3.1 Effects of biostimulation on TPH, saturates and aromatics degradation
The degradation tendency of petroleum hydrocarbons during 90 days of biostimulation remediation is shown in Fig. 1. Addition of deionized water, mineral nutrition and compost reduced the petroleum hydrocarbon contents from 24166 to 21438, 18480 and 18232 mg kg− 1 soil, which represented 11.29%, 23.53% and 24.55% of aged hydrocarbons removal, respectively (Fig. 1(a)).
Compost or KNO3 amendment was more effective for stimulating the activities of indigenous microorganisms than that of deionized water. In our previous studies, hydrocarbons removal rates were 16% and 18% after 10 and 8 weeks of remediation by 20% soil moisture contents, respectively (Wu et al., 2019, 2017). Nwankwegu et al. (2016) reported the organic compost as the source of nutrients and bulking agent promoted 93.31 ± 3.98% of diesel oil degradation, which was more effective than nitrogen and phosphorus fertilizers (71.36 ± 5.60% hydrocarbons degradation rate) amendments in the diesel contaminated soil for eight weeks of incubation. In this study, removal rates of petroleum hydrocarbons were lower than the those of the previous study. The reason for this may be that the aged hydrocarbons are difficult to degrade by soil microorganisms in the loessal soil.
The saturate hydrocarbon fractions in the untreated oil-contaminated soil was 16394 ± 160 mg kg− 1, and they decreased to 15509 ± 124, 13597 ± 136 and 9076 ± 148 mg kg− 1 in the CK, MS and SC treatments after 90 days of biostimulation, representing removal rates of 3.25%, 17.8% and 45.1%, respectively (Fig. 1(b)). The aromatic fractions decreased from 4554 mg kg− 1 in the untreated soil to 3607 ± 130, 1942 ± 143 and 3507 ± 151 mg kg− 1 in the CK, MS and SC treatments, representing degradation rates of 20.8%, 57.4% and 23.0%, respectively. Thus, KNO3 amendment was more effective for removing aromatics while compost addition could degrade more the saturates.
3.2 Influence of biostimulation on soil bacterial community
Table 2 shows that the number of OTUs based on the 97% similarity threshold were CK (5531) > MS (5317) > SC (5025) in the soil microcosms after 90 days of bioremediation. The ACE index were 10849, 10018 and 9824 and Chao 1 index were 8954, 8176 and 7785 in the CK, MS and SC soil samples, respectively, which indicated that enhancing soil moisture was beneficial to the survival of various of soil microorganisms, while KNO3 and compost supplementation could reduce the richness of soil microorganisms. It can also be concluded from Table 2 that the soil microbial community rehabilitated by deionized water addition was the most evenly distribution because the ranking of the Simpson index was CK (0.0026) < MS (0.0301) < SC (0.0405). The inorganic nitrogen fertilizer and compost had great interference on the uniformity distribution of microbial community. Nwankwegu et al. (2016) confirmed that it was a failure to evaluate the degradation potential of diesel oil by the increase of total heterotrophic bacterial in the soil, because only specific hydrocarbon utilizing microorganisms were related to the removal rate of diesel oil. Our study revealed that although KNO3 and compost reduced soil microbial species and the community distribution uniformity, the existence of functional hydrocarbon degrading bacteria may be enhanced which could lead to higher petroleum hydrocarbons degradation in the MS and SC treatments.
Table 2
Bacterial taxonomy and diversity indices in the CK, MS and SC remediation soil.
