Effect of Plant Growth-Promoting Rhizobacteria Inoculation on Sweet Potato Yield and Nutritional Quality in Northeast Thailand

Abstract

Excessive fertilizer use poses environmental risks in the long term. Thus, plant growth-promoting rhizobacteria (PGPR) have been proposed as a complementary approach to reduce fertilizer use and prevent nutrient stress. Moreover, PGPR’s efficacy in sandy soil for sweet potato production in Northeast Thailand has not yet been reported. This study tested the hypothesis that PGPR inoculation could reduce fertilizer dependency while maintaining yield in nutrient-poor sandy soils. In this research, a field study was conducted from 2023 to 2024 in Chonnabot District, Khon Kaen, Thailand, to evaluate the effects of Azospirillum brasilense, Azotobacter vinelandii, Beijerinckia mobilis, and inorganic fertilizer on the Okinawan Orange and Carrot native sweet potato varieties. Treatments followed a factorial randomized complete block design with two factors, PGPR inoculation and fertilizer level (0%, 25%, 50%, 75%, and 100%), and were replicated three times. The results showed that PGPR dipping had no statistically detectable effect on sweet potato growth and yield (p > 0.05). However, a notable finding was that PGPR significantly increased the protein and fiber content of tubers (p < 0.01) while reducing carbohydrate content, which may have implications for the taste and the nutritional quality of sweet potatoes. In addition, the application of inorganic fertilizer had a significant effect on yield. The Carrot native variety achieved the highest yield (13,481.00 kg ha⁻¹) with 75% fertilizer, while the Okinawan Orange variety attained the highest yield (8866.00 kg ha⁻¹) with 100% fertilizer. These data could be used to assist farmers in determining their fertilizer usage.

1. Introduction

As the global population continues to grow and climate change intensifies, tuber crops like the sweet potato, known for their resilience and adaptability, are increasingly recognized as a cornerstone of future food security, particularly in vulnerable tropical and subtropical regions. Sweet potato (Ipomoea batatas L.) is the seventh most important economic crop in the world, after rice, wheat, potato, maize, barley, and cassava [1]. It is a fast-growing tuberous crop. It also has high nutritional value, being rich in carbohydrates; protein; fiber; beta carotene; anthocyanins; and vitamins A, B, C, and E, along with various minerals such as potassium (K), copper (Cu), manganese (Mn), iron (Fe), and zinc (Zn), particularly in the orange- and purple-fleshed sweet potatoes [2]. In addition, the aboveground biomass of sweet potatoes is a useful food source for fish farming due to its high protein and carbohydrate content [3]. Currently, sweet potato cultivation covers approximately 9.39 million ha distributed across various continents worldwide, producing about 133.2 million tons of sweet potato. In Thailand, sweet potato cultivation is increasing, covering a total area of 3300 ha and yielding an average of 15,893.75 kg ha⁻¹ due to consumption demand. However, Thailand’s sweet potato yield remains lower than major producers being 6286 kg ha⁻¹, particularly China, where the average yield is 22,000 kg ha⁻¹ [4]. Originally, native sweet potatoes in Thailand had moderate yields, but they were not colorful, and they were less sweet and soft compared to Japanese varieties. In 2022, the Okinawan Orange variety provided the highest tuber fresh weights of 15,500 kg ha⁻¹ in Phu Wiang district and 8080 kg ha⁻¹ in Nong Ruea district, with a higher return than other Japanese varieties in Northeast Thailand [3]. The agricultural areas in the northeast region are generally characterized by sandy soil, which is suitable for the growth of tuberous plants. This specific soil type presents challenges in farming other types of crops due to its inherently low fertility and high susceptibility to rapid leaching of essential nutrients, especially nitrogen (N) and phosphorus (P) [5].

Enhancing the nutritional quality and productivity of crucial staple crops under resource-limited conditions is paramount to achieving Zero Hunger while limiting ongoing environmental stress. To maintain good sweet potato growth and yield for commercial production, high levels of inorganic and organic fertilizers are often needed. Inorganic fertilizers improve the growth of plants and increase yields in a relatively short period; however, they are costly and can lead to environmental pollution [6,7]. The reliance on high levels of synthetic inorganic fertilizers globally contributes substantially to greenhouse gas emissions and water pollution, necessitating an urgent pivot toward climate-smart agricultural practices. Biological fertilization methods, such as the application of plant growth-promoting rhizobacteria (PGPR), offer an alternative approach, thereby reducing the environmental footprint of commercial crop production systems worldwide. PGPR are a group of bacteria that colonize the rhizosphere and enhance crop performance. Direct mechanisms include facilitating nutrient cycling, such as N fixation, solubilization of P and K [8], and promoting root uptake of N and P [9]. Their secondary function involves stimulating plant growth through the production of various phytohormones, including auxins [10], gibberellins [11], and cytokinins [12], which specifically promote root elongation and branching, essential for maximizing nutrient uptake and crucial tuber initiation in sweet potato. Furthermore, PGPR modulate plant physiology to mitigate hormone stress responses [13,14,15,16]. Notably, high ethylene concentrations [13] often inhibit root growth and limit tuber yield. However, effective PGPR strains possessing ACC deaminase minimize this stress-induced damage, ensuring sustained root elongation and efficient energy partitioning toward the storage root [17]. Additionally, PGPR modulate abscisic acid [14], the primary signal for drought response and stomatal closure. Finally, PGPR bolster systemic defense mechanisms by modulating salicylic acid [15] against biotrophic pathogens and jasmonic acid [16] against herbivores and necrotrophic pathogens, improving overall crop resilience. Yu et al. [18] reported that the application of PGPR increased their yield of purple sweet potato by 26%, increased the product’s nutritive quality of soluble sugar by 8%, increased vitamin C content by 14%, and increased anthocyanin content by 11%. Using PGPR significantly retarded the depletion of P and K elements during crop cultivation, promoted soil bacterial diversity, and altered the bacterial community structure. Thus, developing methods to preserve crop yield and reduce environmental impact while limiting inorganic fertilizer use is an urgent necessity.

For this study, the Department of Agriculture, Bangkok Province, Thailand, suggested the PGPR strains Azospirillum brasilense, Azotobacter vinelandii, and Beijerinckia mobilis based on their demonstrated capabilities in N fixation [19,20], P solubilization [21], and cytokinin synthesis [22,23], which are critical for nutrient availability and crop growth. This experiment aimed to reduce inorganic fertilizer use in sweet potato production based on the hypothesis that the application of PGPR would synergistically interact with reduced inorganic fertilization to sustain or improve sweet potato yield and quality under nutrient-poor conditions. The objective was to evaluate the synergistic effects of PGPR and inorganic fertilizer application on sweet potato growth, yield performance, and tuber nutritional quality. Two sweet potato cultivars, the Okinawan Orange and the Carrot (native variety), were used as model crops to assess the impact of PGPR and inorganic fertilizer under field conditions. The findings from this experiment may reduce the use of inorganic fertilizers and increase productivity, thereby contributing to the development of sustainable and environmentally friendly agriculture.

2. Materials and Methods

2.1. Site and Environmental Conditions

A field study was conducted from 2023 to 2024 for two cropping seasons in Chonnabot district, Khon Kaen province, Northeast Thailand. The study site was located at an elevation of 168 m, 16.098813° N latitude, and 102.600195° E longitude. The annual precipitation recorded was 861.30 mm in 2023 and 1087.10 mm in 2024, and the average maximum and minimum temperatures were 33.33 °C and 23.08 °C in 2023 and 34.73 °C and 25.70 °C in 2024, respectively. The distribution of rainfall during 2023 and 2024 in Chonnabot district, Khon Kaen province, Thailand, is shown in Figure 1. Soil samples were collected from the experimental site before the start of the experiment. Samples were taken to a depth of 0–30 cm from three points per plot. After collection, the samples were air-dried, ground, passed through a 2 mm sieve, and then packed in plastic bags for laboratory analysis at the Northeast Agriculture Research and Development Center, Faculty of Agriculture, Khon Kaen University. Soil chemical analyses assessed soil pH at a 1:1 ratio [24]; organic matter (OM) [25]; electrical conductivity (EC) using a 1:5 slurry [26]; total nitrogen (N) via the Kjeldahl method; and available phosphorus (P) via Bray II extraction and the molybdenum blue method [27]. Exchangeable potassium (K), calcium (Ca), and magnesium (Mg) were extracted with 1 N NH₄OAc (pH 7) and the concentration in the extraction was determined using a flame photometer. The micronutrients Fe, Mn, Cu, and Zn were extracted using DTPA (pH 7.3) and measured via atomic absorption spectrophotometry. Total heavy metals (Cr, Cd, and Pb) and metalloids (As) were measured using wet digestion (nitric perchloric acid) and ICP-OES.

