PreprintPDF Available

Spatial patterns in phage-Rhizobium coevolutionary interactions across regions of common bean domestication

July 2020

DOI:10.1101/2020.07.21.214783

License
CC BY-NC-ND 4.0

Authors:

Rosa Isela Santamaría

National Autonomous University of Mexico

Show all 8 authorsHide

Bacteriophages play significant roles in the composition, diversity, and evolution of bacterial communities. Despite their importance, it remains unclear how phage diversity and phage-host interactions are spatially structured. Local adaptation may play a key role. Nitrogen-fixing symbiotic bacteria, called rhizobia, have been shown to locally adapt to domesticated common bean at its Mesoamerican and Andean origin sites. This may affect phage-rhizobium interactions. However, knowledge about the diversity and coevolution of phages with their respective Rhizobium populations is lacking. Here, through the study of four phage-Rhizobium communities in Mexico and Argentina, we show that both phage and host diversity are spatially structured. Our cross-infection experiments demonstrate that phage infection rates were overall higher on sympatric than on allopatric rhizobia, except for one Argentinean community, indicating phage local adaptation and host maladaptation. Phage-host interactions were shaped by the genetic identity and geographic origin of both the phage and the host. Phages ranged from specialists to generalists, revealing a nested network of interactions. Our results suggest a key role of local adaptation to resident host bacteria communities in shaping phage genetic and phenotypic composition, following a similar spatial pattern of diversity and coevolution as the host.

Available via license: CC BY-NC-ND 4.0

Content may be subject to copyright.

Spatial patterns in phage-Rhizobium coevolutionary interactions across regions of 1

common bean domestication 2

Running title: Spatial patterns in phage-Rhizobium interactions. 3

Jannick Van Cauwenberghe 1,2 *, Rosa I. Santamaría 1, *, Patricia Bustos 1, Soledad Juárez 1, 4

Maria Antonella Ducci 3, Trinidad Figueroa Fleming 4, Angela Virginia Etcheverry 4, and 5

Víctor González 1. 6

1 Centro de Ciencias Genómicas, Universidad Nacional Autonóma de México, Mexico. 7

2 Department of Integrative Biology, University of California, Berkeley, CA, USA 8

3 Instituto Nacional de Tecnología Agropecuaria, Universidad Nacional de Salta, Argentina. 9

4 Facultad de Ciencias Naturales, Universidad Nacional de Salta, Salta, Argentina. 10

* These authors contributed equally to this work. 11

Keywords: Rhizobium, phages, local adaptation, common bean. 13

Correspondence to: 14

Jannick Van Cauwenberghe 15

jvancau@berkeley.edu 16

Víctor González 18

vgonzal@ccg.unam.mx 19

.CC-BY-NC-ND 4.0 International licensewas not certified by peer review) is the author/funder. It is made available under a The copyright holder for this preprint (whichthis version posted July 22, 2020. . https://doi.org/10.1101/2020.07.21.214783doi: bioRxiv preprint

Abstract. 23

Bacteriophages play significant roles in the composition, diversity, and evolution of bacterial 24

communities. Despite their importance, it remains unclear how phage diversity and phage-25

host interactions are spatially structured. Local adaptation may play a key role. Nitrogen-26

fixing symbiotic bacteria, known as rhizobia, have been shown to locally adapt to 27

domesticated common bean at its Mesoamerican and Andean sites of origin. This may affect 28

phage-rhizobium interactions. However, knowledge about the diversity and coevolution of 29

phages with their respective Rhizobium populations is lacking. Here, through the study of 30

four phage-Rhizobium communities in Mexico and Argentina, we show that both phage and 31

host diversity is spatially structured. Cross-infection experiments demonstrated that phage 32

infection rates were higher overall in sympatric rhizobia than in allopatric rhizobia except for 33

one Argentinean community, indicating phage local adaptation and host maladaptation. 34

Phage-host interactions were shaped by the genetic identity and geographic origin of both the 35

phage and the host. The phages ranged from specialists to generalists, revealing a nested 36

network of interactions. Our results suggest a key role of local adaptation to resident host 37

bacterial communities in shaping the phage genetic and phenotypic composition, following a 38

similar spatial pattern of diversity and coevolution to that in the host. 39

Introduction. 40

Bacterial viruses or bacteriophages are the most diverse and abundant biological entities on 41

earth [1, 2]. They play a significant role in bacterial ecology and evolution, enabling 42

horizontal gene transfer, and influencing bacterial diversity through their lytic or lysogenic 43

cycles [3–6]. Nevertheless, phage biogeography, including patterns of dispersal, 44

establishment, and community assembly, is very poorly understood [7], and its study could 45

contribute to the advancement of microbiome engineering in agricultural and medical 46

settings. 47

Most of the available knowledge about phage genetic diversity and spatiotemporal dynamics 48

comes from metagenomic studies on marine cyanophages and gut microbiomes [8–10]. These 49

studies have shown both cosmopolitan [1, 11–13] and habitat-specific phage lineages [14–50

21]. Given the predatory nature of phages, the presence or absence of suitable bacterial hosts 51

shapes their distribution [8, 22–24]. 52

Bacteriophages tend to be locally adapted to sympatric bacteria [25–28]. It is commonly 53

predicted that local adaptation is a significant underlying factor of compositional differences 54

across phage communities [16, 24, 26, 27, 29]. Local adaptation is a process that results in a 55

local population of a given species exhibiting higher fitness in its local environment than in 56

allopatric populations and takes place when environmentally driven selection is more robust 57

than migration [30, 31]. The implied lower fitness experienced by immigrants may limit gene 58

flow and increase genetic differentiation across populations (i.e., “isolation-by-adaptation,” 59

or more general “isolation-by-environment”) [32–35]. Host bacteria are a key environmental 60

driver of phage local adaptation, or more precisely, “local coadaptation”, as phage adaptation 61

(e.g., increased host range) readily invokes the counter- or coadaptation of bacteria (e.g., 62

increased resistance) [25–27, 36–38]. 63

While bacteriophages coadapt with their bacterial hosts, symbiotic bacteria coadapt with their 64

eukaryotic hosts [39–41]. For instance, it is well established that rhizobia coadapt with 65

legumes and form a mutualistic relationship in which they provide legumes with a steady 66

supply of plant-usable nitrogen via nitrogen fixation [42, 43]. Aguilar et al. [44] found that 67

Mesoamerican and Andean common bean (Phaseolus vulgaris) genotypes were preferentially 68

associated with Mesoamerican and Andean rhizobia, respectively, indicating local 69

coadaptation [44] across the two domesticated common bean gene pools [45, 46]. 70

Furthermore, the Rhizobium communities were genetically differentiated, as the relative 71

abundance of different types of the symbiotic nodC gene varied across the two bean gene 72

pools [44]. Additionally, in other legume-rhizobium systems, host legume population 73

differentiation has led to rhizobium population differentiation [47, 48] and even local 74

adaptation [49, 50]. 75

Symbiotic organisms have significant reciprocal evolutionary effects on each other, which in 76

turn affect third-party interactions [51–59]. Bacterial adaptation to plant hosts may affect 77

bacteria-phage interactions (e.g., by altering the expression of surface receptors for plant 78

interactions, which serve as anchor points of phages [60–62]). Here, we predicted that the 79

genetic differentiation and local adaptation experienced by rhizobia across common bean 80

gene pools shape the interaction and genetic differentiation of communities of phages 81

infecting common bean-nodulating rhizobia. Even under the assumption of high phage 82

dispersal capabilities, we expected communities of phages to be locally adapted, showing 83

higher infection rates for sympatric rhizobia than for allopatric rhizobia. Hence, we aimed to 84

elucidate whether the genomic identities and host ranges of phages infecting common bean-85

nodulating rhizobia are geographically structured. Our approach was to sample Rhizobium 86

strains and associated bacteriophages from two common bean fields in Mexico and 87

Argentina, corresponding to independent areas of common bean domestication [45, 46]. Host 88

species identity and bacteriophage genomic types were determined by Sanger sequencing and 89

by both Sanger sequencing and whole-genome sequencing, respectively. Host range 90

assessment results were analyzed for biogeographic signals and local adaptation. More 91

specifically, we aimed to (i) assess bacteriophage diversity associated with common bean-92

nodulating rhizobia; (ii) compare the bacteriophage community composition across common 93

bean gene pools; (iii) determine how bacteriophage host range pertains to bacteriophage 94

provenance and the host species community composition; and (iv) obtain evidence of local 95

adaptation of bacteriophages. 96

Materials and methods. 97

Soil and Rhizobium sampling. 98

The sites of rhizobia and phage sampling were common bean (Phaseolus vulgaris) 99

agriculture fields located in Mexico (Tepoztlán and Yautepec, Morelos) and Argentina 100

(Chicoana and Salta, Salta) (Fig. S1a). The bean fields were named after the municipalities in 101

which they were located except for the ‘Salta’ field, which was located in Cerrillos near the 102

border with the city of Salta. The distance between Tepoztlán and Yautepec is c. 7.4 km; the 103

distance between Salta and Chicoana is c. 20.8 km. In each bean field, we collected >1 L of 104

rhizosphere soil from three plants separated by 10 m. Each soil sample was mixed and split 105

into two aliquots, one used for phage isolation and one used to “trap” rhizobia in the root 106

nodules of bean plants. In Tepoztlán, three bean plants (P. vulgaris var. Negro Veracruz) 107

were collected directly from the field, and their nodules were used to isolate rhizobia. To trap 108

rhizobia from Yautepec, we used P. vulgaris var. Negro Veracruz, while rhizobia from 109

Argentina were caught using P. vulgaris var. Alubia Cerrillos. These cultivars belong to the 110

Mesoamerican and Andean gene pools of domesticated common bean, respectively. 111

Previously, axenically germinated seedlings were planted in presterilized vermiculite pots 112

and inoculated with a 100 mL soil sample. Nodules were harvested after eight to twelve 113

weeks of growth in a greenhouse under natural environmental conditions regarding light and 114

temperature (Supplementary Method S1). Rhizobium strains were obtained from surface-115

sterilized nodules via the squashing method and streaked on PYNal agar plates (Supplementary 116

method S2; [63]). Three subculture steps were used to purify all isolated rhizobial strains. 117

The isolates were stored at -80°C in 50% glycerol for long-term storage. 118

Isolation and purification of bacteriophages. 119

Bacteriophages were isolated from soil samples via the enrichment protocol described 120

previously [64] using both Rhizobium isolates from local sites (local collection, or LC) and 121

Rhizobium from the laboratory collection (standard collection, or SC; Table S1). Briefly, dry 122

sieved soil was suspended in PY-Nal medium at a ratio of 1:2 and incubated overnight at 123

30°C with shaking (250 rpm). Subsequently, the soil solution was centrifuged (10,000 g, 10 124

min, 4°C) to remove large particles, and chloroform was added to eliminate the remaining 125

bacteria. Chloroform was removed by centrifugation, and the solution was filtered through a 126

0.22 µm Millipore filter. The soil filtrate was used to inoculate one pair-member of each 127

Rhizobium strain. The other pair-member served as an uninoculated control. LC rhizobia 128

were inoculated with the filtrate from the soil of origin, while SC rhizobia were inoculated 129

with soil filtrate that was pooled from each of the three soil samples per bean field. Cultures 130

of 1 ml in PYNal were incubated at 30°C (250 rpm) to an optical density of 0.2 at 620 nm 131

(OD620; Beckman DU650 spectrophotometer) in 96-deep-well plates. After 20 hours of 132

incubation, the cultures were centrifuged, and the supernatant was used to reinoculate new 133

cultures of the same pairs of rhizobial strains. This enrichment process was repeated five 134

times; each time, the OD620 value was compared between the control and the inoculated pair 135

member. A decrease in the OD620 was indicative of cell lysis. The presence of phages was 136

confirmed by plaquing a dilution series of the filtrate mixed with 2.5 mL of molten soft PY 137

(0.65% agar) on lawns of rhizobia over solid PYNal plates, as in Carlson (2005)[65]. Lytic 138

plaques were picked and inoculated again in the respective bacterial cultures. Two additional 139

dilution series were performed to ensure the purity of the bacteriophage stocks. Finally, the 140

phage dilution was treated with chloroform to remove the remaining bacteria and stored at 141

4°C. 142

Phylogeny and taxonomic Rhizobium species identification. 143

Two chromosomal genes, dnaB and recA, were partially amplified by colony PCR. The 144

primers and PCR protocols used are described in Table S2. The PCR products were purified 145

using the Exo-SAP cleanup protocol [66]. The purified products were sent to Macrogen for 146

sequencing (Macrogen Inc, Seoul, Korea). Sanger sequences were edited and assembled in 147

