Gross phosphorus fluxes in a calcareous soil inoculated with Pseudomonas protegens CHA0 revealed by 33P isotopic dilution

Soil Biology & Biochemistry 104 (2017) 81e94

Contents lists available at ScienceDirect

Soil Biology & Biochemistry journal homepage: www.elsevier.com/locate/soilbio

Gross phosphorus fluxes in a calcareous soil inoculated with Pseudomonas protegens CHA0 revealed by 33P isotopic dilution €der b, A. Oberson a, * G. Meyer a, b, E.K. Bünemann a, b, E. Frossard a, M. Maurhofer c, P. Ma a

Institute of Agricultural Sciences, Group of Plant Nutrition, ETH Zurich, 8092 Zurich, Switzerland Research Institute of Organic Agriculture FiBL, 5070 Frick, Switzerland c Institute of Integrative Biology, Group of Plant Pathology, ETH Zurich, 8092 Zurich, Switzerland b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 June 2016 Received in revised form 29 September 2016 Accepted 2 October 2016

Inoculation with phosphorus (P) solubilizing bacteria is being proposed to increase P availability for plants by mineralization and solubilization of non-available soil and fertilizer P. Solubilization of inorganic P compounds by bacterial strains has repeatedly been shown on agar plates and in liquid media. However, the effects of inoculation on P availability to plants growing in soils, either in pot or field studies, are inconsistent and do not allow to separate between direct effects on P availability and indirect effects such as improved plant health. This differentiation could be achieved using 33P isotopic labeling. We applied the 33P isotopic dilution method in a pot and in an incubation experiment to study gross P fluxes in a calcareous soil inoculated with the P solubilizing bacteria Pseudomonas protegens CHA0. We hypothesized that the inoculant dilutes the specific activity (33P/31P) in the soil solution or in the plant shoots because of P solubilization beyond the P mobilization by the endogenous microbial biomass. To this end, we conducted a plant growth experiment with Lolium multiflorum var. Gemini and an incubation experiment. In both experiments, the soil was amended or not with a calcium P rich sewage sludge ash, and both treatments were conducted with and without inoculation. The inoculant was able to solubilize P from sewage sludge ash under controlled conditions in liquid media. However, it did not enhance P release from soil or from sewage sludge ash in the incubated soil. Inoculation of the soil reduced organic P mineralization by the soil microbial biomass, which was supported by a simultaneous decrease in soil respiration. Thus, any inorganic P solubilized by the inoculant might have been offset by less basal organic P mineralization. Increased P uptake of inoculated Lolium multiflorum at first harvest was attributed to an indirect effect, since the specific activity in shoots of inoculated Lolium multiflorum was not decreased. Although sewage sludge ash contained very little water-soluble P, an increase in P availability following sewage sludge ash addition could be shown using 33P isotopic dilution, while biological processes remained unchanged. While in this study, the inoculant did not increase P availability, the approach presented here can give insight into the mechanisms underlying beneficial effects of inoculants. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Calcareous soil Phosphorus Isotopic dilution Microbial mobilization Phosphorus solubilizing bacteria Recycling fertilizer

1. Introduction Inoculation of soil-plant systems with phosphorus (P) solubilizing bacteria (PSB) has been proposed since decades to increase P availability for plants (Kucey et al., 1989; Goldstein, 2007; Cornish,

Abbreviations: PSB, P solubilizing bacteria; SA, specific activity; IEK, isotopic exchange kinetics; SSA, sewage sludge ash; Inoc, inoculated treatments; þP, P fertilized treatments with SSA. * Corresponding author. E-mail address: [email protected] (A. Oberson). http://dx.doi.org/10.1016/j.soilbio.2016.10.001 0038-0717/© 2016 Elsevier Ltd. All rights reserved.

2009; Owen et al., 2015). Such bio-inoculants seem particularly useful when soils and/or fertilizers contain P forms of limited P availability. The P availability to plants in calcareous soil is often limited due to the reaction of P with calcium containing phases (Frossard et al., 1995). Rock phosphates (Fardeau et al., 1988) and some recycling fertilizers, e.g., ashes recovered after the incineration of sewage sludge (Nanzer et al., 2014b), contain Ca bound P forms such as apatite. Those P forms are not water soluble and are of low effectiveness to crops, particularly when applied to soils with neutral to alkaline pH (Nanzer et al., 2014a; Brod et al., 2015). In these situations where soil and fertilizer characteristics limit the

82

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94

crop P supply, the enhancement of biological P mobilization by plants and microorganisms is seen as promising option to increase the P use efficiency in the cropping system (Simpson et al., 2011). In vitro studies on agar or in liquid culture have shown the ability of bacteria, e.g., Pseudomonas, Bacillus and Rhizobium (Rodríguez and Fraga, 1999) and fungi, e.g., Aspergillus and Penicillium spp. (Richardson and Simpson, 2011) to solubilize tri-calcium phosphate. Despite the fact that many commercial products containing P solubilizing microorganisms are available on the market, the performances of such bio-inoculants in soil-plant systems remain inconsistent (Kucey et al., 1989; Owen et al., 2015). Inoculation with a PSB is expected to provide an added value to the endogenous soil microorganisms, which already mineralize and potentially solubilize P (Oberson and Joner, 2005). Phosphorus mobilizing bacteria increase P availability through organic P mineralization, e.g., by increased phosphatase activity (Molla et al., 1984), and mineral P solubilization due to exudation of organic acid anions and protons (Rodríguez and Fraga, 1999). For some PSB it is known that solubilization activity depends on the transformation of glucose to gluconic acid, but that glucose can also be transformed to anti-microbial compounds (De Werra et al., 2009). Thus, inoculation with PSB can increase plant P uptake as direct response to P solubilized by the PSB, or as an indirect response via enhanced plant growth because of reduced pathogen pressure in the rhizosphere. Isotope dilution techniques based on the use of a radioisotope P (e.g., 33P) might help to differentiate between direct and indirect effects of PSB on plant P uptake. Indeed, any P mobilized from otherwise non-plant available P would cause a dilution of the specific activity (33P/31P) in shoots of plants grown on a soil labeled with 33P (Asea et al., 1988; Barea et al., 2007; Frossard et al., 2011). Isotopic dilution has been widely applied to study soil P availability, P fluxes in the soil and P uptake from different soil and fertilizer P sources (Frossard et al., 2011). By applying the tracer 33P in soil incubation experiments, the contribution of physicochemical processes delivering available P in the soil can be differentiated from biological P processes and the respective gross P fluxes can be calculated (Oehl et al., 2001a; Bünemann et al., 2004c; Bünemann et al., 2007). The dilution of the specific activity of soil solution P due to physiochemical P exchange processes are determined in short-term (~100 min) isotopic exchange kinetics (IEK) batch experiments and then extrapolated for the time of interest (Fardeau et al., 1991). The dilution of the 33P tracer in the soil solution due to both physico-chemical and biological P processes, is measured in an incubation experiment, usually for no longer than 40 days (Oehl et al., 2001b). The difference of measured and extrapolated 33P dilution in soil solution has been assigned to gross organic P mineralization (Oehl et al., 2001b; Bünemann et al., 2007, 2012; Achat et al., 2010; Randriamanantsoa et al., 2015), but could also involve P mobilization from non-isotopically exchangeable inorganic P pools (Asea et al., 1988). The objective of our study was to assess gross P fluxes using 33P isotopic dilution in a calcareous soil with and without PSB inoculation, using a PSB with known P mobilization mechanisms, and with and without addition of a non-water soluble recycling P fertilizer produced from sewage sludge ash. To this end, first the potential of the inoculant to solubilize P from tri-calcium phosphate and the recycling P fertilizer was measured in vitro by solubilization assays. Then the fraction of P derived from the soil and the fertilizer was quantified as affected by inoculation in Lolium multiflorum (ryegrass) growing in a soil labeled with 33P. Finally, the 33P isotope dilution method was applied to detect gross P solubilization during a soil incubation. Also the specific activities of soil P pools sequentially extracted from the incubated soil were followed to determine the source of any solubilized P. We assumed that any P solubilized by the inoculant would cause a dilution of the specific

activity of P in plant shoots and in the soil solution additional to the dilution caused by activities of the endogenous soil microbial biomass. In consequence, we expected gross P fluxes to be greater in the inoculated than in the non-inoculated soil. This is the first study assessing gross P fluxes in soil inoculated with a PSB. 2. Material and methods Three types of experiments were carried out: i) an in vitro liquid culture experiment to determine the P solubilization potential by the PSB from the sewage sludge ash, ii) a plant growth study with ryegrass lasting 69 days and iii) a soil incubation study lasting 68 days (Fig. 1). Experiments ii) and iii) comprised 33P isotopic dilution applied to the same treatments, which were soil amended (þP) or not (0P) with sewage sludge ash, each with (Inoc) or without inoculation with the P solubilizing strain Pseudomonas protegens CHA0. In addition, soil respiration measurements were conducted throughout the entire incubation period. The 33P isotopic exchange kinetics (IEK) were determined in short term (up to 90 min) lasting batch experiments using incubated, non-33P labeled soil, at the onset (0P treatment) and at the end of the incubation experiment (0P and þP treatment). 2.1. Soil and P fertilizer For the incubation and plant experiments a calcaric Cambisol from an arable field in Rümlang, Switzerland (Table 1) was used. The soil was taken at 0e20 cm soil depth in August 2013 from the non-fertilized border strip of a field experiment described by Gallet et al. (2003). Field moist soil was sieved at 4 mm for the plant growth experiment and at 2 mm for the incubation experiment and stored at 4  C. Soil was reactivated in the dark at 22  C at 30% water holding capacity (WHC, 159 g H2O kg1 soil) for five weeks prior to the experiments. The recycling P fertilizer used in the incubation and plant experiment is a sewage sludge ash (SSA) that had been thermochemically treated with MgCl2 (Table 2). The availability of P in this SSA to plants is high in acidic but very little in alkaline soils (Nanzer et al., 2014a). The P species in the thermo-chemically treated SSA consist mainly of Ca4Mg5(PO4)6, Mg3(PO4)2, Ca5(PO4)3 (Adam et al., 2009). The treated SSA was milled and sieved at 150 mm. 2.2. Inoculant, P solubilization assay and cell culturability of inoculant Pseudomonas protegens CHA0 was isolated in 1984 from a tobacco field suppressive to Thielaviopsis basicola in the region of Morens, Switzerland (Stutz et al., 1986). It produces antifungal compounds such as 2,4-diacetylphloroglucinol and pyoluteorin (Haas and Defago, 2005). In addition, it is able to solubilize TCP (Ca3(PO4)2) by exudation of gluconic acid (De Werra et al., 2009). The strain used in this study is a spontaneous rifampicin-resistant mutant of P. protegens strain CHA0 (CHA0-Rif). The mutant is equivalent to the wild-type in terms of grow in vitro, stress resistance and root-colonization (Troxler et al., 2012) and is from here on referred to as CHA0. Cultivation of the strain was done consecutively on King's B medium (King et al., 1954) and in Luria broth (Sambrook and Russell, 2001). To make sure that exclusively strain CHA0 was growing, the antibiotic rifampicin was added to the first two growth media at a concentration of 100 mg Rif L1. Harvested cells were washed and re-suspended in 0.9% sterile NaCl solution. The final bacterial cell concentration was set by measuring the optical density according to De Werra et al. (2009) to reach a final