soil samples
|
CK
|
MS
|
SC
|
sequencing analysis
|
|
|
|
OTUs num
|
5531
|
5317
|
5025
|
Diversity indices
|
|
|
|
Shannon index
|
7.16
|
6.14
|
5.22
|
Simpson index
|
0.0026
|
0.0301
|
0.0405
|
Ace index
|
10849
|
10018
|
9824
|
Chao1 index
|
8954
|
8176
|
7785
|
Figure 2 reflects the compositions of microbial community in soils in terms of dominant phyla, classes and genera. Proteobacteria, Actinobacteria, Acidobacteria and Planctomycetes phyla were dominant in all three treatments but the relative abundance changed in different treatments (Fig. 2(a)). The relative abundance of the Proteobacteria, Actinobacteria, Acidobacteria and Planctomycetes phyla in the CK treatment were 25.21%, 23.40%, 16.96% and 9.08%, respectively. In the MS and SC treatments, the relative abundance of Proteobacteria were 23.54% and 23.10% respectively, which were not significantly different from that in the CK treatment. Compared to the CK treatment, the relative abundance of Actinobacteria, Acidobacteria and Planctomycetes phyla changed obviously in the MS and SC treatments with the values of 40.0%-31.5%, 15.0%-8.21% and 4.20%-4.21, respectively. In addition, percentage distribution of Firmicutes phylum in the SC treatment (27.3%) was the most, which had a significant increase compared with the CK (1.68%) and MS (7.08%) treatments. The dominant classes affiliated to phyla of Proteobacteria, Actinobacteria, Acidobacteria and Planctomycetes were Alphaproteobacteria, Actinobacteria, Acidobacteria_Gp7 and Planctomycetia, respectively. Compost and KNO3 addition had minor effect on the microbial community at the class level (Fig. 1(b)-(e)). Firmicutes, Proteobacteria, Acidobacteria and Actinobacteria phyla have proven to be important in mitigating the long-chain alkane, PAHs and other types of hydrocarbons pollution in the soil environment (Zhang et al., 2012; Wu et al., 2019; Wu et al., 2020). Wolf et al. (2019) found that soil polluted by pyrene substantially increased the microorganisms belonging to Firmicutes phylum in the clay and sandy loam soils. In addition, phyla of Actinobacteria increased while Acidobacteria and Proteobacteria decreased in the soils remediated by amendments of KNO3 or sterilized nutrition, indicating that the hydrocarbon degradation may be attributed to the increase of Actinobacteria phylum which could utilize petroleum hydrocarbon as their energy source for cellular metabolic activity.
The influences of the biostimulation treatments on the genus level were great (Fig. 1(f)). Steroidobacter, Intrasporangium, Blastopirellula, Saccharibacteria_genera_incertae_sedis, Parcubacteria_genera_incertae_sedis, Aquisphaera and Rhodococcus were new genera appeared in the MS and SC treatments compared with the CK soil. Specially, Intrasporangium also became a second dominant genus (5.6%) in the SC treatment, which may be related to hydrocarbons degradation. Genera of Solirubrobacter, Spartobacteria, incertae_sedis and WPS-1_genera_incertae_sedis with the relative abundance more than 1% in the CK treatment that was not appeared in the MS and SC treatments. Gp6 was the most dominant genus in the CK (6.4%), while the relative abundance of Gemmata was the greatest in the MS (3.21%) and SC (5.90%) treatments. Therefore, it can be concluded that the compositions of indigenous flora in oil-contaminated soil was influenced by nutrition amendments (Seklemova et al., 2001), which may be related to the higher hydrocarbon removal rates in the MS and SC treatments.
3.3 Carbon and nitrogen depletion
The contents of carbon and nitrogen elements in the contaminated soils were detected by elemental analyzer. The soil total carbon content was 32327 ± 30 mg kg− 1 in the initial oil-contaminated soil (Table 3), and TPH content was determined by ultrasonic extraction and gravimetric method with the value of 24166 ± 40.46 mg kg− 1 (Table 1). According to the description of Margesin et al. (2007), the proportion of carbon content of TPH was 20541 mg kg− 1 which derived from 85% TPH concentration. Thus soil organic carbon in polluted soil was 11786 mg kg− 1 which are calculated by subtracting the carbon proportion of TPH from soil total carbon.
Table 3
Soil total carbon (STC) and soil total nitrogen (STN) contents, δ13C and δ15N in the CK, MS and SC remediation soil.
