Identification and expression analyses of the alanine aminotransferase (AlaAT) gene family in poplar seedlings

Alanine aminotransferase (AlaAT, E.C.2.6.1.2) catalyzes the reversible conversion of pyruvate and glutamate to alanine and α-oxoglutarate. The AlaAT gene family has been well studied in some herbaceous plants, but has not been well characterized in woody plants. In this study, we identified four alanine aminotransferase homologues in Populus trichocarpa, which could be classified into two subgroups, A and B. AlaAT3 and AlaAT4 in subgroup A encode AlaAT, while AlaAT1 and AlaAT2 in subgroup B encode glutamate:glyoxylate aminotransferase (GGAT), which catalyzes the reaction of glutamate and glyoxylate to α-oxoglutarate and glycine. Four AlaAT genes were cloned from P. simonii × P. nigra. PnAlaAT1 and PnAlaAT2 were expressed predominantly in leaves and induced by exogenous nitrogen and exhibited a diurnal fluctuation in leaves, but was inhibited in roots. PnAlaAT3 and PnAlaAT4 were mainly expressed in roots, stems and leaves, and was induced by exogenous nitrogen. The expression of PnAlaAT3 gene could be regulated by glutamine or its related metabolites in roots. Our results suggest that PnAlaAT3 gene may play an important role in nitrogen metabolism and is regulated by glutamine or its related metabolites in the roots of P. simonii × P. nigra.

Nitrogen is an essential nutrient element for plant growth. The use of nitrogen directly affects plant growth and development, biomass and grain yield. Poplar has great potential applications in CO 2 mitigation and biofuel production 1 , and is perhaps more often used for pulpwood and nowdays as a biomass crop 2 . Poplar can exchange N with the environment by opening or closing the N cycle 3 , and thus plays a critical role in the ecosystem N cycle 3,4 . However, usually acting as a shelter forest, poplar is often established on marginal lands where the soil N is limited 5 . To achieve sustainable high productivity and decrease N fertilization, it is important to obtain a better understanding of the molecular regulatory mechanisms of N utilization. Nitrate (NO 3 − ) and ammonium (NH 4 + ) are the main sources of inorganic N in the soil. They can be absorbed by roots through at least two transport systems 6,7 . NO 3 − is transported into roots by nitrate transporters (NRT), and then reduced to NH 4 + by nitrate reductase (NR) and nitrite reductase (NiR). NH 4 + is transported into roots by ammonium transporters (AMT), assimilated into glutamine and glutamate through the glutamine synthetase (GS) and glutamate synthase (GOGAT) cycle, and further incorporated into other amino acids by aminotransferase 8 . NH 4 + and NO 3 − have different effects on plant growth, as the pH of the medium is reduced after NH 4 + is absorbed and increased after NO 3 − is absorbed, which affects the availability of other nutrients 9 . When NH 4 + is supplied as the sole N source, many plants showed negative effects, such as reduced leaf area, relative growth rate and dry matter yield [10][11][12] . In contrast to NH 4 + , the presence of NO 3 − stimulated the germination of dormant seeds of Arabidopsis thaliana 13 , regulated shoot-root allocation in tobacco and floral induction in A. thaliana 14,15 and inhibited root growth of maize 16 . Many Populus species showed better growth on NO 3 − than on NH 4 +3 , but some authors have reported a preference for NH 4 +17 . Because different N forms have different effects on plant 1

Results
Identification of AlaAT genes in P. trichocarpa. According to the methods of Wang et al. 39 and Chai et al. 40 , the Hidden Markov Model (HMM) profile "PF00155" was searched against the P. trichocarpa genome to identify AlaAT genes. Four sequences (XM_002315639, XM_002312643, XM_006369021, XM_002304219) located on different chromosomes were found in the P. trichocarpa genome. The total length of each of the four sequences was 1446 bp, encoding 481 amino acids ( Table 1). The Populus AlaAT genes had high sequence similarity to the previously characterized AlaAT genes from A. thaliana 18 , M. truncatula 22 and G. max 37 . We aligned the full-length amino acid sequences of the AlaATs with ClustalW (http://www.ebi.ac.uk/Tools/msa/clustalw2/) ( Figure S1). A phylogenetic tree was constructed using the neighbor-joining method and Poisson correction model with the MEGA5 software 41 . The phylogenetic tree showed that the AlaAT family was clearly separated into two subgroups, with two Populus AlaAT homologues per subgroup (Fig. 1). PtAlaAT1 and PtAlaAT2 were clustered together in subgroup B close to A. thaliana AtGGAT1 and AtGGAT2, whereas PtAlaAT3 and PtAlaAT4 were grouped together in subgroup A close to AtAlaAT1 and AtAlaAT2. PtAlaAT1 and PtAlaAT2 shared the same gene structure, as did PtAlaAT3 and PtAlaAT4 (Fig. 2).