2.2. Experimental Design

Carrot and Okinawan Orange sweet potato slips were used to evaluate the effects of PGPR produced by the Department of Agriculture, Bangkok Province, Thailand. The PGPR product contained 3 types of bacteria (Azospirillum brasilense, Azotobacter vinelandii, and Beijerinckia mobilis) [28] and was combined with different percentages of the commercial NPK fertilizer (46.88 kg N ha⁻¹ + 46.88 kg P ha⁻¹ + 159.38 kg K ha⁻¹) recommended by the Department of Agriculture, Bangkok Province, Thailand. Two components, PGPR (No PGPR, D1 and Dip PGPR, D2) and recommended inorganic fertilizer levels (0% (T1), 25% (T2), 50% (T3), 75% (T4), and 100% (T5)) were used in the treatments and are detailed in

Table 1. Grouping of experimental treatments.

PGPR (D)	Inorganic Fertilizer (T)	Treatment Combinations
No PGPR dipping (D1)	No fertilizer (T1, Control)	No PGPR dipping + No fertilizer (D1 × T1)
	25% inorganic fertilizer (T2)	No PGPR dipping + 25% inorganic fertilizer (D1 × T2)
	50% inorganic fertilizer (T3)	No PGPR dipping + 50% inorganic fertilizer (D1 × T3)
	75% inorganic fertilizer (T4)	No PGPR dipping + 75% inorganic fertilizer (D1 × T4)
	100% inorganic fertilizer (T5)	No PGPR dipping + 100% inorganic fertilizer (D1 × T5)
PGPR dipping (D2)	No fertilizer (T1, Control)	PGPR dipping + No fertilizer (D2 × T1)
	25% inorganic fertilizer (T2)	PGPR dipping + 25% inorganic fertilizer (D2 × T2)
	50% inorganic fertilizer (T3)	PGPR dipping + 50% inorganic fertilizer (D2 × T3)
	75% inorganic fertilizer (T4)	PGPR dipping + 75% inorganic fertilizer (D2 × T4)
	100% inorganic fertilizer (T5)	PGPR dipping + 100% inorganic fertilizer (D2 × T5)

T1 = no fertilizer (control); T2 = 11.72 kg N ha⁻¹ + 11.72 kg P ha⁻¹ + 39.85 kg K ha⁻¹; T3 = 23.44 kg N ha⁻¹ + 23.44 kg P ha⁻¹ + 79.69 kg K ha⁻¹; T4 = 35.16 kg N ha⁻¹ + 35.16 kg P ha⁻¹ + 118.69 kg K ha⁻¹; T5 = 46.88 kg N ha⁻¹ + 46.88 kg P ha⁻¹ + 159.38 kg K ha⁻¹.

. The experiment had a factorial randomized complete block design (RCBD) with three replications to investigate the effects of two main factors on sweet potatoes, as shown by the red arrows in Figure 2. Furthermore, the figure clearly specifies the time points for data collection and harvest, including measurements at 30 days after planting (30 DAP), 60 days after planting (60 DAP), and the final harvest stage (Harvest). This experimental design enables researchers to compare the effectiveness of various management methods and monitor plant growth throughout the cultivation period. The resulting data will be valuable for developing sustainable and efficient sweet potato production systems.

2.3. Field Management

The soil was prepared with cattle manure at a rate of 4.38 t ha⁻¹ for two weeks before the sweet potatoes were planted. The plot size was 2 × 10 m, and raised beds of 50 cm with 30 plots per variety were maintained under rainy conditions and with sufficient water supply to maintain field capacity. The supplemental irrigation was applied as needed based on routine visual and tactile assessments of soil moisture in the surface zone. When plots appeared dry, irrigation was supplied for 30–45 min per event, resulting in an average single event application of 39–59 m³ ha⁻¹. Irrigation was managed to maintain soils near field capacity without causing waterlogging. Raised beds helped sweet potatoes grow and develop tubers, facilitated drainage, and contributed to ease of harvesting [29]. Carrot and Okinawan Orange slips (25–30 cm) were soaked in a solution recommended by the Department of Agriculture containing 1 kg of PGPR per 20 L of water with an approximate bacterial population of 10⁶ CFU g⁻¹ for 30 min in shade before planting. In this study, the PGPR included A. brasilense, A. vinelandii, and B. mobilis. A. brasilense is characterized by its capacity for efficient N fixation [19] and the synthesis of cytokinin hormones [22]. This specific strain has previously demonstrated successful survival and growth-promoting effects when applied to sweet potato [30]. A. vinelandii is a versatile strain, contributing significantly to N fixation [31], P solubilization [21], and cytokinin biosynthesis [23]. Finally, B. mobilis is primarily utilized for its robust nitrogen-fixing capabilities within the rhizosphere environment of the crop [20]. The plant spacing was 50 cm between rows and 50 cm between plants. The sweet potato slips were planted at an angle of 45 degrees at a depth of 10–15 cm in the soil. Weed control was performed manually for 1 month after planting and then regularly every month. At 30 and 45 days after the planting of the sweet potatoes, fertilizer was applied via the drilling method according to the experimental design.

2.4. Data Collection

Vine length, node length, number of nodes per plant, SPAD chlorophyll meter reading (SCMR), total chlorophyll, chlorophyll A, chlorophyll B, carotenoids, and leaf area were measured from 5 plants per plot at 30 and 60 days after planting and at the harvest stage. The vine length was measured from the base of the stem to the tip of the shoot. SCMRs were taken in the middle of the vine with the fully grown leaves of 5 randomized plants per plot, and the average was calculated. Chlorophyll measurement was modified according to the method of Arnon [32] in 2.5. The leaf area (LA) was estimated using the LI-Cor 3100 leaf area meter (LI-COR Inc, Lincoln, NE, USA). At the harvest stage, several parameters were recorded: the number of tubers per plant, tuber length, tuber diameter, tuber fresh weight, fresh and dry weight of aboveground biomass, and harvest index (HI). Additionally, total N was determined according to Bremner and Mulvaney’s method [33], total P using wet digestion and the spectrophotometer method [34,35], and total K using wet digestion (nitric perchloric) and an atomic absorption spectrophotometer [35]. Aboveground biomass, ash, and fiber were determined using the Weende method [36]; the nutritional content of the tuber, such as protein, was determined via combustion (AOAC official method 990.03), and carbohydrates were determined via calculation by subtracting the sum of the percentage moisture, crude protein, ether extract, crude fiber, and ash content from 100 [37].

2.5. Chlorophyll and Carotenoid Analysis

The sample leaves of the sweet potatoes growing in field conditions were taken from leaf position numbers 3–4. The collected leaves were cut into small 100 mg pieces, leaving the mid-ribs, and ground with 10 mL of 80% acetone (v/v). The sample solution was filtered through Whatman No. 1 filter paper, and the final volume was collected. The absorption of the carotenoids, chlorophyll a, and chlorophyll b was read at 440, 645, and 663 nm with a UV-Vis spectrophotometer (Model i3, Hanon, Jinan, China) using 80% acetone as a blank. The following formulas were used to calculate total chlorophyll content, chlorophyll a, chlorophyll b, and carotenoids [38], with all the photosynthetic pigment contents expressed in mg gFW⁻¹:

Chlorophyll a (mg gFW⁻¹) = [(12.7 × OD663) − (2.69 × OD645)] × [V/(1000 × W)]

(1)

Chlorophyll b (mg gFW⁻¹) = [(22.9 × OD645) − (4.68 × OD663)] × [V/(1000 × W)]

(2)

Total chlorophyll (mg gFW⁻¹) = [(20.2 × OD645) + (8.02 × OD663)] × [V/(1000 × W)]

(3)

Carotenoid (mg gFW⁻¹) = (4.69 × OD440) − [0.268(20.2 × OD645) + (8.02 × OD663)] × [V/(1000 × W)]

(4)

where

OD440 = Absorbance at 440 nm;

OD645 = Absorbance at 645 nm;

OD663 = Absorbance at 663 nm;

V = Final volume of 80% acetone in chlorophyll extract;

W = Fresh weight of leaf materials used (mg).

2.6. Statistical Analysis

The data were analyzed via two-way Analysis of Variance (ANOVA) using the Statistix 10 software. The correlation coefficients of the sweet potatoes’ agronomic traits and economic yield were analyzed using Pearson correlation. Comparative analysis of the results was carried out using Fisher’s least significant difference (LSD) at a 5% level of significance; that is, a p-value ≤ 0.05 was considered statistically significant. Microsoft Excel 2021was used to plot the graphs.

3. Results

3.1. Soil Chemical Properties Before Planting Sweet Potatoes

The nutritional content of the soil samples collected before planting sweet potatoes is shown in

Table 2. Soil chemical properties before planting sweet potatoes.

