Effect of Lycium barbarum L. Root Restriction Cultivation Method on Plant Growth and Soil Bacterial Community Abundance

Abstract

The root restriction of protected cultivation has been widely used to increase productivity and sustainability in modern agriculture. However, there have been few studies of wolfberry (Lycium barbarum) root restriction, and it is cultivated mostly by clean tillage. In this study, we measured the growth of Lycium barbarum and the composition and diversity of the bacterial community and soil properties of L. barbarum under different cultivation methods with root restriction. The results showed that the X60 root-restriction treatment significantly increased the canopy size (east–west), leaf length, leaf width, the number and length of new branches, and the concentrations of chlorophyll and K in L. barbarum. The concentrations of N, P, and K in the root-restriction groups were all higher than those in CK. However, the ratio of N:P was greatest in the CK plants and least in X80, indicating that X80 had a relatively weak effect on the balance of N:P. In addition, root restriction improved fruit quality by increasing soil organic matter and organic carbon and also improved fertilization efficiency to promote plant growth. Moreover, high-throughput sequencing showed that the abundance of soil bacteria under root-restriction cultivation was significantly higher than that in CK. Furthermore, the total abundance of the top 10 bacterial genera was greatest in the X60 treatment. Redundancy analysis showed that total N, total P, total K, and total organic matter were the major soil factors that affected the bacterial community. A comprehensive comparison showed that root-restriction cultivation improved the growth of L. barbarum but reduced the abundance and diversity of the soil bacteria. The X60 treatment yielded the best results on plant growth. Our findings provide an empirical reference for root-restriction cultivation of L. barbarum of an appropriate width.

1. Introduction

Lycium barbarum L. (Solanaceae) is often referred to as the Ningxia Wolfberry, a general term for the fruit of species in the genus Lycium, including L. chinense [1]. Lycium barbarum is valuable as its wolfberries are rich in nutrients, such as vitamins, proteins, minerals, bioactive polysaccharides, polyphenols, carotenoids, and amino acids [2]. It is an edible medicinal plant that can reduce blood lipids, blood pressure, and blood sugar [3]. Moreover, wolfberries are widely used in Chinese traditional medicine and as antioxidants, with anti-aging and anti-tumor functions; they nourish the liver to improve visual acuity, maintain beauty, and strengthen immunity [4]. To satisfy the year-round demand for fresh wolfberries, new cultivation technology is needed to improve the productivity of the L. barbarum fruit industry.

Root-restriction cultivation uses a physical or ecological method to restrict the root system within a certain space or volume [5]. It is a new technique that regulates the vegetative and reproductive growth of the aboveground shoots by controlling the growth of the root system [6]. Although root-restriction cultivation inhibits root activity, it can promote biomass accumulation and potassium (K) uptake by L. barbarum [7]. It is well known that the vegetative growth of L. barbarum can promote growth, with concrete manifestations in its rapid growth of new branches, increase in crown diameter, and growth redundancy with only leaf development and no flowers. Zhang showed that restricting root growth by growing L. barbarum in a container could significantly increase photosynthesis in the leaves, stimulate the accumulation of soluble sugars and starch, and increase the mineral content in persimmon leaves [8]. Hong studied the effect of root-restriction cultivation on the growth of sweet cherry tree seedlings and showed that root restriction could control the vegetative growth, promote tree stunting, and stimulate lateral branch growth. In this way, root restriction promoted floral bud differentiation, advanced the flower formation, and increased the yield of high-quality sweet cherries [9]. Restricting the root area significantly improved the fruit color of the ‘Jumeigui’ grape, increased the mass fraction of soluble sugars, and increased the activity of enzymes involved in sugar accumulation and conversion in the fruit [10].

Root restriction is widely used in protected cultivation [11], which promotes the L. barbarum sprout growth and absorption of soil phosphorus. Thus, the yield of L. barbarum fresh fruit is impaired. There are many reports on the effects of root restriction on the temperature of the roots, photosynthetic characteristics, biomass accumulation, and nutritional values of plants [12,13]. However, the effect of root restriction of different widths on the bacterial community in the rhizosphere soil of L. barbarum has rarely been reported. Particularly, the mechanism by which root restriction affects the bacterial diversity and community composition in the rhizosphere soil needs to be elucidated. Thus, this study compared the growth characteristics of L. barbarum and the composition, distribution, and diversity of the bacterial community in the rhizosphere soil between plants cultivated under root restriction of different widths and no root restriction. Our goals were (1) to provide empirical evidence for root-restriction cultivation as a means of manipulating L. barbarum growth and (2) to determine the most effective width of root-restriction cultivation.

2. Materials and Methods

2.1. Plant Material and Study Site

The experimental plants were three-year-old ‘Ningqi No. 7’ trees of L. barbarum. All tree seedlings were planted in the same year and were managed together. The experimental site was in the Agricultural Technology Demonstration Park of Dryland Farming and Water Saving, Wangtuan Village, Tongxin County, Ningxia Hui Autonomous Region, China. The location is in the arid zone in central Ningxia, 36°58′48″ N, 105°54′24″ E, an area connecting the Ordos Plateau and the north Loess Plateau (Figure 1). The climate is a temperate continental climate, arid, with concentrated rainfall, strong evaporation, strong wind, and sufficient daylight. The average annual precipitation is 250–300 mm, concentrated in June–September (and accounting for 60% of the total annual precipitation). Annual evaporation is 2200–2600 mm. The average number of days with strong wind (wind speed ≥ 17 m/s) is 8–46 d, mostly in winter and spring. The annual average temperature is 8.7 °C, annual average thermal radiation is 600 kJ/cm², and the hours of daylight are 2700–3000 h per year. The site has loam soil with high viscosity and low permeability.

