1 Azoxystrobin degradation by plant-associated bacteria

ed
2 Katharina Kraxbergera, Livio Antoniellib, Tanja Kostićb, Thomas Reichenauerb; Angela Sessitschb*

3 a Multikraft Produktions- und HandelsgmbH, Sulzbach 17, 4632 Pichl/Wels, Austria

iew
4 b AIT Austrian Institute of Techonology, GmbH, Center for Health & Bioresources, Bioresources

5 Unit, Konrad-Lorenz-Straße 24, 3430 Tulln, Austria

6 *Corresponding Author: e-mail address: angela.sessitsch@ait.ac.at

v
7

re
8

9
er
10
pe
11

12 CRediT authorship contribution statement:

13 Katharina Kraxberger: Funding acquisition, Conceptualisation, Methodology, Investigation, Writing –

ot

14 original draft. Livio Antonielli: Data curation. Tanja Kostić: Methodology, Writing – review & editing.
tn

15 Angela Sessitsch: Conceptualisation, Methodology, Writing – review & editing. Thomas Reichenauer:

16 Methodology, Writing – review & editing.

17
rin

18
ep
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 19 ABSTRACT

ed
20 Concerns about the possible effects of pesticide residues on both the environment and human health

21 have increased worldwide. Bioremediation by the use of microorganisms to degrade or remove these

22 residues has emerged as a powerful technology. However, the knowledge about the potential of different

iew
23 microorganisms for pesticide degradation is limited. This study focused on the isolation and

24 characterisation of bacterial strains with the potential to degrade the active fungicide ingredient

25 azoxystrobin. Potential degrading bacteria were tested in vitro and in the greenhouse. The genomes of

v
26 the best degrading strains were sequenced and analysed.

re
27 We identified and characterised 59 unique bacterial strains, which were further tested in vitro and in

28 greenhouse trials for their degradation activity. The best degraders from a foliar application trial in the

29 greenhouse were identified as Bacillus subtilis strain MK101, Pseudomonas kermanshahensis strain
er
30 MK113 and Rhodococcus fascians strain MK144 and analysed by whole genome sequencing. Genome

31 analysis revealed that these three bacterial strains encode several genes predicted to be involved in the
pe
32 degradation of pesticides e.g., benC, pcaG, pcaH, however we could not find any specific gene

33 previously reported to be involved in azoxystrobin degradation e.g., strH. Genome analysis pinpointed

34 to some potential activities involved in plant growth promotion.

ot

35
tn

36 GRAPHICAL ABSTRACT

37
rin

38 Keywords: Fungicide, Microorganisms. Bacteria, Rhizosphere, Endosphere, Lettuce

39
ep
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 40 1. Introduction

ed
41 Currently, over 500 pesticides are registered and used worldwide in agriculture (Parte et al., 2017). Due

42 to spray drifts, surface runoff and leaching processes, most of the applied pesticides (70-80%) inevitably

43 move off from target sites, contaminating soils and water and pose a potential risk for humans and other

iew
44 life forms (Ju et al., 2019; Parte et al., 2017). Physical treatments like adsorption or percolator filters, as

45 well as chemical treatments such as advanced oxidation or photodegradation for bioremediation of

46 polluted sites, are cost-ineffective, monitoring-intensive and incomplete due to the generation of more

v
47 persistent or/and more toxic metabolites and often only operate as ex situ clean-up technologies (Parte

48 et al., 2017). Under favourable conditions, bacteria from different genera have been reported to degrade

re
49 pesticides without the production of intermediates by using them as a source of carbon, sulphur or

50 electron donors (Bhagobaty and Malik, 2008; W. Li et al., 2009; Munazza et al., 2012).

51
er
Fungicidal active substances like azoxystrobin are regularly detected in different vegetables and fruits

52 (Medina-Pastor and Triacchini, 2020; Sun and Lueckl, 2018). Azoxystrobin binds to the quinol
pe
53 oxidation site of cytochrome b, which is located in the inner mitochondrial membrane of fungi. Thereby,

54 the electron transfer is blocked, leading to the inhibition of ATP production (Bartlett et al., 2002).

55 Azoxystrobin belongs to the β-methoxyacrylate and strobilurin family, which represents 25% of the
ot

56 fungicides sold worldwide (Nofiani et al., 2018). Abiotic as well as microbial degradation of

57 azoxystrobin is known to involve the hydrolysis of the carboxyl ester group of the parent structure
tn

58 (Clinton et al., 2011; Katagi, 2006; Wang et al., 2022). A subtilisin-like carboxypeptidase group of

59 enzymes (E.C. 3.4.21-) may play an essential role in this microbial degradation (Clinton et al., 2011).

60 Microbial azoxystrobin degradation was reported in Ochrobactrum anthropi SH14, which transforms
rin

61 the active ingredient by hydrolysis of the ester linkage and also cleavage of the aromatic ring (Feng et

62 al., 2020b). In another microorganism, Hyphomicrobium sp. DY-1, the esterase gene strH was found to
ep

63 be responsible for the de-esterification and detoxification of strobilurin fungicides (Jiang et al., 2021).

64 The chemical structure of azoxystrobin (National Center for Biology Information, 2022) and the

65 potential points of action for microbial degradation of this strobilurin are shown in Fig. 1. Frequently,
Pr

66 genes responsible for pesticide degradation are located on plasmids (Bhatt et al., 2021a).

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 67 Only a limited number of azoxystrobin-degrading bacteria has been isolated so far. Isolated strains

ed
68 include members of the genera Cupriavidus, Rhodanobacter (Howell et al., 2014), Bacillus (Baćmaga

69 et al., 2017) and Ochrobactrum (Feng et al., 2020b). These strains were isolated from pesticide

70 contaminated soils (Baćmaga et al., 2017; Howell et al., 2014) or sludge from wastewater treatment

iew
71 systems (Feng et al., 2020b), but not from plants.

72 In plants, numerous interactions with different kinds of microorganisms occur (Hardoim et al., 2015).

73 As described above, some of these bacteria are potential pesticide degraders. Some bacteria are also

v
74 known as plant growth-promoting bacteria (PGPB), capable of increasing plant growth. For example,

75 indole acetic acid (IAA), as a product of L-tryptophan metabolism of bacteria, can stimulate and

re
76 facilitate plant growth (Mohite, 2013). Furthermore, some bacteria have the enzyme 1-

77 aminocyclopropane-1-carboxylic acid (ACC) deaminase that acts by degrading ACC, which is the

78
er
precursor of ethylene synthesis, a plant stress hormone. This mechanism was shown to induce plant

79 tolerance to salt stress and also has an impact on plant growth (Orozco-Mosqueda et al., 2020). Some
pe
80 bacteria are also known for their production of siderophores supporting nutrient mobilization (Miethke

81 and Marahiel, 2007). Quorum sensing is a cell-based communication system, that regulates gene

82 expression for many microbial activities under the control of cell-population density. For example N-
ot

83 acyl-homoserine lactones (AHL) are frequently used as auto-inducers (Schikora et al., 2016).

84 Depending on environmental conditions, soil parameters and plant type, different microorganisms can
tn

85 establish in the plant environment (Berg and Smalla, 2009; Bonatelli et al., 2021). In the leaf

86 environment, diverse microbial communities inhabit the surface and the interior of leaves, known as

87 epiphytes and endophytes, respectively (Kandel et al., 2017; Rastogi et al., 2013). The leaf surface
rin

88 represents with a titer of ~106–107 bacteria cm2 leaf area one of the most abundant habitats for bacteria

89 (Leveau, 2018). In the rhizosphere, i.e., the zone with high biological activity surrounding the root
ep

90 system, numerous plant-microbe interactions occur (Philippot et al., 2013).

91 Microbial products for plants and the garden can provide microorganisms establishing a beneficial

92 interaction with plants. Products like EM Active and BB Foliar contain a mixture of different bacteria
Pr

93 and yeasts, and their application usually leads to a stronger plant development with improved

94 germination, flowering, fruit growth, ripening and yield quality. First trials on vegetable farms have

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 95 indicated that these products also have the ability to degrade pesticides (personal communication with

ed
96 vegetable farmers).

97 Lettuce was shown to be a crop containing relatively high amounts of pesticide residues, including

98 azoxystrobin, compared to other crop types like fruits, cereals, or tubers (BVL, 2007; Sevigné Itoiz et

iew
99 al., 2012). Azoxystrobin-degrading, plant-associated bacteria may influence azoxystrobin residue levels

100 in lettuce and other crops and may also represent a source of valuable strains to be applied for reducing

101 azoxystrobin contamination in plant produce. As so far little is known about azoxystrobin degradation

v
102 by plant-associated bacteria, the objective of this study was to identify and characterize azoxystrobin-

103 degrading bacterial strains from lettuce plants grown on contaminated soil as well as from multi-microbe

re
104 products. Furthermore, we aimed to gain insight into the degradation of the pesticide in cultivated plants

105 grown in non-sterile soil and to obtain insight on responsible pathways.

106
er
107 2. Material and methods
pe
108 2.1. Isolation, de-replication and characterisation of fungicide-degrading strains

109 2.1.1. Set-up of the greenhouse experiment and sampling of plants

110 Lettuce seeds (Lactuca sativa L., Austrosaat AG, Austria) were sown in seed trays containing a mixture
ot

111 (3:1 v/v) of commercial potting soil (unit earth SP ED63T, Sinntal-Altengronau, Germany) and perlite

112 (2-6 mm, GBC Gartenbaucentrum, Austria). After reaching the growth stage 13 of the BBCH scale,
tn

113 which is the third true leaf unfolded (Meier, 2001), the plants were trans-planted to 3 L pots (two plants

114 per pot, six biological replicates, randomized layout). Pots were filled with azoxystrobin-contaminated

115 soil, collected from a field in England-Sleaford (52°99'69.5"N; -°25'94.9"E) treated with the fungicide
rin

116 Azoxystar (200 g/L azoxystrobin, Life Scientific GmbH, Germany) for several years. Soil samples were

117 collected from 20-25 cm depth and stored at 4°C prior to use. The temperature in the greenhouse was
ep

118 set to 22°C with a relative humidity of 30% during the day. The light value was kept between 10 klx

119 and 40 klx by turning on artificial light or shading. During the night, the overall temperature was 20°C

120 with 15% relative humidity.

Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 121 Plants were watered with tap water twice a week from below using saucers to avoid fungicide washout.

ed
122 The pot arrangement, greenhouse settings and information about watering apply to all greenhouse

123 experiments described in this paper.

124 After ten days of growth, the number of plants per pot was reduced from two to one. After reaching the

iew
125 growth stage 49 of the BBCH scale, which is the typical size, form and firmness of lettuce heads, three

126 replicates were randomly selected and carefully removed from the pot. Two compartments (leaves and

127 rhizosphere) were sampled. The rhizosphere soil was removed from the root system by carefully shaking

v
128 it off, whereas leaves were obtained by cutting the plants above the roots with a sterile scalpel. The

129 tissue samples were cut into small pieces and mixed with sterile 0.9% NaCl (1:10 dilution) in a sterile

re
130 pulsifier bag (PUL512 Pulsifier II® Bags, Microgen Bioproducts Ltd, UK). These samples were used

131 immediately for the isolation of bacteria.

