Transcriptome and Low-Affinity Sodium Transport Analysis Reveals Salt Tolerance Variations between Two Poplar Trees

Salinity stress severely hampers plant growth and productivity. How to improve plants’ salt tolerance is an urgent issue. However, the molecular basis of plant resistance to salinity still remains unclear. In this study, we used two poplar species with different salt sensitivities to conduct RNA-sequencing and physiological and pharmacological analyses; the aim is to study the transcriptional profiles and ionic transport characteristics in the roots of the two Populus subjected to salt stress under hydroponic culture conditions. Our results show that numerous genes related to energy metabolism were highly expressed in Populus alba relative to Populus russkii, which activates vigorous metabolic processes and energy reserves for initiating a set of defense responses when suffering from salinity stress. Moreover, we found the capacity of Na+ transportation by the P. alba high-affinity K+ transporter1;2 (HKT1;2) was superior to that of P. russkii under salt stress, which enables P. alba to efficiently recycle xylem-loaded Na+ and to maintain shoot K+/Na+ homeostasis. Furthermore, the genes involved in the synthesis of ethylene and abscisic acid were up-regulated in P. alba but downregulated in P. russkii under salt stress. In P. alba, the gibberellin inactivation and auxin signaling genes with steady high transcriptions, several antioxidant enzymes activities (such as peroxidase [POD], ascorbate peroxidase [APX], and glutathione reductase [GR]), and glycine-betaine content were significantly increased under salt stress. These factors altogether confer P. alba a higher resistance to salinity, achieving a more efficient coordination between growth modulation and defense response. Our research provides significant evidence to improve the salt tolerance of crops or woody plants.


Introduction
Soil salinity is a detrimental environmental stress that constrains plant growth and crop production. High salinity induces osmotic stress, ionic toxicity, and oxidative stress in plants [1,2]. Saline soils limit the capacity of plant roots to absorb water, promoting water-deficiency and osmotic stress, which results in the inhibition of plant growth. Furthermore, when an excess of sodium (Na + ) and/or chlorine (Cl − ) ions enter through the transpiration stream into the plant leaves, it causes ionic imbalance and cellular damage, hampering plant growth and development [3,4]. These osmotic and ionic stresses may induce further secondary oxidative effects, promoting retardation of plant growth, metabolic and developmental changes, and reduction of productivity [1,4,5].
The transport of Na + into plant roots presents a biphasic behavior [6]: when external Na + concentration is at a very low level, Na + transport is mediated by a high-affinity transport system (HATS); at higher Na + concentrations, it is mediated via a low-affinity system (LATS). Extensive research of the adverse effects of high salinity on agricultural 2 of 21 production has, during the last decades, consolidated LATS as a well-acknowledged mechanism in the field of plant physiology [7][8][9]. Electrophysio-and pharmacological analyses have long demonstrated that the LATS of Na + uptake in glycophytes or halophytes (e.g., Arabidopsis and Suaeda maritime) mainly consist of non-selection cation channels (NSCC) [7,9] as well as several Na + /K + transporters, i.e., the low-affinity cation transporter 1 (LCT1), high-affinity K + transporter (HKT), and K + channels Arabidopsis K + Transporter 1 (AKT1) [8][9][10][11][12]. HKT1 mediates Na + transport in both high-and low-affinity systems and may function as either an Na + uniporter or Na + /K + symporter [4,13]. In Arabidopsis, there is only evidence of one HKT ortholog gene AtHKT1;1 performing Na + exclusion from leaves and regulating Na + distribution between roots and shoots [14,15]. In other plant species, a great number of studies have also revealed that the HKT1 ortholog protein plays a key role in salt tolerance in both dicots and monocots via Na + translocation [15][16][17]. Although the function of HKTs in herbs or grass plants has been thoroughly studied, however, little is yet known about their function in perennial woody plants.
Poplars are widely distributed across the northern temperate and cold-temperate zones of the world. Different Populus species evolutionarily adapted to these diverse ecological environments have thus generated long-term variations in responsiveness and tolerance to salinity [18][19][20]. Differences in salt tolerance between Populus species possibly result from different Na + transporting or translocation capacities in the plant organ/tissue or cell, including the following: (1) Na + and/or K + uptake capacities are clearly different between salt-tolerant and salt-sensitive Populus species under salinity stress [18,19].
(2) Excess Na + remains in roots rather than translocating into the salt-sensitive photosynthetic tissues, which is an efficient strategy to tolerate salinity stress in plants [1,2]. For example, the AtHKT1;1 in Arabidopsis and HKT1;5 in rice or wheat perform a similar function in the storge of Na + in roots, unloading Na + from the xylem sap into the surrounding parenchyma cells and thereby protecting the photosynthetic organs from Na + toxicity [14][15][16]21]. (3) Differences are present in the Na + compartmentation and Na + detoxification in cells subjected to salinity stress. The sequestration of Na + within the vacuole through the Na + /H + antiporters (NHXs) and the extrusion of Na + from the cytoplasm through the plasma membrane-localized Na + /H + antiporter SOS1 (Salt Overly Sensitive) are key mechanisms to improve plants' salt tolerance at cellular levels [2,22,23].
Moreover, the capacity of K + uptake and maintaining the K + /Na + ratio in plant cells is crucial to improve salt tolerance [18,19]. Since Na + competes with K + and induces K + efflux from cytoplasm, affecting various enzymatic processes and metabolic pathways, it can cause nutrient ion imbalance [1,24]. Our recent study shows that maintaining the K + /Na + ratio plays a vital role against salinity stress under different nitrogen levels [25]. As Na + enters into the plant cell, K + /Na + ratio maintenance is an essential route to tolerate salt-induced osmotic stress to keep water availability inside the cytoplasm via concerted regulation of metabolic processes and strengthened biosynthesis of osmoprotectants such as proline, glycine-betaine, etc. [1,2,26]. Furthermore, excess Na + triggers the generation of reactive oxygen species (ROS) inside the plant cell; the capacity to activate prompt redox responses and enhance the activity of antioxidant enzymes that scavenge the excess ROS is paramount to resist oxidative stress induced by high salinity [1,27]. Na + transportation under high salinity conditions plays an important role in plant growth and development [2,18]. However, still little is known about Na + transport and the underlying molecular processes of salt tolerance in poplar plants. The regulation of Na + transport mainly depends on the transcriptional level, which is energy-saving and highly plastic, and also depends on the capacity of Na + /K + transporters and related ion channels. In this study, we performed RNA-sequencing and combined the pharmacological analysis with a hydroponic-cultivation condition to analyze the Na + transportation and root transcriptomic profiles in two Populus species (Populus alba and Populus russkii, separately belong to the Leuce section and the Aigeiros section in the Populus genus). Our results show distinct features of Na + transportation and extensive variations in physiological and molecular responses to salt stress between the two Populus.