2.2. Experimental Design and Sample Collection

The field experiment was conducted in an orchard in 2018. Lycium barbarum trees, planted in 2018, were almost one-meter-high super seedlings, with a plant spacing of one meter and row spacing of three meters. The ditching machine was used to open the planting ditch, and the ditch depth was 70 cm, with both sides of the ditch covered with shed film. The decomposed sheep manure was mixed into the soil in a ratio of 4:1, and the soil with good organic fertilizer was filled into the planting ditch. The ditch width of three cultivation system treatments were 40 cm, 60 cm, and 80 cm (X40, X60, and X80, respectively) for root restriction. Each treatment was replicated three times. The control group was conventional cultivation with no root restriction (CK). In August 2018, the rhizosphere soil at 20 cm depth was collected from three randomly selected plants for all four treatment groups. Samples were collected from five points around the tree and mixed into one sample per tree. The mixed sample was divided into two samples: one was immediately cleaned to remove impurities, divided into multiple sub-samples, and frozen in liquid nitrogen, then removed from liquid nitrogen and stored at −80 °C. These samples were used to extract DNA from soil microorganisms for high-throughput sequencing. The other half of the mixed sample was dried, then stored in the dark and used to determine its physical and chemical properties. Three parallel samples were obtained for each treatment (coded as CKA, CKB, and CKC; X40A, X40B, and X40C; X60A, X60B, and X60C; and X80A, X80B, and X80C).

2.3. Measurement of Plant and Soil Parameters

2.3.1. Measurement of Growth Parameters

Measurements were taken in summer (mid-June, mid-July, and mid-August), which is the fruit-setting season. In the spring during budding, five trees were randomly selected, marked, and measured for each treatment group. The growth parameters measured were: plant height, trunk diameter, crown diameter, leaf area, leaf length, leaf width, and the number and length of current-growth branches, which are important for wolfberry yield/production. Each treatment was replicated three times. Each replicate contained 3 rows, with 80 trees in each row.

2.3.2. Measurement of Chlorophyll Content

Measurements were taken in early June, which is the budding and flowering season. A portable chlorophyll meter (SPAD-502, Konica Minolta Sensing, Inc., Tokyo, Japan) was used to measure the chlorophyll in leaves of L. barbarum between 8:00 and 8:30 am. Each treatment was replicated three times. Each replicate contained 3 rows, containing 6 trees in each tree, and one leaf was taken from each tree.

2.3.3. Measurement of Leaf N, P, and K Content

Measurements were taken in early June, which is the budding and flowering season. Samples were deactivated at 105 °C for 30 min and baked at 75 °C in a baking oven until a constant weight was reached. Samples were then treated with H₂SO₄/H₂O₂. Plant total N and P contents were measured using the Continuous-Flow AutoAnalyzer III (AA3, SEAL Company, Hamburg, Germany). Plant total K content was measured by proton magnetic resonance spectroscopy.

2.3.4. Measurement of Soil Physical and Chemical Properties

Measurements were taken in early June, which is the budding and flowering season. Soil water content (WC) was measured using time–domain reflectometry (TDR). Soil physicochemical properties were determined and analyzed according to the standard methods in China (Agricultural Chemistry Committee of China 1983). In brief, soil pH was determined with a glass electrode after suspending 10 g soil in a 25 mL deionized water suspension (1:2.5, w/v). The total N was measured by acid digestion according to the Kjeldahl method. Total P was determined by sodium hydroxide alkali fusion–molybdenum–antimony scandium colorimetric method. Total K was determined by atomic absorption spectrophotometry. Soil organic matter was determined via the K₂Cr₂O₇ oxidation–reduction titration method.

2.3.5. Determination of Soil Bacterial Communities

Measurements were taken in early June, which is the budding and flowering season. Soil bacterial communities were determined by sequencing on the Illumina HiSeq platform (Biomarker Cloud Technologies Co. Ltd. Beijing, China). Total genomic DNA from the soil samples was extracted using the conventional cetyltrimethylammonium bromide (CTAB) method, and DNA purity and concentration were determined using 1.0% agarose gel electrophoresis. DNA was diluted to 1 ng·μL⁻¹ and stored at −80 °C. Primers for amplification were: 515 f/806r (5′-GTGCCAGCMGCCGCGCGGTA-3′ and 5′-GGACTACHVGGGT WT-3′). Polymerase chain reaction (PCR) mix (30 μL) was: Phusion Master Mix (2×) 15 μL, primers (2 μM) 1.5 μL each, gDNA (1 ng·μL⁻¹) 10 μL, and H₂O 2 μL. The PCR program was: 98 °C 1 min; 98 °C 10 s, 50 °C 30 s, and 72 °C 30 s, 30 cycles; and 72 °C 5 min. PCR products were detected by 2% agarose gel electrophoresis. After purification and quality assessment, quantified and normalized PCR products were used for the construction of a DNA library. Library construction and sequencing were completed at Beijing Biomarker Cloud Technologies Co., Ltd. (Huangdao District, Qingdao, China). Sequences were clustered as operational taxonomic units (OTUs) on the basis of a similarity of ≥97%. OTUs were annotated using the Silva bacterial database. On the basis of alignment with the microbial reference database, the taxonomic information corresponding to each OTU was obtained, and the taxonomic rank of species was analyzed. Alpha diversity was calculated and analyzed on the basis of the OTU data.