132
er
133 2.1.2. Isolation of bacteria from plants and from multi-microbe products
pe
134 Microbial cells were dislodged from plant tissue by using the Pulsifier (Pulsifier II®, Microgen

135 Bioproducts Ltd, UK) two times for 15 s. Dilutions of the supernatants were made with sterile 0.9%

136 NaCl, and 100 µl of the selected dilutions, 10-1 for the leaf and 10-3 for the rhizosphere samples, were
ot

137 plated in triplicates on diverse solid media containing fungicide as sole carbon source. Media were (i)

138 200 mL/L minimal medium M9 (Yi et al., 2016) (33.9 g/L Na2HPO4, 15 g/L KH2PO4, 2.5 g/L NaCl, 5
tn

139 g/L NH4Cl) + 25 mg/L azoxystrobin dissolved in ethyl acetate (Azoxystrobin PESTANAL®, Merck,

140 Germany) + 15 g/L agar, (ii) 500 mL/L minimal salts medium MMS (dos Santos et al., 2017) (0.1 g/L

141 MgCl2*6H2O, 0.01 g/L FeCl3*6H2O, 0.005 g/L CaCl2, 0.05 g/L NaCl, 0.5 g/L (NH4)2SO4) + 25 mg/L
rin

142 azoxystrobin + 15 g/L agar, (iii) 200 mL/L minimal media M9 + 25 mg/L azoxystrobin + 30 g/L gellan

143 gum, (iv) 500 mL/L minimal salts medium MMS + 25 mg/L azoxystrobin + 30 g/L gellan gum.
ep

144 For the isolation of bacterial strains from multi-microbe products, EM Active (lot# 42038.1, Multikraft,

145 Austria) and BB Foliar (lot# 42016.01, Multikraft, Austria), 100 µl of the undiluted product were plated

146 (triplicates) on all media listed above.

Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 147 Plates were incubated for 72 h at 28°C. Ten and five morphologically different colonies from each of

ed
148 the four isolation media were selected from the plant multi-microbe products, respectively. Isolates were

149 further cultivated on 200 mL/L minimal media M9 + 25 mg/L azoxystrobin + 15 g/L agar.

150 For gDNA extraction, a loop of cells grown on an agar plate was re-suspended in 200 µL sterile 0.9%

iew
151 NaCl, pelleted (18,407 rcf, 5 min) and stored at -20°C. gDNA of each isolate was extracted using the

152 nexttec 1-Step DNA Isolation Kit for Bacteria (Biozym Scientific GmbH, Germany) according to the

153 manufacturer´s instructions.

v
154

155 2.1.3. Characterisation and de-replication of isolates

re
156 Restriction fragment length polymorphism (RFLP) analysis of the 16S-23S rRNA intergenic spacer

157 (IGS) region (IGS-RFLP-analysis) was performed to de-replicate the isolates and thus to identify unique

158 bacterial strains. IGS PCR was

erperformed with the primer pair P23SR01 (5′-

159 GGCTGCTTCTAAGCCAAC-3′) and pHr (5′-TGCGGCTGGATCACCTCCTT-3′) (Massol-Deya et

pe
160 al., 1995). A conventional PCR amplification in 50 µl PCR reaction mix containing 2.5 mM MgCl2, 0.2

161 mM dNTPs, 0.3 mM of each primer, 4 µl DNA template, 1 U Taq DNA polymerase (Soils BioDyne,

162 Estonia) and 1 x PCR reaction buffer was performed. An initial denaturation step at 95°C for 5 min was
ot

163 followed by 30 cycles of denaturation at 95°C for 45 s, annealing at 54°C for 60 s and elongation at

164 72°C for 120 s, followed by a final extension at 72°C for 10 min. Electrophoresis (100 V, 45 min) on
tn

165 1% (w/v) agarose gels was used to test the quality of the fragments (expected amplicon size in the range

166 1,500 to 2,500 bp depending on the organism).

167 Restriction digest of IGS PCR products was performed using 200 ng amplified DNA, 5 U AluI or HhaI
rin

168 endonuclease (FastDigest, Thermo Fisher Scientific, USA) and 1x Buffer Tango (Thermo Fisher

169 Scientific, USA). Digestion was performed for 3 h at 37°C in a thermocycler. Electrophoresis (100 V,
ep

170 120 min) on 2% (w/v) agarose gels was used to separate DNA fragments. RFLP patterns were compared

171 (Reiter et al., 2002) and isolates with identical IGS-RFLP patterns were excluded.

172 The 16S rRNA genes were amplified using the primers 8F (5′-AGAGTTTGATCCTGGCTCAG-3′)
Pr

173 (Weisburg et al., 1991) and 1520R (5′-AAGGAGGTGATCCAGCCGCA-3′) (Edwards et al., 1989).

174 Conventional PCR amplification was performed, as described above, but in 25 µl PCR reaction mix and

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 175 using 2 µl DNA template. The annealing temperature was set to 52°C. The expected amplicon size of

ed
176 the 16S rRNA amplicons is 1,500 bp.

177 Partial 16S rRNA sequencing was performed by LGC (LGC Group, UK) using the primer 926R (5′-

178 CCGTCAATTCCTTTGAGTTT-3′) (Liu et al., 1997) and 1492R (5′- GGTTACCTTGTTACGACTT-

iew
179 3′) (Lane, 1991).

180 To obtain complete 16S rRNA gene sequences of selected strains, the ends of partial sequences were

181 trimmed and assembled using Geneious Prime® 2022.1.1. Decipher (Wright, 2016) was used to check

v
182 for chimeric sequences. The closest species match of assembled sequences was identified using Basic

183 Local Alignment Search Tool (BLAST) analysis.

re
184

185 2.2. Selection of strains for azoxystrobin degradation trials

186
er
The strains for the in vitro azoxystrobin degradation trials were selected based on available information

187 on genera known to degrade azoxystrobin or other strobilurin fungicides (such as trifloxystrobin and
pe
188 pyraclostrobin). Therefore, strains of the genera Arthrobacter (five strains), Bacillus (two strains) were

189 selected as well as of Pseudomonas (one strain) were selected based ob published evidence (Baćmaga

190 et al., 2017; X. Chen et al., 2018; Clinton et al., 2011). In addition, strains from other isolated genera
ot

191 were selected, including Ensifer (one strain), Variovorax (one strain), Microbacterium (one strain),

192 Pseudoarthrobacter (two strains), Rhodococcus (one strain), Nocardioides (one strain), Domibacillus
tn

193 (one strain) and Flavobacterium (one strain). Furthermore, two reference strains, obtained from

194 Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSMZ), were included in the degradation

195 tests, i.e., Bacillus subtilis DSM1970 (Clinton et al., 2011) and Rhodococcus pyridinovorans
rin

196 DSM44555 (Yoon et al., 2000).

197
ep

198 2.2.1. In vitro azoxystrobin degradation tests on lamb’s lettuce leaves

199 First, the degradation capacity of selected strains was tested in vitro using lamb´s lettuce (Valeriana-

200 Motovilec, Samen Maier GmbH, Austria). Lamb´s lettuce seeds were sown in seed trays containing a
Pr

201 mixture (3:1 v/v) of commercial potting soil and perlite. After reaching growth stage 13 of the BBCH

202 scale, the lettuce leaves were cut with a sterile scalpel. Round filters (Rotilabo ® type 115A, 90 mm Ø,

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 203 Carl Roth GmbH+Co KG, Germany) were placed in a Petri dish (90 mm Ø) and moisturised with 1 mL

ed
204 sterile water. Each wetted round filter was covered with 1 g lettuce leaves (top up) and 200 µL

205 azoxystrobin stock solution per Petri dish (treatment amount was 200 mg AI/kg lettuce) were applied

206 using an airbrush (pressure: 0.1 bar, nozzle: 0.5 mm) (Airbrush Kit with Compressor ABPST05,

iew
207 Timbertech, Airgoo Pneumatic BV, Netherlands). The samples were incubated at room temperature for

208 2 h. Then, lettuce leaves were treated with 100 µL microbial suspension (106 cfu/g plant material) using

209 an airbrush (pressure: 0.1 bar, nozzle: 0.5 mm) and incubated at room temperature overnight. All

v
210 treatments were carried out in triplicates.

211 Sample preparation for the quantitative determination of azoxystrobin concentration was done according

re
212 to Ali et al. (2015). Briefly, 1 g of cut leaves was transferred to a 5 mL tube and 1.5 mL high performance

213 liquid chromatography (HPLC)-grade acetonitrile (Fisher Scientific, UK) containing 1% (v/v) of acetic

214
er
acid was added. The sample was mixed manually for 1 min, followed by 15 s at the highest speed on

215 the Vortex (Vortex-Genie™ 2, Scientific Industries SI, USA). To extract powder, 0.6 g magnesium
pe
216 sulfate (Sigma-Aldrich, USA) and 0.15 g anhydrous sodium acetate (Sigma-Aldrich, USA) were added,

217 mixed manually for 1 min, vortexed for 15 s and centrifuged for 5 min at 1,503 rcf. The supernatant (0.8

218 mL) was transferred to a new 5 mL tube containing 120 mg magnesium sulfate and 40 mg primary and
ot

219 secondary amine exchange material (PSA) (Restek Corporation, USA). No graphitized carbon black

220 (GBC) was added to the samples (L. Li et al., 2009). The sample was shaken immediately manually for
tn

221 1 min, vortexed for 30 s, and centrifuged for 5 min at 2,348 rcf. Afterwards, 0.5 mL of the upper layer

222 were taken by syringe and needle, filtered using sterile syringe filter (PVDF Rotilabo®, 0.45 µm, Carl

223 Roth GmbH+Co KG, Germany) and stored at -20°C. Azoxystrobin concentrations were determined by
rin

224 HPLC. The filtered supernatant was thawed, and 1 mL of each sample was pipetted into a 2 mL screw-

225 top glass HPLC vial (Thermo Fisher Scientific, USA).

ep

226 Samples were analysed using an Agilent 1100 series HPLC with a diode array detector (DAD) and

227 LiChrospher ® 100 RP-18e (5 µm) column (column length: 125 mm, pore size: 100 Å, internal diameter:

228 4 mm, Agilent, UK). 25µl of each sample were injected and analysed using the method of Howell et al.
Pr

229 (2014); briefly: the mobile phase consisting of 75% acetonitrile (HPLC-grade, Fisher Scientific, UK)

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 230 and 25% distilled water was set to a flow rate of 1.30 mL/min. Azoxystrobin concentration was

ed
231 determined by measuring the absorbance at 238 nm in dublicate.

232 Azoxystrobin recovery from the control leaf samples (samples with pesticide treatment and without

233 bacterial treatment) was measured. Treatments showing azoxystrobin concentrations (mean of

iew
234 triplicates) in the range of the control were evaluated as not degraded. The isolates leading to the lowest

235 azoxystrobin concentrations were selected for trials in the greenhouse.

236

v
237 2.2.2. Azoxystrobin degradation in greenhouse tests

238 Selected strains (seven) were tested for their degradation capacity of foliar-applied fungicide using

re
239 lamb´s lettuce. Lamb´s lettuce plants were grown as described above. At the growth stage 13 of the

240 BBCH scale, the plants were trans-planted to 3 L pots filled with commercial potting soil (three plants

241
er
per pot, four biological replicates). One day after trans-planting, 1 mL Azoxystar stock solution per pot

242 (treatment amount 240 mg AI/kg lettuce) were applied on leaves by using an airbrush (pressure: 0.1 bar,
pe
243 nozzle: 0.5 mm). Subsequently, 2 mL bacterial suspension (106 cfu/mL) per plant and treatment were

244 applied two and seven days after the fungicide treatment by using an airbrush. At growth stage 17 of the

245 BBCH scale, i.e., the seventh true leaf unfolded, the leaves were harvested and prepared for HPLC
ot

246 analysis. For leaf sampling, three plants of each pot were pooled into one sample and stored at -20°C.

247 One gram of each sample was further used for quantitative analysis, as explained above. Azoxystrobin
tn

248 recovery from the control leaf samples (samples with pesticide treatment and without bacterial

249 treatment) was measured. The percentage of degradation was calculated by dividing the recovered

250 azoxystrobin amount from treated samples by the recovered azoxystrobin amount from the control
rin

251 samples. Recursive partitioning analysis was used for grouping the strains regarding their degradation

252 abilities. The bacterial strains, which showed the highest azoxystrobin degradation in this trial, were
ep

253 selected for bacterial genome sequencing and analysis.

