This is the accepted version of the following article: Zheng, B. et al. Straw biochar increases the abundance of inorganic phosphate solubilizing bacterial community for better rape (Brassica napus) growth and phosphate upteke in Science of the total environment (Ed. Elsevier), vol. 647 (Jan. 2019), p. 1113-1120. Which has been published in final form at DOI 10.1016/j.scitotenv.2018.07.454 © 2018. This manuscript version is made available under the CC- BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc- nd/4.0/ 1 Straw biochar increases the abundance of inorganic phosphate solubilizing 2 bacterial community for better rape (Brassica napus) growth and phosphate 3 uptake 4 Bang-Xiao Zheng1,2,3,4, Kai Ding1*, Xiao-Ru Yang1, Mohammed A.M. Wadaan5, Wael 5 N. Hozzein5, Josep Peñuelas3,4, Yong-Guan Zhu1,6 1 6 Key Laboratory of Urban Environment and Health, Institute of Urban Environment, 7 Chinese Academy of Sciences, Xiamen 361021, PR China; 2 8 University of Chinese Academy of Sciences, Beijing 100049, PR China; 3 9 Consejo Superior de Investigaciones Científicas (CSIC), Global Ecology Unit, 10 Centre for Ecological Research and Forestry Applications (CREAF)–CSIC– 11 Universitat Autonoma de Barcelona (UAB), Bellaterra, 08193 Barcelona, Catalonia, 12 Spain; 4 13 CREAF, Cerdanyola del Vallès, 08193 Barcelona, Catalonia, Spain; 5 14 Bioproducts Research Chair, Zoology Department, College of Science, King Saud 15 University, Riyadh 11451, Kingdom of Saudi Arabia; 6 16 State Key Laboratory of Urban and Regional Ecology, Research Center for Eco- 17 Environmental Sciences, Chinese Academy of Sciences, Beijing 100085, China. 18 Emails: 19 Bang-Xiao Zheng: bxzheng@iue.ac.cn; b.zheng@creaf.uab.cat 20 Kai Ding: kding@iue.ac.cn 21 Xiao-Ru Yang: xryang@iue.ac.cn 22 Mohammed A.M. Wadaan: wadaan@ksu.edu.sa 23 Wael N. Hozzein: whozzein@ksu.edu.sa 24 Josep Peñuelas: Josep.Penuelas@uab.cat 25 Yong-Guan Zhu: ygzhu@iue.ac.cn 26 27 * Corresponding Author 28 Key Laboratory of Urban Environment and Health, Institute of Urban Environment, 29 Chinese Academy of Sciences, Xiamen 361021, People’s Republic of China. Phone: 30 (+86) 592 6190560; Fax: (+86) 592 6190977. 1 31 Abstract 32 The direct application of inorganic-phosphate-solubilizing bacteria (iPSBs) for 33 improving the efficiency of phosphorus (P) use leads to a low rate of bacterial survival. 34 Biochar is a good inoculum carrier for microbial survival, and diverse feedstocks can 35 have different effects. We generated an iPSB community using seven selected iPSB 36 strains with various phylogenic taxonomies and P-solubilizing abilities. Biochar was 37 then inoculated with the iPSB community and applied to soil in pots seeded with rape 38 (Brassica napus). Growth of the rape for four weeks and the effects of biochars 39 produced from six raw feedstocks, rice straw, rice husks, soybean straw, peanut shells, 40 corn cobs and wood, were compared. The synthetic iPSB community had a larger 41 capacity to solubilize inorganic P and exude organic anions than any of the individual 42 strains. The structure of the iPSB community was analyzed by high-throughput 43 sequencing four weeks after inoculation. All seven iPSB strains were detected, 44 dominated by Arthrobacter defluvii 06-OD12. The abundance of the iPSB community 45 was significantly correlated with rape biomass, P content and P uptake (P < 0.05). The 46 biochar amendments conferred 6.86-24.24% survival of the iPSB community, with the 47 straw biochars conferring the highest survival. The available-P content of the biochar 48 rather than soil pH was the dominant factor for iPSB community structure, suggesting 49 that the biochar material was critical for the survival and functioning of the iPSB 50 community. Our study demonstrates the feasibility of biochar-assisted iPSB 51 improvement of crop growth and P uptake. 2 52 Importance 53 Inorganic phosphate solubilizing bacteria (iPSBs) are commonly proved to be useful as 54 biofertilizer for plant growth and phosphorus (P) uptake. Previous studies dedicated to 55 screen iPSB isolates with high solubilizing capacity. However, the exact limitation 56 would attribute to the low survival rate after direct inoculation, dramatically shortening 57 functioning periods of iPSB strains. A synthetic iPSB community with 7 different 58 generic strains were first-time applied with plant inoculation. By observing the survival 59 condition, including abundance and community structure, the results proved the straw 60 biochar, which compared with other 5 feedstock biochars, was optimal for iPSB 61 community survival. 62 Keywords: straw biochar; Brassica napus; phosphorus; inorganic phosphate 63 solubilizing bacteria; phosphate uptake 64 3 65 Introduction 66 Phosphorus (P) is an essential macronutrient for plant growth. P, however, is a 67 non-renewable element, so the availability of rock P has been dramatically decreased 68 to satisfy the increased demands for P fertilizers (Elser and Bennett, 2011; Elser et al., 69 2007; Penuelas et al., 2013). Applied P fertilizers are readily immobilized due to the 70 strong fixation in soil solutions, which renders P unavailable for plant uptake and 71 hinders the movement of P in terrestrial systems (Kochian, 2012). Developing new 72 strategies to increase P-use efficiency are greatly needed. 