Introduction

Sweet potatoes are an essential food crop in many developing countries, and may hold potential for creating and maintaining food security in the future (Ma et al. 2012; Khan et al. 2016). The yield, stability, and improvement of sweet potatoes and other crops are essential for plant cultivators and breeders (Mukhopadhyay et al. 2011). Soil moisture content and N are the main limiting factors for the growth of sweet potato roots, and essential for growth (Marschner et al. 1996; Khan et al. 2016). Under drought conditions, sweet potato plants undergo changes in root morphology and physiological metabolism which help to adapt environmental stresses (Liu and Cheng 2011). The response mechanism of plant roots to N under different soil moisture conditions is a topic which has raised widespread concern among the agricultural, ecological, and environmental fields (Villordon et al. 2014).

The current understanding of the morphological characteristics of sweet potato roots is based on the explicit deployment of root axes to determine the spatial configuration of the root system, which exhibits great plasticity in response to external environmental conditions such as nutrient availability and soil moisture content (Lynch 1995; Bao et al. 2014). Recent studies have shown that the spatial and temporal fluctuations of N concentrations and morphology in soil environment, known as local N regimes, trigger a systemic signal that regulates plant root growth and development (Xu et al. 2012). It is apparent that localized N supply is critical for sweet potato root development (Lima et al. 2010); in ideal N conditions, the metabolic activity of the nitrogen-donating root system is increased, as is the distribution of the assimilate to the N-donating root system (Granato and Jr 1989).

Under drought conditions, the root morphology of sweet potatoes adapts in order to resist environmental stress (He and Dijkstra 2014). Sweet potatoes have been identified as a moderately drought tolerant crop; however, very sensitive to water deficit in the initial growth stage of storage roots (Mukhopadhyay et al. 2011; Villordon et al. 2012). Previous studies have shown that N can improve the drought resistance of crops, and regulating N levels may help reduce the impact of drought stress (Guo et al. 2003); it has also been acknowledged that root morphology is closely related to the efficient absorption of water and nutrients (Wright and Wright 2004). The root system can activate complex regulatory networks to cope with fluctuations in the rhizosphere soil environment, thereby promoting the efficient uptake and utilization of water and nutrients (Xuan et al. 2017). A sound understanding of the possible relationship between the local N supply and the plant’s ability to thrive in various moisture conditions is essential, as this knowledge can contribute to the development and testing of management practices that improve nutrient and water use efficiency and promote root development (Wang et al. 2013).

The objective of present study was to investigate the effect of variability in local N availability measured by the root development of sweet potatoes under different water conditions, examining the effects of drought and N availability on the morphological characteristics of roots. This information will provide further insight into the external environmental stimuli of plant roots, such as soil moisture variability and nutrient availability, which can promote or hinder root formation.

Materials and Methods

Biological materials and soil

The sweet potato genotype Yanshu 25, mainly grown in northern China, was used as an experimental plant. Soil was collected from the top 20 cm of the soil profile at Jiaozhou Experimental Station, Qingdao Agricultural University (36.3°N, 120.3°E) in Shandong Province. The sampled soil was air dried and passed through a 2.0 mm sieve. The tested soil had the following characteristics with total N 0.3 mg·kg⁻¹, available P (Olsen-P) 6.64 mg·kg⁻¹, available K (NH₄OAc-K) 32.4 mg·kg⁻¹, organic C 10.1 mg·kg⁻¹, and pH of 7.24. After drying at room temperature, the test substrate was prepared by mixing soil and river sand in a ratio of 1:1 (v/v). The nutrient solution was added once as a base fertilizer to ensure the normal growth of the test plants.

Culture device and experimental design

A modified two-compartment culture system was constructed (Villordon et al. 2013). A 35 cm high PVC pipe with a diameter of 12 cm was cut vertically into two equal parts with the same height and width. The pipe was divided by a 3 mm thick acrylic strip, separating it into a left root compartment and a right root compartment. The acrylic strip was secured with PVC glue to minimize the lateral movement of water. A notch at the top end of the acrylic strip allowed a sweet potato cutting to be set such that the vine on either side of the basal leaf gap or node were directed toward separate compartments.