254 Furthermore, the same selected strains were tested for their capacity to degrade azoxystrobin when both

255 microorganisms and fungicide were applied to the soil. A mixture (3:1 v/v) of commercial potting soil
Pr

256 and perlite was treated with 25 mg/kg azoxystrobin (by using Azoxystar). The treated soil was

257 equlibrated for two weeks at 4°C in the dark, before sowing the seeds. Organic lamb´s lettuce seeds

10

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 258 were surface-sterilised on a tube roller (RS-TR5, Phoenix Instrument GmbH, Germany) with 70%

ed
259 ethanol for 2 min followed by 0.6% sodium hypochlorite for 5 min and rinsed three times in sterile water

260 for 1 min. Followingly, the seeds were imbibed with a 108 cfu/mL bacterial suspension for 2 h at 28°C;

261 for the negative control sterilised seeds were incubated in 1 x PBS buffer. At the growth stage 13 of the

iew
262 BBCH scale, plants were trans-planted to 3 L pots (filled with soil/perlite/azoxystrobin mixture, three

263 plants per pot, four biological replicates). At this stage, bacteria were applied again by immersing the

264 root system for 10 min in 108 CFU/mL bacterial suspension. At the growth stage 17 of the BBCH scale,

v
265 the leaves and the soil (expanded rhizosphere) were harvested and prepared for HPLC analysis. Leaf

266 sampling and quantitative azoxystrobin determination were done as described above.

re
267 The expanded rhizospheres of the three plants per pot were mixed, and two 10 g aliquots were stored at

268 -20°C. Sample preparation of the soil was done according to Howell et al. (2014). Briefly, 10g of soil

269
er
were extracted in 50 ml conical bottom tube using 20 mL of a solution containing 75% HPLC-grade

270 acetonitrile and 25% distilled water. The tubes were shaken for 1 h on a tube roller. Subsequently, the
pe
271 tubes were left 30 min to settle and centrifuged at 3,489 rcf for 2 min. Afterwards, the samples were

272 filtered, stored, analysed and grouped as described above.

273
ot

274 2.3. Bacterial genome sequencing and bioinformatic analysis

275 Bacterial genomic DNA from three strains selected for whole genome sequencing was isolated using a
tn

276 modified nexttec 1-Step DNA Isolation Kit for Bacteria (Biozym Scientific GmbH, Germany) kit

277 protocol. Concisely, cells were grown in 10% tryptic soy broth overnight up to 1.5 OD600. Bacterial

278 cultures (1.5 mL) were transferred to 2 mL tubes and collected by centrifugation (13,845 rcf, 5 min).
rin

279 Bacterial pellets were re-suspended in 250 µL lysis buffer LB1 and incubated at 56°C for 15 min with

280 shaking at 1,200 rpm in a thermomixer comfort (Eppendorf, Germany). Modified lysis buffer (250 µL)
ep

281 LB2 (220 µL SDS Solution, 25 µL Proteinase K and 5 µL EDTA) were added to each sample, mixed

282 by using a Vortex and incubated in a thermomixer (1,200 rpm, 56°C, 45 min). DNA was extracted twice

283 using an equal volume of phenol-chloroform-isoamylalcohol (25:24:1) (Carl Roth GmbH+Co KG,
Pr

284 Germany) and collected by centrifugation (13,845 rcf, 4°C, 10 min). The upper phase was re-extracted

285 with an equal volume of chloroform and collected by centrifugation (13,845 rcf, 4°C, 10 min). To

11

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 286 precipitate DNA 0.1 volume of 3 M sodium acetate (pH 5.2) and 2 volumes of 100% ice-cold ethanol

ed
287 were added. The tubes were incubated at 4°C overnight. Afterwards, tubes were centrifuged (13,845 rcf,

288 4°C, 30 min), supernatant discarded and the air-dried pellets re-suspended in water. Whole genome

289 sequencing was performed on an Illumina MiSeq platform (GATC Biotech, Konstanz, Germany).

iew
290 Illumina MiSeq reads were checked for the presence of PhiX using Bowtie 2 (v2.3.4.3) (Langmead and

291 Salzberg, 2012) and adapters were removed with fastp v0.19.5 (S. Chen et al., 2018). Sequence quality

292 and length distribution were checked with FastQC (Andrews, 2010). Genome assembly was carried out

v
293 with SPAdes v3.14.0 and low-abundant (<2×) contigs were filtered out (Bankevich et al., 2012). The

294 presence of contaminant contigs was assessed using BLAST against the entire NCBI nt database

re
295 (downloaded in September 2021) and BlobTools (Laetsch et al., 2017). Genome assembly quality was

296 then inferred using QualiMap v2.2 (Okonechnikov et al., 2016) and QUAST v5.0.0 (Gurevich et al.,

297
er
2013). The presence of plasmids was investigated with Platon (Schwengers et al., 2020). Gene prediction

298 and annotation was performed using Prokka v1.14.5 (Seemann, 2014). Further functional annotation
pe
299 was carried out with eggNOG v2.1.6 (Huerta-Cepas et al., 2019). Genome completeness was assessed

300 with BUSCO v4.0.6 (Simão et al., 2015) and CheckM v1.1.0 (Parks et al., 2015), respectively. Contigs

301 were screened for the presence of antimicrobial resistance and virulence genes with ABRicate v1.0.0
ot

302 (Seemann, 2021). Taxonomic classification was assigned with GTDB-Tk v1.5.0 (Chaumeil et al., 2020)

303 using the Genome Database Taxonomy (GTDB R06-RS202, latest release 27th April 2020) as a
tn

304 reference.

305

306 2.4. Detailed genome analysis: pathways/genes/proteins related to fungicide degradation and genes
rin

307 related to plant-microbe interactions

308 Pathways/genes/proteins related to azoxystrobin degradation and genes related to known beneficial
ep

309 plant-microbe interactions were analysed in the whole genome sequences of Bacillus subtilis strain

310 MK101, Pseudomonas kermanshahensis strain MK113 and Rhodococcus fascians strain MK144.

311 In the first step, strobilurin-degrading pathways in bacteria were searched using MetaCyc (Caspi et al.,
Pr

312 2014) and the microorganism search of Biocatalysis/Biodegradation Database was used to retrieve

313 biodegradation pathways (Gao et al., 2009).

12

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 314 In the second step, the three bacterial genomes were screened for esterases, hydrolases, oxygenases or

ed
315 other genes or proteins (using blastn for genes and blastp for proteins) which are described in the

316 literature for the breakdown of azoxystrobin or similarly structured compounds.

317 In the third step, all genes from whole genome sequencing were allocated to different systems by using

iew
318 the databases Gene Ontology (GO) (Ashburner et al., 2000; The Gene Ontology Consortium, 2021),

319 Kyoto Encyclopedia of Genes and Genomes (KEGG) (Kanehisa et al., 2022), Clusters of Orthologous

320 Genes (COG) (Tatusov et al., 2000) and Carbohydrate Active Enzymes database (CAZy) (Drula et al.,

v
321 2022). Using KEGG, genes responsible for the category 1.11 xenobiotics biodegradation and

322 metabolism were examined. In the COG, the three categories energy production and conversion (C),

re
323 carbohydrate transport and metabolism (G) and secondary metabolites, biosynthesis, transport and

324 catabolism (Q) were compared within the three genomes. By using CAZy the EC number of the enzymes

325
er
were found and were compared to relevant literature.

326 In the fourth step, genes related to known beneficial plant-microbe interactions were analyzed. Genes
pe
327 responsible for ferric-siderophore complexes, indole acetic acid (IAA) synthesis pathways, quorum

328 sensing and 1-aminocyclopropane-1-carboxylic acid (ACC) deaminase in bacteria were searched in the

329 three bacterial genomes.

ot

330

331 2.5. Statistical analysis

tn

332 Statistical analysis for the selection of strains from in vitro trials was performed in Microsoft Excel

333 (Office 2016). Precisely, data from azoxystrobin measurements were expressed as the mean over

334 triplicates ± standard error of the mean (SEM). Variance was calculated and one-way ANOVA was
rin

335 applied for the analysis of potential degrading strains.

336 All data from plant experiments were analysed using RStudio software (version 1.4.1717). A linear
ep

337 regression analysis was carried out using the pesticide content as target, while genera and strains were

338 used as predictors. More in detail, a model was built with HPLC values as dependent variable and genera

339 and strains as independent variables, with the latter nested in the former one. A second model was built
Pr

340 with HPLC values and strains, only. The performance of the two models was assessed by repeated K-

341 fold cross-validation, with 10 random splits repeated 10 times. Root Mean Squared Error (RMSE), R-

13

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 342 squared (R2) and Mean Absolute Error (MAE) were used to evaluate the quality of the models (Kuhn,

ed
343 2022). The model with lower RMSE was chosen for the Analysis of Variance (ANOVA). Pairwise

344 comparisons with Estimated Marginal Means (EMMs) followed as post-hoc analysis (Graves et al.,

345 2019; Lenth R, 2022). A more robust but less sensitive approach was performed as confirmatory test by

iew
346 using a non-parametric regression with recursive partitioning analysis (Hothorn and Zeileis, 2015).

347 Boxplots of pesticide content (mg/l ) by strain were realized using the ggplot2 package (Wickham et al.,

348 2016).

v
349 Generally, the average of the two HPLC measurements per sample was taken for statistical analysis.

350 Statistical significance was defined as p<0.005.

re
351

352 2.6. Nucleotide sequence accession numbers

353
er
Sequence data are available at NCBI database under the accessions SUB12110983: OP583678 to

354 OP583724 for assembled 16S rRNA sequences, SUB12111171: OP583666 to OP583677 for partial 16S
pe
355 rRNA gene sequences and SUB12111202: BioProject number PRJNA886999 for whole genome

356 sequences.

357
ot

358 3. Results

359 3.1. Isolation, de-replication and characterisation of fungicide-degrading strains

tn

360 In total, 43 different strains were obtained from lettuce plant isolates after excluding isolates with

361 identical IGS-RFLP patterns. One Arthrobacter strain (Arthrobacter sp. strain MK97), one Bacillus

362 strain (Bacillus sp. strain MK158) as well as one Ensifer strain (Ensifer sp. strain MK141) were obtained
rin

363 from rhizosphere and from leaf samples. Twenty-four different strains (belonging to 13 genera) were

364 isolated from lettuce rhizosphere, whereas 22 strains (belonging to eleven genera) were obtained from
ep

365 leaves. Potential degraders from rhizosphere samples were bacteria from the genera Ensifer,

366 Neorhizobium, Rhizobium, Acidovorax, Pelomonas, Variovorax, Pseudomonas, Rhizobacter,

367 Arthrobacter, Lentzea, Streptomyces, Bacillus and Flavobacterium. Potential degraders from leaf
Pr

368 samples were bacteria from the genera Ensifer, Microbacterium, Arthrobacter, Pseudoarthrobacter,

14

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 369 Rhodococcus, Nocardioides, Streptomyces, Bacillus, Domibacillus, Paenibacillus and Dyadobacter

ed
370 (Table 1).

371 From the multi-microbe product EM Active, nine different strains were obtained. Strains isolated from

372 EM Active belonged to the six different genera Bacillus, Brevibacillus, Cohnella, Paenibacillus, Niallia

iew
373 and Lacticaseibacillus. From the multi-microbe product BB Foliar, seven different strains were obtained

374 belonging to the genera Streptomyces and Paenibacillus (Table 1).

