Introduction

The cell wall is a semi-rigid structure surrounding plant cells, mainly composed of cellulose, hemicellulose, and pectin, and plays an important role in plant growth and development (Showalter 1993; Willats et al. 2000; Debra 2008). Many genes are involved in plant cell wall synthesis, including glycosyltransferase (GT) (Kong et al. 2011). Galacturonosyltransferase-like (GATL) is a member of the GT family. GATL is similar to α-1, 4-galacturonidasetransferase, which transfers galacturonic acid from uridine 5’-diphosphogalacturonic acid to the pectic polysaccharide homogalacturonan (Sterling et al. 2001). According to previous reports, GATLs might be involved in the biosynthesis of cell wall polysaccharides (Yin et al. 2010). Multiple sequence alignment (MSA) analysis showed that the amino acid sequences of GATLs in many species were highly similar, and the genomic sequences almost contained no introns (Cheng et al. 2018). GATL contained several conserved domains, including motifs related to nucleoside diphosphate (NDP)-sugar donor binding (DXD) (Wiggins and Munro 1998), as well as enzyme catalytic sites (HXXGXXKPW, LPP) and GLG motifs (Sterling et al. 2006).

In Arabidopsis thaliana, nine of ten AtGATL genes (except AtGATL4) were expressed in roots, stems, leaves, and flowers (Kong et al. 2011). However, AtGATL4 was only expressed in flowers. Previous research suggested that AtGATL1 was involved in xylan synthesis (Brown et al. 2007; Lee et al. 2007; Kong et al. 2009). The contents of GalA (galacturonic acid) in the cell wall of the atgatl3, atgatl6, and atgatl9 mutant plants decreased, suggesting that AtGATLs were involved in the process of cell wall biosynthesis (Kong et al. 2011). GhGATLs could regulate plant growth and fiber elongation through the synthesis of pectin in Gossypium hirsutum (Zheng et al. 2020).

In Glycine max and Oryza sativa, many abiotic stresses could induce the expression of GATL genes (Liu et al. 2016). SaGATL9 was induced by oxidative stress in Sedum alfredii (Han et al. 2016). Most EgGATLs were induced by drought, temperature, and abscisic acid (ABA) in Eucalyptus grandis (Cheng et al. 2018). In Populus deltoides, PdGATL1.1 and PdGATL1.2, the closest orthologs to the Arabidopsis GATL1 gene, had also been proven related to xylan synthesis (Kong et al. 2009).

Exogenous nitrogen (N) had an effect on the biosynthesis of plant cell walls (Euring et al. 2012; Lu et al. 2019). High nitrogen conditions could lead to increase in cell wall polysaccharide synthesis (Pitre et al. 2007, 2010). Carbon dioxide (CO2) assimilated by photosynthesis was the primary source of carbon in cell wall polysaccharides (Delmer and Haigler 2002). The synthesis of the cell wall was regulated by the carbon supply (Fujimoto et al. 2015). Although the concentration of exogenous N and CO2 affected the synthesis of polysaccharides in cell wall, as well as the role of GATL in regulating plant cell wall polysaccharide content had been studied, there were no relevant reports on the relationship between PtGATL genes and N and CO2 concentration. In this study, we identified 12 PtGATL genes in Populus trichocarpa genome. We characterized their phylogenetic relationships and analyzed the gene structure, chromosomal location, tandem repeats, and expression patterns. Finally, we investigated the PtGATL expression patterns in response to different concentrations of N and/or CO2 by RT-qPCR. The results laid a theoretical foundation for the study of the function of PtGATL genes in synthesis of cell wall polysaccharide in P. trichocarpa.