Regulatory regions in the poplar AlaAT genes. To get insight into the functions of the AlaAT genes in poplar, the putative regulatory elements in their 5ʹ -upstream regions were investigated (Fig. 3). In all four genes, regulatory elements were found to be concentrated in the promoter region about 700-900 bp upstream of the translation initiation site. Abscisic acid (ABA) responsive elements were identified in the promoters of PtAlaAT2 and PtAlaAT3, MeJA-responsive elements were found exclusively in the PtAlaAT1 promoter, a gibberellin-responsive element (GA element) were found only in the PtAlaAT3 promoter, and salicylic acid responsive elements were present in the promoters of PtAlaAT1 and PtAlaAT3. All of the promoters contained and PnAlaAT4 each encoded proteins of 481 amino acid residues. The cDNA sequences were aligned and the AlaAT gene sequences were found to be highly homologous (99%) between P. simonii × P. nigra and P. trichocarpa. The percentage identity (94%) between PnAlaAT1 and PnAlaAT2 was highest, followed by that (93%) between PnAlaAT3 and PnAlaAT4 ( Figure S2). However, the percentage identity between PnAlaAT1/PnAlaAT2 and PnAlaAT3/PnAlaAT4 was low, only about 50%. According to Igarashi et al. 18 , the carboxy-terminal tripeptides of PnAlaAT1 (SRL) and PnAlaAT2 (SRL) are conserved peroxisome -targeting signal-like (PTS1-like) sequences.   PnAlaAT4 in leaves was higher than in roots, but significantly lower than the levels of PnAlaAT1 and PnAlaAT2 in leaves. In roots, the expression level of PnAlaAT3 was higher than those of PnAlaAT1 and PnAlaAT2 in roots. for 72 h. Compared with L1, PnAlaAT1 abundance in L2 was induced to a high level, irrespective of N form or concentration. Expression of PnAlaAT1 was the highest in L3 compared with L1 and L2, but was not induced by the N sources. However, PnAlaAT1 expression in roots was strongly inhibited by different N sources. In stems, PnAlaAT1 expression levels were low and effectively negligible compared with that in other organs. The expression patterns of PnAlaAT2 were similar to those of PnAlaAT1, but the expression levels in the former were clearly lower than that in the latter ( Figure S3). Expression of PnAlaAT3 gene was strongly induced by exogenous N sources in roots irrespective of N forms (Fig. 6). Notably, the expression levels of PnAlaAT3 increased by more than 100 times when the plants were  treated with 10 mM NH 4 + for 12 h and 72 h in roots. The expression levels of PnAlaAT3 in leaves were very weak, but increased significantly under high-N treatment. PnAlaAT3 expression in stems was also induced by exogenous N, but at a level significantly lower than that in roots. In contrast, PnAlaAT4 was expressed at a negligible level in all tested conditions ( Figure S4).