375

v
376 3.2. Azoxystrobin degradation by selected microbial strains

377 First, the degradation potential of the seventeen selected strains and the two reference strains was tested

re
378 in an in vitro assay with lamb’s lettuce leaves (Table 1). Azoxystrobin recovery from control leaf

379 samples (samples with pesticide treatment and without bacterial treatment) was 72% of the applied

380
er
concentration (theoretically 200 mg/kg; av=144 mg/kg in uninoculated control, sd=5) (Fig. 2). Most

381 strains showed no or only low degradation of azoxystrobin after spray application on lettuce leaves.
pe
382 However, six strains showed degradation rates between 29% and 72%, i.e., Flavobacterium sp. strain

383 MK95 (29% degradation), Pseudomonas sp. strain MK113 (32%), Arthrobacter sp. strain MK97 (56%),

384 Rhodococcus sp. strain MK144 (62%), Ensifer sp. strain MK141 (62%) and Bacillus sp. strain MK101
ot

385 (72%) (Fig. 2). Reference strain R. pyridinovorans DSM44555 showed 2% degradation of azoxystrobin,

386 whereas reference strain B. subtilis DSM1970 showed 47% degradation (Fig. 2). These six strains and
tn

387 the reference strain B. subtilis DSM1970 were further analysed for their degradation potential on leaves

388 of lamb’s lettuce grown in the greenhouse. Here, azoxystrobin recovery from control leaf samples in the

389 plant system was 64% of the applied concentration (153±19 mg/kg of 240 mg/kg applied). All bacterial
rin

390 strains showed degradation of azoxystrobin, and the best degrading strains were Bacillus sp. strain

391 MK101 (>90% degradation), Rhodococcus sp. strain MK144 (>90%), Pseudomonas sp. strain MK113
ep

392 (90%) and Ensifer sp. strain MK141 (83%). Flavobacterium sp. strain MK95 and Arthrobacter sp. strain

393 MK97 showed 23% and 11% degradation, respectively. Reference strain Bacillus subtilis DSM1970

394 also showed high degradation of the fungicide on leaves (>90%) (Fig. 3).
Pr

395 The six degrading strains were also assessed for their degradation of the fungicide (in soil and leaves)

396 when applied as soil inoculum. Here, azoxystrobin recovery from control soil samples in the plant

15

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 397 system was 92% of the applied concentration (23±2 mg/kg of 25 mg/kg applied). Azoxystrobin was

ed
398 translocated upward within the lettuce. In the leaves of the control plants, we measured 2% (0.48±0.01

399 mg/kg) of azoxystrobin that was amended to the soil. Also in this set-up, all bacterial strains showed

400 degradation of azoxystrobin. Rhodococcus sp. strain MK144 and Pseudomonas. sp. strain MK113

iew
401 showed the highest degradation of the fungicide in soils, i.e., 52% and 51% degradation, respectively,

402 whereas Arthrobacter sp. strain MK97 (12%), Flavobacterium sp. strain MK95 (14%) and Ensifer sp.

403 strain MK141 (17%) showed only low degradation rates (Fig. 4A). Bacillus sp. strain MK101 degraded

v
404 19% of azoxystrobin in soil. The reference strain Bacillus subtilis DSM1970 showed 28% azoxystrobin

405 degradation in soil. In this experiment, the best degrading strains in leaves were Arthrobacter sp. strain

re
406 MK97 (>90% degradation) and Pseudomonas sp. strain MK113 (72%), whereas Bacillus sp. strain

407 MK101 (39%), Ensifer sp. strain MK141 (37%) and the reference strain Bacillus subtilis DSM1970

408
er
(38%) showed only moderate degradation (Fig. 4B).

409 Based on the degradation activities, particularly after spraying the fungicide on leaves, Bacillus sp. strain
pe
410 MK101, Rhodococcus sp. strain MK144 and Pseudomonas sp. strain MK113 were selected for bacterial

411 whole genome sequencing and analysis.

412
ot

413 3.3. Bacterial genome sequencing and general genome analysis

414 The azoxystrobin degraders Bacillus sp. strain MK101, Pseudomonas sp. strain MK113 and
tn

415 Rhodococcus sp. strain MK144 had a total genome size of 4.0, 6.3 and 5.4 Mb with an average G+C

416 content of 45%, 63% and 65%, respectively. The use of the quality assessment tool (QUAST) showed

417 good assembly characteristics of Bacillus sp. strain MK101 and Pseudomonas sp. strain MK113. More
rin

418 than 60% of assembled contigs had a length of ≥25,000 bp. The assembly quality of Rhodococcus sp.

419 strain MK144 was poorer as over 50% of the contigs had a length of ≤5,000 bp. QUAST predicted 4,040
ep

420 genes in Bacillus sp. strain MK101, 5,535 genes in Pseudomonas sp. strain MK113 and 4,953 genes in

421 Rhodococcus sp. strain MK144. The majority of predicted genes (>80%) were in the range of 300 –

422 3,000 bp. Prokka-based rapid annotation of genomes revealed that >97% of the assembled sequences
Pr

423 represented coding sequences. Potential antimicrobial resistance genes and virulence factors were

424 identified. Bacillus sp. strain MK101 harboured multiple resistance genes with 100% homology to the

16

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 425 reference sequences. Pseudomonas sp. strain MK113 showed multiple hits related to virulence factors,

ed
426 although the sequence similarity of these genes was less than 90%. Rhodococcus sp. strain MK144

427 showed only low similarity to the reference sequences of identified antibiotic resistance genes. No

428 plasmids were found in the analysed genomes. The genomic features of the three genomes are

iew
429 summarised in Table 2.

430 The three strains were classified taxonomically as (i) Bacillus subtilis, (ii) Pseudomonas

431 kermanshahensis and (iii) Rhodococcus fascians. B. subtilis strain MK101 and R. fascians strain MK144

v
432 were both isolates from lettuce leaves grown in an azoxystrobin-contaminated soil, whereas P.

433 kermanshahensis strain MK113 was isolated from the lettuce rhizosphere (Table 1).

re
434 Analysis of strobilurin degrading pathways with MetaCyc revealed that no azoxystrobin, dimoxystrobin,

435 fluoxastrobin, kresoxim-methyl, metominostrobin, orysastrobin, picoxystrobin, pyraclostrobin and

436
er
trifloxystrobin degradation pathway is known. By searching in the Biocatalysis/Biodegradation

437 Database, Bacillus spp. showed 14 pathways predicted to be involved in biodegradation, but B. subtilis
pe
438 showed zero pathways in this database. Rhodococcus spp. retrieved 30 involved biodegradation

439 pathways, but R. fascians strain showed zero pathways. Pseudomonas spp. retrieved 150 pathways, but

440 P. kermanshahensis strain showed zero pathways. However, 36 pathways predicted to be involved in
ot

441 biodegradation were from Pseudomonas putida, which is a close relative of Pseudomonas

442 kermanshahensis (Table S1).

tn

443 In the literature, sequences for 39 potential degrading genes and two potential degrading proteins which

444 are known to degrade azoxystrobin related substances were found (Table S2). None of these genes,
rin

445 neither the esterase gene strH, responsible for de-esterification and detoxification of strobilurin

446 fungicides by strain Hyphomicrobium DY-1, were found in our genome sequences using the BLAST

447 search.
ep

448 Since we found no definitive genes for azoxystrobin degradation, the search was expanded and revealed

449 that the three bacterial strains contain numerous genes predicted to be involved in pesticide degradation.
Pr

450 Based on Prokka and eggNOG annotation in KEGG Pathways responsible for xenobiotics

451 biodegradation and metabolism (1.11.), 561 genes potentially involved in degrading xenobiotics were

452 found (Table S3). In the genomes of the three strains, the top seven degrading groups (out of 21 groups)

17

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 453 were for benzoate (125 genes), aminobenzoate (63), chlorocyclohexane and chlorobenzene (40), drug

ed
454 metabolism - other enzymes (41), naphthalene (37), styrene (34) and caprolactam (34) degradation (bold

455 numbers in Table S3).

456 The genes responsible for benzoate, naphthalene, styrene and caprolactam degradation can play a role

iew
457 in azoxystrobin degradation by degradation of aromatic rings.

458 In B. subtilis strain MK101 100 genes were found to be connected to xenobiotics biodegradation and

459 metabolism. The top four degrading groups were benzoate (22 genes), aminobenzoate (14 genes),

v
460 chlorocyclohexane and chlorobenzene (13 genes) and chloroalkane and chloroalkene (12 genes). In P.

re
461 kermanshahensis strain MK113 224 genes were found to be connected to xenobiotics biodegradation

462 and metabolism. The top four degrading groups were benzoate (59 genes), aminobenzoate (21 genes),

463 chlorocyclohexane and chlorobenzene (17 genes) and styrene (17 genes). In R. fascians strain MK144
er
464 237 genes were found to be connected to xenobiotics biodegradation and metabolism. The top four

465 degrading groups were benzoate (44 genes), aminobenzoate (28 genes), caprolactam (25 genes) and
pe
466 naphthalene (20 genes).

467 P. kermanshahensis strain MK113 and R. fascians strain MK144 were found to harbour genes for

468 aromatic ring degradation. benC (P. kermanshahensis MK113), pcaG (R. fascians MK144) and pcaH
ot

469 (both strains) could play an important role in the degradation of the two aromatic rings found in

470 azoxystrobin (Fig. 1).

tn

471 Three COG categories connected to metabolism were compared in P. kermanshahensis strain MK113,

472 R. fascians strain MK144 and B. subtilis strain MK101. P. kermanshahensis strain MK113 has a higher
rin

473 number of genes (3,072 genes) in COG category C (energy production and conversion) than R. fascians

474 strain MK144 (2,129) and B. subtilis strain MK101 (1,650). In the COG categories G (carbohydrate

475 transport and metabolism) and Q (secondary metabolites, biosynthesis, transport and catabolism) R.
ep

476 fascians strain MK144 has the highest numbers of genes (G: 1,620 and Q: 1,100). In COG group G B.

477 subtilis strain MK101 has 1,592 genes and P. kermanshahensis strain MK113 has 1,313 genes. In P.
Pr

478 kermanshahensis strain MK113 genome 913 genes and in B. subtilis strain MK101 608 genes were

479 found within the COG group Q. This indicates that the genomes contain potential genes for azoxystrobin

480 degradation. Genes coding for the enzymes in the lactoylglutathione lyase family (COG0346, COG

18

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 481 group Q) are involved in the degradation of aromatic compounds. Nineteen genes from this COG0346

ed
482 were found in the three pesticide-degrading strains, precisely 12 genes in B. subtilis strain MK101, six

483 genes in P. kermanshahensis strain MK113 and one gene in R. fascians strain MK144 genome (Table

484 S4).

iew
485 After removing signal peptidases, Clp, Lex and Lon proteases, 14 subtilase-like serine proteases (E.C.

486 3.4.21.-) were found in the genome of B. subtilis strain MK101, no subtilase-like serine proteases were

487 found in R. fascians strain MK144 and P. kermanshahensis strain MK113 (Table S5).

v
488 In the last step, genes related to plant-microbe interactions were analyzed. The analysis showed that the

re
489 strains do not have genes responsible for siderophore production, however P. kermanshahensis strain

490 MK113 has genes encoding TonB-denpendent iron receptors (bfrG, btuB, cirA, exbB, fcuA, fecA, fepA,

491 fiuA, foxA, fpvA, iutA, oprC, phuR, ronB2, sftP, tonB, viuA and yncD). In the genomes of B. subtilis
er
492 strain MK101, P. kermanshahensis strain MK113 and R. fascians strain MK144 genes for IAA synthesis

493 pathway (trpA and trpB) were found. Concerning quorum sensing-related genes, B. subtilis strain
pe
494 MK101 and P. kermanshahensis strain MK113 showed related genes in their genomes. Precisely,

495 autoinducer-2 system was identified in B. subtilis strain MK101 genome, whereas the AHL-based

496 systems were identified in P. kermanshahensis strain MK113. The gene acdS was found in the genomes
ot

497 of P. kermanshahensis strain MK113 and R. fascians strain MK144, but not in B. subtilis strain MK101

498 (Table S6).

tn

499

500 4. Discussion
rin

501 Azoxystrobin-degrading bacteria were isolated from pesticide contaminated soils (Baćmaga et al., 2017;

502 Howell et al., 2014) or sludge from wastewater treatment systems (Feng et al., 2020b). However, to our

503 knowledge, azoxystrobin-degrading bacteria have not yet been isolated from the plant environment
ep

504 despite the fact that azoxystrobin is widely used as fungicide and residues are commonly found in plants.