Na + and K + Uptake in the Two Populus and Their Distinct Physiological Responses to Salt Stress
To analyze whether there were differences in salt tolerance between P. alba and P. russkii and explore the characteristics of ion uptake/transport under salinity stress, we cultivated the cuttings of the two poplars using the hydroponic technique (see methods section) and subsequently treated them with 100 mM NaCl. We analyzed the dynamics of Na + and K + contents in the two Populus leaves before NaCl treatment (0 h) and at three treatment time points during the treatment (for 24 h, 7 days, and 12 days). Firstly, we detected that Na + accumulation in P. russkii leaves was significantly higher than that of P. alba (Figure 1b,c), albeit Na + concentration gradually increased in both species during the NaCl treatment. Secondly, K + content was higher in P. alba than in P. russkii (Figure 1d), while the K + /Na + ratio in P. alba was also significantly higher than in P. russkii (Figure 1e). Thirdly, we observed severe Na + toxicity and osmotic stress in the leaf phenotype of P. russkii, whose leaves were dehydrated and shrunk after the NaCl treatment for 12 days, whereas no obvious variation was observed in P. alba (Figure 1a). These results altogether indicate that there was a substantial interspecific divergence in salt tolerance and Na + and K + transport between P. alba and P. russkii.
tion and root transcriptomic profiles in two Populus species (Populus alba and Populus russkii, separately belong to the Leuce section and the Aigeiros section in the Populus genus). Our results show distinct features of Na + transportation and extensive variations in physiological and molecular responses to salt stress between the two Populus.

Na + and K + Uptake in the Two Populus and Their Distinct Physiological Responses to Salt Stress
To analyze whether there were differences in salt tolerance between P. alba and P. russkii and explore the characteristics of ion uptake/transport under salinity stress, we cultivated the cuttings of the two poplars using the hydroponic technique (see methods section) and subsequently treated them with 100 mM NaCl. We analyzed the dynamics of Na + and K + contents in the two Populus leaves before NaCl treatment (0 h) and at three treatment time points during the treatment (for 24 h, 7 days, and 12 days). Firstly, we detected that Na + accumulation in P. russkii leaves was significantly higher than that of P. alba (Figure 1b,c), albeit Na + concentration gradually increased in both species during the NaCl treatment. Secondly, K + content was higher in P. alba than in P. russkii (Figure 1d), while the K + /Na + ratio in P. alba was also significantly higher than in P. russkii (Figure 1e). Thirdly, we observed severe Na + toxicity and osmotic stress in the leaf phenotype of P. russkii, whose leaves were dehydrated and shrunk after the NaCl treatment for 12 days, whereas no obvious variation was observed in P. alba (Figure 1a). These results altogether indicate that there was a substantial interspecific divergence in salt tolerance and Na + and K + transport between P. alba and P. russkii.  Red arrows indicate the shrunken leaves of P. russkii. (b) Dynamic of Na + content in the two Populus leaves following NaCl treatment. (c) Na + content, (d) K + content, and (e) K + /Na + ratio in leaves of P. alba and P. russkii before treatment and after treatment for 24 h, 7 d, and 12 d. Data shown are the mean ± SD (n = 3). Different letters above the bars indicate the statistical significance at the p < 0.05 level among different time-points of NaCl treatment according to LSD test.

