wow24 tag‘ leaky mutant of This study, Chapter 2
BD413

Wow25 wax'tag' leaky mutant of This study, Chapter 2
BD413

Wow26 wax'tag’ leaky mutant of This study, Chapter 2
BD413

Wow27 wax’tag' leaky mutant of This study, Chapter 2
BD413

Wow28 wax“ null mutant of This study, Chapter 2

BD413

 

27

1.0 g NH,SO4 per liter. For other purposes such as DNA isolation, A. calcoaceticus
was grown and maintained on LB medium (10 g Bacto tryptone, 5 g Bacto yeast
extract and 10 g NaCl per liter, pH 7.0). For chemical complementation experiments,
the above minimal medium were used with the addition of 0.3% of the substrate
(hexadecane, hexadecanol, Tween-40 (polyoxyethylenesorbitan monopalmitate) or cis-
ll-hexadecenal). Hexadecane and hexadecanol were sonicated for approximately 2
minutes in the medium to generate a suspension. A. calcoaceticus cultures were
typically grown overnight with shaking at 30°C. For the grth of larger volume
cultures (i.e. 50 mls and larger), 3 m1 overnight cultures were collected by
centrifugation, the cell pellet was washed in the new medium, recentrifuged and then
added to the fresh medium, before being incubated at 30°C overnight. Maintenance
and growth of Escherichia coli strains was on LB with appropriate antibiotics.
Antibiotics were used in the following concentrations: ampicillin 100 ug/ml,
chloramphenicol 50 ug/ml, kanamycin 25 ug/ml, rifampicin 50 ug/ml and tetracycline
15 ug/ml.

Chemical mutagenesis. A 25 ml culture of Acinetobacter calcoaceticus strain
BD413 was inoculated with a 0.5 ml sample of overnight culture. Incubation
continued at 30°C for 45 minutes until the observed optical density at 600 nm was
approximately 0.6. Three milliliters was removed for determination of the number of
viable cells at the beginning of the experiment. At this point, 2.3 mg of
nitrosoguanidine (NTG) (0.1 mg/ml final concentration) was added and the culture was
allowed to incubate for an additional 50 minutes at room temperature (in a chemical

safety hood) with shaking. Every 10 minutes (time=0, 10, 20, 30, 40 and 50 minutes)

28

a 3 ml sample was removed and placed on ice. In between the collection of every
sample, the previous sample was centrifuged to pellet the cells and the supernatant was
removed and replaced with fresh LB broth without NTG. Each sample was diluted (a
thousand fold, one hundred thousand fold and ten million fold) and plated onto LB
plates to determine the number of viable colonies. The remainder of the samples was
spun again to collect the cells, the medium containing NTG was removed, and another
3 mls of fresh LB medium added. These 3 ml cultures were incubated at 30°C
overnight. The next day, 1 ml of each culture was frozen in 7% DMSO and stored at
-80°C as a stock. Another 1 ml from each time point was collected and plated onto
LB plates containing 100 ug/ml rifampicin to determine the spontaneous, and induced
rates of resistance, thus giving some indication of the mutagenic effect of NTG on A.
calcoaceticus. Samples from 30 and 50 minutes were then spread onto LB plates and
the resulting colonies were transferred onto LB master plates in arrays of 100 colonies
per 100 mm petri plate.

Mutant screening using the lipophilic dye Sudan black B. The master plates
were replica plated onto low nitrogen minimal medium and the replicas were incubated
overnight at 30°C to induce wax formation. The induced colonies were then stained
by irrigating the plates with sudan black B (0.02% in 50:45:5 DMSO:ethanol:water)
and gently shaking for approximately 20 minutes. The stain was aspirated away and
the plates were carefirlly washed with 70% ethanol and gently shaking them for
approximately 2 minutes. Lighter staining colonies were identified from the stained
plates and the corresponding colony from the master plate was subsequently analyzed

by thin layer chromatography.

29

Thin layer chromatography (TLC). For mutant screening, A. caIcoaceticus
samples to be analyzed by TLC were typically grown in 3 ml cultures in low nitrogen
minimal medium. Fifty milliliter cultures were used at other times. Samples were
collected by centrifugation (5 minutes at 3000 x g) and the medium removed. Pellets
were washed in additional medium, centrifiiged, and the supernatant removed. Neutral
lipids were isolated from 50 ml cultures by extracting the cells with 3 ml (75 [.11 for 3
ml cultures) chloroformzmethanol (50:50) followed by phase separation using 1 ml (25
pl for 3 ml cultures) 1 M potassium chloride in 0.2 M phosphoric acid. Samples were
centrifuged at 2000 x g in a clinical centrifuge at room temperature for 2 minutes.

The chloroform phase was then transferred to a new glass tube and dried under
nitrogen, or in the case of the 3 ml cultures used during screening, the chloroform
phase was directly spotted onto a TLC plate. Samples were subsequently resuspended
in amounts of chloroform proportional to their cellular wet weights (typically 0.5 ml
for a 50 ml culture). Twenty microliters of these samples were spotted onto 19-
channel Si-250 TLC plates containing preadsorbant layers (Baker) that had been
charged by incubating at 120°C for 10 minutes. Ten microliters of 2 mg/ml standards
of known lipid species were also loaded for comparison during subsequent analysis.
Lipids were resolved by running the plates in hexane:ethyl ether:acetic acid (90:10zl).
Samples were visualized by spraying the plates with 50% sulfuric acid and charring at
160° for approximately 5 minutes, or by iodine staining by immersing the plate in

iodine vapor until visualization of the lipids was possible.

30
RESULTS AND DISCUSSION

Comparison of neutral lipid accumulation in various strains of A.
calcoaceticus. To determine which strain of A. calcoaceticus to work with, four
different strains (ACB #14987, ATCC #23055, ATCC #19606 and ATCC #33305)
were examined to determine their degree of wax ester accumulation under low nitrogen
conditions. The four strains were selected based on their use by other researchers
working with Acinetobacter. Neutral lipids from 50 ml cultures, grown under
noninducing (high nitrogen) and inducing (low nitrogen) conditions, were isolated by
chloroformzmethanol extraction. The lipids were spotted proportionately based on
cellular wet weight onto a TLC plate, separated, and visualized by ang (Figure 2-
1). Two of the four strains exhibited a strong, inducible accumulation of wax esters
under low nitrogen conditions. Additionally, one of these strains, ATCC #33305
commonly referred to as A. calcoaceticus strain BD413, also showed the accumulation
of another neutral lipid under low nitrogen conditions. This lipid species comigrates
with a known triacylglycerol standard. The identity of this lipid as triacylglycerol is
consistent with several previous reports of the kinds of lipids that accumulate in A.
calcoaceticus under low nitrogen conditions (Makula et al., 1975; Scott & Finnerty,
1976; Singer et al., 1985). These findings together with a report that strain BD413 is
naturally competent for genetic transformation (Juni & Janik, 1969; Juni, 1972; Palmen
et al., 1993), suggested to me that A. calcoaceticus strain BD413 was the best

available strain for the genetic analysis of wax ester biosynthesis in A. calcoaceticus.

31

Am ATCXI ATCC Aim
14987 23055 33305

wF—jflF—jflw

'PG 3 , - __ 9"
L‘ H h— .L—L 5..
e I- . r e s"
0

Figure 2-1: A comparison of wax ester accumulation from four strains
of A. calcoaceticus. High nitrogen, non-inducing (+) and low
nitrogen, inducing (-) conditions. Std, 0.2 mg of oleic acid stearyl
ester; SF, solvent front; W, wax esters; TG, triacylglycerol; O, origin.

32

Conditions for mutagenesis. In order to determine the optimum level of
mutagenesis, the effect of the duration of NTG treatment for a given amount of NTG
was measured in a manner similar to Benning (Benning, 1991). Mutagenesis was
evaluated by plotting the number of viable colonies versus time on a graph. This
graph was then overlaid with a plot showing the number of colonies that were
observed to be resistant to rifampicin, following the mutagenic treatment for each time
point (Figure 2-2). Based on these data we selected the samples that were removed at
30 and 50 minutes time points for the mutant screening. The 30 minute time point
was selected because it represented a time point in which an acceptable proportion of
the culture was surviving the chemical treatment, yet at the same time had a high
degree of mutation, as witnessed by the amount of Rif colonies. The 50 minute time
point represented the maximum amount of mutagenesis. This is important because of
the method with which I have chosen to screen for mutants. TLC analysis of
individual colonies is a tedious process and therefore it was important to maximize the
number of mutations per colony. Therefore, the 50 minute time point which represents
a high kill rate, and is presumably equivalent to a high mutation rate, was also used.

Mutant screening. 1 developed a quick method for screening for mutants
affected in lipid accumulation. Mutagenized colonies were first plated onto a LB
plate, thus creating a master plate. Next, this master plate was replica plated onto a
minimal medium plate with low nitrogen to induce wax ester and triacylglycerol
accumulation. Replica plating onto the minimal medium not only served to induce
wax ester and triacylglycerol biosynthesis, but it also eliminated auxotrophic mutants

generated by the mutagenesis. Next, these low nitrogen plates were flooded with a

Vleble ctu'e (per ml)

33

 

1 .OOE+08

1.00E+O7 ‘

 

 

1 .OOE+06

 

1 .00E+05

 

 

1 .OOE+04

 

1 .OOE+03

 

1 .00E+02

l

I

 

1 .00E+01
I

 

 

 

 

_.‘- Viable cfu's

- I- am ctu's

 

 

 

 

 

 

 

1.00E+OO i
o

 

10

 

20

“me (min)

40

50

FigureZ-Z: Efiectofdurationomeueaunentcncellviabilityover'time
followingNTGmntagenesisandindncednnmba'ofRifcclcnies.

3
1311 cfu

“a (peg ml)

0

34

solution containing Sudan black B, a nonspecific lipophilic dye. Because wax esters
and triacylglycerol are very nonpolar, colonies containing these compounds are darker
in color, after being stained with Sudan black B, than colonies lacking neutral lipid
accumulation. The results from such a staining procedure are shown in Figure 2-3.
Following such a procedure approximately 10% of the mutagenized colonies were
selected as "lighter staining". To determine if these colonies were truly affected in
wax ester or triacylglycerol accumulation, the colonies from the master plate were
restreaked and then cultured in low nitrogen medium for analysis by TLC.

Mutant phenotypes. A total of more than 6400 chemically mutagenized
colonies were screened using the lipophilic dye Sudan black B in combination with
TLC analysis. A total of 25 mutants were recovered (Table 2-1). These 25 mutants
can be divided into 3 phenotypic classes (Figure 2-4). Class 1 mutants are (wax'tagi),
class II mutants are (waxitag') and class III mutants (wax'tag‘). Chemical mutagenesis
produced a wide variety of phenotypes within the three classes. The amounts of
neutral lipids in the 25 mutants are illustrated in Figure 2-5. The first mutant that was
isolated was a class I mutant. Because of its inability to accumulate wax esters the
mutant was given the name Wowl, for without wax. Subsequent mutants were
isolated, not all of which were solely deficient in wax ester accumulation. However
for the sake of simplicity in dealing with the large number of mutants that were
generated, the wow designation was adopted and a number was assigned to each
subsequent mutant that was isolated, despite it's phenotype.

Chemical complementation as a basis for characterization. Based on

previous observations by other researchers it was known that A. calcoaceticus could be

35

 

Figure 2-3: Example of staining for mutants affected in neutral lipid accumulation using
the lipophilic dye Sudan black B. Arrows indicate lighter staining colonies that would
have been selected for further study. In this example, the known wax' mutant, Wowl
was placed in the comers of the plate for comparison.

36

MutantClasses

 

Stth I IIIII
Wr-.
TG --
0 .

 

 

 

Pigme2-4:TLCp1eteilhrsnatingtheflneegmalc1ueesofmutents.
Classlmutantsfiiltoaccunmlatewaxestu'sunda'nin'ogenwax
inducing)lirnitedconditicm.C1asslImntantsfiiltoecclnmrlate
tiacylglycerolandcluslllnnrtantsaccnnndateneitha’waxescrui-
bysprayingthe'lLCplatewithSMsulfmicacidandchmingat

160°C. Waxestaandtriacylglycemlstandards,0.2mgeuh(8td),

A. calcoaceticus strainBD413 (Wt),wuestas(W),triacy131yceml(TG)
andthecrigin(0).

37

onfiiaaxggsggg
Egggaén_3m§s§81§easua§§_§€
abuses-sea? .oSEsuaauEufiufiaasassnfissa—aé
sauna-533383233 gaggvgsasui
E35339...» ggéaigifisg
gagggfiaiiqflg “mafia:

.. .l 0.1.-.. a..- .... t3

38

cultured in the presence of alkanes and fatty alcohols as a carbon source (Makula et
al., 1975; Dewitt & Ervin, 1982; Singer et al., 1985). These previous studies also
showed that when A. calcoaceticus is incubated in the presence of these substrates they
are directly utilized in wax ester accumulation. I conducted similar types of
experiments on the mutants that had been isolated in order to characterize the sites of
the biochemical lesions in the various mutants. The rationale for these experiments is
based upon the proposed pathway for wax ester biosynthesis (Figure 2-6). An
experiment in which the mutants were fed hexadecanol and subsequently there was an
accumulation of wax esters would be indicative of a lesion in the pathway at either of
the two reductase steps. If waxes were not formed then it might be concluded that the
mutation had affected the acyl-CoA fatty alcohol transferase. By growing the mutants
in the presence of hexadecane, it was hoped to reveal if the mutant phenotype was the
result of a lesion upstream of the outlined wax ester pathway that blocked the pathway
from getting the necessary substrate.

Culturing the mutants in the presence of hexadecanol (a 16 carbon alcohol) as a
substrate showed that all of the class 1 mutants were able to synthesize wax esters
(Figure 2-7). However, when the class I mutants were grown in the presence of
hexadecane no evidence for complementation of the phenotypes was apparent (data
not shown). Growth of the class II and class III mutants in medium containing
hexadecanol and hexadecane had no affect on the mutant phenotypes (data not shown).
These data are summarized in Table 2-2. Unfortunately, at the time of these
experiments a commercial source of fatty aldehydes was not available. However,

recently this has changed and further experiments will be carried out with

39

0

0011

\s/

Acyl-CoA
Reductase

C
WVWW/‘H

Fatty Aldehyde
Reductase

Woman

Fatty Alcohol \
Fatty Acyl-CoA Transferase

O=O

 

 

Figure 2-6: Proposed pathway for wax ester biosynthesis in Acinetobacter
calcoaceticus strain BD413.

4o

dyes—«'5‘.

88 . 83.53839qu . Evan—.83». . elk-al.1— onafinoofluiggéyb .
Eggémvgfizsgvfisuaooé_§ne§s§h guarding?
___u§ooo~oci§. zéigig. nova—083% gfiflcuoiafl
.8363 .55?85an. figgagufl—guoggi "blue—5E

an o

. lu-.----.. “0".-- .I .
0 t1 0-.....-

FI-‘

OOOOOIOC'O I I. O... 3

.mm

efiaaefiaq fieficfiefi a. ace-afloat as... an. «mam an. In. an a...

ac

41

Table 2-2: Effect of inclusion of hexadecane (0.3% w/v) and hexadecanol (0.3% w/v)
on neutral lipid composition of various mutants.

 

Strains Phenotype Phenotype on Phenotype on

 

Unsupplemented Hexadecane Hexadecanol

Wild type wax”tag+ wax“tag+ waxitag+
strain BD413

Wowl wax‘tag+ wax'tag+ waxitag+
Wow2 wax’tag+ wax'tag+ waxitag+
Wow3 wax'tag' wax'tag' wax’tag'
Wow4 wax’tag' wax'tag‘ wax’tag'
Wow5 wax'tag' wax'tag‘ wax’tag'
Wow6 wax*tag' wax+tag' waxitag‘
Wow7 waxitag' waxitag' waxitag'
Wow9 wax'tag' wax'tag' wax‘tag’
Wow10 wax'tag’ wax'tag' wax'tag’
Wowll wax*Mg' waxitag' waxitag'
Wow12 wax‘tag' wax'tag‘ wax'tag'
Wowl3 wax‘tag” wax‘tag" waftag"
Wow14 wax’tag“ wax'tag+ wax”tag+
Wow15 wax‘tag“ wax'tag“ waxitag”
Wow16 wax’tag‘ wax'tag' wax‘tag‘
W ow l 7 waxitag' waxing wax+tag'
Wowl 8 waxitag‘ waxitag' waxitag'
Wow20 wax*tag' wax+tag' “fag
Wow22 waxItag’ wax*tag' wax‘tag'
Wow23 wax'tag' wax'tag' wax'tag'
Wow24 waxitag' wax*tag' wax+tas
Wow25 wax‘tag‘ wax'tag’ wax'tag'
W ow26 wax'tag' wax'tag’ wax'tag'
Wow27 wax'tag' wax'tag' wax'tag’

42

Table 2-2 (cont'd)

Wow28 wax'tag+ wax'tag+ wax”tag+

 

cis-l l-hexadecenal to firrther characterize the mutant phenotypes.

Further chemical complementation experiments on Wow15. In order to
determine if the Wow15 mutant was able to convert a fatty aldehyde to a fatty alcohol
additional chemical complementation experiments were carried out on Wow15.
Previously described experiments illustrated the ability of hexadecanol to complement
the mutant phenotype, while incubation in the presence of hexadecane had no effect.
Chemical complementation studies specific to Wow15 were carried out using two more
substrates involved in wax ester biosynthesis.

Incubation of the mutant in the presence of cis-l l-hexadecenal (a 16 carbon,
monounsaturated aldehyde) restored wax ester production (Figure 2-8). This indicates
that the wow15 mutation inactivates an enzyme that converts either acyl-CoA, or acyl-
ACP to the corresponding aldehyde.

Tween-40, a sucrose polyester of palmitic acid, is a metabolically active and
transportable form of palmitic acid for many organisms (Shintani & Ohlrogge, 1995).
It would be predicted to act as a source of hexadecanoic acid in chemical
complementation experiments of Wow15. As would be predicted based on chemical
complementation experiments with hexadecane, when the mutant was grown in the
presence of Tween-40, it failed to accumulate wax esters (Figure 2-8). This lends
further evidence that a mutation has occurred in the step involving the reduction of the

acyl-CoA, or acyl-ACP, to its corresponding aldehyde or alcohol.

43

Wild'l‘ypo Wow15
81234512345

 

 

w .‘ "’10! H

Figu'ez-S: MWofflreWowlsm Cellcultrnesfi'crn
Ammmlflwfldtypehndmenanomemin
vuiousrnedie. Mundhpidswaemmdwidrchlmofilmmspottod
Wybuedcncelhrlerwetweiglnmflremm
sepaatedmdvinnfinedbyqnyingflrepletewiflrSMmlfiniceddndch-fing
dléo’C. SunpleswuogrowninLBMel),IB+03%TVrem—40(lme2),low
nihogmmhfimelmedieflme3),10wniflogmnfinhnelmedie+0.3%1m-40
(1ene4)end1owniuogmminime1medie+0.3%eis-ll-hendeceno1(1me5). We!
“MMWCTQMWGDWMS
wacloededdOlrngeech.

CONCLUSION

The neutral lipids of four different strains of A. calcoaceticus were analyzed by
TLC. One of these strains, BD413, accumulated the greatest amount of wax esters
when induced under nitrogen starvation conditions. There was no accumulation of
wax esters evident by TLC analysis when the bacteria were grown in LB medium.
This observation reaffirms findings that wax ester accumulation in A. calcoaceticus is
induced under nutrient starvation conditions. It was also observed that when strain
BD413 is grown under nitrogen limited conditions, it accumulates a substantial amount
of another neutral lipid that comigrates with triacylglycerol. This finding is consistent
with reports describing the accumulation of triacylglycerol in A. calcoaceticus under
similar growth conditions (Makula et al., 1975; Scott & Finnerty, 1976; Singer et al.,
1985). These two findings, together with the fact that A. calcoaceticus strain BD413
is naturally competent for genetic transformation, led me to select it for further study
(Juni & Janik, 1969; Palmen et al., 1993). It was my hope that by mutagenizing strain
BD413 I might identify mutants not only in wax ester biosynthesis, but also in
triacylglycerol accumulation.

Chemical mutagenesis using the guanidine analog, nitrosoguanidine, was carried
out on A. calcoaceticus strain BD413 to generate mutants affected in wax ester and
triacylglycerol production. A total of 25 mutants were recovered. The mutants could
be separated into three distinct classes based on their neutral lipid composition. Class
I mutants fail to accumulate wax esters under nitrogen limited conditions. Class II
mutants fail to accumulate triacylglycerol, and class III mutants are affected in their

ability to accumulate both wax esters and triacylglycerol under nitrogen starvation

45

conditions. As seen in Figure 2-5, several of the mutant phenotypes are leaky.
Leakiness in a mutant phenotype is a very common occurrence, especially when using
a mutagen that generates point mutations. There could be several different
explanations for the observation. These could include mutations in the promoter
elements of genes involved in wax ester biosynthesis, such that these genes are not
being transcribed as efficiently as in the wild type. Another likely scenario is that a
missense mutation has occurred that reduces the activity of the enzyme or causes
destabilization of the protein or mRNA. Finally, another possibility is that the
mutation has not occurred in a wax ester or triacylglycerol structural gene. Instead the
mutation has affected another gene upstream of this pathway, in nitrogen sensing and
regulation, for example. These upstream mutations may cause a decrease in wax and
triacylglycerol accumulation, because the bacteria do not respond to the environmental
conditions.

A limited characterization of the mutant phenotypes was carried out using
chemical complementation experiments. Hexadecane is known to be taken up by A.
calcoaceticus and directly incorporated into wax esters (Makula et al., 1975). Thus, it
might be expected that culturing the class I (wax') mutants in the presence of
hexadecane might give some indication of whether or not the mutations were due to
mutations in structural genes involved in wax ester biosynthesis or in genes involving
pathways prior to wax ester biosynthesis, like nitrogen regulation. When the mutants
were grown in medium supplemented with hexadecane it was observed that none of
the mutants (class I, II and III mutants) exhibited a wild type phenotype. Although

these results are difficult to interpret by themselves, they become more meaningful in

46

conjunction with results from the experiment outlined below.

Incubation of the class I (wax’) mutants with hexadecanol resulted in normal
levels of wax accumulation in all cases. This implies that since wax biosynthesis from
hexadecanol is still possible in these mutants the acyl-CoA (or acyl-ACP) fatty alcohol
transferase has not been affected. Thus, the fact that the mutants did not accumulate
wax when grown on hexadecane suggests that the class I (wax') mutants may be
deficient in either of the first two reductase steps (Figure 2-6). Another interpretation
is that the mutants have been affected in their ability to take up and utilize
hydrocarbons. Since it is believed that wax ester biosynthesis starts with either an
acyl-CoA or acyl-ACP substrate, it is hard to imagine that a mutation in the synthesis
of either one of these compounds would be compatible with cell viability. Therefore,
it seems likely that the 8 class I mutants were the results of mutations in one of the
two reductase steps.

Wow15 was firrther characterized by additional chemical complementation
experiments involving 0.3% Tween-40 or 0.3% hexadecenal. These experiments were
carried out at a much later date than previously described experiments involving
hexadecane and hexadecanol, but were included here for the sake of consistency.
Growth of this mutant on both hexadecane and Tween-40 (a sucrose polyester of
palmitic acid) resulted in no production of wax esters. However, incubation of this
mutant in the presence of hexadecanol resulted in the production of wax esters. This
indicates that either the reductase involved in converting acyl-CoA (or acyl-ACP) to
aldehyde, or the second reductase which catalyzes the conversion of aldehydes to

alcohols, had been mutagenized. Growth of the Wow15 mutant on cis-l l-hexadecenal

47

(a commercial source of fatty aldehyde was not available until after the chemical
complementation experiments were completed) restored the ability of this mutant to
synthesize wax esters. This shows that the genetic lesion in this mutant is an inability
to catalyze the conversion of acyl-CoA (or acyl-ACP) to the corresponding fatty
aldehyde.

Triacylglycerol levels in the class II mutants (tag') did not respond to the
addition of either alkanes or hexadecanol to the growth medium. This was predicted
since it is assumed that the mutations that are being observed in the class II mutants
are the result of a mutation in a gene involved in triacylglycerol biosynthesis or the
regulation of triacylglycerol biosynthesis. The possibility that these mutations are the
result of a defect in nitrogen sensing and response seems remote since they accumulate
significant amounts of wax esters, indicating that the nitrogen response systems are
intact. Incubation of wild type stain BD413 in the presence of either compound results
in a greater accumulation of both wax esters and triacylglycerol implying that substrate
availability may be a limiting factor in wax ester accumulation.

Chemical complementation studies of the class III mutants with both
hexadecane and hexadecanol indicated that wax and triacylglycerol accumulation in
these mutants is not restored by nutritional supplementation. I interpreted this to mean
that these mutants may be the result of mutations in genes that permit the organism to
sense and respond to nitrogen levels in the medium. The possibility that the class III
mutants are defective in an enzyme involved in wax or triacylglycerol biosynthesis is
unlikely because there is no known enzyme involved in both pathways after the steps

involved in acyl-ACP biosynthesis and no such enzyme can be readily envisioned.

48

Thus, the class III mutants are either due to two mutations in structural genes for
pathway specific biosynthetic enzymes, or in regulatory genes that participate in

regulation of both pathways.

REFERENCES

Benning, C.. 1991. Genetic Analysis of the Pathway for the Biosynthesis of the Plant
Sulfolipid in the Purple Bacterium Rhodobacter sphaeroides. Ph.D. Dissertation,
Michigan State University.

Bergogne-Bérézin and Joly-Guillou, M. L.. Antibiotic Resistance Mechanisms in
Acinetobacter. In The Biology of Acinetobgcter: Taxonomy, Clinical Immrtance,
Moleculpr Biology, Physiology, Industrial Relevance. Plenum Press, New York, New
York. 1991. Eds. Towner, K. J., Bergogne-Bérézin, E., and C. A. Fewson.

Dewitt, S., Ervin, J. L., Howes-Orchison, D., Dalietos, D., and Neidleman, S. L..
1982. Saturated and Unsaturated Wax Esters Produced by Acinetobacter sp. H01-N
Grown on Cm-C20 n-Alkanes. JAOCS 59(2): 69-74.

Ervin, J. L., Geigert, J ., Neidleman, S. L., and Wadsworth, J.. Substrate-Dependent
and Growth Temperature-Dependent Changes in the Wax Ester Compositions Produced

by Acinetobacter sp. HOl-N. In Biotechnology for the Oils and Fats Indu_s_tgy,
American Oil Chemists Society, 1984. Eds. Ratledge, C., Dawson, P. and Rattray, L..

 

F ixter, L. M., Nagi, M. N., McCormack, J. G. and Fewson, C. A.. 1986. Structure,
Distribution and Function of Wax Esters in Acinetobacter caIcoaceticus. J. Gen.
Micro. 132: 3147-3157.

F ixter, L. M. and Sherwani, M. K.. Energy Reserves in Acinetobacter. In .13
Biology of Acinetobacter: Taxonomy. Clinical Importance, Molecular Biology,
Physiolog, Industrial Relevance. Plenum Press, New York, New York. 1991. Eds.
Towner, K. J., Bergogne-Bérézin, E., and C. A. Fewson.

Juni, E. and Janik, A.. 1969. Transformation of Acinetobacter calco-aceticus
(Bacterium anitratum). J. Bact. 98(1): 281-288.

Juni, E. 1972. Interspecies Transformation of Acinetobacter: Genetic Evidence for a
Ubiquitous Genus. J. Bact. 112:917-931.

Kates, M.. 1986. Techniques of Lipidology: Isolation, Analysis and Identification of
Lipids. Elsevier Science Pub. Co., New York, New York.

49

Makula, R. A., Lockwood, P. J. and F innerty, W. R.. 1975. Comparative Analysis of
Lipids of Acinetobacter Species Grown on Hexadecane. J. Bact. 121( 1): 250-258.

Palmen, R., Vosman, B., Buijsman, P., Breek, C. K. D., and K. J. Hellingwerf. 1993.
Physiological Characterization of Natural Transformation in Acinetobacter
calcoaceticus. J. Gen. Micro. 139:295-305.

Scott, C. C. L. and F innerty, W. R.. 1976. Characterization of Intracytoplasmic
Hydrocarbon Inclusions from the Hydrocarbon-Oxidizing Acinetobacter Species H01-
N. J. Bact. 127(1): 481-489.

Shintani, D. K. and Ohlrogge, J. B.. 1995. Feedback Inhibition of Fatty Acid
Synthesis in Tobacco Suspension Cells. The Plant Journal 7(4): 577-587.

Singer, M. E., Tyler, S. M. and Finnerty, W. R.. 1985. Growth of Acinetobacter sp.
Strain HOl-N on n-Hexadecanol: Physiological and Ultrastructural Characteristics.
Mol. Gen. Genetics 162: 162-169.

CHAPTER 3

ISOLATION AND CHARACTERIZATION OF A CLONE COMPLEMENTING THE
WAX ESTER MUTANT WOW] OF A. CALCOACETIC US

ABSTRACT

Wax esters are present in a variety of different organisms. They perform many
different functions such as providing desiccation tolerance in the form of epicuticular
waxes on the surfaces of plants, or serving as a means of carbon storage in microbes.
Acinetobacter calcoaceticus strain BD413 is an example of a microorganism that
accumulates wax esters as a means of carbon storage when it undergoes nutrient
starvation in the presence of excess carbon, particularly nitrogen starvation. Mutants
that failed to accumulate wax esters under nitrogen limited conditions have been
described previously. In this chapter it was discovered that by providing cis-l 1-
hexadecenal (a monounsaturated 16 carbon fatty aldehyde) in the growth medium of
one of these mutants, Wow15, it was possible to restore wax accumulation in the
mutant. This suggests that the mutant is defective in the synthesis of fatty aldehyde
from acyl-CoA (or acyl-ACP). Two of the mutants, Wowl and Wow15, were
complemented with a cosmid genomic library. The ability of the cosmid to
complement the phenotype was localized to a single gene (acrl) encoding a protein
that is 44% identical (over 264 amino acids) to an open reading frame identified in

Mycobacterium tuberculosis that is thought to encode an enzyme involved in mycolic

50

prod
lillOc

The

liter
step:

m

N

libra

mat

TLC

had

ITans
COSm

inabi

51

acid metabolism. Expression of the acrl gene in Escherichia coli resulted in
production of a functional enzyme that catalyzes the reduction of an acyl-CoA
thioester directly to its corresponding long chain alcohol via an aldehyde intermediate.
The reduction of acyl-CoA to its corresponding fatty alcohol by a single enzyme is a
unique enzymatic function and has not previously been detailed in the scientific
literature. Observation of this activity dispels the notion that three separate enzymatic
steps would be directly involved in wax biosynthesis, thus reducing the number of

enzymes needed to two.

INTRODUCTION

In the previous chapter, the isolation and characterization of mutants in wax
ester and triacylglycerol accumulation were described. Here I describe the isolation of
a gene that complements the wow15 phenotype.

