bactrocera dorsalis (diptera: tephritidae) mazarin akami · 1 1 host fruit as a suitable bacteria...

1

Host fruit as a suitable bacteria growth substrate that promotes larval development 1

of Bactrocera dorsalis (Diptera: Tephritidae) 2

Mazarin Akami1,2,3

, Xueming Ren1, Yaohui Wang

1, Abdelaziz Mansour

1,4, Shuai Cao

1, 3

Xuewei Qi1, Albert Ngakou

3 and Chang-Ying Niu

1* 4

Authors’ Affiliations 5

1Department of Plant Protection, College of Plant Science & Technology, Huazhong 6

Agricultural University, Wuhan 430070, China 7

2Department of Plant Protection, Institute of Vegetables and Flowers, Chinese Academy 8

of Agricultural Sciences, Beijing, China, 9

3Department of Biological Sciences, Faculty of Science, University of Ngaoundere, P.O 10

Box 454 Ngaoundere, Cameroon 11

4Department of Economic Entomology and Pesticides, Faculty of Agriculture, Cairo 12

University, 12613 Giza, Egypt 13

14

15

16

17

*Authors for correspondence: CYN, [email protected] 18

.CC-BY-NC 4.0 International licenseunder anot certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available

The copyright holder for this preprint (which wasthis version posted November 7, 2019. ; https://doi.org/10.1101/834119doi: bioRxiv preprint

https://doi.org/10.1101/834119http://creativecommons.org/licenses/by-nc/4.0/

2

Abstract 19

The ability of a host plant to act as a substrate or media for larval development may 20

depend on how good it is at offering suitable nutrients for bacterial growth. In this study, 21

we hypothesized that the suitability of a fruit type for fruit fly larval development is 22

positively correlated with the ability of that fruit to act as a substrate/media for fruit fly 23

symbiotic bacterial growth. We allowed a single female fruit fly to lay eggs on five 24

different host fruits, then we monitored the larval development parameters across five 25

generations and analyzed the bacterial community structure of larvae developing in 2 of 26

these hosts (apple and banana) at the first and fifth generations. Results indicate that the 27

larval length and dry weight did not vary significantly across experimental generations, 28

but were greatly affected by fruit types and larval stages. The larval development time 29

was extended considerably in apple and tomato but shortened in banana and mango. 30

There was a significant shift in bacterial community structure and composition across 31

fruits and generations. The bacterial community of larvae within the same fruit (apple and 32

banana) clustered and was similar to the parental female (with the predominance of 33

Proteobacteria), but there was a shift at the fifth generation (dominance of Firmicutes). 34

Banana offered a suitable better development and growth to larvae and bacteria, 35

respectively, compared to apple in which reduced larval development and bacterial 36

growth were recorded. Although additional experiments are needed to adequately show 37

that the differences in microbiome seen in fruit fly larval guts are the actual driver of 38

different developmental outcomes of larvae on the different fruits, at the very least, our 39

study has provided intriguing data suggesting interaction between the diets and gut 40

microbial communities on insect development. 41




3

Key words: Bactrocera dorsalis, Host use, Foraging choice, Gut microbiome, Larval 42

development. 43

Importance and Significance of the study 44

Tephritid fruit flies entertain complex interactions with gut bacteria. These bacteria are 45

known to provide nutritional benefits to their hosts, by supplementing missing nutrients 46

from the host diets and regulating energy balance. Foraging for food is a risky exercise 47

for the insect which is exposed to ecological adversities, including predators. Therefore, 48

making beneficial choice among available food substrates is a question of survival for the 49

flies and bacteria as well. Our study demonstrates interactions between the host fly and 50

its intestinal bacteria in sustaining the larval development while foraging optimally on 51

different fruit types. These findings add a novel step into our understanding of the 52

interactions between the gut microbial communities and B. dorsalis and provide avenues 53

for developing control strategies to limit the devastative incidence of the fly. 54

55

Introduction 56

Gut bacteria and fruit flies are thought to have a close evolutionary and biological 57

relationships with a wide degree of interdependence [1-3]. This interaction shapes the 58

host fitness and the abundance of gut microbiota [2], and is modulated by the availability 59

and nutritional quality of host fruits. Different fruit differ in their suitability for fruit fly 60

larvae. If fruit fly larvae gain many of their primary nutritional needs from bacterial 61

break-down products, then maybe the quality of fruit for fruit fly larvae depends on how 62

good the host is for bacterial growth and survival, with the fruit’s value to fruit fly larvae 63

being of secondary importance. 64




4

Some tropical fruits such as banana and apple have been shown to contain a good 65

amount of pectin intensively used as food additive [4-6]. Pectin from fruit has been 66

reported to be involved in defense mechanisms against external aggressions and plant 67

pathogens [7]. In addition, fruits are generally low in protein content. The larvae of the 68

oriental fruit fly Bactrocera dorsalis require a large amount of sugar and nitrogen content 69

diet for their growth [8]. In order to develop in nutritionally poor fruits, gut bacteria may 70

come into play in supplying missing sugar metabolites and marginal amino acid residues 71

from the various fruit types to sustain the larval development [9,10]. 72

The ability of gut bacteria to modify the diet composition [11] and host 73

transcriptome [12] allows aphids to survive on plants with poor nutrient values [13], 74

allows the higher termites Nasutitermes takasagoensis to digest cellulose from wood 75