73 Microorganisms are active participants in numerous biogeochemical cycles (van 74 der Heijden et al., 2008). Inorganic-phosphate-solubilizing bacteria (iPSB) are major 75 drivers of the biomobilization of soil P, which releases many organic ions or protons 76 to effectively liberate immobilized P and preserve P for being fixed again (Richardson 77 et al., 2009; Richardson and Simpson, 2011). iPSB microbes are also commonly 78 accepted as plant-growth-promoting rhizobacteria (PGPR) and are commonly used as 79 biological inoculants to improve P assimilation and plant productivity (Collavino et 80 al., 2010; Kumar et al., 2012). The low survival rate of the bacteria after being 81 introduced into the soil, however, prevents the sustainability of PGPR inoculation 82 (Bashan et al., 1995; Malusá et al., 2012). 83 Biochar obtained by the pyrolysis of various raw feedstocks is commonly used as 84 a stable and economic inoculum carrier (Sun et al., 2016). Generally, biochar showed 85 beneficial effects on the abundance, diversity and functional improvement of 86 inoculated microbial community (Lehmann et al., 2011). Biochar with internal 4 87 porosity provides a protective habitat to prolong PGPR survival, and its strong ability 88 to absorb nutrients can sustainably support the growth of inoculants (Zimmerman et 89 al., 2011), which has been extensively demonstrated (Ahmad et al., 2015; Akhtar et 90 al., 2015). The biochar addition was proved to be of great benefits for regulating 91 phosphate-solubilizing bacteria community (Wei et al., 2018; Wei et al., 2016). 92 Biochars produced from different materials, however, differ in their ability to maintain 93 PGPR due to their different physicochemical properties (Cantrell et al., 2012). A 94 previous study demonstrated the strong influence of biochars with different 95 characteristics on PGPR survival after incorporation into soil (Khan et al., 2014). The 96 selection of biochars is therefore of great importance for PGPR inoculations. 97 We selected biochars pyrolyzed from six raw feedstocks for maintaining the 98 effect of iPSBs on P mobilization to improve plant growth: rice straw (RS), rice husks 99 (RH), soybean straw (SS), peanut shells (PN), corn cobs (CC) and wood (WD). The 100 growth of rape (Brassica napus) needs large amounts of P (Belimov et al., 2002) and 101 was thus chosen as the test plant. Rape growth and the P status of the plants and the 102 rhizospheric soil were determined. An iPSB community containing seven microbial 103 strains with varying characteristics was used as a microbial inoculant, and the 104 structures of the communities after application of the six inoculated biochars were 105 analyzed. The relationships between rape growth, biochar addition and iPSB 106 community structure are also discussed. 107 5 108 Materials and Methods 109 Soil sampling and characterization 110 Soil samples (0-15 cm layer) were collected in May 2015 after crop harvest from 111 the Changshu AgroEcological Experimental Station in Jiangsu Province, China 112 (31°32'N, 120°41'E). The soil contained 0.78% clay, 18.38% silt and 80.84% sand. 113 The samples were air-dried, sieved (0.2 mm) and stored at 4 °C and -20 °C until use 114 and analysis. 115 Soil pH was determined in a suspension (1:2.5 dry soil:sterilized water (w/v)) 116 using a XL60 pH meter (Fisher Scientific, Asheville, USA). Soil moisture content (%) 117 was calculated as the difference in weight between fresh soil and soil dried in an oven 118 at 300 °C for 24 h. Carbon, nitrogen and sulfur contents were measured with an 119 elemental analyzer (vario MAX CNS, Elementar, Hanau, Germany). P and available-P 120 contents were determined after digestion by strong acid (sulfuric acid) and extraction 121 by sodium bicarbonate using the molybdate-blue method (Murphy and Riley, 1962; 122 Olsen et al., 1954; Parkinson and Allen, 1975), respectively. The acid-digested soils 123 were further used to determine the concentrations of metals (potassium, calcium, 124 sodium, magnesium, iron and aluminum) by inductively coupled plasma optical 125 emission spectrometry (ICP-OES, Optima 7000DV, Perkin Elmer, Waltham, USA). 126 The physical and chemical properties of the Changshu soil are listed in Table S1. 127 Experimental design and analysis 6 128 The diagram of experimental design was presented in Fig.1. In brief, the soil was 129 used for new iPSB strain isolation as well as rape growth with its sterilized version. 130 A common iPSB community is composed of strains from genus Arthrobacter, 131 Streptomyces, Bacillus and Beta- and Gamma-proteobacteria (Zheng et al., 2017). 