Materials and methods

Identification and analysis of P. trichocarpa GATL family members

The HMM (hidden Markov model) file Glyco_transf_8 (PF01501) of PtGATLs was obtained in the Pfam database (Finn et al. 2006). The poplar GATLs were searched with AtGATL sequences (obtained from the Arabidopsis TAIR database) as a reference using the hmmsearch command of the HMMER v3.1 software (Potter et al. 2018). The basic characteristics of PtGATL amino acids were analyzed with an online ExPASy program (Wilkins et al. 1999). The subcellular localization of these proteins was predicted with the online tool WOLF PSORT (Horton et al. 2007). Gene ontology (GO) annotation of PtGATLs was conducted using Blast2GO v5.2 software. All PtGATL protein sequences were uploaded to Blast2GO and BLAST in the NCBI database. After drawing and annotating, GO results and visualization images were downloaded. All procedures were conducted with parameters as default.

Gene structure, MSA and phylogenetic analysis

The genomic and coding sequence (CDS) of PtGATL genes were downloaded from Phytozome and the distribution patterns of introns and exons were analyzed using GSDS (Hu et al. 2015). The conserved motifs of PtGATL protein sequences were predicted using MEME software (Bailey et al. 2009). The PtGATL amino acid sequences were aligned and the conserved motifs of DXD, LPPF, GLG and HXXGXXKPW were checked via Clustal X (Thompson et al. 1997). The phylogenetic tree was constructed using the GATL sequences of P. trichocarpa, A. thaliana (At3g06260, At1g13250, At1g19300, At3g50760, At1g02720, At3g62660, At3g28340, At1g24170, and At1g70090), and O. sativa (Os03g18890, Os03g24510, Os07g45260, Os04g44850, Os02g50600, Os03g47530, and Os06g13760) via neighbor-joining (NJ) algorithm with MEGA v7.0.14 software (Kumar et al. 2016).

Chromosome distribution and gene duplication of PtGATL genes

The chromosomal location of PtGATLs was retrieved from Phytozome and PopGenIE (Sjodin et al. 2009), which were used to construct the chromosome distribution map of PtGATLs using the MG2C 2.0 tool. Multicollinearity scanning toolkits (Krzywinski et al. 2009) were used to analyze gene duplication events on a Linux system.

Plant materials and treatments

The plant material P. trichocarpa was obtained from the State Key Laboratory of Tree Genetics and Breeding (Northeast Forestry University, Harbin, China). The seedlings, which were almost 15 cm in height, were cultured for 21 days after being rooted in hydroponic culture. Then, the seedlings were moved into a hydroponic box filled with modified 1/2 nitrogen-free Hoagland nutrient solution (Liu et al. 2015). The culture was carried out in a greenhouse under the condition of a 16 h light/8 h darkness cycle and a stable temperature of 25 °C. Finally, the seedlings were treated for 28 days and the methods were as follows: 0.1 or 5 mM NH4NO3 was added to Hoagland nutrient solution (Ehlting et al. 2007; Euring et al. 2014); at the same time, the CO2 concentration was adjusted to 400 ppm or 800 ppm under each nitrogen concentration treatment (Klaiber et al. 2013; Caldera et al. 2017). During this period, the 1 mM NH4NO3 and 400 ppm CO2 served as the control (Liu et al. 2015; Zhang et al. 2021), and the nutrient solution was renewed every 3 days. The poplar roots, stems, and leaves were sampled. The collected tissue samples were immediately frozen in liquid nitrogen and stored in a − 80 °C freezer for further analysis. The biological replicates were repeated in triplicate to ensure the reliability of the results.

Analysis of PtGATLs expression patterns

The tissue-specific expression patterns of PtGATLs in mature leaves, young leaves, roots, nodes, and internodes were retrieved from PopGenIE (Sjodin et al. 2009), and the visual images were constructed. The bar graphs of PtGATL genes in P. trichocarpa were generated by RT-qPCR data. Expression values of roots (R), stems (S), and leaves (L) were normalized to the relative expression of PtUBQ7.

