Bacterial Communities of Drosophila suzukii (Matsumura, 1931) (Diptera: Drosophilidae) Damaged in Strawberry in Turkey

Drosophila suzukii (Matsumura, 1931) (Diptera: Drosophilidae) is an invasive species originating from Southeastern Asia and spreads in a fast manner. It is among major threats in soft-shell fruit cultivation in the whole world. It was detected in 2014 in Turkey. According to international criteria, it is considered that it has the potential of threatening the fruit cultivation in Turkey where garden plants are grown widely. In this study, a total of 39 bacterial strains were isolated from 100 mature Drosophila suzukii individuals. Gram staining characteristics, catalase, oxidase and nitrate reductase activities and chitinase enzyme activities and hypersensitivity reaction of these strains were determined by using microscopical and visual inspection. The bacterial strains were identified according to their fatty acid methyl esters (FAME) analysis by using Sherlock Microbial Identification System (MIS). The identification test results of the bacterial strains were also confirmed by phylogenetic analysis and their closely related species based on the 16S rRNA sequence. The most abundant bacterial species were Paenibacillus alvei (31.57%) and Bacillus amyloliquefaciens (47.36%) according to the MIS and 16S rRNA sequence analysis results, respectively. According to the MIS results, a total of 6 strains identified as Paenibacillus alvei were identified as Bacillus amyloliquefaciens according to the 16S rRNA sequence analysis results. A total of three Paenibacillus macerans strains identified in MIS system were also identified as Bacillus amyloliquefaciens according to the 16S rRNA sequence analysis. Morphological and biochemical characteristics results of all of Bacillus amyloliquefaciens strains showed the some results. According to the 16S rRNA sequence analysis results, the other bacterial strains consist of 1 Bacillus atrophaeus (5.2%), 1 Bacillus safensis (5.2%), 1 Paenibacillus motobuensis (5.2%) and 1 Staphylococcus epidermidis (5.2%) strains. To our knowledge, this is the first study characterizing the bacterial communities of Drosophila suzukii.

Although the primary hosts of D. suzukii are cherry, sour cherry, strawberries, blackberries, raspberries and blueberries, a very wide spectrum of fruits can be affected. It can cause also serious damage to fruits such as fig, apricot, peach, plum, grape, medlar, greenhouse mandarin, kiwi, persimmon, and fallen or cracked apple and orange [2,4,23,24]. It is estimated that these losses have affected 14% of all potential fruit production worldwide [25].
Due to the wide range of host fruit selection and rapid spreading, it is stated that this species is an important pest that is likely to cause major losses to the European and 36 Bacterial Communities of Drosophila suzukii (Matsumura, 1931) (Diptera: Drosophilidae) Damaged in Strawberry in Turkey American fruit industry in the near future [3][4].
While other Drosophila species feed on rotten fruits, D. suzukii prefers newly ripening fruits on the tree, and using their saw-shaped ovipositor, female individuals lay their eggs inside the fruits prior to their maturation for harvesting, which all make this organism a very important agricultural pest [2]. While the larvae feed on the rich protein content of the flesh of the fruit, the synthesis of inherent ethylene is increased in areas where tissue integrity is compromised. Ethylene synthesis locally accelerates maturity and causes collapse/softening (rotting) of the flesh of the fruit. As a result, these products lose their market value within a short period of time. In addition, the wounds that these flies open on the fruit for laying their eggs lead to additional losses caused by pathogens, including fungi and bacteria [26].
Few studies have been carried out to identify the microflora of this pest. The first study that was conducted to identify the bacterial flora of D. suzukii revealed species belonging to the genera Gluconobacter and Acetobacter [27]. Another study reported that Wolbachia spp.
(Rickettsiales: Rickettsiaceae) could be used in biological control of this pest [28][29]. To our knowledge, there is not any study characterizing the microbial communities of D. suzukii.
The present study aims to determine the microflora of D. suzukii which continues to spread rapidly and cause economic losses.

Pest Samples
The study was conducted on 100 healthy adult D. suzukii individuals collected on July 25th-26th, 2016 in Erzurum province of Turkey using traps prepared with cider vinegar (Figure 1) in strawberry trial fields where neither insecticide nor fungicide was used. Collected adult individuals were put into tubes and brought to Atatürk University Plant Protection Department, Plant Clinical Laboratory.

Isolation of Bacterial Strains
Superficial sterilization was applied to D. suzukii adults in tens with 95% ethyl alcohol for 5 minutes. Adults were homogenized by pulverizing in a sterile mortar with sterile saline solution and serial dilutions were obtained from this homogenate [30] ( Figure 2). The dilutions prepared from the adults were inoculated on Nutrient Agar (NA) for bacterial growth. Then, Bacterial cultures were incubated at 30°C, and at the end of 24-72 hours. The bacterial strains with the dominant character were selected and purified [31]. Each of single colonies was prufied and streaked on agar plate. For each pure culture was given a separate code number, and information regarding the isolation conditions (location, altitude, insect form, date, etc.) was noted. The samples were kept at -86°C in stock growth media containing 30% glycerol and Loria Broth (LB) for routine use.