505 Ochrobactrum anthropi SH14, as an isolate from wastewater treatment systems, was able to degrade
Pr

506 azoxystrobin in laboratory tests (Feng et al., 2020b). However, this strain was described as an

507 opportunistic pathogen that may cause infections in immunocompromised persons (Aguilera-Arreola et

508 al., 2018).

19

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 509 Azoxystrobin, as a fungicide of the β-methoxyacrylate group, is likely to be taken up by crop roots and

ed
510 translocated to aboveground plant tissues (Briggs et al., 1982; Romeh, 2015). We therefore isolated and

511 characterised potentially degrading strains from lettuce leaf and rhizosphere samples grown on

512 azoxystrobin contaminated soil as well as from multi-microbe products due to their fungicide degrading

iew
513 potential (personal communication with vegetable farmers).

514 Overall, a highly diverse strain collection was obtained comprising 59 unique bacterial strains belonging

515 to the Proteobacteria, Actinobacteria, Firmicutes and Bacteroidetes. From the rhizosphere, mostly

v
516 Proteobacteria and few Actinobacteria were obtained, whereas the leaf environment was dominated by

517 Actinobacteria and few Firmicutes. Multi-microbe products contained mostly Firmicutes. So far, only

re
518 few bacterial taxa, i.e., few Bacillus spp. (Baćmaga et al., 2017), Cupriavidus spp. and Rhodanobacter

519 spp. (Howell et al., 2014) have been reported to degrade azoxystrobin, whereas the range of bacterial

520
er
taxa generally degrading strobilurin fungicides is larger and comprises Bacillus spp., Pseudomonas spp.,

521 Klebsiella spp., Stenotrophomonas spp., Arthrobacter spp., Rhodanobacter spp., and Cupriavidus spp.
pe
522 (Feng et al., 2020). Cultivation-independent analysis also indicated, that Pseudomonas spp. are involved

523 in the degradation of pyractostrobin (Chen et al., 2018). In line with these results, several Bacillus and

524 Arthrobacter as well as Pseudomonas species were isolated. However, many isolated taxa have not been
ot

525 reported so far to degrade azoxystrobin or strobilurin fungicides. Interestingly, a high number of

526 Streptomyces spp. was isolated, which are primarily known for the production of a range of secondary
tn

527 metabolites (Lacey and Rutledge, 2022) but also for the degradation of various organic, aromatic

528 compounds like pesticides (Chen et al., 2012; dos Santos et al., 2017; Fuentes et al., 2013).

529 We selected strains from the genera Arthrobacter, Bacillus, Domibacillus Ensifer, Flavobacterium,
rin

530 Microbacterium, Nocardioides, Pseudoarthrobacter, Pseuodomonas, Rhodococcus and Variovorax

531 for in vitro azoxystrobin degradation tests on lamb’s lettuce leaves. Strains from the genera
ep

532 Flavobacterium, Pseudomonas, Arthrobacter, Rhodococcus, Ensifer and Bacillus showed high

533 degradation rates in vitro (ascending order). In contrast to bacteria from the genera Pseudomonas,

534 Arthrobacter and Bacillus (Baćmaga et al., 2017; X. Chen et al., 2018; Feng et al., 2020a), bacteria from
Pr

535 the genera Rhodococcus, Flavobacterium, and Ensifer have not been reported so far to degrade

536 azoxystrobin or strobilurin fungicides. However, these genera are also known for their potential to

20

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 537 degrade other pesticides. Flavobacterium spp. are known for the degradation of herbicides, i.e., 2,4-

ed
538 dichlorophenoxyacetic acid and atrazine and also for the degradation of cyclic ether 1,4-dioxane (Silva

539 et al., 2007; Smith et al., 2005; Sun et al., 2011). Bacteria from genera Ensifer are known as nitrogen-

540 fixing bacteria (Fagorzi et al., 2020), but there are also publications that show that Ensifer spp., in

iew
541 particular Ensifer adhaerens strains, are able to degrade insecticides i.e., thiamethoxam or flonicamid

542 (Zhao et al., 2021; Zhou et al., 2013).

543 Strains with high degradation rates in vitro were further tested in greenhouse tests. Here, Arthrobacter

v
544 sp. strain MK97 was a notable degrader in the lettuce leaves when both bacteria and fungicide were

545 applied to soil. Surface-sterilised lettuce seeds were imbibed with bacterial suspension and bacteria were

re
546 applied again by immersing the root system in bacterial suspension. Due to the fact that degradation of

547 azoxystrobin was high in leaves, we can speculate that Arthrobacter sp. strain MK97 might colonize the

548
er
leaves predominantly from the inside of plants after seed inoculation and root immersion. Some

549 Arthrobacter spp. are also known to colonize as endophytes (Orozco-Mosqueda and Santoyo, 2021).
pe
550 Some Arthrobacter strains are known to degrade trifloxystrobin (Howell et al., 2014), but there are also

551 publications that show that Arthrobacter spp. are able to degrade pesticides like DDT, endosulfan, PCP

552 or toxaphene O (reviewed by Parte et al., 2017).

ot

553 There are only a few studies describing the uptake of the active ingredient within a plant system. In this

554 study, 2% of the active ingredient which was amended to the soil were detected in leaves of control
tn

555 plants (not inoculated with bacteria). Ju et al. (2019) described the uptake mechanism and translocation

556 characteristics of azoxystrobin in crop plants using wheat under laboratory conditions and used a

557 partition-limited model for predicting the uptake and distribution in the plants. Although the results
rin

558 showed that the major site for azoxystrobin accumulation in wheat plants were the roots, around 12%

559 of the azoxystrobin from the solution accumulated in the leaves and remained stable. The results were
ep

560 predicted well by the partition-limited model (Chiou et al., 2001). Azoxystrobin tends to remain in the

561 soil phase and not to migrate down into groundwater (Wu et al., 2016). More uptake studies with

562 azoxystrobin would be needed to get a better understanding of the translocation factor of azoxystrobin
Pr

563 from the soil to the leaves under different conditions.

21

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 564 Based on their comparably high azoxystrobin degradation activities, Bacillus subtilis strain MK101,

ed
565 Pseudomonas kermanshahensis strain MK113 and Rhodococcus fascians strain MK144 were selected

566 for whole genome sequencing to elucidate more information on the potential mechanisms involved in

567 the degradation.

iew
568 Due to the complex structures of strobilurin-derived substances, their catabolic reactions follow multiple

569 pathways. To the best of our knowledge, no universal pathway for the microbial degradation of

570 strobilurin fungicides is available. Pathways for strobilurin degradation involve ester hydrolysis, ring

v
571 hydroxylation, double bond reduction and/or photolytic reaction. The carboxylester hydrolysis via

re
572 esterase is the primary degradation mechanism of strobilurins in microorganisms (reviewed by Feng et

573 al., 2020). In detail, a bacterial consortium was able to degrade azoxystrobin through hydrolysis under

574 the catalysis of carboxylesterases and showed 2-{2-[6-(2-cyano-phenoxy)-pyrimidin-4-yloxy]-phenyl}-

er
575 3-methoxy-acrylic acid as metabolite (Wang et al., 2022). Microbial azoxystrobin degradation was also

576 reported by hydrolysis of the ester linkage and cleavage of the aromatic ring (Feng et al., 2020b). In the
pe
577 three strains analysed in this study, we found pathways connected to the degradation of cyclic

578 compounds. Most hits for pathways predicted to be involved in biodegradation were found in the

579 genome of the P. kermanshahensis strain (Table S1). Pathways found in this genome included benzoate,
ot

580 aminobenzoate, chlorobenzene, naphthalene, styrene and caprolactam degradation. These results are in

581 line with the literature, strains from the genus Pseudomonas have already been reported to degrade
tn

582 strobilurin fungicides (X. Chen et al., 2018; Wang et al., 2022). The R. fascians hosted, for example, the

583 styrene pathway connected to the degradation of cyclic compounds. The members of the genus
rin

584 Rhodococcus have been described for decades as being able to degrade numerous aromatic compounds

585 (Rast et al., 1980; Warhurst and Fewson, 1994), and strains degrading herbicides (i.e., chlorimuron-

586 ethyl) or components of pesticides (i.e. pyridine) (Yoon et al., 2000; Zang et al., 2019). Within the B.
ep

587 subtilis genome the lowest number of pathways connected to the degradation of cyclic compounds was

588 found. However, pathways for the degradation of benzothiopene or other xenobiotics were found (Table
Pr

589 S1). Bacterial consortia containing bacteria of the genus Bacillus e.g., B. cereus, B. weihenstephanensis,

590 B. megaterium have already been reported to degrade azoxystrobin in soil (Baćmaga et al., 2017).

22

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 591 Several genes for benzoate degradation (i.e., catA, fadB or pcaC and aminobenzoate i.e., pdhC, ubiX or

ed
592 vanB) may be responsible for azoxystrobin degradation by cleaving aromatic rings and were found in

593 all three analysed genomes (Table S3). Furthermore, genes coding for the enzymes in the

594 lactoylglutathione lyase family are known to be involved in the degradation of aromatic compounds by

iew
595 employing dioxygenase enzymes to activate and cleave the aromatic ring (Mesarch et al., 2000). Genes

596 encoding the enzymes in the lactoylglutathione lyase family (COG0346) (Elufisan et al., 2020) were

597 found in all three sequenced bacterial genomes, particularly in the B. subtilis strain MK101 genome

v
598 (COG0346; Table S4).

re
599 Enzymes are grouped according to their biochemical activity as defined by their enzyme commission

600 (EC) classification (Scott et al., 2008). Oxidoreductases (EC 1), hydrolases (EC 3) and lyases (EC 4)

601 form the basis of several bioremediation strategies (Scott et al., 2008), and esterases, in particular, play
er
602 a prominent role in the degradation of ester-containing pesticides (Bhatt et al., 2021b). In the genome

603 of B. subtilis strain MK101 we found fourteen potential strobilurin-degrading proteases (Table S5).
pe
604 However, we could not find any specific gene previously reported to be involved in pesticide

605 degradation.

606 Besides their capacity to degrade azoxystrobin, their capability to increase plant growth was analysed.
ot

607 Therefore, genes known to be involved in plant growth promotion were analysed, and were found in the

608 genomes of all three strains. B. subtilis strain MK101, P. kermanshahensis strain MK113 and R. fascians
tn

609 strain MK144 host genes responsible for the production of indole-3-acetic acid (IAA), the major

610 naturally occurring auxin. Auxins are frequently found in plant-associated bacteria and play a central
rin

611 role in plant growth and plant body development (Mitter et al., 2013). Furthermore, P. kermanshahensis

612 strain MK113 has genes encoding TonB-dependent iron receptors, which can promote plant growth

613 indirect by the transport of ferric-siderophore complexes (Mitter et al., 2013). According to its genome,
ep

614 this strain is also able to produce 1-aminocyclopropane-1-carboxylate (ACC) deaminase, which

615 degrades ACC, the precursor of ethylene, in plants to 2-oxobutyrate and ammonia. ACC deaminase
Pr

616 reduces growth-retarding effects of ethylene produced by plants in response to biotic and abiotic stress

617 (Glick, 2005). Quorum sensing is involved in the regulation of many microbial activities, and

618 particularly N-acyl-homoserine lactones (AHL) are key regulating molecules in quorum sensing

23

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 619 (Hartmann et al., 2021) in Proteobacteria. AHLs produced by soil bacteria have also been found to

ed
620 induce a plant response, such as resistance towards pathogens (Liu et al., 2020; Schikora et al., 2016).