[14], Bombyx mori to degrade pectin from mulberry leaves [15] and Anoplophora 76

glabripennis to feed on multiple hosts by disrupting the expression levels of numerous 77

genes involved in digestion and detoxification [12]. Insects are exposed to several 78

ecological adversities (predators and abiotic factors) and the gut bacterial isolates were 79

conjected to help B. dorsalis to making beneficial compromises between the feeding 80

time, nutrient ingestion and fitness [16,17]. These ecological compromises (nutritional 81

tradeoffs) result in differences in B. dorsalis gut microbiome, whose community structure 82

and diversity vary or shift from one fruit to the other in response to the nutritional 83

adaptation, not only for the host but also for the gut microbiotas. In this event, B. dorsalis 84

may rely on its gut-associated bacteria for fitness and survival, which in turn may shape 85

the foraging of the host according to their nutritional requirements [16,17]. 86

In this study, we evaluated the extent of the interactions between the diets and the 87




5

microbial communities on larval development. We predicted that the suitability of a fruit 88

type for fruit fly larval development is positively correlated with the ability of that fruit to 89

act as a substrate/media for fruit fly symbiotic bacterial growth. The method consisted of 90

allowing a single female fly to lay eggs on five different fruit types for 24h and 91

monitoring the larval population dynamics across five generations. At each generation, 92

the larval development parameters (length, weight and development time) were evaluated 93

within and between fruit types. We thus presumed that larval development parameters 94

would be highly enhanced in fruits, thus offering optimal nutrients for development [18]. 95

Data generated from this study allowed us to conject that the gut microbiome can shift its 96

population in response to nutritional adaptations of fly larvae. 97

Materials and methods 98

Insect rearing and maintenance 99

All experiments were conducted in a controlled environment (25±1.5°C, 65±10% 100

RH and 16:8 light: dark cycle). Newly emerged lab-reared flies were fed artificial full 101

diet consisting of Tryptone (25 g/L), Yeast extract (90 g/L), Sucrose (120 g/L), Agar 102

powder (7.5 g/L), Methyl-p-hydroxybenzoate (4 g/L), Cholesterol (2.3 g/L), Choline 103

chloride (1.8 g/L), Ascorbic acid (5.5 g/L) in 1 L of distilled water . 104

Host fruits preparation 105

Ripe fruits (mango, banana, apple, citrus, and tomato) were bought from a local 106

supermarket (Wuhan, China) and surface sterilized by soaking them in 2% sodium 107

hypochlorite for 20 minutes and rinsed with deionized distilled water. The sterilized fruits 108

were air dried under a laminar flow hood to avoid airborne contamination. Five of each 109




6

type of fruit were needed for the bioassays (therefore the total number of fruit were 5 x 5 110

= 25), and five replicates were used for each fruit type (125 fruits in total). 111

Bioassays 112

One mating couple (11-15 day-old) was extracted from the rearing cage and 113

introduced into a new cage (15x15x15cm). The different fruits were separately added into 114

the cage in which the single female was allowed to lay eggs for 24 hours before replacing 115

the fruit with another one. The process was repeated for the five fruit types (Fig 1). 116

To determine how bacteria and fruit type affect larval development, the larval 117

developmental parameters were monitored in all fruit types following the oviposition, 118

starting from the parental female. Fruits bearing eggs were incubated individually in 1L 119

transparent plastic cups and sealed with a fine mesh (25±1.5°C, 65±10% RH and 16:8 120

light: dark cycle). Eggs and developing larvae were extracted periodically from each fruit 121

type at 3, 6 and 9 days (corresponding to the 1st, 2nd, and 3rd instar larvae). The body 122

length of the extracted larvae (to the nearest 0.03 mm), the larval dry weight and 123

developmental time were measured before the larvae were anesthetized in cold 95% 124

ethanol. All Larvae (15) were kept in 95% ethanol and were either used immediately or 125

preserved at −80°C until used for further analyses (measurement of larval length, dry 126

weight and microbiome analyses). The remaining larvae from each fruit were allowed to 127

develop till adult emergence. The adults that emerged were maintained under artificial 128

diet as described above till the flies reached sexual maturity. A single female from the 129

cohort of each fruit was allowed to lay eggs on new fruit to produce subsequent 130

generation and the process was repeated till five generations. 131




7

Microbial analyses 132

Genomic DNA extraction and amplicon generation 133

Total genome DNA from samples was extracted using the CTAB/SDS method 134

[19]. DNA quality was checked on 1% agarose gels using a ladder, and the purity was 135

checked as above. DNA was diluted to 1ng/µL with sterile distilled water. 136

The V1-V3 variable region of the bacterial 16S rDNA gene was amplified to construct a 137

gene library using bar-coded and broadly conserved primers for the PCR reaction: 138

27F_5′ CCTATCCCCTGTGTGCCTTGGCAGTCTCAGAGAGTTTGATCCTGGCTC139

AG-3′, and 533R_5′-140

CCATCTCATCCCTGCGTGTCTCCGACGACTNNNNNNNNTTACCGCGGCTGCT 141

GCAC -3′ [20]. This contains the A and B sequencing adaptors (454 Life Sciences) to 142

facilitate pooling, segregation, sequencing and amplification of ~536 bp region of the 143

mentioned gene (“Ns” represent the 8 nt barcode sequence for multiple samples while 144

the underlined sequences represent the A-adaptor). All PCR reactions were carried out 145

with Phusion® High-Fidelity PCR Master Mix with GC Buffer (New England Biolabs, 146

Ipswich, MA, USA) and high-fidelity polymerase (New England Biolabs). 147

The PCR conditions were as follows: initial denaturation at 94°C for 2 min, 148

followed by 30 cycles of denaturation at 94°C for 1 min, annealing at 60°C for 30 s, and 149

extension at 72°C for 1 min and a final extension step of 10 min at 72°C. PCRs of 150

DNA-free samples were run to check potential contamination of buffers and primers 151