Total RNA was extracted from roots, stems, and leaves of P. trichocarpa seedlings using the OMEGA kit (Plant RNA Kit; OMEGA). The RNA concentration was measured by spectrophotometer (NanoDrop 2000/2000c) and the RNA quality was examined using 1% agarose gel electrophoresis. cDNA was synthesized using extracted RNA (1 µg) with Prime Script™ RT kit (including RNase-free DNase I) (Takara Bio, Dalian, China). Based on the full-length cDNA sequence of PtGATLs, PtUBQ7 and PtCDC2 in P. trichocarpa genome database, oligonucleotide primers were designed by Primer Premier 5.0 software. The primer sequences used in this study for RT-qPCR are presented in Table S1. As the expression profiles of PtUBQ7 and PtCDC2 were stable, the PtUBQ7 was selected as the reference gene (Wu et al, 2015). Based on the SYBR Green fluorescence program, the RT-qPCR experiment was performed using the UltraSYBR Mixture reagent (CWBIO, Beijing, China). The total reaction volume was 20 μL. The specific reaction systems were as follows: 95 °C for 10 min, then 95 °C for 15 s, 60 °C for 1 min for 45 cycles. Meanwhile, each reaction was repeated in triplicate (Chen et al. 2020a, b; Leng et al. 2021). The 2−ΔΔCT method was used to analyze the RT-qPCR amplification data (Livak and Schmittgen 2001). TBtools (Chen et al. 2020a, b) was used to generate heat maps of gene expression.

Results

Identification and analysis of the PtGATL gene members in P. trichocarpa

To identify the GATL genes in P. trichocarpa, we used GATLs of A. thaliana as probe sequences. 12 PtGATLs were identified in P. trichocarpa genome database. PtGATL1–PtGATL12 were named in terms of their positions on the chromosome. The physical and chemical properties of the 12 PtGATL genes were analyzed (Table 1). The results showed that the putative PtGATL protein sequences contained 342 ~ 383 aa. The isoelectric points ranged from 5.36 to 9.26. According to the GRAVY scores, 3 (PtGATL3, PtGATL4, PtGATL11) of 12 PtGATLs were hydrophobic proteins, while the others were hydrophilic. The molecular weight of GATL proteins ranged from 39 to 44 kDa. The amino acid sequences similarity of the PtGATL family members was from 48.40 to 91.16% (Table S2), while PtGATL11 had the highest similarity with PtGATL1. The predicted proteins were localized in chloroplasts, vacuoles, and the extracellular spaces (Table 1). MSA of the 12 PtGATLs was performed and the conserved motifs were confirmed. The PtGATL family contained four conserved motifs, including DXD, LPPF, GLG and HXXGXXKPW (Fig. 1). In addition, all PtGATLs contained a conserved serine residue (Ser308) upstream of the LPPF motif. However, the function of the GLG motif was not yet clear and required further experimental investigation.

Table 1 Parameters of the 12 identified PtGATL genes and deduced polypeptides present in the P. trichocarpa genome
Full size table
Fig. 1
figure 1

Multiple alignments of the 12 identified PtGATL proteins. Conserved DXD, LPPF, GLG, and HXXGXXKPW motifs are represented by boxes and the blue arrow represents the putative serine (Ser308)

Full size image

The GO annotations were divided into three categories: cellular components, molecular functions, and biological processes. Detailed information about the GO annotation of the PtGATL proteins was provided in Table S3. For cellular components, 12 proteins were located in the cytoplasm, intracellular anatomical structure, organelle, and endomembrane. Four proteins were located in the membrane and intrinsic component of the membrane. In the molecular function category, all PtGATLs family members participated in transferase activity (Fig. S1). These results indicated that the PtGATLs might play a significant role in transferase activity in poplar. However, there was no information about biological processes in GO annotations of PtGATL genes.

Gene structure, MSA and phylogeny analysis of the PtGATL genes

To further understand the evolutionary relationships, we analyzed the PtGATL family members in P. trichocarpa using phylogenetic analysis (Fig. 2A). In addition, all members (except PtGATL8) had no intron structures and the gene lengths were short (Fig. 2B), which was similar to the absence of introns in the AtGATL genes. Eleven conserved motifs were analyzed in PtGATL protein sequences using MEME (Fig. 2C; Table S4), and all PtGATL proteins had the same 11 motifs. The conserved motifs GLG and HXXGXXKPW were identified in motif 1. The conserved motifs LPPF and DXD were found in motifs 2 and 3, respectively. The nucleophilic receptor Ser308 was identified in motif 2. These results indicated that the conserved structural sequences were highly consistent with predicted conserved motifs of PtGATL proteins.

