Copyright © 2014, American Society for Microbiology. All ...aem.asm.org/content/early/2014/06/03/AEM.01188-14.full.pdf47 Harmful algal blooms ... From the genomic DNA, ... The PCR

1

Elucidation of insertion elements encoded on plasmids and in vitro construction of 1

shuttle vectors from the toxic cyanobacterium Planktothrix 2

Guntram Christiansen1, Alexander Goesmann2, Rainer Kurmayer1* 3

1 University of Innsbruck, Research Institute for Limnology, Mondseestrasse 9, 5310 4

Mondsee, Austria 5

2 Bielefeld University, Computational Genomics, CeBiTec/BRF, 33594 Bielefeld, Germany 6

7

Running title: Shuttle vectors from the cyanobacterium Planktothrix 8

9

Keywords: mobile elements, cyanotoxins, mutagenisation, recombination, whole genome 10

sequencing 11

12

13

*Corresponding author: 14

Rainer Kurmayer 15

University of Innsbruck 16

Research Institute for Limnology 17

Mondseestrasse 9 18

A-5310 Mondsee 19

20

Tel.: 0043-512-507-50242 21

E-mail: [email protected] 22

23

AEM Accepts, published online ahead of print on 6 June 2014Appl. Environ. Microbiol. doi:10.1128/AEM.01188-14Copyright © 2014, American Society for Microbiology. All Rights Reserved.

on July 6, 2018 by guesthttp://aem

.asm.org/

Dow

nloaded from

http://aem.asm.org/

2

Abstract 24

Several gene clusters that are responsible for toxin synthesis in bloom-forming cyanobacteria 25

have been found to be associated with transposable elements (TEs). In particular, insertion 26

(IS) elements were shown to play a role in the inactivation or recombination of the genes 27

responsible for cyanotoxin synthesis. Plasmids have been considered as important vectors of 28

IS element distribution to the host. In this study, we aimed to elucidate the IS elements 29

propagated on the plasmids and the chromosome of the toxic cyanobacterium Planktothrix 30

agardhii NIVA-CYA126/8 by means of high throughput sequencing. In total, five plasmids 31

(pPA5.5, 14, 50, 79, 115 kbp) were elucidated and two plasmids (pPA5.5, 115 kb) were found 32

to propagate full IS element copies. Large stretches of shared DNA information between 33

plasmids were constituted of TEs. Two plasmids (pPA5.5, 14 kbp) were used as candidates 34

for engineering shuttle vectors (named pPA5SV, pPA14SV) in vitro by PCR amplification 35

and the subsequent transposition of the Tn5 cat transposon, including the R6Kγ origin of 36

replication of E. coli. While pPA5SV was found fully segregated, pPA14SV consistently co-37

occurred with its wild type plasmid even under the highest selective pressure. Interestingly, 38

the Tn5 cat transposon became transferred by homologous recombination into another 39

plasmid pPA50. The availability of shuttle vectors is considered to be of relevance in 40

investigating the genome plasticity as a consequence of homologous recombination events. 41

Combining the potential of high throughout sequencing and in vitro production of shuttle 42

vectors makes it simple to produce species-specific shuttle vectors for many cultivable 43

prokaryotes. 44

45


.asm.org/

Dow

nloaded from

http://aem.asm.org/

3

Introduction 46

Harmful algal blooms (HABs) formed by freshwater cyanobacteria have been frequently 47

linked to the occurrence of diseases in humans and livestock. The best characterized HAB-48

toxin is the hepatotoxin microcystin (MC). In addition, the peptides nodularin, the alkaloids 49

cylindrospermopsin and (homo)anatoxin-a, and the saxitoxins are reported frequently (1). The 50

genus Planktothrix is of quantitative importance in lakes and reservoirs and frequently 51

involved in bloom formation. Notably, Europe red-pigmented Planktothrix is always toxic 52

due to MC production (2), while the production of homoanatoxin-a (3), (4), and saxitoxin (5) 53

has been reported only occasionally. The reasons leading to the sudden appearance of certain 54

toxin producers are unclear. 55

High throughput sequencing has enabled the exploration of the genetic information of 56

prokaryotes from both isolates and communities at an unprecedented scale e.g. (6). 57

Consequently, it is hoped that genome wide comparisons can reveal the interdependence of 58

toxin production and other ecophysiological adaptations, e.g. (7). For a long time, it has been 59

shown that, in prokaryotes, plasmids are a major source of genetic variation and novel 60

ecophysiological adaptations, such as resistance to antibiotics and heavy metals, toxin 61

production, and gas vesicles, e.g. (8), (9). Surprisingly, up to date, none of the gene clusters 62

encoding toxin synthesis in cyanobacteria has been found on a plasmid. Indeed, the vast 63

majority of plasmids elucidated among cyanobacteria are considered cryptic. Only for a few 64

plasmids could the biological role of their genetic information, such as pANL, the large 65

plasmid of Synechococcus elongatus PCC7942, be elucidated (10). 66

Within cyanobacteria, the majority of the investigated plasmids is found to contain genes 67

encoding transposable elements (TEs), so-called insertion (IS) elements ranging from 1-2 kbp 68

in size and typically consisting of 1-3 ORFs, including a transposase (11), (12). It is striking 69

that the percentage of the genetic information devoted to TEs on a particular plasmid is higher 70

when compared with the chromosome, (e.g. (12), Table 2). Already during the time prior to 71


.asm.org/

Dow

nloaded from

http://aem.asm.org/

4

the era of genome sequencing, plasmids have been considered essential in the dissemination 72

of TE to individual E. coli strains (13), (14). Using a branching-process model, it could be 73

shown that correlations among unrelated TEs can result from the dissemination of TEs by 74

infectious plasmids (15). Those plasmids would accumulate TEs through time, and when they 75

are transferred they would infect the host simultaneously with two or more unrelated TEs. 76

This hypothesis has been confirmed by genome sequencing, e.g. for Rhizobium etli 77

populations (16). 78

In previous work, we described different recombinations and TEs affecting the production of 79

MC in Planktothrix. For example, the TE called ISPlr1 has inactivated MC synthesis 80

repeatedly when it was inserted into the genes of the MC synthetase (mcy) gene cluster (2). It 81

is interesting to note that only ISPlr1 could be found regularly among the red-pigmented 82

strains of Planktothrix, while among the green-pigmented strains the occurrence of ISPlr1 was 83

variable (R. Kurmayer, unpublished). A second type of TE (ISPlag1) has been shown to 84

induce the loss of the entire 50 kbp mcy gene cluster via site specific recombination (17). 85

Analogously, TEs or their remainders have been observed in association with the loss or the 86

inactivation of the mcy gene cluster in Microcystis (18), (19) or in Anabaena (20). 87

In this study, we aimed to elucidate the TEs propagated on the plasmids of the toxic 88

cyanobacterium Planktothrix agardhii NIVA-CYA126/8 by means of high throughput 89

sequencing. This strain has been repeatedly found to be amenable to genetic manipulation by 90

electroporation in several laboratories and has been of significance in the elucidation of gene 91

clusters encoding the synthesis of various toxic and bioactive peptides [(17), (21), (22), (23), 92

(24)]. From two of the plasmids, shuttle vectors were constructed in vitro by PCR 93

amplification and the subsequent transposition of a DNA fragment encoding the Tn5 94

transposase including the R6Kγ origin of replication of E. coli. This new technique is 95

considered to be of potential for enabling genetic manipulation methods for non-model 96

organisms such as bloom-forming cyanobacteria. This is a significant advance, since 97


.asm.org/

Dow

nloaded from

http://aem.asm.org/

5

cyanobacteria forming HABs have so far been mostly elusive to genetic manipulation 98

techniques, e.g. (25). 99

100

Materials and Methods 101

Organisms and growth conditions. The green pigmented cyanobacterium P. agardhii 102

NIVA-CYA126/8 was isolated by Olav Skulberg (Norwegian Institute of Water Research, 103

Oslo) from Lake Langsjön in 1984. Cells were grown in BG11 (26) under 30 µmol m-2 s-1 at 104