2.4. Data Analysis

Microsoft Excel 2007 was used for data analysis and graph generation. SPSS software (Version 17.0, SPSS Inc., Chicago, IL, USA) was used for statistical analysis. Duncan’s multiple range test was conducted. Origin2017 was used to draw figures of N, P, and K content and N:P ratio in L. barbarum leaves under different root-restriction treatments. Data homogeneity and normality variances were examined by Levene’s and Shapiro–Wilk’s tests. One-way ANOVA and group means of data were compared using Tukey’s test (p < 0.05). The difference among means was determined using the least significant difference (LSD) (p < 0.05), as indicated by different letters. The Alpha diversity indices were calculated in QIIME (Quantitative Insights into Microbial Ecology). The data of soil physicochemical properties, bacterial total abundances, Alpha diversity indices, and bacterial taxa (phyla) in CK, X40, X60, and X80 systems were compared by Student’s t-test. A Canonical Redundancy Analysis (RDA) was performed to identify variables discriminating among treatments. In RDA, microbiological indicators were correlated with soil physicochemical and plant physiological to identify patterns across treatments.

3. Results and Analyses

3.1. The Growth of L. barbarum under Different Root-Restriction Widths

3.1.1. Growth Parameters of L. barbarum

Compared with CK, L. barbarum Ningqi No. 7 trees under at least one of the three root-restriction cultivation patterns grew significantly differently except in two parameters: leaf area and leaf length (

Table 1. Growth parameters of L. barbarum trees cultivated under different root-restriction treatments and under no root-restriction cultivation (CK) (mean ± SE).

Treatment	Plant Height	Trunk Diameter	Crown Diameter		Leaf Area (cm2)	Leaf Length (cm)	Leaf Width (cm)	Number of New Branches	Length of New Branches (cm)
			East–West (cm)	North–South (cm)
CK	170.88 ± 1.23 a	1.74 ± 0.03 a	60.06 ± 3.36 b	69.66 ± 4.09 b	15.11 ± 0.49 a	8.97 ± 0.44 a	2.08 ± 0.06 b	9.23 ± 0.17 c	43.28 ± 1.27 c
X40	153.91 ± 0.26 c	1.60 ± 0.07 ab	62.56 ± 1.61 b	55.43 ± 2.76 b	14.13 ± 0.38 a	8.58 ± 0.62 a	2.50 ± 0.13 ab	10.58 ± 0.42 b	51.90 ± 2.24 b
X60	169.54 ± 0.49 a	1.58 ± 0.03 b	73.06 ± 2.84 a	63.15 ± 1.58 a	14.72 ± 0.46 a	9.46 ± 0.25 a	2.64 ± 0.20 a	13.33 ± 0.45 a	59.61 ± 1.90 a
X80	164.79 ± 2.00 b	1.63 ± 0.03 ab	68.83 ± 3.47 ab	67.28 ± 3.53 a	14.64 ± 0.72 a	9.41 ± 0.27 a	2.51 ± 0.12 ab	9.53 ± 0.12 bc	54.41 ± 1.59 ab

Note: X40, 40 cm root restriction; X60, 60 cm root restriction; X80, 80 cm root restriction. Different lower-case letters represent a significant difference at p < 0.05.

). The average plant height, trunk diameter, and crown diameter (north–south) were greater in CK, but the crown diameter (east–west), leaf width, new branch number, and new branch length were less in CK. Overall, trees cultivated under X60 root restriction grew the best in the growth parameters important for wolfberry fruit production. The crown diameter (east–west), leaf length, leaf width, and new branch number and length were the highest in X60, followed by X80. The plant height, crown diameter, and leaf area were lowest in X40. An investigation of the levels of all nine growth parameters showed that some parameters in CK were higher than those under the root-restriction treatment. In the year under study, root-restriction treatment significantly increased the number of new branches and new branch length, which was a significant productivity advantage compared with CK. Overall, treatment of root-restriction cultivation promotes the growth and development of wolfberry, indicating that root restriction was more conducive to the production of L. barbarum trees to a certain extent.

3.1.2. Chlorophyll Content in L. barbarum Leaves (SPAD)

As shown in Figure 2, different root-restriction treatments had different effects on the chlorophyll content (Soil Plant Analysis Development (SPAD)) in L. barbarum leaves. The root restriction treatments all had higher SPAD than CK, in the order X60 > X40 > X80 > CK. Among them, the chlorophyll content in X60 was significantly higher than that in X40, X80, and CK (p < 0.05), while the differences among X40, X80, and CK was not significant (p > 0.05).

3.1.3. N, P, and K Contents and N:P Ratio in L. barbarum Leaves

The levels of N, P, and K in the leaves of Ningqi No. 7 during the summer fruit-setting season were different among root-restriction treatments (Figure 3). The N, P, and K levels in the three root-restriction treatments were all significantly higher than those in CK (p < 0.05). N and P were highest in X40, and K was highest in X60, indicating that root-restriction cultivation could promote nutrient uptake in the leaves and the growth of Ningqi No. 7 trees. The N:P ratio followed the order of CK > X80 > X60 > X40; its value in CK was 36.58%, 20.26%, and 9.14% higher than that in X40, X60, and X80, respectively (Figure 3). In the root-restriction treatments, X80 had the lowest effect on the N:P balance, indicating that root-restriction cultivation could affect the N:P balance in the leaves. X80 treatment had the best effect among root restriction treatments, followed by X60.