Treatment	pH	EC (dS·m−1)	OM (g·kg−1)	N (g·kg−1)	P	K	Ca	Mg	Fe	Mn	Zn	Cu	As	Cr	Cd	Pb
					(mg·kg−1)
Soil before planting	6.39	0.031	4.50	0.24	9.00	119.66	159.72	32.40	65.25	5.05	0.37	0.40	19.45	12.60	0.610	5.46

. The soil was classified as sandy loam, with a neutral pH of 6.39. Furthermore, the non-saline EC was measured at 0.031 dS m⁻¹. Several essential fertility nutrients, such as OM, total N, and available P, were found to be low. Conversely, exchangeable K, Ca, and Mg were found to be sufficiently high, measuring 119.66, 159.72, and 32.40 mg kg⁻¹, respectively. Micronutrients were present at a critical level. However, the total concentration of heavy metals in this experimental site was within the normal range for plant growth [39].

3.2. Growth and Chlorophyll Content of Sweet Potatoes

During the initial 30 DAP, the Carrot native sweet potato variety exhibited a significant response to PGPR dipping, showing an increase in node length (

Table 3. Growth parameters of the Carrot native sweet potato.

Treatment	30 DAP					60 DAP					Harvest
	Vine Length (cm)	Node Length (cm)	No. of Nodes per Plant−1	SCMR	Leaf Area per Plant−1 (cm3)	Vine Length (cm)	Node Length (cm)	No. of Nodes per Plant−1	SCMR	Leaf Area per Plant−1 (cm3)	Vine Length (cm)	Node Length (cm)	No. of Nodes per Plant−1	SCMR	Leaf Area per Plant−1 (cm3)
PGPR (D)
No PGPR (D1)	18.17	1.28 b	8.00	40.84	267.00	52.38	3.89	41.00 b	43.32	2519.00 b	66.83	3.48	71.00	35.97	3053.00
Dip PGPR (D2)	20.39	1.44 a	9.00	41.66	307.00	52.09	3.77	48.00 a	44.64	3155.10 a	68.98	3.68	66.00	39.12	3246.00
F-test (D)	ns	*	ns	ns	ns	ns	ns	*	ns	**	ns	ns	ns	ns	ns
Fertilizer (T)
No fertilizer (T1, Control)	17.76	1.39	8.00	39.82	203.00 c	52.67	3.81	34.00 b	45.05	2466.00 b	64.29	3.37	66.00	39.62	3595.00
25% fertilizer (T2)	18.58	1.38	8.00	43.59	335.00 a	49.16	3.59	44.00 a	42.37	2528.00 ab	62.97	3.54	62.00	34.03	2932.00
50% fertilizer (T3)	21.35	1.47	8.00	43.00	327.00 ab	51.95	4.02	49.00 a	46.04	3182.00 a	68.68	3.79	74.00	37.92	3126.00
75% fertilizer (T4)	18.21	1.38	8.00	39.21	257.00 bc	55.33	3.82	48.00 a	41.80	2853.00 ab	65.87	3.51	73.00	36.57	3069.00
100% fertilizer (T5)	20.51	1.19	8.00	40.62	315.00 ab	52.05	3.92	48.00 a	44.63	3156.00 a	77.73	3.68	70.00	39.58	3027.90
F-test (T)	ns	ns	ns	ns	**	ns	ns	*	ns	*	ns	ns	ns	ns	ns
D × T
D1 × T1	18.83	1.23	8.00	36.57	205.00	53.66	3.83	32.00	43.55	2659.00 cde	59.98	3.42	71.00	36.20	3356.00
D1 × T2	14.97	1.41	7.00	42.95	279.00	52.22	3.77	42.00	43.27	2336.00 de	63.52	3.38	63.00	34.08	2761.00
D1 × T3	19.21	1.47	8.00	43.96	271.00	47.23	4.11	45.00	45.11	2516.00 de	64.11	3.44	76.00	34.34	2907.00
D1 × T4	18.26	1.33	8.00	40.52	273.00	56.33	3.65	43.00	40.88	2136.00 e	66.88	3.72	71.00	35.70	2854.00
D1 × T5	19.57	0.96	7.00	40.19	309.00	52.44	4.08	44.00	43.76	2949.00 bcd	79.66	3.42	76.00	39.53	3391.00
D2 × T1	16.69	1.54	8.00	43.06	201.00	51.68	3.79	36.00	46.55	2272.00 de	68.60	3.31	60.00	43.04	3835.00
D2 × T2	22.18	1.34	9.00	44.23	391.00	46.11	3.40	46.00	41.47	2721.00 cde	62.41	3.70	61.00	33.97	3104.00
D2 × T3	23.49	1.46	9.00	42.04	384.00	56.66	3.92	53.00	46.96	3849.00 a	73.24	4.15	72.00	41.51	3345.00
D2 × T4	18.16	1.42	8.00	37.91	240.00	54.33	4.00	53.00	42.72	3571.00 ab	64.85	3.31	74.00	37.43	3283.00
D2 × T5	21.45	1.42	9.00	41.05	321.00	51.66	3.76	51.00	45.49	3363.00 abc	75.81	3.94	64.00	39.62	2665.00
F-test (D × T)	ns	ns	ns	ns	ns	ns	ns	ns	ns	*	ns	ns	ns	ns	ns
CV (%)	15.80	14.14	11.67	15.11	20.68	12.18	13.02	17.33	7.21	16.63	14.03	11.32	20.95	12.02	17.58

CV = coefficient of variation; ns = not significant; * = significant difference at p < 0.05; ** = significant difference at p < 0.01. Different letters within the same column indicate significant differences based on the least significant difference (LSD) test at p < 0.05. T1 = no fertilizer (control); T2 = 11.72 kg N ha⁻¹ + 11.72 kg P ha⁻¹ + 39.85 kg K ha⁻¹; T3 = 23.44 kg N ha⁻¹ + 23.44 kg P ha⁻¹ + 79.69 kg K ha⁻¹; T4 = 35.16 kg N ha⁻¹ + 35.16 kg P ha⁻¹ + 118.69 kg K ha⁻¹; T5 = 46.88 kg N ha⁻¹ + 46.88 kg P ha⁻¹ + 159.38 kg K ha⁻¹.

). In contrast, PGPR dipping showed no statistically significant effects on the Okinawan Orange sweet potato variety during this period (

Table 4. Growth parameters of the Okinawan Orange sweet potatoes.

Treatment	30 DAP					60 DAP					Harvest
	Vine Length (cm)	Node Length (cm)	No. of Nodes per Plant−1	SCMR	Leaf Area per Plant−1 (cm3)	Vine Length (cm)	Node Length (cm)	No. of Nodes per Plant−1	SCMR	Leaf Area per Plant−1 (cm3)	Vine Length (cm)	Node Length (cm)	No. of Nodes per Plant−1	SCMR	Leaf Area per Plant−1 (cm3)
PGPR (D)
No PGPR (D1)	31.41	1.86	12.00	43.35	744.00	52.18	2.56	33.00	44.61	1961.00 b	57.54	2.44	57.00	39.98	2363.00
Dip PGPR (D2)	29.36	1.91	12.00	42.09	708.00	53.43	2.47	33.00	44.08	2277.00 a	57.79	2.47	64.00	39.90	2240.00
F-test (D)	ns	ns	ns	ns	ns	ns	ns	ns	ns	*	ns	ns	ns	ns	ns
Fertilizer (T)
No fertilizer (T1, Control)	25.95 c	1.92	9.00 c	41.24	485.53 b	45.99 c	2.40	26.00 c	43.77	1458.00 d	49.07 c	2.41	54.00	40.00	1634.00 b
25% fertilizer (T2)	29.96 bc	1.84	13.00 ab	44.53	684.00 ab	55.60 ab	2.73	38.00 a	44.56	2031.00 bc	56.35 bc	2.47	65.00	39.60	2388.00 a
50% fertilizer (T3)	27.97 bc	1.83	12.00 b	42.08	703.00 ab	50.31 bc	2.44	29.00 bc	43.93	1928.00 c	55.97 bc	2.48	54.00	40.67	2427.00 a
75% fertilizer (T4)	32.38 ab	1.78	12.00 ab	42.98	824.00 a	53.43 ab	2.40	34.00 ab	44.98	2401.00 ab	61.13 ab	2.33	65.00	39.75	2469.00 a
100% fertilizer (T5)	35.65 a	2.04	14.00 a	42.78	933.00 a	58.71 a	2.60	37.00 a	44.50	2777.00 a	65.81 a	2.57	65.00	39.69	2606.00 a
F-test (T)	**	ns	**	ns	**	**	ns	**	ns	**	**	ns	ns	ns	**
D × T
D1 × T1	27.30	1.90	9.00	41.64	500.00	45.76	2.55	27.00	44.52	1290.00	48.89	2.44	53.63	39.45	1618.00
D1 × T2	29.39	1.82	12.00	44.53	734.00	56.50	2.68	38.00	44.11	1922.00	59.19	2.47	59.00	39.41	2624.00
D1 × T3	30.19	1.75	13.00	43.30	721.00	50.00	2.53	32.00	45.02	2049.00	56.86	2.46	55.00	40.65	2510.00
D1 × T4	32.19	1.71	13.00	44.00	819.00	50.53	2.36	35.00	45.51	2115.00	60.39	2.33	64.00	41.14	2515.00
D1 × T5	37.94	2.11	14.00	43.30	943.00	58.11	2.66	34.00	43.90	2429.00	62.39	2.48	56.00	39.26	2580.00
D2 × T1	24.60	1.94	10.00	40.85	471.00	46.22	2.25	26.00	43.02	1625.00	49.24	2.37	55.00	40.56	1649.00
D2 × T2	30.53	1.86	13.00	44.53	634.00	54.69	2.78	38.00	45.01	2140.00	53.50	2.47	71.00	39.78	2152.00
D2 × T3	25.75	1.90	11.00	40.86	684.00	50.61	2.36	26.00	42.83	1807.00	55.08	2.50	54.00	40.68	2345.00
D2 × T4	32.56	1.85	11.00	41.95	828.00	56.33	2.44	33.00	44.46	2686.00	61.88	2.33	65.00	38.36	2423.00
D2 × T5	33.36	1.97	14.00	42.27	923.00	59.30	2.53	41.00	45.11	3125.00	69.23	2.65	74.00	40.11	2632.00
F-test (D × T)	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns
CV (%)	11.67	13.42	10.93	6.77	24.61	10.05	10.74	16.75	5.08	16.72	10.09	7.67	17.05	3.44	18.09

