Morphological and physiological response of rice roots to the application of calcium polyamino acid under saline-alkali soil conditions

ABSTRACT Although application of calcium polyamino acid (Ca-polyAA) is one of the effective ways to improve saline-alkali soil and promote plant growth, the clear mechanisms are unknown. A two-year field experiment with four Ca-polyAA application treatments (added at rates of 0, 7.5, 15.0, and 22.5 × 103 kg ha−1) was conducted on saline-alkali soil in Songnen Plain of China. In the first year (2016), Ca-polyAA increased soil porosity and decreased pH, electrical conductivity (EC1:5) and exchangeable sodium percentage (ESP). Root Na+/K+ and Na+/Ca2+ ratio were reduced whereas hydraulic conductivity was increased. The hormone concentration and enzyme activity in the root were enhanced at tillering stage. Root structure development at the root tip was promoted after the application of Ca-polyAA. The total root length, surface area, volume and biomass were increased by increasing Ca-polyAA added rate. Similar results were obtained in the second year (2017). Path analysis demonstrate that Ca-polyAA can promote the accumulation of root biomass mainly by decreasing Na+/ K+ ratio, increasing root hydraulic conductivity and the activity of carbon metabolism enzymes, and the development of the root cortex. These results revealed that Ca-polyAA, as an efficient ameliorator, is capable to enhance agronomical and morphophysiological traits of rice in saline-alkali soil.


