Microbial Agents Enhance Sugar Beet Yield and Quality as an Alternative to Chemical Fertilizers

Abstract

Sugar beet (Beta vulgaris L.) is an important economic crop and a primary source of sugar in northern China, characterized by strong stress tolerance and high nutritional value. Microbial inoculants can promote crop growth by regulating soil enzyme activities, enriching dominant beneficial bacterial genera in rhizosphere soil, and improving the availability of soil nutrients. This study aimed to investigate the role of microbial inoculants in sugar beet production and their potential to replace chemical fertilizers and put forward the scientific hypothesis that microbial inoculants can increase soil nutrients and improve the soil microenvironment. A two-year field experiment was conducted: in 2022, treatments with different application rates of Bacillus subtilis and Trichoderma spp. inoculants were set up to screen the optimal inoculant and its dosage (M1); in 2023, based on this optimal inoculant (M1), treatments with reduced chemical fertilizer input were established to explore the mechanisms underlying the maintenance of sugar beet yield and quality. The results showed that the M1N2 (75 kg/ha fertilizer and 20% less nitrogen fertilizer) treatment significantly increased nitrogen, phosphorus, and potassium agronomic use efficiencies by 91.48%, 51.94%, and 53.50%, respectively, compared with the control (CK). Soil urease, catalase, and sucrase activities were significantly enhanced by 14.57%, 66.84%, and 222.46%, respectively. The treatment also significantly increased the relative abundance of beneficial bacterial genera such as JG30-KF-CM45 and KD4-96, while sugar beet yield was significantly increased by 5.53% relative to the CK. This study provides a theoretical basis for the application of microbial inoculants and the reduction in chemical fertilizers in sugar beet production.

1. Introduction

Sugar beet (Beta vulgaris L.) is an important economic crop and a major source of sugar [1]. The increase in sugar beet yield is highly dependent on the application of chemical fertilizers [2]. However, their overuse leads to soil compaction, groundwater pollution [3], and low fertilizer use efficiency [4]. Microbial agents offer a promising solution to these issues. The application of microbial agents can modulate the abundance of microbial taxa—for instance, increasing the abundance of Proteobacteria at the phylum level from 37.69% to 75.86%, while reducing the abundance of non-dominant taxa such as Vermaphilobacteria from 13.34% to 1.98% [5]. This helps improve the soil micro-environment, further activating soil microbial activity. These microorganisms decompose nutrients that are not directly available to plants and convert them into bioavailable forms, thereby enhancing nutrient uptake efficiency. As a result, the available phosphorus content in soil increases significantly by 29.39%, and soil organic matter content rises by 14.19% [6,7]. In addition, microbial agents can secrete phytohormones such as auxins and gibberellins, promoting root cell division [8,9] and ultimately improving crop yield. Currently, microbial agents are being increasingly applied in sugar beet cultivation. For example, Pseudomonas fluorescens (B1) and Bacillus coagulans (B2) have been shown to significantly enhance sugar beet seedling survival rates [10]. Pseudomonas fluorescens Pf0-1 and Pseudochloromonas CECT 462 contribute to notable increases in sugar beet yield [11]. Saccharomyces cerevisiae, Candida albicans, and Saccharomyces vini exhibit significant efficacy in suppressing sugar beet powdery mildew [12]. Commercial bioformulations such as Fitosporin-M, Albit, and Vitaplan, developed based on Bacillus cohnii, can activate dominant microbial communities in the rhizosphere soil, significantly reduce the abundance and incidence of pathogenic fungi, enhance the activity of oxidoreductases and hydrolases, and improve stress tolerance in sugar beet to some extent [13]. Beyond the microbial agents mentioned above, Bacillus subtilis and Trichoderma spp. have also attracted extensive research attention [14].

During the seedling stage of crops, Trichoderma spp. directly interact with the root system, stimulating the production of phytohormone-like substances. By synthesizing indole-3-acetic acid (IAA, auxin), Trichoderma promotes cell division in the meristematic tissues of pepper plants and accelerates shoot growth, ultimately resulting in a significant 17.28% increase in plant height. Concurrently, the dominant fungal strain Trichoderma harzianum C2 becomes substantially enriched [15] and secretes abundant cysteine protease QID74. This enzyme regulates root architecture—significantly increasing root length by 59% compared to the control group (CK). The enrichment of Trichoderma harzianum C1 can activate metabolic pathways related to pest and disease resistance in crops, demonstrating notable efficacy in inactivating nematodes. It also secretes chitinases, glucanases, and proteases, which degrade nematode egg shells and body walls, thereby inhibiting egg hatching and the activity of second-stage juveniles (J₂) [16]. Furthermore, Trichoderma can establish endophytic communities in the rhizosphere and interact with other microorganisms [17], improving the rhizospheric microenvironment. By secreting organic acids such as gluconic acid and fumaric acid, it reduces the pH of the rhizosphere soil [18]. Trichoderma spp. significantly enhance crop yield. To some extent, Trichoderma can serve as a substitute for chemical fertilizers. Under reduced fertilizer application, inoculation with Trichoderma results in better performance in root length, root number, plant height, and yield of wheat compared to chemical fertilizer application alone, along with a significant increase in soil enzyme activity [19,20]. Studies by Molla et al. showed that, compared to the control, this approach not only reduces the need for synthetic fertilizers but also improves nutrient uptake efficiency and dry matter accumulation [21].

Bacillus subtilis improves soil aggregate structure and enhances the soil’s water and nutrient retention capacity. Following the application of this bacterial agent, the relative water content in crop leaves increases significantly by 19% compared to the control group (CK) [22]. Concurrently, it significantly enhances plant homeostatic regulatory capacity [23]. After the application of Bacillus subtilis, the content of indole-3-acetic acid (IAA, auxin) in the soil increases significantly by 92% [24], and soil sucrase and phosphatase activities are also markedly enhanced [25]. Furthermore, this bacterial agent optimizes the rhizosphere microbial community composition and improves nitrogen transformation efficiency [26], thereby promoting crop nutrient absorption and utilization [27]. Additionally, Bacillus subtilis exerts positive effects under abiotic stress conditions. For example, its strain SN3 can alleviate salt stress toxicity by increasing the activities of superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT), while also reducing the accumulation of reactive oxygen species (ROS) and malondialdehyde (MDA) [28].