Effects of the diurnal cycle on
PnAlaATs expression in different organs of P. simonii × P. nigra. It is well documented that the transcriptional levels of several plant genes are subject to diurnal control 42,43 . To investigate whether the diurnal cycle affects the PnAlaATs, diurnal changes of expression level in leaves during a day/night cycle were determined. PnAlaAT1 and PnAlaAT2 expression fluctuated in different leaves during the diurnal cycle, and had the same periodicity (Fig. 7). PnAlaAT3 and PnAlaAT4 showed a similar  expression pattern in leaves, with low fluctuation and expression levels during the diurnal cycle. In addition, the expression levels of PnAlaAT1 were clearly higher than that of PnAlaAT2.
To evaluate the effect of light induction on PnAlaAT gene family members, the transcript levels in plants kept for 2 days in the dark or 2 days in the light were examined (Fig. 8). The mRNA levels of PnAlaAT1 increased significantly in L1 and L3 after 2 days of continuous dark, and the mRNA abundance of PnAlaAT1 was highest in L2 after this treatment. The mRNA level of PnAlaAT2 was lower than that of PnAlaAT1 in all tested sections, and didn't change significantly except in L2. PnAlaAT3 and PnAlaAT4 showed low expression levels in all conditions.  Effects of MSX on PnAlaATs expression in P. simonii × P. nigra roots. To clarify whether the expression of PnAlaAT3 is dependent on NH 4 + or glutamine or its related metabolites, MSX was applied to inhibit GS activity. GS functions in the glutamine synthetase/glutamine: α -oxoglutarate aminotransferase cycle (GS/ GOGAT cycle) to generate glutamine from NH 4 + and glutamate 44 . MSX blocks the enzyme activity of GS and prevents glutamine synthesis. NH 4 + application followed the N starvation increased PnAlaAT3 mRNA levels significantly in roots (Fig. 9). However, PnAlaAT3 expression showed no significant change when MSX was applied alone or with NH 4 + . In contrast, PnAlaAT3 mRNA levels were significantly induced by MSX with Gln. These treatments had no impact on PnAlaAT1, PnAlaAT2 and PnAlaAT4. This suggested that glutamine rather than NH 4 + itself controlled the expression of PnAlaAT3 gene in roots.

Discussion
The present work is the first report of AlaAT homologues in Populus. We identified four AlaAT genes in P. trichocarpa, namely PtAlaAT1, PtAlaAT2, PtAlaAT3 and PtAlaAT4. The four genes were classified into two subgroups ( Fig. 1) based on comparison with the sequences of AlaAT genes from A. thaliana, G. max and M. truncatula 22 . Subgroup A contained PtAlaAT3 and PtAlaAT4, which were closely related to A. thaliana alanine aminotransferase (AtAlaAT1 and AtAlaAT2), G. max alanine aminotransferase (GmAlaAT1 and GmAlaAT4) and M. truncatula alanine aminotransferase (MtmAlaAT); subgroup B contained PtAlaAT1 and PtAlaAT2, which were closely related to A. thaliana glutamate:glyoxylate aminotransferase (AtGGAT1 and AtGGAT2), G. max alanine aminotransferase (GmAlaAT2 and GmAlaAT3) and M. truncatula alanine aminotransferase (MtcAlaAT). According to Tuskan et al. 45 , some segments on chromosomes I and III and chromosomes VIII and X are presumed to have arisen from the salicoid-specific genome duplication. PtAlaAT1-4 are located in these duplicated segments. This indicates that the two members of each subgroup might derive from a duplication event.
Many metabolic processes occur in leaves, such as synthesis of organic compounds, photosynthesis and photorespiration. Several studies have shown the N concentration of leaves generally decreases with increasing plant age 46 . In the chaparral shrub Lepechinia calycina growing in its natural habitat, photosynthetic capacity, leaf N content and stomatal conductance decreased with increasing leaf age 47 . In Portulaca oleracea L., the absolute amount of both ribulose bisphosphate carboxylase/oxygenase (rubisco) and phosphoenolpyruvate carboxylase was lower in senescent leaves than in mature leaves, and rubisco activity was reduced to a lesser degree 48 . In Nicotiana tabacum, metabolic, biochemical and molecular events occur during leaf ageing, with a particular emphasis on N metabolism. The sink/source transition also occurs at a particular leaf stage 49 . Additionally, the concentration of N supplied has an effect on leaf senescence 50 . On the basis of these results, we took leaf development and senescence into account. In our test conditions, we observed that the lower, old leaves wilted first, the uppermost, younger leaves expanded gradually and the middle leaves remained active for a long time. We believed these three sections of leaves represented different developmental periods, though the mechanism of leaf development is unclear. We therefore divided the leaves into three groups in our study.