Introduction
Salinized soils are widely distributed all over the world, especially in arid and semiarid areas. The total area of saline-alkali soil is 11.74 million km 2 according to Hassania et al. (2020). High salinity leads to poor structure and low nutrient levels, eventually evolving into desertification issue. Planting rice is one of the effective ways to reclaim saline-alkali soils and restore ecological stability (Xu et al. 2020). During rice planting, irrigation water rinses salt ions from the topsoil into subsoil and improves the survival rate of crops efficiently, which leading to the rice yield was higher than that of other land crops grown under conditions of less frequent irrigation in saline-alkali soil (Qadir et al. 2003). Although rice planting is the most common agricultural mode to improve soil quality in heavy saline-alkali soil, the biomass was strongly reduced and the rice yield was considerably decreased due to the suppressed physiological metabolism under salt stress (Rawat et al. 2012).
The rice root system, as the prominent part in contact with saline-alkali soil, has a strong physiological response to an unfavorable growth environment. Under salt stress, rice roots absorb excessive Na + , resulting in an imbalance in Na + /K + ratio, which lead to the toxicity of sodium ions and osmotic stress and eventually inhibited root development (Yang et al. 2008). Meanwhile, the decreased total root length, root volume, root number, and darkening of the root tip are manifest (Neves et al. 2010). Excessive salt concentration reduces the water potential of soil solution and worsens water stress, so it is difficult for rice roots to absorb water under the reduced root vigor. In addition, increase of relative salt concentration in cells leads to imbalances in the physiological metabolism. So carbon metabolism enzymes activity in the root system decreases, which reduces the carbon assimilation ability (Nam et al. 2012) and limits the growth of the root system. Furthermore, salt stress signals affect hormone synthesis and transport, including cytokinins, ethylene and abscisic acid (ABA), thus root system architecture is influenced consequently (Osmont et al. 2007). On the other hand, increase of ABA and other hormones in roots can effectively alleviate the inhibition of salt stress on root differentiation (Galvan-Ampudia and Testerink 2011). The combination of the above factors leads to the differentiation and growth of rice roots in severe sodic alkaline soil hindered. Therefore, alleviation of rhizosphere salt stress is critical for rice growth in saline-alkali environment.
Rice root system also has strong morphological response to adverse growth environment. High salinity stress inhibits the degree of lignification, transport tissue, and epidermal cell (Gowda et al. 2011). Specifically, the cortical tissue, one of the primary structures of fine roots, occupies a large proportion of the root cross section and affects root absorption and radial transport (Gowda et al. 2011). During the normal growth of rice root, the parenchyma cells of the root cortex die and disintegrate, and the radial wall of the dead cells gather together to form a larger cavity, which is quickly occupied by the air to become aerenchyma (Kawai et al. 1998). The aerenchyma provides a smooth passage for oxygen supply in the aerobic respiration of the root, which limits the potential damage to the plant (Zhang et al. 2016). Some oxygen in the aerenchyma is transported to the root tip and leak to the surrounding soil, which provides an aerobic environment for rhizospheric microbes and effectively prohibits the accumulation of salt damage in root cells (Akcin et al. 2015). However, Krishnamurthy et al. (2009) studied the root system of rice under salt stress by fluorescence microscopy and found that the outer cortex cell wall was substantially thickened and a large amount of cork had accumulated, and the assimilation and transport of harmful ions by roots was reduced. Fernandez-Garcia et al. (2009) found that the cortical cells of plant roots could reduce the migration of Na + to the stele with excessive salt concentration. Consequently, the lignification degree of phloem parenchyma cell wall can directly reflect the adaptability of root system to salt stress.
To relieve salinity stress, the desalting material was added to the soil. The chemical remediation method was adopted globally for its rapid desalination compared with bioremediation method, and its easy operation compared with the physical restoration strategy. Chemical amendments are noted for its high desalination efficiency by replacing Na + with cations under severe saline-alkali soil. For instance, Gypsum or flue gas desulfurization gypsum is the most widely used modifier in view of the wide range of industrial access, but it also brings a large amount of SO 4 2to soil ). Considering the numerous environmental problems caused by the various inefficient soil modifiers, an environmentally friendly amendment is highly desired. Known as a non-toxic and natural biodegradable biopolymer, polyamino acid (polyAA) is often used in the pharmaceutical, biological and food fields. In recent years, polyAA has been used as a modifier to repair soil (Yang et al. 2018). In addition, polyAA is also used as a fertilizer synergistically reduce fertilizer application. Therefore, we propose to use the polyamino acid react with Ca 2+ , as novel organic soil amendments in saline sodic soil remediation. Firstly, calcium polyamino acid could accelerate soil desalination process in root zone via replacement of Na + with Ca 2+ and improvement of soil condition. Additionally, amino acid derivatives, as an organic substance widely used in agriculture, can regulate the soil nutrient by providing active C, N elements to improve crop growth, and decrease the dispersion of soil particles by increasing soil organic carbon (Moreira et al. 2015). In the current study, two consecutive years of field experiment were established by adding Ca-polyAA in a severe saline-alkali soil. We explored the effect of Ca-polyAA on rice root growth in saline-alkali soils by analyzing the physical and chemical properties of soil, morphological and the anatomical structures of root system, as well as the change of endogenous hormonal level.

Site description and experimental design
The field experiment was conducted at Lianhe town in Da'an city, Jilin province, China (45°28′13′′N, 124°06′0.7′′E) in April 2016. The region has a temperate monsoon climate with the rainfalls of 655 and 570 mm in 2016 and 2017. The soil is a typical soda-type saline-alkali soil with pH of 10.42 and EC 1:5 of 3.20 dS m −1 and the initial soil properties are described in detail in Table S1.
The amendment used in this study was commercial calcium polyamino acid (Ca-polyAA) ( Figure S1) which purchased from Beijing Dezhixin Science & Technic Co., Ltd. (Beijing, China). The amendment with a purity of 90% (The pH of 5% aqueous solution is 4.0 to 7.0), have the following elemental content: Ca 14.0%, C 29.7%, N 9.5%; K 0.5%; Mg 0.2%. The tested rice variety was Baidao 8. The field experiment with four treatments was conducted with Ca-polyAA adding rates of 0 (control), 7.5, 15.0, and 22.5 × 10 3 kg ha −1 . The powdered Ca-polyAA was evenly sprinkled to the surface soil and mixed it with 20 cm soil by rotary tiller before planting, followed with irrigating field, then the water layer was kept about 5 cm for 3 days before discharge. The area of each plot was 0.05 ha, and plots randomly arranged in a block design with four replications. The space between plots was 1.5 m width per test interval. Before planting, 100 kg compound fertilizer (N-P-K nutrient content: 13 kg-18 kg-14 kg) was applied. In addition, 15 kg (NH 4 ) 2 SO 4 was applied to each plot at the green stage (7 days after planting), tillering stage (25 days after planting), and panicle stage (45 days after planting) of the rice growing season. Planting was conducted using rice transplanter machine with density of 2.7 × 10 5 points ha −1 .

