Effect of Acinetobacter sp on Metalaxyl Degradation and Metabolite Profile of Potato Seedlings (Solanum tuberosum L.) Alpha Variety

One of the most serious diseases in potato cultivars is caused by the pathogen Phytophthora infestans, which affects leaves, stems and tubers. Metalaxyl is a fungicide that protects potato plants from Phytophthora infestans. In Mexico, farmers apply metalaxyl 35 times during the cycle of potato production and the last application is typically 15 days before harvest. There are no records related to the presence of metalaxyl in potato tubers in Mexico. In the present study, we evaluated the effect of Acinetobacter sp on metalaxyl degradation in potato seedlings. The effect of bacteria and metalaxyl on the growth of potato seedlings was also evaluated. A metabolite profile analysis was conducted to determine potential molecular biomarkers produced by potato seedlings in the presence of Acinetobacter sp and metalaxyl. Metalaxyl did not affect the growth of potato seedlings. However, Acinetobacter sp strongly affected the growth of inoculated seedlings, as confirmed by plant length and plant fresh weights which were lower in inoculated potato seedlings (40% and 27%, respectively) compared to the controls. Acinetobacter sp also affected root formation. Inoculated potato seedlings showed a decrease in root formation compared to the controls. LC-MS/MS analysis of metalaxyl residues in potato seedlings suggests that Acinetobacter sp did not degrade metalaxyl. GC–TOF–MS platform was used in metabolic profiling studies. Statistical data analysis and metabolic pathway analysis allowed suggesting the alteration of metabolic pathways by both Acinetobacter sp infection and metalaxyl treatment. Several hundred metabolites were detected, 137 metabolites were identified and 15 metabolic markers were suggested based on statistical change significance found with PLS-DA analysis. These results are important for better understanding the interactions of putative endophytic bacteria and pesticides on plants and their possible effects on plant metabolism.


Introduction
Potato is a food with the fourth highest consumption in the world. The annual production is approximately 320 million tons, and it continues to increase when compared to the production of maize, wheat and rice [1]. Potato production is affected by different diseases caused by fungi, bacteria, viruses, nematodes and insects [1,2]. One of the most serious diseases in potato cultivars is caused by the pathogen Phytophthora infestans (late blight). It affects leaves, stems and tubers and disperses rapidly when climatic conditions in the field are favorable (100% humidity and 12-15uC) [3,4]. Chemical control with pesticides such as metalaxyl is an effective method against late blight [5]. Metalaxyl [(R,S)-methyl-N-(2-methoxyacetyl)-N-(2,6-xylyl)-d,l-alaninate] is an important acylanilide fungicide first manufactured by the Ciba-Geigy Corporation in 1977. It is a systemic, apoplastically transported fungicide that is highly active against fungi of the order Peronosporales, by selectively interfering with the synthesis of ribosomal RNA. Metalaxyl is photostable and resistant to heat. Due to its low vapor pressure (3.3 mPa at 25uC), it is very stable in water within a pH range of 1.0 to 8.5 [6]. In soil, its lifetime is from 5 to 35 days and almost all residues are found in the first 10 cm of soil [7]. Amounts of metalaxyl applied in fields vary in different countries. For example, Mexican farmers apply 1.2 kg/ha and farmers in Belgium apply 0.3 kg/ha [8]. Around 15 applications might be necessary during the growing season but in some cases farmers exceed this number of applications. In Mexico, however, farmers typically apply metalaxyl from 33 to 35 times, with the last application 14 days before harvest [9]. This represents a risk to consumers. Although Metalaxyl does not affect reproduction in animals and is not a teratogenic or mutagenic compound, this fungicide provokes cell alterations in mouse liver at 2.5 mg kg 21 day 21 . In dogs at 0.8 mg kg 21 day 21 , it alters alkaline phosphatase levels in blood and causes an increase in liver and brain weight. The LD 50 of metalaxyl in mice is 669 mg kg 21 (oral) and .3100 mg kg 21 (subcutaneous) [10]. Metalaxyl has been detected in several agricultural products (grape, tomato, potato, onion, lettuce, sunflower seeds, spinach, etc.) in countries such as New Zealand, United States, Spain, France, Italy, Germany, Brazil and Belgium [7]. In Mexico, there are no records related to the residues of metalaxyl in vegetables and fruits.
In plants, metalaxyl is taken up by roots, translocated, and extensively metabolized [11,12,13,14]. Metabolism involves oxidation of a ring-methyl group, hydrolysis of the methyl ester and methyl ether bonds, and ultimately conjugation to glucose. Ringmethyl hydroxylation was found predominantly in cell suspension cultures of lettuce and grapevine [14]. Other prominent products arise from O-dealkylation, ester hydrolysis and ester hydrolysis of O-dealkylated product. The principal metalaxyl metabolite found is the acid metabolite (N-(2,6-dimethylphenyl)-N-(methoxyacetyl) alanine) [15].
There are several studies related to the use of endophytic bacteria that degrade toxic organic compounds such as pesticides in the environment [16,17,18]. Plant-associated endophytic bacteria and rhizospheric bacteria have been shown to biodegrade toxic organic compounds in contaminated soil and can promote phytoremediation [17]. Almost all 300,000 plant species identified contain at least one species of endophytic bacteria [19,20,21,22]. Germaine et al. [18] reported that when pea plants (Pisum sativum) were inoculated with Pseudomonas endophytes that were isolated from hybrid poplars P. trichocarpa X P. deltoids cv. Hoogvorst capable of degrading 2,4-D, the pea plant showed no accumulation of 2,4-D and showed little or no signs of phytotoxicity when compared to inoculated controls. McGuinness et al. [16] inoculated pea plants with a bacterium expressing a specific bacterial glutathione-Stransferase (GST) isolated from Burkholderia xenovorans LB400, BphKLB400, wild type and mutant (Ala 180Pro). This bacterium was capable of dehalogenating toxic chlorinated organic pesticides such as chloromequat chloride. Bailey and Coffey [23] showed that eight strains of fungi and six strains of bacteria degraded metalaxyl in liquid medium. The objective of the present study was to evaluate the ability of a bacterial strain isolated from the rhizosphere of potato plants cultivated in commercial fields to degrade metalaxyl in potato seedlings alpha variety cultured in vitro, and identify potential metabolic biomarkers and disturbed metabolic pathways emerging in response to impact of Acinetobacter sp and metalaxyl on potato seedling.