Treatments
|
Time
(days)
|
STC
(mg/kg)
|
STN
(mg/kg)
|
C/N
|
δ13C
(‰)
|
δ15N
(‰)
|
CK
|
0d
|
32327 ± 30
|
1762 ± 8.2
|
10.97 ± 0.2
|
-25.16 ± 0.3
|
5.79 ± 0.2
|
15d
|
31876 ± 25
|
1751 ± 7.1
|
10.10 ± 0.1
|
-26.95 ± 0.1
|
5.81 ± 0.1
|
30d
|
31369 ± 24
|
1747 ± 9.2
|
9.37 ± 0.2
|
-26.84 ± 0.5
|
5.85 ± 0.1
|
45d
|
29987 ± 28
|
1730 ± 10.1
|
10.01 ± 0.3
|
-27.01 ± 0.2
|
5.92 ± 0.3
|
60d
|
30128 ± 32
|
1715 ± 13.5
|
10.40 ± 0.2
|
-26.81 ± 0.1
|
6.01 ± 0.1
|
90d
|
30102 ± 29
|
1698 ± 6.5
|
10.10 ± 0.1
|
-27.18 ± 0.4
|
6.21 ± 0.2
|
MS
|
0d
|
32327 ± 27
|
1919 ± 5.4
|
10.07 ± 0.1
|
-25.16 ± 0.2
|
8.13 ± 0.2
|
15d
|
29158 ± 24
|
1768 ± 5.2
|
10.02 ± 0.1
|
-26.90 ± 0.1
|
12.59 ± 0.2
|
30d
|
30003 ± 18
|
1731 ± 5.1
|
10.43 ± 0.1
|
-26.71 ± 0.1
|
12.71 ± 0.1
|
45d
|
27905 ± 30
|
1718 ± 5.3
|
8.67 ± 0.1
|
-27.04 ± 0.1
|
13.04 ± 0.1
|
60d
|
27189 ± 26
|
1701 ± 5.2
|
8.88 ± 0.2
|
-27.47 ± 0.2
|
13.68 ± 0.1
|
90d
|
27227 ± 21
|
1693 ± 4.9
|
8.73 ± 0.1
|
-28.01 ± 0.1
|
13.74 ± 0.2
|
SC
|
0d
|
32327 ± 24
|
1847 ± 10.2
|
10.46 ± 0.1
|
-25.16 ± 0.5
|
6.25 ± 0.2
|
15d
|
32130 ± 30
|
1763 ± 12.1
|
8.73 ± 0.4
|
-26.13 ± 0.2
|
7.45 ± 0.1
|
30d
|
32482 ± 32
|
1726 ± 10.5
|
8.76 ± 0.1
|
-26.16 ± 0.3
|
8.06 ± 0.1
|
45d
|
31024 ± 16
|
1715 ± 8.3
|
8.67 ± 0.1
|
-28.18 ± 0.2
|
8.86 ± 0.1
|
60d
|
29153 ± 26
|
1696 ± 6.4
|
8.72 ± 0.2
|
-29.81 ± 0.4
|
10.21 ± 0.1
|
90d
|
28595 ± 16
|
1690 ± 7.9
|
8.63 ± 0.3
|
-29.92 ± 0.2
|
10.48 ± 0.2
|
Two kinds of method including alkaline potassium persulfate digestion-UV spectrophotometry and elemental analyzer were respectively used to determine soil total nitrogen. The content of soil total nitrogen (STN) was 1005 ± 12.21mg kg− 1 (Table 1) by using UV spectrophotometry and 1762 ± 8.2 mg kg− 1 (Table 3) by elemental analyzer. One possible reason for an obvious difference between them was that there were some special nitrogen forms in the polluted soil which can’t be digested by alkaline potassium persulfate.
The contents of soil total carbon (STC) obviously decreased with time in the MS and SC treatments compared with the CK treatment. This result was consistent with the TPH degradation trends in our study (Fig. 1(a)).
KNO3 and compost treated soil samples (MS and SC samples) had lower C:N ratios on the 90th days of bioremediation than earlier days (Table 3), while there was less reduction of the C:N ratio in the deionized water-treated soil. Previous literatures reported that soil microorganisms were suitable for living at a nutritional level with a C:N ratio of 100:10 (Prescott et al., 2002). In this study, the C:N ratios had been maintained at about 10:1 in the CK treatment, and these ratios were close to 10 in the first 30 days during the bioremediation and then decreased to about 8 in the MS treatment, while the C:N ratios were reduced from about 10 to 8 on the 15th day in the SC treatment. Additionally, the TPH content was reduced in all three treatments, which was the main reason for the reduction of soil total carbon and the C:N ratios. The reason for the reduced C:N ratios in the SC and MS treatment may be that the degradation rates of TPH by the amendment of KNO3 or compost were faster than that in the CK treatments.