132 Based on that, we compared the isolated iPSB strains with a previous isolated iPSB 133 database to select representative iPSB strains based on their morphology, taxonomic 134 diversity and biochemical characterization. Then, 7 representative iPSB strains, 135 including Arthrobacter defluvii 06-OD12 (KU647200), Burkholderia cepacia 51- 136 Y1415 (KU647244), Bacillus megaterium CS22 (MG430229), Pseudomonas 137 frederiksbergensis 11-D3 (KU647205), Rhodanobacter sp. 25-Y8 (KU647218), 138 Streptomyces prasinopilosus 34-Y1 (KU647227) and Variovorax paradoxus 19-D4 139 (KU647212), were chosen to generate iPSB community, which was validated by 140 biocompatibility and P solubilization capacity tests. Besides, six raw feedstocks, 141 including RS, RH, SS, CC, PN and WD, were used to form biochars. The pH, carbon 142 and nitrogen percentages, available P concentration, Brunauer-Emmett-Teller (BET) 143 surface area and pore size of biochar were determined. To see the effect of biochar on 144 iPSB community-assisted crop growth, six iPSB community-adhered biochars (3%) 145 were inoculated into Changshu soil with rape (Brassica napus) growth while a soil 146 with iPSB community inoculation but without biochar amendment was treated as 147 control (CK). Before and after 4-week rape growth, the pH, available-P concentration 148 and total P amount of rhizospheric soil, biomass, total P content and P uptake of plant 149 and iPSB community survival rate were measured. The iPSB community structure 7 150 was also analyzed by 16S rRNA high-throughput sequencing with iPSB database 151 alignment. Besides, to discriminately quantify the impact of biochar and iPSB 152 community on the plant growth and its P uptake, we conducted another 21-day rape 153 growth with four treatments: rape growth without biochar or iPSB (R-CK), only 3% 154 RS biochar addition (R-B), with only iPSB community inoculation (R-P) and with 155 iPSB community-adhered RS addition (R-BP). The P-related parameters were 156 measured before and after rape growth. The detailed explication could be found in 157 supplementary materials and methods. 158 Statistical analysis 159 One-way correlation and variance were determined using SPSS Statistics 21 160 (IBM, New York, USA). Heatmapping was generated by the PHEATMAP package 161 (http://rpackages.ianhowson.com/cran/pheatmap/) based on R studio (version 3.2.3). 162 Redundancy analysis (RDA), principle component analysis (PCA), variance partition 163 analysis (VPA) and Monte Carlo permutation test were also performed by using R 164 studio (version 3.2.3) with and VEGAN (Oksanen et al., 2007) packages. 8 165 Results 166 Physical and chemical properties of the biochars 167 All physical and chemical parameters of the six biochars were highly variable 168 between treatments (Table S2). RS had the highest pH (10.30 ± 0.34), and RH had the 169 lowest pH (8.92 ± 0.40) but the highest BET surface area (43.00 ± 3.47 m2 g-1). Pore 170 size differed significantly among the biochars (P < 0.05). C content was significantly 171 higher for CC (74.80 ± 0.87%) and WD (75.93 ± 5.86%) than the other biochars, and 172 N and Olsen-P contents were highest for RS (2.22 ± 0.21% and 685.80 ± 52.14 mg 173 kg-1, respectively). 174 Selection and characterization of the iPSB strains 175 We selected seven iPSB strains based mainly on their P-solubilizing ability and 176 phylogenic diversity to artificially simulate an iPSB community. We selected from 177 133 iPSB strains, 76 from a previous study and 57 from our study. B. megaterium 178 CS22 (Firmicutes) had the highest hydroxyapatite content (139.35 μg mL-1, Table S3) 179 and ability to solubilize calcium-phosphate (Ca-P) (119.37 ± 2.80 μg mL-1, Table S5) 180 and was thus selected for the community. Six other strains from different classes with 181 good Ca-P-solubilizing ability were also selected: Rhodanobacter sp. 25-Y8 182 (Gammaproteobacteria, 97.34 ± 2.00 μg mL-1), P. frederiksbergensis 11-D3 183 (Gammaproteobacteria, 51.65 ± 16.41 μg mL-1), V. paradoxus 19-D4 184 (Betaproteobacteria, 26.62 ± 4.00 μg mL-1), S. prasinopilosus 34-Y1 (Actinobacteria, 9 185 17.93 ± 8.98 μg mL-1), B. cepacia 51-Y1415 (Betaproteobacteria, 15.14 ± 1.03 μg 186 mL-1) and A. defluvii 06-OD12 (Actinobacteria, 13.15 ± 0.17 μg mL-1). 187 We comprehensively characterized these seven strains before generating the 188 iPSB community. They had various cellular sizes and shapes (0.5-10 μm and spherical 189 or rod-shaped, Fig. 2a). All these strains could survive in the LB, R2A and PVK (with 190 NH4+ or NO3- as the sole nitrogen source) media but with different enzymatic 191 activities and abilities to hydrolyze carbohydrates and form acids (Table S4). For 192 example, B. cepacia 51-Y1415 hydrolyzed 10 of 11 carbohydrates, and B. 193 megaterium CS22 hydrolyzed only seven but hydrolyzed potassium gluconate, which 194 is different from the other strains. B. megaterium CS22 could also form organic acids 195 from 23 of 49 organic compounds, but Rhodanobacter sp. 25-Y8 did not form 196 detectable amounts of any acid. The seven strains formed colonies with different 197 morphologies in LB medium (Fig. 2b). The experiments with bromocresol-purple 198 staining indicated that anion exudation differed substantially between strains that were 199 cultivated in PVK-NH4+ and PVK-NO3 media (Fig. 2c, d). The pHs of the PVK media 200 and ability to secrete organic anions after 72 h of cultivation also varied among the 201 strains (Fig. 2e and Table S6). B. megaterium CS22 produced the lowest medium pH 202 (4.73) without secreting lactic, acetic or malic anions but secreted the most succinic 203 anions (196.60 ± 9.98 ppm). A. defluvii 06-OD12 produced the highest medium pH 204 (8.04) and secreted abundant organic anions and the most malic anions (107.33 ± 5.71 205 ppm). 