Sampling and measurement
Soil samples were collected after rice harvest in October 2016 and 2017, respectively. Five randomly soil cores were collected in 'S' pattern from each plot and mixed to form a composite sample. Soil sampled by the quaternion method, then were air-dried, ground, uniformly mixed, and sieved through 2-mm and 1-mm sieves. Therefore, each treatment had four soil samples per year. The soil properties determined according to the methods described by Tiquia et al. (2002). The soil pH and EC 1:5 was determined according to the mass ratio of soil/water at 1:5. Soil Na + concentrations were determined using a Flame Atomic Absorption Spectrophotometer (4530 F, INESA, Beijing, China). The exchangeable sodium percentage (ESP) was calculated by the equation described by Feng et al. (2019). Soil porosity was calculated from soil particle density and dry bulk density of the sample according to Glab and Kulig (2008).
Tillering stage is the vulnerable period for the growth of rice to survive particularly in hard salinealkali soil, because the high salt concentration is a fatal threat to the survival of rice. During the tillering stage of rice in 2016 and 2017, 15 rice plants with uniform growth were selected in each plot, and the whole plant was removed in each plot. After washing with distilled water, the total root length, root surface area, root volume, root mean diameter, and diameter distribution of rice roots were analyzed by Root Scanning Software (Expression 10000XL 1.0, Epson Inc. Japan and Regent Instruments Inc., Canada). The shoot and root parts of rice were put into oven to dry to constant weight, as weighed by electronic scales. The sucrose synthase (SS) and sucrose phosphate synthase (SPS) enzyme activity were determined with the method described by Douglas and Prescott (1989). After freezing, grinding, centrifugation, and extraction of the rice root system, indole-3-acetic acid (IAA), abscisic acid (ABA), gibberellin A3 (GA3), and zeatin riboside (ZR) were detected by HPLC-MS /MS (AH352, Mato, Chicago, USA). Root vigor was determined using the triphenyl tetrazolium chloride (TTC) according to the methods described by Liu et al. (2008). Root hydraulic conductivity (L pr ) was calculated by dividing Root hydraulic conductance (K r ) by the root volume according to the methods described by Kamaluddin and Zwiazek (2002). Dry rice roots were crushed and boiled with sulfuric acid. Then Na + , K + , Ca 2+ concentration of root in the supernatant was determined by Flame Atomic Absorption Spectrophotometer (4530 F, INESA, Shanghai, China).
The root tip of rice was immersed in glutaraldehyde (3%) and fixed with 0.1 mol L −1 phosphate buffer twice for more than 2 h. Next, the 0.1 mol/L phosphate buffer solution was used to rinse for 10 min, then dehydrated with acetone gradient, pure acetone dehydrated twice with EMBED 812 resin permeated and subsequently placed in a dry place for 8 h. The anatomical structure of root tip was observed by Transmission Electron Microscope (LIBRA 120, Berlin, Germany). Analysis of root vessel area, stele area, epidermis thickness, and total cortical thickness data were conducted with CellSence Standard Software (Olympus Corporation, Osaka, Japan).

Statistical analysis
Analysis of variance was performed with the general linear model procedure by SPSS v.20 (IBM Corp., Armonk, NY, USA). The statistical model used includes sources of variation due to year or the rate of addition of Ca-polyAA, and interactions of year × rate of addition of Ca-polyAA. Multiple comparisons of means were compared based on the least significant difference (LSD) test at the 0.05 significance level. All the data were presented as mean plus or minus standard error (SE) for quadruplicate replicates. Path analysis was performed using DPS V15.10 data processing system (Ruifeng, Hangzhou, China). In the process of path analysis used the 'progressive regression model' and the relationship between the variables is linear and additive causality. While, the effect of the direct/ indirect action is expressed in the absolute value. The charts were developed using Origin 9.1 (OriginLab, Northampton, MA, USA).