A. thaliana has four AlaAT homologues. AtAlaAT1 and AtAlaAT2 encode alanine aminotransferase (E.C.2.6.1.2), whereas AtGGAT1 and AtGGAT2 contain peroxisome-targeting signal (PTS) sequences and have glutamate:glyoxylate aminotransferase activity (GGAT, E.C.2.6.1.4) 18,36 . PTSs were also found in PnAlaAT1 and PnAlaAT2 (Supplementary Figure 1). To reveal the biological roles of the PnAlaAT genes, their expression profiles were precisely analyzed. PnAlaAT1 was expressed in all organs, with very high levels in leaves. Similarly, PnAlaAT2 was expressed mainly in leaves, with the highest level in L3, but was negligibly expressed in stems and roots. In A. thaliana, the expression of AtGGAT1 and AtGGAT2 was much higher in green leaves than in other organs, but the AtGGAT2 mRNA level was lower than that of AtGGAT1 in all organs 18 . The very high similarity of PnAlaAT1/PnAlaAT2 to AtGGAT1/AtGGAT2 indicated that they might be peroxisomal proteins and should have the same function in the photorespiratory process, catalyzing the reaction of glutamate and glyoxylate to α -oxoglutarate and glycine. However, it needs to be further confirmed. Given the expression pattern of PnAlaAT1 in response to light induction (Fig. 8), it seems interesting that PnAlaAT1 expression was higher at night, while photorespiration only happens during the daytime. It is possible that PnAlaAT1 is highly expressed at night and its products Gly would be used during the subsequent day. The further studies were needed to examine this hypothesis.
PnAlaAT1 expression in L1 was affected by different NO 3 − concentrations, but different NH 4 + concentrations did not cause a significant change (Fig. 5). NO 3 − reduction is related to photorespiration, which is the light-stimulated oxidation of photosynthesis intermediates to CO 2 51 . This process occurs primarily in higher plants that fix CO 2 via the C3 pathway of photosynthesis. Photorespiration protects C3 plants from photooxidation 52 and occurs in the chloroplast, peroxisome and mitochondrion. In the peroxisome, glutamate:glyoxylate aminotransferase catalyzes the reaction of glutamate and glyoxylate to α -oxoglutarate and glycine 53 . Two isoforms exist in A. thaliana, with GGAT1 representing the major form in leaves 18,36 . Photorespiration stimulates provision of a reductant source for nitrate reductase 51 . Most NO 3 − is reduced in leaves 54 and is supplied to L1 predominantly. Additionally, NO 3 − is considered not only a major macronutrient, but also a powerful signal molecule. NO 3 − triggered signals can be rapidly and specifically sensed by plant cells and then the expression of a large set of genes regulating plant metabolism and growth are induced or inhibited 55 . In our results, PnAlaAT1 expression in L1 was affected by different NO 3 − concentrations and reached a peak after 12 h of 1 mM NO 3 − supply. However, this kind of response didn't occur in other organs. Based on the above results, we speculated that photorespiration in L1 was affected predominantly when NO 3 − was supplied. The expression level of PnAlaAT1 and PnAlaAT2 exhibited a diurnal fluctuation in leaves (Fig. 8). This periodic fluctuation may be controlled by an endogenous circadian clock, whose phase can be entrained by light, possibly through the phytochrome system 42,56 . The presence of putative light-regulation and circadian elements in the promoter regions of PnAlaAT1 and PnAlaAT2 is consistent with our data and may partially explain the expression patterns of these genes in leaves [57][58][59] . Previous studies showed that both AlaAT and GGAT activities were present in etiolated wheat seedlings but their activity was half of that observed in light-grown seedlings, and exposure of etiolated seedlings to light caused an increase in enzyme activities and upregulated GGAT1 gene, while AlaAT1 gene didn't respond 60 . But in our study, the expression of PnAlaAT1 and PnAlaAT2 exhibited a diurnal fluctuation in leaves (Fig. 8) and PnAlaAT1 increased significantly in L1 and L3 after 2 days of continuous dark (Fig. 9), while PnAlaAT3 and PnAlaAT4 didn't exhibit these characteristics. The regulatory mechanism of PnAlaAT1 and PnAlaAT2 need to be further studied. Additionally, OsAlaAT1 plays an essential role in the regulation of starch storage in rice endosperm 61 . This is consistent with our finding that endosperm expression elements existed in the promoter regions of the PnAlaAT genes.