[21]. PCR amplicons were later subjected to electrophoresis on a 2% agarose gel, 152

stained with ethidium bromide, and the targeted fragment size (400-450bp) was 153




8

extracted, purified with Qiagen Gel Extraction Kit (Qiagen, Germany) and quality 154

checked before pyrosequencing. Amplicon libraries were generated using TruSeq® 155

DNA PCR-Free Sample Preparation Kit (Illumina, USA) following the manufacturer's 156

recommendations and index codes were added. Library quality was assessed on the 157

Qubit@ 2.0 Fluorometer (Thermo Scientific) and Agilent Bioanalyzer 2100 system 158

using a DNA1000 lab chip (Agilent), respectively. The library was then amplified by 159

emulsion PCR before 454 pyrosequencing was performed from the A-end on an 160

Illumina HiSeq2500 platform using a GS FLX Titanium system according to the 161

manufacturer's instructions (Roche 454 Life Sciences) and 250 bp paired-end reads 162

were generated. 163

Bioinformatics analyses 164

Paired-end reads were assigned to samples based on their unique barcode and 165

truncated by cutting off the barcode and primer sequence. Paired-end reads were 166

merged using FLASH 1.2.7 [22]. Quality filtering on the raw tags was performed under 167

specific filtering conditions to obtain the high-quality clean tags [23] using QIIME1.7.0 168

[24] quality controlled process (Edgar et al. 2011). The tags were compared with the 169

reference database using UCHIME algorithm to detect and remove chimera sequences 170

before obtaining active tags [25]. 171

Operational taxonomic units (OTUs) analyses were performed by UPARSE 172

7.0.1001 [26]. Sequences with ≥97% similarity were assigned to the same OTUs. The 173

representative sequence for each OTU was screened for further annotation, and the 174

GreenGene Database [27] was used based on the Ribosomal Database Project (RDP) 175

classifier 2.2 algorithm [28] to annotate the taxonomic information. OTUs abundance 176




9

information was normalized using a standard of sequence number corresponding to the 177

sample with the least sequences. The Good's coverage, the abundance-based coverage 178

estimator (ACE), the bias-corrected Chao1 richness estimator, the jackknife estimator of 179

species richness, and the Shannon-Weaver and Simpson diversity indices, were 180

calculated with the Mothur package. 181

Statistical analyses 182

All data on developmental parameters (larval length, larval dry weight and 183

development time) were tested for homogeneity of variances using Levene’s tests. The 184

crucial factors that affect the developmental parameters were checked by using the 185

regression model (SPSS) with host fruits and experimental generations as effects. The 186

one-way analysis of variance (ANOVA) was used to analyze differences in 187

developmental data and sequencing data. New Duncan’s Multiple Range Test (NDMRT) 188

test at P = 0.05 significance, was used for mean separations within and between samples. 189

The results of sequencing parameters were presented as means of the three biological 190

replicates. Statistical analyses were carried out using SPSS 20.0 software (Statsoft Inc, 191

Carey, J, USA). OriginPro 8.5.1 software was used to construct graphs. We used the 192

Analysis of similarity (ANOSIM) to determine whether the differences of B. dorsalis 193

bacterial community between treatments (P, A1, A5, B1 & B5) are significantly higher 194

than those in the group and whether the grouping is meaningful. 195

Results 196

Larval development parameters 197

Larval length 198

The larval length and dry weight were not affected by the experimental 199




10

generations (Regression Model, F = 146.194; df = 4; R2 = 0.857; t = 5.013; P = 0.236 and 200

F = 257.234; df = 4; R2 = 0. 798; t = 7.835; P = 0.184, respectively, but were significantly 201

affected by fruit types and larval stages (Regression Model, F = 71.217; df = 4; R2 = 202

0.958; t = 1.464; P ˂ 0.0001 and F = 71.217; df = 2; R2 = 0.958; t = 1.464; P ˂ 0.0001, 203

respectively) (Fig 2 & 3). 204

The larval growth increased significantly across stages in all the experimental 205

fruits and generations (Regression Model, F = 116.643; df = 2, 4; R2 = 0.837; t = 7.322; P 206

˂ 0.0001) and the highest larval length recorded in Banana, peach and mango in 207

comparison to apple and tomato (ANOVA, F = 70.007; df = 4; P ˂ 0.0001 and F = 208

139.185; df = 4; P ˂ 0.001, respectively) (Fig 2). Except for tomato whose larval 209

development stopped at the first instar of the fifth generation, all the host fruits allowed 210

complete larval development up to the fifth generation (Fig 2). 211

Larval weight 212

The larval growth positively correlated with larval dry weight across generations 213

in all host fruits (R2 = 0. 978; P ˂ 0.0001) (Fig 3) and the two parameters were 214

proportional to each other. Moreover, a paired analysis of larval growth and dry weight 215

revealed a significant interaction between the two parameters in selecting the suitability 216

of host fruits by B. dorsalis (F = 40.074; df = 1, 4; R2 = 0.9984; P ˂ 0.0001) (Fig 3). 217

Developmental growth time of larvae 218

The larval development (length and weight) was inversely proportional to 219

developmental duration (time from oviposition to adult emergence). The higher the larval 220

length, the shorter the development time and vice versa. The larval development time was 221

significantly extended in apple and tomato (F = 78.384; df = 1, 4; P ˂ 0.001), but was 222




11

shortened in banana (F = 97.154; df = 4; P ˂ 0.0001) in comparison to peach and mango 223

(Table 1). 224

Gut microbial analyses 225

The diversity of operational taxonomic units (OTUs) 226

A total of 653,932 raw reads were generated from all samples: 38,467±6564 227

reads on average for each sample and 30,637 reads based on the minimum number of 228

trimmed sequences from each sample. The OTU composition varied and increased 229

significantly across host fruits and experimental generations compared to the parental fly. 230