Salt Stress Effects on the Antioxidant System of the Two Populus
The activities of four anti-oxidative enzymes (POD, catalase [CAT], APX, and GR) were determined in the two Populus leaves under salt stress (Figure 2a-d). In contrast to the control, POD, APX, and GR activities significantly increased in P. alba under salt stress, but no significant changes were found for P. russkii; meanwhile, the CAT activity in the two Populus showed the opposite trend to the abovementioned three enzymes (Figure 2b). Moreover, malondialdehyde (MDA) concentration significantly increased in P. russkii leaves under salt stress but did not increase in P. alba (Figure 2e), while glycine-betaine content augmented significantly in both Populus leaves under salt stress and increased ca. four times in P. alba (Figure 2f).
Populus leaves following NaCl treatment. (c) Na + content, (d) K + content, and (e) K + /Na + ratio in leaves of P. alba and P. russkii before treatment and after treatment for 24 h, 7 d, and 12 d. Data shown are the mean ± SD (n = 3). Different letters above the bars indicate the statistical significance at the p < 0.05 level among different time-points of NaCl treatment according to LSD test.

Salt Stress Effects on the Antioxidant System of the Two Populus
The activities of four anti-oxidative enzymes (POD, catalase [CAT], APX, and GR) were determined in the two Populus leaves under salt stress (Figure 2a-d). In contrast to the control, POD, APX, and GR activities significantly increased in P. alba under salt stress, but no significant changes were found for P. russkii; meanwhile, the CAT activity in the two Populus showed the opposite trend to the abovementioned three enzymes (Figure 2b). Moreover, malondialdehyde (MDA) concentration significantly increased in P. russkii leaves under salt stress but did not increase in P. alba (Figure 2e), while glycine-betaine content augmented significantly in both Populus leaves under salt stress and increased ca. four times in P. alba (Figure 2f). alba and P. russkii leaves under NaCl treatment for 12 days. Bar indicates the mean ± SD (n = 3), *, p < 0.05, **, p < 0.01, ***, p < 0.001, ns, p > 0.05, two-tailed t-test.

Differentially Expressed Genes (DEGs) in P. alba and P. russkii Roots under Salt Stress
To investigate the molecular basis of the divergent salt tolerance in the two Populus, we conducted RNA-sequencing on the roots of P. alba and P. russkii just after applying the short-term (24 h) and long-term (12 d) salt stress treatments (100 mM NaCl). Two biological replicates were sequenced for each treatment group and control. We obtained an average of 23.3 million reads from twelve cDNA libraries. After removing the adapters and low-quality reads, we obtained high-quality clean reads for each library (Table S1). alba and P. russkii leaves under NaCl treatment for 12 days. Bar indicates the mean ± SD (n = 3), *, p < 0.05, **, p < 0.01, ***, p < 0.001, ns, p > 0.05, two-tailed t-test.

Differentially Expressed Genes (DEGs) in P. alba and P. russkii Roots under Salt Stress
To investigate the molecular basis of the divergent salt tolerance in the two Populus, we conducted RNA-sequencing on the roots of P. alba and P. russkii just after applying the short-term (24 h) and long-term (12 d) salt stress treatments (100 mM NaCl). Two biological replicates were sequenced for each treatment group and control. We obtained an average of 23.3 million reads from twelve cDNA libraries. After removing the adapters and low-quality reads, we obtained high-quality clean reads for each library (Table S1). The clean reads were aligned on the reference genome (v4.1, PhytozomeV13) of P. trichocarpa, the reference model poplar species from Populus genus; the average mapping rate of P. russkii was 74.85% that higher than that of P. alba (60.6%), suggesting that the genetic relationship between P. russkii and P. trichocarpa was closer than that between P. alba and P. trichocarpa. The numbers of the expressed transcripts (RPKM >1) detected in each library were comparable (average of 22,391, Table S1). The correlation of the two biological replicates for each treatment group was high (Pearson's r = 0.8~1.0, Figure S1). Principal component analysis (PCA) revealed important differences between P. alba and P. russkii transcriptomes on PC1 (31.4% of variation, Figure 3a). With regard to PC2 (16.1% variation), we observed that the transcriptome of short-and long-term salt stress groups and CK was clearly separated in P. russkii, whereas it was not observed in the case of P. alba (Figure 3a). One control sample of P. alba was far away from other samples, which was attributed to a batch effect of RNA-seq, and therefore was removed for further analysis.
DEGs were identified in P. alba and P. russkii using a cutoff of fold change ≥ 2 and FDR < 0.01. The numbers of up-regulated genes were more than double those of downregulated genes in the P. alba root under either short-(CK (Control) vs. S24h, 1070/493) or long-term salt stress (CK vs. S12d, 609/179) (Figure 3b; Supplementary Data S1). In the P. russkii root, the number of up-regulated genes was comparable with that of downregulated genes under short-term salt stress, whereas downregulated genes were two times higher than up-regulated genes under long-term salt stress (2482/1090) (Figure 3b; Supplementary Data S1).
GO enrichment analysis of the up-or downregulated genes in four comparisons (PrCK vs. PrS24h, PrCK vs. PrS12d, PaCK vs. PaS24h, and PaCK vs. PaS12d) identified several biological processes uniquely/commonly overrepresented ( Figure 3c). The GO term of the carbohydrate metabolic process (GO:0005975) was significantly enriched in nearly all comparisons (except for downregulated genes in PaCK vs. PaS12d). Several other metabolic processes were significantly enriched in the downregulated genes in P. russkii and P. alba, such as catabolic (GO:0009056), secondary metabolic (GO:0019748), and lipid metabolic (GO:0006629) processes ( Figure 3c). The GO term of response to stress (GO:0006950) was enriched in the up-regulated genes in P. russkii (PrCK vs. PrS12d) but in the downregulated genes in P. alba (PaCK vs. PaS24h and PaCK vs. PaS12d) (Figure 3c).
To further dissect the characteristics of gene transcription in P. russkii and P. alba roots under salt stress, we analyzed the overlapping of DEGs that were identified in the four comparisons. We found that numerous DEGs overlapped between the short-term and longterm salt stress comparisons in P. russkii. For instance, 737 downregulated genes overlapped in PrCK vs. PrS24h and PrCK vs. PrS12d, and 329 up-regulated genes were also overlapped ( Figure 4a). The 737 downregulated genes were significantly enriched in response to stress, while the 329 up-regulated genes were mainly enriched in metabolic processes (GO:0019784, GO:0005975, and GO:0006629) ( Figure 4b). A total of 308 up-regulated genes overlapped in PaCK vs. PaS24h and PaCK vs. PaS12d that were significantly enriched in carbohydrate metabolic, biosynthetic, and nucleobase-containing compound metabolic processes. A total of 65 downregulated genes overlapped in PaCK vs. PaS24h and PaCK vs. PaS12d that were enriched for transport ( Figure 4b).