A completely randomized design with a 2×3 factorial arrangement of treatments was used. The factors were soil moisture contents and N combinations in left and right compartments. The experiment established two conditions: normal (75–80% of field capacity) and drought (45–50% of field capacity). Three different N combinations were also tested in the left and right compartments; 0 and 50 mg·kg⁻¹; 0 and 150 mg·kg⁻¹; 50 and 150 mg·kg⁻¹. Therefore, the following six treatments were tested: N₀/N₅₀ under normal soil moisture; N₀/N₁₅₀ under normal soil moisture; N₅₀/N₁₅₀ under normal soil moisture; N₀/N₅₀ under drought condition; N₀/N₁₅₀ under drought condition and N₅₀/N₁₅₀ under drought condition. This process was repeated four times for each experiment. The pot study was carried out during the pre-growing season of sweet potato from March to May in a greenhouse at Qingdao Agricultural University, with 14 hours of light and 10 hours of darkness comprising each 24-h photoperiod, day/night temperatures of 25°C/16°C, and 60% relative humidity.

Sampling and analysis

During the first two weeks of our experiment, normal moisture conditions were maintained and soil moisture was maintained at 75 ± 5 % of field capacity and drought (soil moisture maintained at 45 ± 5 % of field capacity) conditions for the next three weeks. The plants were harvested after a total growth period of five weeks.

At harvest, the roots were floated on a waterproof tray and scanned using a dedicated Epson v700 scanner. The image acquisition parameters and the analysis accuracy were both set to "High", using the WinRHIZO (Version 2009) software for image acquisition and analysis.

Data analysis

Analysis of variance was carried out using the S.P.S.S. software, version 19.0. Duncan's multiple range or Fisher's LSD was used to show significant differences between treatment means at P < 5%. Relationships between treatments were tested by Pearson's correlation analyses. The redundancy analysis (RDA) used the Canoco version 4.5 software package. Significance of the first and of all ordinations axes was calculated by the Monte Carlo permutation test.

Results

Fig. 1: Some statistical descriptive data of root biomass of sweet potato plants in two-compartment culture system. Paired t test was performed to determine if roots in each compartment varied in lateral root biomass

Fig. 2: Adventitious root average diameter and root of the total volume in sweet potato as influenced in two-compartment culture system under two moisture conditions

Root biomass (RB) was significantly affected by N combinations in both compartments under both soil moisture contents (P < 0.05) (Fig. 1). Under both moisture conditions, the dry weight of RB in the N₀/N₅₀ treatment was higher in the N₅₀ compartment than in N₀ compartment, and the increase between left and right compartments (N₀/N₅₀) under normal conditions was twice that of drought conditions. However, the RB in the N₀/N₁₅₀ treatment showed the complete opposite effect under both water conditions: the RB in the N₁₅₀ compartment was significant higher than in the N₀ compartment under normal moisture conditions, while N₁₅₀ compartment decreased RB compared to the N₀ compartment under drought stress. The RB showed no significant differences between N₅₀ and N₁₅₀ compartments under normal moisture conditions. However, the plants in the N₁₅₀ compartment decreased RB by half compared to those in the N₅₀ compartment under drought stress.

The root length (RL), root surface area (RSA), and root tip numbers (RTN) of sweet potato roots determine nutrient and water absorption efficiency. Under normal moisture conditions, RL and RSA in the N₅₀ or N₁₅₀ compartment were all significantly higher than in the corresponding N₀ compartment (Table 1). Drought stress decreased the RL and RSA of all N combinations. The RL and RSA of the plants in the N₀/N₁₅₀ compartment showed the opposite trend under both moisture conditions. In the N₅₀/N₁₅₀ treatment, the RL and RSA showed no significant differences from each other, while under drought stress, the RL and RSA in the N₁₅₀ compartment showed a significant decrease compared to the N₅₀ compartment.

The root average diameter (RAD) and root volume (RV) were used to characterize root differentiation. The effects of different N concentrations in left and right compartments on root differentiation differed significantly under both soil moisture conditions (Fig. 2A and B). Under normal moisture conditions, RV in right compartments (N₅₀ or N₁₅₀) were significant higher than in left compartment (N₀). In the N₅₀/N₁₅₀ treatment, N₅₀ compartment increased RV compared to the N₁₅₀ compartment. Under drought stress, RV in the N₅₀ compartment was significantly higher than in other compartments (N₀ or N₁₅₀). However, root morphological characteristics in the compartment with N₁₅₀ showed reduced RV compared with the N₀ compartment.