Overall, 64 OTUs were detected in the original parent (P), 252 and 427 OTUs in Banana 231

at F1 and F5 generations (B1 & B5), respectively, and 421 and 1,059 in Apple at F1 and F5 232

generations (A1 & A5), respectively. The reads length average was 426±4 bp, and the 233

minimum and maximum number of sequence read per OTU were 201 bp and 503 bp, 234

respectively. Although the majority of OTUs from B5 were shared among other samples, 235

it contained the highest number of OTUs per sample. Only 27 core OTUs were detected 236

in all samples (Fig 4). 237

Bacterial richness and diversity 238

The community diversity (Shannon and Simpson) and richness (sobs, Chao, and 239

ace) evaluated at 97% similarity, showed different comparative trends in the prediction of 240

the number and diversity of OTUs from all samples. The Shannon diversity index 241

provides not only species richness (i.e., the number of species present) but how the 242

abundance of each species is distributed (the evenness of the species) among all the 243

species in the community. Here, we recorded significantly higher bacterial diversity 244

across generations in both tested fruits compared to the parental fly (P) (ANOVA, F = 4, 245




12

df = 11.173, P = 0.017) (Table 1). For example, the Shannon indices of banana were 246

0.22±0.26 and 2.5±0.13 at F1 and F5, respectively (t-test, P = 0.00017±0.0), while those 247

of apple were 1.17±0.23 and 2.5±0.13 at F1 and F5, respectively (t-test, P = 0.0118±0.02). 248

The comparison between fruits revealed higher bacterial community diversity in apple at 249

F1 (Shannon = 1.17±0.23) compared to banana at the same generation (Shannon = 250

0.22±0.26) (t-test, P = 0.0095) (Table 1). However, at F5, the community diversity was 251

similar in both fruits (t-test, P = 0.3782±0.38). 252

Rarefactions 253

The rarefaction is a computational analysis of species accumulation based on 254

the repeated re-sampling of all clusters. Based on the good coverage (>0.999) (Table 1), 255

almost all the 16S sequences have been annotated. Moreover, the flattening of the 256

rarefaction curves is an indication that the sampling depth was sufficient enough to detect 257

all OTUs from our samples and an additional sampling effort would only produce few 258

extra OTUs (Fig 5). 259

Bacterial community composition 260

The original bacterial community was represented by two phyla only: 261

Proteobacteria accounted for 95.67%, and Firmicutes accounted for 4.19%. The structure 262

of the vast majority of the gut bacterial community did not change at the first generation 263

(F1) of both host fruits, but there was an emergence of Bacteroidetes which was not 264

detected from the original community. At F1, Proteobacteria represented 96.04% and 265

99.08% in apple and banana, respectively, and Firmicutes represented 1.64% and

13

and 47.36% in apple and banana, respectively, in comparison with the to F1 and P, while 269

the populations of Firmicutes increased considerably at F5 (49.82% and 37.11% in apple 270

and banana, respectively). Also, Actinobacteria populations were detected in both fruits 271

(Fig 6). The heatmap showing the relative abundance of taxa in different samples and 272

groups on a certain taxonomic level can be found in Fig S9. 273

Twenty-one bacterial families in total were annotated from all samples, and the 274

number of families increased across generations. Enterobacteriaceae (91.75%) were the 275

most dominant families in the parental female (P), followed by Pseudomonadaceae 276

(3.81%) and Enterococcaceae (2.89%) in lesser proportions. At F1, Pseudomonadaceae 277

(67.31%) and Enterobacteriaceae (27.07%) were highly detected in apple while 278

Enterobacteriaceae (97.24%) was predominant in banana (Fig S1, Supplementary). At F5, 279

the bacterial families were more diversified in both fruits (12 and 17 in apple and banana, 280

respectively). For instance, Leuconostocacceae (22.66%), Streptococcaceae (19.38%) and 281

Acetobacteriaceae (17.64%) were detected in higher proportions in apple compared to F1, 282

while Enterobacteriaceae (27.62%), Bacillaceae (11.93%), Streptococcaceae (10.84%) 283

and Pseudomonadaceae (10.04%) were highly represented in banana compared to their 284

proportions at F1 (Fig S1, Supplementary). 285

At the genus level, Enterobacter (85.05%) was dominant in the parental female. 286

At F1, Pseudomonas (67.31%) and Enterobacter (26.68%) were highly represented in 287

apple while Morganella (96.82%) was the predominant genus in banana. However, at 288

F5, Lactococcus (19.12%), Fructobacillus (18.58%), Gluconobacter (16.99%) and 289

Morganella (22.23%) emerged abundant in apple, while Morganella (22.23%), Bacillus 290

(11.90%), Lactococcus (10.74%) and Pseudomonas (10.64%) were dominantly 291




14

represented in banana. Overall, in apple, Enterobacter proliferated significantly at F5 292

compared to Pseudomonas. However, in banana, Morganella was consistent across 293

generations (Fig S1, Supplementary). 294

Bacterial structure distribution 295

On the basis of Bray-Curtis distance algorithm, the gut bacterial community 296

structure varied significantly between host fruits and across generations in all samples 297

(Bray-Curtis distance statistics, A = 0.009014; P = 0.0012 & A = 0.000852; P = 0.01) (Fig 298

7. However, the clustering of different samples into groups indicates that the bacterial 299

communities within groups are highly similar (ANOSIM: R = 0.4259; P = 0.019) (Fig 7). 300