3.1.4. Yield and Fruit Quality of L. barbarum under Different Root-Restriction Widths

The weight of single fruit decreased, but the single plant yield of L. barbarum under different root-restriction widths were increased, especially in X60 and X80. In addition, the overall nutritional quality of fruit increased, especially in total sugar, flavonoids, betaine, and ascorbic acid under different root-restriction widths compared with no root restriction, and the most prominent of all the treatments was X60 (

Table 2. Size and nutritional quality of L. barbarum fruit under different root-restriction treatments and no root-restriction cultivation.

Treatment	Longitudinal Diameter (mm)	Transverse Diameter (mm)	Single Fruit Weight (g)	Single Plant Yield (g)	Polysaccharides (g kg−1)	Total Sugar (g kg−1)	Carotenoid (g kg−1)	Flavonoid (g kg−1)	Betaine (g kg−1)	Ascorbic Acid (g kg−1)
CK	22.72 ± 1.67 a	12.97 ± 0.72 a	1.36 ± 0.02 a	255.66 ± 9.66 d	3.11 ± 0.51 a	119.21 ± 801 b	3.91 ± 0.07 a	1.05 ± 0.02 c	3.32 ± 0.06 c	0.21 ± 0.01 d
X40	19.52 ± 0.88 b	12.89 ± 0.69 a	1.18 ± 0.03 b	389.07 ± 3.87 c	2.59 ± 0.44 b	118.09 ± 4.56 b	3.92 ± 0.11 a	1.07 ± 0.04 c	9.91 ± 1.31 a	0.43 ± 0.02 c
X60	19.92 ± 1.02 b	12.00 ± 1.01 ab	1.22 ± 0.07 b	592.74 ± 13.56 a	2.84 ± 0.39 ab	126.42 ± 1.77 a	3.14 ± 0.08 b	1.33 ± 0.01 a	8.88 ± 0.61 b	0.56 ± 0.02 a
X80	19.80 ± 0.56 b	11.61 ± 0.82 b	1.01 ± 0.02 c	538.95 ± 16.33 b	3.01 ± 0.07 a	116.89 ± 8.07 b	3.21 ± 0.06 b	1.19 ± 0.02 b	8.61 ± 0.98 b	0.49 ± 0.01 b

Note: Mean values (x ± SE) followed by the same letter were not significantly different using LSD (p < 0.05) based on ANOVA of square-root-transformed data. Fruit size of L. barbarum (n = 20); Nutritional quality of L. barbarum (n = 3).

). Longitudinal diameter and single fruit weight under the X40, X60, and X80 treatments were significantly lower than under no root restriction by 14%, 12%, and 13% and 10%, 13%, and 26%, respectively. The transverse diameters showed no significant differences except for a 10% reduction in X80 treatment. However, total sugar, flavonoids, betaine, and ascorbic acid content were significantly increased by 6%, 27%, 167%, and 166%, respectively, and carotenoid was significantly decreased by 36% in fruit under the X60 treatment. Moreover, flavonoids, betaine, and ascorbic acid content were significantly increased by 13%, 159%, and 133%, respectively, and total sugar and carotenoid were significantly decreased by 2% and 34%, respectively, in fruit under X80 treatment. Polysaccharide content in wolfberry fruit was not significantly different among root-restriction treatments except for a significant decrease of 17% in fruit under X40 treatment.

3.2. Soil Physicochemical Properties of L. barbarum under Different Root-Restriction Widths

Eleven soil physicochemical properties were determined, as shown in

Table 3. Soil physicochemical properties in different root-restriction treatments and no root-restriction cultivation.