To complement the mutated genes from wowl and wow15, a cosmid genomic
library was constructed. This library was used to transform the mutants by triparental
mating. The resulting exconjugates were screened for complementation by staining the
samples with Sudan black B followed by examination of the neutral lipid profiles by
TLC. The ability of the mutants to synthesize wax esters indicated that the mutation
had been complemented by genes localized on the cosmids.

The cosmids containing the complementary genes were mutagenized with the
transposon mini-TnlOCm to delineate the regions of interest. The mutagenized
cosmids were then used to transform the mutants, wowl and wow15, respectively. The

inability of a mutagenized cosmid to complement the mutant phenotype was evidence

118

C01

we

d‘J:

COI

up

all 1

Ill

desc

this

52

that the transposon had inserted into a region containing a complementary gene. The
restriction pattern of 4A-55, the cosmid which complements the wow15 phenotype,
was compared to the restriction patterns of 4A-55 cosmids which were unable to
complement the mutant phenotype because of the transposon insertions. Fragments
were identified in mutagenized cosmids that showed an increase in molecular weight
due to the presence of the transposon. The corresponding fragment from the
complementary cosmid was subcloned and sequenced.

DNA sequencing of the fragment allowed the identification of an open reading
frame of interest. The open reading frame was subcloned and the encoded protein was
expressed in E. coli. Expression of the protein in E. coli allowed the development of

an in vitro assay which demonstrated the protein's function.

MATERIALS AND METHODS

Bacterial stains and plasmids. The bacterial strains used in the experiments
described in this chapter are shown in Table 3-1. The source of the plasmids used in
this chapter, or in the construction of novel plasmids is presented in Table 3-2.

Growth and culture conditions. Low nitrogen minimal medium (per liter, 2.0
g KH2P04, 1.18 g succinic acid, 0.1 g NH,SO4, pH adjusted to 7.0 with solid KOH;
after autoclaving add 20 ml of sterilized 2% MgSO,) was used in experiments with
Acinetobacter calcoaceticus strain BD413 to induce wax ester formation. High
nitrogen minimal medium, which was the same as the above except for the addition of
1.0 g NH,SO, per liter. For other purposes such as DNA isolation, A. calcoaceticus

was grown and maintained on LB medium (10 g Bacto tryptone, 5 g Bacto yeast

53

Table 3-1: Bacterial strains used in Chapter 3.

 

Bacterial Strains

Relevant
Characteristics

Source or Reference

 

A. calcoaceticus

ATCC #33305, strain
BD413

Ac412

Wowl

Wowl :Rif

Wow2

Wowl3

Wowl4

Wow15

Wow15zRif

Wow28

E. coli
HBlOl
DHSa

MM294
13132103133)

wild type used during
these studies,
unencapsulated mutant

of A. calcoaceticus strain

BD4

TrpE’ mutant of strain
BD413

wax' null mutant of
BD413

spontaneous Rif mutant
of wowl

wax' null mutant of
BD413

wax' null mutant of
BD413

wax' null mutant of
BD413

wax' null mutant of
BD413

spontaneous Rif mutant
of wow15

wax' null mutant of
BD413

F' proA2 recA13 mch

F'/endAl recAl
A(lacZYA-argF) U169
(¢80dlacA(lacZM15)

F- endAl hst17 thi-l

F- ompT hstB(rB'mB')
gal dcm(DE3)

ATCC stock center

A gift from Dr. Elliot
Juni (Juni, 1972)

This study, chapter 2
This study, Chapter 3
This study, Chapter 2
This study, Chapter 2
This study, Chapter 2
This study, Chapter 2

This study, Chapter 3

This study, Chapter 2

(Maniatis et al., 1982)
(Raleigh et al., 1989)

(Bachmann, 1987)

Novagen

lat

.VI

Pb;

54

Table 3-1 (cont'd)

M01655 wild type, for carrying
out Tn mutagenesis

Phage

ANK l 324 mini-Tn] 0Cm

A gift from Dr. Nancy
Kleckner

(Kleckner et al., 1991)

 

pE'

pSl

pSl

pSl

pSl

p51

55

Table 3-2: Plasmid sources and derivations for Chapter 3.

 

Plasmid

Description or
Construction

Source or Reference

 

pBS-KS‘“, pBS-KS‘
pRK2013

pLA2917

pET2 1

pSER2
pSRl

pSR2

pSR6

pSER2zacrl

Bluescript Vector

Km' self-transmissible
RK2 derivative
containing ColEl
replicon and transfer
functions to mobilize
RK2 derivatives

Cosmid Vector (T et')
derived from RK2

Transcriptional

expression vector for use

in E. coli.

A. calcoaceticus/E. coli
shuttle vector (T et’Km')

Bluescript with EcoRV
fragment from 4A-55

Bluescript with EcoRV
fiagment from 4A-55.

Gave rise to sequence
ID#4-9

Bluescript with EcoRV
fragment from 4A-55.
Gave rise to sequence
ID#1-3

pSER2 derivative

containing PCR fragment

amplified using primers
P5 and P6 (Figure 3-8)

Stratagene

(Figurski & Helinski,
1979)

(Allen & Hanson,
1985)

Novagen

Figure B-l, Appendix
B, this study

this study

this study, chapter 3 for
construction and
appendix A for
sequence information

this study, chapter 3 for
construction and
appendix A for
sequence information

this study

4A

SE]

SE!

SE

SER

Table 3-2 (cont'd)

pET2 1 :acrI

lA-3F

4A-5 5

2A-87

SER101

SER102

SER103

SER104

56

pET2] derivative for
protein expression of
acrl in E. coli strain
BL21(DE3). Contains
PCR fragment amplified
using primers P7 and P8
(Figure 3-8)

pLA2917 derived cosmid
clone complementing the
Wowl mutant

pLA29l7 derived cosmid
clone complementing the
Wow15 mutant

pLA2917 derived cosmid
clone complementing the

Wow15 mutant

mini-Tn l 0Cm'
containing derivative of
4A-55 that is no longer
able to complement the
Wow15 mutant

mini-TnlOCm'
containing derivative of
4A-55 that is no longer
able to complement the
Wow15 mutant

mini-TnlOCm‘
containing derivative of
4A-55 that is no longer
able to complement the
Wow15 mutant

mini-Tn l 0Cm'
containing derivative of
4A-55 that is no longer
able to complement the
Wow15 mutant

this study

Figure 34, this study

Figure 3-3, this study

this study

this study

this study

this study

this study

Table 3-2 (cont'd)

SER105

SER106

SER107

SER108

SER109

SER110

57

mini-TnlOCmr
containing derivative of
4A-55 that is no longer
able to complement the
Wow15 mutant

mini-Tn 1 0C mr
containing derivative of
4A-55 that is no longer
able to complement the
Wow15 mutant

mini-TnIOCmr
containing derivative of
4A-55 that is no longer
able to complement the
Wow15 mutant. Sample
contains rearrangement
or deletion as observed
by restriction analysis.

mini-TnIOCm‘
containing derivative of
4A-55 that is no longer
able to complement the
Wow15 mutant

mini-Tn 1 0Cmr
containing derivative of
4A-55 that is no longer
able to complement the
Wow15 mutant. Sample
contains rearrangement
or deletion as observed
by restriction analysis.

mini-Tn] 0Cmr
containing derivative of
4A-55 that is no longer
able to complement the
Wow15 mutant

this study

this study

this study

this study

this study

this study

 

58

extract and 10 g NaCl per liter, pH 7.0). A. calcoaceticus cultures were typically
grown overnight at 30°C. In the case of larger volume cultures (i.e. 50 mls and
larger), 3 ml overnight cultures were collected by centrifugation, the cell pellet was
washed in the new medium, recentrifuged and then added to the fresh medium, before
being incubated at 30°C overnight. Maintenance and growth of Escherichia coli
strains was on LB with appropriate antibiotics. Escherichia coli strain BD21(DE3) for
expression studies was grown in LB medium with ampicillin (100 ug/ml). Three
milliliter overnight culutures were collected, washed and used to inoculate larger 50 ml
cultures. When the cultures reached an optical density of 0.6 at 600 nm, they were
induced to synthesize the protein of interest by adding IPTG to a final concentration of
0.1 mM. Antibiotics were commonly used in the following concentrations: ampicillin
100 ug/ml, chloramphenicol 50 ug/ml, kanamycin 25 ug/ml, rifampicin 50 ug/ml and
tetracycline 15 ug/ml.

Complementation Screening Using the Lipophilic Dye Sudan Black B. The
master plates were replica plated onto low nitrogen minimal medium and were
incubated overnight at 30°C to induce wax formation. The induced colonies were then
stained by irrigating the plates with sudan black B (0.02% in 50:45:5
DMSO:ethanol:water) and gently shaking for approximately 20 minutes. The stain
was aspirated away and the plates were carefully washed with 70% ethanol and gently
shaking them for approximately 2 minutes. Lighter staining colonies were identified
from the stained plates and the corresponding colony from the master plate was
subsequently analyzed by thin layer chromatography.

Thin layer chromatography (TLC). TLC was carried out in the same manner

59
as described in Chapter 2.

Library Construction. A. calcoaceticus genomic DNA for the construction of
the cosmid library was prepared in the following manner. A 200 ml culture of A.
calcoaceticus strain BD413 was grown overnight. Cells were collected by
centrifugation and resuspended in 16 mls of buffer (8% sucrose, 50 mM Tris pH 8.0,
50 mM EDTA pH 8.0). Lysozyme (Sigma) was added to a final concentration 2
mg/ml. Cells were incubated at 30°C for 30 minutes to make spheroplasts. Twenty
four milliliters of lysis buffer (3% SDS, 0.5 M Tris-HCI pH 8.0, 0.2 M EDTA pH 8.0)
was added and the sample was incubated at 65°C for 30 minutes. The sample was
then cooled on ice. A sucrose step gradient was prepared for centrifugation in the
following manner. Five milliliters of 50% sucrose on the bottom, 10 mls of 20%
sucrose in the middle, and 10 mls of 15% sucrose on top in the lysis buffer described
above. Ten milliliters of the sample was then layered on top of the sucrose solutions
and the gradient was centrifuged in an SW222 (Beckman) swing out rotor at 27,000
rpm for 3 hours at 15°C. The DNA was recovered as a pellet above the 50% sucrose
step, from where it was removed to a dialysis bag. The DNA was dialyzed in 1.5 L of
1x TE overnight, with change of the solution. The DNA was then gently extracted
with 1 volume of phenol then extracted with 1 volume of chloroform/isoamyl alcohol.
The DNA was dialyzed in 4 L of TE for 2 days then was partially digested with
SauIIIA to a mean size of approximately 25 kb. Fragments of approximately 25 kb
and larger were size selected by nmning the partially digested DNA on a 0.6% agarose
gel, cutting the band containing the fragments of interest from the gel, and isolating

the DNA by electroelution (Maniatis et al., 1982). The fragments were ligated directly

60

to BglII-digested cosmid, pLA2917, and packaged using Gigapack Gold packaging
extracts (Stratagene). Approximately 1300 E. coli HBlOl primary transfectants were
selected and transferred to 96 well plates. Examination of 19 randomly chosen
cosmids from the library by restriction analysis indicated that 68% of the transfectants
contained inserts.

Triparental filter matings. Identification of a complementary cosmid was
carried out by triparental filter matings of the cosmid genomic library to Wowl and
Wow15 in the presence of MM294, a helper strain. Matings were carried out for all
the mutants in the manner described for Wow15 below. The cosmid library was
contained in E. coli strain HBIOl which was Tet' due to the presence of the cosmid.
MM294 is a strain containing pRK2013, a helper plasmid that enables triparental
mating through its mob genes. The pRK2013 plasmid imparts Km'. Finally, the
mutant strain Wow15zRif was the target of the mating and was a Rif’ derivative of
Wow15. It was obtained by plating Wow15 on rifampicin plates and selecting for the
presence of spontaneous Rif colonies. The neutral lipid profiles of these Rif strains
were examined to make sure they matched those of the original Wow15 mutant. At
the end of the mating, Wow15:Rif' mutants containing the cosmids were selected by
plating the product of the cross onto LB medium containing rifampicin, to select for
Wow15zRif', and tetracycline to select for the presence of the cosmid. MM294 and the
HBlOl donor failed to grow under these conditions.

The Wow15 strain and MM294 were grown overnight as 3 ml cultures, and the
library was grown in 96-well plates as replicates of the original. The day of the

mating, 0.5 ml of MM294 was used to inoculate a 50 ml of culture, and 3 ml of

61

Wow15 was used to inoculate a 50 ml culture. Cultures were collected at an
ODm=0.6, washed and then resuspended in 50 ml of LB. Twenty five milliliters of
MM294 was combined with 50 ml of the Wow15 culture. Ten milliliters of this
mixture was drawn onto a sterilized 45p filter (85 mm in diameter) via a vacuum
apparatus creating an even lawn of bacteria. The filter was then removed and placed
on to an LB plate. Using sterilized set of prongs arranged to match 48 wells of the 96
well titer plates containing the library, the library was stamped onto the lawn of
bacteria and matings were allowed to incubate at 30°C overnight. Filters were then
transferred to selective medium containing rifampicin (100 pg/ml) and tetracycline (15
ug/ml) to select for Wow15:Rif containing the cosmids. The resulting patches were
then restreaked onto a master plate containing the selective medium, before being
replica plated onto minimal, low nitrogen, wax-inducing medium for subsequent
analysis by lipophilic staining and TLC.

Transposon mutagenesis of the complementary cosmid. Phage vehicle
XNK1324 was used to transfect E. coli strain MG1655 containing the cosmids that
complemented the Wow15 and Wowl phenotypes. Protocols for transfection, growth
and maintenance of ANK1324 have been published by Kleckner et. a1. (Kleckner et al.,
1991). Resulting colonies were selected on tetracycline (15 ug/ml) for the presence of
the cosmid and chloramphenicol (50 pg/ml) for the presence of the transposon. To
separate insertion events that were located in the genome versus the desired insertions
in the cosmid, approximately 3000 transfectants were pooled and cosmid DNA was
isolated. The resulting cosmids were used to transform E. coli strain DH5a. Cosmids

with transposon insertions were selected by plating the transforrnants on medium

62

containing tetracycline and chloramphenicol. One hundred and ninety six of the
resulting transforrnants were transferred to 96-well plates and used for triparental
matings with Wow15 and Wowl as previously described. The resulting exconjugates
were then screened by TLC looking for insertions that resulted in the loss of the ability
of the cosmid to complement the mutant phenotypes.

Extended PCR. To delineate the region on cosmid 4A-55 that was responsible
for complementation of the wow15 phenotype, extended PCR was used in conjunction
with transposon mutagenesis of the cosmid. To map the location of transposon
insertions relative to the cosmid, primers were constructed on either end of the BglII
site of pLA2917 facing in toward the insert. Additionally, primers for both ends of the
transposon element were constructed to face out, toward the edge of the insert DNA.
Using one anchor on the cosmid in conjunction with the primers specific to the
transposon (Table 3-3), it was possible to amplify the DNA in between the primers
allowing us to assign a distance of the transposon relative to the anchoring (cosmid
specific) primer (Figure 3-3). Transposon mapping of the cosmids was possible using
Boehringer Mannheim's Expand Long Template PCR System. Reactions contained
350 pM of dNTP's, 300 nM of each primer, 5 pl of 10x buffer 1 (17.5 mM MgC12,
500 mM Tris-HCl, pH 9.2, 160 mM (NH4)ZSO,), 0.15 pg of template DNA and 0.75
pl of the supplied enzyme mix in a total volume of 50 pl. Cycling was carried out in
the following manner:

Step 1: 94°C for 2 minutes
Step 2: 94°C for 10 seconds

Step 3: 65°C for 30 seconds

63

Table 3-3: List of synthetic oligonucleotides used in Chapter 3.

 

Name of primer

Sequence (5'-3')

Description

 

P1

P2

P3

P4

P5

P6

P7

CTTTCTTGCCGCCA
AGGATCTGATG

GGCCGGAGAACCT
GCGTGCAAT

GACGGGGTGGTGC
GTAACGGC

CAGGCTCTCCCCGT
GGAGGTAAT

GCAGGATCCTTGGG
ATTGAACATATTG

GCAGGATCCGGTGC
GAI I IATGATGTA

GCAGGATCCAAAA
CATTGGTAAI I ICA
GATACT

Primer used in mapping
the location of Tn
insertions in 4A-55.
Specific to 5' side of
BglII site of pLA2917 in
the Km cassette which
originated from Tn-S.
Faces toward BglII site

Same as 5' Km Tn-S but
specific to the 3' side of
the BglII site.

Primer used in mapping
the location of Tn
insertions in 4A-55.
Specific to 5' end of
mini-TnlOCm facing
out.

Same as 5' Cm Tn-lO,
but specific to 3' end of
the transposon.

Used with 3'End of
Reductase to amplify
PCR product to generate
pSER2:acr1. Contains
BamHl linker at 5' end.

Used with 5'End of
Reductase to amplify
PCR product to generate
pSER2zacr1. Contains
BamHI linker at 5' end

Used with 3' Reductase
+EcoRI to amplify PCR
product to generate
pET21:acr1. Contains
BamHI linker at 5' end.

64

Table 3-3 (cont'd)

P8 GCAGAATTCGGTGC Used with 5' Reductase
GATTTATGATGTA +BamHI to amplify PCR
product to generate
pET21zacr1. Contains
EcoRI linker at 5' end

 

Step 4: 68°C for 7 minutes

Step 5: repeat from step 2 10 more times

Step 6: 94°C for 10 seconds

Step 7: 65°C for 30 seconds

Step 8: 68°C for 7 minutes + 20 seconds each cycle

Step 9: repeat from step 6 15 more times

Step 10: 68°C for 7 minutes

Step 11: Hold at 4°C

Genomic DNA preparation. Genomic DNA for Southern blot analysis was

prepared by growing 3 ml cultures of the bacteria overnight in LB at 30°C. Half of
the culture was collected by centrifugation in a microfuge at maximum speed for 5
minutes. The pellet was washed with 500 pl of 10 mM Tris-HCl (pH 7.6), 5 mM
EDTA (pH 8.0). The sample was centrifuged, and resuspended in 350 pl of the above
buffer. To this 50 pl of 10% SDS and 100 pl of 2.5 mg/ml stock of pronase (Sigma)
was added. The samples were incubated at 37°C for 1 hour. Samples were then
drawn through a 1 ml syringe with an 18 gauge needle attached to it 3 times to
partially shear the DNA. Samples were extracted once in 1 volume of phenol, twice in

1 volume of phenol/chloroform (50:50) and once in 1 volume of chloroform. The

65

supernatant was removed and the DNA precipitated from solution by the addition of 2
volumes of 100% ethanol. DNA was recovered by spindling it out of solution with a
capillary tube that had been scaled and bent using a bunsen burner. As much ethanol
was removed as possible by gently touching the sample to the side of the eppendorf
tube and letting the ethanol drain off. DNA was gently transferred to 100 pl of the
above buffer and allowed to enter solution by incubating the sample overnight at 4°C.

Colony lifts for hybridization. Ninety five millimeter nitrocellulose filters
(Amersham) were placed on top of LB plates containing 15 pg/ml of tetracycline.
Replicas of the genomic library were stamped onto the filters using a sterile prong
device that matched the 96 well array containing the library. The colonies on the
filters were allowed to grow for approximately 6 hours at 37°C, or until colonies were
just evident. Filters were removed and placed onto 3 mm Whatman paper saturated
with 0.5 M NaOH for 5 minutes. The filters were then transferred to 3 mm Whatrnan
paper saturated in 1 M Tris-HCL, pH 8.0 for 3 minutes. Next, the filters were
neutralized on 3 mm Whatrnan paper soaked in 1 M Tris-HCL, pH 8.0 plus 1.5 M
NaCl for 3 minutes. Filters were briefly washed in 2x SSC and then dried on 3 mm
Whatman paper before being baked at 80°C for 30 minutes to fix the DNA to the
filter.

Restriction digestions and southern blot analysis. Restriction digestions were
carried out using Pharmacia's One-Phor-All buffer system and restriction enzymes.
Southern blot analysis and detection was performed in keeping with Boehringer
Mannheim's instructions for their Genius system with luminescent and nitro-blue-

tetrazolium detection. In brief, probes were labelled by incorporation of digoxigenin-

66
ll-UTP. In detecting the probe, the filter was washed twice in 2x SSC, 0.5% SDS at

room temperature. Two high stringency washes were carried out at 65°C for 15
minutes in 0.5x SSC, 0.5% SDS. Finally, to detect the probe, the filter is incubated in
the presence of anti-digoxigenin antibody conjugated to alkaline phosphatase. The
conjugated antibody was then visualized by either Lumi-Phos 530 or nitro-blue-
tetrazolium.

Nested deletions and DNA sequencing. To expedite the process of DNA
sequencing, nested deletions were constructed using Promega's Erase-a-Base system.
In brief, 1.0 pg of DNA was digested with EcoRV and Kpnl to yield one blunt end
which was susceptible to digestion with DNA exonuclease III. Samples were removed
every 30 seconds, yielding a nested set of deletions varying by approximately 250 hp
at each time point. The treated DNA was then recircularized and used to transform E.
coli strain DHSa. The nested deletions were sequenced on a Perkin Elmer AB1310
automated sequenator using dye deoxyterrninator reactions.

Database analysis of DNA sequences- To determine if identified DNA
sequences shared any similarity to previously characterized proteins, DNA sequences
were typically compared by BLASTX (Altschul et al., 1990) alignment to GenBank
release 92.0 via an electronic mail server.

Overexpression of acrl in E. coli, protein purification and separation. The
acrl gene was PCR amplified from the complementary cosmid using synthetic
oligonucleotide primers, P7 and P8, containing BamHI and EcoRI linkers, respectively,
for directional cloning (Table 3-3). The PCR product was gel purified, digested and

subcloned into Novagen's pET21 transcriptional expression vector to produce plasmid

67

pET21:acrl which was used to transform E. coli strain BL21(DE3). E. coli strain
BD21(DE3) for expression studies was grown in LB medium with ampicillin
(lOOpg/ml). Three milliliter overnight cultures were collected, washed and used to
inoculate 50 ml cultures. When the cultures reached an optical density of 0.6 at 600
nm, they were induced to synthesize the protein of interest by adding IPTG to a final
concentration of 1 mM. Cultures were grown for 2.5 hours before being collected and
processed.

To isolate the protein from cells containing pET21:acr1, or just pET21 as a
control, cells were harvested by centrifugation at 5000 x g for 10 minutes. Cells were
resuspended in 500 mM sodium phosphate buffer (pH 7.4) and incubated 30°C for 15
minutes in the presence 100 pg/ml of lysozyme (Sigma). Cells were then sonicated
for two 40 second bursts at maximum power. Soluble proteins were separated from
cell walls and insoluble materials by centrifugation at 35,000 x g for 30 minutes at
4°C. The soluble fraction was collected as fraction 1. The insoluble fraction was
resuspended and resonicated as before. The sample was centrifuged at 35,000 x g for
30 minutes at 4°C and the aqueous layer was collected as fraction II. The resulting
pellet was resuspended in a minimal amount of phosphate buffer to make a suspension
as the insoluble fraction (fiaction III). SDS-PAGE analysis of proteins was carried out
using Pharmacia's Phast Gel system using 12.5% homogeneous gels as per the
manufacturers instructions for SDS-PAGE gels and silver staining.

Enzymatic assays for reductase activity. To test for enzymatic activity of the
expressed protein from transformed E. coli, an assay was developed using radiolabelled

palmitoyl-l-“C-coenzyme A (44.4 mCi/mmol, 30 pM final concentration per reaction,

68

(Dupont). Reactions were run in 30 pl volumes containing 167 mM sodium phosphate
buffer (pH 7.4) in the presence of 100 pM NADPH and 13.5 pg of protein at 30°C for
15 minutes. Components were added in the following order: water, buffer, NADPH,
palmitoyl-l-“C-coenzyme A and finally the protein. The assays were then extracted
with 75 pl of chloroformzmethanol (50:50), vortexed for 10 seconds and then
centrifuged for 20 seconds for phase separation at maximum speed in a microfuge.
The chloroform phase was then removed and spotted onto a TLC plate where the
lipids were separated using a hexanezethyl ether:acetic acid (90:10:1) solvent system.
The TLC plate was then removed and allowed to dry before being exposed to a
phosphorimaging cassette.

Synthesis of acyl-ACP. In order to determine if acyl-ACP was used as a
substrate by acrl, it was necessary to synthesize this substrate fi'om l-"C-palmitic acid
and ACP using acyl-ACP synthase (a kind gift from Dr. Jan Jaworski). Thirteen and a
half microcuries (77 pl, 55.5 mCi/mmol, 3.54x104 dpm/pl) of 1-"C-palmitic acid in
hexane was dried under nitrogen gas and resuspended in 100 pl of ethanol. Two drops
of concentrated ammonia was added, the mixture was incubated at 65°C for 5 minutes
and dried under nitrogen gas. The sample was then resuspended in 250 pl (final
concentration 600 pM, 1.1x10‘ dpm/pl) of 20% oxidant free Triton X-100 and heated
to 65°C for 5 minutes.

Oxidant free Triton X-100 was prepared in the following manner. One hundred
microliters of Triton X-100 was mixed into solution with 5 ml of 10 mM Tris-HCI, pH
8.0. Approximately 100 mg of NaBH4 was added to the solution. The solution was

sealed in a teflon lined screw capped tube, shaken vigorously and incubated at 37°C

69

for 30 minutes. Concentrated HCI was added dropwise with vortexing until the
addition of acid did not produce anymore foaming. The solution was then extracted
twice with 2 ml of chloroform. The chloroform phases were dried down under
nitrogen at 55°C until there was no remaining smell of chloroform.

To synthesize the acyl-ACP, the following were combined in a screw cap
microcentrifuge tube. Fifty microliters of 5x TML solution (0.5 M Tris-HCl (pH 8.0),
25 mM MgCl,, 2 M LiCl), 5 pl of 0.1 M DTT, 12.5 p1 of 0.1 M ATP (pH 7.6), 25 pl
20% oxidant free Triton X-100 (described above), 25 pl of 600 pM 1-"C-palmitic acid
(the ammonium salt, described above, total of 2.7x10’ dpm), 25 pl of acyl-ACP
synthetase (a gifi fi'om Dr. Jan Jaworski), and 8 pg of ACP protein (Sigma)
resuspended in 22.3 pl of 10 mM Tris-HCl (pH 8.0). This reaction mixture was
incubated at 37°C for 3 hours. At the end of the reaction the mixture was diluted 10x
with H20 to reduce the LiCl concentration.

A gravity flow column was prepared with a 100 pl bed of DEAE cellulose in a
0.25" diameter, 3 ml column. The reaction was gently layered on top of the bed and
allowed to gravity flow through the resin. The bed was then washed 4 times with 2
ml of wash solution (19.6 ml H20, 80 ml isopropanol, 0.4 ml of 5 M NaCl, 0.2 ml of
1 M KZHPO4 (pH 6.0)). The bed was then washed with 2 ml of 50 mM Tris-HCl (pH
7.6). The column was then centrifuged to dryness at 1000 rpm in a table top
centrifuge for 2 minutes. The acyl-ACP was then eluted from the bed of the column
by washing the cellulose four times, each time with 100 p1 of 0.4 M LiCl in 50 mM
Tris-HCl (pH 7.6). Between each application of the 100 pl of LiCl solution, the

column was centrifuged for 2 minutes at 1000 rpm and the fraction collected. The

BE

1h:
EX

a l

l},

70

radioactivity present in ten microliter samples from each fraction was determined by
scintillation counting. The first fraction was the most concentrated, at 1.26x10‘ dpm
per 10 pl. The total counts for the entire volume collected measured 2.66x10’, making
the efficiency of the reaction about 97%. A total of 1.0x10‘ dpm (approximately
5.6x10" pmol of labelled acyl-ACP) of fiaction 1 was used for each reaction testing

for acrl activity from the E. coli protein extracts.

RESULTS AND DISSCUSION

Library construction and testing. In order to complement the mutant
phenotypes, a cosmid genomic library was prepared by partially digesting A.
calcoaceticus genomic DNA with SauIIlA. Fragments greater than 25 kb were size
selected fi'om a gel, purified and ligated to pLA2917, a broad host range cosmid with a
unique BglII site for the insertion of foreign DNA (Figure 3-1). To determine the
number of cosmid clones that contained insertions, cosmid DNA from 19 clones was
prepared. It was observed that 13 of 19 clones contained insert DNA of approximately
25-30 kb in size. Thus, only 68% of the colonies selected contained insertions. Using
the following equation it is possible to predict the number of colonies (e.g.
exconjugates following mating to the mutants) that would need to be screened to give

a 99% coverage of the genome (Zilsel et al., 1992):

N: ln(1-P) _ ln(l -o.99)

ln(l _£) ln(1- 25000
J’ 4x106

 

 

)

Where N is the number of colonies that need to be screened, p is the percentage

71

 

 

 

 

3' I. C u '1 I.

I) t : L - I s. ; ’1' ‘1'
.......... “a“.m‘--.

0" Tc - ‘“-“~.
.1 f, :4. ; t 5 “ ‘t ‘ -
r I! s C Is to s. FI It at I! at
P! h
t: .- $
30" us cum “I "(I ”I" 111“!"

0101!“

Figure 3-1: Detailed restriction map of pLA2917 (taken from Allen, 1985). The vector
is shown linearized at one of the SphI sites. Abbreviations: Bg, BglII; Bs, BstEII; c,
cos site; H, HindIII; Hp, HpaI; Km, kanamycin; Ps, PstI; P, Pqu; R1, EcoRI; RV,
EcoRV; S, SalI; Sp, SphI; and Te, tetracycline. (a) pLA2917; (b) expanded view of
the region between the MI site in the tetracycline cassette and one of the EcoRI sites;
(c) same as (b) showing antiobiotic resistance determinants, cos site and unique
restriction sites. SauIIIA partially digested Genomic DNA fragments from A.
calcoaceticus were inserted into the Bng site located at the 5' end of the Km' cassette.

72

chance of covering the genome (99%), x is the insert size (25 kb) and y is the genome
size (assume 4x10° bp). This makes N=746 colonies. I recovered approximately 1300
colonies, almost twice as many as calculated above, and arranged them in 96-well
plates. In order to test the utility of the library, it was mated to Ac412, a known
tryptophan auxotrophic mutant of A. calcoaceticus (Juni, 1972). Five trp+ exconjugates
were recovered implying that the gene was adequately represented in the library. This
indicated that the library was adequately representative of the genome.

Complementation of Wowl and Wow15. The library was then mated to
Wowl and Wow15 (both class I, wax' mutants). The lipophilic dye Sudan black B was
used to identify darker staining colonies, which might contain greater amounts of
neutral lipids than the mutants. These darker staining colonies were then further
investigated by TLC analysis to confirm whether or not they contained normal levels
of wax esters. Following examination of 50 exconjugates by TLC, a cosmid clone,
lA-3F was found to complement the wowl phenotype (Figure 3-2). After searching
through 350 exconjugates two cosmids, 2A-87 and 4A-55, were found to complement
the wow15 phenotype. Two of these original cosmids, lA-3F and 4A-55, were used in
all subsequent work.

