Please cite this paper as: Y. Wei, Y. Zhao, Y. Fan, Q. Lu, M. Li, Q. Wei, Y. Zhao, Z. Cao and Z.

Wei (2017). "Impact of phosphate-solubilizing bacteria inoculation methods on phosphorus

transformation and long-term utilization in composting." Bioresour Technol 241: 134-141.

Impact of phosphate-solubilizing bacteria inoculation methods on

phosphorus transformation and long-term utilization in composting

Yuquan Weia, Yue Zhaoa, Yuying Fana, Qian Lua, Mingxiao Lib, Qingbin Weic, Yi Zhaoa,

Zhenyu Caoa, Zimin Weia,*

a College of Life Science, Northeast Agricultural University, Harbin 150030, China

b
State Key Laboratory of Environment Criteria and Risk Assessment, Chinese Research Academy of

Environmental Sciences, Beijing 100012, China

c
Heilongjiang Province Environmental Monitoring Centre, Harbin 150056, China

Abstract: This study aimed to assess the effect of phosphate-solubilizing bacteria (PSB)

application and inoculation methods on rock phosphate (RP) solubilization and bacterial

community during composting. The results showed that PSB inoculation in different

stages of composting, especially both in the beginning and cooling stages, not only

improved the diversity and abundance of PSB and bacterial community, but also

distinctly increased the content of potential available phosphorus. Redundancy analysis

indicated that the combined inoculation of PSB in the initial stage with higher

inoculation amount and in the cooling stage with lower inoculation amount was the best

way to improve the inoculation effect and increase the solubilization and utilization of

RP during composting. Besides, we suggested three methods to improve phosphorus

transformation and long-term utilization efficiency in composts based on biological

1
fixation of phosphates by humic substance and phosphate-accumulating organisms.

Keywords: Composting, Different stages inoculation, Phosphate-solubilizing bacteria

(PSB), Denaturing gradient gel electrophoresis (DGGE), Potential available phosphorus

(PAP).

*Corresponding author

Tel/Fax: +86 451 55190413

E-mail: weizimin@neau.edu.cn or weizm691120@163.com

1. Introduction

In China, the improvement of living standards as a result of its rapid economic

development and population expansion means the growing output of municipal solid

waste (MSW), which affects the urban environment (Wang et al., 2015a). To achieve the

goals of resource conservation and reducing environmental and health impacts,

composting is a widely used method to effectively manage MSW though a biological

stabilization process (Wei et al., 2007). Microbes play key roles in all the events related

to the biotransformation of substrates during composting (Zhao et al., 2016a).

Insufficient quantity or limited biological activities of the indigenous microbial

communities could lead to undesirable composting efficiency. Numerous studies have

shown that inoculation or strengthening indigenous microorganisms may improve the

composting process (Wei et al., 2016a; Xi et al., 2015).

Phosphorus (P) is an essential element for plant growth and is widely applied as

inorganic fertilizer for agricultural purposes (Wei et al., 2015). Rock phosphate (RP) is

2
the main base raw material from which inorganic fertilizers can be made in large

quantities (Owen et al., 2015). However, considering that the reserves of RP are

relatively finite, it is crucial to enhance the availability of P and avoid potential risk for

P loss in runoff and leachate. A number of studies showed that most of the organic

wastes composting from different sources could influence the content and distrib ution

of P fractions (Wei et al., 2015; Hashimoto et al., 2014; Ngo et al., 2013), thus, some

organic wastes with low P content are always used for composting amended with RP for

increasing the quantity of organic matter and available P. Phosphate-solubilizing

bacteria (PSB) have fundamental roles in P cycling in natural and agricultural

ecosystems (Wei et al., 2016a). Given that composting ecosystem exhibited rich

bacterial diversity with different enzymatic capabilities but the complicated

environmental conditions, e.g., temperature fluctuations and limited nutrients, could

inhibit microbial activities, numerous studies have been conducted to apply inoculants

for adjusting the correlation between microbial communities with different biochemical

functions, such as, enhancing degradation of cellulose, improving humification and

reducing ammonia emission (Zhao et al., 2016b; Zhang et al., 2016; Xi et al., 2015;

Hachicha et al., 2012), while there is little information about the relationship among the

inoculation period of PSB inoculants, indigenous bacteria and P-solubilization

efficiency.

However, higher P-solubilization may lead to an increased potential risk for P loss in

runoff and leachate and be possibly linked to the activity of PSB with a negative

3
feedback (Wei et al., 2015; Mander et al., 2012). So, to improve the long-term fertility

of composts with RP, it is inevitable to discuss how to stabilize redundant labile P or

control P-solubilization process and P forms. Composting is also a controlled biological

process for rapid stabilization and humification of organic matter as an increase of

humic substances (Zhao et al., 2017; Wei et al., 2007). It is known that humic matter

can affect the solubility of different P-compounds in soils through its chelation capacity

or metal bridging, which is considered as a form of biological fixation of phosphates

(Borggaard et al., 2005). If considered together, the period of PSB inoculation,

P-solubilization process, and changes of indigenous microbial community as well as the

formation of humic substances may provide powerful new insights into the

transformation of P fractions during composting based on biological and

physicochemical method, which would truly enhance the efficiency of P use.