Determination of Morphological and Biochemical Properties of Bacterial Strains
The colony morphology and color of the bacterial strains were determined microscopical and visual inspection.
The Gram staining characteristics of the bacteria were determined according to the method described by Sands [32]. Presence of catalase and oxidase enzymes [33] and nitrate enzyme was assessed by the method of Harley and Prescott [34]. Chitinase enzyme activity was determined according to the method reported by Ortucu [35].

Polymerase Chain Reaction (PCR) of the Bacterial Strains
The strains were also identified in the molecular system. Total 19 bacterial strains selected from 39 bacterial strains obtained from D. suzukii adults according to high colony density were identified by sequencing a fragment of genome [36]. Bacterial DNA was amplified by a two-step PCR targeting the 16S rDNA gene with primers 27F and 907R, designed to include Illumina adaptor and barcode sequences. Sequencing was performed on an IlluminaMiSeq at the UC Davis Genomics Core Facility generating 963 basepair paired-end reads. Samples were OTUs are identified by their closest hit in the SILVA SSU Reference Database Release 111. Number of sequences is after all quality-control steps.

Phylogenetic Relationship of the Bacterial Strains
The sequences obtained were used to perform BLAST searches using the NCBI GenBank database to confirm strain identification Altschul et al. [37]. Evolutionary relationships of the 19 bacterial strains were evaluated. Cluster analyses of the sequences were performed using BioEdit (version 7.09) with Clustal W followed by neighbor joining analysis on aligned sequences performed with MEGA 6.0 software [38] Reliability of dendograms was tested by bootstrap analysis with 1000 replicates using MEGA 6.0.
The evolutionary history was inferred using the Neighbor-Joining method [39]. The optimal tree with the sum of branch length = 0.56341095 is shown. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) is shown next to the branches [40]. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the Maximum Composite Likelihood method [41] and are in the units of the number of base substitutions per site. The analysis involved 16 nucleotide sequences. Codon positions included were 1st+2nd+3rd+Noncoding. All positions containing gaps and missing data were eliminated. There were a total of 823 positions in the final dataset. Evolutionary analyses were conducted [38].

Results
The identification test results of the isolated bacteria and their morphological and biochemical characteristics were given in Table 1.  (Figure 3).
The bacterial strains with high colony density that were obtained from D. suzukii adults were identified molecular analysis. Molecular diagnostic results are given in Table 1. According to the 16S rRNA sequence analysis results, a total of 9 strains were identified as Bacillus amyloliquefaciens. The other bacterial strains consist of 3 Proteus myxofaciens, 1 Bacillus atrophaeus, 1 Bacillus safensis, 1 Paenibacillus motobuensis and 1 Staphylococcus epidermidis. Percent identify of all the identified bacterial strains were 99%. But, a total of three strains were not identified (Table 1).

Discussion
Chandler et al. [27], in their microbiota study on D. suzukii, isolated the genus Tatumella from 99% of both the adults and the larvae and Dunitz et al. [42] isolated Tatumella sp. from the larvae. Although this genus is not commonly found in Drosophila species, another study identified it also in D. melanogaster, which feeds on apples [43]. Again, Chandler et al. [27] reported that Acetobacteraceae and Orbus species were also associated with Drosophila population. However, Broderick and Lemaitre [43] noted that Orbus species were not associated with Drosophila. Brummel et al. [44] conducted molecular analyses on strains from the whole body of D. melanogaster adults, and they identified the genera Lactobacillus, Gluconabacter, Enterobacter, Anaerococcus, while Cox and Gilmore [45]  It suggests that differences in species distribution in microbial flora studies of the same insect species may be due to the differences in the body part where strains were obtained, the biological period of the pest, and dietary differences. Indeed, it has been stated that, in comparison to isolations made from the whole body, isolations from the gut yielded less microorganisms, that the sample preparation time may also influence variation due to the shorter passage time in the gut [46][47], and that microbiota could differ depending on the variations of consumed food, and whether the individual fed in its natural environment or in laboratory setting [47].
As it can be seen from all these studies, there is ongoing research on microbiota of D. suzukii species. It is thought that microbiota studies will guide biological pest control studies. The discovery of new pathogens and parasites of pest insects offers a chance to find organisms that may be useful for biological control [48]. For this purpose it has become increasingly important to identify microorganisms which are present in the microbiota of harmful organisms, and can be used in control of these pests, and to study them in biological control research. Identification of microorganisms that have the potential to be used in biological pest control against the invasive species D. suzukii will be possible detailed studies on this subject.

Conclusions
To our knowledge, this is the first study characterizing the bacterial communities of Drosophila suzukii. In conclusion, we think that expecially Bacillus amyloliquefaciens strains can be used as biological control agents against this economically important pest. Biological control studies will be planned with bacterial strains stored in the Atatürk University, Plant Protection Laboratory in future.