Treatment	0–20 cm				20–40 cm
	CK	X40	X60	X80	CK	X40	X60	X80
Electrical Conductivity (EC: ms/cm)	2.14 ± 0.03 d	4.48 ± 0.04 b	3.26 ± 0.08 c	4.26 ± 0.16 b	2.59 ± 0.04 d	6.24 ± 0.04 a	2.79 ± 0.07 d	2.79 ± 0.05 d
pH	8.15 ± 0.01 a	7.39 ± 0.17 d	7.98 ± 0.03 b	7.80 ± 0.17 d	8.10 ± 0.01 ab	7.73 ± 0.02 c	8.06 ± 0.20 ab	7.83 ± 0.01 c
Total Nitrogen (TN: g kg−1)	0.43 ± 0.01 de	0.86 ± 0.07 a	0.55 ± 0.00 c	0.54 ± 0.01 c	0.43 ± 0.02 de	0.36 ± 0.04 e	0.71 ± 0.09 b	0.47 ± 0.01 cd
Total Phosphorus (TP: g kg−1)	0.66 ± 0.01 bcd	0.83 ± 0.02 d	0.75 ± 0.02 abc	0.69 ± 0.01 bc	0.58 ± 0.02 d	0.76 ± 0.01 ab	0.76 ± 0.04 ab	0.65 ± 0.03 cd
Total Potassium (TK: g kg−1)	19.06 ± 0.13 c	21.00 ± 0.27 a	19.89 ± 0.29 bc	21.80 ± 1.40 a	20.83 ± 0.41 ab	19.08 ± 0.89 c	21.28 ± 1.32 a	19.92 ± 0.56 bc
Rapidly Available Potassium (AK: mg kg−1)	288.67 ± 13.66 cd	437.67 ± 61.23 a	374.56 ± 29.88 b	253.49 ± 14.11 de	291.33 ± 2.17 c	441.67 ± 21.46 a	442.33 ± 13.33 a	237.43 ± 20.45 e
Rapidly Available Phosphorus (AP: mg kg−1)	4.46 ± 0.02 c	5.28 ± 0.05 b	5.54 ± 0.14 b	4.84 ± 0.19 c	4.26 ± 0.03 c	6.92 ± 0.02 a	7.16 ± 0.03 a	4.60 ± 0.08 c
Organic Carbon (TOC: g kg−1)	5.55 ± 0.34 c	6.61 ± 0.25 b	6.96 ± 0.48 ab	7.11 ± 0.05 a	5.51 ± 0.34 c	6.94 ± 0.35 ab	7.33 ± 0.42 a	6.57 ± 0.92 b
Organic Matter (OM: g kg−1)	9.56 ± 0.58 c	11.39 ± 0.43 b	11.99 ± 0.83 ab	12.25 ± 0.18 ab	9.49 ± 0.57 c	11.96 ± 0.62 ab	12.64 ± 0.72 a	12.48 ± 0.24 ab
Ammonium Nitrogen (AN: mg kg−1)	27.06 ± 0.19 d	35.77 ± 2.23 a	33.55 ± 0.75 ab	31.57 ± 1.45 bc	30.25 ± 1.86 bc	29.08 ± 0.56 cd	29.01 ± 1.07 cd	26.85 ± 0.56 d
Nitrate Nitrogen (NN: mg kg−1)	4.60 ± 0.04 e	6.35 ± 0.06 b	5.71 ± 0.03 bc	5.36 ± 0.36 cd	7.41 ± 0.14 a	5.11 ± 0.34 cde	4.96 ± 0.11 de	4.61 ± 0.10 e

Note: Mean values (x ± SE) followed by the same letter were not significantly different using LSD (p < 0.05) based on ANOVA of square-root-transformed data (n = 3).

. Root-restriction method improved EC, TOC, and OM at 0–20 cm soil depth (EC:X40 by 109%, X60 by 52%, and X80 by 99%; TOC: X40 by 19%, X60 by 25%, and X80 by 28%; and OM: X40 by 19%, X60 by 26%, and X80 by 28%) and 20–40 cm soil depth (EC: X40 by 109%, X60 by 52%, and X80 by 99%; TOC: X40 by 26%, X60 by 33%, and X80 by 19%; and OM: X40 by 26%, X60 by 33%, and X80 by 32%). N, P, and K content were significantly higher than CK at the soil depth from 0–40 cm of rhizosphere at the overall level but had differences under different root-restriction patterns. TN, TP, and TK were significantly improved under root restriction at the soil depth of 0–20 cm (TN: X40 by 100%, X60 by 28%, and X80 by 26%; TP: X40 by 26%, X60 by 14%, and X80 by 5%; and TK: X40 by 10%, X60 by 4%, and X80 by 14%) and there were no significant differences at the soil depth of 20–40 cm. AN and NN content were significantly improved under root-restriction at the soil depth of 0–20 cm (AN: X40 by 100%, X60 by 28%, and X80 by 26% and NN: X40 by 26%, X60 by14%, and X80 by 5%), and there was no obvious downward trend at the soil depth of 20–40 cm. Root-restriction method improved AP at 0–20 cm soil depth (X40 by 18%, X60 by 24%, and X80 by 9%) and 20–40 cm soil depth (X40 by 62%, X60 by 68%, and X80 by 8%). AK content under X40 and X60 root-restriction methods had obvious improvement at the soil depth of 0–40 cm (0–20 cm: X40 by 52% and X60 by 30% and 20–40 cm: X40 by 52% and X60 by 52%). AK content under X80 root-restriction method had an obvious decrease at the soil depth of 0–40 cm (0–20 cm: X80 by 12% and 20–40 cm: X80 by 19%).

3.3. Correlation Analysis of L. barbarum Growth-Related Factors Affected by Different Root-Restriction Treatments

The correlation analysis of L. barbarum growth-related factors and soil physicochemical properties found that leaf N, P, and K content were significantly correlated with the N, P, and K content of soil (Figure 4). Leaf N content with ammonium nitrogen, total nitrogen, nitrate nitrogen, and electrical conductivity were significantly positively correlated and significantly negatively correlated with leaf N:P ratio and polysaccharides. Leaf P content was positively correlated with total phosphorus, betaine, and leaf N content and was significantly negatively correlated with leaf N:P ratio, polysaccharides, crown diameter (N-S), and leaf area. Number and length of new branches and leaf width were significantly negatively correlated with leaf K content. Organic matter and organic carbon with fruit nutrition indicators were significantly positively correlated with the total sugar, flavonoid, betaine, and ascorbic acid level. In addition, the nutrition indicators of fruit were positively correlated with the leaf N, P, and K content of leaf.