Genius Pro v. 6.1.2. Multiple alignments were performed using MUSCLE [67], followed by 148

manual correction to remove ambiguously aligned regions. Phylogenetic trees were 149

reconstructed and edited with MEGA 7 using the maximum likelihood (ML) method based 150

on the GTR+G+ I model and 1000 bootstrap replicates. Rhizobia were clustered into 151

sequence types (STs) based on 100% sequence identity of dnaB and recA sequences. 152

Genome sequencing. 153

We obtained the genome sequences of 100 phages from the collection chosen according to 154

their host range differences. Phage genomic DNA was purified from phage stocks propagated 155

in the corresponding host Rhizobium strains. Host DNA and RNA were eliminated using 156

DNase and RNase. Subsequently, phages were precipitated using a PEG-8000/NaCl solution. 157

After centrifugation (10,000 g, 20 min, 4°C), the pellet was suspended in Tris-EDTA buffer. 158

Proteins were hydrolyzed using 4% SDS and proteinase K and precipitated by adding 3 M 159

potassium acetate. Phage genomic DNA was precipitated with 100% isopropanol and washed 160

with 70% ethanol twice. 161

Phage genome sequencing was performed with Illumina technology in a Nextgen 500 system 162

(Unidad Universitaria de Secuenciación Masiva de DNA (UUSMD-UNAM). Genomes were 163

assembled from trimmed [68] sequence reads using the Spades v. 3.13.1 [69], Velvet v. 164

1.2.10 [70] and Phred/Phrap/Consed v. 23.0 [71] software packages. 165

The remaining 96 phages were identified via PCR and Sanger sequencing of phage marker 166

genes used to distinguish between phage genomic types (PGTs). These genes were identified 167

using BPGA software to obtain common genes within the PGTs and to check their presence 168

in the remaining PGTs [72]. The extracted sequences were used to design primers with 169

Primer3 [73]. The selected genes, primers, and PCR protocol are described in Table S2. 170

Comparative genomics. 171

The average nucleotide identity (ANI) of all pairs of phage genomes was calculated with 172

pyani v.0.2.9 using the ANIm MuMmer method [74, 75]. PGTs were clustered based on 80% 173

nucleotide identity and 60% coverage of the smaller genome. The assignation of PGTs to the 174

phage morphological families of tailed bacterial viruses (Siphoviridae, Myoviridae, and 175

Podoviridae) was performed using the VirFam server (http://biodev.cea.fr/virfam/) [76]. 176

PGTs assigned to Microviridae were recognized by their short genome length and through 177

BLASTn searches against the virus database of the NCBI 178

(https://www.ncbi.nlm.nih.gov/genome/viruses/). 179

Rhizobium susceptibility and host range assessment. 180

The infectivity of all isolated phages was evaluated by the spot test procedure [77] in the 181

rhizobia isolated from all four bean fields. Although this method could overestimate the lytic 182

properties of phages and underestimate infections due to low phage titers or different 183

adsorption efficiencies [77, 78], the spotting of phages on lawns is an affordable method for 184

testing a large number of pairwise interactions [79]. Rhizobium ‘lawns’ in double-agar-layer 185

plates were spotted with 5 µl of each bacteriophage solution prepared from the respective 186

phage Rhizobium lysates. After overnight incubation at 30°C, the plates were assessed for 187

lysis. At least three replicates of each Rhizobium-phage combination were performed to 188

ensure reproducibility, and at least two replicates with the same lytic or resistant phenotype 189

were considered to indicate a positive or negative result. Spots resulting from lysis with a 190

translucent appearance rather than a transparent appearance were recorded as ‘partial lysis’ 191

but were treated equivalently to transparent spots for the statistical and BiMat analyses. The 192

binary interaction matrix is available from Github (github.com/jvancau/interactiondata). 193

With this information, a matrix was constructed based on Bray-Curtis dissimilarity calculated 194

with the vegdist function of the vegan package in R [80]. To compare the phenotypic 195

composition with the genetic composition, we clustered the bacterial hosts showing >80% 196

similarity in terms of their susceptibility to phages into Rhizobium phenotype groups (RPGs, 197

Table S3). Then, phages that infected a common range of hosts (Bray-Curtis similarity > 198

80%) formed phage phenotype groups (PPGs, Table S3). 199

Network structure of phage-bacterium interactions. 200

To analyze the modularity and nestedness properties of the phage-bacterium bipartite 201

network, we employed the BiMat program [81], which maximizes the similarities in the 202

bacteria-phage lytic interaction matrix. The program ran in the MATLAB environment and 203

was used according to the author’s start guide [81] (https://www.github.com/cesar7f/BiMat). 204

Modularity was tested using the Adaptive Brim algorithm, and nestedness was tested using 205

the NODF (Nestedness metric based on Overlap and Decreasing Fill) and NTC (Nestedness 206

Temperature Calculator) algorithms. Statistical analysis was performed using 1000 replicates 207

and the equiprobable null model. 208

Local adaptation. 209

Rhizobium susceptibility rates were calculated for each Rhizobium strain as the number of 210

phages able to infect the strain divided by the total number of tested phages. Similarly, phage 211

infection rates were calculated as the proportion of rhizobia that a given phage could infect 212

among the total number of strains tested. These values were calculated for sympatric and 213

allopatric phage-strain combinations. The statistical significance of the differences between 214

sympatric and allopatric susceptibility/infection rates was tested using a generalized linear 215

model in R with a quasi-binomial model [82]. 216

Phage local adaptation was also calculated according to Vos et al. [27] as the mean difference 217

in the mean of the rates of sympatric (S) and allopatric (A) phage infections. 218

Statistical analyses. 219

To test the significance of compositional differences in Rhizobium sequence types (STs) and 220

phage genomic types (PGTs) across common bean fields, we performed PERMANOVA tests 221

based on Jaccard and Bray-Curtis distances with 999 permutations using the ecodist and 222

vegan packages [80]. Principal coordinates of neighbor matrices (PCNM), which are 223

orthogonal spatial variables derived from a spatial distance matrix, were calculated from the 224

geographical coordinates using the ‘pcnm’ function of the vegan package for R [83]. The first 225

PCNM value was used as a proxy for spatially related variation across the two regions and 226

was fit on a principal coordinates analysis (PCoA) using envfit (‘vegan’) with 9999 227

permutations. This was approach was employed to assess the significance of the distance 228

between the two regions in explaining the compositional differences among bean fields. For 229

each distance matrix type, the correlations among all datasets of the genetic and phenotypic 230

composition across bean fields (ST, PGT, RPG and PPG) were assessed using Mantel tests 231

(999 permutations, vegan package). 232

A Mantel test was also used to correlate Rhizobium and phage genetic distances with the 233

corresponding susceptibility range and host range distances across all isolated rhizobia and 234

phages. PERMANOVAs were used to test the significance of the effects of the origin and 235

genetic or taxonomic identity of rhizobia and phages on their susceptibility range and host 236

range, respectively. 237

Phage and Rhizobium nucleotide accessions. 238

Phage genome sequences are available from GenBank with IDs MN988459 to MN988558. 239

Rhizobium nucleotide sequences are available under: MT756388 – MT756428 (dnaB) and 240

MT756429 – MT756469 (recA). 241

Results. 242

Rhizobium and phage community sample composition. 243

We studied 229 Rhizobium strains isolated from four agricultural plots in Central Mexico 244

(Tepoztlán and Yautepec) and Northwest Argentina (Salta and Chicoana) (Fig. S1a-b). The 245

rhizobial strains from each site (LC, local collections) were employed to trap phages from the 246

soil of the same locality by the enrichment method [64]. In parallel, the same soils were 247

pooled for each plot and used to search for phages using a standard collection (SC) of 94 248

Rhizobium strains of diverse geographic origins maintained in our laboratory (Table S1). A 249

total of 196 phages were obtained with this protocol, 110 from LC and 86 from SC (Fig. 250

S1b). 251

Genetic differentiation of Rhizobium populations between agricultural fields. 252

Our first aim was to investigate whether the collected Rhizobium strains represent 253

geographically structured populations. Phylogenetic analysis of the partial recA-dnaB 254

sequences of 229 rhizobial strains identified R. etli as the predominant species at Mexican 255

sampling sites (74.6%), while R. phaseoli was dominant in Argentina (80%) (Fig. 1; Fig. 256

S2a). According to the nucleotide variations in recA and dnaB, all of the isolated rhizobia 257

were grouped into 41 chromosomal sequence types (STs) (Table S4). PERMANOVA tests 258

employing two distance measures (Jaccard and Bray-Curtis) indicated that the rhizobial 259

communities differed significantly in terms of genetic composition and the relative 260

abundance of genotypes (STs) among different bean fields and regions (Fig. 2a-c; Table S4-261

S5). The geographic distance between the regions, represented by a PCNM vector, was 262

correlated with the ST composition among the rhizobia isolated from the four common bean 263

fields (Table S5). At the species level, R. etli and R. phaseoli STs differed significantly 264

among the sites of origin of the common bean fields. Between regions, only the R. etli 265

populations differed in terms of the ST composition, whereas the R. phaseoli ST composition 266

changed only marginally across regions (Table S5). The abundance of ST5 in R. phaseoli, the 267

only ST of the 41 total STs found across the four common bean fields, might explain this last 268

result (Fig. 2c; Table S4). 269

Diversity of phage communities. 270

To assess phage diversity, we obtained the complete genome sequences of 100 out of 196 271

phages. The phage genomes displayed a wide range of lengths (from 4.8 to 207.6 kb; median 272

54.4) and GC contents (from 41% to 61%; median 57%). They were clustered into 29 PGTs 273

(defined by ANIm, see Methods), 18 of which had two or more individual genomes, and 11 274

were singletons. Within PGTs, the genomes exhibited nucleotide variation ranging from 85.8 275

to 99.7%, with a coverage of approximately 64.1 to 100% (Fig. S3). Additionally, 90 phages 276

among the remaining 96 phages were assigned to PGTs by the PCR identification of specific 277

phage genes conserved within the members of the 29 PGTs, followed by Sanger sequencing 278

(see Methods) (Table S2). 279

Phages were also classified into morphological families using VirFam predictions [76]. Most 280

of them (26/29) belonged to the order Caudovirales, represented by the families Podoviridae 281

(12 PGTs), Siphoviridae (8 PGTs), and Myoviridae (6 PGTs). Three PGTs were identified as 282

members of the Microviridae family. 283

Most phage families were present at the four sampling sites, but the Salta community was 284

dominated by Myoviridae (69%) and the Chicoana community by Siphoviridae (62%) (Fig. 285

S2b; Table S6). Remarkably, the Microviridae family (F02), defined by small-genome phages 286

(4.8 – 6.2 kb), was dominant in Tepoztlán (60%), whereas it showed low abundance or was 287

absent in the other populations (0-23%; Table S6). 288

Phage population differentiation across agricultural sites. 289

Following the differences in the rhizobial ST composition per sampling site, we found that 290

the composition of PGTs also differed significantly among common bean fields and regions 291

(Fig. 2 d-f, Table S6). The differences were significantly correlated with geographic distance 292

(Table S5). The phage communities were also significantly different between the Mexican 293

bean fields within regions based on Jaccard distances (F1,5 = 3.744, P = 0.024), but they were 294

not significantly different between Mexican bean fields based on Bray-Curtis distances (F1,5 295

= 3.679, P = 0.055) or between Argentinian bean fields (Jaccard: F1,5 = 1.132, P = 0.384; 296

Bray-Curtis: F1,5 = 2.204, P = 0.149). Mantel tests showed that the genetic composition 297

differences among phage communities were significantly correlated with the differences in 298

the Rhizobium community genetic composition among bean fields (Table S5). Some PGTs 299

coexisted at two of the sampling sites, whereas PGTs rarely coexisted at three sites and never 300

at four (Fig. 2f). Moreover, 52% of the 29 PGTs occurred solely in one bean field, and 69% 301

were restricted to a particular region. A spatial pattern distinction was also shown by the ANI 302

values, as the average ANI of allopatric phages belonging to the same PGTs was 88%. In 303

comparison, the average ANI of sympatric phages belonging to the same PGT was 96%. 304

Structure of the phage-bacterium interaction network. 305

To define the rhizobium-phage interactions within and between the four communities, we 306

tested the infectivity of 196 phages against 229 Rhizobium strains by the spot assay (see 307

methods). We registered the following phenotypes: complete lysis (transparent spots), partial 308

lysis (translucent spots), and resistance (absence of lysis), in three independent experiments. 309

A total of 44,884 interactions were examined in triplicate experiments; 19,474 plaques 310

showed full or partial lytic phenotypes, recorded as positive interactions (1 in the bipartite 311

network of Fig. 3) [81, 84]. The rhizobium-resistant phenotypes were recorded as 0 in the 312

corresponding binary matrix (Fig. 3). 313

Overall, the BiMat network showed high connectance (0.43), indicating that there were 314

approximately 4/10 effective phage-bacterium interactions in the community, and weak 315

modular differentiation (Qb = 0.21; Fig. 3a). Three large modules, with high internal 316

connectance in comparison with the outside modules, were detected (Fig. 3a). Phages from 317