10 206 Assessment of iPSB-community generation 207 The biocompatibility experiment by colony steaking indicated that these seven 208 strains survived well together (data not shown). The abilities of the community 209 generated by mixing the seven strains at the same cell density (109 CFU mL-1) to 210 solubilize Ca-P (146.77 μg mL-1) and ferric-phosphate (Fe-P) (29.16 μg mL-1) were 211 significantly higher than those of the individual strains (P < 0.05, Table S5). The 212 features of community growth indicated an initial large release of inorganic P (78.28 ± 213 6.89 μg mL-1 at 12 h) and then a continuous increase to 146.77 ± 15.81 μg mL-1 after 214 144 h of cultivation. All six organic anions were detected from the cultivation of the 215 iPSB community (Table S6). Malic and lactic anions were the two most abundant 216 organic anions (80.48 ± 10.28 and 64.03 ± 5.94 μg mL-1 after 144 h, respectively), 217 which increased with the release of inorganic P. 218 The bacterial, soil and plant responses after four weeks of rape growth 219 Parameters of the microbes, rhizospheric soil and plants associated with the four 220 weeks of rape growth are presented in Table 1. The bacterial population was largest 221 for RS (2.88 ± 0.45 × 108 CFU g-1 dry soil), followed by RH and SS (1.89 ± 0.18 and 222 1.52 ± 0.39 × 108 CFU g-1 dry soil, respectively). iPSB survival after four weeks of 223 rape growth was significantly higher for RH (30.34 ± 1.49%) than CK, followed by 224 RS, SS and PN at 24.24 ± 2.15, 19.58 ± 1.29 and 18.79 ± 2.40%, respectively (P < 225 0.05). The pH of the rhizospheric soil was about 6.5-6.9 for all treatments. The total P 226 content was slightly higher for RH, CC and PN than CK. The amount of Olsen P, 11 227 however, was significantly higher with biochar addition and was highest for RS 228 (34.93 ± 0.72 mg kg-1, P < 0.05). RS and SS had significantly higher plant biomass 229 (6.10 ± 0.57 and 6.14 ± 0.70 g pot-1, respectively), P content (0.66 ± 0.03 and 0.68 ± 230 0.05 g kg-1, respectively) and P uptake (4.05 ± 0.55 and 4.19 ± 0.36 mg pot-1, 231 respectively) than the other treatments (P < 0.05). 232 Response of the iPSB community after four weeks of rape growth 233 A total of 61 816 reads of the 16S rRNA gene were obtained after filtering for 234 quality and removing chimeric reads. Of these, 15 640 sequences with 235 similarities >99% with the iPSB database were accepted as potential iPSB species. 236 The percentage of iPSBs in each biochar treatment ranged from 6.44 ± 1.55% (CK) to 237 62.57 ± 17.27% (RS), with an average of 33.81 ± 18.58%. The abundance of the iPSB 238 community was significantly higher for RS and SS (50.73 ± 12.02%) than the other 239 biochar-addition treatments (Fig. 3a, P < 0.05), and the Shannon diversity index of the 240 community was significantly higher for CK (2.38 ± 0.01) and CC (2.36 ± 0.03) than 241 the other treatments (Fig. 3b, P < 0.05). 242 The pattern of the structures of the iPSB communities is shown in Fig. 4a. Thirty 243 species were aligned, and the seven selected iPSB strains were successfully detected 244 in all treatments. A. defluvii 06-OD12 was the dominant species with percentages of 245 the total iPSB populations of 1.04 ± 0.30, 4.61 ± 1.08, 3.99 ± 0.98, 3.13 ± 0.50, 38.28 246 ± 11.20, 9.65 ± 2.54 and 28.61 ± 7.77% for CK, WD, PN, CC, RS, RH and SS, 247 respectively. The average abundances of B. megaterium CS22, B. cepacia 51-Y1415, 12 248 P. frederiksbergensis 11-D3, Rhodanobacter sp. 25-Y8, S. prasinopilosus 34-Y1 and 249 V. paradoxus 19-D4 were 4.19 ± 2.23, 2.03 ± 1.83, 0.19 ± 0.25, 1.68 ± 0.65, 4.65 ± 250 3.14 and 2.09 ± 2.14%, respectively. The PCA of different iPSB strains across 251 different biochar treatments showed that six strains from stimulated iPSB community 252 were well separated from other strains (Figure S1). 253 The RDA indicated that the first two axes explained 90.72% of the effect of the 254 soil and biochar characteristics on the iPSB community, with RDA1 and RDA2 255 explaining 85.07 and 5.65%, respectively (Fig. 4b). Biochar available P (AP) and soil 256 AP were the most important factors affecting the iPSB communities, supported by 257 their significant correlations (P < 0.0001) in the Monte Carlo permutation tests (Table 258 S7). Other significantly correlated factors were pore size, biochar N content, soil total 259 P content, biochar pH and BET surface area (P < 0.0001). 260 The contribution of each factor and their interactions to the structure of the iPSB 261 communities was quantified, and the percentages of the variance due to one or two 262 variables resolved by the VPA are listed in Table S8. Biochar and soil AP contributed 263 the most, at 0.352 and 0.219%, respectively. The interaction between soil AP and 264 biochar AP explained most of the variance (9.367%) for the iPSB community, 265 followed by the interactions between pore size and pH (3.273%) and between biochar 266 C and N content (3.045%). The analyses of correlations between iPSB community 267 abundance with diversity and the plant parameters are presented in Fig. 5. The 268 abundance of the iPSB community was significantly correlated negatively with its 13 269 diversity (P < 0.001) but positively with plant biomass, plant P and plant P uptake (P 270 < 0.001). 