In this work, five MSW composting experiments were conducted with different PSB

or RP-addition disposals. The principal goals of this study were to (1) analyze the

microbial biomass and bacterial structure in different composting treatments, (2)

compare the influence of different inoculation methods on PSB activity and P fractions,

(3) characterize the interactions between inoculants, indigenous bacteria, P forms and

treatments, and (4) present a biological and physicochemical method for future studies

to better control the P transformation and utilization efficiency in composts.

2. Materials and methods

4
2.1. Composting experimental design

Composting materials were mainly residual MSWs with inorganic materials such as

metals, plastics, and glasses removed before composting, which were collected from

Northeast Agricultural University (Harbin, China). Straw chopped into small pieces

(20-30 mm) was used to adjust the initial C/N ratio of composting materials. The basic

chemical characterization of the raw materials were shown in Table 1. The ratio of water

content is approximately 60% and C/N ratio is about 25:1. Composting materials were

put in the special small cylinder compost reactor, which referenced to Zhao et al.

(2016b). The changes of reactor temperature see supplementary Fig. S1. The actual

temperature of compost was basically the same as reactor. During composting, moisture

was maintained at 50-60% and the ventilation rate was 0.5 L min-1. The initial and final

conditions of composting were shown in the Table S1.

Table 1 Basic characterization of the composting materials.

Resource materials Organic matter (%) C/N Moisture content (%) pH

municipal solid waste 49.1± 1.0 16.5± 0.9 57.6± 1.5 7.8± 0.2
straw 70.3± 2.1 53.6± 2.4 5.2± 0.1 6.5± 0.1

Five groups of composting experiments were carried out and all of the treatments

were replicated three times. MSW was chosen as the major raw material because it has a

relatively low P content (Wei et al., 2015) and therefore allowed the effects of the PSB

and RP on composting to be readily observed. The first group is blank control group

(CK). Nothing is added during composting. For the other four groups, RP was added

5
with 5%, which was obtained from KaiLin company, Guizhou, with 28.27% P2O5, 3.55%

Olsen P and 5.21% soluble P in 2% citric acid. The group with only RP addition was

prepared as negative control (CP). There are three PSB treatment groups, which are

inoculated in the initial stage (0d) as CMP1, in the cooling stage (8d) as CMP3 and in

the above two stages as CMP2, respectively. The inoculant of composting in this study

was a kind of compound PSB agent, which was described in a separate paper (Wei et al.,

2016b). Inoculant in each group was at the level of 1.5% in dry weight with the same

inoculation amount, and the concentrations of the composite PSB strains were 1×108

CFU mL-1. The whole time of the experiment in this work is 20 days and the samples

collected every two days. Some samples were stored at -20°C for DNA extraction and

others were used for physical-chemical analysis and P content, while those for microbial

quantification were freshly processed.

2.2. Microbiological enumeration and Molecular biological analyses

The cultivable bacteria and PSB were estimated using a standard dilution-plating

procedure as described by Wei et al. (2016a). Bacteria were cultured in standard

Nutrient Agar (Cultimed, Spain) for 48h at 30°C. PSB was cultivated in National

Botanical Research Institute’s phosphate growth medium devoid of yeast extract

(NBRIY) medium supplemented with 1.5% Bacto-agar (Difco Laboratories, Detroit, MI,

USA).

DNA was isolated for dominant microorganism using soil DNA kit (Omega Biotek,

6
Inc.). To analyze the DNA of the bacterial community, 16S rRNA genes were amplified

using the prokaryotic primers 341F/534R (Xi et al., 2015). A GC clamp was attached to

forward primers to prevent complete separation of the strands during DGGE.

Polymerase chain reaction (PCR) conditions for each 50-μL reaction mixture and touch

down PCR protocol were performed as described by Wei et al. (2016a). Each PCR

concoction was set in a polyacrylamide (8%) gels at a 35-60% denaturant gradient. A

Gene Mutation Detection System (Bio-Rad Laboratories, Inc.) was run at 150 V for 4 h

at 60°C to separate the fragments. After electrophoresis, gels were stained in

3×GelRed™ nuclear acid gel stain (Biotium, USA) and photographed with a UVP

Imaging System (UVP Inc., USA). This procedure was performed in triplicate for each

sample with different gels. Representatives of bands that were clear and had high

intensity were excised from DGGE gels and transferred to 30-μL Milli-Q water

(Millipore, USA), incubated overnight for elution of DNA at 4°C. Then, PCR was

performed to re-amplify the DNA using primers 534R and 341F without a GC-clamp.

After sequencing, the results were compared with the GenBank from the National

Center of Biotechnology Information (NCBI) using BLAST.