Extensive Variation in Gene Transcription between the Two Populus under Salt Stress
We identified a large number of DEGs between P. russkii and P. alba in CK and under both short-term and long-term salt stress (Figure 5a; Supplementary Data S2). Numbers

Extensive Variation in Gene Transcription between the Two Populus under Salt Stress
We identified a large number of DEGs between P. russkii and P. alba in CK and under both short-term and long-term salt stress (Figure 5a; Supplementary Data S2). Numbers of up-regulated and downregulated genes were comparable in CK (P. russkii vs. P. alba), but under salt stress the up-regulated genes were about two times greater than downregulated genes, suggesting a substantial difference in gene expression between the two Populus under salt stress. In the different comparisons for these genes, GO enrichment indicated that up-regulated genes were mainly involved in metabolic and cellular processes, such as generation of precursor metabolites and energy (GO:0006091); lipid metabolic (GO:0006629), catabolic (GO:0009056), and carbohydrate metabolic (GO:0005975) processes; translation (GO:0006412); biosynthetic process (GO:0009058); and transport (GO:0006810) (Figure 5b). The downregulated genes were significantly enriched in metabolic processes (GO:0005975, GO:0019748), response to stress (GO:0006950), and response to biotic stimulus (GO:0009607) (Figure 5b). We found 1080 high-expressing genes and 598 low-expressing genes in P. alba versus P. russkii regardless of whether they were in CK or under salt stress ( Figure S2a,b). These genes might be the interspecific expression genes in the two Populus.
Since the metabolism-related GO term was the most enriched among the DEGs between the two Populus, thereafter we analyzed the expression patterns of the genes corresponding to the phenylpropanoid pathway and secondary metabolism. Heatmap analysis showed that most of the key enzyme genes in flavonoids biosynthesis were differentially expressed between the two Populus (Figure 5c). Under short-term salt stress, most of these genes were commonly downregulated in the two Populus. In addition, we found that many laccases (LAC) encoding genes involved in lignin biosynthesis were differentially expressed in the two Populus, i.e., AtLAC3, AtLAC4, and AtLAC5 homolog genes being up-regulated in P. russkii compared with P. alba (Figure 5c). Under short-term salt stress, the homolog genes of AtLAC4, AtLAC11, and AtLAC17 were up-regulated in P. russkii, suggesting these genes were associated with the rapid response to salt stress in Populus.

Differential Expression Patterns of the Na + /K + Transporters between the Two Populus under Salt Stress, and Low-Affinity Na + Transport Analysis
We analyzed the transcription level of Na + /K + transporters in P. alba and P. russkii root transcriptomes, the genes showing distinct expression patterns between the two Populus under salt stress (Figure 7). For instance, the Na + transporter gene HKT1;1 (Potri.018G132200) was significantly downregulated in the P. russkii root under salt stress

