ENGINEERING ACTINOBACILLUS SUCCINOGENES FOR SUCCINATE
PRODUCTION- A FOCUS ON SUCCINATE TRANSPORTERS AND SMALL RNAs
By
Rajasi Virendra Joshi

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Microbiology and Molecular Genetics- Doctor of Philosophy
2017

ABSTRACT
ENGINEERING ACTINOBACILLUS SUCCINOGENES FOR SUCCINATE
PRODUCTION- A FOCUS ON SUCCINATE TRANSPORTERS AND SMALL RNAS
By
Rajasi Virendra Joshi
An important aspect of any industrial scale bio-based production is the choice of
biocatalyst used. Many commercially relevant microorganisms and industrial strains have been
engineered to optimize the production of bio-based chemicals. One such chemical is succinate,
listed as one of the top 12 building block chemicals from biomass by the US Department of
Energy. Succinate is considered an important platform chemical as it has a number of
applications and, most importantly, is a precursor to high-volume value-added commodity
chemicals. Bio-based succinate is currently being produced at industrial scale levels using
engineered microorganisms such as E. coli and S. cerevisiae. Actinobacillus succinogenes is one
of the best natural succinate producers, which can grow on a wide variety of substrates, and, with
the advance in genetic tools, can possibly be engineered for increased succinate production.
Very few studies have focused on using succinate exporters as metabolic engineering
targets for succinate production. Only a handful of studies have been carried out in E. coli and C.
glutamicum with none whatsoever in A. succinogenes. With a combination of proteomics and
transcriptomics we have identified candidate succinate transporters in A. succinogenes. Four of
the top hits in our proteomics analysis were Asuc_1999, Asuc_0142, Asuc_2058 and
Asuc_1990-91. To carefully tune the expression of these membrane proteins, we generated a
library of promoters covering a large range of strengths below the strong, constitutive promoter
(ppckA) we had been using. Some of the promoters were truncated versions of ppckA, and others
were identified from our transcriptomics data. These promoters were tested using lacZ as the

reporter gene in an A. succinogenes ∆lacZ background. Promoters ranged from ppckA as the
highest down to pAsuc_0701 with a strength 209-fold lower than ppckA. The four succinate
transporter candidates over-expressed under ppckA-92, a truncated version of ppckA, increased the
succinate yield in glucose cultures compared to the control strain carrying the empty vector.
Synthetic small RNAs (sRNAs) are another tool for metabolically engineering industrially
relevant microorganisms. However, no sRNAs have been identified in A. succinogenes and only
a few have been identified in other members of the Pasteurellaceae family. We identified sRNAs
in A. succinogenes grown anaerobically on glucose and microaerobically on glycerol by RNA
sequencing. We found 260 sRNAs in total, of which 39 were predicted by at least one of five
computational programs. We validated 14 sRNAs identified from sequencing with RT-PCR.
Additionally we identified probable Hfq-binding sRNAs through their Rho-independent
terminators, a key feature of Hfq-binding sRNAs. Using additional characteristics of Hfqbinding sRNAs, we designed synthetic sRNAs targeting lacZ expression as a proof of concept.
One plasmid-borne synthetic sRNA caused a 32% decrease in β-galactosidase activity in lactosegrown cultures. Using the same sRNAs scaffolds, we generated synthetic sRNAs that targeted
the ackA and pta mRNAs to decrease the production of acetate, one of the major by-products of
succinate production. One of the synthetic sRNAs targeting ackA caused a 14% decrease in the
acetate yield of glucose-grown cultures.
In summary, we have identified candidate succinate transporters and seen an increase in
succinate production upon their overexpression. In the process, we have developed a promoter
library for tunable expression of genes in A. succinogenes. We have also shown that sRNAs can
be used as a tool for metabolic engineering in A. succinogenes, although additional studies are
needed to make it more tunable and robust.

This dissertation is dedicated to my entire family, especially my
husband, Ashwin for his patience, constant encouragement and
morale boosting pep talks. I thank my parents, sister, brother-in
law and in-laws for always believing in me and supporting me
throughout this journey.

iv

ACKNOWLEDGEMENTS

This dissertation has been possible only because of the contributions of several people who
have guided and helped me throughout these years. First and foremost, I would like to express
my sincere gratitude to my mentor, Dr. Claire Vieille for giving me the opportunity to work in
her lab during my Master’s and for my PhD. I thank her for being a great mentor and for always
taking out the time to discuss any idea, big or small, despite being busy with numerous other
things. Her intellect, passion for science and positive attitude has always inspired and motivated
me, especially during the more trying times of this project. I will always be indebted to her, for
her guidance and thoughtful input throughout my PhD.
I am thankful to all the members of the Vieille lab- Maris Laiveneiks, Dr. Bryan
Schindler, Dr. Justin Beauchamp, Dr. Nikolas Mcpherson and Peggy Wolf for a great work
atmosphere with lots of support and help. I am truly grateful for the time taken by Dr. Bryan
Schindler to train me during my initial years in the lab even when he was busy with his own
experiments. I am also grateful to Maris Laivenieks, who was always willing to help me,
especially during my initial struggles while building a chemostat and mixing gas tanks. I would
like to thank Dr. Nikolas McPherson for being an excellent lab member and for all the helpful
discussions we had while working with Actinobacillus succinogenes during the course of both of
our PhDs. I must acknowledge Dr. Justin Beauchamp for helping me with the intial RNAseq
analysis. I wish to thank Barry Tigner, who has saved my experiments numerous times, by fixing
last minute instrument fails.
I must thank my committee members, Drs. Gemma Reguera, Michael Bagdasarian,
Shannon Manning, Yair Shachar-Hill for their expert guidance and constructive feedback. I

v

would like to thank Dr. Donna Koslowsky for letting me use her lab and Laura Kirby, her grad
student for helping me set up Northern blots.
I would like to thank all the undergraduates who helped me with my project, Alex Yang,
Jean Kim, Maeva Bottex, Matthias Schulz, Kanupriya Tiwari, Joel Tencer, Jacob Gibson, Aaron
Walters, Jack Peleman and Corey Kennelly. It was a pleasure interacting with each of them and
training them has made me a better researcher.
I would like to thank Betty Miller, graduate secretary of Microbiology and Molecular
Genetics for helping me navigate through the paperwork required and other assistance
throughout these years.
I am grateful for the lasting friendships I have made here at MSU. I feel fortunate to have
had a strong support system in them, sharing the ups and downs of graduate life.

vi

TABLE OF CONTENTS

LIST OF TABLES .......................................................................................................................... x
LIST OF FIGURES ...................................................................................................................... xii
Chapter 1 Introduction .................................................................................................................... 1
1.1 Bio-based succinate production ............................................................................................ 2
1.2 Substrates and feedstocks for succinate production .............................................................. 4
1.3 Applications for succinate and its derivatives ....................................................................... 4
1.4 Succinate-producing microorganisms ................................................................................... 5
1.4.1 Native succinate producers ............................................................................................. 5
1.4.2 Engineered succinate producers ................................................................................... 12
1.5 C4-dicarboxylate transporters ............................................................................................. 21
1.5.1 E. coli C4-dicarboxylate transporters ........................................................................... 21
1.5.2 C4-dicarboxylate transporters in other succinate producers ........................................ 23
1.6 sRNAs as synthetic regulators for metabolic engineering .................................................. 24
1.7 Identification of small RNAs in Pasteurelleaceae ............................................................... 28
1.8 Introduction to chapters ....................................................................................................... 28
REFERENCES .......................................................................................................................... 31
Chapter 2 Identification of succinate transporters in Actinobacillus succinogenes ...................... 42
2.1 Abstract ............................................................................................................................... 43
2.2 Introduction ......................................................................................................................... 44
2.3 Materials and Methods ........................................................................................................ 47
2.3.1 Strains, media, and culture conditions. ......................................................................... 47
2.3.2 Plasmids, DNA manipulations, and electroporations. .................................................. 47
2.3.3 Preparation of A. succinogenes cytoplasmic membrane-enriched fractions for
proteomics analysis. .............................................................................................................. 49
2.3.4 Proteomic analysis. ....................................................................................................... 50
2.3.5 Total RNA purification and RNA sequencing ............................................................. 52
2.3.6 Identification of A. succinogenes’s pckA transcription start site. ................................. 53
2.3.7 Construction of truncated pckA promoters. .................................................................. 53
2.3.8 Construction of an A. succinogenes ΔlacZ mutant strain. ............................................ 54
2.3.9 Preparation of crude extracts, β-galactosidase assays, and protein assays. .................. 55
2.3.10 Construction of a promoter library based on RNAseq results. ................................... 55
2.3.11 Construction of strains expressing putative succinate transporters. ........................... 56
2.3.12 High-performance liquid chromatography (HPLC) analysis of fermentation media
from strains expressing transporters. ..................................................................................... 57
2.4 Results and Discussion ........................................................................................................ 57
2.4.1 Potential C4-dicarboxylate exporters in A. succinogenes ............................................. 57
2.4.2 Proteomic analysis of A. succinogenes cytoplasmic membrane-enriched fractions .... 59
2.4.3 Transcript levels of A. succinogenes putative C4-dicarboxylate transporters .............. 62
2.4.4 Construction of an expression reporter system. ........................................................... 64

vii

2.4.5 Identification of ppckA’s TSS by RLM-RACE and construction of truncated ppckA
promoters ............................................................................................................................... 65
2.4.6 Identification of promoters of different strengths from the transcriptomics results. .... 67
2.4.7 Overexpression of putative succinate transporter candidates in A. succinogenes and
succinate production. ............................................................................................................. 68
2.4.8 The A. succinogenes succinate efflux transporters differ from those of E. coli and C.
glutamicum ............................................................................................................................ 71
2.5 Conclusion........................................................................................................................... 73
2.6 Acknowledgements ............................................................................................................. 73
APPENDIX ............................................................................................................................... 74
REFERENCES .......................................................................................................................... 78
Chapter 3 Identification of sRNA in Actinobacillus succinogenes 130Z ..................................... 82
3.1 Abstract ............................................................................................................................... 83
3.1 Introduction ......................................................................................................................... 84
3.2 Materials and Methods ........................................................................................................ 87
3.2.1 Bacterial strains and growth conditions. ...................................................................... 87
3.2.2 RNA isolation, library preparation, and sequencing. ................................................... 89
3.2.3 Trimming of reads, mapping, and sRNA identification. .............................................. 89
3.2.4 Small RNA analysis. .................................................................................................... 90
3.2.5 Computational prediction of sRNAs. ........................................................................... 91
3.2.6 RT-PCR validation of small RNAs. ............................................................................. 92
3.2.7 Construction of synthetic small RNA constructs. ........................................................ 92
3.2.8 Beta-galactosidase assays. ............................................................................................ 93
3.2.9 Determination of fermentation balances. ..................................................................... 94
3.3 Results and Discussion ........................................................................................................ 95
3.3.1 Detection of sRNAs from RNAseq data using RockHopper. ...................................... 95
3.3.2 Comparison of RNAseq-identified to computationally predicted sRNAs. ................ 104
3.3.3 sRNAs with Rho-independent terminators. ................................................................ 106
3.3.4 RT-PCR validation of small RNAs. ........................................................................... 106
3.3.5 Scaffold selection for synthetic sRNA design based on predicted Hfq-dependent small
RNAs. .................................................................................................................................. 110
3.3.6 Testing of synthetic sRNA designs with lacZ as target gene. .................................... 111
3.3.7 Synthetic sRNAs for inhibiting acetate production. ................................................... 114
3.4 Conclusion......................................................................................................................... 117
3.5 Acknowledgements ........................................................................................................... 117
APPENDIX ............................................................................................................................. 119
REFERENCES ........................................................................................................................ 131
Chapter 4 Development of a markerless knockout method for Actinobacillus succinogenes .... 137
4.1 Abstract ............................................................................................................................. 140
4.2 Introduction ....................................................................................................................... 141
4.3 Materials and Methods ...................................................................................................... 144
4.3.1 Strains, media, culture conditions, and chemicals. ..................................................... 144
4.3.2 Plasmids, DNA manipulations, and electroporations. ................................................ 146
4.3.3 Construction of plasmid pLGZ924. ............................................................................ 147

viii

4.3.4 Construction of Δfrd::icd and ΔpflB::icd mutants. ..................................................... 148
4.3.5 Construction of the ∆pflB ΔlacZ::icd, ∆pflB Δcit::icd, and ∆pflB Δacn::icd double
mutants. ............................................................................................................................... 149
4.3.6 Natural transformation................................................................................................ 150
4.3.7 Removal of the icd marker. ........................................................................................ 151
4.3.8 Curing of plasmid pCV933. ....................................................................................... 151
4.3.9 Enzyme assays. ........................................................................................................... 152
4.3.10 Analysis of fermentation products............................................................................ 153
4.4 Results ............................................................................................................................... 153
4.4.1 A positive selection method for recombination events in A. succinogenes. .............. 153
4.4.2 Natural transformation of A. succinogenes using a Δfrd::icd construct. .................... 154
4.4.3 Confirmation of the Δfrd::icd knockout strain. .......................................................... 155
4.4.4 Excision of the selection marker. ............................................................................... 156
4.4.5 Curing pCV933 from A. succinogenes Δfrd(pCV933). ............................................. 156
4.4.6 Construction of strain ΔpflB. ...................................................................................... 157
4.4.7 Effects of the DNA construct length and of the length of the homologous regions on
the efficiency of homologous recombination. ..................................................................... 159
4.4.8 Construction of double knockout mutants. ................................................................. 161
4.4.9 Development of a selective solid medium.................................................................. 163
4.4.10 Knockout mutants via electroporation...................................................................... 164
4.5 Discussion ......................................................................................................................... 165
4.6 Acknowledgements ........................................................................................................... 167
REFERENCES ........................................................................................................................ 172
Chapter 5 Respiratory glycerol metabolism of Actinobacillus succinogenes 130Z for succinate
production ................................................................................................................................... 176
Chapter 6 Conclusions and future directions .............................................................................. 178
6.1 Introduction ....................................................................................................................... 179
6.2 Succinate transporters in A. succinogenes......................................................................... 179
6.3 Identification of sRNAs in A. succinogenes...................................................................... 180
6.4 Future directions ................................................................................................................ 181
6.4.1 Experimental identification of Hfq-binding sRNAs................................................... 181
6.4.2 Continuing the development of synthetic sRNAs in A. succinogenes ....................... 181
6.4.3 Identifying the major succinate transporter(s) in A. succinogenes ............................. 182
6.4.4 Construction of a ∆ackA knockout mutant ..................................................................... 182
6.4.5 Combination of different approaches to increase succinate production ..................... 183
REFERENCES ........................................................................................................................... 184

ix

LIST OF TABLES

Table 1.1 Succinate-producing strains with their yields, productivities, and titers ...................... 20
Table 2.1 Strains and plasmids used in this study ........................................................................ 48
Table 2.2 A. succinogenes putative C4-dicarboxylate transporters .............................................. 58
Table 2.3 Proteomics and transcriptomics data for succinate transporters in A. succinogenes .... 61
Table 2.4 Fermentation balances of strain 130Z over-expressing candidate succinate transporters.
....................................................................................................................................................... 70
Table A2.1 Primers used in this study .......................................................................................... 75
Table A2.2 Transcript levels of 40 candidate genes with low variation across several conditions,
compared to pckA .......................................................................................................................... 77
Table 3.1 Strains and plasmids used in this study ........................................................................ 88
Table 3.2 Statistics on read alignments by Rockhopper ............................................................... 96
Table 3.3 A. succinogenes sRNAs detected by RNAseq and Rockhopper analysis in
anaerobically-grown glucose cultures and microaerobically-grown glycerol cultures ................ 98
Table 3.4 Rho-independent smRNAs predicted by ARNold ...................................................... 107
Table 3.5 smRNA sequences that fit the criteria for Hfq-dependent sRNAs ............................. 111
Table 3.6 Fermentation balances of strain 130Z expressing synthetic sRNA constructs. .......... 116
Table A3.1 Manually identified smRNAs using alignment files created by Rockhopper.......... 120
Table A3.2 smRNAs common between SIPHT and RNAseq .................................................... 124
Table A3.3 smRNAs common between BSRD and RNAseq..................................................... 125
Table A3.4 smRNAs common between INFERNAL and RNAseq ........................................... 126
Table A3.5 smRNAs common between RNAz and RNAseq ..................................................... 127
Table A3.6 smRNAs common between Rfam and RNAseq ...................................................... 128
Table A3.7 Primers used for RT-PCR validation of sRNAs ...................................................... 129

x

Table A3.8 Primers and g-blocks used in this study ................................................................... 130
Table 4.1 Strains and plasmids used in this study ...................................................................... 145
Table 4.2 Frequencies of knockout mutations introduced by natural transformation in A.
succinogenes strains 130Z and ΔpflB ......................................................................................... 162
Table A4.1 Oligonucleotide primers used in this study .............................................................. 169

xi

LIST OF FIGURES

Figure 1.1 Routes to succinate production ...................................................................................... 3
Figure 1.2 Metabolic map of Actinobacillus succinogenes. ......................................................... 16
Figure 1.3 Pathways to succinate production by mixed acid metabolism in bacteria .................. 19
Figure 1.4 Hfq-dependent sRNA organization and structure in E. coli. ....................................... 27
Figure 2.1 Construction of the ΔlacZ strain .................................................................................. 64
Figure 2.2 Identification of the pckA transcription start site. ........................................................ 65
Figure 2.3 Reporter constructs with lacZ under control of truncated versions of pckA’s 5’UTR. 66
Figure 2.4 β-galactosidase activity of promoter reporter constructs............................................. 68
Figure 3.1 Flow chart of the approach used for sRNA identification in A. succinogenes. ........... 95
Figure 3.2 Classification of identified sRNAs. ............................................................................. 97
Figure 3.3 Example of 3’UTR sRNA overlapping with the mRNA transcript and downstream
intergenic region with both mRNA and sRNA encoded on the reverse strand. ......................... 103
Figure 3.4 smRNA76 is seen as a single transcript in IGV. ....................................................... 105
Figure 3.5 RT-PCR validation of small RNAs. .......................................................................... 109
Figure 3.6 A. succinogenes smRNA8’s genomic locus .............................................................. 111
Figure 3.7 Synthetic sRNAs targeting lacZ expression. ............................................................. 113
Figure 3.8 β-galactosidase activity of strain 130Z expressing synthetic sRNAs targeting lacZ
expression. .................................................................................................................................. 113
Figure 3.9 Synthetic sRNA designs targeting expression of ackA and pta. ............................... 115
Figure 4.1 Construction of the icd selection cassette and construction of strain ∆frd. ............... 148
Figure 4.2 Physical maps of pCR2.1-ΔpflB::icd and its truncated constructs ........................... 157
Figure 4.3 Construction of strain ∆pflB (A) and natural transformation with the truncated
products ΔpflB::icd-600, ΔpflB::icd-400, and ΔpflB::icd-200 (B). ............................................. 158

xii

Figure 4.4 UV spectra of the HPLC profiles of fermentation supernatants of A. succinogenes
130Z, and ΔpflB grown on AM3 in the presence of 25 mM NaHCO3. ..................................... 159
Figure 4.5 Phenotypic characterization of double knockout mutant ∆pflB-∆lacZ::icd. ............. 163
Figure A4.1 Verification of the construction of strains ∆pflB ∆cit and ∆pflB ∆acn by PCR. .... 171

xiii

Chapter 1

Introduction

1

1.1 Bio-based succinate production
A huge push towards bio-based succinate production has taken place in the last ten years.
Succinate, a C4-dicarboxylic acid, has consistently been among the top ten bio-based chemicals
since 2004 (1-3). Succinate is currently produced by two methods. The majority of succinate is
produced by catalytic hydrogenation of petroleum-based maleic anhydride, but bio-based
succinate production by fermentation of glucose by natural/engineered succinate producers is
also catching up, as a number of companies are either in the process of setting up or have already
started large-scale fermentation facilities for succinic acid. The global production rate for
succinate is between 30,000-50,000 metric tons per year. The market volume for succinate in
2011 was 40,000 metric tons, out of which only 1,150 metric tons were bio-based succinate (4).
Market price for petroleum-derived succinate is between $6,000 and $9,000/metric ton (5). If all
announced large-scale fermentation projects come to fruition, an estimated total of 140,000
metric tons per year of bio-based succinate will be available globally (5). Large-scale production
facilities are estimated to produce a minimum of 642,450 metric tons in 2020.
As of today, four major companies have commercial-scale production facilities for bio-based
succinic acid—BioAmber, Myriant, Succinity GmbH, and Reverdia. BioAmber is a partnership
between DNP Green Technology and the French agricultural cooperative. In 2008, they started a
succinic acid plant in Pomacle, France, with an annual capacity of 2,000 metric tons (Chemicals
from Biomass) using Escherichia coli (licensed DuPont technology). In August 2015, BioAmber
announced the opening of the world’s largest succinate plant in Sarnia, Canada, as a joint venture
with Mitsui & Co., Ltd. The company has hit its target for operational milestones, which were
laid out at the start, and sold succinic acid worth $3.7 million in the third quarter of 2016

2

(Michael McCoy, ACS). BioAmber has two additional production plants planned or underway
with an annual capacity of 65,000 metric tons in USA or Brazil and in Thailand (5).

Figure 1.1 Routes to succinate production

Myriant started its flagship production facility in Lake Providence, Louisiana, in 2013 with
a 13,607 metric tons succinate production capacity per year. It had plans to expand this
production facility to have an additional 63,502 metric tons capacity per year (6) by 2015. In
partnership with ThyssenKrupp Uhde, Myriant also started a production facility in Leuna,
Germany, with a 1,500 metric tons production capacity per year (6). A jointly owned facility in
Nanjing, China, between Myriant and China Nation BlueStar is being planned with a potential to
scale up capacity to 100,000 metric tons per year (5).
Succinity GmbH is a joint venture between BASF and Corbian Purac. BASF has patents on
a genetically engineered strain of Basfia succiniciproducens to produce succinate from either
glycerol or mixed sugars while fixing the greenhouse gas CO2. A production facility in
Montmelό, Spain, has the capacity to produce 10,000 metric tons of succinate per year using
crude glycerol as the feedstock. A larger 50,000 metric tons per year facility was planned but no
specifics are known thus far (5).

3

Reverdia is a joint venture between Royal DSM, Netherlands, and Roquette Frѐres, France.
In 2012, Reverdia started a production facility to produce 10,000 metric tons of succinate per
year in Cassano Spinola, Italy, which uses the yeast Saccharomyces cerevisiae to convert
commodity sugars to succinate (5).

1.2 Substrates and feedstocks for succinate production
Substrates and feedstocks used in fermentations for succinate production can significantly
contribute to production costs and play a role in the economic feasibility of succinate production.
Abundance, cost, and availability of substrates are some of the important factors that make a
substrate more or less attractive for production. Recently, the focus has been more on using
renewable sources as substrates to lower the costs of production. Succinate production has been
studied from varied substrates such as glucose, sucrose, lactose, fructose, glycerol,
lignocellulosic hydrolysates, whey, cassava, and sugarcane molasses (7-15).

1.3 Applications for succinate and its derivatives
Applications for succinate and its derivatives are many. Succinate is used to make a wide variety
of products such as paints, coatings, adhesives, sealants, foods and flavors, cosmetics, nylons,
industrial lubricants, phthalate-free plasticizers, dyes and pigments, as well as pharmaceuticals
compounds (16). As bio-based succinate becomes more inexpensive, it can be used as a building
block for a variety of chemicals such as adipic acid, 1,4-butanediol, γ-butyrolactone,
tetrahydrofuran, n-methylpyrrolidone, poly-butylene succinate, 2-pyrrolidone, and polyamides
(17, 18). About two-thirds of the bio-based succinate produced by year 2020 is expected to be
used as an intermediate for producing 1,4-butanediol, tetrahydrofuran, and polyesters (19).

4

1.4 Succinate-producing microorganisms
Succinate is an important metabolite in cellular metabolism, as an intermediate metabolite in the
tricarboxylic acid (TCA) cycle and the glyoxylate shunt. Some microbes also produce it as an
end product during anaerobic fermentations. Most bacterial species that naturally produce large
amounts of succinate, such as Actinobacillus succinogenes, Mannheimia succiniciproducens, and
B. succiniciproducens were isolated from the cow rumen (20-22). These natural succinate
producers have been studied and engineered to further enhance succinate production. E. coli has
been engineered for succinate production by mimicking the metabolism of these natural
producers. E. coli, Corynebacterium glutamicum, and S. cerevisiae have also been engineered for
succinate production by disrupting the TCA cycle and forcing flux through the glyoxylate shunt.
Huge advances have been made in both natural producers and engineered strains towards
succinate production.

1.4.1 Native succinate producers
1.4.1.1 Anaerobiospirillum succiniciproducens
Anaerobiospirillum succiniciproducens was one of the first bacterial species studied for
succinate production. When grown at pH 6.2 and high CO2 concentrations, A. succiniproducens
produced high amounts of succinate, as opposed to lactate (23). Continuous cultures of A.
succiniciproducens grown on glycerol (supplemented with yeast extract), yielded 1.17 mol
succinate per mol glycerol with a productivity of 2.1 g L-1 h-1. Using a three-stage continuous
cell recycle bioreactor, the succinate yield increased to 1.35 mol mol-1 glucose. Production rate
and titer were 10.4 g L-1 h-1 and 83 g L-1, respectively (24).

5

Several features make this Gram-negative species unfit for industrial succinate production,
though. It is highly sensitive to oxygen, making it difficult to handle. It is known to cause rare,
but potentially lethal cases of bacteremia and diarrhea in humans (25), and no genetic tools are
available, making it an unsuitable host for engineering.

1.4.1.2 Actinobacillus succinogenes
A. succinogenes is a gram-negative, non-motile, osmotolerant, capnophilic, facultatively
anaerobic bacterium that was isolated from a bovine rumen. It has the ability to grow on a broad
range of substrates, such as glucose, lactose, xylose, arabinose, fructose, and glycerol (20). It
belongs to the Pasteurellaceae family, along with Haemophilus, Aggregatibacter, Mannheimia,
and Pasteurella. A. succinogenes’s fermentation pathways are well studied. A. succinogenes
produces phosphoenolpyruvate (PEP) from glucose via the Embden-Meyerhof Parnas pathway.
PEP is then either dephosphorylated to pyruvate by pyruvate kinase, or converted into
oxaloacetate by PEP carboxykinase (PEPCK, encoded by pckA) (26). These two branches are
interconnected by two decarboxylating enzymes, malic enzyme and oxaloacetate decarboxylase
(27). Oxaloacetate is further reduced to malate, malate to fumarate, and fumarate is eventually
reduced to succinate (Figure 1.2). A. succinogenes has an incomplete TCA cycle that lacks
citrate synthase and isocitrate dehydrogenase (28). It also lacks the glyoxylate shunt (27).
Metabolic flux analyses and genome annotation have led to a deeper understanding of A.
succinogenes’s pathways and metabolism (27, 29, 30). A. succinogenes is capable of natural
transformation and a knockout method has been developed (31).
A. succinogenes does not ferment glycerol, but is can grow on glycerol by respiration. With
nitrate as the terminal electron acceptor A. succinogenes produced acetate and CO2 as the main

6

products, together with small amounts of succinate, but no ethanol. In contrast, succinate became
the main secreted product of glycerol-grown cultures with dimethyl sulfoxide (DMSO) as the
electron acceptor, with a yield of 59% of the maximum theoretical yield, with formate, acetate,
and CO2 as by-products. Similarly, succinate yields reached as high as 67% of the maximum
theoretical yield under microaerobic conditions (1% O2, batch), with acetate, CO2 and formate as
by-products. Deleting the pyruvate-formate lyase (ΔpflB) increased the succinate yield to 76% in
microaerobic cultures (1% O2, batch) (15). Carvalho et al. grew A. succinogenes on glycerol with
DMSO as the external electron acceptor under fed-batch conditions, and obtained a succinate
titer of 49.6 g L-1 and a succinate yield of 0.5 mol mol-1 glycerol (32).
The Nicol group developed a novel extended recycle biofilm reactor (continuous system)
for fermentative succinate production. Using glucose as the substrate and CO2 to maintain
anaerobic conditions, fermentations were carried out with media supplemented with yeast extract
and corn steep liquor. At a 0.56 h-1 dilution rate, the succinate yield reached 1.05 mol mol-1
glucose with a 6.35 g L-1 h-1 productivity (33). The succinate yield was shown to be an
increasing function of glucose consumption in that system, and the succinate yield could be
increased to 1.39 mol mol-1 glucose with a 48.5 g L-1 succinate titer (34). In a separate study,
Guettler et al. overexpressed the native glucose-6-phosphate dehydrogenase (G6PDH) in a
pyruvate–formate lyase mutant strain to direct more flux through the pentose phosphate pathway.
Under anaerobic conditions, the recombinant strain produced succinate with productivity, titer,
and yield of 2.19 g L-1 h-1, 100.6 g L-1, and 1.36 mol mol-1 glucose, respectively (35).
Vlysidis et al. first conducted batch experiments using a wide range of initial glycerol
concentrations to test substrate inhibition on the process. One of the best results was found with a
starting concentration of 37 g L-1 glycerol and yeast extract supplementation. These batch

7

cultures produced a succinate titer of 29.3 g L-1, a productivity of 0.27 g L-1 h-1, and a yield of
0.62 mol mol-1 glycerol. Next Vlysidis et al. developed an unstructured model by fitting a set of
kinetic equations to the experimental data they found in the batch experiments done in small
anaerobic reactors. This model was developed to predict experimental behavior and was
validated by performing experiments in a scaled up bench top reactor. The model was able to
predict the bench-top experiments without any additional fitting for a variety of conditions and
can safely be used for future experiments (36, 37). An unstructured model was also developed to
predict A. succinogenes fermentations on mixtures of C5 and C6 sugars found in spent sulfite
liquor (38). Batch anaerobic fermentations were carried out using a mixture of xylose (72.6%),
galactose, glucose, mannose, and arabinose with yeast extract. Succinate was produced with a
titer of 27.4 g L-1, productivity of 0.45 g L-1 h-1, and yield of 0.7 g g-1 total sugars. Simulations
were carried out using the kinetic parameters obtained from these experimental studies. Scaled
up experiments using a 2-L lab-scale bioreactor using varying mixed sugar concentrations (15-50
g L-1) were performed to validate the predictive nature of their model. The R2 value of the model
validations with the experimental results was 0.93, indicating that this model can effectively
predict batch fermentations (38).
The Bechkam group has studied the behavior of A. succcinogenes on xylose-enriched
hydrolysates, a feedstock being widely studied as a substrate for production of value-added
chemicals. This group grew A. succinogenes on deacetylated dilute acid-pretreated corn stover
hydrolysate and obtained a succinate yield of 0.74 g g-1 sugars. The maximum succinate
productivity and titer were 1.27 g L-1 h-1 and 42.8 g L-1, respectively (39). More recently, they
have made ΔackA, ΔpflB and ΔpflBΔackA strains of A. succinogenes. When grown on mock
biomass hydrolysates all strains showed a decrease in succinate titer, yield and productivity

8

compared to the wild-type strain indicating that, removal of heterofermentative pathways may
not lead to increase in flux towards succinate (40). In the same study, they also overexpressed the
PEP carboxykinase, malate dehydrogenase and fumarase in the wild-type strain as well as in all
three of the above strains. They did not see any significant increase in yield, productivity and
titers for either of ΔackA, ΔpflB and ΔpflBΔackA strains compared to the wild-type. All three of
the wild-type strains expressing PEP carboxykinase, malate dehydrogenase and fumarase did
show a slight increase in succinate yield and titer as compared to the wild-type strain. It is
interesting to note that acetate accumulation was seen after a lag period in all ackA mutant
backgrounds, suggesting an alternate route to acetate in A. succinogenes. Many other studies
have tested for succinate production by A. succinogenes on different renewable carbon sources,
such as straw hydrolysate, crop stalk waste, corn stover, cheese whey, and sugarcane bagasse
(12, 41-44).

1.4.1.3 Mannheimia succiniciproducens
In 2002, Lee et al. reported the isolation and characterization of a novel succinateproducing Pasteurellaceae species, Mannheimia succiniciproducens MBEL55E, from cow
rumen. M. succiniciproducens produces succinate as its major fermentation product, along with
acetate, formate, and lactate as by-products (22). M. succiniciproducens’s genome was
sequenced and its metabolism was extensively studied (45). Genome-based metabolic
engineering of M. succiniciproducens was carried out to study its metabolism and increase
succinate production (45). In-silico flux analysis identified PEPCK as the major CO2-fixing
enzyme instead of PEP carboxylase. PEPCK is the anaplerotic enzyme in the reductive TCA,

9

thus playing an important role in succinate production. Other enzymes responsible for formation
of by-products during fermentation were also identified.
During anaerobic fermentation on glucose in batch conditions, the wild-type strain
produced 10.5 g L-1 of succinate, 4.9 g L-1 of acetate, 4.1 g L-1 of formate, and 3.5 g L-1 of lactate
(45). LPK7 is a metabolically engineered strain of M. succiniciproducens, carrying ldhA, pflB,
pta, and ackA deletions (45). In fed-batch cultures this strain produced succinate with yield,
productivity, and titer of 1.16 mol mol-1glucose, 1.8 g L-1 h-1, and 52.4 g L-1, respectively. With
glycerol as the carbon source, the PALK strain (ΔldhA Δpta-ackA) had a succinate yield of 0.88
mol mol-1 glycerol with an overall productivity of 0.13 g L-1 h-1. The PALK strain grown in fedbatch conditions using glucose and glycerol as co-substrates along with 6.84 M magnesium
hydroxide and 1.57 M ammonia for pH control gave a succinate yield of 1.15 mol mol-1 glucose
equivalent, with an overall productivity of 3.5 g L-1 h-1 and a titer of 90.7 g L-1, together with
acetate (2.3 g L-1) and pyruvate (4.0 g L-1) as byproducts (46).
Recent simulations (47) based on omics studies and metabolism reconstructions predicted
that sucrose and glycerol used as co-substrates would increase succinate production while
reducing by-product formation. In this study, the PALKF (ΔldhA Δpta-ackA ΔfruA) strain was
constructed to divert the majority of the flux towards succinate production, minimize by-product
formation, and deregulate catabolite repression. The PALKF strain was grown in fed-batch
conditions as low and medium density cultures. The succinate yields obtained were 1.56 and
1.64 mol mol-1 glucose equivalent with overall productivities of 2.50 and 6.02 g L-1 h-1 for low
and medium density cultures, respectively. The PALKG strain (expressing the allosteric
inhibition-free E. coli glpK22) (48, 49) was also developed, which gave a higher productivity
(3.34 g L-1 h-1) than the PALKF strain, but produced more by-products. A membrane cell recycle

10

bioreactor system was also developed in which the PALKF strain had a succinate productivity of
38.6 g L-1 h-1 (47). Overall, Dr. Sang Yup Lee’s group was able to make a homo-succinate
production possible with high productivities using sucrose and glycerol as co-substrates.

1.4.1.4 Basfia succiniciproducens
B. succiniciproducens DD1 was isolated from the bovine rumen and characterized as
another succinate-producing Pasteurellaceae species (50). B. succiniciproducens DD1 and M.
succiniciproducens MBEL55E had similar genome sizes, and their homologous proteins shared
95% similarity (21). When grown on glucose or sucrose as the carbon source, B.
succiniciproducens DD1 produced succinate with a 5.8 g L-1 titer, a 1.5 g L-1 productivity, and a
0.6 g g-1 sucrose yield. With glycerol as the carbon source the succinate titer was higher, at 8.4 g
L-1, with a productivity of 0.9 g L-1 h-1 and a yield of 1.2 g g-1 glycerol (50). A continuous
cultivation process was developed on crude glycerol supplemented with yeast extract and was
maintained for 80 days. At a dilution rate of 0.018 h-1 the succinate titer was 5.21 g L-1, the
productivity was 0.094 g L-1 h-1, and the yield was 1.02 g g-1 glycerol (50, 51). Metabolic flux
analysis showed unwanted fluxes through pyruvate-formate lyase and lactate dehydrogenase. A
ΔldhA ΔpflD strain was developed that showed a succinate yield (1.08 mol mol-1 glucose)
reaching 62% of the maximum theoretical yield (52). Succinity has a patent for the engineered
strain and is using it for commercial scale production of succinate in its facilities.

11

1.4.2 Engineered succinate producers
1.4.2.1 Escherichia coli
E. coli is metabolically versatile and can consume glucose both anaerobically and
aerobically. Under anaerobic conditions it carries out a mixed acid fermentation when grown on
glucose to produce formate, lactate, ethanol, acetate, and succinate, where succinate is only a
minor fermentation product (53). Under aerobic conditions it only produces succinate if the
glyoxylate shunt is operational (54). The theoretical maximum yield for succinate in E. coli
grown fermentatively on glucose is 1 mol mol-1 glucose (54). Wild-type E. coli produces only
0.11 mol succinate per mol glucose in these conditions (55). The succinate yield on glucose is
generally limited by the availability of reducing equivalents (54). Although E.coli is not a natural
succinate producer, strategies have been developed to improve succinate production in E. coli by
engineering, evolution, or optimizing the production conditions. In one of the earlier studies by
Millard et al., overexpressing PEP carboxylase (PEPC) increased succinate production 3.75-fold,
but concomitantly decreased glucose uptake (56). A wild-type E. coli strain overexpressing a
heterologous pyruvate carboxylase (from Rhizobium etli) produced 1.77 g L-1 succinate
compared to 1.18 g L-1 by the wild-type strain (57). Strain AFP111 carrying knockout mutations
in pflB and ldhA was unable to ferment glucose (53). A spontaneous mutation in the ptsG gene
restored the ability of AFP111 derivative NZN111 to ferment glucose while producing 1 mol of
succinate, as well as 0.5 mol of acetate and ethanol each per mol of glucose (58). Strain
AFP111(pTrc99A-pyc), which overexpresses the R. etli pyruvate carboxylase, produced 99.2 g
L-1 of succinate with yield and productivity of 1.74 mol mol-1 glucose and 1.3 g L-1 h-1,
respectively, when grown in dual phase conditions (cultures grown aerobically to generate
biomass, then transferred to anaerobic conditions conducive to succinate production) (59). Kim

12

et al., 2004 demonstrated that PEPCK can replace PEPC as the PEP-carboxylating enzyme in E.
coli. Overexpressing the A. succinogenes 130Z PEPCK in E. coli K-12 ppc::kan increased
succinate production 6.5-fold (60).
Dr. Lonnie Ingram’s group combined metabolic evolution and gene knockouts to develop a
good biocatalyst for succinate production. First the genes responsible for the formation of major
by-products were knocked out, namely acetate kinase (ackA), lactate dehydrogenase (ldhA), and
alcohol dehydrogenase (adhE). This strain, called KJ012, grew poorly on glucose minimal
medium under anaerobic conditions. In this strain, however, ATP production was coupled to
succinate production. Strain KJ012 was evolved to grow anaerobically in glucose minimal
medium for over 2,000 generations to allow for growth-based selection. The evolved strain grew
much better on glucose minimal medium under anaerobic conditions, and produced more
succinate than its parent strain. Additionally, the genes focA and pflB were deleted as well,
increasing the succinate production levels even higher. This new strain, KJ060, produced a
succinate yield of 1.41 mol mol-1 glucose, with productivity and titer of 0.9 g L-1 h-1 and 86.6 g
L-1 respectively. The succinate yield reached 1.61 mol mol-1 glucose when higher inoculums
were used (61). KJ134 (ΔldhA ΔadhE ΔfocA-pflB ΔmgsA ΔpoxB ΔtdcDE ΔcitF ΔaspC ΔsfcA
Δpta-ackA) is a derivative of KJ060. This strain had additional deletions in pta, tdcE, and tdcD to
further reduce acetate production. The carbon flux to oxaloacetate was also increased by
knocking out genes coding for aspartate aminotransferase (aspC) and NADP-linked malic
enzyme (sfcA). KJ134 produced 1.53 mol succinate per mol of glucose with a productivity of
0.75 g L-1 h-1 and a titer of 71.56 g L-1 (62). Additionally, strain KJ134 showed an 80% reduction
in acetate production as compared to KJ073 (ΔldhA, ΔadhE, ΔackA, ΔfocA, ΔpflB, ΔmgsA,

13

ΔpoxB). When grown on 2% glycerol, KJ073 produced 15 g L-1 succinate from 16.9 g L-1
glycerol, a yield of 89% of the theoretical maximum (61).
During the evolution of strains KJ060 and KJ073, the authors noticed changes in
metabolism that led to increased succinate production and ATP yield per glucose (63). They
found a spontaneous mutation in the promoter of pckA, encoding gluconeogenic PEPCK, that
allowed PEPCK to replace PEPC as the main carboxylating enzyme. In addition, glucose uptake
through the PEP-dependent phosphotransferase system had been inactivated (spontaneous
mutation) and replaced by galactose permease and glucokinase. These E. coli strains now had a
pathway similar, for succinate production, to the natural succinate-producing organisms found in
cow rumens.
The Ingram group confirmed that these changes led to higher succinate amounts by
engineering an E. coli strain with the mutations seen above. Strain XZ647, with only two
mutations (ptsI truncation and constitutive pckA), was able to produce succinate with a yield of
0.89 mol mol-1 of glucose. An additional pflB deletion yielded strain XZ721, whose succinate
yield increased to 1.25 mol mol-1glucose (64). When grown on glycerol, strain XZ721 had a
yield of 0.8 mol mol-1 of glycerol (63, 65).
Dr. Ramon Gonzalez’s group also engineered E. coli for succinate production from
glycerol. Their approach included blocking the pathways to major by-products and
overexpressing a heterologous pyruvate carboxylase to increase succinate production by driving
the flux from pyruvate to oxaloacetate. The final strain, ΔadhE Δpta ΔpoxB ΔldhA Δppc [pZSpyc], overexpressed the Lactobacillus lactis pyruvate carboxylase and produced succinate with a
yield of 0.54 mol mol-1 glycerol in microaerobic conditions (14).

14

1.4.2.2 Corynebacterium glutamicum
C. glutamicum is a gram-positive, non-motile, spore-forming, facultatively anaerobic
bacterium that belongs to the Actinomycetes subdivision of eubacteria. It has been widely
studied and used as an industrial strain for the production of amino acids and other organic acids
(66). Okino et al. observed that C. glutamicum incubated in glucose-mineral medium under
oxygen-limited conditions (without growth) produced succinate, lactate, and acetate (67).
Increased bicarbonate concentrations led to an increased succinate yield and a decreased lactate
yield. The ability of this species to produce organic acids in arrested growth conditions allowed
for a bioreactor design in which high cell density led to a high succinate volumetric productivity
of 11.7 g L-1 h-1 (67). The C. glutamicum strain ΔldhA-pCRA717, a lactate dehydrogenase
knockout that overexpresses pyruvate carboxylase, produced 146 g L-1 succinate in 46 h and the
succinate and acetate yields were 1.4 and 0.29 mol mol-1 glucose, respectively (68). The
succinate productivity was 3.2 g L-1 h-1.
Litsanov et al. constructed the BOL-1 strain, in which all known pathways leading to
acetate and lactate synthesis were deleted, and the BOL-2 strain where the pyruvate carboxylase
gene was integrated chromosomally into the BOL-1 strain. Integrating the Mycobacterium
vaccae NAD+-coupled formate dehydrogenase into the BOL-2 chromosome gave strain BOL-3.
Additionally, a metabolic blockage of glycolysis caused by NADH inhibition of GAPDH activity
in glucose-grown BOL-3 was relieved by overexpressing the native glyceraldehyde-3-phosphate
dehydrogenase gene on a plasmid. This final strain grown in the presence of glucose and formate
in fed-batch conditions produced 134 g L-1 of succinate in 53 h. The succinate yield was 1.67
mol mol-1 glucose with very little accumulation of other by-products (0.1 mol mol-1 glucose)
(69).