Numerous studies have confirmed the positive effects of bio-inoculants on crop growth, yield, and soil nutrient status. However, systematic research on plant growth-promoting microorganisms (PGPMs) in sugar beet cultivation systems—especially those conducted under conditions of reduced chemical fertilizer input—remains scarce. This study proposes two scientific hypotheses: (1) Application of Bacillus subtilis or Trichoderma spp. inoculants at appropriate rates can improve the agronomic use efficiency of nitrogen (N), phosphorus (P), and potassium (K) in sugar beet by altering soil enzyme activities and soil nutrient contents; (2) Application of microbial inoculants at suitable dosages can maintain or even enhance sugar beet yield and quality under reasonable chemical fertilizer reduction by improving rhizosphere microbial community composition and optimizing the soil rhizosphere microecological environment. Accordingly, the objectives of this study are as follows: (1) to identify the optimal application scheme for Bacillus subtilis and Trichoderma spp. inoculants and (2) to elucidate the mechanism underlying the yield-promoting effect of Trichoderma spp. on sugar beet under reduced chemical fertilizer input.

2. Materials and Methods

2.1. Site Description

This experiment was a two-year field trial conducted in 2022 and 2023 at the experimental base of the Ulanqab Agriculture and Animal Husbandry Science Academy in the Inner Mongolia Autonomous Region (40°55′23.5″ N, 113°07′10.6″ E). The study area is characterized by a typical temperate continental semi-arid climate, with an average annual frost-free period of 113 days. Detailed precipitation and temperature data are presented in

Table 1. Precipitation and temperature at the experimental site in 2022.

Precipitation in 2022 (mm)	April	May	June	July	August	September
	15.4	17	82.6	57.3	191	11.2
Max. temp. (°C)	26.2	31.1	33.6	33.5	32.5	26.8
Min. temp. (°C)	−4.3	0.1	7.4	12.2	5.5	0.1

and

Table 2. Precipitation and temperature at the experimental site in 2023.

Precipitation in 2023 (mm)	April	May	June	July	August	September
	77.1	23.7	26.6	125.4	53.6	38.8
Max. temp. (°C)	26.8	28.8	31.9	32.7	30.4	28.5
Min. temp. (°C)	—6.4	2.2	7.3	11.8	8.7	2.8

. The soil type in the experimental area is classified as chestnut soil (According to the FAO World Reference Base for Soil Resources it corresponds to Kastanozems) [29], and the initial physicochemical properties of the 0–30 cm tillage layer are detailed in

Table 3. Nutrient profile of the test plots (2022).

Organic Matter (g·kg−1)	Total N (g·kg−1)	Total P (g·kg−1)	Total K (g·kg−1)	Avail N (g·kg−1)	Avail P (g·kg−1)	Avail K (g·kg−1)	pH
18.21	0.71	0.46	16.31	111.07	9.23	153.01	7.71

and

Table 4. Nutrient profile of the test plots (2023).

Organic Matter (g·kg−1)	Total N (g·kg−1)	Total P (g·kg−1)	Total K (g·kg−1)	Avail N (g·kg−1)	Avail P (g·kg−1)	Avail K (g·kg−1)	pH
20.2	0.67	0.39	12.47	109.27	11.44	163.01	8.02

(baseline soil fertility data for 2022–2023).

2.2. Experimental Design

The application rates of microbial inoculants in this experiment were determined based on the results of single-factor preliminary experiments conducted earlier by the research group, so as to confirm the application threshold and guide the corresponding gradient design. The tested microbial inoculants were Bacillus subtilis and Trichoderma spp. (viable count: 1 × 10⁸ CFU·g⁻¹; provided by Hebei Cangzhou Xingye Co., Ltd. (Cangzhou, China), all commercial microbial inoculants). In 2022, the following treatments were established: non-inoculant control (CK); Bacillus subtilis application rates of 375 kg·ha⁻¹ (K1), 750 kg·ha⁻¹ (K2), and 1125 kg·ha⁻¹ (K3); and Trichoderma spp. application rates of 75 kg·ha⁻¹ (M1), 150 kg·ha⁻¹ (M2), and 225 kg·ha⁻¹ (M3), three blank controls were set up: no nitrogen fertilizer (N0), no phosphorus fertilizer (P0), and no potassium fertilizer (K0). A total of 10 treatments were designed, with 3 replicates per treatment, resulting in 30 experimental plots. Each plot was 6 m in length and 5 m in width. Chemical fertilizers (pure nitrogen: 39.9 kg·ha⁻¹, phosphorus: 59.9 kg·ha⁻¹, potassium: 99.6 kg·ha⁻¹) and microbial inoculants were all applied once as base fertilizer before transplanting. The sugar beet variety used in the experiment was ‘IM1162’, which was raised by the paper tube seedling method [1] and transplanted to the field on 22 May 2022, with a row spacing of 60 cm and a plant spacing of 20 cm. During the growth period, irrigation was performed three times via a drip irrigation system (once every 30 days after transplanting, with a single irrigation rate of 450 m³·ha⁻¹, local precipitation was able to compensate for the insufficiency of irrigation, and the precipitation data was sourced from https://www.yanshou100.com/, accessed on 6 December 2025). Harvesting and determination of yield and quality were carried out on 3 October 2022.

The 2023 experiment was designed based on the results of 2022 (The experiment was conducted over two years, in 2022, the optimal biological inoculant and its optimal application rate were screened out), with Trichoderma spp. applied at 75 kg·ha⁻¹ (M1, the microbial inoculant and application rate with the best yield-increasing effect) as base fertilizer. The control group (CK) maintained conventional fertilization without reduction. According to the previous research results of the research group, a potassium application rate of 180 kg·ha⁻¹ in the soil can meet the nutrient requirements of sugar beet throughout the entire growth period [30,31]. Combined with the actual situation that the available potassium content in the experimental field was all above 150 mg·kg⁻¹, nitrogen and phosphorus reduction experiments were conducted. Experimental gradients were designed based on the results of preliminary experiments by the research group. The treatments included nitrogen reduction (M1N1: −10%, M1N2: −20%, M1N3: −30%) and phosphorus reduction (M1P1: −10%, M1P2: −20%, M1P3: −30%), with a total of 7 treatments. Plot design, irrigation scheme, and other management parameters were consistent with those in 2022. Harvesting and determination of yield and quality were carried out on 4 October 2023.