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94

83

Fig. 1. Time scale of the plant growth experiment and the incubation study.

concentration of 107 CFU g1 dry soil in the incubation and plant experiment. Total P introduced with the inoculum resulted in 66 ± 1 mg P kg1 dry soil, as determined by total digestion with ammonium persulfate (Tiessen and Moir, 1993) on a subsample containing 108 cells. Culturable P. protegens CHA0 cells in soil and on roots were determined by plating soil or root suspension on King's B agar containing antibiotics (cycloheximide, chloramphenicol, ampicillin and rifampicin at concentration of 1 ml L1, 0.13 ml L1, 0.4 ml L1 and 2 ml L1, respectively) (Meyer et al., 2011). In order to test the

Table 1 Characteristics of the soil used in the incubation experiment and plant growth study with ryegrass. Calcareous soil Rümlang Soil type

Calcaric Cambisol

Clay (g kg1)a Silt (g kg1)a Sand (g kg1)a Soil pHb Total P (mg kg1)c Inorganic P (mg kg1)d E1min (mg kg1)e P in soil solution Pw (mg kg1) Organic C (g kg1)f Microbial C (mg kg1) Dissolved organic C (mg kg1) Total CaCO3 (g kg1)g Active Ca2þ (g kg1)h WHCmax (g H2O kg1)i Cation exchange capacity (cmolþ kg1)h

209 322 469 7.7 960 480 2.6 0.5 32 209 71 41 18 520 18.8

a

Gravimetric measurement (Agroscope, 2004). pH in H2O (1:2.5) (Agroscope, 2004). c Total P obtained by digestion with H2O2/HNO3 in a microwave. d Inorganic P obtained using the ignition/extraction method of Saunders and Williams (1955). e P exchangeable within 1 min (E1min), determined by isotopic exchange kinetics (Fardeau et al., 1991). f Organic C determined according to Agroscope (2004). g Total CaCO3 obtained by measuring the released volume of gas (Agroscope, 2004). h Titration (Agroscope, 2004). i Water holding capacity (WHC), maximal soil saturation with H2O without external pressure. b

recovery of added cells to the soil, an unfertilized soil was inoculated and a soil suspension obtained after 1 day was plated to determine cell cultivability of strain CHA0. Subsamples of all incubated soil treatments were plated after 40 days and at the end of the experiment. In the growth trial, culturable cells of strain CHA0 on root samples were determined after the third cut, i.e., 69 days after inoculation. In order to confirm the ability of P. protegens CHA0 to solubilize P from the SSA, a quantitative P solubilization assay was conducted with different glucose concentrations, as strain CHA0 requires glucose to produce gluconic acid (De Werra et al., 2009). For this, a modified liquid National Botanical Research Institute's phosphate growth medium (NBRIP) (Nautiyal, 1999) was amended with either SSA or with TCP, with two and three glucose concentrations, respectively. The P concentration in the media was 1 g P L1 in the form of 5 g L1 of TCP or 16 g L1 of SSA. Glucose was added in different concentrations: 10 g L1, 5 g L1 and 1 g L1 in the TCP treatments and 10 g L1 and 5 g L1 in the SSA treatments. Before autoclaving, the pH of the NBRIP liquid media was adjusted to pH 7.2 using NaOH. Glass vials (30 ml) containing 10 ml of the modified NBRIP liquid media were inoculated with 340 ml of bacteria suspension (108 CFU ml1) with eight replicates per treatment. Noninoculated control treatments were performed in quadruplicates. The vials were incubated on a rotary shaker at 150 rpm at 27  C during 120 h. Then the solution was filtered using a 0.2 mm Millipore membrane before determination of pH and the solution P concentration using the colorimetric method of Ohno and Zibilske (1991).

Table 2 Characteristics of the thermo-chemically treated sewage sludge ash (SSA) (Nanzer et al., 2014a). Thermoechemically treated sewage sludge ash Total P (g P kg1)a Water solubility (%P) Resin extractable P (%P) HCl extractable P (%P) Crystallinity (%)b a

63 <1 17 74 44

XRF measurement after Li2B4O7/LiBO2 fusion (PANalytical Axios Advanced). Determined by Rietveld refinement of XRD spectra (X'Pert PRO PANalytical diffractometer) (Nanzer et al., 2014a). b

84

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94

2.3. Plant growth experiment A growth trial was conducted in a glasshouse using 33P labeling of isotopically exchangeable soil P (Morel and Fardeau, 1989) to obtain quantitative information on the fraction of P in ryegrass derived from soil or the recycling fertilizer P that has been mobilized by P. protegens CHA0. The experimental design consisted of two non-inoculated control treatments (0P, þP) and two inoculated treatments (0P Inoc, þP Inoc) with four replicates each. Moist soil portions equivalent to 1 kg soil dry weight were labeled with carrier-free 33P-orthophosphate (Hartmann Analytic, Braunschweig, Germany) at a rate of 1.4 MBq kg1 soil, transferred into bottom sealed pots to prevent leaching and incubated for 15 days to reach an isotopic equilibrium. Then, an equivalent of 50 mg P kg1 soil of SSA was mixed into the soil of the þP treatments (þP, þP Inoc). Pots were sown with 0.7 g kg1 soil of ryegrass seeds (Lolium multiflorum var. Gemini). The CHA0 treatments (0P Inoc, þP Inoc) were inoculated with 107 CFU g1 soil at germination (5 days after sowing) by adding to the soil surface 50 ml distilled water containing re-suspended cells. The same procedure was done with non-inoculated treatments (0P, þP), but without re-suspended cells. The plants were grown in a glasshouse with a mean day temperature of 22  C and night temperature of 16  C and a photoperiod of 16 h with artificial light intensity of maximum 500 mmol m2 s1 and at atmospheric humidity of 65%. Soil water content was kept between 40 and 60% WHC and pots were randomized every second day. All pots received a P-free nutrient solution with N, K and Mg providing twice the expected nutrient uptake: at sowing and after each cut the plants were fertilized with 230 mg N, 280 mg K and 24 mg Mg kg1 soil and with adequate amounts of S, Ca, Fe, B, Mn, Cu, Mo and Zn. The plants were harvested three times, at 29, 48 and 69 days after sowing. Because P derived from the seed is an additional non-33P labeled P source which can result in an overestimation of P taken up from the non-labeled fertilizer or of P mobilized from non-isotopically exchangeable soil P pools, the seed P contribution has to be accounted for. To this end, two additional treatments with 50 mg P kg1 soil as water soluble P (Ca(H2PO4)2*H2O) were included. In one treatment, soil exchangeable P (indirect) and in the other treatment the P fertilizer (direct) was labeled. The P in plant shoots of the 1st and 2nd harvest derived from seed could then be estimated from the difference in the P derived from (Pdf) the fertilizer obtained with these two approaches (Oberson et al., 2010). The same Pdf seed of 0.4 mg P kg1 soil was used for all treatments, because the P concentration in the shoots was similar in all treatments. 2.4. Incubation experiment The incubation experiment had the same treatments as the plant growth experiment. Soil exchangeable P was labeled by mixing carrier-free 33P-orthosphosphate with moist soil equivalent to 3.4 kg dry soil, resulting in a soil label (R) of 27.5 kBq g1 soil. The labeled soil was then split into four portions equivalent to 0.840 kg dry soil. Two of these portions were fertilized with 50 mg P kg1 soil by adding SSA (þP), while the remaining two did not receive any P (0P). None of the treatments were supplemented with nutrients other than P. One þP and one 0P soil portions were inoculated with the P. protegens CHA0 cell suspension to reach the target concentration of 107 CFU g1 soil. The inoculation was done by mixing the diluted cell suspension in to the soil. The same procedure was done with non-inoculated treatments (0P, þP), but without re-suspended cells. The water content of the four portions was adjusted with double distilled water (ddH2O) to 50% WHC. Aliquots of 120 g soil per treatment and sampling time points were