Fig. 2
figure 2

Phylogeny and structure analysis of the PtGATLs in P. trichocarpa. A The phylogenetic tree was generated based on the full-length sequences of PtGATLs using the neighbor-joining method. B Structure of the corresponding PtGATLs. CDSs and the upstream/downstream sequences are represented by the yellow and blue lines, respectively. C Motifs in PtGATLs sequences predicted by online MEME tool

Full size image

The GATLs of P. trichocarpa, rice, and Arabidopsis were analyzed, and a phylogenetic tree was constructed (Fig. 3). All these GATL proteins were divided into six subclasses: GATL-a, GATL-b, GATL-c, GATL-d, GATL-e and GATL-f. The number of members in each subclass ranged from two to seven. Among these, GATL-c only included two OsGATLs and GATL-d included four PtGATLs, while the rest of the subclasses included two PtGATLs.

Fig. 3
figure 3

Phylogenetic analysis of GATL proteins in P. trichocarpa, O. sativa, and A. thaliana. The phylogenetic tree was divided into six distinct subclasses. The red diamond represents PtGATLs, the blue dot represents AtGATLs, and the green triangle represents OsGATLs

Full size image

Chromosome distribution and synonymy analysis of PtGATL genes

To understand the evolutionary process of the PtGATL genes, their chromosome locations were mapped in the poplar genome (Figure S2) and the PtGATL duplication was analyzed using MCScanX (Fig. 4). The results showed that 12 PtGATL genes were mapped onto six chromosomes (chr2, 5, 7, 8, 10, and 14) in P. trichocarpa. However, tandem duplications were not identified in these genes. The segmental duplication analysis results showed that 11 segmental duplications of the 12 PtGATL genes were located on six of the chromosomes. For example, PtGATL1/12, PtGATL1/4, PtGATL1/3, PtGATL2/11, PtGATL2/12, PtGATL3/11, PtGATL3/4, PtGATL4/11, PtGATL5/10, PtGATL6/9 and PtGATL7/8 had segmental duplications on Chr2/14, Chr2/7, Chr1/5, Chr2/14, Chr2/14, Chr5/14, Chr5/7, Chr7/14, Chr8/10, Chr8/10 and Chr8/10, respectively. These results suggested a part of PtGATLs expansion through segmental duplication, which was the main way of duplication of PtGATL genes.

Fig. 4
figure 4

Schematic representations for segmental duplications of PtGATL genes. Gray and thick green lines indicate all syntenic blocks between each chromosome in P. trichocarpa genome and the duplicated PtGATL gene pairs, respectively. The black font at the end of the green line and the scale bar marked on the chromosomes represent the genes name and the length of the chromosome (Mb), respectively. Chromosome numbers are shown at the bottom of each chromosome

Full size image

In addition, we constructed four comparative syntenic maps of the GATL gene family of P. trichocarpa associated with four dicotyledonous plants (A. thaliana, Glycine max, Medicago truncatula, and Brassica oleracea) (Fig. 5). Most PtGATL genes showed a syntenic relationship with those four species. The numbers of genes with syntenic relationships between P. trichocarpa and the other four dicotyledonous plants were ten (A. thaliana), six (M. truncatula), nine (B. oleracea), and ten (G. max), respectively.