Mexico (Tepoztlán and Yautepec) generally interacted best with Rhizobium isolates from 318

Mexico. It was less common for these phages to infect Rhizobium from Argentina. These 319

results suggest large-scale modularity dominated by sympatric interactions. However, some 320

exceptions were observed; for instance, phages from Chicoana were very promiscuous, since 321

a significant number of Mexican strains were cross-infected by these Argentinian phages 322

(Fig. 3a). This may in part explain the overall nested structure shown by the BiMat matrix, 323

quantified by the independent NODF (0.68) and NTC (0.73) algorithms. This suggests that 324

the phage communities consist of a range of specialist to generalist phages (Flores et al. 2013; 325

Fig. 3b). 326

Rhizobium susceptibility and phage host range. 327

Although all Rhizobium strains were closely related in the recA-dnaB phylogenetic tree, they 328

varied widely in their susceptibility range, with some rhizobia being infected by 329

approximately 10.2% to 74.0% of the phages (average rate of infection = 43.4%). Hence, we 330

sought to investigate whether the susceptibility of rhizobia was associated with geographic 331

origin, taxonomic affiliation or genetic identity. When we compared the variation in the 332

susceptibility range between individual rhizobia, we found that the susceptibility of rhizobia 333

was significantly different not only between species (F1,217=10.296, P=0.001) (Fig. 4a) but 334

also among bean fields (F3,217=30.964, P=0.001), and we observed the interaction of the two 335

factors (F3,217=6.3575, P=0.001; Fig. 4a). The last finding indicates that the effect of 336

Rhizobium species identity on the susceptibility range depends on the geographic origin of 337

the rhizobia. When species were split into STs, the Rhizobium susceptibility range was found 338

to differ significantly among bean fields (F3,178= 53.458, P=0.001) and STs (F40,178=6.056, 339

P=0.001), although their interaction was not significant (F6,178=0.955, P=0.542) due to the 340

limited geographic spread of most STs. Susceptible strains were clustered in 48 RPGs 341

(Rhizobium Phenotype Groups; see methods). Twenty-nine RPGs were singletons, whereas 342

19 were formed by more than two strains (maximum of 40) (Table S3). Thirty-two RPGs 343

belonged to R. etli, twelve RPGs corresponded to R. phaseoli (RPG4, 13, and 19), and the 344

other four RPGs belonged to both species (RPG1, 2, 6, and 11). The most abundant RPGs (1 345

to 4) were composed of the frequently identified STs of R. phaseoli (ST5) and R. etli (ST10 346

and ST34). We found that the RPG composition was strongly correlated with the ST 347

composition (Table S7). Additionally, Rhizobium genetic distance and susceptibility range 348

similarity were significantly correlated (r= 0.3125, P= 0.001). 349

We examined the host range of phages to identify those with similar infection spectra. Phage 350

infection rates, or the proportion of hosts that a given phage isolate could infect, varied 351

considerably from 2.2% to 92.6% (average infection rate 43.4%). All but three phages were 352

able to infect both R. etli and R. phaseoli; however, the phages were able to infect only 51% 353

of STs on average. The host range of the phages was significantly affected by the bean field 354

of origin (F3,142=22.938, P=0.001), phage genomic type (PGT; F27,142=6.930, P=0.001), and 355

the interplay between these factors (F14,142=2.882, P=0.001). The last finding indicates that 356

the effect of PGT on the host range depends on the geographic origin of the phage. Similar 357

results were obtained when the phages were grouped by taxonomic family: phage host range 358

was significantly affected by the bean field of origin (F3,172=14.158, P=0.001), family 359

(F3,172=8.355, P=0.001), and the interplay between them (F8,172=3.872, P=0.001) (Fig. 4b). 360

Based on Bray-Curtis dissimilarity < 20%, 139 phages (70% of the total) were clustered into 361

24 PPGs; the remaining phages (57 phages) exhibited a unique PPG. Within the PPGs, two 362

profiles accounted for 53% of the phages, whereas 22 profiles exhibited fewer than ten 363

phages from similar PPGs. The PPG composition across common bean fields was 364

significantly correlated with the RPG composition based on Bray-Curtis distances, but not 365

based on Jaccard distances (Table S7). The PPG composition was also significantly 366

correlated with PGT composition (Table S7). Accordingly, host range similarity was 367

considerably correlated with phage ANI (r= 0.2890, P= 0.001). 368

Rhizobium-phage local adaptation. 369

The above results suggest that phage communities are adapted to Rhizobium in their areas of 370

origin. To examine this issue, we estimated the infectivity rates of phage isolates and the 371

susceptibility rates of Rhizobium isolates in sympatric and allopatric combinations. Despite 372

ample variation, phage infection rates were significantly greater for sympatric infections 373

(mean = 0.55 ± 0.01; CV = 55%; σ2 =k 0.092) than for allopatric infections (mean = 0.39 ± 374

0.02; CV = 64%; σ2 = 0.062; F1,390 = 23.234, P = 2.06e-6) (Fig. 5a), suggesting a trend of 375

local adaptation. This was true for both phages isolated using the standard collection (SC 376

phages; F1,170= 5.656, P= 0.0185) and phages isolated using local rhizobia (LC phages; 377

F1,218= 20.133, P= 1.17e-05). However, sympatric phage infection rates were significantly 378

higher than allopatric infection rates for Mexican communities (0.52 ± 0.02 versus 0.35 ± 379

0.02; F1,252= 30.210, P = 9.49e-08; SC: F1,94= 7.053, P= 9.30e-03; LC: F1,156= 26.405, P= 380

8.17e-07), but the difference was not statistically significant for Argentinean communities 381

(0.60 ± 0.04 versus 0.47 ± 0.04; F1,136= 3.542, P = 0.062; SC: F1,74= 0.989, P= 0.323; LC: 382

F1,60= 2.730, P= 0.104) (Fig. 5b). This was largely due to the fact that the phages isolated 383

from Chicoana (Argentina) showed no differences between the rates of infection of local 384

rhizobia and those of nonlocal rhizobia isolated from the other three sites (Fig. 5e). Although 385

the Mexican phages were locally adapted overall, phage cross-infection was detected at 386

similar rates between the communities of Tepoztlán and Yautepec (Mexico), indicating 387

nonadaptive phage differentiation between these communities (Fig. 5e). Indeed, the infection 388

rates of Mexican phages in Argentinian rhizobia were significantly lower, suggesting an 389

effect of geographic distance on local adaptation (Fig. 5b). 390

On average, the rhizobia were more susceptible to sympatric phages (mean 0.54 ± 0.01; CV= 391

36.9%; σ2 = 0.039) than to allopatric phages (mean 0.40 ± 0.01; CV= 49.7%; σ2 = 0.040), 392

indicating that the bacterial populations were maladapted to the local phages (F1,456= 46.652, 393

P= 2.72 e-11; Fig. 5c). Similar results were obtained regardless of whether susceptibility rates 394

were calculated for each Rhizobium isolate using only infections by phages isolated by using 395

the standard collection (SC phages; F1,456= 38.942, P= 9.99e-10) or phages isolated by using 396

local rhizobia (LC phages; F1,456= 45.901, P= 3.85e-11). All of these results suggest that 397

rhizobia are maladapted to coexisting phages. However, the high susceptibility of rhizobia to 398

local phages appeared to hold only for Argentinian Rhizobium communities (0.57 ± 0.02 399

versus 0.30 ± 0.02; F1,212= 64,277, P= 7.20e-14; SC: F1,212= 102.51, P< 2.2e-16; LC: F1,212= 400

42.256, P= 5.64E-10), since the Mexican Rhizobium isolates showed no significant 401

differences in susceptibility to sympatric and allopatric phage infection (0.51 ± 0.02 versus 402

0.49 ± 0.01; F1,242= 1.232, P= 0.268; SC: F1,242= 0.440, P= 0.508; LC: F1,242= 1.999, P= 403

0.159) (Fig. 5d; Fig. 5f). When we considered the susceptibility of Rhizobium to sympatric 404

phages or allopatric phages by site, the Chicoana Rhizobium community was found to be very 405

susceptible to its own phages but resistant to phages from the other sites (Fig. 5f). In contrast, 406

the rhizobial strains from Tepoztlán, Yautepec, and Salta were very susceptible to Chicoana 407

phages (Fig. 5f). 408

In most cases, the difference between the mean sympatric rates of phage infection (S) and the 409

mean allopatric rates of infection (A) indicated phage local adaptation in the four rhizobium-410

phage communities (Fig. 5g). In contrast, resident rhizobia were more susceptible to 411

sympatric than to allopatric phages, except for the rhizobia from Yautepec that were also 412

efficiently infected by Tepoztlán phages (Fig. 5h). The four populations of rhizobia were 413

susceptible to the Chicoana phages, but the Chicoana rhizobia were mostly resistant to 414

allopatric phages (Fig. 5h). 415

Discussion. 416

Overall, our data indicate that the genetic and phenotypic diversity of phages and their 417

Rhizobium hosts is spatially structured and that phages are adapted to their local host 418

communities. Previous research has shown that rhizobia are spatially structured and can 419

locally adapt to their legume hosts and other local environmental factors [49, 50, 85, 86]. For 420

instance, across the areas of common bean domestication, rhizobia receive a greater 421

competitive benefit when nodulating sympatric common beans (Phaseolus vulgaris) than 422

nonnative common bean varieties [44]. In turn, our results show that sympatric phage 423

communities have locally adapted to these rhizobia. 424

We found that each of the four phage communities analyzed was dominated by a particular 425

taxonomic family, prominently Microviridae in Tepoztlán, Myoviridae in Salta, and 426

Siphoviridae in Chicoana. Moreover, phage genome sequencing revealed high genetic 427

diversity within taxonomic families. Phage genetic diversity varied considerably within and 428

between communities, and phages were clustered into phage genomic types (PGTs). 429

Approximately 52% of the 29 PGTs occurred solely in a single phage community. Similarly, 430

Scola et al. [87] found that 66.4% of Namib Desert soil phage OTUs were exclusive to a 431

single sampling site. Other PGTs (31%) occurred in both Mexican and Argentinian bean 432

fields, with an average ANI of 88% across regions. Moreover, phages of the same PGT 433

exhibited an 8% higher ANI within regions than between regions on average. Highly similar 434

phages have previously been found across multiple distant aquatic ecosystems around the 435

world, in human virome samples [88, 89] and even in more distinct ecosystems [11]. The 436

results indicate an emerging pattern in which a higher fraction of phage community members 437

present a limited geographic range, while a significant minority of relatively closely related 438

phages are distributed globally [16, 87, 90]. It remains unclear to what extent such 439

observations are due to dispersal limitation [8]. The Rhizobium communities also showed 440

spatial structure. R. etli was the predominant species nodulating common bean in the 441

analyzed agricultural fields in Mexico, and R. phaseoli predominated in the Argentinian 442

fields, although both species were mainly characterized by spatially restricted STs. 443

Phage community (PGT) spatial patterns were correlated with the compositional differences 444

among Rhizobium (ST) communities. Similar correlations between host-phage communities 445

have been seen in aquatic systems [16, 17, 22, 91–93], which seems to verify the common 446

assumption that the relative abundance of phages within a community depends largely on 447

host abundance and susceptibility [8, 23]. However, the presence of susceptible rhizobium 448

lineages did not imply the presence of specific phages or vice versa. For example, the 449

omnipresent ST-5 was susceptible to all but two members of the phage lineages, but most 450

phage lineages were spatially limited. Similarly, F06 phages could infect members of all STs, 451

yet their presence was limited to Mexican bean fields. Although our study provides detailed 452

genetic information on phages, the Rhizobium lineages were broadly defined by 453

housekeeping gene markers (recA, dnaB). It is expected that much of the still-uncharacterized 454

phenotypic and genetic microdiversity within bacterial species [94] may explain the spatial 455

heterogeneity of the phage community composition and host-range patterns better than the 456

presence or absence of a suitable host (defined by species or ST). 457

Host range breadth varied considerably among phages, from generalists to specialists, 458

resulting in a nested structure of the inferred whole cross-interaction network. Based on the 459

extensive analysis of experimental cross-infection data, Flores et al. [95] concluded that 460

nestedness is the characteristic profile in most cases. A modular network structure may be 461

significant at large phylogenetic scales [84], but genotype-to-genotype interactions are most 462

frequent within narrow phylogenetic ranges and result from coevolutionary processes of 463

susceptibility and resistance. High host genetic similarity may underlie the nested structure of 464

our network. 465

Nevertheless, we found that phage-rhizobium interactions were significantly affected by the 466

genetic identity of both phages and rhizobia as well as their geographic origin. This may 467

explain the detection of three large modules with high internal connectivity and suggests 468

ongoing local adaptation. Indeed, we showed that phages infecting common bean-nodulating 469

rhizobia experienced higher infection rates in sympatric rhizobia than in allopatric rhizobia; 470

hence, they were locally adapted. Furthermore, sympatric phages showed more similar host 471

ranges than allopatric phages, and sympatric rhizobia shared similar susceptibility ranges. 472