then filled in 250 ml plastic bottles with screw caps. The water content was maintained at 50% WHC throughout the experiment. The bottles were kept in a dark box in the glasshouse at 65% atmospheric humidity with mean temperatures of 22 and 16  C during day and night, respectively. Soil was sampled immediately after inoculation and thereafter six times during the incubation lasting 69 days, except for the isotopic exchange kinetics (IEK) batch experiment (see below). Unlabeled moist soil was prepared as for the incubation experiment to measure soil respiration weekly and IEK parameters at the beginning and end of the incubation period using four analytical replicates. 2.5. Analysis 2.5.1. Shoot P content Shoots of ryegrass were cut 29 and 48 days after sowing at 4 cm above the ground and at final harvest after 69 days at 0.5 cm above the soil and dried at 60  C for 72 h to determine dry matter (DM) yield. For determination of P contained in shoots, incinerated samples were digested with HNO3 following the method described by Nanzer et al. (2014a). The P concentration in the extract was then determined colorimetrically by malachite green (Ohno and Zibilske, 1991) and the 33P with the liquid scintillation counting (TRI-CARB 2500 TR, liquid scintillation analyzer, Packard Instruments, Meriden, CT). 2.5.2. Respiration Soil respiration was determined weekly by trapping CO2 released from 20 g dry soil equivalent in a NaOH trap, followed by titration with HCl (Alef, 1995). 2.5.3. Isotopic exchange kinetics The IEK were conducted according to Fardeau et al. (1991) in a batch experiment at the beginning of the incubation experiment in the 0P treatment and at the end in all treatments. Moist soil equivalent to 10 g dry weight was weighed into 250 ml Nalgene bottle, 99 ml ddH2O was added and placed on an end-over-end shaker at 15 rpm for 16 h at room temperature. Then, bottles were placed on a magnetic stirring plate and at time zero 1 ml carrier-free 33P-orthosphophate (around 45 kBq) was added. At 1, 4, 10, 30, 60, and 90 min solution was sampled with a syringe and filtrated (0.2 mm) before determining the ortho-P (Cp, mg P L1) and the radioactivity (rCp) in the soil solution. 2.5.4. Sequential soil P fractionation of incubated soil labeled with 33 P Soil P was sequentially extracted to study the effect of P addition and inoculation on the size and SA of P pools. The extraction method was based on Hedley et al. (1982), as modified by Tiessen and Moir (1993) and was done at 1, 12, 27, 40, 55 and 69 days after labeling. All measurements were done with four analytical replicates. Soil water extractable P (Pw, mg P kg1 soil) and radioactivity (rPw) of the incubated soil were determined by extracting moist soil corresponding to 10 g soil dry weight, at a 1:10 soil to water ratio, under same extraction conditions (shaking and filtration) as for the IEK. The corresponding sampling time was the time of filtering at day 1, 6, 12, 27, 40, 55, 69 after labeling. Resin extractable soil P (Pres) and hexanol labile microbial P (Pmic) were determined on 2 g dry soil equivalent according to Kouno et al. (1995), using hexanol (McLaughlin et al., 1986) instead of chloroform. We did not use a conversion factor which accounts for incomplete recovery of microbial P during fumigationextraction. The hexanol fumigated samples were further

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94

extracted with 0.1 M NaOH and 1 M HCl following the procedure of Bünemann et al. (2004b). Total (NaOH-Ptot), inorganic P (NaOH-Pi) and organic P (NaOH-Porg) were determined according to Tiessen and Moir (1993). The P in the extracts was measured colorimetrically by malachite green method (Ohno and Zibilske, 1991) with pH adjustment when needed and 33P by liquid scintillation counting (TRI-CARB 2500 TR, liquid scintillation analyzer, Packard Instruments, Meriden, CT). The 33P in NaOH-Pi pool was measured in the acidified NaOH extract (Tiessen and Moir, 1993). Corrections for quenching were not required in any extract, as tested by 33P additions.

2.6.1. General 31P calculations Water extractable P (Pw) is calculated as follows:


Pw mg P kg1 ¼ 10  Cp

(1)

where Cp stands for P concentration in solution (mg P L1) and the factor 10 derives from soil-solution ratio. Microbial P (Pmic) is calculated as the difference between P in the fumigated (Pfum) and non-fumigated (Pres) subsamples as follows:


Pfum  Pres Pmic mg P kg1 ¼ 31 Prec

(2)

2.6.2. General calculations when using the indirect isotope dilution approach The specific activity (SA) in a P pool is expressed as follows:


 33 P SA Bq mg1 P ¼ 31 P

(3)

where 31P stands for P concentration in a given P pool in mg P kg1 soil and 33P for the activity in Bq kg1 soil. The activity in the microbial P pool has to be corrected for 33P sorption according to McLaughlin et al. (1988). The correction procedure has often been described in detail, recently by Bünemann et al. (2016). In brief, 33Pfum is first corrected by 33P released from the soil solid phase due to isotopic exchange reaction. This release is determined from the increased 33P in non-fumigated samples after 31P-spike addition. Thereafter, the activity is corrected for sorption of released 33P during fumigation.

33

Phex

33 Pfum   Bq kg1 soil ¼

recoveryð%Þ ¼




a

33 P spike

33 Prec

 þb

(4)

where 33Pfum stands for the activity (Bq kg1 soil) measured in fumigated samples, a is the slope and b the intercept of the 33P increase due to 31P-spike addition and 33Prec stands for the fraction of 33P spike recovered in non-fumigated subsamples. The 33Prec was averaged over all sampling points and per treatment and was 68%, 67%, 72% and 75% for 0P, 0P CHA0, þP and þP CHA0, respectively.

r  100 R

(5)

where R stands for total introduced activity and r for the activity of a given P pool, both expressed in Bq kg1 soil. The relative proportion of P derived from fertilizer (Pdff) in a given pool when using the indirect labeling approach is calculated as:

 1

SAþP SA0P

 (6)

where SAþP denotes the SA obtained for the P fertilized treatment and SA0P stands for the non-P fertilized treatment. Multiplying the relative Pdff by the pool (plant, soil) size in the þP treatment equals P derived from fertilizer in mg P kg1 soil (Frossard et al., 2011).

2.6.3. Calculations of isotopic dilution in soil-plant system The quantity of isotopically exchangeable P in soil can also be derived from the SA of a plant grown in labeled soil, the L-value (Larsen, 1952), and is calculated according to Truong and Pichot (1976) and Sibbesen (1984) as follows:




Ln mg P kg

The sorption correction factor 31Prec stands for the recovery of a31P-spike in non-fumigated soil samples (Brookes et al., 1982; Bünemann et al., 2004c) and was determined for each treatment at every sampling time. Since there was no significant effect of inoculation on sampling time on 31Prec, values of 77% and 78% were used for the non-P fertilized and P fertilized treatments, respectively.




The fraction of total introduced activity (recovery, %) in a given P pool is calculated as follows:

Pdffð%Þ ¼

2.6. Calculations

85

1

 soil ¼

n1 X

R

!, ri

i¼1

þ

n1 X

Pn

ðPi  Pseedi Þ

rn  Pseedn



(7)

i¼1

where R stands for total introduced radioactivity (MBq kg1soil) at t ¼ 0, rn for the amount of 33P in the shoots (MBq kg1 soil), Pn for the quantity of 31P (mg P kg1 soil) in the shoot and Pseedn for the 31 P (mg P kg1 soil) in shoots derived from the seed (see plant P growth experiment) at the nth harvest. The sum n1 i¼1 ri stands for 33 the amount of P taken up in the shoots between the first and the Pn1 (n-1)th harvest, while the sum i¼1 ðPi  Pseedi Þ stands for the 31 quantity of P in the shoots corrected by seed P between the first and the (n-1)th harvest. The L-value can then be compared with E(t)extrapolated (see below) (Frossard et al., 1994).

2.6.4. Calculation of gross P fluxes The decrease of the 33P radioactivity r(t) in soil solution over time due to physico-chemical process can be described by the nonlinear equation (Fardeau et al., 1991; Frossard et al., 2011):

i h 1 n 10  Cp rt =R ¼ m t þ mn þ Pi

(8)

where t stands for time (min), Pi for total inorganic P in soil and Cp for P concentration in solution (mg P L1), which is multiplied by 10 to obtain the concentration as Pw in mg P kg1 soil. Parameters m and n are estimated by non-linear regression. The starting values of parameter estimation by non-linear regression analysis were obtained by the following log-function (Fardeau et al., 1991):

log½rt =R ¼ log½r1 =R  n  logðtÞ

(9)

where t denotes the sampling time (min) in the IEK batch experiment and rt/R the measured fraction at time t. Finally, the isotopically exchangeable P within t minutes (Evalue, (E(t)extrapolated)) is extrapolated (Frossard et al., 1994):

86

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94


 EðtÞextrapolated mg P kg1 soil ¼ 10  Cp  R=rt

the standard deviation with N ¼ 4 unless otherwise stated.

(10) 3. Results

where R/rt is the inverse of rt/R and is estimated using equation (8). Likewise, an E(t)measured can be calculated from the Pw and rt/R measured in the incubation study:


 EðtÞmeasured mg P kg1 soil ¼ Pw  R=rt

(11)

Gross P mobilization is then derived from the differences between E(t)measured in the incubation experiment and E(t)extrapolated (Oehl et al., 2001b):


 gross P mobilization mg P kg1 soil ¼ EðtÞmeasured  EðtÞextrapolated

(12)

Dividing gross P mineralization by time t (days) results in daily gross P mineralization rates. Microbial P immobilization is calculated as follows:

P immobilization


 SAPmic mg P kg1 soil ¼  Pmic SAPres

(13)

with SAPmic corrected for 31P and 33P sorption according to McLaughlin et al. (1988), see equations (2) and (4). Daily P immobilization rates are calculated by dividing the immobilization rate by time t (days).

3.1. In vitro P solubilization assay The inoculant P. protegens CHA0 solubilized inorganic P from SSA and TCP (Table 3). At a glucose concentration of 10 g L1, 41 ± 1% of P from the SSA and 90 ± 1% of P from TCP was solubilized, while a decrease in pH to about 4 was measured in both treatments. Providing only half of the glucose (i.e., 5 g L1) led to the solubilization of 37% of the P added as SSA and of 42% of the P added as TCP and to a pH reduction to about 4.8 in both treatments. A further reduction in available glucose to 1 g L1 did not cause a pH decrease and only 6% of P added with TCP was solubilized. 3.2. Persistence of strain CHA0 in soil and on roots Immediately after inoculation, 83% of the inoculated cells (107 CFU g1 dry soil) of strain CHA0 could be re-isolated from unfertilized soil. After 40 days, the cells concentration of strain CHA0 in soil had decreased to 6  104 CFU g1 dry soil equal to less than 1% of the initial concentration, while in non-inoculated soil no strain CHA0 cells were detected. By the end of the experiment, bacterial strain CHA0 concentrations were not detectable anymore, neither in incubated soil nor on roots of ryegrass. 3.3. Soil respiration

2.7. Statistics Statistical analysis was performed with R (R Core Team, 2014). For each sampling date of the plant growth and incubation experiment, data were tested by two-way ANOVA with factors P fertilizer (P), inoculant (I) and P fertilizer  inoculant (P  I) interaction. Data from respiration measurements were tested by repeated measure ANOVA with factors P fertilizer (P), inoculant (I) and time (t). Multiple comparisons using Tukey Honest Significant Differences tests were performed whenever the ANOVA indicated significant differences (p  0.05) of factor P fertilizer or inoculant. When the interaction term was significant, the interaction plot was consulted and no Tukey HSD on factor P fertilizer or inoculant was done. A Student's t-test was performed to check significant differences of extrapolated and measured E-values and parameters m and n between þP and 0P treatments. Fractions (r(t)/R, root to shoot ratio) were arcsin(sqrt()) transformed and data which did not fulfill assumption of ANOVA were transformed using Box-Cox transformations for linear models (Box and Cox, 1964). Error bars denote

Daily respiration rates decreased over time on average over all treatments from 27 to 5 mg C kg1 soil day1 (Fig. 2). Inoculation lowered daily respiration rates significantly by 11% on average compared to non-inoculated soil from day 2 until day 23. Soil amendment with SSA did not affect respiration rates. The resulting cumulated soil respiration after 72 days of incubation was on average 615 mg C kg1 soil, which is about 1.9% of total organic C. Over the entire experimental period about 9% (p < 0.01) less C was respired in inoculated than in non-inoculated soils. 3.4. Plant growth experiment Inoculating ryegrass did not increase shoot DM yield, regardless of P fertilization (Table 4). However, P fertilization increased total shoot DM from 6.7 to 7.4 g DM kg1 soil compared to non-P fertilized plants. This difference in DM production was most pronounced in the first cut. Fertilization decreased the root to shoot ratio significantly from 0.24 to 0.19.