Materials and Methods
All solvents were HPLC grade and were obtained from either Fisher Scientific (USA) or JT Baker (USA). Metalaxyl (99.5% purity), N-Methyl-N (trimethylsily) trifluoroacetamide, Murashige and Skoog medium, potato dextrose agar and tissue culture agar were from Sigma-Aldrich-Fluka (SAF, Deisenhofen, Germany

Isolation of rizhospheric bacteria
Samples for bacterial isolation were collected from potato fields of Zuckerman Family Farms, Inc. in Stockton, CA (USA). All necessary permits were obtained from Mr. Kenneth N. Jochimsen, Vice President of Zuckerman Family Farms, Inc. Fields were selected based on the background of treatments with metalaxyl (3 to 10 applications in 2007 and 2008). Potato plants were extracted with part of the rizhospheric soil. Under aseptic conditions using a sterile sponge, part of the rizhospheric soil was removed using the stamping technique [24] and impressions were made with the sponge over the surface of Petri dishes containing a minimal media with metalaxyl (50 mg kg 21 of media). Minimal media (pH 7.0) used in the study contained 2 g (NH 4 [25]. Once the bacterial strains were purified, only one strain was selected based on its growth on minimal media that was fortified with metalaxyl.

Degradation of metalaxyl
All biodegradation experiments were carried out aerobically. Liquid minimal media (55 mL) containing 50 mg kg 21 of metalaxyl was inoculated with the selected bacteria (7.8610 6 6 1.643 CFU mL 21 ) and incubated at 30uC in a bath (Versa Bath, Fisher Scientific, USA) with horizontal agitation at 150 rpm for a period of 75 days. To determine if the bacteria degrades metalaxyl, two controls were included. One control consisted of liquid minimal media with metalaxyl and no bacteria added, while the other control included liquid minimal media with bacteria and no metalaxyl. Treatment and controls had five replicates. Aliquots (1 mL) were taken every 2 weeks from treatments and controls. Ethanol (1 mL) was added to inactivate samples. Samples were kept at 220uC until analysis for metalaxyl residues by liquid chromatography-tandem mass spectrometry (LC-MS/MS) and total protein content.