3.4. Structure and Diversity of Bacterial Communities in the Rhizosphere Soil of L. barbarum Ningqi No. 7 under Different Root-Restriction Cultivation Patterns

3.4.1. Sequencing Quality Assessment

After sample quality control and analysis, clean tags were obtained. Compared with CK, X40, X60, and X80, all had sequencing library coverage above 94%. This indicated that the majority of bacterial sequences were detected, and the sequencing results exhibited good representation (

Table 4. Statistics of tags in each rank in rhizosphere soil samples of L. barbarum among different root-restriction treatments.

Sample	Kingdom	Phylum	Class	Order	Family	Genus	Species
CKA	9791	8245	7868	7421	6240	5070	4352
CKB	13,613	11,300	10,663	9941	8569	6847	5954
CKC	12,017	10,225	9704	9185	7903	6434	5645
X40A	7176	5930	5514	5082	4215	3336	2693
X40B	7539	6623	6348	5861	5131	4240	3716
X40C	10,082	8470	8025	7516	6425	5257	4489
X60A	9470	8101	7614	6913	5897	4842	3943
X60B	8248	6991	6612	6186	5221	4268	3609
X60C	8083	6909	6597	6077	5235	4411	3688
X80A	13,060	10,727	10,280	9662	8358	6993	5796
X80B	12,775	10,505	9966	9316	7891	6367	5238
X80C	9747	7930	7450	6959	5580	4345	3707

).

3.4.2. OTU Clustering

The tags were clustered at the 97% similarity level to obtain OTUs, and the species composition was obtained by aligning to the database. The results showed that under the four root-restriction treatments, the bacterial communities detected in the rhizosphere soil of L. barbarum Ningqi No. 7 could be classified into 1 kingdom, 27 phyla, 42 classes, 89 orders, 123 families, 210 genera, and 255 species. The number of species in each rank is listed in

Table 5. Statistics of bacterial species at each rank in the rhizosphere soil samples of L. barbarum under different root restriction treatments.

Sample	Kingdom	Phylum	Class	Order	Family	Genus	Species
CKA	1	26	38	84	111	181	213
CKB	1	27	42	86	113	177	203
CKC	1	26	41	85	112	178	211
X40A	1	24	38	75	99	139	156
X40B	1	26	37	77	106	169	198
X40C	1	27	40	85	115	189	218
X60A	1	25	37	78	105	162	194
X60B	1	26	40	84	112	172	199
X60C	1	26	37	78	105	162	187
X80A	1	27	40	87	118	190	226
X80B	1	27	41	85	114	185	223
X80C	1	27	40	86	114	171	193
Total	1	27	42	89	123	210	255

. The common and unique OTU numbers in the samples were directly visualized by Venn diagrams. As shown in Figure 5a, in the soil samples from the four different treatments, the number of unique OTUs was 11, 6, 7, and 3 in CK, X40, X60, and X80, respectively. There were 1576 OTUs shared in common by all four groups. As shown in Figure 5b, the number of common OTUs between root restriction and CK was 1956; there were 11 unique OTUs in CK and 150 unique OTUs in root-restriction cultivation. This indicated that the soil bacterial species were more abundant in the root-restriction cultivation than in CK.

3.4.3. Soil Microbial Community Diversity and Structure

The relative abundance of soil bacterial community composition in different root restriction treatments was visualized on the phylum (Figure 6a,b) and genus level (Figure 6c,d). Among all sequences, the dominant bacterial phyla were Proteobacteria, Planctomycetota, Bacteroidota, Acidobacteriota, Gemmatimonadota, Actinobacteriota, Verrucomicrobiota, Chloroflexi, and Nitrospirota, with contributions of 31.1%, 14.9%, 8.7%, 6.8%, 5.1%, 4.3%, 2.1%, 2.1%, and 2.0%, respectively. The abundance of the top three bacterial genera in CK (Vicnamibacter, Sphingomonas, Tepidisphaera) was higher than that in the root restriction treatment; these three bacterial genera together made up more than 50% of the species in CK (Figure 6a). Proteobacteria, Bacteroidota, and Gemmatimonadota were significantly increased under X40 treatments, whereas Acidobacteriota, Actinobacteriota, and Nitrospirota significantly decreased; Proteobacteria, Planctomycetota, Bacteroidota, and Chloroflexi were significantly increased under X60 treatments, whereas Acidobacteriota, Gemmatimonadota, and Actinobacteriota significantly decreased; Proteobacteria, Actinobacteriota, and Nitrospirota were significantly increased under X80 treatments, whereas Planctomycetota, Bacteroidota, and Acidobacteriota significantly decreased (Figure 6b). The species distribution graph of bacteria from the soil samples of the three root-restriction treatments showed that Vicinamibacter, Sphingomonas, Tepidisphaera, and Gemmatimonas were the dominant bacterial communities, and the overall trend of the top 10 bacterial communities was decreasing at the genera level (Figure 6c). The nine important genus-level microorganisms all showed a downward trend under the root-limiting condition, but X60 treatment had no significant difference (Figure 6d).

3.4.4. Alpha Diversity Analysis

A comparison analysis of the four Alpha diversity indices of the soil bacterial communities in root restriction and CK was conducted (Figure 7). The coverage, which reflected the captured diversity, was greater than 95% for all samples. The results are shown in Figure 6. The numbers of OTUs, and the ACE, Chao1, Simpson and Coverage indices of soil bacterial communities were higher in the CK and X80 system than in the X40 and X60 systems, although the diversity of soil bacterial communities in root-restriction cultivation was not significantly different from that in CK (p > 0.05); there was no significant difference among the diversities of soil bacterial communities in three root-restriction treatments (p > 0.05). The abundance-based coverage estimator (ACE) and Chao1, which reflected species richness, were both higher in CK than in root-restriction treatments. The Shannon index for species evenness was higher only in X80 compared with that in CK.

