Direct effects of nitrogen addition on seed germination of eight semi‐arid grassland species

Abstract Seed germination plays an important role in mediating plant species composition of grassland communities under nitrogen (N) enrichment. Shifts of plant community structure with N‐enhanced deposition in terrestrial ecosystems have occurred globally. Despite numerous studies about the effects of enhanced N deposition on mature plant communities, few studies have focused on seed germination. Using a laboratory experiment, we report the effects of five N concentrations, including 0, 5, 10, 20, and 40 mM N (NH4NO3) on seed germination of eight semi‐arid grassland species. Results showed that low N concentrations (5‐ and 20‐mM N) promoted mean final germination proportion of all eight species by 4.4% and 6.4%, but high concentrations (40 mM N) had no effect. The mean germination rate was decreased 2.1% and 5.1% by higher N concentration (20‐ and 40‐mM N) levels, but germination start time showed the opposite trend, delayed by 0.7, 0.9, and 1.8 d for the 10, 20, and 40 mM N treatments. Final germination proportion, mean germination rate, and germination start time were significantly different among species in response to N concentration treatments. The final germination proportion of Allium tenuissimum and Chenopodium glaucum were suppressed by increased N concentration, whereas it increased for Potentilla bifurca, Plantago asiatica, and Setaria viridis. Our findings provide novel insights into N deposition‐induced species loss based on seed germination factors in semi‐arid grassland communities.


| INTRODUC TI ON
Nitrogen deposition resulting from human activities has become a global environmental issue Deng et al., 2017;Liang et al., 2020;Liu et al., 2013). The availability of nitrogen is a critical control on plant diversity and growth in many terrestrial ecosystems, including grasslands in Northern China (Bobbink et al., 2010;Wang et al., 2020). Plant diversity is mediated at different stages of plant growth, including seed germination, seedling growth, age of senescence, and offspring performance. Most studies focus on the responses of mature or offspring plant species diversity to N deposition (Bobbink et al., 2010;Hu & Wan, 2019;Liu et al., 2019;Xia & Wan, 2013;Zhou et al., 2019). It is also necessary to study how natural seed germination directly responds to N deposition.
Seed germination is the first crucial stage of plant maturation, affecting the growth and development of individual plants, thereby affecting community dynamics (Dürr, Dickie, Yang, & Pritchard, 2015;Jia, Chen, Fan, Li, & Zhang, 2020). The proportion of seeds germinating and the germination timing affect species establishment and colonization success and play a critical role in promoting species coexistence within communities. Since seed germination determines when and where seedling growth begins, the success of seedling establishment depends on seed germination (Tobe, Zhang, & Omasa, 2005). Many plants have dormancy mechanisms that suppress seed germination until environmental conditions are favorable (Adondakis & Venable, 2004).
Under favorable conditions, seeds absorb water and exhibit radicle protrusion, but under others, they may become dormant or die (Miransari & Smith, 2014). Nitrogen deposition is likely more crucial than other abiotic variables in determining suitable conditions for a plant (Bobbink et al., 2010;Clark et al., 2007). Nitrogen is an essential nutrient and many plants in terrestrial ecosystems are adapted to conditions of low N availability; in addition to serving as a basic nutrient, N also promotes seed germination through function as a signaling molecule (Osuna et al., 2015).
Yet, high N availability may suppress seed germination and plant development by changing the levels of metal ions, abscisic acid, phytochromes, or seed water absorption Grubišić & Konjević, 1990;Sun, Wang, He, & Hao, 2018;Tian et al., 2016;Yan et al., 2016). Furthermore, the negative effects of N deposition on plant diversity may be depended on plant functional group, and result in loss of perennial forbs, or annuals and biennials (Gao, Wang, Fu, & Zhao, 2017;Zhong, Miao, Han, & Wang, 2019). However, few studies have investigated how N deposition affects seed germination of different functional groups, especially under variable N deposition levels in semi-arid grassland ecosystems.
The grasslands in Northern China are a critical component of the Eurasian steppe, providing important ecological and economic services for sustainable development (Kang, Han, Zhang, & Sun, 2007;. The amount of N deposition is increasing due to increased human activity, posing a threat to plant diversity in the semi-arid region (Liu et al., 2013;Tian et al., 2016;Wang et al., 2020).
This study aimed to investigate the effects of N deposition on seed germination of two functional groups (perennial forbs, four species; annuals and biennials, four species). The following questions were posed: (a) Does seed germination respond to N deposition? and (b) How does N deposition affect seed germination of two functional groups?

| Study site and seed selection
Our experimental seeds were collected from a temperate steppe of Inner Mongolia, Northern China (42°02′N, 116°17′E, 1, 324 m a. s. l.). The long-term mean annual precipitation is 385.5 mm, with ~ 90% distributed May-October. The mean annual temperature is 2.1°C, with a range from −17.5°C in January to 18.9°C in July. According to FAO classification, the soil type is Haplic Calcisols, with an average of 16.95% clay, 20.30% silt, and 62.75% sand. Four perennial forbs (Artemisia frigida, Allium tenuissimum, Potentilla tanacetifolia, Potentilla bifurca) and four annuals and biennials (Chenopodium aristatum, Plantago asiatica, Chenopodium glaucum, Setaria viridis) those co-occur in the grassland were selected Sagar, Li, Singh, & Wan, 2019;Song, Niu, & Wan, 2016). Plant seeds were collected in the field in September 2015 and germination occurred in November 2015.