621 In this study, the P. kermanshahensis strain MK113 genome suggests the production of AHLs. In

622 addition, B. subtilis strain MK101 and R. fascians strain MK144 possess genes responsible for IAA

iew
623 production, the genome of R. fascians strain MK144 also suggest ACC deaminase activity in this strain.

624 These results indicate that all three strains have the potential to enhance the stress resilience and improve

625 the growth of plants, however, further studies are required to confirm these functions.

v
626 In summary, we conclude that the rhizosphere and the leaves of lettuce plants grown on fungicide

re
627 contaminated soils are a reservoir for bacteria that can potentially degrade active fungicide ingredients.

628 Three strains, B. subtilis strain MK101, P. kermanshahensis strain MK113 and R. fascians strain MK144

629 were found to be efficient azoxystrobin degraders and their genomes suggest multiple pathways
er
630 potentially involved in degradation and plant growth promotion.

631
pe
632 Appendix A. Supplementary data

633 Supplemented data to this article can be found online at

634
ot

635 Declaration of competing interest

636 The authors declare that they have no known competing financial interests or personal relationships that
tn

637 could have appeared to influence the work reported in this paper.

638
rin

639 Acknowledgements

640 This work was supported by the Austrian Research Promotion Agency (FFG), Grant No. 25080812.
ep

641
642 References
Pr

643 Aguilera-Arreola, M.G., Ostria-Hernandez, M.L., Albarran-Fernandez, E., Juarez-Enriquez, S.R.,

644 Majalca-Martinez, C., Rico-Verdin, B., Ruiz, E.A., Ruiz-Palma, M.D., Morales-Garcia, M.R.,

645 Contreras-Rodriguez, A., 2018. Correct Identification of Ochrobactrum anthropi From Blood

24

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 646 Culture Using 16rRNA Sequencing: A First Case Report in an Immunocompromised Patient in

ed
647 Mexico. Front. Med. 5, 1–6. https://doi.org/10.3389/fmed.2018.00205

648 Ali, A.A.I., Ashour, M., Tohamy, M., Ragheb, D.A., 2015. Azoxystrobin residues on tomato leaves

649 and fruits. Zaganzig J. Plant Prot. Res. 42, 1547–1553.

iew
650 Andrews, S., 2010. FastQC: a quality control tool for high throughput sequencing data [WWW

651 Document]. URL https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ (accessed

652 8.28.20).

v
653 Ashburner, M., Ball, C.A., Blake, J.A., Botstein, D., Butler, H., Cherry, M., Davis, A.P., Dolinski, K.,

re
654 Dwight, S.S., Eppig, J.T., Harris, M.A., Hill, D.P., Issel-Traver, L., Kasarskis, A., Lewis, S.,

655 Matese, J.C., Richardson, J.E., Ringlwald, M., Rubin, G.M., Sherlock, G., 2000. Gene Ontology:

656
er
tool for the unification of biology. Nat Genet 25, 25–29. https://doi.org/10.1038/75556.Gene

657 Baćmaga, M., Wyszkowska, J., Kucharski, J., 2017. Bioaugmentation of Soil Contaminated with
pe
658 Azoxystrobin. Water, Air Soil Pollut. 228, 1–9. https://doi.org/10.1007/s11270-016-3200-9

659 Bankevich, A., Nurk, S., Antipov, D., Gurevich, A.A., Dvorkin, M., Kulikov, A.S., Lesin, V.M.,

660 Nikolenko, S.I., Pham, S., Prjibelski, A.D., Pyshkin, A. V., Sirotkin, A. V., Vyahhi, N., Tesler,
ot

661 G., Alekseyev, M.A., Pevzner, P.A., 2012. SPAdes: A New Genome Assembly Algorithm and

662 Its Applications to Single-Cell Sequencing. J. Comput. Biol. 19, 455–477.

tn

663 https://doi.org/10.1089/cmb.2012.0021

664 Bartlett, D.W., Clough, J.M., Godwin, J.R., Hall, A.A., Hamer, M., Parr-Dobrzanski, B., 2002. The
rin

665 strobilurin fungicides. Pest Manag. Sci. 58, 649–662. https://doi.org/10.1002/ps.520

666 Berg, G., Smalla, K., 2009. Plant species and soil type cooperatively shape the structure and function
ep

667 of microbial communities in the rhizosphere. FEMS Microbiol. Ecol. 68, 1–13.

668 https://doi.org/10.1111/j.1574-6941.2009.00654.x
Pr

669 Bhagobaty, R.K., Malik, A., 2008. Utilization of chlorpyrifos as a sole source of carbon by bacteria

670 isolated from wastewater irrigated agricultural soils in an industrial area of western Uttar

671 Pradesh, India. Res. J. Microbiol. 3, 293–307. https://doi.org/10.3923/jm.2008.293.307

25

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 672 Bhatt, P., Bhandari, G., Bhatt, K., Maithani, D., Mishra, S., Gangola, S., Bhatt, R., Huang, Y., Chen,

ed
673 S., 2021a. Plasmid-mediated catabolism for the removal of xenobiotics from the environment. J.

674 Hazard. Mater. 420, 126618. https://doi.org/10.1016/j.jhazmat.2021.126618

675 Bhatt, P., Zhou, X., Huang, Y., Zhang, W., Chen, S., 2021b. Characterization of the role of esterases

iew
676 in the biodegradation of organophosphate, carbamate, and pyrethroid pesticides. J. Hazard.

677 Mater. 411, 125026. https://doi.org/10.1016/j.jhazmat.2020.125026

678 Bonatelli, M.L., Lacerda-Júnior, G.V., dos Reis Junior, F.B., Fernandes-Júnior, P.I., Melo, I.S.,

v
679 Quecine, M.C., 2021. Beneficial Plant-Associated Microorganisms From Semiarid Regions and

re
680 Seasonally Dry Environments: A Review. Front. Microbiol. 11, 1–16.

681 https://doi.org/10.3389/fmicb.2020.553223

682
er
Briggs, G.G., Bromilow, R.H., Evans, A.A., 1982. Relationships between lipophilicity and root uptake

683 and translocation of non‐ionised chemicals by barley. Pestic. Sci. 13, 495–504.
pe
684 https://doi.org/10.1002/ps.2780130506

685 BVL, 2007. Berichte zur Lebensmittelsicherheit 2007. Tabellenband zum Bericht über die Montoring-

686 Ergebnisse des Jahres 2007. https://doi.org/10.1007/978-3-7643-8913-0

ot

687 Caspi, R., Altman, T., Billington, R., Dreher, K., Foerster, H., Fulcher, C.A., Holland, T.A., Keseler,

688 I.M., Kothari, A., Kubo, A., Krummenacker, M., Latendresse, M., Mueller, L.A., Ong, Q., Paley,
tn

689 S., Subhraveti, P., Weaver, D.S., Weerasinghe, D., Zhang, P., Karp, P.D., 2014. The MetaCyc

690 database of metabolic pathways and enzymes and the BioCyc collection of Pathway/Genome
rin

691 Databases. Nucleic Acids Res. 42, 459–471. https://doi.org/10.1093/nar/gkt1103

692 Chaumeil, P.A., Mussig, A.J., Hugenholtz, P., Parks, D.H., 2020. GTDB-Tk: a toolkit to classify
ep

693 genomes with the Genome Taxonomy Database. Bioinformatics 36, 1925–1927.

694 https://doi.org/10.1093/bioinformatics/btz848

695
Pr

Chen, S., Luo, J., Hu, M., Lai, K., Geng, P., Huang, H., 2012. Enhancement of cypermethrin

696 degradation by a coculture of Bacillus cereus ZH-3 and Streptomyces aureus HP-S-01.

697 Bioresour. Technol. 110, 97–104. https://doi.org/10.1016/j.biortech.2012.01.106

26

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 698 Chen, S., Zhou, Y., Chen, Y., Gu, J., 2018. fastp: an ultra-fast all-in-one FASTQ preprocessor.

ed
699 Bioinformatics 34, i884–i890. https://doi.org/10.1093/bioinformatics/bty560

700 Chen, X., He, S., Liang, Z., Li, Q.X., Yan, H., Hu, J., Liu, X., 2018. Biodegradation of pyraclostrobin

701 by two microbial communities from Hawaiian soils and metabolic mechanism. J. Hazard. Mater.

iew
702 354, 225–230. https://doi.org/10.1016/j.jhazmat.2018.04.067

703 Chiou, C.T., Sheng, G., Manes, M., 2001. A partition-limited model for the plant uptake of organic

704 contaminants from soil and water. Environ. Sci. Technol. 35, 1437–1444.

v
705 https://doi.org/10.1021/es0017561

re
706 Clinton, B., Warden, A.C., Haboury, S., Easton, C.J., Kotsonis, S., Taylor, M.C., Oakeshott, J.G.,

707 Russell, R.J., Scott, C., 2011. Bacterial degradation of strobilurin fungicides: A role for a

708
er
promiscuous methyl esterase activity of the subtilisin proteases? Biocatal. Biotransform. 29,

709 119–129. https://doi.org/10.3109/10242422.2011.578740

pe
710 dos Santos, J.C., Batista, I.H., Barroso, H.S., Lima, J.M.S., Costa Neto, P. de Q., Ghelfi, A., Pereira,

711 J.O., 2017. Biodegradation of the fungicide carbendazim by bacteria from Coriandrum sativum

712 L. rhizosphere. Acta Sci. - Biol. Sci. 39, 71–77.

ot

713 https://doi.org/10.4025/actascibiolsci.v39i1.34067

714 Drula, E., Garron, M.L., Dogan, S., Lombard, V., Henrissat, B., Terrapon, N., 2022. The
tn

715 carbohydrate-active enzyme database: functions and literature. Nucleic Acids Res. 50, D571–

716 D577. https://doi.org/10.1093/nar/gkab1045

rin

717 Edwards, U., Rogall, T., Blöcker, H., Emde, M., Böttger, E.C., 1989. Isolation and direct complete

718 nucleotide determination of entire genes. Characterization of a gene coding for 16S ribosomal
ep

719 RNA. Nucleic Acids Res. 17, 7843–7853.

720 Elufisan, T.O., Rodríguez-Luna, I.C., Oyedara, O.O., Sánchez-Varela, A., Hernández-Mendoza, A.,

721
Pr

Gonzalez, E.D., Paz-González, A.D., Muhammad, K., Rivera, G., Villalobos-Lopez, M.A., Guo,

722 X., 2020. The Polycyclic Aromatic Hydrocarbon (PAH) degradation activities and genome

723 analysis of a novel strain Stenotrophomonas sp. Pemsol isolated from Mexico. PeerJ 2020.