In the case of 1A-3F, the end fragments from the insert region of the cosmid
were used as probes to screen the genomic library by colony hybridization to identify
other overlapping cosmids that also complemented the wowl phenotype. These are
summarized in Table 3-4. Only the two original cosmids that were found to
complement the wow15 phenotype were isolated.

The cosmids were restriction mapped with a limited number of restriction

73

Wowl
4.

Std Wt Wowl lA-3F

w w
, ’ 3
a3

 

 

 

 

Merl-2: Woflbwl. This‘l'LCplmeilluetruedreebilltyoMe
connid.lA-3Ftocornplarmflremlphmotype. Surpleewuegrownmda
ninogmlimitedcouditicne. IipidewueviufiudbyepreyingmenCphewiflr
mmwmmampmerwc. Waxeetu'mm
(TG)stndmtk,0.2mgeech(8tl),A.cdcoamelninBD413(Wt)mdtho
edema)

74
Table 3-4: Cosmids that share homology to 1A-3F.

 

 

Name of Cosmid Phenotype of transformed wowl
1A-3F waxi
23-2A wax"
3A-3C wax”
3A-3D wax+
3A-3F wax‘
3A-6F wax‘
4A-1D wax'
5A-4E wax'
5A-4H wax’
6B-4D wax‘
7A-3C wax'
7A-4B wax’“
7A-4D wax”
7B-3C wax'
10A-2C wax‘
10A-3C wax"
llB-2A wax+
12A-2D wax+
123-5E wax'
13A-3F wax+
13A-5E wax'

13B-4D wax'

were
hon:

I0»

deter

com;

Io»

75

enzymes due to their large size (Figures 3-3 and 3-4). It was found that many of the
restriction enzymes that were tried (i.e. HindIII, EcoRV and PstI) resulted in a large
number of bands making it very difficult to order them on the map. Comparison of
the restriction patterns of the two different cosmids (IA-3F and 4A-55) when they
were digested with a variety of different enzymes indicated that the cosmids were
distinct and had no obvious overlap. Additionally, when the end fragments of 1A-3F
were used as hybridization probes against the cosmid 4A-55 (data not shown), no
homology was evident. This implies that the cosmids do not overlap, and therefore,
Wowl and Wow15 represent mutations at two different loci.

Next, the cosmid 1A-3F was mated into all of the different Wow mutants to
determine if it was able to complement any of the other mutants. Besides
complementing Wowl, the cosmid was also observed to complement Wowl3 and
Wow14. Both of these mutants were class 1 null mutants (wax'), the same phenotype
as Wowl. This leaves only two other class 1, null mutants, Wow2 and Wow28 that
have not been complemented. A similar experiment using 4A-55, the cosmid that was
found to complement Wow15 has not yet been completed.

Transposon mutagenesis of complementary cosmids. Because of the large
sizes of the cosmids (55 kb), the genes were localized on the cosmids by mutagenizing
the cosmids with a transposon. The mutagenized cosmids were then screened for
insertions that eliminated the ability of the cosmid to complement the mutant
phenotype. Mutagenesis of 4A-55 with a Tn-10 derived transposon, produced a total
of 10 insertions out of a total of 192 that resulted in the loss of the ability of the

cosmid to complement the wow15 phenotype. Two of the 10 mutations resulted in

76

A)

 

2gb E S S K

1
k

A

   

A

Figure 3-3: Restriction map of cosmid 4A-55 and locations of the transposon
insertions. The transposon, mini-TnIOCm, illustrated in Panel A (Kleckner et al.,
1991) was used to generate mutants of 4A-55 which were unable to complement the
wow15 phenotype. The approximate position of primers used in mapping the
positions of the insertions by extended PCR are illustrated as P3 and P4. The
triangles at the ends of the transposon represent the inverted repeats of the
transposon.. Panel B represents the cosmid 4A-55, which was found to complement
the wow15 phenotype. The insert DNA is represented by the lighter shaded line,
while the vector portion of the cosmid is the darker shaded line. The arrows facing
the insert portion of the cosmid (P1 and P2) symbolize the primers that were
constructed with specificity to the Km' cassette of pLA2917 and were used with P3
and P4 (above) to map the location of the transposon insertions. The small triangles
indicate the sites of transposon insertions that were found to inactivate the ability of
4A-55 to complement the mutant phenotype. The approximate location of the
transposon insertions were determined by extended PCR. The restriction sites are
EcoRI (E), BamHI (B), Hpal (H), Kpnl (K), SalI (S), ScaI (Sc), SmaI (Sm) and XhoI
(X).

77

 

5 kb
A) B B x HSm
_ I I l h
S Sm S B
s B 33 K 3 KB Ba x
B)

4kb

 

Figure 3-4: Restriction map of cosmid 1A-3F. This cosmid was found to complement
the wowl mutation. Panel A shows the insert DNA in respect to the cosmid vector,
pLA2917 (darker line). The insert was subcloned into the Bng site of pLA2917 as a
SauIIIA fragment, destroying the site. Panel B is an enlargement of the insert region
showing the following restriction sites, BamHI (B), Bng (Bg), deI (H), Kpnl (K),
SalI (S), SmaI (Sm) and XhoI (X).

78

some sort of deletion or rearrangement in the cosmids that were evident when they
were analyzed by restriction analysis. This left 8 insertions of interest. Mutagenesis
of the other cosmid, 1A-3F, has resulted in a total of 3 insertions that inactivate the
ability of the cosmid to complement the wowl phenotype.

The approximate location of the insertions relative to the restriction map have
been highlighted in Figure 3-3 for the cosmid 4A-55. The location of the insertions in
1A-3F have not yet been determined. Delineation of the transposons on the map was
possible by using extended PCR primed with oligonucleotides that were
complementary to the transposon (P3 and P4, Table 3-3) in combination with primers
(PI and P2, Table 3-3) sharing homology to the region surrounding the BglII site of
pLA2917 (Figure 3-3). It was observed that the transposons had inserted 2.0-6.5 kb
away from P2 primer by determining the size of the extended PCR products on an
agarose gel. The localization of all of the insertions to a small region of the cosmid's
DNA indicated that the wow15 gene probably did not reside within an operon, or if it
did, it was near the beginning of the operon. Localization of the transposons on the
restriction map showed that there was no single restriction fragment of practical size
that would encompass the region containing all of the transposon insertions (Figure 3-
3).

Delineation of the complementary region from cosmid 4A-55. In order to
subclone a fragment corresponding to the region that contained all of the transposon
insertions, the insertional mutants of cosmid 4A-55 were digested with several different
enzymes (i.e. EcoRV, Clal, and Nhel that are not on the restriction map) and the

restriction pattern compared to that of the wild type cosmid. Digestions with EcoRV

'.'.
ti 3
.

(“In

5123

323'
ECO

difi

HES
€10

inlt‘

79

resulted in a shift (of the predicted size) of a band in the transposon mutagenized
cosmids, that was not present in the wild type cosmid, 4A-55 (Figure 3-5). As seen in
Figure 3-5 the fragment that was shifted, did not disappear from the mutagenized lines,
indicating that there were two or more EcoRV fragments of approximately the same
size.

Wild type cosmid DNA was digested with EcoRV, resolved on a gel, and the
approximately 4.0 kb band was electroeluted. The resulting DNA was ligated to
EcoRV digested Bluescript vector and transformed into E. coli strain DHSa. Two
different isolates, designated pSR2 and pSR6, were identified based on restriction
patterns with enzymes other than EcoRV. These two different samples were used as
hybridization probes against wild type cosmid DNA and transposon mutagenized
cosmid DNA digested with EcoRV (Figure 3-6). It can be seen from panels A and B
of Figure 3-6 that the transposons have inserted into two different EcoRV fragments
that were subcloned as constructs pSR2 and pSR6. This indicated that the region of
interest spanned these two fragments.

In order to expedite the process of sequencing the two EcoRV fragments,
nested deletions of the fragments were made using Promega's Erase-a-Base
exonuclease kit. Single strand passes through the fragments provided enough sequence
information to distinguish a total of six open reading frames which were identified
based on their sequence similarity to identified open reading frames deposited in
GenBank (Appendix A, this text). The six sequence similarities reported by GenBank
are summarized in the table of Figure 3-7. The open reading fiame that was thought

to encode the gene of interest was localized to the end of the EcoRV fragment used in

mfihmwmmflwm“

80

12345678910111213

 

Figrne3-5: EcoRVdigesticmofM—SSendtnnepoeonccntainingdaivefives

of4A-55. EwRVrflgesfiomofM—SSMZLflIecosnddwhichcomplemmls

the wow15 phmotype, endthe4A-55 duivetivee, SERIOI- 110 (lanes 3-12,
)whlch ccnteinmini-TnlOCm. Theeetreneposonineertionswue

ngelpnifiedfiumEeoRVdigeefionsoftheccmplemmnycomnd,
4A-55,endsubsequartlysubclcned.

81

A)

1234567891011

6.5 kb — ,
'° - "I' ..
4.3 kb — , a T- u -

 

 
    
 

B) 1234567891011
6.5kb- _ _
4.3kb- .. .--_b...

Figure 3-6: Southern analysis of transposon mutagenized cosmids. Cosmids
containing the mini-Tn10Cm from ANK1324 which inactivates

their ability to complement the wow 1 5 phenotype were digested with Bg/Il.
The blots were probed with the subcloned EcoRV fragment, pSR6 (panel A),
and pSR2 (panel B), each of which had been subcloned from the original
complementary cosmid 5A-Fl. Wild type cosmid DNA, 4A-55, that is able
to complement the wow 1 5 phenotype has been loaded in lanes labelled lane 1.
Cosmids SER101-110 (lanes 2-11, respectively) contain the transposable
element.

82

Figure 3-7: Map showing sequence ID#1-9 in respect to one another. The boxed
regions highlight regions of homology that were detected between the sequence and
GenBank release 92.0 by BLASTX analysis (Altschul et al., 1990). These data are
summarized in the table below the map. The numbers surrounding the boxes indicate
the start of the similarity between the DNA query sequence (the numbers above the
line) and the amino acid numbers in the matching protein from GenBank (the numbers
below the line). This is to give some sense of the encoded protein over the whole
restriction fragment, and also helps to determine the amount of DNA sequence missing
between gaps. The gene, acrl, spans sequence ID#3-4 with the promoter residing on
the very end of ID#4, and was first found to match cmal from Mycobacterium
tuberculosis. Sequence ID#1-3 are from pSR2 and ID#4-9 come from pSR6. The
table summarizes the GenBank information giving the length of the matching protein,
accession number of the matching protein, the BLASTX score, probability score and
the name of the protein as reported by GenBank. Cases where the description of the
first match is ambiguous, second and third matches are included. Contig maps and
actual DNA sequences can be found in Appendix A of this text.

83

 

 

 

 

 

an ass".

 

 

 

 

 

 

 

 

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mm: ski—x00 now $.00 2&me 890... u m
t imMml llfleocsé mmm :8 .m 80: ommExO 082:5 moonom E m
wvm 3.9066 own :00 .m E0: 30205 Swain 8031
i illli mmcoEogoml E0: :2 mm
liiwmm li il $.9me l ova .m omstmom. 5 522m 9. mdv .8350 I mamoK «mi \.
iii «:2 ll (\zlil «.2 access. 02 $2 S2 0
9:. $906.». mmm 68> E9302 0 55¢. ostoono£ oo<-O mopoom E m
iii magi: gootoEm E0:
ll an sotfim 3 385 No v
vmm $303.0 mum can 0. 0.0 o $839003 Eatoacmeco 3: $6an 0 m
iii SmEm: message: m
new 9.036...“ 9: vmtmm
0mm 9.3me mm: 30va m N
F lili .
En l $953 mmm 0082 -odobu £3: Secs Newman? 350.. < F
W28 oEE£ 9.00m 0.60m xcmmoeow a 530 we tonfizz o>on< no.
.35 mcEiofis. :0 3.3.30.5 ._.m<4m n .836 on :82: :0 oEoz . 576023 02... =... E .mom
5 co.— EoEE mesons—2 :29:
a N a E E N a E i. a an a 5 E a
an 2. E .2 2 a t = 323 E a E a a _
I

8

as:

mu

CUT;

cm}
H!
pm
{0 c
L531

{30

(AI

0R

im:
C011

“'33

84

the construction of pSR6, and the promoter region of the gene resided on the end of
pSR2. This is in keeping with the previous finding that the region of interest,
delineated by the transposon insertions, spanned the two EcoRV fragments as
witnessed by the results presented in Figure 3-6, panels A and B. Additionally, the
open reading frame was of interest because of its strong similarity to an open reading
frame identified in Mycobacterium tuberculosis that is thought to be involved in
mycolic acid biosynthesis. Therefore, this open reading frame and its surrounding
sequence was completely sequenced on both strands (Figure 3-8 and Appendix A for
contig maps).

Comparison of the fully sequenced open reading We identified from A.
calcoaceticus with ORF 2 from Mycobacterium tuberculosis resulted in an optimized
FASTA score of 609 (Figure 3-9) indicating a very strong similarity between the two
proteins. ORF2 from Mycobacterium tuberculosis was sequenced as part of an effort
to characterize cmal, cyc10propane mycolic acid synthase. It is an open reading frame
that resides just downstream of cmal. ORF2 is reported to be homologous to ActIlI
(30% identity over 188 amino acids), a B-ketoacyl reductase from Streptomyces
cinnamonensis (a gene involved in chain elongation during polyketide biosynthesis)
(Arrowsmith et al., 1992). Based on this similarity and its location between cmal and
ORF3, an open reading frame with 35% identity (over 278 amino acids) to a
trifunctional hydratase/dehydrogenase/epimerase from Candida topicalis which is
involved in peroxisomal degradation of fatty acids, Yuan et. al. (Yuan et al., 1995)
concluded that the probable role of ORF2 was in mycolic acid metabolism. A three-

way alignment of wow15, ORF2 from Mycobacterium tuberculosis and ActIII from

85

Figure 3-8: DNA and protein sequence of the region containing acrl. A conserved
Shine Delgamo sequence is indicated in bold, underlined type (346-349 hp), While
possible promoter elements are highlighted in underlined italics (-10 box) and in italics
(-35 box). The EcoRV site which divides the two EcoRV fragments which were
subcloned to give pSR2 and pSR6 is highlighted at 336-341 bp. Priming sites used for
PCR to generate pSER2zacrl and pET21zacrl are double underlined (P5, 25-42 hp;

P7, 202-225 bp; P6 and P8, 1325-1342 bp (complementry strand».

61
121
181
241

301

361

421

22

481
42

S41
62

601
82

661
102

721
122

781
142

841
162

901
182

961
202

1021
222

1081
242

1141
262

1201
282

1261
1321
1381
1441
1501
1561

1621

IRR
CRR
RRC

Asn

GGT
Gly

GIR
Val

CAA
Gln

IIR
Lau

CGI
Arg

RIG
Mat

AAG
Lys

cor
Arg

GCC
Ala

CCA
Pro

GCA
Ala

CGI

RRG
RRC

RII
CCR

ICC

Lys

GCR
Ala

TTA
Lau
Gly

TCA
Sar

ICG
Sar
Gln

cor
Arg

TTT
Pha

GAG
Glu

ATG
Mat

GRI
ASP

CIG

RIR
RRG

RGR

RRR

RRR
Lys

TCT
Sar

TTG
Lau

GGA
Gly

CAA
Gln

AIT
Ila

CTG
Lau

RRR
Lys

TCT
Sar

GIR
Val

ATC
Ila

CTC
Lau

GCG

IGG

CRT

86

m 661' m CGG W m

TGG CCR CRG III GRG CGI RRR III TAT.RRR RRR CCI CIG

,ATI IGC III RII RIC GIR IGR IGI ICR IRR IIG

IRI

GCR
CRC

IRC
CII

Lau

RGI
Sar

GII
Val

CAC
Gln

CAA
Gln

CGC
Arg

RRI
Asn

AAT
Asn

GCT
Ala

CTC
Lau

GCA
Ala

ATT
Ila

ICR

RII IGR IIG

EB

RIC IRG

m m w W m CAC m

RII IIR III

III RGR IIR RCI IIR GII CIG
RRI RRI GRC RGC CII IRC RGI
GRR GCI CIC IIC CGR GRG RRI

Glu Ala Lau Pha Arg Glu Asn

GGA ATC GGT TTG ACG ATT GCA
Gly Ila Gly Leu Thr Ila Ala

GCC CGA ACC CAA GAA ACA CTG
Ala Arg Thr Gln Glu Thr Lau

GCC TCT ATT TTT CCT TGT GAC
Ala Sar Ila Pha Pro Cys Asp

RII RIG GCC RGI GIC GRI CRT
Ila Mat Ala Sar Val Asp His

CGT GCC GTA CAC GAG TCG TTT
Arg Ala Val His Glu Sar Pha

IRC III GGI GCG GIR CGI IIR
Tyr Pha Gly Ala Val Arg Lau

GGC CAG ATC ATC AAT ATC AGC
Gly Gln Ila Ila Asn Ila Sar

TAT GTC GCG TCT AAA GCT GCG
Tyr Val Ala Sar Lys Ala Ala

AAG CAT AAA ATC TCA ATT ACC
Lys His Lys Ila Sar Ila Thr

CCC ACC AAA ATT TAT AAA TAC
Pro Thr Lys Ila Tyr Lys Tyr

GTC TAC GCC ATT GTG AAA CGT
Val Tyr Ala Ila Val Lys Arg

RII RCC IRI GCC RIC GCR CCR

GRR

IIG

GIR
Val

AAA
Lys

GRR
Glu

CTG
Lau

GTC
Val

car
ASP

GTG
Val

TCT
Sar

CTG
Lau

ICG
Sar

GTG
Val

CCA
Pro

GRC

EcoRV

RRR
Lys

AGA
Arg

GAI
Asp

ATT
Ila

CCC
Pro

ACA
Thr

RIC

GGI
Gly

ATT
Ila

GIG
Val

GAC
Asp

TTC
Pha

TTC
Pha

RRI
Asn

GGT
Gly

GCC
Ala

TAT
Tvr

ACG
Thr

cor
Arg

RRI

RIR ICR RIC

RRR
Lys

GCT
Ala

AAA
Lys

ATG
Mat

CTG
Lau

CAI
His

TTA
Lau

GTA
Val

TTC
Pha

RIG
Mat

CTT
Lau

RII
Ila

RRI

CCC III RRC

ROG GRR RRR

GTG GCT TTG
Val Ala Lau

GCG GCA GGT
Ala Ala Gly

GCI GCR RII
Ala Ala Ila

RRI GCG RII
Asn Ala Ila

ATC AAT AAT
Ila Asn Asn

CAT TTT GAA
Asp Pha Glu

CTG CCA CAT
Lau Pro His

TTG GCC AAT
Lau Ala Asn

RGI CGC IGI
Sar Arg Cys

CCA TTG GTG
Pro Lau Val

TCC CCA GAA
Sar Pro Glu

GCG ACG CAC
Ala Thr His

RII CIG RIG

3%

RIR
RII

CGI
CGC
RIC

Ila

GCI
Ala

GRR
Glu

one
Asp

GCA
Ala

coc
Arg

ATG
Mat

GCG
Ala

CTT
Lau

cor
Arg

GRR
Glu

TTG
Lau

ICG

RIC

RIR

GTG
Mat

ACT
Thr

CAT
His
Gln

Glu

Gly

ACC
Thr

ATT
Ila

RCC
Ihr

TCA
Sar

RCC
Thr

GCC
Ala

GGT
Gly

RII

Arg Lau Ala Sar Ila Thr Tyr Ala Ila Ala Pro Asp Ila Asn Asn Ila Lau Mat Sar Ila

GGA
Gly

CTA
Lau

CCI
RCI
RIR

GCR

ICG

GIR

III
Pha

CAA
Gln

CIC
III
ICI
GIC
CCR
GRR

IGI

Figure 3-8

AAC
Asn

cor
Arg

RIR
CCI
IIR
RCC
RCC
RIC

CCC

CTA
Lau

GCC
Ala

CCG
GII
RGI
IGR
GRR
IGC

TTC CCA AGC TCA ACG GCT GCA
Pha Pro Sar Sar Thr Ala Ala

TAT GCC CGC TTG TTC CCA GGC
Tyr Ala Arg Lau Pha Pro Gly

AGA soc III III ATG GIT ace
TTI.BEE.I§B.IBfl-BIS.§SB.SSA
ACA GAR CTA TGC IGR ATA are
TCA ATA AAI GCT TTG CTT AAT
rec ATG AGT soc CCA AGC ace
TIA TGT rec ART err AAC rec

RCC RGR RIC RCI IIG GRR CCI

CTG
Lau

GRR
Glu

RCC
RCR
IRI
ICR
IRI
CRI

IIG

GGT
Gly

CAC
His

RIC
RIR
IGR
CGC
IGG
GCG
GCI

GRR
Glu

TGG
Trp

RGC
ICR
RIR
GGI
RIG
CIG
IGR

GAG
Gln

IRR

CRG
RII
IIR
IGC
IGI
GIG

CRG

GAA AAA TTG
Glu Lys Lau

RRI IIR IRR

RII IRG RGG
ICI IIG CGR
IRG IGR RCR
RIR RIR ICR
GGC RII ICC
RCC GCR CCR
GCR RR

AAT
Asn

RRG

RRR
ICC
IIC
RRI

CTG
Lau

RRG

IIG
RGC
RIR
RCR
RIR
ICG

‘F‘0

“3...:

«y- .- .
0...: g

o...

has:

A I.
i?..'.

ACRE

V l“ n
he;

~-
I
p.“

87

No. Target file Definition Match} Over. INIT OPT 1
ORPZ.AMI 44.7 264 597 609
10 20 30 40 so so
ACAR1.AMI VNKKLEALFRKNVKGKVALITGASSGIGLTIAKRIAAAGAHVLLVARTQETLBEVKAAIB
.X::::::: .......
ORF2.AMI HLDPDRARRNDPLLGRHVIITGASSGIGRASAIAVAKRGATVFALARNGNALDELVTBIR
so 90 100 110 120 130
70 so 90 100 110 120
ACAR1.AMI QQGGQASIPPCDLTDMNAIDQLSQQIMASVDHVDFLINNAGRSIRRAVKBSPDRFKDFER
ORP2.AMI AHGGQAHAFTCDVTDSASVEHTVKDILGRFDHVDYLVNNAGRSIansvvusroaLnnraa
140 150 160 170 180 190
130 140 150 160 170 180
ACAR1.AMI TMQLNYFGAVRLVLNLLPHMIKRKNGQIINISSIGVLANATRFSAYVRSKAALDAFSRCL
ORFZ.AHI vuAvuvPGAvRMVLALLpnwasaarcnvvuvsSAGVQARprrssriérxAALoAéonv °°°°
200 210 220 230 240 250
190 200 210 220 230 240
ACAR1.AMI SAEVLKHKISITSIYMPLVRTPMIAPTKIYKYVPTLSPBEAADLIVYAIVKRPTRIAIHL
ORF2.AMI ASETLSDHIrrruiuuéLVATpuvasARLupvaAisfiéRAAAfiviRéLvéiéAaibréi
250 270 280 290 300 310
250 260 270 280 290
ACAR1.AMI GRLASITYAIAPDINNILMSIGFNLPPSSTAALGEQBKLNLLQRAYARLPPGBHW
...... x
ORF2.AMI GTLAEAGNYVAPRLSRRILHQLYLGYPDSAAAQGISRPDADRPPAPRRPRRSARA
320 330 340 350 360

Figure 3-9: Optimized FASTA alignment between acrland ORF2. An optimized
FASTA alignment between acrl (ACARIAMI) and ORF2 (ORF2.AMI) from
Mycobacterium tuberculosis was prepared using the Lipman-Pearson algorithm. The
Optimized FASTA 'score for these two protein sequences is observed to be 609 and is
44.7% identical over 264 amino acids. Symbols used in the above alignment are (z)
indicating a full query/target match, (.) denotes a partial match and (X) represents
where the homologous region between the two sequences starts and stops.

inten

t:-
(1"

km

sequ
"him

246:'

Lilli;
(PS.

The

M

88

Streptomyces cinnamonensis is shown in Figure 3-10. It can be observed from this
alignment that the open reading We identified from A. calcoaceticus (labelled
ACAR.1) is more similar to ORF2 than ActIII (optimized FASTA alignment between
acrl and ActIII is observed to produce a score of 274, 29.8% identity over 245 amino
acids, Figure 3-11). Thus based on the similarity between the identified open reading
frame and ORF2 from Mycobacterium tuberculosis, it was my feeling that the gene of
interest may encode some sort of B-ketoacyl reductase. This would be in keeping with
the biochemical model outlined in Figure 3-12, where acyl-CoA (or acyl-ACP) is
acting as the substrate for the encoded enzyme and the ketone group, which is a B-
ketone, is being reduced.

Cloning of acrl: Examination of the open reading frames present in the gene
sequence indicate that the largest protein, 32468 KDa, would be encoded using GTG
(base pairs 358-360, Figure 3-8) as a translational initiation codon as compared to a
24655 KDa protein when ATG (base pairs 583-585 on Figure 3-8) is used. Use of
GTG as an initiation codon is further supported by the presence of a strongly
conserved Shine Delgarno sequence of AGG from -10 to -13 bp (base pairs 346-349 of
Figure 3-8) upstream of the predicted start site, while there is no such conserved
sequence upstream of the first ATG codon.

The open reading frame predicted by the use of GTG as the translational
initiation codon was subcloned by PCR from the complementary cosmid using primers
(PS, P6, P7 and P8) specific to the regions illustrated in Figure 3-8 (also Table 3-3).
The amplified product resulting from primers P5 and P6 was subcloned into the

Baml-II site of pSER2 (construction of pSER2 is detailed in appendix B of this text),

‘34-. .1177 .1 'I‘f‘v.’l!‘.’l“i-TE*:\Z 31121:;.'.-.'3«‘.".'.-'.A i-llll.?)'r‘fi‘f~".3.l-‘l4 .‘qtépztspm't
------------------- -- ---- ~ “H-SF-XII-
12 0 7 30 1-73

181 TIIIIMI lilti- V-TO IIIEIVKAAI Pt". I PII-MN
191 A4“ .PI ITI IIG NPIIILV'T- IM-w‘l -SASV
131E'1-IUI:I' 1" "FMIAI I101»r ?‘
.3.)

.9121: 151 DQIS GINA; 'I-FI I — mFI "01-
F 151 IHTIKCIMJ F-Y- .av bI-YI .1-
" - 151 “my. 1'<::1'.- -va:(':21 [IAINI- 13:1113'

4'. . . . - ‘4 ..
Mi 201 III- I: I W II-I-L WATPM'
1:: ~11“? «"701 “- WP'PFIi11'M FNPFYISIL .-
:0: Fl I};- c an t it: - comm-I- amt

C.

G

,.
ilIl

‘~-v
a ‘U

llu

m mls-MK Kasai?- ” x roam Its-{IQ
- is: .111; -‘ =3: DWASII‘ISD mm. Ms I— R NP'iAI "MI
-23 mMu .lw c u Id-xl- mass mm mm

a? 34:
9:91. . .1 YAI'KE‘T- AI-s- 7::At-D1NNH - LMSIGFN: F "1.1.”
71 1 acrla- III-EA Gr-R SF Hot-CY II-CIIS
- »-. -------- Iv III---— Ids-”A m» I c Imm-

360 3'7: 39:“ 39: 4;:
2.7-. .11". 4. 1 lriI:_".-.I-<I.N1.L clA- - - - . - - - - YA. If“. . . .......... .1
i-F". F?"T.»\E)E<E1‘?>.»".P P.PPF.~‘3‘.I~4A(§ -- 9.212;. m. .. .......... .;
.1127 4-1 ------------------ 1_m.~mr"a .-—-r:Nv_ . _ ....... , .z 1

mylm'l‘lno-wafimnofaal (ACARIAWORFZMW
mmmmmndmamsummumcnm
mafimuisbuedonflnfliggimmdmpdgodflmmlflsmflmingh
DNASISforWindowuofiwuapachgeaMSoflm)wimflnfiollowing
Mzgqpmahyofi, topdiagonalsofS, fixodmpunltyoflo, finding”
pmhyoflo, K—upboflndawindowdnofs.

90

No. Target file Definition Matcht Over. INIT OPT
2 ACT3.AMI 29.8 245 237 274
10 20 30 40 50 60
ACAR1.AMI RENVKGKVALITGASSGIGLTIAKRIAAAGAHVLLVARTQBTLBBVKAAIEQQGGQASIF
.X::.:::.:::: ...............
ACT3.AMI MATQDSEVALVTGATSGIGLEIARRLGKBGLRVFVCARGEEGLRTTLKBLREAGVBADGR
10 20 3O 40 50 60
7O 80 90 100 110 120
ACAR1.AMI PCDLTDMNAIDQLSQQIMASVDHVDFLINNAGRSIRRAVHESFDRP- HDPBRTMQLNYFG
ACT3.AMI TCDVRSVPBIBALVAAVVERYGPVDVLVNNAGRPGGGATABLADBLWLDVVBTNLTGVPR
70 80 90 100 110 120
130 140 150 160 170 180
ACAR1.AMI AVRLVLNLLPHMIKRKNGQIINISSIGVLANATRFSAYVASKAALDAFSRCLSABVLKHK
ACT3.AMI VTKQVLK- AGGMLERGTGRIVNIASTGGKQGVVHAAPYSASKHGVVGFTKALGLBLARTG
130 140 150 160 170
190 200 210 220 230 240
ACAR1.AMI ISITSIYMPLVRTPMIAPTK-IYKYVPTLSPBBAADLIVYAIVKRPTRIATHLGRLASIT
....... :x ..:.::: .....
ACT3.AMI ITVNAVCPGFVBTPMAASVREHYSDIWRVSTBBAFDRITARVPIGRYVQPSEVABH--VA
180 190 200 210 220 230
250 260 270
ACAR1.AMI YAIAPDINNILMSIGFNLFPSSTA
ACT3.AMI YLIGPGAAAVTAQALNVCGGLGNY
240 250 260

Figure 3-11: Optimized FASTA alignment between acrl and ActIII. An optimized
FASTA alignment between acrl (ACAR1.AMI) and ActIII (ACT3.AMI) from
Streptomyces cinnamonensis was prepared using the Lipman-Pearson algorithm. The
Optimized FASTA score for these two protein sequences is observed to be 274 and is
29.8% identical over 245 amino acids. Symbols used in the above alignment are (z)
indicating a full query/target match, (.) denotes a partial match and (X) represents
where the homologous region between the two sequences starts and stops.

91

0
ll

W\/\/\/\/\/\/c\s/COA

Acyl-CoA
Reductase

o
5
\/\/\/\/\/\/\/\/\H

Fatty Aldehyde
Reductase

W\/\/\/\/\/\/CH20H

Fatty Alcohol \
Fatty Acyl-CoA Transferase

O=O

 

 

Figure 3-12: Proposed pathway for wax ester biosynthesis in Acinetobacter
calcoaceticus strain BD413.

