The seeds used for planting were of the high-quality, high-yielding cereal ‘Jingu 21,’ the most recent variety grown by the Shanxi Academy of Agricultural Science and the Industrial Crop Institute; the soil used for planting was grass charcoal substrate soil (for seedlings). And according to previous research (Sharma et al., 2020), the reagent used in the experiments was 1 mmol∙L⁻¹ of Cr⁶⁺ (K₂Cr₂O₇) to ensure that high concentrations of Cr were toxic to cereal crops.

In this study’s greenhouse pot experiment, the test soil was cleaned of stones and other contaminants before being left aside and air-drying for 3 days in the laboratory. Selected high-quality seeds were cleaned with tap water, shielded from light, and immersed in water for approximately 48 h to promote germination. For the planting process, 18 pots (15 cm × 20 cm) were used, and the soil moisture was regulated to 60% of the water-holding capacity required for crop growth. Each pot included 1,300 g of soil and 50–60 seeds that were uniformly dispersed, and the top layer was covered with soil to protect the seeds from light and promote germination and was wrapped with plastic wrap to prevent contamination by airborne bacteria. The pots were then placed in a growth chamber at a constant temperature of 23°C with 16–8 h of light and dark per day. After 15–20 days of incubation, each pot received 300 ml of 1 mmol·L⁻¹ Cr⁶⁺ (The Cr⁶⁺ solution was added to the trays and applied every other day at a fixed time. And distilled water was sprayed on the soil surface daily to ensure soil moisture, no fertilizer was applied to the soil during the experiment).

Three groups made up the experiment: the group used before treatment with Cr solution are the control (CK, equal amount of distilled water was applied to the control group), Cr stress treatment for 6 h (Cr_6h), and Cr stress treatment for 6 days (Cr_6d) (Leng et al., 2020). The setting of the sampling time was done because the microbial community in the soil will respond to the heavy metal stress to a certain extent when Cr stress is applied for approximately 6 h. Due to its inherent resistance and resilience, the microbial community eventually displayed a stable state after around 6 days of the Cr stress treatment. No follow-up analysis was done since heavy metal pollution is a long-term problem. Each experimental group had six biological replicates, and the soil in each pot was randomly selected as the sampling target for rhizosphere soil, for a total of 18 samples. The sampling process was carried out by gently grating the soil with a small spade to expose all the roots of the plant as much as possible and to ensure the integrity of the root soil. The cereal seedlings were removed as a whole and the bulk soils of the root system were shaken off, and about 1 g of soil was collected from each plant. The collected soil was stored in sealed sterile bags and marked properly. The sterile bags were placed in a biological sample sampling box for low temperature storage (in advance in ice packs or dry ice). All samples were then stored in a −80°C refrigerator until use; some were used for DNA extraction and some for determination of soil physical and chemical indicators.

The physical and chemical properties of soil were measured with a JXBS-3001-SCPT-SC (Weihai, Shandong Province, China), which is a high-precision environmental sensor that determines available nitrogen (N), available phosphorus (P), available potassium (K), electrical conductivity (EC) and pH in soil.

A Beijing Zhongke Weihe chlorophyll meter (TYS-3 N, Haidian District, Beijing, China), a pressure-head sensor that emits red and infrared light with peak wavelengths of 650 mm and 940 mm, respectively, was used to measure the chlorophyll content and nitrogen content of the leaves of living seedlings. The samples were measured for stem length, root length, and fresh weight; dried and heated at 105°C for 30 min, then dried at 60°C until a constant weight to calculate their dry weight, after which they were photographed to record seedling growth.

A TIANamp Bacteria DNA kit (Tiangen Biochemical Technology Co., Beijing, China) was used to extract and purify genomic DNA from the samples, which totaled 1 g of soil. Agarose gel electrophoresis was used to assess the extracted DNA, and a enzyme-labeled instrument (Epoch, Xinghui Biotechnology Co., Ltd., Tianjin, China) was used to gauge the extracts’ purity and quantity.