27

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 724 https://doi.org/10.7717/peerj.8102

ed
725 Fagorzi, C., Ilie, A., Decorosi, F., Cangioli, L., Viti, C., Mengoni, A., di Cenzo, G.C., 2020. Symbiotic

726 and Nonsymbiotic Members of the Genus Ensifer (syn. Sinorhizobium) Are Separated into Two

727 Clades Based on Comparative Genomics and High-Throughput Phenotyping. Genome Biol.

iew
728 Evol. 12, 2521–2534. https://doi.org/10.1093/GBE/EVAA221

729 Feng, Y., Huang, Y., Zhan, H., Bhatt, P., Chen, S., 2020a. An Overview of Strobilurin Fungicide

730 Degradation:Current Status and Future Perspective. Front. Microbiol. 11, 1–11.

v
731 https://doi.org/10.3389/fmicb.2020.00389

re
732 Feng, Y., Zhang, W., Pang, S., Lin, Z., Zhang, Y., Huang, Y., Bhatt, P., Chen, S., 2020b. Kinetics and

733 New Mechanism of Azoxystrobin Biodegradation by an Ochrobactrum anthropi Strain SH14.

734
er
Microorganisms 8, 1–17. https://doi.org/10.3390/microorganisms8050625

735 Fuentes, M.S., Briceño, G.E., Saez, J.M., Benimeli, C.S., Diez, M.C., Amoroso, M.J., 2013. Enhanced
pe
736 Removal of a Pesticides mixture by Single Cultures and Consortia of Free and Immobilized

737 Streptomyces Strains. Biomed Res. Int. 2013, 1–9. https://doi.org/10.1155/2013/392573

738 Gao, J., Ellis, L.B.M., Wackett, L.P., 2009. The University of Minnesota Biocatalysis/Biodegradation
ot

739 Database: Improving public access. Nucleic Acids Res. 38, 488–491.

740 https://doi.org/10.1093/nar/gkp771
tn

741 Glick, B.R., 2005. Modulation of plant ethylene levels by the bacterial enzyme ACC deaminase.

742 FEMS Microbiol. Lett. 251, 1–7. https://doi.org/10.1016/j.femsle.2005.07.030

rin

743 Graves, S., Piepho, H., Selzer, L., 2019. multcompView: Visualizations of Paired Comparisons. R

744 package version 0.1-8.

ep

745 Gurevich, A., Saveliev, V., Vyahhi, N., Tesler, G., 2013. QUAST: quality assessment tool for genome

746 assemblies. Bioinformatics 29, 1072–1075. https://doi.org/10.1093/bioinformatics/btt086

Pr

747 Hardoim, P.R., van Overbeek, L.S., Berg, G., Pirttilä, A.M., Compant, S., Campisano, A., Döring, M.,

748 Sessitsch, A., 2015. The Hidden World within Plants: Ecological and Evolutionary

28

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 749 Considerations for Defining Functioning of Microbial Endophytes. Microbiol. Mol. Biol. Rev.

ed
750 79, 293–320. https://doi.org/10.1128/mmbr.00050-14

751 Hartmann, A., Klink, S., Rothballer, M., 2021. Importance of n-acyl-homoserine lactone-based

752 quorum sensing and quorum quenching in pathogen control and plant growth promotion.

iew
753 Pathogens 10. https://doi.org/10.3390/pathogens10121561

754 Hothorn, T., Zeileis, A., 2015. partykit : A Toolkit for Recursive Partytioning. J. Mach. Learn. Res.

755 16, 3905–3909.

v
756 Howell, C.C., Semple, K.T., Bending, G.D., 2014. Isolation and characterisation of azoxystrobin

re
757 degrading bacteria from soil. Chemosphere 95, 370–378.

758 https://doi.org/10.1016/j.chemosphere.2013.09.048

759
er
Huerta-Cepas, J., Szklarczyk, D., Heller, D., Hernández-Plaza, A., Forslund, S.K., Cook, H., Mende,

760 D.R., Letunic, I., Rattei, T., Jensen, L.J., Von Mering, C., Bork, P., 2019. eggNOG 5.0: a
pe
761 hierarchical, functionally and phylogenetically annotated orthology resource based on 5090

762 organisms and 2502 viruses. Nucleic Acids Res. 47, D309–D314.

763 https://doi.org/10.1093/nar/gky1085
ot

764 Jiang, W., Gao, Q., Zhang, L., Liu, Y., Zhang, M., Ke, Z., Zhou, Y., Hong, Q., 2021. Detoxification

765 Esterase StrH Initiates Strobilurin Fungicide Degradation in Hyphomicrobium sp. Strain DY-1
tn

766 87, 1–13. https://doi.org/10.1128/AEM.00103-21

767 Ju, C., Zhang, H., Yao, S., Dong, S., Cao, D., Wang, F., Fang, H., Yu, Y., 2019. Uptake,
rin

768 Translocation, and Subcellular Distribution of Azoxystrobin in Wheat Plant (Triticum aestivum

769 L.). J. Agric. Food Chem. 67, 6691–6699. https://doi.org/10.1021/acs.jafc.9b00361

ep

770 Kandel, S.L., Joubert, P.M., Doty, S.L., 2017. Bacterial endophyte colonization and distribution within

771 plants. Microorganisms 5, 9–11. https://doi.org/10.3390/microorganisms5040077

Pr

772 Kanehisa, M., Furumichi, M., Sato, Y., Kawashima, M., Ishiguro-Watanabe, M., 2022. KEGG for

773 taxonomy-based analysis of pathways and genomes. Nucleic Acids Res. 1–6.

774 https://doi.org/10.1093/nar/gkac963

29

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 775 Katagi, T., 2006. Behavior of pesticides in water-sediment systems. Rev. Environ. Contam. Toxicol.

ed
776 187, 133–251. https://doi.org/10.1007/0-387-32885-8_4

777 Kuhn, M., 2022. caret: Classification and Regression Training. R package version 6.0-93.

iew
778 Lacey, H.J., Rutledge, P.J., 2022. Recently Discovered Secondary Metabolites from Streptomyces

779 Species. Molecules 27. https://doi.org/10.3390/molecules27030887

780 Laetsch, D.R., Koutsovoulos, G., Booth, T., Stajich, J., Kumar, S., 2017. BlobTools v1.0.1. Zenodo

v
781 [WWW Document]. https://doi.org/10,5281/zenodo.845347

re
782 Lane, D., 1991. Nucleic acid techniques in bacterial systematics. New York.

783 Langmead, B., Salzberg, S.L., 2012. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–

784 359. https://doi.org/10.1038/nmeth.1923

er
785 Lenth R, 2022. emmeans: Estimated Marginal Means, aka Least-Squares Means. R package version
pe
786 1.8.2.

787 Leveau, J.H.J., 2018. Microbial Communities in the Phyllosphere. Annu. Plant Rev. 334–367.

788 https://doi.org/10.1002/9781119312994.apr0239
ot

789 Li, L., Li, W., Quin, D., Jiang, S., Liu, F., 2009. Application of Graphitized Carbon Black to the

790 QuEChERS Method for Pesticide Multiresidue Analysis in Spinach. J. AOAC Int. 92, 538–547.
tn

791 Li, W., Dai, Y., Xue, B., Li, Y., Peng, X., Zhang, J., Yan, Y., 2009. Biodegradation and detoxification

792 of endosulfan in aqueous medium and soil by Achromobacter xylosoxidans strain CS5. J. Hazard.
rin

793 Mater. 167, 209–216. https://doi.org/10.1016/j.jhazmat.2008.12.111

794 Liu, F., Zhao, Q., Jia, Z., Song, C., Huang, Y., Ma, H., Song, S., 2020. N-3-oxo-octanoyl-homoserine
ep

795 lactone-mediated priming of resistance to Pseudomonas syringae requires the salicylic acid

796 signaling pathway in Arabidopsis thaliana. BMC Plant Biol. 20, 1–13.

797 https://doi.org/10.1186/s12870-019-2228-6
Pr

798 Liu, W.T., Marsh, T.L., Cheng, H., Forney, L.J., 1997. Characterization of Microbial Diversity by

799 Determining Terminal Restriction Fragment Length Polymorphisms of Genes Encoding 16S

30

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 800 rRNA. Appl. Environ. Microbiol. 63, 4516–4522. https://doi.org/10.1128/aem.63.11.4516-

ed
801 4522.1997

802 Massol-Deya, A.A., Odelson, D.A., Hickey, R.F., Tiedje, J.M., 1995. Bacterial community

803 fingerprinting of amplified 16S and 16–23S ribosomal DNA gene sequences and restriction

iew
804 endonuclease analysis (ARDRA). Mol. Microb. Ecol. Man. 3.3.2, 1–8.

805 https://doi.org/10.1007/978-94-011-0351-0_20

806 Medina-Pastor, P., Triacchini, G., 2020. The 2018 European Union report on pesticide residues in

v
807 food. EFSA J. 18, 1–103. https://doi.org/10.2903/j.efsa.2020.6057

re
808 Meier, U., 2001. BBCH Monograph. Quedlinburg. https://doi.org/10.5073/20180906-074619

809 Mesarch, M.B., Nakatsu, C.H., Nies, L., 2000. Development of Catechol 2,3-Dioxygenase-Specific
er
810 Primers for Monitoring Bioremediation by Competitive Quantitative PCR. Appl. Environ.

811 Microbiol. 66, 678–683. https://doi.org/10.1128/AEM.66.2.678-683.2000

pe
812 Miethke, M., Marahiel, M.A., 2007. Siderophore-Based Iron Acquisition and Pathogen Control.

813 Microbiol. Mol. Biol. Rev. 71, 413–451. https://doi.org/10.1128/mmbr.00012-07

ot

814 Mitter, B., Brader, G., Afzal, M., Compant, S., Naveed, M., Trognitz, F., Sessitsch, A., 2013.

815 Advances in Elucidating Beneficial Interactions Between Plants, Soil, and Bacteria. Adv. Agron.
tn

816 121, 381–445. https://doi.org/10.1016/B978-0-12-407685-3.00007-4

817 Mohite, B., 2013. Isolation and characterization of indole acetic acid (IAA) producing bacteria from
rin

818 rhizospheric soil and its effect on plant growth. J. Soil Sci. Plant Nutr. 13, 638–649.

819 https://doi.org/10.4067/S0718-95162013005000051

820 Munazza, A., Sheikh, A.R., Sikander khan, S., Tasneem, A.A., 2012. High Profile Chlorpyrifos
ep

821 Degrading Pseudomonas putida MAS-1 from Indigenous Soil: Gas Chromatographic Analysis

822 and Molecular Characterization. Int. J. Basic Med. Sci. Pharm. 2, 58–61.
Pr

823 National Center for Biology Information, 2022. PubChem Compound Summary for CID 3034285,

824 Azoxystrobin [WWW Document]. URL

31

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 825 https://pubchem.ncbi.nlm.nih.gov/compound/Azoxystrobin (accessed 7.18.22).

ed
826 Nofiani, R., de Mattos-Shipley, K., Lebe, K.E., Han, L.C., Iqbal, Z., Bailey, A.M., Willis, C.L.,

827 Simpson, T.J., Cox, R.J., 2018. Strobilurin biosynthesis in Basidiomycete fungi. Nat. Commun.

828 9, 1–11. https://doi.org/10.1038/s41467-018-06202-4

iew
829 Okonechnikov, K., Conesa, A., García-Alcalde, F., 2016. Qualimap 2: advanced multi-sample quality

830 control for high-throughput sequencing data. Bioinformatics 32, 292–294.

831 https://doi.org/10.1093/bioinformatics/btv566

v
832 Orozco-Mosqueda, M. del C., Glick, B.R., Santoyo, G., 2020. ACC deaminase in plant growth-

re
833 promoting bacteria (PGPB): An efficient mechanism to counter salt stress in crops. Microbiol.