Soil characteristics
Changes in soil pH, EC 1:5 , ESP and porosity after the application of Ca-polyAA are the important parameters showing the amelioration effects. Table 1 demonstrates that application of Ca-polyAA can obviously affect these parameters in two consecutive years. With the increased rate of Ca-polyAA, soil porosity increased while pH, EC 1:5 , ESP decreased. Specifically, compared with the control, the soil pH with Ca-polyAA applied were significantly (P < 0.05) decreased by 2.9%-8.4%, EC 1:5 decreased by 12.5%-44.5%, ESP decreased by 19.3%-43.2%, porosity increased by 3.5%-15.7% in 2016. Similar results were observed in 2017. Compared with control, 22.5 × 10 3 kg ha −1 treatment has the best performance to improve soil properties.

Root Na + , K + , Ca 2+ concentration
The addition of Ca-polyAA drastically affected the Na + , K + , Ca 2+ concentration of rice root in saline-alkali soil (Table 2). Compared with control, the rice root Na + concentration was significantly decreased by 8.3%, 22.9%, and 31.2% under the 7.5, 15.0, and 22.5 × 10 3 kg ha −1 treatments in 2016, and by 7.8%, 21.1%, 30.6% in 2017, respectively. The results demonstrate that application of Ca-polyAA reduced the accumulation of Na + into the root system. Meanwhile Ca-polyAA increased root K + and Ca 2+ concentration as well, thus, root Na + /K + and Na + /Ca 2+ ratio were decreased.

Root biomass, shoot biomass and total biomass
The addition of Ca-polyAA can dramatically affect the shoot and root biomass of rice in saline-alkali soil at the tillering stage ( Figure 1). For 2016, compared with control, the rice shoot biomass increased by 47.6%, 89.9%, and 101.2% under the 7.5, 15.0, and 22.5 × 10 3 kg ha −1 treatments, respectively. The rice root biomass increased by 62.2%, 96.2%, and 109.4%, respectively. For 2017, compared with control, the rice shoot biomass increased by 40.5%, 106.6%, and 116.3% under the 7.5, 15.0, and 22.5 × 10 3 kg ha −1 treatments. The rice root biomass increased by 54.1%, 83.6%, and 98.3%, respectively. Therefore, the total biomass of rice all dramatically increased over the two years. There was no significant (P > 0.05) difference in shoot biomass, root biomass, and total biomass between the 15.0 × 10 3 and the 22.5 × 10 3 kg ha −1 treatments. Two-factor analysis showed that the biomass of rice was not significantly (P > 0.05) affected by the year of cultivation, which suggested that there is no significant (P > 0.05) interaction between the factors of planting years and the rate of addition of Ca-polyAA. Whereas the biomass of rice was significantly (P < 0.05) affected by the rate of addition of Ca-polyAA.

Root length, surface area, root volume, and average diameter
Normally, the root length, surface area, root volume, and root mean diameter are utilized to characterize the growth of rice as listed in Table 3   There was no significant (P > 0.05) difference of root length and mean root diameter between the 22.5 × 10 3 and the 15.0 × 10 3 kg ha −1 treatments. Two-factor analysis showed that the root length, root surface area, root volume, and average diameter were no significantly (P > 0.05) affected by the year of planting. Based on two consecutive year experiments, Ca-polyAA affected the rice root significantly (P > 0.05) in root length, root surface area, root volume, and average diameter. Furthermore, the repetitive results indicated that the Ca-polyAA application rate had significantly affected these parameters regardless of planting year.

