Understanding of Model Plant Arabidopsis thaliana Response to the Coexistence of BPA and TiO2-NPs in low environmental concentration: A Comparative Proteomics Study


 This work presents the most extensive proteomic description of Arabidopsis thaliana in the knowledge of its responses to BPA and TiO2-NPs. Previous studies have reported that nanoparticles (NPs) and Bisphenol A (BPA) are toxic to the environment. However, the jointed toxicity is not yet well understood. This study was aimed to investigate the combined toxicity of BPA and TiO2-NPs to plants. Model plant Arabidopsis thaliana was selected as the target plant. The seedlings were randomly separated into 5 groups and treated with BPA (1000, 100, 10 and 0 µg/kg) and TiO2-NPs (100, 10, 1 and 0 mg/kg). The plant height, biomass and root length indicated no significant toxicity of low concentration of BPA and TiO2-NPs to the growth. In the results of comparative proteomics, both positive and negative effects were observed in root growth, plant development and energy metabolism, et.al, according to GO and KEGG analysis.


Introduction
During the last decades, about 7.2 million dry tons of biosolids were generated in the U.S. per year, 55% of which (about 3.9 million dry tons) were applied to soil for nutrients recycles, soil health enhance or degraded lands restoration. 1 In European countries, the number is about 2.39 million dry tons biosolids generated per year. 2 Biosolids, mostly based on the urban wastewater sewage sludge, were applied as fertilizer providing abundant plant nutrients and organic matter to soils and will improve and stimulate plant growth. 3 It was considered an important biological resource for sustainable agriculture, which is economical and bene cial. The only disadvantage of the application of biosolids in agriculture is their pollutant load, the contaminants of emerging concern (CECs), including heavy metals, organic compounds, pharmaceuticals, and nanoparticles. Questions have been raised about the coexistence of organic toxicants and nanoparticles in biosolids, such as bisphenol A (BPA) 4 and titanium dioxide nanoparticles (TiO 2 -NPs) 5 which were at insigni cant levels, could affect the plant growth and its biosystem.
According to recent researches on crop plants effects of BPA, this endocrine-disrupting chemical, once be absorbed by roots, can act as an estrogen in the plant system. BPA was reported toxicity to mung bean by inhibiting shoot and root development, decreasing chlorophyll content and stomata size, and also, photosynthetic activity appeared to decrease(although not statistically signi cant). 6 In Wang, S., et al.'s study on the mechanism of BPA affects soybean growth. BPA can regulate the levels of single endogenous hormones and the ratios of growth and stress hormones in plant roots and showed a dosedependent manner. 7 Treatment with lower BPA concentration(1.5 mg•L − 1 ) improved soybean seedling growth, while in higher concentrations of BPA treatment(from 3 to 96 mg•L − 1 ), inhibitions of seedling growth were observed. Although nanoparticles naturally exist in the environment, the wide usage of TiO 2 -NPs in manufacturing and commercial products could certainly enhance the TiO 2 -NPs to reach agricultural soils. 8 Especially the precise engineered nano-materials with high reactive and sensitive capacity may exhibit toxicity to the environment. 9 Movafeghi, A., et al., fund TiO 2 -NPs were uptake by the aquatic plant Spirodela polyrrhiza when exposed. Subsequently, all the plant growth parameters were signi cantly decreased, and changes in antioxidant enzyme activities were fund, such as the signi cantly increased of superoxide dismutase. 10 Nanoparticles with high surface reactivity and large surface area, have a high capability to amalgamate with other pollutants and transit into the organism, accompany with interactive effects, such as bioconcentration enhancement, endocrine disruption and developmental neurotoxicity. 11,12 Experimentally analyzes of the effects of BPA and TiO 2 -NPs combine toxicity to animals have been done.
Zebra sh share a high gene similarity (about 87%) with human. Studies have demonstrated that interaction between BPA and TiO 2 -NPs results in zebra sh sex endocrine disruption and reproductive impairment, 11 and gut microbiota dysbiosis. 13 But what will happen when plants exposed to these two contaminants at the same time? How plants respond to these abiotic stresses. A comprehensive understanding of plant responses to the combine toxicity of TiO 2 -NPs and BPA is required. But to the best of our knowledge, plant responses to this kind of stress have not been well studied. Our group was deeply interested in the proteome responding process of plant proteomics. As a target, the model plant, Arabidopsis thaliana was introduced in this study, because of its simple genome but rich genetic resources. 14 We hereby provide new insights into the molecular mechanisms associated with response and tolerance to combine toxicity of BPA and TiO 2 -NPs in the model plant Arabidopsis thaliana roots.