Genomic DNA was amplified by polymerase chain reaction (PCR). The following components made up the amplification system (50 μl): Taq DNA polymerase (5 U∙μL-1) 1.0 μl, 10 × PCR buffer 10.0 μl, dNTP solution 8.0 μl, template DNA 10.0 μl and primers (50 μmol∙L-1). PCR reaction conditions: pre-denaturation at 98°C for 1 min, followed by 98°C for 10 s, 50°C for 30 s, and 72°C for 30 s, for a total of 30 cycles; with a final extension at 72°C for 5 min (Zeng and An, 2021).

The PCR products were then delivered to Shanghai Meiji Biotechnology Engineering Co., Ltd. for Illumina MiSeq PE300 platform high-throughput sequencing analysis. The bacterial community in soil was sequenced using the sequence amplification region 338F_806R (338F: ACTCCTACGGGAGGCAGCAG; 806R: GGACTACHVGGGTWTCTAAT) (Caporaso et al., 2011), and the raw data from the sequencing platform were homogenized. The Flash (1.2.11¹) and Fastp (0.19.6²) software programs were used to perform double-end sequence splicing, filter the reads’ tails for bases with quality values below 20, and eliminate reads that contained N bases to produce high-quality sequences. The overlap between PE readings was used to guide the splicing of the sequence. Uparse (7.0.1090¹) software was used to cluster non-repeated sequences (excluding single sequences) into OTUs based on 97% similarity, and chimeras were eliminated during the process to generate representative sequences of OTUs (operational classification units). Taxonomic analysis of representative sequences of OTUs with a 97% similarity level was carried out using the RDP classifier (2.11²) Bayesian technique. Based on OTU abundance data, a annotation analysis was carried out to determine the makeup of the bacterial community in soil samples. Qiime (1.9.1³) software was used to create a table of the taxonomic abundance and the calculation of the β-diversity distance. Mothur (1.30.2⁴) software was used to conduct the α-diversity study.

The target segments of five nitrogen cycle functional genes, i.e., nitrogen-fixing enzyme (nifH), bacterial ammonia monooxygenase (AOB-amoA), archaeal ammonia monooxygenase (AOA-amoA), nitrate reductase (narG), and nitrite reductase (nirK), were quantified by qPCR. In Table 1, the specifics of the primer sequences for the nitrogen cycle functional genes are displayed.

An ABI Model 7,300 Fluorescent Quantitative PCR instrument (Applied Biosystems, United States) and ChamQ SYBR Color qPCR Master Mix (2X) reagent (Nanjing Novozymes Biotechnology Co., Ltd.) were used to perform real-time fluorescence quantitative PCR (q-PCR). The final reaction mixture volume for the quantification technique was 20 μl and contained 10 μl 2X Taq Plus Master Mix, 1 μl template DNA, 0.8 μ primer F (5 μM), 8 μl primer R (5 μM), and 7.4 μl ddH₂O. Each qPCR reaction was run in triplicate for each set of primers, and the non-template control served as a negative control. Furthermore, q-PCR was carried out using a three-step thermal cycling method that involved 35 cycles of 95°C pre-denaturation for 5 min, 95°C denaturation for 30 s, 58°C annealing for 30 s, and 72°C extension for 1 min (Li et al., 2022). Standard curves were created by diluting pMD18-T, a plasmid for five functional genes, in a 10-fold gradient. Additionally, the R² value of each standard curve exceeded 0.95, demonstrating a linear relationship throughout the concentration range used in this work, with amplification effectiveness ranging from 89.28 to 103.81% (mean 96.31%).