271 To discriminate the impact of biochar or iPSB community for the rape growth 272 and its P uptake, we conducted another experiment for 21 days. Results (Table S9) 273 showed that solo biochar addition (R-B) or iPSB inoculation (R-P) significantly 274 increased rhizosphere soil and plant P content and increased plant biomass compared 275 with R-CK treatment, while R-P was measured with significantly higher P amounts 276 than R-B in rhizosphere soil (P < 0.05). Combined amendment with biochar and iPSB 277 (R-BP) performed significantly higher soil P concentration and plant parameters (P < 278 0.05). 279 14 280 Discussion 281 Plant biomass and the ability to take up P are thus important indices for assessing 282 the effect of an iPSB community on plant growth. Plant biomass, P content and P 283 uptake in our study were all significantly correlated (P < 0.001) with the abundance of 284 the iPSB community (Fig. 5b-d), suggesting that the inoculation of the biochar with 285 the iPSB community positively contributed to rape growth. 286 Inoculation of microorganism with biochar should perform with improvement 287 effect since the physical and chemical characters of biochar will provide appropriate 288 habitats for microbial community, including its abundance, diversity, nutrient 289 acquirement, mobilization and cycling (Lehmann et al., 2011; Quilliam et al., 2013). 290 All biochars tested in this study were beneficial as inoculum carriers for the survival 291 of the iPSB community, but their efficacy varied, perhaps due to the physical and 292 chemical properties of the individual biochars. RS was the best feedstock as an 293 inoculum carrier; the maintenance of the bacterial population was significantly 294 highest for RS (2.88 ± 0.45 × 108 CFU g-1 dry soil, P < 0.05; Table 1), perhaps due to 295 its surface area and porosity. BET surface area (35.29 ± 3.02 m2 g-1) and pore size 296 (27.86 ± 1.22 μm) were significantly higher (P < 0.05) for RS than the other biochars 297 except RH (Table S2). The larger pore size would provide a habitat niche for microbes 298 and have positive effects on the capacity to retain nutrients by binding both cations 299 and anions to the surface (Atkinson et al., 2010; Liang et al., 2006). A systematic 15 300 study of 32 biochars found that the survival of Rhizobium tropici had a quadratic 301 dependence on biochar pore size (Vanek et al., 2016). 302 The porosity of RS also allowed the retention of significantly (P < 0.05) more N 303 (2.22 ± 0.21%) and AP (685.80 ± 52.14 mg kg-1) than the other biochars, which 304 provided a good habitat for the survival of the iPSB community. The correlation 305 analysis (Table S7) found that the structure of the iPSB community depended mostly 306 on biochar AP content (R2 = 0.881), followed by biochar N content (R2 = 0.652), pore 307 size (R2 = 0.656) and BET surface area (R2 = 0.352), suggesting that the survival of 308 the community may dependent on the retention of nutrients by the biochar. Biochar is 309 a good material for the retention of key macronutrients such N and P due to its surface 310 area and availability of anionic and cationic charges (Lehmann, 2007). The nutrient 311 status of biochars, however, depends on the origin of the biochar feedstock and its 312 mode of production. RS was the best feedstock in our study for nutrient storage and 313 bacterial survival. A previous study found that RS biochar could hold more organic C 314 and organic matter for the survival of Bradyrhizobium and sulfur-reducing bacteria 315 (Khan et al., 2014). Another study also reported that the benefits of RS biochar to 316 microbes contributed to the significantly higher survival than for other biochars of 317 Bacillus mucilaginosus, which had higher carbohydrate contents after four weeks of 318 incubation (Sun et al., 2015). 319 We used seven strains to generate the iPSB community. All strains were detected 320 by high-throughput sequencing after four weeks of cultivation (Fig. 4a). A. defluvii 16 321 06-OD12 was the most abundant iPSB strain, followed by S. prasinopilosus 34-Y1 322 and B. megaterium CS22. This finding was consistent with previous studies reporting 323 that Arthrobacter, Streptomyces and Bacillus were commonly isolated iPSB genera in 324 rhizospheric soils that promoted plant growth (Aislabie et al., 2006; Gopalakrishnan et 325 al., 2011; Taha et al., 1969; Xuan et al., 2011). Pseudomonas has been extensively 326 used as an iPSB genus due to a specific gene for solubilizing inorganic phosphate 327 (Kwak et al., 2015; Miller et al., 2010). Pseudomonas in our study, however, was less 328 competitive, perhaps because the biochars provided alkaline environments (pH >8, 329 Table S2), and Pseudomonas prefers acidic environments (Garbeva et al., 2004). 