Based on the classification criteria of (Noh et al. 2013) sweet potatoes, their roots can be divided into the following categories: fibrous, secondary, and tuberous roots. Under normal moisture, total fibrous and tuberous root volume in the N₁₅₀ compartment was significant higher than that in N₀ compartment (Table 2). Under drought stress, total fibrous and tuberous RV in the N₀/N₁₅₀ treatment showed the complete opposite trend. There were no differences in total fibrous RV between the N₅₀ compartment and N₁₅₀ compartments. Under drought stress, total fibrous RV in the N₁₅₀ compartment was significant lower than in other compartments (N₀ or N₅₀).

RDA analysis was performed to calculate the contribution of factors of local N supply and the correlation between various explanatory variables under different moisture conditions. The cosine of the angle between the Table 1: Effects of different nitrogen rates on root morphological characteristic of sweet potatoes under different moisture conditions

Moisture treatment	Local nitrogen supply		Length of lateral root		Root surface area		Tips
	treatment combination		(cm·pot^-1)		(cm^2·pot^-1)		(numbers)
Normal	N0	N50	853.04 b	2098.47 a	123.13 b	340.66 a	1461.03b	2475.33a
	N0	N150	1094.65 b	2465.38 a	146.73b	374.23a	1861.11 b	3400.66 a
	N50	N150	2048.49 a	2249.06 a	345.57 a	348.59 a	3381.67 a	2972.33 a
Drought	N0	N50	1073.65b	1559.37 a	153.05b	248.97a	1850.55 a	1747.66 a
	N0	N150	1359.59a	919.60 b	177.09 a	123.6b	1916.44 a	1438.33b
	N50	N150	1603.26 a	1222.16 b	264.41 a	159.38b	2291.33 a	1804.00 b

Table 2: Effect of nitrogen supply roots on root diameter and distribution of sweet potato roots under different moisture conditions

Moisture treatment	Local supply treatment	V ≤ 1.50 mm	1.50 < V ≤ 3.00 mm	3.00 < V ≤ 4.50 mm	V > 4.50 mm
Normal	N0	0.98b	0.48b	0.48b	0.54b
	N50	3.32a	1.11a	0.65a	4.04a
	N0	1.23b	0.39b	0.44a	0.67b
	N150	3.68a	1.27a	0.40a	1.04a
	N50	2.78a	1.27a	0.50a	3.40a
	N150	2.96a	1.06ab	0.42a	0.20b
Drought	N0	1.39b	0.32b	0.25a	0.85b
	N50	2.38a	1.04a	0.40a	1.81a
	N0	1.89a	0.72a	0.74a	1.34a
	N150	1.06b	0.63a	0.54a	0.65b
	N50	2.69a	1.45a	0.73a	1.83a
	N150	1.48b	0.46b	0.42ab	0.02b

Fig. 3: Representative adventitious root samples under normal and drought conditions from two-compartment culture system. The adventitious roots that were still attached to the plant were placed on red cloth to facilitate image capture

explanatory variables in the RDA analysis graph indicated the correlation. Under normal moisture conditions, the explanatory variables were mostly concentrated between N₅₀ and N₁₅₀ treatments, indicating that appropriate level of N fertilizer can induce root development and promote root differentiation and enlargement (Fig. 4A). However, with the variation of the soil moisture, the explanatory variables also changed. Under drought conditions, the explanatory variables were mostly concentrated around the N₅₀ treatment, indicating it as most suitable amount. While N₅₀ seemed to alleviate drought stress, the excessive application of N (N₁₅₀) appeared to increase drought stress. RDA analysis showed a significant positive correlation between N₅₀ and root growth and differentiation, but there was a significant negative correlation with N₀ and N₁₅₀ treatments (Fig. 4B).