Only a few field studies have provided evidence that phages locally adapt to their bacterial 473

hosts in nature [27, 84, 96]. Although the Rhizobium communities were generally maladapted 474

to the local phages, they may be adapted to their local environment as a whole. A nodule may 475

provide an isolated niche for rhizobia where they may survive competition with and the 476

antagonistic effects of other bacteria or, more directly, phages. In free-living conditions, 477

depredation by phages may change the population structure of rhizobia. Phages are usually 478

considered to be slightly ahead of their host in their coevolutionary arms race due to the 479

higher selective pressure they experience and their greater evolutionary potential [27, 97]. 480

Although local bacterial adaptation to phages has been described multiple times in 481

coevolutionary in vitro experiments [36, 98, 99], evidence in nature appears to be lacking 482

[26]. This discrepancy is probably due to the relatively high availability of resources to hosts 483

in vitro, which sways the arms race to the benefit of the host [100]. 484

In our model, the degree of local adaptation was spatially inconsistent. Argentinean phages 485

(mainly from Chicoana) infected approximately as many local as nonlocal rhizobia, while 486

Mexican phages were more infectious in local rhizobia than in nonlocal rhizobia. Spatial 487

asymmetry in phage local adaptation is believed to be the result of the effects of nutrients on 488

phage-host encounter rates, mutation rates and the cost of resistance [38, 101] or the local 489

mode of coevolutionary dynamics (i.e., arms race or fluctuating selection [102]). Although 490

we did not detect local phage maladaptation and we assumed that the differences in 491

productivity across the sampled bean fields were insignificant, these studies show how 492

environmental differences create spatially different intensities of phage local adaption. 493

Indeed, the spatial heterogeneity of environmental factors results in a geographic mosaic of 494

different evolutionary pressures [103]. Local adaptation to various environmental conditions 495

can undermine the colonization success of allopatric individuals and limit gene flow (i.e., 496

“isolation-by-adaptation,” or more general “isolation-by-environment” [33, 35, 101, 103]). 497

Zhang & Buckling [34] found that host bacteria grown in the presence of phages in 498

heterogeneous environments were limited in their ability to migrate across environments as a 499

result of maladaptation. The limiting effect of local adaptation on phage migration has not 500

been tested explicitly, although it has long been predicted [24, 101]. 501

The spatial structure in the genetic composition of phage communities is probably due to the 502

interplay of a variety of factors (e.g., historical contingencies, abiotic selection, genetic drift, 503

and, potentially, dispersal limitation [8, 15]. Our results indicate that the presence of suitable 504

hosts may play a role in shaping phage biogeography and that suitability is determined not 505

only by the genetic identity of the host but also by local adaptation. The spatial patterns are 506

analogous to those observed in Rhizobium-common bean interactions and suggest that the 507

local adaptation of rhizobia to common bean may have shaped the spatial differences in the 508

phage-rhizobium interactions. Through isolation-by-adaptation, local adaptation may 509

reinforce spatial patterns in the phage community composition. Strong local adaptation of 510

phages has been found across much shorter distances than in our present study [27, 96], and it 511

is as yet unclear to what extent phage local adaptation leads to limited migration and at which 512

scale this may occur. At smaller scales, spatial heterogeneity is probably under greater 513

pressure due to higher viral migrant densities. However, across broad scales, local adaptation 514

may be a significant barrier to successful long-distance phage migration. 515

Acknowledgments. 516

PAPIIT-UNAM IN209817 and the CCG-UNAM budget to VG funded the work. JVC 517

received a Postdoctoral Scholarship from DGAPA-UNAM (2016-2018) and was partly 518

supported by NSF grants DEB-1457508 and IOS-1759048, both awarded to E.L. Simms. We 519

thank Gabriela Guerrero, José Espíritu, Alfredo Hernández, and Víctor del Moral for 520

bioinformatics support, Alfonso Leija and Georgina Hernández for their help in greenhouse 521

experiments, and Mario Marquina for providing access to the agricultural fields at Tepoztlán 522

(Finca Xochitlamila) and Yautepec. We thank Ellen Simms, Olga María Pérez Carrascal, and 523

Ellie Harrison for comments on previous versions of the manuscript. Special thanks go to 524

Joshua Weitz for his advice on the BiMat application. 525

Conflict of interest. 526

The authors declare that they have no conflicts of interest. 527

528

References. 529

1. Breitbart M, Rohwer F. Here a virus, there a virus, everywhere the same virus? Trends 530

Microbiol 2005; 13: 278–284. 531

2. Hatfull GF. Dark matter of the biosphere: the amazing world of bacteriophage 532

diversity. J Virol 2015; 89: 8107–8110. 533

3. Bouvier T, Del Giorgio PA. Key role of selective viral-induced mortality in 534

determining marine bacterial community composition. Environ Microbiol 2007; 9: 535

287–297. 536

4. Canchaya C, Fournous G, Chibani-Chennoufi S, Dillmann ML, Brüssow H. Phage as 537

agents of lateral gene transfer. Curr Opin Microbiol 2003; 6: 417–424. 538

5. Howard-Varona C, Hargreaves KR, Solonenko NE, Markillie LM, White RA, Brewer 539

HM, et al. Multiple mechanisms drive phage infection efficiency in nearly identical 540

hosts. ISME J 2018; 12: 1605–1618. 541

6. Weinbauer MG, Rassoulzadegan F. Are viruses driving microbial diversification and 542

diversity? Environ Microbiol 2004; 6: 1–11. 543

7. Thurber RV. Current insights into phage biodiversity and biogeography. Curr Opin 544

Microbiol 2009; 12: 582–587. 545

8. Chow C-ET, Suttle CA. Biogeography of viruses in the sea. Annu Rev Virol 2015; 2: 546

41–66. 547

9. Roux S, Brum JR, Dutilh BE, Sunagawa S, Duhaime MB, Loy A, et al. Ecogenomics 548

and potential biogeochemical impacts of globally abundant ocean viruses. Nature 549

2016; 537: 689–693. 550

10. Shkoporov AN, Khokhlova E V, Fitzgerald CB, Stockdale SR, Draper LA, Ross RP, et 551

al. ΦCrAss001 represents the most abundant bacteriophage family in the human gut 552

and infects Bacteroides intestinalis. Nat Commun 2018; 9: 4781. 553

11. Breitbart M, Miyake JH, Rohwer F. Global distribution of nearly identical phage-554

encoded DNA sequences. FEMS Microbiol Lett 2004; 236: 249–256. 555

12. Dutilh BE, Cassman N, McNair K, Sanchez SE, Silva GGZ, Boling L, et al. A highly 556

abundant bacteriophage discovered in the unknown sequences of human faecal 557

metagenomes. Nat Commun 2014; 5: 4498. 558

13. Jameson E, Mann NH, Joint I, Sambles C, Mühling M. The diversity of 559

cyanomyovirus populations along a North-South Atlantic Ocean transect. ISME J 560

2011; 5: 1713–1721. 561

14. Delong EF, Preston CM, Mincer T, Rich V, Hallam SJ, Frigaard N, et al. Community 562

genomics among stratified microbial assemblages in the ocean’s interior. Science (80- 563

) 2006; 311: 496–503. 564

15. Finke JF, Suttle CA. The environment and cyanophage diversity: Insights from 565

environmental sequencing of DNA polymerase. Front Microbiol 2019; 10: 167. 566

16. Hanson CA, Marston MF, Martiny JB. Biogeographic variation in host range 567

phenotypes and taxonomic composition of marine cyanophage isolates. Front 568

Microbiol 2016; 7: 983. 569

17. Huang S, Zhang S, Jiao N, Chen F. Marine cyanophages demonstrate biogeographic 570

patterns throughout the global ocean. Appl Environ Microbiol 2015; 81: 441–452. 571

18. Marston MF, Taylor S, Sme N, Parsons RJ, Noyes TJE, Martiny JBH. Marine 572

cyanophages exhibit local and regional biogeography. Environ Microbiol 2013; 15: 573

1452–1463. 574

19. Paez-Espino D, Eloe-Fadrosh EA, Pavlopoulos GA, Thomas AD, Huntemann M, 575

Mikhailova N, et al. Uncovering Earth’s virome. Nature 2016; 536: 425–430. 576

20. Winter C, Matthews B, Suttle CA. Effects of environmental variation and spatial 577

distance on bacteria, archaea and viruses in sub-polar and arctic waters. ISME J 2013; 578

7: 1507–1518. 579

21. Luo E, Aylward FO, Mende DR, Delong EF. Bacteriophage distributions and temporal 580

variability in the ocean’s interior. MBio 2017; 8: e01903-17. 581

22. Brum JR, Ignacio-espinoza JC, Roux S, Doulcier G, Acinas SG, Alberti A, et al. 582

Patterns and ecological drivers of ocean viral communities. Science (80- ) 2015; 348: 583

1261498-1–11. 584

23. Dennehy JJ. What Ecologists Can Tell Virologists. Annu Rev Microbiol 2014; 68: 585

117–135. 586

24. Held NL, Whitaker RJ. Viral biogeography revealed by signatures in Sulfolobus 587

islandicus genomes. Environ Microbiol 2009; 11: 457–466. 588

25. Ashby B, Boots M. Multi-mode fluctuating selection in host–parasite coevolution. 589

Ecol Lett 2017; 20: 357–365. 590

26. Koskella B, Brockhurst MA. Bacteria-phage coevolution as a driver of ecological and 591

evolutionary processes in microbial communities. FEMS Microbiol Rev 2014; 38: 592

916–931. 593

27. Vos M, Birkett PJ, Birch E, Griffiths RI, Buckling A. Local adaptation of 594

bacteriophages to their bacterial hosts in soil. Science (80- ) 2009; 325: 833. 595

28. Gomez P, Buckling A. Coevolution with phages does not influence the evolution of 596

bacterial mutation rates in soil. Isme J 2013; 7: 2242–2244. 597

29. Kraemer SA, Boynton PJ. Evidence for microbial local adaptation in nature. Mol Ecol 598

2017; 26: 1860–1876. 599

30. Kawecki T, Ebert D. Conceptual issues in local adaptation. Ecol Lett 2004; 7: 1225–600

1241. 601

31. Lenormand T. Gene flow and the limits to natural selection. Trends Ecol Evol 2002; 602

17: 183–189. 603

32. Nosil P, Egan SP, Funk DJ. Heterogeneous Genomic Differentiation between 604

Walking-Stick Ecotypes : " Isolation by Adaptation " and Multiple Roles for Divergent 605

Selection. Evolution (N Y) 2008; 62: 316–336. 606

33. Orsini L, Vanoverbeke J, Swillen I, Mergeay J, De Meester L. Drivers of population 607

genetic differentiation in the wild: Isolation by dispersal limitation, isolation by 608

adaptation and isolation by colonization. Mol Ecol 2013; 22: 5983–5999. 609

34. Zhang Q-G, Buckling A. Migration highways and migration barriers created by host–610

parasite interactions. Ecol Lett 2016; 19: 1479–1485. 611

35. Wang IJ, Bradburd GS. Isolation by Environment. Mol Ecol 2014; 23: 5649–5662. 612

36. Buckling A, Rainey PB. Antagonistic coevolution between a bacterium and a 613

bacteriophage. Proc Biol Sci 2002; 269: 931–6. 614

37. Kunin V, He S, Warnecke F, Peterson SB, Garcia Martin H, Haynes M, et al. A 615

bacterial metapopulation adapts locally to phage predation despite global dispersal. 616

Genome Res 2008; 18: 293–7. 617

38. Lopez Pascua L, Gandon S, Buckling A. Abiotic heterogeneity drives parasite local 618

adaptation in coevolving bacteria and phages. J Evol Biol 2012; 25: 187–195. 619

39. Baumann P. Biology of endosymbionts of plant sap-sucking insects. Annu Rev 620

Microbiol 2005; 59: 155–89. 621

40. Levy A, Gonzalez IS, Mittelviefhaus M, Clingenpeel S, Paredes SH, Miao J, et al. 622

Genomic features of bacterial adaptation to plants. Nat Genet 2018; 50: 138–150. 623

41. Bäckhed F, Ley RE, Sonnenburg JL, Peterson DA, Gordon JI. Host-bacterial 624

mutualism in the human intestine. Science (80- ) 2005; 307: 1915–1920. 625

42. Heath KD, Tiffin P. Context dependence in the coevolution of plant and rhizobial 626

mutualists. Proc Biol Sci 2007; 274: 1905–12. 627

43. Koch M, Delmotte N, Rehrauer H, Vorholt JA, Pessi G, Hennecke H. Rhizobial 628

adaptation to hosts, a new facet in the legume root-nodule symbiosis. 2010; 23: 784–629