92

an A. calcoaceticus/E. coli shuttle vector, to give pSER2:acrl. This construct was
used to transform E. coli strain DHSa for amplification and analysis, followed by
subsequent transformation into the A. calcoaceticus, mutant Wow15. Transformants
were grown under wax inducing conditions, collected and extracted with
chloroform/methanol for isolation of lipid compounds. Resolution of the lipid fraction
by TLC indicated that Wow15 transformed with pSER2zacrl was now able to
synthesize wax esters (Figure 3-13), indicating that correct open reading frame had
been identified, and that the gene responsible for the wow15 mutation had been cloned.

Hydropathy analysis. Hydropathy analysis of the protein sequence via the
Kyte and Doolittle prediction (DNASIS for Windows, Hitachi Software) (Kyte &
Doolittle, 1982) is illustrated in Figure 3-14. Panel A shows the hydropathy plot when
a window of 21 amino acids was used to identify putative membrane-spanning
domains. Since it has been observed that wax ester inclusions are surrounded by a
single membrane (Scott & Finnerty, 1976), hydropathy prediction was repeated using
a window of 10 amino acids (Panel B, Figure 3-14). The use of a 20 amino acid
window showed the protein to be very hydrophobic, suggesting the possibility that the
majority of the protein is associated with a hydrophobic environment. Use of a 10
amino acid window gave strong indication of several potential membrane spanning
domains.

In vitro reductase activity assays. To test the enzymatic function of the
wow15 gene product in vitro, the gene was subcloned into pET21 . The P7 and P8

primers (Table 3-3) specific to the region illustrated in Figure 3-8 were used to

f I‘ .‘IV E I‘ ‘1 Ir. ,l

93

1234

TG

Alc -

Figure 3-13: Complementation of the mutant Wow15 with pSER22acrl. Wild type
(lane 2), the mutant Wow15 (lane 3) and the mutant transformed with pSER2:acrl
(lane 4) were grown under nitrogen limited conditions. The neutral lipids were
extracted with chlorform1methanol, spotted onto the TLC and separated. The lipids
were visualized by spraying the plate with 50% sulfuric acid and charring the plate at
160C. Lane I contains 0.2 mg of wax esters (W), triacylglycerol (TG), and
hexadecanol (Alc) standards. The ori gin is shown at (O) and the solvent front is at
(SF).

94

A) File: 0071.81!!! 1 - 295
Table: K‘yte s Doolittle
Window: 21 Average: 0.14 Threshold Line: 0.00
5.00
4.0d
3.00
2.00
1.00
0.00
—l.00
-2
-3
-4
-5

 

.00
.00
.od
.0

 

I I I I I I I I
b

 

 

l 101 201 295

Table: Kyte & Doolittle

Window: 10 xveraqe: 0.14 Threshold Line: 0.00
5.00
4.00

3.00

2

l

B) File: acr1.ami 1 - 295

 

I I I I

 

0.
-l.
-2.
-3.

 

 

 

l 101 201 295

FigureS-M: KytemdDoolittleplotsoftheproteinsequencefi'omacrl. Awindow
sizeonI—oacidaflielengdlofmalphahelixneededmspanadonblephospho-
lipidmembumwasusedinpanelA. 'l‘hewindowsizewasredncedtolOamino
acidsinpanelederdIemumpfionthattheproteinmaybelocafizedtodlem
estu'inclusionbodies. Waxeswrinclusionbodieshavebeenobservedtocontainonly
asinglephospholipidmembrane. KyteandDoolittleplotsweregenentedusingthe
sofiwarepachgeDNASISforWindowsa-IitachiSoflwm).

95

remove the native promoter of the gene. The PCR product was subcloned
directionally into the BamHI, EcoRI sites of pET21 to give the pET21zacrl construct.
In this expression system the gene is under the transcriptional control of a T7lac
promoter which can be induced by the addition of IPTG. Based on the sequence
information for the wow15 gene, it was predicted that the size of the induced protein
should be approximately 32 kDa. Following induction of the system with IPTG, a
protein of the predicted size was observed primarily in the insoluble protein fraction
(Figure 3-15) implying the protein is being localized to the membrane, or is in the
form of inclusion bodies. Assays for acyl-CoA, acyl-ACP and palmitic acid reductase
activity were run on all the protein fractions by incubating the extracts in the presence
of 1‘C labelled substrate. The greatest amount of enzymatic activity was associated
with the insoluble fraction where the induced protein was observed to be localized
(Figure 3-16). Additionally, enzymatic activity was observed in the presence of acyl-
CoA, but not acyl-ACP, implying the enzyme is specific for acyl-CoA (Figure 3-17).
Experiments aimed at determining the cofactor specificity of the enzyme showed that
catalytic activity was observed when the enzyme was incubated in the presence of
NADPH but not NADH (Figure 3-18). “C labelled fatty alcohol was also observed to
be associated with protein extracts derived from E. coli transformed with pET21:acrl.
This implies that the enzyme is not only able to catalyze the reduction of acyl-CoA to
the corresponding aldehyde, but it is also able to reduce the aldehyde to the
corresponding alcohol. To verify this result, a time course was run using the same
enzymatic conditions, but removing a fraction of the reaction every 2.5 minutes (and

every 30 seconds in a separate experiment). The results of that experiment are shown

43-

30-‘ ‘

20.1-i.
I I

Figure 3-15: SDS-PAGE gel showing protein induction. Protein fractions from

E. coli containing pET21 was compared to protein fractions prepared from E. coli
transformed with pET21:acr1, following induction with 0.1 M IPTG. Lanes 1

and 2 show soluble fraction 1 from E coli transformed with pET21 and

pET2 l :acrl , respectively. Lanes 3 and 4 show the insoluble protein fraction

from induced E. coli samples transformed with pET21 and pET21:acr1,
respectively. Lanes were loaded with 1 ug of protein. The gel was electophoresed
and silver stained as per materials and methods. Size standards are shown to the
left of the gel and are in kilodaltons.

l 2 3 4
94- I
67-

97

1234

FA H.
”My:
6

Figure 3-16: In vitro acyl-CoA reductase assay. Soluble (lane 1) and
insoluble (lane 2) protein fractions from E. coli containing pET2]

were compared to soluble (lane 3) and insoluble (lane 4) protein

preparations from E. coli transformed with pET212acr1. Reactions contained
13.5 ug of pretein and were incubated in the presence of 0.04 uCi of
palmitoyl-l-“C-coenzyme A (30 uM) at 30°C for 15 minutes. Neutral

lipids were separated by TLC and radiolabelled products were visualized by
exposing the TLC plate to a phosphorimaging cassette. Fatty aldehyde (FA)
and fatty alcohol (Alc).

Acyl-CoA Acyl-ACP Free FA
ABCDEABCDEABCDE 123

 

 

 

FA .

we..-

%
l

m

M:

a at:

ll:

- lull
.
9

O

i
i

Pigle3-l7: Wofnhmuuymmwvhy
ofaerlwithdifiumtndiolsbelledmbsuutes. Sohibleaudimoluble
poteinexuaetsfi'omE. eolistrainDE3CBL21)trInlinlmedwith
pEI'leacrlineomp-isoutoptoteinfiactiomfiummunusfionued
Ecolisuaianaiml). Wownfiugofmeindieded
poteineflraetSOuMpahuitoyl—l-“C—eomzymeA.0.83mMATP,100uM
NADHmleOuMNADPHmincubatedaflO‘CfirlSmim.
Namallipidswaeeolleetedbyehklofomwumuedouand
spottedontoTLCplates. Radiolsbelledlipidswuevisualizedbyupos-
ingflnflflpletetoaphesphosimsgiugmmleproteinm
1(A),solubleaxtnetfnetion2(B)audflleiusolubleennet(C)wue
flumE. eolistniuDE3(BL21)umsfouuedwilhpE'l‘21:aa-l. Soluble
poteinahual(D)mdflIeinsohibleennet(E)wuefiumuuu-nsfomed
Ecol! strinDE3CBL21). Pahnitoyl—l-“C-eomzymeA(30uM),pahnitoyl-l-
”C-ACP(10000dpmeompcedto70000dpmfirpslmitoyl—CoA)ndl-"C-
pahniticacid(65uM)waeusedssmbsuutesintheindieetedseu. '
ouehalfofthevolmnemedforeaehresctiouwssspottedinlnesl,2nd3,
respectively. thyaldehydeGA)ndfityaleohol(Ale).

 

 

 

 

 

 

123456

FA u.-

-—~-—

Pigtail-18: Cohetnrdepudmeeofscyl—CoAm. hmm
assayswueeuriedmnwithflleedifl'euuueom Theimolubleprotdn
fiaetiou(l3.5ugpuresetiou)wasineubetedinfl|epresmeeof30uM
pahnituyl—l-“C-eoauymeAfimlSmiuutestO‘C. Inadditiontotbepotein
Wfiumflwflumsfimuedwiflipmlwlwasflieaddifionofm
mamlL033mMATPaaue2). 100uMNADHaane3), lOOuM
NADPI-Iane4), lNuMNADthulMuMNADPHGmSndOfiSmM
AIRIOOuMNADH-flloouMNADPHMG). FutyaldellydewA)”
MalcoanAle).

100

in Figure 3-19. It was observed that fatty aldehyde accumulation can be detected as
early as 1.5 minutes into the assay, while fatty alcohol formation is not observed until
about 7.5 minutes into the assay.

The ability of the Acrl protein to convert fatty aldehyde to the alcohol was
tested directly. Radiolabelled aldehyde was recovered from the TLC plate by
scrapping off the silica, eluting with chloroform/methanol, then drying under nitrogen
gas. The resulting radiolabelled fatty aldehydes were resuspended in 2% Triton-X100
(0.2% final concentration) for use as the substrate in an enzyme assay under the
conditions detailed above. The results of this experiment are shown in Figure 3-20
where the presence of fatty alcohols is associated solely with protein extracts prepared
from E. coli transformed with pET21zacrl.

Protein motifs. Analysis of the protein sequence using the program
MotifFinder (Ogiwara, unpublished), which searches the primary sequence for amino
acid motifs that are conserved among families of proteins, indicated the presence of a
region that is conserved among short chain alcohol dehydrogenase proteins. Figure 3-
21 shows predicted secondary structure using DNASIS for Windows (Hitachi
Software). Amino acids that were noted to be conserved by the family of short chain
alcohol dehydrogenases by Person et. al. are highlighted in the Figure 3-21 (Persson et
al., 1991). The conserved glycine residues at positions 22, 26 and 28 are consistent
with a nucleotide binding domain, in this case NADPH. These observations support
the idea that the isolated gene encodes a protein which is able to catalyze the

production of fatty alcohol from an acyl-CoA substrate.

101

30 sec. intervals 2.5 min. intervals
0 ——> 2.5 0 —-—-> 15.0
FA . i 531 .1

‘~'HU~H~ ”W

.“_H

-~*-mh—-u

0009000 eeoasie

Pigtail-19: hmacyl-CoAreduet-etimeeme. Atotalofl35ugof
illoluableproteiufi'omE. eolitrnsformedwilhpE'l‘leacrlwasincubatediu
memofiOuMpahnitoyl-l-“C-eomzymeAatM’Cfiorlsm
MMWeolleeteddwaeeoudMuptozSnumesuMnou
dielefisideofthefiglle. 'I‘hesameeouditiousmusediuueeoud
mmmmmmwuzsmmwms
Mummfiefiwsideoffllefim Fivemicmlitmoffllereeedou
mmedateechtimepoifimdtheneuuallipidsmisolmdby
chlorofiemmedmol(50:50)exuuetionndspottedomfieILCpUe.
Pommmerafiowmmfinfiudbym
bMpl‘etoaphesphorimsgiugm thyalddmlesCFAhudMy
aleohoh(Ale).

102

FA H

“In!

Alc

Figure 3-20: In vitro enzyme assay testing for reductase activity using

1-1 'C-palmitoyl aldehyde. Reactions contained 13.5 ug of protein and were
incubated in the presence 100 uM NADPH and the aldehyde for 15

minutes at 30”C. Samples were then chloroformzmethanol (50:50)

extracted and separated on a TLC plate before being visualized using a
phosphorimager. Lane 1 contains the radiolabelled aldehyde that was used

as a substrate, lane 2 represents the reaction run with proteins prepared

from E. coli transformed with pET21. Lane 3 is from the reaction

containing proteins from E. coli transformed with pET21zacrl. Fatty aldehyde
(FA) and fatty alcohol (Alc).

103

File: acr1.ami

Size: 295 aa
Seq: 1 - 295
Function: Chou and Fasman
10 20 3O 40 SO 60 70
fit tit * Q
VNKXLEALFRENVKGKVALITGASSGIGLTIAKRIAAAGAHVLLVARTQETLEEVKAAIEQQGGQASIFP
HELIX HHHHHHHHHHHh hhhhhH HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHh hH
SHEET sssss sSSSSs sSSSsssss sssssssssssssss 333 388833
TURN TTTTT TTTTTT TTTTT TT
COIL
80 90 100 110 120 130 140

t t f it. *

CDLTDMNAIDQLSQQIMASVDHVDFLINNAGRSIRRAVHESFDRFHDFERTMQLNYFGAVRLVLNLLPHH
HELIX HHHHHHHHHHHHHHHHHHhhhhH

SHEET ssSSss 538353333833 88885 SSSSSSSssssSSSSSSSSSSSS
TURN TT TTTTTT TTTT
COIL C C
150 160 170 180 190 200 210
it it iii 0'

IKRKNGQIINISSIGVLANATRFSAYVISEAALDAFSRCLSAEVLKHKISITSIYMPLVRTPMIAPTKIY
HELIX HH hhHHHHHHhhhhHHHHHHHHhhhhhHHHHHHhhhhhhH

SHEET 8 83833888888 88858 $88888 SSSSSSSSSSSSSSSSSSSBSBBSS
TURN TTTTT TTTT
COIL

220 230 240 250 260 270 280

KYVPTLSPEEAADLIVYAIVKRPTRIATHLGRLASITYAIAPDINNILMSIGFNLFPSSTAALGEQEKLN
HELIX HhhhhHHHhhhhhhhhhhhhhhhhhhhhhhhhhhHh HHHHHHHHhh

SHEET SSSSSS SSSSSSSSSSSSSSSSSSSSSSSSSSS SBSSSSSSSSSE SS
TURN TTTT TTTT TTTTTT TTTTT
COIL
290

LLQRAYARLFPGEHW
HELIX hhhhhhhhhhh hh
SHEET SSSSSSSSSS
TURN TTTTTT
COIL

Figure 3-21: Secondary structure prediction for acrland conserved amino acid
residues specific to the family of short chain alcohol dehydrogenases. This figure
shows predicted or-helixes (H and h), B-sheets (B and b), turns (T and t) and coils (C
and c) in context to the protein sequence. Upper case letters represent a higher
probability of the indicated secondary structure, while lower case letters represent a
possibility of the structural element being present. This secondary prediction was
calculated using the Chou-Fasman algorithm with a beta-turn probability value of
7.5x10‘ using the DNASIS for Windows software package (Hitachi Software). Amino
acid residues that are highlighted with (*) were found to share homology and spacing
requirements observed by Persson et. al. (Persson et al., 1991) to be associated with
short-chain alcohol dehydrogenase family of enzymes. Amino acids in bold type are
the same as those found by Persson to be 100% conserved throughout the family.
Also of particular interest are glycine residues 22, 26 and 28 which make up the
nucleotide binding motif and are associated with a helix-turn-helix region of the
predicted secondary structure as described by Persson et. al. (Persson et al., 1991).

104

Enzyme mechanism. The time course experiment outlined above provided
useful information about the mechanism of the Acrl protein. The observation that
alcohols do not accumulate in parallel with aldehydes suggests several things. This
result might be interpreted to mean that there is a single active site which carries out
both the conversion of acyl-CoA to aldehyde and aldehyde to alcohol. The emergence
of alcohols later in the assay would suggest that acyl-CoA is a preferred substrate over
fatty aldehyde. Another possible explanation is that the enzyme possesses two active
sites, one for the conversion of acyl-CoA to aldehyde, and another which catalyzes the
conversion of the fatty aldehydes to alcohol. To address these questions, the time
course experiment was repeated in the presence of unlabelled cis-l l-hexadecenal. If
the single site on the enzyme is substrate specific, an overabundance of unlabelled
aldehyde should be able to out compete acyl-CoA for the reactive site, and the
formation of radiolabelled aldehydes should be inhibited. If there is a second reaction
site, then addition of an overabundance of aldehyde should have no affect on the
accumulation of radiolabelled aldehyde during the timecourse. The timecourse
experiment was repeated in saturating amounts of unlabelled cis-l l-hexadecenal. The
results of the experiment are shown in Figure 3-22. It can be seen that aldehyde
accumulation is not inhibited even in the presence of saturating amounts of unlabelled
aldehyde. As would be predicted, radiolabelled alcohols are absent. This suggests that
the enzyme might contain two active sites, one that converts acyl-CoA to aldehyde,
and a second site that catalyzes the conversion of aldehyde to alcohol. This finding is
supported by the phenotype of the Wow15 mutant during nutritional supplementation

experiments. Observation that the mutant is unable to synthesize waxes when grown

105

1234567

FA

aw
Ale

t 0H“

‘ fir“ h'vi.

Pigtail-22: hmacyl-CoAreduet-ealsy euiedomiutllepesmeeof
mlsbelledcls-ll-hendeeml. Amlof13.5ugofpoteinwuinairuediu
fiepsumeeof30uMpalmitoyl—l-“O-eomyrueAd30‘CfirlS

ruirultes. mmmamoaflnummnm
mane3)t7.5min.(lne4),lo.0min.(lne5).12.5min.(lane6)and15.0
moan ‘I‘henerlnllipirbmeolleetedbychlesofiounmeflmrl(50:50)
mmwmmmm Foflowhgmme
radiolabelledproductawuevisualiudbyexposimfln‘l‘lflplatetos
Wm thyalddrydesGA)udfltyaleohols(Ale).

106

in the presence of hexadecane and Tween-40, but accumulates wax esters when grown
with cis-l l-hexadecenal, is consistent with an enzymatic model involving two active

sites.

CONCLUSIONS

A cosmid genomic library was constructed and used to complement two class I,
wax‘ mutants, Wowl and Wow15. DNA from the ends of one of the cosmid's inserts
was labelled and used as a probe against digested DNA from the other cosmid. No
cross hybridization was evident, indicating that the insert DNAs in the cosmids
represented two different areas of the A. calcoaceticus genome and that the mutants
were affected in two different loci. This was further proven by the observation that
the two different cosmids had different restriction patterns.

In order to delineate the complementary region and subclone the gene of
interest, I mutagenized the complementary cosmids with a TnlO derived transposon,
mini-TnlOCm. Three insertions in 1A-3F led to an inability of that cosmid to
complement the mutant phenotype of Wowl. These insertions have not yet been
mapped on the cosmid. Eight insertions in 4A-55 eliminated the ability of the cosmid
to complement the Wow15 mutant. These mutagenized cosmids were used to delineate
the gene of interest by comparing their restriction patterns to that of the wild type
when it was digested with EcoRV. The insertions were found to span two different
EcoRV fragments, which were subcloned and sequenced.

An open reading frame of interest was identified based on its location on the

EcoRV fragments and its strong similarity to an open reading frame from

107

Mycobacterium tuberculosis, ORF 2, which is thought to encode a B-ketoacyl reductase
involved in mycolic acid biosynthesis. The open reading frame from A. calcoaceticus
was PCR amplified from the complementary cosmid, subcloned into pSER2, an E.
coli/A. calcoaceticus shuttle vector constructed for this purpose, and used to transform
Wow15. The transformed mutant was able to synthesize wax esters indicating that the
complementary gene, designated acrl, had been isolated.

Because of the strong sequence similarity of acrl to a B-ketoacyl reductase,
more chemical complementation experiments were carried out to determine if the
mutant was unable convert acyl-CoA to its corresponding aldehyde. Incubation of the
mutant, Wow15, in the presence of Tween-40, under wax inducing conditions, resulted
in no wax production. Growth of Wow15 in the presence of cis-l l-hexadecenal
resulted in the production of wax esters. These observations, together with the
previously noted observation that waxes were not synthesized by Wow15 when it was
cultured in the presence of hexadecane led to the conclusion that Wow15 was most
likely the result of a mutation in the gene encoding the enzyme for acyl-CoA
reductase.

To investigate the enzymatic properties of the cloned acrl gene, it was
transformed into an inducible E. coli transcriptional expression system. An induced
protein of the appropriate size was observed to be present in the insoluble fraction of
proteins prepared from these induced transforrnants. This enzyme catalyzed the
conversion of acyl-CoA to the corresponding alcohol, via an aldehyde intermediate.
The enzyme was substrate specific for acyl-CoA, and not acyl-ACP, and utilized

NADPH (but not NADH) as a cofactor. The primary sequence of the protein

108

contained recognizable motifs that match signatures of the short chain alcohol
dehydrogenase family.

I have named the gene encoding this enzyme acrl, for acyl-CoA reductase. It
represents a novel enzyme that has not been detailed previously in the scientific
literature.

Experiments are currently under way to ascertain the function of the gene
residing on the cosmid that was found to complement Wowl. Because of the
discovery that acrl is able to catalyze both of the reduction steps involved in wax
ester biosynthesis, it seems unlikely that the Wowl mutation is affected in either one
of these steps. This is in contradiction to the chemical complementation experiments
which characterized it as a reductase mutant. The other activity left in the pathway
involves the acyl-CoA fatty alcohol transferase. However, experiments in which the
Wowl mutant was fed hexadecanol indicated that it was still able to synthesize waxes,
implying this gene had not been disrupted. It is possible that this mutant has been
affected in a regulatory protein that is specific to controlling wax ester production and
not triacylglycerol production. Although this does not seem very likely, it is one
explanation that can be put forth with the evidence in hand. It will be interesting to

pursue this mutant firrther and determine what gene has been disrupted.

REFERENCES

Allen, L. N. and Hanson, R. S.. 1985. Construction of Broad-Host-Range Cosmid
Cloning Vectors: Identification of Genes Necessary for Growth of Methylobacterium
organophilum on Methanol. .I. Bact. 16l(3):955-962.

Altschul, S. F., Gish, W., Miller, W., Myers, E. W. and Lipman, D. J.. 1990. Basic
Local Alignment Search Tool. J. Mol. Biol. 215: 403-410.

109

Arrowsmith, T. J., Malpartida, F., Sherman, D. H., Birch, A., Hopwood, D. A. and
Robinson, J. A.. 1992. Characterization of actl-Homologous DNA Encoding

Polyketide Synthase Genes from the Monensin Producer Streptomyces cinnamonensis.
Mol. Gen. Genet. 234:254-264.

Bachmann, B. J. 1987. In Archer-icky coli and Salmonella whimurium, Cellular and
Molecular Biology. Eds. Neidhardt F. C. et. al., ASM.

Figurski, D. and Helinski, D. R.. 1979. Replication of an Origin-Containing

Derivative of Plasmid RK2 Dependent on a Plasmid Function in trans. Proc. Natl.
Acad. Sci, U.S.A. 76:1648-1652.

F ixter, L. M. and Sherwani, M. K.. Energy Reserves in Acinetobacter. In m
Biology of Acinetobacter: Taxonomy, Clinical lmmrtance. Molecular Biology,
Physiology, Industrial Relevance. Plenum Press, New York, New York. 1991. Eds.
Towner, K. J., Bergogne-Bérézin, E., and C. A. Fewson.

Juni, E. 1972. Interspecies Transformation of Acinetobacter: Genetic Evidence for a
Ubiquitous Genus. J. of Bact. 112:917-931.

Kleckner, N., Bender, J. and Gottesman, S.. 1991. Uses of Transposons with
Emphasis on Tn10. In Methods of Enzvmology: Bacterial Genetic Systems. Vol.
204. Miller, J. H. (ed.). Academic Press Inc., San Diego, California.

Kyte, J. and Doolittle, R. F .. 1982. A Simple Method for Displaying the Hydropathic
Character of a Protein. J. Mol. Biol. 157: 105-132.

Maniatis, T., Fritsch, E. F., and Sambrook, J.. 1982. In Molecular Cloning, A
Laboratog Manual. Cold Spring Harbor Laboratory Press, New York, New York.

Ogiwara, A. MotifFinder can be accessed via the World Wide Web at the following
address, which is current at the time of this manuscript:
http://www.genome.ad.jp/SIT/MOTIF.html. Unpublished.

Persson, B., Krook, M., and Jomvall H.. 1991. Characteristics of Short-Chain
Alcohol Dehydrogenases and Related Enzymes. Eur. J. Biochem. 200:537-543.
Raleigh, E. A., Lech, K. and Brent, R.. 1989. In Current Protocols in Molecular
Biology. Eds. Ausubel, F. M. et al. Publishing Associates and Wiley Interscience,
New York.

Scott, C. C. L. and F innerty, W. R.. 1976. Characterization of Intracytoplasmic
Hydrocarbon Inclusions from the Hydrocarbon-Oxidizing Acinetobacter Species H01-
N. J. Bact. 127(1):481-489.

Yuan, Y., Lee, R. E., Besra, G. S., Belisle, J. T. and Barry, C. E.. 1995.

110

Identification of a Gene Involved in the Biosynthesis of Cyclopropanated Mycolic
Acids in Mycobacterium tuberculosis. Proc. Natl. Acad. Sci, U.S.A. 92:6630-6634.

Zilsel, J., Ma, P. H. and Beatty, J. H.. 1992. Derivation of a Mathematical
Expression Useful for the Construction of Complete Genomic Libraries. Gene 120:89-
92.

CHAPTER 4

ISOLATION AND CHARACTERIZATION OF TRANSPOSON MUTANTS FROM
A. CALCOACETIC US STRAIN BD413

ABSTRACT

It has previously been observed that when A. calcoaceticus strain BD413 is
grown under nitrogen limited conditions it accumulates both wax esters and
triacylglycerol. In the previous chapter I described the isolation of mutants of A.
calcoaceticus that were deficient in triacylglycerol accumulation via chemical
mutagenesis. However, it was not possible to complement the mutant phenotypes with
the cosmid genomic library that had been constructed. Therefore, another round of
mutagenesis was performed on A. calcoaceticus strain BD413, using a transposon
derivative of Tn10 called mini-TnIOPtth. A total of four mutants were isolated, two
of these were class 11 (mg) mutants, and the other two were class III (wax‘tag')
mutants. One of these tag' mutants, II-C7, was selected for further study.

DNA flanking the transposon insertion was amplified by extended inverse PCR
(IPCR). The IPCR product was subcloned and used to identify cosmids from the
genomic library that shared homology to the flanking DNA regions. A BgIII fragment
from one of these cosmids (SA-F1) was observed to share homology to the flanking
sequences surrounding the transposon insertion. The BglII fragment from the cosmid

was subcloned and used as a probe against genomic DNA prepared from wild type

111

112

A. calcoaceticus and the mutant, ll-C7. A shift in molecular weight, appropriate to
the size of the expected transposon was observed. The BglII fragment was sequenced
and three open reading frames could be identified. The approximate location of the
insertion was mapped via extended PCR. From this information it was determined that
the transposon inserted into an open reading frame that shares considerable homology
to a gene called glnE that has been identified in many different organisms. Since gInE
has been implicated in the regulation of nitrogen, it is considered likely that the 11-C7
phenotype is not the result of a mutation in a triacylglycerol structural gene, but rather

is the result of a mutation to a gene involved in nitrogen regulation or response.

INTRODUCTION

During the course of selecting a strain of A. calcoaceticus to work with in
studying wax ester biosynthesis it was observed that A. calcoaceticus strain BD413
also accumulated triacylglycerol when grown under low nitrogen conditions (Chapter
2, this text). This represented an opportunity to not only screen for wax ester mutants,
but at the same time to screen for mutants that failed to accumulate triacylglycerol.

It was our hope to identify mutants that were affected in triacylglycerol
production, but were unaffected in wax ester accumulation. This would help eliminate
the possibility that mutants affected in nitrogen sensing and response would be
recovered. In recovering such mutants we hoped to identify mutants that had been
affected in structural genes for enzymes involved in triacylglycerol biosynthesis,
particularly the gene encoding the enzyme diacylglycerol acyl transferase (DGAT).

DGAT plays a very important part in triacylglycerol production. It is

113

responsible for the final conversion of diacylglycerol to triacylglycerol via the addition
of an acyl group to the sn3 position of diacylglycerol. It is believed that this might be
a key regulatory step in triacylglycerol production in higher organisms, primarily
plants, since it would regulate the flow of carbon between phospholipid production and
triacylglycerol accumulation.

A similar approach to that outlined in the previous chapter for the isolation and
characterization of the wax mutant, Wow15, was also carried out for chemically
induced mutants affected in triacylglycerol accumulation. Complementation using the
genomic library that was known to complement two class I mutants and a tryptophan
auxotrophic mutant, was used to try to complement Wow7, a class 11 (mg) mutant, but
was unsuccessful. This raised the possibility that the cosmid library was incomplete, or
that the wow7 phenotype was the result of a dominant mutation that could not be
complemented by the library. Therefore, a new approach involving mutagenizing A.
calcoaceticus with a transposon was tried in an attempt to generate new tag' mutants.

There are several benefits of using a transposon mutagenesis compared to
chemical mutagenesis. One reason is that by generating mutants with a transposon,
only a single mutational event occurs that is the result of the transposon inserting into
the gene of interest. By contrast, chemical mutagenesis can cause many mutational
events throughout the genome depending on the amount of mutagen used and the
duration of exposure. Another reason is that once the desired phenotype is isolated,
the transposons can be rescued from the genome of the mutant, usually together with
flanking genomic sequence. This can be done either by plasmid rescue, if the

transposon contains an origin of replication, or by inverse PCR. The flanking

114

sequence can then be used as a probe to identify the full length gene.

The reason transposon mutagenesis was not used initially, was simply because
there had not been a transposon available for use with A. calcoaceticus. That changed
when a Tn10 derivative, mini-TnlOPtth, was used to generate lipase mutants in A.
calcoaceticus strain RAG-1 (Leahy et al., 1993). Another reason not to use transposon
mutagenesis is simply the amount of work required to screen for the desired
phenotype. Because the mutagenesis typically produces one insertional event per
colony (i.e. one mutagenic event), it is necessary to screen more samples than in the
chemical mutagenesis to find the desired phenotype.

In this chapter I describe the mutants isolated by transposon mutagenesis of A.
calcoaceticus strain BD413. I will discuss the characterization of these mutants by
Southern analysis, and finally describe the isolation of flanking DNA surrounding the

insertion in the tag‘ mutant ll-C7 and what this DNA possibly encodes.

MATERIALS AND METHODS

Bacterial stains and plasmids. The bacterial strains used in the experiments
described in this chapter are shown in Table 4-1. The source of the plasmids used in
this chapter, or in the construction of novel plasmids is presented in Table 4-2.

Growth and culture conditions. Conditions have been previously described in
Chapter 2.