Fig. 5
figure 5

Synteny analysis of GATLs between P. trichocarpa, A. thaliana, G. max, B. oleracea, and M. truncatula. Gray lines in the background and red lines represent collinear blocks among P. trichocarpa and other plant genomes, as well as GATL gene pairs, respectively. Red or green lines represent chromosomes which are marked with the chromosome number at the top or bottom. The species names are on the left

Full size image

Tissue-specific expression analysis of GATL genes in the P. trichocarpa genome

To analyze the possible roles of the PtGATLs in the developmental processes of P. trichocarpa, tissue-specific expression of the PtGATL genes was analyzed (Fig. 6). We used the RT-qPCR method to further verify the previous microarray data (Fig. 6). The results from these two analyses were roughly the same. The expression traits of PtGATL2/8/12 were similar, with high expression in leaves. On the other hand, PtGATL7 had the lowest expression in roots but higher level in leaves. The expression of PtGATL1/2/6/8/9/10/11/12 was higher in leaves, while PtGATL3/4 were higher in stems. The expression level of PtGATL5 was higher in roots and stems.

Fig. 6
figure 6

Visual images of the tissue-specific expression traits of PtGATL genes. Tissue-specific expression data of roots, internodes, nodes, mature leaves, and young leaves derived from PlantGenIE. Tissue-specific expression analysis on roots, stems, and leaves was performed using RT-qPCR and the bar graphs were generated. The expression level of each gene in plant roots was normalized to 1.0, and significant differences in gene expression level are indicated by letters

Full size image

To study the responses of PtGATL genes to different concentrations of N and CO2, the expression patterns of PtGATLs were analyzed under different treatments (Fig. 7). The expression levels PtGATL1/2/9/10/11/12 were significantly up-regulated, while PtGATL3/4 were significantly down-regulated under ambient CO2 and low N treatments in roots. Under high CO2 and low N treatments, the expression of only three genes (PtGATL1/2/12) was up-regulated significantly in roots. Under ambient CO2 and high N treatments, the transcription levels of PtGATL3/4/10 were significantly down-regulated in roots. Under high CO2 and high N treatments, almost all PtGATLs, except PtGATL2/9, were significantly down-regulated in roots. The expression of PtGATL5/6 was significantly induced only under ambient CO2 and high N treatments in roots. The expression of PtGATL3/4 was significantly down-regulated, and PtGATL2 was significantly up-regulated under all treatments in roots. In stems, the expression of PtGATL7/10 was significantly decreased only under high N treatments. The expression of PtGATL1/2 was significantly increased and the expression patterns were similar under high CO2 and low N, as well as high CO2 and high N treatments. The transcription level of PtGATL3 was significantly down-regulated under the four treatments in stems. In leaves, the transcription level of PtGATL1 was significantly down-regulated under high CO2 treatments. Under ambient CO2 and low N conditions in leaves, the expression of only PtGATL6 was significantly down-regulated.

Fig. 7
figure 7

Expression patterns of PtGATL genes in different tissues under different nitrogen and CO2 treatments. Expression patterns of PtGATL genes in roots, stems, and leaves are shown. The concentration of N treatment is 0.1 mM NH4NO3 (low nitrogen concentration) and 5 mM NH4NO3 (high nitrogen concentration), and the concentration of CO2 treatment is 400 ppm (adequate carbon concentration) and 800 ppm (high carbon concentration). The 1 mM NH4NO3 and 400 ppm CO2 served as the control. The 2−ΔΔCT method was used to calculate the transcription level of PtGATLs, and the log2 (sample/control) value of each PtGATL was used to indicate its relative expression level. In heat map, the right side is a scale bar, and different colors indicate that the gene expression level in the treated sample is up-regulated or down-regulated compared to the control

Full size image

Discussion

Nitrogen could affect the biosynthesis of plant cell walls (Aerts et al. 1995; Euring et al. 2012; Lu et al. 2019). The cell wall, as the plant’s largest carbon pool, is also regulated by carbon supply (Showalter 1993; Delmer and Haigler 2002; Verbancic et al. 2018). Under the co-treatment of CO2 and N, carbon distribution in the plant and the cell wall structure was changed (Luo et al., 2005, 2010). In this study, we identified 12 PtGATL genes via bioinformatics methods and studied the changes of their transcription levels under different concentrations of exogenous N and CO2. Our results provided important information about the PtGATL genes expression by N and CO2 treatments, and provided new insights into how carbon and nitrogen affected polysaccharides synthesis in the cell wall during plant developments.