| Germination experiments
The experiments were conducted in an automatic greenhouse at Henan University. Germination experiments were conducted in 9 cm diameter plastic Petri dishes on 2 layers of filter paper. Distilled water was added until seeds floated, but they were not inundated.
Seed germination of all species was evaluated under five N concentrations: 0, 5, 10, 20, and 40 mM N (NH 4 NO 3 ). Each N treatment level had 3 replicates of 50 seeds. NH 4 NO 3 is a common source of N in grassland experiments. The emergence of the radicle was the criterion for germination (Lai et al., 2019). Germination of seeds was monitored daily over 30 days, and germinated seeds removed promptly to allow for more likely germination of other seeds (Lai et al., 2019).
The FGP is the proportion of sown seeds that germinated (Lai et al., 2019).

GR is calculated as follows:
where G i is the number of seed germinated on day t i (t i = 0, 1, 2, 3, …) and n is the number of seeds used in an experiment (Daws et al., 2002).
Higher GR values represent more rapid seed germination.
The GS is time (day) between seed sowing and the start of germination (Nasr & Shariati, 2005).
Two-way analysis of variance (ANOVA) was used to determine the statistical significance of N treatment, species, and their interactions on FGP, GR, and GS. The Duncan's post hoc tests were used for multiple comparisons when a significant effect was detected. Oneway ANOVA was used for multi-comparisons of the effect of N concentration on FGP, GR, and GS of each species, followed by Duncan's post hoc tests. A linear model (y = a + bx) was used to determine the relationship between pairs of FGP, GR, and GS. All statistical analyses were performed using the R statistical software environment (R Core Team 2018, version 3.5.2).

| RE SULTS
Two-way ANOVA showed that both N addition and species had a significant effect on final germination (FGP), germination rate (GR), and starting time (GT). Moreover, there was an interaction effect of N and species on FGP, GR, and GT (Table 1)

| Effects of nitrogen concentration on final germination proportion
Nitrogen treatment had a significant effect on mean FGP of all species, as well as perennial forb (PF), and annuals and biennials (AB) functional groups (all p < .05, Figure 1 inset). Compared to the control, mean FGP was enhanced 4.4% and 6.4% (absolute change) under N1 and N3 treatments, whereas mean FGP of the PF functional group was suppressed by 5.1% under the N4 treatment and mean FGP of the AB functional group was enhanced by 12.7% under the N3 treatment (all p < .05; Figure 1) Figure 1).

| Effects of nitrogen concentration on germination rate
Nitrogen treatment had a significant effect on the mean GR of all species, as well as PF and AB functional groups (Table 1, Figure 2).
Compared to the control, mean GR of all species were suppressed by 2.1% and 5.1% (absolute change) under the N3 and N4 treatments, mean GR of PF functional group was suppressed by 2.0% and 3.3% under the N3 and N4 treatments, and mean GR of AB functional group was suppressed by 7% under the N4 treatment (all p < .05; Figure 2). When analyzed by species, N treatment had a significant effect on GR of all species, except S. viridis (Figure 2). Compared to the control, we found the following: GR of A. frigida was suppressed by 4.2% and 4.3% under the N3 and N4 treatments; GR of A. tenuissimum was suppressed by 3.4% and 6.1% under the N3 and N4 treatments; GR of P. tanacetifolia was suppressed by 2.1%, 1.8%, 1.6%, and 2.8% under the N1, N2, N3, and N4 treatments; GR of P. bifurca was enhanced by 3.1% and 1.7% under the N1 and N2 treatments; GR of C. aristatum was suppressed by 12% under the N4 treatment; GR of P. asiatica was enhanced by 6.5%, 4.0%, 6.4%, and 4.9% under the N1, N2, N3, and N4 treatments; and GR of C. glaucum was suppressed by 10.7%, 12.6%, 15.3%, and 23.7% under the N1, N2, N3, and N4 treatments, respectively (Figure 2).

| Effects of nitrogen concentration on the germination start time
Nitrogen treatment had a significant effect on GS of all species, as well as for PF and AB functional groups (all p < .05, Figure 3).
Compared to the control, results include the following: mean GS

| Relationships of final germination proportion, germination rate, and germination start
Across treatments, mean FGP of all species and the PF functional group were negatively correlated with mean GS, but that was not the case for the AB functional group (Figure 4a).  Nitrogen can be used by plants not only as a nutrient, but it also acts as a germination signal (Tiansawat & Dalling, 2013;Yan et al., 2016).

| D ISCUSS I ON
A previous study found that reduction of abscisic acid levels is controlled in a nitrate-dependent manner, specifically by proteins (which are transcription factors) binding to a promoter of a gene coding an abscisic acid catabolic enzyme (Wang et al., 2015;Yan et al., 2016).
In another study, nitrogen addition enhanced germination in 9 of 53 species representative of the flora in Central-Eastern Spain (Luna & Moreno, 2009); the optimal nitrogen concentration seems to promote germination by lowering the abscisic acid/gibberellins ratio (Song, Xiang, et al., 2016;Yan et al., 2016). Nitrates that naturally occur in the soil can override light requirements in some cases (Daws et al., 2002).