15

Figure 1.2 Metabolic map of Actinobacillus succinogenes. AcCoA, acetyl CoA; CoA,
Coenzyme A; OAA, oxaloacetate; and PEP, phosphoenolpyruvate.

Another study came up with a dual synthesis approach for succinate production without
using formate. The glyoxylate pathway citrate synthase genes were overexpressed to direct more
carbon toward the glyoxylate pathway. The succinate exporter SucE was overexpressed as well.
This strain produced 109 g L-1 succinate with an overall volumetric productivity of 1.11 g L-1 h-1
and a yield of 1.32 mol mol-1 glucose (70). All these studies were carried out under anaerobic
conditions.
More recently, a few studies have focused on aerobic production of succinate in C.
glutamicum. Strain BL-1 (pAN6-pycP458Sppc) has in-frame deletions of pqo, pta-ackA, sdhCAB,
and cat, and overexpresses the pyruvate carboxylase (pyc) and PEPC (ppc) genes. This strain is
the first known to produce succinate aerobically in minimal glucose medium. Grown under

16

conditions that decoupled succinate production from growth, this strain produced a titer of 10.6 g
L-1 succinate with a yield of 0.45 mol mol-1 glucose (71). Strain BL-1(pVWEx1-glpFKD)
overexpresses the E. coli glycerol facilitator, glycerol kinase, and glycerol-3-phosphate
dehydrogenase. When grown aerobically on glycerol minimal medium, it produced 9.3 g L-1 of
succinate with a volumetric productivity of 0.42 g L-1 h-1 (72). In another study, B. subtilis
acetyl-coA synthetase was overexpressed on a plasmid in the base strain ZX1 to recycle the
carbon in acetate. The ZX1 strain has deletions of ldhA, pqo, cat, and pta genes along with
replacement of the native promoters of the pyc and ppc with the sod promoter. Strain
ZX1(pEacsA) did not secrete any acetate, and the succinate yield was 0.5 mol mol-1 glucose.
Overexpressing citrate synthase (encoded by gltA) led to an additional 22% increase in succinate
yield. When strain ZX1(pEacsAgltA) was grown in fed-batch conditions on glucose it produced
28.4 g L-1 of succinate with a volumetric productivity of 0.41 g L-1 h-1 and a yield of 0.63 mol
mol-1 glucose. This study, however, did use rich medium components in their production
medium (73).
A recent study reported the development of an engineered strain of C. glutamicum, S071
(ΔldhA Δpta-ackA ΔactA ΔpoxB pycP458S Ptuf::ppc Δpck_Ptuf::pckG ΔptsG) able to produce 152.2
g L-1 of succinate with a yield of 1.67 mol mol-1 glucose by over-expressing the transcriptional
regulator NCgl0275. Over-expressing NCgl0275 allowed the release of an end-product inhibition
due to succinate by increasing the glucose consumption rate (74).

1.4.2.3 Saccharomyces cerevisiae
S. cerevisiae is the only yeast that has been well studied for succinate production. Succinate
is the main component that imparts flavor to Sake during fermentation (75). Arikawa et al.

17

studied the metabolic pathways leading to succinate production in S. cerevisiae and established
that succinate could be produced by either n-ketoglutarate oxidation or fumarate reduction. In
one of the first engineering studies of S. cerevisiae, inactivation of succinate dehydrogenase (sdh
and sdh1b deletions) generated a strain with reduced malate productivity, and a succinate
productivity about double that of the wild-type strain (76). Another group deleted sdh1, sdh2,
idh1, and idp1 to completely abolish isocitrate dehydrogenase and succinate dehydrogenase
activities, diverting the TCA cycle flux into the glyoxylate pathway, and increasing succinate
production. In glucose-shake flask cultures, this strain produced 3.62 g L-1 of succinate with a
yield of 0.11 mol mol-1 glucose (77). Ito et al. (78) deleted sdh1 and sdh2 (encoding succinate
dehydrogenase subunits) to allow for aerobic succinate production. To further increase succinate
production they deleted the ethanol-producing genes, adh1, adh2, adh3, adh4, and adh5
(encoding alcohol dehydrogenases). This strain, S149sdh12, produced 20-fold more succinate
than the control strain, with a yield of 0.22 mol mol-1 glucose. Since succinate was accumulating
intracellularly, they overexpressed the Schizosaccharomyces pombe malic acid transporter
(mae1) in this engineered strain and further increased the succinate yield to 0.24 mol mol-1
glucose.
Yan et al. engineered S. cerevisiae for succinate production through the reductive branch of
the TCA cycle. In S. cerevisiae, fum1 encoding fumarate hydratase irreversibly converts
fumarate to malate, which poses a problem for reductive succinate production. The authors
deleted fum1 and overexpressed E. coli fumC in its place, in a pyruvate decarboxylase-deficient
(TAM) strain. The pyc2, mdh3, and frd1 genes were also overexpressed to increase pyruvate
decarboxylase, malate dehydrogenase, and fumarate reductase activity, respectively. GPD1
(glycerol-3-phosphate dehydrogenase) is responsible for NADH-dependent glycerol production

18

in aerobic conditions (79). Since production of 1 mol of succinate consumes 2 mol of NADH
through the reductive TCA branch, the authors suspected that inactivating GPD1 would
potentially increase succinate production. When grown in a bioreactor with optimal CO2 (10%)
and medium conditions, the final strain, PMCFfg, produced succinate with a titer of 13 g L-1 and
a yield of 0.21 mol mol-1 glucose at a pH of 3.8 (79).

Figure 1.3 Pathways to succinate production by mixed acid metabolism in bacteria.
Abbreviations are the same as in Figure 1.2.

19

Table 1.1 Succinate-producing strains with their yields, productivities, and titers
Strain

Conditions/Substrate

Succinate
Yield
(mol mol-1)

Productivity
(g L-1 h-1)

Other
products

Reference

31.7
24.4
49.6
48.5

0.75
0.74
0.50
1.39

0.14
2.13
0.96
ND

A, F
A, F
A, F
A, F

(15)
(32)
(32)
(34)

90.7
78.4

1.15a
1.64a

3.49
6.03

P, A
P, A

(46)
(47)

13.2

1.22a

38.6

P, A

(47)

69.8

1.70

2.91

46.3

1.33

ND

A, P, L, M,
F
A

(78, 80,
81)
(80)

99.2
86.6
11.7
71.5
8.75

1.74
1.41
0.89
1.53
0.77

1.3
0.9
ND
0.75
ND

A, E
A, M, L
ND
A, P, M
ND

(59)
(61)
(61)
(62)
(65)

146
134

1.4
1.67

3.2
2.48

A, L, M, P
K, M, A, F,
P

(68)
(69)

ND
3.62
13.0

0.24
0.11
0.21

ND
ND
ND

G, E, Fum
K, P, C
G, P

(78)
(77)
(79)

Titer
(g L-1)
Actinobacillus succinogenes
130Z
130Z
130Z
130Z

D, CC, glycerol, Mi
D, glycerol, B, DMSO, An, YE
D, glycerol, FB, DMSO, An, YE
D, glucose, YE, CSL, An

LU15224

Mannheimia succiniciproducens
D, FB, An, glucose + glycerol
D, FB, An, sucrose + glycerol, high cell concentration (Initial OD 600 of
9.03)
D, MRCB (dilution rate of 2.93l h-1), An, sucrose + glycerol
Basfia succiniciproducens DD1
D, B, An, glycerol + maltose

LU15224 pJFF224 (icl ms Y.m.)

D, B, An, glucose, YE

PALK
PALKF
PALKF

Escherichia coli
AFP111(pTrc99A-pyc)
KJ060
KJ134
XZ465

D, FBc, YE, T
D, B, glucose
D, B, glycerol
D, B, glucose
D, B, glycerol
Corynebacterium glutamicum

ΔldhA-pCRA717
BOL-3/pAN6-gap (Δcat, Δpqo, Δpta-ack,
Δldh, pycP4585, fdh)

Db, B, OD, glucose
Db, FB, An, glucose + formate

S149sdh12/pNV11-mae1
Δsdh1Δsdh2Δidh1Δidp1
PMCFfg

D, B, Ae, glucose
D, B, Ae, glucose
D, B, Ae, glucose

Saccharomyces cerevisiae

Abbreviations: A, acetate; Ae, aerobic; An, anaerobic; B, batch; C, Citrate; CC, continuous culture; CSL, corn steep liquor; D,
defined; DMSO, dimethyl sulfoxide; E: ethanol; F, formate; Fum, fumarate; FB, fed-batch; G, Glycerol; K, α-ketoglutarate; L, lactate;
M, Malate; Mi, microaerobic; MCRB, membrane cell recycling bioreactor; ND: Not determined; OD, oxygen deprivation (dissolved
oxygen lower than 0.01 ppm); P, pyruvate; T, tryptone; VMD, vacuum membrane distillation; and YE, yeast extract.
a
mol mol-1 glucose equivalent
b
aerobic cell propagation and transfer to anaerobic succinate production conditions
c
dual phase fed-batch where cells were grown in aerobic conditions, then oxygen-free CO2 was sparged at transition point

20

1.5 C4-dicarboxylate transporters
Engineering of microbial species for succinate production has mostly focused on targets
within intracellular metabolic pathways. Very few engineering studies have focused on
expressing succinate transporters to increase succinate production. E. coli and C. glutamicum are
two commercially relevant strains in which succinate transporters have been characterized and
used to increase succinate production. C4-dicarboxylates such as succinate, fumarate, and malate
are intermediates in the TCA cycle and serve as carbon/energy sources during aerobic growth
and therefore require uptake systems (82). During anaerobic growth, all C4-dicarboxylates other
than succinate are used as electron acceptors in fumarate respiration. In contrast, succinate is
exported via antiport against the uptake of malate or fumarate (82). Many different types of
transporters are involved in C4-dicarboxylate transport and, with a few exceptions, most studies
to identify these transporters have focused on E. coli and on C4-dicarboxylate uptake.

1.5.1 E. coli C4-dicarboxylate transporters
So far, E. coli has seven known C4-dicarboxylate transporters —DauA, DctA, DcuA,
DcuB, DcuC, YjjPB, and CitT. Unidirectional uptake of fumarate, succinate, malate, and
aspartate takes place under aerobic conditions, while exchange, uptake, and efflux of these C4dicarboxylates take place in anaerobic conditions (83). DauA and DctA are primarily known to
function in aerobic conditions while all others are active in anaerobic conditions. DauA
(dicarboxylic acid uptake system A) has been characterized as a succinate uptake transporter
active under acidic pH conditions (84). DctA, a dicarboxylate amino acid-cation symporter
family (DAACS) transporter is used for aerobic growth in many bacteria such as E. coli, Bacillus
subtilis, Rhizobium leguminosarum, and C. glutamicum (85-87). It mediates the uptake of

21

succinate, malate and fumarate in E. coli under aerobic conditions. DcuA, DcuB, and DcuC are
antiporters found only in anaerobic and facultatively anaerobic bacteria and are involved in
fumarate respiration. The DcuAB and DcuC families are independent from each other, but
members of both families are capable of uptake, exchange, and efflux of C4-dicarboxylates. The
main functions of DcuA, DcuB, and DcuC are still not completely elucidated. An early study
determined that DcuAB carriers mostly functioned as uptake or exchange carriers whereas DcuC
carriers likely mostly operated as efflux carriers (82). Still, an E. coli DcuC mutant grown
anaerobically was not affected in succinate production, indicating that DcuC is not the main
succinate exporter in anaerobic conditions (88). Another E. coli study showed that individual
deletions of dcuB and dcuC decreased succinate titer by 15 and 11%, and a ΔdcuBΔdcuC double
mutant showed a 90% decrease in succinate titer in a glucose fermentation. Modulating the
expression of these two transporters in E. coli also increased the succinate titer by 34% (89).
Another family of transporters involved in C4-dicarboxylate transport is the TRAP
(tripartite-ATP independent periplasmic) family. In TRAP carriers, the driving force for solute
accumulation is an electrochemical ion gradient (90). TRAP carriers consist of two membrane
integral proteins (DctQM) and a periplasmic solute-binding protein (DctP). The first
characterized TRAP transporter was the R. capsulatus DctPQM system and homologs of this
system were found to be present in a wide range of bacteria and archaea (90). The TRAP family
carriers are involved in solute uptake and are considered unidirectional (91). For this reason, it is
unlikely that they could be involved in succinate efflux.
Another known E. coli C4-dicarboxylate transporter is CitT, however CitT is not restricted
to only C4-dicarboxylate transport. E. coli CitT is involved in citrate uptake in exchange for
succinate, fumarate, or tartarate under anaerobic conditions (82, 92). More recently, a study by

22

Fukui et al. (88) demonstrated that E. coli YjjP and YjjB form a single complex involved in
succinate export under both aerobic and anaerobic conditions. A YjjPB mutant showed a 70%
decrease in succinate production suggesting that YjjPB is involved in succinate export. YjjPB
were found to contain different structures and conserved domains when compared to other
known C4-dicarboxylate transporters (88).

1.5.2 C4-dicarboxylate transporters in other succinate producers
BlastP searches with E. coli YjjP in A. succinogenes, M. succiniciproducens, and E.
aerogenes predicted homologs with 50% (87%), 54% (89%) and 88% (90%) identity
(similarity), respectively (88). Similar BlastP searches with E. coli YjjB identified homologs in
A. succinogenes, M. succiniciproducens, and E. aerogenes with 50% (81%), 52% (83%), and
83% (97%) identity (similarity) (88). Whether these homologs function in succinate export is
unknown yet.
The C. glutamicum proteome does not contain any homologs of the E. coli Dcu family
succinate transporters. In a study of C. glutamicum transmembrane proteins, 57 membrane
proteins with at least four trans-membrane domains were found to be significantly similar to A.
succinogenes and M. succiniciproducens proteins. Eliminating those with predicted or known
functions brought the list of candidates down to twenty. Construction of deletion mutants of
twelve of these candidates allowed the identification of SucE. The sucE mutant produced 70% of
the wild-type levels of succinate and thus may not be the main succinate exporter in anaerobic
conditions. SucE is a secondary carrier like the Dcu family carriers in E. coli (93). Another study
identified SucE through transcriptomics studies for C. glutamicum grown in anaerobic and
microaerobic conditions (94). Unlike the Dcu family carriers SucE is only involved in succinate

23

export (93). In another study, SucE-overexpressing C. glutamicum produced 109 g L-1 of
succinate and a yield of 1.32 mol mol-1 glucose as mentioned above (70). SucE homologs were
found in E. coli, A. succinogenes, and M. succiniciproducens (93), but their function has not
been tested.
The only identified succinate transporter in A. succinogenes is Asuc_0304. It was identified
by complementation studies done in a C4-dicarboxylate transport deficient mutant E. coli strain
IMW213. Asuc_0304 was identified to be an uptake C4-dicarboxylate transporter capable of
fumarate or malate uptake with Na+ symport, although it may not be the sole or main transporter
under aerobic conditions (95).

1.6 sRNAs as synthetic regulators for metabolic engineering
Industrially-relevant microorganisms have been subject to heavy modifications to optimize
yield, productivity, and titers of their products. Most engineering strategies are aimed at
identifying target genes and either knocking them out or over-expressing them. Many groups
also take a rational approach for evolving strains for growth under certain conditions. However,
it is often not easy to manipulate microorganisms and therefore conduct any type of metabolic
engineering. In addition, these methods can only target a limited number of genes at a time and
are therefore unsuitable for applications on a genomic scale (96). Another disadvantage of
overexpressing genes is the limited control over expression level and/or stability in the particular
organism. There is a growing need for strategies that enable researchers to modulate gene
expression or repression at the translational level.
Synthetic sRNAs are gaining tremendous momentum in research, as using them would
allow us to overcome many of the challenges listed above. One of the major advantages of

24

synthetic sRNAs is the possibility of designing them with a rational approach, making it
unnecessary to build individual strains to test the intended phenotypes of the knockout or
overexpression of genes. A recent explosion of studies has started to decipher the different ways
in which sRNAs modulate different genes in nature. The increased understanding of how sRNAs
function has led to an interest in using them as synthetic regulators by mimicking in vivo gene
regulation to target genes of interest. Synthetic small RNAs have been designed for E. coli in
several studies (42, 96-102). Desai et al. (103) demonstrated that expressing non-coding
antisense sRNAs (asRNAs) complementary to the butyrate kinase, phosphotransbutyrylase, and
phosphotransacetylase mRNAs inhibited these enzymes’ activities and increased acetone and
butanol production. Kim et al. designed antisense RNAs to lower the expression of acetate kinase
and phosphotransacetylase. This strategy partially reduced the mRNA levels of the target genes
and blocked the synthesis of the enzymes (97). Nakashima et al. designed an asRNA with pairedend termini for conditional silencing of acetate kinase and phosphotransacetylase. The pairedend termini increased the asRNA stability and improved the conditional silencing in E. coli (98,
99). Negrete et al. also demonstrated that overexpressing the GadY sRNA in E. coli decreased
acetate production, which in turn decreased the inhibitory effect acetate had on the strain’s
growth (99).
Several of the early discovered sRNAs were found to be Hfq-dependent. Hfq is an RNAbinding protein that acts as a chaperone by enabling the binding of sRNAs to their target RNAs.
Hfq is also known to stabilize sRNAs and their target mRNAs, thereby increasing their half-lives
(104). E. coli Hfq-dependent sRNAs have an mRNA base-pairing region, Hfq-binding region,
and a Rho-independent terminator (Figure 1.4). The Hfq-binding region contains a consensus
secondary structure that provides a scaffold for Hfq to bind. In one of the first studies that

25

implemented sRNAs for gene repression, artificial sequences complementary to the ribosomal
binding sites and start codons of ipp, ompA, and ompC were inserted in between the stem loops
at the 5’ and 3’ termini of the natural hfq-dependent MicF sRNA. These synthetic sRNAs could
inhibit the production of OmpC, OmpA, and Ipp (100). In another E. coli study the known sRNA
RhyB (involved in regulation of iron metabolism) was overexpressed with an arabinose inducible
promoter. Overexpressing RhyB during growth on glucose led to a 7-fold increase in succinate
production without citrate production. RhyB was shown to regulate sdhCDAB and acnB, which
encode succinate dehydrogenase and aconitase, respectively, in the TCA cycle (101).
Overexpressing RhyB in the E. coli DALRA strain (engineered for 5-aminolevulinic acid
production) allowed for 16% higher 5-aminolevulinic acid production than the parent strain (42).
Na et al. designed 130 synthetic sRNAs by using two separate parts—the scaffold sequence and
the mRNA or target-binding sequence. Using this approach the authors were able to substantially
increase the production of cadaverine and tyrosine in E. coli. At high cell densities, the best
engineered strain produced a tyrosine titer of 21.9 g L-1, whereas cadaverine production went up
to 12.6 g L-1, compared to 9.6 g L-1 for the starting strain (96).

26

Figure 1.4 Hfq-dependent sRNA organization and structure in E. coli.

Designing sRNAs in E. coli is much more straightforward now with the vast amount of
knowledge gained over the past decade. However, for other organisms where sRNAs are still
being identified and the Hfq-binding scaffold sequence is not clearly understood or identified,
designing synthetic, Hfq-dependent sRNAs is much more difficult. To engineer B. subtilis, Liu et
al. (105) overcame these hurdles by designing synthetic sRNAs based on the E. coli MicC
scaffold, and by coexpressing them with E. coli Hfq in B. subtilis. With this approach, the
authors repressed the expression of pfk (encoding 6-phosphofructokinase) and glmM (encoding
phosphoglucosamine mutase) to increase N-acetylglucosamine production. Cho et al. used the
same approach in Clostridium acetobutylicum PJC4BK to repress phosphotransacetylase
expression. Strain PJC4BK (pPta-HfqEco) produced 16.9 g L-1 of butanol, which was higher than
the starting strain as acetate production was inhibited (106). Research is underway to discover
new sRNAs in many bacterial species.

27

1.7 Identification of small RNAs in Pasteurelleaceae
Pasteurellaceae are a family of Gram-negative bacteria that are mostly commensals on the
mucosal surfaces of birds and mammals. The Pasteurellaceae family consists of eighteen genera,
some of the major ones being Haemophilus, Actinobacillus, and Pasteurella. Only a handful of
papers mention sRNAs in Pasteurelleaceae. Subashchandrabose et al. demonstrated the role of
Hfq in pathogenesis of Actinobacillus pleuropneumoniae. A Δhfq mutant of A.
pleuropneumoniae was defective in biofilm formation and sensitive to oxidative stress; however
not much more is known regarding the sRNAs involved (107). Three sRNAs have been
identified in Aggregatibacter actinomycetemcomitans that are iron- and fur-regulated (108).
Baddal et al. (109) identified seventeen novel putative sRNAs in intergenic regions, six of them
homologous to known sRNA families in Haemophilus influenzae. HrrF, a fur-regulated sRNA
was the first sRNA to be identified in any Haemophilus species (110).
More recently, the Bazzolli group conducted an extensive study to identify small RNAs in A.
pleuropneumoniae, an organism known to cause porcine pleuropneumonia. The authors used
four different algorithms to identify small RNAs and were able to experimentally confirm
seventeen of the 23 sRNAs found by all four algorithms by Northern blotting, RT-PCR, and
RNA sequencing. They also found that the sequences of these seventeen sRNAs were well
conserved in the species that are evolutionarily close to A. pleuropneumoniae (111). These early
studies have shed some much needed light in this unexplored area of sRNAs in Pasteurellaceae.
However, much remains to be discovered.

1.8 Introduction to chapters
The main aims of this dissertation are to:

28

(i) provide insight into small RNAs and make progress in using small RNAs as a valuable tool
for increasing succinate production in A. succinogenes.
(ii) make a promoter library for A. succinogenes for tunability of gene expression
(iii) carry out a proteomics- and transcriptomics-based search for succinate export transporters
and see the effect of their overexpression on succinate production.
Chapter 2 describes a proteomics- and transcriptomics-based search for succinate exporters
in A. succinogenes. Transporters with the most hits in our omics data were further tested by
overexpression for succinate production. Tunability for expressing these transporters was
required and so far no such system was available in A. succinogenes. For this purpose, a reporter
assay system was developed in A. succinogenes to test promoter strength. Several promoters
were tested for strength to generate a library of promoters of increasing strengths. The four
candidate succinate exporters identified were expressed under control of a few of the weaker
promoters to identify the transporter expression levels that would not interfere with A.
succinogenes growth and test their effect on succinate production. .
Chapter 3 describes a sequencing study of A. succinogenes small RNAs conducted on
cultures grown anaerobically on glucose and microaerobically on glycerol. The main aim of this
study was to find Hfq-dependent small RNAs and compare their sequences to be able to design a
synthetic small RNA for repression of mRNA in A. succinogenes.
Chapter 4 describes a method for making markerless knockout mutants in A. succinogenes
using natural transformation.
Chapter 5 describes the work I did as a part of a previously published paper. I developed a
continuous culture system for A. succinogenes growth on glycerol under microaerobic
conditions.

29

In chapter 6, I summarize what was learned about succinate transporters and sRNAs in A.
succinogenes. I also discuss future work for gaining more insight and further engineering of A.
succinogenes for succinate production.

30

REFERENCES

31

REFERENCES

1.

Werpy TA, Petersen G. 2004. Top value added chemicals from biomass for the U.S.
Department of Energy (DOE) by National Renewable Energy Laboratory (NREL).

2.

Patel MK, Crank M, Dornburg V, Hermann B, Roes L, Hüsing B, Overbeek L,
Terragni FR, E. 2006. The BREW project: Medium and Long-term Opportunities and
Risks of the Biotechnological Production of Bulk Chemicals from Renewable Resources–
the Potential of White Biotechnology. Technology and Society (STS)/Copernicus
Institute, Utrecht, the Netherlands.

3.

Bozell JJ, Petersen GR. 2010. Technology development for the production of biobased
products from biorefinery carbohydrates—the US Department of Energy’s “top 10”
revisited. Green Chem 12:539-554.

4.

WP 8.1. Determination of market potential for selected platform chemicals. Itaconic acid,
succinic acid, 2,5-furandicarboxylic acid. Weastra, S. R. O.

5.

Biddy MJ, Scarlata C, Kinchin C. 2016. Chemicals from Biomass: A Market
Assessment of Bioproducts with Near-Term Potential. NREL (National Renewable
Energy Laboratory, Golden, CO, USA.

6.

www.myriant.com. Accessed

7.

Sawisit A, Jantama SS, Kanchanatawee S, Jantama K. 2014. Efficient utilization of
cassava pulp for succinate production by metabolically engineered Escherichia coli
KJ122. Bioproc Biosyst Eng 38:175-187.

8.

Chen C, Ding S, Wang D, Li Z, Ye Q. 2014. Simultaneous saccharification and
fermentation of cassava to succinic acid by Escherichia coli NZN111. Bioresour Technol
163:100-105.

9.

Chatterjee R, Millard CS, Champion K, Clark DP, Donnelly MI. 2001. Mutation of
the ptsG gene results in increased production of succinate in fermentation of glucose by
Escherichia coli. Appl Environ Microbiol 67:148-154.

10.

Wang J, Zhu J, Bennett GN, San KY. 2011. Succinate production from sucrose by
metabolic engineered Escherichia coli strains under aerobic conditions. Biotechnol Prog
27:1242-1247.

11.

Liu R, Liang L, Cao W, Wu M, Chen K, Ma J, Jiang M, Wei P, Ouyang P. 2013.
Succinate production by metabolically engineered Escherichia coli using sugarcane
bagasse hydrolysate as the carbon source. Bioresour Technol 135:574-577.

32

12.

Borges ER, Pereira Jr N. 2011. Succinic acid production from sugarcane bagasse
hemicellulose hydrolysate by Actinobacillus succinogenes. J Ind Microbiol Biotechnol
38:1001-1011.

13.

Samuelov NS, Datta R, Jain MK, Zeikus JG. 1999. Whey fermentation by
Anaerobiospirillum succiniciproducens for production of a succinate-based animal feed
additive. Appl Environ Microbiol 65:2260-2263.

14.

Blankschien MD, Clomburg JM, Gonzalez R. 2010. Metabolic engineering of
Escherichia coli for the production of succinate from glycerol. Metab Eng 12:409-419.

15.

Schindler BD, Joshi RJ, Vieille C. 2014. Respiratory glycerol metabolism of
Actinobacillus succinogenes 130Z for succinate production. J Ind Microbiol Biotechnol
41:1339-1352.

16.

www.bio-amber.com.

17.

Chen Y, Nielsen J. 2016. Biobased organic acids production by metabolically
engineered microorganisms. Current Opin Biotechnol 37:165-172.

18.

Ahn J-H, Sang B-I, Um Y. 2011. Butanol production from thin stillage using
Clostridium pasteurianum. Bioresource Technol 102:4934-4937.

19.

Cavani F, Albonetti S, Basile F, Gandini AA. 2016. Chemicals and Fuels from BioBased Building Blocks. Wiley Online Library.

20.

Guettler MV, Rumler D, Jain MK. 1999. Actinobacillus succinogenes sp. nov., a novel
succinic-acid-producing strain from the bovine rumen. Int J Syst Bacteriol 49:207–216.

21.

Kuhnert P, Scholten E, Haefner S, Mayor D, Frey J. 2010. Basfia succiniciproducens
gen. nov., sp. nov., a new member of the family Pasteurellaceae isolated from bovine
rumen. Int J Syst Evol Microbiol 60:44-50.

22.

Lee PC, Lee SY, Hong SH, Chang HN. 2002. Isolation and characterization of a new
succinic acid-producing bacterium, Mannheimia succiniciproducens MBEL55E, from
bovine rumen. Appl Microbiol Biotechnol 58:663–668.

23.

Samuelov NS, Lamed R, Lowe SE, Zeikus JG. 1991. Influence of CO2-HCO3- levels
and pH on growth, succinate production, and enzyme activities of Anaerobiospirillum
succiniciproducens. Appl Environ Microbiol 57:3013–3019.

24.

Meynial‐Salles I, Dorotyn S, Soucaille P. 2008. A new process for the continuous
production of succinic acid from glucose at high yield, titer, and productivity. Biotechnol
Bioengin 99:129-135.

33

25.

Secchi C, Cantarelli VV, Pereira Fde S, Wolf HH, Brodt TC, Amaro MC, Inamine
E. 2005. Fatal bacteremia due to Anaerobiospirillum succiniciproducens: first description
in Brazil. Braz J Infect Dis 9:169-172.

26.

van der Werf MJ, Guettler MV, Jain MK, Zeikus JG. 1997. Environmental and
physiological factors affecting the succinate product ratio during carbohydrate
fermentation by Actinobacillus sp. 130Z. Arch Microbiol 167:332–342.

27.

McKinlay JB, Shachar-Hill Y, Zeikus JG, Vieille C. 2007. Determining Actinobacillus
succinogenes metabolic pathways and fluxes by NMR and GC-MS analyses of 13Clabeled metabolic product isotopomers. Metab Eng 9:177–192.

28.

McKinlay JB, Zeikus JG, Vieille C. 2005. Insights into Actinobacillus succinogenes
fermentative metabolism in a chemically defined growth medium. Appl Environ
Microbiol 71:6651-6656.

29.

McKinlay JB, Vieille C. 2008. 13C-metabolic flux analysis of Actinobacillus
succinogenes fermentative metabolism at different NaHCO3 and H2 concentrations.
Metab Eng 10:55–68.

30.

McKinlay JB, Laivenieks M, Schindler BD, Mckinlay AA, Siddaramappa S,
Challacombe JF, Lowry SR, Clum A, Lapidus AL, Burkhart KB, Harkins V, Vieille
C. 2010. A genomic perspective on the potential of Actinobacillus succinogenes for
industrial succinate production. BMC Genomics 11:680.

31.

Joshi RJ, Schindler BD, McPherson NR, Tiwari K, Vieille C. 2014. Development of a
markerless knockout method for Actinobacillus succinogenes. Appl Environ Microbiol
80:3053-3061.

32.

Carvalho M, Matos M, Roca C, Reis MAM. 2014. Succinic acid production from
glycerol by Actinobacillus succinogenes using dimethylsulfoxide as electron acceptor. N
Biotechnol 31:133-139.

33.

Van Heerden CD, Nicol W. 2013. Continuous succinic acid fermentation by
Actinobacillus succinogenes. Biochem Engin J 73:5-11.

34.

Bradfield MFA, Nicol W. 2014. Continuous succinic acid production by Actinobacillus
succinogenes in a biofilm reactor: steady-state metabolic flux variation. Biochem Engin J
85:1-7.

35.

Guettler M, Hanchar R, Kleff S, Jadhav S. 2015. Recombinant microorganisms for
producing organic acids.

36.

Vlysidis A, Binns M, Webb C, Theodoropoulos C. 2009. Utilisation of glycerol to
platform chemicals within the biorefinery concept: a case for succinate production. Chem
Eng Trans 18:537-542.

34

37.

Vlysidis A, Binns M, Webb C, Theodoropoulos C. 2011. Glycerol utilisation for the
production of chemicals: conversion to succinic acid, a combined experimental and
computational study. Biochem Eng J 58:1-11.

38.

Pateraki C, Almqvist H, Ladakis D, Liden G, Koutinas AA, Vlysidis A. 2016.
Modelling succinic acid fermentation using a xylose based substrate. Biochem Eng J
114:26-41.

39.

Salvachúa D, Mohagheghi A, Smith H, Bradfield MFA, Nicol W, Black BA, Biddy
MJ, Dowe N, Beckham GT. 2016. Succinic acid production on xylose-enriched
biorefinery streams by Actinobacillus succinogenes in batch fermentation. Biotechnol
Biofuels 9:28.

40.

Guarnieri MT, Chou YC, Salvachua D, Mohagheghi A, St John PC, Peterson DJ,
Bomble YJ, Beckham GT. 2017. Metabolic Engineering of Actinobacillus succinogenes
Provides Insights into Succinic Acid Biosynthesis. Appl Environ Microbiol 83.

41.

Zheng P, Dong J-J, Sun Z-H, Ni Y, Fang L. 2009. Fermentative production of succinic
acid from straw hydrolysate by Actinobacillus succinogenes. Bioresour Technol
100:2425-2429.

42.

Li F, Wang Y, Gong K, Wang Q, Liang Q, Qi Q. 2014. Constitutive expression of
RyhB regulates the heme biosynthesis pathway and increases the 5-aminolevulinic acid
accumulation in Escherichia coli. FEMS Microbiol Lett 350:209-215.

43.

Zheng P, Fang L, Xu Y, Dong J-J, Ni Y, Sun Z-H. 2010. Succinic acid production
from corn stover by simultaneous saccharification and fermentation using Actinobacillus
succinogenes. Bioresour Technol 101:7889-7894.

44.

Wan C, Li Y, Shahbazi A, Xiu S. 2008. Succinic acid production from cheese whey
using Actinobacillus succinogenes 130 Z. Appl Biochem Biotech 145:111-119.

45.

Lee SJ, Song H, Lee SY. 2006. Genome-based metabolic engineering of Mannheimia
succiniciproducens for succinic acid production. Appl Environ Microbiol 72:1939–1948.

46.

Choi S, Song H, Lim SW, Kim TY, Ahn JH, Lee JW, Lee M-H, Lee SY. 2016.
Highly selective production of succinic acid by metabolically engineered Mannheimia
succiniciproducens and its efficient purification. Biotechnol Bioeng 113:2168-2177.

47.

Lee JW, Yi J, Kim TY, Choi S, Ahn JH, Song H, Lee M-H, Lee SY. 2016. Homosuccinic acid production by metabolically engineered Mannheimia succiniciproducens.
Metab Eng 38:409-417.

48.

Pettigrew DW, Liu WZ, Holmes C, Meadow ND, Roseman S. 1996. A single amino
acid change in Escherichia coli glycerol kinase abolishes glucose control of glycerol
utilization in vivo. J Bacteriol 178:2846-2852.

35

49.

Lee KS. 2004. Anaerobic hydrogen production with an efficient carrier-induced granular
sludge bed bioreactor. Biotechnol Bioeng 87:648-657.

50.

Scholten E, Dägele D. 2008. Succinic acid production by a newly isolated bacterium.
Biotechnol Lett 30:2143-2146.

51.

Scholten E, Renz T, Thomas J. 2009. Continuous cultivation approach for fermentative
succinic acid production from crude glycerol by Basfia succiniciproducens DD1.
Biotechnol Lett 31:1947-1951.

52.

Becker J, Reinefeld J, Stellmacher R, Schäfer R, Lange A, Meyer H, Lalk M, Zelder
O, von Abendroth G, Schröder H. 2013. Systems-wide analysis and engineering of
metabolic pathway fluxes in bio‐succinate producing Basfia succiniciproducens.
Biotechnol Bioeng 110:3013-3023.

53.

Chatterjee R, Millard CS, Champion K, Clark DP, Donnelly MI. 2001. Mutation of
the ptsG gene results in increased production of succinate in fermentation of glucose by
Escherichia coli. Appl Environ Microbiol 67:148–154.

54.

Thakker C, Martínez I, San K-Y, Bennett GN. 2012. Succinate production in
Escherichia coli. Biotechnol J 7:213-224.

55.

Sawers RG, Clark DP. 2004. Fermentative pyruvate and acetyl-coenzyme A
metabolism. In Bock A (ed), Escherichia coli and Salmonella: Cellular and Molecular
Biology. ASM Press.

56.

Millard CS, Chao YP, Liao JC, Donnelly MI. 1996. Enhanced production of succinic
acid by overexpression of phosphoenolpyruvate carboxylase in Escherichia coli. Appl
Environ Microbiol 62:1808–1810.

57.

Gokarn RR, Eiteman MA, Altman E. 2000. Metabolic analysis of Escherichia coli in
the presence and absence of the carboxylating enzymes phosphoenolpyruvate
carboxylase and pyruvate carboxylase. Appl Environ Microbiol 66:1844–1850.

58.

Chatterjee T, Datta AG. 1973. Anaerobic formation of succinate from glucose and
bicarbonate in resting cells of Leishmania donovani. Exp Parasitol 33:138-146.

59.

Vemuri GN, Eiteman MA, Altman E. 2002. Succinate production in dual-phase
Escherichia coli fermentations depends on the time of transition from aerobic to
anaerobic conditions. J Ind Microbiol Biotechnol 28:325-332.

60.

Kim P, Laivenieks M, McKinlay J, Vieille C, Zeikus JG. 2004. Construction of a
shuttle vector for the overexpression of recombinant proteins in Actinobacillus
succinogenes. Plasmid 51:108–115.

36

61.

Jantama K, Haupt MJ, Svoronos SA, Zhang XL, Moore JC, Shanmugam KT,
Ingram LO. 2008. Combining metabolic engineering and metabolic evolution to develop
nonrecombinant strains of Escherichia coli C that produce succinate and malate.
Biotechnol Bioeng 99:1140-1153.

62.

Jantama K, Zhang XL, Moore JC, Shanmugam KT, Svoronos SA, Ingram LO.
2008. Eliminating side products and increasing succinate yields in engineered strains of
Escherichia coli C. Biotechnol Bioeng 101:881-893.

63.

Zhang XL, Jantama K, Moore JC, Jarboe LR, Shanmugam KT, Ingram LO. 2009.
Metabolic evolution of energy-conserving pathways for succinate production in
Escherichia coli. Proc Natl Acad Sci USA 106:20180-20185.

64.

Zhang XL, Jantama K, Shanmugam KT, Ingram LO. 2009. Reengineering
Escherichia coli for succinate production in mineral salts medium. Appl Environ
Microbiol 75:7807-7813.

65.

Zhang X, Shanmugam KT, Ingram LO. 2010. Fermentation of glycerol to succinate by
metabolically engineered strains of Escherichia coli. Appl Environ Microbiol 76:23972401.

66.

Inui M, Murakami S, Okino S, Kawaguchi H, Vertès AA, Yukawa H. 2004.
Metabolic analysis of Corynebacterium glutamicum during lactate and succinate
productions under oxygen deprivation conditions. J Mol Microbiol Biotechnol Bioeng
7:182-196.

67.

Okino S, Inui M, Yukawa H. 2005. Production of organic acids by Corynebacterium
glutamicum under oxygen deprivation. Appl Microbiol Biotechnol 68:475-480.

68.

Okino S, Noburyu R, Suda M, Jojima T, Inui M, Yukawa H. 2008. An efficient
succinic acid production process in a metabolically engineered Corynebacterium
glutamicum strain. Appl Microbiol Biotechnol 81:459-464.

69.

Litsanov B, Brocker M, Bott M. 2012. Toward homosuccinate fermentation: metabolic
engineering of Corynebacterium glutamicum for anaerobic production of succinate from
glucose and formate. Appl Environ Microbiol 78:3325-3337.

70.

Zhu N, Xia H, Yang J, Zhao X, Chen T. 2014. Improved succinate production in
Corynebacterium glutamicum by engineering glyoxylate pathway and succinate export
system. Biotechnol Lett 36:553-560.

71.

Litsanov B, Kabus A, Brocker M, Bott M. 2012. Efficient aerobic succinate production
from glucose in minimal medium with Corynebacterium glutamicum. Microb Biotechnol
5:116-128.

37

72.

Litsanov B, Brocker M, Bott M. 2013. Glycerol as a substrate for aerobic succinate
production in minimal medium with Corynebacterium glutamicum. Microb Biotechnol
6:189-195.

73.

Zhu N, Xia H, Wang Z, Zhao X, Chen T. 2013. Engineering of acetate recycling and
citrate synthase to improve aerobic succinate production in Corynebacterium
glutamicum. PLoS One 8:e60659.

74.

Chung SC, Park JS, Yun J, Park JH. 2017. Improvement of succinate production by
release of end-product inhibition in Corynebacterium glutamicum. Metab Eng 40:157164.

75.

Arikawa Y, Kobayashi M, Kodaira R, Shimosaka M, Muratsubaki H, Enomoto K,
Okazaki M. 1999. Isolation of sake yeast strains possessing various levels of succinateand/or malate-producing abilities by gene disruption or mutation. J Biosci Bioeng
87:333-339.

76.

Kubo Y, Takagi H, Nakamori S. 2000. Effect of gene disruption of succinate
dehydrogenase on succinate production in a sake yeast strain. J Biosci Bioeng 90:619624.

77.

Raab AM, Gebhardt G, Bolotina N, Weuster-Botz D, Lang C. 2010. Metabolic
engineering of Saccharomyces cerevisiae for the biotechnological production of succinic
acid. Metab Eng 12:518-525.

78.

Ito Y, Hirasawa T, Shimizu H. 2014. Metabolic engineering of Saccharomyces
cerevisiae to improve succinic acid production based on metabolic profiling. Biosci
Biotechnol Biochem 78:151-159.

79.

Yan D, Wang C, Zhou J, Liu Y, Yang M, Xing J. 2014. Construction of reductive
pathway in Saccharomyces cerevisiae for effective succinic acid fermentation at low pH
value. Bioresour Technol 156:232-239.

80.

Scholten E, Haefner S, Schröder H. 2014. Bacterial cells having a glyoxylate shunt for
the manufacture of succinic acid. US Patent 8,877,466 B2.

81.

Schroder H, Haefner S, Von Abendroth G, Hollmann R, Raddatz A, Ernst H,
Gurski H. 2014. Microbial succinic acid producers and purification of succinic acid.

82.

Janausch IG, Zientz E, Tran QH, Kroger A, Unden G. 2002. C4-dicarboxylate
carriers and sensors in bacteria. Biochim Biophys Acta 1553:39-56.

83.

Zientz E, Six S, Unden G. 1996. Identification of a third secondary carrier (DcuC) for
anaerobic C4-dicarboxylate transport in Escherichia coli: roles of the three Dcu carriers
in uptake and exchange. J Bacteriol 178:7241-7247.

38

84.

Karinou E, Compton EL, Morel M, Javelle A. 2013. The Escherichia coli SLC26
homologue YchM (DauA) is a C4-dicarboxylic acid transporter. Mol Microbiol 87:623640.

85.

Six S, Andrews SC, Unden G, Guest JR. 1994. Escherichia coli possesses two
homologous anaerobic C4-dicarboxylate membrane transporters (DcuA and DcuB)
distinct from the aerobic dicarboxylate transport system (Dct). J Bacteriol 176:64706478.

86.

Davies SJ, Golby P, Omrani D, Broad SA, Harrington VL, Guest JR, Kelly DJ,
Andrews SC. 1999. Inactivation and regulation of the aerobic C4-dicarboxylate transport
(dctA) gene of Escherichia coli. J Bacteriol 181:5624-5635.

87.

Youn JW, Jolkver E, Kramer R, Marin K, Wendisch VF. 2009. Characterization of
the dicarboxylate transporter DctA in Corynebacterium glutamicum. J Bacteriol
191:5480-5488.

88.

Fukui K, Nanatani K, Hara Y, Yamakami S, Yahagi D, Chinen A, Tokura M, Abe
K. 2017. Escherichia coli yjjPB genes encode a succinate transporter important for
succinate production. Biosci Biotechnol Biochem 81:1837-1844.

89.

Chen J, Zhu X, Tan Z, Xu H, Tang J, Xiao D, Zhang X. 2014. Activating C4dicarboxylate transporters DcuB and DcuC for improving succinate production. Appl
Microbiol Biotechnol 98:2197-2205.

90.

Kelly DJ, Thomas GH. 2001. The tripartite ATP-independent periplasmic (TRAP)
transporters of bacteria and archaea. FEMS Microbiol Rev 25:405-424.

91.

Mulligan C, Fischer M, Thomas GH. 2011. Tripartite ATP-independent periplasmic
(TRAP) transporters in bacteria and archaea. FEMS Microbiol Rev 35:68-86.

92.

Pos KM, Dimroth P, Bott M. 1998. The Escherichia coli citrate carrier CitT: a member
of a novel eubacterial transporter family related to the 2-oxoglutarate/malate translocator
from spinach chloroplasts. J Bacteriol 180:4160-4165.

93.