Differential Expression Patterns of the Na + /K + Transporters between the Two Populus under Salt Stress, and Low-Affinity Na + Transport Analysis
We analyzed the transcription level of Na + /K + transporters in P. alba and P. russkii root transcriptomes, the genes showing distinct expression patterns between the two Populus under salt stress (Figure 7). For instance, the Na + transporter gene HKT1;1 (Potri.018G132200) was significantly downregulated in the P. russkii root under salt stress but did not change in P. alba under short-term salt stress. Many K + channels (AKT5, AKT2, KT1, KT2a/b, KCO1) and calcium-binding protein (CML15, CIPK11) encoding genes were downregulated in P.
russkii under salt stress, whereas they did not change in P. alba (Figure 7). Furthermore, the transcription levels of Na + /H + exchanger genes (NHX1/2/6, SOS1a, SOS2/3) in the P. alba root had a higher expression than in P. russkii; meanwhile, under long-term salt stress some of them were downregulated in P. russkii but did not change in P. alba (Figure 7). We designed gene-specific primers for the above transporter genes (HKT1;1, AKT1, NHX1/2/3, SOS1, CBL9, and CIPK23/25) and conducted quantitative real-time PCR (qRT-PCR) analysis to detect their transcription levels and verify the transcriptome data ( Figure S4, Table S2). The qRT-PCR results were consistent with the transcriptome data. This confirmed our transcriptome data are accurate and reliable.  To investigate the Na + uptake/transport features of the two Populus under salt stress, we conducted a pharmacological analysis of the low-affinity transport system by the inhibitors of the K + channels (tetraethylammonium chloride [TEA + ], Cs + , and Ba 2+ ), Na + /K + transporter (Ba 2+ ), and non-selective cation channels (Gd 3+ and La 3+ ) [8,9,44]. The results show that Ba 2+ had strongly inhibited Na + uptaking in P. alba but less so in P. russkii, that the NSCCs blockers (Gd 3+ and La 3+ ) largely constrained Na + uptake in both Populus, and that the K + channel blockers severely inhibited Na + uptaking in P. alba more than in P. russkii (Table 1). TEA + is a specific K + inward rectifier channels blocker, and its suppression rate in the two Populus show that K + channels play a dominant role in Na + transport. Its higher inhibition in P. alba than in P. russkii suggests the K + channels of P. alba have a stronger capacity than those of P. russkii. The inhibition of TEA + from Ba 2+ (i.e., Na + transportation via HKT transporter), which shows a higher suppression value in P. alba than P. russkii, reveals that P. alba HKT has superior Na + transportation than P. russkii (Table 1). Based on the pharmacological analysis and transcriptome data, we infer that the HKT function of P. alba may be divergent from that of P. russkii. To confirm this hypothesis, we cloned the HKT1;1 and HKT1;2 from P. alba and P. russkii and obtained the full-length coding sequences. The alignment of the HKT1;1 and HKT1;2 sequences of the two Populus showed extensive SNP variations ( Figures S5 and S6). To study whether these sequence variations affected or not the function of Na + transporting, we conducted an additional Na + transport analysis by separately transforming the PrHKT1;1, PaHKT1;1, PrHKT1;2, and PaHKT1;2 to complement the yeast Na + transport loss-function mutant G19 (Figure 8a). The yeast growth curves showed that the function of PaHKT1;1 and PrHKT1;1 presented no difference, while the Na + transportation of PaHKT1;2 was significantly higher than PrHKT1;2 (Figure 8a,b), and the yeast growth was inhibited even under control conditions containing low Na + levels.

Transcriptomic Profiles Reveal the Interspecific Difference in Salt Responsiveness between the Two Populus
Our transcriptome data comprehensively depict the salt response profiles of the two Populus under either short-(24 h) or long-term (12 d) salinity stress; the reliability of the data was verified by qRT-PCR ( Figure S4). Our data reveal the distinct strategies of the two Populus responses to the short-term salt stress. The up-regulated DEG numbers substantially surpassed the downregulated in P. alba, whereas in P. russkii they were much more comparable (Figure 3b). Numerous DEGs detected between P. alba and P. russkii in the control and up-regulated genes (P. russkii versus P. alba) are mainly involved in the generation of pre-metabolites and energy, transport, and carbohydrate metabolic processes (Figure 5b). This energy reserve endows P. alba with a higher defense response to salt in the short term. Under long-term salt stress, phenotypic observation indicates that P. alba was able to resist salinity to keep growing, whereas the P. russkii leaves were severely dehydrated and numerous genes involved both in multiple metabolic (GO:0009056; 0019748; 0006629) and cellular processes were downregulated (Figure 3c). In contrast, in P. alba the up-regulated DEGs were three times as much as the downregulated, these genes being significantly enriched in response to biotic stimulus (GO:0009607) and carbohydrate metabolic (GO:0005975) and catabolic processes (GO:0009056), which suggests that P. alba effectively mobilizes a series of defenses and energy to combat salt stress. This response to salinity is highly similar with that of P. euphratica, a relative species of P. alba [18,19].