Transposon Mutagenesis. Fifty milliliter cultures of A. calcoaceticus strain

BD413 and the transposon donor E. coli strain SMIO containing the transposon mini-

115

Table 4-1: Bacterial strains used in Chapter 4.

 

Bacterial Strains

Relevant
Characteristics

Source or Reference

 

A. calcoaceticus

ATCC #33305, strain
BD413

Wow7

3-A9

6A-H5

9-C3

11-C7

1 1-C7:Rif’

ll-D12

30-F10

30-F10:Rif

wild type used during
these studies,
unencapsulated mutant

of A. calcoaceticus strain
BD4

tag' null mutant of
BD413

mini-TnIOPtth mutant
of strain BD413 that is
wax'tag' and found to
have multiple insertions

mini-TnIOPtth mutant
of strain BD413 that is
wax‘ and found not to
have an insertion

mini-TnlOPtth mutant
of strain BD413 that is
tag' and found to have
multiple insertions

mini-TnIOPtth mutant
of strain BD413 that is
tag‘

spontaneous Rif mutant
of 11-C7

mini-TnlOPtth mutant
of strain BD413 that is
tag' and found not to
have an insertion

mini-TnlOPtth mutant
of strain BD413 that is
tag'

spontaneous Rif mutant
of 30-F10

ATCC stock center

this study, chapter 2

this study

this study

this study

this study

this study

this study

this study

this study

Table 4-1 (cont'd)

30-G9

35-G5

E. coli
HB 101
DHSa

MM294
SMIO Apir

116

mini-TnIOPtth mutant
of strain BD413 that is
wax'tag‘

mini-TnlOPtth mutant
of strain BD413 that is

wax'tag‘

F’ proA2 recA13 mch

F'/endAl recAl
A(lacZYA-argF) U169
(¢80dlacA(lacZM15)

F- endAI hst17 thi-l

Kmr thi-l thr leu tonA
lacY supE recAzsz4-2-
Tc::Mu lpir

this study

this study

(Maniatis et al., 1982)
(Raleigh et al., 1989)

(Bachmann, 1987)

(Miller & Mekalanos,
1988)

 

117

Table 4-2: Plasmid sources and derivations for Chapter 4.

 

Plasmid

Description or
Construction

Source or Reference

 

sz-Kst, pBS-KS'
pRK2013

pLA2917

5A-F1

pLOFPtth

pGEM-T

pSR10

pSRl 1

Bluescript Vector

Km’ self-transmissible
RK2 derivative
containing ColEl
replicon and transfer
functions to mobilize
RK2 derivatives

Cosmid Vector (T et')
derived from RK2

pLA2917 derived cosmid
clone with homology to
flanking sequences
surrounding the insertion
found in the strain ll-C7

Plasmid vehicle carrying
mini-Tnl OPtth

T-vector for facilitating
the subcloning of PCR
fragments

pGEM-T derivative
containing flanking
sequence surrounding the
insertion site of the Tn
from the ll-C7 mutant

Bluescript derivative
containing the Bglll
fragment from the
cosmid SA-FI that
shared homology with
flanking sequence
surrounding the insertion
site of the Tn from the
11-C7 mutant. Gave
rise to sequence ID#IO-
ll

Stratagene

(F i gurski & Helinski,
1979)

(Allen & Hanson,
1985)

this study

(Herrero et al., 1990)

Promega

this study

this study, chapter 4 for
construction and
appendix A for
sequence information

118

Table 4-2 (cont'd)

pSR12 Bluescript derivative this study, chapter 4 for
containing a Bglll construction and
fragment from 5A-F1 appendix A for
that lies upstream of the sequence information
Bglll fragment contained
in pSRl 1. Gave rise to
sequence ID#12-l4

 

TnlOPtth on plasmid pLOFPtth were grown in LB at 30°C to an OD600 of 0.6.
The cells were collected by centrifugation, washed twice with LB medium, and
resuspended in the original volume. Fifiy milliliters of A. calcoaceticus culture was
then mixed with 25 m1 of the transposon donor strain and collected on 0.45 in sterile
85 mm cellulose acetate filter producing an even lawn of bacteria. The resulting filter
was transferred to an LB plate containing 500 uM IPTG to induce the mobilization of
the transposon. The diparental mating was incubated at 30°C for 8 hours. The
bacteria were then removed from the filter by washing the filter in 10 m1 of 10 mM
MgSO,. The sample was concentrated by centrifugation and resuspended in 10 mM
MgSO,. Transconjugates containing the transposon were selected by plating the cells
on LB plates containing rifampicin to select against the presence of E. coli, plus
kanamycin to select for the presence of the transposon in A. calcoaceticus.

Mutant and complementation screening using the lipophilic dye Sudan
black B. Methods have been previously described in Chapter 3.

Thin layer chromatography (TLC). Method has been previously described in
Chapter 2.

Library Construction. Library construction was described in Chapter 3.

119

Triparental filter matings. Matings were carried out in the manner previously
described in Chapter 3. The cosmid library was contained in E. coli strain HB101 and
was Tetr due to the presence of the cosmid. MM294 is a strain containing pRK2013, a
helper plasmid that enables triparental mating through its mob genes. The pRK2013
plasmid imparts Km‘. Finally, the mutant strain 11-C7:Rif is the target of the mating
and is a Rif strain of ll-C7. It was obtained by plating ll-C7 on rifampicin plates
and looking for the presence of spontaneous Rif colonies. The neutral lipid profiles of
these Rif‘ strains were examined to make sure they matched those of the original 11-
C7 mutant. At the end of the mating, ll-C7:Rif mutants containing the cosmids can
be selected by plating the product of the cross onto LB medium containing rifampicin,
to select for 11-C7:Rif, and tetracycline to select for the presence of the cosmid.
MM294 and the HB101 donor will fail to grow because they are not resistant. The
ll-C7 strain and MM294 were grown overnight as 3 m1 cultures, and the library was
grown in 96 well titer plates as replicates of the original. The day of the mating, 0.5
ml of MM294 was used to inoculate a 50 m1 of culture, and 3 ml of 11-C7 was used
to inoculate a 50 ml culture. Cultures were collected at an ODwo=0.6, washed and
then resuspended in 50 m1 of LB. Twenty five milliliters of MM294 was combined
with 50 m1 of the ll-C7 strain. Ten milliliters of this mixture was drawn through a
sterilized 45}: filter (85 mm in diameter) via a vacuum apparatus creating an even lawn
of bacteria. The filter was then removed and placed on to an LB plate. Using a
sterilized set of prongs arranged to match 48 wells of the 96 well titer plates
containing the library, the library was stamped onto the lawn of bacteria and matings

were allowed to incubate at 30°C overnight. Filters were then transferred to selective

120

medium containing rifampicin (100 ug/ml) and tetracycline (15 ug/ml) to select for 11-
C7 :Rif containing the cosmids. The resulting patches were then restreaked onto a
master plate containing the selective medium, before being replica plated onto
minimal, low nitrogen, triacylglycerol inducing medium for subsequent analysis by
lipophilic staining and TLC.

Inverse PCR. In the hope of creating intramolecular recircularization events,
genomic DNA from the class II mutants, 11-C7 and 30-F10 were digested to
completion with Bglll and Xbal, respectively. Fragments were size selected by
separating the digested products on a 0.8% agarose gel, cutting out an agarose block
containing the appropriately sized fragments, and purifying the DNA fragments away
from the agarose by electroelution (Maniatis et al., 1982). The DNA fragments were
recircularized by diluted them to a concentration of 0.6 ng/ul in a 200 pl ligation
volume containing 3 units of T4 DNA ligase (Boehringer Mannheim) and 20 ul of 10x
ligation buffer (660 mM Tris-HCl (pH 7.6), 50 mM MgC12, 10 mM dithioerythritol
and 10 mM ATP (pH 7.5)). Three microliters of this ligation reaction was used with
Boehringer Manneheim's Expand Long Template PCR System. Reactions contained
350 uM of dNTP's, 300 nM of each primer, 5 ul of 10x buffer 1 (17.5 mM MgC12,
500 mM Tris-HCI, pH 9.2, 160 mM (NH,)ZSO,), 3.0 ul of the ligation reaction
described above was used as template DNA and 0.75 [.11 of the supplied enzyme mix
in a total volume of 50 ul. Cycling was carried out in the following manner:

Step 1: 94°C for 2 minutes
Step 2: 94°C for 10 seconds

Step 3: 65°C for 30 seconds

121
Step 4: 68°C for 7 minutes

Step 5: repeat from step 2 10 more times

Step 6: 94°C for 10 seconds

Step 7: 65°C for 30 seconds

Step 8: 68°C for 7 minutes + 20 seconds each cycle

Step 9: repeat from step 6 15 more times

Step 10: 68°C for 7 minutes

Step 11: Hold at 4°C

Extended PCR. Mapping the location of the Bglll fragment that was found to

share homology to flanking sequences surrounding the transposon insertion in 11-C7 in
respect to the cosmid was possible using Boehringer Mannheim's extended PCR.
Primers were constructed on either end of the Bglll site of pLA29l7 facing in toward
the insert. Additionally, primers specific to both ends of the Bglll fragment were
constructed to face out, toward the edge of the insert DNA. Using one anchor on the
cosmid in conjunction with the primers specific to the Bglll sequence (Table 4-3), it
was possible to amplify the DNA in between the primers allowing me to assign a
distance from the end of the fragment relative to the anchoring (cosmid specific)
primer (Figure 4-1). Long PCR products were obtainable using Boehringer
Mannheim's Expand Long Template PCR System. Reactions contained 350 uM of
dNTP's, 300 nM of each primer, 5 pl of 10x buffer 1 (17.5 mM MgC12, 500 mM Tris-
HCl, pH 9.2, 160 mM (NH4)ZSO4), 0.15 pg of template DNA and 0.75 ul of the
supplied enzyme mix in a total volume of 50 ul. Cycling was carried out in the

following manner:

122
Step 1: 94°C for 2 minutes

Step 2: 94°C for 10 seconds

Step 3: 65°C for 30 seconds

Step 4: 68°C for 7 minutes

Step 5: repeat from step 2 10 more times

Step 6: 94°C for 10 seconds

Step 7: 65°C for 30 seconds

Step 8: 68°C for 7 minutes + 20 seconds each cycle
Step 9: repeat from step 6 15 more times

Step 10: 68°C for 7 minutes

Step 11: Hold at 4°C

Colony lifts for hybridization. Method was described in Chapter 3.
Restriction digestions and southern blot analysis. Methods were previously
described in Chapter 3.

Nested deletions and DNA sequencing. Methods were previously described
in Chapter 3.

Database analysis of DNA sequences. Methods were previously described in

Chapter 3.

123

Table 4-3: List of synthetic oligonucleotides used in this Chapter 4.

 

Name of primer

Sequence (5'-3')

Description

 

P9

P10

P11

P12

P13

CTTTCTTGCCGCCA
AGGATCTGATG

GGCCGGAGAACCT
GCGTGCAAT

TCGCGGCCTCGAGC
AAGACGT

CTGCCTCGGTGAGT
l l ICTCC

GATTCGGCCACCGC
TTCCAAA

Primer used in mapping
the location of Bglll
fragment containing
glnE in 4A-55. Specific
to 5' side of Bglll site of
pLA2917 in the Km
cassette which originated
from Tn-S. Faces
toward Bglll site

(Figure 4-1).

Same as 5' Km Tn-S but
specific to the 3' side of
the Bglll site (Figure 4-
1).

Primer used for inverse
PCR experiments and for
mapping the location of
Tn with respect to Bglll
fragment from 5A-F1.
Primer is specific to 5'
end of Km cassette in
mini-TnIOPtth. Km
cassette is from Tn903.
Primer faces out (Figure
4-1).

Same as SER-S', except
primer is located firrther
into Km cassette (Figure
4-1).

Used for the same
purposes and in
conjunction with SER-S'
and SER-3'. Specific to
5' region of the Ptt
cassette of mini-
TnlOPtth

Table 4-3 (cont'd)

P14

P15

P16

P17

124

CTTTTGCGGTGTGG
TGTATAA

ACACCGCAAAAGA
GAAAGGTT

TTCGGCTCGTTCAC
GTAGATA

AGCCGAAAGCCAC
CGTTTAGC

Primer designed with
specificity to end of
Sequence ID#IO. Primer
faces out toward Bglll
site of pLA29l7 (Figure
4-1).

Primer designed with
specificity to end of
Sequence ID#10. Primer
faces in toward mini-
TnlOPtth (Figure 4-1).

Primer designed with
specificity to end of
Sequence ID#l 1. Primer
faces out toward Bglll
site of pLA29l7 (Figure
4.1).

Primer designed with
specificity to end of
Sequence ID#l 1. Primer
faces in toward mini-
TnlOPtth (Figure 4-1).

 

125

A) E H ES Sm

— P9 914 m p10

 

Bgm mm

 

 

 

from SA—Fl
B)
500 bp
B glII — B glII B glII
P15 P12 P11 P17
-> 4- -> 1 <—
s— Km -— ~s— Ptt
3 2 l
<— e—— 4—

Figure 4-1: Positional mapping of the B glII fragment in the cosnrid SA-Fl. The
location of the Bng fragment cloned from cosmid 5A-F1, in relation to 5A-Fl '8
insert DNA, was determined by extended PCR using primers specific to the
cosmids Km' cassette (P9 and P10) in conjunction with primers specific to
identified sequence from the B glII fragment (P14 and P16) (Panel A). The location
of the transposon insertion relative to the Bng fragment was determined by PCR
mapping using genomic DNA as a template and primers specific to the

transposon, mini-TnlOPtth (P11 and P12) in combination with primers specific
to determined sequence from the ends of the B gIII fragment (P15 and P17)

(Panel B). By testing the different combinations of primers it was possible to
finally come to the orientation shown above and determine that the transposon had
inserted into the open reading frame encoding glnE. The approximate positions of
the open reading frames present on the Bglll fragment illustrated in panel B are
represented by arrows below the fragment showing direction of transcription.
These are glutarnate-ammonia-ligase adenylyltransferase from E. coli (1), putative
branched-chain-amino-acid transaminase from E. coli (2) and an unidentified open
reading frame (3).

1 26
RESULTS AND DISCUSSION

Transposon mutagenesis. To measure the frequency of transposition, it was
determined that the number of recipient cells at the time of infection with the
transposon measured 1.3x107 cfu/ml. The number of resulting colonies after selection
for the presence of the transposon in the recipient, A. calcoaceticus strain BD413:Rif
was observed to be 3000 cfu/ml. By dividing the number of chloramphenicol resistant
transforrnants by the number of recipients, the frequency of transposition was estimated
to be 2.3x10'5. It was noted that when the chloramphenicol resistant lines resulting
from the mutagenesis were plated out onto selective medium, a variety of colony
morphologies were evident. This was one indication of the success of the
mutagenesis. Additionally, when the transposon mutagenized colonies were transferred
to low nitrogen, minimal medium plates for staining with Sudan black B, several
colonies were unable to grow. This suggested the presence of auxotrOphic mutants.
To confirm if transposition had occurred and whether it was random, nine colonies
were selected and genomic DNA was prepared. The genomic DNA from these
suspected transposon mutagenized lines was digested with HindIII, an enzyme known
to cut internal to the transposon, and separated on a 0.8% agarose gel. The samples
were then probed with the Mlul fragment from the Km' marker from the transposon
(Figure 4-2, panel A). The Mlul fragment hybridized to two bands in each sample as
predicted due to the presence of a HindlII site in the transposon (Figure 4—2, panel B).
It was also observed that the fragments were all of different sizes implying that each
colony carried an insertion at a different site.

Mutant screening. A total of over 3000 transposon-mutagenized colonies

127

A)

15103 Km P" I kb

ISIOR probe Ptt Km 10mm

 

Figure 4-2: Southern blot analysis of 9 random colonies following mutagenesis with
mini-TnIOPtth transposon. Genomic DNA was digested with HindIII and probed
with the cassette illustrated in panel A. (A) An abbreviated restriction map of
pLOFPtth, the plasmid carrying mini-TnIOPtth the transposon used in the
mutagenesis. The region between the two Mlul sites was used as a probe against the
blot illustrated in panel B (B, Bglll, E, EcoRI, Hp, Hpal, M, Mlul and Xh, Xhol).
The two small rectangles in panel A represent the inverted repeats of the transposon.
Note that there is a HindIII site internal to the transposon, thus giving rise to the
presence of two bands on the blot.

128

were directly screened by TLC analysis of their neutral lipid composition. A total of 8
mutants were recovered. These 8 mutants could be divided into the 3 different
phenotypic classes that were similarly observed in the chemically mutagenized
samples. One class 1 mutant (wax‘), four class II mutants (tag‘) and three class III
mutants (wax'tag') were recovered. Genomic DNA from the eight mutants was
prepared and then digested with PstI, which is known not to cut the transposon. The
DNA was then Southern blotted and probed with the transposon (Figure 4—3). The
transposon was absent in 2 of the 8 mutants, indicating that their phenotype's were
either the result of a spontaneous mutation, or more likely, that the transposon excised
itself from the genome. Two of the mutants were observed to contain multiple bands
on the Southern blot indicating multiple insertions. This left a total of four mutants
with single insertions. Although the mutants containing multiple insertions could be
analyzed, I decided to simplify the matter and investigate the mutants harboring single
insertions. Two of these were class II mutants and the other two were class III
mutants. The class III mutants were dismissed as nitrogen regulatory mutants and only
the class II mutants were pursued for further analysis.

It is important to note that when the two class II mutants, ll-C7 and 30-F10,
were cultured under low nitrogen conditions, little to no growth occurred.
Identification of the mutant phenotype was possible by scrapping colonies off of an LB
plate and then culturing the cell paste in low nitrogen medium. In this manner, it was
possible to identify them as class II mutants in a reproducible manner.

Inverse PCR. Genomic DNA from the two class II mutants, Il-C7 and 30-

F10, was digested with several different restriction enzymes, the fragments were

129

23 kb —

9.4 kb —

6.5 kb —

4.3 kb —

 

Figure 4-3: Southern analysis of isolated mutants carrying mini-TnlOPtth.
Genomic DNA from the isolated mutants was digested with Pstl and probed

with the MINI fragment of the transposon (Figure 4-2, panel A).

There are no Pstl sites internal to the transposon as witnessed in lane 10 which
contains pLOFPtth digested with Pstl. A. calcoaceticus strain BD413 wild

type DNA (lane 1), 3-A9 (lane2), 6A-H5 (lane 3), 9-C3 (lane 4), l l-C 7 (lane 5),
ll-D12 (lane 6), 30-FlO (lane 7), 30-G9 (lane 8), 35-GS (lane 9). The identification
of multiple bands in lanes 2 and 4 indicates the presence of multiple insertions in
these strains, while the absence of bands in lanes 3 and 6 can be interpreted to mean
that these mutants have lost the transposable element.

 

130

separated by electrophoresis through an agarose gel, and a Southern blot was probed
with the transposon (Figure 4-4). The purpose of this experiment was to identify a
fragment of convenient size that could be size-selected and recircularized for inverse
PCR (IPCR). Digestion of 30-F10 with Xbal and ll-C7 with Bglll, respectively,
produced the desired results. A restriction fragment of approximately 6.5 kb was
observed for 30-F10 when it was digested with Xbal and a fragment of about 4.0 kb
was seen for ll-C7 when it was cut with Bglll.

Genomic DNA was digested with the appropriate enzyme, and fragments of
approximately 6.5 kb in the case of 30-F10 and 4.0 kb for 11-C7 were diluted to a
concentration of 0.6 ng/ul and then recircularized for IPCR. By taking advantage of
extended PCR using Boehringer Mannheim's Expand Long Template PCR polymerase
and buffer mixture, it was possible to amplify a single band of the appropriate size
from each of the samples (Figure 4-5). The resulting 3.5 kb band amplified from the
11-C7 sample was successfully subcloned into Promega's pGEM-T vector producing
pSRIO. However, problems arose in trying to subclone 30-F 10's IPCR product,
probably because of it's larger size (approximately 6 kb). Therefore, because of the
ease in subcloning the fragment from ll-C7 and in the interest of proceeding forward,
further analysis was focused on ll-C7. To verify that the subcloned IPCR fragment in
pSRIO was the correct product, genomic DNA from the wild type A. calcoaceticus and
the tag' mutant ll-C7 was cut with BglII. The DNA was Southern blotted and probed
with pSRIO. A shift in molecular weight of the predicted size was observed in the
mutant indicating the presence of the transposon and verifying that the subcloned IPCR

fragment in pSRIO was the right product (Figure 46).

131

12 3456 7 8910111213

 

23kb—

9.4kb-
6.51:1)—

43kb—

23kb-

FigIIe4-4: Badman-lysis ofgnemicDNAfiom 30-F10sndll-C7. Bothtaa'
mmumsgmanrxmamzymmstmauo
IndMflnS). MUHAfiumll-C‘lwuabowwiflrtbme
mmmnmmnmmnmmnmm
(hem). Theblotwasllohedwidr MW mommi-
TnlWflKmM4-ZA). Imelleunaimlkbladdcsiumndlne
lZealaiusLsrndsDNAeuwiflrM Pinally,lnel3kaposlliveoorllol

mmwmmu

132

 

FigureA—S: ExteudedIPCRofll-C7and30-F10. ExtmdeleCRreaetionsm
runonreeircularizedDNAsamplesfiom ll-C7(he2)nd30-F10(lane3).Lne
leoutaimOWugoflkbhddumdlm4shawsO4ugofLamda/M
mmmmwhmzmamuwmm
insubsequutckning.

133

6.5 kb '-
4.3 kb —

 

2.3 kb -

Figure 4-6: Southern anlaysis of l l-C7 with pSR10. Southern
blot analysis of A. calcoaceticus wild type genomic DNA (lane
1) and the tag' mutant, ll-C7 genomic DNA (lane 2) cut with
Bglll. The blot was probed with pSRIO.

134

Library screening. The subcloned IPCR fragment was labelled with
digoxigenin-I l-UTP and used to probe filter replicas of the cosmid genomic library.
In the first round of screening, colorimetric detection was used via nitro-blue-
tetrazolium to identify cosmids that shared homology with the probe. This resulted in
the identification of a single cosmid (SA-F 1) that shared homology to the IPCR
fragment. Subsequent screening, taking advantage of luminescent detection, has
allowed the identification of 10 other cosmids that share homology to the subcloned
fragment (1B-B6, 2B-F3, 2B-G4, 2B-G5, 4A-E4, 4A-F4, 4A-F5, 4A-GZ, 4A-GS and
9B-BS). Attempts to complement the mutation with the first cosmid (SA-F 1) that was
found was unsuccessful. Cosmid DNA preparations of the mutant, ll-C7, transformed
with the cosmid indicated that the cosmid is present in the mutant. Examination of all
the exconjugates transformed with the other identified cosmids has been hampered by
the poor growth response of the exconjugates in low nitrogen medium. The
observation that the exconjugates fail to grow under low nitrogen conditions, could be
interpreted to indicate that the cosmids do not complement the mutant phenotype.
Another possibility is that there is a second site mutation, the result of a spontaneous
event, that is contributing to the mutants poor growth response under low nitrogen
conditions.

Based on these assumptions I decided to sequence the region disrupted by the
transposon. To do this, the cosmid that was found to share homology with pSR10 was
digested with several restriction enzymes and the products were separated on an
agarose gel to make a Southern blot. This blot was then probed with pSRIO to find a

fragment of convenient size for sequencing. As can be seen in Figure 4-7, digestion

135

12345678910

23kb-

9.4kb -
6.5kb III

43kb-

2.3kb '-

 

Figure4-7: WMMSA-Fl oosmidDNA. Theoosruid. 5A-F1,
whichhornologytotlreIPCRproarafiumthetlg‘mm, 11-C7wu
wmmmmmwumm
pSRlo. BanHI(lauel),BgllI(he2).aal(l-re3).leolu(lane4).3eoRV
m%)M(lme6).mane7).Salla-re8),mm9)ndml

136
of SA-Fl with BglII yielded such a fragment. This fragment was subcloned into

Bluescript producing pSRl 1. To ensure that the proper fragment had been subcloned,
pSRll was used as a probe against Bglll digested wild type, mutant (1 l-C7) and
cosmid (SA-F1) DNA. The results of the experiment are shown in Figure 4-8. Based
on the shift in molecular weight observed in the mutant lane, it was concluded that the
proper fragment had been subcloned. At this same time another band of approximately
the same size was also subcloned in bluescript to give pSR12. This fragment was
originally subcloned because it was believed to be the Bglll fragment of interest, but
was later discovered not to be by Southern analysis. However, upon sequencing this
fragment (see below) it was determined that this fragment resided next to the Bglll
fiagment subcloned in pSRl 1. It was possible to determine this because they both
shared homology to the same gene identified in GenBank release 92.0 (glutamate-
ammonia-ligase adenylyltransferase from E. coli, accession #P30870).

Exonuclease digestions of pSRll and pSR12 were carried out to create nested
deletions. The deletion series was then sequenced allowing for the identification of 3
open reading frames on the DNA (Figure 4-9). One of these shared no homology to
anything in the GenBank database. The other two showed significant homology to
proteins involved in nitrogen regulation and amino acid biosynthesis. One of these
was homologous to E. coli glnE, the gene which encodes glutamate-ammonia-ligase
adenylyltransferase (BLAST score of 317 with a probability score of 1.6x10"”). The
other open reading frame matched a branched-chain-amino-acid transaminase from E.

coli (BLAST score of 448 with a probability score of 7.1x10’69)

137

  

2.3H3—I

Figure4-8: Smnhunmalys'uofll-C7wiflrpSRll. Souflunblotanalysis
ofguromicDNAplepuedfiumwfldtypeAcaleoaeedauflanel),m'
mutant, ll-C7(lane2)andtheeosrnidclone5A-Fl(lane3). DNA“
probedwithpSRll,acloneeanyirrgtlreBgllIfiamfiom5A-Fl.

138

Figure 4-9: Map showing sequence ID#10-l4 in respect to one another. The boxed
regions highlight regions of homology that were detected between the sequence and
GenBank release 92.0 by BLASTX analysis (Altschul et al., 1990). These data are
summarized in the table below the map. The numbers surrounding the boxes indicate
the start of the similarity between the DNA query sequence (the numbers above the
line) and the amino acid numbers in the matching protein from GenBank (the numbers
below the line). This is to give some sense of the encoded protein over the whole
restriction fragment, and also helps to determine the amount of DNA sequence missing
between gaps. The gene, glnE, spans sequence ID#I 1-14. Sequence ID#IO-ll were
from pSRll and Sequence ID#12-l4 are from pSR12. The table summarizes the
GenBank information giving the length of the matching protein, accession number of
the matching protein, the BLASTX score, probability score and the name of the
protein as reported by GenBank. Cases where the description of the first match is
ambiguous, second and third matches are included. Contig maps and actual DNA
sequences can be found in Appendix A of this text.

139

 

3. 05mm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 

 

  

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a8 .m Eo:
mvm $95.4” mum $339.9: _ coon omm =-mEoEEm-9mEmS_0 050mm mI 3.1.
35.6225 mEE 08%: So:
:3 5.0 cm 0.0.9.229: _ coca omm __-m_coEEm-9mEmS_0 mpvevm NI or
:8 .m :5:
mvm $2me to 39295:. _ coca own =-mEoEEm-9mEmS_G 030mm F: NF
:00 .m E0:
m3 2.26m cm: 282 v: c
.130 .m So:
8m $0 Q P s 94 «massages 28-8_Em-c_£o-uococsm 88mm 0 2
IlawwwcaioeavllT Boom Boom xeamoeou n .320 mm .3532 o>on< no.
- sealants. 31 a See on can: .a 252 agfitfi
11553.. 5205 3:822 lcgmwmi
Sign: flee 3%: ER: AER:
a an 5 an an E .3 a; a. ma
illEIv E 191' E o1.
E 2.: IV _ a. IlV an $2
a an as a

3%

140

Mapping the location of the transposon relative to the Bglll fragment. To
determine why the cosmid (SA-F1) is unable to complement the mutant phenotype,
and to determine which of the open reading fiarnes was directly affected by the
insertion of the transposon, I mapped the location of the insertion in the genome and
extrapolated its position in respect to the cosmid. One simple explanation for why the
cosmid does not complement the ll-C7 phenotype would be if the gene was at the
extreme end of the cosmid's insert DNA. Another possibility would be if the
complementary gene resided in an operon and the operon was at an the extreme end of
the cosmid's insert so that the complementary gene was not, in fact, on the cosmid. If
either of these situations were the case, then it might be possible that the rest of the
coding sequence of the gene, or the operon, is not present on the insert DNA of
cosmid SA-Fl. Thus, the reason that it can not complement 11-C7 is because the
entire coding region of the gene or the operon simply is not present. To determine if
this was the case, extended PCR was carried out as outlined below to map the location
of the Bglll fragment in the cosmid, and at the same time, determine the location of
the transposon relative to the Bglll fragment.

To map the location of the sequenced region relative to the end of the insert
DNA of the cosmid, primers that faced in toward the cosmid's insert DNA (P9 and
P10, Table 4-3) were used in conjunction with primers that were specific to the ends
of the Bglll fragment that faced out (P14 and P16, Table 4-3) (Figure 4-1, Panel A).
Using extended PCR with the cosmid, SA-Fl as a template, it was possible to amplify
a fragment of approximately 6.0 kb when using a primer specific for the transposon in

conjunction with an anchoring primer specific for the cosmid. This maps one end of

141

the Bglll fragment that was sequenced approximately 6.0 kb from the end of the
cosmids insert DNA.

Mapping the location of the transposon relative to the Bglll fragment to
determine which open reading frame it inserted into involved using primers specific to
the transposon that faced out (PII and P12, Table 4-3), in combination with primers
with specificity to the ends of the sequence obtained from the Bglll fragment (P15 and
P17, Table 4-3) (Figure 4-1, Panel B). Using genomic DNA prepared from the
mutant, 11-C7, it was determined that the transposon had inserted into the third open
reading frame. This open reading frame shares considerable homology to glnE,‘
glutamate ammonium ligase adenylyltransferase, that has been identified from many

different organisms.

CONCLUSION

Transposon mutagenesis based on a Tn10 derivative was used to identify
mutants in triacylglycerol biosynthesis in A. calcoaceticus. This second attempt at
isolating triacylglycerol deficient (tag') mutants was performed following attempts to
complement tag' mutants generated by NTG mutagenesis had failed for unknown
reasons. It is possible that the mutants that were generated using NTG resulted in
dominant mutations that could not be complemented through the addition of the wild
type gene. Therefore, I decided to try to generate tagged mutants in triacylglycerol
biosynthesis, subclone the flanking regions around the transposon insertion and identify
the gene of interest by DNA sequencing.

The transposon mutagenesis was successful. A total of 8 mutants were

I42

identified, however only 4 of these were clearly due to a single insertion of the
transposon. The other 4 mutants were removed from study because they did not
contain the transposon or had multiple insertions. Two of the remaining four mutants
were class 11 (mg) mutants and the other two were class III (wax'tag‘) mutants. The
class II mutants were further pursued because of their interesting phenotype. The class
III mutants were again dismissed based on the belief that they might be nitrogen
regulatory mutants since they do not accumulate either waxes or triacylglycerol when
induced under low nitrogen conditions.