Huhn S, Jolkver E, Kramer R, Marin K. 2011. Identification of the membrane protein
SucE and its role in succinate transport in Corynebacterium glutamicum. Appl Microbiol
Biotechnol 89:327-335.

94.

Fukui K, Koseki C, Yamamoto Y, Nakamura J, Sasahara A, Yuji R, Hashiguchi K,
Usuda Y, Matsui K, Kojima H, Abe K. 2011. Identification of succinate exporter in
Corynebacterium glutamicum and its physiological roles under anaerobic conditions. J
Biotechnol 154:25-34.

39

95.

Rhie MN, Yoon HE, Oh HY, Zedler S, Unden G, Kim OB. 2014. A Na+-coupled C4dicarboxylate transporter (Asuc_0304) and aerobic growth of Actinobacillus
succinogenes on C4-dicarboxylates. Microbiology 160:1533-1544.

96.

Na D, Yoo SM, Chung H, Park H, Park JH, Lee SY. 2013. Metabolic engineering of
Escherichia coli using synthetic small regulatory RNAs. Nat Biotechnol 31:170-174.

97.

Kim JYH, Cha HJ. 2003. Down-regulation of acetate pathway through antisense
strategy in Escherichia coli: improved foreign protein production. Biotechnol Bioeng
83:841-853.

98.

Nakashima N, Tamura T, Good L. 2006. Paired termini stabilize antisense RNAs and
enhance conditional gene silencing in Escherichia coli. Nucleic Acids Res 34:e138.

99.

Negrete A, Majdalani N, Phue J-N, Shiloach J. 2013. Reducing acetate excretion from
E. coli K-12 by over-expressing the small RNA SgrS. N Biotechnol 30:269-273.

100.

Coleman J, Green PJ, Inouye M. 1984. The use of RNAs complementary to specific
mRNAs to regulate the expression of individual bacterial genes. Cell 37:429-436.

101.

Kang Z, Wang X, Li Y, Wang Q, Qi Q. 2012. Small RNA RyhB as a potential tool
used for metabolic engineering in Escherichia coli. Biotechnol Lett 34:527-531.

102.

Ding W, Weng H, Du G, Chen J, Kang Z. 2017. 5-Aminolevulinic acid production
from inexpensive glucose by engineering the C4 pathway in Escherichia coli. J Ind
Microbiol Biotechnol 44:1127-1135.

103.

Desai RP, Papoutsakis ET. 1999. Antisense RNA strategies for metabolic engineering
of Clostridium acetobutylicum. Appl Environ Microbiol 65:936-945.

104.

Boudry P, Gracia C, Monot M, Caillet J, Saujet L, Hajnsdorf E, Dupuy B, MartinVerstraete I, Soutourina O. 2014. Pleiotropic role of the RNA chaperone protein Hfq in
the human pathogen Clostridium difficile. J Bacteriol 196:3234-3248.

105.

Liu Y, Zhu Y, Li J, Shin H-D, Chen RR, Du G, Liu L, Chen J. 2014. Modular
pathway engineering of Bacillus subtilis for improved N-acetylglucosamine production.
Metab Eng 23:42-52.

106.

Cho C, Lee SY. 2016. Efficient gene knockdown in Clostridium acetobutylicum by
synthetic small regulatory RNAs. Biotechnol Bioeng 114:374-383.

107.

Subashchandrabose S, Leveque RM, Kirkwood RN, Kiupel M, Mulks MH. 2013.
The RNA chaperone Hfq promotes fitness of Actinobacillus pleuropneumoniae during
porcine pleuropneumonia. Infect Immun 81:2952-2961.

40

108.

Amarasinghe JJ, Connell TD, Scannapieco FA, Haase EM. 2012. Novel iron‐
regulated and Fur‐regulated small regulatory RNAs in Aggregatibacter
actinomycetemcomitans. Mol Oral Microbiol 27:327-349.

109.

Baddal B, Muzzi A, Censini S, Calogero RA, Torricelli G, Guidotti S, Taddei AR,
Covacci A, Pizza M, Rappuoli R. 2015. Dual RNA-seq of nontypeable Haemophilus
influenzae and host cell transcriptomes reveals novel insights into host-pathogen cross
talk. mBio 6:e01765-15.

110.

Santana EA, Harrison A, Zhang X, Baker BD, Kelly BJ, White P, Liu Y, Munson Jr
RS. 2014. HrrF is the Fur-regulated small RNA in nontypeable Haemophilus influenzae.
PloS one 9:e105644.

111.

Rossi CC, Bossé JT, Li Y, Witney AA, Gould KA, Langford PR, Bazzolli DMS.
2016. A computational strategy for the search of regulatory small RNAs in Actinobacillus
pleuropneumoniae. RNA 22:1373-1385.

41

Chapter 2

Identification of succinate transporters in Actinobacillus succinogenes

42

2.1 Abstract
Succinate, listed as one of the top 12 building block chemicals by the U.S. Department of
Energy has been produced from maleic anhydride, a petrochemical until recently. A few
companies, including BioAmber, Myriant, and Reverdia, have started producing industrial scale
amounts of bio-based succinate using yeast and E. coli strains. A. succinogenes, along with other
succinate producing bacteria have been studied and engineered with the aim of increasing
succinate production. Succinate exporters have been studied in very few of these bacteria,
especially with the aim of using them as targets in engineering for increased succinate
production. E. coli and C. glutamicum are the only strains where succinate transporters have
been studied. Several succinate exporter candidates have been overexpressed in E. coli and in C.
glutamicum and caused an increase in succinate efflux. So far, A. succinogenes has not been
studied for succinate exporters.
In this study, we sought to identify succinate transporters using a combined proteomics and
transcriptomics approach. The proteins with the most hits in our proteomics studies—
Asuc_1999, Asuc_0142, Asuc_2058 and Asuc_1990-91 were selected for overexpression in the
wild-type strain. Since we are limited by our only expression vector which harbors the native,
strong, and constitutive promoter ppckA, we developed a library of expression vectors with
promoters weaker than ppckA by truncating ppckA and searching for additional promoter candidates
from our RNA sequencing analysis. We looked for candidate genes that were expressed at a
similar level across growth conditions tested and that had several fold lower transcript levels than
pckA. Finally, we overexpressed our candidate succinate exporters under the control of
promoter’s ppckA-92 and ppckA-103, and observed an increase in succinate production in all the
strains tested as compared to the control strain. The highest increase was seen when Asuc_1999

43

was expressed under the control of ppckA-103. To the best of our knowledge this is the first study
focusing on identifying succinate exporters in A. succinogenes. This is also the first report of a
promoter library being constructed for A. succinogenes, further increasing the tunability of gene
expression.

2.2 Introduction
Succinate is a dicarboxylate molecule with many possible applications. It can be used as a
precursor to make products such as nylons, resins, paints, cosmetics, flavors, and adhesives (1).
The conventional route to produce succinate has been by hydrogenation of maleic anhydride, a
petroleum derivative, in the presence of a metal catalyst (2). Bio-based succinate production has
gained momentum in recent years, though, as awareness about the benefits of renewable
products and downsides of burning fossil fuels have increased. A number of companies, such as
BioAmber, Myriant, and Succinity, have started selling bio-based succinate produced by
engineered microorganisms (e.g., Escherichia coli, Basfia succiniciproducens, and yeast). Biobased succinate is produced naturally by anaerobes such as Actinobacillus succinogenes and
Mannheimia succiniproducens via the reductive pathway of the tricarboxylic acid cycle.
Industrial workhorse microorganisms such as E. coli, Saccharomyces cerevisiae, and
Corynebacterium glutamicum also have been or are being engineered to produce succinate (2-5).
One of the major hurdles still lies in making bio-based succinate as cost–effective as
conventionally produced succinate.
A. succinogenes is one of the best natural succinate producers, yet our knowledge of its
uptake and efflux mechanisms remains limited. For example, the system or systems involved in
glucose uptake as well as the one or ones involved in succinate export remain unknown, and the
succinate export system has not yet been targeted for increasing succinate production.

44

Several succinate transporters have been identified and characterized in E. coli, but most
studies have focused on uptake and antiport transporters. The dicarboxylate uptake (Dcu) carriers
DcuA, B, and C are the best-characterized succinate transporters. DcuA and DcuB are involved
in fumarate: succinate antiport during anaerobic respiration on fumarate. They are also capable
of uptake and efflux of C4-dicarboxylates (6, 7). Zeintz et al. (8) suggested that, even though
DcuC can carry out the same transport functions as DcuA and DcuB, its major function likely is
succinate export. E. coli dcuD shows similarity to dcuC, but DcuD mutants show no effect on E.
coli growth under a number of conditions, and do not affect succinate uptake or efflux (9). DauA
is the sole transporter for aerobic succinate uptake at acidic pH (10). Another transporter, CitT,
catalyzes citrate/succinate antiport during citrate fermentation (11). DctA, a H+/C4-dicarboxylate
symporter is responsible for C4-dicarboxylate uptake in aerobic conditions (6, 12) at neutral pHs.
In one report, a quintuple dctA dcuA dcuB dcuC citT E. coli mutant did not show any deficiency
in succinate efflux as compared to the wild-type strain in glucose-grown anaerobic cultures (13).
In another report, while deleting dcuA and dcuD did not have any effect on succinate efflux,
individual dcuB and dcuC deletion mutants showed decreases in succinate titers of 15% and
11%, respectively (14). A double dcuBC deletion resulted in a 90% decrease in succinate titer,
and modulating expression of both genes increased the succinate titer by 34% (14). Thus no clear
picture of the involvement of Dcu transporters in succinate efflux is available. DauA is the sole
characterized E. coli succinate transporter that has not been tested for succinate efflux during
glucose fermentation.
Two studies using different approaches identified the same succinate exporter, SucE, of
the aspartate:alanine exchanger family, in Corynebacterium glutamicum (15, 16). This
transporter functioned in anaerobic and microaerobic conditions and was responsible for about

45

30% of the succinate export in microaerobic conditions, indicating that SucE is not the sole
succinate exporter in C. glutamicum. In a recent study aimed at identifying the E. coli anaerobic
succinate exporter, C. glutamicum strain AJ110655 ΔsucE, deficient in succinate export, was
partially rescued for succinate export by expressing the E. coli yjjPB genes (17). Additionally, an
E. coli yjjPB knockout mutant produced 70% less succinate than the parental strain, indicating
that YjjPB may be one of the main E. coli anaerobic succinate exporters. There has been no
report on whether or not the sole E. coli SucE homolog participates in succinate export.
In this study, we conducted transcriptomic studies of A. succinogenes grown on multiple
carbon sources as well as proteomics studies on cytoplasmic membrane-enriched fractions of A.
succinogenes grown anaerobically on glucose, xylose, and fructose (conditions favoring
succinate production) or on glycerol-nitrate (less succinate is produced). Of the putative C4
dicarboxylate transporters identified in the annotated genome, locus tags Asuc_1999,
Asuc_1990-1991, Asuc_0142, and Asuc_2058 had the most hits in our proteomics and
transcriptomics results, and we focused on these four transporters for further analysis. Currently,
to overexpress A. succinogenes genes we only have one strong constitutive native promoter
(ppckA, (18) at our disposal, which is not optimal for the expression of membrane proteins. We
constructed a library of promoters of varying strengths to overcome this problem and modulate
the expression of these transporters. We show that overexpression of any of these four putative
C4-dicarboxylate transporters significantly increases succinate production.

46

2.3 Materials and Methods
2.3.1 Strains, media, and culture conditions.
E. coli strains (Table 2.1) were cultivated in lysogeny broth (LB) and on LB agar plates and
were supplemented with 100 µg mL-1 ampicillin or 50 µg mL-1 kanamycin when required for
plasmid maintenance (19). A. succinogenes type strain 130Z was obtained from the American
Type Culture Collection. A. succinogenes liquid cultures were grown in AM3 defined medium
(20), Medium B (21), or Bacto brain heart infusion medium (BHI; Becton, Dickinson and Co.,
Franklin Lakes, NJ). Mutant strains of A. succinogenes were grown on AM2-isocitrate or AM16isocitrate after natural transformations (21). Liquid cultures were grown anaerobically in 28-mL
anaerobic tubes flushed with N2 at 37°C with shaking at 250 rpm. Optical densities were
measured at 660 nm using a Spectronic 20 (Bausch and Lomb, Rochester, NY) or a DU650
spectrophotometer (Beckman, Fullerton, CA). To isolate A. succinogenes colonies, strains were
grown on LB plates containing 10 g L-1 glucose (22) and 10 µg mL-1 kanamycin (Km), 600 mg
L-1 polyvinyl alcohol (PVA) and 10 mg L-1 CaCl2. Agar plate cultures of A. succinogenes strains
were grown under a CO2-enriched atmosphere for 24-48 h. All strains used in this study are
listed in Table 2.1.

2.3.2 Plasmids, DNA manipulations, and electroporations.
PCR products were cloned into pCR2.1 using the TA TOPO cloning kit (Invitrogen, Carlsbad,
CA) and transformed into One Shot TOP10 chemically competent cells (Invitrogen) as per
manufacturer’s protocols. Shuttle plasmid pLGZ920 was used to express native or foreign genes
in A. succinogenes (18). Electrocompetent A. succinogenes cells, prepared as described (21),
were transformed with plasmid constructs and spread on LB Glucose PVA CaCl2 Km plates.

47

DNA manipulations were carried out according to Ausubel et al. (19). PCR products were
amplified using Phusion High-Fidelity DNA polymerase (New England Biolabs, Ipswich, MA).
Restriction enzymes were from New England Biolabs. Plasmids and DNA fragments were
purified using the Wizard SV miniprep kit and Wizard SV Gel and PCR Clean-Up system,
respectively, (Promega, Madison, WI). Primers used in this work were synthesized by Integrated
DNA Technologies (Coralville, IA) and are listed in Table S2.1 in supplementary material. DNA
sequencing was performed by GENEWIZ, Inc. (South Plainfield, NJ). Colonies were screened
by colony PCR using Taq polymerase.

Table 2.1 Strains and plasmids used in this study
Description

Source

DH5α

F- φ80lacZΔM15 Δ(lacZYA-argF) U169 recA1 endA1 hsdR17

Laboratory
collection

TOP10

(rk-, mk+) phoA supE44 λ- thi-1 gyrA96 relA1 F- mcrA Δ(mrrhsdRMS-mcrBC) φ80lacZΔM15 ΔlacΧ74 recA1 araD139 Δ(ara-leu)
7697 galU galK rpsL (StrR) endA1 nupG λ-

Invitrogen

130Z (ATCC55618)

Wild-type strain

ATCC

ΔlacZ::icd

130Z derivative, contains the AscI-FRT-ppckA-icd-FRT-AscI cassette in
the lacZ deletion

This study

ΔlacZ

130Z derivative, lacZ deletion, contains one FRT site

This study

pCR2.1 TOPO

AmpR, KmR, lacZα, cloning vector

Invitrogen

pCR2.1-ΔlacZ

pCR2.1 derivative, lacZ deletion (2.1-kb fusion product of frdup and
frdCD)

(21)

pCR2.1-icd

pCR2.1 derivative, E. coli icd under control of A. succinogenes ppckA,
flanked by FRT repeats and AscI restriction sites

(21)

pCV933

pLGZ920 derivative, S. cerevisiae flp under control of ppckA

(21)

pLGZ920-lacZ

pLGZ920 derivative containing lacZ under ppckA

This study

pLGZ920-ppckA-92-lacZ

pLGZ920 derivative containing lacZ under ppckA-92

This study

E. coli

A. succinogenes

Plasmids

48

Table 2.1 (cont’d)
pLGZ920-ppckA-103-lacZ

pLGZ920 derivative containing lacZ under ppckA-103

This study

pLGZ920-ppckA-134-lacZ

pLGZ920 derivative containing lacZ under ppckA-134

This study

pLGZ920-ppckA-164-lacZ

pLGZ920 derivative containing lacZ underppckA-164

This study

pLGZ920*-pAsuc_0701

pLGZ920 derivative with the HindIII site 3’ of ColE1 ori replaced with This study
BamHI and ppckA-0701

pLGZ920*-pAsuc_0701-lacZ

pLGZ920* derivative containing lacZ under pAsuc_0701

This study

pLGZ920*-pAsuc_2109-lacZ

pLGZ920* derivative containing lacZ under pAsuc_2109

This study

pLGZ920*-pAsuc_0391-lacZ

pLGZ920* derivative containing lacZ under pAsuc_0391

This study

pLGZ920*-pAsuc_0289-lacZ

pLGZ920* derivative containing lacZ under pAsuc_0289

This study

2.3.3 Preparation of A. succinogenes cytoplasmic membrane-enriched fractions for
proteomics analysis.
A. succinogenes cultures in AM3 glucose, fructose, and xylose were grown in 300-mL
volumes and harvested in the middle of the exponential phase at an OD660 of 0.8 to 0.9. Because
A. succinogenes does not grow fermentatively on glycerol, glycerol cultures were grown by
nitrate respiration. The glycerol-nitrate cultures were grown in 600-mL volumes and harvested at
an OD660 no higher than 0.4 to 0.5 to avoid succinate accumulation after nitrate depletion
(Schindler et al, 2014). Cytoplasmic membrane-enriched fractions were prepared using a method
adapted from (23). Cells were harvested by centrifugation at 12,000 × g for 10 min and washed
twice with 100 mM Tris-HCl (pH 8.0). Bacterial pellets were resuspended in 15 mL of buffer A
(100 mM Tris-HCl [pH 8.0] containing 20% sucrose and Complete Mini, EDTA-free, protease
inhibitor cocktail [Sigma-Aldrich]). ReadyLyse lysozyme (EpiCentre, Madison, WI) was added

49

at a final concentration of 104 U mL-1, and bacterial suspensions were stirred on a magnetic plate
at the lowest setting for 8 min. Ethylenediaminetetraacetic acid (EDTA, 1.5 mL of 100 mM, pH
8.0) (1:10 v/v EDTA:cells) was added drop by drop over 2.5 min to avoid lysis, and suspensions
were stirred for 8 more min. The outer membrane and periplasmic fractions were separated from
the spheroplasts by centrifugation at 12,000 × g for 10 min.
The pellets, corresponding to the spheroplast-enriched fractions, were washed once with 15
mL buffer A, and resuspend in 15 mL buffer A containing 225,000 U DNase I and 0.2 mg RNase
A. The spheroplasts were lysed by passing twice though a French press at 1,400 psi (high
setting). The intact spheroplasts were removed by centrifugation at 12,000 × g for 10 min. The
supernatants were then ultracentrifuged at 150,000 × g for 2 h at 4°C. The pellets were
resuspended in buffer A and ultracentrifuged again to wash off the cytoplasmic proteins. The
cytoplasmic membrane-enriched fractions were finally resuspended in 100 mM Tris-HCl (pH
8.0) containing 1 mM EDTA and protease inhibitors to a concentration of 4 mg mL-1.

2.3.4 Proteomic analysis.
To submit the cytoplasmic membrane-enriched fractions for proteomic analysis, 50 µl of
extracts (i.e., 200 µg protein) were loaded on sodium dodecyl sulfate-polyacrylamide gels (miniPROTEAN TGX gel, 10% acrylamide, Bio-Rad, Hercules, CA) and subjected to electrophoresis
until the samples had entered the stacking gel. The proteins were visualized by Coomassie blue
staining. Gel bands were digested in-gel according to Shevchenko et al.1 with modifications.
Briefly, gel bands were dehydrated using 100% acetonitrile and incubated with 10 mM
dithiothreitol in 100 mM ammonium bicarbonate (pH 8.0) at 56 °C for 45 min, dehydrated again
and incubated in the dark with 50 mM iodoacetamide in 100 mM ammonium bicarbonate for 20

50

min. Gel bands were then washed with 100 mM ammonium bicarbonate and dehydrated again.
Sequencing-grade modified trypsin (50 μL at 0.01 μg μL-1 in 50 mM ammonium bicarbonate)
was added to each gel band to completely submerge the band. Bands were then incubated at 37
°C overnight. Peptides were extracted from the gel by water bath sonication in a solution of 60%
acetonitrile/1% trichloroacetic acid and vacuum-dried to ~2μL.
Peptides were then re-suspended in 2% acetonitrile/0.1% trifluoroacetic acid to 25 μL.
Five μL were automatically injected by a Thermo (www.thermo.com) EASYnLC 1000 onto a
Thermo Acclaim PepMap RSLC 0.075 mm x 250 mm C18 column and eluted over 185 min with
a 174-min gradient of 5% B to 25% B. The gradient was then raised to 100% B in 1 min and
held at 100% B for the duration of the run (Buffer A = 99.9% water/0.1% formic acid, Buffer B
= 99.9% acetonitrile/0.1% formic Acid).
Eluted peptides were sprayed into a Q-Exactive mass spectrometer (Thermo Fisher
Scientific, Waltham, MA) using a FlexSpray spray ion source. Survey scans were taken in the
Orbi trap (35,000 resolution, determined at m/z 200) and the top ten ions in each survey scan
were subjected to automatic higher energy collision induced dissociation with fragment spectra
acquired at 17,500 resolution.
The resulting MS/MS spectra were converted to peak lists using Mascot Distiller, v2.5.1
(www.matrixscience.com) and searched against a protein sequence database containing entries
for A. succinogenes (downloaded on May 22, 2015 from NCBI, www.ncbi.nlm.nih.gov) and
appended with common laboratory contaminants (downloaded from www.thegpm.org, cRAP
project) using the Mascot searching algorithm, v 2.5. The following parameters were used: allow
up to two missed tryptic sites, fixed modification of carbamidomethyl cysteines, variable
modification of methionine oxidation and of asparagine and glutamine deamidation; +/- 5 ppm

51

peptide tolerance, 0.3 Da MS/MS tolerance, and false discovery rate (FDR) calculated using
randomized database search. The Mascot output was then analyzed using Scaffold, v4.4.3
(www.proteomesoftware.com) to probabilistically validate protein identifications. Assignments
validated using the Scaffold 1% FDR confidence filter were considered true. NSAF values (24)
were calculated within Scaffold for protein quantitation.

2.3.5 Total RNA purification and RNA sequencing
Total RNA was purified from A. succinogenes cultures and quality-controlled as described
(25). In short, cultures harvested in the middle of the exponential phase were immediately mixed
with one volume of -20 °C methanol and stored at -20 °C for at least 30 min to quench all
enzymatic activity. After centrifugation, RNA was purified from the cell pellets with the
QIAGEN RNeasy Midi Kit (QIAGEN, Hilden, Germany), including the optional on-column
DNase digestion. RNA samples were quantified on a Qubit Fluorometer (Life Technologies,
Grand Island, NY), and RNA quality was validated visually on a 1% agarose gel. RNA was
mixed 50% v/v with formaldehyde before loading on the gel.
Total RNA sequencing was performed by the Michigan State University Research
Technology Support Facility as described (25). In short, libraries were prepared using the
Illumina TruSeq Stranded Total RNA Library Preparation kit and rRNAs were depleted with the
Illumina Ribo-Zero Gram-Negative Bacteria kit. Libraries were validated and quantified using
the Qubit dsDNA assay, Caliper LabChipGX, and Kapa Illumina Library Quantification qPCR
kits. The sequenced libraries (three biological replicates each) were from A. succinogenes 130Z
fermenting glucose, xylose, and fructose, as well as A. succinogenes respiring glucose with
nitrate as the electron acceptor, glycerol with nitrate as the electron acceptor, and glycerol in

52

microaerobic conditions. Pools of 12 libraries were loaded on separate lanes of an Illumina
HiSeq 2500 High Output (v4) flow cell and sequenced in 1x50bp single end format using HiSeq
SBS reagents. Base calling was done by Illumina Real Time Analysis v1.18.64. Output was
demultiplexed and converted to FastQ format using Illumina Bcl2fastq v1.8.4. Sequencing
results were analyzed using the SPARTA pipeline (26).

2.3.6 Identification of A. succinogenes’s pckA transcription start site.
The transcriptional start site for pckA was identified with First Choice RNA ligasemediated rapid amplification of cDNA ends (RLM-RACE) kit (Ambion, Inc.) following the
manufacturer’s instructions. DNase-treated total RNA (10 µg) extracted from microaerobic
glycerol cultures (27) was treated with calf intestine phosphatase. The RNA was phenolchloroform extracted and then treated with tobacco acid pyrophosphatase (TAP) to remove the
cap from full-length mRNAs. TAP-treated RNA was then used for 5’ RACE adapter ligation.
The sample was reverse-transcribed and used as the template for nested PCR. Outer 5’RLMRACE PCR was carried out using the 5’ RACE outer primer and pckA-specific outer primer. The
outer PCR product was used as the template for carrying out the inner 5’RLM-RACE PCR with
the 5’RACE inner primer and pckA-specific inner primer. A-overhangs were added to the inner
PCR product before cloning into pCR2.1. Colonies were screened by PCR using the inner
5’RLM-RACE PCR primers, and insert sequences were verified by sequencing.

2.3.7 Construction of truncated pckA promoters.
Plasmid pLGZ920-lacZ was constructed to use lacZ as the reporter gene to test the
strength of truncated pckA promoters. Primers CV624 and CV625 were used to amplify lacZ

53

from A. succinogenes 130Z genomic DNA. The PCR product was cloned into pLGZ920 under
control of the ppckA promoter by Gibson Assembly (New England Biolabs, Ipswich, MA). The
cloning reaction was transformed into cells provided with the kit. Colonies were screened by
PCR with primers GZ1302 and GZ1303, and the insert sequence was verified by sequencing.
Expression of lacZ from this plasmid construct was verified by β-galactosidase assay of
recombinant strain ΔlacZ (pLGZ920-lacZ) (see below).
The ColE1 origin of replication (ori) was amplified from pLGZ920 using primers CV469
and CV471. Primer pairs CV473-CV443, CV474-CV443, CV475-CV443, and CV476-CV443
were used to amplify truncated pckA promoters ppckA-92, ppckA-103, ppckA-134, and ppckA-164 with
pLGZ920-lacZ as the template. The ColE1 ori and truncated pckA promoters were fused by PCR
using primers CV469 and CV470. The four fusion PCR products were gel purified and cloned
into the pLGZ920-lacZ HindIII and XbaI sites by Gibson assembly (New England Biolabs,
Ipswich, MA) as described (Gibson et al., 2009). Note that restriction by HindIII removes the
ColE1 ori initially present in pLGZ920. The cloning reactions were transformed into TOP10
cells. Colonies were screened by PCR using primers CV565 and CV568. Positive plasmids were
verified by sequencing. The constructs were then transformed into strain ΔlacZ.

2.3.8 Construction of an A. succinogenes ΔlacZ mutant strain.
The A. succinogenes ΔlacZ mutant strain was constructed as described (21). Briefly, the
ΔlacZ::icd strain was built by natural transformation of A. succinogenes 130Z with a ΔlacZ::icd
linear DNA fragment. Double recombinants were screened by colony PCR using primers CV376
and CV377. The icd marker was excised by the yeast flippase recombinase carried by plasmid
pCV933. Transformants were screened for the loss of icd by colony PCR using primers CV376

54

and CV377. Plasmid pCV933 was cured from the ΔlacZ(pCV933) strain using acridine orange
as described (21). Strain ΔlacZ was used as the host strain to test the promoter strength of the
reporter promoter library constructs.

2.3.9 Preparation of crude extracts, β-galactosidase assays, and protein assays.
Ten-mL cultures were grown in AM3 supplemented with 50 mM glucose and 150 mM
NaHCO3. Cultures were harvested in exponential phase by centrifugation, washed, and
resuspended in 0.5 mL of 50 mM sodium phosphate buffer (pH 7.0). Cells were lysed by
sonication (Branson S-450A probe sonifier, Danbury, CT) with six 20-sec repetitions, 50% duty
cycle, and a power level of 3. Cell extracts were centrifuged for 5 min (2,000 × g) at room
temperature and the supernatant was stored on ice until used.
Beta-galactosidase assays were conducted in 96-well plates in a PowerWave HT microplate
reader at room temperature (BioTek Instruments Inc., Winooski, VT). Reactions (200 µl)
contained 20 µl cell extract mixed with 150 µl of Z-buffer (60 mM Na2HPO4 , 40 mM NaH2PO4,
10 mM KCl, 1 mM MgSO4, 38 mM 2-mercaptoethanol) and 30 µL of 2-nitrophenyl β-Dgalactopyranoside (4 mg mL-1) in Z-buffer. Enzyme activity was calculated using the linear slope
obtained from 2-nitrophenol production recorded at 420 nm. The extinction coefficient of 2nitrophenol was 4.8 mM-1 cm-1. Total cell protein was quantified using the Bio-Rad protein assay
dye reagent using bovine serum albumin as the standard. Activities were reported as an average
of three biological replicates.

2.3.10 Construction of a promoter library based on RNAseq results.
An initial promoter-lacZ reporter plasmid was assembled in pLGZ920 using the promoter

55

of Asuc_0701 (pAsuc_0701). Other promoter-lacZ reporter plasmids were then constructed by
substituting pAsuc_0701 with other promoters. First, ColE1 ori was amplified using primers CV615
and CV616 with pLGZ920 as the template. A. succinogenes 130Z genomic DNA was used as the
template to amplify pAsuc_0701 with primers CV 617 and CV618. ColE1 ori and ppckA were
removed from pLGZ920 by digestion with HindIII and XbaI. ColE1 ori and pAsuc_0701 were then
cloned into pLGZ920’s HindIII and XbaI sites by Gibson Assembly to create plasmid
pLGZ920*- pAsuc_0701. In this plasmid, the HindIII site 3’ of ColE1 ori is replaced with a BamHI
site to facilitate promoter replacement. Colonies were screened by PCR with primers CV615 and
CV616 and positive plasmids were verified by sequencing. The lacZ gene was amplified from A.
succinogenes 130Z genomic DNA using primers CV624 and CV625, and cloned into pLGZ920*
-pAsuc_0701’s XbaI and SacI sites by Gibson cloning. Colonies were screened by PCR using
primers CV636 and CV640. A plasmid with the correct lacZ sequence (sequencing with primers
listed in Table S2.1) was called pLGZ920*-pAsuc_0701-lacZ.
Promoters pAsuc_2109, pAsuc_0391, and pAsuc_0289 were amplified using primers CV 569 and
CV570, CV571 and CV572, and CV578 and CV579, respectively (Table S2.1), cut with BamHI
and XbaI, and cloned into pLGZ920*-pAsuc_0701-lacZ’s BamHI and XbaI sites (these restrictions
excise pAsuc_0701) by ligation using T4 DNA ligase (New England Biolabs, Ipswich, MA).
Colonies were screened using CV615 and the reverse primers used for amplifying each promoter
fragment. Positive clones were verified by sequencing. All four plasmid constructs were then
transformed into strain ΔlacZ.

2.3.11 Construction of strains expressing putative succinate transporters.
Asuc_0142, Asuc_1990-1991, Asuc_1999, and Asuc_2058 were amplified from A.
succinogenes genomic DNA using primers CV579 and CV580, CV582 and CV586, CV587 and

56

CV588, and CV589 and CV590, respectively. The four PCR fragments were cloned in
pLGZ920-ppckA-92-lacZ and pLGZ920-ppckA-103-lacZ cut with XbaI and SacI (which remove lacZ)
by Gibson cloning. Colonies were screened by PCR with GZ1302 and GZ1303 and positive
clones were verified by sequencing. All constructs were then transformed into A. succinogenes
130Z.

2.3.12 High-performance liquid chromatography (HPLC) analysis of fermentation media
from strains expressing transporters.
A. succinogenes growth in liquid cultures was monitored by optical density at 660 nm
(OD660) on a DU 650 spectrophotometer (Beckman, Fullerton, CA). OD660 values were used to
calculate carbon balances. Glucose and fermentation products were quantified in culture
supernatants by HPLC (Waters, Milford, MA) using an Aminex HPX-87H column (Bio-rad).
Samples were run at room temperature with 4 mM H2SO4 as the eluent at a 0.6 mL min-1 flow
rate. Glucose and ethanol were quantified on a Waters 410 differential refractometer. Organic
acids were quantified on a Waters 2487 UV detector at 210 nm.

2.4 Results and Discussion
2.4.1 Potential C4-dicarboxylate exporters in A. succinogenes
Succinate as a fermentation product is produced by fumarate reductase, whose active site
faces the cytoplasm. Succinate has then to be exported out of the cell. A. succinogenes succinate
exporters are completely unknown. Table 2.2 lists the putative C4-dicarboxylate transporters
identified in the A. succinogenes 130Z genome by BlastP using known C4-dicarboxylate
transporters from other bacterial species. A. succinogenes contains homologs of E. coli DcuA,

57

DcuB, DcuC, DauA, CitT, and YjjPB, as well as one homolog of C. glutamicum SucE, and
multiple homologs of C. glutamicum DcsT. It also contains a protein of the telluriteresistance/carboxylate transport (TDT) family, up to nine tripartite ATP-independent periplasmic
(TRAP family) transporters, and up to five divalent anion:Na+ symporters (DASS family).

Table 2.2 A. succinogenes putative C4-dicarboxylate transporters
Asuc locus
0020
0023
0074
0142
0146
0147
0148
0156
0157
0158
0183
0270
0271
0272
0273
0304
0366
0367
0368
0715
0716
1063
1163
1164
1165
1482
1568
1577
1578
1579
1781
1921
1922
1923
1957
1958
1988
1989
1990
1991
1999
2058

Putative function
Anion transporter, ArsB/NhaD family
Putative transporter, aspartate:alanine exchanger family, 32% identical to C. glutamicum SucE
C4-dicarboxylate transporter/malic acid transport protein
Anaerobic C4-dicarboxylate membrane transporter, 43% identical to E. coli DcuA
Possible TRAP C4-dicarboxylate transporter solute receptor, DctP
Possible TRAP C4-dicarboxylate transporter, small permease component, DctQ
Possible TRAP C4-dicarboxylate transporter, large permease component, DctM
Possible TRAP C4-dicarboxylate transporter, large permease component, DctM
Conserved hypothetical protein (TRAP, small permease component, DctQ)
Possible TRAP C4-dicarboxylate transporter solute receptor, DctP
Anion transporter; Anion transporter, ArsB/NhaD family; 33% identical to E. coli CitT
Conserved hypothetical protein (TRAP, small permease component, DctQ)
Possible TRAP C4-dicarboxylate transporter, large permease component, DctM
Possible TRAP C4-dicarboxylate transporter solute receptor, DctP
Possible TRAP C4-dicarboxylate transporter solute receptor, DctP
Anion transporter, DASS family, involved in fumarate uptake
Possible TRAP C4-dicarboxylate transporter solute receptor, DctP
Conserved hypothetical protein (TRAP, small permease component, DctQ)
Possible TRAP C4-dicarboxylate transporter, large permease component, DctM
Putative Thr/Ser exporter, 52% identical to E. coli YjjB
Putative Thr/Ser exporter, 50% identical to E. coli YjjP
Anaerobic C4-dicarboxylate antiporter, DcuC family
Possible TRAP C4-dicarboxylate transporter solute receptor, DctP
Possible TRAP C4-dicarboxylate transporter, small permease component, DctQ
Possible TRAP C4-dicarboxylate transporter, large permease component, DctM
Anion transporter, ArsB/NhaD family; 31% identical to E. coli CitT
Anion permease ArsB/NhaD, 36% identical to C. glutamicum DcsT
Possible TRAP C4-dicarboxylate transporter, large permease component, DctM
Conserved hypothetical protein (TRAP, small permease component, DctQ)
Possible TRAP C4-dicarboxylate transporter solute receptor, DctP
Hypothetical protein; form of DctA or dicarboxylate transport protein found in many bacterial families
Possible TRAP C4-dicarboxylate transporter, large permease component, DctM
Possible TRAP C4-dicarboxylate transporter, small permease component, DctQ
Possible TRAP C4-dicarboxylate transporter solute receptor, DctP
TRAP transporter, 4TM/12TM fusion protein , DctQM
TRAP transporter solute receptor, TAXI family
TRAP transporter solute receptor, TAXI family
UspA domain protein, involved in stress response
TRAP transporter, 4TM/12TM fusion permease protein, DctQM
TRAP transporter solute receptor, TAXI family
Putative anaerobic C4-dicarboxylate transporter, 75% identical to E. coli DcuB (contains a 100-residue
insertion)
Sulfate permease (sulP) family, 58% identical to E. coli DauA

58

Some of the TRAP transporters could be involved in C4-dicarboxylate transport, but TRAP
transporters are involved in solute uptake and are considered mostly unidirectional (28), thus
they are unlikely to be involved in succinate efflux. DASS transporters are also involved in
solute uptake, and the A. succinogenes DASS transporter SdcA (Asuc_0304) was shown to
restore growth on C4-dicarboxylates in a C4-dicarboxylate-transport-negative E. coli strain (29).
Other putative C4-dicarboxylate transporters tested (Asuc_0074, _0142, _1063, _1482, and
_1999) did not (29). Because DASS transporters work in the direction of a decreasing Na+
gradient, Asuc_0304 is unlikely to be involved in anion efflux. In the same study, qPCR
reactions suggested that none of the genes tested (i.e., Asuc_0074, _0142, _0304, _1063, _1482,
and _1999) were induced during anaerobic growth on glucose compared to other growth
conditions. Asuc_1063 was upregulated 4 and 6 times during aerobic and anaerobic growth on
fumarate, respectively, suggesting that this transporter is involved in C4-dicarboxylate uptake as
well (29).

2.4.2 Proteomic analysis of A. succinogenes cytoplasmic membrane-enriched fractions
Based solely on the putative C4-dicarboxylate transporters identified in the genome and on
the functions of the different transporters already characterized in E. coli and C. glutamicum, it is
impossible to predict which A. succinogenes putative transporters are the best candidates for
succinate export. For this reason, we performed a proteomics analysis of cytoplasmic membraneenriched fractions from A. succinogenes anaerobically grown on different carbon sources (i.e.,
glucose, fructose, xylose, and glycerol-nitrate). Note that our cytoplasmic membrane fractions
were only enriched in cytoplasmic membrane. They did not consist of not entirely purified
cytoplasmic membranes. For this reason the fractions still contained outer membrane,

59

cytoplasmic, and periplasmic proteins. Of the top 250 hits, though, 101 were associated with the
cytoplasmic membrane, with 14 associated with the outer membrane, and 103 cytoplasmic
proteins, indicating that these fractions could still provide good information on cytoplasmic
membrane protein abundance.
The putative transporter candidates with the most hits (in NSAF format) are listed in Table
2.3. Asuc_1999, a homolog of E. coli DcuB, had the highest number of hits across all conditions
tested, followed by Asuc_1990, Asuc_2058, Asuc_0142, and Asuc_1163. Asuc_1990 is the
fused permease component (DctQM) of a TRAP-family transporter. Note that the putative solute
receptor likely associated with Asuc_1990, Asuc_1991, was detected with high probability (rank
963) in all growth conditions tested, even though it is a periplasmic protein. Asuc_2058 belongs
to the sulfate permease family, pfam00916. All characterized members of this family so far are
sulfate uptake transporters (30). Asuc_0142 is a DcuA homolog. Asuc_1163 is the putative
solute receptor of a TRAP C4-dicarboxylate transporter. Surprisingly, this periplasmic solute
receptor was detected in all growth conditions, with a high ranking, while its putative
cytoplasmic membrane partners, Asuc_1164 and Asuc_1165, were not detected in any growth
condition tested (Table 2.3). Peptides of other putative C4 dicarboxylate transporters were also
detected in our samples, but typically at lower levels and not consistently across all growth
conditions. Among those, DcuC homolog Asuc_1063 was detected at very low levels across all
growth conditions. Asuc_0304, which had been shown to be involved in succinate uptake (29),
was not detected in any growth condition. Neither were Asuc_0023, a C. glutamicum SucE
homolog, nor Asuc_0715-16, homologs of E. coli YjjBP (Tables 2.3 and 2.4).

60

Table 2.3 Proteomics and transcriptomics data for succinate transporters in A. succinogenes
Glucose anaerobic
Ranka Protein putative function
46
125
963
216
272
342
554
672
751
918
1122
1138
1258
1289
1372
ND
ND
ND
ND
ND
ND

Anaerobic C4-dicarboxylate transporter, DcuB
homolog
TRAP-TAXI transporter, 4TM/12TM fusion
protein
TRAP transporter solute receptor, TAXI family
C4-dicarboxylic acid transporter DauA
Anaerobic c4-dicarboxylate transporter, DcuA
homolog
TRAP C4-dicarboxylate transporter solute
receptor, DctP
TRAP C4-dicarboxylate transporter solute
receptor, DctP
C4-dicarboxylate ABC transporter
C4-dicarboxylate ABC transporter, DASS family
C4-dicarboxylate ABC transporter, DcuC homolog
Hypothetical protein
Anion permease ArsB/NhaD
C4-dicarboxylate ABC transporter permease
TRAP family transporter membrane component,
DctQ
TRAP transporter, 4TM/12TM fusion protein,
DctQM
Corynebacterium glutamicum SucE homolog
Dicarboxylate transporter/tellurite-resistance
protein
Anion transporter, DASS family
TRAP transporter solute receptor, TAXI family
Anion transporter, similar to Asuc_1482
Anion permease ArsB/NhaD

Asuc_
locus
1999

Proteinb

Fructose anaerobic

RNAc,d

Proteinb

RNAc

Xylose anaerobic
Proteinb

RNAc

Glycerol nitrate
Proteinb

RNAc

Glycerol
microaerobic
RNAc

0.164 ± 0.015 2698 ± 310 0.171 ± 0.015

1230 ± 98 0.199 ± 0.022 2980 ± 133 0.116 ± 0.017 919 ± 193

1990

0.058 ± 0.023

104 ± 25

195 ± 35 0.038 ± 7E-04

1991
2058
0142

0.037 ± 0.007
0.023 ± 0.007
0.031 ± 0.04

752 ± 222 0.052 ± 5E-04 1094 ± 333 0.050 ± 0.014 702 ± 54 0.028 ± 0.004 887 ± 483
377 ± 17 0.025 ± 0.001
163 ± 9 0.028 ± 0.002 199 ± 9 0.024 ± 0.001 159 ± 19
407 ± 84 0.026 ± 0.008 387 ± 48 0.054 ± 0.007 1025 ± 76 0.072 ± 0.009 377 ± 78

1163

0.009 ±0.003

12 ± 1

0.001 ± 0.002

6±1

0.016 ± 0.002

14 ± 2

0.035 ± 0.007

19 ± 9

11 ± 4

0366

0.002 ± 7E-04

21 ± 3

0.002 ± 2E-05

5±1

0.004 ± 5E-04

31 ± 1

0.003 ± 0.000

18 ± 5

34 ± 2

1781
1482
1063
0272
0020
1577
0367

0.004 ± 0.004
0.002 ±2E-04
0.002 ± 0.004

56 ± 5
98 ± 9
55 ± 2
93 ± 12
87 ± 55
40 ± 3
13 ± 2

0.005 ± 0.001
0.003 ± 0.002
0.007 ± 0.002

1957

6E-04 ± 0.001

0.003 ± 0.002

0.080 ± 0.007

0.003 ± 0.001

120 ± 2

65 ± 2
0.012 ± 0.003
72 ± 1
164 ± 44 0.009 ± 6E-04 236 ± 13
30 ± 2
0.000 ± 0.000
39 ± 2
16 ± 1
47 ± 2
43 ± 13
136 ± 23
36 ± 3
0.001 ± 0.002
87 ± 0
5±1
0.003 ± 0.004
18 ± 1

0.020 ± 0.03

134 ± 47

0.000 ± 0.000 114 ± 36
0.003 ± 0.001 507 ± 87
0.000 ± 0.000 121 ± 7
0.012 ± 0.002 92 ± 22
115 ± 72
64 ± 9
0.006 ± 0.001 13 ± 3

1407 ± 219
88 ± 20
368 ± 85
210 ± 37
234 ± 113

63 ± 5
140 ± 28
44 ± 8
75 ± 23
115 ± 72
57 ± 19
24 ± 1

9±1

13 ± 2

18 ± 0

60 ± 35

13 ± 3

0023
0074

342 ± 38
101 ± 11

146 ± 37
96 ± 18

245 ± 12
98 ± 3

326 ± 140
84 ± 29

33 ± 3
93 ± 3

0304
1988
0183
1568

236 ± 9
140 ± 26
13 ± 2
29 ± 6

212 ± 17
278 ± 61
11 ± 1
23 ± 3

289 ± 9
189 ± 12
12 ± 1
28 ± 3

266 ± 26
179 ± 63
39 ± 9
103 ± 30

309 ± 23
121 ± 30
9±2
247 ± 96

a

Rank is the overall rank of each protein in the entire normalized proteomics data set
ND: not detected
b
Proteomics data are in NSAF format
c
Transcriptome data are expressed per nt
d
Glu RNA data are average ± standard deviations from three independent biological
replicates.