Table 3 In vitro P solubilization assay: pH, P in solution and net P solubilized (%) of total added P as tri-calcium phosphate (TCP) or sewage sludge ash(SSA) at a rate of 1 g P L1 in nonbuffered NBRIP liquid media culture with P. protegens CHA0 in dependency of glucose content of 1, 5 and 10 g L1 after 120 h of inoculation. Mean ± standard deviation.

ŧ

P source

Inoculant

Replicates N

Total P added mg P L1

Glucose added g L1

pHŧ

TCP TCP TCP TCP TCP TCP SSA SSA SSA SSA

no CHA0 no CHA0 no CHA0 no CHA0 no CHA0

4 8 4 8 4 8 4 8 4 8

1000 1000 1000 1000 1000 1000 1000 1000 1000 1000

1 1 5 5 10 10 5 5 10 10

7.2 7.3 7.2 4.8 7.2 4.0 7.7 4.9 7.6 3.9

Different letters denote significant differences at p < 0.05 within a P source and glucose treatment. ŧŧ Different letters denote significant differences at p < 0.05 within a P source treatment.

P in solution mg P L1 ± ± ± ± ± ± ± ± ± ±

0.04 0.02 0.04 0.1 0.01 0.01 0.01 0.04 0.01 0.02

16 ± 0 56 ± 1 14 ± 0 418 ± 13 15 ± 0 887 ± 12 25 ± 0 362 ± 11 27 ± 0 396 ± 9

Net P solubilized by strain CHA0ŧŧ % 6% ± 0% a 42% ± 1% b 90% ± 1% c 37% ± 1% a 41% ± 1% b

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94

87

Fig. 2. Soil respiration as affected by soil amendment P fertilizer with sewage sludge ash or inoculation with P. protegens CHA0. Daily (mg C kg1 day1) (left) and cumulated (right) soil respiration (mg C kg1 soil) of non-fertilized (0P), non-fertilized inoculated (0P Inoc), P fertilized (þP) and P fertilized inoculated (þP Inoc) soils during the period of incubation (from day 2 till 72). Error bars represent standard deviation of four replicates. Significance levels of repeated measure ANOVA with factors P fertilizer (P), inoculant (I) and the interaction P fertilizer  inoculant (P  I) per time point are indicated.

Shoots of ryegrass fertilized with SSA had significantly lower specific activities (SAshoots) (Fig. 3) in cuts 2 and 3 than shoots of non-P fertilized plants, revealing 0.6 (p < 0.01) and 0.3 mg P kg1 soil (p < 0.001) in shoots derived from the added fertilizer (equation (6)). The effect of inoculation on SAshoots depended on P fertilization. A significantly greater SAshoots was measured in inoculated, non-P fertilized plants of cut 1. Then, at final harvest, an increase of 4% (p < 0.05) in SAshoots was observed in inoculated plants of both the 0P and þP treatment. Inoculation increased the P uptake of ryegrass significantly only in the first cut by 0.5 mg P, from 3.5 to 4.0 mg P kg1 soil (Fig. 4). However, total P taken up in the three cuts of ryegrass was similar for all treatments. The L-values were significantly increased with P fertilization by 7.3, 7.8 and 9.3 mg P kg1 soil at cut 1, cut 2 and cut 3, respectively (Table 5). At cut 1, inoculation decreased the L-values of non-P fertilized plants by 8.0 mg (p < 0.05), while at cut 3, inoculation decreased L-values, independent of P fertilization by 3.7 mg P kg1 soil (p < 0.05). 3.5. Gross P fluxes in incubated soil 3.5.1. IEK parameters and extrapolated E1min The parameters of IEK in soil determined in batch experiments were affected by P fertilization (Table 6). Addition of P significantly

increased Pw and significantly decreased parameter n. In consequence, extrapolated E1min in P fertilized soil was significantly greater with 3.2 in the P fertilized soil compared to 2.8 mg P kg1 soil in the unfertilized soil. Thus, different parameters for E-value extrapolation were used for fertilized and unfertilized soils. Inoculation did not affect E1min (data not shown). 3.5.2. Gross P mobilization The measured E-values were greater in the P fertilized than in the 0P soils (Fig. 5). Inoculation slightly reduced the E-values compared to the non-inoculated treatment. Greater measured Evalues of the þP than of the 0P soil resulted from higher Pw concentration and lower SAPw (Fig. 6). Measured Pw concentrations were almost constant over time. Fertilization with P increased Pw on average by 21% (p < 0.01), from 0.50 to 0.60 mg P kg1 soil throughout the experiment compared to non-fertilized soil. The added P diluted SAPw in average by 7%, but most strongly during the first three weeks. Measured E-values in non-inoculated 0P and þP soils were between day 6 and 13 significantly (p < 0.05) greater, on average by about 13%, than E-values extrapolated from the IEK for the 0P and þP soil, respectively (Fig. 5). From these differences, daily gross P mobilization rates of 1.2 mg P kg1 soil day1 were derived at day 1 and decreased continuously to 0.17 mg P kg1 soil day1 till day 27 in the 0P soil

Table 4 Dry matter (DM) of ryegrass as affected by P fertilization with sewage sludge ash or inoculation with P. protegens CHA0. Shoot DM after three cuts and root DM after the third cut and root to shoot ratios of non-P fertilized (0P), inoculated non-P fertilized (0P Inoc), P fertilized (þP) and inoculated P fertilized (þP Inoc). Mean ± standard deviation. Shoot

Root

Cut 1

Cut 2

Cut 3

Total

g DM kg1 soil 0P 0P Inoc þP þP Inoc Significance level (p) P fertilizer (P) Inoculant (I) PI

2.4 2.5 2.6 2.7 *** ns ns

± ± ± ±

0.2 0.1 0.0 0.1

Root to shoot ratio

Total g DM kg1 soil

2.5 2.3 2.6 2.7 ns ns ns

ns: not significant, *p < 0.05, **p < 0.01, ***p < 0.001.

± ± ± ±

0.3 0.2 0.1 0.2

2.0 2.0 1.7 2.2 ns ns ns

± ± ± ±

0.2 0.2 0.3 0.5

6.9 6.5 7.3 7.6 * ns ns

± ± ± ±

0.5 0.6 0.2 0.7

1.9 1.3 1.4 1.4 ns ns ns

± ± ± ±

0.4 0.2 0.1 0.3

0.28 0.20 0.19 0.18 * ns ns

± ± ± ±

0.07 0.03 0.01 0.02

88

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94

25

Table 5 L-values of ryegrass based on specific activity in shoots as affected by P fertilization with sewage sludge ash (þP) or inoculation with P. protegens CHA0 (Inoc). Mean ± standard deviation.

Specific activity SAshoot (Bq μg-1 P)

0P 0P Inoc

20

15

+P

L-value

+P Inoc

Cut 1

Cut 2

Cut 3

mg P kg1 soil 0P 0P Inoc þP þP Inoc Significance level (p) P fertilizer (P) Inoculant (I) PI

10

5

P I P×I

0 0

*** ns ns

*** * * 20

40

*** * ns 60

80

76 68 79 79

± ± ± ±

4 3 1 3

80 80 90 86

*** * *

± ± ± ±

4 3 4 1

91 84 97 97

*** ns ns

± ± ± ±

3 2 3 2

*** * ns

ns: not significant, *p < 0.05, **p < 0.01, ***p < 0.001.

days Fig. 3. Specific activity (SA) in shoots of ryegrass as affected by soil amendment P fertilizer with sewage sludge ash or inoculation with P. protegens CHA0. SAshoot (Bq mg1 P) of cut 1 (29 days after sowing), cut 2 (48 days after sowing) and cut 3 (69 days after sowing) of non-P fertilized (0P), non-P fertilized inoculated (0P Inoc), P fertilized (þP) and P fertilized inoculated (þP Inoc) ryegrass. Significance levels of two-factorial ANOVA with factors P fertilizer (P), inoculant (I) and the interaction P fertilizer  inoculant (P  I) per time point are indicated.

Table 6 Isotopically exchangeable P in unfertilized (0P) and P fertilized (þP) soil with sewage sludge ash determined in batch experiments. Number of replicates (N), P in solution (Pw, mg P kg1 soil), isotopic exchange kinetics parameters n and m and isotopically exchangeable P after 1 min (E1min, mg P kg1 soil) in unfertilized (0P) and P fertilized (þP) soils. Parameters m and n derived from short-term isotopic exchange kinetics (IEK) batch experiments and E1min by extrapolation using the non-linear equation (equation (8)). Means ± standard deviation. Treatment N Pw

18

P uptake (mg P kg-1 soil)

16 14 12

0P 7 0.53 ± 0.03 þP 4 0.65 ± 0.01 Significance level (p) P fertilizer (P) **

0P

0P Inoc +P +P Inoc

8

4

n

E1min mg P kg1 soil

0.195 ± 0.013 0.317 ± 0.002 2.8 ± 0.3 0.204 ± 0.002 0.310 ± 0.004 3.2 ± 0.1 ns

***

**

ns: not significant, *p < 0.05, **p < 0.01, ***p < 0.001.

10

6

m

mg P kg1 soil

B A B A

2 0 29 DAS

48 DAS

69 DAS

cut 1

cut 2

cut 3

total

Fig. 4. P uptake (mg P kg1 soil) of ryegrass as affected by soil amendment P fertilizer with sewage sludge ash or inoculation with P. protegens CHA0. P uptake at cut 1, cut 2, cut 3 and in total of non-P fertilized (0P), non-P fertilized inoculated (0P Inoc), P fertilized (þP) and P fertilized inoculated (þP Inoc) ryegrass. DAS; days after sowing. Error bars represent standard deviation of four replicates. Different capital letters show significant factor effects (p < 0.05).