Rice root enzyme activity and hormone concentrations
The root growth of rice is a process of photosynthetic product accumulation under the catalysis of carbon metabolism enzyme (Kraemer et al. 1998). The added rate of Ca-polyAA had dramatic effect on root vigor, SS activity, and SPS activity in saline-alkali soil ( , respectively. Two-factor analysis showed that the application of Ca-polyAA affected the root vigor, SS enzyme activity, and SPS enzyme activity. However, there was no significant (P > 0.05) interaction between the year of planting and the application rate of Ca-polyAA. These results show that the addition of Ca-polyAA in saline-alkali soil can drastically increase the activity of carbon metabolism enzymes in rice roots that promote the carbon assimilation by roots, and improve the redox ability of roots and their ability to obtain nutrients and water. The differentiation of rice roots is regulated by endogenous hormones. The changes of endogenous hormones in rice roots under different application rate of Ca-polyAA are shown in Figure 3. There was no clear effect of increasing the application rate of Ca-polyAA on ZR and GA3 of rice roots, but the effects on IAA and ABA were apparent. Compared with control, the rice root IAA concentration of 7.5, 15.0, and 22.5 × 10 3 kg ha −1 treatments increased by 7.3%, 33.9%, and 35.5% in 2016, and by 0.7%, 33.6%, and 37.8% in 2017, respectively. The rice root ABA content of the three treatments increased by 13.6%, 18.4%, and 20.6% in 2016, and increased by 6.5%, 18.2%, and 21.1% in 2017, respectively. Two- Table 4. Rice root vigor, sucrose synthase (SS), and sucrose phosphate synthase (SPS) enzyme activity at the tillering stage in saline-alkali soil with different addition rates of Ca-polyAA in 2016 and 2017.

Root structure
The anatomical structure of the root tip of rice can directly reflect the development of rice root system in saline-alkali soil (Pater and Schilperoort 1992). The root tip cross sections under different Ca-polyAA application rates are shown in Figure 4. The cells in the control and 7.5 × 10 3 kg ha −1 treatments were disorderly distributed, and no clear stele structures were observed (Figure 4(a), Figure 4(b)). With the increase of adding rate, the primary xylem conduit and epiphytic xylem conduit were differentiated more orderly. Under the 22.5 × 10 3 kg ha −1 treatment, a mature stele structure and root differentiation were observed which were very close to unstressed state (Figure 4(d)).
During the normal development of rice roots, the cells of the root cortex are programmed to die and form ventilatory tissue, which provides oxygen for respiration and metabolism of root system and can be used to characterize the development of rice root system. The parenchyma cells of rice root were arranged closely, and the cell volume was smaller treated with 0 and 7.5 × 10 3 kg ha −1 without obvious ventilation tissue (Figure 4(e, f)). With the increase of adding rate, the cell wall of parenchyma cells ruptured and formed evident ventilation tissue. Particularly in 22.5 × 10 3 kg ha −1 treatment, the expanded and slender aerenchyma distributed in radial direction along the root (Figure 4(h)), and the ventilation tissues were connected with each other.
The root phloem cell is one of the most important cells in the root of rice, and its activity can directly affect the formation of mature root tissue (Pater and Schilperoort 1992). The structure of root phloem cells in rice treated with 0 and 7.5 × 10 3 kg ha −1 was relatively simple, and organelle development was relatively retarded (Figure 4(i _ n)). Increasing the application rate promoted the development of root phloem cells. The mitochondria and Golgi apparatus were clearly presented in the cells treated with 22.5 × 10 3 kg ha −1 (Figure 4(p)), and the whole development of the cells was mature.
Quantitative analysis of root vessel area, stele area, epidermis thickness, and total cortical thickness of rice under different Ca-polyAA application rates is presented in Table S2. A mature catheterized column could not be detected in plants grown under the control and 7.5 × 10 3 kg ha −1 treatments; however, those grown under 22.5 × 10 3 kg ha −1 treatment demonstrated the largest catheter area and stele area (4.87 × 10 3 µm 2 and 42.33 × 10 3 µm 2 , respectively). The epidermis thickness and total cortical thickness increased with the increase of application rate of Ca-polyAA. Compared with control, the rice epidermis thickness of plants grown under the 7.5, 15.0, and 22.5 × 10 3 kg ha −1 treatments increased by 27.8%, 32.5%, and 81.7%, respectively. The total cortical thickness increased by 33.9%, 46.1%, and 71.5%, respectively. Therefore, the application of Ca-polyAA can promote the development of the whole root system (Table S2). Correlation analysis was conducted between root biomass and vessel area, stele area, epidermis thickness and total thickness ( Figure S2). There was a significant positive correlation between root biomass and each parameter, with the greatest positive correlations found between the total thickness and the root biomass (R 2 = 0.7575).
The primary reason that crops are difficult to grow in saline-alkali soil is that serious salt stress environment reduced the L pr (Meng and Fricke 2017). The addition of Ca-polyAA notably affected rice root hydraulic conductivity . Compared with control, the L pr under the 7.5, 15.0, and 22.5 × 10 3 kg ha −1 treatments increased by 36.1%, 59.5%, and 76.2% in 2016, respectively, and increased by 37.0%, 61.3%, and 81.4% in 2017, respectively. The application of Ca-polyAA in saline-alkali soil alleviates the stress of crop physiological water shortage by increasing the water conductivity of root system (Figure 2).