2.3. Identification of total P and P fractions during composting

The total phosphorus (TP) and P fractions were determined for compost samples

collected every 2 days. Extraction of TP, Olsen P, water soluble phosphorus (WSP) and

citric acid phosphorus (CAP) were performed following the procedure of Wei et al.

7
(2015). The P concentration in the different extracts was assayed colorimetrically using

the ascorbic acid/molybdate reagent blue color method (Murphy and Riley, 1962).

Olsen P, WSP and CAP, extracted readily and moderately soluble compounds, which

are relatively active P pool and available to plants and microorganisms immediately or

in a long time (Li et al., 2011; Turner and Leytem, 2004). However, considering that

WSP could also be extracted in the method of both Olsen P and CAP, potential available

P (PAP) during composting was estimated as: Olsen P + CAP - WSP. The remaining P

fractions that could be not available P (NAP) were calculated as the difference between

TP and PAP.

2.4. Statistical analysis

All physical-chemical data were analyzed using Origin pro 8.0 and SPSS 19.0. The

DGGE images were analyzed for the relative intensity data of DNA bands in the gel.

The Shannon-Wiener Diversity Index was determined using Quantity one software

(version 4.5, Bio-Rad laboratories, USA). Canoco (version 5.0) was used for

determination of multivariate relationships between bacterial community compositions

and different addition treatments as well as P fractions. The length of the first detrended

correspondence analysis ordination axis was 2.2 for bacterial species data. Therefore,

redundancy analysis (RDA) was performed to ordinate the compositions of bacterial

community to treatments and assess the relationship between P fractions and the

bacterial community. The inter-species distances for the DGGE profiles were analyzed

8
by forward manual selection and using 499 unrestricted Monte Carlo permutation tests

(Qiu et al., 2016). Statistical analyses were performed using Statistix 8 by the LSD

All-Pairwise Comparisons Test. The significant level of differences in the research was

set as p <0.05.

3. Results and Discussion

3.1. Overview of microbiological analyses during composting

The succession of the cultivable bacterial and PSB populations during MSW

composting is illustrated in Table 2. On the one hand, the changes of bacteria and PSB

populations were similar during MSW composting. The amount of bacteria and PSB

increased at the beginning of composting and reached a maximum in the thermophilic

phase, but decreased until the end of the trial, ranging from 4.5ｘ107 to 8.5ｘ108 CFU/g

for bacteria and from 1.25ｘ106 to 1.07ｘ107 CFU/g for PSB, respectively. On the other

hand, the counts of bacteria and PSB varied distinctly in different treatments. The

amounts of bacteria in CP was significantly lower than that in CK (p < 0.05) but the

amounts of PSB was significantly higher than in CK (p < 0.05) during composting, i.e.

notably increasing the proportion of PSB in bacteria (p < 0.05). The results suggested

that the addition of RP, mainly including high proportion acid-soluble P but hard to be

utilized directly for microbes (Wei et al., 2016b), may suppress bacterial growth but

activate potential PSB during composting, which may also depend on P status and

availability as the viewpoint of Mander et al. (2012). The abundance of bacterial and

9
 PSB at the end of composting in CMP1, CMP2 and CMP3 were higher than that in CK

and CP and that in CMP2 was the highest among the five groups, suggesting that

composite PSB may promote indigenous bacterial growth and adapt rapidly to

complicated composting environment, especially inoculating both in the initial stage

and the cooling stage.

Table 2. Cultivable bacteria and phosphate-solubilizing bacteria (cfu) isolated from

different compost samples in different composting periods.

Bacteria PSB
Composts Mean cfu g-1 (ｘ107) Mean cfu g-1 (ｘ106)
0day 4day 10day 20day 0day 4day 10day 20day
CK 12.2± 0.8 55.3± 2.4 19.1± 3.3 6.8± 0.5 2.9± 0.3 3.8± 0.4 1.9± 0.2 1.3± 0.1

CP 8.0± 0.4 36.3± 2.8 12.2± 1.1 4.5± 0.3 4.2± 0.6 6.1± 0.5 3.6± 0.2 2.5± 0.1

CMP1 21.8± 2.1 84.6± 5.2 42.0± 2.2 13.3± 0.8 5.8± 0.4 10.7± 0.8 5.6± 0.3 4.0± 0.1

CMP2 15.7± 2.9 67.3± 8.5 53.3± 4.5 19.8± 1.4 5.2± 0.4 8.9± 0.7 7.8± 0.4 6.6± 0.3

CMP3 10.6± 1.7 45.7± 3.5 29.3± 1.8 11.7± 0.4 3.9± 0.1 5.9± 0.2 5.6± 0.4 4.2± 0.5

The bacterial DGGE patterns of 16S rDNA gene fragments were shown in Fig. 1 and

Fig. S2. The bacterial communities exhibited a significant difference between the same

compost of different ages and varied among different treatments. Forty-six different

bands were detected in the DGGE profile, and 33% of them except the inoculated PSB

bands were ubiquitous during MSW composting. The inoculated PSB bands except

band 10 were all detected in CK and CP and most of them were classified as Phylum

Firmicutes (Table S2), which have a relatively high abundance and survival time during