(Table 7). Thereafter, mobilization by the soil endogenous microbial biomass was not any more detectable. The daily gross P mobilization was similar for the 0P and þP soils, i.e., not affected by P fertilization, except at day 1 when gross P mineralization rate was greater in the þP than in the 0P soil (p < 0.001). Inoculation decreased the measured E-values (Fig. 5) and in consequence gross P mobilization compared to non-inoculated treatments (Table 7). The observed effects of inoculation were not as consistent as the P fertilizer effect, mainly due to a significant negative interaction of inoculant with P fertilizer at day 1 and 27. The lower measured Evalues in inoculated than non-inoculated soils resulted mostly from reduced Pw and from increased SAPw in the inoculated soils (Fig. 6).

3.5.3. Microbial P immobilization Daily microbial P immobilization rate decreased from a maximum of 3 mg P kg1 soil day1 at the beginning to 0.2 mg P kg1 soil day1 at the end of the experiment (Table 7). Microbial P was on average about 20 mg P kg1 soil and was overall not affected by P fertilization or inoculation, except towards the end of incubation when it was significantly greater in inoculated than in non-inoculated soils (Fig. 6). The SAPmic was affected neither by inoculation nor by P fertilization at any sampling time (Fig. 6), as so the recovery of introduced activity in Pmic was constant over time (Fig. S 1). Across all treatments, resin extractable P averaged 8 mg P kg1 soil and fluctuated little during incubation, with lowest average values of 7 mg P kg1 soil on day 27 (Fig. 6). Fertilization with P significantly increased Pres throughout the experiment in average by 1.3 ± 0.3 mg P kg1 soil. No clear trend of inoculation on Pres was detectable and if so, differences to non-inoculated soil were small and did not affect P immobilization rates. More consistent effects were observed for SAPres. As expected, P amendment decreased SAPres significantly at every time point (Fig. 6). On the other hand, inoculation increased SAPres significantly (p < 0.05) after the first week till the end of the experiment, although average difference to non-inoculated soils of 34 Bq mg1 P was little and did not affect microbial P immobilization rates. 3.5.4. Sequential extraction The NaOH-P and HCl-P pools were not affected by inoculation, while P fertilization increased HCl-P (Table 8). Total NaOH-Porg decreased from 33 to 24 mg P kg1 soil and NaOH-Pi from 4.0 to 2.7 mg P kg1 soil during the time of incubation. These pools were neither affected by inoculation nor by P fertilization, except the NaOH-Porg at day 1. At this day, NaOH-Porg

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94

120

E value (mg P kg-1 soil)

100

*** *** ** *** ** ns

* ** ns

** ** **

*** *** *

ns ** ns

*** * *

89

P I P×I

80 60 40 20 0 0

20

40 days

measured 0P measured 0P Inoc

measured +P measured +P Inoc

extrapolated 95% confidence interval

extrapolated 95% confidence interval

60

80

0

20

40 days

60

80

Fig. 5. Measured E-values during incubation experiment against extrapolated E-values from isotopic exchange kinetics (IEK) batch experiments for the non-P fertilized (left; 0P, 0P Inoc) and the P fertilized (right; þP, þP Inoc) treatments, while Inoc stands for inoculation with P. protegens CHA0. Error bars represent standard deviation of measured E-values of four replicates. Significance levels of two-factorial ANOVA with factors P fertilizer (P), inoculant (I) and the interaction P fertilizer  inoculant (P  I) per time point for measured Evalues are indicated only in the left figure. Extrapolated E-values and their two sided 95% confidence intervals are shown for the non P fertilized (left) and the P fertilized soil (right) as inoculation did not affect extrapolated E-values.

was greater in the inoculated than in the non-inoculated treatments by 2 mg P kg1 soil (p < 0.05). Most of the 33P in the NaOH extract was recovered in the inorganic P fraction. In the P fertilized treatments, the SANaOH-Pi was significantly decreased by 12% between day 13 and day 40 compared to unfertilized soil (data not shown). The HCl-P pool was stable over time and greater in the fertilized treatments by about 46 mg P kg1 soil compared to unfertilized soil from the second week of incubation onwards. Accordingly, P fertilization diluted SAHCl significantly during this time. Still, the difference in SAHCl between non-inoculated (28 ± 1.1 Bq mg1 P) and inoculated soils (31 ± 0.8 Bq mg1 P) was small. 4. Discussion We tested the effect of inoculation with Pseudomonas protegens CHA0 on the availability of P from soil and from the recycling P fertilizer SSA in an incubation and plant growth study by applying 33 P isotopic dilution. Although strain CHA0 was able to solubilize SSA-P in the growth media (Table 3), we found no increase in available P neither from soil nor from fertilizer P which would have resulted in a dilution of the SA of P contained in the soil solution or of P taken up by the plant. 4.1. Effect of inoculation on P availability to plants from soil and fertilizer The soil used in this study was P limiting for crop growth as water extractable Pw of 0.5 mg P kg1 and E1min of 2.8 mg P kg1 soil were below the critical values of 1 mg P kg1 soil (Frossard et al., 2004) and 5 mg P kg1 soil (Gallet et al., 2003), respectively. The P added with the recycling fertilizer did not increase total P uptake by ryegrass, confirming that P contained in the thermo-chemically treated SSA is not effective in alkaline soils (Nanzer et al. (2014a)). This is further supported by Pw of 0.6 mg P kg1 and E1min of the SSA amended soil of 3.2 mg P kg1 (Table 6, Fig. 6). Nevertheless, the isotopic dilution in shoots of the plants fertilized with SSA revealed that 0.6 and 0.3 mg P kg1 soil in shoots of cut 2 and 3, respectively, were derived from SSA (calculated using equation (6)). That SSA affected the P nutrition of ryegrass was also

suggested by the decrease of the root to shoot ratio from 0.24 to 0.19 (Table 4). A decrease in root to shoot ratio from P limiting to P sufficient conditions has been reported (Fageria and Moreira, 2011). Thus, under conditions where soil P availability limits plants growth, we expected that the inoculant would solubilize non-plant available soil or fertilizer P, which would lead to a higher total P uptake of ryegrass and/or a dilution in SAshoots compared to noninoculated plants, i.e., a greater L-value in inoculated soils. This expectation was not confirmed by our results. Even though inoculation increased P uptake by ryegrass in the first cut by 0.5 mg P kg1 soil, the L-values of the inoculated treatments were never higher at any harvest than for non-inoculated treatments. Likewise, the SAPw of inoculated soils was not smaller than SAPw of noninoculated soils. Our results thus suggest that inoculation with strain CHA0 caused no significant P mobilization in addition to that caused by the endogenous microbial biomass. 4.2. Gross P fluxes caused by the endogenous microbial biomass in incubated soil The E-values obtained in the incubation study were significantly greater than the E-values extrapolated from the IEK batch experiments at the beginning of the incubation period (Fig. 5). This indicates that the endogenous microbial biomass mobilized P, although the increase over the extrapolated E-values resulting from physico-chemical processes alone was small. Measured daily gross P mobilization rates of 1.2 mg P kg1 soil day1after one week of incubation (Table 7) were similar to rates reported for an organically managed arable soil incubated for the same time (Oehl et al., 2004). Afterwards, daily rates decreased and were not any more detectable from day 40 on, either due to uncertainties in base line estimation in P deficient soils (Oehl et al., 2001b) or due to a decline in microbial activity, which is indicated by declining respiration rates (Fig. 2). Microbial P immobilization was similar to immobilization rates reported by Bünemann (2015) for a range of arable soils. Microorganisms incorporated 12% of introduced label after 1 day of incubation (Fig. S 1). This proportion remained unchanged thereafter in the microbial P pool throughout the incubation period (Fig. 6). Fast incorporation of 33P together with a constant recovery of the label

90

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94

0.7

0.5

0.4

0P 0P Inoc +P +P Inoc

0.3

0.2 *** 0.1 *** *** *** ** ** ** ns ns ** ns ns 0.0 0 20

*** ** ns

*** ns ns

40 days

*** * ns 60

12

6 4

*** *** *** ns ns ns ns ** ns

0

0

** ns ns 20

*** ** **

** P ns I ns P×I

*** *** ns

40 days

800 600 400 200 *** *** * *** ** *** ** ** ** ns ns ** 0 0 20

40 days

*** P * I * P×I 60

80

60

0P 0P Inoc +P +P Inoc

1000 800 600 400 200 0

** ** ** ns * * ns ns ns 0

80

*** * ns 20

* ** ns

* P * I ns P×I

*** ns ns

40 days

60

80

300

20

15 0P 0P Inoc +P +P Inoc

10

5 ns ns ns 0

ns ns ns

ns ns ns 20

ns ns ns 40 days

ns *** ns

ns *** * 60

P I P×I 80

Specific activity SAPmic (Bq μg-1 P)

Microbial P Pmic (mg P kg-1 soil)

ns ** ns

1200

25

0

*** *** ns

1400

8

2

0P 0P Inoc +P +P Inoc

1000

80 0P 0P Inoc +P +P Inoc

10 Resin P Pres (mg P kg-1 soil)

P I P×I

Specific activity SAPres (Bq μg-1 P)

P in solution Pw (mg P kg-1 soil)

0.6

specific activity SAPw (Bq μg-1 P)

1200

250 200 150 0P 0P Inoc

100

+P +P Inoc

50 0

ns ns ns 0

* ns ns

ns ns ns 20

ns ns ns 40 days

ns ns ns

* P ns I ns P×I 60

80

Fig. 6. Concentrations of water extractable P (Pw), resin extractable P (Pres) and microbial P (Pmic) in mg P kg1 soil and the respective specific activities (SA, Bq mg1 soil) in nonfertilized (0P), inoculated non-fertilized (0P Inoc), P fertilized with sewage sludge ash (þP) and inoculated P fertilized (þP Inoc) soils during time of incubation (day 1 till day 68). Error bars represent standard deviation. Significance levels of two-factorial ANOVA with factors P fertilizer (P), inoculant (I) and the interaction P fertilizer  inoculant (P  I) per time point are indicated.

in Pmic during the entire incubation time was also found in studies of Oehl et al. (2001a) in a non-fertilized Swiss arable soil, and by Oberson et al. (2001) and by Bünemann et al. (2004c) in highly weathered tropical soils. Fast incorporation followed by a constant

recovery of 33P in Pmic is likely caused by the labeling procedure. During soil mixing substrates become available, which in turn trigger microbial activity in nutrient limited soil (De Nobili et al., 2001). It has been supposed that under such conditions

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94

91

Table 7 Microbial gross P fluxes in calcareous soil as affected by P fertilization with sewage sludge ash or inoculation with P. protegens CHA0 at different times of the 27 days lasting incubation period. Extrapolated E-values (mg P kg-1 soil) based on isotopic exchange kinetics determined in batch experiments of unfertilized (0P) and P fertilized (þP) soils, measured E-values (mg P kg1 soil) in incubated soils, microbial gross P mobilization calculated as difference between measured and extrapolated E-value and microbial P immobilization over given incubation time (mg P kg1 soil) and per day (mg P kg1 soil day1) in unfertilized (0P) unfertilized inoculated (0P Inoc), P fertilized (þP) and P fertilized inoculated (þP Inoc) incubated soil. Mean ± standard deviation.