Materials And Methods
Plant culture and exposure Analytical grade (> 98%) of BPA was obtained from Sigma-Aldrich (Missouri, US). Anatase TiO 2 -NPs (15nm average particle size) and agar powder were purchased from Alfa Aesar, Thermo Fisher Scienti c In the plant growth experiment, the effects of BPA and TiO 2 -NPs concentrations were investigated.
Treatments (Table 1) of the two contaminants were set according to the previous report. It was predicted that BPA concentration in sludge application soil was 297 ng/g 4 and TiO 2 -NPs concentrations were reported 136 mg/kg in sewage treatment plant (STP) sludge. 5 . Thus in our experiment, to simulate low contaminants concentrations in real biosolids, BPA concentrations were simpli ed and ranged from 1000 to 10 µg/kg, while TiO 2 -NPs concentrations from 100 to 1 mg/kg. To eliminate other uncertain in uence parameters in biosolids, such as organic matter and soil colloid, all media in the treatment groups and control were supplemented with 1× MS medium and 0.6% (g/g) agar powder. 15 Before each exposure, the stock TiO 2 -NPs suspensions were dispersed in Millipore water (MPW) and homogenized in an ultrasonic bath for 30 min. For germination rate experiment, the Arabidopsis thaliana seeds were rstly vernalized at 4°C for 2 days. Then seeds were evenly sown in petri dishes with culture medium, 100 seeds per group. The germination rates were counted every day for seven days and primary root lengths were determined at the end of week 1 and week 2.
After germination, 30 Arabidopsis thaliana plants per group were transplanted into glass jars and put into an environmental growth chamber (Ohio, US) for ve weeks. Parameters were set with 16 h/8 h day/night photoperiod, 25/22°C day/night temperature, and 65 ~ 70% relative humidity. Plants were regularly watered with 1/4 MS medium solution every two days. For the rst week, each jar was watered 2 mL. The irrigation water volume was then gradually increased to 15 mL per jar until the 5th week due to the need for plant growth.
After ve weeks of growth, Arabidopsis thaliana plants were considered mature. When harvest, plant height was measured and the average height was calculated, plant biomass in fresh weight was recorded and the average calculated too. Plants were then divided into three parts: roots, leaves and stems immediately stored at -20°C for further analyses. For plant growth condition please see SI Fig. 1.
All the research works were done in UTEP, Texas, US. The permission of collecting plant material has been obtained. The plant material collection complies with federal legislation and state law.

Total Protein Extraction
After harvest, the plant roots were carefully cleaned and immediately used to do the total protein extraction by using a total protein extraction kit (G-Biosciences, Missouri, US). Brie y, to lyse to plant roots, every 1g of each sample was added 2 ml TPE Buffer-I and transfer into a grinder to grind until a homogeneous suspension was achieved. This Preparation was carried out on ice to avoid excess heating and protein degradation. The homogenate was transferred to a 15 ml tube and added 240µl TPE Buffer-II, vortex immediately for 30s, put in a boiling hot water bath for another 30s, and vortex again for 30s. Repeat this heating and vortexing process (30s each) until a clear solution was seen. After 10 min of incubation in a boiling water bath, the extracts tubes were centrifuged for 10 min at 16000 g, at − 4°C.
The supernatant was transferred to another tube and stored at -20°C for further analyses. The protein concentrations were quanti ed using a Thermo Pierce bicinchoninic acid (BCA) protein assay kit.
Proteomic Analysis Using 1d Lc-ms/ms Each sample containing 600 µg of protein was reduced with 5 mM dithiothreitol (DTT) for 30 min at 55°C, followed with 10 mM iodoacetamide to alkylate the reduced thiol groups for 30 min in the dark.
Thereafter, it was diluted 8-fold to the nal concentration of 1 M urea/50 mM NH 4 HCO 3 . The proteins were then digested overnight at 37°C with sequencing grade trypsin (Sigma-Aldrich, Missouri, US). The protein digestion was quenched using tri uoro-acetic acid (TFA) to the nal concentration of 0.05% TFA.
The resulting tryptic peptides were then desalted using solid-phase extraction cartridges (Discovery DSC-18 SPE Tube) prior to1D LC-MS/MS analysis.
Each group sample was run in triplicate separated by blanks, using a Dionex Ultimate 3000 RSLC nanosystem online coupled to a Q-Exactive (Thermo Fisher Scienti c, US). Peptides were introduced to the analytical column (Acclaim PepMap RSLC, 75 µm×15 cm, nanoViper, C18, 2 µm, 100 Å) and separated using a 90 min gradient from 5 to 40% solvent B at a ow rate of 300 nL/min (solvent A: 0.1% formic acid 5% acetonitrile, solvent B:0.1% formic acid 80% acetonitrile). The mass spectrometer (MS) was operated in a data-dependent mode. Full scan MS spectra were acquired at a mass resolution of 70000 (mass range 400-1600 m/z and AGC target value of 1E6) in the Orbitrap analyzer. Tandem mass spectra, of the 10 most abundant peaks from the preceding full scan MS spectra, were acquired via fragmentation using higher-energy collisional dissociation (HCD) at a mass resolution 17500 (AGC target value of 2E5, NCE 28%, isolation width 4 m/z). The dynamic exclusion time was set at 15 s. A polysiloxane ion, m/z 445.12003, was used as lock mass throughout.