2.3. Sample Collection

Plant samples: Sampling was conducted four times throughout the growing season at 30-day intervals after transplantation, specifically at the seedling stage (40 days after transplantation), rosette rapid growth stage (70 days after transplantation), root and sugar accumulation stage (100 days after transplantation), and sugar accumulation period (130 days after transplantation). Five uniformly growing sugar beet plants were selected from each plot for the calculation of the agronomic efficiency and recovery efficiency of fertilizers.

Soil samples: Four samplings were performed during the four growth stages of sugar beet. Soil samples from 0 to 30 cm depth were collected, sealed in zip-lock bags, stored at 4 °C, and transported to the laboratory for soil analysis. Approximately 20–30 g of fresh soil was sieved through a 1.25 mm (16-mesh) sieve for determination of soil enzyme activity and nutrient content. All measurements were performed with three biological replicates.

2.4. Measurement Methods for Indicators

2.4.1. Determination of Soil Enzyme Activities

Soil urease was determined by the indophenol blue colorimetric method, invertase by the DNS (3,5-dinitrosalicylic acid) colorimetric method, neutral phosphatase by the disodium phenyl phosphate colorimetric method, and catalase by the potassium permanganate titration method; with reference to the method described in Soil Agrochemical Analysis [30].

2.4.2. Determination of Soil Nutrient Contents

Soil organic matter was determined by the potassium dichromate volumetric method (dilution-heating method); alkaline hydrolysable nitrogen was measured via the alkaline hydrolysis-diffusion method (with a pH 4.8 boric acid solution as the absorption indicator); available phosphorus was analyzed using the sodium bicarbonate method; and available potassium was determined by flame photometry, with reference to the method described in Soil Agrochemical Analysis [30].

2.5. Analysis of Microbial Diversity in Rhizosphere Soil

For the purpose of SCI paper writing, the high-throughput sequencing on the Illumina NextSeq 2000 platform (targeting the bacterial 16S rRNA gene and fungal ITS region) was commissioned to Shanghai Majorbio Biomedical Technology Co., Ltd. (Shanghai, China). The sequencing samples included two groups with three biological replicates each: the control group (CK) without microbial inoculant application and the treatment group (M1N2) treated with a microbial inoculant of the Trichoderma genus.

2.5.1. Sample DNA Extraction

The extraction of total genomic DNA from the microbial community in rhizosphere soil samples was performed following the manufacturer’s standard protocol of the E.Z.N.A.^® Soil DNA Kit (Omega Bio-tek, Inc., Norcross, GA, USA). The preliminary quality assessment of the extracted genomic DNA, including its integrity and degradation level, was conducted via 1% agarose gel electrophoresis (Figure 1). Meanwhile, the concentration and purity of the genomic DNA were determined using a NanoDrop 2000 nucleic acid analyzer (Thermo Fisher Scientific Inc., Waltham, MA, USA).

2.5.2. PCR Amplification and Construction of Sequencing Libraries

Using the genomic DNA extracted above as the amplification template, PCR amplification was performed, with the primer sequences listed in

Table 5. Primer Design.

Sequencing Region	Primer Name	Primer Sequence
338F_806R	338F	ACTCCTACGGGAGGCAGCAG
	806R	GGACTACHVGGGTWTCTAAT
ITS1F_ITS2R	ITS1F	CTTGGTCATTTAGAGGAAGTAA
	ITS2R	GCTGCGTTCTTCATCGATGC

. The PCR amplification system and reaction program are detailed in

Table 6. Reaction System.

20 μL Reaction System
5× FastPfu Buffe	4 μL
2.5 mM dNTPs	2 μL
Forward Primer (5 μM)	0.8 μL
Reverse Primer (5 μM)	0.8 μL
FastPfu Polymerase	0.4 μL
BSA	0.2 μL
Template DNA	10 ng
Add ddH2O to	20 μL

and

Table 7. Reaction Program.

PCR Procedure Steps	Number of Cycles
95 °C for 3 min	1
95 °C for 2s	30
55 °C for 30 s	30
72 °C for 45 s	30
72 °C for 10 min	1

, respectively. The PCR products were examined via 2% agarose gel electrophoresis, and the corresponding electrophoresis results are presented in Figure 2. The target PCR products were recovered and purified using a DNA Gel Extraction and Purification Kit (PCR Clean-Up Kit, Yuhua Biotech, Shanghai, China). Subsequently, the concentration and quality of the purified products were quantified and verified using a Qubit 4.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA, USA). Library construction for the purified PCR products was conducted with the NEXTFLEX Rapid DNA-Seq Kit, following the four-step protocol below: (1) Adapter ligation, where sequencing adapters were covalently linked to the ends of purified PCR fragments; (2) Removal of adapter-dimer fragments through magnetic bead-based size selection to eliminate non-specific ligation products; (3) Enrichment of library templates via supplementary PCR amplification to enhance the concentration of qualified library fragments; (4) Recovery of the amplified library products using magnetic beads to obtain the final sequencing-ready libraries. High-throughput sequencing of the constructed libraries was carried out on the Illumina NextSeq 2000 platform, with the service commissioned to Shanghai Majorbio Biomedical Technology Co., Ltd. (Shanghai, China).

2.5.3. Sample Quality Control

Quality control and preprocessing of the paired-end raw sequencing reads were performed using the fastp software [31] (access link: https://github.com/OpenGene/fastp, version: 0.19.6, accessed on 6 December 2025). Specifically, a 50 bp sliding window was applied to evaluate base quality; if the average base quality within the window was lower than Q20, all downstream bases starting from the window’s initial position were truncated. Meanwhile, bases with quality values below Q20 at the read tails were filtered out. Subsequently, short fragments with lengths less than 50 bp after quality control and invalid sequences containing N bases were removed. Thereafter, the FLASH read merging tool [32] (access link: https://ccb.jhu.edu/software/FLASH/, version: 1.2.11, accessed on 6 December 2025) was employed to assemble paired-end (PE) reads based on their overlap characteristics. The minimum overlap length was set to 10 bp, and the maximum mismatch ratio in the overlapping region of merged reads was limited to 0.2 to eliminate abnormal assembly products. Sample differentiation and sequence orientation correction were then completed according to the barcode tags and primer sequences at the 5′ and 3′ ends of the reads, where exact matches (allowing 0 mismatches) were required for barcode sequences, and a maximum of 2 mismatches were permitted for primer sequences.