10
composting. But three of the dominant PSB disappeared after the 10th day of

composting, i.e. band 33, 36 and 44, belonging to Proteobacteria and Actinobacteria,

which might be attributed to the exhaustion of nutrient (easily degradable organic

matter). As the comprehensive parameters for microbial diversity, the changes of bands

numbers and Shannon-Wiener index for the DGGE profile were evaluated and shown in

Fig. 1. The trends of Shannon-Wiener index in all piles were similar to the changes seen

in bands numbers, which increased to a maximum value on the 10th day after

composting then fell until the end of composting. The mean value among different

treatments, ranging from about 21 to 33 for bands numbers and from 2.99 to 3.45 for

Shannon-Wiener index, was higher than that of Wang et al. (2015b), which may be the

result of more complex composition in MSW and PSB inoculation. To better compare

the similarity of bacterial community present in CK, CP, CMP1, CMP2 and CMP3, a

hierarchical cluster analysis based on UPGAMA was conducted (Fig. S3). The bacterial

cluster analysis showed that the similarity of five groups was above 70% in the same

day. This is possibly because same raw material contains a similar intrinsic microbial

community and the PSB agent is screened from composting samples. However, the

bacterial homology coefficient of samples in different days of composting was low, such

as 39% in that of samples from day 10 and 20. The above results suggest that PSB

inoculation may not inhibit indigenous microbial growth activities and diversity

(biological safety), and the change of physicochemical environment (e.g., temperature,

pH, organic matter content, etc.) in different stages is more crucial that could lead to the

11
bacterial community composition succession.

Fig. 1. The DGGE fingerprint diagram of bacterial community, bands numbers and

Shannon-Wiener index during composting for different treatments. Bands numbers and

Shannon-Wiener index are represented by points and rectangles.

3.2. TP and P fractions in different treatments

TP content in all the composts exhibited a significant increase during composting for

20 days (Fig. 2a), which may be due to the “concentration effect” (Wei et al., 2015).

The concentration ranged from 5.8 to 10.9 g/kg for CK, from 7.2 to 13.7 g/kg for CP,

and from 7.1 to 18.7 g/kg for composts with PSB and RP addition. There were distinct

differences in concentration of TP at 20 days of composting (p < 0.05), which decreased

in the order: CMP2 > CMP3, CMP1 > CP > CK. This may be attributed due to the

contribution of P from RP in P-enriched composts (Nishanth and Biswas, 2008) and

12
increased decomposition caused by the activation of microbial community related to

PSB inoculation.

Fig. 2c and Fig. 2e illustrates the concentrations of Olsen P and CAP in different

treatments. As the mobile and easily mineralizable fraction of P for plant use, the

concentration of Olsen P in the end of composting increased significantly as compared

to the values at the beginning of composting in all composts (p < 0.05). The final

amount of Olsen P varied from 4.2 to 8.1 g kg-1 (38.2-43.1% of TP). Olsen P content of

three inoculation groups was all significantly higher than that in CP and CK on the 20th

day (p < 0.05). The variation in CAP shows a similar uptrend to that observed for Olsen

P among all composts, whereas the ratio of CAP fraction relative to TP was higher than

Olsen P in the end of composting, ranging from 41.4% to 52.4% on average, highest for

CMP2, CMP3, and CMP1 (all above 50%) and lowest for CK. Moreover, the content of

CAP had a more obvious increase by around 75.4% in the 20th day compared with that

in the 8th day than the increase from day 0 to 8, especially in CMP3. The results might

be related to the suitable condition after the thermophilic phase of composting for the

activity of PSB to solubilize RP, particularly the exogenous inoculants. CAP was

closely correlated with the available P that can be used by crops from chemical

fertilizers, including labile and moderately labile P pool, which might be also related to

high humic acid formation during composting by coadsorption or metal bridging

(Borggaard et al., 2005; Cheng et al., 2004). WSP is the phosphate in solution, which

could be regarded as the most bioaccessible and labile form of P (Siciliano et al., 2016).

13
There was a gradual decrease from the beginning to the 8th day of composting and then

an increase in WSP content in all the treatments until the end of composting, the most

significant increase being found in CMP2, CMP3, CMP1 and CP (Fig. 2d). Combined

with the results shown in DGGE, the increase of WSP may be associated with

increasing microbial population as the decomposition of organic matter after the

thermophilic phase of composting, which led to more demand for labile P. We suspect

that the need of WSP might stimulate inoculated or indigenous PSB for releasing

organic acids or increasing the H+ activity to solubilize larger amount of P in compost

with RP. The WSP content on the 20th day of composting was lowest for CK (1.07 g

kg-1) and highest for CMP2, CMP3 and CMP1 (2.00 g kg-1, 1.84 g kg-1, 1.76 g kg-1,

respectively), followed by CP (1.32 g kg-1). In contrast, there is a large decrease in the

proportion of WSP to TP during composting, ranging from 4.3% to 18.3%, which was

similar to the results reported by Wei et al. (2015), suggesting that RP and PSB

inoculation could enhance CAP and Olsen P but might not improve the ratio of WSP to

TP in MSW composts.