CV = coefficient of variation; ns = not significant; * = significant difference at p < 0.05; ** = significant difference at p < 0.01. Different letters within the same column indicate significant differences based on the least significant difference (LSD) test at p < 0.05. T1 = no fertilizer (control); T2 = 11.72 kg N ha⁻¹ + 11.72 kg P ha⁻¹ + 39.85 kg K ha⁻¹; T3 = 23.44 kg N ha⁻¹ + 23.44 kg P ha⁻¹ + 79.69 kg K ha⁻¹; T4 = 35.16 kg N ha⁻¹ + 35.16 kg P ha⁻¹ + 118.69 kg K ha⁻¹; T5 = 46.88 kg N ha⁻¹ + 46.88 kg P ha⁻¹ + 159.38 kg K ha⁻¹.

). Inorganic fertilizer application significantly affected the leaf area per plant⁻¹ of the Carrot native variety, particularly at the 25%, 50%, and 100% application levels, while the Okinawan Orange sweet potato variety required 100% inorganic fertilizer to achieve optimal growth, especially in terms of vine length, no. of nodes per plant⁻¹, and leaf area per plant⁻¹, achieving 35.65 cm, 14, and 933 cm² per plant, respectively; however, these values were not significantly different compared to those achieved with the application of 75% inorganic fertilizer. At 30 DAP, neither sweet potato variety showed any interaction between PGPR and the inorganic fertilizer in their growth parameters or photosynthetic pigments (Table 3 and Table 4, Figure 3a,b).

By 60 DAP, PGPR dipping had significantly boosted the leaf area per plant to 3155 cm² in the Carrot native variety and 2277 cm² in the Okinawan Orange variety (Table 3 and Table 4). Also, PGPR dipping had significantly increased the number of nodes and significantly reduced the carotenoid content in the leaves of the Carrot native variety (Figure 4a). In terms of inorganic fertilizer management for the Carrot native variety, the 50% inorganic fertilizer application resulted in the highest no. of nodes per plant⁻¹ and the largest leaf area per plant⁻¹, but the values were not significantly different from those achieved with the 25%, 75%, and 100% inorganic fertilizer applications (Table 3). The Okinawan Orange variety continued to show a higher nutrient demand than the Carrot native variety, and the 100% inorganic fertilizer application still provided the best results in terms of vine length and leaf area per plant⁻¹, though, again, they did not differ significantly from those achieved with the 75% inorganic fertilizer application (Table 4). At 60 DAP, a significant interaction was observed between PGPR dipping and the 50% inorganic fertilizer application in the Carrot native variety, leading to a maximum leaf area of 3849 cm² per plant, whereas, in the Okinawan Orange variety, no interaction between PGPR and inorganic fertilizer application was detected for any growth parameter except chlorophyll a content (Figure 4b).

At harvest, PGPR dipping resulted in only a slight increase in the growth parameters of both sweet potato varieties. The pigment content in the Carrot native variety did not differ (Figure 5a), and only the chlorophyll b content in the leaves of the Okinawan Orange variety significantly increased (Figure 5b). In terms of inorganic fertilizer management, the Carrot native variety did not show significant differences among fertilizer treatments (Table 3), whereas, in the Okinawan Orange variety, a significant response in terms of growth was still observed. Specifically, the 100% inorganic fertilizer application provided the best vine length and leaf area per plant (Table 4), and the 50% inorganic fertilizer maximized chlorophyll a and total chlorophyll content (Figure 5b). Neither sweet potato variety showed a significant interaction between PGPR dipping and inorganic fertilizer management.

3.3. Yield Components, Economic Yields, and Aboveground Biomass

Yield components and economic yield, including the number of tubers per plant, tuber length, tuber diameter, tuber fresh weight, and harvest index (HI), were measured at the harvest stage. The results showed that PGPR application did not significantly affect the yield components of either variety (

Table 5. Yield components, economic yield, and aboveground biomass of Carrot native sweet potatoes.

Treatment	No. of Tubers per Plant−1	Tuber Length (cm)	Tuber Diameter (cm)	Tuber Fresh Weight (kg·ha−1)	Fresh Weight of Aboveground Biomass (kg·ha−1)	Dry Weight of Aboveground Biomass (kg·ha−1)	HI
PGPR (D)
No PGPR (D1)	4.31	9.83	3.06	9812.70	11,815.00	1889.20	0.48
Dip PGPR (D2)	4.05	10.39	3.17	9925.50	12,502.00	1768.20	0.46
F-test (D)	ns	ns	ns	ns	ns	ns	ns
Fertilizer (T)
No fertilizer (T1, Control)	3.79 b	10.11	3.00	7665.00 c	9399.00 b	1388.90 b	0.47 a
25% fertilizer (T2)	3.77 b	10.55	3.19	8709.00 bc	8233.00 b	1382.30 b	0.52 a
50% fertilizer (T3)	4.70 a	10.25	3.36	10,518.00 ab	12,734.00 ab	1756.30 b	0.49 a
75% fertilizer (T4)	4.82 a	9.66	3.14	13,481.00 a	14,291.00 a	2181.60 a	0.50 a
100% fertilizer (T5)	3.82 b	10.00	2.89	8972.00 bc	16,134.00 a	2434.40 a	0.36 b
F-test (T)	**	ns	ns	**	*	**	**
D × T
D1 × T1	4.00	9.48 abc	2.92 c	6973.00	7472.00 d	1292.00 cd	0.51
D1 × T2	3.81	10.07 abc	2.88 c	8125.00	7295.00 d	1220.30 d	0.53
D1 × T3	5.03	9.15 bc	3.15 abc	10,350.00	8820.00 d	1662.90 cd	0.55
D1 × T4	4.42	10.90 ab	3.55 a	13,913.00	18,716.00 a	2743.30 a	0.42
D1 × T5	4.30	9.54 abc	2.80 c	9703.00	16,771.00 ab	2527.70 a	0.37
D2 × T1	3.59	10.73 ab	3.08 abc	8357.00	11,327.00 bcd	1485.80 cd	0.44
D2 × T2	3.72	11.02 ab	3.50 ab	9293.00	9170.00 cd	1544.30 cd	0.50
D2 × T3	4.36	11.35 a	3.56 a	10,687.00	16,649.00 ab	1849.80 bc	0.43
D2 × T4	5.23	8.43 c	2.73 c	13,050.00	9867.00 cd	1619.90 cd	0.57
D2 × T5	3.33	10.45 ab	2.98 bc	8241.00	15,496.00 abc	2341.10 ab	0.35
F-test (D × T)	ns	*	**	ns	*	**	ns
CV (%)	14.48	10.85	9.77	18.17	31.68	17.93	15.74

HI = harvest index; CV = coefficient of variation; ns = not significant; * = significant difference at p < 0.05; ** = significant difference at p < 0.01. Different letters within the same column indicate significant differences based on the least significant difference (LSD) test at p < 0.05. T1 = no fertilizer (control); T2 = 11.72 kg N ha⁻¹ + 11.72 kg P ha⁻¹ + 39.85 kg K ha⁻¹; T3 = 23.44 kg N ha⁻¹ + 23.44 kg P ha⁻¹ + 79.69 kg K ha⁻¹; T4 = 35.16 kg N ha⁻¹ + 35.16 kg P ha⁻¹ + 118.69 kg K ha⁻¹; T5 = 46.88 kg N ha⁻¹ + 46.88 kg P ha⁻¹ + 159.38 kg K ha⁻¹.

and

Table 6. Yield components, economic yield, and aboveground biomass of Okinawan Orange sweet potatoes.