Sample preparation and metalaxyl/protein quantification
The samples were centrifuged (Eppendorf, model 5415 D, USA) at 16.1 RCF for 15 min and the supernatant was filtered through a syringe filter (0.45 mm, 4 mm diameter, Millipore, USA). Quantification of analytes was made by LC-MS/MS (Sciex API 2000 triple-quadrupole MS system, Perkin-Elmer, Shelton, CT). The pellet was used for measurement of total protein content. The resulting pellet was lysed according to the protocol proposed by Massoud et al. [26]. Protein content was determined following the Microtiter Plate Protocols from BioRad [27].

Characterization of bacterial strain by gen ribosomal 16S DNAr
The phenotypic characterization of the bacteria consisted of a Gram stain [28] and for the genotypic characterization we used a Wizard Genomic DNA purification kit to obtain bacterial DNA. The ribosomal gen 16S DNAr was amplified with the initiators 27f and 1492r [29] using the enzyme GoTaq DNA polymerase Promega. The polymerase chain reaction (PCR) was conducted with a thermocycler Mastercycler gradient, consisting of an initial cycle of denaturation (94uC, 5 min), 35 cycles of amplification (94uC, 0.3 min; 53uC, 0.3 min: 72uC, 1.3 min) and finally one cycle of elongation (72uC, 7 min). The fragment of the ribosomal gen 16S amplified (ca. 1.5 kb) was digested with 5 U of the restriction enzymes AluI, HaeIII y HhaI. The restriction profile was determined and compared to different bacterial strains with electrophoresis using 3% agarose gels. The amplified fragment of the 16S DNAr gen was cloned in the vector PCR2.1, using a commercial TA cloning kit. The sequence of the 16S DNAr gen was obtained using an Applied Biosystems sequencer (Model 3730), using the universal initiators M13 present in the vector. The sequence was compared with the NCBI data base (National Center for Biotechnology Information from USA). A phylogenetic analysis was performed, where the tree topology was inferred with the UPGMA method (Unweighted Pair Group Method with Arithmetic Mean) using the MEGA version 4 program.

Inoculation of potato seedlings
Potato seedlings (21 day-old) were inoculated with a bacterial suspension (4610 9 67.549 UFC mL 21 ) following the protocol of Zuno et al. [24], Sahay and Varma [30], Barka et al. [31], Martinez et al. [32], Dini et al. [33], Anderote et al. [34] and Ma et al. [35]. The seedling containers were sealed and placed in photoperiod as previously described growth chamber. The containers were monitored daily to verify the bacterial-root association, through the formation of a biofilm surrounding the roots.

Quantitation of endophytic bacteria in potato seedlings
The protocol of Rosenblueth and Martinez-Romero [20], Reinhold-hurek and Hurek [21] Zuno et al. [24], Dini et al. [33], Anderote et al. [34] and Ma et al. [35] was followed to determine the number of endophytic bacteria in potato seedlings (n = 5). A serial dilution plating technique was followed (until decimal dilution 1610 5 ) and a pouring plate technique was used. Potato dextrose agar medium was added (25 mL) to each petri dish and the dishes were incubated at 30uC for 24 h, after which a colony count was conducted.  Under aseptic conditions, the inoculated potato seedlings were removed from the containers and the Murashige & Skoog media attached to the roots was rinsed off with sterile distilled water. Excess water was removed and the individual seedlings were deposited in sterile glass test tubes containing 5 mL of liquid Murashige and Skoog media containing metalaxyl (50 mg kg 21 of media). Bacteria-free potato seedlings were used as controls and were treated in the same way as inoculated seedlings. The potato seedlings were exposed to metalaxyl for 30 days. To check for potential photodegradation of metalaxyl during the photoperiod,  tubes containing 5 mL of Murashige and Skoog media with metalaxyl (50 mg kg 21 of media) were included as controls. After 30 days, the potato seedlings were transferred to tubes containing 10 mL of Murashige & Skoog media (metalaxyl free) and incubated for another 60 days. During this time, growth parameters were recorded (fresh weight, length increase and vigor). For treatments and controls, 10 replicates (consisting of one potato seedling) were included. Treatment groups consisted of seedlings that were inoculated with bacteria (PB) and inoculated with bacteria plus metalaxyl (PMB). The control group consisted of seedlings without any treatment (P) and a treatment only with metalaxyl (PM).