F I G U R E 3
Effects of nitrogen concentration on germination start (GS) of all species, perennial forb (PF), annuals and biennials (AB), and each species individually. Different letters over the bars show significant differences among treatments based on Duncan's multiple range tests (p < .05). Species abbreviations are given in Figure 1 F I G U R E 4 Relationships among mean final germination proportion (FGP), mean germination start (GS), and mean germination rate (GR). Each data point represents the mean value of each species under each treatment The positive responses of nitrogen on seed germination are related to phytochromes (Grubišić & Konjević, 1990). Nitrate may enhance the number of Pfr-receptors or may act as a Pfr cofactor (Grubišić & Konjević, 1990). However, higher nitrogen concentrations can also result in a toxic effect on seed germination for certain species in specific environmental contexts (Pérez-Fernández, Calvo-Magro, Montanero-Fernández, & Oyola-elasco, 2006). The sensitivity of different species germination to nitrogen addition is various (Davis, 2007). Previous studies found that the N application could delay the seed germination rate of Bromus inermis (Zhu, Wang, Yan, Mao, & Mao, 2018). Therefore, the seed germination responses to N treatments were species-specific, mainly positive or unimodal (Ochoa-Hueso & Manrique, 2010).
We also found that higher nitrogen reduced mean GR of all species, as well as for the PF and AB functional groups. The GR of five species were reduced at higher concentrations, but not for one PF (P. bifurca) and two AB (P. asiatica and S. viridis). Moreover, higher nitrogen significantly delayed the mean GS of all species, as well as for the PF and AB functional groups. Water uptake is a fundamental requirement for the initiation and completion of seed germination . Increased nitrogen concentration can slow or inhibit seed water absorption, inhibiting GR and GS (Wen et al., 2017).
Differences in germination time and related variables can help optimize long-term success by increasing the probability that seedlings will emerge and grow under more favorable environmental conditions (Rice & Dyer, 2001). Correlation analysis showed that mean FGP was negatively correlated with GS of PF, but not correlated with GS of AB. This finding suggests species that can germinate early may have a competitive advantage, especially for perennial plants.
However, many annuals and biennials have a bet-hedging strategy for germination and germinate at very high rates under suitable environmental conditions (Gremer & Venable, 2014).
Resources utilization varied greatly among the plant functional groups (Wang, Zhang, Zhu, Yang, & Li, 2018). Variation in strategies of seed germination among species may contribute to coexistence at a community level, because it allows for temporal partitioning of resources, as well as providing a buffer against local species extirpation (Chesson, 2000). In the process of seed germination, other environmental factors also affect FGP, such as temperature and water potential (Lai et al., 2019). Correlation analysis showed that the mean GR of seed germination was negatively correlated with GS in all species and the two functional groups (Figure 4). Seedlings that rapidly establish would have an advantage when there is a competition for resources.
However, many laboratory experiments, including the present study, have shown that low levels of nitrogen addition can have positive effects on seed germination (Tiansawat & Dalling, 2013;Yan et al., 2016 in one study N deposition increased mobilization of soil Mn, with a 10-fold greater accumulation of Mn in forbs than in grasses, resulting in a reduction of forbs abundance (Tian et al., 2016). Nitrogen addition can also increase the concentration of ferric iron and aluminum in soils through ion exchange processes, that can be toxic to seeds and seedlings (Liu, Zhang, & Lal, 2016;Roem, Klees, & Berendse, 2002). Light asymmetry reduces the abundance of light-adapted species under the condition of nutrient enrichment (DeMalach et al., 2017). Nitrogen addition can indirectly affect the composition of soil microbial and animal communities (Kim et al., 2015;Shao et al., 2017;Zhao et al., 2018), which may lead to degradation of seeds by microorganisms (Chee-Sanford, Williams, Davis, & Sims, 2006). Our results show the role of nitrogen in seed germination, providing new insights into species loss under various N addition scenarios.

| CON CLUS IONS
Our findings showed that N enrichment increased seed germination of eight species, suggesting that seed germination is sensitive to atmospheric N deposition in this semi-arid grassland. Compared to responses of seed germination to N deposition in pot and in situ experiments, our results may provide a new perspective for the study of the reduction of diversity by N deposition. The indirect effect of nitrogen deposition on seed germination was greater or even opposite than the direct effect of nitrogen in pot and in situ studies.
Further studies on the direct and indirect effects of N deposition on seed germination are necessary to provide more comprehensive insight into ecological mechanisms structuring these communities.

ACK N OWLED G M ENTS
This study was financially supported by the National Natural

CO N FLI C T O F I NTE R E S T
None declared.

DATA AVA I L A B I L I T Y S TAT E M E N T
A copy of the data will be archived using the DRYAD international repository (https://doi.org/10.5061/dryad.w3r22 80nd).