20°C under sterile conditions. The filaments of NIVA-CYA126/8 were purified according to 105

Rippka (26). The status of microbial purity was tested prior to sequencing by (i) selective 106

media in the dark (27), (ii) DAPI staining of contaminant bacteria on membrane filters, (iii) 107

16S rRNA gene amplification using general primers (27F, 1492R of E. coli, (28) and cloning 108

and RFLP analysis of 16S rRNA products. None of the tests revealed any evidence of 109

bacterial contamination. Transformed P. agardhii (see below) was grown in BG11 as 110

indicated above and supplemented with 1 µg ml-1 chloramphenicol (Cm). In general, 111

Planktothrix is sensitive to Cm and this concentration of Cm has previously been used for 112

genetic manipulation and growing genetically modified strains (21). In order to quantify 113

copies of plasmids as well as the transformed shuttle vectors (see below) cultures of wildtype 114

and transformants were grown in BG11 supplemented with 0-10 µg ml-1 of Cm at 20°C under 115

16:8 h light:dark conditions (10 µmol m-2 s-1). 116

117

Genome sequencing. The axenic strain P. agardhii NIVA-CYA126/8 was harvested by 118

centrifugation and high molecular weight DNA was extracted and sequenced using 454 119

pyrosequencing (GS20) by Roche (Penzberg, Germany) and assembled using the Newbler 120

Metrics software (23× coverage, estimated genome size 5 Mbp, Version v. 1.1.02.09). The 121

sequencing resulted in 805 contigs that were reduced to 23 scaffolds by paired end 122

sequencing. The construction of a fosmid genome library (CopyControl™ Fosmid Library 123


.asm.org/

Dow

nloaded from

http://aem.asm.org/

6

Production Kit, Epicentre, Biozym, Vienna) was critical for finally combining the 23 124

scaffolds to one scaffold. By fosmid end sequencing (475 fosmids) and long-range PCR the 125

integrity of the assembled chromosome was confirmed (99.6% of the total chromosome). 126

During this process, five plasmids (5, 5.9, 50, 70, 115 kbp) were elucidated and confirmed by 127

overlapping PCR amplification using the Phusion High-Fidelity DNA Polymerse (Finnzymes, 128

Thermo Scientific, Vienna, Austria), (Suppl. Table 1). This polymerase has a high 129

processivity and proofreading accuracy resulting in the ability to amplify long templates (> 10 130

kbp) following the conditions of the manufacturer. The sequences of the chromosome and 131

plasmids were submitted to NCBI (PRJNA163669). The Whole Genome Shotgun project has 132

been deposited at DDBJ/EMBL/GenBank under the accession number ASAK00000000. The 133

version described in this paper is the first version, ASAK01000000. 134

135

Genome annotation. The annotation of the genome was performed using the automatic 136

annotation by the gene prediction tools Glimmer + Critica (GenDB), (29). The automatic 137

genome annotation was corrected using manual annotation with regard to the plausibility of 138

the proposed ORFs. In general, the annotation of ORFs was not approved when (i) the amino 139

acid sequence was < 33, (ii) no conserved domains were found by homology searches using 140

Psi-BLAST against COG and BlastP, (iii) it appeared to be part of a larger ORF. The 141

individual plasmids were analyzed using the Vector NTI software package (Invitrogen, 142

Germany). TEs were classified using the IS finder database http://www-is.biotoul.fr/is.html 143

using an e-value (< 1e-30), (30). The inverse repeats (IR) were either identified from the 144

closest homologs (< 1e-30) or by means of the EINVERTED program with default settings on 145

the bioinformatics portal Mobyle (http://mobyle.pasteur.fr/cgi-bin/portal.py). 146

147

Shuttle vector construction. From the genomic DNA, two of the five plasmids were 148

amplified by PCR using the Phusion High-Fidelity Polymerase and the amplicons were used 149


.asm.org/

Dow

nloaded from

http://aem.asm.org/

7

for self-ligation. The PCR products of the correct sizes were isolated using a gel extraction kit 150

(Qiagen, Hilden, Germany). The PCR products were 5’ prime-phosphorylated using T4 151

polynucleotide kinase (Thermo Scientific, MBI Fermentas, St. Leon-Rot, Germany) under 152

standard conditions, column purified, and diluted to 5 ng µl-1 for self-ligation. Self-ligation 153

was performed using T4 DNA ligase (MBI Fermentas) at 22°C for 1 h in a total volume of 154

500 µl reaction mix (containing 1× ligase buffer, 5% polyethylenglycol 4000, 50 units of 155

enzyme). After column purification (Qiagen), an equimolar amount of plasmid (pPA5.5, 156

pPA14) and the constructed transposon (0.0025 pmol µl-1) were used in 10 µl containing 1× 157

reaction buffer and 1 unit of EZ-Tn5 Transposase (Epicentre) following the manufacturer’s 158

instructions. The transposon was constructed by the insertion of a cat (chloramphenicol acetyl 159

transferase) resistance marker gene isolated from the pACYA184 vector (NEB, Frankfurt am 160

Main, Germany) into the SmaI restriction site of the multi-cloning site of the pMOD5 vector 161

(Epicentre). After successful integration, the transposon containing the R6Kγ origin of 162

replication of E. coli was isolated from the pMOD vector by PCR using the manufacturer’s 163

primers (Epicentre). The PCR product was purified and subsequently used for in vitro 164

transposition as outlined above. 165

One µl of the in vitro transposition mix was used for electroporation into Transformax 166

EC100D pir+ electrocompetent E. coli (Epicentre). After overnight incubation (37°C) on LB 167

agar (12.5 µg ml-1 Cm), approx. 250 colonies were obtained. Plasmids were purified from 168

twelve randomly chosen colonies and used for DNA sequencing using the SqFP (5´-169

GCCAACGACTACGCACTAGCCAAC-3´) and SqRP (5´-170

GAGCCAATATGCGAGAACACCCGAGAA-3´) primers (Epicentre) to elucidate the 171

insertion site of the transposon into the plasmid pPA5.5 or pPA14. The sequences of the 172

pPA5.5 and pPA14 shuttle vectors were submitted to Genbank/NCBI/DDBJ (Access. No. 173

JX134573, JX134574). 174


.asm.org/

Dow

nloaded from

http://aem.asm.org/

8

For the transformation of P. agardhii 30 ml of a mid-log phase culture (culture conditions as 175

above, OD880nm = 0.4) was treated using the “Hammer, Cork, and Bottle" method (31) to 176

destroy the gas vesicles and enable sedimentation during centrifugation (10,000 g, 10 min, 177

RT). The pellet was washed five times with 1 mM sterile HEPES (Sigma, Vienna, Austria) 178

and finally resuspended in 100 µl of 1 mM HEPES. Ten µg of the shuttle vector plasmid 179

DNA were added, and vigorously mixed with the cell suspension by vortexing under sterile 180

conditions. Electroporation conditions using a gene pulser Xcell (Biorad, Vienna, Austria) 181

were identical as published previously (21). 182

Ten ml of the P. agardhii transformant culture (OD = 0.4) were used for plasmid prep 183

(Thermo Scientific, MBI Fermentas) following alkaline lysis (32). Only a small amount of the 184

total plasmid DNA was isolated (2 ng µl-1) subsequent to column elution. One µl was used for 185

the electroporation of Transformax EC100D pir+ electrocompetent E. coli cells and after 186

overnight incubation (37°C) on LB agar (12.5 µg ml-1 Cm) >1,000 colonies were obtained. In 187

order to elucidate the potential recombination events that could have happened during 188

propagation in Planktothrix, plasmids were isolated from twelve randomly chosen colonies 189

and used for RFLP analysis (TruII). Only one restriction type was found. 190

191

Quantification of the copy number of plasmids and the stability of shuttle vectors 192