834 Res. 235, 126439. https://doi.org/10.1016/j.micres.2020.126439

835
er
Orozco-Mosqueda, M. del C., Santoyo, G., 2021. Plant-microbial endophytes interactions:

836 Scrutinizing their beneficial mechanisms from genomic explorations. Curr. Plant Biol. 25, 1–10.
pe
837 https://doi.org/10.1016/j.cpb.2020.100189

838 Parks, D.H., Imelfort, M., Skennerton, C.T., Hugenholtz, P., Tyson, G.W., 2015. CheckM: assessing

839 the quality of microbial genomes recovered from isolates, single cells, and metagenomes.
ot

840 Genome Res. 25, 1043–1055. https://doi.org/10.1101/gr.186072.114

tn

841 Parte, S.G., Mohekar, A.D., Kharat, A.S., 2017. Microbial degradation of pesticide: A review. African

842 J. Microbiol. Res. 11, 992–1012. https://doi.org/10.5897/ajmr2016.8402

rin

843 Philippot, L., Raaijmakers, J.M., Lemanceau, P., Van Der Putten, W.H., 2013. Going back to the

844 roots: The microbial ecology of the rhizosphere. Nat. Rev. Microbiol. 11, 789–799.

845 https://doi.org/10.1038/nrmicro3109
ep

846 Rast, H.G., Engelhardt, G., Wallnöfer, P.R., 1980. Degradation of aromatic compounds in the

847 actinomycete-genus Rhodococcus. FEMS Microbiol. Lett. 7, 1–6. https://doi.org/10.1016/S0378-

Pr

848 1097(80)80048-6

849 Rastogi, G., Coaker, G.L., Leveau, J.H.J., 2013. New insights into the structure and function of

32

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 850 phyllosphere microbiota through high-throughput molecular approaches. FEMS Microbiol. Lett.

ed
851 348, 1–10. https://doi.org/10.1111/1574-6968.12225

852 Reiter, B., Pfeifer, U., Schwab, H., Sessitsch, A., 2002. Response of Endophytic Bacterial

853 Communities in Potato Plants to Infection with Erwinia carotovora subsp. atroseptica. Appl.

iew
854 Environ. Microbiol. 68, 2261–2268. https://doi.org/10.1128/AEM.68.5.2261-2268.2002

855 Romeh, A.A., 2015. Evaluation of the phytoremediation potential of three plant species for

856 azoxystrobin-contaminated soil. Int. J. Environ. Sci. Technol. 3509–3518.

v
857 https://doi.org/10.1007/s13762-015-0772-7

re
858 Schikora, A., Schenk, S.T., Hartmann, A., 2016. Beneficial effects of bacteria-plant communication

859 based on quorum sensing molecules of the N-acyl homoserine lactone group. Plant Mol. Biol. 90,

860
er
605–612. https://doi.org/10.1007/s11103-016-0457-8

861 Schwengers, O., Barth, P., Falgenhauer, L., Hain, T., Chakraborty, T., Goesmann, A., 2020. Platon:
pe
862 identification and characterization of bacterial plasmid contigs in short-read draft assemblies

863 exploiting protein sequence-based replicon distribution scores. Microb. Genomics 6, 1–12.

864 https://doi.org/10.1099/mgen.0.000398
ot

865 Scott, C., Pandey, G., Hartley, C.J., Jackson, C.J., Cheesman, M.J., Taylor, M.C., Pandey, R.,

866 Khurana, J.L., Teese, M., Coppin, C.W., Weir, K.M., Jain, R.K., Lal, R., Russell, R.J.,
tn

867 Oakeshott, J.G., 2008. The enzymatic basis for pesticide bioremediation. Indian J. Microbiol. 48,

868 65–79. https://doi.org/10.1007/s12088-008-0007-4

rin

869 Seemann, T., 2021. tseemann/abricate [WWW Document]. URL https://github.com/tseemann/abricate

870 (accessed 3.10.21).

ep

871 Seemann, T., 2014. Prokka: rapid prokaryotic genome annotation. Bioinformatics 30, 2068–2069.

872 https://doi.org/10.1093/bioinformatics/btu153
Pr

873 Sevigné Itoiz, E., Fantke, P., Juraske, R., Kounina, A., Vallejo, A.A., 2012. Deposition and residues of

874 azoxystrobin and imidacloprid on greenhouse lettuce with implications for human consumption.

875 Chemosphere 89, 1034–1041. https://doi.org/10.1016/j.chemosphere.2012.05.066

33

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 876 Silva, T.M., Stets, M.I., Mazzetto, A.M., D.Andrade, F., Pileggi, S.A.V., Fávero, P.R., Cantú, M.D.,

ed
877 Carrilho, E., Carneiro, P.I.B., Pileggi, M., 2007. Degradation of 2,4-D herbicide by

878 microorganisms isolated from Brazilian contaminated soil. Brazilian J. Microbiol. 38, 522–525.

879 https://doi.org/10.1590/S1517-83822007000300026

iew
880 Simão, F.A., Waterhouse, R.M., Ioannidis, P., Kriventseva, E. V., Zdobnov, E.M., 2015. BUSCO:

881 assessing genome assembly and annotation completeness with single-copy orthologs.

882 Bioinformatics 31, 3210–3212. https://doi.org/10.1093/bioinformatics/btv351

v
883 Smith, D., Alvey, S., Crowley, D.E., 2005. Cooperative catabolic pathways within an atrazine-

re
884 degrading enrichment culture isolated from soil. FEMS Microbiol. Ecol. 53, 265–275.

885 https://doi.org/10.1016/j.femsec.2004.12.011

886
er
Sun, B., Ko, K., Ramsay, J.A., 2011. Biodegradation of 1,4-dioxane by a Flavobacterium.

887 Biodegradation 22, 651–659. https://doi.org/10.1007/s10532-010-9438-9

pe
888 Sun, H., Lueckl, J., 2018. Ergebnisse des nationalen Kontrollprogramms Pestizide 2018, in: Pestizid-

889 Rückstände in Pflanzlichen und Tierischen Lebensmitteln. AGES Österreichische Agentur für

890 Gesundheit und Ernährungssicherheit, Austria.

ot

891 Tatusov, R.L., Galperin, M.Y., Natale, D.A., Koonin, E. V., 2000. The COG database: a tool for

892 genome-scale analysis of protein functions and evolution. Nucleic Acids Res. 28, 33–36.
tn

893 https://doi.org/10.1093/nar/28.1.33

894 The Gene Ontology Consortium, 2021. The Gene Ontology resource: Enriching a GOld mine. Nucleic
rin

895 Acids Res. 49, D325–D334. https://doi.org/10.1093/nar/gkaa1113

896 Wang, W., Zhao, Z., Yan, H., Zhang, H., Li, Q.X., Liu, X., 2022. Carboxylesterases from bacterial
ep

897 enrichment culture degrade strobilurin fungicides. Sci. Total Environ. 814, 152751.

898 https://doi.org/10.1016/j.scitotenv.2021.152751
Pr

899 Warhurst, A.M., Fewson, C.A., 1994. Biotransformations Catalyzed by the Genus Rhodococcus. Crit.

900 Rev. Biotechnol. 14, 29–73. https://doi.org/10.3109/07388559409079833

34

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 901 Weisburg, W.G., Barns, S.M., Pelletier, D.A., Lane, D.J., 1991. 16S Ribosomal DNA Amplification

ed
902 for Phylogenetic Study. J. Bateriology 173, 697–703.

903 Wickham, H., Navarro, D., Lin Pedersen, T., 2016. ggplot2: Elegant Graphics for Data Analysis.

904 Springer-Verlag, New York.

iew
905 Wright, E.S., 2016. Using DECIPHER v2.0 to Analyze Big Biological Sequence Data in R. R J. 8,

906 352–359. https://doi.org/10.32614/rj-2016-025

v
907 Wu, P., Wu, W.Z., Han, Z.H., Yang, H., 2016. Desorption and mobilization of three strobilurin

908 fungicides in three types of soil. Environ. Monit. Assess. 188. https://doi.org/10.1007/s10661-

re
909 016-5372-6

910 Yi, S.Y., Cui, Y., Zhao, Y., Liu, Z.D., Lin, Y.J., Zhou, F., 2016. A Novel Naturally Occurring Class I
er
911 5-Enolpyruvylshikimate-3-Phosphate Synthase from Janibacter sp. Confers High Glyphosate

912 Tolerance to Rice. Sci. Rep. 6, 1–11. https://doi.org/10.1038/srep19104

pe
913 Yoon, J.H., Kang, S.S., Cho, Y.G., Lee, S.T., Kho, Y.H., Kim, C.J., Park, Y.H., 2000. Rhodococcus

914 pyridinivorans sp. nov., a pyridine-degrading bacterium. Int. J. Syst. Evol. Microbiol. 50, 2173–

915 2180. https://doi.org/10.1099/00207713-50-6-2173

ot

916 Zang, H., Wang, H., Miao, L., Cheng, Y., Zhang, Y., Liu, Y., Sun, S., Wang, Y., Li, C., 2019.
tn

917 Carboxylesterase, a de-esterification enzyme, catalyzes the degradation of chlorimuron-ethyl in

918 Rhodococcus erythropolis D310-1. J. Hazard. Mater. 387, 1–10.

919 https://doi.org/10.1016/j.jhazmat.2019.121684
rin

920 Zhao, Y.X., Wang, L., Chen, K.X., Jiang, N.D., Sun, S.L., Ge, F., Dai, Y.J., 2021. Biodegradation of

921 flonicamid by Ensifer adhaerens CGMCC 6315 and enzymatic characterization of the nitrile
ep

922 hydratases involved. Microb. Cell Fact. 20, 1–13. https://doi.org/10.1186/s12934-021-01620-4

923 Zhou, G.C., Wang, Y., Zhai, S., Ge, F., Liu, Z.H., Dai, Y.J., Yuan, S., Hou, J.Y., 2013. Biodegradation
Pr

924 of the neonicotinoid insecticide thiamethoxam by the nitrogen-fixing and plant-growth-

925 promoting rhizobacterium Ensifer adhaerens strain TMX-23. Appl. Microbiol. Biotechnol. 97,

926 4065–4074. https://doi.org/10.1007/s00253-012-4638-3

35

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274
 927 Figure legends

ed
928 Fig. 1. Structure of azoxystrobin. The chemical compound consists of a 4,6-diphenoxypyrimidine

929 skeleton (one of the phenyl rings is cyano-substituted at C-2, the other carries a 2-methoxy-1-

930 (methoxycarbonyl)vinyl substituent at C-2). The last named substituent is the toxophoric group of the

iew
931 chemical compound (rectangle). Some microorgnisms possibly harbour the ability to degrade

932 azoxystrobin by hydrolysis of the carboxyl ester group (arrow A) or cleavage of the bond between the

933 aromatic ring and toxophoric group (arrow B).

v
934 Fig. 2. Azoxystrobin content in the leaves (mg/kg) of the in vitro test (mean values of the triplicates ±

re
935 SEM) after treatment with selected bacterial strains. The negative control are uninoculated lettuce

936 samples without azoxystrobin treatment and the positive control are uninoculated lettuce samples with

937 azoxystrobin treatment. Strains with filled bars showed no or only low degradation of azoxystrobin.
er
938 Strains with empty bars showed degradation rates between 29% and 72% and were selected for

939 greenhouse experiments.

pe
940 Fig. 3. Azoxystrobin concentrations in the leaves (mg/kg) after spraying the fungicide and the different

941 bacterial strains at the leaves (n=4). Dots represent the measured values. Lines within boxes represent
ot

942 median values. Different letters represent significant differences between measured pesticide contents

943 (recursive partitioning analysis; p<0.005).

tn

944 Fig. 4. Azoxystrobin concentrations in the soil (A) (n=8) and in the leaves (B) (n=4) after soil application

945 and translocation to the shoot. Lines within boxes represent median values. Different letters represent
rin

946 significant differences between measured pesticide contents (recursive partitioning analysis; p<0.005).

947
ep

948

949
Pr

36

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4394274