R (4.1.2), SPSS (19.0), and Excel 2019 were used for data analysis and graphing. The variations in the soil physicochemical parameters, as well as the compositional structure and diversity of the soil bacterial communities and the nitrogen cycle functional genes, in the Cr stress time series were evaluated using ANOVA. The ‘ggtern’ package in R was used to create ternary phase diagrams to illustrate the distribution variations in the phylum and genus levels of the bacterial communities in different samples (Hamilton and Ferry, 2018). Mothur (V1.30.2) was used to calculate the α-diversity indices (Shannon index and Simpson index) to reflect the variation in the α-diversity of the soil bacterial communities and the functional genes involved in the nitrogen cycle. QIIME software was used to calculate the weighted (Bray–Curtis) and unweighted (UniFrac) diversity indices in order to compare and contrast the makeup of the soil bacterial communities and functional genes involved in the nitrogen cycle in the Cr stress time series. Non-metric multidimensional scale analysis (NMDS) was used to rank and analyze the temporal distribution patterns of the soil bacterial communities and nitrogen cycle functional genes in the Cr stress time series; the results were verified by stress values, and the graphs were typically thought to have some explanatory significance when the stress<0.2. Redundancy analysis (RDA) was used to examine the relationship between the bacterial community, nitrogen cycle functional genes, and environmental factors (Oksanen et al., 2020). The ‘ggcor’ package in R was used to map the relationships between the dominant bacterial phylum, functional genes involved in the nitrogen cycle, and soil physicochemical parameters. Mantel analysis and the PLS-EM model were used to further investigate the strength of the correlations between the soil physicochemical parameters, bacterial communities, and functional genes involved in the nitrogen cycle (Li Y. et al., 2019).

The results of ANOVA (Figure 1) of the physicochemical parameters of the soil revealed that there was no significant difference (p > 0.05) between the CK and Cr_6h groups, but that there was a difference between the Cr_6h and Cr_6d groups (p < 0.05). At the Cr_6d stage, the soil physicochemical parameters pH and EC value, N, P, K content increased substantially by 135.79, 131.02, 135.34, and 3.55%, respectively (p < 0.05).

In order to assess the impact of Cr stress on grain growth, measurements of the plant stem length, root length, dry weight, fresh weight, chlorophyll content, and nitrogen content were conducted (Supplementary Figure S1). ANOVA (Supplementary Table S1) revealed that stem length and chlorophyll of cereals significantly decreased by 22.40 and 19.07% (p < 0.05) at Cr_6h stage; stem length significantly increased by 8.67% (p < 0.05) at Cr_6d stage; and root length significantly decreased by 26.31% (p < 0.05) at Cr_6d stage compared to the CK.

According to the experimental findings, there were significant differences between the soil bacterial communities at the phylum and genus levels in the Cr stress time series in terms of their composition, structure, and diversity.

Similar to the findings of Zhang et al. (2021), Actinobacteriota, Firmicutes, and Proteobacteria were the three taxa with the highest abundance at the phylum level, and Planifilum, Longispora, and Actinomadura were the three with the highest abundance at the genus level. At the Cr_6d stage, the abundances of Actinobacteriota, Firmicutes, and Chloroflexi in the soil significantly decreased as the Cr stress time increased, whereas the abundances of Proteobacteria and Gemmatiminadota significantly increased. This may be due to the ability of soil microorganisms to adjust to various levels of heavy metal contamination by altering the structure of the microbial community (Li et al., 2017). Previous studies have demonstrated that the bacterial communities transferred across contamination levels in Cr-affected soils are predominantly Proteobacteria, Actinobacteriota, and Chloroflexi (Liu et al., 2019). Proteobacteria, the dominant phylum in soils contaminated with Cd and many other heavy metals, have considerable resistance to heavy metals (Jiang et al., 2019), and some of them have good Cr(VI) biotransformation efficiency (Garavaglia et al., 2010). Thus, Proteobacteria are frequently utilized as markers of Cr contamination, and their abundance increased significantly in the Cr_6d phase of Cr stress.

In addition, Bacillus in the Firmicutes and Streptomyces in the Actinobacteriota can reduce the toxic effect of Cr by reducing Cr(VI) (Liu et al., 2019) and have strong Cr tolerance, making them the dominant genera in Cr-stressed soils, and their abundance is more stable over time when Cr stress is present. At the stage of Cr_6d stress, however, the abundance of Actinobacteriota and Firmicutes decreased considerably, most likely because the concentration of heavy metals was higher than the community could tolerate, which inhibited growth and reduced the abundance of the community. Both Chloroflexi and Myxococcota have been demonstrated to be dominant in a variety of heavy metal-contaminated soils due to their tolerance and resistance to heavy metal stress (Berg et al., 2012; Hao et al., 2021); Chloroflexi, as low-nutrient bacteria, have an advantage in nutrient poor soils (Islam et al., 2019). Additionally, Actinobacteriota have been demonstrated to play a role in soil nutrient cycling (Narendrula-Kotha and Nkongolo, 2017), while Acidobacteriota are a dominant phylum in acidic environments or locations that are contaminated with heavy metals, using a variety of nitrogenous compounds as nitrogen sources for their growth (Eichorst et al., 2018). Therefore, as the main bacterial phyla, Acidobacteriota and Actinobacteriota may control the N cycle in Cr-stressed soils.