330 The RS and SS biochars that held abundant AP were the best for population 331 growth of the iPSB community (Fig. 4a). The size of the community was significantly 332 correlated with biochar properties and soil AP (P < 0.05, Fig. 4b and Table S7). The 333 interaction between soil and biochar AP contributed the most to the structure of the 334 community (Table S8). Soil pH has a strong impact on the abundance of bacterial 335 communities (Shen et al., 2013), and iPSB strains are sensitive to soil pH, with 336 abundance increasing with pH (Zheng et al., 2017). The interaction between soil pH 337 and pore size or N content of biochar would also make greater contribution for iPSB 338 community structuring, which may suggest that the combined effect of soil pH and 339 biochar would provide a better environment for the survival of iPSB community, even 340 for its better functioning. Our results, however, indicated that AP had a larger effect, 341 suggesting that the biochar provided a protective environment for the survival and 342 growth of the iPSB community. The combined effect of biochar and soil AP was 17 343 proved to be given the best contribution for the formation and functionality of iPSB 344 community, suggesting that available P was the critical dependent factor for iPSB 345 community. Exogenously amended inoculants may not survive better due to strong 346 competition with native microbes for long periods, so the storage of abundant 347 nutrients by biochars would be critical for increasing community size (Lehmann et al., 348 2011). Biochars with nearly neutral internal pHs would benefit microbial survival, 349 suggested by the significant correlation between iPSB community abundance and 350 biochar pH rather than soil pH (P < 0.05). The biodiversity of the iPSB community 351 indicated by the Shannon index was significantly lower (P < 0.05) for RS and SS than 352 the other treatments (Fig. 4b), and the biodiversity of all treatments was significantly 353 negatively correlated (P < 0.05) with iPSB community abundance (Fig. 5a), perhaps 354 because the strains of the inoculated iPSB community survived well and became 355 dominant with the assistance of the RS and SS biochars. 356 The combined usage of RS biochar and iPSB community inoculation was most 357 beneficial for rape growth. In terms of rhizospheric soil (Table S9), RS biochar 358 significantly increased the P contents while the effect of iPSB community was much 359 higher (P < 0.05), suggesting that the iPSB community inoculation had the effect of P 360 mobilization and was better functioning on soil P releasing than biochar. As for the 361 impact on plant growth, we could not discriminate significant differences from the 362 effect of solo biochar or iPSB community on the promotion of plant biomass and its P 363 content and uptake; however, the combined application of biochar and iPSB 364 performed significantly higher promotion (P < 0.05), which suggested that the RS 18 365 biochar may provide a better habitat for P releasing by iPSB community to promote 366 rape growth. The biochar was extensively reported with beneficial effects on soil 367 nutrient mobilization and plant nutrient uptake (Atkinson et al., 2010), which was also 368 found in our study (nearly 2 times higher in soil Olsen-P concentration and 2 times of 369 plant P uptake). If we considered the enhancing effect of biochar for P releasing is 370 permanent, the remaining increasing effect observed in R-BP treatment (almost 1.5 371 times higher in available P enhancement and 1.5 times of plant P uptake than R-B 372 treatment) should be attributed to the functioning of iPSB community; however, the 373 actual effect of solo iPSB community inoculation was far less. One reasonable 374 explanation to this phenomenon may be that RS biochar provide a well environment 375 for iPSB survival and strengthen its ability for P mobilization. Although we have 376 found RS biochar was the best for the survival of iPSB community (Table 1), we 377 could not conclude that the biochar was the only reason for iPSB survival since there 378 is not a longer period (more than 3 month) to observe its living status. A similar study 379 of 20 weeks was reported that the biochar addition showed significant impact on the 380 population increase of Bacillus mucilaginosus within 4 weeks; however, the survival 381 rate was gradually declined afterwards (Sun et al., 2015). Another study also reported 382 the short-term effect of biochar amendment for the better survival of soil bacteria 383 (Hale et al., 2013). Both of those studies indicate that the biochar was effective but 384 may hardly give a long-term support for bacterial survival. A possibility is that there 385 would be not enough nutrients (such as available N and P) retained in biochar after a 19 386 long time, and the starving bacteria may enter into a non-culturable state to keep them 387 survive without functioning or even cell division (Sun et al., 2016). 388 The iPSB community benefited crop growth, and the biochar provided suitable 389 conditions for iPSB community survival due to its strong absorption of AP. All 390 biochar feedstocks in this study benefited the iPSB community, but the straw biochars 391 (RS and SS) were best for increasing iPSB community abundance and its functioning 392 in P mobilization. 393 20 394 Acknowledgements 395 The authors wish to acknowledge Dr. Jian-Qiang Su (Institute of Urban Environment, 396 Chinese Academy of Sciences) and Dr. Yu Yan (College of Chemical Engineering, 397 Huaqiao University) for their helpful suggestions and assistance with figure 398 production. This work was financially supported by the Strategic Priority Research 399 Program of the Chinese Academy of Sciences (XDB15020402), the Natural Science 400 Foundation of China (41571130063, 41430858, 41701299), the National Key 401 Research and Development Program of China (2017YFD0200201). JP acknowledged 402 European Research Council Synergy grant ERC-SyG-2013-610028 IMBALANCE-P. 403 MAMW and WNH are grateful to the Deanship of Scientific Research, King Saud 404 University for funding through Vice Deanship of Scientific Research Chairs. 