To investigate the stability of the shuttle vectors, transformants were grown under a gradient 193

of Cm concentrations ranging from 0-10 µg Cm ml-1. Cultures were harvested by filtering on 194

glass fiber filters and DNA was quantitatively extracted as described previously (33). Aliquots 195

were filtered onto membrane filters in parallel, stained with DAPI, and enumerated using 196

epifluorescence microscopy as described (34). The chromosome and all five plasmids were 197

quantified using the established qPCR protocols (35). Briefly, 100 ng of DNA template were 198

added to Sybr Green qPCR master mix (Thermo Scientific, Fermentas) and primer pairs 199

specifically amplifying the loci of the chromosome, and the five plasmids were added (Suppl. 200


.asm.org/

Dow

nloaded from

http://aem.asm.org/

9

Table 2). All of the measurements were performed in triplicate in a total reaction volume of 201

12 µl using an Eppendorf Master Cycler Ep Realplex system. The initial denaturation of 10 202

min at 95°C was followed by 40 cycles of a three-step PCR with denaturating, annealing, and 203

elongation temperature of 95°C (15 s), 60°C (15 s), and 72°C (20 s), respectively, followed by 204

a standard melting curve protocol (95°C, 15 s, 60°C, 15 s, 60-95°C, 20 min). For each gene 205

locus, the fluorescence threshold values (Ct) were calibrated using a plasmid carrying the gene 206

fragment of interest (approx. 100 bp, which was ligated into a standard cloning vector) and 207

relating the template DNA concentration (0.1 pg–1 ng) to the obtained Ct value using 208

regression curves (Suppl. Table 2). 209

210

Rescue cloning 211

In order to find out whether the Tn5 cat transposon has been transferred by homologous 212

recombination into other plasmids that share identical sequence regions (pPA14 and pPA50 or 213

pPA79), genomic DNA was isolated from both pPA5.5SV and pPA14SV transformant 214

cultures (see above). Three µg of isolated DNA were incubated with PacI following the 215

guidelines of the manufacturer (NEB). After digestion, the sample was precipitated and the 216

DNA pellet was washed three times with 70% EtOH. After all the liquids have been 217

evaporated the pellet was resuspended in 50µl water. 500 ng of the digested DNA was used 218

for self-ligation and precipitated (see above). 100 ng of the self-ligated DNA was used to 219

transform electrocompetent pir+ cells. The obtained colonies were screened by PCR using 220

pMOD based primer pair SqFP and SqRP to amplify DNA sequences flanking the transposon 221

Tn5. Obtained PCR products were screened by RFLP (DraI) and colonies representing 222

different restriction patterns were cultivated overnight and used for plasmid isolation 223

procedures. Isolated plasmid DNA was sequenced according to standard conditions. 224

225

Results 226


.asm.org/

Dow

nloaded from

http://aem.asm.org/

10

Identification of plasmids. In total, five plasmids (pPA5.5, 14, 50, 79, 115) were elucidated 227

(Table 1). The plasmids pPA5.5 + 14 and the plasmids pPA50 + 79 were previously 228

misassembled and incorrectly fused into scaffolds using the Newbler Metrics software. The 229

reason for the mis-assemblage was either due to a TE shared by pPA5.5 + 14 or a plasmid 230

partitioning protein (parA) shared by pPA50 + 79. In order to prove the physical existence of 231

each of the plasmids, DNA fragments of 10,000 bp were amplified by designing 232

forward/reverse primer pairs in overlapping position and allowing for the amplification of the 233

whole plasmid without interruption. For all the plasmids pPA50, 79, 115, all of the primer 234

sites (Suppl. Table 1) revealed PCR products in the expected size (Fig. 1). For plasmids 235

pPA5.5 and pPA14, the expected smaller sized PCR products (3 kbp) were obtained. It is 236

concluded that the DNA molecules were indeed circular. 237

Aligning the DNA sequences of the plasmids with each other revealed large stretches of 238

shared DNA (> 80% similarity). For example, 55% and 38% of the DNA sequence 239

information found in pPA14 were also found in pPA50 and pPA79 (Table 2). Forty percent of 240

the DNA of pPA5.5 was also contained in pPA115 and in the chromosome. Consequently, 241

based on the shared DNA sequence, two groups of DNA molecules were observed: pPA14, 242

50, 79 were more closely related to each other when compared with pPA5.5, pPA115, and the 243

chromosome. 244

245

Annotation of plasmids. The five plasmids contained 4-102 ORFs (Table 1, Fig. 2). Except 246

for pPA14, the GC dinucleotide content per ORF on plasmids was higher and the AC 247

dinucleotide content per ORF on plasmids was lower than when compared with the ORFs 248

located on the chromosome (GC dinucleotide content: p = 0.2, not significant, AC 249

dinucleotide content: p < 0.001, nonparametric Kruskal Wallis ANOVA on ranks). In total, 95 250

(46%) of the 208 ORFs located on the plasmids were of an unknown function. The majority 251

of the ORFs located on all the plasmids had cyanobacterial homologs (78% cyanobacteria, 252


.asm.org/

Dow

nloaded from

http://aem.asm.org/

11

11% bacteria, 12% no match). Within cyanobacteria, the most abundant closest homologs 253

were from Oscillatoria PCC6506 (28%), Cyanothece PCC7822 (18%), Nostoc PCC7120 254

(10%), and Lyngbya PCC8106 (9%). 255

In total, 2,932 ORFs (68%) were assigned to COG categories (36). Except for pPA14, the 256

plasmids showed a higher proportion of ORFs assigned to replication, recombination, and 257

repair (COG category L, 33-50%) than to the chromosome (7%). The other COG categories 258

did not differ significantly in proportion between the plasmids and chromosome (data not 259

shown). The largest ORFs were encoding polyketide synthases located on pPA79 (pPA79_30: 260

7.6 kbp, pPA79_37: 8.4 kbp). Those genes encoded two complete PKS type I modules each 261

including a keto-synthase, acyl-transferase, dehydrogenase, enoylreductase, ketoreductase, 262

and acyl carrier domain. The ORF pPA79_37 contained a second acyl carrier domain. 263

Associated genes included a putative glycosyltransferase (pPA79_33), putative oligoketide 264

cyclase (pPA79_34), and putative free standing ketosynthase (pPA79_35). 265

266

Transposable elements. Including the chromosome and the plasmids, a total of 91 ORFs 267

could be unequivocally assigned to TE constituting 1.4% of the genome. All of the plasmids 268

were found to contain at least the remainders of TEs. Except for pPA14, the percentage of the 269

genetic information attributable to TE was higher on the plasmids than on the chromosome 270

(Table 1). Furthermore, the GC dinucleotide content and the AC dinucleotide content of the 271

91 ORFs assigned to TEs was significantly different from the rest of the ORFs located on 272

both plasmids and the chromosome, i.e. a GC dinucleotide content of 35.6 (33.9, 38) within 273

TEs vs. 39.6 (36.6, 43.4) within non TEs (p < 0.001, Mann-Whitney U test) or a AC 274

dinucleotide content of 52.4 (50.8, 53.9) within TEs vs. 49.5 (47.2, 51.6) within non TEs (p < 275

0.001, Mann-Whitney U test). Consequently, the differences in the dinucleotide content 276

between TEs and non-TEs ORFs in combination with the higher proportion of TEs on 277


.asm.org/

Dow

nloaded from

http://aem.asm.org/

12

plasmids can explain the dinucleotide differences as observed between plasmids and the 278

chromosome. 279

TEs significantly contributed to the DNA sequence information shared between the plasmids. 280

Between pPA5.5, pPA115, and the chromosome, 88-100% of the shared information was 281

attributable to TEs. Between pPA14, pPA50, and pPA79, 0-83% of the shared information 282

consisted of TEs (Table 2). Out of the 91 ORFs assigned to TEs, 29 ORFs (32%) were 283

identified as full copies of IS elements. Among these, seven ORFs (32%) were located on two 284

of the plasmids (pPA5.5, pPA115). The plasmids contained 13 (27%) of all the partial TE 285

copies. The TEs were classified into eleven groups comprising 2-16 copies (Table 3). Only 286

the IS element groups I, II, and IV showed the lowest genetic variability on the nucleotide 287

level (< 2.1%) and more than two full copies suggesting a relatively recent insertion activity, 288

e.g. (11), (16). The IS element (group I) that is flanking the mcy gene cluster at the 289

downstream end (21) occurred in seven full copies that are all located on the chromosome. 290