Treatment	No. of Tubers per Plant−1	Tuber Length (cm)	Tuber Diameter (cm)	Tuber Fresh Weight (kg·ha−1)	Fresh Weight of Aboveground Biomass (kg·ha−1)	Dry Weight of Aboveground Biomass (kg·ha−1)	HI
PGPR (D)
No PGPR (D1)	3.74	9.43	2.78	7040.10	9314.80	1548.20	0.46
Dip PGPR (D2)	3.60	9.25	3.11	7374.90	9649.30	1702.40	0.45
F-test (D)	ns	ns	ns	ns	ns	ns	ns
Fertilizer (T)
No fertilizer (T1, Control)	3.37	9.11	3.00	5215.80 c	7413.00 b	1025.40 c	0.46
25% fertilizer (T2)	3.22	9.17	3.13	6304.60 bc	8546.00 b	1437.40 bc	0.43
50% fertilizer (T3)	4.03	9.17	2.74	7531.10 ab	8686.00 b	1581.30 b	0.50
75% fertilizer (T4)	4.04	9.98	3.07	8120.10 a	11,478.00 a	1861.10 ab	0.45
100% fertilizer (T5)	3.69	9.25	2.78	8866.00 a	11,287.00 a	2222.00 a	0.46
F-test (T)	ns	ns	ns	**	**	**	ns
D × T
D1 × T1	3.07	9.35	2.89	5042.10	7873.00	1002.30	0.44
D1 × T2	3.11	9.28	2.88	6036.90	8699.00	1569.10	0.42
D1 × T3	4.44	10.17	2.90	7905.20	8401.00	1436.60	0.53
D1 × T4	3.82	9.41	2.70	7877.80	11,214.00	1700.30	0.46
D1 × T5	4.26	8.93	2.52	8338.40	10,387.00	2032.90	0.46
D2 × T1	3.66	8.88	3.11	5389.60	6952.00	1048.60	0.47
D2 × T2	3.33	9.06	3.39	6572.20	8394.00	1305.60	0.45
D2 × T3	3.62	8.18	2.59	7157.10	8970.00	1725.90	0.46
D2 × T4	4.26	10.56	3.44	8362.30	11,743.00	2021.80	0.43
D2 × T5	3.13	9.58	3.04	9393.20	12,188.00	2410.10	0.46
F-test (D × T)	ns	ns	ns	ns	ns	ns	ns
CV (%)	17.13	12.23	17.52	17.97	20.14	22.65	11.47

HI = harvest index; CV = coefficient of variation; ns = not significant; ** = significant difference at p < 0.01. Different letters within the same column indicate significant differences based on the least significant difference (LSD) test at p < 0.05. T1 = no fertilizer (control); T2 = 11.72 kg N ha⁻¹ + 11.72 kg P ha⁻¹ + 39.85 kg K ha⁻¹; T3 = 23.44 kg N ha⁻¹ + 23.44 kg P ha⁻¹ + 79.69 kg K ha⁻¹; T4 = 35.16 kg N ha⁻¹ + 35.16 kg P ha⁻¹ + 118.69 kg K ha⁻¹; T5 = 46.88 kg N ha⁻¹ + 46.88 kg P ha⁻¹ + 159.38 kg K ha⁻¹.

), while inorganic fertilizer management had significant effects on both varieties. In the Carrot native variety, the 75% inorganic fertilizer application resulted in the highest number of tubers per plant (4.82), tuber fresh weight (13,481.00 kg ha⁻¹), fresh and dry aboveground biomass (14,291.00 and 2181.60 kg ha⁻¹), and HI (0.50), though these values were not significantly different from the values achieved with the 50% inorganic fertilizer application. The interaction between PGPR dipping and 50% inorganic fertilizer application produced the highest tuber length and diameter of 11.35 and 3.56 cm, respectively (Table 5). In the Okinawan Orange variety, the 100% inorganic fertilizer yielded the highest tuber fresh weight (8866 kg ha⁻¹) and fresh and dry aboveground biomass (11,287 and 2222 kg ha⁻¹), though these values were not significantly different from the values achieved with the 75% inorganic fertilizer application. No significant interaction was found between PGPR and inorganic fertilizer in yield components, economic yields, or aboveground biomass (Table 6).

3.4. Nutritional Content in the Tubers of Sweet Potatoes

Next, we considered the effect of PGPR and inorganic fertilizer on the nutritional value of the sweet potatoes. The results showed that PGPR dipping significantly increased protein, fiber, and ash content in the Carrot native variety. Similar results were obtained in the Okinawan Orange variety, except in the case of ash. In both varieties, 100% inorganic fertilizer provided the highest carbohydrate content: 85.21% and 86.87%, respectively. The highest protein (4.47%) and ash (1.16%) contents were found for 50% inorganic fertilizer application in the Carrot variety, while the 75% inorganic fertilizer application produced the highest protein (5.91%) and ash (1.38%) contents in the Okinawan Orange variety. The interaction between PGPR and inorganic fertilizer was found to be significant for all nutritional values for both varieties (

Table 7. Nutritional content in the tubers of the sweet potatoes.

Treatment	Carrot Native Variety				Okinawan Orange Variety
	Carbohydrate (%)	Protein (%)	Fiber (%)	Ash (%)	Carbohydrate (%)	Protein (%)	Fiber (%)	Ash (%)
PGPR (D)
No PGPR (D1)	85.45 a	3.70 b	4.02 b	0.99 b	86.62 a	5.57 b	2.19 a	1.35 a
Dip PGPR (D2)	82.10 b	4.26 a	6.22 a	1.09 a	86.21 b	5.81 a	2.11 b	1.32 b
F-test (D)	**	**	**	**	**	**	**	**
Fertilizer (T)
No fertilizer (T1, Control)	84.37 b	3.54 d	5.25 c	0.94 e	86.39 c	5.55 c	2.09 c	1.23 d
25% fertilizer (T2)	83.22 c	4.00 c	5.72 a	1.12 b	86.32 c	5.52 c	2.14 b	1.33 c
50% fertilizer (T3)	82.80 d	4.47 a	5.41 b	1.16 a	85.93 d	5.73 b	2.32 a	1.39 a
75% fertilizer (T4)	83.26 c	4.41 b	4.93 d	1.01 c	86.56 b	5.91 a	2.10 bc	1.38 a
100% fertilizer (T5)	85.21 a	3.50 d	4.28 e	0.96 d	86.87 a	5.75 b	2.12 bc	1.36 b
F-test (T)	**	**	**	**	**	**	**	**
D × T
D1 × T1	86.18 a	3.42 e	3.71 g	0.88 f	87.11 b	4.99 d	2.09 e	1.18 g
D1 × T2	84.75 b	3.62 d	4.72 e	0.99 d	86.57 d	5.49 c	2.11 de	1.36 d
D1 × T3	84.18 c	4.43 bc	4.23 f	1.24 a	86.46 de	5.50 c	2.44 a	1.50 a
D1 × T4	85.98 a	3.68 d	3.69 g	0.88 f	86.70 c	5.92 b	2.10 e	1.31 e
D1 × T5	86.15 a	3.37 e	3.73 g	0.94 e	86.25 f	5.97 b	2.23 b	1.42 c
D2 × T1	82.57 d	3.65 d	6.79 a	1.00 d	85.66 h	6.12 a	2.08 e	1.28 f
D2 × T2	81.70 e	4.38 c	6.72 a	1.26 a	86.07 g	5.55 c	2.16 cd	1.30 ef
D2 × T3	81.42 f	4.51 b	6.59 b	1.07 c	85.40 i	5.96 b	2.21 bc	1.27 f
D2 × T4	80.54 g	5.13 a	6.17 c	1.14 b	86.42 e	5.89 b	2.10 e	1.46 b
D2 × T5	84.27 c	3.64 d	4.84 d	0.98 d	87.49 a	5.52 c	2.00 f	1.29 ef
F-test (D × T)	**	**	**	**	**	**	**	**
CV (%)	0.18	1.29	1.30	1.47	0.08	1.11	1.55	1.11

CV = coefficient of variation; ** = significant difference at p < 0.01. Different letters within the same column indicate significant differences based on the least significant difference (LSD) test at p < 0.05. T1 = no fertilizer (control); T2 = 11.72 kg N ha⁻¹ + 11.72 kg P ha⁻¹ + 39.85 kg K ha⁻¹; T3 = 23.44 kg N ha⁻¹ + 23.44 kg P ha⁻¹ + 79.69 kg K ha⁻¹; T4 = 35.16 kg N ha⁻¹ + 35.16 kg P ha⁻¹ + 118.69 kg K ha⁻¹; T5 = 46.88 kg N ha⁻¹ + 46.88 kg P ha⁻¹ + 159.38 kg K ha⁻¹.

).