After sequence quality control and merging, denoising of the optimized reads was implemented using the built-in DADA2 module [33] under the QIIME2 analysis pipeline [34] with default parameters, yielding Amplicon Sequence Variants (ASVs). To eliminate the interference of sequencing depth differences on subsequent Alpha diversity analysis, the number of ASV sequences in all samples was rarefied to 20,000. After rarefaction, the Good’s coverage of each sample still reached 99.09%. Finally, taxonomic annotation of ASVs was accomplished using the Naive Bayes classifier integrated into the QIIME2 platform.

2.5.4. Analysis Pipeline

All bioinformatics analyses in this study were performed on the Majorbio Cloud Analysis Platform [35] (access address: https://cloud.majorbio.com, accessed on 6 December 2025), and the specific analysis pipeline was as follows: The diversity calculation module built into the mothur software [8] (http://www.mothur.org/wiki/Calculators, version: 1.48.0, accessed on 6 December 2025) was used to calculate Alpha diversity indices including Chao1 and Shannon. Principal Coordinate Analysis (PCoA) was conducted based on the Bray–Curtis distance algorithm to characterize the overall similarity of microbial community structures among different samples; meanwhile, the PERMANOVA non-parametric test was coupled to further verify the statistical significance of differences in microbial community structures among sample groups. The Linear discriminant analysis Effect Size (LEfSe) method [36] (access address: http://huttenhower.sph.harvard.edu/LEfSe, version: 1.0.8, accessed on 6 December 2025) was adopted, with the LDA discriminant threshold set to >2 and the significance level set to p < 0.05, to screen and identify specific bacterial taxa with significant abundance differences among different treatment groups at the taxonomic levels from phylum to genus.

2.6. Determination of Sugar Beet Yield and Quality

Sugar beet yield was measured at harvest by harvesting four rows per treatment per replication, with the results converted to yield per hectare. The sampling and cutting method of tuberous roots referred to the protocol described by Zhang Bowen et al. [37]

Sugar Content (%): Fifteen sugar beet plants were selected from each yield measurement plot for sugar extraction and pressing. Sugar degree was measured using an ATAGO refractometer (The refractometer used for determining sugar content was produced by ATAGO Co., Ltd. The city of origin of this device is Tokyo Metropolis, and its country of origin is Japan).

Sugar Yield (kg/hm²) = Yield (kg/hm²) × Sugar Content (%)

Nitrogen Agronomic Efficiency (kg/kg) = (Yield under nitrogen treatment − Yield under no nitrogen treatment)/Nitrogen application rate

Nitrogen Recovery Efficiency (%) = (Plant nitrogen uptake in nitrogen-fertilized plot − Plant nitrogen uptake in non-nitrogen-fertilized plot)/Nitrogen application rate × 100

The determination methods for the agronomic use efficiency and absorption use efficiency of nitrogen (N), phosphorus (P), and potassium (K) in sugar beets were adapted from the protocols reported by Li, Z. and Marschner, P. et al., and further optimized in accordance with the actual experimental conditions [38,39]. The calculation formulas for phosphorus and potassium fertilizer efficiency are the same as above.

2.7. Data Analysis

These graphs were generated using Origin 2018 software. Statistical analysis was per-formed in SPSS 19.0.25.0, with significant differences analyzed by the LSD test and relationships between variables were assessed using Pearson correlation analysis (p ≤ 0.05).

3. Results

3.1. The Effect of Microbial Agents on the Yield and Quality of Sugar Beet

Following the application of microbial agents in the 2022 trial, we found that sugar beet yield under all treatments was significantly higher than that in the CK (p ≤ 0.05). For Bacillus subtilis applications, the K2 treatment yielded the best results, with a significant increase in yield of 6.97% and a significant increase in sugar yield of 8.50% compared to CK (p ≤ 0.05; Figure 3a,b). For Trichoderma spp. applications, the M1 treatment was optimal, resulting in a significant yield increase of 8.50% and a significant sugar yield increase of 16.48% compared to CK (p ≤ 0.05; Figure 3a,b). The sugar content, however, remained generally consistent with CK across treatments (Figure 3c). Based on yield and sugar yield, the M1 treatment was identified as the most effective among the two bio-inoculants.

In the 2023 trial, the previously identified optimal agent and application rate (M1) served as the baseline. Building upon this, fertilizer reduction treatments were implemented. In the nitrogen (N) reduction experiment, yield exhibited an initial increase followed by a decrease as the N reduction level increased (Figure 3d). Specifically, compared to CK, yield was significantly reduced by 10.31% under M1N1 (p ≤ 0.05), significantly increased by 5.53% under M1N2 (p ≤ 0.05), and significantly decreased by 5.40% under M1N3 (p ≤ 0.05). The trend in sugar yield mirrored that of overall yield (Figure 3e), with M1N1 showing a significant decrease of 2.14% (p ≤ 0.05), M1N2 a significant increase of 9.87% (p ≤ 0.05), and M1N3 a significant decrease of 1.48% (p ≤ 0.05) compared to CK. In the phosphorus (P) reduction experiment, yield under the M1P1 treatment was significantly reduced by 9.61% compared to CK (p ≤ 0.05). While a subsequent increasing trend in yield was observed with further P reduction, the final yield and sugar yield under these treatments were not significantly different from CK (Figure 3d,e). Notably, sugar content across all fertilizer reduction treatments was not compromised and remained consistent with CK.