The IS element ISPlag1 (group II) that caused the deletion of the mcy gene cluster resulting in 291

nontoxic Planktothrix (17) occurred in five copies (located on pPA115 and the chromosome) 292

and seven nonfunctional remainders. Group IV (located on pPA5.5, pPA115, and the 293

chromosome) contained six full copies that are highly identical to a transposase (anaH) 294

associated with the gene cluster encoding (homo)anatoxin-a synthesis in Oscillatoria 295

PCC6506 (37). In summary, two plasmids (pPA5.5, pPA115) carried putatively active TEs 296

(groups II, IV) while four plasmids carried partial TE copies from IS element groups I, II, IX, 297

X, and unassigned residues. 298

299

Application of shuttle vectors. Two of the five isolated plasmids (pPA5.5, pPA14) were 300

produced in vitro and subsequently used for shuttle vector construction (pPA5SV, pPA14SV). 301

Both shuttle vectors were successfully shuttled back and forth between E. coli and P. 302

agardhii. In both cases, the PCR experiments using primer pairs (pPA5SV: TpmChk+/-, 303


.asm.org/

Dow

nloaded from

http://aem.asm.org/

13

pPA14SV: TpmChkN+/-) binding in the proximity of the inserted Tn5 region (1295 bp) 304

revealed the expected PCR product in P. agardhii (pPA5SV: 1697 bp; pPA14SV: 1921 bp) 305

when compared with the product size obtained directly from the shuttle vector DNA (Fig. 3). 306

However, for pPA14SV, the amplification of the Tn5 insertion site revealed two PCR 307

products (626 bp, 1921 bp) indicating the coexistence of both wildtype and modified plasmid 308

in P. agardhii (Fig. 3). The sequencing of these PCR products for both shuttle vectors 309

revealed an insertion of the complete Tn5 region (pPA5SV: at nt 112, pPA14SV: at nt 4339, 310

1295 bp) with left and right inverse repeat sequences (5’-CTGTCTCTTATACACATCT-3’) 311

and a 9-bp direct repeat sequence (pPA5SV: 5’- GCTCTACTG-3’, pPA14SV: 5’- 312

GCAATAAAC-3’), resulting from the DNA repair after Tn5 insertion (Suppl. Fig. 1). 313

Moreover, it became apparent that in pPA5SV the Tn5 transposon inserted at nt 112, which 314

was a sequence region unique to pPA5.5 (< nt 1097, > nt 3160). In contrast, in pPA14SV the 315

insertion site at nt 4339 (> nt 3508, < nt 850) was identical to a gene region that is also 316

located in pPA50 (Suppl. Fig. 2), with 44% identity to a hypothetical protein Osc7112_6402 317

of Oscillatoria nigro-viridis PCC7112 possibly related to the TOPRIM (topoisomerase-318

primase) domain. 319

320

Copy number of plasmids and the stability of shuttle vectors. In order to investigate the 321

stability of shuttle vectors, both plasmids and shuttle vectors were quantified and compared in 322

numbers with the chromosome. On average, one WT cell contained 4.2 ± 0.4 (SD) 323

chromosome copies, and plasmids pPA5, 14, 50, 79, 115 occurred with a frequency of 8.2 ± 324

0.4, 4.5 ± 1, 5.6 ± 0.3, 2.2 ± 0.6, and 5.6 ± 0.1 fold the chromosome copy, respectively (Fig. 325

4). 326

In order to eliminate the WT plasmid from the transformant culture of the pPA14SV, the Cm 327

selection pressure was raised and the ratio of each SV molecule to the chromosome was again 328

monitored by qPCR. For pPA5SV, the ratio of pPA5.5/chromosome and Cm-resistance 329


.asm.org/

Dow

nloaded from

http://aem.asm.org/

14

gene/chromosome remained stable and the transformant grew up to 10 µg Cm ml-1 (Suppl. 330

Fig. 3). The pPA14SV transformant was less resistant to Cm and did not grow above 7 µg Cm 331

ml-1. The lower resistance of the pPA14SV transformant could be explained by the lower 332

copy number and the observation that the pPA14SV transformation culture could not be fully 333

segregated. Surprisingly, however, pPA14SV also showed rather unaltered ratios of 334

pPA14SV/chromosome and Cm-resistance gene/chromosome under increasing Cm selective 335

pressure. Even the long-term cultivation of pPA14SV transformant under 7 µg Cm ml-1 did 336

not lead to a complete segregation. 337

338

Rescue cloning. Since plasmids pPA5.5 and pPA14 both contained large stretches of DNA 339

information shared with other plasmids/chromosome (see above), we were interested to see 340

whether the Tn5 cat transposon has been translocated by homologous recombination. In most 341

cases, the DNA sequences obtained through rescue cloning represented the shuttle vector 342

pPA14SV. In one case, however, the DNA sequence also showed identity to the native 343

plasmid pPA50 indicating the occurrence of a homologous recombination event between 344

pPA14SV and plasmid pPA50 (Suppl. Fig. 5). The flanking regions of the Tn5 cat transposon 345

were amplified and sequenced using the pPA50 specific primer pairs pPA50reco+/pMOD+ 346

and pPA50reco-/pMOD-. In both cases, the amplicons of the expected sizes were detected: 347

2.6 kbp (pPA50reco+/pMOD+) and 0.8 kbp (pPA50reco-/pMOD-), (Fig. 5). Unspecific 348

amplification products were visible for the primer pair pPA50reco+/pMOD+ indicating the 349

repetitive occurrence of the pPA50reco+ primer binding sites. Thus, we could show a 350

dynamic homologous recombination activity between two genetic elements. 351

352

Discussion 353

Genome copy numbers. The copy number of the chromosome was estimated as 4.2 ± 0.4 per 354

cell, which might be at odds with the expectation of one copy per cell. Following the 355


.asm.org/

Dow

nloaded from

http://aem.asm.org/

15

terminology of Griese et al. (38), P. agardhii NIVA-CYA126/8 would be classified as 356

oligoploid. The maintenance of several genome copies might be a more common feature in 357

cyanobacteria (38). Typically, in cyanobacteria, the estimates of copy numbers vary between 358

1 and 10, while a few strains even have higher copy numbers, e.g. Microcystis strain HUB524 359

(39) or Synechocystis PCC6803 (38). Griese et al. (38) concluded that polyploid 360

cyanobacteria exist not only under laboratory conditions but also in nature. 361

An alternative explanation to the observed variation in copy number within cyanobacteria is 362

the physiologically induced variability. For example, during the transfer from the exponential 363

to the stationary phase in batch culture, Microcystis strain HUB524 showed a tenfold 364

variation in the genome copy number (35). Under general (batch) culture conditions, cell 365

growth is only rarely synchronized and, consequently, it is the variation in physiological state 366

of the cells that might contribute to the observed variability. The ratio of the plasmid copy 367

numbers to the chromosome, however, should not be affected by such physiological 368

variability on the cellular level. 369

370

Whole genome sequencing. Up to date, very few harmful algal bloom-forming species such 371

as toxic cyanobacteria have been totally sequenced. Only the Kazusa DNA sequencing 372

institute in Japan succeeded to form one contiguous DNA molecule of the toxin-producing 373

cyanobacterium Microcystis (40). The large insert genome library sequencing efforts 374

performed during this study were essential to test for the integrity of the automatically 375

assembled genome. Indeed, it could be shown by fosmid end sequencing that multiple copies 376

of IS elements and the parA gene led to incorrect assemblies such as those of pPA5.5 + 377

pPA14 and pPA50 + pPA79. This approach has been emphasized, as automatic assemblies 378

are frequently insufficient and error prone (41), (42). Total genomes are invaluable for 379

emerging fields such as “ecogenomics”, when the plasticity of microorganisms to changing 380

environmental conditions needs to be investigated. So far, it has been speculated that TEs in 381