3.5. Macronutrient Content in Aboveground Biomass of Sweet Potatoes

Measurement of the macronutrient content in the aboveground parts of both sweet potato varieties showed that PGPR dipping tended to significantly enhance the accumulation of P and K in the Okinawan Orange variety. Conversely, in the Carrot native variety, the accumulation of P and K was significantly enhanced without PGPR dipping. In both sweet potato varieties, the application of inorganic fertilizer significantly increased the accumulation of N, P, and K compared to no fertilizer application. The application of 100% inorganic fertilizer produced the highest accumulation of N in the Carrot native variety and the highest accumulation of P and K in the Okinawan Orange variety, while the application of 50 and 75% inorganic fertilizer produced the highest accumulation of K and P, respectively, in the Carrot native variety. In addition, there was an interaction between the use of PGPR and inorganic fertilizer management. PGPR dipping with 50% inorganic fertilizer application produced the highest accumulation of N at 1.99% in the Carrot native variety, while no PGPR dipping with 75% and 25% inorganic fertilizer produced the highest accumulations of P and K (0.099% and 1.65%), respectively. For the Okinawan Orange variety, no PGPR dipping with the application of 75% inorganic fertilizer produced the highest accumulation of N (1.87%), and PGPR dipping with 100% inorganic fertilizer application produced the highest accumulation of P and K (0.152% and 2.73%, respectively) (

Table 8. Macronutrient content in aboveground biomass of sweet potatoes.

Treatment	Carrot Native Variety			Okinawan Orange Variety
	N (%)	P (%)	K (%)	N (%)	P (%)	K (%)
PGPR (D)
No PGPR (D1)	1.73 b	0.078 a	1.36 a	1.78 a	0.087 b	1.87 b
Dip PGPR (D2)	1.80 a	0.077 b	1.34 b	1.65 b	0.097 a	2.04 a
F-test (D)	**	**	**	**	**	**
Fertilizer (T)
No fertilizer (T1, Control)	1.61 d	0.076 c	1.06 e	1.45 d	0.081 d	1.52 d
25% fertilizer (T2)	1.73 c	0.080 b	1.49 b	1.82 a	0.980 b	2.18 a
50% fertilizer (T3)	1.83 a	0.076 c	1.54 a	1.79 b	0.077 d	2.08 b
75% fertilizer (T4)	1.81 b	0.085 a	1.27 d	1.74 c	0.084 c	1.82 c
100% fertilizer (T5)	1.84 a	0.073 d	1.41 c	1.78 b	0.119 a	2.18 a
F-test (T)	**	**	**	**	**	**
D × T
D1 × T1	1.58 i	0.074 f	0.93 h	1.54 f	0.077 g	1.46 i
D1 × T2	1.65 h	0.081 c	1.65 a	1.85 ab	0.105 b	2.38 b
D1 × T3	1.67 g	0.070 gh	1.47 c	1.84 b	0.077 g	1.92 e
D1 × T4	1.86 c	0.099 a	1.30 e	1.87 a	0.088 cd	1.94 de
D1 × T5	1.89 b	0.069 h	1.46 c	1.80 c	0.086 de	1.64 g
D2 × T1	1.65 gh	0.079 d	1.18 g	1.36 g	0.085 e	1.57 h
D2 × T2	1.81 d	0.078 d	1.32 e	1.80 c	0.090 c	1.98 d
D2 × T3	1.99 a	0.083 b	1.61 b	1.74 d	0.077 g	2.24 c
D2 × T4	1.76 f	0.071 g	1.23 f	1.60 e	0.080 f	1.69 f
D2 × T5	1.79 e	0.077 e	1.36 d	1.76 d	0.152 a	2.73 a
F-test (D × T)	**	**	**	**	**	**
CV (%)	0.68	0.81	1.26	0.90	1.84	1.37

CV = coefficient of variation; ** = significant difference at p < 0.01. Different letters within the same column indicate significant differences based on the least significant difference (LSD) test at p < 0.05. T1 = no fertilizer (control); T2 = 11.72 kg N ha⁻¹ + 11.72 kg P ha⁻¹ + 39.85 kg K ha⁻¹; T3 = 23.44 kg N ha⁻¹ + 23.44 kg P ha⁻¹ + 79.69 kg K ha⁻¹; T4 = 35.16 kg N ha⁻¹ + 35.16 kg P ha⁻¹ + 118.69 kg K ha⁻¹; T5 = 46.88 kg N ha⁻¹ + 46.88 kg P ha⁻¹ + 159.38 kg K ha⁻¹.

).

3.6. The Correlation Coefficients of the Sweet Potatoes’ Agronomic Traits and Economic Yield

3.6.1. Carrot Native Variety

Through Pearson correlation analysis involving 17 agronomic traits of the Carrot native variety (

Table 9. The matrix correlation coefficients of agronomic traits in the Carrot native sweet potato.

Agronomic Traits	Vine Length	SCMR	Leaf Area per Plant−1	No. of Tubers per Plant−1	Tuber Length	Tuber Diameter	Tuber Fresh Weight Ha−1	Aboveground Fresh Weight ha−1	Aboveground Dry Weight ha−1	HI	Carbohydrate	Protein	Fiber	Ash	N	P	K
Vine Length	1
SCMR	0.587 **	1
Leaf Area per Plant−1	0.203 ns	0.393 *	1
No. of Tubers per Plant−1	−0.091 ns	−0.119 ns	−0.115 ns	1
Tuber Length	0.074 ns	0.047 ns	0.123 ns	−0.012 ns	1
Tuber Diameter	−0.168 ns	0.041 ns	0.048 ns	0.015 ns	0.549 **	1
Tuber Fresh Weight Ha−1	−0.066 ns	−0.074 ns	−0.235 ns	0.65 6 **	0.052 ns	0.206 ns	1
Fresh Weight of Aboveground Biomass ha−1	0.455 *	0.277 ns	−0.226 ns	0.233 ns	0.295 ns	0.162 ns	0.436 *	1
Dry Weight of Aboveground Biomass ha−1	0.471 **	0.209 ns	−0.342 ns	0.199 ns	0.087 ns	0.117 ns	0.424 *	0.819 **	1
HI	−0.551 **	−0.457 *	0.098 ns	0.091 ns	−0.323 ns	−0.058 ns	0.070 ns	−0.829 **	−0.695 **	1
Carbohydrate	0.063 ns	−0.146 ns	−0.185 ns	−0.128 ns	−0.054 ns	−0.141 ns	−0.189 ns	0.158 ns	0.329 ns	−0.270 ns	1
Protein	−0.163 ns	−0.099 ns	−0.001 ns	0.447 *	−0.177 ns	0.119 ns	0.445 *	−0.158 ns	−0.209 ns	0.43 9 *	−0.831 **	1
Fiber	−0.018 ns	0.235 ns	0.245 ns	−0.088 ns	0.233 ns	0.191 ns	0.013 ns	−0.101 ns	−0.340 ns	0.107 ns	−0.923 **	0.572 **	1
Ash	−0.166 ns	−0.199 ns	−0.022 ns	0.223 ns	−0.100 ns	0.155 ns	0.098 ns	−0.324 ns	−0.320 ns	0.405 *	−0.686 **	0.772 **	0.543 **	1
N	0.416 *	0.190 ns	−0.022 ns	0.084 ns	0.312 ns	0.384 *	0.427 *	0.609 **	0.576 **	−0.387 *	−0.223 ns	0.230 ns	0.183 ns	0.054 ns	1
P	−0.056 ns	−0.034 ns	−0.160 ns	−0.138 ns	0.478 **	0.502 **	0.328 ns	0.360 *	0.313 *	−0.196 ns	0.149 ns	−0.188 ns	−0.038 ns	−0.377 ns	0.295 ns	1
K	0.230 ns	−0.041 ns	−0.224 ns	0.045 ns	0.161 ns	0.148 ns	0.114 ns	0.176 ns	0.130 ns	−0.085 ns	−0.115 ns	0.149 ns	0.077 ns	0.264 ns	0.484 **	0.116 ns	1

ns = not significant; * = significant difference at p < 0.05; ** = significant difference at p < 0.01. The value of correlation: low (r ≤ 0.359), moderate (r ≥ 0.360–0.709), and strong (r ≥ 0.710). HI = harvest index; SCMR = SPAD chlorophyll meter reading; N, P, K = nitrogen, phosphorus, potassium content in aboveground biomass. The color coding in this table represents the following correlation types: = significant moderate positive correlation; = significant strong positive correlation; = significant moderate negative correlation; = significant strong negative correlation.