Spectra Data Processing And Database Search
The resultant MS/MS spectra were searched against Arabidopsis thaliana UniProt sequence databases with 77369 entries (downloaded on 2019.10.09). In the database search, mass tolerances were set to 10 ppm and 0.02 Da for precursor and fragment ions, respectively, and up to two missed cleavages were permitted. Cysteine carbamidomethylation was included as a xed modi cation. Oxidation of methionine and deamidation of asparagine and glutamine were allowed as variable modi cations. Peptide and protein identi cation was performed using PeptideProphet and ProteinProphet in the Trans-Proteomics Pipeline respectively, with the target false discovery rate (FDR) at 1%. 16,17 Protein quanti cation was performed using a label-free approach called spectral counting, 18 which tallies the number of MS/MS spectra assigned to any peptide belonging to each protein. 19 Spectral counting was performed using the Abacus tool, 20 since the algorithm provides an automated spectral counting with robust treatment of ambiguity due to shared peptides in homologous proteins. The resulting spectral count data was analyzed using a proteomics software, Scaffold (version 4.10.0), 21 which performs model-based statistical analysis of differential protein expression. The spectral count data was transformed by log2 transformation and was centered by the median of the three controls in each protein.
The software reports statistical signi cance scores (Fisher's Exact Test, p < 0.05) among treatments and control. To visualize the protein expression changes with varying of different group of treatments, a heat map of the spectral count data was drawn after performing hierarchical clustering with average linkage and Euclidean distance metric.

GO Functional, Pathway Enrichment and Protein Interaction Network Analysis
In order to comprehensively analyze the biological functions, gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) analysis were carried by the database of QuickGO (https://www.ebi.ac.uk/QuickGO/) and the UniProt (https://www.uniprot.org/). The GO Term analysis included the biological process, cellular component and the molecular function.
SRTING version 11.0 (https://string-db.org/).was used to analysis the protein interaction information of all signi cant differentially expressed protein, and their biological functions were also obtained. Then, the interaction network of these proteins was then visualized.

Statistical Analysis
All experimental data obtained were using at least 3 independent biological replicates. Signi cant differences for all statistical tests were evaluated at the level of P ≤ 0.05 with one-way ANOVA test (LSD method). All data analyses were conducted using SPSS Statistics, Version 26.0.