3.2. Effect of Microbial Agents on the Agronomic Use Efficiency of Fertilizers in Sugar Beet

Both microbial agents significantly enhanced the fertilizer agronomic efficiency (FAE) of nitrogen (N), phosphorus (P), and potassium (K) in sugar beet (Figure 4a–c). Under the application of Bacillus subtilis, the FAE of N, P, and K initially increased and then decreased with increasing Bacillus subtilis application rates. Specifically, in the K1 treatment, the FAE of N, P, and K increased significantly by 166.08%, 884.32%, and 103.92%, respectively, compared to CK (p ≤ 0.05). The highest improvement was observed in the K2 treatment, with significant increases of 388.28%, 220.48%, and 219.67%, respectively (p ≤ 0.05). Under the K3 treatment, the FAE remained significantly higher than CK, with increases of 286.78%, 164.52%, and 163.92%, respectively (p ≤ 0.05). In contrast, under Trichoderma spp. application, the FAE of N, P, and K showed a gradual decreasing trend with increasing Trichoderma spp. application rates. The most substantial enhancement was achieved with the M1 treatment, which resulted in significant increases of 906.43%, 514.48%, and 512.69% in the FAE of N, P, and K, respectively, compared to CK (p ≤ 0.05). Between the two bio-inoculants, the M1 treatment demonstrated the best overall performance in improving fertilizer agronomic efficiency.

In the fertilizer reduction experiment, under nitrogen reduction treatments, the FAE of N, P, and K initially increased and then decreased as the N reduction level increased (Figure 4d–f). The M1N2 treatment showed the most pronounced effect, with significant increases of 918.48%, 519.40%, and 535.00% in the FAE of N, P, and K, respectively, compared to CK (p ≤ 0.05). In the phosphorus reduction experiment, all treatments either showed FAE values significantly lower than CK or were not significantly different from CK (p > 0.05; Figure 3d–f).

3.3. Effect of Microbial Agents on the Fertilizer Uptake Efficiency in Sugar Beet

Both microbial agents significantly enhanced the fertilizer absorption and utilization efficiency (FAUE) of nitrogen (N), phosphorus (P), and potassium (K) in sugar beet across all treatments. For Bacillus subtilis applications, the FAUE of N, P, and K initially increased and then decreased with increasing Bacillus subtilis application rates (Figure 5a–c). Specifically, under the K1 treatment, the FAUE of N, P, and K increased significantly by 12.39%, 24.61%, and 38.19%, respectively, compared to CK (p ≤ 0.05). The greatest improvement was observed under the K2 treatment, with significant increases of 35.51%, 43.25%, and 70.60%, respectively (p ≤ 0.05). Under the K3 treatment, the FAUE remained significantly higher than CK, showing increases of 17.35%, 17.34%, and 39.42%, respectively (p ≤ 0.05). In contrast, under Trichoderma spp. applications, the FAUE of N, P, and K exhibited a decreasing trend with increasing Trichoderma spp. application rates (Figure 5a–c). The most substantial enhancement occurred under the M1 treatment, which resulted in significant increases of 107.30%, 41.46%, and 183.28% in the FAUE of N, P, and K, respectively, compared to CK (p ≤ 0.05).

In the fertilizer reduction experiment, the M1N2 treatment demonstrated the highest improvement in FAUE (Figure 5d–f), with significant increases of 82.76%, 58.38%, and 203.23% in the FAUE of N, P, and K, respectively, compared to CK (p ≤ 0.05). The remaining treatments generally showed no significant difference from CK (p > 0.05).

3.4. Effect of Microbial Agents on Soil Enzyme Activity

Soil enzyme activity serves as a key biological indicator of soil ecosystem functionality and an important basis for evaluating soil nutrient transformation efficiency [40]. The sugar beet growth cycle is divided into four stages: seedling stage (40 days after transplanting), rosette rapid growth stage (70 days after transplanting), Root and sugar accumulation stage (100 days after transplanting), and sugar accumulation stage (130 days after transplanting). Dynamic monitoring of soil enzyme activity across these stages was conducted to elucidate the underlying mechanisms for yield enhancement. Analysis of the 2022 soil enzyme activity data (Figure 6a–d) revealed that soil urease activity under Trichoderma treatments was generally superior to that under Bacillus subtilis treatments (Figure 6a). The application of Trichoderma enhanced soil urease activity, peaking during the critical nitrogen demand period (70–100 days, from rosette rapid growth to Root and sugar accumulation stage). Under the M1 treatment, soil urease activity increased significantly by 16.28% and 14.29% (p ≤ 0.05) compared to CK at 70 and 100 days, respectively. This provided a stable nitrogen foundation for high sugar beet yield. In contrast, Bacillus subtilis treatments had a limited effect on improving urease activity, demonstrating weaker potential in enhancing nitrogen use efficiency. This finding not only clarifies the central role of Trichoderma in substituting chemical nitrogen fertilizers with microbial agents but also establishes a theoretical basis for subsequent fertilizer reduction experiments. In the fertilizer reduction experiment (Figure 6e), the synergistic effect between Trichoderma and nitrogen reduction measures was significantly superior to that with phosphorus reduction. Under the M1N2 treatment, soil urease activity was significantly higher than CK (p ≤ 0.05) throughout the growth period (except at 100 days, where an increase was observed, though not statistically significant [p > 0.05]), with respective increases of 28.29%, 26.20%, 9.37%, and 14.57% (p ≤ 0.05) compared to CK. In contrast, the combination of Trichoderma with phosphorus reduction generally did not significantly promote urease activity (p > 0.05), indicating limited synergistic potential.

Soil neutral phosphatase activity was generally significantly lower than (p ≤ 0.05) or showed no significant difference from CK (p > 0.05) under all Bacillus subtilis treatments. In contrast, under Trichoderma spp. treatments, a decreasing trend was observed with increasing Trichoderma spp. application rates. However, under the M1 treatment, the activity was significantly higher than CK (p ≤ 0.05) throughout the growth stages (except at 70 days, where an increase was observed but was not statistically significant [p > 0.05]), with respective increases of 13.05%, 14.10%, 8.00%, and 20.17% compared to CK. In the fertilizer reduction experiment (Figure 6f), soil neutral phosphatase activity was significantly higher than CK (p ≤ 0.05) only at the 70-day growth stage across all treatments (except for M1P3), with significant increases of 55.45%, 55.85%, 59.93%, 54.93%, and 45.00% (p ≤ 0.05), respectively. During the rest of the growth period, the activities in all other treatments were either not significantly different from (p > 0.05) or significantly lower than CK (p ≤ 0.05), indicating a limited synergistic effect between fertilizer reduction and soil neutral phosphatase activity.