3.4.5. Factor Analysis of Soil Bacterial Communities Affected by Sampling Depth and Different Root-Restriction Treatments

CK and different root-restriction treatments could be clearly distinguished by soil microorganisms and physicochemical properties under redundancy analysis (RDA). The RDA analysis of bacterial taxa at phylum and genus level and soil physicochemical parameters showed that the soils from the same root-restriction treatments were clustered together and the soils from different root-restriction treatments were separated (Figure 8 and Figure 9). This indicates that the effects of root restriction on the composition of soil bacterial communities are significantly different. Redundancy analysis (RDA) showed that individual bacterial phyla and genus responded differently to changes in the soil physicochemical properties. Soil physicochemical properties accounted for 69.12% of the variation in bacterial community composition at the phylum level in the soil at 0–20 cm (Figure 8a) and 99.95% of the variation at 20–40 cm depth (Figure 8b). Soil physicochemical properties accounted for 81.83% of the variation in bacterial community composition at the genus level in the soil at 0–20 cm (Figure 9a) and 89.33% of the variation at 20–40 cm depth (Figure 9b). This indicated that, in combination, these factors affected the composition of the soil bacterial community under different root-restriction treatments, and CK, X40, X60, and X80 were differentiated. In addition, the soil properties TP, AK, AP, AN, and NN were related to changes in Planctomycetota, Bacteroidota, and Chloroflexi (Figure 8a) at the soil depth of 0–20 cm. AP, AK, TP, TN, and EC soil properties were related to changes in Bacteroidota, Chloroflexi, Verrucomicrobiota, Proteobacteria, and Planctomycetota (Figure 8b) at the soil depth of 20–40 cm. In genus, TOC, OM, TP, AK, and AP were related to changes in Terrimonas and Chryseolinea at the soil depth of 0–40 cm (Figure 9).

4. Discussion

4.1. The Effect of Root-Restriction Cultivation Methods on the Growth Characteristics of L. barbarum

In this study, growing L. barbarum under different root-restriction treatments in their third year after planting resulted in greater numbers and length of new branches in that year than under no root restriction growth. The crown diameter (east–west), leaf length, leaf width, new branch number, and new branch length were the highest in X60, followed by X80, with X40 being the lowest. This indicated that root-restriction cultivation regulates root density to a certain extent, leading to a nutrient supply that depends on the need by tree growth. As a result, the nutrient uptake by the roots increased, promoting plant reproductive growth. Plant height and leaf area under root-restriction cultivation were less than under no root restriction due to the limited space for root growth and, therefore, nutrient uptake, as found by other researchers [14], which is in accord with the observation in past years. However, because L. barbarum exhibits continuous and active growth characteristics, the number of flowers and fruits and the leaf area of L. barbarum during vegetative growth and reproductive growth were different; therefore, the vegetative growth and reproductive period need to be separated. Future studies of L. barbarum growth parameters among different root restriction treatments should include nutrient distribution and utilization at different locations, as well as the soil characteristics.

Although this study showed that chlorophyll content in L. barbarum leaves in the root-restriction groups was significantly increased in X60 only, and its content in the root-restriction groups was higher than that in CK on the whole. This may be due to that fact that root restriction can effectively control the vegetative growth of the aboveground part, which can promote the germination and flower of fruit trees, ultimately leading to the growth and development of L. barbarum trees, which was better than that under CK, and the crown diameter (east–west), leaf length and leaf width were higher in X60. This was consistent with studies showing that the SPAD value (chlorophyll concentration) was positively correlated with the N content in the leaves and the percentage of leaves [15,16].

Nutrient elements are closely related to plant growth and development and are indispensable to complete plant life cycle. However, restricted-root cultivation is used to control the root system within a certain range, which will inevitably affect the absorption of nutrients. We found that different root-restriction treatments had effects on the levels of N, P, and K in the leaves, which were overall significantly lower in CK than in root-restriction treatments. It was possible that the nutrients were absorbed and utilized in the larger root system rather than transported to the shoots and leaves in the no-restriction treatment. This was consistent with the finding that the N, P, and K uptake in the underground portion of plants was higher than that in the aboveground portion [17]. Additionally, X60 treatment led to high K content and N:P ratio, while the N:P ratio was the highest in X80, indicating that X80 treatment had the smallest impact on the N:P balance. Our results showed that restricting root growth could reduce the N, P, and K uptake by the roots so more nutrients could be transported to the aboveground portion, consistent with the results from Zhi et al. [18,19].