0P 0P Inoc þP þP Inoc 0P 0P Inoc þP þP Inoc 0P 0P Inoc þP þP Inoc 0P 0P Inoc þP þP Inoc

Time

Extrapolated E-value

Measured E-valuez

Microbial gross P mobilizationzz

Microbial P immobilizationzz

days

mg P kg1 soil

mg P kg1 soil

mg P kg1 soil

mg P kg1 soil day1

mg P kg1 soil

mg P kg1 soil day1

1

24.3 ± 2.1

25.2 25.5 29.5 26.7 51.4 47.9 56.4 52.5 62.8 59.7 68.4 62.0 72.6 72.4 81.3 73.4

0.9 ± 0.7 a 1.2 ± 0.6 a 3.0 ± 0.3 a 0.2 ± 0.7 b 6.9 ± 1.4 A 3.4 ± 0.7 B 8.6 ± 1.1 A 4.7 ± 0.2 B 7.4 ± 2.1 A 4.3 ± 3.2 B 9.1 ± 0.8 A 2.6 ± 3.4 B 4.4 ± 1.2 ab 4.3 ± 2.0 ab 8.6 ± 2.7 a 0.7 ± 1.4 b

1.2 ± 0.9 a 1.5 ± 0.8 a 4.0 ± 0.5 b 0.3 ± 0.9 a 1.2 ± 0.23 AB 0.6 ± 0.11 C 1.5 ± 0.20 A 0.8 ± 0.04 BC 0.58 ± 0.16 A 0.34 ± 0.25 B 0.72 ± 0.06 A 0.21 ± 0.27 B 0.17 ± 0.05 ab 0.16 ± 0.07 ab 0.32 ± 0.10 a 0.03 ± 0.05 b

2.8 ± 2.2 ± 3.1 ± 2.7 ± n.a. n.a n.a n.a 8.0 ± 8.5 ± 7.9 ± 7.7 ± 7.5 ± 8.0 ± 7.6 ± 6.9 ±

4.2 ± 0.7 2.9 ± 1.7 4.1 ± 0.5 3.6 ± 1.0 n.a n.a n.a n.a 0.64 ± 0.02 0.68 ± 0.09 0.62 ± 0.04 0.61 ± 0.07 0.28 ± 0.02 0.30 ± 0.07 0.28 ± 0.03 0.26 ± 0.02

26.5 ± 0.5 44.5 ± 3.5

6

47.8 ± 1.2 55.4 ± 4.2

13

59.3 ± 1.7 68.1 ± 4.9

27

72.7 ± 2.2

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.7 0.6 0.3*** 0.7 1.4** 0.7# 1.1*** 0.2** 2.1* 3.2 0.8* 3.4 1.2# 2.0 2.7* 1.4

0.8 1.3 0.4 0.8

0.3 1.2 0.5 0.8 0.6 1.9 0.7 0.6

na. not available. Microbial gross P mobilization rates derived from non-significant differences between extrapolated and measured E-values are subjected to uncertainties. z Symbols showing significant differences between extrapolated and measured E-value per sampling time; #p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001. zz Different capital letters per time point within columns show significant factor effects (p < 0.05), while different lower case letters show significant P fertilizer  inoculant interaction effects (p < 0.05).

Table 8 Sequential extraction of incubated calcareous soil as affected by P fertilization with sewage sludge ash or inoculation with P. protegens CHA0 at the beginning and at day 40 of the incubation period. Organic (Porg) and inorganic (Pi) extractable NaOH-P and their specific activities (SA) and total (Ptot) extractable HCl-P and SAHCl-P in unfertilized (0P) unfertilized inoculated (0P Inoc), P fertilized (þP) and P fertilized inoculated (þP Inoc) incubated soil. Means ± standard deviation.

0P 0P Inoc þP þP Inoc 0P 0P Inoc þP þP Inoc

Time

NaOH-Pz Porg

SAPorg

Pi

SAPi

Ptot

SAPtot

days

mg P kg1 soil

Bq mg1 P

mg P kg1 soil

Bq mg1 P

mg P kg1 soil

Bq mg1 P

1

31.5 34.3 31.9 33.6 24.0 26.6 25.3 24.3

26.8 ± 2.6 24.7 ± 1.2 28.3 ± 4.5 25.4 ± 2.3 3.7 ± 1.6 6.3 ± 2.4 9.3 ± 3.5 5.9 ± 0.3

3.9 3.8 3.9 4.3 2.5 3.0 3.1 3.1

40

± ± ± ± ± ± ± ±

1.1 1.3 1.8 1.7 1.0 1.7 1.7 1.9

HCl-Pz

A B A B

± ± ± ± ± ± ± ±

0.6 0.6 0.4 0.5 0.4 0.3 0.1 0.2

136 114 134 120 518 445 408 501

± ± ± ± ± ± ± ±

6A 5A 4B 2B 109 25 14 21

411 407 432 439 406 407 467 478

± ± ± ± ± ± ± ±

23 65 19 14 16 17 15 51

A A B B

20.5 21.4 18.7 20.3 33.0 33.4 28.0 28.8

± ± ± ± ± ± ± ±

1.5 3.8 0.9 1.5 3.5 1.1 0.7 2.5

A A B B

z Different capital letters per time point within columns show significant factor effects (p < 0.05) while different lower case letters show significant P fertilizer  inoculant interaction effects (p < 0.05).

microorganisms return into a resting state soon after exposure to small amounts of nutrients, i.e., mainly C (Bünemann et al., 2004a). The strongest decrease in daily respiration rates of 52% was measured between 2 days and one week after soil labeling (Fig. 2). Thereafter, respiration rates declined slowly by 10% per week on average. Under such conditions, efficient internal cycling of 33P within the microbial biomass, with minimal fluxes into other soil P pools, is assumed (Bünemann et al., 2004a). 4.3. Gross P fluxes as affected by soil amendment with sewage sludge ash in incubated soil Addition of P with SSA neither affected gross P mobilization rates nor microbial P immobilization rates as revealed by 33P isotopic dilution. Spohn and Kuzyakov (2013) likewise reported that inorganic P addition to the soil without organic C does not increase microbial P mobilization. This is supported by similar daily

respiration rates in P fertilized and non-fertilized incubated soil (Fig. 2). Although less than 1% of the P added with SSA was water soluble (Table 2), a dilution of SAPw and of SAshoot was measureable, highlighting the sensitivity of the 33P isotopic dilution approach to detect small changes in SA. The SAPw at day 1 in the incubated soil revealed that 14% of P in the Pw pool of soil fertilized with SSA derived from the added P, i.e., 0.09 mg P kg1 soil. Plants derived 11% and 6% of P in shoots of second and third cut, respectively, from the SSA, equal to 0.6 and 0.3 mg P kg1 soil (equation (6)). According to the L-value of P fertilized plants, 18% of SSA-P was exchangeable and therefore potentially plant available. 4.4. Gross P fluxes affected by inoculation of soil with P. protegens CHA0 Gross P mobilization rates were not increased by inoculation.

92

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94

The measured E-values during the incubation for inoculated soils were lower, or similar to those of non-inoculated soils at every time point (Fig. 5). Since the extrapolated E-values were identical for inoculated and non-inoculated soils, this shows that strain CHA0 caused no additional dilution in SAPw over that caused by the endogenous microbial biomass (Fig. 6). The SAPw was on average even 10% higher in inoculated than non-inoculated soils (Fig. 6). This implies that either less P was mobilized or that inorganic P was solubilized from P pools having a higher SA than the Pw pool. However, no P pool assessed by sequential fractionation had a higher SA than the Pw pool (Fig. 6, Table 8). Higher SAPw derived therefore from lower Pw values in the inoculated than noninoculated soils. Increased SA after inoculation was also observed in shoots of non-P fertilized ryegrass in cut 1 and independent of P fertilization in cut 3. Inoculation reduced gross P mobilization most strongly between week two and three by 53% on average compared to the noninoculated soils (Table 7). At the same time, microbial P immobilization was not affected by the inoculant. Lower gross P mobilization and reduced daily respiration rates in inoculated soils suggest inhibition of the endogenous microbial activity by strain CHA0, because respired CO2 mg1 Pmic declined while the P mic remained the same. The inoculant P. protegens CHA0 is known to suppress plant pathogens in soil by several antimicrobial compounds, such as 2,4-diacetylphloroglucinol and pyoluteorin (Keel et al., 1992; De Werra et al., 2009), HCN and pyrrolnitrin (Haas and Defago, 2005). The release of an antimicrobial compound is likely to occur under high pressure, i.e., microbial diversity (Jousset et al., 2013) and predator-prey competition (Pedersen et al., 2009). It has been shown, that pyrrolnitrin inhibits the fungal respiratory chains without causing cell lysis. At a concentration of 2 mg pyrrolnitrin ml1 respiration is reduced by 21% already after 10 min of incubation (Tripathi and Gottlieb, 1969). As the inoculation reduced the respiration rates most during the first 23 days, an effect of antimicrobial compound production by strain CHA0 seems therefore the most probable reason. The slight increase in organic P in the NaOH-extractable P pool of inoculated soils could therefore origin from some lysed cells of microbes affected by exudates of strain CHA0. The suppression of gross P mobilization by the endogenous microbial biomass due to inoculation resulted in non-detectable P mobilization from day 13 on, suggesting that strain CHA0 did not mobilize P from non-plant available P pools. 4.5. No P solubilization by P. protegens CHA0 in incubation and plant growth experiment Even though the inoculant P. protegens CHA0 was able to solubilize about 40% of P contained in the added SSA and TCP at a glucose concentration of 5 g L1 (Table 3), no additional dilution in SAPw and SAshoot of inoculated and SSA-fertilized soil and plant was found. The P solubilization assay has shown that strain CHA0 requires glucose in order to solubilize inorganic P. However, it is unlikely that glucose availability impaired the production of gluconic acid in the rhizosphere. Plants exude about 10e30% of assimilated C of which about 50% are carbohydrates, depending on plant species and age, and about half of the carbohydrates are composed by glucose (Hütsch et al., 2002). In the pot experiment, the ryegrass was inoculated with strain CHA0 at germination stage. Thus, glucose availability to strain CHA0 should not have been a constraint for the production of gluconic acid. However, nutrient competition between strain CHA0 and the endogenous microbial community may have favored the production of antimicrobial compounds rather than gluconic acid by strain CHA0 (Jousset et al., 2013). Due to suppressed respiration in incubated soil we can