Using IPCR it was possible to subclone the DNA flanking the transposon
insertion from the mutant 11-C7. By labelling the IPCR product and using it as a
probe against the cosmid genomic library it was possible to identify 9 cosmids that
shared homology to the DNA. One of these cosmids, 5A-F l was selected for further
analysis based on the simple fact that it was identified first. Attempts at
complementing the mutant, ll-C7, with any of the cosmids has been complicated by
the fact that the mutant does not grow, or grows poorly, when cultured under low
nitrogen conditions. The reason that the mutant was identified at all was probably the
result of culturing the mutant on LB plates and then transferring a large amount of cell
paste to liquid medium for analysis by TLC. Attempts at culturing (in low nitrogen
medium) the exconjugates obtained from mating the cosmids with homology to the
IPCR fragment into the mutant has not been successful. Most attempts have been tried
with the cosmid 5A-Fl, but complementation of the mutant phenotype, in terms of
growth defect, or triacylglycerol accumulation has never been observed. Additional

attempts at complementing the mutant with the other cosmids will need to be repeated

143

before they can be completely ruled out.

Through extended PCR it was determined that the transposon inserted into an
open reading frame that is highly homologous to glnE. This gene product adenylates
glutamate synthase, thus down regulating it's activity in the presence of high nitrogen
levels. Interestingly, glnE was found to be strongly repressed in the presence of
excess nitrogen and carbon limited conditions in E. coli. One phenotypic characteristic
of glnE mutants in E. coli, is their inability to grow in low nitrogen medium after
being transferred from a high nitrogen medium. This is a growth characteristic that is
also true of ll-C7 and 30-F10, the two class II mutants that were identified, loosely
implying that the phenotypes that are observed are due to mutations in nitrogen
sensing and response. This growth characteristic together with the observation that the
transposon inserted into an open reading fiarne encoding an enzyme with a high degree
of similarity to glnE of E. coli is evidence, although not conclusive, that the ll-C7
mutation is a result of a mutation in glnE. However, this interpretation is complicated
by the fact that a cosmid, 5A-F1, presumably containing a wild type copy of glnE does
not complement the mutant phenotype (either its ability to grow in low nitrogen
medium, or its ability to accumulate triacylglycerol).

Examination of the region in the cosmid where the transposon inserted, shows
the presence of three open reading frames that are all being transcribed in the same
direction. Additionally, the first two genes are involved in amino acid biosynthesis
(Figure 4-9). The first open reading frame is glnE and the second shares strong
homology to a branch chained amino acid transaminase. These two observations hint

at the possibility that these genes are organized in an operon. If this is so, then the

144

possibility exists that the operon is not entirely located on the cosmid's insertion. To
determine if this was a possibility the insertion site of the transposon was mapped
relative to the cosmid. First, the transposon was situated relative to the Bglll fragment
that was cloned and sequenced, and then the Bglll fragment was placed relative to it's
location in the cosmid. The results of this mapping experiment placed the Bglll
fragment approximately 6.0 kb from the end of the cosmids insert, and therefore the
third open reading frame identified in this fragment is 6.0 kb from the end of the
cosmid.

The results from this experiment fall in the gray area of interpretation.
Although the distance from the end of the last open reading frame to the end of the
insert DNA is considerable (a distance of 6.0 kb), there are many examples of much
larger operons. There are 4 known examples of operons in A. calcoaceticus. The
lengths of these operons range from about 3.5 kb to over 10 kb. Thus, it seems likely
that the cosmid does not complement the phenotype because the entire operon is not
present and therefore is not being fully transcribed.

The fact that the third open reading fiame does not share strong homology to
anything in the database leaves open the possibility that it could be a gene involved in
triacylglycerol accumulation. Since there are no known genes for diacylglycerol
acyltransferase (DGAT), it is possible that this unknown open reading frame is DGAT.
It can be imagined that the gene encoding DGAT would be situated in an operon that
might be transcriptionally controlled by nitrogen levels in the medium, since it is
observed that A. calcoaceticus accumulates triacylglycerol under low nitrogen

conditions. However, it would not be straight forward to test the function of this open

145

reading frame and there is a high probability that the outcome would not be the
identification of DGAT, but rather another gene involved in amino acid biosynthesis.
Therefore, because of the lack of promise that lies in the direction of this project, I

decided to bring it to a close.

REFERENCES

Allen, L. N. and Hanson, R. S.. 1985. Construction of Broad-Host-Range Cosmid
Cloning Vectors: Identification of Genes Necessary for Growth of Methylobacterium
organophilum on Methanol. J. Bact. l61(3):955-962.

Altschul, S. F., Gish, W., Miller, W., Myers, E. W. and Lipman, D. J.. 1990. Basic
Local Alignment Search Tool. J. Mol. Biol. 215: 403-410.

Bachmann, B. J. 1987. In Egcherichg coli and Salmonella gmhimurium, Cellular and
Molecular Biology. Eds. Neidhardt F. C. et. al., ASM.

Figurski, D. and Helinski, D. R.. 1979. Replication of an Origin-Containing
Derivative of Plasmid RK2 Dependent on a Plasmid Function in trans. Proc. Natl.
Acad. Sci, U.S.A. 76:1648-1652.

Herrero, M., de Lorenzo, V. and Timmis, K. N.. 1990. Transposon Vectors
Containing Non-Antibiotic Reisistance Selection Markers for Cloning and Stable
Chromosome Insertion of Foreign Genes in Gram-Negative Bacteria. J. Bact. 172,
6557-6567.

Leahy, J. G., Jones-Meehan, J. M., Pullias, E. L. and Colwell, R. R.. 1993.
Transposon Mutagenesis in Acinetobacter calcoaceticus RAG-l. J. Bact. 175(6):
1838-1840.

Maniatis, T., Fritsch, E. F., and Sambrook, J.. 1982. In Molecular Cloning, A
@oratorv Manual. Cold Spring Harbor Laboratory Press, New York, New York.

Miller, V. and Mekalanos, J.. 1988. A Novel Suicide Vector and its Use in
Construction of Insertion Mutations: Osmoregulation of Outer Membrane Proteins and
Virulence Determinants in Vibrio cholerae Requires toxR. J. Bact. 170, 2575-2583.

Raleigh, E. A., Lech, K. and Brent, R.. 1989. In Current Protocols in Molecular
Biology. Eds. Ausubel, F. M. et al. Publishing Associates and Wiley Interscience,
New York.

CHAPTER 5

CONCLUSIONS AND PERSPECTIVES

The work described here has provided new insight into wax ester biosynthesis.
It has long been believed that in A. calcoaceticus, wax ester biosynthesis is alkane
oxidation run in reverse, requiring three separate enzymatic steps. This work dispels
this notion through biochemical and genetic evidence. It is now evident that wax ester
biosynthesis is carried out in two enzymatic steps. With the cloning of acrl, I have
shown that the starting substrate for wax ester biosynthesis is acyl-CoA, that reduction
is dependent on a NADPH cofactor, and that a single enzyme catalyzes the conversion
of acyl-CoA to fatty alcohol.

This is a very interesting observation and provides the potential for many
interesting experiments. It would be nice to know how this enzyme catalyzes both
reactions. Future experiments could address whether the reaction is carried out by a
single catalytic site, or if their are two separate sites on the enzyme. Steps in this
direction have already been taken by running timecourse experiments and substrate
competition experiments. It might be important to know if the enzyme is acting alone
as a single subunit or as a multimer. Based on the substrate competition assays it
appears that aldehyde does not compete with acyl-CoA. Thus, it appears, at this time,

that the enzyme contains two active sites.

146

147

Other biochemical experiments that should be conducted involve measuring the
kinetics and substrate specificity of the enzyme. Kinetic studies will be difficult to
pursue given the extremely hydrophobic nature of the substrates. Fatty aldehydes and
fatty alcohols are very insoluble in an aqueous environment. This makes
determination of the actual free substrate concentration nearly impossible. Experiments
aimed at substrate specificity are much easier to carry out. By using different carbon
length acyl-CoA's it will be possible to determine which substrates can be utilized by
the enzyme and preferred substrates can be determined by measuring the amount of
radiolabelled product. Substrates can be varied based on the length of the acyl chain
and the degree of unsaturation. Highly unsaturated substrates may even provide some
rudimentary insight into the organization of the active site. For instance, once the
preferred carbon length substrate is determined (most likely palmitoyl-GOA since
waxes made from this is substrate are the major product in wild type), multiple cis and
trans double bonds can be added to bend the acyl chain. Because of these double
bonds the acyl chain is predominantly in a bent position, and it might be predicted that
if the active site is a narrow opening then "bent", unsaturated acyl chains will not be
readily converted because of their inability to reach the active site.

As described, the Wow15 mutant was the result of a point mutation in the gene
acrl. This mutation effected the ability of the Acrl protein to convert acyl-CoA to
aldehyde, but had no effect on the enzymes ability to convert aldehyde to alcohol. An
important experiment aimed at determining the in viva activity of the Acrl protein
would be to subclone the mutant acrl gene from the Wow15 mutant. Based on the

results described in this study, the mutant allele should encode a protein product that is

148

able to convert fatty aldehyde to alcohol, but should not be able to catalyze the
conversion of acyl-CoA to fatty aldehyde. Additionally, by carrying out nutritional
supplementation experiments on the other wax' mutants that were isolated, it may be
possible to identify mutants which contain point mutations in the acrl gene which
effect the proteins ability to catalyze the conversion of aldehydes to alcohols, but has
not effect on the conversion of acyl-CoA to aldehyde.

Alteration of the pET21zacr1 construct to eliminate the translational stop region
from the native coding sequence, would allow the protein product to be His tagged
(there is a His tag encoded on the pET21 vector allowing for C-terminal His tag
fusions). This would provide an easy means of purifying large amounts of
recombinant protein from E. coli. With purified protein it would be possible to raise
antibodies specific to acrl. This could lead to the localization of the protein in A.
calcoaceticus either by cell fractionation or by in situ hybridization using immunogold
labelling. Because of the novel activity of this protein, possible presence of two active
sites and potential industrial applications, further structural studies may provide
important insights into the catalytic domain. This could be coupled with experiments
dealing with site directed mutagenesis to produce altered enzyme specificities.

One of the desired goals of this work was to be able to transfer the knowledge
gained from A. calcoaceticus wax biosynthesis and apply it to other organisms,
particularly Arabidopsis. Efforts in this direction have already begun in the form of
identifying ESTs (expressed sequence tags) from the GenBank library that share
homology to acrl. The results from such a search have already yielded an EST

(Figure 5-1) that has an optimized BLASTX score of 137 (Figure 5-2). This EST has

FILE NAME

SEQUENCE

SGSBP;

T21872.SEQ

*** SEQUENCE LIST ***

5 I

10

20

AGAAAAAAAA AAGATTGAAA

7O
GATTGTNACG
130
AGGCGCTTCC
190
ACTCAAATCC
250
TCGCCGCAAT
310
TAACGCTGCT
370
CAAGCTTTTG
430
CACCTTGAGN
490
AAGGNNCTAT
550
CTTCCTAGAT

8O
GCTTCGACGC
140
GTCGTCGTCT
200
AAAGGGATTG
260
CTTGTCGAAA
320
GCCAATCCAT
380
GGAAATCAAT
440
AGGGGTTCTT
500
TNGCTATGTA
560
GCCCCGATCA

154 A;

149

121

30
TGGAGAAGAA
90
AAGGGATTGG
150
CTTCCCGCAA
210
ATGCATATGG
270
AGACGGTTCA
330
CTACAGACCC
390
GTCAAATCCT
450
CTGTTATCTT
510
TNGNGCACTA

NGGTC 3'

C;

(SINGLE)

4O
GCTACCGAGA
100
TTTCGGAATC
160
ACAGGCAAAT
220
AATCGTCTGT
280
GAAATATGGG
340
AATCTTGTCT
400
CTATACTTCT
460
CATAACTTCC
520
AAACTNCNTT

Figure 5-1: DNA Sequence of Arabidopsis EST T21872.

123 G;

50
AGATTGGAAG
110
ACTGAGCGTT
170
GTTGATGAGG
230
CATGTTTCCA
290
AAGATAGATA
350
AGCAAAGAAG
410
CCTGCAAGGA
470
TATTTGCTGG
530
NTNNGNCTAA

150 T.

60
GCAAAGTCGC
120
TCGGTCTCGA
180

240
ATGCTCAACA
300
TCGTTGTCTG
360
CTGTTCTTGA
420
TATGGCTCCT
480
NTTTCACCAC
540
CTAAGNGNTT

150

No. Target file Definition Match? Over. 181? OPT
1 T21872.AMI 25.8 97 130 137
10 20 30 40 SO 60
ACR1.AMI LEALFRENVKGKVALITGASSGIGLTIAKRIAAAGAHVLLVARTQBTLBBVKAAIBQQGG
T21872.AMI MEKKLPRRLEGKVAIVTASTQGIGFGITBRFGLBGASVVVSSRKQANVDBAVAKLKSKGI
10 2O 30 40 SO 60
7O 80 90 100 110 120
ACA1.AMI QASIFPCDLTDMNAIDQLSQQIMASVDHVDFLINNAGRSIRRAVHESFDRFHDFBRTMQL
T21872.AMI DAYGIVCHVSNAQHRRNLVBKTVQKYGKIDIVVCNAAANPSTDPILSSKBKVLDKLLGNQ
7O 80 90 100 110 120
130 140 150
ACA1.AMI NYFGAVRLVLNLLPHMIKRKNGQIINISSIGVLAN

T21872.AMI CQILYTSPARIWLLTLXRGSSVIFITSYLLXFTTR
130 140 150

Figure 5-2: Optimized FASTA alignment between Acrl and the Arabidopsis EST,
T21872. F ASTA alignment was done using the software package, DNASIS for
Windows (Hitachi Software).

151

been classified as a putative ketoacyl reductase, because of its similarity to granaticin
polyketide synthase putative ketoacyl reductase. Rough mapping of this EST has
already been done, and was found to map to the upper portion of chromosome 4, near
the hy4 locus. None of the mapped cer mutants have been localized to this region of
the genome. It is important to note that only 9 of the 21 known cer mutants have
mapped. Therefore, although it is disappointing that the EST does not map near a cer
locus, it definitely is not conclusive that this EST is not a gene involved in wax ester
biosynthesis. The EST can also be used to make an antisense construct which could
be used to transform Arabidopsis. Resulting plants could be easily screened for the
glossy phenotype which is indicative of a cer mutation. Additionally, it would be very
interesting to transform Arabidopsis with acrl and see what affects this might have on
the plant. It would be expected that the Acrl protein product would reduce acyl-CoA
to its corresponding fatty aldehydes and fatty alcohol. It is not clear what sort of
phenotypic affect this might have on the plant, however, it might be predicted that
fatty aldehydes and fatty alcohols, which are not normally found within plant cells,
may accumulate in significant levels to produce inclusion bodies in the cells. If
expressed in leaves, alterations in epicuticular wax composition might occur. Although
the mechanism of wax deposition on leaves is not known, it is possible that by
increasing the amount of fatty aldehydes and fatty alcohols present in epicuticular
cells, that epicuticular wax composition might be altered. This might lead to a better
understanding of the mechanisms involved in epicuticular wax deposition by
examining where Acrl is localized. Additionally, by measuring the lipid composition

of the epicuticular waxes in transgenic plants, it might be determined if alterations in

152

fatty aldehyde and fatty alcohol concentrations have any affects on the concentrations
of other wax components, such as wax esters and free fatty acids. Observed
alterations in wax composition could lead to investigations of this effect on insect and
pathogen resistance. Finally, acrl could also be expressed in seeds where it might
cause a decrease in triacylglycerol accumulation by siphoning off acyl-CoA and
converting it to fatty aldehydes and alcohols. This might lead to a decrease in seed
viability because of the lack of an available carbon source for growth.

Other directions leading from this work deal with more immediate questions
dealing with wax biosynthesis in A. calcoaceticus. As described in Chapter 2, Wow15,
was not the only wax' mutant complemented by the cosmid genomic library. The
mutant Wowl was complemented by the cosmid 1A-3F. With the observation that this
cosmid is able to complement two other wax‘ mutants besides Wowl it is imperative
that further analysis be conducted on this cosmid. Transposon insertions have already
been identified which inactivate the ability of this cosmid to complement the mutant
phenotype. Additionally, there have been several other cosmids which have been
found that share homology with this cosmid, some of which do and do not
complement the mutant phenotype. With all of these resources available, discovery of
the region of interest should be straightforward. It will be interesting to discover what
the gene of interest encodes. Nutritional supplementation experiments on this mutant
suggested that it is deficient in one of originally proposed reductase steps. With the
discovery of acrl this now seems unlikely. Possible roles for the mutated gene may
involve nitrogen response and regulation that is specific to wax ester biosynthesis or

perhaps the gene product serves as an acyl-CoAzACP transacylase which converts

153

acyl-ACP to acyl-CoA, which then gets fed into wax ester biosynthesis.

Two other wax‘, null mutants, Wow2 and Wow28 have not been complemented.
It is possible that cosmids which complements the Wow15 mutant, might also
complement these mutants. Investigation of these mutants may yield more insight into
wax ester biosynthesis of A. calcoaceticus.

It is disturbing that a mutant was never identified that seemed deficient in an
acyl-CoAzfatty alcohol transferase reaction. It is possible, although unlikely, that the
same enzyme which catalyzes this reaction also catalyzes the formation of
triacylglycerol. This idea might prompt a reexamination of the class III mutants (wax’
tag') that have been isolated, in terms of experiments aimed at trying to complement
some of these mutants. The transposon-mutagenized tag‘ mutants may be an excellent
place to start since flanking sequence can be easily cloned and sequenced by inverse
PCR. Additionally, nutritional supplementation experiments should be carried out on
all of the transposon marked lines. With a source of readily available, inexpensive
fatty aldehyde (cis-l l-hexadecenal, Aldrich), chemical complementation experiments
using this compound as a carbon source, should be performed for all of the mutants to
further characterize their deficiencies.

Although not promising in terms of understanding triacylglycerol production,
the tag' transposon mutant, 11-C7, should be pursued one further step. The
observation that the region surrounding the transposon insertion contains genes
involved in amino acid biosynthesis, and that these genes might be organized as an
operon leaves the possibility of another gene, perhaps involved in triacylglycerol

biosynthesis, further downstream in the operon. Subcloning and expressing glnE in the

154

11-C7 mutant can bring this question to a close. If this gene is found to complement
the mutation, then there is little doubt that the tag‘ phenotype is the result of the
insertion in the glnE gene. If not, then experiments aimed at identifying and
subcloning open reading frames further downstream may provide the answer. A
valuable tool for such experiments will be pSERZOO-l and pSER200-4, two
transcriptional expression vectors that can be used in A. calcoaceticus and E. coli. The
construction of these vectors is detailed in Appendix B of this text. Experiments
aimed at investigating the 30-F10 phenotype (the other tag', transposon generated
mutant) should probably be avoided, if the goal is to investigate structural genes
involved in triacylglycerol production. The reason for this caution is the observation
that this mutant exhibits poor growth under nitrogen limited conditions, similar to 11-
C7. This hints at the possibility that this mutant may also be affected in nitrogen
sensing and response.

In conclusion it is apparent that this work has answered some of the questions
dealing with wax biosynthesis in A. calcoaceticus. It has provided a tool for the
investigation of similar pathways in other organisms, and opened new avenues for
scientific investigation. It is my honest wish that work will continue to be focused on
wax ester and triacylglycerol biosynthesis in A. calcoaceticus so that more insights and

discoveries are gained.

APPENDIX

APPENDIX A

APPENDIX A

DNA SEQUENCE INFORMATION AND CONTIG MAPS

In the course of identifying the genes of interest involved in wax ester and
triacylglycerol biosynthesis, a number of open reading frames were identified. This
appendix lists the sequence data that was generated in the course of sequencing the
fragments that contained the genes of interest (Figures A-l through A-14). For the
most part they represent single stranded reads that were obtained using Perkin Elmer's
automated sequenator, the ABI310. The contigs that were sequenced are highlighted
in Figures A-15 through A-18. Contigs were assembled using Perkin Elmer's
Sequence Navigator software. Double stranded sequencing was conducted on the

region encoding acrl and is indicated by reverse arrows.

155

FILE NAME

SEQUENCE

IDl.

7468p;

DNA

*** SEQUENCE LIST ***

SI

10
ATGCAGCTCG
7O
GTGCATCAGG
130
CATAAAGTGG
190
CGTAAGCGTG
250
CGACCACCAC
310
CCACAGGCTG
370
CCACAAGAGC
430
CTAAATAATT
490
TGCATTGACG
550
AAGATCAACA
610
CTATGAGCCA
670
TTCGATATAA
730
ACCCCCAGGT

20
AAGGTGTTGT
80
ACGGTCTGGT
140
TGAAACCAGG
200
TCAATTTGAG
260
GTCGTGAGCA
320
CACGTTCAAA
380
AAATTTTGGT
440
TTAGTCCTGT
500
CTGCTTTAAA
560
AAAAAAAGTG
620
TAAAAAATGT
680
GAAACTCAAG
740
CAGAAACAAG

Figure A-l: Sequence ID#l

243 A;

156

146

30
GACTAATGTG
90
ACACATTTCT
150
TCAGATTGTG
210
TATGCGTCCA
270
ACAAGAACAG
330
TGATCAGGCT
390
GGGTTGGGCG
450
TTAAGAAGAA
510
TTAAATTTAT
570
AAAACATTCG
630
CAGACAAGAT
690
GAATGCGTCA

TTTCAT 3'

C;

(SINGLE)

40
ACTAACTTTG
100
GAGCTAGCTA
160
CAGGTTCGTG
220
GAAGGTGCAG
280
CGCGGTGAAC
340
AAAAAGCCAC
400
CATTATTGCT
460
AAAACCACCT
$20
TGTCATGAGA
580
GCATCGTCAC
640
CACACCGCTT
700
ACTTTTTTAT

172 G;

50
GCGCGTTTGT
110
ACGAGTTTGT
170
TGATTCAAGT
230
AAGCTCCAGC
290
GTAAGCCGCA
350
AACGTGCCAA
410
ACAAGCTGGA
470
TCGGGTGGTT
530
ATAAAGATGA
590
AACAGCACAG
650
GAGATCAAAA
710
TGAATGGTTA

185 T.

60
GGATATTGGT
120
GTCAGATCCA
180
CGATGCAGAG
240
GAAGACACAA
300
GCCAAAACGT
360
GCAAGAAAAA
420
ATTACTGGTT
480
TTTTCTTATG
540
GGCGATTCTG
600
CATAACTCAA
660
ACATCGGGTT
720
TTTAATGACA

FILE NAME

SEQUENCE

ID2.

93OBP;

DNA

*** SEQUENCE LIST ***

SI

10
GCGCTATTCA
70
TTTAATTGCA
130
CTGGGCGATT
190
TGTGATGATC
250
TACGCTGGCA
310
TGACCCTCAA
370
TATCGTGTCA
430
CAATGCCAAT
490
ACTGGCAGTT
550
GATCCTGATT
610
ACAAGGCTGG
670
CCTTCTCATC
730
TTTGCATAGA
790
TTGAGTTGCC
850
CCGACTTCAT
910
TGATATTCAT

2O
ATTCTGGCTT
80
GTCGTGGCTG
140
CAGCCTAATA
200
TATCCAGATT
260
GTTGTTGCTT
320
AAACGTTCTC
380
GGCTTCATTG
440
ACTGGTGCGC
500
TTATGGGGTG
560
CTTACAGGTT
620
AAGCAGTTTA
680
GGAATGTTGA
740
GGGCGTGCGC
800
GACTCAAGTA
860
CGCAATGAAA
920
CAGGTGATAC

Figure A-2: Sequence ID#2

238 A;

157

202

30
GGGACAGTAG
90
CTGCGCTGCT
150
ATCTGATTCA
210
TCAGTGCGCT
270
CTTTAGAAAC
330
CACCACCCAA
390
GTGGAATGCC
450
GTAGCAAATG
510
TGCCATTAAT
570
TTATACTTAC
630
TACCATTTAT
690
TTGGGCTAGC
750
GTGTACAAGG
810
ACTTTTCTAA
870
AGCTGGTGAT
930
AGGATTATCA

C;

(SINGLE)

4O
TCCACTCAAA
100
CAATTATCTA
160
GTTACCCAAA
220
CAGTAATCCA
280
GTTATTAAAT
340
TCGTGAATTA
400
AATCACTTCG
460
TTCAACGATT
520
GAATATGATT
580
TCATCCCAAG
640
CATTACGCTA
700
CACCAGTATC
760
AAAAACATTT
820
CATCGTGGCG
880
TGATGCCACC

3|

202 G;

50
AAATGGGTAT
110
CTGGTTTATT
170
ATCCTAGACG
230
CTCATTTACA
290
CTTGAAGCAG
350
TGGGCGCAAG
410
GTCATTGTGC
470
ATTCATGGCG
530
CCATTATCCG
590
ATGTTTAAAA
650
GTTGGAATGT
710
GCTTTTATTC
770
GCATGGTGTC
830
GTGCTTATTT
890
CAGTCAGATA

287 T.

60
TGCCTTCTGC
120
TTCAGTCACC
- 180
CACCTGAAAG
240
CAGGTGCTAT
300
CAGACAAGCT
360
GTACAGGCAA
420
GTAGTTCGGT
480
TATTACTGTT
540
CACTGGCTGC
600
NGCTGTATCA
660
TGATCACCGA
720
TGTACGGAAA
780
ATAACACGTA
840
CTGCGCTAGA
900
GCATTGATTC

FILE NAME
SEQUENCE

SI

10
GTGGGCTGTA
70
TTATTTAGTT
130
GTTTTGCCTG
190
GTGCGGTCAC
250
AAATGCCACA
310
ATATTATGCA
370
TTCACTATAA
430
GCAAAGAAAT
490
TCTAAATCTG
550
ATAAATTTTA
610
ATTTTTCCTG
670
TCAGAATATT
730
GCGTCGCAAT
790
CTGGGGAAAG
850
CCAATGGCAT
910
AGCGACTGAA
970
TGGCCAATAC
1030
GTGGCAGTAA
1090
CAAAATCATG
1150
TATTGATCAG
1210
TCGCATTCAT
1270
TTGCAGCTTT
1330
CTGCCGCAGC
1390
AAGCCACTTT
1450
TTTCCCTGAT
1510
TAAAGGGAAA
1570
AGTGTTAAGT
1630
GCTAGATCAA
1690
GCAGAGGTTT

ID3.
17OBBP;

DNA

20
TGGATTCACG
80
TGAGAATCGC
140
TCAAGCCAAA
200
CAGCGCATGC
260
CATCCAATAC
320
ACCGCGTGAA
380
TATTCAATAC
440
TGATATTGTT
500
GCTGATGGTC
560
CCAGTGTTCG
620
TTCACCCAGT
680
ATTGATGTCT
740
ACGTGTTGGA
800
CGTGGGCACG
860
ATAAATCGAG
920
GGCATCCAGC
980
ACCAATAGAG
1040
ATTTAACACT
1100
GAAGCGATCA
1160
GAAATCGACA
1220
GTCAGTCAGG
1280
CACTTCTTCC
1340
AATTCTTTTT
1400
ACCTTTTACA
1460
TGATATCAAA
1520
TAAAATTTCC
1580
ATCTGAAATT
1640
TTATGAACAT
1700
TTTATAAAAT

Figure A-3: Sequence ID#3

499 A;

158

210
CAGTTAAGAT
270
GTGCTTGGGC
330
TTAAGCAAAG
390
ATTATTCAGC
450
GGTGCGATTT
510
GTAACCATAA
570
CCTGGGAACA
630
GCAGCCGTTG
690
GGTGCGATGG
750
CGTTTCACAA
810
TATTTATAAA
870
GTAATTGAGA
930
GCAGCTTTAG
990
CTGATATTGA
1050
AAACGTACCG
1110
AACGACTCGT
1170
TGATCGACAC
1230
TCACAAGGAA
1290
AGTGTTTCTT
1350
GCAATCGTCA
1410
TTCTCTCGGA
1470
CTGTAAAGGC
1530
AGAACTAAAG
1590
ACAAATGTTT
1650
CATACGATAA

TTACGCTC 3'

C;
40
GAAATGGTGT
100
TGCGGGGCAG
160
TGATTCTGGT
220
TGCAACATAA
280
CACTCATGCA
340
CATTTATTGA
400
ATAGTTCTGT
460
ATGATGTAAA
520
AAAAGCCTCT
580
AGCGGGCATA
640
AGCTTGGGAA
700
CATAGGTAAT
760
TGGCGTAGAC
820
TTTTGGTGGG
880
TTTTATGCTT
940
ACGCGACATA
1000
TGATCTGGCC
1060
CACCAAAGTA
1120
GTACGGCACG
1180
TGGCCATAAT
1240
AAATAGAGGC
1300
GGGTTCGGGC
1360
AACCGATTCC
1420
AGAGAGCTTC
1480
TGTCATTATT
1540
TTAATCTAAA
1600
TTAGAGTAAT
1660
TAAAGCAAAT

383 G;

50
TTGATGTGGG
110
AAAGTCCTTG
170
TTTGGGACAT
230
GCAGATTTCC
290
TTCGGTTGGC
350
TCAGGTGACT
410
ACTTAAAGAT
470
AACAGGAAAA
530
CGGTATGAGA
590
GGCACGTTGT
650
TAGGTTAAAT
710
TGACGCCAGA
770
AATGAGATCT
830
TGCGATCATT
890
GAGTACCTCG
950
AGCAGAAAAA
1010
ATTTTTACGC
1070
ATTCAGCTGC
1130
GCGAATCGAA
1190
TTGTTGTGAT
1250
CTGTCCCCCT
1310
AACCAATAAT
1370
ACTAGATGCA
1430
AAGTTTTTTG
1490
GTATTTGGAT
1550
GTGATTTGGT
1610
TGCTCTAATT
1670
ATAAAAAAAC

504 T.

60
CATTGGTGAC
120
GATCATTGGA
180
ACCGACTGTG
240
GATATTCAGG
300
AGTGTATTTG
360
GCTATGAATG
420
ATGCTGGATC
480
GTCAATTTCC
540
GGCTTCTTTT
600
AGCAGATTCA
660
CCAATCGACA
720
CGACCCAAGT
780
GCGGCTTCTT
840
GGGGTACGCA
900
GCTGAAAGAC
960
CGGGTCGCAT
1020
TTAATCATAT
1080
ATGGTGCGTT
1140
CGCCCTGCAT
1200
AACTGGTCAA
1260
TGCTGTTCAA
1320
ACATGAGCAC
1380
CCAGTGATCA
1440
TTCACGCGTT
1500
TGAAGACGGT
1560
TAAAAAATCT
1620
TTTATTATTT
1680
CTCTGAAATT

159

FILE NAME ID4.DNA
SEQUENCE 33SBP; 125 A; 52
*** SEQUENCE LIST ***
10 20 30
5' ATCAAACTGT AAAGGCTGTC ATTATTGTAT
70 80 90
ATTTCCAGAA CTAAAGTTAA TCTAAAGTGA
130 140 150
GAAATTACCA ATGTTTTTAG AGTAATTGCT
190 200 210
GAACATCATA CGATAATAAA GCAAATATAA
250 260 270
TAAAATTTAC GCTCAAACTG TGGCCAATGC
310 320 330
ATGTTCAATC CCAACCGATA ACCGAACCAT

Figure A-4: Sequence ID#4

FILE NAME

SEQUENCE

IDS.