61

2.4.3 Transcript levels of A. succinogenes putative C4-dicarboxylate transporters
As seen in Table 2.3, transcript levels for the putative C4 dicarboxylate transporters do not
always agree well with detected protein levels, but this observation is not particularly surprising
due to post transcriptional regulation, the typically higher stability of proteins compared to RNA,
and the large variability in protein detectability by mass spectrometry. This variability can
explain why some transporter genes were clearly transcribed (e.g., Asuc_0023, Asuc_0304, or
Asuc_1988) but the corresponding protein was not detected in any growth condition. For each
putative transporter listed, though, trends in transcript levels were relatively uniform across
growth conditions, and four of the top six transporters (including Asuc_1991) identified by
proteomics, had the highest transcript levels in anaerobic glucose cultures (i.e., Asuc_1999,
Asuc_2058, Asuc_0142, and Asuc_1991).
As observed in our proteomics results, Asuc_1999 has the highest transcript levels of all
putative C4-dicarboxylate transporters in all growth conditions tested, even in the presence of
nitrate. It also ranked very high relative to all transcript levels in the cell (e.g., rank of 101 on
glucose). Succinate production during growth on glycerol-nitrate is much lower than during
anaerobic growth on glucose (27). Interestingly, asuc_1999’s transcript levels decreased almost
3-fold and its ranking fell almost 4-fold in glycerol-nitrate compared to anaerobic glucose
cultures (Table 2.3). Based on these results, Asuc_1999 is a good candidate as a transporter
involved in succinate export.
Transcript levels for asuc_1990 and asuc_1991 were relatively steady across growth
conditions, with transcript levels for asuc_1991 always higher than those for asuc_1990,
suggesting that the two genes are not cotranscribed. The facts that the intergenic region between

62

asuc_1990 and asuc_1991 is 336-nt long and that very few RNA reads mapped to this region
support this conclusion.
Transcripts levels for asuc_2058 were higher in anaerobic glucose cultures than in any
other conditions, but they remained at least 100-fold higher than the lowest transcript levels
detected for other genes in any conditions. Transcripts levels for asuc_0142 were among the
highest for putative C4 dicarboxylate transporters across all growth conditions, in good
agreement with the proteomics results.
While the Asuc_1163 protein was consistently detected by proteomics in the growth
conditions tested, the corresponding transcript levels were very low across all growth conditions,
and transcript levels for the putative associated membrane components Asuc_1164 and
Asuc_1165 were extremely low as well.
Our combined proteomics and RNAseq data suggest that Asuc_1999, Asuc_2058,
Asuc_0142, and Asuc_1990-91 could be involved in succinate export. Of these, only Asuc_1999
and Asuc_0142 are homologs of known succinate exporters, DcuB and DcuA, respectively
(Table 2.2). Our proteomics and RNAseq data also seem to exclude Asuc_0304, DcuC homolog
Asuc_1063, SucE homolog Asuc_0023, and E. coli YjjBP homologs Asuc_0715-16 as succinate
exporters. Quantitative real-time PCR studies by Rhie et al (2014) in A. succinogenes grown
anaerobically on glucose did not show any induction of genes asuc_0142, asuc_1063,
asuc_1482, and asuc_1999. Our RNAseq results do not agree with these results for Asuc_1999
and Asuc_0142 since we saw high transcript levels for both of these genes.

63

2.4.4 Construction of an expression reporter system.
E. coli promoters are typically poorly or not functional at all in Pasteurellaceae species. At
the beginning of this study, the only A. succinogenes promoter that had been tested was that of
the phosphoenolpyruvate carboxykinase gene, pckA, which is a strong, constitutive promoter. No
transcriptomic data were available for A. succinogenes to give us a sense of which promoters
were strong or weak and when they were functional. Because recombinant membrane protein
overexpression often requires a well-calibrated, weak-to-moderately strong promoter, and
because this single available A. succinogenes native promoter did not allow for any tunability of
gene expression, we sought to decrease the strength of ppckA by first identifying the pckA
transcription start site (TSS) and then constructing truncated versions of the promoter. To test
these truncated promoters, though, we needed a reporter expression system.

lacZ region in strain 130Z

lacZ

lacZup

lacZdown

(4.5 kb)

ΔlacZ construct in pCR2.1

lacZdown

lacZup

(1.5 kb)

ΔlacZ::icd construct in pCR2.1 and ΔlacZ::icd strain

A

A

lacZup

FRT

ppckA

icd

FRT

lacZdown

(3.3 kb)

ΔlacZ strain

lacZup

130Z

FRT

ΔlacZ

lacZdown

(1.6 kb)

(A)

ΔlacZ(pLGZ920::lacZ)

(B)

Figure 2.1 Construction of the ΔlacZ strain. (A) Schematic of the construction procedure.
FRT, flippase recognition target site; A, AscI restriction site. (B) A. succinogenes 130Z, ΔlacZ,
and ΔlacZ(pLGZ920::lacZ) strains grown on LB-glucose supplemented with X-gal (40 µg mL-1).

64

To test promoter strength in A. succinogenes, we used lacZ as the reporter gene. A lacZ
deletion mutant of A. succinogenes was constructed (strategy described in Figure 2.1A) to be
used as the background strain. The ΔlacZ strain was confirmed by testing lacZ expression on
LB-glucose plates supplemented with 5-bromo-4-chloro-3-indolyl-β-D-galactoside (X-gal, 40 µg
mL-1) (Figure 2.1B). Plasmid pLGZ920::lacZ, with lacZ under control of ppckA, was used as the
positive control to test the reporter system (Figure 2.1B).

2.4.5 Identification of ppckA’s TSS by RLM-RACE and construction of truncated ppckA
promoters
Out of the 14 clones sequenced, six clones indicated the TSS (+1) of pckA as a T nucleotide
32 nt upstream of the start codon. Sequences similar to E. coli promoter -10 and -35 consensus
sequences were found 14 nt and 27 nt upstream of the TSS, respectively, with a 21-nt spacer
sequence between the probable -10 and -35 sites (Figure 2.2).

ATTTGAAACGGATCACAAATCATGAAAAAAATACGTTCAAATTAGAACTAATTATCGAAAATTTGATCTA
GTTAACATTTTTTAGGTATAAATAGTTTTAAAATAGATCTAGTTTGGATTTTTAATTTTAAATTATCAAT
GAGGTGAAGTATGACTGACTTAAACAAACTCGTTAAAGAACT

Figure 2.2 Identification of the pckA transcription start site. 5’ region of the A. succinogenes
pckA gene. Yellow highlight: probable -35 region; green highlight: probable -10 region; red and
bold: transcription start site; underline: ribosome binding site; bold characters: start of the pckA
coding region.

In our earlier studies, we routinely used a 231-nt sequence upstream of the pckA start codon
to express genes under the control of ppckA. We built truncated versions of this 231-nt sequence to
decrease the strength of the promoter (Figure 2.3), and tested the strength of these truncated
promoters using lacZ as the reporter gene. Activity of the truncated promoters decreased with the

65

length of the promoter region, and ranged from 1.4-fold (ppckA-164 promoter) to 20-fold lower
(ppckA-92 promoter) than ppckA (Figure 2.4). Almost no β-galactosidase activity was detected with
promoter ppckA-75 (data not shown), which suggests that the identified putative -35 region is
indeed needed for promoter function.

Figure 2.3 Reporter constructs with lacZ under control of truncated versions of pckA’s
5’UTR. H, HindIII; X, XbaI; yellow highlight: probable -35 region; green highlight: probable 10 region; red and bold: transcription start site; underline: ribosome binding site; italics: indicate
restriction sites—HindIII (5’ end) and XbaI (3’ end); vertical arrows indicate start of truncated
promoters. Drawings of DNA fragments are not to scale.

66

2.4.6 Identification of promoters of different strengths from the transcriptomics results.
To expand the expression range of our promoter library toward expression levels lower
than from ppckA-92, we used our RNAseq results. Candidate promoters were selected from the
genes whose transcript levels did not vary between growth conditions using the RNAseq results
from anaerobic cultures grown on glucose, xylose, fructose, and mannose, as well as glycerol
microaerobic cultures, which all favor succinate production. Genes whose log2 fold-change was
between -0.2 and 0.2 in all growth conditions compared to glucose were compiled and their
transcript levels (using expression levels per nucleotide) (Table S2.2) were ranked, together with
that of pckA, for further analysis. Candidates (individual genes or the first genes in an operon)
were selected that covered a large range of decreasing strengths compared to ppckA. Promoters of
candidate genes asuc_0391, asuc_0289, asuc_0701, and asuc_2109 were selected, whose
transcript levels were 16-, 24-, 29-, and 62-fold lower than pckA’s. DNA fragments 247 bp
upstream of candidate genes asuc_0701, asuc_0289, and asuc_0391 and 261-bp upstream of
asuc_2109 were cloned in front of lacZ to be tested as promoters.
Promoter strength was tested by β-galactosidase assays using ΔlacZ as the expression
strain, with lacZ under control of ppckA as the positive control (Figure 4). Promoter pAsuc_0701 was
the weakest promoter, 209-fold weaker than ppckA, which was the strongest. To the best of our
knowledge this is the first promoter library constructed for A. succinogenes 130Z, making it
easier to test expression of different genes in this strain for succinate production.

67

Specific activity
(µmol/min/mg)

100000
90000
80000
70000
60000
50000
40000
30000
20000
10000
8000
7000
6000
5000
4000
3000
2000
1000
0
1

2

3

4

5

6

7

8

9

10

Figure 2.4 β-galactosidase activity of promoter reporter constructs. All promoter constructs
were tested in the ΔlacZ strain. 1, ΔlacZ(pLGZ920); 2, ΔlacZ(pLGZ920*-pAsuc_0701-lacZ); 3,
ΔlacZ(pLGZ920*-pAsuc_2109-lacZ); 4, ΔlacZ(pLGZ920*-pAsuc_0391-lacZ); 5, ΔlacZ(pLGZ920-ppckA92-lacZ); 6, ΔlacZ(pLGZ920*-pAsuc_0289-lacZ); 7, ΔlacZ(pLGZ920-ppckA-103-lacZ); 8,
ΔlacZ(pLGZ920-ppckA-134-lacZ); 9, ΔlacZ(pLGZ920*-ppckA-164-lacZ); 10, ΔlacZ(pLGZ920-lacZ).
Results are the average activity ± standard deviation from three independent biological
replicates.

2.4.7 Overexpression of putative succinate transporter candidates in A. succinogenes and
succinate production.
Succinate transporter candidates with the most hits in the proteomics analysis and the
highest transcript levels were over-expressed to determine their effects on succinate production.
We focused on Asuc_1999, Asuc_1990-1991, Asuc_2058, and Asuc_0142. Even though

68

Asuc_1990 and Asuc_1991 seemed to be transcribed independently, they are likely part of the
same transport system, so the two genes were expressed in a single construct. For each putative
transporter, we tested several weak promoters to maximize protein expression while avoiding
toxicity to the cells. Expression of all four transporters under control of ppckA-134 was toxic to the
cells, thus we tested expression of these proteins under control of ppckA-92 and ppckA-103.
Asuc_2058 and Asuc_1990-1991 expressed under control of ppckA-103 proved toxic to the cells as
well, but expression of Asuc_0142 and Asuc_1999 under the same promoter did not inhibit cell
growth. Promoter ppckA-92 allowed expression of all four transporters without inhibiting growth of
the 130Z recombinant strain.
Fermentation balances for A. succinogenes 130Z carrying the six expression constructs are
shown in Table 2.4. All six expression constructs caused A. succinogenes 130Z’s succinate
yields to increase significantly, compared to the empty vector, in anaerobic glucose cultures.
Much of the succinate yield increase seemed to be at the expense of biomass production. Note
that expressing Asuc_1999 and Asuc_0142 genes under control of ppckA-103 increased the
succinate yield compared to ppckA-92 for both transporters, although the increase was not
statistically significant.
The strain with the highest succinate yield, 130Z (pLGZ920-ppckA-103-Asuc_1999), also
produced almost the highest acetate yield, but produced the least biomass. In contrast, the second
best succinate producing strain, 130Z (pLGZ920-ppckA-92-Asuc_1990-1991), produced almost as
much succinate, plus it did not produce more acetate than the control strain, making 130Z
(pLGZ920-ppckA-92-Asuc_1990-1991) possibly a more suitable candidate for succinate production
and further engineering.

69

Table 2.4 Fermentation balances of strain 130Z over-expressing candidate succinate transporters.
Products (mmol/100 mmol glucose consumed)

Strain

Succinate

Formate

Acetate

Ethanol

130Z (pLGZ920)
130Z (pLGZ920-ppckA-92-Asuc_1999)
130Z (pLGZ920-ppckA-92-Asuc_0142)
130Z (pLGZ920-ppckA-103-Asuc_1999)
130Z (pLGZ920-ppckA-103-Asuc_0142)
130Z (pLGZ920-ppckA-92-Asuc_2058)
130Z (pLGZ920-ppckA-92-Asuc_1990-91)

47.3 ± 2.3
55.0 ± 0.8
53.7 ± 1.4
58.5 ± 2.8
56.4 ± 1.1
55.7 ± 2.4
57.2 ± 2.2

98.0 ± 5.4
103 ± 1
98.5 ± 2.4
101 ± 6
101 ± 3
98.1 ± 5.2
96.9 ± 0.6

65.5 ± 2.8
72.4 ± 1.1
64.6 ± 2.1
73.0 ± 3.3
73.9 ± 1.6
68.4 ± 2.9
65.0 ± 0.8

30.6 ± 2.8
28.5 ± 2.0
25.5 ± 4.2
25.2 ± 3.2
29.8 ± 0.9
28.5 ± 0.9
26.8 ± 1.7

CO2

Biomassa

0.0 ± 0.0 193 ± 10
0.0 ± 0.0 157 ± 2
0.0 ± 0.0 188 ± 6
0. 0± 0.0 138 ± 5
0.5 ± 0.3 152 ± 8
0.0 ± 0.0 142 ± 2
0.0 ± 0.0 154 ± 2

%
increased
Carbon succinate
recoveryb
yield
104 ± 5
104 ± 1
104 ± 3
101 ± 4
105 ± 3
100 ± 4
101 ± 1

NA
16**
14*
24**
19**
18*
21**

Doubling time
2.04 ± 0.03
2.22 ± 0.05
2.27 ± 0.02
2.74 ± 0.2
2.46 ± 0.01
2.49 ± 0.06
2.26 ± 0.05

Results are an average of three biological replicates ± standard deviations.
a
Biomass was determined using assumed values of 567 mg dry cell weight/mL per OD660(31) and a cell composition of CH2O0.5N0.2
(24.967 g/mol) (32)
b
Carbon balance is the carbon in products/carbon in glucose consumed. It is assumed that one CO2 is fixed for each molecule of
succinate produced.
CO2 was calculated using the following formula:
CO2 (in mM) = acetate (in mM) + ethanol (in mM) - formate (in mM)
* Significantly different from 130Z (pLGZ920) (p < 0.05, two-tailed student’s t-test)
** Significantly different from 130Z (pLGZ920) (p < 0.01, two-tailed student’s t –test)

70

When attempting to express all the constructs in the ΔpflB strain (a pyruvate-formate lyase
deletion mutant) (21), which does not produce any formate, we were either unable to obtain any
transformants or the strains grew extremely poorly, indicating that these constructs are toxic in
ΔpflB and that weaker promoters are likely needed in this strain (data not shown).

2.4.8 The A. succinogenes succinate efflux transporters differ from those of E. coli and C.
glutamicum
A. succinogenes contains homologs of many E. coli succinate transporters—DcuA, DcuB,
DcuC, DauA, CitT, and YjjPB. It also contains one homolog of C. glutamicum SucE, along with
multiple homologs of C. glutamicum DcsT.
E. coli DcuC can perform different functions, but, in particular, it has been shown to
participate in succinate efflux under anaerobic conditions (9). Asuc_1063 is the A. succinogenes
DcuC homolog, but it does not show high transcriptional levels or proteomics hits in our studies,
eliminating it as a succinate efflux transporter. E. coli YjjP and YjjB constitute a major succinate
efflux transporter (17) but Asuc_0715 and Asuc_0716, which are 52% and 50% identical to YjjB
and YjjP, respectively, were not detected in our proteomics analysis.
Golby et al. (33) determined that E. coli DcuB is subject to catabolite repression, repressed
by nitrate, and strongly induced by C4-dicarboxylates. These authors suggested that it is mostly
involved in C4-dicarboxylate transport during fumarate respiration. A quintuple or sextuple
deletion mutant, including ΔdcuB, did not show decreased succinate efflux (13), but then a single
dcuB mutant affected succinate efflux during growth on glucose (9), thus DcuB could contribute
to succinate efflux in E. coli. Asuc_1999, a putative anaerobic C4-dicarboxylate transporter that
is 75% identical to E. coli DcuB increases the succinate yield in A. succinogenes when

71

overexpressed. Asuc_1999 had the most hits in our proteomics analysis and had the highest
transcript levels of all putative C4-dicarboxylate transporters across all conditions tested,
suggesting that it plays a major role in succinate efflux in A. succinogenes.
Overexpressing Asuc_1990-1991 in A. succinogenes caused an increase in succinate yield.
This result is rather surprising, since Asuc_1990-1991 belongs to the TRAP family of
transporters, which are primarily involved in uptake transport. Asuc_1990-1991 had the second
highest hits in our proteomics analysis and had high transcriptional levels across all the
conditions we tested.
Asuc_0142, an anaerobic C4-dicarboxylate membrane transporter, is 43% identical to E.
coli DcuA. Overexpression of Asuc_0142 in A. succinogenes also caused an increase in
succinate yield. E. coli DcuA is known to function as both a succinate uptake and efflux
transporter, as confirmed using double mutants expressing only one of the DcuA, DcuB, or
DcuC carriers at a time (8).
Asuc_2058, showing similarity to E. coli DauA, also led to an increase in succinate yield
upon overexpression. This result suggests that Asuc_2058 has a function completely different
from that of E. coli DauA, which has been shown to be the main succinate uptake transporter
under acidic conditions and to be inactive at pH 7 (10).
SucE has been identified as one of the C. glutamicum transporters involved in succinate
efflux under microaerobic and anaerobic conditions. Asuc_0023, a homolog of C. glutamicum
SucE had moderate transcript levels but was not detected in our proteomics studies, indicating
that the C. glutamicum and A. succinogenes homologs do not have the same function.

72

Thus succinate efflux transporters in A. succinogenes are different from the major succinate
exporters of E. coli and C. glutamicum, and even include transporters (e.g., TRAP-family
transporters Asuc_1990-91) that have so far never been shown to function in export.

2.5 Conclusion
This study is the first to focus on A. succinogenes dicarboxylate efflux transporters with the
aim of increasing succinate production. Combining proteomics and transcriptomics approaches,
we identified the four putative C4-dicarboxylate transporters with the highest expression levels.
Overexpressing all four transporters individually increased succinate production. While, based
on our expression data, Asuc_1999 is a likely candidate to be a major succinate exporter, we
cannot identify a single major succinate exporter in A. succinogenes. Our results suggest instead
that more than one transporter is involved in succinate export in A. succinogenes, similar to the
situation observed in E. coli. Knockout mutants of these transporters (construction unsuccessful
so far) would shed more light on the relative involvement of these transporters in succinate
efflux in A. succinogenes.
We also developed a library of A. succinogenes promoters covering a large expression
range across multiple growth conditions, which can be used for modulating protein expression in
A. succinogenes as needed for strain engineering.

2.6 Acknowledgements
I thank Dr. Claire Vieille and Peggy Wolf for carrying out all work related to proteomics. I thank
Dr. Nikolas McPherson for obtaining and analyzing all RNA sequencing data.

73

APPENDIX

74

Table A2.1 Primers used in this study
Primer
P1
P2
CV376
CV377
CV443
CV469
CV470
CV471
CV473
CV474
CV475
CV476
CV569
CV570
CV571
CV572
CV577
CV578
CV579
CV580
CV586
CV582
CV587
CV588
CV589
CV590
CV615
CV616
CV617
CV618
CV624
CV625
GZ1302
GZ1303
CV431

Sequencea (restriction site)
CGTGGTTAACGTCCCTTTATCG
GGTTTCTTCCTCGAAAAGTTGTTC
CTTGCCAAACCGACCGAAAG
TATTGATAATGAAAATCCGACCGCACTTGGCAGTACCGGCGTATTCCTC b
CACACCATTCCCAAACAAAAC
GACCATGATTACGCCAAGCTTTCTGCTAATCCTGTTACCAGTGG
TAGGTTGGCAGAATCATCTAGATCACCTCATTGATAATTTAAAATTAAAAATCCAAACT
AGATCTATTTTAAAAC
AAGCTTTCTTCCGCTTCCTCGCTCACTG
GAGGAAGCGGAAGAAAGCTTCGAAAATTTGATCTAGTTAACATTTTTTAGG
GAGGAAGCGGAAGAAAGCTTGAACTAATTATCGAAAATTTGATCTAG
GAGGAAGCGGAAGAAAGCTTCACAAATCATGAAAAAAATACGTTC
GAGGAAGCGGAAGAAAGCTTCATTTACCGCCATAAAAATTTGAAAC
TGAGCGAGGAAGCGGAAGAGGATCCGGTCAAACTCCTACGAATTTC
AAGTAGGTTGGCAGAATCATCTAGAGCTCACCAACAGGCTTGA
TGAGCGAGGAAGCGGAAGAGGATCCGAGAGAACAGGACAACAGTTTTATTG
AAGTAGGTTGGCAGAATCATCTAGAAATAATAACTCTTAATATAGAAAAAAACGATTG
TGAGCGAGGAAGCGGAAGAGGATCCGACGAGCTACTCCCTGGTTTG
AAGTAGGTTGGCAGAATCATCTAGAGATCCTTTTAAAAAAAACGATGAAAAAAT
TAAATTATCAATGAGGTGATCTAGATGACTGCAATGTTTATTATCC
ACGGCCAGTGAATTCGAGCTCTTAGATAAACAGATTCGCAAATAAC
TAAATTATCAATGAGGTGATCTAGATGAAAAAATTATTTAAACTTTCTCTTGTC
ACGGCCAGTGAATTCGAGCTCTATTCCGGACGACGACG
TAAATTATCAATGAGGTGATCTAGATGGATTTTTTGATGAATCTAAG
ACGGCCAGTGAATTCGAGCTCTTAAAGATAACCGTATAAGCCTG
TAAATTATCAATGAGGTGATCTAGATGCTAAATAAATGGTTTTTAACC
ACGGCCAGTGAATTCGAGCTCTTAATGACGTAATTCCGATTC
GACCATGATTACGCCAAGCTTTCTGCTAATCCTGTTACCAGTGGC
ATCACCTTAGGATCCCCGCATCAGGCGCTCTTC
TGATGCGGGGATCCTAAGGTGATTTATAGTCTGGACGG
GTACCCGGGGATCCTCTAGATTCACCTCGAAACAGATAAAAAAATC
AAATTATCAATGAGGTGATCTAGATGATTCTGCCAACCTACTTTGAAAATCC
CCAGTGAATTCGAGCTCTTATTCAAAGTGAATATCGAAACGACAGTCAAATTTTG
CGTTGTAAAACGACGGCC
AATTTTAAATTATCAATGAGGTG
GGTTTCGCCCACTCGTATTCC

75

Specificityb (directionc)
5’RLM-RACE pckA-specific outer primer (R)
5’RLM-RACE pckA-specific inner primer (R)
ΔlacZ-fusion, nested (F)
ΔlacZ-fusion, nested (R)
Truncated ppckA constructs (R)
colE1 ori (F)
ppckA with lacZ overhangs (R)
ColE1 ori (R)
ppckA-92 (F)
ppckA-103 (F)
ppckA-134 (F)
ppckA-164 (F)
pAsuc_2109 (F)
pAsuc_2109 (R)
pAsuc_0391 (F)
pAsuc_0391 (R)
pAsuc_0289 (F)
pAsuc_0289 (R)
Asuc_0142 (F)
Asuc_0142 (R)
Asuc_1990-1991 (F)
Asuc_1990-1991 (R)
Asuc_1999 (F)
Asuc_1999 (R)
Asuc_2058 (F)
Asuc_2058 (R)
ColE1 ori (F)
ColE1 ori (R), replaces ori’s 3’ HindIII site with BamHI
pAsuc_0701 (F)
pAsuc_0701 (R)
lacZ Gibson cloning (F)
lacZ Gibson cloning (R)
pLGZ901 specific primer downstream of SacI site (R)
pLG901 specific primer upstream of XbaI site (F)
lacZ sequencing primer-1

Table A2.1 (cont’d)
CV445 GCTCATATGAAATTACATAGCCTGTTTTCGGATCGC
CV446 CTATGTAATTTCATATGAGCAGGCGTTAGTTGATTTG
CV636 GGATATGGTGAAAACTTCGAAG
CV637 CCATTGTTTGGTTTACTCTCC
a
F, forward primer; R, reverse primer
b
Primer P4 contains the USS ACCGCACTT

lacZ sequencing primer-2
lacZ sequencing primer-3
lacZ sequencing primer-4
lacZ sequencing primer-5

76

Table A2.2 Transcript levels of candidate genes with low variation across several
conditions, compared to pckA
Gene ID
Average expression
Expression
Fold lower expression
per nucleotide
compared to pckA
than pckA
Asuc_2109
Asuc_0708
Asuc_1818
Asuc_1044
Asuc_1536

88
93
101
125
130

0.016
0.017
0.019
0.023
0.024

61.7
58.7
53.8
43.4
41.8

Asuc_1879
Asuc_1430
Asuc_0701

145
170
189

0.027
0.031
0.035

37.5
32.1
28.8

Asuc_0289
Asuc_1432
Asuc_2046
Asuc_1945
Asuc_1699
Asuc_1071

227
228
246
272
285
310

0.042
0.042
0.045
0.050
0.052
0.057

24.0
23.9
22.2
20.0
19.1
17.6

Asuc_0391
Asuc_1037
Asuc_0983
Asuc_1877
Asuc_2007
Asuc_2066

332
469
472
483
549
1423

0.061
0.086
0.087
0.089
0.101
0.261

16.4
11.6
11.5
11.3
9.9
3.8

77

REFERENCES

78

REFERENCES

1.

Zeikus JG, Jain MK, Elankovan P. 1999. Biotechnology of succinic acid production
and markets for derived industrial products. Appl Biochem Biotechnol 51:545-552.

2.

Sanchez AM, Bennett GN, San KY. 2005. Efficient succinic acid production from
glucose through overexpression of pyruvate carboxylase in an Escherichia coli alcohol
dehydrogenase and lactate dehydrogenase mutant. Biotechnol Prog 21:358-365.

3.

Jantama K, Haupt MJ, Svoronos SA, Zhang XL, Moore JC, Shanmugam KT,
Ingram LO. 2008. Combining metabolic engineering and metabolic evolution to develop
nonrecombinant strains of Escherichia coli C that produce succinate and malate.
Biotechnol Bioeng 99:1140-1153.

4.

Litsanov B, Brocker M, Bott M. 2012. Toward homosuccinate fermentation: metabolic
engineering of Corynebacterium glutamicum for anaerobic production of succinate from
glucose and formate. Appl Environ Microbiol 78:3325-3337.

5.

Ito Y, Hirasawa T, Shimizu H. 2014. Metabolic engineering of Saccharomyces
cerevisiae to improve succinic acid production based on metabolic profiling. Biosci
Biotechnol Biochem 78:151-159.

6.

Six S, Andrews SC, Unden G, Guest JR. 1994. Escherichia coli possesses two
homologous anaerobic C4-dicarboxylate membrane transporters (DcuA and DcuB)
distinct from the aerobic dicarboxylate transport system (Dct). J Bacteriol 176:64706478.

7.

Engel P, Kramer R, Unden G. 1994. Transport of C4-dicarboxylates by anaerobically
grown Escherichia coli. Energetics and mechanism of exchange, uptake and efflux. Eur J
Biochem 222:605-614.

8.

Zientz E, Six S, Unden G. 1996. Identification of a third secondary carrier (DcuC) for
anaerobic C4-dicarboxylate transport in Escherichia coli: roles of the three Dcu carriers
in uptake and exchange. J Bacteriol 178:7241-7247.

9.

Janausch IG, Unden G. 1999. The dcuD (former yhcL) gene product of Escherichia coli
as a member of the DcuC family of C4-dicarboxylate carriers: lack of evident expression.
Arch Microbiol 172:219-226.

10.

Karinou E, Compton EL, Morel M, Javelle A. 2013. The Escherichia coli SLC26
homologue YchM (DauA) is a C4-dicarboxylic acid transporter. Mol Microbiol 87:623640.

79

11.

Pos KM, Dimroth P, Bott M. 1998. The Escherichia coli citrate carrier CitT: a member
of a novel eubacterial transporter family related to the 2-oxoglutarate/malate translocator
from spinach chloroplasts. J Bacteriol 180:4160-4165.

12.

Davies SJ, Golby P, Omrani D, Broad SA, Harrington VL, Guest JR, Kelly DJ,
Andrews SC. 1999. Inactivation and regulation of the aerobic C4-dicarboxylate transport
(dctA) gene of Escherichia coli. J Bacteriol 181:5624-5635.

13.

Janausch IG, Kim OB, Unden G. 2001. DctA- and Dcu-independent transport of
succinate in Escherichia coli: contribution of diffusion and of alternative carriers. Arch
Microbiol 176:224-230.

14.

Chen J, Zhu X, Tan Z, Xu H, Tang J, Xiao D, Zhang X. 2014. Activating C4dicarboxylate transporters DcuB and DcuC for improving succinate production. Appl
Microbiol Biotechnol 98:2197-2205.

15.

Huhn S, Jolkver E, Kramer R, Marin K. 2011. Identification of the membrane protein
SucE and its role in succinate transport in Corynebacterium glutamicum. Appl Microbiol
Biotechnol 89:327-335.

16.

Fukui K, Koseki C, Yamamoto Y, Nakamura J, Sasahara A, Yuji R, Hashiguchi K,
Usuda Y, Matsui K, Kojima H, Abe K. 2011. Identification of succinate exporter in
Corynebacterium glutamicum and its physiological roles under anaerobic conditions. J
Biotechnol 154:25-34.

17.

Fukui K, Nanatani K, Hara Y, Yamakami S, Yahagi D, Chinen A, Tokura M, Abe
K. 2017. Escherichia coli yjjPB genes encode a succinate transporter important for
succinate production. Biosci Biotechnol Biochem 81:1837-1844.

18.

Kim P, Laivenieks M, McKinlay J, Vieille C, Zeikus JG. 2004. Construction of a
shuttle vector for the overexpression of recombinant proteins in Actinobacillus
succinogenes. Plasmid 51:108–115.

19.

Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K.
1993. Current Protocols in Molecular Biology. Greene Publishing & Wiley-Interscience,
New York.

20.

McKinlay JB, Zeikus JG, Vieille C. 2005. Insights into Actinobacillus succinogenes
fermentative metabolism in a chemically defined growth medium. Appl Environ
Microbiol 71:6651-6656.

21.

Joshi RJ, Schindler BD, McPherson NR, Tiwari K, Vieille C. 2014. Development of a
markerless knockout method for Actinobacillus succinogenes. Appl Environ Microbiol
80:3053-3061.

80

22.

Lee SJ, Song H, Lee SY. 2006. Genome-based metabolic engineering of Mannheimia
succiniciproducens for succinic acid production. Appl Environ Microbiol 72:1939–1948.

23.

Tai SP, Kaplan S. 1985. Intracellular localization of phospholipid transfer activity in
Rhodopseudomonas sphaeroides and a possible role in membrane biogenesis. J Bacteriol
164:181-186.

24.

Zhang Y, Wen Z, Washburn MP, Florens L. 2010. Refinements to label free proteome
quantitation: how to deal with peptides shared by multiple proteins. Anal Chem 82:22722281.

25.

McPherson N. 2017. Novel Insights Into Sugar and Succinate Metabolism of
Actinobacillus succinogenes from Strains Evolved for Improved Growth on
Lignocellulose Hydrolysate Sugars. Doctoral. Michigan State University.

26.

Johnson BK, Scholz MB, Teal TK, Abramovitch RB. 2016. SPARTA: Simple
program for automated reference-based bacterial RNA-seq transcriptome analysis. BMC
Bioinformatics 17:66.

27.

Schindler BD, Joshi RJ, Vieille C. 2014. Respiratory glycerol metabolism of
Actinobacillus succinogenes 130Z for succinate production. J Ind Microbiol Biotechnol
41:1339-1352.

28.

Mulligan C, Fischer M, Thomas GH. 2011. Tripartite ATP-independent periplasmic
(TRAP) transporters in bacteria and archaea. FEMS Microbiol Rev 35:68-86.

29.

Rhie MN, Yoon HE, Oh HY, Zedler S, Unden G, Kim OB. 2014. A Na+-coupled C4dicarboxylate transporter (Asuc_0304) and aerobic growth of Actinobacillus
succinogenes on C4-dicarboxylates. Microbiology 160:1533-1544.

30.

Aguilar-Barajas E, Diaz-Perez C, Ramirez-Diaz MI, Riveros-Rosas H, Cervantes C.
2011. Bacterial transport of sulfate, molybdate, and related oxyanions. Biometals 24:687707.

31.

McKinlay JB, Shachar-Hill Y, Zeikus JG, Vieille C. 2007. Determining Actinobacillus
succinogenes metabolic pathways and fluxes by NMR and GC-MS analyses of 13Clabeled metabolic product isotopomers. Metab Eng 9:177–192.

32.

van der Werf MJ, Guettler MV, Jain MK, Zeikus JG. 1997. Environmental and
physiological factors affecting the succinate product ratio during carbohydrate
fermentation by Actinobacillus sp. 130Z. Arch Microbiol 167:332–342.

33.

Golby P, Kelly DJ, Guest JR, Andrews SC. 1998. Transcriptional regulation and
organization of the dcuA and dcuB genes, encoding homologous anaerobic C4dicarboxylate transporters in Escherichia coli. J Bacteriol 180:6586-6596.

81

Chapter 3

Identification of sRNA in Actinobacillus succinogenes 130Z

82

3.1 Abstract
Small RNAs (sRNAs) are powerful tools in metabolic engineering and synthetic biology.
Synthetic sRNAs are being designed using Hfq-dependent sRNA scaffolds. Hfq is a chaperone
protein known to facilitate binding of sRNAs to their mRNA targets and also protecting them
from degradation. Recent studies have focused on the identification of sRNAs in a few
commercially-relevant organisms, particularly Escherichia coli, to develop them as tools for
metabolic engineering. Actinobacillus succinogenes is one such industrially-relevant organism,
known to be one the best natural succinate producers. In this study, we identified sRNAs in A.
succinogenes to gain better insight into their sequence, size, structure, and function. Our goal
was to gain enough information to be able to design synthetic sRNAs that would act as posttranscriptional regulators in our strain. We performed RNAseq analysis of the wild-type strain
grown microaerobically on glycerol and anaerobically on glucose. RNAseq data were analyzed
using Rockhopper and manually using the Integrated Genomics Viewer. Using ARNold, we
identified sRNAs with Rho-independent terminators, a key characteristic of Hfq-dependent
sRNAs. We also looked for other known features of Hfq-dependent sRNAs, which allowed us to
narrow our focus down to only a few candidates for our synthetic sRNA design. Scaffolds from
two sRNAs—smRNA8 and smRNA28, were used in synthetic sRNA constructs to test inhibition
of β-galactosidase expression in A. succinogenes. A 32% decrease in β-galactosidase activity
was seen with one of the constructs. The target binding region for lacZ was then replaced by the
target binding regions for ackA and pta. One of four constructs we tested caused a 14% decrease
in acetate yield. To our knowledge, this is the first study identifying sRNAs in A. succinogenes.
We have also provided a proof of concept for using synthetic sRNAs in A. succinogenes as a

83

metabolic engineering tool. However, more optimization of these synthetic sRNAs in A.
succinogenes is needed to develop a robust and tunable system for future engineering efforts.

3.1 Introduction
Small RNAs (~30-500 nucleotides) are regulatory RNAs known to regulate mRNA
transcript levels and translation in bacteria. Since they were first discovered 30 years ago, a large
number of small RNAs (sRNAs) have been identified in several bacterial species, in part due to
advances in deep RNA sequencing and computational tools (1, 2). It is now known that sRNAs
have a significant role in gene regulation, but understanding of their mechanisms of action lags
behind, and many of their targets remain unknown. sRNAs are most commonly classified by
their mechanism of action as reviewed by Gottesman et al. (3). Antisense sRNAs can be
classified as cis-encoded sRNAs or trans-encoded sRNAs. Most commonly studied are transencoded antisense sRNAs (encoded by a locus different from their target[s]) that bind to their
target mRNA(s) with limited and varying degrees of complementarity. Due to their limited
binding capabilities these sRNAs may require the assistance of Sm-like protein Hfq (4) Cisencoded antisense sRNAs are usually encoded by the same locus as the mRNA but on the
opposite strand, making them entirely complementary to their target mRNA, in contrast to transencoded antisense sRNAs. Riboswitches are another group of cis-encoded and -acting RNA
elements, which sense small molecules and change conformation to allow or block translation
(5). In some cases they have been shown to act in trans (6). CRISPRs (clustered regularly
interspaced short palindromic repeats) are a newly identified group of sRNAs playing an
important role in bacterial immunity against foreign DNA (7), (3). Many small RNAs are also
synthesized from the 3’ regions of mRNA. These small RNAs are produced by mRNA

84

processing or transcribed separately with a shared terminator (8). A couple of studies have
shown that 3’-UTR derived sRNAs acted as trans regulators for mRNAs (9, 10). DapZ was the
first example of a small RNA that was transcribed by an independent promoter from the sense
3’UTR of a gene, dapB (10). DapB is responsible for the catalyzing the second step of lysine
biosynthesis (11). The function of DapZ was found to be regulation of major ABC transporters.
Thus the gene and the small RNA derived from the sense 3’UTR of the gene had completely
different functions. Another example; CpxQ is a sRNA that is generated by RNase E cleavage of
CpxP, a stress chaperone. CpxQ, an Hfq-dependent sRNA was found to repress production of
several inner membrane proteins (9). However, in this case, the gene and sRNA both were
transcribed from the same promoter and therefore transcribed under the same conditions.
Most of the studies that have identified and characterized sRNAs have focused on bacterial
members of the Enterobacteriaceae family, mainly in Escherichia coli. Sixty one sRNAs have
been annotated in Ecogene for E. coli, but many more have been predicted. Many research
groups are now investigating how these newly identified sRNAs function in gene regulation. A
few studies have been conducted on sRNAs in other bacterial families. Most of them focus on
human pathogens, such as Clostridium difficile, Neisseria gonorrhoeae, Acinetobacter
baumannii, Pseudomonas aeruginosa, Rickettsia prowazekii, Burkholderia cenocepacia,
Streptococcus pneumoniae, Listeria monocytogenes, and Mycobacterium tuberculosis, to
identify virulence-related sRNAs (12-20). Very few studies have focused on the Pasteurellaceae
family, even though many members of this family are known to be veterinary pathogens (e.g.,
Actinobacillus pleuropneumoniae, Haemophilus parasuis, Mannheimia paralytica, and
Pasteurella multocida), and two, Haemophilus influenzae and Aggregatibacter
actinomycetemcomitans, cause disease in humans (19). A majority of the Pasteurellaceae species

85

are difficult to culture outside their host environment, and they are known to have relatively
small genomes possibly due to adaptation to their host environment (21). Very little is known
about the sequence, structure, and function of sRNAs in Pasteurellaceae, apart from a few studies
in H. influenzae and A. pleuropneumoniae. HrrF was the first sRNA identified in a Haemophilus
species. It is expressed at high levels when iron availability is low, is regulated by Fur, and is
conserved in several other Pasteurellaceae species (22). Eighteen sRNAs have been identified in
H. influenzae by RNA sequencing (RNAseq), of which HrrF has been experimentally verified
(22). Of these eighteen sRNAs, seven belong to known sRNA families in the Rfam database
(23). Seven sRNAs possibly interacting with the carbon storage regulatory protein CsrA were
computationally identified in Haemophilus spp. genomes (24). Three sRNAs were also identified
in A. actinomycetemcomitans that are regulated by iron and Fur. Most recently, A.
pleuropneumoniae sRNAs were identified using four prediction algorithms—BLAST/Rfam,
SIPHT, Infernal, and RNAz. sRNAs predicted by two or more programs were considered to be
candidates for further analysis. Seventeen of the 23 sRNAs identified using this approach were
confirmed by northern blotting (25).
Actinobacillus succinogenes is a non-pathogenic member of the Pasteurellaceae family.
This capnophilic organism, isolated from a cow’s rumen, is among the best natural succinate
producers (26). We are interested in engineering A. succinogenes for succinate production.
Although knockout methods and overexpression vectors exist, they are not always the best
means to engineer the organism for industrial level production. Many a times a more fine-tuned
approach is needed. Synthetic sRNAs have being designed and used to modulate gene expression
in a couple of industrially relevant microorganisms (27, 28). We sought to identify sRNAs in A.

86

succinogenes and gain further information on their sequences, structure, size, and function, to
then be able to design synthetic sRNAs for engineering purposes.
In this study, we carried out deep sequencing of sRNAs from A. succinogenes grown on
two different carbon sources—glucose and glycerol. We analyzed the RNAseq data using
Rockhopper and by manually reviewing the data on the Integrative Genomics Viewer (IGV). We
compared the data obtained from our sequencing results to data obtained from the predictions of
a few computational programs. We also sought to identify Hfq-dependent sRNAs in A.
succinogenes, and tested them as scaffolds in the design of synthetic regulatory sRNAs.

3.2 Materials and Methods
3.2.1 Bacterial strains and growth conditions.
E. coli strains (Table 3.1) were cultivated in lysogeny broth (LB) and plates and were
supplemented with 100 µg mL-1 ampicillin for plasmid maintenance where required. A.
succinogenes type strain 130Z (ATCC 55618) was grown in 60-mL AM3 medium containing 50
mM glucose and 150 mM NaHCO3 under a nitrogen atmosphere in 150-mL anaerobic bottles.
AM3 is a phosphate-based defined medium containing vitamins and the three amino acids
cysteine, methionine, and glutamate (29). Pre-cultures were grown in the same medium in 10-mL
volumes in 28-mL anaerobic tubes. Anaerobic tubes and bottles were flushed with O -free N gas
2

2

(Airgas, Independence, OH) for 10 min, stoppered with rubber bungs, flushed for another 10 min
and sealed with aluminum crimps. For growth on glycerol, A. succinogenes was grown in AM3
supplemented with 150 mM glycerol and 150 mM NaHCO in continuous culture conditions as
3

described (30). Cultures were quenched by mixing 25-mL of culture at an OD660 of ~ 0.9-1.0
with 25-mL cold methanol and stored at -20 °C until RNA purification. Two independent

87

biological replicates were harvested for each condition. For strains expressing the synthetic
sRNA constructs grown on lactose, cultures were supplemented with 50 mM lactose and 40 μg
mL-1 ampicillin. All strains used in this study are listed in Table 3.1.