Fig. 4: Independent and interactive action of growth substrate moisture and nitrogen rate (N) on the properties of the root morphological characteristics in ordination diagrams from redundancy analysis (RDA). Under normal moisture conditions, the coordinate from the first ordination axes explained 66.5% of the variance. The significance (according to Monte Carlo permutation tests) of all canonical axes was p = 0.024, indicating that the presence of N₅₀ and N₁₅₀ had a significant influence on the sweet potato root morphological characteristics. Under drought conditions, the coordinate from the first ordination axes explained 45.5% of the variance. The significance (according to Monte Carlo permutation tests) of all canonical axes was P=0.044, indicating that the presence of N₅₀ had a significant influence on the sweet potato root morphological characteristics. N₀=without nitrogen supply; N₅₀=50mg·N kg^-1; N₁₅₀= 150 mg·N kg^-1

Discussion

The spatial fluctuations of N concentrations (local N supply) are critical for sweet potato root signal transduction, growth and development (Zhang and Forde 1998; Lima et al. 2010; Xu et al. 2012). The root phenotype is based on the response of the root system to the local supply of nutrients, or the competitive response to local changes in nutrients (Zhang et al. 2007). In the present study, different N supplies in each compartment could significantly affect the root development of sweet potatoes; when the N concentrations on both sides were different, the plants in each compartment showed different levels of N competition. Our studies suggested a competitive advantage relative to the compartment with local N supply of (N₅₀), compared to the compartments with absent N or very high N, in terms of access to nutrients and soil moisture. Those were similar to earlier work (Kim et al. 2002; Villordon et al. 2012) which provided evidence that within a certain range of N application, the total volume of roots in the early stage of sweet potato development increased with the increase of N application rate; however, the total amount of root differentiation gradually decreased.

The ability of plant roots to proliferate preferentially in nutrient-rich soil has been well documented in the literature (Zhang et al. 2007). N deficiency triggers a “foraging” response, wherein the roots continue to deepen as though they were seeking N deeper in the soil. Meanwhile, N saturation signals the roots to enrich in the soil surface (Okamoto et al. 2013; Tabata et al. 2014). Previous studies on wheat roots showed that the increase in water infiltration depth caused by high N application directly affected the growth and distribution of crop roots. When the N application rate is high, the roots of the crops are mainly distributed in the upper soil to absorb the surface moisture. When N levels are low, the crop roots expand to the lower layers of soil and increase the absorption of the deeper soil moisture (Wang et al. 2001). The present study results showed that the root morphology of plants with access to a suitable N supply shows uniform dispersal growth, while in the case of N deficiency or N excess, the root morphology shows a pronounced pattern of "long and thin" roots or "short and fat" roots, respectively, under normal water conditions (Fig. 3 A). Water and nutrients are not only the main stress factors affecting dryland agricultural production, but also a pair of factors which are complementary and interactive. Soil moisture affects the transformation and availability of nutrients in soil; in turn, these nutrients also affect the ability of plants to uptake water efficiently and can mitigate or exacerbate drought stress (Huang et al. 2002).

Sweet potatos are more drought-tolerant than other crops, but their rooting, branching, and tuber stages are relatively sensitive to moisture conditions (Zhang et al. 1999). Adequate root growth after transplanting requires adequate water supply (Belehu and Hammes 2004). Drought stress will adversely affect adventitious root differentiation of sweet potatoes and hinder the formation of tuber roots, ultimately resulting in the reduction of the number of tuber roots (Villordon et al. 2012). The present study showed that the growth and differentiation of the corresponding root system of N₀ and N₅₀ were not significantly reduced under drought stress, while the root biomass in the N₁₅₀ compartment was significantly decreased (Fig. 3B). Our results were similar to previous researches on wheat (Liang and Chen 1996; Zhang and Zhang 2001), which provided evidence that under drought conditions, excessive application of N fertilizer would lead to a significant decrease in root volume and biomass. Li and Shao (2000) also reported that under water stress, excessive application of N fertilizer to wheat led a significant increase in the rate of root cell membrane damage, deterioration of the root water environment, decreased water retention capacity, and reduced drought resistance. Possible reasons were that excessive N application promotes the increase of root biomass in the surface layer of soil, which has little significance for actual drought resistance as it does not provide sufficient water storage capacity, and the lack of deep roots leaves plants unable to draw moisture from deeper levels of soil when there is little water accessible at the surface (Jackson et al. 2008). Nitrogen supply significantly increases root moisture content and enhances roots’ ability to absorb and retain water (Passioura 1983). Appropriate amounts of N can increase the total root weight in deep soil, thereby enhancing the water absorption capacity of the roots, decreasing the cell membrane damage rate, and improving the ability to resist dehydration and maintain turgor pressure. Recent studies have shown that in order to reduce the effects of drought stress on N uptake after plant roots sense drought signals, plants activate specific signals to promote N uptake (He and Dijkstra 2014). In particular, drought stress caused mutations in certain proteins, which reduced plants’ N absorption but enhanced drought resistance (Guo et al. 2003; Castaings et al. 2009; Marchive et al. 2013), indicates that plants can adjust their N-collecting activity under drought stress to maintain survival. Rational application of N fertilizer can meet plants’ needs for soil nutrients, promote the growth and development of sweet potatoes, and increase biomass and yield. It can also improve the physiological functions of sweet potatoes, as well as reducing the adverse effects of drought stress. However, these outcomes are greatly affected by the amount of N fertilizer applied. With the increase of the amount of N fertilizer applied, the beneficial effects of N gradually weaken.