Instrumental Conditions
Metalaxyl and its acid metabolite were analyzed using an AB Sciex API 2000 triple-quadrupole-tandem mass spectrometer (AB Sciex Instruments, Foster City, CA, USA) with Atmospheric Pressure Chemical Ionization (APCI). The instrument was controlled using ''Analyst'' software (version 1.4.2; Applied Biosystems, Foster City, CA). The high performance liquid chromatography (HPLC) equipment consisted of a Hewlett-Packard Model 1100 series with a quaternary pump system, auto sampler and in-line degasser. The analytes were separated using a C 18 VARIAN Pursuit column (10063.0 mm, i.d., 3 mm particle size, Santa Clara, CA) fitted to a C 18 pre-column (463.0 mm i.d., Phenomenex, Torrance, CA). The injection volume was 10 mL and the HPLC was operated at a flow rate of 600 mL min 21 . A gradient program was used, consisting of 60% water/40% acetonitrile/0.1% acetic acid for 0.50 min, followed by 10% water/90% acetonitrile/0.1% acetic acid for 4.0 min, and 60% water/40% acetonitrile/0.1% acetic acid for 4.0 min. The total run time per sample was 8.0 min. The retention times were 2.56 min for metalaxyl and 1.6 min for the acid metabolite. The system configuration was as follows: temperature 425uC, dwell time 300 ms and nebulizing current 3 kV. The curtain gas, gas 1, gas 2 were set at 35, 40 and 15, respectively.
The declustering potential (DP), focusing potential (FP), entrance potential (EP), collision energy (CE), collision cell exit potential (CXP), and collision cell entrance potential (CEP) for each compound are listed in Table 1. Quantitation of metalaxyl and its acid metabolite was done using external standards by comparison to a ten and eight point calibration curve, respectively. The area of the analyte response was plotted against the amount of analyte injected. A linear regression was used to generate calibration curves over a range of concentrations (0.20-100 mg kg 21 , metalaxyl, and 0.2-20 mg kg 21 , acid metabolite. These ranges spanned the concentration range found in potato seedlings and in all cases r 2 .0.999. MRM parameters were developed by infusing 100 mg mL 21 solutions of each compound into the MS with mobile phase at flow rate of 25 mg mL 21 . The most abundant transition ion with the lowest accompanying background was selected for each analyte (280.1R220.1 for metalaxyl and 266.1R206.1 for the acid metabolite). Metalaxyl and its acid metabolite had LODs of 0.2 mg kg 21 and 0.04 mg kg 21 , respectively. Metalaxyl and its acid metabolite had LOQs of 0.4 mg kg 21 and 0.3 mg kg 21 , respectively. These results were generated with intra-assay (same day) coefficients of variation less than 5%. In the presence of sample matrix, the metalaxyl acid metabolite showed a gain of 50% in signal due to ion enhancement. To overcome this matrix effect, accurate quantitation of this analyte was achieved by the preparation of a standard curve in the matrix.