.asm.org/

Dow

nloaded from

http://aem.asm.org/

16

general are an important source of physiological plasticity in bloom-forming cyanobacteria 382

such as M. aeruginosa (43), (44). The influence of TEs on the restructuring of genomes and 383

the inactivation/rearrangement of genes can only be investigated when completely assembled 384

genome information is available. For example, Bickhart et al. (45) used a 250 kb-sliding 385

window analysis in order to identify those gene regions indicative of (potential) genome 386

flexibility. The TE frequency of occurrence and the occurrence of genome 387

rearrangements/deletion events was then compared statistically by plotting the number of TEs 388

against the variable genome regions as identified by genome comparison. This approach has 389

thus far not been possible for bloom-forming cyanobacteria. It is hoped that the P. agardhii 390

NIVA-CYA126/8 genome sequence will form a suitable reference for this type of genomic 391

variability analysis. 392

393

Plasmid encoded IS elements and toxin synthesis. In this study, we show that P. agardhii 394

contains only relatively few groups of IS elements occurring with more than two copies: 395

Groups I, II, IV. Notably, full copy representatives or residues of each of these groups were 396

associated with cyanotoxin synthesis previously: Group I was associated with the mcy gene 397

cluster of P. agardhii (21), Group II was associated with the loss of the mcy gene cluster (17), 398

and group IV was associated with the (homo) anatoxin gene cluster in the closely related 399

Oscillatoria PCC6506 (37). In contrast, any IS element ISPlr1 that has been shown to 400

inactivate MC synthesis by insertion into the mcy gene cluster in numerous red-pigmented 401

Planktothrix strains (2) was detected in the P. agardhii NIVA-CYA126/8 chromosome or in 402

the plasmids. We could show earlier by multi locus sequence typing (MLST) that the red-403

pigmented Planktothrix strains occurring in European lakes are relatively distantly related to 404

P. agardhii NIVA-CYA126/8 (17). Consequently, the elucidation of the TEs propagated on 405

plasmids from both red- and green pigmented strains can in accordance with phylogenetic 406

analysis (e.g. by MLST) help to explain the distribution of putatively active IS elements. For 407


.asm.org/

Dow

nloaded from

http://aem.asm.org/

17

P. agardhii NIVA-CYA126/8, the three IS element groups I, II, IV differ in their frequency of 408

occurrence among plasmids and the chromosome (Table 3). While full copies of group I were 409

located on the chromosome, full copies and residues of groups II and IV were located on 410

plasmids and the chromosome. Consequently, groups II and IV should have a wider 411

distribution when compared with group I. Accordingly, anaH described from the 412

(homo)anatoxin that a gene cluster in Oscillatoria (37) is almost identical (74-78% similarity 413

on nucleotide level, 515 bp) to group IV described in this study. In total, one full copy [78-414

82% similarity on the nucleotide (1924 bp) and protein level (554 aa)] and two partial copies 415

of group IV can be found in the Oscillatoria PCC6506 draft genome (46). In only a few cases 416

has Planktothrix been reported to produce (homo)anatoxin a (3), (4). In contrast, numerous 417

Oscillatoria strains have been reported to produce anatoxin-a (47). Given the close genetic 418

and ecological relationship between benthic Oscillatoria sp and both benthic and planktonic 419

Planktothrix, e.g. (48), it is possible that the IS elements part of group IV are involved in the 420

translocation of (homo)anatoxin-a synthesis gene cluster among these two genera in certain 421

habitats. 422

423

Applicability of shuttle vectors. Shuttle vectors propagate in two different species allowing 424

all the DNA manipulation steps in one model organism like E. coli and subsequent transfer 425

into the organism of interest. As cyanobacteria from section III are typically motile, the 426

transformation efficiency cannot be determined by counting CFUs. Therefore, the 427

applicability of the shuttle vector pPA5.5 and pPA14 was confirmed by the amplification of 428

the cat gene in both E. coli and Planktothrix under chloramphenicol pressure. Typically for 429

the construction of shuttle vectors, high amounts of plasmids are required, which is a process 430

that is often laborious, including mass-cultivation, caesium-chloride ultra-centrifugation, 431

cloning, and sequencing processes, e.g. (49), (50). With the dawn of second generation whole 432

genome sequencing, it became a fast and inexpensive standard tool. Usually a part of the 433


.asm.org/

Dow

nloaded from

http://aem.asm.org/

18

sequence data represents plasmids that can be identified by careful annotation. Combining the 434

potential of next generation sequencing and in vitro production of vectors makes it simple to 435

produce species-specific shuttle vectors for many cultivable prokaryotes. Consequently, this 436

technique is considered of high potential to enable genetic manipulation methods for non-437

model organisms. 438

We were able to observe recombination activity in vivo using the Tn5 cat transposon as a 439

tracer. In general, double recombination events of linearized plasmids have been observed in 440

unicellular or filamentous cyanobacteria (using linearized plasmids) while the single 441

recombination of circular plasmids have been observed during conjugation (25). The 442

recombination system utilizes sequence similarity within two DNA molecules, and in E. coli, 443

does not discriminate between perfect and imperfect matches of sequence until a fraction of 444

mismatch of 10% (51). Thus, all the repeated sequences within the genome constitute 445

potential sites of site-specific recombination that can be visualized using the rescue cloning 446

experiments applied in this study. It is interesting to note that plasmids pPA5.5 and pPA115 447

showed the highest sequence similarity with each other (40%) and with the chromosome 448

(pPA5.5: 40%; pPA115: 15%), while pPA14 showed the highest similarity with pPA50 (55%) 449

and pPA79 (39%), (Table 2). Thus, there is some likelihood that there are genes located on 450

plasmids pPA14, 50, 79 although translocated in-between they become less integrated into the 451

chromosome and vice versa, when compared with the genes located on pPA5.5, pPA115. 452

Accordingly, only pPA5.5 and pPA115 show full copies of TEs group II, IV located on the 453

chromosome. Therefore, constructing SVs from elucidated plasmids and using a tracer during 454

rescue cloning would elucidate the routes of recombination in-between various groups of 455

plasmids and the chromosome both with a strain or between strains representing different 456

phylogenetic lineages (17). 457

458

Acknowledgements 459


.asm.org/

Dow

nloaded from

http://aem.asm.org/

19

We are grateful to the comments of the three anonymous reviewers of an earlier draft of the 460

manuscript. We thank Sabine Münch-Gatthof, Andreas Mausolf, Olaf Kaiser (Roche 461

Penzberg), and Peter Hufnagl (Roche Vienna) for performing the GS20 de novo genome 462

sequencing study. The excellent technical assistance of Maria Reischauer, Katharina 463

Moosbrugger, Josef Knoblechner, and Johanna Schmidt is greatly acknowledged. This study 464

was financially supported by grants from the Austrian Science Fund (P20231, P24070) to 465

R.K. 466

467

References 468

1. Hudnell H. 2008. Proceedings of the interagency, international symposium on 469

cyanobacterial harmful algal blooms (ISOC-HAB): State of the science and research 470

needs. Advances in Experimental Medicine & Biology:924pp. 471

2. Kurmayer R, Christiansen G, Fastner J, Börner T. 2004. Abundance of active and 472

inactive microcystin genotypes in populations of the toxic cyanobacterium 473

Planktothrix spp. Environ. Microbiol. 6:831-841. 474

3. Skulberg OM, Carmichael WW, Andersen RA, Matsunaga S, Moore RE, 475

Skulberg R. 1992. Investigations of a neurotoxic oscillatorialean strain 476

(Cyanophyceae) and its toxin. Isolation and characterization of homoanatoxin-a. 477

Environ. Toxicol. Chem. 11:321-329. 478

4. Viaggiu E, Melchiorre S, Volpi F, DiCorcia A, Mancini R, Garibaldi L, Crichigno 479

G, Bruno M. 2004. Anatoxin-A toxin in the cyanobacterium Planktothrix rubescens 480

from a fishing pond in Northern Italy. Environ. Toxicol. 19:191-197. 481

5. Pomati F, Sacchi S, Rossetti C, Giovannardi S, Onodera H, Oshima Y, Neilan 482

BA. 2000. The freshwater cyanobacterium Planktothrix sp. FP1: molecular 483


.asm.org/

Dow

nloaded from

http://aem.asm.org/

20

identification and detection of paralytic shellfish poisoning toxins. J. Phycol. 36:553-484