The distribution of PAP and NAP of samples at the initial and final stages of

composting was presented in Fig. 2b. In general, composting is associated with

increasing the potential availability of P, leading to the similar distribution in the end of

composting in different treatments, which is consistent with the earlier report of Ngo et

al. (2013). The proportion of PAP to TP was lower than NAP in the beginning of

composting, averaging just 45.4%, while PAP was the main fraction in the TP content

14
in the end composting (above 69% on average) and a decrease of the more resistant

NAP occurred. There were distinct differences in the proportion of PAP to TP on day 20

of composting among different treatments, which increased in the order: CK < CP <

CMP1, CMP3 < CMP2 (p < 0.05). The RP enriched composts in CMP1, CMP2 and

CMP3 had significantly larger amounts of PAP, which may be a result of solubilization

of P by PSB inoculant. Low molecular weight organic acids (LOAs) are incompletely

decomposed products of relatively low molecular weight degradable organic matter

(Reyes et al., 2006). Given that LOAs could dissolve the mineral phosphate in RP and

precipitates through chelation reactions and ligand exchange reactions with phosphate

anions and decreasing pH (Reyes et al., 2006; Guppy et al., 2005), the P-solubilization

by PSB inoculant is probably related to the production of LOAs. Moreover, the above

results suggest that there is little inhibition on the function and activity of composite

PSB inoculum when directly applied into MSW due to the source of PSB inoculants and

appropriate environments, and two-stage inoculation of PSB (inoculating both in the

initial and cooling stages) could further accelerate the effective transformation of P in

RP and MSW compost, increasing 34.8% of that in CP.

15
Fig. 2. Changes in TP and PAP in different treatment groups during composting. (a) TP

(total phosphorus) content; (b) Distribution of PAP (potential available phosphorus) and

NAP (not available phosphorus) ; (c) Olsen P (Olsen phosphorus) content; (d) WSP

(water-soluble phosphorus) content; (e) CAP (citric acid phosphorous) content.

3.3. Interaction between P forms, treatments and biological factors

There are many confounding factors that mask PSB inoculation effects under

composting conditions. Therefore, considering that microbial communities were clearly

linked to P forms variation (Wei et al., 2016a; Mander et al., 2012), we used constrained

RDA to test what extent the treatments including RP addition and PSB inoculation

affected the bacterial diversity firstly and then to determine which inoculation method

was major external controller of community composition during composting in this

study (Fig. 3). Table S3 shows that the total variation in the species data explained by all

the external treatments accounted for 43.2% statistically. Both the addition of P (F value

= 3.6, p = 0.008) and PSB inoculation (F value = 2.9, p = 0.020) exerted highly

16
significant influences on the composting bacterial community structure as inferred from

Monte Carlo permutation tests. Although many previous studies have demonstrated the

importance of inoculation on indigenous microbial populations by collaborative

symbiosis or inhibition (Zhao et al., 2016b; Xi et al., 2015), high P addition has rarely

been reported as an important amended factor of bacterial community in composting,

which may stimulate microbial activity and respiration (Malik et al., 2012). The results

of variation partitioning analysis, which was conducted to extract the changes in

microbial communities explained by different inoculation methods without the effects

of others, suggested that inoculation for two times in both initial and cooling stages of

composting was clearly significant, solely explaining 19.6% (F value = 6.8, p = 0.004).

The above results also could explain the reason why the obvious differences of

microbial biomass and diversity among three inoculation method in the DGGE

fingerprint diagram.

17
Fig. 3. RDA of bacterial community composition and treatments. Solid symbols indicate

the DGGE bands. Inoculation in the initial stage or cooling stage or both two stages of

composting are presented by Initial, Cooling and Mix, respectively. Significant

treatments are indicated by solid lines (p < 0.05) while others with dashed lines are not

significant.

Distance-based redundancy analysis (db-RDA) combined with principal component

analysis (PCA) was used to investigate interactions between different bacterial

populations and availability of P fractions in different treatments. Fig. 4 shows an

ordination graph where the first two principal components accounted for 84.06% of the

variance. Multivariate analysis indicated that bacterial communities were different at

each composting stage but similar in different treatments since samples from each phase

were dispersed but mainly clustered into 3 groups in different areas of the ordination

18
graph in agreement with Fig. S3. Moreover, bacterial communities were significantly

related to TP and CAP based on forward selection (p < 0.01), and PAP is the most

important factor affecting community structure (Table S4), which could explain 45.6%

variance. However, Olsen P and WSP had little effect on community composition. Thus,

PAP, a new index for potential bio-available fraction of P taken by plant and

microorganisms immediately or in a long time, likely exert a kinetic influence on the

interaction of indigenous microbial community and inoculated PSB, which limits

microbial P-solubilizing phenotypes. When the angle between environmental factors

and the straight lines of different points representing samples in the ordination diagram

was considered according to Siciliano et al. (2016), it provided powerful new insights

into the comprehensive relationship between P forms, treatments and biological factors.

