Homeostatic responses and growth of Leymus chinensis under incrementally increasing saline-alkali stress

The experimental site

We performed experiments in an outdoor terrace on the campus of the College of Earth Sciences, Jilin University, Changchun, Jilin Province, northeast China. This is in the north temperate continental climatic zone, which has four distinct seasons. The mean annual temperature and precipitation are 4.8 °C and 567 mm, respectively, and mean winter and summer temperatures are -14 and 24 °C, respectively (Zhang et al., 2017).

Experimental design

N and P addition

After the seeds germinated, a 200 ml portion of Hoagland’s nutrient solution with adjusted N and P concentrations was added to each pot once every two days. The elemental composition of Hoagland’s solution is described in detail elsewhere (Li et al., 2018). In five sets of nine pots, the N concentration was adjusted using ammonium nitrate to establish five N concentration treatments (2, 4, 8, 16, and 24 mmol/L) in which the concentration of P was fixed at one mmol/L. In another five sets of nine pots, the P concentration was adjusted using potassium dihydrogen phosphate to establish five P concentration treatments (0.3, 0.6, 1.2, 2.4, and 4.8 mmol/L) in which the concentration of N was fixed at 14 mmol/L. The experiment thus involved 10 nutrient treatments and a total of 90 pots (10 treatments ×3 replicates ×3 salinity stress intervals) in a randomized complete block design. The N and P concentration gradients were set on the basis of previously reported results (Yu et al., 2011).

Saline-alkali stress

On 7 Aug 2017, after the L. chinensis plants had grown to an average height of about 15 cm, plants in triplicate pots were subjected to the saline-alkali stress treatments shown in Table 1, by adding NaHCO₃ to the previously described treatment solution to simulate various degrees of land salinization encountered in the western regions of Jilin Province. To avoid excessively stressing the plants, the pH was increased at intervals of 14, 17, or 23 days. The pH was raised from 7.5 to 8.1, and then to 8.4, 8.7, and finally 9.3. Aboveground parts of L. chinensis plants subjected to these treatments were harvested (cutting from the ground level) according to the schedule shown in the Table 1.

Sample collection and testing methods

Aboveground parts of L. chinensis plants were cut according to the harvesting schedule shown in Table 1. The samples were rinsed with distilled water, dried with absorbent paper, deactivated for 15 min in a dry air oven (105 °C) and dried for 12 h at 65 °C to eliminate water completely. After cooling the samples to room temperature, they were weighed, crushed, screened with a 100 mesh sieve, and finally quartered. The resulting sub-samples were then packed in sealed bags for analysis.

Total N and total P in the L. chinensis samples were respectively determined by Chinese standard methods LY/T 1269-1999 (using a SAN++ Continuous Flow Analyzer; Skalar, Netherlands) and LY/T 1270-1999 (using an ICPS-7500 inductively coupled plasma atomic emission spectrometer; Shimadzu, Kyoto, Japan).

Data analysis

The acquired data were statistically analyzed using SPSS 24 (SPSS Inc., USA). Homeostasis indices were calculated using the stoichiometric homeostasis model $y=c{x}^{\frac{1}{H}}$. Here, x is the measured content of an element in the soil, y is the measured content of the same element in L. chinensis, c is a constant coefficient, and H is the homeostasis index. The results were visualized using Sigmaplot 12.5 (Systat Software, Inc.). We calculated specific growth rates based on recorded changes in dry mass and instantaneous growth rates using the equation u = ln(Mt∕M0)∕t, where u is the specific growth rate (day⁻¹), M_t and M₀, are the final and initial dry mass, respectively, and t is experiment duration in days. Details of several statistical analyses are included in Appendix 1 and Appendix 2.

Effects of incremental increases in saline-alkali stress on N, P contents and N:P ratios

The 14-day stress increment yielded the highest N and P contents in the plants when the N concentration in the substrate was low (2–8 mmol/L), but not when the N concentration in the substrate was higher (16 or 24 mmol/L). The variation in the N and P contents, and the N:P ratio, was similar under all three stress intervals: the N content and N:P ratio were highest with 17-day intervals, while the P content was highest with 14-day intervals (Fig. 1).

Homeostasis characteristics of L. chinensis

The homeostasis index of N (H_N) in L. chinensis ranged from 2.35 to 7.25, and first rose then fell as salinity-alkalinity increased, independently of interval length (Fig. 2). Under the 14-, 17- and 23-day intervals, H_N ranges were 3.10–7.25, 2.35–5.36 and 3.19–3.87, respectively. It was consistently highest at pH 8.4, and higher at pH 8.1 and 8.7 than at pH 7.5 and 9.3 (Table 2). Under the 14-day stress intervals, the changes in H_N were moderate, and H_N was high at all pH values. The index was lower at both pH 8.1 and 8.7 under the 17- and 23-day intervals.