Metabolite Profiling by GC-TOF-MS
The extraction and analysis of potato seedlings for metabolomic profiling were carried out at the Metabolomics Core, Genome Center of University of California, Davis (Davis, CA, USA). Seedlings were harvested after 30 days of exposure to metalaxyl. Each group consisted of 10 seedlings. Sample preparation and extraction followed the protocol reported by Fiehn et al. [36]. Briefly, seedlings (,30 mg) for each treatment were harvested and transferred to a 2 mL Eppendorf tube, immediately frozen in liquid nitrogen and crushed in a Restch,mill (Hann, Germany). A 1 mL aliquot of pre-chilled extraction solvent mixture (acetonitrile:isopropanol:water 3:3:2) at 220uC was added, and the pH was adjusted between 5 and 6. The samples were vortexed for about 10 s at room temperature and then shaken for 4-6 min at 4uC using an orbital mixing chilling/heating plate (Thermomixer R., Cole-Parmer, USA). The tubes were centrifuged for 2 min at 14,000 g. The supernatant was decanted into a new Eppendorf tube and the extracts were evaporated to complete dryness overnight at room temperature (26uC) using a Labconco Centrivap cold trap concentrator (Labconco, Corporation, Kansas City, MO). An aliquot of 500 mL of acetonitrile:water (50% v/v) was added and the extracts were vortexed. The extracts were centrifuged at 14,000 g, 4uC for 3 min. The eluent was transferred to a new Eppendorf tube and evaporated to dryness overnight at room temperature. (26uC) using a Labconco Centrivap cold trap concentrator. To the dried samples 20 mL of 40 mg/mL methoxylamine hydrochloride in pyridine was added, and tubes with the samples were agitated at 30uC for 30 min. Subsequently, 180 mL of trimethylsilylating agent N-methyl-N-trimethylsilyltrifluoroacetamide (MSTFA) was added, and samples were agitated at 37uC for 30 min. GC-TOF-MS analysis was performed using an Agilent 6890 N gas chromatograph (Palo Alto, CA, USA) interfaced to a time-of-flight (TOF) Pegasus III mass spectrometer (Leco, St. Joseph, MI). Automated injections were performed with a programmable robotic Gerstel MPS2 multipurpose sampler (Mülheim an der Ruhr, Germany). The GC was fitted with both an Agilent injector and a Gerstel temperature-programmed injector, cooled injection system (model CIS 4), with a Peltier cooling source. An automated liner exchange (ALEX) designed by Gerstel was used to eliminate cross-contamination from sample matrix occurring between sample runs. Multiple baffled liners for the GC inlet were deactivated with 1-mL injections of MSTFA. The Agilent injector temperature was held constant at 250uC while the Gerstel injector was programmed (initial temperature 50uC, hold 0.1 min, and increased at a rate of 10uC/s to a final temperature of 330uC, hold time 10 min). Injections of 1 mL were made in split (1:5) mode (purge time 120 s, purge flow 40 ml/ min). Chromatography was performed on an Rtx-5Sil MS column (30 m60.25 mm i.d., 0.25 mm film thickness) with an Integra-Guard column (Restek, Bellefonte, PA). Helium carrier gas was used at a constant flow of 1 mL/min.
The GC oven temperature program was 50uC initial temperature with 1 min hold time and ramping at 20uC/min to a final temperature of 330uC with 5 min hold time. Both the transfer line and source temperatures were 250uC. The Pegasus III TOF (Leco, St. Joseph, MI) mass spectrometer ion source operated at 270 kV filament voltage with ion source. After a solvent delay of 350 s, mass spectra were acquired at 20 scans per second with a mass range of 50 to 500 m/z. Resulting GC-TOF-MS data were processed following the methods outlined by Fiehn et al. [36]. In brief, initial GC-TOF-MS peak detection and mass spectrum deconvolution were performed with ChromaTOF software version 2.25 (Leco). A reference chromatogram was defined that had a maximum of detected peaks over a signal/noise threshold of 20 and used for automated peak identification based on mass spectral comparison to standard and in-house customized mass spectral libraries. Mass spectra were searched against custom spectrum libraries (e.g. the Fiehn library of 713 unique metabolites) and identified based on retention index and spectrum similarity match. All known artifactual peaks caused by column bleeding or phthalates and polysiloxanes derived from N-methyl-N-trifluoroacetamide (MSTFA) hydrolysis were manually identified and removed. Resulting data for each sample were normalized using the total summed metabolite concentration and then logarithmi- cally transformed (base = 10). For each metabolite, transformed values greater than six standard deviations from mean across sample groups were set to missing data.

Statistical Analysis
Processing of the raw data yielded 137 identified metabolites from potato seedling samples. Statistical analysis was applied to GCMS data from P, PB, PM and PMB sample groups. Statistical analysis was performed by the submission of previously normalized data to web based service for metabolomic data analysis: MetaboAnalyst (http:// www.metaboanalyst.ca/MetaboAnalyst/faces/Home.jsp). Additional statistical analysis was performed with the use of MarkerView TM software versión 1.2 (AB Sciex, Foster City, CA, USA) and SAS Software version 9.0 (SAS Institute Inc., Cary, NC, USA). The primary objective of the statistical analysis was to identity metabolites whose concentrations differentiate significantly between control (untreated) plants and infected/treated plants. To identify metabolites as metabolic markers, we wanted to identify metabolites which were capable of sample groups' discrimination, and further to identify sets of relevant metabolites that act synergistically within functionally defined pathways. Specifically, to identify metabolites whose expression was associated with the plant treatment, we performed differential analysis based on a general linear model. Analysis of variance (ANOVA) was used to assess groups' effects for each metabolite. Unsupervised principal component analysis (PCA) was applied that best explain the variance in a data set without referring to group labels. Supervised discriminant analysis PLS-DA (MetaboAnalyst) and PCA-DA (MarkerView TM 1.2) which use multivariate regression techniques to extract via linear combination of original variables the information that can predict the group membership was further applied to visualize samples clustering and identify the most important features.