Interspecific Difference in Poplar Salt Tolerance Is Related to Na + Transportation Capacity and to Its Regulation
Previous studies show that high tolerance to salinity is related to a lower Na + concentration accumulated in shoots or leaves as well as to K + /Na + homeostasis in plants [10,18,45]. In our study, pharmacological analysis reveals that the capacity of Na + uptake in P. alba via K + inward rectifier channels and HKT transporters is higher than that in P. russkii (Table 1). The transcription of the K + inward rectifier channel genes, such as AKT2, AKT5, KT1, KT2a/b, and KT12, were steady expressed in P. alba, but their expression was reduced in P. russkii under salt stress (Figure 7). The downregulation of these genes contributes to a decreased Na + uptake and a decreased K + uptake, producing the negative effect of maintaining a high K + /Na + cytoplasmic ratio, thereby declining salt tolerance. The HKT transporter function in Na + uptake in roots plays a key role in Na + translocation and K + /Na + homeostasis maintenance in xylem parenchyma cells [14][15][16]21]. In P. trichocarpa, there is one HKT1 ortholog gene together with a HKT1-like pseudogene. However, in the desert poplar P. euhratica, the gene has expanded to four members [20], indicating it plays a functional role in salt tolerance for poplar plants. In this study, we found there were two ortholog genes of HKT1, i.e., HKT1;1 and HKT1;2, in both P. alba and P. russkii. We detected only HKT1;1 expressed in the two Populus roots and its transcription decreased under salt stress, which serves as a defense strategy in response to salinity. HKT1;2, an orthology gene of HKT1;1, was mainly expressed in Populus leaves and plays a critical role in Na + transport/translocation in photosynthetic apparatus. As the yeast analysis shows, PaHKT1;2 presents a higher Na + transportation activity than PrHKT1;2, suggesting that PaHKT1;2 might play an important function in Na + transportation/translocation or salt tolerance in P. alba. Extensive SNP variations detected with the HKT1;2 gene coding sequence between the two Populus also suggest that the HKT allele of P. alba has a superior performance than P. russkii and might have a significant role in salt tolerance; therefore, its function needs further research. Na + uptake via the NSCCs pathway was dominant in P. russkii under high salinity rather than in P. alba (Table 1). Gd 3+ and La 3+ serve as the broad-spectrum inhibitors; although they can block NSCCs, they also play important roles in inhibiting Ca 2+ channels [46]. The changes in Ca 2+ transportation can directly impact Na + and K + channels via signal transmission [10,45,47]. The uptake of Na + mediated by NSCCs depends on the external Na + concentration and can be-to a lesser extent-in competition with K + , suggesting that NSCCs may serve as a passive way to mediate Na + absorption. In this study, the NSCCs play a major role in LATS Na + uptake in P. russkii, which explains the weak resistance to high salinity.
The distribution of Na + in the cells and its translocation in different organs are critical to plant salt tolerance when it enters into root cells. Compartmentalizing Na + into vacuoles via the Na + /H + antiporter NHX1 and squeezing out the protoplast by SOS1 is a very important strategy for Na + detoxification in plants [1,23,48,49]. Our results indicate that the transcription of NHX1 and SOS1a can maintain a high level in P. alba while maintaining low levels in P. russkii under salt stress; this is another explanation for P. russkii being highly sensitive to salt stress.