Under our treatments, H_P of L. chinensis ranged from 2.60 to 5.33. Under all three saline-alkali stress intervals, it first rose and then declined (Fig. 2). Under the 14-, 17-, and 23-day saline-alkali stress intervals, the ranges of H_P were 2.84–5.33, 2.65–4.90 and 2.60–4.10, respectively, and the highest values of H_P occurred at pH 8.1, 8.7, and 8.1, respectively (Table 2). At pH 8.1, H_P was significantly higher under 14-day intervals than under longer intervals. At pH 8.4, it was highest under 17-day intervals, and at pH 8.7 it declined in the order 17-day >14-day >23-day intervals.

Plants have strong self-regulating mechanisms, and generally keep their N:P ratios within a narrow range by adjusting their N and/or P contents in response to environmental changes. For example, the H_N:P of L. chinensis varied from 3.06 to 9.23 in plants exposed to different P concentrations, while its ranges under the 14-, 17- and 23-day stress interval regimes were 3.15–9.23, 3.11–6.83 and 3.06–6.44, respectively, and its peak values occurred at pH 8.4, 8.4, and 8.1, respectively (Table 2). L. chinensis plants are thus able to effectively regulate their N:P ratio in the pH range 8.1 to 8.7, but less so at pH 7.5 and 9.3.

The specific growth rate of L. chinensis

The specific growth rate of the plants rapidly increased then decreased as the N concentration in the substrate increased, peaking at 16 mmol/L (Fig. 3). This indicates that when L. chinensis plants lack N, initial N addition will be extremely efficient, but saturating or excessive amounts will not promote (and may even hinder) further growth. Conversely, the plants’ specific growth rate first increased, then decreased, and then increased again with increases in environmental P concentration. Specifically, it was lowest at a P concentration of 2.4 mmol/L, and higher at both 1.2 and 4.8 mmol/L. Thus, high addition of P is beneficial for plant growth but may lead to wastage of resources. Therefore, the optimum P concentration under our experimental conditions was 1.2 mmol/L.

In summary, the threshold concentrations of N and P in the substrate solution at which the biomass of L. chinensis began to decrease (or stopped increasing) were 16 and 1.2 mmol/L, respectively. Moreover, plants exposed to 1.2–4.8 mmol/L P were mainly restricted by N, while those exposed to 16 and 24 mmol/L N were mainly restricted by P, and those exposed to lower concentrations of P and N were restricted by P and N, respectively (Fig. 4). According to the results of the correlation analysis, the specific growth rate of L. chinensis and the plant N:P ratio is positively correlated in both N-limited and P-limited environments, and the correlation is slightly stronger under P-restricted conditions, but neither of them passed the significance test. In the N and P co-limited environment, the specific growth rate of L. chinensis has a positive correlation with the plant N:P ratio, and the correlation coefficient is 0.46, which is significant at the level of 0.01.

N and P contents and the N:P ratio in L. chinensis

N and P are key nutrients that play major roles in myriad processes in plants and have important effects on one-another’s uptake and metabolism (Han et al., 2009). N:P ratios in plants fluctuate within relatively narrow ranges, and variations in their N and P contents are restricted by homeostatic mechanisms. Although P and N contents can influence each other, N affects P contents more strongly than vice versa, putatively due to the overall higher abundance of N in plants (Yu, 2005). We found that L. chinensis plants absorbed limited amounts of P from soils with a low N content, and changes in the soil’s N contents between 2 and eight mmol/L had minor effects on the P content of their aboveground parts. However, their P contents were substantially increased by higher N concentrations in the soil. Similarly, we found that N contents of aboveground parts of L. chinensis are influenced by the environmental P concentration: they changed greatly when the soil’s P content was low but remained very similar at high P concentrations. In summary, P content is the main constraint on plant growth when the soil’s P content is low, but N content becomes the main constraint of plant growth when the P content is high.

Our experiment examined saline-alkaline environments with different pH values. The N and P contents in aboveground parts of L. chinensis were both high at pH 8.4, indicating that the plants maintained strong control over their N:P ratios at this pH and confirming the previously reported finding that this pH promotes growth of L. chinensis (Li et al., 2018). L. chinensis can resist mild pH stress but it is inhibited by strong saline-alkali stress (Yan et al., 2006). At high pH (8.7 and 9.3), we found that its P content was relatively high when there were high soil concentrations of either N (16 or 24 mmol/L) or P (2.4 or 4.8 mmol/L). However, plants need more P to cope with pH stress. Plants’ nutrient adsorption is related to pH, together with indications that luxury uptake of N and P under moderate pH stress enables some plants (e.g., Rhizoma acori graminei and Lythrum salicaria) to resist strong pH stress (Cheng et al., 2017). The N and P contents in aboveground parts of L. chinensis we observed at various pH values indicate that N contents in plants are more sensitive to saline conditions than their P contents.