Conclusion

The present study investigated the effects of three different N combinations on root morphological traits under two levels of moisture in sweet potatoes. Results of the present study demonstrated that variation in N rate and local availability profoundly affect root architecture development in sweet potatoes. The appropriate N application rate (N₅₀) can both promote the development of roots and induce roots to differentiate into storage roots in normal or drought condition. However, excessive application of N fertilizer (N₁₅₀) might aggravate drought stress, the amount of N fertilizer should be appropriately reduced under drought conditions. The results of this research can guide the revision or enhancement of current management practices.

Acknowledgments

This research was funded by the China Agriculture Research System (No. CARS-10-B10). We are grateful to Prof. Qing Chen for valuable comments on an earlier version of this manuscript. We are also thankful for the constructive comments received from anonymous reviewers and the editors.

References

Bao Y, P Aggarwal, NE Robbins, CJ Sturrock, MC Thompson, HQ Tan, C Tham, L Duan, PL Rodriguez, T Vernoux, SJ Mooney, MJ Bennett, JR Dinneny (2014). Plant roots use a patterning mechanism to position lateral root branches toward available water. Proc Natl Acad Sci 111:9319‒9324

Belehu T, PS Hammes (2004). Effect of temperature, soil moisture content and type of cutting on establishment of sweet potato cuttings. South Afr J Plant Soil 21:85‒89

Castaings L, A Camargo, D Pocholle, V Gaudon, Y Texier, Stéphanie Boutet-Mercey, L Taconnat, JP Renou, F Daniel-Vedele, E Fernandez, C Meyer, A Krapp (2009). The nodule inception-like protein 7 modulates nitrate sensing and metabolism in Arabidopsis. Plant J 57:426‒435

Granato TC, RC Jr (1989). Proliferation of maize (Zea mays L.) roots in response to localized supply of nitrate. J Exp Bot 40:263‒275

Guo FQ, J Young, NM Crawford (2003). The nitrate transporter atnrt1.1 (chl1) functions in stomatal opening and contributes to drought susceptibility in Arabidopsis. Plant Cell 15:107‒117

He M, FA Dijkstra (2014). Drought effect on plant nitrogen and phosphorus: a meta-analysis. New Phytol 204:924‒931

Huang ML, XP Deng, DZ Bai (2002). Progress on compensative effects of nitrogen and phosphorus on physiological processes and yield formation of wheat in dryland. Acta Agron Sin 22:74‒78 (In Chinese)

Jackson LE, M Burger, TR Cavagnaro (2008). Roots, nitrogen, transformations, and ecosystem services. Annu Rev Plant Biol 59:341‒363

Khan MA, DC Gemenet, V Arthur (2016). Root system architecture and abiotic stress tolerance: current knowledge in root and tuber crops. Front Plant Sci 7; Article 1584

Kim SH, K Mizuno, S Sawada, T Fujimura (2002). Regulation of tuber formation and adp-glucose pyrophosphorylase (AGPase) in sweet potato [Ipomoea batatas (L.) Lam] by nitrate. Plant Growth Regul 37:207‒213

Liang YL, PY Chen (1996). Effects of soil moisture and nitrogen and phosphorus nutrition on root growth of winter wheat. Acta Agron Sin 22:476‒482 (In Chinese)

Li YY, MA Shao (2000). Physiological and ecological responses of wheat roots to water and nitrogen fertilizers. J Plant Nutr Fert 6:383‒388