), 136 significant correlations were identified, comprising 80 positive and 56 negative relationships. According to the classification criteria of [40], these included 54 weak (r ≤ 0.359), 24 moderate (r = 0.360–0.709), and 2 strong correlations (r ≥ 0.710). Specifically, strong correlations were observed primarily between the fresh and dry weights of aboveground biomass ha⁻¹ (r = 0.819 **, p < 0.01). Furthermore, tuber carbohydrate content showed strong negative correlations with protein (r = −0.831 **, p < 0.01) and fiber (r = −0.923 **, p < 0.01) content. Regarding key yield-related traits, tuber fresh weight ha⁻¹ exhibited moderate positive correlations with the number of tubers per plant⁻¹ (r = 0.656 **, p < 0.01), nitrogen content in aboveground biomass (r = 0.427 *, p < 0.05), protein content in tubers (r = 0.445 *, p < 0.05), the fresh weight of aboveground biomass ha⁻¹ (r = 0.436 *, p < 0.05), and the dry weight of aboveground biomass ha⁻¹ (r = 0.424 *, p < 0.05). In terms of vegetative growth relationships, vine length showed moderate positive correlations with the SCMR (r = 0.587 **, p < 0.01), the fresh weight of aboveground biomass ha⁻¹ (r = 0.455 *, p < 0.05), and the dry weight of aboveground biomass ha⁻¹ (r = 0.471 **, p < 0.01) and a negative correlation with harvest index (r = −0.551 **, p < 0.01). Supporting this, the harvest index also exhibited a strong negative correlation with the fresh weight of aboveground biomass ha⁻¹ (r = −0.829 **, p < 0.01) and a moderate negative correlation with the dry weight of aboveground biomass ha⁻¹ (r = −0.695 **, p < 0.01). In terms of nutritional quality, the strong negative correlations between carbohydrate, protein, and fiber contents suggest a metabolic trade-off in nutrient composition. In addition, protein content showed a strong positive correlation with ash content (r = 0.772 **, p < 0.01). In terms of nutrient accumulation, nitrogen and potassium contents in aboveground biomass were moderately correlated (r = 0.484 **, p < 0.01). These findings indicate that maximizing tuber yield in the Carrot native variety requires balanced aboveground biomass development. This is consistent with the nitrogen accumulation in aboveground biomass, demonstrating the importance of strategic nitrogen management for sweet potato production.

3.6.2. Okinawan Orange Variety

Through Pearson correlation analysis of 17 agronomic traits of the Okinawan Orange variety (

Table 10. The matrix correlation coefficients of agronomic traits in the Okinawan Orange sweet potato.

Agronomic Traits	Vine Length	SCMR	Leaf Area per Plant−1	No. of Tubers per Plant−1	Tuber Length	Tuber Diameter	Tuber Fresh Weight ha−1	Aboveground Fresh Weight ha−1	Aboveground Dry Weight ha−1	HI	Carbohydrate	Protein	Fiber	Ash	N	P	K
Vine Length	1
SCMR	−0.121 ns	1
Leaf Area per Plant−1	0.575 **	0.076 ns	1
No. of Tubers per Plant−1	0.029 ns	0.150 ns	0.134 ns	1
Tuber Length	0.419 *	−0.085 ns	0.090 ns	0.030 ns	1
Tuber Diameter	0.275 ns	−0.187 ns	−0.141 ns	−0.115 ns	0.664 **	1
Tuber Fresh Weight ha−1	0.576 **	0.139 ns	0.449 *	0.482 **	0.281 ns	0.171 ns	1
Fresh Weight of Aboveground Biomass ha−1	0.619 **	0.024 ns	0.479 **	0.132 ns	0.243 ns	0.245 ns	0.618 **	1
Dry Weight of Aboveground Biomass ha−1	0.700 **	0.034 ns	0.638 **	0.218 ns	0.196 ns	0.046 ns	0.720 **	0.859 **	1
HI	−0.208 ns	0.271 ns	−0.166 ns	0.426 *	0.025 ns	−0.083 ns	0.313 ns	−0.485 **	−0.324 ns	1
Carbohydrate	0.398 *	−0.097 ns	0.170 ns	−0.196 ns	0.330 ns	0.067 ns	0.259 ns	0.368 *	0.303 ns	−0.092 ns	1
Protein	0.147 ns	0.169 ns	0.189 ns	0.376 *	−0.115 ns	−0.028 ns	0.289 ns	0.211 ns	0.300 ns	0.026 ns	−0.625 **	1
Fiber	−0.090 ns	0.170 ns	0.229 ns	0.474 **	0.097 ns	−0.133 ns	0.129 ns	−0.112 ns	−0.029 ns	0.290 ns	−0.327 ns	0.051 ns	1
Ash	0.356 ns	−0.108 ns	0.495 **	0.542 **	0.361 *	0.081 ns	0.454 *	0.265 ns	0.370 *	0.125 ns	−0.073 ns	0.305 ns	0.612 **	1
N	0.427 *	0.109 ns	0.615 **	0.080 ns	0.007 ns	−0.226 ns	0.396 *	0.253 ns	0.354 ns	0.059 ns	0.216 ns	−0.122 ns	0.352 ns	0.341 ns	1
P	0.545 **	0.026 ns	0.292 ns	−0.343 ns	0.042 ns	0.058 ns	0.340 ns	0.322 ns	0.474 **	−0.114 ns	0.593 **	−0.158 ns	−0.507 **	−0.143 ns	0.224 ns	1
K	0.497 **	0.118 ns	0.461 *	−0.281 ns	−0.040 ns	−0.017 ns	0.350 ns	0.255 ns	0.454 *	−0.055 ns	0.264 ns	−0.075 ns	−0.183 ns	−0.020 ns	0.544 **	0.774 **	1

ns = not significant; * = significant difference at p < 0.05; ** = significant difference at p < 0.01. The value of correlation: low (r ≤ 0.359), moderate (r ≥ 0.360–0.709), and strong (r ≥ 0.710). HI = harvest index; SCMR = SPAD chlorophyll meter reading; N, P, K = nitrogen, phosphorus, potassium content in aboveground biomass. The color coding in this table represents the following correlation types: = significant moderate positive correlation; = significant strong positive correlation; = significant moderate negative correlation.

), 136 significant correlations were identified, comprising 99 positive and 37 negative relationships. Initially, strong correlations were observed primarily between the fresh and dry weights of aboveground biomass ha⁻¹ (r = 0.859 **, p < 0.01), but additional strong relationships were found between phosphorus and potassium contents in aboveground biomass (r = 0.774 **, p < 0.01), as well as between tuber fresh weight ha⁻¹ and dry weight of aboveground biomass ha⁻¹ (r = 0.720 **, p < 0.01). Regarding key yield-related aspects, tuber fresh weight ha⁻¹ exhibited moderate-to-strong positive correlations with dry weight of aboveground biomass ha⁻¹ (r = 0.720 **, p < 0.01), fresh weight of aboveground biomass ha⁻¹ (r = 0.618 **, p < 0.01), and vine length (r = 0.576 **, p < 0.01). In terms of vegetative growth relationships, vine length demonstrated moderate positive correlations with dry weight of aboveground biomass ha⁻¹ (r = 0.700 **, p < 0.01), fresh weight of aboveground biomass ha⁻¹ (r = 0.619 **, p < 0.01), and leaf area per plant⁻¹ (r = 0.575 **, p < 0.01), while leaf area per plant⁻¹ displayed moderate positive correlations with the dry weight of aboveground biomass ha⁻¹ (r = 0.638 **, p < 0.01) and the nitrogen content in aboveground biomass (r = 0.615 **, p < 0.01). In terms of nutritional quality, carbohydrate content showed a moderate negative correlation with protein content (r = −0.625 **, p < 0.01), while fiber content displayed moderate positive correlations with ash content (r = 0.612 **, p < 0.01). Concerning nutrient accumulation, the strong positive correlation between phosphorus and potassium contents in the aboveground biomass (r = 0.774 **, p < 0.01) indicates synchronized uptake and accumulation of these macronutrients. These findings indicate that tuber yield improvement in the Okinawan Orange variety is strongly associated with enhanced aboveground biomass production, particularly dry matter accumulation and vigorous vine growth.

4. Discussion

4.1. Response of Sweet Potatoes to PGPR and Inorganic Fertilizer Management

Optimizing nutrient use efficiency (NUE) in highly demanding tuber crops is critical for global sustainability. Our findings indicate that PGPR can significantly improve early vegetative growth, highlighting a viable strategy for precision nutrient management in the early growth stage. During the 30–60 DAP period, the results of this experiment showed that PGPR significantly promoted internode length at 30 DAP in the Carrot native variety and the no. of nodes per plant⁻¹ and leaf area at 60 DAP in both varieties. This physiological enhancement aligns with the known functions of the selected PGPR strains. The observed increase in aboveground biomass, reflected by improved leaf area, is partly attributable to the nitrogen-fixing capabilities of A. brasilense [19] and B. mobilis [20]. Furthermore, some PGPR strains, such as A. brasilense [22] and A. vinelandii [23], can synthesize the hormone cytokinin, which stimulates cell division and vine elongation [41]. These effects likely contributed to the increased number of nodes per plant and internode length observed in the Carrot native variety during the early growth phase. Several studies have suggested that the use of PGPR in cropping systems can promote plant growth and increase crop yield. For example, Yasmin et al. [42] found that PGPR can significantly increase the growth and yield of sweet potato.