According to the α-diversity results, the Shannon and Simpson indices significantly increased and then decreased, and initially decreased and then increased, respectively, which also revealed that the diversity of bacterial communities in soil exhibited a phase change from stress to stability over the Cr stress time series. On the one hand, this might be related to the fact that the total amount of heavy metals in soils contaminated with Cd, zinc (Zn), Pb, and Cr decreases the biomass and diversity of bacterial communities and even inhibits bacterial enzymatic activities (Chen et al., 2020; Tang et al., 2022). The stage-specific changes in bacterial communities can be characterized by the extensive substrate utilization properties and high metabolic activity of some bacterial communities, which are better adapted to the toxic effects in heavy metal-contaminated soils. On the other hand, changes in α-diversity may be closely related to soil nutrients such as N, P, and SOC (Soil Organic Carbon), as soil nutrients may be the main source of energy in promoting the growth and metabolic processes of bacterial communities (Li et al., 2021). Additionally, soil microbes are crucial for material and nutrient cycling in plant and soil ecosystems, but the mechanisms driving the composition and diversity of microbial communities vary depending on the kind of vegetation. According to research (Xu et al., 2021), vegetation traits and soil physicochemical features may synergistically regulate microbial communities in soils. Furthermore, because heavy metals have more complex natural origins and soil biogeochemical characteristics and result in complex interactions between microbial communities, multiple factors combine to determine whether an ecosystem stimulated by metals would produce a positive or negative response (Abdu et al., 2017). In summary, bacterial communities in soils adapt to Cr contamination through phase changes in composition, structure, and diversity; rather than solely being influenced by the toxic effects of heavy metals, these phase shift characteristics of bacterial community diversity in soil may be influenced by a number of factors, including heavy metal contamination, soil nutrients, and vegetation conditions.

According to the experimental data, there were significant differences in the composition structure and diversity of functional genes involved in the nitrogen cycle in soil under conditions of Cr stress.

The AOA-amoA and AOB-amoA abundances exhibited a significant decrease and then an increase (p < 0.05). In particular, the abundance of the AOB-amoA gene increased by 8.30% in the Cr_6d stage compared with the Cr_6h stage; the abundances of nifH, narG, and nirK showed an increasing tendency; and the gene abundance was ranked as denitrification > nitrogen fixation > nitrification. The results of α-diversity analysis revealed that from Cr_6h to Cr_6d, the diversity of the nitrogen cycle functional genes significantly decreased. Similar to the bacterial community diversity, the functional gene abundance displayed a stress to stability phase transition. This is most likely because high Cr pollution temporarily inhibits the nitrogen cycle function in soil, which is caused by the soil microbial community’s immediate damage response to external stimuli and the gradual recovery of the nitrogen cycle function with increasing Cr stress time (Zhou et al., 2016). It has been demonstrated that Cd stress promotes N cycling activities in soil and that the degree of stress, microbial response mechanisms, and other mechanisms change with time (Zhao et al., 2021), which is consistent with the present study’s findings. The initial and rate-limiting phase of soil nitrification is the ammonia oxidation process, which is mediated by AOA and AOB genes with significant changes in abundance. As specialized chemoautotrophic bacteria, AOA and AOB can oxidize ammonia to nitrite and generate huge amounts of ATP, which promotes the survival of microbial communities in soil (Prosser et al., 2020). Similar to other research, the results showed that AOA was more abundant than AOB, probably because AOA is the dominating microbe in soil ammonia oxidation (Leininger et al., 2006) and is better adapted to the challenging environment of low oxygen and oligotrophic growth (David et al., 2014). Although AOA and AOB populations exhibit ecological niche differentiation and functional complementarity (Prosser and Nicol, 2012), AOB populations are more vulnerable to Cr stress than AOA, and therefore their abundance changes more significantly.