The expression of AlaAT genes was diverse in different species. In soybean, GmAlaAT1 and GmAlaAT4 were expressed only in the roots of non-nodulated plants, with very low expression in the roots of nodulated plants 37 . In M. truncatula, m-AlaAT was expressed at very similar levels in roots, stems and leaves of adult plants and in the embryo axes of young seedlings 22 . In our study, PnAlaAT3 and PnAlaAT4 were expressed in all organs, while PnAlaAT3 was expressed at a much higher level in roots than in the other organs (Fig. 4). NO 3 − and NH 4 + are absorbed by roots through NRT and AMT, respectively 6,7 . NO 3 − is reduced to NH 4 + by NR and NiR, and then enters the glutamine synthetase (GS) and glutamate synthase (GOGAT) cycle. NH 4 + is mainly assimilated in roots, whereas most NO 3 − reduction occurs in the leaves of poplars 54 . Additionally, N concentration affects the NH 4 + content and NR activities in poplar roots 62 . It has been confirmed that GS activity in roots is promoted by ammonium 63 . Ammonium has been identified as a signaling molecule 64 . In our study, PnAlaAT3 expression in roots was increased more than 100 times after treatment with 10 mM NH 4 + (Fig. 6). To determine whether this was due to the effect of the ammonium signal or the promotion of GS activity, we designed an experiment in which the GS activity was inhibited.
Glutamine, one of the N-assimilation products, can be synthesized from ammonium and glutamate by GS. Glutamine is the main transportable form of organic N and is a N-storing compound in plants 44 . As a major amino donor for the synthesis of amino acids and other N-containing compounds, glutamine can be taken up from the soil 65 . In addition to its role in nutrition and metabolism, glutamine can also function as a signal molecule inducing the expression of at least 35 genes involved in metabolism, transport, signal transduction, and stress responses within 30 min in rice 66 . Can Gln induce the expression of AlaAT genes? We showed in this study that Gln affected the expression level of PnAlaAT3 in roots, but not the other three PnAlaAT genes. That is, NH 4 + participated in the GS/GOGAT cycle to synthesize glutamine after being absorbed by the roots, and then glutamine or its related metabolites induced the expression of PnAlaAT3. PnAlaAT1, PnAlaAT2 and PnAlaAT4 genes were not significantly influenced by NH 4 + or Gln in roots. In conclusion, PnAlaAT3 was expressed at a higher level than other PnAlaAT genes in roots, and only the expression of PnAlaAT3 was promoted by NH 4 + or Gln or its related metabolites in roots. These results suggest that PnAlaAT3 might play an important role in nitrogen metabolism.