405 21 406 References 407 Ahmad MT, Asghar HN, Saleem M, Khan MY, Zahir ZA. Synergistic effect of 408 rhizobia and biochar on growth and physiology of maize. Agron J 2015; 107: 409 2327-2334. 410 Aislabie JM, Chhour K-L, Saul DJ, Miyauchi S, Ayton J, Paetzold RF, et al. 411 Dominant bacteria in soils of Marble Point and Wright Valley, Victoria Land, 412 Antarctica. Soil Biol Biochem 2006; 38: 3041-3056. 413 Akhtar SS, Andersen MN, Naveed M, Zahir ZA, Liu F. Interactive effect of biochar 414 and plant growth-promoting bacterial endophytes on ameliorating salinity 415 stress in maize. Funct Plant Biol 2015; 42: 770-781. 416 Atkinson CJ, Fitzgerald JD, Hipps NA. Potential mechanisms for achieving 417 agricultural benefits from biochar application to temperate soils: a review. 418 Plant Soil 2010; 337: 1-18. 419 Bashan Y, Puente ME, Rodriguez-Mendoza MN, Toledo G, Holguin G, Ferrera- 420 Cerrato R, et al. Survival of Azospirillum brasilense in the Bulk Soil and 421 Rhizosphere of 23 Soil Types. Appl Environ Microbiol 1995; 61: 1938-1945. 422 Belimov AA, Safronova VI, Mimura T. Response of spring rape (Brassica napus var. 423 oleifera L.) to inoculation with plant growth promoting rhizobacteria 424 containing 1-aminocyclopropane-1-carboxylate deaminase depends on 425 nutrient status of the plant. Can J Microbiol 2002; 48: 189-199. 22 426 Cantrell KB, Hunt PG, Uchimiya M, Novak JM, Ro KS. Impact of pyrolysis 427 temperature and manure source on physicochemical characteristics of biochar. 428 Bioresour Technol 2012; 107: 419-428. 429 Collavino MM, Sansberro PA, Mroginski LA, Aguilar OM. Comparison of in vitro 430 solubilization activity of diverse phosphate-solubilizing bacteria native to acid 431 soil and their ability to promote Phaseolus vulgaris growth. Biol Fert Soils 432 2010; 46: 727-738. 433 Elser J, Bennett E. Phosphorus cycle: a broken biogeochemical cycle. Nature 2011; 434 478: 29-31. 435 Elser JJ, Bracken ME, Cleland EE, Gruner DS, Harpole WS, Hillebrand H, et al. 436 Global analysis of nitrogen and phosphorus limitation of primary producers in 437 freshwater, marine and terrestrial ecosystems. Ecol Lett 2007; 10: 1135-1142. 438 Garbeva P, van Veen JA, van Elsas JD. Assessment of the diversity, and antagonism 439 towards Rhizoctonia solani AG3, of Pseudomonas species in soil from 440 different agricultural regimes. FEMS Microbiol Ecol 2004; 47: 51-64. 441 Gopalakrishnan S, Humayun P, Kiran BK, Kannan IGK, Vidya MS, Deepthi K, et al. 442 Evaluation of bacteria isolated from rice rhizosphere for biological control of 443 charcoal rot of sorghum caused by Macrophomina phaseolina (Tassi) Goid. 444 World J Microbiol Biotechnol 2011; 27: 1313-1321. 445 Hale SE, Jensen J, Jakob L, Oleszczuk P, Hartnik T, Henriksen T, et al. Short-term 446 effect of the soil amendments activated carbon, biochar, and ferric 23 447 oxyhydroxide on bacteria and invertebrates. Environ Science Technol 2013; 448 47: 8674-8683. 449 Khan TF, Ahmed MM, Huq SMI. Effects of biochar on the abundance of three 450 agriculturally important soil bacteria. J Agr Chem Environ 2014; 3: 31. 451 Kochian LV. Rooting for more phosphorus. Nature 2012; 488: 466-467. 452 Kumar P, Dubey R, Maheshwari D. Bacillus strains isolated from rhizosphere showed 453 plant growth promoting and antagonistic activity against phytopathogens. 454 Microbiol Res 2012; 167: 493-499. 455 Kwak Y, Jung BK, Shin JH. Complete genome sequence of Pseudomonas 456 rhizosphaerae IH5T (=DSM 16299T), a phosphate-solubilizing 457 rhizobacterium for bacterial biofertilizer. J Biotechnol 2015; 193: 137-8. 458 Lehmann J. Bio‐energy in the black. Frontiers Ecol Environ 2007; 5: 381-387. 459 Lehmann J, Rillig MC, Thies J, Masiello CA, Hockaday WC, Crowley D. Biochar 460 effects on soil biota–a review. Soil Biol Biochem 2011; 43: 1812-1836. 461 Liang B, Lehmann J, Solomon D, Kinyangi J, Grossman J, O'neill B, et al. Black 462 carbon increases cation exchange capacity in soils. Soil Sci Soc Am J 2006; 463 70: 1719-1730. 464 Malusá E, Sas-Paszt L, Ciesielska J. Technologies for beneficial microorganisms 465 inocula used as biofertilizers. Sci World J 2012; 2012. 466 Miller SH, Browne P, Prigent‐Combaret C, Combes‐Meynet E, Morrissey JP, O'Gara 467 F. Biochemical and genomic comparison of inorganic phosphate solubilization 468 in Pseudomonas species. Environ Microbiol Rep 2010; 2: 403-411. 24 469 Murphy J, Riley JP. A modified single solution method for the determination of 470 phosphate in natural waters. Anal Chim Acta 1962; 27: 31-36. 471 Oksanen J, Kindt R, Legendre P, O’Hara B, Stevens MHH, Oksanen MJ, et al. The 472 vegan package. Community Ecology Package 2007; 10. 473 Olsen SR, Cole CV, Wantanabe FS, Dean LA. Estimation of available phosphorus in 474 soils by extraction with sodium bicarbonate: Washington, USDA, 1954. 