790. 630

44. Aguilar OM, Riva O, Peltzer E. Analysis of Rhizobium etli and of its symbiosis with 631

wild Phaseolus vulgaris supports coevolution in centers of host diversification. Proc 632

Natl Acad Sci U S A 2004; 101: 13548–53. 633

45. Bitocchi E, Bellucci E, Giardini A, Rau D, Rodriguez M, Biagetti E, et al. Molecular 634

analysis of the parallel domestication of the common bean (Phaseolus vulgaris) in 635

Mesoamerica and the Andes. New Phytol 2013; 197: 300–313. 636

46. Koenig R, Gepts P. Allozyme diversity in wild Phaseolus vulgaris: further evidence for 637

two major centers of genetic diversity. Theor Appl Genet 1989; 78: 809–817. 638

47. Melkonian R, Moulin L, Béna G, Tisseyre P, Chaintreuil C, Heulin K, et al. The 639

geographical patterns of symbiont diversity in the invasive legume Mimosa pudica can 640

be explained by the competitiveness of its symbionts and by the host genotype. 641

Environ Microbiol 2014; 16: 2099–111. 642

48. Tian CF, Young JPW, Wang ET, Tamimi SM, Chen WX. Population mixing of 643

Rhizobium leguminosarum bv. viciae nodulating Vicia faba: the role of recombination 644

and lateral gene transfer. FEMS Microbiol Ecol 2010; 73: 563–76. 645

49. Burdon JJ, Thrall PH. Spatial and temporal patterns in coevolving plant and pathogen 646

associations. Am Nat 1999; 153: S15–S33. 647

50. Van Cauwenberghe J, Visch W, Michiels J, Honnay O. Selection mosaics differentiate 648

Rhizobium-host plant interactions across nitrogen environments. Oikos 2016. 649

51. Guimarães PR, Pires MM, Jordano P, Bascompte J, Thompson JN. Indirect effects 650

drive coevolution in mutualistic networks. Nature 2017; 550: 511–514. 651

52. Heath KD, Lau JA. Herbivores alter the fitness benefits of a plant–rhizobium 652

mutualism. Acta Oecologica 2011; 37: 87–92. 653

53. Rogers HS, Buhle ER, HilleRisLambers J, Fricke EC, Miller RH, Tewksbury JJ. 654

Effects of an invasive predator cascade to plants via mutualism disruption. Nat 655

Commun 2017; 8: 6–13. 656

54. Delmas E, Besson M, Brice MH, Burkle LA, Dalla Riva G V., Fortin MJ, et al. 657

Analysing ecological networks of species interactions. Biol Rev 2019; 94: 16–36. 658

55. Gaiarsa MP, Guimarães PR. Interaction strength promotes robustness against 659

cascading effects in mutualistic networks. Sci Rep 2019; 9: 1–7. 660

56. Sih A, Crowley P, McPeek M, Petranka J, Strohmeier K. Predation, competition, and 661

prey communities: a review of field experiments. Annu Rev Ecol Syst 1985; 16: 269–662

311. 663

57. Parratt SR, Barrès B, Penczykowski RM, Laine AL. Local adaptation at higher trophic 664

levels: contrasting hyperparasite–pathogen infection dynamics in the field and 665

laboratory. Mol Ecol 2017; 26: 1964–1979. 666

58. Hatcher MJ, Dick JTA, Dunn AM. How parasites affect interactions between 667

competitors and predators. Ecol Lett 2006; 9: 1253–1271. 668

59. Hutchinson MC, Bramon Mora B, Pilosof S, Barner AK, Kéfi S, Thébault E, et al. 669

Seeing the forest for the trees: Putting multilayer networks to work for community 670

ecology. Funct Ecol 2019; 33: 206–217. 671

60. Koskella B, Taylor TB. Multifaceted impacts of bacteriophages in the plant 672

microbiome. Annu Rev Phytopathol 2018; 56: 361–380. 673

61. Labrie SJ, Samson JE, Moineau S. Bacteriophage resistance mechanisms. Nat Rev 674

Microbiol 2010; 8: 317–327. 675

62. Evans TJ, Ind A, Komitopoulou E, Salmond GPC. Phage-selected lipopolysaccharide 676

mutants of Pectobacterium atrosepticum exhibit different impacts on virulence. J Appl 677

Microbiol 2010; 109: 505–514. 678

63. Perez Carrascal OM, Vaninsberghe D, Juárez S, Polz MF. Population genomics of the 679

symbiotic plasmids of sympatric nitrogen-fixing Rhizobium species associated with 680

Phaseolus vulgaris. Environ Microbiol 2016; 18: 2660–2676. 681

64. Santamaría RI, Bustos P, Sepúlveda-Robles O, Lozano L, Rodríguez C, Fernández JL, 682

et al. Narrow-host-range bacteriophages that infect Rhizobium etli associate with 683

distinct genomic types. Appl Environ Microbiol 2014; 80: 446–454. 684

65. Carlson K. Working with bacteriophages: Common techniques and methodological 685

approaches. In: Bacteriophages: Biology and Applications. In: Kutter E, Sulakvelidze 686

A (eds).2005. CRC Press, USA. 687

66. Werle E, Schneider C, Renner M, Völker M, Fiehn W. Convenient single-step, one 688

tube purification of PCR products for direct sequencing. Nucleic Acids Res 1994; 22: 689

4354–4355. 690

67. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high 691

throughput. Nucleic Acids Res 2004; 32: 1792–7. 692

68. Bolger AM, Lohse M, Usadel B. Trimmomatic: A flexible trimmer for Illumina 693

sequence data. Bioinformatics 2014; 30: 2114–2120. 694

69. Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, et al. 695

SPAdes: A new genome assembly algorithm and its applications to single-Cell 696

sequencing. J Comput Biol 2012; 19: 455–477. 697

70. Zerbino DR, Birney E. Velvet: Algorithms for de novo short read assembly using de 698

Bruijn graphs. Genome Res 2008; 18: 821–829. 699

71. Gordon D, Green P. Consed: a graphical editor for next-generation sequencing. 700

Bioinformatics 2013; 29: 2936–2937. 701

72. Chaudhari NM, Gupta VK, Dutta C. BPGA- an ultra-fast pan-genome analysis 702

pipeline. Nat Publ Gr 2016; 6: 24373. 703

73. Untergasser A, Cutcutache I, Koressaar T, Ye J, Faircloth BC, Remm M, et al. Primer3 704

— new capabilities and interfaces. 2012; 40: e115. 705

74. Richter M, Rosselló-Móra R. Shifting the genomic gold standard for the prokaryotic 706

species definition. Proc Natl Acad Sci U S A 2009; 106: 19126–19131. 707

75. Pritchard L, Glover RH, Humphris S, Elphinstone JG, Toth IK. Genomics and 708

taxonomy in diagnostics for food security: soft-rotting enterobacterial plant pathogens. 709

Anal Methods 2016; 8: 12–14. 710

76. Lopes A, Tavares P, Petit M, Guérois R, Zinn-justin S. Automated classification of 711

tailed bacteriophages according to their neck organization. BMC Genomics 2014; 15: 712

1027. 713

77. Hyman P, Abedon ST. Phage host range and efficiency of plating. In: Clokie MRJ, 714

Kropinski AM (eds). Bacteriophages, methods and protocols. Vol. I: Isolation, 715

characterization, and interactions. 2009. Totowa, NJ: Humana Press, pp 175–202. 716

78. Holmfeldt K, Solonenko N, Howard-Varona, C Moreno M, Malmstrom RR, Blow MJ, 717

Sullivan MB. Large-scale maps of variable infection efficiencies in aquatic 718

Bacteroidetes phage-host model systems. Environ Microbiol 2016; 18: 3949–3961. 719

79. Kauffman KM, Hussain FA, Yang J, Arevalo P, Brown JM, Chang WK, et al. A major 720

lineage of non-tailed dsDNA viruses as unrecognized killers of marine bacteria. Nature 721

2018; 554: 118–122. 722

80. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, Glinn D, et al. Community 723

Ecology Package. Retrieved from https://cran.r-project.org, 724

https://github.com/vegandevs/vegan. 2019. 725

81. Flores CO, Poisot T, Valverde S, Weitz JS. BiMat : A MATLAB package to facilitate 726

the analysis of bipartite networks. Methods Ecol Evol 2016; 7: 127–132. 727

82. Consul PC. A simple urn model dependent on predetermined strategy. Sankhyā Indian 728

J Stat Ser B 1974; 36: 391–399. 729

83. Borcard D, Legendre P. All-scale spatial analysis of ecological data by means of 730

principal coordinates of neighbour matrices. Ecol Modell 2002; 153: 51–68. 731

84. Flores CO, Valverde S, Weitz JS. Multi-scale structure and geographic drivers of 732

cross-infection within marine bacteria and phages. ISME J 2013; 7: 520–32. 733

85. Porter SS, Chang PL, Conow CA, Dunham JP, Friesen ML. Association mapping 734

reveals novel serpentine adaptation gene clusters in a population of symbiotic 735

Mesorhizobium. ISME J 2016; 11: 248–262. 736

86. Greenlon A, Chang PL, Damtew ZM, Muleta A, Carrasquilla-Garcia N, Kim D, et al. 737

Global-level population genomics reveals differential effects of geography and 738

phylogeny on horizontal gene transfer in soil bacteria. Proc Natl Acad Sci 2019; 116: 739

15200–15209. 740

87. Scola V, Ramond JB, Frossard A, Zablocki O, Adriaenssens EM, Johnson RM, et al. 741

Namib desert soil microbial community diversity, assembly, and function along a 742

natural xeric gradient. Microb Ecol 2018; 75: 193–203. 743

88. Short CM, Suttle CA. Nearly identical bacteriophage structural gene sequences are 744

widely distributed in both marine and freshwater environments. Appl Environ 745

Microbiol 2005; 71: 480–486. 746

89. Edwards RA, Vega AA, Norman HM, Ohaeri M, Levi K, Dinsdale EA, et al. Global 747

phylogeography and ancient evolution of the widespread human gut virus crAssphage. 748

Nat Microbiol 2019; 4: 1727–1736. 749

90. Culley AI, Steward GF. New genera of RNA viruses in subtropical seawater, inferred 750

from polymerase gene sequences. 2007; 73: 5937–5944. 751

91. Hurwitz BL, Brum JR, Sullivan MB. Depth-stratified functional and taxonomic niche 752

specialization in the ‘core’ and ‘flexible’ Pacific Ocean Virome. ISME J 2015; 9: 472–753

484. 754

92. Mühling M, Fuller NJ, Millard A, Somerfield PJ, Marie D, Wilson WH, et al. Genetic 755

diversity of marine Synechococcus and co-occurring cyanophage communities: 756

evidence for viral control of phytoplankton. Environ Microbiol 2005; 7: 499–508. 757

93. Sun Y, Zhang S, Long L, Dong J, Chen F, Huang S. Genetic diversity and 758

cooccurrence patterns of marine cyanopodoviruses and picocyanobacteria. Appl 759

Environ Microbiol 2018; 84: e00591-18. 760

94. Chase AB, Arevalo P, Brodie EL, Polz MF, Karaoz U, Martiny JBH. Maintenance of 761

sympatric and allopatric populations in free- living terrestrial bacteria. MBio 2019; 10: 762

e02361-19. 763

95. Flores CO, Meyer JR, Valverde S, Farr L, Weitz JS. Statistical structure of host – 764

phage interactions. Proc Natl Acad Sci 2011; 108: E288. 765

96. Koskella B, Thompson JN, Preston GM, Buckling A. Local biotic environment shapes 766

the spatial scale of bacteriophage adaptation to bacteria. Am Nat 2011; 177: 440–451. 767

97. Koskella B, Parr N. The evolution of bacterial resistance against bacteriophages in the 768

horse chestnut phyllosphere is general across both space and time. Phil Trans R Soc B 769

Biol Sci 2015; 370: 20140297. 770

98. Morgan AD, Gandon S, Buckling A. The effect of migration on local adaptation in a 771

coevolving host-parasite system. Nature 2005; 437: 253–256. 772

99. Gómez P, Paterson S, De Meester L, Liu X, Lenzi L, Sharma MD, et al. Local 773

adaptation of a bacterium is as important as its presence in structuring a natural 774

microbial community. Nat Commun 2016; 7: 12453. 775

100. Zhang Q-G, Buckling A. Resource-dependent antagonistic coevolution leads to a new 776

paradox of enrichment. Ecology 2016; 97: 1319–1328. 777

101. Lopez-Pascua LDC, Buckling A. Increasing productivity accelerates host-parasite 778

coevolution. J Evol Biol 2008; 21: 853–60. 779

102. Gurney J, Aldakak L, Betts A, Gougat-Barbera C, Poisot T, Kaltz O, et al. Network 780

structure and local adaptation in co-evolving bacteria–phage interactions. Mol Ecol 781