Lima JE, S Kojima, H Takahashi, NV Wiren (2010). Ammonium triggers lateral root branching in Arabidopsis in an ammonium transporter1; 3-dependent manner. Plant Cell 22:3621‒3633

Liu ZL, D Cheng (2011). Plant drought-resistant physiology research progress and breeding. Chin Agric Sci Bull 27:249‒252

Lynch J (1995). Root architecture and plant productivity. Plant Physiol 109:7‒13

Ma DF, Q Li, QH Cao, FX Niu, YP Xie, J Tang, HM Li (2012). Development and prospect of 337 sweet potato industry and its technologies in China. Jiangsu J Agric Sci 28:969‒973 (In Chinese)

Marchive C, F Roudier, L Castaings, V Bréhaut, E Blondet, V Colot, C Meyer, A Krapp (2013). Nuclear retention of the transcription factor NLP7 orchestrates the early response to nitrate in plants. Nat Commun 4; Article 1713

Marschner H, EA Kirkby, I Cakmak (1996). Effect of mineral nutritional status on shoot-root partitioning of photoassimilates and cycling of mineral nutrients. J Exp Bot 47:1255‒1263

Mukhopadhyay SK, A Chattopadhyay, I Chakraborty, I Bhattacharya (2011). Crops that feed the world 5. Sweet potato. Sweet potatoes for income and food security. Food Secur 3:283‒305

Noh SA, HS Lee, YS Kim, KH Paek, JS Shin, JM Bae (2013). Down-regulation of the\r, ibexp1\r, gene enhanced storage root development in sweet potato. J Exp Bot 64:129‒142

Okamoto S, H Shinohara, T Mori, Y Matsubayashi, M Kawaguchi (2013). Root-derived CLE glycopeptides control nodulation by direct binding to HAR1 receptor kinase. Nat Commun 4; Article 2191

Passioura JB (1983). Roots and drought resistance. Agric Water Manage 7:265‒280

Tabata R, K Sumida, T Yoshii, K Ohyama, H Shinohara, Y Matsubayashi (2014). Perception of root-derived peptides by shoot LRR-RKs mediates systemic n-demand signaling. Science 346:343‒346

Villordon AQ, I Ginzberg, N Firon (2014). Root architecture and root and tuber crop productivity. Trends Plant Sci 19:419‒425

Villordon AQ, D LaBonte, N Firon, C Edward (2013). Variation in nitrogen rate and local availability alter root architecture attributes at the onset of storage root initiation in ‘beauregard’ sweet potato. HortScience 48:808‒815

Villordon AQ, D LaBonte, J Solis, N Firon (2012). Characterization of lateral root development at the onset of storage root initiation in Beauregard sweet potato adventitious roots. HortScience 47:961‒968

Xu GH, XR Fan, AJ Miller (2012). Plant nitrogen assimilation and use efficiency. Annu Rev Plant Biol 63:153‒182

Xuan W, T Beeckman, GH Xu (2017). Plant nitrogen nutrition: sensing and signaling. Curr Opin Plant Biol 39:57‒65

Wang FH, Ren DC, Wang XQ, Cao HX, YU SL, YU ZW (2001). Effect of fertilization on wheat root activity delaying flag leaf senescence and yield. J Trit Crops 20:51‒54 (In Chinese)

Wang G, TJ Fahey, S Xue, F Liu (2013). Root morphology and architecture respond to N addition in Pinus tabuliformis, west China. Oecologia171:583‒590

Wright AN, RD Wright (2004). Horhizotron: a new instrument for measuring root growth. HortTechnology 14:560‒563

Zhang GS, RS Zhang (2001). Effects of nitrogen and phosphorus nutrition on wheat root development under water stress. Chin Gansu J Agric Coll 2:163‒167 (In Chinese)

Zhang H, H Rong, D Pilbeam (2007). Signalling mechanisms underlying the morphological responses of the root system to nitrogen in Arabidopsis thaliana. J Exp Bot 58:2329‒2338

Zhang H, BG Forde (1998). An Arabidopsis mads box gene that controls nutrient-induced changes in root architecture. Science 279:407‒409

Zhang MS, F Tan, QT Zhang (1999). Relationship between physiological changes and drought resistance of sweet potato under water stress. HortSeed 19:35‒39