475 Parkinson JA, Allen SE. A wet oxidation procedure suitable for the determination of 476 nitrogen and mineral nutrients in biological material. Commun Soil Sci Plan 477 1975; 6: 1-11. 478 Penuelas J, Poulter B, Sardans J, Ciais P, van der Velde M, Bopp L, et al. Human- 479 induced nitrogen–phosphorus imbalances alter natural and managed 480 ecosystems across the globe. Nat Commun 2013; 4. 481 Quilliam RS, Glanville HC, Wade SC, Jones DL. Life in the ‘charosphere’–Does 482 biochar in agricultural soil provide a significant habitat for microorganisms? 483 Soil Biol Biochem 2013; 65: 287-293. 484 Richardson AE, Barea J-M, McNeill AM, Prigent-Combaret C. Acquisition of 485 phosphorus and nitrogen in the rhizosphere and plant growth promotion by 486 microorganisms. Plant Soil 2009; 321: 305-339. 487 Richardson AE, Simpson RJ. Soil microorganisms mediating phosphorus availability 488 update on microbial phosphorus. Plant Physiol 2011; 156: 989-996. 25 489 Shen C, Xiong J, Zhang H, Feng Y, Lin X, Li X, et al. Soil pH drives the spatial 490 distribution of bacterial communities along elevation on Changbai Mountain. 491 Soil Biol Biochem 57: 204-211. Soil Biol Biochem 2013; 57: 204-211. 492 Sun D, Hale L, Crowley D. Nutrient supplementation of pinewood biochar for use as 493 a bacterial inoculum carrier. Biol Fert Soils 2016; 52: 515-522. 494 Sun D, Meng J, Liang H, Yang E, Huang Y, Chen W, et al. Effect of volatile organic 495 compounds absorbed to fresh biochar on survival of Bacillus mucilaginosus 496 and structure of soil microbial communities. J Soils Sediments 2015; 15: 271- 497 281. 498 Taha SM, Mahmoud SAZ, El-Damaty AH, El-Hafez AMA. Activity of phosphate- 499 dissolving bacteria in Egyptian soils. Plant Soil 1969; 31: 149-160. 500 van der Heijden MG, Bardgett RD, van Straalen NM. The unseen majority: soil 501 microbes as drivers of plant diversity and productivity in terrestrial 502 ecosystems. Ecol Lett 2008; 11: 296-310. 503 Vanek S, Thies J, Wang B, Hanley K, Lehmann J. Pore-Size and Water Activity 504 Effects on Survival of Rhizobium tropici in Biochar Inoculant Carriers. J 505 Microb Biochem Technol 2016; 8: 296-306. 506 Wei Y, Zhao Y, Shi M, Cao Z, Lu Q, Yang T, et al. Effect of organic acids production 507 and bacterial community on the possible mechanism of phosphorus 508 solubilization during composting with enriched phosphate-solubilizing 509 bacteria inoculation. Bioresour Technol 2018; 247: 190-199. 26 510 Wei Y, Zhao Y, Wang H, Lu Q, Cao Z, Cui H, et al. An optimized regulating method 511 for composting phosphorus fractions transformation based on biochar addition 512 and phosphate-solubilizing bacteria inoculation. Bioresour Technol 2016; 221: 513 139-146. 514 Xuan Y, Xu L, Tian HZ, Liu GH, Cui M. Isolation and characterization of phosphate- 515 solubilizing bacteria from walnut and their effect on growth and phosphorus 516 mobilization. Biol Fertil Soils. Biol Fert Soils 2011; 47: 437-446. 517 Zheng BX, Hao XL, Ding K, Zhou GW, Chen QL, Zhang JB, et al. Long-term 518 nitrogen fertilization decreased the abundance of inorganic phosphate 519 solubilizing bacteria in an alkaline soil. Sci Rep-UK 2017; 7: 42284. 520 Zimmerman AR, Gao B, Ahn MY. Positive and negative carbon mineralization 521 priming effects among a variety of biochar-amended soils. Soil Biol Biochem 522 2011; 43: 1169-1179. 523 27 524 Figure legends 525 Figure 1. The diagram of experimental design. The soils sampled from Changshu was 526 used in new iPSB strain isolation and its sterilized one were used for rape growth with 527 different treatments. The asterisk (*) indicates that those samples were physically or 528 chemically analyzed. The pound sign (#) suggests that those strains were being 529 biochemically determined. The detail method was explicated in Supplementary 530 Information. 531 Figure 2. (a) Morphology of the iPSBs by scanning electron microscopy and colonies 532 after 72 h of cultivation at 30 °C on (b) LB medium, (c) PKV medium with NH4+ as the 533 N source and (d) PKV medium with NO3- as the N source. E. coli without P-solubilizing 534 ability was used as a negative control. (e) Heatmap of the profiles of medium pH and 535 organic-anion exudation by the iPSBs after 144 h of cultivation based on the average 536 of three replicate experiments. 537 Figure 3. Abundance (a) and Shannon diversity (b) of the biochars inoculated with the 538 iPSB community. Error bars represent the standard deviation of the replicates (n=4). 539 Different letters indicate significant differences at P < 0.05. 540 Figure 4. Heatmap and redundancy analysis (RDA) of the iPSB communities with the 541 biochars after four weeks of rape growth based on Bray-Curtis distances. (a) Profile of 542 iPSB community composition and (b) the effect of the physical and chemical properties 543 of the soil and biochar on the iPSB communities. The plotted values are natural- 28 544 logarithm transformations of relative iPSB community abundance. The columns in (a) 545 are labeled with biochar names and replicate numbers. The seven strains used to 546 generate the iPSB community are identified in red, and the other strains are identified 547 as species or genera. 548 Figure 5. Relationships between iPSB community abundance and the Shannon 549 diversity index (a), plant biomass (b), plant P (c) and plant P uptake (d). 29 550 30