3.4 Carbon and nitrogen isotope effect
The δ13C and δ15N of soil samples detected by stable isotope ratio mass spectrometer are shown in Table 3. According to Equations 3 and 4, carbon isotope enrichment factors respectively were εc = 28.3 ± 0.3‰, εc = 16.6 ± 0.8‰ and εc = 38.8 ± 1.1‰ in the CK, MS and SC treatments after 90 days of remediation (Fig. 3). Previous study have illustrated that the carbon isotope of petroleum hydrocarbons has almost no fractionation effect in the process of petroleum hydrocarbon biodegradation, which means that the carbon isotope enrichment factor is not shift by hydrocarbon degradation (Aggarwal and Hinchee, 1991). During the bioremediation, the δ13C values of inorganic carbon in loessal soil was closely originated from the decomposition of organic matter (Ehleringer et al., 2000; Bouchard et al., 2008) and soil total carbon isotope enrichment was the result of the transformation of soil total carbons including inorganic and organic fractions. A significant changes in δ13C of soil total carbons can be used as an indicator of the petroleum hydrocarbon degradation in the contaminated soil. Figure 1(b) shows KNO3 supplement was more effective for removing the aromatics while compost addition were beneficial to degrade the saturates. Thus the significant difference of carbon enrichment factors of εc = 16.6 ± 0.8‰ in the MS and εc = 38.8 ± 1.1‰ in the SC treatments maybe caused by the degradation of aromatic and saturated hydrocarbons, respectively.
The δ15N value of soil total nitrogen was variable in three different treatments during 90 days of incubation. And the nitrogen isotope effects (\(\varepsilon N\)) were respectively εN = -7.23 ± 4.08‰, εN = -24.20 ± 9.54‰ and εN = -79.49 ± 16.41‰ in the CK, MS and SC treatments at the end of the bioremediation. Previous study reported that humidity had a significant effect on the transformation of nitrogen in the soil through the mineralization of organic nitrogen by microorganisms, and promoted the metabolic activities of petroleum hydrocarbon degrading bacteria (Jia et al., 2019; Liu et al., 2020). Additionally, nitrogen interactive priming effect between the amendments and soil could greatly enhanced by input of exogenous nitrogen (Fiorentino et al., 2019). Therefore, it can be inferred that nitrogen isotope effects occurred during the bioremediation are attributed to both mineralization of organic nitrogen by indigenous microorganism and utilization of nitrogen for hydrocarbons degradation after addition of the deionized water, KNO3 and compost in the oil-contaminated soil.
To further investigate the enrichment trend of carbon and nitrogen isotopes, carbon and nitrogen dual-isotope relationship in the three different treatments are shown in Fig. 4. The carbon and nitrogen dual stable isotopes showed clear differences due to the different hydrocarbons degradation and nitrogen utilization mechanisms by microbial population among the three treatments (Fig. 4). A inverse significant carbon isotope effect (εc = 28.3 ± 0.3‰) and a normal significant nitrogen isotope effect (εN = -7.23 ± 4.08‰) in the CK treatment. On the other hand, a inverse significant carbon isotope effect (εc = 16.6 ± 0.8‰) and a strong significant nitrogen isotope effect (εN = -24.20 ± 9.54‰) in the MS treatment, and a significant carbon isotope effect (εc = 38.8 ± 1.1‰) as well as a strong significant nitrogen isotope effect (εN = -79.49 ± 16.41‰) in the SC treatment. Thus, carbon isotope enrichment factors are inverse significant while nitrogen isotope enrichment factors are normal significant in all three treatments. This difference was caused by different degradation mechanisms of petroleum hydrocarbon pollutants.
Additionally, the nitrogen isotope effects based on the δ15N of KNO3 (1.008‰) and compost (5.3‰) made the enrichment factors variable in the MS (εN = -24.20 ± 9.54‰) and SC (εN = -79.49 ± 16.41‰) treatments, respectively. The nitrogen enrichment factors in the SC treatment was the largest, followed by the MS treatment, and the CK treatment was the smallest (Fig. 3a), which was consistent with the degradation rate of the TPH in the three treatments, implying that petroleum hydrocarbons were degraded accompanied by the utilization and transformation of nitrogen. The ranking of carbon isotope enrichment factors among the three treatments was the SC > CK > MS, which was inconsistent with the degradation rate of TPH, indicating that the enrichment of carbon isotope in the soils may be significantly affected by the different fractions of petroleum degradation and inorganic carbon transformation in loessal soil.
In this study, carbon and nitrogen isotopes showed different enrichment effects after introducing various of stimulants to the soil. Microbial diversity and uniformity also changed by different treatments. The isotope enrichment effects are related to the change of the soil microbial community structure or the soil microbial biomass by the amendments (Ma et al., 2020). It is necessary to investigate soil microbial biomass for understanding isotope enrichment mechanisms during bioremediation of oil-contaminated soil in future studies.