Table 3.1 Strains and plasmids used in this study
Description

Source

E. coli
DH5α

F- φ80lacZΔM15 Δ(lacZYA-argF)U169 recA1 endA1 hsdR17

Laboratory
collection

TOP10

(rk-, mk+) phoA supE44 λ- thi-1 gyrA96 relA1 F- mcrA Δ(mrrhsdRMS-mcrBC) φ80lacZΔM15 ΔlacΧ74 recA1 araD139 Δ(araleu) 7697 galU galK rpsL (StrR) endA1 nupG λ-

Invitrogen

A. succinogenes
130Z (ATCC55618)

Wild-type strain

ATCC

Plasmids
pCR2.1 TOPO

AmpR, KmR, lacZα, cloning vector

pLGZ920

E. coli-A. succinogenes shuttle vector; AmpR; A. succinogenes ppckA

pLGZ920*-ppckA-92smRNA_lacZ1

pLGZ920 derivative containing synthetic sRNA smRNA_lacZ1
downstream of ppckA-92 promoter

This study

pLGZ920*-ppckA-92smRNA_lacZ2

pLGZ920 derivative containing synthetic sRNA smRNA_lacZ2
downstream of ppckA-92 promoter

This study

pLGZ920*-ppckA-92smRNA_lacZ3

pLGZ920 derivative containing synthetic sRNA smRNA_lacZ3
downstream of ppckA-92 promoter

This study

pLGZ920*-ppckA-92smRNA_lacZ4

pLGZ920 derivative containing synthetic sRNA smRNA_lacZ4
downstream of ppckA-92 promoter

This study

pLGZ920*-ppckA-92smRNA_ackA1

pLGZ920 derivative containing synthetic sRNA smRNA_ackA1
downstream of ppckA-92 promoter

This study

pLGZ920*-ppckA-92smRNA_ackA2

pLGZ920 derivative containing synthetic sRNA smRNA_ackA2
downstream of ppckA-92 promoter

This study

pLGZ920*-ppckA-92smRNA_pta1

pLGZ920 derivative containing synthetic sRNA smRNA_pta1
downstream of ppckA-92 promoter

This study

pLGZ920*-ppckA-92smRNA_pta2

pLGZ920 derivative containing synthetic sRNA smRNA_pta2
downstream of ppckA-92 promoter

This study

88

Invitrogen
(31)

3.2.2 RNA isolation, library preparation, and sequencing.
The 50-mL bacterial suspensions in methanol were centrifuged at 4,500×g for 15 min at 4
°C. The cell pellets were used to isolate the total RNA, including sRNAs, using the miRNeasy
mini kit (Qiagen, Valencia, CA). Contaminating DNA was removed using DNase I (Ambion,
Austin, TX). The RNA was tested for chromosomal DNA contamination by PCR and was then
cleaned up using Zymo RNA Clean & Concentrator-5 (Zymo Research, Irvine, CA). RNA
quantity and quality were analyzed on a Qubit fluorometer (Invitrogen, Carlsbad, CA) and 2100
Bioanalyzer (Agilent Technologies; Santa Clara, CA), respectively. Total RNAs that passed the
quality check were submitted to the Michigan State University Research Technology Support
Facility for sequencing. Small RNA libraries were created using the Illumina small RNA Sample
Prep Kit and were purified from 6% native polyacrylamide gels by excising fragments in the 22to-500 bp range. The sRNA libraries were quality controlled by validating them using Qubit
dsDNA, Caliper LabChipGX, and Kapa qPCR assays. The libraries were pooled in equimolar
amounts for multiplexed sequencing, and loaded onto one lane of an Illumina HiSeq 2500 Rapid
Run flow cell (v2). Sequencing was performed in a 2×100bp paired-end format using HiSeq
Rapid SBS reagents. Bases were called using Illumina Real Time Analysis (RTA) v 1.18.64. The
RTA output was demultiplexed and converted to FASTQ format with Illumina Bcl2fastq v1.8.4.

3.2.3 Trimming of reads, mapping, and sRNA identification.
Demultiplexed raw FASTQ files were received in paired-end format (R1 files containing
left reads and R2 files containing right reads). These paired FASTQ files were trimmed to
remove low-quality bases and adapters using Trimmomatic (v0.33) (32). Illumina clip was run in
palindrome mode and specified to keep both reads. The sliding window was set at 4 bp, and

89

reads were discarded when the quality dropped below 20 or when read length was below 15
bp. FastQC analysis was done before and after trimming of the reads. Reads that passed filtering
were aligned with the genome (Genbank accession no. NC_009655) using Rockhopper (33, 34)
in the reference-based transcript assembly mode. The aligned reads were then analyzed with
Rockhopper twice with different expression parameters. In run 1, the maximum number of bases
between paired-end mates and the minimum expression of UTRs and ncRNAs were set at 300
and 0.5, respectively (referred to as 300_0.5 in text ahead). In run 2, the maximum number of
bases between paired end mates was 300, and the minimum expression of UTRs and ncRNAs
was 0.1 (referred to as 300_0.1 in text ahead). Small RNAs were also manually identified by
visually inspecting the Rockhopper alignment files in IGV. Transcript start and stop coordinates
were manually determined by finding the maximum expression level and extending the transcript
start and stop until the expression level dropped below 10% of the maximum.

3.2.4 Small RNA analysis.
All sRNA sequences were input into ARNold (Rho-independent terminator prediction
program) to detect rho-independent terminators. If Rockhopper-determined sRNAs were shorter
than the manually-annotated transcripts in IGV, the coordinates used were those observed in
IGV. The program was run under default conditions. The free energy for the positive sequences
was noted. IntaRNA (35, 36), run using default parameters, was used to identify target mRNAs
for the sRNAs. RNAfold (37) was used to determine sRNA secondary structures wherever
needed.

90

3.2.5 Computational prediction of sRNAs.
Small RNAs were downloaded from the web-based platform SIPHT (available at
http://newbio.cs.wisc.edu/sRNA/). SIPHT can search for intergenic loci that contain putative
Rho-independent terminators in any of the bacterial replicons in the NCBI database. The
parameters used were those mentioned by the authors (38). The sRNAs predicted for A.
succinogenes were also downloaded from the Bacterial Small Regulatory RNA Database
(BSRD, http://www.bac-srna.org/BSRD/taxonomyIndexNew.jsp) (39). The BSRD parameters
were a maximum E-value of 1 × 10-15, TransTerm confidence value of 87%, maximum
RNAMotif score of -9, FindTerm score of -10, and minimum and maximum lengths for
predicted loci of 50 and 500, respectively.
INFERNAL (40), RNAz (41), and Blast against Rfam (blasts the genome against
previously described sRNAs)(42)—all available through the platform RNAspace
(www.rnaspace.org/) (43) were also run. INFERNAL, which searches for homologs of structural
RNAs in sequence databases like the Rfam 10.0 database by building covariance models (CM),
was run with the default parameters. RNAz is a program that can detect structurally conserved,
functional, thermodynamically stable RNA in genome-wide screens or multiple alignments.
RNAz was run using the default parameters, with a probability cutoff of 0.7, slice alignments
longer than 300, window size of 200, and stepsize of 50. RNAz used the annotated genomes of
A. pleuropneumoniae L20 and H. influenzae Rd KW20 for alignment with the A. succinogenes
genome, using BLAST with default parameters in RNAspace. In the same RNAspace run,
sequence aggregation was also done using CG-seq using default parameters (i.e., score lambda
parameter of 1, minimal and maximal lengths of a conserved region of 30 and 500, minimum and

91

maximum identity thresholds of 60 and 100, respectively). CRISPRs were identified in the
genome using CRISPRFinder (44) with default parameters.

3.2.6 RT-PCR validation of small RNAs.
First-strand cDNA was synthesized from 1 µg of total RNA with random hexamers and
SuperScript II reverse transcriptase (Invitrogen, Carlsbad, CA) according to the manufacturer’s
instructions. The reverse transcriptase reaction was used as the template for PCR reactions using
primers designed for each sRNA candidate to be verified. As a positive control, A. succinogenes
genomic DNA was used as the template for PCR. To verify the absence of genomic DNA,
DNase-treated total RNA (not subjected to reverse transcriptase) was used as the template. PCR
reactions were done in a 20-µL volume containing Taq polymerase, 1X Taq buffer, 50 µM of
each dNTP, 1.5 mM MgCl2, and 1.25 µM of each primer. Thermocycler conditions were as
follows: 95 °C for 3 min; 32 cycles of 95 °C for 30 sec, 60 °C for 30 sec and 72 °C for 1 min; 72
°C for 5 min. PCR products were visualized on a 3% agarose gel with a low molecular weight
ladder (New England Biolabs, Ipswich, MA). All sRNAs were tested using cDNA synthesized
from total RNA from glucose-grown cultures, except for smRNAs 47 and 126, which were tested
with cDNA synthesized from RNA obtained from glucose and glycerol cultures.

3.2.7 Construction of synthetic small RNA constructs.
Synthetic sRNA constructs designed to inhibit lacZ expression—smRNA_lacZ1,
smRNA_lacZ2, smRNA_lacZ3, and smRNA_lacZ4, were constructed by annealing
oligonucleotides CV662 and CV663, CV664 and CV665, CV666 and CV667, and CV668 and
CV669 respectively (Table S3.1). Annealed oligonucleotides were used as templates and

92

amplified using primers CV677 and CV678 (smRNA_lacZ1), and primers CV677 and CV 679
(smRNA_lacZ2, smRNA_lacZ3, and smRNA_lacZ4). PCR products were cloned into
pLGZ920*-pAsuc_0701-lacZ’s BamHI and SacI sites using Gibson cloning (New England Biolabs,
Ipswich, MA). pLGZ920*- pAsuc_0701-lacZ is a derivative of pLGZ920 expressing lacZ, in which
the pckA promoter was replaced by the promoter of Asuc_0701, and the HindIII restriction site
has been removed (see chapter 2). Digesting pLGZ920*- pAsuc_0701-lacZ with BamHI and SacI
removes pAsuc_0701 and lacZ. Ligation mixtures were transformed into E. coli Top10 (Invitrogen)
chemically competent cells and plated on LB agar plates containing 100 µg mL-1 ampicillin.
Transformant colonies were tested for the presence of the insert by colony PCR with primers
CV677 and CV640, followed by plasmid sequencing with primer CV640 (GENEWIZ, Inc.,
South Plainfield, NJ). Positive clones were transformed into A. succinogenes strain 130Z by
electroporation as described (31).
Synthetic sRNA constructs targeting acetate kinase (AckA) and phosphotransacetylase
(Pta) expression—smRNA_ackA1, smRNA_ackA2, smRNA_pta1, smRNA_pta2, were ordered
as g-blocks from Integrated DNA Technologies (Coralville, IA) (Table S3.1). All g-blocks were
digested with BamHI and XbaI and ligated into pLGZ920*-pAsuc_0701’s BamHI and XbaI sites as
described above. Colonies were screened by colony PCR using primers CV677 and CV640, and
plasmid inserts were verified by sequencing (GENEWIZ, Inc.). Positive clones were transformed
into A. succinogenes 130Z.

3.2.8 Beta-galactosidase assays.
Strain 130Z carrying constructs smRNA_lacZ1, smRNA_lacZ2, and smRNA_lacZ4 was
grown in AM3 supplemented with 50 mM lactose, 150 mM NaHCO3, and 40 µg mL-1

93

ampicillin, and harvested in mid-exponential phase. The positive control was strain 130Z
(pLGZ290). Cultures (200 µL) were transferred to 96-well plates and optical density at 600 nm
(OD600) was recorded on a PowerWave HT microplate reader (BioTek Instruments Inc.,
Winooski, VT). OD600 values were used as a measure of cell biomass. One hundred µL of
cultures were transferred to a new 96-well plate and lysed by adding 100 µL of Bugbuster
Protein Extraction Reagent (Millipore Sigma, Billerica, MA) and incubating for 5 min. The
lysate was used as the crude extract for β-galactosidase assays. β-galactosidase assays were
conducted in 96-well plates in the PowerWave HT microplate reader. Reactions (200 µL)
contained 102.5 µL cell extract mixed with 67.5 µL of Z-buffer (60 mM Na2HPO4 , 40 mM
NaH2PO4 , 10 mM KCl, 1 mM MgSO4, 38 mM 2-mercaptoethanol, pH 7.0) and 30 µL of 2nitrophenyl β-D-galactopyranoside (4 mg mL-1 in Z-buffer). Enzyme activity was calculated
using the linear slope obtained from 2-nitrophenol production recorded at 420 nm for 20 min at
room temperature. Specific activity values are reported in ΔA420 min-1 ml-1 OD600-1 and are the
average of three independent biological replicates.

3.2.9 Determination of fermentation balances.
Growth of A. succinogenes in liquid cultures was monitored by measuring OD660 on a DU
650 spectrophotometer (Beckman, Fullerton, CA). Samples collected early (OD660 ~ 0.5) and late
(OD660 ~ 1.5) in the exponential phase were used to determine the carbon balance. Glucose and
organic acids were quantified in the 10-fold diluted, filtered culture supernatants by HPLC
(Waters, Milford, MA) using an Aminex HPX-87H column (Bio-rad). Samples were run with 4
mM H2SO4 as the eluent at a 0.6 mL min-1 flow rate at room temperature. Glucose and ethanol
were quantified on a Waters 410 differential refractometer while organic acids were quantified

94

on a Waters 2487 UV detector at 210 nm. OD660 values were used to calculate the biomass as
described (45).

3.3 Results and Discussion
3.3.1 Detection of sRNAs from RNAseq data using RockHopper.
The overall approach we took for identifying sRNAs is described in Figure 3.1. A.
succinogenes sRNAs were sequenced in cultures grown anaerobically on glucose and
microaerobically on glycerol, with two independent biological replicates for each condition.
Over 95% of the read pairs remained after trimming and quality control, the details of which are
mentioned in Table 3.2.

Figure 3.1 Flow chart of the approach used for sRNA identification in A. succinogenes.

95

Table 3.2 Statistics on read alignments by Rockhopper
Sample

Glucose-1
Glucose-2
Glycerol-1
Glycerol-2

No of read
Read pairs
Successful
Alignment to
Alignment to
pairs
remaining after alignment by protein coding protein coding
before
trimming
Rockhopper genes (sense) genes (antisense)
trimming (% surviving) (% alignment)
(%)
(%)
12,495,687
13,066,506
85
16
1
(95.6%)
11,809,343
12,423,762
85
16
1
(95.0%)
10,452,076
10,842,533
86
9
2
(96.4%)
10,236,278
10,637,886
85
11
2
(96.2%)

96

Alignment to
rRNAs
(%)

Alignment
to tRNAs
(%)

Alignment to
unannotated
regions
(%)

65

5

13

66

5

12

76

5

7

73

5

10

The Rockhopper-predicted sRNAs were manually inspected by visualization in IGV. Three
types of Rockhopper-predicted sRNAs were discarded or modified: i) sRNAs in sense direction
located in 3’UTR or 5’UTR region of annotated mRNA, whose transcript levels did not differ
much from the corresponding mRNA (sense) transcript levels were deleted; ii) sRNAs that were
shorter than 30 nt were deleted, unless they appeared to be longer in IGV; and iii) two or more
separate RNAs that were immediately consecutive but that, upon visual inspection, seemed to
belong to one bigger sRNA transcript, were listed as a unique, larger transcript.
After analysis of the Rockhopper results (from 300_0.5 and 300_0.1), a total of 145 sRNAs
were detected in glucose and glycerol cultures combined (Table 3. 3). Of these 145 sRNAs, five
were found in glucose-grown cultures only, while 40 were found in glycerol-grown cultures only
(Table 3.3). All others were found in both culture conditions. An additional 115 more sRNAs
that had been missed by Rockhopper were found upon manual inspection in IGV (Table S3.1). In
total, these 260 sRNAs were classified into six separate categories, as described in Figure 3.2.

Figure 3.2 Classification of identified sRNAs.

97

Table 3.3 A. succinogenes sRNAs detected by RNAseq and Rockhopper analysis in anaerobically-grown glucose cultures and
microaerobically-grown glycerol cultures
Candidates
smRNA1
smRNA2
smRNA3
smRNA4
smRNA5
smRNA6
smRNA7
smRNA8
smRNA9
smRNA10
smRNA11
smRNA12
smRNA13
smRNA14**
smRNA15
smRNA16
smRNA17
smRNA18
smRNA19
smRNA20
smRNA21
smRNA22
smRNA23
smRNA24
smRNA25
smRNA26
smRNA27

Rockhopper parameters; max bases between paired end mates=300 ; minimum expression of UTRs and nc RNAS=0.5
Rfam
Transcription Transcription
Predicted by algorithms
Size Strand
Neighborhood
Annotation
start
stop
134,044
134,063
19
+
ASUC_RS00615/ASUC_RS00620
368,282
368,340
58
+
ASUC_RS01695/ASUC_RS01700
369,316
369,371
55
+
ASUC_RS01700/ASUC_RS01705
PyrR
431,715
431,814
99
ASUC_RS01960/ASUC_RS01965
I
517,122
517,162
40
+
ASUC_RS02365/ASUC_RS02370
Z
553,139
553,170
31
ASUC_RS02590/ASUC_RS02595
609,640
609,664
24
+
ASUC_RS02865/ASUC_RS02870
744,502
744,593
91
ASUC_RS03550/ASUC_RS03555
785,724
785,705
19
ASUC_RS03760/ASUC_RS03765
Z
837,968
838,075
107
ASUC_RS04020/ASUC_RS04025
MicF
981,582
981,628
46
+
ASUC_RS04740/ASUC_RS04745
I
1,000,490
1,000,464
26
ASUC_RS04845/ASUC_RS04850
1,013,544
1,013,530
14
ASUC_RS04900/ASUC_RS04905
1,044,143
1,044,177
34
ASUC_RS05040/ASUC_RS05045
1,045,717
1,045,764
47
ASUC_RS05045/ASUC_RS05050
1,048,744
1,048,726
18
ASUC_RS05055/ASUC_RS05060
1,136,086
1,136,035
51
ASUC_RS05410/ASUC_RS05415
SRP_bact
1,215,624
1,215,735
111
+
ASUC_RS05760/ASUC_RS05765
B, I, Z, M
1,237,630
1,237,591
39
ASUC_RS05855/ASUC_RS0560
1,276,490
1,276,511
21
+
ASUC_RS06025/ASUC_RS06030
Z
1,276,603
1,276,695
92
+
ASUC_RS06025/ASUC_RS06030
Z
1,367,934
1,367,964
30
+
ASUC_RS06370/ASUC_RS06375
1,376,784
1,376,856
72
+
ASUC_RS06445/ASUC_RS11015
Z
1,441,632
1,441,660
28
+
ASUC_RS06770/ASUC_RS06775
S
1,525,386
1,525,407
21
ASUC_RS07160/ASUC_RS07165
1,545,573
1,545,663
90
+
ASUC_RS07260/ASUC_RS07265
Z
GcvB
1,862,491
1,862,517
26
+
ASUC_RS08780/ASUC_RS08785
B, I, M

98

Table 3.3 (cont’d)
smRNA28
smRNA29
smRNA30
smRNA31

2,003,880
2,003,940
60
+
ASUC_RS09455/ASUC_RS09460
S
2,089,221
2,089,240
19
+
ASUC_RS09840/ASUC_RS09845
Z
2,204,631
2,204,672
41
ASUC_RS10330/ASUC_RS10335
2,275,657
2,275,690
33
ASUC_RS10685/ASUC_RS10690
Rockhopper parameters; max bases between paired end mates=300 ; minimum expression of UTRs and nc RNAS=0.1
smRNA32
22
81
59
ASUC_RS10935/ASUC_RS00005
smRNA33*
16,626
16,657
31
ASUC_RS00085/ASUC_RS00090
smRNA34*
16,749
16,849
100
ASUC_RS00085/ASUC_RS00090
smRNA35*
17,297
17,329
32
ASUC_RS00090/ASUC_RS00095
smRNA36
44,059
44,089
30
+
ASUC_RS00210/ASUC_RS00215
smRNA37
74,727
74,758
31
+
ASUC_RS00320/ASUC_RS00325
smRNA38*
91,637
91,689
52
ASUC_RS00400/ASUC_RS00405
smRNA39
132,913
132,945
32
ASUC_RS00610/ASUC_RS00615
smRNA40
134,004
134,071
67
ASUC_RS00615/ASUC_RS00620
smRNA41
278,628
278,675
47
ASUC_RS01270/ASUC_RS01275
smRNA42
289,411
289,455
44
ASUC_RS01335/ASUC_RS01340
smRNA43
313,893
314,010
117
+
ASUC_RS01445/ASUC_RS01450
smRNA44*
330,475
330,511
36
ASUC_RS01530/ASUC_RS01535
smRNA45
330,852
330,883
31
ASUC_RS01535/ASUC_RS01540
smRNA46**
353,040
353,070
30
+
ASUC_RS01650/ASUC_RS01655
smRNA47**
356,657
356,723
66
ASUC_RS01655/ASUC_RS01660
smRNA48**
359,036
359,095
59
ASUC_RS01660/ASUC_RS01665
smRNA49
412,230
412,339
109
+
ASUC_RS01890/ASUC_RS01895
smRNA50
501,364
501,412
48
+
ASUC_RS02295/ASUC_RS02300
smRNA51
508,217
508,265
48
ASUC_RS02325/ASUC_RS02330
smRNA52
520,787
520,823
36
ASUC_RS02400/ASUC_RS02405
smRNA53
522,418
522,483
65
ASUC_RS02420/ASUC_RS02425
smRNA54
525,832
525,866
34
ASUC_RS02460/ASUC_RS02465
smRNA55
Alpha_RBS
527,930
528,046
116
+
ASUC_RS02475/ASUC_RS02480
S, B, I, Z, M
smRNA56
596,463
596,584
121
ASUC_RS02810/ASUC_RS02825
smRNA57 Parecho_CRE
599,496
599,569
73
+
ASUC_RS02835/ASUC_RS02840
I

99

Table 3.3 (cont’d)
smRNA58*
smRNA59*
smRNA60
smRNA61*
smRNA62*
smRNA63
smRNA64
smRNA65**
smRNA66*
smRNA67
smRNA68
smRNA69*
smRNA70*
smRNA71*
smRNA72*
smRNA73
smRNA74*
smRNA75
smRNA76
smRNA77
smRNA78
smRNA79
smRNA80
smRNA81
smRNA82*
smRNA83*
smRNA84
smRNA85
smRNA86*
smRNA87

Glycine
riboswitch
Thr_leader
Parecho_CRE

604,037
604,515
620,650
632,299
632,547
643,335
643,644
656,185
657,344
694,517
705,794
743,678
744,115
744,307
744,450
769,202
785,488
787,773

604,135
604,583
620,681
632,398
632,583
643,371
643,675
656,204
657,404
694,526
705,810
743,708
744,089
744,380
744,438
769,245
785,512
787,816

98
68
31
99
36
36
31
19
60
9
16
30
26
73
12
43
24
43

+
+
+
+
+
+
+
+
+
+
+

ASUC_RS02845/ASUC_RS02850
ASUC_RS02845/ASUC_RS02850
ASUC_RS02915/ASUC_RS02920
ASUC_RS02980/ASUC_RS02985
ASUC_RS02980/ASUC_RS02985
ASUC_RS03035/ASUC_RS03040
ASUC_RS03035/ASUC_RS03040
ASUC_RS03080/ASUC_RS03085
ASUC_RS03085/ASUC_RS03090
ASUC_RS03285/ASUC_RS03290
ASUC_RS03340/ASUC_RS03345
ASUC_RS03545/ASUC_RS03550
ASUC_RS03545/ASUC_RS03550
ASUC_RS03550/ASUC_RS03555
ASUC_RS03550/ASUC_RS03555
ASUC_RS03670/ASUC_RS03680
ASUC_RS03755/ASUC_RS03760
ASUC_RS03770/ASUC_RS03775

806,342

806,566

224

+

ASUC_RS03870/ASUC_RS03875

B, I

858,732
869,489
871,031
883,325
906,434
924,096
926,654
936,865
938,190
1,012,701
1,014,494

858,775
869,546
871,049
883,345
906,465
924,135
926,690
936,906
938,323
1,012,743
1,014,571

43
57
18
20
31
39
36
41
133
42
77

+
+
+
+
+
+
+
+
+

ASUC_RS04140/ASUC_RS04145
ASUC_RS04185/ASUC_RS04190
ASUC_RS04190/ASUC_RS04195
ASUC_RS04250/ASUC_RS04255
ASUC_RS04380/ASUC_RS04385
ASUC_RS04460/ASUC_RS04465
ASUC_RS04465/ASUC_RS04470
ASUC_RS04530/ASUC_RS04535
ASUC_RS04535/ASUC_RS04550
ASUC_RS04895/ASUC_RS04900
ASUC_RS04905/ASUC_RS04910

M
I

100

Table 3.3 (cont’d)
smRNA88
smRNA89*
smRNA90*
smRNA91
smRNA92
smRNA93
smRNA94
smRNA95*
smRNA96
smRNA97*
smRNA98
smRNA99*
smRNA100*
smRNA101*
smRNA102*
smRNA103
smRNA104*
smRNA105
smRNA106*
smRNA107*
smRNA108
smRNA109
smRNA110
smRNA111*
smRNA112
smRNA113
smRNA114
smRNA115
smRNA116*
smRNA117
smRNA118

1,057,772
1,067,837
1,069,547
1,073,915
1,083,430
1,117,501
1,119,016
1,146,049
1,165,100
1,240,529
1,253,225
1,250,832
1,252,103
1,252,811
1,264,237
1,265,927
1,266,017
1,266,795
1,267,226
1,267,779
1,296,921
1,298,854
1,300,630
1,368,446
1,370,455
1,376,675
1,377,939
1,400,489
1,415,979
1,424,398
1,455,834

1,057,841
1,067,870
1,069,772
1,073,974
1,083,410
1,117,539
1,119,057
1,146,091
1,165,133
1,240,560
1,253,175
1,250,866
1,252,198
1,252,842
1,264,348
1,265,963
1,266,073
1,266,826
1,267,256
1,267,810
1,296,999
1,298,937
1,300,662
1,368,552
1,370,499
1,376,702
1,377,967
1,400,682
1,416,016
1,424,495
1,455,884

69
33
225
59
20
38
41
42
33
31
50
34
95
31
111
36
56
31
30
31
78
83
32
106
44
27
28
193
37
97
50

+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

ASUC_RS05080/ASUC_RS05085
ASUC_RS05120/ASUC_RS05125
ASUC_RS05125/ASUC_RS05130
ASUC_RS05155/ASUC_RS05160
ASUC_RS05215/ASUC_RS05220
ASUC_RS05330/ASUC_RS05335
ASUC_RS05330/ASUC_RS05340
ASUC_RS05450/ASUC_RS05455
ASUC_RS05545/ASUC_RS05550
ASUC_RS05865/ASUC_RS05870
ASUC_RS05925/ASUC_RS05930
ASUC_RS05915/ASUC_RS05920
ASUC_RS05920/ASUC_RS05925
ASUC_RS05925/ASUC_RS05930
ASUC_RS05980/ASUC_RS05985
ASUC_RS05990/ASUC_RS05995
ASUC_RS05990/ASUC_RS05995
ASUC_RS05990/ASUC_RS05995
ASUC_RS05990/ASUC_RS05995
ASUC_RS05995/ASUC_RS06000
ASUC_RS06085/ASUC_RS06090
ASUC_RS06085/ASUC_RS06090
ASUC_RS06085/ASUC_RS06090
ASUC_RS06370/ASUC_RS06375
ASUC_RS06390/ASUC_RS06395
ASUC_RS06445/ASUC_RS11015
ASUC_RS11015/ASUC_RS06455
ASUC_RS06585/ASUC_RS06590
ASUC_RS06650/ASUC_RS06655
ASUC_RS06685/ASUC_RS06690
ASUC_RS06840/ASUC_RS06845

101

S

Table 3.3 (cont’d)
smRNA119*
1,493,409
1,493,453
44
+
ASUC_RS07025/ASUC_RS07030
smRNA120
1,499,794
1,499,829
35
ASUC_RS07050/ASUC_RS07055
smRNA121
1,528,991
1,528,963
28
ASUC_RS07175/ASUC_RS07180
smRNA122
1,543,327
1,543,374
47
+
ASUC_RS07245/ASUC_RS07250
smRNA123
1,586,321
1,586,358
37
ASUC_RS07425/ASUC_RS07430
smRNA124
1,623,706
1,623,739
33
+
ASUC_RS07620/ASUC_RS07625
smRNA125
1,671,194
1,671,207
13
+
ASUC_RS07840/ASUC_RS07845
smRNA126*
1,677,571
1,677,675
104
ASUC_RS07870/ASUC_RS07875
smRNA127
1,700,597
1,700,770
173
ASUC_RS07990/ASUC_RS07995
smRNA128
1,737,609
1,737,650
41
ASUC_RS08165/ASUC_RS08170
smRNA129*
1,901,859
1,901,893
34
ASUC_RS08940/ASUC_RS08945
smRNA130*
1,902,651
1,902,682
31
ASUC_RS08945/ASUC_RS08950
smRNA131*
1,927,500
1,927,538
38
ASUC_RS09070/ASUC_RS09075
smRNA132
1,974,287
1,974,272
15
ASUC_RS09300/ASUC_RS09305
smRNA133
1,977,589
1,977,719
130
ASUC_RS09310/ASUC_RS09315
smRNA134*
1,978,600
1,978,637
37
+
ASUC_RS09320/ASUC_RS09325
smRNA135
1,981,750
1,981,830
80
ASUC_RS09340/ASUC_RS09345
smRNA136
1,989,202
1,989,252
50
ASUC_RS09380/ASUC_RS09385
smRNA137
2,099,513
2,099,588
75
+
ASUC_RS09885/ASUC_RS09890
smRNA138
2,202,223
2,202,295
72
+
ASUC_RS10320/ASUC_RS10325
smRNA139
2,220,187
2,220,167
20
ASUC_RS10410/ASUC_RS10415
smRNA140
2,266,375
2,266,447
72
ASUC_RS10650/ASUC_RS10655
smRNA141
2,291,971
2,291,996
25
+
ASUC_RS10785/ASUC_RS10790
smRNA142*
2,300,851
2,300,963
112
+
ASUC_RS10825/ASUC_RS10830
smRNA143*
2,307,624
2,307,693
69
ASUC_RS10870/ASUC_RS10875
smRNA144
2,314,470
2,314,449
21
ASUC_RS10900/ASUC_RS10905
smRNA145*
2,315,613
2,315,645
32
ASUC_RS10910/ASUC_RS10915
**smRNA expressed in glucose cultures only
* smRNA expressed in glycerol cultures only.
A minimum cutoff of 50 for transcript level was used to determine whether or not a sRNA was expressed.
B, BSRD; I, INFERNAL; M, Rfam; S, SIPHT; Z, RNAz

102

Z

5’UTRs were defined as transcripts whose 3’ end preceded a coding sequence by fewer
than 50 nt. In some cases riboswitch transcripts did not end until well within the coding
sequence. For ease of classification, these riboswitches were still grouped in the 5’UTR category.
3’UTRs were defined as transcripts whose 5’ end started within the coding region or
immediately after the stop codon. Many transcripts started well within the coding sequence
(sense direction) and extended beyond the coding sequence, usually accompanied by lower
transcript levels of the gene itself (Figure 3.3). Antisense sRNAs were divided into three
categories: as5’UTRs, cis-asRNAs, and as3’UTRs. Transcripts antisense to the 5’UTR of the
coding sequence that started within 50 nt of the start codon were called as5’UTRs. Cis-asRNAs
were defined as antisense and started within the coding sequence.

Figure 3.3 Example of 3’UTR sRNA overlapping with the mRNA transcript and
downstream intergenic region with both mRNA and sRNA encoded on the reverse strand.
-, reverse strand; +, forward strand; red: reads on the reverse strand; blue: reads on the forward
strand; white arrowheads in gene indicate direction of the gene. Figure modified from screenshot
in IGV.

Transcripts antisense to the coding sequence that started within 50 nt of the stop codon
were called as3’UTRs. All others were called trans-encoded sRNAs. Using this classification,

103

the 260 sRNAs were catalogued as 28 5’UTRs, 71 3’UTRs, 2 as5’UTRs, 81 asRNAs, 10
as3’UTRs, and 68 trans-encoded sRNAs.

3.3.2 Comparison of RNAseq-identified to computationally predicted sRNAs.
SIPHT predicted 45 sRNAs, of which 40 were unknown sRNAs and five were known
sRNAs or riboswitches annotated in Rfam. Eleven of the SIPHT-predicted sRNAs were also
identified by sRNAseq, but some of them were combined since they looked like a single
transcript in IGV, giving a total of six sRNAs shared between the two—smRNA24, smRNA28,
smRNA55, smRNA88, smRNA175, and smRNA251 (Table S3.2). SIPHT predicted two sRNAs
that were already annotated in A. succinogenes as rRNAs or tRNAs—asuc_R0067 and
asuc_RS0071 (NCBI old locus tags). These were omitted in our analysis. For smRNA24, SIPHT
predicted nine different transcripts, five of which started from the same coordinate but had
different stop coordinates. The other four all had different start coordinates but shared the same
stop coordinate. Because smRNA24 corresponded to a continuous transcript in IGV, it was
considered to be a single transcript. SIPHT also predicted smRNA175 as two separate
transcripts, but since IGV showed it as a continuous transcript, it was deemed to be one sRNA.
Two of the sRNAs predicted by SIPHT and detected in RNAseq are already annotated in Rfam
as Alpha_RBS (smRNA55) and CRISPR-DR34 (smRNA175) while the other four are unknown.
BSRD (Bacterial Small Regulatory RNA Database) lists eighteen sRNA candidates for A.
succinogenes 130Z, all of which are orthologs of known sRNAs in other bacterial species.
Twelve of these were detected in our RNAseq study, of which three (not listed in Table S3.3) are
already annotated in A. succinogenes: asuc_R0067 (NCBI old locus tag) (annotated as tmRNA in
Rfam), ASUC_RS11040 (annotated as Rnase P classA in Rfam), and ASUC_RS09385

104

(annotated as S15 in Rfam). These three sRNAs were not considered for further analysis. The
remaining nine are listed in Table S3.3. BSRD lists two separate sRNAs spanning coordinates
806377‒806453 and 806464‒806566. Rockhopper predicted a single sRNA instead (smRNA76),
with coordinates 806342‒806566. smRNA76 was considered to be one single sRNA after visual
confirmation in IGV (Figure 3.4).

Figure 3.4 smRNA76 is seen as a single transcript in IGV. +, forward strand; -, reverse strand;
blue: reads detected in RNAseq; green: Rockhopper-predicted sRNA in both glucose and
glycerol libraries; red bars, two transcripts predicted by BSRD. Figure modified from screenshot
in IGV.

INFERNAL, RNAz, and Blastn against Rfam predicted 167, 108, and 115 putative non
coding RNAs (including tRNAs and rRNAs) respectively. Fifteen were common between
INFERNAL and RNA-seq, all of which are annotated in Rfam (Table S3.4). INFERNAL
predicted two separate transcripts for smRNA76, which we described above as a single transcript
(Figure 3.4 and Table S3.4).
Twenty sRNAs were common between RNAz and RNA-seq, three of which are annotated
in Rfam (Table S3.5). RNAseq and Blastn against Rfam had ten sRNAs in common, with
annotations listed in Table S3.6. CRISPRFinder predicted two CRISPR elements in the A.

105

succinogenes genome, which were both detected in RNAseq. One of the CRISPR elements is
annotated in Rfam as CRISPR –DR34 (smRNA175) (predicted as 7 separate trascripts in Rfam
spanning coordinates 436,472 and 436,903), while the other is unannotated (smRNA24).
smRNA24 had previously been described in our A. succinogenes genome paper (46) using
CRISPRs web service. Overall, thirteen of small RNAs detected in RNAseq were also detected
by at least one of the computational programs. Twenty-nine of the sRNAs detected by RNAseq
were predicted by at least one of the computational programs, three were predicted by two
computational programs, four were predicted by three of the programs, and two were predicted
by four programs, while only one was predicted by all five computational programs.

3.3.3 sRNAs with Rho-independent terminators.
Many sRNAs depend on Hfq, a chaperone protein, to bind to their target mRNAs and for
stability (4). Hfq typically binds to the U-rich region immediately preceding the poly(U) tail
found in RNAs with Rho-independent terminators. Thus sRNAs with Rho-independent
terminators are possibly Hfq-dependent. Twenty four Rho-independent terminators were
identified among the 260 sRNAs detected in RNAseq (Rockhopper and manually identified)
using ARNold (Table 3.4).

3.3.4 RT-PCR validation of small RNAs.
We selected fourteen sRNAs for validation with RT-PCR—smRNA4, 8, 11, 18, 21, 27, 28,
47, 49, 55, 76, 126, 135, and 187. smRNA4 was selected because it was predicted by a single
computational algorithm, INFERNAL. smRNA8 and smRNA27 were not predicted by any of the
computational algorithms. smRNA11 was selected because it is annotated as MicF in Rfam and
was also predicted by INFERNAL. smRNA18 was selected because it was antisense to ffs

106

Table 3.4 Rho-independent smRNAs predicted by ARNold
sRNA
candidate

Transcription Transcription
starta
stopa

Size

Strand

Neighborhood

Free
energyb
(kcal/mol )

smRNA6

553,132

553,188

56

-

ASUC_RS02590/ASUC_RS02595

-5.06

smRNA8

744,498

744,593

95

-

ASUC_RS03550/ASUC_RS03555

-4.80

smRNA11

981,582

981,635

53

+

ASUC_RS04740/ASUC_RS04745

-12.7

smRNA13

1,013,483

1,013,544

61

-

ASUC_RS04900/ASUC_RS04905

-11.1

smRNA15

1,045,695

1,045,765

70

-

ASUC_RS05045/ASUC_RS05050

-8.45

smRNA16

1,048,685

1,048,748

63

-

ASUC_RS05055/ASUC_RS05060

-12.26

smRNA17

1,135,901

1,136,086

185

-

ASUC_RS05410/ASUC_RS05415

-11.2

smRNA25

1,525,370

1,525,415

45

-

ASUC_RS07160/ASUC_RS07165

-12.3

smRNA27

1,862,486

1,862,541

55

+

ASUC_RS08780/ASUC_RS08785

-12.86

smRNA28

2,003,880

2,003,947

67

+

ASUC_RS09455/ASUC_RS09460

-8.8

smRNA31

2,275,649

2,275,708

59

-

ASUC_RS10685/ASUC_RS10690

-10.67

smRNA77

858,732

858,865

133

+

ASUC_RS04140/ASUC_RS04145

-13.1

smRNA79

871,031

871,079

48

+

ASUC_RS04190/ASUC_RS04195

-9.3

smRNA88

1,057,768

1,057,841

73

-

ASUC_RS05080/ASUC_RS05085

-11.1

smRNA113

1,376,675

1,376,719

44

+

ASUC_RS06445/ASUC_RS11015

-3.4

smRNA132

1,974,240

1,974,280

40

-

ASUC_RS09300/ASUC_RS09305

-9.2

smRNA153

102,498

102,589

91

+

ASUC_RS003455/ASUC_RS00460

-11.64

smRNA175c

436,421

436,943

522

-

ASUC_RS01975/ASUC_RS01980

-10.4

smRNA185

691,158

691,339

181

+

ASUC_RS03270/ASUC_RS03275

-8.99

smRNA207

1,169,658

1,169,834

176

+

ASUC_RS05560/ASUC_RS05565

-11.1

smRNA209

1,185,693

1,185,896

203

-

ASUC_RS05625/ASUC_RS05630

-11.2

smRNA227

1,520,160

1,520,316

156

-

ASUC_RS07125/ASUC_RS07130

-14.2

smRNA238

1,724,668

1,724,828

160

-

ASUC_RS08120/ASUC_RS08125

-10.3

smRNA240
1,825,558
1,825,646
88
ASUC_RS08605/ASUC_RS08610
-7.5
sRNA sizes predicted by Rockhopper were often shorter than the transcripts seen in IGV and did not include the
rho-independent terminators. Rockhopper coordinates were extended at the 3’end until they included the rhoindependent terminator, but only until the transcript ends seen in IGV. These extended corodinates were used as
input in ARNold and are listed in this table.
a

b

Free energy of the predicted terminator stem-loop structure using RNAfold (47), can be used as a confidence value
for predicted terminators.
c

Rho-independent terminator is also predicted about 265 nts into the transcript of smRNA175, which is annotated as
CRISPR-DR34 with a free energy of stem loop region of -6.10 kcal/mol.

(encoding the signal recognition particle 4.5S RNA), was predicted by most algorithms, and is
annotated as SRP_bact in Rfam (Table 3.2). smRNA21 and smRNA137 were only predicted by
RNAz. smRNA27 is annotated as GcvB in Rfam and was selected because it was identified as a

107

regulatory ncRNA in E.coli and other enteric bacteria. It regulates several protein-encoding
genes involved in metabolic pathways (48). smRNA47 was detected only in the glucose-grown
cultures in our RNAseq analysis (any sRNA expression level below 50 was considered as not
expressed). smRNA49 was selected as it was classified as a 3’UTR sRNA in our analysis.
smRNA55 was classified as a 5’UTR and annotated in Rfam as Alpha_RBS. smRNA76 was
selected because BSRD predicts it as 2 separate transcripts, but it looks like a single transcript on
IGV (Figure 3.3). smRNA126 was classified as an as3’UTR and was detected only in the
glycerol cultures. smRNA187 was selected as it is annotated as 6S RNA in Rfam.
All sRNAs tested by RT-PCR showed a PCR product of the expected size (Figure 3.5).
smRNA47 was tested with cDNA synthesized from RNA extracted from both growth conditions.
Even though smRNA47 was not detected in our RNAseq data of glycerol cultures, it produced a
PCR band in both conditions. This may be due to a small amount of transcript being expressed in
AM3-glycerol that may have been missed by RNAseq, or the expression level was below our
cutoff of 50. Note that RT-PCR is not quantitative and that even trace amounts of DNA can be
amplified to produce intense bands. It is not due to contaminating genomic DNA in the sample,
as the negative control with the RNA sample does not produce a PCR band (lane 11).
(Note that the band seen at ~50 bp in lane 11 [DNase-treated total RNA used as template] is
likely due to primer dimers, as the same band is not seen with another primer set tested with the
same template [lane 17]). A similar result (a PCR product of the expected size for both glucose
and glycerol samples, lanes 15-16), with a similar interpretation, was obtained for smRNA126,
for which we expected a PCR product only with the cDNA originating from the glycerol sample.
For smRNA76, the primers were designed to amplify a product only if the two sRNAs predicted

108

by BSRD corresponded to a single transcript. The 240 bp PCR product seen for smRNA76 (lane
14) confirms our RNAseq results.
As described above, we chose the 14 small RNAs to cover a wide variety of sRNAs
belonging to different categories in our analysis. We were able to confirm all these sRNAs by
RT-PCR. Based on these results, most other sRNAs detected by RNAseq are likely to be valid
sRNAs.

Figure 3.5 RT-PCR validation of small RNAs. Lane 1, low molecular weight ladder; Lane 2,
smRNA4; Lane 3, smRNA8; Lane 4, smRNA11; Lane 5, smRNA18; Lane 6, smRNA21; Lane
7, smRNA27, Lane 8, smRNA28; Lane 9, smRNA47 (cDNA from Glucose-1); Lane 10,
smRNA47 (cDNA from Glycerol-1), Lane 11, smRNA47 (DNAse treated RNA control from
Glycerol-1); Lane 12, smRNA49; Lane 13, smRNA55; Lane 14, smRNA76; Lane 15,
smRNA126 (cDNA from Glucose-1); Lane 16, smRNA126 (cDNA from Glycerol-1); Lane 17,
smRNA126 (DNase treated RNA control from Glucose-1); Lane 18, smRNA135; Lane 19,
smRNA187.