2017; 26: 1764–1777. 782

103. Thompson JN. The geographic mosaic of coevolution. 2005. Univ. Chicago Press. 783

784

785

Figure legends. 786

Fig. 1. Phylogenetic tree of all rhizobia collected from Mexico and Argentina (LC, n = 229) 787

and from the standard laboratory collection (SC, n = 94). The tree was constructed using the 788

maximum likelihood method and is based on the concatenated sequences of two 789

chromosomal genes (dnaA-recA). The tree bar scale indicates the number of nucleotide 790

substitutions per site. Insets on the left side explain the contents of the four concentric circles. 791

From the inner to the outermost circles: taxonomic classification of rhizobia, Rhizobium 792

chromosomal STs (sequence types; no STs were assigned to SC strains), field of origin of the 793

strains, and squares indicating the origin of the phages isolated using the corresponding 794

strain. 795

Fig. 2. Differentiation of Rhizobium and phage communities. A and B, PCoA plots 796

illustrating the differences in chromosomal composition between Rhizobium communities 797

across common bean populations based on Jaccard distances (presence-absence) and Bray-798

Curtis distances (relative abundance), respectively. D and E, PCoA plots showing the 799

differences in the phage genomic type composition between phage communities across 800

common bean populations based on Jaccard distances and Bray-Curtis distances. Bean fields 801

are indicated by different colors: Tepoztlán (T), light blue; Yautepec (Y), dark blue; Salta (S), 802

orange; Chicoana (C), red. Phages isolated by inoculating pooled samples from a given 803

location into rhizobia from the laboratory’s standard collection of rhizobia (SC) are indicated 804

with a star symbol. C and F, Venn diagrams of the distribution of Rhizobium chromosomal 805

STs and phage genomic types (PGTs), respectively. 806

Fig. 3. Structure of phage-Rhizobium interactions. Phage–bacterium infection network of 807

interactions between 196 phages (columns) and 229 Rhizobium strains (rows) performed with 808

BiMat [81]. Either full (red) or partial (blue) lytic interactions were recorded as positive 809

interactions in at least 2/3 of independent tests; blank cells indicate the absence of interaction. 810

In the bottom row, colored bars indicate the bean field of origin of the phages: Tepoztlán (T, 811

Mexico) = light blue; Yautepec (Y, Mexico) = dark blue; Salta (S, Argentina) = orange; 812

Chicoana (C, Argentina) = red. The first column on the right indicates the site of origin of 813

rhizobia with the same colors of the bars used for the phages. (A) Modular sorting of the 814

interaction data, visualizing the presence of three modules. (B) Nested sorting of the 815

interaction data, visualizing the spectrum of generalists to specialist phages. 816

Fig. 4. Principal coordinates analysis (PCoA) plot showing the Bray-Curtis dissimilarity in 817

the Rhizobium susceptibility range among common bean fields and Rhizobium species in A 818

and the Bray-Curtis dissimilarity in the phage host range among regions and phage 819

taxonomic families in B. Each axis explains a certain fraction of dissimilarity, given within 820

parentheses. Different Rhizobium species and phage taxonomic families are represented by 821

symbols. The bean fields of origin are indicated by different colors: Tepoztlán (T), light blue; 822

Yautepec (Y), dark blue; Salta (S), orange; Chicoana (C), red. 823

Fig. 5. Box plots of phage infection rates and Rhizobium susceptibility rates indicating phage 824

local adaptation and Rhizobium maladaptation. Phage infection rates between sympatric 825

versus allopatric combinations averaged across all locations (A), averaged across populations 826

within regions (B), or among populations (E). Rhizobium susceptibility rates between 827

sympatric versus allopatric combinations averaged across all locations (C), averaged across 828

populations within regions (D), or among populations (F). The mean is given for each box 829

plot (black diamond). Differences between the means of sympatric (“S”) versus allopatric 830

(“A”) phage infection rates (G) and the rates of the susceptibility of Rhizobium (H) are also 831

shown. The origins of the phage and Rhizobium samples are indicated by T = Tepoztlán 832

(Mexico), Y = Yautepec (Mexico), S = Salta (Argentina), and C = Chicoana (Argentina). 833

834

N2.11

C21 N841

TM1.19

TM2.3

N3.5

Y1.17

TM2.6

C46 IE1004

TM1.27

N3.23

TM1.3

i1.4

i2.3

TM1.31

C78 BAT 95

Y3.44

N1.12

N2.1

TM1.6

TM1.7

C20 N324

N3.21

TM1.10

C7 R634

TM1.25

Y1.11

C18 R711

N3.3

i3.9

C56 IE6854

i3.16

Y3.51

Y1.10

Y2.4

TM1.24

N3.16

N1.17

TM1.29

i2.9

Y3.34

Y3.27

Y2.8

C64 MIM7 4

TM1.17

C86 14C2

C44 IE4872

N1.6

N2.9

i1.13

7.2i

i3.1

TM1.12

C91 GR10

Y3.1

i3.7

Y3.23

Y3.9

i1.22

i1.2

C62 NXC14

Y1.6

Y3.31

N2.13

TM3.4

i1.20

TM3.2

C30 N941

Y3.38

C77 BAT 32

Y2.17

N1.7

C31 N1341

i1.5

TM3.19

TM2.5

TM1.1

C87 8NJ2

Y1.20

C41 Mim1

Y2.9

i2.2

C79 D5

TM1.9

i1.21

Y2.10

i1.6

C57 IE6868

Y2.18

Y3.37

Y3.43

i1.23

C75 BAT 4

N1.4

TM1.37

C52 IE6794

TM1.4

Y1.1

N3.13

N2.12

i1.17

C92 GR14

C26 N4311

TM1.22

i2.8

TM3.9

Y2.6

N3.15

Y2.1

Y1.15

C9 R635

TM1.16

C23 N621

C53 IE6833

TM2.1

Y3.41

N1.5

Y3.36

TM3.10

N2.14

Y2.15

C51 IE6766

i2.6

TM1.28

139N 92C

Y3.45

N3.22

C36 GR56

TM1.8

i3.4

i2.11

TM3.18

C32 N871

C76 BAT 49

TM1.21

C38 Bra5

N1.18

TM2.2

C42 IE4771

C67 TUX10P

Y3.21

i3.10

TM1.20

C4 N771

C60 TAL182

N3.6

N3.12

C6 R72

i3.8

C45 IE951

C88 17NJ2

TM1.5

N1.16

TM1.14

N1.11

C13 R723

N3.20

C83 PS14

C69 TUX1712m

C74 B27

C80 D11

N3.7

C34 CFN42

i1.10

Y2.13

C37 Kim5

C70 TUX250P

i1.18

C54 IE6840

N3.10

TM3.16

i3.2

Y1.23

i2.4

C93 GR62

i1.16

138N 72C

Y3.19

i3.15

TM1.30

Y2.16

i1.15

C12 R650

TM3.14

N1.13

C14 R611

Y1.7

41.1Y

N2.3

C47 IE1006

i3.6

N1.14

TM1.2

i3.5

C39 S20

i2.5

i1.9

i3.3

C85 14C1

C24 N113

Y1.22

C63 MIM2

TM1.18

Y1.4

N1.1

Y1.12

C3 N561

C82 D21

i2.1

C61 NXC12

C28 N731

i3.19

Y1.21

N2.2

C84 6C1

i1.3

N2.8

N2.6

Y3.17

Y3.5

TM1.15

Y3.15

N3.1

i1.7

i1.19

i1.24

C49 IE4777

C55 IE6845

N3.17

C94 GR75

C43 IE4803

TM1.23

C11 N261

N3.11

C5 N161

i2.10

Y2.14

C48 IE2737

C17 R630

C15 R620

i3.11

TM3.5

N3.18

i1.1

TM3.15

C72 343

i3.18

Y1.19

C68 TUX16P

N1.9

Y3.42

TM3.7

C40 CIAT894

N3.19

i3.17

Y1.8

C16 R693

TM3.13

i1.12

TM2.4

C50 IE4810

C25 N6212

C1 N1314

C89 21NJ2

i1.11

TM1.36

C59 CNPAF512

i3.14

C58 IE6896

Y3.12

C2 N741

N1.8

C35 CIAT652

TM3.17

TM1.32

N1.2

C81 D13

Y3.25

Y1.16

N1.10

i1.14

Y3.56

TM1.11

TM1.13

Y2.3

C65 INC1 2

C33 R339

N3.2

TM1.38

C8 R744

TM3.12

Y3.47

TM3.11

N1.15

N2.10

i3.13

TM1.33

Y3.20

TM1.34

N1.3

i3.12

6.3MT

TM3.3

Y3.46

C10 N671

Y2.19

N3.14

TM1.35

N2.4

N2.5

Y2.7

0142hC 17C

Y3.40

C90 6PR1

Y3.52

TM1.26

C66 INC2 5

C22 N541

N3.4

Y2.11

N2.7

C73 F6

C19 N122

i2.20

0.9

7.0

0.7

0.9

1.0

0.7

0.8

1.0

0.7

1.0

0.6

1.0

0.5

1.0

0.8

0.7

1.0

0.9

0.6

0.8

1.0

0.7

1.0

0.8

1.0

0.9

1.0

0.6

0.7

0.9

0.8

1.0

0.7

0.6

0.5

0.1

0.8

1.0

0.9

0.7

0.5

0.8

1.0

0.6

0.8

0.7

0.6

0.9

0.7

0.6

0.8

0.7

0.6

0.1

1.0

0.7

0.8

1.0

0.5

7.0

0.5

0.6

1.0

0.8

1.0

Tree scale: 0.01

Species

R.etli

R.phaseoli

R.gallicum

R.leguminosarum

Rhizobia ST’s

ST5 - 82 strains

ST34 - 42 strains

ST22 - 19 strains

ST16 - 17 strains

ST18 - 12 strains

ST’s - 2-10 strains

ST’s - 1 strain

Rhizobia origin

Tepoztlan

Yautepec

Chicoana

Salta

Phages origin

Tepoztlan

Yautepec

Chicoana

Salta

-0.2

0.0

0.2

0.4

0.6

-0.4 -0.2 0.2

PCoA axis 1 (27.3%)

PCoA axis 2 (15.3%)

0.0 0.4

PCoA axis 1 (42.5%)

PCoA axis 2 (18.5%)

-0.2

0.0

0.2

0.4

0.6

0.8

-0.2 0.2

0.0 0.4

-0.4-0.6

Tepoztlan

Yautepec

Chicoana

Salta

ST9

ST1

ST6

ST2

ST7

ST8 ST13

ST14 ST15

ST16 ST17

ST18 ST19

ST28 ST39

ST40

ST20 ST25

ST41

ST3

ST12

ST32

ST4

ST21

ST23

ST26 ST27

ST29 ST30

ST31 ST35

ST36

ST37

ST38

ST5

ST10

ST22

ST11

ST33

ST24

ST34

PCoA axis 1 (20.1%)

PCoA axis 2 (17.6%)

-0.4

-0.2

0.0

0.2

0.4

0.6

-0.2 0.2

0.0 0.4

-0.4-0.6

PCoA axis 1 (25.0%)

PCoA axis 2 (23.6%)