Homeostasis characteristics of N, P, and N:P in L. chinensis

Homeostasis theory postulates that plants can adjust their growth rate, and rates of both resource allocation and utilization in mechanisms that maintain their internal homeostasis (Mendez & Karlsson, 2005). Homeostasis is stronger in animals than in plants, and in higher plants than lower plants (Zeng et al., 2016), suggesting that homeostatic mechanisms have become more powerful over the course of evolutionary history. We found that the homeostasis indices for N content, P content and the N:P ratio in L. chinensis were within reported ranges for algae, fungi (Sterner & Elser, 2002), and animals (Yu, 2005; Karimi & Folt, 2006). On the basis of experiments on Inner Mongolian grassland and sand culture, Yu (2009) concluded that aboveground parts of L. chinensis have slightly higher H_N, H_P and H_N:P values (5.88–8.80, 3.37–6.67 and 4.49–9.46, respectively) than those observed in our study, possibly due to differences in climatic conditions.

We also found that all three homeostasis indexes first increased then decreased with increases in pH. At pH 8.1 and 8.4, they were all high, showing that L. chinensis can maintain stable levels of nutrients in weakly alkaline environments. However, the indices were lower under more strongly alkaline conditions (pH > 8.7), indicating that such conditions have toxic effects. This is consistent with previous observations that saline and alkaline stress has both physiological and biochemical effects on roots and leaves of L. chinensis (Liu et al., 2014).

Another finding is that H_N:P>H_N>H_P, indicating that the stronger homeostasis of N content than of P content in plants is mainly due to their significantly higher N content. Accordingly, homeostasis in zooplankton is reportedly highest for macronutrients, followed by micronutrients, and lowest for non-essential elements (Karimi & Folt, 2006). Since high N contents in plant tissues promote high P contents and plants’ N:P ratios vary less than their N and P contents, the N:P ratio may be subject to stronger homeostasis than either N or P individually (Sterner & Elser, 2002). Plants maintain N:P stability by their adjusting resource allocation and utilization of resources. The degree of stability may reflect their environmental adaptability, so the N:P ratio appears to be more important than N and P contents in this respect.

The plants’ homeostasis is related to the duration of the incremental increases in salinity-alkalinity stress. With 14-day intervals (rapid changes in alkalinity), the positive effects of a weakly alkaline environment (pH 7.5–8.4) on leaves of L. chinensis are stronger than those observed with longer intervals: the homeostasis indices of the aboveground parts are relatively high, and further increases in alkalinity (to pH 8.7 or 9.3) induce relatively little damage to leaves. H _N:P and H _P are strongly affected by changes in the duration of the incremental increases. For example, the variation in H _N:P of aboveground parts is higher with 23-day intervals than with either 14- or 17-day intervals (rising to a peak and then quickly falling with increases in pH, as shown in Fig. 2).

Specific growth rate

Specific growth rates are crucial indicators of organisms’ adaptation to (and thus ability to survive and reproduce within) an environment. Clearly, variations in growth capacity can be linked to variations in elemental demand (Moody et al., 2017). Our results confirm previous findings (and intuitive expectations) that adding N can significantly enhance growth of N-deficient plants. However, they also show that adding N can inhibit plants’ growth if the environmental content of N exceeds the plants’ requirements. The optimal N and P contents in the soil under our experimental conditions were 16 and 1.2 mmol/L, respectively. Adding excess P will not have inhibitory effects, and may even further stimulate growth slightly, but would be a waste of resources.

Plants can absorb elements in soils selectively. Moreover, The N:P ratio has a relatively complex relationship with specific growth rates: they are positively correlated at low specific growth rates, but negatively correlated once the specific growth rate exceeds a certain threshold (Agren, 2004). Additionally, specific growth rates of belowground parts of plants were positively correlated with N:P under N constraints, but negatively correlated with N:P under P constraints (Yu, 2009). In our experiment, growth of L. chinensis was restricted by both N and P. The specific growth rate of aboveground parts was positively related to N:P under N constraint or P constraint, but the positive correlation was strengthened under the co-constraint of N and P. This is consistent with the conclusions of Yu et al. with respect to the effects of N constraint, but not those of P constraint. We speculate that this was due to the plant’s generally low specific growth rate in the experiment, and also possibly to effects of salinity-alkalinity on the relationships between plant growth and other environmental variables. Although effects of plants’ nutrient storage on their specific growth rates have been discussed (Sterner & Elser, 2002; Yu, 2009), the relationship between their specific growth rates and C:N:P ratios clearly requires further study.

According to the specific growth rate hypothesis, high amounts of ribosomal RNA (rRNA), and thus P, are needed to synthesize the large quantities of proteins required to sustain high specific growth rates (Sterner & Elser, 2002). Therefore, organisms with high specific growth rates have relatively high P contents and low N:P ratios. This hypothesis is supported by both theoretical considerations and empirical observations of zooplankton, arthropods, and bacteria (Elser et al., 2003; Watts et al., 2006; Hessen et al., 2007). However, the relationships may be more complex in higher plants. Accordingly, we observed the positive correlation between the specific growth rate and N:P ratio of L. chinensis, rather than the negative correlation predicted by the hypothesis. This may have been because we monitored adult L. chinensis plants rather than juveniles, and/or because the specific growth rate of the studied plants was strongly influenced by the variation of several environmental factors (N content, P content, and pH) and thus does not reflect their intrinsic potential specific growth rates.