MSA showed that the conserved structure sequences of the PtGATLs included DXD, LPPF, GLG, and HXXGXXKPW (Fig. 1, Table S4). The DXD motif and HXXGXXKPW motif were believed to interact with Mn2+ and bind to NDP-sugar donors (Persson et al. 2001; Gibbons et al. 2002; Yin et al. 2010). LPPF was a motif unique to GATL, which might have some important roles in maintaining the integrity of the binding pocket and/or in catalysis. All PtGATL proteins contained a conserved serine residue (Ser308), which might be the nucleophilic receptor for GATL proteins (Yin et al. 2010). GLG motif might be involved in protein–protein interactions during dimer formation (Persson et al. 2001; Gibbons et al, 2002).

Whole-genome analysis showed that the 12 genes located on the six chromosomes had segmental duplication to varying degrees and without tandem duplication (Fig. 4, S2). These results indicated that segmental duplication was the main driving force for the expansion of the PtGATL genes. The synonymous map of the GATLs of P. trichocarpa and the four dicot species showed that at least six genes were collinear with other species (Fig. 5), suggesting that these PtGATL genes might have played an essential role in the evolution of the GATL gene family in poplar.

Previous studies had found that there might be a signal peptide in the amino acid sequence of OsGATL (Sterling et al. 2001; Liu et al. 2016). Therefore, we performed sequence analysis on the members of PtGATL proteins (Table S6), and the results showed that all 12 PtGATLs contained signal peptides at the N-terminus. Previous research had shown that AtGATL3 was related to cell wall polysaccharide synthesis in Arabidopsis (Kong et al. 2011). AtGATL5 might play a role in regulating rhamnogalacturonan I synthesis (Kong et al. 2013). AtGATL6 was involved in the initiation of primary cell wall synthesis or secondary cell wall synthesis, and also related to pectin synthesis (Kong et al. 2011). Phylogenetic analysis showed that PtGATL6/9 and AtGATL3, PtGATL2/12 and AtGATL5/6/7 belonged to the same subclass, respectively, and their expression patterns were similar (Figs. 3, 6). Therefore, PtGATLs might be related to the synthesis of cell wall polysaccharides, and the relationship still need experimental verification. AtGATL1, PdGATL1.1 and PdGATL1.2, which were homologous genes, involvement in xylan synthesis had been previously confirmed (Brown et al. 2007; Lee et al. 2007; Kong et al. 2009). The two genes, PtGATL1 and PtGATL11, were in the same subclass as AtGATL1. We, therefore, deduced that the genes might be related to the synthesis of xylan.

The carbon sequestration capacity of trees was limited by the availability of soil nitrogen under higher CO2 concentrations (Oren et al. 2001; Sigurdsson et al. 2001). Previous research reported that a few genes played an important role in C/N balance, and many physiological and molecular studies had been conducted (Zheng 2009). In this study, we found that the expression of PtGATLs respond to CO2 and N conditions. For example, under all treatment conditions, the transcription levels of PtGATL3/4 were significantly inhibited in roots and stems (Fig. 7). The transcription level of some genes did not change significantly when only C or N concentration was modified. For example, under ambient CO2 treatments, the expression of PtGATL8 was not change significantly, but it was significantly down-regulated under high C and high N concentration in roots. We speculated that PtGATL8 might play an important function in the C/N balance.

Conclusion

We identified 12 PtGATL genes via bioinformatics methods with similar gene structures and conserved motifs. Phylogenetic analysis showed that PtGATLs were divided into six classes. The chromosomal locations and duplication of PtGATL genes were predicted. We found that under different N and CO2 treatments, the expression patterns of PtGATL genes were different. This study provided a better understanding of the evolution of the PtGATL genes, and laid the foundation for further detailed analysis of the gene family.