109

3.3.5 Scaffold selection for synthetic sRNA design based on predicted Hfq-dependent small
RNAs.
Among the twenty four Hfq-dependent small RNAs predicted by ARNold, we short-listed
a few candidates with other characteristic features of Hfq-dependent small RNAs known to be
important for binding to Hfq: (i) Rho-independent terminator with a long poly(U) tail, (ii)
internal hairpin, and (iii) U-rich sequence preceding the hairpin (49, 50). Seven candidates
(Table 3.5) met the three criteria. Out of these, smRNA8 and smRNA28 were selected as
scaffolds for synthetic small RNA design. In E. coli, sRNA MicA (51) is located near the 5’ end
of the luxS gene in the antisense orientation. In A. succinogenes, smRNA8 is similarly positioned
upstream of luxS (ASUC_RS03555) in the opposite orientation (Figure 3.6). Although these two
sRNAs do not share any sequence identity, they may be functionally similar. When we searched
for target genes for smRNA8, OmpA was one of the top predicted targets (data not shown). In E.
coli, MicA has been confirmed to be an antisense OmpA regulator and Hfq-dependent. We chose
the smRNA8 scaffold for testing our synthetic small RNA for these reasons. We chose
smRNA28 as a second scaffold, because it is conserved in at least three other Pasteurellaceae
species—Mannheimia succiniciproducens, Haemophilus somnus 2336, and Haemophilus
somnus 129PT.

110

Table 3.5 smRNA sequences that fit the criteria for Hfq-dependent sRNAs
sRNA candidates
smRNA8 (-)
smRNA11 (+)
smRNA13 (-)
smRNA16 (-)
smRNA27 (+)

smRNA28 (+)
smRNA77 (+)

Sequences

GCAGUUGUGAUUAAUAAUAAAAAAAUUGGUUCCUUAGGUUAUAUUCACCGCUCAAUUCCGCA
AGGAAAAGAGCGGUUUUUUUU
UUAUUCAUAAAACCCCUUUUUACGCCGGAUUCCCUUAGUCCGGCUUUUUU
UGGCUAAACCGAUUUCACCCAGUAAUUCGCCGGCAUUAGUACCGACAAUCGCACCACCUAAA
AUACGGUGCGAUUCUUUGUC
UCCUAAAUAAAAUAAUCAUAAAUAAAUAGCUAAACUACUUUCCUACCGUCCUUUUGGACGGU
UUUUUUUC
UUUCUAGUUUGUCCGCUCUGCUUUCUUUUUCUACAAUACGCGCAUACUUAAUGACUGGUAAU
UCCUUAAUUGAUUAAGAGUUGAAUCUUUUAGUUAAGUAUUAUGUUGUGUUUGCAUAUUGUUU
GGGUAACCAAACAAAAGUAAUUAAUCCUUCUAUUUAAUUACUUAUUAACUUCCUGUAUAUUU
ACUACCUAAUUUUAGGUUAUUGGCACCGCGCUUAAACUCCAAAAAAGUGCGGUGUUUUUU
AUUCAAACAAUAAUAGAUAAUCACUCCAACUUUCGGCGUUUCUCUCCCCACAAGGAAACGCC
UUUUUCU
AUUCGUUUUUAACGGAAAAAACACCAUGAAAUCCGACCGCACUUUUACCAUGAUGACGAUUA
CCACCAUUAUGACCUUUAUAAUGGCGGGGUAGUGCGAACGAAAGAUAUUCAAUUAAGCCCGC
AACCUGAAAAGUGCGGGCUUUUUU

+ and – signs indicate the coding strand. Blue: sequences that may be complementary to target
mRNA(s); red: inverted repeats in the hairpin; underline: AU-rich regions before the inverted
repeats; and black and italics: poly(U) tails.

Figure 3.6 A. succinogenes smRNA8’s genomic locus (Glucose-1 sample shown as a
representative sample). -, reverse strand; red, transcript level from RNAseq; green, sRNA
predicted by Rockhopper (in both glucose and glycerol libraries).

3.3.6 Testing of synthetic sRNA designs with lacZ as target gene.
Yoo et al. (52) designed a synthetic sRNA in E. coli using the scaffold from MicC, which
is a well-characterized E. coli sRNA. Yoo et al. simply changed the complementary target gene
binding regions. Addition of a terminator sequence was optional since the scaffold sequence has

111

a hairpin at the 3’ end. In that study, a target binding sequence of 20-30 nt with binding energy
between -30 and -40 kcal mol-1 worked best (52). Using these guidelines, we constructed four
synthetic sRNA constructs targeting lacZ using smRNA8 and smRNA28 as scaffolds (Figure
3.7). smRNA_lacZ1 contains the smRNA8 scaffold sequence. smRNA_lacZ2 contains the
smRNA28 scaffold sequence with a region complementary to the initial target mRNA kept intact
(in blue) (Figure 3.7). smRNA_lacZ3 contains the smRNA28 scaffold sequence without any
region complementary to the initial target mRNA. smRNA_lacZ1, smRNA_lacZ2, and
smRNA_lacZ3 contain the same lacZ target sequence. smRNA_lacZ4 has the same scaffold as
smRNA_lacZ3, but has a different lacZ target binding region. All constructs were cloned under
control of a promoter of moderate strength (constructed in chapter 2).
We were unable to obtain A. succinogenes transformants with pLGZ920-ppckA-92smRNA_lacZ3. Strains 130Z (pLGZ920-ppckA-92-smRNA_lacZ1), 130Z (pLGZ920-ppckA-92smRNA_lacZ2), and 130Z (pLGZ920-ppckA-92-smRNA_lacZ4) were grown on lactose to
determine whether or not lacZ was expressed. Although all three strains were able to grow on
AM3-lactose, 130Z (pLGZ920-ppckA-92-smRNA_lacZ1) and 130Z (pLGZ920-ppckA-92smRNA_lacZ2) grew slower than 130Z (pLGZ920-ppckA-92-smRNA_lacZ4) and 130Z
(pLGZ920), the positive control. Crude extracts of strain 130Z (pLGZ920-ppckA-92smRNA_lacZ1) had a 32% decrease in β-galactosidase activity compared to the positive control
(p = 0.007) (Figure 3.8). Crude extracts of strain 130Z (pLGZ920-ppckA-92-smRNA_lacZ2)
showed a slight decrease in β-galactosidase activity that was not significant (p = 0.12), and βgalactosidase activity of strain 130Z pLGZ920-ppckA-92-smRNA_lacZ4) was not affected.

112

CGAAAAUUUGAUCUAGUUAACAUUUUUUAGGUAUAAAUAGUUUUAAAAUAGAUCUAGUUUGGAUUCAAAGUAGGUUGGCAGAAUCA
UUCAAUUCCGCAAGGAAAAGAGCGGUUUUUUUU
(A)
CGAAAAUUUGAUCUAGUUAACAUUUUUUAGGUAUAAAUAGUUUUAAAAUAGAUCUAGUUUGGAUUCAAAGUAGGUUGGCAGAAUCA
UAAUCACUCCAACUUUCGGCGUUUCUCUCCCCACAAGGAAACGCCUUUUUCU
(B)
CGAAAAUUUGAUCUAGUUAACAUUUUUUAGGUAUAAAUAGUUUUAAAAUAGAUCUAGUUUGGAUUCAAAGUAGGUUGGCAGAAUCA
UAACUUUCGGCGUUUCUCUCCCCACAAGGAAACGCCUUUUUCU
(C)
CGAAAAUUUGAUCUAGUUAACAUUUUUUAGGUAUAAAUAGUUUUAAAAUAGAUCUAGUUUGGAUAGAAUCAUGCUGAACUCCUUAA
ACUUUCGGCGUUUCUCUCCCCACAAGGAAACGCCUUUUUCU
(D)
lacZ target binding regions
5’ AUGAUUCUGCCAACCUACUUUGAA 3’

(E)

5’ UAAGGAGUUCAGCAUGAUUCU 3’

(F)

Figure 3.7 Synthetic sRNAs targeting lacZ expression. (A) smRNA_lacZ1; (B)
smRNA_lacZ2; (C) smRNA_lacZ3; (D) smRNA_lacZ4; (E) Target binding region in synthetic
small RNAs in constructs A, B, and C; (F) Target binding region in synthetic sRNA construct D.
Underlined:ppckA-92 promoter; green highlight: sequence complementary to lacZ mRNA; italics:
scaffold sequence from smRNA8 (construct A) and smRNA28 (constructs B, C, and D); blue:
sequences that may be complementary to initial target mRNA(s); red: inverted repeats preceded
by AU-rich region and followed by poly(U) tail; bold: start codons in panels E and F. Binding
energy for complementary regions is -36 kcal mol-1 for constructs A, B, and C, and -32.4 kcal
mol-1 for construct D.

ΔA420 min-1 ml-1 OD600-1

800
700
600

**

500
400
300
200
100
0

1

2

3

4

Figure 3.8 β-galactosidase activity of strain 130Z expressing synthetic sRNAs targeting lacZ
expression. 1, 130Z (pLGZ920), positive control; 2, 130Z (pLGZ920-ppckA-92-smRNA_lacZ1); 3,
pLGZ920-ppckA-92-smRNA_lacZ2; and 4, pLGZ920-ppckA-92-smRNA_lacZ4. Results are the
average ± standard deviation based on three independent biological replicates.

113

3.3.7 Synthetic sRNAs for inhibiting acetate production.
Since a 32% decrease in β-galactosidase activity was observed using a synthetic sRNA
based on the smRNA8 scaffold sequence, we used the same scaffold to produce synthetic sRNAs
and decrease expression of the ackA-pta genes. The ackA and pta genes encode acetate kinase
and phosphate acetyltransferase, respectively, which are responsible for acetate production from
acetyl-CoA. We also used the smRNA28 scaffold of the smRNA_lacZ2 construct, which also
caused a slight decrease in β-galactosidase activity. In both scaffolds, we replaced the lacZ target
binding region specific for lacZ with ackA and pta target binding regions (Figure 3.9) to generate
four constructs. Fermentation balances of strains 130Z (pLGZ920-ppckA-92-smRNA_ackA1),
130Z (pLGZ920-ppckA-92-smRNA_ackA2), 130Z (pLGZ920- ppckA-92-smRNA_pta1) and 130Z
(pLGZ920-ppckA-92-smRNA_pta2) grown anaerobically on glucose are shown in Table 3.6. Only
strain 130Z (pLGZ920-ppckA-92-smRNA_ackA2) showed a significant decrease, 14%, in acetate
production compared to 130Z (pLGZ920). Surprisingly, the two sRNA constructs based on the
smRNA28 scaffold caused significant increases in acetate and succinate production at the
expense of ethanol and biomass production.

114

CGAAAAUUUGAUCUAGUUAACAUUUUUUAGGUAUAAAUAGUUUUAAAAUAGAUCUAGUUUGGACAGUUAAGAAUUAAAACUAAU
UUGGACAUUCAAUUCCGCAAGGAAAAGAGCGGUUUUUUUU
(A)
CGAAAAUUUGAUCUAGUUAACAUUUUUUAGGUAUAAAUAGUUUUAAAAUAGAUCUAGUUUGGACAGUUAAGAAUUAAAACUAAU
UUGGACAUAAUCACUCCAACUUUCGGCGUUUCUCUCCCCACAAGGAAACGCCUUUUUCU
(B)
CGAAAAUUUGAUCUAGUUAACAUUUUUUAGGUAUAAAUAGUUUUAAAAUAGAUCUAGUUUGGAGGGAUAAGAAUAAAUGUACGAGA
CAUUCAAUUCCGCAAGGAAAAGAGCGGUUUUUUUU
(C)
CGAAAAUUUGAUCUAGUUAACAUUUUUUAGGUAUAAAUAGUUUUAAAAUAGAUCUAGUUUGGAGGGAUAAGAAUAAAUGUACGAGA
CAUAAUCACUCCAACUUUCGGCGUUUCUCUCCCCACAAGGAAACGCCUUUUUCU
(D)
ackA target binding region
5’ AUGUCCAAAUUAGUUUUAAUUCUUAACUG 3’

(E)

pta target binding region
5’ AUGUCUCGUACAUUUAUUCUUAUCCC 3’

(F)

Figure 3.9 Synthetic sRNA designs targeting expression of ackA and pta. (A)
smRNA_ackA1; (B) smRNA_ackA2; (C) smRNA_pta1; (D) smRNApta2; (E) ackA target
binding regions in synthetic sRNAs A and B; (F) pta target binding regions in synthetic sRNAs
C and D. Underlined: ppckA-92 promoter; green highlight: sequences complementary to ackA
mRNA (A and B) and to pta mRNA (C and D) (binding energy is -35.2 kcal mol-1 [A and B] and
-36.7 kcal mol-1 [C and D]); italics: scaffold sequence from smRNA_lacZ1 (A and C) and
smRNA_lacZ2 (B and D); red: inverted repeats preceded by AU-rich region and followed by
poly(U) tail; bold: start codons.

115

Table 3.6 Fermentation balances of strain 130Z expressing synthetic sRNA constructs.
Products (mmol/100 mmol glucose consumed)
Strain
Succinate
130Z (pLGZ920)
47.9 ± 1.4
130Z (pLGZ920-ppckA-92-smRNA_ackA1) 58.9 ± 4.2*
130Z (pLGZ920-ppckA-92-smRNA_ackA2) 46.5 ± 1.4
130Z (pLGZ920-ppckA-92-smRNA_pta1) 60.8 ± 4.2*
130Z (pLGZ920-ppckA-92-smRNA_pta2) 56.6 ± 5.4

Formate
100 ± 2
106 ± 10
81.1 ± 4.2
103 ± 10
97.2 ± 8.8

Acetate
65.5 ± 1.6
76.8 ± 6.5*
56.1 ± 2.2**
73.6 ± 3.8*
68.2 ± 5.4

Ethanol
36 ± 3.4
25.3 ± 3.6
36.5 ± 16.9
25.5 ± 12.5
22.9 ± 7.6

CO2
2.06 ± 1.87
0.0±0.0
0.0±0.0
0.0±0.0
0.0 ±0.0

Biomassa
193 ± 10
165 ± 8
186 ± 3
146 ± 3
149 ± 8

Carbon
recoveryb
104 ± 6
104 ± 1
105 ± 3
102 ± 2
100 ± 9

Doubling
time
2.05 ± 0.03
2.45 ± 0.15
2.17 ± 0.05
2.51 ± 0.01
2.40 ± 0.09

Results are an average of three biological replicates ± standard deviations.
a
Biomass was determined using assumed values of 567 mg dry cell weight/mL per OD660 (45) and a cell composition of CH2O0.5N0.2
(24.967 g/mol) (53)
b
Carbon balance is the carbon in products/carbon in glucose consumed. It is assumed that one CO2 is fixed for each molecule of
succinate produced.
CO2 is calculated using the following formula: CO2 (in mM)= Ethanol (in mM)+ Acetate (in mM) - Formate (in mM)
* Significantly different from 130Z (pLGZ920) (p < 0.05, two-tailed student’s t-test)
** Significantly different from 130Z (pLGZ920) (p < 0.01, two-tailed student’s t-test)

116

3.4 Conclusion
In this study, we have successfully identified sRNAs in A. succinogenes and verified select
ones using RT-PCR. In all, 260 sRNAs were identified using Rockhopper and manual annotation
using IGV, in A. succinogenes grown anaerobically on glucose and/or microaerobically on
glycerol. Combined with RNAseq, we used five computational programs to identify additional
small RNAs in A. succinogenes. Thirty-nine of the sRNAs detected by RNAseq were also
predicted by at least one, if not more, of the computational programs we used. We validated 14
small RNAs from different categories in our analysis using RT-PCR.
We have also shown that synthetic sRNAs can be used in A. succinogenes to decrease gene
expression. While our first attempt at designing and using a synthetic sRNA was successful as a
proof of concept, much optimization is needed in the synthetic sRNA design and promoter
strength as well. Further optimization of these synthetic sRNAs would likely entail changing the
length and binding energy of the target regions or even testing several different scaffold
sequences. One experiment that is likely to give us more insight into Hfq-binding sRNAs is a copurification experiment using Hfq-his6. We plan to sequence the Hfq-binding RNAs obtained
from this experiment to confirm the Hfq-binding sRNAs identified in this study, maybe identify
more Hfq-binding sRNAs, and identify some of the target mRNAs.
Our study is a great first step towards understanding sRNAs and realizing their potential
as metabolic tools for engineering A. succinogenes and studying other pasteurellaceae species.

3.5 Acknowledgements
I thank Justin Beauchamp for his hands on help with the initial troubleshooting for analysis
of the RNAseq. Rhiannon LeVeque for sharing the Mulk’s laboratory RNA purification protocol.

117

I thank Dr. Martha Mulks for her helpful discussions in the initial planning phase of this study. I
thank Dr. Ben Johnson and John Johnson for helpful discussions regarding RNA analysis. I am
grateful to Jeff Landgraf of the MSU Genomics Core for making the small RNA libraries and
sequencing.

118

APPENDIX

119

Table A3.1 Manually identified smRNAs using alignment files created by Rockhopper
Candidate
smRNA146
smRNA147
smRNA148
smRNA149
smRNA150
smRNA151
smRNA152
smRNA153
smRNA154
smRNA155
smRNA156
smRNA157
smRNA158
smRNA159
smRNA160
smRNA161
smRNA162
smRNA163
smRNA164
smRNA165
smRNA166
smRNA167
smRNA168
smRNA169
smRNA170
smRNA171
smRNA172

Rfam
annotation

Transcriptio
n start

Transcription
stop

Size
(nt)

Coding
strand

Neighborhood

tRNA

30,584
31,859
39,498
45,565
48,319
83,116
102,424
102,498
102,542
113,439
115,771
130,180
138,915
180,930
194,098
195,459
199,493
199,497
214,454
214,473
218,099
248,074
293,615
298,474
306,273
332,486
339,352

30,640
31,907
39,529
45,614
48,364
83,215
102,470
102,582
102,588
113,530
115,918
130,330
139,034
181,103
194,153
195,540
199,572
199,570
214,513
214,508
218,172
248,108
293,774
298,517
306,369
332,548
339,413

56
48
31
49
45
99
46
84
46
91
147
150
119
173
55
81
79
73
59
35
73
34
159
43
96
62
61

+
+
+
+
+
+
+
+
+
+
+
+
+

ASUC_RS00135/ASUC_RS00140
ASUC_RS00140/ASUC_RS00145
ASUC_RS00185/ASUC_RS00190
ASUC_RS00220/ASUC_RS00225
ASUC_RS00230/ASUC_RS00235
ASUC_RS00360/ASUC_RS00365
ASUC_RS00455/ASUC_RS00460
ASUC_RS00455/ASUC_RS00460
ASUC_RS00455/ASUC_RS00460
ASUC_RS00530/ASUC_RS00535
ASUC_RS00535/ASUC_RS00540
ASUC_RS00600/ASUC_RS00605
ASUC_RS00630/ASUC_RS00635
ASUC_RS00810/ASUC_RS00815
ASUC_RS00875/ASUC_RS00880
ASUC_RS00880/ASUC_RS00885
ASUC_RS00895/ASUC_RS00900
ASUC_RS00895/ASUC_RS00900
ASUC_RS00965/ASUC_RS00970
ASUC_RS00965/ASUC_RS00970
ASUC_RS00975/ASUC_RS00980
ASUC_RS01110/ASUC_RS01115
ASUC_RS01350/ASUC_RS01355
ASUC_RS01375/ASUC_RS01380
ASUC_RS01415/ASUC_RS01420
ASUC_RS01540/ASUC_RS01545
ASUC_RS01585/ASUC_RS01590

120

Predicted
by algorithms
M

Table A3.1 (cont’d)
smRNA173
smRNA174
smRNA175
smRNA176
smRNA177
smRNA178
smRNA179
smRNA180
smRNA181
smRNA182
smRNA183
smRNA184
smRNA185
smRNA186
smRNA187
smRNA188
smRNA189
smRNA190
smRNA191
smRNA192
smRNA193
smRNA194
smRNA195
smRNA196
smRNA197
smRNA198
smRNA199
smRNA200
smRNA201
smRNA202
smRNA203

His_leader
6S RNA

Parecho_CRE

420,735
429,830
436,423
442,689
445,536
445,542
449,450
553,135
606,527
635,596
662,062
663,380
691,158
701,823
718,766
731,105
744,521
785,398
831,957
837,804
845,242
866,148
868,810
936,289
967,596
978,135
979,854
1,000,431
1,052,438
1,059,515
1,064,546

420,767
429,921
436,943
442,725
445,584
445,585
449,585
553,183
606,776
635,688
662,237
663,492
691,339
701,942
718,947
731,329
744,551
785,589
832,102
838,075
845,452
866,268
868,875
936,468
967,728
978,207
980,036
1,000,539
1,052,472
1,059,731
1,064,610

32
91
520
36
48
43
135
48
249
92
175
112
181
119
181
224
30
191
145
271
210
120
65
179
132
72
182
108
34
216
64

121

+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

ASUC_RS01910/ASUC_RS01915
ASUC_RS01955/ASUC_RS01960
ASUC_RS01975/ASUC_RS01980
ASUC_RS02010/ASUC_RS02015
ASUC_RS02030/ASUC_RS02035
ASUC_RS02030/ASUC_RS02035
ASUC_RS02055/ASUC_RS02060
ASUC_RS02590/ASUC_RS02595
ASUC_RS02850/ASUC_RS02855
ASUC_RS02995/ASUC_RS03000
ASUC_RS03105/ASUC_RS03110
ASUC_RS03115/ASUC_RS03120
ASUC_RS03270/ASUC_RS03275
ASUC_RS03320/ASUC_RS03325
ASUC_RS03405/ASUC_RS03410
ASUC_RS03475/ASUC_RS03480
ASUC_RS03550/ASUC_RS03555
ASUC_RS03755/ASUC_RS03760
ASUC_RS04000/ASUC_RS010990
ASUC_RS04020/ASUC_RS04025
ASUC_RS04055/ASUC_RS04060
ASUC_RS04165/ASUC_RS04170
ASUC_RS04180/ASUC_RS04185
ASUC_RS04530/ASUC_RS04535
ASUC_RS04675/ASUC_RS04680
ASUC_RS04730/ASUC_RS04735
ASUC_RS04735/ASUC_RS04740
ASUC_RS04845/ASUC_RS04850
ASUC_RS05065/ASUC_RS05070
ASUC_RS05085/ASUC_RS05090
ASUC_RS05105/ASUC_RS05110

S, M

B, I
B, I, Z, M

Z

Z
Z
Z
I

Table A3.1 (cont’d)
smRNA204
smRNA205
smRNA206
smRNA207
smRNA208
smRNA209
smRNA210
smRNA211
smRNA212
smRNA213
smRNA214
smRNA215
smRNA216
smRNA217
smRNA218
smRNA219
smRNA220
smRNA221
smRNA222
smRNA223
smRNA224
smRNA225
smRNA226
smRNA227
smRNA228
smRNA229
smRNA230
smRNA231
smRNA232
smRNA233
smRNA234

TPP

TPP riboswitch

1,101,346
1,136,235
1,148,049
1,169,658
1,178,550
1,185,698
1,192,341
1,192,343
1,226,566
1,249,629
1,254,045
1,258,601
1,259,001
1,350,512
1,356,570
1,377,923
1,378,631
1,379,749
1,412,500
1,430,292
1,438,223
1,440,698
1,498,787
1,520,162
1,543,389
1,607,502
1,625,036
1,628,813
1,644,162
1,654,562
1,665,202

1,101,403
1,136,327
1,148,185
1,169,830
1,178,645
1,185,896
1,192,408
1,192,388
1,226,635
1,249,785
1,254,220
1,258,640
1,259,063
1,350,633
1,356,750
1,377,967
1,378,725
1,379,827
1,412,583
1,430,344
1,438,385
1,440,777
1,498,840
1,520,316
1,543,436
1,607,619
1,625,112
1,628,940
1,644,218
1,654,858
1,665,386

57
92
136
172
95
198
67
45
69
156
175
39
62
121
180
44
94
78
83
52
162
79
53
154
47
117
76
127
56
296
184

122

+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
-

ASUC_RS05270/ASUC_RS05275
ASUC_RS05410/ASUC_RS05415
ASUC_RS05455/ASUC_RS05460
ASUC_RS05560/ASUC_RS05565
ASUC_RS05600/ASUC_RS05605
ASUC_RS05625/ASUC_RS05630
ASUC_RS05655/ASUC_RS05660
ASUC_RS05655/ASUC_RS05660
ASUC_RS05800/ASUC_RS05805
ASUC_RS05910/ASUC_RS05915
ASUC_RS05930/ASUC_RS05935
ASUC_RS05950/ASUC_RS05955
ASUC_RS05955/ASUC_RS05960
ASUC_RS06255/ASUC_RS06260
ASUC_RS06285/ASUC_RS06290
ASUC_RS06450/ASUC_RS06455
ASUC_RS06455/ASUC_RS06460
ASUC_RS06465/ASUC_RS06470
ASUC_RS06640/ASUC_RS06645
ASUC_RS06720/ASUC_RS06725
ASUC_RS06755/ASUC_RS06760
ASUC_RS06770/ASUC_RS06775
ASUC_RS07045/ASUC_RS07050
ASUC_RS07125/ASUC_RS07130
ASUC_RS07245/ASUC_RS07250
ASUC_RS07530/ASUC_RS07535
ASUC_RS07625/ASUC_RS7630
ASUC_RS07635/ASUC_RS07640
ASUC_RS07700/ASUC_RS07705
ASUC_RS07745/ASUC_RS07750
ASUC_RS07805/ASUC_RS07810

I

Z
Z
B, I, M

Table A3.1 (cont’d)
smRNA235
TPP riboswitch
1,668,651
smRNA236
1,707,974
smRNA237
1,712,810
smRNA238
1,724,670
smRNA239
1,741,704
smRNA240
1,825,558
smRNA241
1,837,348
smRNA242
MOCO_RNA_motif 1,838,604
smRNA243
1,851,093
smRNA244
1,862,282
smRNA245
1,881,063
smRNA246
1,968,657
smRNA247
1,985,767
smRNA248
2,001,685
smRNA249
2,013,568
smRNA250
2,038,722
smRNA251
2,089,369
smRNA252
2,144,262
smRNA253
2,144,312
smRNA254
2,177,961
smRNA255
2,181,502
smRNA256
2,222,240
smRNA257
2,258,917
smRNA258
2,271,792
smRNA259
2,282,160
smRNA260
2,286,315
B, BSRD; I, INFERNAL; M, Rfam; S, SIPHT; Z, RNAz

1,668,974
1,708,024
1,712,890
1,724,828
1,741,752
1,825,646
1,837,609
1,838,746
1,851,234
1,862,350
1,881,151
1,968,878
1,985,884
2,001,800
2,013,612
2,038,821
2,089,514
2,144,302
2,144,415
2,178,043
2,181,720
2,222,496
2,259,021
2,271,888
2,282,232
2,286,393

323
50
80
158
48
88
261
142
141
68
88
221
117
115
44
99
145
40
103
82
218
256
104
96
72
78

123

+
+
+
+
+
+
+
+
+
+
+

ASUC_RS07825/ASUC_RS07830
ASUC_RS08025/ASUC_RS08030
ASUC_RS08055/ASUC_RS08060
ASUC_RS08120/ASUC_RS08125
ASUC_RS08185/ASUC_RS08190
ASUC_RS08605/ASUC_RS08610
ASUC_RS08675/ASUC_RS08680
ASUC_RS08680/ASUC_RS08685
ASUC_RS08730/ASUC_RS08735
ASUC_RS08780/ASUC_RS08785
ASUC_RS08865/ASUC_RS08870
ASUC_RS09250/ASUC_RS09255
ASUC_RS09360/ASUC_RS09365
ASUC_RS09445/ASUC_RS09450
ASUC_RS09495/ASUC_RS09500
ASUC_RS09605/ASUC_RS09610
ASUC_RS09840/ASUC_RS09845
ASUC_RS10070/ASUC_RS10075
ASUC_RS10070/ASUC_RS10075
ASUC_RS10215/ASUC_RS10220
ASUC_RS10325/ASUC_RS10240
ASUC_RS10420/ASUC_RS10425
ASUC_RS10620/ASUC_RS10625
ASUC_RS10665/ASUC_RS10670
ASUC_RS10725/ASUC_RS10730
ASUC_RS10750/ASUC_S10755

B, I, M

B, I, M
Z
Z

S
Z

Table A3.2 smRNAs common between SIPHT and RNAseq
Candidate
smRNA24
smRNA28
smRNA55
smRNA88
smRNA175
smRNA251

Rfam
annotation
Alpha_RBS
CRISPR_DR34
-

Predicted by SIPHT
Transcription Transcription
start
stop
1,441,909
1,441,669
2,003,809
2,003,941
527,920
527,984
1,057,774
1,057,957
436,427
436,904
2,089,126
2,089,365

124

Size
(nt)
465
132
64
183
477
239

Coding
strand
+
+
+
+

Table A3.3 smRNAs common between BSRD and RNAseq

Candidate
smRNA18
smRNA27
smRNA55
smRNA76
smRNA185
smRNA187
smRNA222
smRNA235
smRNA242

Rfam
annotation
SRP_bact
GcvB
Alpha_RBS
Glycine riboswitch
His_leader
6S RNA
TPP riboswitch
TPP riboswitch
MOCO_RNA_motif

Listed in BSRD
Transcription
Transcription
Start
Stop
1,215,630
1,215,727
1,862,334
1,862,537
527,966
528,079
806,377
806,566
691,203
691,334
718,765
718,947
1,412,495
1,412,588
1,668656
1,668,748
1,838,596
1,838,749

125

Size
(nt)
98
204
114
180
132
183
94
221
154

Coding
strand
+
+
+
+
+
+
-

Table A3.4 smRNAs common between INFERNAL and RNAseq
Candidate

Rfam annotation

smRNA4
smRNA11
smRNA18
smRNA27
smRNA55
smRNA57
smRNA76
smRNA78
smRNA185
smRNA187
smRNA202
smRNA214
smRNA222
smRNA235
smRNA242

PyrR
MicF
SRP_bact
GcvB
Alpha_RBS
Parecho_CRE
Glycine
Parecho_CRE
His_leader
6S
Parecho_CRE
TPP
TPP
TPP
MOCO_RNA_motif

Predicted in INFERNAL
Transcription
Transcription
Size (nt)
start
stop
431,716
431,819
104
981,537
981,628
92
1,215,630
1,215,730
101
1,862,334
1,862,537
204
527,966
528,079
114
599,496
599,595
100
806,377
806,566
189
869,491
869,593
103
691,203
691,334
132
718,765
718,947
183
1,059,499
1,059,618
120
1,254,038
1,254,175
138
1,412,495
1,412,588
94
1,668,656
1,668,748
93
1,838,596
1,838,749
154

126

Coding
strand
+
+
+
+
+
+
+
+
+
+
+
-

Table A3.5 smRNAs common between RNAz and RNAseq
Candidate
smRNA5
smRNA9
smRNA18
smRNA20
smRNA21
smRNA23
smRNA26
smRNA29
smRNA55
smRNA135
smRNA187
smRNA190
smRNA199
smRNA200
smRNA201
smRNA219
smRNA221
smRNA243
smRNA244
smRNA252

Rfam annotation

SRP_bact

Alpha_RBS
6S

Predicted by RNAz
Transcription
Transcription
start
stop
517,020
51,7211
785,646
785,837
1,215,619
1,215,737
1,276,338
1,276,512
1,276,596
1,276,762
1,376,647
1,376,896
1,545,554
1,545,663
2,089,122
2,089,319
527,919
528,046
1,981,636
1,981,823
718,771
718,946
785,408
785,590
979,919
980,050
1,000,433
1,000,492
1,052,422
1,052,593
1,377,838
1,377,966
1,379,748
1,379,829
1,851,057
1,851,225
1,862,313
1,862,473
2,144,253
2,144,349

127

Size (nt)
192
192
119
175
167
250
110
198
128
188
176
183
132
60
172
129
82
169
161
97

Coding
strand
+
+
+
+
+
+
+
+
+
+
+
-

Table A3.6 smRNAs common between Rfam and RNAseq
Candidate
smRNA18
smRNA27
smRNA55
smRNA77
smRNA146
smRNA175
smRNA187
smRNA222
smRNA235
smRNA242

Rfam
annotation
SRP_bact
GcvB
Alpha_RBS
Thr_leader
tRNA
CRISPR-DR34
6S
TPP
TPP
MOCO_RNA_motif

Predicted by BLAST against Rfam
Transcription Transcription
Size (nt)
start
stop
1,215,630
1,215,730
101
1,862,318
1,862,528
211
5,279,66
5,28,079
114
8,58,744
8,58,862
119
30,593
30,634
42
4,36,472
4,36,903
36
718,765
718,947
183
1,412,505
1,412,583
79
1,668,661
1,668,738
78
1,838,596
1,838,749
154

128

Coding
strand
+
+
+
+
+
-

Table A3.7 Primers used for RT-PCR validation of sRNAs
Primer

Sequence

smRNA4_F
smRNA4_R
smRNA8_F
smRNA8_R
smRNA11_F
smRNA11_R
smRNA18_F
smRNA18_R
smRNA21_F
smRNA21_R
smRNA27_F
smRNA27_R
smRNA28_F
smRNA28_R
smRNA47_F
smRNA47_R
smRNA49_F
smRNA49_R
smRNA55_F
smRNA55_R
smRNA76_F
smRNA76_R
smRNA126_F
smRNA126_R
smRNA135_F
smRNA135_R
smRNA187_F
smRNA187_R

ATTGTAATGGCACTGCGAAATG
AAAAAACGATGCCCCCTGAG
AAGCAACTTGCAGTTGTG
AAAAAAAACCGCTCTTTTCCTTG
AACGTTGTGATGATGTTAAACG
AAAAAAAGCCGGACTAAGGGAAT
ACTAAGCCGGTGTGCGAAAG
GAAACCTCCCCAGTGATTC
AAAATTAACCGCACTTTATGTTC
ACAAGTAAACTCGCTACGC
TATTTCTAGTTTGTCCGCTCT
AACACCGCACTTTTTTGGAG
ATTCAAACAATAATAGATAATCACTCC
AAGAAAAAGGCGTTTCCTTGTG
TGCAATTCGTCGGAGATTTG
ATTGAACAATCCGAAAGTTCC
ATGCCGACTTTAGGACAAAAG
CTTACGGCGAGGGTATTC
ACAGGTTGAGCAGTTATACTG
GCACTCCTATATTTTAACTAATTTG
CAATAAAAAACTGTTCGGACGAAGG
GGCAGAAGAGATAAAAATGATAGG
GAATTTGGCAGAGAAGTAGTAC
AATACCCAAGCGGTCCTC
TTTAGCAAGTGCGGTCAG
AAAGGCAACCCTGCTTTAC
ATTACCTGAGATGCTCGCCAGC
GGAGTCAGTTGTAACCGTTTTTAGG

129

Expected product size
(n)
99
91
85
146
105
244
70
120
141
114
240
93
193
128

Table A3.8 Primers and g-blocks used in this study
Primer Sequence

Specificity (direction)

CV662a CATACGGATCCCGAAAATTTGATCTAGTTAACATTTTTTAGGTATAAATAGTTTT
AAAATAGATCTAGTTTGGAAGAATCATGCTGAACTCCTTAAACTTTCGGCGTTTC
TCTCCCCACAAGGAAACGCCTTTTTCTTCTAGAATCGA
CV663a TCGATTCTAGAAAAAAAAACCGCTCTTTTCCTTGCGGAATTGAATGATTCTGCCA
ACCTACTTTGAATCCAAACTAGATCTATTTTAAAACTATTTATACCTAAAAAATG
TTAACTAGATCAAATTTTCGGGATCCGTATG
CV664a CATACGGATCCCGAAAATTTGATCTAGTTAACATTTTTTAGGTATAAATAGTTTT
AAAATAGATCTAGTTTGGATTCAAAGTAGGTTGGCAGAATCATAATCACTCCAAC
TTTCGGCGTTTCTCTCCCCACAAGGAAACGCCTTTTTCTTCTAGAATCGA
a
CV665 TCGATTCTAGAAGAAAAAGGCGTTTCCTTGTGGGGAGAGAAACGCCGAAAGTTGG
AGTGATTATGATTCTGCCAACCTACTTTGAATCCAAACTAGATCTATTTTAAAAC
TATTTATACCTAAAAAATGTTAACTAGATCAAATTTTCGGGATCCGTATG
a
CV666 CATACGGATCCCGAAAATTTGATCTAGTTAACATTTTTTAGGTATAAATAGTTTT
AAAATAGATCTAGTTTGGATTCAAAGTAGGTTGGCAGAATCATAACTTTCGGCGT
TTCTCTCCCCACA AGGAAACGCCTTTTTCTTCTAGAATCGA
CV667a TCGATTCTAGAAGAAAAAGGCGTTTCCTTGTGGGGAGAGAAACGCCGAAAGTTAT
GATTCTGCCAACCTACTTTGAATCCAAACTAGATCTATTTTAAAACTATTTATAC
CTAAAAAATGTTAACTAGATCAAATTTTCGGGATCCGTATG
CV668a CATACGGATCCCGAAAATTTGATCTAGTTAACATTTTTTAGGTATAAATAGTTTT
AAAATAGATCTAGTTTGGATAGAATCATGCTGAACTCCTTAAACTTTCGGCGTTT
CTCTCCCCACAAG GAAACGCCTTTTTCTTCTAGAATCGA
CV669a TCGATTCTAGAAGAAAAAGGCGTTTCCTTGTGGGGAGAGAAACGCCGAAAGTTTA
AGGAGTTCAGCATGATTCTATCCAAACTAGATCTATTTTAAAACTATTTATACCT
AAAAAATGTTAACTAGATCAAATTTTCGGGATCCGTATG
CV677 AGCGCCTGATGCGGGGATCCCGAAAATTTGATCTAGTTAACATTTTTTAG

smRNA_lacZ1 (F)

CV678

CCAGTGAATTCGAGCTCAAAAAAAACCGCTCTTTTC

CV679

CCAGTGAATTCGAGCTCAGAAAAAGGCGTTTCCTTG

CV640
CV695a,b
CV696a,b
CV697a,b
CV698a,b

a
b

smRNA_lacZ1 (R)

smRNA_lacZ2 (F)

smRNA_lacZ2 (R)

smRNA_lacZ3 (F)

smRNA_lacZ3 (R)

smRNA_lacZ4 (F)

smRNA_lacZ4 (R)

smRNA_lacZ1 (F)
smRNA_lacZ1 (R)

smRNA_lacZ2, smRNA_lacZ3,
smRNA_lacZ4 (R)
CTTCGCTATTACGCCAGCTG
Sequencing primer for pLGZ920
(90 bp downstream of SacI site)
CATACGGATCCCGAAAATTTGATCTAGTTAACATTTTTTAGGTATAAATAGTTTT smRNA_ackA1
AAAATAGATCTAGTTTGGACAGTTAAGAATTAAAACTAATTTGGACATTCAATTC
CGCAAGGAAAAGAGCGGTTTTTTTTTCTAGAATCGA
CATACGGATCCCGAAAATTTGATCTAGTTAACATTTTTTAGGTATAAATAGTTTT smRNA_ackA2
AAAATAGATCTAGTTTGGACAGTTAAGAATTAAAACTAATTTGGACATAATCACT
CCAACTTTCGGCGTTTCTCTCCCCACAAGGAAACGCCTTTTTCTTCTAGAATCGA
CATACGGATCCCGAAAATTTGATCTAGTTAACATTTTTTAGGTATAAATAGTTTT smRNA_pta1
AAAATAGATCTAGTTTGGAGGGATAAGAATAAATGTACGAGACATTCAATTCCGC
AAGGAAAAGAGCGGTTTTTTTTTCTAGAATCGA
CATACGGATCCCGAAAATTTGATCTAGTTAACATTTTTTAGGTATAAATAGTTTT smRNA_pta2
AAAATAGATCTAGTTTGGAGGGATAAGAATAAATGTACGAGACATAATCACTCCA
ACTTTCGGCGTTTCTCTCCCCACAAGGAAACGCCTTTTTCTTCTAGAATCGA

Restriction sites BamHI and XbaI underlined
Sequences for g-blocks.

130

REFERENCES

131

REFERENCES

1.

Itoh T, Tomizawa J-I. 1980. Formation of an RNA primer for initiation of replication of
ColE1 DNA by ribonuclease H. Proc Natl Acad Sci USA 77:2450-2454.

2.

Stougaard P, Molin S, Nordström K. 1981. RNAs involved in copy-number control
and incompatibility of plasmid R1. Proc Natl Acad Sci USA 78:6008-6012.

3.

Gottesman S, Storz G. 2011. Bacterial small RNA regulators: versatile roles and rapidly
evolving variations. Cold Spring Harb Perspect Biol 3:1-16.

4.

Moller T, Franch T, Hojrup P, Keene DR, Bachinger HP, Brennan RG, ValentinHansen P. 2002. Hfq: a bacterial Sm-like protein that mediates RNA-RNA interaction.
Mol Cell 9:23-30.

5.

Mandal M, Boese B, Barrick JE, Winkler WC, Breaker RR. 2003. Riboswitches
control fundamental biochemical pathways in Bacillus subtilis and other bacteria. Cell
113:577-586.

6.

Loh E, Dussurget O, Gripenland J, Vaitkevicius K, Tiensuu T, Mandin P, Repoila
F, Buchrieser C, Cossart P, Johansson J. 2009. A trans-acting riboswitch controls
expression of the virulence regulator PrfA in Listeria monocytogenes. Cell 139:770-779.

7.

Brouns SJJ, Jore MM, Lundgren M, Westra ER, Slijkhuis RJH, Snijders APL,
Dickman MJ, Makarova KS, Koonin EV, Van Der Oost J. 2008. Small CRISPR
RNAs guide antiviral defense in prokaryotes. Science 321:960-964.

8.

Chao Y, Papenfort K, Reinhardt R, Sharma CM, Vogel J. 2012. An atlas of Hfq‐
bound transcripts reveals 3′ UTRs as a genomic reservoir of regulatory small RNAs.
EMBO J 31:4005-4019.

9.

Chao Y, Vogel J. 2016. A 3' UTR-derived small RNA provides the regulatory noncoding
arm of the inner membrane stress response. Mol Cell 61:352-363.

10.

Chao Y, Papenfort K, Reinhardt R, Sharma CM, Vogel J. 2012. An atlas of Hfqbound transcripts reveals 3' UTRs as a genomic reservoir of regulatory small RNAs.
EMBO J 31:4005-4019.

11.

Bouvier J, Stragier P, Morales V, Remy E, Gutierrez C. 2008. Lysine represses
transcription of the Escherichia coli dapB gene by preventing its activation by the ArgP
activator. J Bacteriol 190:5224-5229.

12.

Soutourina OA, Monot M, Boudry P, Saujet L, Pichon C, Sismeiro O, Semenova E,
Severinov K, Le Bouguenec C, Coppee JY, Dupuy B, Martin-Verstraete I. 2013.

132

Genome-wide identification of regulatory RNAs in the human pathogen Clostridium
difficile. PLoS Genet 9:e1003493.
13.

McClure R, Tjaden B, Genco C. 2014. Identification of sRNAs expressed by the human
pathogen Neisseria gonorrhoeae under disparate growth conditions. Front Microbiol
5:456-456.

14.

Sharma R, Arya S, Patil SD, Sharma A, Jain PK, Navani NK, Pathania R. 2014.
Identification of novel regulatory small RNAs in Acinetobacter baumannii. PloS one
9:e93833.

15.

Livny J, Brencic A, Lory S, Waldor MK. 2006. Identification of 17 Pseudomonas
aeruginosa sRNAs and prediction of sRNA-encoding genes in 10 diverse pathogens
using the bioinformatic tool sRNAPredict2. Nucleic Acids Res 34:3484-3493.