Phytohormone-Regulated Responses Are Related to Plant Salt Tolerance
In this study, the transcriptions of NCED3 and ACO1 as rate-limiting enzymes in abscisic acid (ABA) and ethylene biosynthetic pathways were significantly downregulated in P. russkii but up-regulated in P. alba under long-term salt stress, respectively. Correspondingly, the ethylene response factors ERF6, ERF71, and HRA1 were also significantly decreased in P. russkii but up-regulated in P. alba (Figure 6b). The expression patterns of these genes shows that P. alba presents a higher level of ethylene and ABA syntheses than P. russkii. Previous studies demonstrate that ethylene improves plants' salt tolerance, mainly via inducing antioxidant defense and ROS detoxification and by maintaining K + /Na + homeostasis [50,51]. The ERF6, ERF71, and HRA1 factors were reported to play an important role in resistance to abiotic stresses in Arabidopsis and crops plants [37,52,53]. ABA modulates plants responses to salt stress largely through osmotic regulation and by inducing stomatal closure. Therefore, the differences in both ethylene and ABA syntheses between P. alba and P. russkii are identified here as critical factors that cause the differential salt resistance under long-term salt stress.
GA2-oxidases (GA2oxs) regulate the deactivation of bioactive GAs and are very important to plant development and stress responses [54]. GASA (Gibberellic Acid Stimulated in Arabidopsis) genes encoding cysteine-rich peptides are involved in plant development and environmental adaption [55]. It has been reported that overexpression of GA2ox and GASA14 genes in Arabidopsis might enhance plants' resistance to high salinity [56,57]. In our study, the GA2ox3, GA2ox6, and GASA14s maintained a steady high expression in P. alba but decreased in P. russkii under long-term salt stress (Figure 6c), which indicates the GAs' metabolism in P. alba differs from that in P. russkii. Furthermore, we found many genes in auxin biosynthetic or signaling pathways, such as IAAs, ARFs, SAURs, and PINs, showing a reverse expression pattern between P. alba and P. russkii (Figure 6c). Aux/IAA, ARF, and SAURs are key regulators of auxin responses that also play important roles in plant development and their responses to environmental cues [58][59][60]. A recent study shows that the heterologous expression of grapevine VvIAA18 in tobacco can significantly enhance salt tolerance [61]. In this study, these auxin response genes presented a steady differential expression between the two Populus and were hardly affected by salinity, indicating they are mostly regulated by auxin concentration rather than directly responding to salt stress. The expression patterns of these genes reveals that the differential local auxin levels between the two Populus is one of the key factors shaping its salt tolerance.
Under salt stress, the activities of POD, APX, and GR increased in P. alba but did not significantly change in P. russkii, and their encoding genes had a higher expression level in P. alba than in P. russkii ( Figure S3), which underlines the transcriptional regulation effects on enzyme activities resulting from protein abundance. The transcription of galactinol synthase1 (GolS1) and GolS4 genes increased in P. alba but were diminished in P. russkii under salt stress. GolS is a key enzyme catalyzing the biosynthesis of raffinose family oligosaccharides (RFOs) in plants [38]. Overexpressed GolS can significantly enhance the tolerances to salt, cold, and drought stresses in poplar and cucumber [62][63][64]. In addition, most of flavonoid-or anthocyanin-related biosynthetic genes maintained a high expression under long-term salt stress in P. alba but were downregulated in P. russkii. Many studies show that overexpression of these genes can improve salt resistance [65,66]. In contrast, laccases (LACs) genes had higher expression levels in P. russkii than those in P. alba in control, whereas under long-term salt stress they were downregulated in P. russkii although they did not change in P. alba. These genes are involved in lignin synthesis and their transcription is positively related to their growth rate [67,68]. Although previous studies have found that the growth rate of P. russkii was higher than P. alba under normal conditions, its salt or drought tolerance was weaker than P. alba [25,27]. The expression pattern of LACs explains this and also reveals the different opportunistic approaches to trade-offs between growth and defense in these two poplar species: P. russkii adopts an opportunistic growth strategy while P. alba adopts an opportunistic defense strategy.
This work reveals the molecular mechanism underlying the differential salt-tolerance between two poplar species under salinity stress. Compared to the salt-sensitive poplar (P. russkii), the salt-tolerant poplar (P. alba) shows a high transcriptional level of cellular energy metabolism processes, thus effectively initiating the stress defense responses under salt stress. Moreover, PaHKT1;2 has a strong Na + transportation ability that effectively recycles xylem Na + transport in P. abla under salt stress. On a physiological level, it is the higher antioxidant enzyme activities, osmoregulation, and stress-related hormone levels (ABA and ethylene) that altogether confer P. alba high salt tolerance. This study has significant implications for improving salt resistance in crop plants.

Plant Material and Experimental Treatment
One-year-old saplings of P. alba and P. russkii were collected at their natural habitat (44.29  The hydroponic culture solution was refreshed every 3 days. An air pump supplying air to plant roots was installed for the hydroponic water culture. The cuttings were cultivated inside a controlled climate chamber at 25/18 • C with a 14/10 h light/dark photoperiod with a maximum photon flux density of 400 µmol m −2 s −1 (photosynthetically active radiation) supplied by fluorescent lights; the relative humidity was within a 65~70% range. The cuttings were rooting and sprouting after a week.
Four weeks after the start of the hydroponic culture, we selected sprouting cuttings with uniform shoots, with an average height of 32 cm for P. alba and 35 cm for P. russikii, and then initiated the experiment. The treatments included both a control (CK) and a salt stress group (with 100 mmol L −1 NaCl) for each of the two Populus. The control was cultured in the Hoagland solution without NaCl, whereas the salt stress treatment group was cultured inside a Hoagland solution containing 100 mmol L −1 NaCl. Each treatment group in both Populus species consisted of three biological replications (six plants per replicate). The fourth to sixth fully expanded leaves of the treatment group and control were collected for ion content determination and experimental analyses. Intact roots of the salt stress treatment group and control were collected and frozen immediately using liquid nitrogen and subsequently stored at −80 • C for RNA-sequencing and physiological experimental analyses.

Measurement of Na + and K + Content in Poplar Leaves
Na + and K + concentration in the two Populus species were measured with an atomic absorption spectrophotometer (Perkin-Elmer, AA700, Waltham, MA, USA) using the method described previously [19]. Leaves of the salt-treated and control plants were oven-dried at 65 • C during more than 3 days to achieve complete dehydration. The dry samples were ground into fine powder (100 mg) and incubated into a 10 mL 0.5 M HNO 3 solution subjected to shaking extraction for 48h at room temperature. Afterwards, we used centrifugation at 12,000 rpm for 10 min to settle the solid material down into the tube bottom, and then we transferred the extracts and diluted them with deionized H 2 O (Milli-Q, Merck KGaA, Darmstadt, Germany) to measure the Na + or K + content.
Glycine-betaine content was measured using a plant Betaine content kit (Cominbio, Suzhou, China, Cat.# TCJ-2-G) following the manufacturer's protocol. Absorbance was measured at 525 nm and expressed as µg g −1 dry weight (DW). Cellular membrane lipid peroxidation analysis was based on MDA concentration according to our previous study [27]. Absorbance was measured at 450 nm, 532 nm, and 600 nm using a spectrophotometer. MDA concentration was calculated using the formula: MDA (µmol L −1 ) = 6.45 (A 532 − A 600 ) − 0.56A 450 .