Discussion 301

Gut bacteria and fruit flies share close evolutionary relationships which mutually 302

influence the physiology and ecological adaptations of both parties. In this interaction, 303

the larval stages play a dominant role in the survival of the insect whose capacity to stand 304

across multiple generations depends on host suitability for larval growth sustained by its 305

gut microbiome. If fruit fly larvae gain many of their primary nutritional needs from 306

bacterial metabolic activity, then maybe the quality of fruit for fruit fly larvae depends on 307

how good it is for bacterial growth and larval development. 308

In this study, we assessed how host suitability might shape the larval performance 309

and gut bacterial growth over time. A single female fly was allowed to lay eggs on five 310

different host fruit types and we monitored the larval population dynamics across five 311

generations and bacterial community structure at first and fifth generations. 312

Host suitability for larval development 313




15

The larval development parameters (length, dry weight, and development duration) 314

increased across larval stages depending on the suitability of host fruits. The highest 315

larval performance was recorded in banana, mango, and peach, while larvae that 316

developed in apple and tomato had the shortest length and dry weight, and extended the 317

development duration across generations. Within the context of adaptive abilities of the 318

fly larvae, these results suggest that the host fruit quality (capacity to offer a readily 319

available source of nutrients) is a critical factor that determines the completion of larval 320

development and its propensity over time [29]. While the adult flies require a diet 321

consisting of protein and sugar for their maintenance, larvae must feed on high sugar 322

content diet for their growth and development [8,10]. Therefore, a suitable substrate for 323

larval growth should be able to offer the necessary carbohydrates to fuel the urgent 324

nutritional requirements of larvae. Larvae under continuous development in tomato did 325

not reach the second instar of their development at the fifth generation. The abortion of 326

larvae may partly be due to the high moisture content in the tomato flesh that led to the 327

drowning of the larvae, and death. In addition, the lack of sufficient sugar metabolites, 328

and free amino acids in tomato may be another cause of the abortion of the larvae over 329

time [30]. 330

Host suitability for bacterial growth 331

The gut microbiome plays a prominent role in the nutritional adaptations of insects 332

[18]. When the fly is found in a nutritionally imbalanced environment in terms of 333

nutrients availability, the metabolic activity of gut bacteria may constitute an additional 334

source of nutrients, for their ability to provide amino acids, carbohydrates, vitamins and 335

digestive enzymes to foster and reinforce the adaptive capacity of the host fly [18]. Many 336




16

previous studies have reported the importance of gut microbiota in host nutrition and 337

physiology [9,10,31-33], and the microbial community divergence at different life stages 338

and geographic areas have been established [8]. However, how the gut bacterial 339

community of larvae developing on different host fruits varies across generations to 340

maintain optimal larval development is not fully understood. 341

Here, we explored the bacterial community structure and composition of larvae 342

developing in banana (suitable substrate for larval growth and development) and in apple 343

(hostile substrate), highlighted the core bacteria in each fruit and evaluated the extent of 344

their variations across generations compared to the single parental community. Very 345

similar structure of bacterial communities was observed between the parental female and 346

the offspring of the first generation (F1), which was dominated by Proteobacteria (≥95%). 347

The Proteobacterial populations decreased significantly, and Firmicutes emerged as the 348

dominant taxa at F5. This could be understood at two scales. First, the similarity of 349

bacterial community structure at F1 with the parental fly (but different at F5) could imply 350

that the community structure changes over time to allow the larvae to adapt in an 351

environment with either no food available or less suitable food for their optimal growth. 352

Secondly, although originated from the same parent, the differences observed between 353

fruits at F5 may come at the cost of the different degree of suitability of the substrate and 354

the survival emergency in the long run. Therefore, Proteobacteria and Firmicutes 355

appeared as the driving forces of this behavior due to their prevalence and persistence 356

over time [8,34]. Similarly, the stability of these same bacterial phyla (Proteobacteria and 357

Firmicutes) were shown to be responsible for maintaining optimal larval development 358

across life stages of B. minax [29]. 359




17

Furthermore, when compared to the parental female, huge bacterial community diversity 360

and richness was recorded across fruits (higher in apple at F1 but similar in both fruits at 361

F5) and across generations (higher at F5 in comparison to F1). This finding puts to the 362

light the plasticity of gut bacteria which can amend its population to help the host 363

circumventing challenges linked to the fruit phenology and nutrient availability. As 364

shown above, banana and apple offered different larval development facilities. While 365

banana allowed the larvae to develop optimally and grow faster (due to their high sugar 366

content and vitamins), apple extended the larval development time and reduced the 367

measured development parameters. The quality of ingested food influences the structure 368

and composition of gut microbiota in insects, some bacterial taxa are advantaged and 369

increase in size while others are disfavored and maintain their population at the survival 370

threshold (Douglas, 2015). The higher bacterial diversity (with the abundance of the 371

family Pseudomonadaceae and the genera Pseudomonas) in apple at F1 could be an 372

indication that, at early generation, either the apple offered all the required nutrients for 373

bacterial normal growth, or these gut bacteria possess the ability to digest potential 374

recalcitrant dietary constituents from the fruit. Previously, symbiotic bacteria of the 375

genera Burkholderia and Parabacteroides were suggested to mediate the degradation of 376

hemicellulose and the recycling nitrogen in Melolontha hippocastani [35]. At F5, 377

Enterobacteriaceae (genus Enterobacter) became the most abundant bacterial community 378

in apple, while the same family Enterobacteriaceae (genus Morganella) persisted across 379

generations in banana. This suggests to some extent the dependence of the larvae upon 380

Enterobacteriaceae for their optimal development across generations. In this fashion, the 381

larvae can directly use the available resources from the fruits, or use the byproducts of 382




18

bacterial metabolic activities to compensate the lack of nutrients from their diet [36]. 383