Building favorable rhizosphere soil environment for rice
High salt content, high alkalinity, and low nutrient content are the main soil factors restricting crop growth in saline-sodic soils. The soils in the current study were selected from typical soda-alkali soil in Songnen Plain of China with average pH of 10.42, total water-soluble salt of 9.21 g kg −1 , and ESP of 65.7%. To guarantee the survival rate of rice, the critical step in soil remediation is to build a tillage layer with low salt content. Herein, soil amendment was applied to paddy field and irrigated with 10 cm water layer for several days before transplanting seedlings. Firstly, according to Oades's (1984) research, soil macroaggregate (>250 μm) and micro-aggregate (<250 μm) depend on organic matter for stability and the stability of microaggregates is also enhanced by multivalent cations, which act as bridges between organic colloids and clays. High levels of Ca-polyAA as organic materials can highly promote aggregating soil colloid and increase soil porosity and hydraulic conductivity (Table 1). Secondly, the amendment provided large amount of active Ca 2+ to replace excess Na + on the cation exchange complex thereby improving the physical condition of the soil and increasing water infiltration (Poonia and Bhumbla 1973). As a result, the increase of soil porosity and Lpr promoted the leaching of Na + , and effectively inhibit the upward migration of Na + . We speculate that part of exchanged salt were dissolved and discharged from the field with the runoff of a lot of irrigation water, and rest of dissolved salt were leached into the deeper soil layer under the hydrostatic pressure of the water according to research findings by Ma et al. (2008), leading to the rapid desalination of salt in the surface layer of saline-alkali soil. Our results revealed that soil EC 1:5 and ESP in 0-20 cm layer were decreased 44.9% and 44.5% after reclamation with Ca-ployAA for two years, and the effect was more notably with the adding levels increasing (Table 1). This may be due to the fact that high-level Ca-polyAA provide more Ca 2+ , which can replace more Na + and thus reduce soil EC and ESP. The decrease of salinity and alkalinity of topsoil would ameliorate the adverse effects of salt on plant and soil solution relations (Rengasamy 2002). In addition, nutrients from Ca-polyAA can greatly compensate for the deficiency of plant nutrient in the severe saline-alkali soil (Table S3), which was beneficial for root growth. Previous studies also have proved that polyAA derivatives will be decomposed into low molecular mass matters when added into soil and eventually mineralized into a considerable amount of inorganic N, C and other nutrients, which improved soil fertility (Zhou et al. 2019). The application of Ca-polyAA can provide a suitable rhizosphere microenvironment for the development of the rice root system not only by ameliorating salt stress but also enhancing nutrient supply.