In our previous study, we found that long-term application of different forms of nitrogen may cause morphological changes of poplar roots. However, we did not find significantly differentially expressed genes related to N metabolism pathway, mitochondrial electron transport/ATP synthesis and mineral nutrition in our previous report of global gene expression analysis utilizing RNA-seq. On the contrary, we found that the significantly differentially expressed genes are largely associated with fermentation, glycolysis, and tricarboxylic acid cycle (TCA), secondary metabolism, hormone metabolism and transport processing 67 . In the study of Beatty et al. 33 , Scientific RepoRts | 7:45933 | DOI: 10.1038/srep45933 the alanine aminotransferase (AlaAT) gene was transferred into rice plants and ectopically expressed under the control of a tissue-specific promoter to investigate their functions in uses of nitrogen sources. Consistent with our findings, the authors found the transgenic plants displayed a strong N use efficiency but less changes in the transgenic transcriptome compared with the controls, with only 0.11 and 0.07% differentially regulated genes in roots and shoots, respectively. We speculate that N metabolism related genes might play an important role in the regulation of short-time N metabolism, and affect morphology changes of poplar roots through regulating fermentation, glycolysis and tricarboxylicacidcycle (TCA), secondary metabolism, hormone metabolism and transport processing.

Methods
Tissue culture and growth of plants. Young leaves were collected from cuttings of P. simonii × P. nigra grown at Northeast Forestry University Forest Farm, Harbin, China. Explants were surface sterilized with 70% absolute ethyl alcohol for 1 min and 0.5% NaOCl (Purui, Shanghai, China) solution for 7 min, and then rinsed three times with sterile double-distilled water. The leaves were cut into squares (1 cm 2 ). The leaf squares were cultivated in Petri dishes (diameter 9 cm) on MS medium 68 with 0.5 mg/L 6-benzyl-aminopurine (PhytoTechnology, Lenexa, USA) and 0.05 mg/L β -naphthaleneacetic acid (PhytoTechnology, Lenexa, USA), shoots were induced on MS medium with 0.1 mg/L 6-benzyl-aminopurine and 0.05 mg/L β -naphthaleneacetic acid, and roots were induced on MS medium with 0.2 mg/L indole-3-butyric acid (PhytoTechnology, Lenexa, USA). When they reached a height of 10 cm, the plantlets were transferred to a greenhouse with a photosynthetic photon flux density (PPFD) of 100 μ mol m −2 s −1 , a 16-h-light/8-h-dark photoperiod, and a temperature of 22 °C. There were 114 plants from the hybrids of P. simonii × P. nigra were studied in this research and 546 samples (leaves, stems and roots) were collected for all the analysis.  (Table S1) 69 . The plant tissues (1 st -3 rd (L1) and 4 th -6 th (L2) leaves from the top of the plant, 1 st -3 rd leaves (L3) from the bottom of the plant, stems and roots) were harvested after 0 h (control), 3 h, 12 h and 72 h treatment (Fig. 10). The plants were treated at different times, however, at the same time harvested at 11 o' clock. Three repeated samples were frozen in liquid nitrogen and stored at − 80 °C for further analysis. Light treatment. Plantlets grown in tissue culture vessels were directly transferred to soil supplied with water every 2 days. To examine the influence of the diurnal cycle, samples were harvested every 3 h over one day. To examine the effect of light, the plants were grown in darkness for 2 days and then transferred to a 16-h-light/8-h-dark photoperiod for 4 days. Three repeated samples were frozen in liquid nitrogen and stored at − 80 °C for analysis.

Identification of AlaAT gene family members in P. trichocarpa. We downloaded the Hidden Markov
Model (HMM) profile file (Aminotran_1_2.hmm) of the Pfam Aminotran_1_2 domain (PF00155) from the Pfam database 70 . The protein sequences of P. trichocarpa were downloaded from Phytozome 9.0 (http://www. phytozome.net/poplar_er.php). We used the HMM modules of PF00155 with the HMMER (v 3.0) software to search the proteome of P. trichocarpa 71 . Proteins with e-values less than 2.2E-34 were included in further analyses. We searched the Aminotran_1_2-domain in all the collected proteins using the Interproscan (http://www.ebi. ac.uk/Tools/pfa/iprscan/) and SMART software 72 . We used the Gene Structure Display Server (GSDS) program to illustrate the exon/intron organization of individual AlaAT genes 73 .