Therefore, gut bacteria and the substrate suitability are crucial to maintaining a balanced 384

food web and the survival of the host across generations. 385

Overall, the variability of the gut bacterial community structure and composition in 386

response to different host fruits observed in this study could also be driven by, but not 387

limited to, the changes in the gut transcriptome that might have induced a regulation 388

mechanisms of genes linked to nutrition, growth and detoxification. However, additional 389

experimental steps are needed to provide support for such a conjecture. Such experiments 390

might, for example, include studies where flies developed on one type of fruit for many 391

generations on a particular fruit that gave smaller larvae would continue to show poorer 392

development when transferred to a different fruit. The use of axenic fruit and flies 393

represent another way forward to untangle these interactions. At the very least, our 394

findings add another layer of understanding on the various bacterial compromises that 395

enable B. dorsalis to feed in a broad range of host fruits. This may explain (but not 396

limited to) the invasive polyphagous state of the fly. 397

Conclusion 398

In this work, we evaluated how different host fruits can act as a suitable substrate 399

for larval development and gut bacterial growth. By allowing a single female fly to lay 400

eggs on different host fruit, we monitored the bacterial population with regards to the 401

parental ones and determined the patterns of their variations across five generations. The 402

results show that the development of larvae depends on the ability of fruits to offer a 403

suitable substrate or media for gut bacterial growth. As such, banana, mango and peach 404

allowed optimal larval development across generations while apple and tomato were less 405




19

suitable substrate for the larval growth. Overall, the bacterial community structure and 406

composition was similar at the early stage of the exposure to the fruits (with a 407

predominance of Proteobacteria), but the variations arose over time depending on how 408

suitable the substrate was at availing nutrients for the larvae. This study adds a new stair 409

of understanding of the effects of interactions between gut bacteria and the quality of 410

substrate (fruits) on B. dorsalis development. 411

412

Data accessibility 413

The metagenomic data are pending submission to sequence reads archive (SRA) of 414

GenBank. The project references will be provided as soon as possible. 415

Funding 416

This study was funded by the National Natural Science Foundation of China 417

(31661143045), International Atomic Energy Agency (CRP No. 17153 and No. 18269), 418

Agricultural public welfare industry research supported by Ministry of Agriculture of 419

People’s Republic of China (201503137) and the Fundamental Research Funds for the 420

Central Universities (2662015PY148). 421

Competing interests 422

The authors declare that they have no competing interests. 423

Author’s contribution 424

CYN conceived and designed the study. MA conducted the experiments and wrote the 425

first draft of the manuscript. AM, XR and YW analyzed metagenomics data. XQ and SC 426

helped in statistical analyses. AN edited and revised the manuscript. All authors read and 427

approved this submission. 428




20

Acknowledgements 429

Authors are grateful to two anonymous reviewers for their insightful comments on the 430

early version of this manuscript. 431

References 432

1. Engel P, Moran NA (2013) The gut microbiota of insects–diversity in structure and 433

function. FEMS Microbiology Reviews 37: 699-735. doi: 10.1111/1574-434

6976.12025 435

2. Gould AL, Zhang V, Lamberti L, Jones EW, Obadia B, Korasidis N, et al. (2018) 436

Microbiome interactions shape host fitness. Proceedings of the National Academy 437

of Sciences 115: E11951-E11960. doi: 438

3. Lauzon C, E. Potter S, J. Prokopy R (2003) Degradation and detoxification of the 439

dihydrochalcone Phloridzin by Enterobacter agglomerans, a bacterium associated 440

with the apple pest, Rhagoletis pomonella (Walsh) (Diptera: Tephritidae). 953-962 441

p.10.1603/0046-225X-32.5.953 442

4. Khamsucharit P, Laohaphatanalert K, Gavinlertvatana P, Sriroth K, Sangseethong K 443

(2018) Characterization of pectin extracted from banana peels of different 444

varieties. Food Science and Biotechnology 27: 623-629. doi: 445

5. Maneerat N, Tangsuphoom N, Nitithamyong A (2017) Effect of extraction condition on 446

properties of pectin from banana peels and its function as fat replacer in salad 447

cream. Journal of Food Science and Technology 54: 386-397. doi: 448

6. Virk BS, Sogi DS (2004) Extraction and characterization of pectin from apple (Malus 449

Pumila. Cv Amri) peel waste. International journal of food properties 7: 693-703. 450

doi: 451




21

7. Voragen AGJ, Coenen G-J, Verhoef RP, Schols HA (2009) Pectin, a versatile 452

polysaccharide present in plant cell walls. Structural Chemistry 20: 263. doi: 453

8. Zhao X, Zhang X, Chen Z, Wang Z, Lu Y, Cheng D (2018) The divergence in bacterial 454

components associated with Bactrocera dorsalis across developmental stages. 455

Frontiers in Microbiology 9. doi: 10.3389/fmicb.2018.00114 456

9. Ben-Yosef M, Pasternak Z, Jurkevitch E, Yuval B (2014) Symbiotic bacteria enable 457

olive flies (Bactrocera oleae) to exploit intractable sources of nitrogen. Journal of 458

Evolutionary Biology 27: 2695-2705. doi: 10.1111/jeb.12527 459

10. Ben-Yosef M, Pasternak Z, Jurkevitch E, Yuval B (2015) Symbiotic bacteria enable 460

olive fly larvae to overcome host defences. Royal Society Open Science 2: 461

150170. doi: 10.1098/rsos.150170 462

11. Bourtzis K, Miller TA (2003) Insect symbiosis. Boca Raton: CRC Press. 463

12. Scully ED, Geib SM, Mason CJ, Carlson JE, Tien M, Chen H-Y, et al. (2018) Host-464

plant induced changes in microbial community structure and midgut gene 465

expression in an invasive polyphage (Anoplophora glabripennis). Scientific 466

Reports 8: 9620. doi: 10.1038/s41598-018-27476-0 467

13. Douglas A (1998) Nutritional interactions in insect-microbial symbioses: aphids and 468

their symbiotic bacteria Buchnera. Annual review of entomology 43: 17-37. doi: 469