Pathway Analysis
Pathway analysis was performed utilizing MetPA: A web-based metabolomics tool for pathway analysis & visualization (http:// metpa.metabolomics.ca/MetPA). MetPA (Metabolomic Pathway Analysis) is a web-based tool that combines result from pathway enrichment analysis with the pathway topology analysis which allowed identifying the most relevant pathways involved in the conditions under currently reported study. Data for identified metabolites detected in all samples was submitted into MetPA with annotation based on common chemical names. Verification of accepted metabolites was conducted manually using HMDB, KEGG, and PubChem DBs. Arabidopsis thaliana (thale cress) pathway library was used for pathway analysis. List of the most impacted pathways was generated in accordance with previously described approaches [37].

Results and Discussion
Samples of five varieties of potato plants (ZM#13-4, CW2912, Alegria, Chieftan and Granola) were collected from potato fields with a history use of metalaxyl application. Five isolates were obtained and their ability to grow in the presence of metalaxyl was assessed by culturing the strains on a minimal media containing metalaxyl. Controls were grown in media without metalaxyl. Each strain and corresponding control included three replicates. After incubating for 24 h at 37uC, strains only grew on media that contained metalaxyl (Figure 1). The bacterial strain isolated from the Chieftan potato plant variety was evaluated for metalaxyl degradation in minimal liquid media. This strain showed a 5-fold increase in total protein content compared to growth in the absence of metalaxyl (Figure 2). The increase in protein content for the strain isolated from ZM#13-4, CW2912, Alegria, and Granola potato varieties was 3 times less than strains isolated from the Chieftan variety (data not shown). The gram stain showed that the bacteria isolated from the Chieftan potato plant variety was a Gram negative coco-bacillus. The genotypic characterization indicated that the bacterial strain had a 98% resemblance to the strain Acinetobacter sp (FJ753401.1). A phylogenetic analysis showed that the closest strain to the one obtained from the potato fields also had a 98% similarity to Acinetobacter sp 40 (GQ289378)  Figure 3). Although an increase in total protein content was observed for the strain isolated from the Chieftan variety, concentration of metalaxyl in the media did not decrease over time (data not shown). It's important to note that the bacterial strain was exposed to three different metalaxyl concentrations in minimal media (155, 50 and 3 mg L 21 , Figure 2B). It was observed that as the metalaxyl concentration increased; microbial growth was favored, reflecting the increase in total protein content, from 0.08260.003 to 0.49860.041 mg mL 21 at the metalaxyl concentration of 155 mg L 21 . This behavior was observed by Pooja et al. [38], Liu et al. [39], Xie et al. [40], Hongsawat and Vangnai [41] and Chanika et al. [42], when they performed similar experiments with species of Acinetobacter to degrade different pesticides. There are studies that report the use of Acinetobacter to degrade pesticides such as atrazine [38], methyl parathion [39], malathion [40], and chloroanilines [41]. However, no studies mention the use of Acinetobacter sp in metalaxyl degradation, although this is a common soil microorganism [38]. Bailey and Coffey [23] studied different bacterial strains for metalaxyl degradation and showed that eight strains of fungi and six strains of bacteria degraded metalaxyl in liquid medium.