5118p;

DNA

*** SEQUENCE LIST ***

SI

10
ATTGTCGACT
70
CGATAATATT
130
CAATTAAACT
190
GTTTGGGGAG
250
CAATATTGTC
310
ATGCCAGCGC
370
CCGTGGTTGG
430
TCTGCACCAT
490
AACAGCTTTC

20
ACCACAGGGT
80
ACCTAATGGA
140
GCGTAATTGC
200
GGTATGCGCA
260
GCCTGCTTCG
320
CAAAGCACCA
380
ATTCATGATT
440
GTTGTGTATT
500
GTGGTTGGTT

Figure A-5: Sequence ID#5

143 A;

111

30
ACGCCATATT
90
TTGCCAATCG
150
TCGGGCTGCT
210
AATAAATTAT
270
GTTATCGTCT
330
ATTCCGCCTT
390
CGAGTATAAA
450
GTCAAAAGCA
510
CTGGTGAATA

C; 50 G; 112 T.
(SINGLE)
40 50
TTGGATTGAA GACGGTTAAA
100 110
TTTGGTTAAA AAATCTAGTG
160 170
CTAATTTTTA TTATTTGCTA
220 230
AAAAACCTCT GAAATTGCAG
280 290
TTGTTCCAGA TCTGCAATCA
ATCTTCTGA 3'
C; 115 G; 142 T.
(SINGLE)
40 50
CATGGGCAAT CTTGGCAATC
100 110
ATTCCACAAA AACCAGTTTG
160 170
GATCATCGAA GAAACGTACC
220 230
ATGTACCCCC ATAGAGTGTA
280 290
GAATCGCATA GGTAATGGCG
340 350
CCAATGCGGC AAGTCGCTGC
400 410
TATCACCCTG TACTTTCAGA
460 470
TAGGAAGTGG TTTGATAAAT
A 3'

60
GGGAAATAAA
120
TTAAGTATCT
180
GATCCATTAT
240
AGGTTTTTTA
300
AATCATCAAT

60
GTTTCCAAGT
120
GTTTTGTCAT
180
TCAATGCCCT
240
GAAACAGATG
300
GCCATACCTG
360
TCGAGCACGG
420
TCAAAATAAA
480
AGGTACAAGC

160

FILE NAME ID6.DNA
SEQUENCE 3513p; 120 A; 59
*** SEQUENCE LIST ***
10 20 30
S' CATATTCTGT ATGCTTAATT TATCTATTTN
70 80 90
NGATTAAAAA TTTACTTCAT TTCTAATGCT
130 140 150
GCCACAAAAA CTTACTCGGC AGTCTGGGGC
190 200 210
TTCAAATCAT GATAAAAAAG CTGGAGAATT
250 260 270
ATTATCATCG GAAGTGCTTG AGCAAGCAGG
310 320 330
CACTTAAACC ATGTCACCGA CTCCATGTCA

Figure A-6: Sequence ID#6

C;

(SINGLE)

40
TATGAGGAGT
100
TGATTTCAAT
160
TAGCAAGACC
220
GACCTAATGA
280
CGTTCAGCGC
340
AAAAGCAAGA

65 G;

50
TCAGCTTTTT
110
TTACATGATG
170
ACTATAAAAA
230
GTAATCAAAA
290
TGTTACGGTA
350
TTGANTGGAT

103 T.

60
TAAGAATAAC
120
TTGAGAGAAC
180
ACNAAAGTAA
240
AGTATCCGAT
300
TTGTGGGTGA

C 3'

FILE NAME

SEQUENCE

ID7.

953BP;

DNA

*** SEQUENCE LIST ***

SI

10
CGGNCCANTG
70
TTGNGCTGTG
130
AGGAAAAATT
190
TTGAACAAAA
250
GGCAAGCTTG
310
GAGCAGGCAG
370
AATCCGATGG
430
CTTTATGCAA
490
AATAATTTCA
550
AAAATGCATG
610
TCAGTCATAA
670
AATAGATGCT
730
GGCCCAATAT
790
CCTCAAGGAT
850
TTGTTCCCTT
910
TGTGTGAAAT

20
CATGCCCCAG
80
TGGCGATGGT
140
ACCGATTAAA
200
AGTCGAAGGC
260
CCAGTGTGAT
320
TTGAAAACTT
380
AACTGGTGAT
440
TTAAGGCCTT
500
TTAAATAAAT
560
ATCTAGCGTC
620
AAACCGATTA
680
ATTTTTCGCT
740
ATTTATACGA
800
ATCAAGCTTA
860
TAGTGAGGGT
920
TGTTATCCGC

Figure A-7: Sequence ID#7

260 A;

161

199

30
GCGCTTGGCG
90
GGATTAGCCA
150
ATTGTGGTAT
210
TTGCTTGATC
270
CGGGCTACAT
330
CTTAAAGCAT
390
GCCGCCAGAT
450
GATGTCGGGT
510
TTANTGAACA
570
ATGCATCCNT
630
AACTCAATCA
690
CATTGCGATG
750
TGAACCGAGT
810
TCGATACCGT
870
TANTTTCGAG
930
TCACAATTCC

C;

(SINGLE)

40
CACAAAAAGC
100
TGTTACTTGG
160
TTAACAACAG
220
ATTATACCGA
280
GGGCAGACTG
340
GATGGTCCAG
400
CCGANTCTGA
460
CGAGTAGATG
520
CAAGTCANCT
580
TATATCAGGA
640
CGCCTTATTT
700
TATTGGTTTT
760
ATCCAAGATG
820
CGACCTCGAG
880
CTTGGCGTAA
940
ACACAACATA

207 G;

50
TTCCCNGAAT
110
TGATTTACTC
170
CTCGCTTAAT
230
CTTGCTCAAT
290
TGACGCATGG
350
CATTACTCAA
410
ATCAAGTCTC
470
ACGTTAAAAA
530
CAGGGCACAN
590
AACANNNNAG
650
TTTTACGGTA
710
TCTATTGAGG
770
GTTCCGCCGA
830
GGGGGGCCCG
890
TCATGGTCAT
950
CGAGCCGGAA

273 T.

60
CGCCAGATTA
120
ACTACGATTC
180
TTTGTTGAGC
240
CCTGATTTTG
300
CGATGGCTTG
360
TGTGCATACC
420
GTCCACTTCA
480
TTTGTTGGTC
540
CTATAAGTTA
600
ACAACATTTA
660
TTCAGCATAT
720
AGTCCTAAGT
780
AGCGCGAAAT
840
GTACCAGCTT
900
AGCTGTTTCC

GCA 3'

162

FILE NAME ID8.DNA
SEQUENCE 681BP; 183 A; 146
*** SEQUENCE LIST ***
10 20 30
5' GCCTTTCGAC GAAATGGNAG GNTTTATNGA
70 80 90
TTTATGCGAA AAATTCGGTC TTCTGTGAAG
130 140 150
TATCATGAGC ATGGCCTGTC AGGCAGCACT
190 200 210
GCCAGCCAAT ATCAGTGAAG CATCGGCTGA
250 260 270
TGAACCTGAT ATGTTGCCAA ACAAAGCTGA
310 320 330
GCATCAAAAA ATTGGAATTT ATGCAGGTGC
370 380 390
TGCGTTTGCT GAAAAGTTAA AAGCCCCTGT
430 440 450
GGAGTACGAT AACCCATACA ACATGGGCAT
490 500 510
TCACACGCTC ATGGATTGTG ATTTACTGAT
550 560 570
GTACTATCCA AGTCATGCCA AATTCTTGCA
610 620 630
ACGTCACCAA TTACACTGGG TGCGGTAGGT
670 680
CCATTGCTTG AAACGCGTCA A 3'

Figure A-8: Sequence ID#8

C;

(S INGLE)

40
CTTTCCGNAA
100
NAAATTACTC
160
CAATAAACGC
220
AGCGGGTCTA
280
ACTGCATCAA
340
GGGGTGTGAA
400
GGCGCATACC
460
GACGGGTATT
520
TTTGCTGGGT
580
GATTGATATT
640
AAAATCTCAT

161 G;

50
TACGTCGATT
110
AACCCTCACA
170
GGTGTCGCTG
230
CCTTTCGTGC
290
ATGGTTGAGC
350
GGCGCACATG
410
TCACGTGCCA
470
GTTGGCAATA
530
GCCGATTTTG
590
GATCCGACAC
650
CGACGCTCGA

184 T.

60
TTAAATCGNG
120
AGCTAGNACA
180
TGGTCATCGT
240
CACGTCATGT
300
TGATTTCCCA
360
ATCAGTTAAT
420
AAGATTTTGT
480
AAGCAGGTTA
540
CATGGGCACA
600
ATTTAGGGCG
660
TGCCTTATTA

163

FILE NAME ID9.DNA
SEQUENCE 354BP; 87 A; 100
*** SEQUENCE LIST ***
10 20 30
5' TCGGTCGTTC GGGTGAGTGA GAGAGGCATC
70 80 90
CACAGTAATC AGGGTGATAA CGCAGGAAAG
130 140 150
AGGAACCGTA AAAAGGCCGC GTTGCTGGCG
190 200 210
GCATCACAAA AATCGACGCT CAAGTCAGAG
250 260 270
CAGGCGTTTC CCCCTGGAAG CTCCCTCGTG
310 320 330
GATACCTGTC CGCTTTCTCC CTTCGGGAAG

Figure A-9: Sequence ID#9

C; 96 G; 71
(SINGLE)

40 50
AGCTCACTCA AAGGGGGTAA
100 110
AACATGTGAG CAAAAGGCCA
160 170
TTTTTCCATA GGCTCCGCCC
220 230
GTGGCGAAAC CGACAGGACT
280 290
CGCTCTCCTG TTCCGACCCT
340 350
CGTGGCGCTT TCTCATAGCT

60
TACGGTTATC
120
GCAAAAGGCC
180
CCCTGGACGA
240
ATAAAGATAC
300
GCCGTTACCG

CACG 3'

FILE NAME
SEQUENCE

ID10.DNA
2341BP;

*** SEQUENCE LIST ***

SI

10
TTTTATAAAT
70
ATCTGTATTA
130
CATATACGCT
190
ACNAAAGACG
250
TATAATGGAT
310
TTAAGCACTT
370
TTATTCACAA
430
AAAATTGTGC
490
TCTACCAGCC
550
AAACCAAAAT
610
TTGAGATGCT
670
ACCAATATTC
730
ACGATCCATT
790
AAACACTATC
850
CTTGAGGGGT
910
AATGTGATAT
970
GATAGTCTCG
1030
GTTTGGAGTC
1090
CGATTGGATT
1150
GCTCAGGAAT
1210
TCGTTTTCGC
1270
CTACCCCCAA
1330
CCCTTCAAGA
1390
ATAACGAGTT
1450
CCATCACTCA
1510
ACATGTGTTT
1570
ATATTTTGCA

20
TTTTCTCTCT
80
ANATTTTTAA
140
GCACTCAAAT
200
CATCCAGATA
260
CGACATACCA
320
CAGGACTCAT
380
TTTTTAAAAT
440
CCATGCGCGC
500
CCGCCGAAAT
560
TTGCAATGTG
620
ACACGACTTG
680
TGCTCAAAAG
740
GCAATCCACA
800
CTCTGTGACC
860
TTCTTCAACA
920
AAAATTTACA
980
AACAATATTT
1040
GCCGCCTAGG
1100
CAGCGCCCCC
1160
AACTGCATAA
1220
CTGAGGATAA
1280
TATAAAATCC
1340
TCGTTAATAA
1400
TATGCATTGC
1460
CATACACAAC
1520
CGCCTTGATC
1580
TTTTTGCCCT

Figure A-lO: Sequence ID#IO

681 A;

164

532

30
CTGGTTCTTT
90
ATTGAATTAA
150
GTGGCTGTTT
210
GGGAGAAAAC
270
AAACGCAATT
330
CAGGTCGCTG
390
TTCTTTGTTG
450
CATGTCACGA
510
TTTACGATCG
570
CAATATTTTC
630
CTTCATTGAG
690
GCAATTGATC
750
CCTACACGGC
810
CATACCGTGG
870
GGAAAGAGCT
930
GGCGTACGTC
990
TGAATAGACG
1050
GCAATCAAAA
1110
TGTGTCACAA
1170
TTAAACCAAT
1230
ATTTTTCCAA
1290
AGGCGGCTGC
1350
TGGCAACTGC
1410
CTGATACAGG
1470
ATGCATATTT
1530
TTAACGGATT
1590
GAACCGCATC

C;
(SINGLE)

40
ATACACCACA
100
GCTTTCAACA
160
TAACTTTGCT
220
GCTTTTACAA
280
TTGGTTTCTG
340
TGCCCGATCA
400
GCCCAACTCG
460
AAACTAAAAT
520
GTGCAGTAGA
580
ATTTAATTAA
640
TGGAGCCGAG
700
AATCATTTGA
760
AGCCTGCAGT
820
CCGCCAACTG
880
GTAATTGCTG
940
GTGAGGTCGT
1000
TTAAAACTTT
1060
TTCGATTAGT
1120
TGACTCGATC
1180
GTATCGGTAA
1240
GTAACCACAC
1300
TTGATCTGTG
1360
TGTAGAGGAA
1420
CCTAAAGCTT
1480
TTTCCTAAAA
1540
CGCTTACTTC
1600
GAAAAAGGTT

432 G;

50
CCGCAAAAGA
110
TTACGATGCC
170
ANATATTCCC
230
AATCCAGCTT
290
GAAACTGCTG
350
ATACCAAATC
410
CACCGAGTTT
470
CGTAAACAAA
530
AATGCGCGCC
590
AAACTCGATG
650
TTTTGCAGCA
710
GTAATTCTAT
770
CAATGCTTCA
830
CATTTGTTCA
890
TGCACAAGGC
950
CAAAATAATA
1010
TTCTTGATTC
1070
GTCGTGACGA
1130
AGATGCAGCC
1190
GCTTGGCTTC
1250
TCGAAAATGC
1310
ATACCTGATG
1370
TTTCCTCAAA
1430
GTGATCGGTG
1490
ACATCAATAA
1550
ACGTAAGTTA
1610
TTCTGGATTT

687 T.

60
GAAAGGTNTT
120
CGATATTGGG
180
CTGCAGTAGT
240
ATGTTCAAGA
300
TTTGATCTGC
360
AGGCTCGAGG
420
TTTGGTTTTA
480
ATGTCCATTT
S40
TTGCTTGTTA
600
AGTCCAATAT
660
CACGCATGAC
720
CAGATTTTAA
780
AAAATCATGG
840
AATATCCAGC
900
AATGTTTTTA
960
TGGGTATCTG
1020
CATTGATGTC
1080
TGCTCATTTA
1140
ACACCATCAT
1200
ATCAAAATAA
1260
GTATGACTTC
1320
TGTAAACAGC
1380
ACGGCTATGG
1440
TCCTGCTTTT
1500
AAAAGGCGAA
1560
ACCAGTGTTC
1620
CTGTGGTAAT

1630
TGGGCCACGG
1690
TGCGGCAGTA
1750
ACGCTCAACC
1810
CCCTTCTAAT
1870
CACGTCCTCG
1930
ATAACCTGAA
1990
TTTTGCCTTA
2050
AATACCACGT
2110
AGCATGAATA
2170
TGGACGCAAG
2230
AAGTGTCGCT
2290
CCTNGTGTGA

C 3'

Figure A- l O (cont'd)

1640
CATCCTTCAC
1700
CCAGTAAAGA
1760
ACTTCGTATC
1820
GCACCACCAG
1880
CCTGCACCTT
1940
CGCGCTACTT
2000
CACATCGTCA
2060
GCCATTGCCT
2120
CTATTTTCAG
2180
TAACAAGATG
2240
TGATCAAATG
2300
TCTTGGANAA

165

1650
CAATTTCACG
1710
ATGCTTCATC
1770
CTAAATCTTT
1830
CCAAATCAGG
1890
GACACACAAA
1950
CCTGATGTGC
2010
CGTTTGGATG
2070
CTTCACCAAG
2130
TGGCAGCAAT
2190
CTAATTTATT
2250
GAACTTTCAT
2310
CAAAAATTGC

1660
GTCATCGTAT
1720
TGCAATGTAG
1780
TGCAATGGTA
1840
GGTATGCAAC
1900
GCCTTGTGGA
1960
CAAGATCGAC
2020
ATGATGCGTA
2080
GTAAGCACCC
2140
ACCTAATTTT
2200
TTCACGAACC
2260
TTGATAAATT
2320
NGTGGCTTTA

1670
TCACGAATTG
1730
AATTCATCAC
1790
ATAACTGTTT
1850
ACGCCATCTT
1910
TCTAGCAACA
1970
ACAGTATAGT
2030
AACGAAGATG
2090
CAGCCCCATG
2150
TCAGAACCGA
2210
ACATCAATTT
2270
TTTGCAGAGT
2330
AGANTTTCAT

1680
GCGTCACTTC
1740
GGGTAATACG
1800
GACGGGTAAT
1860
TAATCAAGAA
1920
TCGCTTCATC
1980
TGCCACATGC
2040
TTTTTACACG
2100
CCGCAACGGC
2160
TCCAGATCAA
2220
GAGCTTGCTC
2280
TTAACAAACG
2340
NAGGGCCGGA

FILE NAME ID11.DNA
SEQUENCE 3258p; 74 A; 76 C; 73 G; 98 T.
*** SEQUENCE LIST *** (SINGLE)
10 20 30 40 50
5' CTCTCGGGTA TGGCGCCACT GTGCAGCATT CACNTGCANG GATTGATTTG
70 80 90 100 110
ACGTGGCTTT CGGCTCGTTC ACGTAGATAA GCATGCATTA ATGCTGTGGC
130 140 150 160 170
GATAAGCAGC CTGCTTGAGC AGCATCTTCT AGAATTCTTA CATTGTCGGA
190 200 210 220 230
AGATCAGGAT TCGCTCCACT CCATGCAAGC ACCATATACT GTGCGATAAA
250 260 270 280 290
ACGATACCAC CTGCATCCTG TTTTAAATGA TGCCATGTTT TTTTTGCTCA
310 320
CCAGATGTCT TTCNTCTTGT GGCGC 3'

Figure A-l 1: Sequence ID#11

166

60
CAAGGGNTAA
120
ATCTTCACTG
180
AAAATGGGCG
240
TTCGATGTCA
300
TTGGATGATC

FILE NAME

SEQUENCE

13OOBP;

ID12.DNA

*** SEQUENCE LIST ***

5U

10
GGCGATTGGC
70
AAAGAAGGTG
130
ATTTATGCGG
190
GCCTGCAACT
250
GTTGCAAAAA
310
TTATTTAGAA
370
TCTTCCAGTC
430
TAGTTCGTAT
490
TCACCAATGT
550
CAGCCAAGCA
610
ATGGTCAGAT
670
ACTCCATCCA
730
CGTAAATCAA
790
TTGCTGGATC
850
TGGGGTGATG
910
CATGGACGCG
970
CACGGCGATG
1030
TCCGCTTTCG
1090
AATATCGCAG
1150
CAGGTATTTC
1210
GTCAGCCTAA
1270
GAAGATGCCT

20
TGTGTAAGCG
80
AAGGTACTGT
140
AAAAGCAGGT
200
GCTATAACTA
260
ACATTAAGAT
320
CAAGATTATC
380
AGTTGAAAAT
440
TCTGCGTGCG
500
GGTGAGTTTG
560
GTTTGCACGT
620
TCANGATTTG
680
GTGATATTGA
740
TTGATGTGCA
800
ATATTACCGC
860
GTTCAGCTTT
920
AATGGGAGCG
980
ACCTGCTGGA
1040
CAGCCATGCG
1100
ATGATATTAA
1160
AGCTTATCTA
1220
ATCATTTAGG
1280
ATTTGTTTTT

Figure A-12: Sequence ID#12

355 A;

167

227

30
GTTTGTCGAT
90
ATTTACGGNG
150
GAGTGAAGAA
210
TGCAAACAAA
270
CAAGTCAGTA
330
AAATTGATCA
390
GAATTAAAAG
450
CGTTTAATGT
510
ACGCGAGAGC
570
GCTCCTTTGG
630
ATCGTCATTG
690
TCTGATTGTT
750
GCAGTGTGTT
810
AGATGGTTTT
870
GGCAATTAGC
930
TTATGCGTGG
990
AATGACCCGA
1050
TGAAATGAAA
1110
GCTAGGCGCG
1170
TGGTGGATCG
1230
TCAGGCTGGC
1290
AAGACGTGTA

C;

(SINGLE)

40
CTTAATCTCG
100
ACCTTACCTA
160
AAATTGGTTT
220
AGGTTGCAGG
280
CGCAGAACAG
340
GTGTTGCTGC
400
ACATTCAGGA
460
TCCGTTGGTT
520
TTTCAGATTT
580
TTGCCAAACA
640
GTATGGGTAA
700
GCCTATGATG
760
GTATNCCCGT
820
GTATTTCGTG
880
CATATGGCAT
940
ATCAAAGCGC
1000
CCATTTGTGT
1060
AGTATGATCG
1120
GGAGGAATAC
1180
AAACGTGAGT
1240
TTACTTCAGT
1300
GAACATNCCA

338 G;

50
GTGCAATTTC
110
TTGCCCAGAA
170
AAATAGAACT
230
TTTTTTATGA
290
GTTTTAGGAT
350
TCCGCTTTCG
410
TGAAAGTCAG
470
CTGGCAAGAT
530
TGCAGATGCC
590
TGGAGAGCCT
650
GCTAGGGGCA
710
AGCAGGGCGA
770
GGGGACAAAA
830
TGGATATGCG
890
TAGAGAAATA
950
GTATCATCAG
1010
TTCGGCGCTA
1070
AACGTGAAGT
1130
GTGAAGTTGA
1190
TGCAAGATCG
1250
CACAAGATGT

3'

376 T.

60
AGTCACCTCC
120
TGTTGCAGAG
180
CCCTAAACAT
240
ATACTGAGCA
300
TACATCAGCT
360
AGAGAGAATA
420
TGGATGCGAG
480
GCCAATCGAC
540
TGCATTTGTG
600
GTAGGATATG
660
CAAGAGCTGA
720
AACAAATGGT
780
GCTGATTTAT
840
ATTACGTCCC
900
TCTGATCCAG
960
TGGTGGAAAG
1020
TGTCGATTAT
1080
GGCACGTCGT
1140
ATTCATTGTT
1200
CCAATGTTTG
1260
GATTGAGTTA

168

FILE NAME ID13.DNA
SEQUENCE 446BP; 140 A; 92
*** SEQUENCE LIST ***
10 20 30
5' CGATCTGTTT GTTTNTANGA CGTGTAGNAA
70 80 90
AACCCAGATG TTACCGATGG AGCCTGAGCT
130 140 150
ACCCAACTTG GAACAATTTT ATTGAGGCTT
190 200 210
AGTTTAAAAA ACTGATTCAG GAAGAAGTCA
250 260 270
AACAACAACT TAATGCTATC TTGGATGAGA
310 320 330
GAGTAATGCA TTAAAACGGC TTCCTTCTAA
370 380 390
TCATTTTATT GAAGCTATTT TACAGNCTGA
430 440

GCCTTGATTG ATCTTGTCAT

Figure A-13: Sequence ID#13

GCGCCG 3'

C; 90 G; 117 T.
(SINGLE)

40 50
CATGCCATTC AAGCCTTGCN
100 110
ACGCCAACGT ATACTCAGAT
160 170
TAAACGAAAA ACGTCATAAA
220 230
CTTCACCAGA CGAAACCGAT
280 290
CAGCCCAAAA TTTAGTGCAT
340 350
AGCGGNACAG CGTCTAAAAG
400 410

ACATCCACAA ATGGCATTTA

60
ATGATCAGCA
120
ACGCTTGAAT
180
GTGAGTGAAC
240
ACAGAGCTAG
300
GATGGCGGCA
360
ATTTCTGGCC
420
TGCGTNTGAT

FILE NAME

SEQUENCE

7868P;

ID14.DNA

*** SEQUENCE LIST ***

SI

10
CCGACACTGG
70
ATCTGGTGAT
130
TGAGCCCCTG
190
CTATGGATGG
250
CTTCGGATTG
310
AGTAATGTCC
370
TCAGATGCAT
430
CAGGCTGTTG
490
GAGCATACAG
550
GGTTCTGATC
610
TCAAAATCCA
670
ATGACAACTC
730
GGAGAAGCAG

ATCGCT 3'

20
CATCTAGCGT
80
GCTGNTGGAA
140
GATTTGTGAG
200
TNGAACTGCC
260
AAATTGATCA
320
TTACCGTTGC
380
CGACCGATAT
440
TTAAAAAACA
500
GTTTTGCAGT
560
TGGACCTGGT
620
TCACAGGTTT
680
AAACCTTGGA
740
GACTACTGGT

Figure A-l4: Sequence ID#14

202 A;

169

155

30
NTGATGTCCT
90
AGTCGAGGTG
150
GAGCTAACGC
210
ACAGCGNAAG
270
GGTCGAAGAC
330
TGCAAGTGAT
390
TGCTGAAGTC
450
TGGTTATCCC
510
GATTGGTTAT
570
TTTTATTCAC
630
TGAATTTGCC
690
TGGTCGTGTG
750
AACAAGTNTA

C;

(SINGLE)

40
TGATTGAATC
100
CCATGCAGCG
160
AATATCCTGT
220
GATCTGGAAG
280
CAGATGCGTG
340
GTTTTAGCCG
400
AGTGTCGCAG
460
AAAGATGCCA
520
GGGAAGCTGG
S80
TATTTTGATG
640
ATGCGTGTGG
700
TATGAAATTG
760
ANAGCATTTG

198 G;

50
TGCCATGCGN
110
ACTGGTCAAA
170
GTTACTGGAT
230
ATTCTTTACG
290
TACTGCGGCT
350
AAAGTCCTCT
410
CAACCTTAAA
470
GTGGTGAACG
S30
GGGGGATTGA
590
AACAGGCCGA
650
NACAAAAGTT
710
ATACTCGACT
770
ACACCACCAC

222 T.

60
CGCACTGTAT
120
ATGGCCACCG
180
GAATTTCTTT
240
TCAGCAGCTT
300
GTTTAAGAAA
360
CATGAAAGTT
420
TTTGGCCTAT
480
TTGCTCACTG
540
GCTGGGTTAT
600
AACAGATGGA
660
TTTATCACTC
720
TCGACCATCT
780
TCAAAAGTNC

170

.5: memoeoscem 05 E 55026 2: Beam $58.2. damn .«o mwzcoo wfioaesaem Hn_-< oaswE

Eman— omem 82
Max: oozes—Sow

A

A

‘P
A
T

 

Ease. Boom
Na ooeoscom

‘)

Guam 8mm 33
IR: eocezcem

171

 

A
V

 

rrrr
(r4

 

 

Sequence ID#4 Sequence ID#5
341 Base Pairs (511 Base Pairs)

 

V

 

Sequence ID#6
(351 Base Pairs)

 

>
‘ ‘
r r

500

—

Sequence ID#7
(952 Base Pairs)

L

r
L
_'

500 >
Sequence ID#8 Sequence ID#9

(680 Base Pairs) (353 Base Pairs)

 

L
'

 

 

SOObp

Figure A-16: Sequencing contigs of pSR2. Arrows show the direction of
the sequencing run.

172

.5: @6533 2: .8 actuate 05 305 $65 ._ Ema we mwzcoo wemoeoacom “S -< oSwE

GEE ommm 30
SEA: eoeoscom

Epsom

 

A

25 8am msé
Gin: ooeoswom

l—lllJlqul]

88 82 08—

i
‘

 

 

 

A§

A4}
A A

k A
‘ A

 

A

L

‘
t
l

A

 

A

 

t

l
} r
‘ l

173

L
r

 

' 500 1000 a

__L_____I—

 

‘
7

Sequence ID#12
(1300 Base Pairs)

 

L
r

 

‘
'

 

Sequence ID#13
(447 Base Pairs)

L
r

 

L L
V 7

500

Sequence ID#14
(786 Base Pairs)

SOObp

Figure A-18: Sequencing contigs of pSR12. Arrows show the direction of
the sequencing run.

APPENDIX B

APPENDIX B

CONSTRUCTION OF A TRANSCRIPTIONAL EXPRESSION VECTOR FOR A.
CALCOACETIC US AND E. COLI

During the course of the examination of the wax ester and triacylglycerol
mutants, a possibility existed that the gene or genes of interest might reside in an
operon. To ascertain which of the genes had been affected by the chemical
mutagenesis, and which was needed to complement the mutant phenotype, it would be
necessary to express the genes outside of the operon. This would necessitate removing
the gene from the environment of its promoter. Therefore to express the isolated gene,
it would need to be expressed in some sort of transcriptional expression vector that
was functional in A. calcoaceticus. Therefore, it was decided to construct such a
vector using pSER2, a pBR322 derivative that is able to replicate in both A.
calcoaceticus and E. coli. This appendix highlights the construction of pSERl and
pSER2 and their subsequent use in constructing pSER200-1 and pSER200-4. Table B-
1 lists the plasmids used for this purpose.

An A. calcoaceticus/E. coli shuttle vector, pWH1274, was previously
constructed by Hunger et. al. (Hunger et al., 1990). It is a pBR322 derivative
containing ampicillin and tetracycline resistance markers and an A. calcoaceticus origin

of replication in addition to the origin of replication fi'om E. coli (Figure B-l).

174

175

Table B-l: Plasmid sources and derivations for Appendix B.