Root development in saline-alkali soil under Ca-polyAA application
As commonly known, rice is highly susceptible to the rhizosphere salinity compared to other cereals after transplanting. Once the seedlings are transferred from population growth to individual growth in the high salt environment, the tillering stage is most vulnerable to salt stress (Li et al. 2003). Once root system is unable to develop normally under stress during that period, rice seedlings will wilt and die (Galvan-Ampudia and Testerink 2011). Kawai et al. (1998) also demonstrated that the organelle structure was destroyed under adversity stress conditions, leading the loss of physiological function. In the present study, the survival rate of rice was at least 95% with application of Ca-polyAA, while that was only 40% without Ca-polyAA in the field investigation (data not presented). Furthermore, Ca-polyAA, as a source of Ca, K, N, and Mg, improves root growth, the root and total biomass of 15.0 × 10 3 kg ha −1 treatment were dramatically higher than that of control and 7.5 × 10 3 kg ha −1 treatment. However, there was no significant increase (P > 0.05) when Ca-polyAA was added to 22.5 × 10 3 kg ha −1 (Figure 1). So, in view of the reclamation cost, we reckon 15.0 × 10 3 kg ha −1 as a relatively appropriate addition. Similar observations have been reported in previous studies. Zhang et al. (2017) found that polyAA greatly strengthened the plant nutrient uptake capacity through enhancing both root biomass and activity. The increases in the root biomass and can be attributed to the decrease in salinity and sodicity and the increase in nutrients in the fields after reclamation with Ca-polyAA. At the same time, compared with control, the root vigor in 22.5 × 10 3 kg ha −1 treatment increased drastically by 11.1 and 10.0 μg g −1 h −1 in 2016 and 2017, so the application of Ca-polyAA could promote root differentiation and enhance the assimilation ability of rice . Transmission electron microscopic picture of rice root tip cross section magnified (a, b, c, d, f: ×1500; e, g, h: ×3000; i, j, k, l, m, n, o, p: ×15,000) at tillering stage with different application rates of Ca-polyAA (a, e, I, j: 0 × 10 3 kg ha −1 ; b, f, j, n: 7.5 × 10 3 kg ha −1 ; c, g, k, o: 15.0 × 10 3 kg ha −1 ; d, h, l, p: 22.5 × 10 3 kg ha −1 ). seedlings via the remediation of salt stress (Table 4). After the application of Ca-polyAA, particularly in 22.5 × 10 3 kg ha −1 treatment, mature vessel and stele structures appeared in the root system of rice, and the aerated tissue was connected with each other, which dramatically promoted the growth of rice seedlings. Our results revealed that the application of Ca-polyAA notably promoted the development of the root system in the saline-alkali environment.