14. Tokuda G, Watanabe H, Matsumoto T, Noda H (1997) Cellulose digestion in the 470

wood-eating higher termite, Nasutitermes takasagoensis (Shiraki): distribution of 471

cellulases and properties of endo-β-1, 4-gIucanase. Zoological Science 14: 83-94. 472

doi: 10.2108/zsj.14.83 473

15. Prem Anand AA, Vennison SJ, Sankar SG, Gilwax Prabhu DI, Vasan PT, Raghuraman 474




22

T, et al. (2010) Isolation and characterization of bacteria from the gut of Bombyx 475

mori that degrade cellulose, xylan, pectin and starch and their impact on 476

digestion. Journal of Insect Science 10: 107. doi: 10.1673/031.010.10701 477

16. Akami M, Andongma AA, Zhengzhong C, Nan J, Khaeso K, Jurkevitch E, et al. 478

(2019) Intestinal bacteria modulate the foraging behavior of the oriental fruit fly 479

Bactrocera dorsalis (Diptera: Tephritidae). PLOS ONE 14: e0210109. doi: 480

10.1371/journal.pone.0210109 481

17. Akami M, Ren X-M, Qi X, Mansour A, Gao B, Cao S, et al. (2019) Symbiotic 482

bacteria motivate the foraging decision and promote fecundity and survival of 483

Bactrocera dorsalis (Diptera: Tephritidae). BMC Microbiology 19: 229. doi: 484

10.1186/s12866-019-1607-3 485

18. Douglas AE (2009) The microbial dimension in insect nutritional ecology. Functional 486

Ecology 23: 38-47. doi: 10.1111/j.1365-2435.2008.01442.x 487

19. William S, Feil H, Copeland A (2012) Bacterial genomic DNA isolation using CTAB. 488

Sigma 50: 6876. doi: 489

20. Andongma AA, Wan L, Dong YC, li P, Desneux N, White JA, et al. (2015) 490

Pyrosequencing reveals a shift in symbiotic bacteria populations across life stages 491

of Bactrocera dorsalis. Scientific Reports 5: 9470. doi: 10.1038/srep09470 492

21. Yun JH, Roh SW, Whon TW, Jung MJ, Kim MS, Park DS, et al. (2014) Insect gut 493

bacterial diversity determined by environmental habitat, diet, developmental 494

stage, and phylogeny of host. Applied and Environmental Microbiology 80: 5254-495

5264. doi: 10.1128/aem.01226-14 496

22. Magoč T, Salzberg SL (2011) FLASH: fast length adjustment of short reads to 497




23

improve genome assemblies. Bioinformatics 27: 2957-2963. doi: 498

10.1093/bioinformatics/btr507 499

23. Bokulich NA, Subramanian S, Faith JJ, Gevers D, Gordon JI, Knight R, et al. (2013) 500

Quality-filtering vastly improves diversity estimates from Illumina amplicon 501

sequencing. Nature methods 10: 57-59. doi: 10.1038/nmeth.2276 502

24. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, et 503

al. (2010) QIIME allows analysis of high-throughput community sequencing data. 504

Nature methods 7: 335-336. doi: 10.1038/nmeth.f.303. 505

25. Haas BJ, Gevers D, Earl AM, Feldgarden M, Ward DV, Giannoukos G, et al. (2011) 506

Chimeric 16S rRNA sequence formation and detection in Sanger and 454-507

pyrosequenced PCR amplicons. Genome Research 21: 494-504. doi: 508

10.1101/gr.112730.110 509

26. Edgar RC (2013) UPARSE: highly accurate OTU sequences from microbial amplicon 510

reads. Nature Methods 10: 996. doi: 10.1038/nmeth.2604 511

27. DeSantis TZ, Hugenholtz P, Larsen N, Rojas M, Brodie EL, Keller K, et al. (2006) 512

Greengenes, a chimera-checked 16S rRNA gene database and workbench 513

compatible with ARB. Appl Environ Microbiol 72: 5069-5072. doi: 514

10.1128/aem.03006-05 515

28. Wang Q, Garrity GM, Tiedje JM, Cole JR (2007) Naive Bayesian classifier for rapid 516

assignment of rRNA sequences into the new bacterial taxonomy. Applied and 517

Environmental Microbiology 73: 5261-5267. doi: 10.1128/aem.00062-07 518

29. Yao Z, Ma Q, Cai Z, Raza MF, Bai S, Wang Y, et al. (2019) Similar shift patterns in 519

gut bacterial and fungal communities across the life stages of Bactrocera minax 520




24

larvae from two field populations. Frontiers in Microbiology 10: 2262. doi: 521

30. Bialczyk J, Lechowski Z, Dziga D, Molenda K (2005) Carbohydrate and free amino 522

acid contents in tomato plants grown in media with bicarbonate and nitrate or 523

ammonium. Acta Physiologiae Plantarum 27: 523-529. doi: 10.1007/s11738-005-524

0058-7 525

31. Andongma AA, Wan L, Dong X-p, Akami M, He J, Clarke AR, et al. (2018) The 526

impact of nutritional quality and gut bacteria on the fitness of Bactrocera minax 527