Under the Bacillus subtilis treatments, soil catalase activity generally exhibited a decreasing trend during the 70–103 day growth period with increasing Bacillus subtilis application rates (Figure 6c). In contrast, the performance of Trichoderma spp. was markedly superior. At 100 days after transplanting, soil catalase activity under the Trichoderma spp. treatments was significantly increased by 35.62%, 37.98%, and 34.73% (p ≤ 0.05) compared to CK, respectively.

In the fertilizer reduction experiment (Figure 6g), soil catalase activity under the M1N1 treatment was significantly higher than other treatments at 100 days, showing a significant increase of 20.26% (p ≤ 0.05) compared to CK. Under the M1N2 treatment, a significant increase of 66.84% (p ≤ 0.05) was observed at 130 days compared to CK. In the phosphorus reduction experiment, the treatments generally exhibited no significant synergistic effect with increasing phosphorus reduction rates.

Soil sucrase activity responded differently to the application of the two microbial agents (Figure 6d). Across the growth stages, with increasing application rates, the Bacillus subtilis treatments generally showed an initial increase followed by a decrease in activity, whereas the Trichoderma spp. treatments exhibited an initial decrease followed by an increase. Among them, the K2 treatment resulted in the greatest enhancement under Bacillus subtilis, while the M3 treatment was most effective under Trichoderma spp. In the fertilizer reduction experiment (Figure 6h), soil sucrase activity under the M1N2 treatment was significantly higher than CK (p ≤ 0.05) at most growth stages (except at 70 and 100 days, where increases were observed but were not statistically significant [p > 0.05]), with increases of 32.04%, 12.58%, 8.05%, and 222.46% (p ≤ 0.05) compared to CK, respectively. In contrast, phosphorus reduction measures showed a significantly inhibitory trend (p ≤ 0.05).

In conclusion, under the treatments of microbial agents and fertilizer reduction measures, the change in soil enzyme activity is not merely a simple increase or decrease, but rather exhibits complex dynamic changes across different growth stages. Dynamic monitoring of soil enzyme activity throughout the entire growth period facilitates a better understanding of the mechanisms influencing the formation of sugar beet yield and quality.

3.5. Effect of Microbial Agents on Soil Nutrients

Soil nutrient content exhibited significant interactive effects between treatment and growth stage in response to different microbial agents and fertilizer reduction measures (Figure 7). Specifically, in both the single-agent and fertilizer reduction experiments (Figure 7e), changes in soil organic matter content were not significantly different from CK (p > 0.05), with slight increases observed. Regarding soil alkali-hydrolyzable nitrogen content, performance under Trichoderma treatments was significantly superior to that under Bacillus subtilis treatments (Figure 7b). The content under all Bacillus subtilis treatments was either significantly lower than (p ≤ 0.05) or not significantly different from CK (p > 0.05). In contrast, following Trichoderma application, the content increased significantly (p ≤ 0.05) by 22.56% and 29.39% compared to CK at 40 and 70 days after transplanting, respectively. In the fertilizer reduction experiment (Figure 7f), the values were not significantly different from CK (p > 0.05).

For soil available phosphorus content (Figure 7c), significant increases (p ≤ 0.05) of 16.12%, 15.46%, and 18.23% compared to CK were observed at 70, 100, and 130 days, respectively, but only under the M3 treatment. All other treatments showed no significant difference from CK (p > 0.05). In the fertilizer reduction experiment (Figure 7g), soil available phosphorus content responded differently to fertilizer type and reduction rate: it showed an initial increase followed by a decrease with increasing nitrogen reduction, and a decreasing trend with increasing phosphorus reduction. Specifically, under the M1N2 treatment, the content increased significantly (p ≤ 0.05) by 22.44% and 15.51% compared to CK at 100 and 130 days, respectively. Soil available potassium content showed no significant difference from CK (p > 0.05) across all treatments and growth stages. However, among the single-agent treatments, the M1 treatment showed the greatest improving effect (Figure 7d), while the M1N2 treatment was the most effective in the fertilizer reduction experiment (Figure 7h).

3.6. Correlation Analysis of Yield and Related Indicators

Correlation analysis of yield and related indicators (Figure 8) revealed that sugar beet yield formation showed significant or highly significant positive correlations with the agronomic efficiency of N, P, and K fertilizers, nutrient absorption and utilization efficiency, and soil available potassium content. Based on these findings, we hypothesize that the primary mechanism by which microbial agents promote sugar beet yield lies in their ability to enhance fertilizer agronomic efficiency and nutrient absorption and utilization efficiency.

3.7. Changes in Rhizosphere Soil Bacterial and Fungal Diversity Under Microbial Agent Treatments

3.7.1. Effects of Microbial Inoculants on Soil Structure and Alpha Diversity Index

To clarify the mechanism underlying the improvement of yield and quality of sugar beets by microbial inoculants, this study employed Illumina high-throughput sequencing technology to compare and analyze the differences in microbial composition between the control group without microbial inoculants (CK) and the rhizosphere soil group treated with Trichoderma spp. (M1N2). As shown in Figure 9a, the bacterial community characteristics revealed that the total number of amplicon sequence variants (ASVs) in the CK group and M1N2 group were 486 and 471, respectively, with 388 shared ASVs; among them, the proportion of unique ASVs accounted for 20.16% in the CK group and 17.62% in the M1N2 group. Regarding the fungal community characteristics (Figure 9b), the total number of ASVs in the CK group and M1N2 group were 244 and 237, respectively, with 198 shared ASVs; the proportion of unique ASVs was 21.31% in the CK group and 18.99% in the M1N2 group. As shown in

Table 8. Effect of Microbial Agents on the Alpha Diversity Index of Soil.