Physiological mechanisms of Ca-polyAA application in saline-alkali soil
The regulation of rice root under the application Ca-polyAA is attributed to many factors. First of all, large amount Ca 2+ loaded by Ca-polyAA were released and provided nutrient in rice growth (Razzaque et al. 2010). Calcium is a substance regulates physiological metabolism and crop growth (Fasani et al. 2019). The Ca 2+ intake keeps the cell membrane system stable and enables the transduction of calcium signaling system, so as to guarantee the growth of the root system (Wu et al. 2017). Furthermore, our results indicated that the reduction of Na + concentration, leading to lower Na + /K + (0.51-0.81) and Na + /Ca 2+ ratio (2.93-5.09) in the rhizosphere and dramatic alleviation the salt toxicity, which is consistent with the findings of Sun et al. (2014), and the effect was more obvious with the application levels increasing. At the same time, our chemical modifier can provide a considerable amount of N, K, Mg, which improved cell membrane stability and nutrient uptake under salinity stress to facilitate root growth (Yildirim et al. 2009). Noteworthy, salinity and sodicity symptoms manifest not only in root but also in shoot (ultimate visual damage symptom). A substantially lower Na + concentration through reduced uptake and translocation to shoots, probably resulting in more favorable Na + -K + homeostasis in functional leaves and actively growing tissues (Rahman et al. 2016). Meanwhile, the increased nutrient contents (N, P, K) in the rice leaves (Table S3) provided by Ca-polyAA relieved shoot stress and promoted root growth based on the feedback mechanism between leaves and roots.
Our results also validated that the physiological basis of root growth is to reduce the Na + /Ca 2+ ratio of root system and increase the hydraulic conductivity in salt condition (Table S4). Besides, Ca-polyAA affects hormone secretion in rice and regulates root growth (Garay-Arroyo et al. 2012). The growth of rice roots is a process of continuous accumulation of photosynthetic products under the condition of carbon metabolism enzymes. We found that the application of Ca-polyAA all dramatically increased the activity of SS (10.5-14.5 mg suc −1 g −1 h −1 ) and SPS (5.5-6.7 mg suc −1 g −1 h −1 ) in the root system (Table 4). The results conforms to previous results that organic materials can promote root growth by increasing the activity of carbon metabolism enzymes (Vieweg 1974). According to the microstructure of rice root phloem cells with Ca-polyAA, the mitochondria and Golgi apparatus can be witnessed in the phloem cells and the development of the cells was close to that of stress-free state, thus to attain the matured tissues ( Figure 4). Similar result has been found by Xie et al. (2009), and it may be one of the physiological mechanisms through which Ca-polyAA promote root growth in rice.
Path analysis was conducted and showed that there were significant correlations between root biomass and root hydraulic conductivity, carbon metabolism enzymes, hormones content, anatomical parameters, and root structure (R 2 = 0.7654, P = 0.0032). Na + / K + ratio (X 1 ), L pr (X 2 ), root activity (X 3 ), SS activity (X 4 ), SPS activity (X 5 ), IAA (X 6 ), ABA (X 7 ), Zr (X 8 ), GA3 (X 9 ), epidermis thickness (X 10 ) and total thickness (X 11 ) were taken as independent variables, and rice root biomass was taken as the dependent variable (Y). The analysis demonstrated a significant relationship between dependent variables and independent variables, with a fitted equation of: Y = −21.674 − 0.011X 1 + 0.0024X 2 + 0.589X 3 + 0.546X 4 + 0.001X 5 + 272X 6 + 3.290X 7 + 0.016X 8 + 1.308X 9 + 0.125X 10 + 0.045X 11 (R 2 = 0.7654, F = 453.96, P = 0.0032) (Table S4). Based on the analysis of the effect of independent variables on dependent variables, Na + / K + ratio, L pr , SPS activity and total cortical thickness had the largest relative direct-action coefficients. This indicates that Ca-polyAA can promote the accumulation of rice root biomass in saline-alkali soil by decreasing Na + / K + ratio to increase root hydraulic conductivity, the activity of carbon metabolism enzymes, and the development of the root cortex.
The above results suggest that Na + / K + ratio, L pr , SPS activity and total cortical thickness contributed considerably on rice root biomass accumulation ( Figure S3). Overall, Ca-polyAA reduced the salt content in the rhizosphere and cut the absorption of Na + in root which promoted root hydraulic conductivity and the accumulation of rice root biomass by increasing the activity of carbon metabolism enzymes, the content of hormones, and the development of the root cortex. However, under the application of Ca-polyAA, the field of rice root exudates in saline-alkali soil may need to be explored further.

Conclusions
The application of Ca-polyAA could promote the growth of rice roots in saline-alkali soil. First, Ca-polyAA improved the desalination efficiency of root zone soil with irrigation water resulting in reduced the salt concentration in rhizosphere soil. Furthermore, Ca-polyAA promoted the accumulation of rice root biomass in saline-alkali soil by increasing root hydraulic conductivity, the activity of carbon metabolism enzymes, the hormones concentrations, and the development of the root cortex, especially by decreasing Na + /K + ratio. The positive role in soil properties and rice root growth of Ca-polyAA in plant-soil system suggested that Ca-polyAA (adding rate of 15.0 × 10 3 kg ha −1 ) could be an effective soil amendment to enhance rice root growth in saline-alkali soil in Songnen Plain of China.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
This work was supported by National Key Research and Development Program of the Ministry of Science and Technology of China (2016YFC0501205, 2016YFC0501208, and 2017YFD0200706) and National Natural Science Foundation of China (21775163).