assume that a glucose-like source was not limiting for gluconic acid production, even in the non-planted bulk soil. The plating assays with suspensions of inoculated soils suggest that population of P. protegens CHA0 decreased down to about 6  104 CFU g1 soil after 40 days, which is a population density commonly observed for this strain 1e2 months after inoculation to soils (Troxler et al., 2012). By the end of the experiments, strain CHA0 was probably outcompeted by resident soil bacteria since no cultivable P. protegens CHA0 cells were detected anymore in the colony plating assay. This matches with the evolution of daily respiration rates of the inoculated soils which were reduced by about 11% during the first 23 days of incubation (Table 1) and thereafter no effect of inoculant was anymore measurable. Low survival rate of Pseudomonas fluorescens following soil inoculation with cell suspension has also been reported by Van Dyke and Prosser (2000). 4.6. The sensitivity of

33

P dilution

Assessing the isotopic dilution in the plant and soil solution allowed detecting small differences of 0.09 mg P kg1 soil, as shown for SSA-amended treatments (discussed above). We can argue that an inoculant would have needed to increase measured E-values and L-value at least to the levels measured in P amended soil with SSA in order to cause a significant dilution in SAPw or SAshoots. Using isotopic dilution, we could as well estimate the gross P mobilization rates by the endogenous soil microbial activity. This estimate is based on the difference between E-values derived from SA measured during the incubation and E-values extrapolated from the isotopic exchange kinetics determined in batch experiments (equations (8) and (10)). The extrapolation is affected by the fitting procedure (Bünemann et al., 2007), mainly parameter n (Frossard et al., 1996) and by the accuracy of measured variables (Cp and Pi) entering equation (8) (Bühler et al., 2003; Frossard et al., 2011; Randriamanantsoa et al., 2013). The Cp values of our soils were with 50 (non-fertilized) and 61 mg P L1 (P fertilized) above the limit of quantification (Randriamanantsoa et al., 2013) and could be determined with a low coefficient of variation (1e3%) at every sampling time. Since the extrapolated E-values were not different from the L-values (data not shown), we conclude that the fitting procedure was appropriate. The endogenous soil microbial activity increased the measured E-values only slightly (though significantly) over the extrapolated E-values. Thus, differences between measured and extrapolated E-values were small. Through the described suppression of the soil microbial biomass, the inoculant had a counteracting effect and has in turn not increased this difference. 5. Conclusion The 33P isotopic dilution approach allowed differentiating between physico-chemical and biological P processes and effects of the inoculant on gross P fluxes were detectable. The P mobilization was not enhanced by P. protegens CHA0 neither in the incubation nor in the pot experiment. In contrast, the reduced respiration rates in inoculated soils during the first weeks of incubation suggested that strain CHA0 counteracted high microbial pressure with the production of antimicrobial compounds. This in turn lowered the microbial P mobilization by the endogenous microbial biomass. Cell culturability towards the end of the incubation experiment indicated that CHA0 had been outcompeted. The greater P uptake by inoculated ryegrass at the first cut could not be assigned to P mobilization from non-plant available P, but was probably due to an indirect effect of the inoculant. This shows the trade-off between beneficial antimicrobial effects against plant pathogens and

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94

inhibition effects on beneficial microorganisms such as soil endogenous P mobilizers. The complex interaction of a PSB with the endogenous microbial community showed clearly the limitation of in vitro P solubilization assays. Acknowledgement This project was largely funded by the CORE Organic II Funding Bodies, being partners of the FP7 ERA-Net project, CORE Organic II (Coordination of European Transnational Research in Organic Food and Farming systems, project no. 249667). We also thank Iris Huber for the help with the plant growth experiment and the two anonymous reviewers for their constructive comments. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.soilbio.2016.10.001. References Achat, D.L., Bakker, M.R., Saur, E., Pellerin, S., Augusto, L., Morel, C., 2010. Quantifying gross mineralisation of P in dead soil organic matter: testing an isotopic dilution method. Geoderma 158, 163e172. Adam, C., Peplinski, B., Michaelis, M., Kley, G., Simon, F.G., 2009. Thermochemical treatment of sewage sludge ashes for phosphorus recovery. Waste Manag. 29, 1122e1128. Agroscope, F., 2004. Referenzmethoden der Forschungsanstalten Agroscope. Forschungsanstalt Agroscope Reckenholz-T€ anikon (ART) and Changins-W€ adenswil €denswil Vol. 1e4. (ACW), Zurich, Changings, Wa Alef, K., 1995. Soil respiration. In: Alef, K., Nannipieri, P. (Eds.), Methods in applied soil microbiology and biochemistry. Academic Press, London, pp. 214e219. Asea, P.E.A., Kucey, R.M.N., Stewart, J.W.B., 1988. Inorganic phosphate solubilization by two Penicillium species in solution culture and soil. Soil Biol. Biochem. 20, 459e464. n, R., 2007. The use of 32P isotopic dilution techniques to Barea, J.M., Toro, M., Azco evaluate the interactive effects of phosphate-solubilizing bacteria and mycorrhizal fungi at increasing plant P availability. In: Vel azquez, E., RodríguezBarrueco, C. (Eds.), First International Meeting on Microbial Phosphate Solubilization. Springer Netherlands, pp. 223e227. Box, G.E.P., Cox, D.R., 1964. An analysis of transformations. J. Royal Statistical Soc. Ser. B (Methodological) 26, 211e252. Brod, E., Øgaard, A., Haraldsen, T., Krogstad, T., 2015. Waste products as alternative phosphorus fertilisers part II: predicting P fertilisation effects by chemical extraction. Nutr. Cycl. Agroecosyst. 103, 187e199. Brookes, P.C., Powlson, D.S., Jenkinson, D.S., 1982. Measurement of microbial biomass phosphorus in soil. Soil Biol. Biochem. 14, 319e329. Bühler, S., Oberson, A., Sinaj, S., Friesen, D.K., Frossard, E., 2003. Isotope methods for assessing plant available phosphorus in acid tropical soils. Eur. J. Soil Sci. 54, 605e616. Bünemann, E.K., 2015. Assessment of gross and net mineralization rates of soil organic phosphorus e a review. Soil Biol. Biochem. 89, 82e98. Bünemann, E.K., Augstburger, S., Frossard, E., 2016. Dominance of either physicochemical or biological phosphorus cycling processes in temperate forest soils of contrasting phosphate availability. Soil Biol. Biochem. 101, 85e95. Bünemann, E.K., Bossio, D.A., Smithson, P.C., Frossard, E., Oberson, A., 2004a. Microbial community composition and substrate use in a highly weathered soil as affected by crop rotation and P fertilization. Soil Biol. Biochem. 36, 889e901. Bünemann, E.K., Marschner, P., McNeill, A.M., McLaughlin, M.J., 2007. Measuring rates of gross and net mineralisation of organic phosphorus in soils. Soil Biol. Biochem. 39, 900e913. Bünemann, E.K., Oberson, A., Liebisch, F., Keller, F., Annaheim, K.E., HugueninElie, O., Frossard, E., 2012. Rapid microbial phosphorus immobilization dominates gross phosphorus fluxes in a grassland soil with low inorganic phosphorus availability. Soil Biol. Biochem. 51, 84e95. Bünemann, E.K., Smithson, P.C., Jama, B., Frossard, E., Oberson, A., 2004b. Maize productivity and nutrient dynamics in maize-fallow rotations in western Kenya. Plant Soil 264, 195e208. Bünemann, E.K., Steinebrunner, F., Smithson, P.C., Frossard, E., Oberson, A., 2004c. Phosphorus dynamics in a highly weathered soil as revealed by isotopic labeling techniques. Soil Sci. Soc. Am. J. 68, 1645e1655. Cornish, P.S., 2009. Research directions: improving plant uptake of soil phosphorus, and reducing dependency on input of phosphorus fertiliser. Crop Pasture Sci. 60, 190e196. De Nobili, M., Contin, M., Mondini, C., Brookes, P.C., 2001. Soil microbial biomass is triggered into activity by trace amounts of substrate. Soil Biol. Biochem. 33, 1163e1170. chy-Tarr, M., Keel, C., Maurhofer, M., 2009. Role of gluconic acid De Werra, P., Pe