Results And Discussion
Morphological and physiological response analysis Plant morphological and physiological responses had been shown in SI Table 1.
To determine how BPA and TiO 2 -NPs affects Arabidopsis thaliana seeds germinate. We rst inspected the morphological responses every day in the rst week. As shown in Fig. 1 panel A and B, seeds germinated on day 2 and generally came to the highest germination rate on day 7. From which, we can see G5 (with 100 µg/kg BPA and 0 TiO 2 -NPs) had the lowest GR with 68.98%, while the highest germination rate (87.97% signi cant, p < 0.05) were obtained in G1 (with 1000 µg/kg BPA and 100 mg/kg TiO 2 -NPs). This indicated that the seed germination was affected by the coexistence of BPA and TiO 2 -NPs and it possibly helped seeds to germinate under low environmental concentration. We believed that it is the TiO 2 -NPs which really affected the increasing of germination rate. This positive effect of nanoscale TiO 2 on germination was also reported in other plants, such as soya bean. 22 When it came to roots growth, the coexistence of these two contaminates seemed not quite helpful. As shown in Fig. 1 panel C and D, We investigated the root growth on the rst and second week and the response varied. With the medium concentration of BPA (G5, 100 µg/kg BPA and 0 TiO 2 -NPs) or TiO 2 -NPs (G4, 0 BPA and 100 mg/kg TiO 2 -NPs) individually, plant roots grew longer and had more branch roots than the others. As in higher concentrations treatment, shorter primary roots and more branch roots were obtained.
After 5 weeks of cultivation, plant height and biomass were recorded. Signi cant height with a 49.2% higher compared to control was found in G3 (10 µg/kg BPA and 1 mg/kg TiO 2 -NPs) and signi cant plant biomass with a 13.8% weightier comparing to control was found in G4 (with 0 BPA and 10 mg/kg TiO 2 -NPs). Other treatments were not signi cant, which suggested that it is hard to know from plant morphology outcome that whether the plant was poisoned or not by the low concentration BPA and TiO 2 -NPs in the environment. More insight and proteomic investigations were needed to be introduced in.
Comparative Proteomics A total of 3985 proteins were identi ed and quanti ed across all the treatment groups and control against Arabidopsis thaliana UniProt sequence database, and among which, 134 proteins (SI Table 2) were found signi cant differentially expressed, as shown in the heat map in Fig. 2.
Venn diagrams in Fig. 3(a) segregate the signi cant differentially expressed proteins in different treatment groups compared to control identi ed using MS/MS analysis. Of the 134 proteins identi ed, there were 116 proteins in G1 compared to control (G1 v.s. Control), 38 proteins in G2 compared to control (G2 v.s. Control), and 12 proteins in G3 compared to control (G3 v.s. Control). This indicated that with the higher exposure concentration of the mixed contaminants, proteins that were involved in the responded process became more, which made sense that with the more strict abiotic stresses more damages occurred and plant needed to invoke more bioprocesses to deal with, as shown in the interaction networks in Fig. 4.
With the same contaminant concentrations, G2, G4 and G5, compared to control the number of signi cant differentially expressed proteins differed. There were 14 and 12 proteins identi ed signi cant differentially expressed in G4 and G5, responded to BPA and TiO 2 -NPs exposure, relatively. In G4, according to the GO analysis, 7 down-regulated proteins were found that responded to BPA. Such as, Glutathione S-transferase F3(GSTF3_ARATH) and Glutathione S-transferase F2 (GSTF2_ARATH) were 2 defense-related compounds during plant stress, which respond for binding a series of heterocyclic compounds and had a detoxi cation role against certain herbicides; 23,24 Beta-glucosidase 23(BGL23_ARATH) may participate in the control of root colonization by P.indica by repressing defense responses such BPA; 25 Jacalin-related lectin 33 (JAL33_ARATH) a sugar-binding protein, respond for carbohydrate binding; 26 Dehydrin ERD14 (ERD14_ARATH) a molecular chaperone, were also downregulated. Another 7 proteins were found up-regulated (3 found in G4 only). Such as Chaperonin 60 subunit beta 2 (CPNB2_ARATH), a ATP binding protein, assisted protein folding that requires ATP hydrolysis; 27 Receptor for activated C kinase 1C (GPLPC_ARATH), involved in multiple hormone responses and developmental processes which positive regulation of protein phosphorylation; Sucrose synthase 4 (SUS4_ARATH) was a sucrose-cleaving enzyme that provided UDP-glucose and fructose for various metabolic pathways; Cell division control protein 48 homolog E (CD48E_ARATH), which probably functions in cell division and growth processes; Protein TIC110 (TI110_ARATH) involved in protein precursor import into chloroplasts.
In G5, according to the GO annotation, 1 in 7 down-regulated proteins was found that responded to TiO 2 -NPs, Beta-glucosidase 23 (BGL23_ARATH), which was also found signi cant down-regulated in G4. Within the 7 up-regulated proteins, Chaperonin 60 subunit beta 2 (CPNB2_ARATH) also found signi cant up-regulated in G4; cell division control protein 48 homolog E (CD48E_ARATH) (a cell division and growth processes), Chaperone protein dnaJ 2 (DNAJ2_ARATH) (plays a continuous role in plant development), Germin-like protein subfamily 2 member 4 (GL24_ARATH) (participated in the regulation of root development). This indicated that in the response of BPA and TiO 2 -NPs stress, plant primary root length would be shorter and more lateral roots were observed, which consistent with the previous result of plant roots growth.
In G2, besides the proteins that were mention in G4 and G5, there were another 26 proteins (13 upregulated, 13 down-regulated) were found signi cant differentially expressed. The down-regulated proteins were, such as nitrile-speci er protein 1 (JAL28_ARATH, participated nitrile biosynthetic process 28 ), A0A384KEJ8_ARATH(a defense response protein, reactions, triggered in response to the presence of a foreign body or the occurrence of an injury, which result in restriction of damage to the organism attacked or prevention/recovery from the infection caused by the attack 29 ), Myrosinase 4 (BGL34_ARATH, for Hydrolyzes sinigrin), Protein sieve element occlusion B (SEOB_ARATH, for phloem development), Probable plastid-lipid-associated protein 1 (PAP1_ARATH, involved in light/cold stress), which suggest that the plant growth would have been affected under higher exposure of the contaminants and re ected in the plant heights and biomasses. But the negative morphology outcomes were not observed in our study (including G1, G2 and G3), which may indicate that the toxic effects of the co-existence of BPA and TiO 2 -NPs were too low in this concentration. The up-regulated proteins were, such as Tubulin beta-6 chain (TBB6_ARATH, involved in microtubule cytoskeleton organization and microtubule-based process), Calmodulin-1 (CALM1_ARATH, metal-binding), Ras-related protein RABA2a (RAA2A_ARATH, intracellular vesicle tra cking and protein transport), Glyceraldehyde-3-phosphate dehydrogenase GAPCP2 (plays a speci c role in glycolytic energy production), 3-oxoacyl-[acyl-carrierprotein] synthase I, (KASC1_ARATH, catalyzes the condensation reaction of fatty acid synthesis), which suggested that plant stimulated the ability of resource transportation and energy metabolism to overcome the stress and maintain or somehow enhance their growth.