In addition, root restriction can increase the content of nutrient elements in leaves, in particular, leaf N, P, and K content were all increased under root restriction (Figure 2). There was a positive correlation between leaf N, P, and K content and soil total N, P, and K content (Figure 4). Moreover, there was a positive correlation between leaf N, P, and K content and soil total N, P, and K content, which further proved that root-restricted treatments were beneficial for promoting plant growth and improving fertilization efficiency [20]. Although the researchers found that proper root restriction helped control vegetative growth [21,22], the leaf nutrition was determined in early June, which is conducive to the delivery of more nutrients to the fruit. Soil organic matter directly affects the fertilizer supply capacity of soil, and higher soil organic matter can promote plant growth and development [23]. In contrast to the CK, the root restriction with three indicators (flavonoid, betaine, and ascorbic acid) showed significant growth trends, especially betaine, which increased twofold (Table 2) in our study. Moreover, root-restriction regulation could significantly increase soil organic matter and organic carbon content, and this change was beneficial in improving the fruit quality of wolfberry (Figure 4), which was similar to results found by other researchers [24]. The reason may be the slow release of nutrients in the process of organic fertilizer decay, which can synchronize with the physiological needs of wolfberry, improve the fertilizer utilization efficiency, increase the mineralization process of organic nutrients, improve the soil nutrient pool source, and increase the content of total and available nutrients in soil [25,26].

4.2. The Effect of Different Root-Restriction Cultivation Methods on the Diversity and Composition of Bacterial Communities

This study utilized high-throughput sequencing technology and compared the composition of the rhizosphere soil bacterial community of L. barbarum grown under different root-restriction cultivation methods and no root restriction. The results showed that root-restriction cultivation caused significant changes in the composition of the bacterial community in soil. This indicated that root-restriction cultivation increased the relative abundance of soil bacteria. In addition, among the root-restriction treatments, X60 had the greatest relative abundance of the top 10 bacterial genera; generally, this impacted the soil microbial community and enzyme activities by changing soil characteristics (e.g., pH, temperature, soil water content, and total nitrogen) under an aerobic environment [27,28,29,30]. Furthermore, X60 treatment had the most significant improvement of the root restrictions, regulating the limited growth environment in a way that was beneficial to plant growth. This was consistent with the results from Li et al., who studied the effect of root-restriction cultivation on the rhizosphere microbial community of marigolds [31].

The three bacterial species with the highest abundance in the root restriction treatments accounted for 50% of all bacterial species, and the abundance of these three species was significantly lower than that in CK. This may be due to the increased soil N content and acidified soil after root-restriction cultivation, leading to a decrease in the abundance of certain bacterial taxa in the rhizosphere soil [32]. In addition, X60 root-restriction cultivation was higher than other treatment of the top 10 bacterial species. The Pirellula, Terrimonas, and Chryseolinea genera had the greatest relative abundance in X60. It was possible that Pirellula, Terrimonas, and Chryseolinea genera are all obligate aerobic bacteria, and with significant increase in 60 cm width treatment, which increased the soil permeability and soil oxygen content [33].

Two unclassified microbial groups were detected, and we speculated that they may be new microbial species in the rhizosphere soil of L. barbarum. The microbial community in the rhizosphere soil is complex, the result of synergistic microbial populations, and the effect of rhizosphere soil microbes on L. barbarum cannot be explained by changes in one or a few populations. Future studies need long-term experiments and observations from multiple aspects combined with other methods to further reveal the effects of different microbes on L. barbarum growth.

Redundancy analysis of the composition of soil bacterial communities and environmental factors showed that some bacterial genera exhibited significant positive or negative correlations with soil TOC, pH, and available N, P, and K. This may be because soil bacteria are present mainly in macropores or aggregates, so they are easily affected by changes in the microclimate [34]. Among them, TK, and pH were positively correlated with the genera Lysobacter, Pirellula, Tepidisphaera, Vicinamibacter, and Sphingomonas. TP, TN, and TOC exhibited a positive correlation with Chryseolinea, Candidatus, and Saccharimonas. WC was positively correlated with the Terrimonas genus. These results showed that soil properties were highly correlated with the composition of the bacterial community, which was consistent with the finding of Yu et al. (2019) that soil-available K greatly influenced the composition of bacterial communities. In addition, root-restriction cultivation did not significantly affect the diversity of soil microbes; rather, it affected their abundance, which was consistent with the results from Gong [35]. Thus, root-restriction treatments can significantly increase the relative abundance of beneficial bacterial genera and significantly decrease the relative abundance of harmful bacterial genera, thereby showing positive regulatory functions on the structure of the microbial community [36].

5. Conclusions

Root-restriction treatments of L. barbarum was the most productive cultivation system; the crown diameter (east–west), leaf length, leaf width, new branch number, and new branch length were markedly improved, and plant height was decreased. Root-restriction treatments not only increased the chlorophyll content but also the N, P, and K contents in L. barbarum leaves, which was conducive to the absorption and accumulation of nutrients. Root restriction improved fruit quality by increasing soil organic matter and organic carbon, which also improved fertilization efficiency to promote plant growth. In X60, K level and top 10 bacteria abundance were significantly increased and performed best on all treatments; therefore, the X60 treatment should be more desirable to farmers. In the root-restriction treatments, the N:P ratio was the lowest in X80, indicating that X80 had a lower impact on the N:P balance. The changes of soil TP, TK, TN, and TOC induced by root restriction were potential factors for the prediction of bacterial diversity. In summary, root-restriction cultivation was beneficial to the growth of L. barbarum Ningqi No. 7 to a certain extent. X60 and X80 had better effects than X40. X60 had the most significant effect on improving L. barbarum growth, increasing chlorophyll content and K content in the leaves, improving the abundance and diversity indices of soil bacteria, and optimizing the composition and abundance of bacteria. Its effects on nutritional regulation, root growth, nutrient utilization, flowering, and fruit setting in L. barbarum trees need to be further investigated.