RNA extraction and reverse transcription.
Total RNA was extracted from leaf, stem, and root tissues using the pBIOZOL plant total RNA Extraction Reagent (BioFlux, Hangzhou, China) according to the manufacturer's protocol. The integrity of the RNA was verified by 1.5% agarose gel electrophoresis. After removing genomic DNA with gDNA Eraser, approximately 2 μ g RNA was used to synthesize the first-strand cDNA using the PrimerScript RT Reagent Kit (Takara Biotechnology, Dalian, China).

Real-time PCR (RT-PCR).
The Primer Premier 5.0 (Premier Biosoft, Palo Alto, CA, USA) software was used to design specific primers for real-time PCR analysis and the primer sequences of each gene were included in Table 2. The following gene-specific primers were used: for AlaAT1. Real-time PCR was performed using a 7500 Real-Time PCR System (Applied Biosystems) with SYBR Green PCR Master Mix (Applied Biosystems) according to the manufacturer's protocol. Each reaction was performed on 5 μ L of a 1:5 (v/v) dilution of the first-strand cDNA, synthesized as described above, with 0.3 μ M of each primer in a total reaction of 20 μ L. The specificity of the PCR amplification procedures was checked with a heat dissociation protocol after the final cycle. The amplification program had three steps: (1) 1 cycle (95 °C, 10 min); (2) 40 cycles, cDNA denaturing (95 °C, 15 s), hybridization and extension (60 °C, 1 min); (3) 1 cycle (95 °C, 15 s; 60 °C, 1 min; 95 °C, 15 s; 60 °C, 15 s) to generate a dissociation curve to confirm the specific amplification of each individual reaction. Each reaction was done in triplicate and the corresponding Ct values were determined. In the expression analyses, transcript levels were normalized to the PtActin2 gene (accession number: XM_002298946) as it is expressed stability independently of tissues, N sources, N concentration and developmental stage. The 2 −ΔΔCT method was used to analyze the relative changes in gene expression 74,75 . Promoter analysis. Regulatory elements in the 5ʹ -upstream regions of the poplar AlaAT genes were predicted starting from the ATG codon for initiation of translation. Sequence identity was analyzed to identify putative cis-acting elements using the PlantCARE database 76 . Sequence stretches of 1500 bp for each gene were compared.
Statistical analysis. Statistical tests were performed with SPSS 19.0 (IBM, USA), and the data were tested to confirm their normality before statistical analysis. For experimental variables, one-way analysis of variance (ANOVA) was used with N-treatment as a factor. Differences between means were considered significant when P < 0.05 according to the ANOVA F-test.

AlaAT4-sense AGTTGTCTCCCGTCTCACAGAG
AlaAT4-antisense CTTCGATGGAGGAGCAATAAAG PtActin2-sense CACAACTGCTGAACGGGAAAT PtActin2-antisense CAGGGCAACGGAAACACTCT Our work demonstrated that the poplar genome contained four genes encoding alanine aminotransferase (PnAlaAT3 and PnAlaAT4) and glutamate:glyoxylate aminotransferase (PnAlaAT1 and PnAlaAT2). PnAlaAT1 and PnAlaAT2 were closely related to AtGGAT1 and AtGGAT2, and contained PTS1-like sequences in their proteins. They were expressed predominantly in leaves and induced by NH 4 + and NO 3 − . Their expression exhibited a diurnal fluctuation in leaves. The expression level of PnAlaAT1 was higher than that of PnAlaAT2 in all conditions. PnAlaAT3 and PnAlaAT4 were expressed in roots, stems and leaves. The expression level of PnAlaAT3 was higher than that of PnAlaAT4. PnAlaAT3 expression was increased significantly by NH 4 + in roots, because of Gln or its related metabolites. We speculated that PnAlaAT1 and PnAlaAT3 might play important roles in leaves and roots, respectively. These results offered new insight into the AlaAT gene family in woody plants and the involvement of AlaAT genes in woody plant responses to exogenous N and light.