-0.6

-0.4

-0.2

0.0

0.2

0.6

-0.4 -0.2 0.2

0.0 0.4

F64

F72_1

F37

F38

F41

F03

F08

F48

F09

F02

F62

F89

F40

F45

F06

F10

F14

Tepoztlan

Yautepec

Chicoana

Salta F12

F53

F42

F92

F75

F95

F98

F108

F112

F72_2

F114

F82

d = 0.2

F48

T3.12

N2.10

N2.1

N2.5

N2.14

N3.13

Y1.14

N3.7

i3.8

i3.9

T1.38

Y2.19

T1.13

Y2.8

Y3.31

N1.14

T3.16

Y3.45

i3.18

N3.2

T3.7

Y3.34

T1.20

Y2.7

i2.8

Y1.19

Y1.22

T1.23

Y3.44

i2.4

T3.11

T3.15

Y2.13

Y1.12

i3.12

T1.2

T1.14

Y3.25

T1.18

Y2.4

Y3.5

i2.5

i2.20

T1.12

T1.16

Y3.41

T1.8

Y2.15

T1.22

T2.4

Y1.11

i3.11

Y3.27

T3.10

T3.14

T1.25

T1.30

Y3.15

Y3.36

i2.6

T1.11

T1.21

Y1.7

Y1.21

Y3.9

T1.32

Y3.51

Y3.52

T3.19

Y2.10

Y1.15

Y1.16

Y1.17

T1.6

T1.19

T2.5

T3.18

Y2.11

Y3.19

i2.7

T1.7

Y3.1

T1.24

T1.4

Y1.10

i2.1

T1.29

T1.34

T1.36

Y3.56

T1.31

T2.6

Y3.43

N1.15

T1.5

T1.17

i2.10

T1.15

T3.4

N1.16

T1.3

T1.10

Y1.1

Y2.9

Y3.38

T1.35

i2.2

T3.9

Y1.6

Y3.23

T1.1

Y1.4

Y3.47

T1.28

T2.1

Y3.12

Y1.23

Y2.17

Y3.21

Y3.20

i1.19

i1.6

i1.10

i1.18

T3.2

T3.5

Y3.40

i1.20

T1.26

i1.14

N3.19

Y2.14

Y3.46

i1.9

i1.7

Y1.20

Y2.3

i1.4

N3.6

i1.5

i1.24

T2.2

Y2.18

i1.3

i1.13

i2.9

i3.19

N3.5

Y2.16

N3.3

N3.10

i3.6

i1.23

i3.1

i3.17

T3.13

Y3.42

i3.5

i3.7

i3.14

i3.15

i3.16

Y2.6

i1.11

i1.15

i1.22

i3.2

i3.3

i3.4

i1.12

N1.18

N2.6

T1.37

N2.12

N3.20

N3.21

T3.17

N1.17

T3.6

Y1.8

i1.2

i1.16

N1.10

N2.11

N3.12

N3.18

N3.22

N3.23

N1.8

N1.9

N2.2

N2.7

N3.1

N3.11

N3.16

i1.1

i3.13

N2.3

N2.8

N2.13

N3.4

i1.17

N1.7

N2.9

T3.3

i1.21

N1.4

N1.11

N1.13

N2.4

N3.14

N1.3

N1.5

N1.12

N3.15

N1.6

N1.2

T1.33

T1.9

N1.1

i3.10

Y2.1

Y3.17

T1.27

N3.17

i2.3

i2.11

Y3.37

T2.3

N45

TM45

N1.14F

N12

N15

N23

N1.2F

N1.10F

N1.11F

N1.16F

N2.7F

N2.9F

N2.13F

N14

N43

N13

N32

N11

N1.4F

N30

TM2.3B

TM41

Y52

TM34

TM21B

Y74

N2.2F

N1.15F

Y2.11F

TM2.3A

TM39

i36

TM46

N1.5F

N1.18F

TM61

TM2.2A

TM3_3.4A

TM36

TM3_3.4B

TM1.2A

Y2.17F

i65

i72

TM1.10B

TM3.4B

N25

TM3.4A

TM3.14B

Y2.18F

TM25B

TM23

Y12

Y3.44F

Y3.47F

Y38

Y3.45F

TM30

TM32

TM2.5A

TM3_3.7A

Y3.37F

TM1.34A

TM1.34B

TM3_3.7B

Y2.6F

Y2.14F

TM24

TM35

TM1.21B

TM15

TM3_3.11B

Y68

Y3.19F

N28

N38

N39

TM25A

Y3.36F

Y3.42F

TM3_3.14A

N2.6F

TM1.7B

TM3.5A

Y3.27F

Y3.56F

TM1.13A

Y2.7F

Y2.19F

Y3.40F

TM3.5B

TM1.7A

TM3.12A

TM1.21A

TM3.12B

Y65

Y1.7F

TM1.13B

N34

TM5B

i34

TM8

i45

i46

TM27A

Y66

N17

TM27B

Y1.1F

i37

TM17

TM22

Y2.4F

TM3_3.5B

TM29

TM3_3.11A

TM3_3.6

TM3_3.5A

TM1.2B

i1.21F

TM1.10A

N10

TM3.14A

i1.22F

TM1.36A

TM2.5B

Y17

i3.8F

i3.11F

Y55

TM3_3.12

TM3_3.16

N2.10F

Y1.11F

i3.18F

N27

TM3_3.9

Y1.21F

TM5A

TM1.36B

i1.13F

Y60

Y1.22F

TM3_3.10

Y1.20F

i17

Y5B

Y1.12F

i1.14F

Y86

Y21

Y5A

Y48

Y67

i1.6F

TM3_3.14B

Y3.5F

Y3.15F

Y3.1F

N42

TM1.20A

TM3_3.3

Y25

Y3.38F

Y1.10F

Y3.43F

Y1.6F

N3.2F

TM26

i1.10F

i1.23F

Y90

i1.11F

i1.9F

i1.18F

N3.13F

TM40

TM3_3.13

TM33

N65

TM2.4B

TM38

TM2.4A

N3.19F

i20

i42

Phages

Rhizobia

i1.6

i1.19

i1.10

i1.18

i1.20

i1.14

i1.9

i2.9

i1.7

T1.15

Y2.14

i1.4

i1.24

i1.13

i1.5

i3.6

i3.1

i1.3

i3.17

i3.7

i3.14

i3.15

i3.16

i3.5

i3.2

i3.3

Y2.18

i1.23

i1.11

i3.4

N2.6

N3.21

Y1.20

i1.15

N1.18

N3.20

i1.22

N2.12

N1.17

N1.10

N3.12

N3.18

N3.22

N3.23

N3.11

N1.8

N2.2

N3.16

N2.3

i3.19

i1.12

N1.9

N2.7

N3.1

N2.13

N3.4

N1.7

Y2.3

Y2.6

i1.16

N2.11

i3.13

N2.8

N2.9

N1.11

N2.4

N3.14

i1.2

i1.1

N1.4

N1.13

N1.3

N1.5

N1.12

N3.15

T1.26

N1.6

T3.17

Y1.8

i1.17

N1.2

T3.6

Y3.42

i1.21

i3.10

T3.13

N1.1

T1.9

Y3.17

T1.27

Y2.1

N3.17

i2.3

i2.11

N2.10

N2.5

N2.1

N2.14

N3.13

N3.7

i3.8

i3.9

Y3.31

N1.14

i2.8

T1.13

i3.18

N3.2

Y2.19

Y2.7

Y3.45

Y3.34

Y3.44

Y1.14

i3.12

i3.11

Y2.13

T1.16

T1.25

T1.36

i2.4

T1.2

Y2.4

Y3.5

Y3.25

i2.20

Y1.11

Y3.27

Y3.9

Y1.16

i2.5

Y2.15

Y3.15

Y3.38

T3.12

Y3.47

T3.7

Y1.19

Y3.36

i2.7

Y1.21

Y3.52

Y3.51

T1.38

T1.20

i2.6

Y3.19

Y2.8

T3.11

Y3.1

T1.23

Y1.12

Y3.41

T2.4

Y3.21

T1.14

Y3.12

Y3.20

T3.14

T3.19

Y1.15

Y3.56

Y3.43

T3.5

Y3.40

i2.1

Y1.1

Y2.11

N1.15

N1.16

Y2.9

i2.10

i2.2

T3.4

Y3.46

Y2.16

N3.19

T1.35

T3.2

T1.10

T2.2

T3.3

Y3.23

T1.33

Y1.23

Y1.6

Y1.4

N3.10

T3.16

T3.10

T1.12

T1.18

T1.21

T1.24

Y1.22

T1.22

T1.8

T1.11

T1.7

T1.5

T3.9

T2.5

T3.18

T1.4

T1.34

T1.31

T2.1

T1.6

T3.15

T1.30

T1.32

Y1.17

Y1.7

T1.19

N3.5

Y1.10

T1.3

N3.6

N3.3

T1.1

T1.29

T1.17

T1.28

Y2.10

T1.37

T2.6

Y2.17

Y3.37

T2.3

i72

i65

Y2.17F

TM39

N25

Y2.7F

TM8

TM30

Y3.37F

Y1.7F

Y66

TM22

Y3.42F

Y12

TM3_3.16

Y55

Y60

Y1.22F

Y1.12F

TM3_3.9

i3.11F

i3.8F

TM2.5B

TM3_3.10

Y25

TM3_3.13

TM40

TM3_3.4A

TM36

TM3.4A

TM3_3.4B

TM3.4B

TM1.2A

TM3_3.14A

TM1.13A

TM15

TM23

TM35

TM1.10B

TM3.14B

TM25B

Y3.47F

Y3.44F

Y3.45F

Y3.36F

Y3.19F

Y3.27F

Y2.19F

TM25A

TM3_3.7A

TM3.5A

TM1.13B

TM3_3.7B

TM3_3.11B

Y3.40F

TM2.5A

TM1.34B

TM1.34A

TM1.21B

TM3.5B

TM32

TM24

Y3.56F

TM3.12A

TM1.7B

TM3.12B

TM27A

TM27B

TM17

TM1.2B

TM1.7A

TM29

TM1.21A

TM1.10A

Y65

TM3.14A

TM1.36A

TM5B

Y1.1F

TM3_3.11A

N10

TM3_3.5B

TM3_3.5A

Y2.4F

Y17

Y1.11F

N17

TM3_3.12

TM5A

N2.10F

Y1.21F

N27

i3.18F

TM1.36B

Y21

TM3_3.14B

Y3.15F

Y3.5F

Y67

Y3.1F

TM1.20A

Y3.38F

TM3_3.3

Y3.43F

Y1.10F

Y1.6F

N3.2F

N3.13F

TM2.4B

N65

TM2.4A

N3.19F

i42

N45

N2.13F

N2.9F

N2.7F

N1.16F

N1.11F

N1.10F

N1.2F

N23

N15

N12

N43

N14

N32

N1.4F

N11

N1.14F

N13

N30

TM45

TM41

TM34

TM2.3B

Y74

TM21B

N2.2F

Y68

Y2.11F

TM2.3A

TM46

N1.15F

N1.18F

Y52

N1.5F

TM2.2A

TM61

Y2.18F

Y2.14F

N34

N2.6F

N38

Y38

N39

TM3_3.6

i34

i46

i45

i36

i37

Y2.6F

i1.22F

Y86

i1.21F

i1.13F

i1.14F

N28

i1.6F

Y48

i17

Y1.20F

Y5B

Y5A

i1.10F

Y90

i1.23F

TM26

i1.11F

i1.18F

i1.9F

TM33

N42

TM38

i20

Phages

Rhizobia

Modular network Nested network

−0.4 −0.2 0.0 0.2

Dim2

−0.4 −0.2 0.0 0.2 0.4

−0.4 −0.2 0.0

Dim1

Dim2

−0.2 0.0 0.2 0.4

Dim2

−0.4 −0.2 0.0 0.2

Dim2

PCoA axis 1 (43.0%)

R. phaseoli

R. etli

PCoA axis 1 (22.2%)

PCoA axis 2 (15.4%)

−0.2 0.0 0.2 0.4

−0.4 −0.2 0.0 0.2 0.4 0

Dim2

−0.2 0.0 0.2 0.4

−0.4 −0.2 0.0 0.2 0.4

Dim1

Dim2

−0.4 −0.2 0.0 0.2 0.4 0.6

Dim2

−0.2 0.0 0.2 0.4

−0.4 −0.2 0.0 0.2 0.4

Dim1

Dim2

Podoviridae

Siphoviridae

Myoviridae

Microviridae

PCoA axis 2 (13.1%)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Sympatric Allopatric

Infection rates

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Infection rates

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Sympatric Allopatric

Susceptibility rates

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Susceptibility rates

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Infection rates

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Susceptibility rates

−0.6

−0.4

−0.2

0.0

0.2

0.4

0.6

0.8

1.0

Difference S − A

−0.8

−0.6

−0.4

−0.2

0.0

0.2

0.4

0.6

0.8

Difference S − A

Sympatric

Argentina

Allopatric

Argentina

Sympatric

Mexico

Allopatric

Mexico

Sympatric

Argentina

Allopatric

Argentina

Sympatric

Mexico

Allopatric

Mexico

C S T Y

Phages

origin

Rhizobium

origin

Rhizobium

origin

Phages

origin

C S T Y C S T Y C S T Y C S T Y

C S T Y

C S T Y C S T Y C S T Y C S T Y

Sympatric

phages

Allopatric

phages

Sympatric

rhizobia

Allopatric

rhizobia

C S T Y

S T Y C T Y C S Y C S T

C S T Y

S T Y C T Y C S Y C S T

ResearchGate has not been able to resolve any citations for this publication.

Maintenance of Sympatric and Allopatric Populations in Free-Living Terrestrial Bacteria

Article

Full-text available

Oct 2019

Due to the promiscuous exchange of genetic material and asexual reproduction, delineating microbial species (and, by extension, populations) remains challenging. Because of this, the vast majority of microbial studies assessing population structure often compare divergent strains from disparate environments under varied selective pressures. Here, we investigated the population structure within a single bacterial ecotype, a unit equivalent to a eukaryotic species, defined as highly clustered genotypic and phenotypic strains with the same ecological niche. Using a combination of genomic and computational analyses, we assessed the phylogenetic structure, extent of recombination, and flexible gene content of this genomic diversity to infer patterns of gene flow. To our knowledge, this study is the first to do so for a dominant soil bacterium. Our results indicate that bacterial soil populations, similarly to those in other environments, are structured by gene flow discontinuities and exhibit distributional patterns consistent with both isolation by distance and isolation by environment. Thus, both dispersal limitation and local environments contribute to the divergence among closely related soil bacteria as observed in macroorganisms.