562. 485

6. Ma Y, Paulsen IT, Palenik B. 2011. Analysis of two marine metagenomes reveals 486

the diversity of plasmids in oceanic environments. Environ. Microbiol. 14:453-466. 487

7. Gobler CJ, Berry DL, Dyhrman ST, Wilhelm SW, Salamov A, Lobanov AV, 488

Zhang Y, Collier JL, Wurch LL, Kustka AB, Dill BD, Shah M, Verberkmoes NC, 489

Kuo A, Terry A, Pangilinan J, Lindquist EA, Lucas S, Paulsen IT, Hattenrath-490

Lehmann TK, Talmage SC, Walker EA, Koch F, Burson AM, Marcoval MA, 491

Tang YZ, Lecleir GR, Coyne KJ, Berg GM, Bertrand EM, Saito MA, Gladyshev 492

VN, Grigoriev IV. 2011. Niche of harmful alga Aureococcus anophagefferens 493

revealed through ecogenomics. Proc. Natl. Acad. Sci. USA 108:4352-4357. 494

8. Lau RH, Sapienza C, Doolittle WF. 1980. Cyanobacterial plasmids - their 495

widespread occurrence, and the existence of regions of homology between plasmids in 496

the same and different species. Mol. Gen. Genet. 178:203-211. 497

9. Aminov RI, Mackie RI. 2007. Evolution and ecology of antibiotic resistance genes. 498

FEMS Microbiol. Lett. 271:147-161. 499

10. Chen Y, Holtman CK, Magnuson RD, Youderian PA, Golden SS. 2008. The 500

complete sequence and functional analysis of pANL, the large plasmid of the 501

unicellular freshwater cyanobacterium Synechococcus elongatus PCC 7942. Plasmid 502

59:176-192. 503

11. Zhou FF, Olman V, Xu Y. 2008. Insertion Sequences show diverse recent activities 504

in Cyanobacteria and Archaea. BMC Genomics 9:36. 505

12. Lin S, Haas S, Zemojtel T, Xiao P, Vingron M, Li RH. 2011. Genome-wide 506

comparison of cyanobacterial transposable elements, potential genetic diversity 507

indicators. Gene 473:139-149. 508


.asm.org/

Dow

nloaded from

http://aem.asm.org/

21

13. Sawyer SA, Dykhuizen DE, Dubose RF, Green L, Mutangaduramhlanga T, 509

Wolczyk DF, Hartl DL. 1987. Distribution and abundance of insertion sequences 510

among natural isolates of Escherichia coli. Genetics 115:51-63. 511

14. Lawrence JG, Ochman H, Hartl DL. 1992. The evolution of insertion sequences 512

within enteric bacteria. Genetics 131:9-20. 513

15. Hartl DL, Sawyer SA. 1988. Why do unrelated insertion sequences occur together in 514

the genome of Escherichia coli. Genetics 118:537-541. 515

16. Lozano L, Hernandez-Gonzalez I, Bustos P, Santamaria RI, Souza V, Young 516

JPW, Davila G, Gonzalez V. 2010. Evolutionary dynamics of insertion sequences in 517

relation to the evolutionary histories of the chromosome and symbiotic plasmid genes 518

of Rhizobium etli populations. Appl. Environ. Microbiol. 76:6504-6513. 519

17. Christiansen G, Molitor C, Philmus B, Kurmayer R. 2008. Nontoxic strains of 520

cyanobacteria are the result of major gene deletion events induced by a transposable 521

element. Mol. Biol. Evol. 25:1695-1704. 522

18. Tooming-Klunderud A, Mikalsen B, Kristensen T, Jakobsen KS. 2008. The 523

mosaic structure of the mcyABC operon in Microcystis. Microbiol. 154:1886-1899. 524

19. Noguchi T, Shinohara A, Nishizawa A, Asayama M, Nakano T, Hasegawa M, 525

Harada K, Nishizawa T, Shirai M. 2009. Genetic analysis of the microcystin 526

biosynthesis gene cluster in Microcystis strains from four bodies of eutrophic water in 527

Japan. J. Gen. Appl. Microbiol. 55:111-123. 528

20. Fewer DP, Halinen K, Sipari H, Bernardová K, Mänttäri M, Eronen E, Sivonen 529

K. 2011. Non-autonomous transposable elements associated with inactivation of 530

microcystin gene clusters in strains of the genus Anabaena isolated from the Baltic 531

Sea. Environ. Microbiol. Rep. 3:189-194. 532


.asm.org/

Dow

nloaded from

http://aem.asm.org/

22

21. Christiansen G, Fastner J, Erhard M, Börner T, Dittmann E. 2003. Microcystin 533

biosynthesis in Planktothrix: genes, evolution, and manipulation. J. Bacteriol. 534

185:564-572. 535

22. Christiansen G, Philmus B, Hemscheidt T, Kurmayer R. 2011. Genetic variation 536

of adenylation domains of the anabaenopeptin synthesis operon and the evolution of 537

substrate promiscuity. J. Bacteriol. 193:3822-3831. 538

23. Ishida K, Christiansen G, Yoshida WY, Kurmayer R, Welker M, Bonjoch J, 539

Hertweck C, Börner T, Hemscheidt T, Dittmann E. 2007. Biosynthetic pathway 540

and structure analysis of aeruginoside 126A and B, cyanobacterial peptide glycosides 541

bearing an unusual 2-carboxy-6-hydroxyoctahydroindole moiety. Chem. Biol. 14:565-542

576. 543

24. Philmus B, Christiansen G, Yoshida W, Hemscheidt T. 2008. Posttranslational 544

modification in microviridin biosynthesis. Chembiochem 9:3066-3073. 545

25. Flores E, Muro-Pastor AM, Meeks JC. 2008. Gene transfer to cyanobacteria in the 546

laboratory and in nature. In: Antonia Herrero and Enrique Flores: The Cyanobacteria: 547

Molecular Biology, Genomics and Evolution. Caister Academic Press, Norfolk, 548

UK.:45-57. 549

26. Rippka R. 1988. Isolation and purification of cyanobacteria. Meth. Enzymol. 167:3-550

27. 551

27. Wu QL, Boenigk J, Hahn MW. 2004. Successful predation of filamentous bacteria 552

by a nanoflagellate challenges current models of flagellate bacterivory. Appl. Environ. 553

Microbiol. 70:332-339. 554

28. Brosius J, Dull TJ, Sleeter DD, Noller HF. 1981. Gene organization and primary 555

structure of a ribosomal RNA operon from Escherichia coli. J. Mol. Biol. 148:107-556

127. 557


.asm.org/

Dow

nloaded from

http://aem.asm.org/

23

29. Meyer F, Goesmann A, McHardy AC, Bartels D, Bekel T, Clausen J, Kalinowski 558

J, Linke B, Rupp O, Giegerich R, Puhler A. 2003. GenDB--an open source genome 559

annotation system for prokaryote genomes. Nucl. Acids Res. 31:2187-2195. 560

30. Siguier P, Perochon J, Lestrade L, Mahillon J, Chandler M. 2006. ISfinder: the 561

reference centre for bacterial insertion sequences. Nucl. Acids Res. 34:D32-D36. 562

31. Klebahn H. 1929. Über die Gasvakuolen der Cyanophyceen. Verh. Internat. Verein. 563

Theoret. Angew. Limnol. 4:408-414. 564

32. Birnboim HC, Doly J. 1979. Rapid alkaline extraction procedure for screening 565

recombinant plasmid DNA. Nucl. Acids Res. 7:1513-1523. 566

33. Kurmayer R, Christiansen G, Chorus I. 2003. The abundance of microcystin-567

producing genotypes correlates positively with colony size in Microcystis and 568

determines its microcystin net production in Lake Wannsee. Appl. Environ. Microbiol. 569