In the Cr stress time series, the abundance of the nitrogen cycle functional genes generally increased. This may be explained by the fact that soil microbial communities in Cr-stressed environments protect themselves from metal toxicity using a variety of resistance mechanisms, such as intercellular isolation and enzymatic detoxification, while dominant phyla such as Actinomycetes and Acidobacteria may enhance the transcriptional processes of genes related to amino acid biosynthesis and uptake pathways for protein detoxification (Viti et al., 2014). Thus, as the succession process advances, the soil microbial community gradually adjusts to external pressures at the physiological, genotypic, or community level (Spang et al., 2012; Azarbad et al., 2015) and inevitably requires more energy and irreversible overall metabolic transformation in order to maintain higher energy expenditures in organism growth and metabolic activities (Singh et al., 2014); thus, the nitrogen use and transformation process may be expedited. However, during the Cr_6h phase, Cr stress inhibited bacterial communities and nitrogen cycle functional gene abundance in soil, probably because most denitrifying bacteria, diazotrophs, and soil bacteria are mostly heterotrophic and primarily take up ammonia to biosynthesize compounds required for metal detoxification. Based on the energy saving strategy of the soil microbial community, in order to save energy costs, the soil microbial community may inhibit processes that require high energy, such as cell division (Chen et al., 2016), so its abundance tends to decrease. In conclusion, Cr stress may accelerate the process of N utilization and transformation in soil, and soil microbial communities may adopt energy production and conservation strategies to survive in Cr-stressed environments.

The findings revealed that at the Cr_6d stage, all soil physicochemical parameters increased significantly (p < 0.05); pH had a greater impact on Chloroflexi, while the N, P, K, and EC contents primarily affected Gemmatimonadota and Bacteroidota, and the functional genes involved in the nitrogen cycle nifH and AOB-amoA were significantly correlated with EC, N, and K, with EC having the strongest effect. Similar to the findings of Xu et al. (2019), pH had an impact on the bacterial community and was significantly correlated with nifH functional genes. First, it is possible that pH directly imposes a certain degree of physiological restriction on soil microorganisms; its value outside of a certain range will change how microorganisms interact, which will have an impact on the abundance of microbial taxa that cannot grow individually (Lauber et al., 2009). Second, previous research has demonstrated that soil pH affects the transport and bioavailability of heavy metals (Liang et al., 2017), and in addition to influencing the composition of the microbial community, it also regulates the diversity and abundance of functional genes involved in the nitrogen cycle (Lauber et al., 2009; Rousk et al., 2010). Additionally, Kandeler et al. (2006) demonstrated that pH had little to no impact on the functional genes involved in denitrification, which is in line with the findings of the present study. A wide-ranging variable in soil, pH is strongly related to properties such as the moisture content and nutrient availability. By modifying soil conditions, pH may cause variations in microbial composition; consequently, other soil physicochemical parameters other than soil pH may also have a limiting impact on the spread of microbial communities. For example, it has been demonstrated that the microbial community and available potassium in soil are significantly and negatively correlated (Pan et al., 2020), which may result from the enrichment of microorganisms that prefer potassium, such as Paenibacillus that dominate the soil, thus reducing microbial diversity.

The nitrification process of soil bacteria is favored by an elevated N content in the soil (Ouyang et al., 2017); hence, AOB is strongly associated with N when the N concentration increases. One study demonstrated that AOB are susceptible to environmental factors, and high pH and EC directly affect the variety of AOB in soil (Wang and Gu, 2014). In contrast, AOA require less energy and ammonium than AOB, and they have mechanisms for adapting to pH and severe settings, making AOA better suited for surviving in soil environments that are stressed by heavy metals. In Cr-contaminated soils, the soil may constitute a boundary habitat affected by heavy metals at the initial stage of stress, when nutrients are scarce and the concentration of effective carbon and nitrogen is low, thus the colonization of nitrogen-fixing bacteria enhances the input of nitrogen into the soil (Schrama et al., 2012). Furthermore, the nifH genes in this study had a high initial abundance and gradually increased, indicating that the input of heterotrophic nitrogen may be crucial in the time series.