Different treatment based on inoculation methods had variable effects, with increase of

inoculated PSB amount in the beginning of composting (the line of CMP1-10 and

CP-10 and the line of CMP2-10 and CP-10) altering the communities at a smaller

included angle with the changes in PAP. In contrast, increase of inoculated PSB amount

in the cooling stages of composting (the line of CMP3-20 and CP-20 and the line of

CMP2-20 and CP-20) altered the microbial community at a larger angle with the PAP,

tending to be orthogonal. Given that a smaller angle represents a stronger correlation in

the ordination graph (Zhao et al., 2016b; Chen et al., 2014), two-stage inoculation of

PSB (both in the initial stage and the cooling stage) affected PAP by regulating

microbial community in composting and there is more demand for PSB inoculation in

19
the beginning of composting but less need in the cooling stage to accumulate PAP,

which may be attributed to more intense competition between PSB inoculants and

indigenous bacteria in the beginning and the difference of microbial populations in

diverse stages. On the other hand, conscious of the effects of P status on abundance or

frequency of bacterial P-solubilization (Mander et al., 2012), the higher content of PAP

measured since the cooling phase of composting may link to the inhibition of growth

and P-solubilizing ability of PSB by negative feedback. Therefore, the combined

inoculation of PSB in the initial stage with higher inoculation amount and in the cooling

stage with lower inoculation amount was the most effective way to improve the

inoculation effect and increase the solubilization and utilization of RP during

composting. The new insights for multivariate analytical methods could be used to

preliminarily evaluate other inoculants for finding suitable inoculation amount and stage

of composting.

20
Fig. 4. Distance-based redundancy analysis (db-RDA) combined with principal

component analysis (PCA) of bacterial species and different phosphorus fractions.

Significant parameters are indicated by solid lines (p < 0.05) while supplementary

parameters are shown using dashed lines. Solid symbols represent samples from the five

different treatments. Ellipses demonstrate groups of communities.

3.4. How to improve P transformation and long-term utilization efficiency in

composts?

The aim of the regulating method for the distribution of P fractions of composts

described in Wei et al. (2016a) was not only to solubilize RP, but also to improve the

long-term utilization efficiency of P in composts by biological and physicochemical

strategies, which could ultimately adjust the status of P fractions in different stages of

21
composting based on the demand. However, considering that bacteria are more

influential due to their metabolic versatility, the potential activities of PSB do not

necessarily correspond to rapid P solubilization in the microenvironment of composting.

Therefore, it is difficult to fully understand the underlying pathways of P

transformations through microbial communities along with the degradation of organic

matter, which may involve extremely complex interrelationships between humification

process, solubilization of any added insoluble P source (e.g. RP), mineralization of

organically-bound P and P uptake mechanisms in microbial cells. Based on a number of

hypotheses have been presented on P cycle (Owen et al., 2015; Malik et al., 2012; Khan

and Joergensen, 2009; Borggaard et al., 2005; Guppy et al., 2005; Cheng et al., 2004), a

possible complex web with regard to P transformation by biological and

physicochemical ways in composts with the addition of RP and PSB is shown in Fig. 5.

Humic compounds as chemically reactive colloids are known to interact with inorganic

and organic components and metal ions, which could modify composting conditions for

microbial growth and activity (Zhao et al., 2017; Wei et al., 2007). Indications are also

present that some labile phosphate in solution is possibly converted into a fulvo-metal-P

or humo-metal-P complex compound by coadsorption or metal bridging with the

participation of the metal cation (Urrutia et al., 2013; Cheng et al., 2004), even though

there may be a competitive effect of LOAs on P adsorption to variable charge metal

surfaces (Guppy et al., 2005). Besides, some functional microbes like

phosphate-accumulating organisms (Saito et al., 2004) are potentially present in the end

22
of composting as the result of the abundant labile phosphate and capable of

accumulating excess amounts of polyphosphate into microbial cells as microbial

biomass P (MBP) but have limited sorption. Therefore, the transformation showed by

red arrows in Fig. 5 could be considered the form of biological fixation of phosphates,

i.e., sorption of phosphates into the cellular material of microorganisms and yielding

humo-metal-P or fulvo-metal-P complexes, which are biologically stable but more

soluble than the fixed Ca-, Fe- and Al-phosphates (Tan, 2014; Guppy et al., 2005).

These P may act as a long-term available pool of P that could be released slowly,

potentially being taken up by plants more efficiently during the process of biomass

turnover (Wei et al., 2016b; Khan and Joergensen, 2009). To reduce the possible

inhibition to PSB and further improve P transformation and long-term utilization

efficiency in composts, it is meaningful to enhance the biological fixation of phosphates

by following methods: (1) inoculating phosphate-accumulating organisms from MSW

composts in the cooling stage of composting after PSB inoculation to transform some

phosphate in solution to MBP, (2) controlling the content of metal cation with the

chelation capacity to increase the formation of fulvo-metal-P or humo-metal-P complex

compound according to the formation of humic matter, and (3) regulating the production

of LOAs to increase RP solubilization in the beginning of composting but decrease the

competition with phosphate since the cooling stage of composting. Further elucidation

of the relationships between HA or FA and P in sorption processes is necessary,

specifically because long-term P availability may result.