Inoculation of potato seedlings with Acinetobacter sp
Acinetobacter sp formed a bacterial biofilm around the potato seedling roots 3 days after inoculation (Figure 4). Several studies have determined the period of time that different bacteria with endophytic potential may need in order to associate the plant roots [43], Pseudomona fluorescens SBW25 associates with wheat roots in a period of 6 to 9 days. Garcia et al. [44], determined that Azospirillum spp and Azotobacter beijerincki colonized wheat Pavon variety after 3 days. Prieto and Mercado-Blanco [45], found that Pseudomonas fluorescens (PICFZ) colonized olive roots in 9 days. Zuno et al. [24] found that Pseudomonas sp takes 7 days to associate with potato Seedling Alpha variety and Rainey [46] reported that root association of different Pseudomonas strains took 14 days after inoculation. According to these studies, the time that a microorganism needs to associate with plant roots depends on many factors; one of the most important is the affinity that a microorganism has for the exudates liberated by the plant [47,48]. The population density of Acinetobacter sp in the interior of potato seedlings was observed after the root-bacteria association (three days) and 3.9360.865 log CFU g 21 of vegetable tissue was observed. After 30 days of incubation of potato seedlings in Murashige and Skoog with and without metalaxyl, the population density of cells increased to an average concentration of 7.1060.2 log CFU g 21 of vegetable tissue and 6.1360.2 log CFU g 21 of vegetable tissue, respectively ( Figure 5D).
Statistical analysis indicated that the density of cells of Acinetobacter sp in the treated potato seedlings with metalaxyl was 14% higher and significantly different than the control group (p#0.05). This value remained constant for several weeks until the conclusion of the experiment. The population densities of Acinetobacter sp in potato seedlings were similar to those reported by other investigators. Sturz and Nowak [49], found a population density of 3 to 8 log CFU g 21 of vegetable tissue. When working with strains of Pseudomonas, Germaine et al. [50] reported a bacterial density in poplar plants of 4 to 5 log CFU g 21 of vegetable tissue. Adachi et al. [51] found a population of 3 to 5 log CFU g 21 of vegetable tissue in sweet potato. Garbeva et al. [52] reported a concentration of endophytic bacteria in potato seedlings of 4 to 6 log CFU g 21 of vegetable tissue. Elvira-Recuenco and van Vuurde [53] found values of 4 to 8 log CFU g 21 of vegetable tissue in pear cultivars and Zuno et al. [24] reported a population of 5.28 log CFU g 21 for Pseudomonas sp in potato seedlings Alpha variety.
The results obtained in the present study indicate that Acinetobacter sp was capable of establishing as a putative endophyte in potato seedlings. However further confirmation that Acinetobacter sp is a true endophyte of potato seedlings is needed using techniques as immunological detection of bacteria, fluorescence tags, confocal laser scanning microscopy or specific oligonucleotide probes [54,55,56].

Response of potato seedlings to metalaxyl and Acinetobacter sp
Our results suggest that metalaxyl did not affect the growth of potato seedlings during 30 days of exposure ( Figure 5A, B and C). The potato seedling length increase, fresh weight gain and vigor were similar in both treated seedlings and controls. Although metalaxyl did not affect the growth of potato seedlings, it did result in a 14% increase in the growth of Acinetobacter sp inside the potato seedlings compared to controls ( Figure 5D). Acinetobacter sp, in turn, strongly affected potato seedling growth, as confirmed by a 40% reduction in length and fresh weight (27%). Acinetobacter sp also affected root formation ( Figure 6). Potato seedlings inoculated with bacteria showed a decrease in root formation compared to the controls. This effect on growth could be due to change in phytohormone production as reported earlier, however further elaborate studies are necessary to corroborate these findings. [57,58,59,60,61] Small cay formed on the roots of the potato seedlings inoculated with Acinetobacter sp, giving the appearance of nodules. Potato seedlings that were not inoculated were not affected in such a manner ( Figure 6). One of the effects observed in the inoculated seedlings was an early tuberization; inoculated potato seedling produced at least 161 tuber at 75 days compare to controls (no tuber). Significant statistical differences were observed between inoculated seedlings and controls when data was analyzed by the Kruskal Wallis test (p,0.0001). The production of tubers in inoculated seedlings was initiated one month before the normal cycle of potato production (three months). It is necessary to conduct studies to determine the effect of Acinetobacter sp on some metabolic pathways in potato seedlings which may have influence on tuber production [62,63,64,65]. The results obtained in the present work are similar to those reported by Frommel et al. [66], Nowak et al. [67] and Sturz [68].
LC-MS/MS was used to analyze metalaxyl and the metalaxyl acid metabolite. The SPE percentage recovery at 50 mg kg 21 was 9569% (n = 5) for metalaxyl and 9269% (n = 5, 50 mg kg 21 ) for acid metabolite. The results obtained by LC-MS/MS indicated that Acinetobacter sp did not contribute to metalaxyl degradation in potato seedlings ( Figure 7A). According to these results the degradation behavior of metalaxyl in potato seedlings follows a quadratic model with an r 2 of 0.93 ( Figure 7B). As shown in Figure 7A, the maximum concentration of metalaxyl absorbed by the seedlings was 2264 mg kg 21 after 15 days of exposure. This was reduced by 97% after 90 days (0.6660.29 mg kg 21 of metalaxyl). Mehta et al. [69], found that 9 mg kg 21 in mustard plants completely dissipated in 60 days. Zadra et al. [15], measured 2.7 mg kg 21 of metalaxyl in sunflower plants 38 days exposure of metalaxyl. This measurement decreased by 67% to 0.9 mg kg 21 after 85 days. Badaway et al. [70] reported a concentration of 19 mg kg 21 of metalaxyl which was reduced by 82% to 0.49 mg kg 21 after 7 days. Overall metalaxyl degrades anywhere from 67% to 100% over a period of 7 to 60 days.
The highest concentration of acid metabolite found in vegetable tissue was 1567 mg kg 21 after 45 days of exposure to metalaxyl. This concentration was reduced by 81% to 2.861.7 mg kg 21 of acid metabolite after 90 days ( Figure 7A). In studies performed with 14 C-metalaxyl did not detect the presence of acid metabolite in potato foliage, but there were found 2.8 mg kg 21 of acid metabolite in tubers [71]. Stingelin (cited by Hamilton [71]) reported a concentration of 0.25 mg kg 21 of acid metabolite in lettuce plants 14 days after exposure decreasing 8% in 21 days to 0.02 mg kg 21 . Zadra et al. [15] observed the formation of acid metabolite in sunflower plants 21 days after treatment with metalaxyl, reaching a concentration of 1.3 mg kg 21 after 85 days of exposure. Acid metabolite shows a clear tendency to degrade over time inside the potato seedlings, which may suggest that it's only one of several metabolites reported to be involved in various metalaxyl degradation routes in plants [6,7].