		ACE	Chao	Shannon	Simpson
Bacteria	CK	1407.09 ± 15.11 a	1392.79 ± 15.65 a	6.59 ± 0.75 a	0.0022 ± 0.0002 a
	M1N2	1398.58 ± 16.69 b	1382.25 ± 13.67 b	6.58 ± 0.85 a	0.0024 ± 0.0004 a
Fungus	CK	477.63 ± 7.78 a	476.76 ± 7.18 a	4.67 ± 0.11 a	0.0201 ± 0.0581 a
	M1N2	449.52 ± 2.01 b	448.42 ± 2.71 b	4.61 ± 0.06 a	0.0191 ± 0.0031 a

Note: Each value represents the mean with n = 3. Lowercase letters following the values in each column indicate statistically significant differences at p ≤ 0.05. CK (no microbial agent applied), M1N2 (20% N fertilizer reduction).

, Alpha diversity can clearly reflect species diversity and abundance; after the application of Trichoderma spp., the ACE index, Chao1 index, and Simpson index of soil bacteria and fungi all exhibited the highest values in the CK group.

3.7.2. Effects of Microbial Inoculants on the Distribution of Rhizosphere Soil Bacteria and Fungi

The bacterial community heatmap revealed that, at the genus level (Figure 10a), the top ten dominant bacterial taxa in relative abundance were: Vicinamibacterales, Vicinamibacteraceae, *JG30-KF-CM45*, *KD4-96*, Gaiellales, Micrococcaceae, Rubrobacter, Sphingomonas, *MB-A2-108*, and Gaiella. Under the M1N2 treatment, the relative abundances of Vicinamibacteraceae, Gaiellales, Rubrobacter, and *MB-A2-108* decreased by 10.6%, 7.0%, 2.4%, and 11.4%, respectively, compared to CK. In contrast, the relative abundances of *JG30-KF-CM45*, *KD4-96*, Micrococcaceae, and Sphingomonas increased by 23.8%, 8.4%, 56.2%, and 5.6%, respectively.

The fungal community heatmap indicated that (Figure 10b), at the genus level, the top ten dominant fungal taxa in relative abundance were: unclassified_k__Fungi, Mortierella, Gibberella, Gibberellulopsis, unclassified_f__Chaetomiaceae, Humicola, Plectosphaerella, Chaetomium, Thielavia, and Pseudombrophial. Under the M1N2 treatment, the relative abundances of Mortierella, Humicola, Chaetomium, Thielavia, and Pseudombrophila decreased by 24.8%, 24.2%, 37.0%, 44.2%, and 9.7%, respectively, compared to CK Meanwhile, the relative abundances of unclassified_k__Fungi, Gibberella, Gibberellulopsis, Chaetomiaceae, and Plectosphaerella increased by 5.5%, 69.3%, 79.0%, 23.2%, and 4.1%, respectively.

3.7.3. Correlation Analysis of Bacteria, Fungi and Environmental Factors Under Microbial Inoculant Treatments

At the genus level of bacterial colonies (Figure 11a), the correlation coefficients between the top 10 dominant microorganisms and environmental factors reveal that organic matter, available potassium, and soil urease exhibit highly significant positive correlations with Vicinamibacteraceae and MB-A2-108, while showing highly significant negative correlations with JG30-KF-CM45 and Sphingomonas. In contrast, alkali-hydrolyzable nitrogen, available phosphorus, soil neutral phosphatase, catalase, and sucrase demonstrate highly significant positive correlations with JG30-KF-CM45 and Sphingomonas, while showing highly significant negative correlations with Vicinamibacteraceae and MB-A2-108.

At the genus level of fungal colonies Fungi (Figure 11b), the correlation coefficients between the top 10 dominant microorganisms and environmental factors indicate that organic matter, available phosphorus, and soil urease exhibit highly significant positive correlations with Chaetomium and Pseudombrophila, while showing highly significant negative correlations with Fungi and Gibberellulopsis. Conversely, alkali-hydrolyzable nitrogen, available phosphorus, soil neutral phosphatase, soil catalase, and sucrase demonstrate highly significant positive correlations with Fungi and Gibberellulopsis, while showing highly significant negative correlations with Chaetomium and Pseudombrophila.

4. Discussion

Through a two-year field experiment, this study clarified the regulatory effects of microbial inoculants including Bacillus subtilis and Trichoderma spp. on sugar beet growth and soil environmental properties, as well as their compatibility with chemical fertilizer reduction strategies, with its core findings as follows: the Trichoderma treatment (M1) achieved a significant 16.29% yield increase in sugar beet in the 2022 experiment, and in the 2023 chemical fertilizer reduction experiment, the M1N2 treatment obtained the highest yield with a 5.53% increase compared with the control group, which verified the feasibility of integrating microbial inoculants with chemical fertilizer reduction; under the M1 treatment, the nitrogen, phosphorus, and potassium absorption efficiencies of sugar beet were increased by 107.30%, 41.46%, and 183.28%, respectively, while under the M1N2 treatment, the increments of the three nutrients reached 82.76%, 58.38%, and 203.23%, significantly enhancing the crop’s nutrient acquisition capacity; the M1 treatment could significantly elevate the activities of urease, catalase, phosphatase, and sucrase during the 100–130-day growth stage of sugar beet, and under the M1N2 treatment, the activities of all the above-mentioned enzymes except neutral phosphatase showed an increasing trend, with a positive correlation observed between neutral phosphatase activity and soil available phosphorus content; microbial inoculants also significantly restructured the rhizosphere microbial community—in the bacterial colonies, the relative abundances of JG30-KF-CM45, KD4-96, Micrococcaceae, and Sphingomonas increased by 23.80%, 8.40%, 56.20%, and 5.60%, respectively, and in the fungal colonies, the relative abundances of unclassified_k__Fungi, Gibberella, Gibberellulopsis, Chaetomiaceae, and Plectosphaerella increased by 5.50%, 69.30%, 79.00%, 23.20%, and 4.10%, respectively; under the M3 treatment, the soil available phosphorus content during the 70–130-day growth stage of sugar beet was increased by 16.12%, 15.46%, and 18.23% compared with the control group, respectively, while under the M1N2 treatment, the soil available phosphorus content during the 100–130-day growth stage of sugar beet was increased by 22.44% and 15.51% relative to the control group, respectively.