23
Fig. 5. Scheme showing the complex web of P transformation by biological and

physicochemical method within composts with the addition of RP and PSB. MBP,

microbial biomass phosphorus; LOAs, low molecular weight organic acids; MW,

molecular weight.

4. Conclusions

This study showed that inoculation with PSB during MSW composting could

increase PAP and improve bacterial community structure and function compared with

CK and CP. Mixed inoculation in the initial stage with higher inoculation amount and

cooling stage with lower inoculation amount had a clear advantage over inoculation in

any alone stage to enhance the solubilization and utilization of RP according to the new

insights for multivariate analyses. Besides, inoculating phosphate-accumulating

24
organisms, controlling the content of metal cation with the chelation capacity, and

regulating the production of LOAs were suggested to improve P transformation and

long-term utilization efficiency in composts.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of

China (No. 51378097), the National Key Technology R&D Program (No.

2012BAJ21B02-02).

References:

1. Borggaard, O.K., Raben-Lange, B., Gimsing, A.L., Strobel, B.W., 2005. Influence

of humic substances on phosphate adsorption by aluminium and iron oxides.

Geoderma, 127, 270-279.

2. Chen, Y., Zhou, W., Li, Y., Zhang, J., Zeng, G., Huang, A., Huang, J., 2014. Nitrite

reductase genes as functional markers to investigate diversity of denitrifying

bacteria during agricultural waste composting. Appl. Microbiol. Biot. 98,

4233-4243.

3. Cheng, W.P., Chi, F.H., Yu, R.F., 2004. Effect of phosphate on removal of humic

substances by aluminum sulfate coagulant. J. Colloid. Interf. Sci. 272, 153-157.

4. Guppy, C.N., Menzies, N.W., Moody, P.W., Blamey, F., 2005. Competitive

sorption reactions between phosphorus and organic matter in soil: a review. Aust. J.

25
 Soil. Res. 43, 189-202.

5. Hachicha, R., Rekik, O., Hachicha, S., Ferchichi, M., Woodward, S., Moncef, N.,

Cegarra, J., Mechichi, T., 2012. Co-composting of spent coffee ground with olive

mill wastewater sludge and poultry manure and effect of Trametes versicolor

inoculation on the compost maturity. Chemosphere, 88, 677-682.

6. Hashimoto, Y., Takamoto, A., Kikkawa, R., Murakami, K., Yamaguchi, N., 2014.

Formations of Hydroxyapatite and Inositol Hexakisphosphate in Poultry Litter

during the Composting Period: Sequential Fractionation, P K-edge XANES and

Solution P-31 NMR Investigations. Environ. Sci. Technol. 48, 5486-5492.

7. Khan, K.S., Joergensen, R.G., 2009. Changes in microbial biomass and P fractions

in biogenic household waste compost amended with inorganic P fertilizers.

Bioresour. Technol. 100, 303-309.

8. Li, S.X., Wang, Z.H., Stewart, B.A., 2011. Differences of some leguminous and

nonleguminous crops in utilization of soil P and responses to phosphate fertilizers.

Adv. Agron. 110, 125-249.

9. Malik, M.A., Marschner, P., Khan, K.S., 2012. Addition of organic and inorganic P

sources to soil - Effects on P pools and microorganisms. Soil Biol. Biochem. 49,

106-113.

10. Mander, C., Wakelin, S., Young, S., Condron, L., O'Callaghan, M., 2012. Incidence

and diversity of phosphate-solubilising bacteria are linked to phosphorus status in

grassland soils. Soil Biol. Biochem. 44, 93-101.

26
11. Murphy, J., Riley, J.P., 1962. A modified single solution method for the

determination of phosphate in natural waters. Anal. Chim. Acta. 27, 31-36.

12. Ngo, P.T., Rumpel, C., Ngo, Q.A., Alexis, M., Vargas, G.V., Gil, M., Dang, D.K.,

Jouquet, P., 2013. Biological and chemical reactivity and phosphorus forms of

buffalo manure compost, vermicompost and their mixture with biochar. Bioresour.

Technol. 148, 401-407.

13. Nishanth, D., Biswas, D.R., 2008. Kinetics of phosphorus and potassium release

from rock phosphate and waste mica enriched compost and their effect on yield and

nutrient uptake by wheat (Triticum aestivum). Bioresour. Technol. 99, 3342-3353.

14. Owen, D., Williams, A.P., Griffith, G.W., Withers, P., 2015. Use of commercial

bio-inoculants to increase agricultural production through improved phosphorus

acquisition. Appl. Soil. Ecol. 86, 41-54.