69:787-795. 570

34. Kosol S, Schmidt J, Kurmayer R. 2009. Variation in peptide net production and 571

growth among strains of the toxic cyanobacterium Planktothrix spp. Eur. J. Phycol. 572

44:49-62. 573

35. Kurmayer R, Kutzenberger T. 2003. Application of real-time PCR for 574

quantification of microcystin genotypes in a population of the toxic cyanobacterium 575

Microcystis sp. Appl. Environ. Microbiol. 69:6723-6730. 576

36. Tatusov RL, Galperin MY, Natale DA, Koonin EV. 2000. The COG database: a 577

tool for genome-scale analysis of protein functions and evolution. Nucl. Acids Res. 578

28:33-36. 579

37. Mejean A, Mann S, Maldiney T, Vassiliadis G, Lequin O, Ploux O. 2009. 580

Evidence that biosynthesis of the neurotoxic alkaloids anatoxin-a and homoanatoxin-a 581

in the cyanobacterium Oscillatoria PCC 6506 occurs on a modular polyketide 582

synthase initiated by L-Proline. J. Am. Chem. Soc. 131:7512-7513. 583


.asm.org/

Dow

nloaded from

http://aem.asm.org/

24

38. Griese M, Lange C, Soppa J. 2011. Ploidy in cyanobacteria. FEMS Microbiol. Lett. 584

323:124-131. 585

39. Schober E, Werndl M, Laakso K, Korschinek I, Sivonen K, Kurmayer R. 2007. 586

Interlaboratory comparison of Taq Nuclease Assays for the quantification of the toxic 587

cyanobacteria Microcystis sp. J. Microbiol. Meth. 69:122-128. 588

40. Kaneko T, Nakajima N, Okamoto S, Suzuki I, Tanabe Y, Tamaoki M, Nakamura 589

Y, Kasai F, Watanabe A, Kawashima K, Kishida Y, Ono A, Shimizu Y, 590

Takahashi C, Minami C, Fujishiro T, Kohara M, Katoh M, Nakazaki N, 591

Nakayama S, Yamada M, Tabata S, Watanabe M. 2007. Complete genomic 592

structure of the bloom-forming toxic cyanobacterium Microcystis aeruginosa NIES-593

843. DNA Res. 14:247-256. 594

41. Birney E. 2011. Assemblies: the good, the bad, the ugly. Nature Meth. 8:59-60. 595

42. Ricker N, Qian H, Fulthorpe RR. 2012. The limitations of draft assemblies for 596

understanding prokaryotic adaptation and evolution. Genomics 100:167-175. 597

43. Frangeul L, Quillardet P, Castets AM, Humbert JF, Matthijs HCP, Cortez D, 598

Tolonen A, Zhang CC, Gribaldo S, Kehr JC, Zilliges Y, Ziemert N, Becker S, 599

Talla E, Latifi A, Billault A, Lepelletier A, Dittmann E, Bouchier C, De Marsac 600

NT. 2008. Highly plastic genome of Microcystis aeruginosa PCC 7806, a ubiquitous 601

toxic freshwater cyanobacterium. BMC Genomics 9:274. 602

44. Kuno S, Yoshida T, Kamikawa R, Hosoda N, Sako Y. 2010. The distribution of a 603

phage-related insertion sequence element in the cyanobacterium, Microcystis 604

aeruginosa. Microb. Environ. 25:295-301. 605

45. Bickhart DM, Gogarten JP, Lapierre P, Tisa LS, Normand P, Benson DR. 2009. 606

Insertion sequence content reflects genome plasticity in strains of the root nodule 607

actinobacterium Frankia. BMC Genomics 10:468. 608


.asm.org/

Dow

nloaded from

http://aem.asm.org/

25

46. Mejean A, Mazmouz R, Mann S, Calteau A, Medigue C, Ploux O. 2010. The 609

genome sequence of the cyanobacterium Oscillatoria sp. PCC 6506 reveals several 610

gene clusters responsible for the biosynthesis of toxins and secondary metabolites. J. 611

Bacteriol. 192:5264-5265. 612

47. Araoz R, Nghiem HO, Rippka R, Palibroda N, De Marsac NT, Herdman M. 613

2005. Neurotoxins in axenic oscillatorian cyanobacteria: Coexistence of anatoxin-a 614

and homoanatoxin-a determined by ligand-binding assay and GC/MS. Microbiol. 615

151:1263-1273. 616

48. Wood SA, Heath MW, Holland PT, Munday R, Mcgregor GB, Ryan KG. 2010. 617

Identification of a benthic microcystin-producing filamentous cyanobacterium 618

(Oscillatoriales) associated with a dog poisoning in New Zealand. Toxicon 55:897-619

903. 620

49. Van den Hondel C, Verbeek S, Vanderende A, Weisbeek PJ, Borrias WE, 621

Vanarkel GA. 1980. Introduction of transposon TN901 into a plasmid of Anacytis 622

nidulans - preparation for cloning in cyanobacteria. Proc. Natl. Acad. Sci. USA 623

77:1570-1574. 624

50. Wallace MM, Miller DW, Raps S. 2002. Characterization of pMa025, a plasmid 625

from the cyanobacterium Microcystis aeruginosa UV025. Arch. Microbiol. 177:332-626

338. 627

51. Bazemore LR, FoltaStogniew E, Takahashi M, Radding CM. 1997. RecA tests 628

homology at both pairing and strand exchange. Proc. Natl. Acad. Sci. USA 94:11863-629

11868. 630

631


.asm.org/

Dow

nloaded from

http://aem.asm.org/

26

632

Table 1: Nucleotide characteristics and dinucleotide frequencies (mean, min, max as 633

calculated from all the ORFs per molecule) in the five plasmids and in the chromosome from 634

Planktothrix agardhii NIVA-CYA126/8. 635

636

Plasmid bp ORFs GC (mean,min,max) AC (mean,min,max) IS elements (%)

pPA5.5 4789 4 38.3 36.1 40.5 53.9 42.3 58.8 8

pPA14 5960 4 39.9 36.5 45.0 47.8 39.6 56.3 0.8

pPA50 50852 43 37.8 27.2 50.2 51.1 40.1 58.2 3.5

pPA79 79107 55 40.2 31.4 48.5 49.8 37.4 62.2 4.3

pPA115 119570 102 38.8 30.7 49.6 51.5 37.7 58.9 6.9

Chromosome 4786776 4159 39.6 24.6 69.8 49.5 24.1 80.3 1.1

TOTAL 5047053 4367 39.5 24.6 69.8 49.6 24.1 80.3 1.4

637

* significantly different at p<0.001 (Kruskal Wallis One Way ANOVA) 638


.asm.org/

Dow

nloaded from

http://aem.asm.org/

27

639

Table 2: Percentage of shared DNA sequence information (> 100 bp) between individual 640

plasmids and the chromosome as determined by BLASTn (min. similarity 80%, e-value < 1 e-641

10), the percentage attributable to TE is given in parentheses. 642

643

Query/subject pPA5.5 pPA14 pPA50 pPA79 pPA115 Chromosome

pPA5.5 _ 0 0 0 40(100) 40(100)pPA14 0 _ 54.8(0) 38.8(0) 0 0pPA50 0 6.4(0) _ 6.3(56) 2.2(24) 2.9(100)pPA79 0 2.9(83) 5.3(83) _ 3.4(95) 5.7(94)pPA115 4.6(100) 0 2.3(89) 4.3(97) _ 15(92)Chromosome 0.2(99) 0 0.2(72) 0.3(73) 0.8(88) _

644

645


.asm.org/

Dow

nloaded from

http://aem.asm.org/

28

Table 3: List of transposable elements occurring in Planktothrix agardhii NIVA-CYA126/8 on plasmids and the chromosome 646

Gro

up N

o (C

opy

No)

No

copi

es

(ful

l len

gth,

Loc

atio

n (f

ull l

engt

h)

No

of p

arti

al

copi

es

(res

idue

s (a

a)

Loc

atio

n (r

esid

ues)

Clo

sest

ho

mol

ogue

(B

LA

STp)

% id

entit

y

Tot

al le

ngth

4

(bp)