Ultimately, although PGPR significantly promoted growth in various parameters during the early period, this advantage was not sustained and did not translate into continued positive effects on growth parameters or final yield at harvest. This was evident in the fresh and dry weights of aboveground biomass and total yield, which showed an upward trend but did not differ significantly from the results for crops grown with no PGPR dipping. These findings are similar to the results of Ollio et al. [43], who found that the use of Azospirillum, Pseudomonas, Azotobacter, and Bacillus did not promote broccoli yield. Moreover, a relevant limitation of the experimental design was the reliance on a single, brief inoculation protocol: 30 min slip soaking without reinoculation. This method is likely insufficient to ensure a robust and persistent bacterial presence in the rhizosphere throughout the growth cycle, especially considering the rapid nutrient leaching and potential runoff in a rainy field environment [44]. This study also lacked any confirmation of viable PGPR cell density post-soaking or, more importantly, confirmation of successful colonization and survival in the field rhizosphere. In addition, the application of cattle manure to all experimental plots prior to planting introduced native microbial communities and additional nutrient sources [45]. This microbial abundance likely resulted in strong competition for niche and resources, making it exceedingly difficult for the low-density inoculum to establish and maintain dominance in the rhizosphere. The lack of a significant response may be attributable to the competitive advantage of native soil microflora over the introduced PGPR [46].

Furthermore, the specificity of the PGPR used may not have been suitable for these two sweet potato varieties, as different root exudates affect microbial responses to the plants [47,48]. The lack of host specificity between the PGPR and the crop might explain the absence of distinct growth promotion results, as other studies have successfully stimulated sweet potato growth using different PGPR strains. For instance, Yu et al. [18] utilized Bacillus cereus AR156, Bacillus subtilis SM21, and Serratia sp. XY21, while Sadeq et al. [49] employed Bacillus tequilensis UPMRB9 and Bacillus subtilis UPMB10. These strains demonstrated statistically significant potential for sweet potato growth. Consequently, the differences in microbial species suggest that the PGPR used in this experiment are not as effective in associating with the two sweet potato cultivars. Additionally, methodological differences in the application may have contributed to the reduced efficacy. Notably, Sadeq et al. [49] soaked cuttings for 48 h and reinoculated biweekly, whereas our protocol involved only 30 min of soaking in shade before planting, with no subsequent inoculation. Ahemad and Kibret [50] reported that environmental conditions during soaking may affect the efficacy of microorganisms, such as in shade or dark conditions. Although soaking was performed in the shade, without direct sunlight, there is a higher probability of a temperature increase, which is a critical factor that reduces PGPR effectiveness. In contrast, soaking under dark conditions helps to maintain a more stable solution temperature, which aligns with the fundamental principles of microbial inoculant storage necessary to preserve maximum viability before true application [51]. However, the sensitivity of PGPR strains to temperature and light stressors varies. Therefore, the reduced efficacy in our experiment highlights the methodological challenge of single inoculation under competitive field conditions.

Overall, the trial showed that there was a linear relationship between the amount of inorganic fertilizer applied and the gain in productivity, and data were available to quantify that effect. The experimental results showed that the 75% and 100% inorganic fertilizer applications had positive effects on the growth and yield of the Carrot native and Okinawan Orange sweet potatoes, reflected in the vine length, leaf area, and tuber fresh weight. In the case of the Carrot native variety, increasing inorganic fertilizer from 75% to 100% did not make a significant difference to the tuber fresh weight, reflecting the potential to reduce the use of inorganic fertilizers without affecting the growth and yield. This result is consistent with those of Kumar et al. [52], who found that, when nitrogen fertilizer was applied at a rate of 80%, sweet potatoes grew and yielded no less than when nitrogen fertilizer was applied at a rate of 100%. On the other hand, the 100% inorganic fertilizer application tended to more significantly increase the Okinawan Orange tuber fresh weight. In addition, genetic variation plays a significant role in the nutrient requirements of plants [53]. This difference in response underscores the necessity of cultivar-specific nutrient management strategies, particularly in nutrient-poor environments like the sandy soils of Northeast Thailand. This indicates that, if farmers are interested in growing Japanese sweet potatoes in rural areas, they should apply 100% inorganic fertilizers.

Lastly, the study confirmed the potential for inorganic fertilizer in the Carrot native variety and underscored the varying nutrient needs of different varieties. The hypothesis concerning the synergistic effect of PGPR and reduced fertilization on final yield was not supported under the conditions tested. The results reinforce the view that applying biofertilizers in real-world agricultural systems remains challenging, as external factors and methodological limitations significantly influence efficacy. To advance the integration of PGPR technology in sweet potato production, future research should focus on (1) the colonization and survival of PGPR in the rhizosphere under diverse environmental conditions, (2) the selection of PGPR with specific characteristics and suitability for sweet potato and local soil types, (3) the investigation of cultivar and host specific mechanisms using molecular methods, (4) the exploration of integrated nutrient management strategies combining biofertilizers with inorganic fertilizers, and (5) controlled environment testing including proper control plots (e.g., without manure application) to accurately separate the effects of the inoculated strains from those of the native microbial community and organic inputs.

4.2. Effect of PGPR and Inorganic Fertilizer on Nutritional Composition of Sweet Potatoes

Although PGPR exerted limited influence on growth and yield, it significantly enhanced the protein and fiber contents in the tubers. Generally, sweet potatoes contain two types of fiber: soluble fiber and insoluble fiber. Soluble fiber absorbs water in the digestive tract, thus softening stool, preventing constipation and, when both types are mixed with a group of good bacteria in the intestines, stimulating the production of short-chain fatty acid compounds as a source of energy in the intestines, consequently strengthening the intestines, supporting digestion, and improving nutrient absorption [54]. At the same time, there was a noticeable decrease in carbohydrate content, contrary to the results of Yasmin et al. [30]. It is noted that the use of PGPR strains containing A. brasilense [19], which involves nitrogen fixation, enhances a plant’s nitrogen uptake for protein synthesis. Furthermore, PGPR strains such as A. brasilense [22] and A. vinelandii [23] can synthesize cytokinin hormones, which stimulate cell division in various plant organs, including roots, stems, and lateral buds [41]. This may contribute to the plant utilizing sugars produced via photosynthesis to synthesize fiber components such as lignin, hemicellulose, and cellulose, which are key constituents of the cell wall required for tuber development. Additionally, the sugars from photosynthesis can serve as precursors for protein synthesis and accumulation in sink organs such as tubers, resulting in an increase in protein and fiber and contributing to the reduction in carbohydrates in sweet potato tubers. This is consistent with the correlation analysis in Table 9 and Table 10, where there is a negative correlation between carbohydrates, protein, and fiber in both varieties. The reduction in carbohydrates could negatively affect taste and overall nutritional value for those who primarily consume plant-based carbohydrates. These changes in nutritional composition were observed with both PGPR and inorganic fertilizer management.

The 50% inorganic fertilizer produced low carbohydrate and high protein contents, which was explained by the optimization of nitrogen for metabolic processes. At moderate nitrogen supply, plants achieved optimal nitrogen use efficiency through balanced carbon-nitrogen metabolism [55]. The reduced carbohydrate content indicates that photosynthetic products were channeled toward protein synthesis under optimal nitrogen conditions. In contrast, high inorganic fertilizer (75% and 100%) may result in plant luxury nitrogen consumption, promoting vegetative growth and biomass accumulation (Table 5 and Table 6). Excess nitrogen may likely be stored as nitrate, free amino acids, or protein in aboveground biomass [3] rather than incorporated into tuber storage. On the other hand, low inorganic fertilizer (0% and 25%) limited protein synthesis. Under nitrogen deficiency, plants shifted carbon partitioning toward carbohydrate and starch accumulation rather than nitrogen-demanding compounds [56]. This compensatory mechanism redirects carbon toward carbohydrate storage when nitrogen deficiency limits protein synthesis. Since inorganic fertilizer influences not only nutritional composition but also overall productivity, optimal fertilizer recommendations should balance protein enhancement against carbohydrate content and economic yield.

5. Conclusions

This study confirms the crucial roles of inorganic fertilizer management and the sweet potato variety in sweet potato production in Northeast Thailand. Adequate inorganic fertilizer management and sweet potato varieties are the main factors in increasing economic yield. For instance, the Carrot native variety with 75% inorganic fertilizer application produced higher yields than the Okinawan Orange variety, which required up to 100% inorganic fertilizer application. Although PGPR dipping tended to increase the growth and yield, the response was inconsistent and did not result in statistically significant differences under the tested experimental conditions. However, PGPR dipping significantly enhanced tuber quality, particularly the protein and fiber contents, but concurrently decreased carbohydrate levels, which may affect the taste and the nutritional quality of sweet potatoes. The limited growth response could be attributed to several factors. First, the tested PGPR strains (A. brasilense, A. vinelandii, and B. mobilis) may not have been fully compatible with the sweet potato genotypes used, as host–microbe specificity strongly influences colonization efficiency and growth promotion. Second, the inoculation protocol consisting of 30 min of slip soaking without subsequent reapplication may have been insufficient to maintain effective bacterial populations in the rhizosphere throughout the growing period. These methodological constraints could explain the conditional ineffectiveness of the PGPR treatments observed in this study. Future research should focus on selecting PGPR with specific characteristics and suitability for sweet potato and local soil types, and exploring the concentration and inoculation times of biofertilizers.