RNA Extraction, cDNA Library Preparation, and RNA Sequencing
Total RNA was extracted from the poplar roots using an RNeasy plant mini kit (Qiagen, Hilden, Germany) following the manufacturer's protocol. Each treatment group or control sample had three biological replicates. RNA concentration and purity was measured using a Qubit 2.0 fluorometer (Invitrogen, Waltham, MA, USA). High-quality RNA was processed for RNA-seq library construction. Two micrograms of total RNA were used for mRNA isolation, while mRNA fragmentation and cDNA library construction were conducted using a TruSeq Stranded mRNA Library Prep Kit (Illumina, San Diego, CA, USA) as stipulated by the manufacturer's instructions. The index codes were added to attribute sequences to each sample. The cDNA libraries were sequenced at the Beijing Genomics institution (BGI, Shenzhen, China) on the Illumina HiSeq 2500 System by 50 bp single-read sequencing.

Analysis of RNA-Seq and Identification of DEGs
RNA-seq raw reads were filtered to remove adapter sequence and low-quality reads by Trimmomatic (version 0.36) software. The clean reads were aligned to the P. trichocarpa reference genome (v4.1, PhytozomeV13) using HISAT2 (version 2.1.0). Gene expression quantification and differentially expressed genes (DEGs) calculation were conducted by Cufflinks (version 2.2.1) with default parameters [72]. DEGs were identified between two comparisons using the following criteria: |log2 (fold change)| > 1 and false discovery rate (FDR) < 0.01. The FDR was generated from an adjusted P-value using the Benjamini-Hochberg method. GO enrichment and heatmap visualization were generated using TBtools (version 1.086) [73].

Reversal Transcription and qRT-PCR
Four micrograms of total RNA was used to synthesize first-strand cDNA with a reversal transcription reagent kit (TaKaRa). Synthesized cDNA was diluted to a final volume of 100 µL, and 1 µL was added as the template for detection in each reaction. qRT-PCR analyses were conducted using SYBR Premix ExTaq (TaKaRa) with a CFX96 real-time system (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Populus elongation factor-1 alpha (EF-1-α) was selected as the internal control to normalize gene transcription. Three biological replicates were performed for each qRT-PCR analysis. The levels of gene expression were evaluated by calculating the threshold cycle (Ct) based on the 2 −∆∆Ct method and were measured in terms of relative quantitative variation.

Pharmacological Analysis of Low-Affinity Na + Transport
After four weeks of hydroponic cultivation, sprouting cuttings of P. alba and P. russkiiwhose shoots had heights of 32 cm and 35 cm-were used to evaluate the effect of ion channel/transporter inhibitors on Na + uptake/transport via the low-affinity transport system. The applied inhibitors in this study included K + channel blockers (TEA-Cl, CsCl, and BaCl 2 ), and non-selective cation channel (NSCCs) blockers (LaCl 3 and GdCl 3 ). The concentration of inhibitors was determined based on the following previous studies: [74] with 10 mM TEA-Cl; [44] with 10 mM CsCl; [75] with 5 mM BaCl 2 ; and [44] with 50 µM LaCl 3 and 50 µM GdCl 3 . The inhibitor was added to the hydroponic culture solution containing 100 mM NaCl, taking the salt stress treatment group (i.e., 100 mM NaCl without any inhibitors) as the control. Each treatment had three replications and each replicate had three plants. The inhibitors treatment lasted 72 h and Na + and K + concentrations in the treated plants leaves were measured at 0 h (before treatment), 8 h, 24 h, 48 h, and 72 h.

Yeast Transformation and Growth Analysis
The yeast (Saccharomyces cerevisiae) strain G19 (Matα ura3 his3 trp7 ade2 ena1∆::HIS3:: ena4∆) [76] was used to analyze the Na + transporting function of HKT1;1 and HKT1;2 in the two Populus. The G19 strain was transformed with the pYES2 plasmid alone (empty vector) or by pYES2 containing the coding sequences of PaHKT1;1, PrHKT1;1, PaHKT1;2, and PrHKT1;2. Transformants were cultured on a synthetic dropout medium without uracil (SD/-Ura), and with 2% glucose as the carbon source for selection and normal growth, or with 2% galactose for inducible expression of HKT genes. Complementary analysis was performed by inoculation of 10 µL drops of cell suspension (OD600 = 1 or its dilution series) onto the surface of SD/-Ura/galactose agar medium supplied with 0, 50, 100, 150, or 200 mM NaCl. The results were recorded after 2 days of incubation at 30 • C. As for the growth curve test, yeast cells were incubated in SD/-Ura/galactose liquid medium with or without 100 mM NaCl at 30 • C with shaking. The absorbance was measured at 600 nm every 4 h.

Statistical Analysis
Both datasets were separately analyzed using SPSS software (v19.0), each bar representing the mean ± SE of at least three replicates. Different letters above the bars indicate significant differences, and values of p < 0.05 represented statistical significance using LSD's test.