% v

aria

bilit

y O

RF

(ful

l IS

)

IR(L

) 5‘

-3‘

IR(R

) 5‘

-3‘

Dir

ect r

epea

t 5‘

-3‘

IS f

amil

y6

I (16)1 7 (359)

Chr 9 (83-287)

Chr, pPA115

hypothetical protein MC7420_6982 [Microcoleus chthonoplastes PCC 7420]

52 1152 0.4 (1.4)

tatagcagtcctaaatcattatc

gaaaatcatttaggattgctata

- -

II (13)2 5 (337-340)

Chr, pPA115

9 (50-272)

Chr, pPA14, pPA79, pPA50

ISPlag1, transposase, IS4 family protein [Synechococcus sp. PCC 7335]1

59 1305 0.7 (0.8)

caggacttacgcaggcacactatatatagtgtgcagtaagccagcgaacgctgccaat

attggtatgcgatcgcctacttttagtacgctatatatagcgtgcttgcgtaagtcctg

ctctt, tctgt, aaacg

IS701

III (5) 2 (326-461)

Chr 3 (41-142)

Chr Transposase [Nodularia spumigena CCY9414], IS605 family

77 1350 10 (10.4)

catctgggagattgaaaactcagtggctttagaccaccaga

agtagttagaatctcagtgtcttcagacctgagagtgtcaa

- IS200/IS6057

IV (16)3 6 (368-547)

Chr, pPA5.5, pPA115

10 (101-239)

Chr conserved hypothetical protein [Oscillatoria sp. PCC 6506], IS4 family

79 2058 2.1 (6.2)

aacccacattccgcagatatttaagtcaatttaattttcaactaaattgttctaaataaaagc

ggtttctattttttagaattagtggaaaatttgacccctatctgcggaatgtgggtt

- IS1634

V (5) 2 (306)

Chr 3 (161-207)

Chr DDE domain transposase [Lyngbya majuscula 3L], IS4 family

54 1151 0.1 (0.1)

cattagacatctccaaaa

tttcggagatgtctattga

- -

VI (8) 1 (433)

Chr 7 (95-187)

Chr transposase, IS605 OrfB family, central region [Lyngbya majuscula 3L]

73 1805 - aaacctgggaagttctcaacttctgtagtggtcattaactgccc

ttatggatacttgggaagaagccagcgaatatgatgcgatggactttagc

- IS200/IS6058

VII (2) 1 (418)

Chr 1 (264) Chr transposase, IS605 OrfB family [Cyanothece sp. PCC 8801]

81 1854 - ctaaagattctttggt

accacacaatctttag - IS200/IS6059


.asm.org/

Dow

nloaded from

http://aem.asm.org/

29

VIII (3) 2 (318-390)

Chr 1 (226) Chr transposase [Microcystis aeruginosa NIES-843]

78 1247 28 (26) attcacaaaaagatatactataatgtgagat

tttacggcggtgaggatgtcaa

- IS200/IS6051

0

IX (2) - - 2 (134-223)

pPA79, pPA50

transposase, IS4 family protein [Nostoc punctiforme PCC 73102]

79 - - - - - IS701

X (2) - - 2 (276) Chr, pPA79

conserved hypothetical protein [Microcoleus chthonoplastes PCC 7420]

63 - - - - - -

XI (2) - - 2 (262-269)

Chr hypothetical protein Cyan7822_0488 [Cyanothece sp. PCC 7822]

72 - - - - - -

Ungrouped

A19Y_757 1 (388)

Chr - - transposase, IS605 OrfB family [Cyanothece sp. PCC 7424]

68 1553 - aaaatatcctggttttctcactccaaaggt

aaaaattagaataatattcagtcccttttaatattg

- IS607

A19Y_919 1(460)

Chr - . Transposase, IS605 OrfB [Nodularia spumigena CCY9414]

64 1549 - aacccgtagtggggtgttcaccaaaaaacgagcgg

tcaagaatcccccgcatttatgcgtggggagtgtcaa

- IS200/IS6058

A19Y_3678 1(390)

Chr - - putative transposase [Arthrospira platensis NIES-39]

86 1701 - cgaaaaaatgggtttaaaaccccgtcgttctacgacggcttttcttgatt

agcgtcaacccgtccagcatttcaacaattgtctcctgagtttgtcga

- IS200/IS6051

1

14 (113-307)

Chr, pPA79, pPA50

Unassigned Transposases IS4, IS200/IS605, IS630

647

1 flanking the mcy gene cluster at the downstream end (21) 648 2 caused the deletion of the mcy gene cluster in nontoxic Planktothrix strains (17) 649 3 short TEs (113-212 bp) that are highly identical (81-84%) to a transposon associated with homoanatoxin-a gene cluster (37) 650 4 Full copies (including IRL and IRR) only 651 5 variability between total TEs 652


.asm.org/

Dow

nloaded from

http://aem.asm.org/

30

6 according to the IS finder database (e value 1e-30), (30) 653 7 has no tandem inverse repeats (11), left end and right end according to the closest homologue ISTosp1 from Tolypothrix (DQ257628) 654 8 Left end and right end according to the closest homologue ISSoc6 from Synechococcus sp. JA-3-3Ab (NC_007775) 655 9 closest homologue ISBce3 from Bacillus cereus (NC_004722) 656 10 Left end and right end according to the closest homologue ISTel3 from Thermosynechococcus elongatus BP-1 (NC_004113) 657 11 Left end and right end according to the closest homologue ISTel2 from Thermosynechococcus elongatus BP-1 (NC_004113) 658

659


.asm.org/

Dow

nloaded from

http://aem.asm.org/

31

Figure legends 660

Fig. 1: Picture of ethidium-bromide stained agarose gels showing PCR amplicons (ten kbp) 661

obtained from Planktothrix agardhii NIVA-CYA126/8 with primer pairs amplifying the 662

entire plasmid without interruption (pPA5.5, 14, 50, 79, 115). For clarity only the nucleotide 663

pos. of the forward primer according to Genbank Access No. ASAK01000000 are indicated 664

(see also Suppl. Table 1). M, DNA size marker in kbp (0.5 – 10 kbp). The expected PCR 665

product size is marked. A few smaller sized PCR amplification byproducts deviating from the 666

expected size represent unspecific amplification. 667

668

Fig. 2: Schematic representation of annotated plasmids occurring in Planktothrix agardhii 669

NIVA-CYA126/8 and construction of the shuttle vectors pPA5.5 and pPA14. The ORFs 670

marked in grey represent proteins putatively involved in polyketide synthesis. The ORFs 671

marked in black represent transposable elements. Small ORFs (< 650 bp) are marked as black 672

bars (transposable elements) and blue bars (other proteins). MCS, multi cloning site. 673

674

Fig. 3: Amplification of cat transposon Tn5 from Planktothrix agardhii NIVA-CYA126/8 675

transformant using pPA5.5 and pPA14 plasmid specific primer pairs (shuttle vector and 676

transformant). M, DNA size marker in kbp (0.5 – 3 kbp). 677

678

Fig. 4: (A) Absolute and (B) relative quantification (mean ± SD) of copies of Planktothrix 679

agardhii NIVA-CYA126/8 chromosome and plasmids per individual cell. 680

681

Fig. 5: Amplification of cat transposon Tn5 from Planktothrix agardhii NIVA-CYA126/8 682

transformant using pPA50 plasmid specific primer pairs. 1, 2 constitute different harvests at 4 683

µg Cm ml-1. 684

685


.asm.org/

Dow

nloaded from

http://aem.asm.org/


.asm.org/

Dow

nloaded from

http://aem.asm.org/


.asm.org/

Dow

nloaded from

http://aem.asm.org/


.asm.org/

Dow

nloaded from

http://aem.asm.org/


.asm.org/

Dow

nloaded from

http://aem.asm.org/


.asm.org/

Dow

nloaded from

http://aem.asm.org/

Copyright © 2014, American Society for Microbiology. All ...aem.asm.org/content/early/2014/06/03/AEM.01188-14.full.pdf47 Harmful algal blooms ... From the genomic DNA, ... The PCR

Documents