Plasmid

Description or
Construction

Source or Reference

 

pWH1274

pSERl

pSER2

pUC4k
pUC l 9

pPRl3

pSER100

pSERl 10

pSER120-1

pSER120-4

pSER200-l

pSER200-4

A. calcoaceticus/E. coli
shuttle vector. pBR322
derivative containing A.

calcoaceticus origin of
replication (Tet'Amp’)

A. calcoaceticus/E. coli
shuttle vector (Tet'Km')

A. calcoaceticus/E. coli
shuttle vector (Tet'Km')

Kmr cloning vector

Source of MCS to give
pSER200 derivatives

Reporter construct
containing full length
lacZYA cassette

pSER2 derivative with
the EcoRI site filled in
to destroy it

pSER100 derivative
containing lacZYA
cassette from pPR13

pSER110 derivative
containing ~1200 bp
insertion homologous to
pntA

pSER110 derivative
containing ~500 bp
insertion with homology
to dihydrofolate
reductase

pSER120-l derivative
with lacZYA removed

pSER120-4 derivative
with lacZYA removed

(Hunger et al., 1990)

Figure B-l, this study

Figure B-1, this study

Dr. Lee McIntosh
New England Biolabs

Dr. Frans DeBruijn
(Ratet et al., 1988)
Figure B-2, this study

Figure B-2, this study

Figure B-2, this study

Figure B-2, this study

Figure B-5, this study

Figure B-8, this study

 

Figure B-l: Restriction maps of pBR322, pWH1274, pSERl and pSER2. The original
A. calcoaceticus/E. coli shuttle vector was derived from pBR322 (top) by the addition

of an A. calcoaceticus origin of replication to give pWH1274 (middle). The plasmids at
the bottom of the figure, pSERl and pSER2 are derivatives of pWH1274 that contain the
Km' cassette from pUC4K (probably derived from Tn903). The location of several
restriction sites are listed by their positions in kb.

177

One problem with this vector is that A. calcoaceticus is naturally resistant to
ampicillin, making selection with that particular marker difficult in this species.
Therefore, it was decided to modify pWH1274 by inserting a kanamycin resistance
marker taken as a Pstl fragment from pUC4k into the Pstl site of the ampicillin
cassette of pWH1274, thus inactivating it. The kanamycin resistance marker fi'om
pUC4k is probably derived from Tn903 as evidenced by the lack of a Bglll site and
the locations of the HindIII and ClaI sites. The new vectors containing the kanamycin
cassette in two different orientations were called pSERl and pSER2 (Figure B-l).

pSER2 was then used to construct the transcriptional expression vector in the
following manner. pSER2 was digested with EcoRI. The site was filled in using
Klenow enzyme and dNTPs and the vector recircularized to produce pSER100. The
full length lac cassette (containing lacZYA) was then removed from pPR13 (a kind gift
from Dr. Frans DeBruijn) via a BamHI+SalI digestion (Ratet et al., 1988). This lac
cassette was then directionally cloned into pSER100, removing the majority of the Tet'
cassette, creating pSERl 10 (Figure B-Z). This provided a unique BamHI site upstream
of the lac cassette for the insertion of random fragments of DNA that might contain
promoter activity. This promoter activity could be detected by the presence of B-
galactosidase activity in A. calcoaceticus. SauIIIA partials of the cosmid 1A-3F (the
cosmid that complements the wowl phenotype) were carried out, and fragments of
approximately 500 bp were size selected. The reason that it was decided to obtain a
promoter element from this cosmid, as opposed to something from the genome, was in
the hope of enriching for a promoter that would be inducible under low nitrogen

conditions. This would allow the expression of the gene of interest under the same

178

.=2§:u:o C8 seam 2m 83. coco—52 582 6553838 .< E 8: C8 €082,
5.32me Becca—Comma: Toommmwmm 28 fl -oommmmm 8.5280 9 :83 an; 2: Co 5:852: .m- m oSmE

128$me
e 782mg

 

=~m 55mm 28m

.2553: <_:_.:..m

G m UDQ we pap—bud. 2.: CO.“

2; 5.5m a £2.63 2

H HRH U BREE QB 203a EC;
82 SE 8:83 :8
_=Emm geom— +E8m :a E 52528

gamma Soc 8552
v.63 ozwmmmu (CfiNodA

Toflmmma
2 Emma S TONEmma
H1} Iv El "Hung
- o
gamma 95
£38826
3:85
a... 8:. 8558 Emma
Seawwmm HI]! All HulJ’
a ESE 82mm; 2;» 2 a a 28m
a = m 58:... 235.com = m is m
2MB

[El

:8 28m EEam

179

conditions that the were needed for expression of the rest of the operon, all of which
would presumably be induced under low nitrogen conditions.

The resulting products containing the SauIIIA partials were transformed into A.
calcoaceticus which was then plated onto medium containing 15 pig/ml Tetracycline
and X-gal (40 ug/ml). Seven blue colonies were identified. These seven were
replated onto low nitrogen selective medium containing X-gal and IPTG. Two of the
seven seemed to be more heavily stained when grown under low nitrogen conditions,
than when plated on LB. No colonies were identified that turned blue under low
nitrogen conditions, that had previously been blue on LB. The two different plasmid
constructs (presumably different because of their promoter fragments) were called
pSER120-1 and pSER120—4 (Figure B-2).

The final products, pSER200-l and pSER200-4 derived from pSER120-l and
pSER120-4 respectively, were the result of digesting the samples with EcoRI+SalI to
remove the lac cassette and replace it with a polylinker from pUC19. This allowed
the presence of a unique BamHI site for cloning foreign fragments behind the
promoter element. Restriction analysis of the pSER200 constructs indicated that the
promoter element located in pSER200-l is approximately 500 bp, while the one
present in pSER200-4 is roughly 1200 bp.

To ascertain the relative strengths of the promoters, assays were conducted to
measure the amount of B-galactosidase in the cells. Activity was measured in extracts
of cells grown in minimal medium under both high and low nitrogen levels.
Additionally, the constructs were transformed into E. coli strain DHSa and [3-

galactosidase levels were measured for transforrnants cultured in LB medium. Assays

180
were carried out in the manner described by J. H. Miller (Miller, 1992). The number

of units of B-galactosidase were calculated based on the absorbance reading of the
sample at 420 nm resulting from the cleavage of OPNG and are presented in Table B-
2.

From these data we see that wild type A. calcoaceticus strain BD413 does not
exhibit any B-galactosidase activity. As a control, A. calcoaceticus was transformed
with pSERl 10, the construct containing the lac cassette without any sort of promoter
element. In this transforrnant, B-galactosidase levels are only slightly higher.
Transformation and introduction of pSER120-1 into A. calcoaceticus resulted in the
presence of 1817 units of B-galactosidase. A. calcoaceticus transformants containing
pSER120-4 have approximately half the levels of pSER120-1 constructs at 950 units of
B-galactosidase. Neither sample was induced to higher levels of B-galactosidase when
grown under low nitrogen conditions, contradicting what appeared like a marked
increase when the colonies were inspected visually on plates. Rather, there appears to
be an approximately 10-20% decrease in the amount of B—galactosidase present in the
cells.

Examination of the constructs in E. coli shows them to still be active, but at
greatly different levels. Because of the genotype of the E. coli strain, it was not
convenient to measure the amount of B-galactosidase under low and high nitrogen
conditions, rather measurements were made afier incubating the cultures in LB
medium. Introduction of pSER120-l in E. coli results in an almost 10 fold increase in
B-galactosidase production, while there is a about a six fold decrease in levels of B-

galactosidase production when pSER120-4 is expressed in E. coli.

181

Table B-2: B-Galactasidase content of cells transformed with pSER120 constructs.

 

Sample
A. calcoaceticus strain BD413
High nitrogen conditions
Wild type
Wild type with pSERl 10
Wild type with pSER120-1
Wild type with pSER120-4
Low nitrogen conditions
Wild type
Wild type with pSERl 10
Wild type with pSER120-1
Wild type with pSER120-4
E. coli strain DHSa
Wild type
Wild type with pSERl 10
Wild type with pSER120-l
Wild type with pSER120-4

Units of B-Galactasidase

3.13
13.65
1817.03
950.48

7.37
15.03
1600.54
759.77

27.96
39.19
10127.3
167.78

 

182

To get some sense of the origin of subcloned promoter elements, a single
stranded sequence read into both ends of the promoter element was performed. The
promoter of pSER200-1 was found to have 100% sequence identity with a pyridine
nucleotide transhydrogenase (pntA) from E. coli (Figure B-3). The only explanation
for this result is that the cosmid DNA that was partially digested and used to identify
promoter elements must have been contaminated with E. coli genomic DNA. This is
easily explainable since cosmid DNA was routinely prepared using Promega's Wizard
DNA preparation kits. These kits employ a resin that the DNA binds to and is later
removed from via salt exchange. Because the cosmid DNA was routinely extracted
from E. coli strain HB101, where it is present in a very low copy number, it is likely
that there was sufficient room on the column for genomic DNA fragments to bind,
resulting in contamination of the DNA sample. Although this is rather unfortunate and
sloppy, it still does not detract from the fact that this promoter element is completely
active in A. calcoaceticus. It also serves to explain the 10 fold increase in B-
galactosidase accumulation when the construct is introduced into E. coli in comparison
to A. calcoaceticus. The known sequence of the upstream region of pntA starts
approximately 200 bp upstream of the lac cassette. Additionally, the gene is oriented
in such a way that it is being transcribed in the opposite direction to the lac cassette.
This implies that the transcriptional activity of the subcloned fragment resides in the
200 bp just upstream of the lac cassette and is oriented away from the pntA gene
toward the cassette. Possible promoter elements are highlighted in Figure B-4.
Finally, the best known restriction map for pSER200-1 is illustrated in Figure B-S.

DNA sequencing of the promoter element in pSER200-4, found it to be

183

FILE NAME : 200-1—5'.DNA
SEQUENCE 347BP; 70 A; 103 C; 86 G; 72 T.
*** SEQUENCE LIST *** (SINGLE)
10 20 30 40 50 60
5' GATCCCCAGC GCGCCATCGC TATTGAATTC CGCGCCCATA CTTTGAACTT GTTCTTTCAC
70 80 90 100 110 120
TTCCGGGNGG GTGTCGAATG CACGCACAAT CGCGCCGAGA CTGNTTGCTG CGCCAATGGC
130 140 150 160 170 180
GGCCAGACCT GCAACACCCG YACCAATCAC CATCACTTTT GCCGGTGGCA CTTTCCCGGN
190 200 210 220 230 240
CGCAGTAATT TGCCCGGTWA AGAAGCGCCC AAATTCATGT GCCGCTTCAA CAATGGNGCG
250 260 270 280 290 300
ATAACCGGNG ATGTNCGCCA TCGAGCTTAG NGNGTCCAGC GNATTGTGCG NGTGANATAC
310 320 330 340
GCGGCACAGA GTCCATCGCC ANCACGGTCA NGTTANGTTC CGCAAGT 3'

Figure B-3: Sequence ID#IS. Sequence of the 5' region of the promoter cloned into
pSER200-1. This sequence was found to be 100% identical to pntA from E. coli.

FILE NAME : 200-1-3'.DNA
SEQUENCE 4148?; 117 A; 99 C; 85 G; 109 T.
*** SEQUENCE LIST tic (SINGLE)
10 20 30 40 50 60
5' CAACATAAAA AGAGGCGGCA CCTTCTTTAT CTGCGCGACG AGAAACGGCT TTCACCGCTT
70 80 90 100 110 120
CGCCAATAGC ATTAAAACGA CCGGTAACCA CTACACGGTC AAAAGGTTTA ACCGCTGCCG
130 140 150 160 170 180
CTTGCTCCGG TGTCAGTTCT GTTGCTGCGT TAACGGAGAA TGCCATAGCA GAAAGCAGTG
190 200 210 220 230 240
CCGACGCCAG GAGGGTGTTC TTAAGCTTCA TAAAAATAAT CCTTCGCCTT GCGCAAACCA
250 260 270 280 290 300
GGTACTGGTA TTGTTATTAA CGAGAAACGT GGCTGATTAT TGCATTTAAA CGGTGTAACT
310 320 330 340 350 360
GTCTGCGTCA TTTTTCATAT CACATTCCTT AAGCCAATTT TAATCCTGCT CAAATGACGN
370 380 390 400 410
CTATGCTTAA AAAACAGCCG NNTCAGCATC ATTACTACTG AAGCAACTGN ATTG 3'

Figure B-4: Sequence ID#16. Sequence of the 3' region of the promoter cloned into
pSER200-1. This sequence was found to be 100% identical to pntA from E. coli.
Base pair 209 through the end of the sequence represents pntA 5' untranslated region,
as it was deposited in GenBank. A possible -10 and -35 boxes are highlighted in bold

type-

184

 

 

H'ndll 0.02
s 9 0.19
‘9' °°° 50081 0.41
Seal 9.29 0.43
0.92
Pst18.04 m1 1.19
thII 1.29
cm 7.57 1.33
Smal 7.39 60081
Sect
Promoter KO”
H'ndlll 7.14 Sum
Km xw
Sal
pSER200-1
Styl 220
9.19 110
13906.54 9.11 227
om 8° om Hincll 2.90
Ndel 5.23
Acct 3.79
Accl 5.00
99'" “1 Ndel 3.91

Figure B-S: Best known restriction map of A. calcoaceticus/E. coli transcriptional
expression vector pSER200-1. The region between 0.62 kb and

1.19 kb was not sequenced so restriction sites in this area are currently unknown.
Features include Km' from the Km' cassette, probably derived from Tn903,
Acinetobacter and E. coli origins of replication, a promoter that is active in both
E. coli and Acinetobacter (10x higher expression in E. colt) and polylinker with
several possible cloning sites. DNA sequence information for the promoter region
of this vector can be found in this text as sequence ID#15 and ID#16.

185
strongly similar to dihydrofolate reductase (BLAST score of 175, Figure B-6), which

has been cloned from several different organisms. The sequence of this promoter
element is shown in Figure B-7. A strong blast score to dihydrofolate reductase starts
approximately 330 bp upstream of the lac cassette and continues up to the lac cassette.
This would most likely place the promoter element in approximately the first 170 bp
of the fragment, about 330 bp away from the lac cassette. Possible promoter elements
in this region are highlighted in Figure B-7. This explains the rather weak
accumulation of B-galactosidase in samples containing this promoter element.
However, the possibility of a cryptic promoter element in the coding sequence of
dihydrofolate reductase cannot be ruled out. Finally, the best known restriction map
for pSER200-4 is illustrated in Figure B-8.

Neither of these transcriptional expression vectors was modified any further. It
would be nice to streamline them by eliminating the excess sequence that is present in
both of the vectors. Approximately 1.1 kb could probably be cut out of pSER200-l
without any adverse affect. Similarly, the 300 bp of sequence that separates the
promoter from the potential cloning sites in pSER200-4 could also be removed. In
this case it might even improve transcriptional expression. Regardless of these minor
imperfections, both of these constructs are new and important tools that could be used

in studying A. calcoaceticus.

186

>sp|P12833|DYR3_SALTY DIHYDROFOLATE REDUCTASB TYPE III. >pir|S|RDEBDT
dihydrofolate reductase (EC 1.5.1.3) type III - Salmonella
typhimurium plasmid pAZl >gp|J03306|PAZDHFRA l Plasmid pAZl type
III dihydrofolate reductase gene, complete C39.
Length . 162

Minus Strand HSPS:

[Plasmid pAZl]

Score a 235 (109.3 bits), Expect - 2.6e-27, P - 2.6e-27
Identities = 42/79 (53%), Positives - 57/79 (72%), Prame - -1
Query: 297 VVAMDQKQCIGKGNALPWHIPADLKHFKEITQDGVVIMGRKTLESMGRTLPKRVNWVITR 118
+ A+ IGK N +PWH+PADL+HFK +T V+MGR+T BS+GR LP R N V++R
Sbjct: 6 IAALAHNNLIGKDNLIPWHLPADLRHFKAVTLGKPVVMGRRTFESIGRPLPGRRNVVVSR 65
Query: 117 DPNWQFEGAKVASSIEAAL 61
+P WQ EG +VA S++AAL
Sbjct: 66 NPQWQAEGVEVAPSLDAAL 84
Figure B-6: Results of BLASTX alignment using sequence ID#17 as a query

sequence. Based on this homology, the promoter element for pSER200-4 was found
to be derived from the promoter of dihydrofolate reductase from A. calcoaceticus strain

BD413.
FILE NAME 200-4-5'.DNA
SEQUENCE 47939; 129 A; 109 C; 113 G; 119 T.
*** SEQUENCE LIST *** (SINGLE)
10 20 3o 40 so 60
5' CGGAGCNNTA TCGACTACGC GATCATGGCG ANCACACCCG TCCTGTGGAT CTGTTTGATT
70 90 90 100 110 120
TTAAATTTGA AGATATTGAA ATTGTGGATT ATCAATCCCA CCCTGCAATC AAAGCCCCTG
130 140 150 160 170 190
TTGCCGTATA AGGAAGCTCA GCAATGGCAT TTCAAGATTT AGAAGTCGTG CATGTGGTTG
190 200 210 220 230 240
CGATGGATCA AAAACAATGT ATTGGTAAAG GCAATGCATT ACCTTGGCAC ATTCCTGCCG
250 260 270 290 290 300
ATCTCAAACA TTTTAAGGAA ATCACTCAAG ATGGCGTCGT GATTATGGGT CGTAAGACAC
310 320 330 340 350 360
TTGAATCCAT GGGGCGCACG CTCCCTAAAC GTGTCAACTG GGTGATTACC CGNGATCCTA
370 380 390 400 410 420
ATTGGCAGTT TGAAGGCGCC AAAGTTGCAT CCAGCATAGA AGCTGCACTT NAAAGGCGCA
430 440 450 460 470
GCTCAGGATC CSGNGAATTC GAGCTCGGTA CCCSGNGNAT CCTCTAGTAG AGTCGACC

Figure B-7: Sequence ID#17. DNA sequence of the promoter element used in the
construction of pSER200-4. Possible -10 and -35 boxes are shown in bold faced type.
Sequence similiarity to dihydrofolate reductase begins at base pair 172 and extends to
the end of the sequence.

187

 

 

 

Ml 0.02
0.18
3.91 goo 0.44
Seal 7.68 0'"
0.78
P911 7.44 80081
Sect
Kpnl
Clal 6.97 Smal
Promoter aunt-C
Salt
801016.78
Sty! 1.80
H'ndlll 6.54 Km Bell 1.67
pSER200-4
8.19 kb
Pstl 5.94
Hinctl 2.30
OR! Ac OR!
Ace! 118
No.1 3.31
Ndel 4.63
A001 4.40 891" 3.81

Figure B-8: Best known restriction map of A. calcoaceticus/E. coli transcriptional
expression vector pSER200-4. Features include Kmr from the Km' cassette,
probably derived from Tn903, Acinetobacter and E. coli origins of replication, a
promoter that is active in both E. coli and Acinetobacter and a polylinker with
several possible cloning sites. DNA sequence information for the promoter region
of this vector can be found in this text as sequence ID#17.

l 88
REFERENCES

Hunger, M., Schmucker, R., Kishan, V. and Hillen, W.. 1990. Analysis and
Nucleotide Sequence of an Origin of DNA Replication in Acinetobacter calcoaceticus
and its use for Escherichia coli Shuttle Plasmids. Gene, 87:45-51.

Miller, J. H.. 1992. Procedures for Working with lac. In A Sm Course in Bacterial
Genetic; A Laboratory Manual and Handbook for Escherichia coli and Related
Bacteria. Cold Spring Harbor Laboratory Press, Plainview, New York.

 

Ratet, P., Schell, J. and de Bruijn, F. J.. 1988. Mini-Mulac Transposons with Broad-
Host-Range Origins of Conjugal Transfer and Replication Designed for Gene
Regulation Studies in Rhizobiaceae. Gene 63:41-52.

APPENDIX C

APPENDIX C

CLONING OF A PARTIAL DNA FRAGMENT OF SN-l ACYLTRANSFERASE
FROM A. CALCOACETIC US

Another attempt at isolating DGAT (diacylglycerol acyltransferase) involved an
approach using degenerate oligonucleotides and PCR. It was our belief that DGAT
might be isolated by constructing degenerate primers based on consensus sequence
generated from other known acyltransferases. Several acyltransferases have been
isolated from a number of different organisms. A careful comparison of those
sequences was done using the GCG sequence analysis program by Dr. Joe Ogas. In
his pileup of the different sequences, Dr. Ogas was able to identify two different
region that were conserved among the acyltransferases. These are shown in Figure C-
1.

Degenerate oligos based on the conserved regions highlighted in Figure C-l
were constructed based on the preferred codon usage table published by White et. al.
(White et al., 1991). These primers and their degeneracy are listed in Table C-l. PCR
was carried out under the following conditions. A total of 100 ng of genomic DNA
from A. calcoaceticus strain BD413 was mixed with 50 pmols of each primer, 2.5 ul
of 2 mM dNTP's 2.5 1.11 of 10x buffer (660 mM Tris-HCl (pH 7.6), 50 mM MgC12, 10

mM dithioerythritol and 10 mM ATP (pH 7.5)) and 1.25 units of TAQ

189

1 50
A. calco. .................. HR WHMDYLLLSY VIYKRGLM.. VPYIAAGDNL
Yst-Snl-pl KNHPEIIKGR SKNPQTTPVN FTKRFSAKSL LGLPDYLS.. NAQIKEIPDD
Yst-Snl-pz K ........ R VIGHDT...H PLTDCMPKGL IGLPKSMG.. FGEIQSIBSD
E.Coli-Sn1 ERVRQLAHDG HELVYVPCHR SHMDYLLLSY VLYHQGLV.. PPHIAAGINL
MousMitSnl EMVKAATETN LPLLFLPVHR SHIDYLLLTF ILPCHNIK.. APYIASGNNL
E.Coli-Sn2 KPTDAESYGN A..IYIANHQ NNYDMVTASN IVQPPTVT.. ...VGKKSLL
Yst-Snz-pl KVVGEENLAK KPYIMIANHQ STLDIFMLGR IFPPGCTV.. ...TAKKSLK
Maize—Snz RSMGKEHA.. ...LIISNHR SDIDWLIGWI LAQRSGCLGS TLAVMKKSSK
Yst-SnZ-pz LSHLKSNS.. ...VAICNHQ IYTDWIFLWW LAYTSNLGAN VPIILKKSLA
Consensus HR SHMDY

$1 100
A. calco. NLPFVGQLLR GGGAFFIRRS PRGNGLYTSV PK ..................
Yst-Snl-pl ETIILSSPFR TSKSKVVBLL TNGTNPKYAB KI ..................
Yst-Snl-pz TSLTLRKEFK MAKPEIKTAL LTGTTYKYAA XV ..................
E.Coli-Sn1 NFWPAGPIFR RLGAFFIRRT PKG ...... N KL ..................
MousMitSnl NIPVFSTLIH KLGGFFIRRR LDBTPDGRKD IL ..................
E.Coli-Sn2 WIPFFGQLYW LTGNLLIDRN NRTKAHGTIA 3V ..................
Yst-SnZ-pl YVPFLGWPMA LSGTYFLDRS KRQEAIDTLN KG ...................
Maize-Snz FLPVIGWSMW FAEYLFLERS WAKDEK.TLK WGLQ ................
Yst-Sn2-p2 SIPILGFGMR NYNFIFMSRK NAQDKI.TLS NSLAGLDSNA RGAGSLAGKS

101 150
A. calco. ............... BYLYS ILSRNTPLEY PPBGTRS... ..........
Yst-Snl-pl ......... D NTETFQSVFD HLHTKGCVGI PPEGGSHDRP SL. LPIKAG
Yst-SnI-pz ......... D QSCVYHRVFE HLAHNNCIGI FPEGGSHDRT NL. LPLKAG
E.Coli-Sn1 ......... Y STVPRBYLGB LPSRGYSVEY FVEGGRSRTG RL. LDPKTG
MousMitSnl ......... Y RALLHGHVVB LLRQQQFLBI FLEGTRSRSG KT. SCARAG
E.Coli-Sn2 ................. VNH FKKRRISIWM FPEGTRSRGR GL. LPFKTG
Yst-SnZ-pl ................. LEN VKKNKRALWV PPEGTRSYTS BLTMLPPKKG
Maize-Snz ................. RLK DFPRPPWLAL PVBGTR.... PTPAKLL
Yst-Sn2-p2 PERITEEGES IWNPBVIDPK QIHWPYNLIL PPBGTN.... LSADTRQ
Consensus PPBGTRS

151 199
A. calco. .................................................
Yst-Snl-pl VAIMALGAVA ADPTMKVAVV PCGLRYPHRN KPRSRAVLEY GBPIVVDGK
Yst-Snl-p2 VAIMALGCMD KHPDVNVXIV PCGMNYPHPH KPRSRAVVRP GDPIBIPKB
E.coli-Sn1 TLSMTIQAML RGGTRPITLI PIYIGYE... ..... HVMEV G...TYAKB
MousMitSnl VLSVVVNTLS SNTIPDILVI PVGISYD... ..... RIIB. G...HYNGB
E.Coli-Sn2 AFHAAIAAGV P ....... II PVCVSTTSNK I..NLNRLHN G...LVIVB
Yst-SnZ-pl AFHLAQQGKI P ....... IV PVVVSNTSTL VSPKYGVPNR G...CNIVR
Maize-Snz AAQE.YAASQ GLPAPRNVLI PRTKG.FVSA VS .................
Yst-SnZ-pz KSAK.YAAKI GKKPPKNVLL PHSTGLRYSL QKLKPSI.BS LYDITIGYS

190

Figure C-l: Pileup of known and putative acyltransferases. Regions used to construct
the degenerate oligos are highlighted with the predicted consensus sequence underneath
the pileup. The consensus sequence used for the construction of oligo 3'-3 is actually
5'-F(P,V,L)EG(T,G)RS-3'. The DNA sequence of the cloned fragment is illustrated
above. The sequences used for the pile up are: (A. calco.), the determined A.
calcocaceticus sequence appearing in sequence ID#18 , (Y st-SnI-pl), a putative Yeast
Snl-acyltransferase, (Yst—Snl-pZ), another putative Yeast Snl-acyltransferase, (E.coli-
Snl), an E. coli Snl-acyltransferase, (MousMitSnl), Mouse mitochondrial Snl-
acyltransferase, (E.coli-Sn2), E. coli Sn2-acyltransferase, (Y st-Sn2-p1), Yeast Sn2-
acyltransferase, (Maize-Sn2), Maize Sn2-acy1transferase and (Yst-Sn2-p2), another
Yeast Sn2-acyltransferase. Alignment was constructed by Dr. Joe Ogas using the
PILEUP program from the GCG software package.

191

Table C-l: Synthetic degenerate oligonucleotides used in Appendix C.

 

Name 01' Sequence (5'-3')' Degenerac
primer y

 

Primer 52-1 CA(T/C) CG(T/C/A) (T/A)(C/G)N CA(T/C) ATG 768
GA(T/C) TA(T/C)

Primer 3'-3 N(G/C)(A/T) (A/G/T)CG N(G/C)(T/C) 221134
(A/G/T)CC (T/C)’I‘C N(G/A)(G/C/A) (A/G)AA

 

FBases in parenthesis represent degeneracy at that position

192

polymerase (Boehringer Mannheim) in a total volume of 25 1.11. The reaction was
cycled under the following conditions:

Step 1: 94°C for 1 minute

Step 2: 65°C for 1 minute -1°C per cycle

Step 3: 72°C for 2 minutes

Step 4: Repeat steps 1-3 ten times

Step 5: 94°C for 1 minute

Step 6: 55°C for 1 minute

Step 7: 72°C for 2 minutes

Step 8: Repeat steps 5-7 30 times

Step 9: 72°C for 5 minutes

Step 10: hold at 4°C

Following the PCR reaction the products were separated on a 1.2% agarose gel
and visualized by staining using ethidium bromide. An expected band of
approximately 250 bp was observed. This DNA was removed from the agarose by
electroelution and subcloned into pGEM-T to produce the construct p51-33. The
subcloned region was sequenced and is shown in Figure C-2. BLASTX alignment of
this sequence against the GenBank database indicated a strong (BLASTX score of 255
with a probability score of 2.2x10'", 51 out of 84 (60%) amino acids were identical)
similarity to sn-glycerol-3-phosphate acyltransferase from E. coli (accession number
K00127) (Figure C-3) (Altschul et al., 1990).
At this point the experiment was drawn to a close. The probability of success

was highly doubtful since it appeared that the degenerate oligos that were synthesized,

193

*** INPUT INFORMATION ***

FILE NAME : ACYL.DNA

SEQUENCE : 252BP; 51 A; 56 C; 57 G; 88 T.
*** SEQUENCE LIST *** (SINGLE)
10 20 30 40 50 60
5' CATCGCTGGC ATATGGATTA TTTGCTGTTG TCCTATGTCA TTTACAAACG CGGCTTGATG
70 80 90 100 110 120
GTTCCGTACA TTGCAGCGGG TGACAATCTT AACTTGCCAT TCGTTGGTCA GCTATTGCGT
130 140 150 160 170 180
GGTGGTGGTG CATTCTTCAT TCGACGTTCT TTCCGTGGTA ATGGCTTATA TACTTCGGTT
190 200 210 220 230 240
TTTAAAGAAT ATCTATACAG TATTTTGTCA CGTAACACGC CGCTTGAATA TTTCCCCGAG
250

GGCACACGCT CC 3'

Figure C-2: Sequence ID#18. Sequenced region of p51-33 that was found to contain
homology to Sn-l acyltransferase from E. coli.

>gp|K00127|ECOPLSB_2 sn-glycerol-3-phosphate acyltransferase [Escherichia coli]
Length 2 807

Plus Strand HSPs:

Score = 255 (117.3 bits), Expect - 2.2e-28, P . 2.2e-28
Identities . 51/84 (60%), Positives = 63/84 (75%), Frame . +1

Query: 1 HRWHMDYLLLSYVIYKRGLMVPYIAAGDNLNLPFVGQLLRGGGAFFIRESFRGNGLYTSV 180
HR HMDYLLLSYV+Y +GL+ P+IAAG NLN G + R GAFFIRR+F+GN LY++V
Sbjct: 306 HRSHMDYLLLSYVLYHQGLVPPHIAAGINLNFWPAGPIFRRLGAFFIRRTFKGNKLYSTV 365

Query: 181 FKEYLYSILSRNTPLEYFPEGTRS 252
F+EYL + SR +EYF EG RS
Sbjct: 366 FREYLGELFSRGYSVEYFVEGGRS 389

Figure C-3: BLASTX alignment between a putative acyltransferase from A.
calcoaceticus and Sn-l acyltransferase from E. coli. Sn-l acyltransferase from A.
calcoaceticus is labelled as Query and the Sn-l acyltransferase from E. coli is the
sequence labelled Sbjct.

194

would most likely result in the amplification of known acyl-transferases. This might
have been predicted since it was acyl-transferase sequences that the consensus
sequences were based on. Therefore subsequent experiments in this direction were

abandoned.

REFERENCES

Altschul, S. F., Gish, W., Miller, W., Myers, E. W. and Lipman, D. J.. 1990. Basic
Local Alignment Search Tool. J. Mol. Biol. 215: 403-410.

White, P. J., Hunter, I. S., and Fewson, C. A.. Codon Usage in Acinetobacter
Structural Genes. In The BiolgLOf Acinetobacter: Taxonomy, Clinical Immrtance,
Molecular Biolggy. Physiology, Industrial Relevance. Plenum Press, New York,'New
York. 1991. Eds. Towner, K. J., Bergogne-Bérézin, E., and C. A. Fewson.

HICHIGAN STATE UNIV. LIBRARIES
(HI WI M llll (I ”W I'll ll ”11 WW W “H l llWl
31293013904408