Multifaceted Impacts of Bacteriophages in the Plant Microbiome

Article

Full-text available

Aug 2018

Plant-associated bacteria face multiple selection pressures within their environments and have evolved countless adaptations that both depend on and shape bacterial phenotype and their interaction with plant hosts. Explaining bacterial adaptation and evolution therefore requires considering each of these forces independently as well as their interactions. In this review, we examine how bacteriophage viruses (phages) can alter the ecology and evolution of plant-associated bacterial populations and communities. This includes influencing a bacterial population's response to both abiotic and biotic selection pressures and altering ecological interactions within the microbiome and between the bacteria and host plant. We outline specific ways in which phages can alter bacterial phenotype and discuss when and how this might impact plant-microbe interactions, including for plant pathogens. Finally, we highlight key open questions in phage-bacteria-plant research and offer suggestions for future study. Expected final online publication date for the Annual Review of Phytopathology Volume 56 is August 25, 2018. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

A major lineage of non-tailed dsDNA viruses as unrecognized killers of marine bacteria

Article

Full-text available

Feb 2018
NATURE

The most abundant viruses on Earth are thought to be double-stranded DNA (dsDNA) viruses that infect bacteria. However, tailed bacterial dsDNA viruses (Caudovirales), which dominate sequence and culture collections, are not representative of the environmental diversity of viruses. In fact, non-tailed viruses often dominate ocean samples numerically, raising the fundamental question of the nature of these viruses. Here we characterize a group of marine dsDNA non-tailed viruses with short 10-kb genomes isolated during a study that quantified the diversity of viruses infecting Vibrionaceae bacteria. These viruses, which we propose to name the Autolykiviridae, represent a novel family within the ancient lineage of double jelly roll (DJR) capsid viruses. Ecologically, members of the Autolykiviridae have a broad host range, killing on average 34 hosts in four Vibrio species, in contrast to tailed viruses which kill on average only two hosts in one species. Biochemical and physical characterization of autolykiviruses reveals multiple virion features that cause systematic loss of DJR viruses in sequencing and culture-based studies, and we describe simple procedural adjustments to recover them. We identify DJR viruses in the genomes of diverse major bacterial and archaeal phyla, and in marine water column and sediment metagenomes, and find that their diversity greatly exceeds the diversity that is currently captured by the three recognized families of such viruses. Overall, these data suggest that viruses of the non-tailed dsDNA DJR lineage are important but often overlooked predators of bacteria and archaea that impose fundamentally different predation and gene transfer regimes on microbial systems than on tailed viruses, which form the basis of all environmental models of bacteria-virus interactions.

Local adaptation at higher trophic levels: Contrasting hyperparasite-pathogen infection dynamics in the field and laboratory

Article

Full-text available

Nov 2016
MOL ECOL

Predicting and controlling infectious disease epidemics is a major challenge facing the management of agriculture, human and wildlife health. Coevolutionarily derived patterns of local adaptation among pathogen populations have the potential to generate variation in disease epidemiology, however studies of local adaptation in disease systems have mostly focused on interactions between competing pathogens or pathogens and their hosts. In nature, parasites and pathogens are also subject to attack by hyperparasitic natural enemies that can severely impact upon their infection dynamics. However, few studies have investigated if this interaction varies across combinations of pathogen-hyperparasite strains, and if this influences hyperparasite incidence in natural pathogen populations. Here, we test if the association between a hyperparasitic fungus, Ampelomyces quisqualis, and a single powdery mildew host, Podosphaera plantaginis, varies among genotype combinations, and whether this drives hyperparasite incidence in nature. Laboratory inoculation studies reveal that genotype, genotype x genotype interactions, and local adaptation affect hyperparasite infection. However, observations of a natural pathogen metapopulation reveal that spatial rather than genetic factors predict the risk of hyperparasite presence. Our results highlight how sensitive the outcome of biocontrol using hyperparasites is to selection of hyperparasite strains. This article is protected by copyright. All rights reserved.

The evolution of bacterial resistance against bacteriophages in the horse chestnut phyllosphere is general across both space and time

Article

Full-text available

Aug 2015
PHILOS T R SOC B

Insight to the spatial and temporal scales of coevolution is key to predicting the outcome of host-parasite interactions and spread of disease. For bacteria infecting long-lived hosts, selection to overcome host defences is just one factor shaping the course of evolution; populations will also be competing with other microbial species and will themselves be facing infection by bacteriophage viruses. Here, we examine the temporal and spatial patterns of bacterial adaptation against natural phage populations from within leaves of horse chestnut trees. Using a time-shift experiment with both sympatric and allopatric phages from either contemporary or earlier points in the season, we demonstrate that bacterial resistance is higher against phages from the past, regardless of spatial sympatry or how much earlier in the season phages were collected. Similarly, we show that future bacterial hosts are more resistant to both sympatric and allopatric phages than contemporary bacterial hosts. Together, our results suggest the evolution of relatively general bacterial resistance against phages in nature and are contrasting to previously observed patterns of phage adaptation to bacteria from the same tree hosts over the same time frame, indicating a potential asymmetry in coevolutionary dynamics.

Automated classification of tailed bacteriophages according to their neck organization

Article

Full-text available

Nov 2014
BMC GENOMICS

The genetic diversity observed among bacteriophages remains a major obstacle for the identification of homologs and the comparison of their functional modules. In the structural module, although several classes of homologous proteins contributing to the head and tail structure can be detected, proteins of the head-to-tail connection (or neck) are generally more divergent. Yet, molecular analyses of a few tailed phages belonging to different morphological classes suggested that only a limited number of structural solutions are used in order to produce a functional virion. To challenge this hypothesis and analyze proteins diversity at the virion neck, we developed a specific computational strategy to cope with sequence divergence in phage proteins. We searched for homologs of a set of proteins encoded in the structural module using a phage learning database. We show that using a combination of iterative profile-profile comparison and gene context analyses, we can identify a set of head, neck and tail proteins in most tailed bacteriophages of our database. Classification of phages based on neck protein sequences delineates 4 Types corresponding to known morphological subfamilies. Further analysis of the most abundant Type 1 yields 10 Clusters characterized by consistent sets of head, neck and tail proteins. We developed Virfam, a webserver that automatically identifies proteins of the phage head-neck-tail module and assign phages to the most closely related cluster of phages. This server was tested against 624 new phages from the NCBI database. 93% of the tailed and unclassified phages could be assigned to our head-neck-tail based categories, thus highlighting the large representativeness of the identified virion architectures. Types and Clusters delineate consistent subgroups of Caudovirales, which correlate with several virion properties. Our method and webserver have the capacity to automatically classify most tailed phages, detect their structural module, assign a function to a set of their head, neck and tail genes, provide their morphologic subtype and localize these phages within a "head-neck-tail" based classification. It should enable analysis of large sets of phage genomes. In particular, it should contribute to the classification of the abundant unknown viruses found on assembled contigs of metagenomic samples.

Statistical structure of host-phage interactions

Article

Full-text available

Jun 2011
P NATL ACAD SCI USA

Interactions between bacteria and the viruses that infect them (i.e., phages) have profound effects on biological processes, but despite their importance, little is known on the general structure of infection and resistance between most phages and bacteria. For example, are bacteria-phage communities characterized by complex patterns of overlapping exploitation networks, do they conform to a more ordered general pattern across all communities, or are they idiosyncratic and hard to predict from one ecosystem to the next? To answer these questions, we collect and present a detailed metaanalysis of 38 laboratory-verified studies of host-phage interactions representing almost 12,000 distinct experimental infection assays across a broad spectrum of taxa, habitat, and mode of selection. In so doing, we present evidence that currently available host-phage infection networks are statistically different from random networks and that they possess a characteristic nested structure. This nested structure is typified by the finding that hard to infect bacteria are infected by generalist phages (and not specialist phages) and that easy to infect bacteria are infected by generalist and specialist phages. Moreover, we find that currently available host-phage infection networks do not typically possess a modular structure. We explore possible underlying mechanisms and significance of the observed nested host-phage interaction structure. In addition, given that most of the available host-phage infection networks examined here are composed of taxa separated by short phylogenetic distances, we propose that the lack of modularity is a scale-dependent effect, and then, we describe experimental studies to test whether modular patterns exist at macroevolutionary scales.

Genetic diversity and cooccurrence patterns of marine cyanopodoviruses and picocyanobacteria

Article

Aug 2018

Picocyanobacteria Prochlorococcus and Synechococcus are abundant in the global oceans and subject to active viral infection. In this study, the genetic diversity of picocyanobacteria and the genetic diversity of cyanopodoviruses were synchronously investigated along water columns in the equatorial Indian Ocean and over a seasonal time course in the coastal Sanya Bay, South China Sea. Using the 16S-23S rRNA internal transcribed spacer (ITS)-based clone library and quantitative PCR (qPCR) analyses, the picocyanobacterial community composition and abundance were determined. Sanya Bay was dominated by clade II Synechococcus during all the seasons, and a typical population shift from high-light-adapted Prochlorococcus to low-light-adapted Prochlorococcus was found along the vertical profiles. Strikingly, the DNA polymerase gene sequences of cyanopodoviruses revealed a much greater genetic diversity than we expected. Nearly one-third of the phylogenetic groups were newly described here. No apparent seasonal pattern was observed for the Sanya Bay picocyanobacterial or cyanopodoviral communities. Different dominant cyanopodovirus lineages were identified for the coastal area, upper euphotic zone, and middle-to-lower euphotic zone of the open ocean. Diversity indices of both picocyanobacteria and cyanopodoviruses were highest in the middle euphotic zone and both were lower in the upper euphotic zone, reflecting a host-virus interaction. Cyanopodoviral communities differed significantly between the upper euphotic zone and the middle-to-lower euphotic zone, showing a vertical pattern similar to that of picocyanobacteria. However, in the surface waters of the open ocean, cyanopodoviruses exhibited no apparent biogeographic pattern, differing from picocyanobacteria. This study demonstrates correlated distribution patterns of picocyanobacteria and cyanopodoviruses, as well as the complex biogeography of cyanopodoviruses.

Population Genomics of the Symbiotic Plasmids of Sympatric Nitrogen-Fixing Rhizobium Species Associated with Phaseolus vulgaris

Article

Jun 2016
ENVIRON MICROBIOL

Cultivated common beans are the primary protein source for millions of people around the world who subsist on low-input agriculture, enabled by the symbiotic N2 -fixation these legumes perform in association with rhizobia. Within a single agricultural plot, multiple Rhizobium species can nodulate bean roots, but it is unclear how genetically isolated these species remain in sympatry. To better understand this issue, we sequenced and compared the genomes of 33 strains isolated from the rhizosphere and root nodules of a particular bean variety grown in the same agricultural plot. We found that the Rhizobium species we observed coexist with low genetic recombination across their core genomes. Accessory plasmids thought to be necessary for the saprophytic lifestyle in soil show similar levels of genetic isolation, but with higher rates of recombination than the chromosomes. However, the symbiotic plasmids are extremely similar, with high rates of recombination and do not appear to have co-evolved with the chromosome or accessory plasmids. Therefore, while Rhizobium species are genetically isolated units within the microbial community, a common symbiotic plasmid allows all Rhizobium species to engage in symbiosis with the same host in a single agricultural plot. This article is protected by copyright. All rights reserved.

Abiotic heterogeneity drives parasite local adaptation in coevolving bacteria and phages

Article

Nov 2011
J EVOLUTION BIOL

Spatial abiotic heterogeneity can result in divergent selection, hence might increase the magnitude of host-parasite local adaptation (the mean difference in fitness of sympatric vs. allopatric host-parasite combinations). We explicitly tested this hypothesis by measuring local adaptation in experimentally coevolved populations of bacteria and viruses evolved in the same or different nutrient media. Consistent with previous work, we found that mean levels of evolved phage infectivity and bacteria resistance varied with nutrient concentration, with maximal levels at nutrient concentrations that supported the greatest densities of bacteria. Despite this variation in evolved mean infectivity and resistance between treatments, we found that parasite local adaptation was greatly increased when measured between populations evolved in different, compared with the same, media. This pattern is likely to have resulted from different media imposing divergent selection on bacterial hosts, and phages in turn adapting to their local hosts. These results demonstrate that the abiotic environment can play a strong and predictable role in driving patterns of local adaptation.

Spatial patterns in phage-Rhizobium coevolutionary interactions across regions of common bean domestication

Abstract

Recommended publications

Spatial patterns in phage-Rhizobium coevolutionary interactions across regions of common bean domest...

Correction to: Spatial patterns in phage-Rhizobium coevolutionary interactions across regions of com...

Bacteria-Phage Interactions across Time and Space: Merging Local Adaptation and Time-Shift Experimen...

Novel lineages of single-stranded DNA phages that coevolved with the symbiotic bacteria Rhizobium