These results indicate that microbial inoculants promote sugar beet growth not merely by means of nutrient supplementation [41], but rather through the multi-pathway synergistic effect of “enzyme activity regulation-nutrient activation-microbial community optimization” to construct a rhizosphere microecosystem suitable for sugar beet growth [42,43,44]. Among all treatments, the M1N2 treatment exhibited the optimal comprehensive effect, which not only achieved stable yield increase under the context of chemical fertilizer reduction, but also simultaneously improved soil fertility and microbial community structure. This demonstrates that this inoculant application mode can effectively balance crop yield demands and farmland ecological benefits, providing a practical and feasible technical scheme for chemical fertilizer reduction and substitution. The findings of this study are highly consistent with those of existing similar research: Singh et al. (2015) [45] confirmed that Trichoderma spp. could increase the nitrogen (N), phosphorus (P), and potassium (K) absorption efficiencies of sugarcane by 27.0%, 65.0%, and 44.0%, respectively, while the magnitude of improvement in nutrient absorption efficiency of sugar beet by inoculants in this study was higher, which is speculated to be related to the inherent nutrient demand characteristics of sugar beet and field cultivation conditions; the study by Wang et al. [46] verified that compared with the control group (CK), the application of Trichoderma spp. could increase the activities of sucrase, catalase, phosphatase, and urease in rhizosphere soil by 387.6%, 16.0%, 130.5%, and 38.0%, respectively, and the increase in enzyme activities under the M1 treatment in this study was consistent with this result, which corroborates the universality of the positive regulatory effect of inoculants on the soil enzyme system. Syafruddin et al. proposed that the core mechanism underlying the yield-promoting effect of microbial inoculants is the supply of available phosphorus [47], and in this study, the M1N2 treatment significantly increased soil available phosphorus content with a positive correlation with yield, verifying the applicability of this mechanism in sugar beet crops. Meanwhile, this study also fills the gap in research on inoculant application in the sugar beet field: previous studies on Bacillus subtilis, Trichoderma spp., and sugar beet mostly focused on disease resistance, whereas this study clarifies their yield-increasing and soil-improving effects under the scenario of chemical fertilizer reduction, thus improving the application value of inoculants in sugar beet production. From the perspective of physiological and ecological mechanisms, the regulatory effect of microbial inoculants on sugar beet can be divided into three levels: first is the nutrient activation mechanism, where Trichoderma spp. can secrete low-molecular-weight organic acids and chitinases to decompose insoluble phosphate through chelation, redox reactions, and acidification processes [48,49], and simultaneously induce roots to synthesize nitrogen transport proteins [45,50] and activate mineral potassium, realizing the synergistic and efficient utilization of N, P, and K nutrients; second is the enzyme-nutrient synergistic mechanism, where inoculants can retain phosphatase via hyphal adsorption and secrete signaling substances to promote the synthesis of urease and other enzymes, and the enhancement of soil enzyme activity further accelerates organic nitrogen mineralization, phosphorus hydrolysis [51,52,53], and carbon cycling [46], forming a positive feedback loop of “inoculant-enzyme-nutrient”; third is the rhizosphere microecosystem regulation mechanism, where inoculants can enrich beneficial rhizosphere bacteria such as JG30-KF-CM45, optimize the microbial community structure, and are speculated to potentially activate the crop JA/ET signaling pathway and promote lateral root elongation [54,55], which not only expands the nutrient absorption range but also enhances plant stress resistance, ultimately achieving the synergistic improvement of growth and nutrient utilization [56,57]. At the scientific and practical levels, this study clarifies the coupling relationship of “microbial inoculant-soil enzyme-rhizosphere microbe-crop nutrient absorption”, improves the theoretical system of inoculant-regulated crop-soil systems, and provides a new theoretical perspective for understanding the synergistic effect of inoculants and chemical fertilizer reduction. Meanwhile, the screened M1 and M1N2 inoculant application modes can maintain or even increase sugar beet yield and improve soil fertility under chemical fertilizer reduction conditions, providing a feasible and implementable technical scheme for the green production of sugar beet, as well as practical reference for chemical fertilizer reduction and efficiency enhancement of cash crops in northern arid areas and the large-scale promotion of microbial inoculants.

This study still has certain limitations. First, the experiment only conducted a two-year field trial without investigating the long-term sustainability of the inoculant effects; additionally, the experimental region was relatively single, so the regional universality of the results remains to be verified. Second, the research focused on mechanism analysis at the physiological and ecological levels, lacking in-depth exploration at the molecular level (such as gene expression and metabolomics), which makes it impossible to clarify the molecular pathways through which microbial inoculants regulate sugar beet growth. Third, no research was carried out on the compatibility of inoculants with different soil types and climatic conditions, which limits the precise promotion of the technical scheme. For future research, multi-regional and long-term in situ experiments can be carried out to clarify the spatiotemporal stability and regional adaptability of inoculant effects. Moreover, transcriptomics and metabolomics technologies can be combined to analyze the molecular mechanisms underlying the regulation of sugar beet nutrient absorption and growth by inoculants, excavate key regulatory genes, and further elucidate the mechanism by which microbial inoculants improve crop yield at the molecular level.

5. Conclusions

The findings of this study demonstrate that the proper application of inoculants containing Bacillus subtilis or Phanerochaete (white-rot fungi) represents an effective strategy to enhance the agronomic efficiency of nitrogen (N), phosphorus (P), and potassium (K) in sugar beet cultivation. Heterogeneous effects were observed in the impacts of microbial inoculants with different types and dosages on sugar beet nutrient use efficiency and soil properties, among which the Trichoderma treatment (M1) achieved the optimal comprehensive benefits. Notably, the M1N2 treatment significantly improved the utilization efficiency of N, P, and K, while concurrently inducing a marked increase in the activities of soil urease, catalase, and sucrase. Additionally, the relative abundances of beneficial rhizosphere bacteria, including JG30-KF-CM45 and KD4-96, were substantially enriched, and sugar beet yield was elevated by 5.53% significantly. Collectively, these results confirm that the evaluated bioinoculants promote key processes associated with nutrient mineralization and microbial activity, thereby fostering a more functional rhizosphere environment that supports sugar beet growth. This study provides an important theoretical basis for the popularization and application of microbial inoculants in the green production of sugar beets.