93

production in the regulation of biocontrol traits of Pseudomonas fluorescens CHA0. Appl. Environ. Microbiol. 75, 4162e4174. Fageria, N.K., Moreira, A., 2011. Chapter four - the role of mineral nutrition on root growth of crop plants. In: Donald, L.S. (Ed.), Advances in Agronomy. Academic Press, pp. 251e331. Fardeau, J.-C., Morel, C., Jahiel, M., 1988. Does long contact with the soil improve the efficiency of rock phosphate? Results of isotopic studies. Fertil. Res. 17, 3e19. tiques de transfert des ions phosphate Fardeau, J., Morel, C., Boniface, R., 1991. Cine tres caracte ristiques. Agronomie 11, du sol vers la solution du sol : parame 787e797. Frossard, E., Achat, D., Bernasconi, S., Bünemann, E., Fardeau, J.-C., Jansa, J., Morel, C., Rabeharisoa, L., Randriamanantsoa, L., Sinaj, S., Tamburini, F., Oberson, A., 2011. The use of tracers to investigate phosphate cycling in soileplant systems. In: Bünemann, E., Oberson, A., Frossard, E. (Eds.), Phosphorus in Action. Springer Berlin Heidelberg, pp. 59e91. Frossard, E., Brossard, M., Hedley, M.J., Metherell, A., 1995. Reactions controlling the cycling of P in soils. In: Tiessen, H. (Ed.), Phosphorus in the Global Environment. John Wiley & Sons, pp. 107e137. Frossard, E., Julien, P., Neyroud, J.-A., Sinaj, S., 2004. Phosphor in €dendStandortbestimmung Schweiz. Bundesamt für Umwelt, Wald und Bo Landschaft, Bern, 174. pez-Hern Frossard, E., Lo andez, D., Brossard, M., 1996. Can isotopic exchange kinetics give valuable information on the rate of mineralization of organic phosphorus in soils? Soil Biol. Biochem. 28, 857e864. Frossard, E., Morel, J.L., Fardeau, J.C., Brossard, M., 1994. Soil isotopically exchangeable phosphorus: a comparison between E and L values. Soil Sci. Soc. Am. J. 58, 846e851. Gallet, A., Flisch, R., Ryser, J.-P., Frossard, E., Sinaj, S., 2003. Effect of phosphate fertilization on crop yield and soil phosphorus status. J. Plant Nutr. Soil Sci. 166, 568e578. Goldstein, A., 2007. Future trends in research on microbial phosphate solubilization: zquez, E., Rodríguez-Barrueco, C. one hundred years of insolubility. In: Vela (Eds.), First International Meeting on Microbial Phosphate Solubilization. Springer Netherlands, pp. 91e96. Haas, D., Defago, G., 2005. Biological control of soil-borne pathogens by Fluorescent Pseudomonads. Nat. Rev. Micro 3, 307e319. Hedley, M.J., Stewart, J.W.B., Chauhan, B.S., 1982. Changes in inorganic and organic soil phosphorus fractions induced by cultivation practices and by laboratory incubations. Soil Sci. Soc. Am. J. 46, 970e976. Hütsch, B.W., Augustin, J., Merbach, W., 2002. Plant rhizodeposition d an important source for carbon turnover in soils. J. Plant Nutr. Soil Sci. 165, 397e407. Jousset, A., Becker, J., Chatterjee, S., Karlovsky, P., Scheu, S., Eisenhauer, N., 2013. Biodiversity and species identity shape the antifungal activity of bacterial communities. Ecology 95, 1184e1190. Keel, C., Schnider, U., Maurhofer, M., 1992. Suppression of root diseases by Pseudomonas fluorescens CHA0-Importance of the bacterial secondary metabolite 2,4-diacetylphloroglucinol. Mol. Plant-Microbe Interact. 5, 4e13. King, E.O., Ward, M.K., Raney, D.E., 1954. Two simple media for the demonstration of pyocyanin and fluorescin. J. Lab. Clin. Med. 44, 301e307. Kouno, K., Tuchiya, Y., Ando, T., 1995. Measurement of soil microbial biomass phosphorus by an anion exchange membrane method. Soil Biol. Biochem. 27, 1353e1357. Kucey, R.M.N., Janzen, H.H., Leggett, M.E., 1989. Microbially mediated increases in plant-available phosphorus. In: Brady, N.C. (Ed.), Advances in Agronomy. Academic Press, pp. 199e228. Larsen, S., 1952. The use of P32 in studies on the uptake of phosphorus by plants. Plant Soil 4, 1e10. McLaughlin, M., Alston, A., Martin, J., 1988. Phosphorus cycling in wheat pasture rotations.II. The role of the microbial biomass in phosphorus cycling. Soil Res. 26, 333e342. McLaughlin, M.J., Alston, A.M., Martin, J.K., 1986. Measurement of phosphorus in the soil microbial biomass: a modified procedure for field soils. Soil Biol. Biochem. 18, 437e443. Meyer, J.B., Frapolli, M., Keel, C., Maurhofer, M., 2011. Pyrroloquinoline quinone biosynthesis gene pqqc, a novel molecular marker for studying the phylogeny and diversity of phosphate-solubilizing pseudomonads. Appl. Environ. Microbiol. 77, 7345e7354. Molla, M.A.Z., Chowdhury, A.A., Islam, A., Hoque, S., 1984. Microbial mineralization of organic phosphate in soil. Plant Soil 78, 393e399. Morel, C., Fardeau, J.C., 1989. The uptake by crops of fresh and residual phosphatic fertilizers by simultaneous measurements with 32P and 33P. Int. J. Radiat. Appl. Instrum. Part A. Appl. Radiat. Isot. 40, 273e278. Nanzer, S., Oberson, A., Berger, L., Berset, E., Hermann, L., Frossard, E., 2014a. The plant availability of phosphorus from thermo-chemically treated sewage sludge ashes as studied by 33P labeling techniques. Plant Soil 1e18. Nanzer, S., Oberson, A., Huthwelker, T., Eggenberger, U., Frossard, E., 2014b. The molecular environment of phosphorus in sewage sludge ash: implications for bioavailability. J. Environ. Qual. 43, 1050e1060. Nautiyal, C.S., 1999. An efficient microbiological growth medium for screening phosphate solubilizing microorganisms. FEMS Microbiol. Lett. 170, 265e270. Oberson, A., Friesen, D.K., Rao, I.M., Bühler, S., Frossard, E., 2001. Phosphorus transformations in an oxisol under contrasting land-use systems: the role of the soil microbial biomass. Plant and Soil 237, 197e210. Oberson, A., Joner, E.J., 2005. Microbial Turnover of Phosphorus in Soil. CABI Publishing, Wallingford, pp. 133e164.

94

G. Meyer et al. / Soil Biology & Biochemistry 104 (2017) 81e94

€der, P., Frossard, E., 2010. Oberson, A., Tagmann, H., Langmeier, M., Dubois, D., Ma Fresh and residual phosphorus uptake by ryegrass from soils with different fertilization histories. Plant Soil 334, 391e407. Oehl, F., Frossard, E., Fliessbach, A., Dubois, D., Oberson, A., 2004. Basal organic phosphorus mineralization in soils under different farming systems. Soil Biol. Biochem. 36, 667e675. Oehl, F., Oberson, A., Probst, M., Fliessbach, A., Roth, H.-R., Frossard, E., 2001a. Kinetics of microbial phosphorus uptake in cultivated soils. Biol. Fertil. Soils 34, 31e41. Oehl, F., Oberson, A., Sinaj, S., Frossard, E., 2001b. Organic phosphorus mineralization studies using isotopic dilution techniques. Soil Sci. Soc. Am. J. 65, 780e787. Ohno, T., Zibilske, L.M., 1991. Determination of low concentrations of phosphorus in soil extracts using malachite green. Soil Sci. Soc. Am. J. 55, 892e895. Owen, D., Williams, A.P., Griffith, G.W., Withers, P.J.A., 2015. Use of commercial bioinoculants to increase agricultural production through improved phosphorous acquisition. Appl. Soil Ecol. 86, 41e54. Pedersen, A., Nybroe, O., Winding, A., Ekelund, F., Bjørnlund, L., 2009. Bacterial feeders, the nematode Caenorhabditis elegans and the Flagellate cercomonas longicauda, have different effects on outcome of competition among the Pseudomonas biocontrol strains CHA0 and DSS73. Microb. Ecol. 57, 501e509. R Core Team, 2014. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Randriamanantsoa, L., Frossard, E., Oberson, A., Bünemann, E.K., 2015. Gross organic phosphorus mineralization rates can be assessed in a Ferralsol using an isotopic dilution method. Geoderma 257e258, 86e93. Randriamanantsoa, L., Morel, C., Rabeharisoa, L., Douzet, J.-M., Jansa, J., Frossard, E., 2013. Can the isotopic exchange kinetic method be used in soils with a very low water extractable phosphate content and a high sorbing capacity for phosphate ions? Geoderma 200e201, 120e129. Richardson, A.E., Simpson, R.J., 2011. Soil microorganisms mediating phosphorus availability update on microbial phosphorus. Plant Physiol. 156, 989e996. Rodríguez, H., Fraga, R., 1999. Phosphate solubilizing bacteria and their role in plant

growth promotion. Biotechnol. Adv. 17, 319e339. Sambrook, J., Russell, D.W., 2001. Molecular Cloning: a Laboratory Manual, 3 ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Saunders, W.M.H., Williams, E.G., 1955. Observations on the determination of total organic phosphorus in soils. J. Soil Sci. 6, 254e267. Sibbesen, E., 1984. Determination of isotopically exchangeable P in soil (L-values) over several crop cuttings. J. Sci. Food Agric. 35, 731e732. Simpson, R., Oberson, A., Culvenor, R., Ryan, M., Veneklaas, E., Lambers, H., Lynch, J., Ryan, P., Delhaize, E., Smith, F.A., Smith, S., Harvey, P., Richardson, A., 2011. Strategies and agronomic interventions to improve the phosphorus-use efficiency of farming systems. Plant Soil 349, 89e120. Spohn, M., Kuzyakov, Y., 2013. Phosphorus mineralization can be driven by microbial need for carbon. Soil Biol. Biochem. 61, 69e75. fago, G., Kern, H., 1986. Naturally occurring Fluorescent Pseudomonads Stutz, E.W., De involved in suppression of black root rot of tobacco. Phytopathology 76, 181e185. Tiessen, H., Moir, J., 1993. Characterisation of available P by sequential extraction. In: Carter, M.R. (Ed.), Soil Sampling and Methods of Analysis. CRC press Inc, Boca Raton Florida USA, pp. 75e86. Tripathi, R.K., Gottlieb, D., 1969. Mechanism of action of the antifungal antibiotic pyrrolnitrin. J. Bacteriol. 100, 310e318. €nne-Loccoz, Y., De fago, G., Troxler, J., Svercel, M., Natsch, A., Zala, M., Keel, C., Moe 2012. Persistence of a biocontrol Pseudomonas inoculant as high populations of culturable and non-culturable cells in 200-cm-deep soil profiles. Soil Biol. Biochem. 44, 122e129. Truong, B., Pichot, J., 1976. Influence du phosphore des graines de la plante test sur termination du phosphore isotopiquement diluable. Agron. Trop. 31, la de 379e386. Van Dyke, M.I., Prosser, J.I., 2000. Enhanced survival of Pseudomonas fluorescens in soil following establishment of inoculum in a sterile soil carrier. Soil Biol. Biochem. 32, 1377e1382.