Statistical Analysis of the Metabolite Profiling Data
Metabolite profiling is a powerful tool that has contributed to the understanding of plant physiology, including phenotypic differences, gene annotations, metabolite regulation, and characterization of stress responses. Metabolic profiling of potato tubers has been accomplished using NMR, HPLC-UV and GC-MS techniques [72,73,74]. In the present research one study statistical analysis was used to investigate modifications to metabolic pathways. However, metabolic profiling of potato seedlings, as well as effect of Acinetobacter sp and metalaxyl on metabolic profiles of potato seedlings was never carried out. Data normalization results, performed with MetaboAnalyst assistance prior statistical analysis, are shown on Figure 8. PCA (not shown), PCA-DA and PLS-DA were applied to normalized GC-MS data from P, PB, PM and PMB sample groups in order to identify possible variations in the metabolite composition between the infected and non-infected samples. Out of the 18 principal components, the first three (PC1, PC2 and PC3) were responsible for major variation (67.6%) in study groups. Score plots clearly demonstrated that all four groups of samples, P, PB, PM and PMB cluster into four very distinct groups (Figure 9). Further PCA-Discriminant Analysis resulted in three components, which contributed equally to 100% of variation. The list of 15 the most important features is shown on Figure 10 alone with the indication of relative metabolite concentrations in each group. When groups compared side by side, there are metabolites that having larger score in the particular pare ( Figure 11). For instance proline and asparagine taking place of glucose in the line of important features illustrating changes characteristic for each group of samples (see Figure 11).

Metabolic Pathway Analysis
Pathway analysis performed with MetPA assistance revealed a number impacted metabolic pathways. The overall pathway impact picture is illustrated with Figure 12. This data presented in greater details in Table 2 where number of hits, p values and KEGG links are depicted. The most impacted appears to be phenylalanine metabolism; glycine, serine and threonine metabolism; galactose metabolism; alanine, aspartate and glutamate metabolism, etc. However impact consequences are not straightforward and metabolic flux is involved. Changes in flux can be illustrated with Figure 13 and Figure 14 where some metabolites do not follow the pattern of key metabolite. This indicates flux redistribution upon influence of bacterial infection and influence of Metalaxyl and its metabolites. This confirms a high complexity and sensitivity of metabolic networks exploited by plant in order to survive and adjust to environmental challenges.
In conclusion, the effect of Acinetobacter sp on metalaxyl degradation in potato seedlings as well the effect of bacteria and metalaxyl on growth of potato seedlings was investigated using GC-TOF-MS and LC-MS/MS techniques. Results from our study suggest that metalaxyl alone did not affect the growth of potato seedlings. However, Acinetobacter sp strongly affected the growth and root formation of inoculated seedlings. LC-MS/MS analysis of metalaxyl residues in potato seedlings suggests that Acinetobacter sp did not degrade metalaxyl in potato seedlings. Based on a review of the literature, we report for the first time that metabolic profiling study followed by statistical and metabolic pathway analyses to demonstrate that the metabolic profile of treated and control plants are very distinct and suggest the significant alteration of metabolic pathways by both Acinetobacter sp infection and metalaxyl treatment.