(Diptera: Tephritidae). Royal Society Open Science 5: 180237. doi: 528

10.1098/rsos.180237 529

32. Ben-Yosef M, Jurkevitch E, Yuval B (2008) Effect of bacteria on nutritional status 530

and reproductive success of the Mediterranean fruit fly Ceratitis capitata. 531

Physiological Entomology 33: 145-154. doi: 10.1111/j.1365-3032.2008.00617.x 532

33. Khaeso K, Andongma AA, Akami M, Souliyanonh B, Zhu J, Krutmuang P, et al. 533

(2017) Assessing the effects of gut bacteria manipulation on the development of 534

the oriental fruit fly, Bactrocera dorsalis (Diptera; Tephritidae). Symbiosis 74: 97-535

105. doi: 10.1007/s13199-017-0493-4 536

34. Aksoy E, Telleria EL, Echodu R, Wu Y, Okedi LM, Weiss BL, et al. (2014) Analysis 537

of multiple tsetse fly populations in Uganda reveals limited diversity and species-538

specific gut microbiota. Applied and Environmental Microbiology 80: 4301. doi: 539

10.1128/AEM.00079-14 540

35. Alonso-Pernas P, Bartram S, Arias-Cordero EM, Novoselov AL, Halty-deLeon L, 541

Shao Y, et al. (2017) In vivo isotopic labeling of symbiotic bacteria involved in 542

cellulose degradation and nitrogen recycling within the gut of the forest 543




25

cockchafer (Melolontha hippocastani). Frontiers in Microbiology 8: 1970. doi: 544

10.3389/fmicb.2017.01970 545

36. Panizzi AR, Parra JR (2012) Insect bioecology and nutrition for integrated pest 546

management: CRC press. 547

548

Figure captions 549

Figure 1 Experimental design. The highlighted ones are samples used for microbial 550

analyses. 551

Figure 2 Effect of host fruits suitability on B. dorsalis larval growth across five 552

generations. 553

Figure 3 Effect of host fruits suitability on B. dorsalis larval dry weight across five 554

generations 555

Figure 4 Venn diagram of OTU distribution of sample-specific and shared OTUs. The 556

different colored circles represent different samples or groups. The overlapped areas 557

represent common species among different samples or groups; non-overlapped areas 558

represent unique species in each sample or group. 559

Figure 5 Multiple samples rarefaction curves based on 16S rRNA genes sequencing. The 560

X-axis represents the number of randomly selected sequences; The Y-axis represents the 561

number of species (such as Sobs) or diversity indices (such as Shannon) of each sample. 562

Legends P: parental female; A1 & A5: Apple at first and fifth generations; B1 & B5: 563

Banana at first and fifth generations. 564




26

Figure 6 Bacterial community compositions at the phylum level as revealed by high 565

throughput sequencing. Different colored bars represent different species, and the length 566

of the column represents the proportion of the species. Legends: P: parental female; A1 & 567

A5: Apple at first and fifth generations; B1 & B5: Banana at first and fifth generations. 568

Figure 7 Principal Coordinates Analysis (PCoA) showing the variations of the gut 569

bacterial community of B. dorsalis as affected by time and host fruits. Note: The X and Y 570

axes represent two selected principal coordinate components. Legends: P: parental 571

female; A1 & A5: Apple at first and fifth generations; B1 & B5: Banana at first and fifth 572

generations. The percentage “%” indicates the contribution proportion of each principal 573

coordinate component to the sample composition variance. The scales on X-axis and Y-574

axis are relative distance with no practical significance. Different groups are marked in 575

different colors and shapes. Close samples have a similar community composition 576

structure. 577

578

Table captions 579

Table 1 Biological cycle duration of B. dorsalis (from first instar larvae to first instar 580

larvae) as affected by host fruit suitability 581

Table 2 Alpha diversity indices, showing the diversity and species richness of gut 582

symbionts of B. dorsalis as affected by host fruits and experimental generations. 583

584

585




27

586

587

588

589

590

591

592

593

List of tables 594

Table 1 Biological cycle duration of B. dorsalis (from first instar larvae to first instar 595

larvae) as affected by host fruit suitability 596

Host fruits Developmental duration (days)

F1 F2 F3 F4 F5

Banana 28±3c 27±1

c 28±2

c 27±3

c 27±3

c

Apple 36±4a 36±2

a 35±3

a 34±5

a 35±4

a

Peach 31±2b 30±2

b 31±2

b 29±3

b 30±2

b

Mango 30±2b 28±1

c 29±2

b 31±2

b 30±3

b

Tomato 33±3a 31±2

b 32±2

ab 30±3

b 31±2

b

Means followed by different letters in the same column and row are significantly 597

different according to ANOVA and LSD test (P < 0.05). 598

599

Table 2 Alpha diversity indices, showing the diversity and species richness of gut 600

symbionts of B. dorsalis as affected by host fruits and experimental generations. 601

Samples Sobs* Shannon Simpson ACE Chao1 Coverage




28

P 69e 1.2236

b 0.4126

b 262 133

d 0.99

A1 188±29c 1.17±0.23

b 0.51±0.01

b 204±21 202±25

c 0.99±0

A5 500±119a 2.86±0.63

a 0.24±0.07

c 525±127 534±128

a 0.99±0

B1 108±96d 0.22±0.26

c 0.94±0.08

a 182±70 150±93

d 0.99±0

B5 203±40b 2.5±0.13

a 0.18±0.05

c 210±40 210±38

b 0.99±0

P: parental female; A1 & A5: Apple at first and fifth generations; B1 & B5: Banana at 602

first and fifth generations. Different letters within columns are statistically different after 603

Student’s t-test at P < 0.05; (*): average number of species observed in each sample 604

(alpha diversity). 605




bactrocera dorsalis (diptera: tephritidae) mazarin akami · 1 1 host fruit as a suitable bacteria...

Documents