16.

Schroeder CL, Narra HP, Sahni A, Rojas M, Khanipov K, Patel J, Shah R, Fofanov
Y, Sahni SK. 2016. Identification and characterization of novel small RNAs in Rickettsia
prowazekii. Front Microbiol 7:859.

17.

Ramos CG, Grilo AM, da Costa PJ, Leitao JH. 2013. Experimental identification of
small non-coding regulatory RNAs in the opportunistic human pathogen Burkholderia
cenocepacia J2315. Genomics 101:139-48.

18.

Mann B, van Opijnen T, Wang J, Obert C, Wang YD, Carter R, McGoldrick DJ,
Ridout G, Camilli A, Tuomanen EI, Rosch JW. 2012. Control of virulence by small
RNAs in Streptococcus pneumoniae. PLoS Pathog 8:e1002788.

19.

Christiansen JK, Nielsen JS, Ebersbach T, Valentin-Hansen P, Sogaard-Andersen
L, Kallipolitis BH. 2006. Identification of small Hfq-binding RNAs in Listeria
monocytogenes. RNA 12:1383-96.

20.

Arnvig KB, Young DB. 2009. Identification of small RNAs in Mycobacterium
tuberculosis. Mol Microbiol 73:397-408.

21.

Kuhnert P, Christensen H. 2008. Pasteurellaceae: biology, genomics and molecular
aspects. Horizon Scientific Press.

22.

Santana EA, Harrison A, Zhang X, Baker BD, Kelly BJ, White P, Liu Y, Munson Jr
RS. 2014. HrrF is the Fur-regulated small RNA in nontypeable Haemophilus influenzae.
PloS one 9:e105644.

23.

Baddal B, Muzzi A, Censini S, Calogero RA, Torricelli G, Guidotti S, Taddei AR,
Covacci A, Pizza M, Rappuoli R. 2015. Dual RNA-seq of nontypeable Haemophilus
influenzae and host cell transcriptomes reveals novel insights into host-pathogen cross
talk. mBio 6:e01765-15.

133

24.

Sridhar J, Sekar K, Rafi ZA. 2009. CsrA interacting small RNAs in Haemophilus spp
genomes: a theoretical analysis. Arch Microbiol 191:451-459.

25.

Rossi CC, Bossé JT, Li Y, Witney AA, Gould KA, Langford PR, Bazzolli DMS.
2016. A computational strategy for the search of regulatory small RNAs in Actinobacillus
pleuropneumoniae. RNA 22:1373-1385.

26.

Guettler MV, Rumler D, Jain MK. 1999. Actinobacillus succinogenes sp. nov., a novel
succinic-acid-producing strain from the bovine rumen. Int J Syst Bacteriol 49:207–216.

27.

Cho C, Lee SY. 2016. Efficient gene knockdown in Clostridium acetobutylicum by
synthetic small regulatory RNAs. Biotechnol Bioeng 114:374-383.

28.

Na D, Yoo SM, Chung H, Park H, Park JH, Lee SY. 2013. Metabolic engineering of
Escherichia coli using synthetic small regulatory RNAs. Nat Biotechnol 31:170-174.

29.

McKinlay JB, Zeikus JG, Vieille C. 2005. Insights into Actinobacillus succinogenes
fermentative metabolism in a chemically defined growth medium. Appl Environ
Microbiol 71:6651-6656.

30.

Schindler BD, Joshi RJ, Vieille C. 2014. Respiratory glycerol metabolism of
Actinobacillus succinogenes 130Z for succinate production. J Ind Microbiol Biotechnol
41:1339-1352.

31.

Kim P, Laivenieks M, McKinlay J, Vieille C, Zeikus JG. 2004. Construction of a
shuttle vector for the overexpression of recombinant proteins in Actinobacillus
succinogenes. Plasmid 51:108–115.

32.

Bolger AM, Lohse M, Usadel B. 2014. Trimmomatic: a flexible trimmer for Illumina
sequence data. Bioinformatics 30:2114-20.

33.

Tjaden B. 2015. De novo assembly of bacterial transcriptomes from RNA-seq data.
Genome biology 16:1.

34.

McClure R, Balasubramanian D, Sun Y, Bobrovskyy M, Sumby P, Genco CA,
Vanderpool CK, Tjaden B. 2013. Computational analysis of bacterial RNA-Seq data.
Nucleic Acids Res 41:e140-e140.

35.

Busch A, Richter AS, Backofen R. 2008. IntaRNA: efficient prediction of bacterial
sRNA targets incorporating target site accessibility and seed regions. Bioinformatics
24:2849-2856.

36.

Wright PR, Georg J, Mann M, Sorescu DA, Richter AS, Lott S, Kleinkauf R, Hess
WR, Backofen R. 2014. CopraRNA and IntaRNA: predicting small RNA targets,
networks and interaction domains. Nucleic Acids Res 42:W119-23.

134

37.

Gruber AR, Lorenz R, Bernhart SH, Neuböck R, Hofacker IL. 2008. The Vienna
RNA Websuite. Nucleic Acids Res 36:W70-W74.

38.

Livny J, Teonadi H, Livny M, Waldor MK. 2008. High-throughput, kingdom-wide
prediction and annotation of bacterial non-coding RNAs. PLoS One 3:e3197.

39.

Li L, Huang D, Cheung MK, Nong W, Huang Q, Kwan HS. 2013. BSRD: a
repository for bacterial small regulatory RNA. Nucleic Acids Res 41:D233-D238.

40.

Nawrocki EP, Eddy SR. 2013. Infernal 1.1: 100-fold faster RNA homology searches.
Bioinformatics 29:2933-2935.

41.

Gruber AR, Findeiss S, Washietl S, Hofacker IL, Stadler PF. 2010. RNAz 2.0:
improved noncoding RNA detection. Pac Symp Biocomput:69-79.

42.

Nawrocki EP, Burge SW, Bateman A, Daub J, Eberhardt RY, Eddy SR, Floden
EW, Gardner PP, Jones TA, Tate J, Finn RD. 2015. Rfam 12.0: updates to the RNA
families database. Nucleic Acids Res 43:D130-D137.

43.

Cros MJ, de Monte A, Mariette J, Bardou P, Grenier-Boley B, Gautheret D, Touzet
H, Gaspin C. 2011. RNAspace.org: An integrated environment for the prediction,
annotation, and analysis of ncRNA. RNA 17:1947-1956.

44.

Grissa I, Vergnaud G, Pourcel C. 2007. CRISPRFinder: a web tool to identify clustered
regularly interspaced short palindromic repeats. Nucleic Acids Res 35:W52-W57.

45.

McKinlay JB, Shachar-Hill Y, Zeikus JG, Vieille C. 2007. Determining Actinobacillus
succinogenes metabolic pathways and fluxes by NMR and GC-MS analyses of 13Clabeled metabolic product isotopomers. Metab Eng 9:177–192.

46.

McKinlay JB, Laivenieks M, Schindler BD, Mckinlay AA, Siddaramappa S,
Challacombe JF, Lowry SR, Clum A, Lapidus AL, Burkhart KB, Harkins V, Vieille
C. 2010. A genomic perspective on the potential of Actinobacillus succinogenes for
industrial succinate production. BMC Genomics 11:680.

47.

Hofacker IL, Fontana W, Stadler PF, Bonhoeffer LS, Tacker M, P. S. 1994. Fast
folding and comparison of RNA secondary structures. Monatshefte für Chemie/Chemical
Monthly 125:167-188.

48.

Pulvermacher SC, Stauffer LT, Stauffer GV. 2009. Role of the Escherichia coli Hfq
protein in GcvB regulation of oppA and dppA mRNAs. Microbiology 155:115-123.

49.

Ishikawa H, Otaka H, Maki K, Morita T, Aiba H. 2012. The functional Hfq-binding
module of bacterial sRNAs consists of a double or single hairpin preceded by a U-rich
sequence and followed by a 3' poly(U) tail. RNA 18:1062-1074.

135

50.

Otaka H, Ishikawa H, Morita T, Aiba H. 2011. PolyU tail of rho-independent
terminator of bacterial small RNAs is essential for Hfq action. Proc Natl Acad Sci U S A
108:13059-13064.

51.

Udekwu KI, Darfeuille F, Vogel J, Reimegard J, Holmqvist E, Wagner EG. 2005.
Hfq-dependent regulation of OmpA synthesis is mediated by an antisense RNA. Genes
Dev 19:2355-2366.

52.

Yoo SM, Na D, Lee SY. 2013. Design and use of synthetic regulatory small RNAs to
control gene expression in Escherichia coli. Nat Protocols 8:1694-1707.

53.

van der Werf MJ, Guettler MV, Jain MK, Zeikus JG. 1997. Environmental and
physiological factors affecting the succinate product ratio during carbohydrate
fermentation by Actinobacillus sp. 130Z. Arch Microbiol 167:332–342.

136

Chapter 4

Development of a markerless knockout method for Actinobacillus succinogenes

137

This work is previously published in Applied and Environmental Microbiology as:

Joshi RV, Schindler BD, McPherson N, Tiwari K, Vieille C (2014) Development of a
Markerless Knockout Method for Actinobacillus succinogenes. Appl Environ Microbiol
80:3053–3061. doi:10.1128/AEM.00492-14

The article can be viewed at the following URL:
http://aem.asm.org/content/80/10/3053

138

The work described in this chapter was partly completed as a requirement for a professional
master’s degree and was submitted as a form of a report titled, “Development of a gene knockout
method for A. succinogenes strain 130Z and engineering of A. succinogenes for aspartate
production”. My contributions to this paper during my PhD are listed as below:


Construction of double knockouts ΔpflB ΔlacZ::icd, ΔpflB Δcit::icd, ΔpflB Δacn::icd.



Training and mentoring Kanupriya Tiwari during her research project in the summer of
2013. She worked on determining the effects of the DNA construct length and of the
length of the homologous regions on the efficiency of homologous recombination in A.
succinogenes.



Drafting parts, reading, and editing manuscript.

139

4.1 Abstract
Actinobacillus succinogenes is one of the best natural succinate-producing organisms, but it still
needs engineering to further increase succinate yield and productivity. In this study we
developed a markerless knockout method for A. succinogenes using natural transformation or
electroporation. The Escherichia coli isocitrate dehydrogenase gene with flanking flippase
recognition target sites was used as the positive selection marker, making use of A.
succinogenes’s auxotrophy for glutamate to select for growth on isocitrate. The Saccharomyces
cerevisiae flippase recombinase (Flp) was used to remove the selection marker, allowing its reuse. Finally, the plasmid expressing flp was cured using acridine orange. We demonstrate that at
least two consecutive deletions can be introduced into the same strain using this approach, that
no more than a total of 1 kb of DNA is needed on each side of the selection cassette to protect
from endonucleases activity during transformation, and that no more than 200 bp of homologous
DNA is needed on each side for efficient recombination. We also demonstrate that
electroporation can be used as an alternative transformation method to obtain knockout mutants,
and that an enriched defined medium can be used for direct selection of knockout mutants on
agar plates with high efficiency. Knockout mutants of the fumarate reductase and the citrate
lyase operons, as well as of the pyruvate formate lyase, β-galactosidase, and aconitase genes
were obtained using this knockout strategy.

140

4.2 Introduction
Dicarboxylic acids are among the top of the US’s Department of Energy list of valueadded chemicals from biomass (1). In particular, if produced economically by fermentation,
succinic acid could replace maleic anhydride as the precursor to many bulk and commodity
chemicals, with a potential market of 25 billion tons per year (2). Actinobacillus succinogenes
strain 130Z (ATCC 55618) is a facultative anaerobe and a member of the Pasteurellaceae. It is
the highest natural succinate producer known (3-5). A. succinogenes produces succinate, acetate,
and formate as its major fermentative products with ethanol as a minor by-product (6). Under
optimized conditions, wild-type A. succinogenes produces 80 g L-1 succinate, while chemically
induced mutant strains resistant to fluoroacetate produce up to 110 g L-1 succinate (3), suggesting
that even higher succinate yields can be obtained with genetically engineered strains. Bio-based
succinate will only be price-competitive with petroleum-based maleic anhydride if alternative
fermentation products can be eliminated. Achieving a homosuccinate fermentation would
drastically reduce the cost of downstream succinate purification (7).
Until recently, genetic tools for A. succinogenes were limited to the expression vector
pLGZ920 and electroporation (8). Plasmid pLGZ920 confers ampicillin (Amp) resistance,
replicates in A. succinogenes and Escherichia coli, and allows high-level expression of foreign
genes in A. succinogenes from the strong, constitutive A. succinogenes pckA promoter (ppckA)
(8). Electroporation of A. succinogenes with pLGZ920 yields transformants with an efficiency of
104‒106 CFU/μg of plasmid, depending on the electroporation parameters (8) (Schindler and
Vieille, unpublished results). Early attempts to construct knockout mutants of A. succinogenes by
allelic exchange used electroporation with suicide vectors containing gene knockout constructs
interrupted by antibiotic resistance genes. The high frequency of spontaneous antibiotic resistant

141

mutants masked the low frequency of double recombination events, and no knockout mutants
were isolated in these studies (McKinlay and Vieille, unpublished results). Another selection
method and possibly other means to introduce DNA into cells are thus needed to develop a
knockout method for A. succinogenes.
In other Pasteurellaceae species conjugation and natural transformation have been
commonly used to construct knockout mutants. Conjugation has been used in A.
pleuropneumoniae (9), Mannheimia haemolytica, Pasteurella multocida, and Haemophilus
somnus (10). Deletions were typically selected for with an antibiotic resistance marker, and
SacB-based sucrose counterselection was used to select for double recombination. The
conjugated plasmids were either suicide vectors or temperature-sensitive shuttle plasmids (10).
One study used the Cre-lox system to remove the selection marker (11). Natural transformation
has been used in many studies of H. influenzae and other naturally competent Pasteurellaceae.
Natural competence is induced by starvation stress. Transformation frequencies can be as high as
10-2 (12). Natural transformation is well-suited for strain engineering since it works best with
linear DNA (e.g., PCR products), which requires double recombination events for complete
chromosome integration.
Naturally competent Pasteurellaceae preferentially take up DNA from their own species
over unrelated DNA (13, 14). They do so with a membrane-bound DNA uptake machinery that
specifically recognizes and binds a conserved uptake signal sequence (USS) (15). Low
transformation efficiencies of A. actinomycetemcomitans were observed with DNA fragments
not containing a USS (16). All Pasteurellaceae genomes sequenced contain USS repeats, in
numbers ranging from 41 to 1,760 (1,690 in A. succinogenes) (17), even though not all these
species are naturally competent (12). A. succinogenes’s genome contains twenty three of the

142

twenty five genes in H. influenzae’s natural competence regulon (17, 18), including those
encoding the competence regulatory proteins Sxy and CRP (cyclic AMP receptor protein). The
missing two genes encode hypothetical proteins with unknown roles in natural competence (12).
The frequency of USS repeats in A. succinogenes’s genome and the likely presence of a
complete natural competence machinery suggest that A. succinogenes is naturally competent.
In this study, we demonstrate that natural competence can be used to create knockout
mutants of A. succinogenes, that the E. coli isocitrate dehydrogenase gene can be used as a
positive selection marker, that the Saccharomyces cerevisiae Flp recombinase (encoded by flp)
(19) can be used in A. succinogenes to remove the positive selection marker flanked by two
direct Flp recognition target sites (FRT) (20), and that more than one deletion can be introduced
in the same strain. We also demonstrate that no more than a total of 1 kb of DNA is needed on
each side of the selection cassette to protect from exonucleases activity during transformation,
and that no more than 200 bp of homologous DNA is needed for efficient recombination. The
genes encoding fumarate reductase and pyruvate formate lyase were chosen as initial targets for
knockout constructs. Fumarate reductase (encoded by frdABCD) converts fumarate to succinate.
Knocking out fumarate reductase would likely be the first step needed to engineer A.
succinogenes into a fumarate or aspartate producer. During A. succinogenes’s fermentative
growth on glucose, pyruvate formate lyase (PFL, encoded by pflB) is the main enzyme
converting pyruvate into acetyl-CoA, with the concomitant production of formate (21, 22).
Acetyl-CoA is then the precursor of acetate and ethanol (23). Fluoroacetate-resistant mutants
devoid of PFL activity are not affected in their growth and produce increased amounts of
succinate (3). A pflB knockout mutation is likely to be the first mutation needed to engineer a
homosuccinate-producing A. succinogenes strain (24-27). Three other sets of genes (encoding β-

143

galactosidase, citrate lyase, and aconitase) deemed non-essential for fermentative growth on
glucose were targeted for deletion as well.

4.3 Materials and Methods
4.3.1 Strains, media, culture conditions, and chemicals.
Strains used in this study are listed in Table 4.1. E. coli strains were cultivated in lysogeny
broth (LB) and on LB agar plates (28). A. succinogenes strains were cultivated in medium B (in
g L-1: yeast extract, 5; bactotryptone, 10; NaH2PO4·H2O, 8.5; K2HPO4, 15.5; NaHCO3, 2.1; and
glucose, 9), AM3 defined medium (29), or Bacto brain heart infusion (BHI; Becton-Dickinson
and Co., Franklin Lakes, NJ). AM3 is a phosphate-based medium containing glucose, NH4+, ten
vitamins, and the amino acids glutamate, cysteine, and methionine. After natural transformations,
A. succinogenes was grown in liquid AM2 (AM3 minus glutamate) in the absence or presence of
isocitrate (30 mM; AM2-isocitrate) or on AM16-isocitrate or AM17 agar plates. AM17 is
modified AM3 containing twice the amount of vitamins, cysteine, and methionine normally
present in AM3, plus an amino acid supplement mix (in mg L-1, final concentrations in AM17:
alanine, 120; aspartate, 105; asparagine·H2O, 120; glycine, 115; histidine, 21; isoleucine, 58;
leucine, 90; lysine(HCl)2, 113; phenylalanine, 45; serine, 44; threonine, 56; tryptophan, 13;
tyrosine, 28; and valine, 82). AM16 is AM17 without glutamate. A. succinogenes liquid cultures
were grown in N2-flushed 28-mL anaerobic tubes at 37°C, with shaking at 250 rpm. OD660 of A.
succinogenes cultures was monitored using a Spectronic 20 (Bausch and Lomb, Rochester, NY).
To isolate single colonies, A. succinogenes strains were grown on LB plates containing 10 g L-1
glucose (27) and 10 μg mL-1 kanamycin (Km, A. succinogenes is naturally resistant to Km). Amp
was added to all growth media for plasmid maintenance when required (40 μg mL-1 for A.

144

succinogenes and 100 μg mL-1 for E. coli). All cultures of A. succinogenes on agar plates were
grown under a CO2-enriched atmosphere. Unless stated otherwise, chemicals were from SigmaAldrich (St. Louis, MO).

Table 4.1 Strains and plasmids used in this study
Description

Source

E. coli
K-12 (ATCC19020) Wild-type strain

Laboratory collection

DH5α

F- φ80lacZΔM15 Δ(lacZYA-argF) U169 recA1 endA1 hsdR17

Laboratory collection

TOP10

(rk-, mk+) phoA supE44 λ- thi-1 gyrA96 relA1 F- mcrA Δ(mrrhsdRMS-mcrBC) φ80lacZΔM15 ΔlacΧ74 recA1 araD139 Δ(araleu) 7697 galU galK rpsL (StrR) endA1 nupG λ-

Invitrogen

A. succinogenes
130Z (ATCC55618)

Wild-type strain

ATCC

Δfrd::icd

130Z derivative, contains the AscI-FRT-ppckA-icd-FRT-AscI cassette This study
in the frdAB deletion

Δfrd

130Z derivative, frdAB deletion, contains one FRT site

ΔpflB::icd

130Z derivative, contains the AscI-FRT-ppckA-icd-FRT-AscI cassette This study
in the pflB deletion

ΔpflB
ΔpflB-ΔlacZ::icd
ΔpflB-ΔcitDEF
ΔpflB-Δacn

130Z derivative, pflB deletion, contains one FRT site
ΔpflB derivative, lacZ deletion, still contains the icd cassette
ΔpflB derivative, citDEF deletion
ΔpflB derivative, acn deletion

This study
This study
This study
This study

pCR2.1

AmpR, KmR, lacZα, cloning vector

Invitrogen

pLGZ920

E. coli-A. succinogenes shuttle vector; AmpR; A. succinogenes ppckA (8)

pCP20

pSC101 derivative, flp+, λ cI857+, λ pR Repts, AmpR, CmR

(36)

pCR2.1-icd

pCR2.1 derivative, E. coli icd under control of A. succinogenes
ppckA, flanked by FRT repeats and AscI restriction sites

This study

pCR2.1-Δfrd::icd

pCR2.1 derivative, A. succinogenes Δfrd::FRT-ppckA-icd-FRT

This study

pCR2.1-Δfrd

pCR2.1 derivative, frdAB deletion (2.1-kb fusion product of frdup
and frdCD)

This study

This study

Plasmids

145

Table 4.1 (cont’d)
pCR2.1-ΔpflB

pCR2.1 derivative, pflB deletion (2-kb fusion product of pflBup and This study
pflBdown)

pCR2.1-ΔpflB::icd

pCR2.1 derivative, A. succinogenes ΔpflB::FRT-ppckA-icd-FRT

This study

pLGZ924

pLGZ920 derivative, E. coli icd under control of A. succinogenes
ppckA promoter

This study

pCV933

pLGZ920 derivative, S. cerevisiae flp under control of A.
succinogenes ppckA
pCR2.1 derivative, lacZ deletion (fusion product of lacZup and
lacZdown)
pCR2.1 derivative, A. succinogenes ΔlacZ::FRT-ppckA-icd-FRT
pCR2.1 derivative, citDEF deletion (fusion product of citDEFup
and citDEFdown)
pCR2.1 derivative, A. succinogenes ΔcitDEF::FRT-ppckA-icd-FRT
pCR2.1 derivative, lacZ deletion (fusion product of acnup and
acndown)
pCR2.1 derivative, A. succinogenes Δacn::FRT-ppckA-icd-FRT

This study

pCR2.1-ΔlacZ
pCR2.1-ΔlacZ::icd
pCR2.1-Δcit
pCR2.1-Δcit::icd
pCR2.1-Δacn
pCR2.1-Δacn::icd

This study
This study
This study
This study
This study
This study

4.3.2 Plasmids, DNA manipulations, and electroporations.
Plasmids used in this study are listed in Table 4.1. PCR products were cloned into pCR2.1
using the TOPO-TA cloning kit (Invitrogen, Carlsbad, CA). pLGZ920 was used to express
foreign genes in A. succinogenes under control of ppckA (8). One Shot TOP10 chemically
competent E. coli cells (Invitrogen) were transformed as described by the manufacturer.
Electrocompetent A. succinogenes cells used for transformation of linear DNA were prepared as
described (28). Electrocompetent A. succinogenes cells used for transformation of circular DNA
were prepared using a method modified from (30) as follows. A 10-mL culture of actively
growing A. succinogenes (OD660 0.5 ± 0.2) in medium B or AM3 was incubated on ice for 10
min. Six mL of culture were harvested by centrifugation (3 min, 4,500 × g) at 4°C. The pellet
was washed three times with 1 mL of cold 272 mM sucrose and finally resuspended in 100 μL of
272 mM sucrose. Electroporation was performed with 1 μL DNA (100−300 ng for circular DNA,

146

700−800 ng for linear DNA) and 100 μL electrocompetent cells in a 2-mm gap width
electroporation cuvette (Bio-rad, Hercules, CA). After electroporation (settings: 25 μF, 400 Ω for
plasmids, 600 Ω for linear DNA, and 2.5 kV on a Bio-Rad GenePulser™), electroporation
mixtures were incubated with 0.5 mL of super optimal broth with catabolite repression (SOC)
outgrowth medium (New England Biolabs, Ipswich, MA) for 1 h at 37°C. The cells were then
centrifuged (3 min, 4,500 × g), resuspended in 100 μL supernatant, and spread on a single plate.
DNA manipulations used standard protocols (28). PCR reactions were performed with
the Advantage HD polymerase kit (Clontech, Mountain View, CA) unless otherwise stated.
Restriction enzymes were from New England Biolabs. Genomic DNA extractions were
performed using the Wizard genomic DNA purification kit, plasmid DNA was purified using the
Wizard SV miniprep kit, and DNA fragments were recovered from PCR mixtures and agarose
gels using the Wizard SV gel and PCR Clean-Up System (Wizard kits from Promega, Madison,
WI). Primers used in this work are listed in Table S4.1. Primers were synthesized by the
Michigan State University Research Technology Support Facility (MSU RTSF) or by Integrated
DNA Technologies (Coralville, IA). PCR and cloning accuracy were confirmed by DNA
sequencing performed by the MSU RTSF or by GENEWIZ, Inc. (South Plainfield, NJ). Colony
PCR was performed with Taq polymerase to confirm cloning steps.

4.3.3 Construction of plasmid pLGZ924.
The isocitrate dehydrogenase gene (icd) was amplified from E. coli K-12 genomic DNA
using primers P1 and P2. The PCR product was cloned into pCR2.1, and then subcloned into the
XbaI and SacI sites of pLGZ920, downstream of ppckA, yielding plasmid pLGZ924.

147

4.3.4 Construction of Δfrd::icd and ΔpflB::icd mutants.
The E. coli icd gene under control of ppckA in pLGZ924 was amplified with primers P3 and
P4. The primers added S. cerevisiae FRT sequences (5’GAAGTTCCTATTCCGAAGTTCCTATTCTCTAGAAAGTATAGGAACTTC-3’) (20) in the
same orientation plus AscI restriction sites on each side of the ppckA-icd cassette to yield a 1.6 kb,
AscI-FRT-ppckA-icd-FRT-AscI, cassette (Figure 4.1). After the PCR reaction, the 100-μL mixture
was heated at 95°C for 20 min. 3’ A-overhangs were added to the PCR product using Taq
polymerase as described in the TOPO-TA cloning kit, before cloning into pCR2.1 to yield
pCR2.1-icd.

A

ppckA

A

icd

FRT

FRT

AscI-FRT-ppckA-icd-FRT-AscI cassette

(A)

A

frdup

1

2

3

4

Δfrd in pCR2.1

frdCD
A

frdup

(B)
M

5.0 kb
3.8 kb

ppckA

A

icd

FRT

FRT

frdCD

Δfrd::icd in pCR2.1
and Δfrd::icd strain

2.1 kb

3.8 kb

frdA

frdup

frdCD

frdB

frdABCD
region in 130Z

5.0 kb

frdup

FRT

frdCD

Δfrd strain

2.1 kb

Figure 4.1 Construction of the icd selection cassette and construction of strain ∆frd. (A)
Physical maps of the intermediary constructs and strains involved. A: AscI; Black triangles:
primers P9 and P10 used to amplify the PCR products shown in (B). (B) PCR verification of
intermediate and final constructs using primers P9-P10. PCR product amplified from the 5.7-kb
Δfrd::icd PCR fragment used for natural transformation (lane 1), from A. succinogenes 130Z
(lane 2), from A. succinogenes ∆frdAB::icd (lane 3), and from A. succinogenes ∆frd (lane 4).
Lane M: DNA molecular markers.

A Δfrd DNA construct was assembled in two steps. First, the 1.1-kb region upstream of
frdABCD (frdup) was amplified with primers P5 and P6, and a 1-kb fragment containing the
frdCD genes was amplified with primers P7 and P8. Primers P6 and P7 overlapped by 19 nt,

148

including an AscI site. Second, the 1.1- and 1.0-kb purified fragments were fused by PCR using
nested primers P9 and P10. The 2.1-kb frdup-AscI-frdCD product (Figure 4.1) was cloned into
pCR2.1, yielding pCR2.1-Δfrd. The AscI-FRT-ppckA-icd-FRT-AscI fragment from pCR2.1-icd
was cloned into the AscI site of pCR2.1-Δfrd, yielding pCR2.1-Δfrd::icd (Figure 4.1). This
plasmid was used as a template with primers P11 and P12 to amplify a 5.4-kb fragment that was
naturally transformed into A. succinogenes 130Z to create strain Δfrd::icd.
A ∆pflB construct was assembled in pCR2.1 in two steps as well. Two 1.4-kb DNA
fragments upstream (pflBup) and downstream (pflBdown) of A. succinogenes pflB were amplified
separately using primer pairs P13-P14 and P15-P16, respectively. Primers P14 and P15
overlapped by 21 nt, including an AscI site. The pflBup and pflBdown purified PCR fragments were
then fused by PCR using nested primers P17 and P18. The 2.1-kb pflBup-AscI-pflBdown product
was cloned into pCR2.1, yielding plasmid pCR2.1-ΔpflB. Finally, the AscI-FRT-ppckA-icd-FRTAscI cassette from pCR2.1-icd was cloned into the AscI site of pCR2.1-ΔpflB, yielding pCR2.1ΔpflB::icd. After restriction digest with XmnI, the 7.6-kb linear plasmid DNA was naturally
transformed into A. succinogenes to create strain ΔpflB::icd.

4.3.5 Construction of the ∆pflB ΔlacZ::icd, ∆pflB Δcit::icd, and ∆pflB Δacn::icd double
mutants.
Strain ΔpflB was used as the host to delete lacZ, the genes encoding citrate lyase (citDEF),
and the aconitase gene (acn). The strategy used was identical to that used to build Δfrd and
ΔpflB. Construction of plasmids pCR2.1-ΔlacZ::icd, pCR2.1-Δcit::icd, and pCR2.1-Δacn::icd
used primers P35 to P40 (ΔlacZ), p41 to p46 (Δcit), and P47 to P52 (Δacn) (Table S4.1). Δcit
was flanked by 0.5 kb of its upstream and downstream regions in pCR2.1-Δcit::icd. Δacn was

149

flanked by 0.4 kb and 0.45 kb of its upstream and downstream regions, respectively, in pCR2.1Δacn::icd. ΔlacZ was flanked by 0.75 kb and 0.85 kb of its upstream and downstream regions,
respectively, in pCR2.1-ΔlacZ::icd. These plasmids were used as the templates with primer pairs
P11-P12 (add 1 kb of vector DNA on each side of the regions flanking the deletion) and P53-P54
(add 0.1 kb of vector DNA on each side of the regions flanking the deletion) to amplify the
fragments used to naturally transform strain ΔpflB. Double recombinations in strain ΔpflB were
screened by colony PCR using primer pairs P39-P40, P39-P26, and P25-P40 (ΔlacZ::icd); P45P46, P45-P25, and P26-P46 (Δcit::icd), and P51-P52, P51-P26, and P25-P52 (Δacn::icd). ΔpflB
ΔlacZ::icd double mutant candidates were further confirmed by patching the colonies on LB
glucose plates containing 40 μL 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-gal).

4.3.6 Natural transformation.
The natural transformation protocol for A. succinogenes closely resembles those for H.
influenzae (31) and A. pleuropneumoniae (32). To prepare competent cells, 400 μL of an
overnight culture of A. succinogenes in BHI were transferred to 35 mL fresh BHI and grown to
an OD600 of 0.2–0.25. Cells were washed twice with AM3 phosphate buffer (per liter:
NaH2PO4·H2O, 8.5 g; K2HPO4, 15.5 g), re-suspended in 10 mL of anaerobic MIV medium (31)
containing 2 mM cAMP, and incubated at 37°C for 100 min with shaking at 100 rpm to induce
competence. For natural transformation, a mixture of 1 μg DNA and 1 mL competent cells was
incubated in a 37°C water bath for 25 min. A negative control without DNA was included. Two
volumes of BHI were added to the transformation mixtures and incubated at 37°C with shaking
(250 rpm) for 100 min. Cells were then harvested (4,500 × g, 4°C, 15 min), washed twice with
0.75 mL AM3 phosphate buffer, and resuspended in 0.75 mL AM3 phosphate buffer. Finally,

150

0.25 mL of cell suspension was used to inoculate one tube each of AM3 (positive control), AM2
(negative control), and AM2-isocitrate (selective medium). Tubes were incubated at 37°C with
shaking until growth was observed in AM2-isocitrate. AM2-isocitrate cultures (0.25 mL) were
streaked onto LB-glucose-Km plates to isolate putative recombinants. A method for plate
selection is evaluated in the Results section.

4.3.7 Removal of the icd marker.
The S. cerevisiae flp gene was cloned into pLGZ920 for expression in A. succinogenes.
The flp gene was amplified from plasmid pCP20 using primers P19 and P20. The purified PCR
product was cloned into the XbaI and SacI sites of pLGZ920 using the In-Fusion cloning system
(Clontech). In the resulting plasmid, pCV933, flp is expressed constitutively under control of A.
succinogenes ppckA. pCV933 was purified and electroporated into A. succinogenes strains
Δfrd::icd, ΔpflB::icd, ∆pflB Δcit::icd, and ∆pflB Δacn::icd. Transformants were screened for
excision of the icd marker by colony PCR using several primer pairs. Excision of icd yielded
strains ΔpflB(pCV933), Δfrd(pCV933), ∆pflB Δcit(pCV933), and ∆pflB Δacn(pCV933), each of
which contains a single FRT scar in the chromosome. The deletions and the presence of the 48-nt
FRT sequence framed by two AscI sites were confirmed by sequencing these regions in the
genomes of strains Δfrd and ΔpflB.

4.3.8 Curing of plasmid pCV933.
An overnight culture of A. succinogenes Δfrd(pCV933) or ΔpflB(pCV933) in medium B
was used to inoculate a series of medium B tubes containing 0, 10, 50, 100, and 200 μg mL-1
acridine orange (AO). After 6.5 h of growth (37°C), cultures from each tube were streaked onto

151

LB-glucose-Km plates. The resulting colonies were patched onto LB-glucose-Amp plates to
screen for the loss of pCV933. Amp-sensitive (Amps) colonies were screened for plasmid curing
by colony PCR using primers P21 and P22, specific for sequences upstream and downstream of
the pLGZ920 multiple cloning site.

4.3.9 Enzyme assays.
Enzyme assays were performed with cell extracts prepared from exponential phase cultures
(OD660 between 0.6 and 1.0) grown in AM3. Cells from 10-mL cultures were harvested by
centrifugation (4,500 × g, 15 min, 4˚C) and washed three times with 50 mM Tris-HCl (pH 7.4) at
room temperature. For ICD and citrate lyase assays, pellets were resuspended in 0.5 mL of 50
mM Tris-HCl (pH 7.4). For fumarate reductase assays pellets were resuspended in 1 mL of 50
mM potassium phosphate (pH 7.0) containing 0.1 mM dithiothreitol. Cells were sonicated with a
Branson Sonifier 450 (Danbury, CT; 50% duty cycle, level 2, 20 sec, 4 times). The lysate was
centrifuged and the supernatant used for the assays as crude cell extract. Total proteins were
quantified using the Bio-Rad (Hercules, CA) Protein Assay Dye Reagent Concentrate with
bovine serum albumin as the standard. One unit of enzyme activity was defined as the amount of
enzyme needed to convert 1.0 μmol of substrate into product per min. ICD and citrate lyase
activities were assayed using a Cary 300 UV/vis spectrophotometer (Varian Instruments, Walnut
Creek, CA) equipped with a Peltier system. ICD activity was assayed by monitoring NADP+
reduction at 340 nm in a 1-mL reaction mixture containing 50 mM Tris-HCl (pH 7.4), 5 mM
DL-isocitric acid trisodium salt, 5 mM MgCl2, 1 mM dithiothreitol, 0.3 mM NADP+, and 90 mM
NaCl. The reaction was started by adding 50 μL crude cell extract. The molar extinction
coefficient of NADPH was 6,200 M-1cm-1. Fumarate reductase activity was assayed in 96-well

152

plates in a Sunrise microplate reader (Tecan, Durham, NC) in an anaerobic glove bag. The 200μL reaction contained 50 mM potassium phosphate (pH 7.0), 5 mM benzyl viologen, 0.4 mM
sodium dithionite, and 5 mM sodium fumarate. The reaction was started by adding 10 μL cell
extract. The activity was measured at 595 nm as the difference between the slopes before and
after adding the cell extract. Measurements were taken every 30 sec with shaking between
measurements. All enzyme assays were performed at 37˚C on three biological replicates.

4.3.10 Analysis of fermentation products.
Overnight cultures of strains 130Z and ΔpflB in AM3-25 mM NaHCO3 were inoculated
(0.25 mL) into 10 mL of fresh AM3-25 mM NaHCO3. Fermentation products in culture
supernatants were quantified by high-performance liquid chromatography (HPLC) (Waters,
Milford, MA) using a 300 × 7.8 mm Aminex HPX-87H column (Bio-Rad, Hercules, CA) at
30°C with 4 mM H2SO4 as the mobile phase (0.6 mL/min flow rate). Organic acids were
detected with a Waters 2478 UV detector at 210 nm or 254 nm.

4.4 Results
4.4.1 A positive selection method for recombination events in A. succinogenes.
To avoid using antibiotic resistance genes as selection markers, a positive selection method
was developed based on A. succinogenes’s metabolism. Because it is missing citrate synthase
and ICD in the citric acid cycle, A. succinogenes is auxotrophic for glutamate (29). A.
succinogenes has glutamate dehydrogenase activity and is able to grow on α-ketoglutarate
instead of glutamate (29). We proposed that A. succinogenes expressing ICD would be able to
grow in minimal medium with isocitrate substituted for glutamate.

153

A. succinogenes 130Z(pLGZ924) could grow in AM2-isocitrate, whereas 130Z could not
(data not shown). Strain 130Z(pLGZ924) grew slower in AM3 and AM2-isocitrate, (generation
times of 2.08 ± 0.03 h and 5.7 ± 0.1 h, respectively) than 130Z did in AM3 (generation time of
1.66 ± 0.01 h). AM3 contains 1.4 mM glutamate. Even with 5.7 mM L-isocitrate (4-fold more
than glutamate in AM3), 130Z(pLGZ924) grew in AM2-isocitrate to a maximum OD660 40%
lower than in AM3. In contrast to glutamate, which can only be a precursor of biomass in 130Z,
isocitrate can also be metabolized to oxaloacetate and acetyl-CoA through aconitase and citrate
lyase, and it seems to be partially diverted to acetate, explaining why 130Z(pLGZ924) grows
slower and to a lower final OD in AM2-isocitrate than in AM3. While the icd gene looks to be a
promising positive selection marker for A. succinogenes, its introduction in A. succinogenes
increases the complexity of the central metabolism, suggesting that the icd gene should be
excised from knockout mutant strains.

4.4.2 Natural transformation of A. succinogenes using a Δfrd::icd construct.
Two USSs were present in the Δfrd::icd construct, 650 nt upstream (in frdup) and 900 nt
downstream (in frdCD) of the FRT-ppckA-icd-FRT cassette, respectively (Figure 4.1). AscI
restriction sites were created on both ends of the FRT-ppckA-icd-FRT cassette and in the center of
the Δfrd construct for easy insertion of the cassette into the Δfrd construct. The A. succinogenes
chromosome contains only 33 AscI sites, allowing our selection cassette to be used for almost
any gene deletion in the genome. DNA starts being degraded by exonucleases as soon as it enters
the cell. To protect the A. succinogenes sequences (~1 kb on each side of the selection marker)
from degradation and maximize the chances of recombination, the PCR product used to construct
strain Δfrd::icd contained 1 kb of pCR2.1 DNA on each side of the knockout construct.

154

To determine whether and when cAMP is needed to induce A. succinogenes competence,
different growth and competence induction conditions were tested. cAMP was completely
omitted in the first experiment. In the second experiment, 2 mM cAMP was added to BHI only.
In the third experiment, 2 mM cAMP was added to MIV only. In each of these experiments the
transformation mixtures were finally incubated in AM3, AM2-isocitrate, and AM2 liquid media.
Only the third experiment led to growth in AM2-isocitrate after four days. Cultures grew
overnight in AM3 (positive control) but no growth was observed after four days in AM2
(negative control). The four-day AM2-isocitrate culture (0.25 mL) was used to inoculate 10 mL
of fresh AM2-isocitrate to enrich for recombinants able to grow on isocitrate. This second
culture was streaked onto LB-glucose-Km plates to isolate putative recombinants.

4.4.3 Confirmation of the Δfrd::icd knockout strain.
Double recombination of the Δfrd::icd construct in the A. succinogenes chromosome was
confirmed by colony PCR. Of six putative recombinant colonies screened by PCR with primers
P9 and P10, one yielded the expected 3.8-kb fragment (Figure 4.1). This putative mutant was
confirmed using colony PCR screens with other primers. PCR with primers P11 and P12
confirmed that this colony did not contain any pCR2.1 sequences, while PCRs with primer pairs
P23-P24 (internal to frdB) and P25-P26 (internal to E. coli icd) confirmed that this colony no
longer contained the frdB gene but that it contained the icd gene (not shown). This mutant strain
was called Δfrd::icd.
Construction of the Δfrd::icd mutant was confirmed with in vitro ICD and fumarate
reductase activity assays. Strain 130Z(pLGZ924) was the positive control for ICD activity. A.
succinogenes 130Z showed almost no detectable ICD activity (0.30 ± 0.49 mU mg-1). With an

155

activity level of 28 ± 7 mU mg-1 Δfrd::icd showed over 90-fold higher ICD activity than 130Z,
and 13-fold lower activity than 130Z(pLGZ924) (360 ± 130 U/mg). Higher ICD activity in
130Z(pLGZ924) reflects the fact that this strain contains multiple copies of icd. Strain Δfrd::icd
lacks the frdAB genes and should be devoid of fumarate reductase activity. Indeed, Δfrd::icd
showed over 500-fold lower fumarate reductase activity (0.0078 ± 0.0011 U mg-1) than 130Z
(3.97 ± 0.08 U mg-1).

4.4.4 Excision of the selection marker.
Strain Δfrd::icd was transformed with pCV933 (expressing the S. cerevisiae FLP
recombinase) and plated on LB-glucose-Amp. One transformant colony was re-isolated on LBglucose-Amp and excision of the icd marker was tested by colony PCR using primers P25-P26.
Out of the twenty potential Δfrd colonies tested, one yielded no PCR product. Colony PCR of
this strain using primers P9-P10 yielded a 2.1-kb product instead of the 3.8-kb product obtained
with Δfrd::icd, confirming that the icd gene was no longer present in that strain (Figure 4.1). The
new strain was called Δfrd(pCV933).

4.4.5 Curing pCV933 from A. succinogenes Δfrd(pCV933).
AO is commonly used to inhibit plasmid replication (33). After AO treatment, colonies
isolated on LB-glucose-Km were replica-plated onto LB-glucose-Km and LB-glucose-Amp
plates to identify Amps colonies. Twenty, ten, and ten Amps colonies from the Δfrd(pCV933)
cultures grown with 50 μg ml-1, 75 μg ml-1, and 200 μg ml-1 AO, respectively, were screened by
colony PCR with primers P21-P22. A single colony originating from the 50 μg ml-1 AO culture
was confirmed to have lost the plasmid (not shown), and was called strain Δfrd.

156

A

pCR2.1::ΔpflB::icd

pflBup

ppckA

FRT

A

pCR2.1::ΔpflB::icd-600

FRT

icd

FRT

ppckA

FRT

P11

A

icd

A

P29

pCR2.1::ΔpflB::icd-400

ppckA

2.9 kb

icd

FRT

A
FRT

P12

P32

P31
A

pCR2.1::ΔpflB::icd-200
P11

P12

P30
A

P11

pCR2.1

pflBdown

FRT

ppckA

2.5 kb

icd

P33

A
FRT

P34

P12

2.1 kb

Figure 4.2 Physical maps of pCR2.1-ΔpflB::icd and its truncated constructs
pCR2.1::ΔpflB::icd-600, pCR2.1::ΔpflB::icd-400, and pCR2.1::ΔpflB::icd-200. A: AscI; black
triangles: primers; vertical arrows: USS sequences.