15. Qiu, L., Cui, H., Wu, J., Wang, B., Zhao, Y., Li, J., Jia, L., Wei, Z., 2016.

Snowmelt-driven changes in dissolved organic matter and bacterioplankton

communities in the Heilongjiang watershed of China. Sci. Total. Environ. 556,

242-251.

16. Reyes, I., Valery, A., Valduz, Z., 2006. Phosphate-solubilizing microorganisms

isolated from rhizospheric and bulk soils of colonizer plants at an abandoned rock

phosphate mine. Plant. Soil. 287, 69-75.

17. Saito, T., Brdjanovic, D., van Loosdrecht, M., 2004. Effect of nitrite on phosphate

uptake by phosphate accumulating organisms. Water. Res. 38, 3760-3768.

27
18. Siciliano, S.D., Chen, T., Phillips, C., Hamilton, J., Hilger, D., Chartrand, B.,

Grosskleg, J., Bradshaw, K., Carlson, T., Peak, D., 2016. Total Phosphate

Influences the Rate of Hydrocarbon Degradation but Phosphate Mineralogy Shapes

Microbial Community Composition in Cold-Region Calcareous Soils. Environ. Sci.

Technol. 50, 5197-5206.

19. Tan, K.H., 2014. Humic matter in soil and the environment: principles and

controversies. CRC Press.

20. Turner, B.L., Leytem, A.B., 2004. Phosphorus compounds in sequential extracts of

animal manures: chemical speciation and a novel fractionation procedure. Environ.

Sci. Technol. 38, 6101-6108.

21. Urrutia, O., Guardado, I., Erro, J., Mandado, M., Garcia-Mina, J.M., 2013.

Theoretical chemical characterization of phosphate-metal-humic complexes and

relationships with their effects on both phosphorus soil fixation and phosphorus

availability for plants. J. Sci. Food. Agric. 93, 293-303.

22. Wang, H.X., Xu, J.L., Yu, H.X., Liu, X.J., Yin, W., Liu, Y.Y., Liu, Z.W., Zhang, T.,

2015a. Study of the application and methods for the comprehensive treatment of

municipal solid waste in northeastern China. Renew. Sust. Energ. Rev. 52,

1881-1889.

23. Wang, X., Cui, H., Shi, J., Zhao, X., Zhao, Y., Wei, Z., 2015b. Relationship

between bacterial diversity and environmental parameters during composting of

different raw materials. Bioresour. Technol. 198, 395-402.

28
24. Wei, Y., Zhao, Y., Xi, B., Wei, Z., Li, X., Cao, Z., 2015. Changes in phosphorus

fractions during organic wastes composting from different sources. Bioresour.

Technol. 189, 349-356.

25. Wei, Y., Wei, Z., Cao, Z., Zhao, Y., Zhao, X., Lu, Q., Wang, X., Zhang, X., 2016a.

A regulating method for the distribution of phosphorus fractions based on

environmental parameters related to the key phosphate-solubilizing bacteria during

composting. Bioresour. Technol. 211, 610-617.

26. Wei, Y., Zhao, Y., Wang, H., Lu, Q., Cao, Z., Cui, H., Zhu, L., Wei, Z., 2016b. An

optimized regulating method for composting phosphorus fractions transformation

based on biochar addition and phosphate-solubilizing bacteria inoculation.

Bioresour. Technol. 221, 139-146.

27. Wei, Z.M., Xi, B.D., Zhao, Y., Wang, S.P., Liu, H.L., Jiang, Y.H., 2007. Effect of

inoculating microbes in municipal solid waste composting on characteristics of

humic acid. Chemosphere, 68, 368-374.

28. Xi, B.D., He, X.S., Dang, Q.L., Yang, T.X., Li, M.X., Wang, X.W., Li, D., Tang, J.,

2015. Effect of multi-stage inoculation on the bacterial and fungal community

structure during organic municipal solid wastes composting. Bioresour. Technol.

196, 399-405.

29. Zhang, Y., Zhao, Y., Chen, Y., Lu, Q., Li, M., Wang, X., Wei, Y., Xie, X., Wei, Z.,

2016. A regulating method for reducing nitrogen loss based on enriched

ammonia-oxidizing bacteria during composting. Bioresour. Technol. 221, 276-283.

29
30. Zhao, X., He, X., Xi, B., Gao, R., Tan, W., Zhang, H., Li, D., 2016a. The evolution

of water extractable organic matter and its association with microbial community

dynamics during municipal solid waste composting. Waste Manage. 56, 79-87.

31. Zhao, Y., Lu, Q., Wei, Y., Cui, H., Zhang, X., Wang, X., Shan, S., Wei, Z., 2016b.

Effect of actinobacteria agent inoculation methods on cellulose degradation during

composting based on redundancy analysis. Bioresour. Technol. 219, 196-203.

32. Zhao, Y., Wei, Y., Zhang, Y., Wen, X., Xi, B., Zhao, X., Zhang, X., Wei, Z., 2017.

Roles of composts in soil based on the assessment of humification degree of fulvic

acids. Ecol. Indic. 72, 473-480.

30