Interaction Networks
Interaction networks of signi cant proteins were shown in Fig. 4. The role of proteins at the core and the relationship between proteins can be easily observed. With the higher concentration exposure of BPA and TiO 2 -NPs, the interaction networks became more complicated (Fig. 4 and SI Table 3, 4). The STRING analysis showed that, in the circumstance of BPA alone in G5, 2 KEGG pathways were found in uenced. The metabolisms of protein processing in the endoplasmic reticulum and GTP hydrolysis were depressed. As a result, plant root growth was depressed. In the case of TiO 2 -NPs alone in G4, 2 affected KEGG pathway were glutathione metabolism and starch and sucrose metabolism, which on the other hand enhance the plant growth at biomass and plant height (although it was not signi cant). As the coexistence of BPA and TiO 2 -NPs and the applied concentration got higher, 17 pathways were got in G1, 5 pathways in G2, and 3 pathways in G3. The involved processes were various. The positive-regulated biological processes were, such as, regulation of protein phosphorylation, regulation of cell growth by extracellular stimulus, amino acid homeostasis, regulation of transcription and DNA-templated, regulation of cell proliferation, pollen tube growth and root hair elongation, et.al. The negative-regulated biological processes were, such as, regulation of cellular response to oxidative stress, ethylene-activated signaling pathway, photoinhibition, regulation of stomatal movement, membrane permeability, pollen tube growth and lateral root formation, et.al. As we can see, plants always got their own way to survive.

Conclusion
To understand the possible effects of applying biosolids in agriculture, it is important to analyze the possible contaminants that exist in biosolids, the behavior and toxicity of BPA and TiO 2 -NPs in the plants, and the response of proteins in roots. Considering these aspects, both positive and negative effects of BPA and TiO 2 -NPs were observed in the model plant, Arabidopsis thaliana. The results suggested that higher concentration of TiO 2 -NPs improved the germination and branch roots growth, but no signi cant help to the growth when the plant has grown to maturity.
The inside view of the comparative proteomics analysis, the impact of biological processes of root growth, plant development and energy metabolism observed. Plants seem to have a pearl of wisdom in their system to maintain a balance. Thus, in our study, the practice of applying biosolids in agriculture, especially contained BPA and TiO 2 -NPs, low concentrations were acceptable. It seemed to do no harm to plant growth. But further examination is needed such as generational observation to have a comprehensive understanding of the long-term impact of biosolids in agriculturing. Figure 1 Plant response to different concentration of combine BPA and TiO2-NPs. A, Dynamic germination rate counted every day in the rst 7 days. B, Seeds germination rate counted at 7th day after sown. C Primary root length measured at day 7 and day 14 after seeds sown. D Image of roots at day 7 and day 14. E Average plant height measured at 5th week. F Total plant biomass of fresh weight determined when harvest at 5th week.