4.4.6 Construction of strain ΔpflB.
The ΔpflB::icd construct contained a single USS in the frdBdown fragment, 33 nt
downstream of the FRT-ppckA-icd-FRT cassette (Figure 4.2, top construct). XmnI-linearized
pCR2.1-ΔpflB::icd, containing 2 kb of pCR2.1 DNA on each side of the knockout construct, was
used to construct strain ΔpflB::icd.. Construction of the A. succinogenes ΔpflB::icd mutant was
confirmed by colony PCR with primers P27-P28, which match sequences flanking the pflB
deletion. Thirty-five of the thirty-six colonies tested yielded the expected 2-kb PCR fragment
(Figure 4.3). PCR with primers P25-P26 confirmed that the putative mutants contained the E.
coli icd gene (Figure 4.3). One isolated mutant was called ΔpflB::icd. The icd marker was
excised by transforming strain ΔpflB::icd with pCV933. Of the thirty-eight colonies tested by
PCR with primers P25-P26, eighteen yielded no PCR product. Excision of the icd gene was
confirmed by colony PCR using primers P27-P28 (Figure 4.3), and one confirmed isolate was
called ΔpflB(pCV933).

157

M
M

11

2
2

3
3

44

3.0 kb 3.0
3.0

2.0
2.0 kb 2.0
1.5
1.5 kb 1.6
1.0 kb 1.0
1.0

A. succinogenes ΔpflB

A. succinogenes ΔpflB::icd

A. succinogenes 130Z

pCR2.1-ΔpflB::icd

A. succinogenes ΔpflB

A. succinogenes ΔpflB::icd

A. succinogenes 130Z

pCR2.1-ΔpflB::icd

Molecular weight markers

P25-P26
Primers
P25-P26

(A)

(A) (kb)
(kb)

Gel images showing positively
screened
colonies
in E.coli
:
Gel images
showing
positively
screened
colonies in E.coli :
ΔpflB::icd-200
ΔpflB::icd-400
ΔpflB::icd-600
ΔpflB::icd-200
ΔpflB::icd-400
ΔpflB::icd-600
LL 1 1 2 32 4 3
L
1
2
L 1 2
L 1 L2 31 42

P27-P28
Primers
P27-P28 (B) M 1
11 22 33 44
(kb)

M

2

M

3

3 .0
2 .0

0.5 kb 0.5
0.5

L – 1 Kb Ladder; 3 – Negative

Figure 4.3 Construction of strain ∆pflB (A) and natural transformation with the truncated
L – 1 Kb Ladder; 3 – Negative
products ΔpflB::icd-600, ΔpflB::icd-400, and ΔpflB::icd-200 (B). (A) Gel electrophoresis of
the PCR products amplified from the ∆pflB::icd cassette (lane 1), from A. succinogenes 130Z
(lane 2), from an A. succinogenes ∆pflB::icd derivative (lane 3), and from a ∆pflB(pCV933)
derivative (lane 4) using primers P25-P26 (internal to icd) and P27-P28 (framing the pflB
deletion). Lane M: DNA molecular markers. (B) Examples of colony PCRs after transformation
with the truncated constructs ΔpflB::icd-600 (lane 1, PCR with primers P29-P30), ΔpflB::icd-400
(lane 2, PCR with primers P31-P32), and ΔpflB::icd-200 (PCR with primers P33-P34). Lane M:
DNA molecular markers.

Strain ΔpflB(pCV933) was cured of pCV933 by AO treatment. Ten and fifty μg mL-1 AO
produced four and ten Amps colonies, respectively. The 100 μg mL-1 and 200 μg mL-1 AO
treatments produced no Amps colonies. Of the fourteen Amps colonies, one from the 10 μg mL-1
AO and two from the 50 μg mL-1 AO treatments were shown to have lost the plasmid after two
replicate colony PCR screens with primers P21 and P22 (not shown). One cured derivative from
the 50 μg mL-1 AO treatment was called strain ΔpflB.
During growth in AM3-25 mM NaHCO3, conditions that normally favor formate production
(6, 29), ΔpflB did not produce any formate (Figure 4.4), confirming that this strain is devoid of
pyruvate formate lyase activity.

158

unitsunits
Absorption
Absorption

0.08
0.07
0.08
0.06
0.07
0.05
0.06
0.04
0.05
0.03
0.04
0.02
0.03
0.01
0.02
0.00
0.01

Absorption units

4

2

4

Absorption units

Absorption units

B
B

2

26

28

26

28

B

0.7
0.07
Pyr
0.08
0.6 B
0.06
0.07
0.5
0.05
0.06
0.4
0.04
0.05
0.3
0.03
0.04
0.2
0.02
0.03
0.01
0.1
0.02
0.0
0.00
0.01
8
10
2
412
6
8 10
12
0.00

Absorption units

0.00

A

0.07
0.8
0.08
130Z
130Z
130Z
0.06 A
0.7
0.07
0.05
0.6
0.06
0.04
0.5
0.05
0.03
Pyr
Suc For Pyr
Suc For
0.04
0.4
0.02
0.03
0.3
SucAce
Pyr
Suc For Ace Pyr
For
0.01
0.02
0.2
Ace
Ace
0.00
0.01
0.1
6
8 10
12
18 12
20 1422 1624 1826 2028 22 24 26 28
2
4
614 816 10
0.00
0.0
Minutes
Minutes
10
124 14
146 16
16 8 18
20 12
22 14
2 12
16 28
18 20 22 24
6
88 10
18 1020
22
24 26
Time (min)
Minutes
Minutes
Pyr
Pyr ∆pflB
0.8
0.08

Absorption units

0.07
0.08
0.06
0.07
0.05
0.06
0.04
0.05
0.03
0.04
0.02
0.03
0.01
0.02
0.00
0.01

0.08

A
A

Absorption units

unitsunits
Absorption
Absorption

0.08

Pyr

Suc

Suc

Suc

Suc

Ace

Ace

Fum
Ace
Ace
Fum
Fum
14 8
16 10
18
20 14
22 16
6
12
18
2028 22 24 26 28
2
4
14
16
18
20
22
24
26
0.00
Time (min)
Minutes
Minutes
2 124 146 16 8 18 1020 12
16 28
18 20 22 24
2
4
6
8 10
22 14
24 26
Minutes
Figure 4.4 UV spectra of the HPLC profiles
of fermentation supernatants
of A.
Minutes

Fum

succinogenes 130Z, and ΔpflB grown on AM3 in the presence of 25 mM NaHCO3.
Supernatant samples were collected immediately after inoculation (baseline) and after 12 h
growth. Ace: acetate; For: formate; Fum: fumarate; Pyr: pyruvate.

4.4.7 Effects of the DNA construct length and of the length of the homologous regions on
the efficiency of homologous recombination.
The DNA fragments used in natural transformation so far contained approx. 1 kb of A.
succinogenes DNA on each side of the selection marker for efficient recombination plus at least
1 kb of non-homologous plasmid DNA on each side to protect the homologous DNA from
degradation by exonucleases. To test whether the non-homologous plasmid DNA on each side of

159

the recombination cassette could be omittted, the 3.8-kb Δfrd::icd cassette (amplified from
pCR2.1-Δfrd::icd with primers P9-P10, Figure 4.1) was used in natural transformation. In this
experiment, the selection marker was flanked by approx. 1 kb of A. succinogenes DNA on each
side. Transformants grew in AM2-isocitrate. Forty percent of recombinant colonies screened by
PCR with primers P9 and P10 yielded the expected 3.8-kb fragment (Table 4.2). This result
indicates that no more than a total of 1 kb of DNA is needed on each side of the marker for
recombination to take place.
To determine what minimum length of homologous DNA is needed on each side of the
selection marker to allow double recombination to take place, homologous recombination
cassettes were generated that contained 200 bp, 400 bp, and 600 bp homologous DNA on each
side of the selection marker. pCR2.1-ΔpflB::icd was used as the template because truncating the
frdup and frdCD fragments would have deleted both USS sequences (Figure 4.1). The ΔpflB::icd600, -400 and -200 fragments were amplified from pCR2.1-ΔpflB::icd with primers P29-P30,
P31-P32, and P33-P34, respectively, and cloned back into pCR2.1 (Figure 4.2). The fragments
used for transformation were generated by PCR with primers P11-P12, including 1 kb of vector
DNA on each side of the recombination cassette. Three independent natural transformations
were performed with each fragment, in which 15 colonies were screened by PCR. While the
standard deviations were large in all cases, averages of 47%, 57%, and 60% colonies contained
the ΔpflB::icd deletion for the transformations with fragments ΔpflB::icd-600, -400, and -200,
respectively. These results suggest that 200 bp of homologous DNA on each side of the selection
marker are still enough to allow double recombination. Conversely, all transformations using the
ΔpflB::icd-600 cassette amplified with primers P30-P31, the ΔpflB::icd-400 amplified with

160

primers P31-P32, or the ΔpflB::icd-200 cassette amplified with primers P33-P34 yielded no
growth in AM2-isocitrate and no ΔpflB::icd deletions.

4.4.8 Construction of double knockout mutants.
To determine whether the same knockout strategy could be used to introduce two successive
deletions in the same strain, three genes deemed non-essential for growth in AM3 glucose were
targeted for deletion in strain ΔpflB. The genes encoding citrate lyase and aconitase were chosen
as targets, because the tricarboxylic acid cycle is incomplete in A. succinogenes. The third target
gene was lacZ. The strategy used to build the three deletions was identical to that used to build
the Δfrd and ΔpflB strains. Citrate lyase is encoded by three genes organized in an operon,
citDEF (Asuc_1194-1196). CitE encodes the subunit with lyase activity. The citrate lyase
deletion, Δcit, encompassed over half of citD, the entire citE, and over a third of citF (not
shown). One USS was present in the upstream fragment. The 2.6-kb aconitase gene (acn,
Asuc_0185) is immediately preceded by the only USS in that area. To preserve that USS and to
preserve the promoter of the gene downstream of acn, the acn deletion encompassed only 1.6 kb
inside the acn gene. The lacZ deletion encompassed the entire 3.0-kb lacZ gene, including the
only four USSs present in the area. For this reason, a USS was introduced into the reverse nested
primer used to fuse the lacZ up and down regions (P40, Table S4.1).
The three double knockout mutant strains were obtained with frequencies ranging between
25% and 100% (Table 4.2). Note that double recombinants were obtained with frequencies of
25% or higher for Δcit::icd in strains ΔpflB and 130Z and for Δacn::icd in ΔpflB with only 0.52
kb to 0.6 kb of total flanking DNA on each side of the selection marker (Table 4.2).

161

Table 4.2 Frequencies of knockout mutations introduced by natural transformation in
A. succinogenes strains 130Z and ΔpflB
Homologous DNA

a

Vector DNA on each Knockout frequency

Transformation

Host

on each side of the

side of the

(positive

cassette

strain

selection cassette

recombination

colonies/colonies

(kb)

cassette (kb)

screened)

Knockout
frequency
(%)

Δfrd::icd

130Z

1

0.8

1/6

17

Δfrd::icd

130Z

1

0

6/15

40

Δfrd::icd

130Z

1

0.8

No growtha

–

ΔpflB::icd

130Z

1

1.9

35/36

97

ΔpflB::icd-600

130Z

0.6

1.0

6/16, 15/15, 6/15

60 ± 35

ΔpflB::icd-400

130Z

0.4

1.0

4/15, 7/15, 15/15

57 ± 38

ΔpflB::icd-200

130Z

0.2

1.0

3/15, 4/15, 14/15

47 ± 41

ΔpflB::icd-600

130Z

0.6

0

No growth

–

ΔpflB::icd-400

130Z

0.4

0

No growth

–

ΔpflB::icd-200

130Z

0.2

0

No growth

–

Δcit::icd

130Z

0.5

1.0

5/12

42

Δcit::icd

130Z

0.5

0.1

3/12

25

ΔlacZ::icd

ΔpflB

0.8

1.0

17/19

89

Δcit::icd

ΔpflB

0.5

1.0

30/30

100

Δcit::icd

ΔpflB

0.5

0.1

6/12

50

Δacn::icd

ΔpflB

0.42

0.1

4/16

25

Experiment repeated five times with the same results.

Figure 4.5 confirms that ΔpflB ΔlacZ::icd transformants no longer have β-galactosidase
activity. ΔpflB Δcit showed 27-fold less in vitro citrate lyase activity than ΔpflB (0.02 μmol min-1
mg-1 in ΔpflB Δcit vs. 0.55 μmol min-1 mg-1 in ΔpflB), confirming the double deletion.

162

ΔpflB

ΔpflB-ΔlacZ::icd

Figure 4.5 Phenotypic characterization of double knockout mutant ∆pflB-∆lacZ::icd. One
colony each of ∆pflB and one ∆pflB-∆lacZ::icd were patched on LB X-gal Amp plates.

4.4.9 Development of a selective solid medium.
A. succinogenes 130Z grows slowly on AM3 plates, with colonies visible only after 4 days.
Even though it contains multiple copies of the E. coli icd gene, 130Z(pLGZ924) did not form
visible colonies on AM2-isocitrate plates. To develop a selective agar-isocitrate medium that
allows strains containing a single copy of E. coli icd to form colonies, all amino acids but
glutamate, glutamine, arginine, and proline (i.e., amino acids derived from glutamate) were
added to the medium. Glutamate was replaced by 30 mM DL-isocitrate. 130Z(pLGZ924) formed
colonies on this medium after two days, but Δfrd::icd did not, even after six days. The same
medium containing additional vitamins, cysteine, and methionine (i.e., medium AM16-isocitrate)
supported growth of Δfrd::icd, with colonies forming after five days. AM16-isocitrate plates
were tested as selective solid medium in two replicate natural transformations of strain 130Z
with 1 μg NcoI-linearized pCR2.1-Δfrd::icd. Transformation mixtures were inoculated into
liquid AM3 and AM2-isocitrate (250 µL in each), and spread on AM16-isocitrate plates (225 µL
spread at 100, 10-1, and 10-2 dilutions). As expected, transformants grew overnight in AM3 and
after four days in liquid AM2-isocitrate. Colonies grew on AM16-isocitrate plates after five
days, with an efficiency of approximately 140 colonies/μg DNA. Eighteen colonies grown on
AM16-isocitrate plates were tested by colony PCR using primers P9-P2. All were Δfrd::icd

163

mutants, confirming that AM16-isocitrate plates can be used to directly select icd-containing
knockout mutant constructs in A. succinogenes.
All attempts to measure transformation and double recombination efficiencies failed
because of a yet unexplained problem with serial dilutions and plating. Starting from fresh dense
precultures, all A. succinogenes strains tested, including 130Z, grew as lawns when plated from
the undiluted preculture, they grew as numerous compact colonies when plated from the 10-1
dilution, but they produced only a few colonies when plated from the 10-2 dilution and no
colonies at higher dilutions. Similar results were observed when the serial dilutions were
prepared with AM3 phosphate buffer, AM3 with and without glucose, LB, or with the
supernatant of the preculture, and when the bacterial suspensions were plated on LB-glucose,
AM3, AM17, or AM16-isocitrate.

4.4.10 Knockout mutants via electroporation.
The knockout mutants obtained using natural transformation showed that linear DNA
fragments could recombine with the A. succinogenes chromosome. Despite previous failed
attempts at constructing knockout mutants using electroporation and antibiotic resistance genes
as selection markers, electroporation was tested again as an alternative transformation method to
create knockout mutants in A. succinogenes.
Electroporations of strain 130Z with NcoI-linearized pCR2.1-Δfrd::icd were performed using
0.1‒1 µg DNA in 100 ng increments. After recovery, electroporation mixtures were washed once
with MIV and resuspended in MIV. One third of each electroporation mixture was inoculated in
liquid AM3, one third was inoculated in liquid AM2-isocitrate, and one third was spread on a
single AM16-isocitrate plate. Electroporations with 300 ng or more linearized pCR2.1-Δfrd::icd

164

consistently yielded colonies on AM16-isocitrate plates. All colonies growing on AM16isocitrate plates that were tested by colony PCR with primers P9-P2 were confirmed as Δfrd::icd
knockouts, showing that electroporation can be used as an alternative to natural transformation to
introduce linear DNA in A. succinogenes for homologous recombination.

4.5 Discussion
The fermentative metabolism of A. succinogenes has been studied in detail using
fermentation balances, in vitro enzyme activity assays, genome sequencing, and metabolic flux
analysis, and how ATP is produced and NAD+ is regenerated in the different fermentative
branches is well understood (6, 17, 21, 22, 29). To enhance the production of a given
fermentative metabolite (e.g., fumarate or succinate), we needed a means to produce knockout
mutants. Here, we demonstrated that natural transformation and electroporation can be used to
introduce DNA in A. succinogenes for recombination, that the E. coli icd gene can be used as a
positive selection marker for recombination events, that the yeast Flp/FRT recombination system
can be used to excise the selection marker, that plasmids can be cured from A. succinogenes
using AO, and that the selection marker can be reused to introduce at least two consecutive
deletions into the same strain. The 64-nt scar (FRT sequence flanked by AscI sites) interrupting
the deletion does not by itself leave the remaining sequences in frame, but if an in-frame deletion
is desired, the deletion itself can be designed with a frameshift to place the remaining sequences
back in frame.
Except for our first transformation to construct the Δfrd::icd strain, at least 25% of colonies
obtained after all other natural transformations were knockout mutants (Table 4.2). The
frequency of knockout colonies increased to 100% when transformants (obtained by

165

electroporation) were selected directly on the enriched defined medium, AM16-isocitrate. Note
that we could not repeat the natural transformation with the Δfrd::icd cassette in at least five
more attempts, suggesting that the efficiency of the method is somewhat gene-specific. The
sequences of the two USS repeats present in the Δfrd::icd cassette were closer to the consensus
USS for A. succinogenes than the single USS in the ΔpflB::icd cassette was (not shown).
Therefore poor recognition of the USSs flanking the Δfrd::icd cassette by the DNA uptake
machinery cannot explain these results. We have also tried repeatedly to knockout other genes
(i.e., ackA, pta, pykA, and zwf) in 130Z and ΔpflB without success, but these genes might be
essential for growth, or we have not yet found conditions in which a knockout strain can grow.
With the ΔpflB::icd, ΔlacZ::icd, Δcit::icd, and Δacn::icd constructs, we demonstrated that a
single USS in the knockout construct is enough to allow DNA uptake by A. succinogenes.
Introducing a USS in the ΔlacZ::icd construct―which would otherwise not have contained any
USS, also allowed double recombination to take place. Linear constructs of varying lengths
could be used for transformation, from the entire plasmid carrying the knockout construct (e.g.,
ΔpflB::icd in 130Z) down to recombination cassettes containing 0.52 kb to 0.6 kb of DNA on
each side of the selection marker (e.g., Δacn::icd in ΔpflB) (Table 4.2). While flanking regions of
0.5 kb to 0.6 kb on each side of the cassette yielded knockouts with frequencies as low as 25%
(or even 0% for ΔpflB::icd-600), 1 kb of flanking DNA on each side was enough to protect
against exonucleases prior to recombination, routinely yielding knockout frequencies between
42% and 100% (Table 4.2). As little as 200 bp of homologous DNA on each side of the selection
marker was enough for efficient double recombination, as long as the homologous DNA was
itself flanked by non-homologous DNA on each side.

166

The development of a markerless knockout method for A. succinogenes will greatly
facilitate future genetic engineering of A. succinogenes. Strain ΔpflB represents an important first
step in engineering a strain that can produce succinate at near maximum theoretical yields.

4.6 Acknowledgements
This work was supported by a grant from the Michigan Economic Development
Corporation, by MSU startup funds, and by grant # 2010-04061 from the USDA National
Institute for Food and Agriculture’s Sustainable Bioenergy Research Program to CV. BS was
supported in part by a research fellowship from the MSU Quantitative Biology Initiative. NM
was supported in part by a recruiting fellowship from the MSU Plant Research Laboratory. KT
was supported by a Khorana scholarship. We thank Reena Jain, Jean Kim, and Maeva Bottex for
their technical assistance, as well as Dr. Michael Bagdasarian for his critical reading of the
manuscript.

167

APPENDIX

168

Table A4.1 Oligonucleotide primers used in this study
Primer
P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19
P20
P21
P22
P23
P24
P25
P26
P27
P28
P29
P30
P31
P32
P33
P34
P35
P36
P37
P38
P39

Sequencea (restriction site)
GTTCTAGAGATGGAAAGTAAAGTAGTTG (XbaI)
CTGAATTCATTACATGTTTTCGATGATC (EcoRI)
GAGGCGCGCCGAAGTTCCTATTCCGAAGTTCCTATTCTCTAGAA
AGTATAGGAACTTCTCGATAAATTGAAAATGCAGCAA (AscI)
CTGGCGCGCCGAAGTTCCTATACTTTCTAGAGAATAGGAACTTC
GGAATAGGAACTTCTTACATGTTTTCGATGATCGC (AscI)
TTAGGTACCGGCAACAAAGG
CCCCTATGCTTGGCGCGCCCGTAAACCTAAACGCGCAAT (AscI)
GGCGCGCCAAGCATAGGGGGAAAGCAAT (AscI)
GTCCGATTTGGGTTTTGCTA
GTAAGCTTACGGCAAACACGATCACATA
CACTCGAGGCACCGCCTGTCACTAAAAT
CCGGATCAAGAGCTACCAAC
CGAAACGATCCTCATCCTGT
CGTTAACCGTGGGAATCAGT
TTACGTTACCCCAGGCGCGCCCTTCCTTTTGCTAGTATTGATAAT
GA (AscI)
GGCGCGCCTGGGGTAACGTAATAAAAATGTAATG (AscI)
TCTCTCCTTCGCGGAATAAA
TGAGCCTGACTGGTAAATCCA
CACATCGACCCCGATAACTT
ATGAGGTGATCTAGATGCCACAATTTGGTATATTATGTAAA
(XbaI)
CGGCCAGTGAATTCGAGCTCTTATATGCGTCTATTTATGTAGG
(SacI)
CGTTGTAAAACGACGGCC
AATTTTAAATTATCAATGAGGTG
TGCGTTACAACCCTGAAACA
TCTTTCGCACTTTCCAGCTT
GAAGGTGATGGAATCGGTGT
GTTACGGTTTTCGCGTTGAT
AATTTGCGGATCTGGATGTC
TGGTGTATTCCGTCAGTTCG
GTTGCGAACTGTTCACCTC
CAATATCGCGGCGTTCCAGC
GCATTGGGCATTTTGTGTAAC
CTAGTAGAAATGCGCGTTTTA
GAGTTTTGGCAGGCAATC
CTTTCCCAGCGTTCCACAAAAAC
GGATAAACTCGGCGTACG
GCGGTTATGGGCGCGCCGCAGAATCATGCTGAACTCC (AscI)
GGCGCGCCCATAACCGCATAAAAATACAGGGCAG (AscI)
CATGAAACCGAACAGTATTGTGC
CTTGCCAAACCGACCGAAAG

169

Specificityb (directionc)
E. coli K-12 icd gene (F)
E. coli K-12 icd gene (R)
AscI-FRT- ppckA (F)
E. coli icd-FRT-AscI (R)
frdup (F)
frdup (R)
frdCD (F)
frdCD (R)
Δfrd-fusion, nested (F)
Δfrd-fusion, nested (R)
pCR2.1 backbone (F)
pCR2.1 backbone (R)
pflBup (F)
pflBup (R)
pflBdown (F)
pflBdown (R)
ΔpflB-fusion, nested (F)
ΔpflB-fusion, nested (R)
S. cerevisiae flp (F)
S. cerevisiae flp (R)
pLGZ920 (F)
pLGZ920 (R)
frdB (F)
frdB (R)
Inside E. coli K-12 icd (F)
Inside E. coli K-12 icd (R)
217 nt upstream of pflB (F)
146 nt downstream of pflB (R)
ΔpflB::icd-600 (F)
ΔpflB::icd-600 (R)
ΔpflB::icd-400 (F)
ΔpflB::icd-400 (R)
ΔpflB::icd-200 (F)
ΔpflB::icd-200 (R)
lacZup (F)
lacZup (R)
lacZdown (F)
lacZdown (R)
ΔlacZ-fusion, nested (F)

Table A4.1 (cont’d)
TATTGATAATGAAAATCCGACCGCACTTGGCAGTACCGGCGTAT ΔlacZ-fusion, nested (R)
TCCTCd
P41
CTTGACGGTAGCAGTGATC
citDEFup (F)
P42
GGCGCGCCGAATTGCTTTCCGAC (AscI)
citDEFup (R)
P43
GAAAGCAATTCGGCGCGCCAATCAGTATCGGCCAGG (AscI)
citDEFdown (F)
P44
CCGTATCGCAGTGACCGC
citDEFdown (R)
P45
CTGGTGGTGACGATCATAC
ΔcitDEF-fusion, nested (F)
P46
GGTATCGATTTCCAATGCGG
ΔcitDEF-fusion, nested (R)
P47
CGGCGAAGTGTTCTACGATG
acnup (F)
P48
GGCGCGCCGTGGTATCCACCTCAAC (AscI)
acnup (R)
P49
GGATACCACGGCGCGCCATATGCCACAGCACGG (AscI)
acndown (F)
P50
CATGCGGTAGAATTACTCGG
acndown (R)
P51
ACGATTATCCGAGTCATGAG
Δacn-fusion, nested (F)
P52
GCCAGATTCGCTCCGCTTG
Δacn-fusion, nested (R)
P53
TGTAAAACGACGGCCAGT
M13 forward
P54
CAGGAAACAGCTATGACC
M13 reverse
a
5’ to 3’ direction; restriction sites are underlined and named at the end of the primer sequences
b
If unspecified, gene names refer to A. succinogenes
c
F, forward primer; R, reverse primer
d
Primer P41 contains the USS ACCGCACTT
P40

170

kb

M

1

2

3

4

M

5

6

M

7

8

3.0
2.0
1.6
1.0
0.5

Figure A4.1 Verification of the construction of strains ∆pflB ∆cit and ∆pflB ∆acn by PCR.
Lanes M: DNA molecular markers. PCR products amplified from 130Z genomic DNA (Lane 1),
∆pflB genomic DNA (Lanes 2, 5, and 7), a ∆pflB ∆acn colony (Lanes 3 and 6), and a ∆pflB ∆cit
colony (Lanes 4 and 8). Lanes 1 to 4: verification that the host for the double mutations is strain
pflB; primers P27-P28 (flanking pflB) amplify a 2.7-kb fragment in 130Z, and a 434-bp fragment
in ∆pflB, ∆pflB ∆acn, and ∆pflB ∆cit). Lanes 5 and 6: verification that acn is deleted in strain
∆pflB ∆acn; primers P51-P52 amplify a 2.4-kb fragment in ∆pflB and a 855-bp fragment in
∆pflB ∆acn. Lanes 7 and 8: verification that citDEF are deleted in strain ∆pflB ∆cit; primers P55P56 amplify a 2.2-kb fragment in ∆pflB and a 524-bp fragment in ∆pflB ∆cit. The PCR products
in lanes 6 and 8 would be 2.35 kb (∆acn) and 2.0 kb (∆cit), if the deletions still contained the icd
marker.

171

REFERENCES

172

REFERENCES

1.

Werpy TA, Petersen G. 2004. Top value added chemicals from biomass for the U.S.
Department of Energy (DOE) by National Renewable Energy Laboratory (NREL).

2.

Bozell JJ, Petersen G. 2010. Technology development for the production of biobased
products from biorefinery carbohydrates—the US Department of Energy’s Top 10
revisited. Green Chemistry 12:539–554.

3.

Guettler MV, Jain MK, Rumler D. 1996. Method for making succinic acid, bacterial
variants for use in the process, and methods for obtaining variants. U.S. patent 5,573,931.

4.

Guettler MV, Jain MK, Soni BK. 1996. Process for making succinic acid,
microorganisms for use in the process and methods of obtaining the microorganisms.
U.S. patent 5,504,004.

5.

Guettler MV, Rumler D, Jain MK. 1999. Actinobacillus succinogenes sp. nov., a novel
succinic-acid-producing strain from the bovine rumen. Int J Syst Bacteriol 49:207–216.

6.

van der Werf MJ, Guettler MV, Jain MK, Zeikus JG. 1997. Environmental and
physiological factors affecting the succinate product ratio during carbohydrate
fermentation by Actinobacillus sp. 130Z. Arch Microbiol 167:332–342.

7.

Kurzrock T, Weuster-Botz D. 2010. Recovery of succinic acid from fermentation broth.
Biotechnol Lett 32:331–339.

8.

Kim P, Laivenieks M, McKinlay J, Vieille C, Zeikus JG. 2004. Construction of a
shuttle vector for the overexpression of recombinant proteins in Actinobacillus
succinogenes. Plasmid 51:108–115.

9.

Oswald W, Tonpitak W, Ohrt G, Gerlach G-F. 1999. A single-step transconjugation
system for the introduction of unmarked deletions into Actinobacillus pleuropneumoniae
serotype 7 using a sucrose sensitivity marker. FEMS Microbiol Lett 179:153–160.

10.

Tatum FM, Briggs RE. 2005. Construction of in-frame aroA deletion mutants of
Mannheimia haemolytica, Pasteurella multocida, and Haemophilus somnus by using a
new temperature-sensitive plasmid. Appl Environ Microbiol 71:7196–7202.

11.

Kim JM, Lee KH, Lee SY. 2008. Development of a markerless gene knock-out system
for Mannheimia succiniciproducens using a temperature-sensitive plasmid. FEMS
Microbiol Lett 278:78–85.

12.

Maughan H, Sinha S, Wilson L, Redfield R. 2008. Competence, DNA uptake, and
transformation in Pasteurellaceae, p 79–98. In Kuhnert P, Christensen H (ed),
Pasteurellaceae Biology, Genomics and Molecular Aspects. Caister Academic Press.

173

13.

Redfield RJ, Findlay WA, Bossé J, Kroll JS, Cameron AD, Nash JH. 2006. Evolution
of competence and DNA uptake specificity in the Pasteurellaceae. BMC Evol Biol 6:82.

14.

Sisco KL, Smith HO. 1979. Sequence-specific DNA uptake in Haemophilus
transformation. Proc Natl Acad Sci USA 76:972–976.

15.

Danner DB, Deich RA, Sisco KL, Smith HO. 1980. An 11-base pair sequence
determines the specificity of DNA uptake in Haemophilus transformation. Gene 11(3–
4):311–318.

16.

Wang Y, Orvis J, Dyer D, Chen C. 2006. Genomic distribution and functions of uptake
signal sequences in Actinobacillus actinomycetemcomitans. Microbiology 152:3319–
3325.

17.

McKinlay JB, Laivenieks M, Schindler BD, McKinlay AA, Siddaramappa S,
Challacombe JF, Lowry SR, Clum A, Lapidus AL, Burkhart KB, Harkins V, Vieille
C. 2010. A genomic perspective on the potential of Actinobacillus succinogenes for
industrial succinate production. BMC Genomics 11:680.

18.

Redfield RJ, Cameron AD, Qian Q, Hinds J, Ali TR, Kroll JS, Langford PR. 2005.
A novel CRP-dependent regulon controls expression of competence genes in
Haemophilus influenzae. J Mol Biol 347:735–747.

19.

Cox MM. 1983. The FLP protein of the yeast 2-microns plasmid: expression of a
eukaryotic genetic recombination system in Escherichia coli. Proc Natl Acad Sci USA
80:4223–4227.

20.

Zhu XD, Pan G, Luetke K, Sadowski PD. 1995. Cleavage-dependent ligation by the
FLP recombinase. Characterization of a mutant FLP protein with an alteration in a
catalytic amino acid. J Biol Biol 270:11646–11653.

21.

McKinlay JB, Shachar-Hill Y, Zeikus JG, Vieille C. 2007. Determining Actinobacillus
succinogenes metabolic pathways and fluxes by NMR and GC-MS analyses of 13Clabeled metabolic product isotopomers. Metab Engin 9:177–192.

22.

McKinlay JB, Vieille C. 2008. 13C-metabolic flux analysis of Actinobacillus
succinogenes fermentative metabolism at different NaHCO3 and H2 concentrations.
Metab Engin 10:55–68.

23.

Sawers RG, Clark DP. 2004. Chapter 3.5.3, Fermentative pyruvate and acetylcoenzyme A metabolism. In Böck A, Curtiss III R, Kaper JB, Karp PD, Neidhardt FC,
Nyström T, Slauch JM, Squires CL, Ussery D (ed), EcoSal—Escherichia coli and
Salmonella: Cellular and Molecular Biology. ASM Press, Washington, DC.

174

24.

Stols L, Donnelly MI. 1997. Production of succinic acid through overexpression of
NAD+-dependent malic enzyme in an Escherichia coli mutant. Appl Environ Microbiol
63:2695–2701.

25.

Jantama K, Haupt MJ, Svoronos SA, Zhang XL, Moore JC, Shanmugam KT,
Ingram LO. 2008. Combining metabolic engineering and metabolic evolution to develop
nonrecombinant strains of Escherichia coli C that produce succinate and malate.
Biotechnol Bioengin 99:1140–1153.

26.

Zhang X, Jantama K, Shanmugam KT, Ingram LO. 2009. Reengineering Escherichia
coli for succinate production in mineral salts medium. Appl Environ Microbiol 75:7807–
7813.

27.

Lee SJ, Song H, Lee SY. 2006. Genome-based metabolic engineering of Mannheimia
succiniciproducens for succinic acid production. Appl Environ Microbiol 72:1939–1948.

28.

Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K
(ed). 1993. Current Protocols in Molecular Biology. Greene Publishing & WileyInterscience, New York.

29.

McKinlay JB, Zeikus JG, Vieille C. 2005. Insights into the Actinobacillus succinogenes
fermentative metabolism in a chemically defined growth medium. Appl Environ
Microbiol 71:6651–6656.

30.

Choi KH, Kumar A, Schweizer HP. 2006. A 10-min method for preparation of highly
electrocompetent Pseudomonas aeruginosa cells: application for DNA fragment transfer
between chromosomes and plasmid transformation. J Microbiol Methods 64:391–397.

31.

Poje G, Redfield RJ. 2003. Transformation of Haemophilus influenzae. Methods Mol
Med 71:57–70.

32.

Bosse JT, Sunita S, Schippers T, Kroll JS, Redfield RJ, Langford PR. 2009. Natural
competence in strains of Actinobacillus pleuropneumoniae. FEMS Microbiol Lett
298:124–130.

33.

Trevors JT. 1986. Plasmid curing in bacteria. FEMS Microbiol Rev 32:149–157.

175

Chapter 5

Respiratory glycerol metabolism of Actinobacillus succinogenes 130Z for succinate
production

176

The work completed in this chapter is part of a collaboration with Dr. Bryan Schindler (primary
author, a former member of the Vieille lab) and Dr. Claire Vieille, published as:

Schindler BD, Joshi RJ, Vieille C. 2014. Respiratory glycerol metabolism of Actinobacillus
succinogenes 130Z for succinate production. J Ind Microbiol Biotechnol 41:1339-1352. doi:
10.1007/s10295-014-1480-x

Article can be viewed at the following URL:
http://rdcu.be/v6Wy

My contribution to this report included:


Developing and optimizing a microaerobic continuous culture system for A. succinogenes
(chemostats) for growth on glycerol.



Conducting continuous culture experiments with A. succinogenes 130Z and carrying out
fermentation balances.



Drafting portions of the manuscript for above bullet points

Reading and reviewing entire manuscript.

177

Chapter 6

Conclusions and future directions

178

6.1 Introduction
A. succinogenes is one of the best natural succinate producers known. Many advances have
been carried out with respect to its metabolism on substrates like glucose, glycerol, and
lignocellulosic hydrolysates (1-4). Microorganisms such as E. coli, S. cerevisiae, and C.
glutamicum are being studied and engineered for succinate production as well. Many of the
succinate production increases obtained in E. coli were obtained by mimicking A. succinogenes
metabolism. Despite much insight into A. succinogenes’s metabolism, we still know very little
about other aspects, such as succinate efflux transporters or small RNAs (sRNAs) in A.
succinogenes. In this dissertation, we have focused on identifying succinate exporters, and
overexpressing four individual transporters increased succinate production. Additionally, we
have also identified sRNAs in A. succinogenes and shown that synthetic sRNAs can be used in
A. succinogenes for inhibiting expression of unwanted pathways. The knowledge gained about
succinate transporters and sRNAs in this dissertation can be used for future metabolic
engineering of A. succinogenes for succinate production.

6.2 Succinate transporters in A. succinogenes
We have identified succinate transporters in A. succinogenes using a combination of
proteomics and RNA sequencing (RNAseq). The transporters with the most hits in our
proteomics studies and high transcripts levels were different from the succinate exporters
identified in E. coli and C. glutamicum. Although homologs of the E. coli and C. glutamicum
succinate exporters exist in A. succinogenes, none of them were among the top hits in our
proteomics and transcriptomics studies. Asuc_1999, Asuc_0142, Asuc_2058, and Asuc_1990-91
were the transporter candidates that topped our proteomics list. Since expressing the transporters

179

under the control of a strong promoter was toxic for A. succinogenes, we developed a promoter
library with a large span of expression strength. Expressing Asuc_1999 under the control of
weaker promoter ppckA-103 increased the succinate yield in A. succinogenes by 24%, mainly at the
cost of biomass production.

6.3 Identification of sRNAs in A. succinogenes
The purpose of this study was to gain some insight into the size, structure, and function of
sRNAs in A. succinogenes, especially of Hfq-dependent sRNAs. We identified 260 sRNAs in A.
succinogenes in glucose- and glycerol-grown cultures, out of which 24 had predicted Rhoindependent terminators, a characteristic feature of Hfq-dependent sRNAs. We were also able to
validate 14 randomly selected sRNAs by RT-PCR. One of the major goals of this study was to
use the knowledge gained from the RNAseq studies of sRNAs and design a synthetic sRNA to
inhibit the translation of select mRNAs in A. succinogenes. We were most interested in this
approach since knockout mutants maybe impossible to construct for certain essential genes, and
this approach offers a much more tunable way to lower gene expression. In this study, we
successfully designed synthetic sRNAs targeting β-galactosidase and acetate kinase expression.
We were able to decrease β-galactosidase activity by 32% and to decrease the acetate yield by
14%.

180

6.4 Future directions
6.4.1 Experimental identification of Hfq-binding sRNAs
In this study we used computational and manual approaches to identify possible Hfqbinding sRNAs. We still need to experimentally verify these Hfq-binding sRNAs, though. One
approach to identifying Hfq-binding sRNAs as well as their target mRNAs would be to compare
transcriptomics data between a Δhfq strain and the wild-type strain. I built a Δhfq::CmR strain,
but that strain was only transiently culturable, not long enough to allow transcriptomic analysis.
Another approach is copurification using a strain expressing His-tagged Hfq, which would allow
us to isolate and sequence Hfq-binding sRNAs. This approach would allow us to confirm a
number of the predicted Hfq-binding sRNAs in our RNAseq studies, as well as identify some
mRNAs regulated by Hfq-dependent sRNA. Because we could not complement our Δhfq::CmR
strain with a plasmid-borne His-tagged hfq gene, we are copurifying His6-tagged Hfq with its
binding RNAs after expression in wild-type A. succinogenes instead. Results from this
experiment will yield additional scaffold sequences from Hfq-binding sRNAs.

6.4.2 Continuing the development of synthetic sRNAs in A. succinogenes
Once we confirm Hfq-binding sRNAs experimentally, it will be possible to compare many
scaffolds sequences and build additional synthetic sRNAs. Additionally, we need more studies
with synthetic sRNAs designed with differing scaffold and targeting sequence lengths to
modulate the binding with and regulation of the target mRNA. A more in depth look at the
secondary structures of the native Hfq-binding sRNAs is required. Mimicking the secondary
structure of native Hfq-binding sRNAs in our synthetic sRNAs may increase stability and

181

improve effectiveness of the designed sRNAs. All these studies will help in building a library of
synthetic sRNA scaffolds, where simply replacing the target binding regions would be possible.

6.4.3 Identifying the major succinate transporter(s) in A. succinogenes
In our proteomics and RNAseq studies we were able to identify a few transporter
candidates that may be responsible for exporting succinate out of the cell. However, we were
unable to identify one major succinate transporter that functions under anaerobic conditions. As
seen in E. coli and C. glutamicum, several succinate exporters may contribute to succinate efflux
under anaerobic conditions. Our attempts to knockout any of the four succinate exporters
identified in this study have been unsuccessful so far. But once we have these knockout mutants,
we will be able to test the knockout strains under anaerobic and microaerobic conditions to shed
more light on succinate exporters active under different conditions in A. succinogenes.

6.4.4 Construction of a ∆ackA knockout mutant
All our attempts at making a ∆pflB∆ackA mutant have been unsuccessful so far.
Recently, the Bechkam group was able to make a ∆pflB∆ackA mutant strain using homologous
recombination (5). The difference in approach was the use of a different selection marker and
growth medium. The authors used an antibiotic resistance gene as their selection marker and
tryptic soy broth supplemented with glucose and the antibiotic as the selection medium. In the
same study, the authors overexpressed phosphoenolpyruvate carboxykinase, malate
dehydrogenase, and fumarase independently and in the same strain. The succinate yields and
titers of strains expressing malate dehydrogenase and all the enzymes together were higher than

182

that of the wild-type strain. However, the same constructs in the ∆pflB∆ackA background did not
cause any increase in succinate titer or yield.
Our lack of success in constructing a ∆ackA strain suggests that the acetate pathway is
essential for A. succinogenes’s growth on a minimal medium such as AM3. We have previously
constructed ∆lacZ::CmR and Δhfq::CmR strains by selecting colonies on LB medium
supplemented with chloramphenicol, so we should be able to obtain a ∆ackA mutant using the
same approach.

6.4.5 Combination of different approaches to increase succinate production
Evolved strains of A. succinogenes that produce more succinate than the wild-type strain on
glucose and xylose have been isolated in our lab. The X strains, evolved for fast growth on
xylose, have mutations in the xylE and fusA genes, encoding the xylose permease and elongation
factor G, respectively. Introducing the xylE mutation in the wild-type strain increased the
succinate yield by 40%. We have a ΔpflB mutant (devoid of pyruvate formate lyase) that no
longer produces formate and produces 69% more succinate than the wild-type strain on glucose
(6). We also have identified succinate exporters that, upon overexpression, increase succinate
production as well. And we now have a library of promoters of a range of strengths,
Combining different mutations, expressing succinate exporter(s) as well as possibly some
of the enzymes in the succinate pathway should allow us to engineer strains that produce record
amounts of succinate with minimum byproducts.

183

REFERENCES

184

REFERENCES

1.

McKinlay JB, Shachar-Hill Y, Zeikus JG, Vieille C. 2007. Determining Actinobacillus
succinogenes metabolic pathways and fluxes by NMR and GC-MS analyses of 13Clabeled metabolic product isotopomers. Metab Eng 9:177–192.

2.

McKinlay JB, Vieille C. 2008. 13C-metabolic flux analysis of Actinobacillus
succinogenes fermentative metabolism at different NaHCO3 and H2 concentrations.
Metab Eng 10:55–68.

3.

Schindler BD, Joshi RJ, Vieille C. 2014. Respiratory glycerol metabolism of
Actinobacillus succinogenes 130Z for succinate production. J Ind Microbiol Biotechnol
41:1339-1352.

4.

McPherson N. 2017. Novel Insights Into Sugar and Succinate Metabolism of
Actinobacillus succinogenes from Strains Evolved for Improved Growth on
Lignocellulose Hydrolysate Sugars. Doctoral. Michigan State University.

5.

Guarnieri MT, Chou YC, Salvachua D, Mohagheghi A, St John PC, Peterson DJ,
Bomble YJ, Beckham GT. 2017. Metabolic Engineering of Actinobacillus succinogenes
Provides Insights into Succinic Acid Biosynthesis. Appl Environ Microbiol 83: e0099617.

6.

Schindler B. 2011. Understanding and improving respiratory succinate production from
glycerol by Actinobacillus succinogenes. Doctoral Dissertation. Michigan State
University.

185