Portal de Periódicos da CAPES

QuantAS: a comprehensive pipeline to study alternative splicing by absolute quantification of splice isoforms

2023; Wiley; Volume: 240; Issue: 3 Linguagem: Inglês

10.1111/nph.19193

1469-8137

Yu‐Chen Song, Mo‐Xian Chen, Kai‐Lu Zhang, Anireddy S. N. Reddy, Fuliang Cao, Fu‐Yuan Zhu,

Plant and Fungal Interactions Research

Alternative splicing (AS) is a mechanism by which cells generate abundant protein diversity from a limited number of genes (Baralle & Giudice, 2017). AS plays a crucial role in regulating various life activities such as growth, development, and aging in plants (Zhu et al., 2017; Godoy Herz & Kornblihtt, 2019; Jabre et al., 2019; Chen et al., 2020; Reddy et al., 2020; Zhang et al., 2020), where it greatly influences plant growth, development, and response to biotic and abiotic stresses (Motion et al., 2015; Laloum et al., 2018; Chaudhary et al., 2019; Chen et al., 2021; Ganie & Reddy, 2021; Saini et al., 2021; Zhu et al., 2023; Supporting Information Fig. S1). The traditional method for the identification of AS is semi-quantitative RT-PCR, which is easy to perform (Palusa et al., 2007; Li et al., 2020; Riegler et al., 2021; Han et al., 2022). Quantitative PCR (qPCR) is also widely used in AS research, as it enables real-time monitoring of fluorescence signals and accurate quantification of isoform copy numbers through the use of specific primers (Hefti et al., 2018; Liu et al., 2018; Huang et al., 2021). With the emergence of digital PCR (dPCR), the identification methods of AS have become more diversified, which disperses each single target fragment into separate droplets as much as possible through the calculation of positive droplets (Fig. S2; Gao et al., 2021). Based on the urgent need for the accurate quantification of various isoforms, an AS detection method called QuantAS was established (Fig. 1), which allows us to accurately quantify all isoforms of genes based on absolute quantification technology and specific primer design. The method utilizes isoform-specific primers to overcome the isoform identification difficulty caused by different AS events and is designed by using the functional coding region as the isoform structure classification unit to ensure isoform independence (Fig. 2a). RT-qPCR enables real-time monitoring of changes in the fluorescence signal, quantification of differences between expression levels, and simultaneous detection of multiple in a single reaction. According to the copy number of different isoforms, isoform expression patterns can be identified by combining with absolute quantitative techniques. This method greatly increases the accuracy of identification and reduces the cost of repeated experiments. Furthermore, the absolute quantification of AS isoforms employing the combination of qPCR and dPCR could provide their respective advantages, thus rapidly obtaining all isoform information of the potential functional genes to be investigated. QuantAS consists of three stages: (1) gene structure assembly and specific primer design, including AS event analysis; (2) accurate quantitative analysis of the isoforms in the treated samples using qPCR and dPCR to obtain the copy number of each isoform; and (3) absolute quantification, which involves data analysis to explore the existence and levels of isoforms (the outline of the protocol for QuantAS is shown in detail in Figs 1, S3). After determining the target genes, we obtain and sort out the gene sequence information. Gene sequence data were obtained through databases such as NCBI (https://www.ncbi.nlm.nih.gov/)/Phytozome (https://phytozome-next.jgi.doe.gov)/TAIR (https://www.arabidopsis.org/), including gDNA, cDNA, and coding sequence (CDS) data corresponding to multiple isoforms contained in the database. Through RNA-Seq data processing, new isoforms of cDNA sequence information can be obtained through Expasy (https://web.expasy.org/translate/), which can acquire information on the splicing of multiple open reading frames (ORFs). The CDS of the isoform can be obtained by comparing it with other cDNAs. Due to the variety of database versions and the revisions of transcriptome data, the isoform number may vary. Therefore, all obtained gene sequences were renamed and ordered according to gene names and the number of isoforms and stored in FASTA files according to gDNA cDNA and CDS respectively. Gene structure was assembled by GSDS (http://gsds.gao-lab.org/), and the sequence (FASTA) option was selected in the format of gene features for sequence alignment. Upload the CDS files of all genes at the CDS (FASTA) input data, upload the gDNA sequence files at the genomic sequence (FASTA), choose SVG as the export format, and obtain the corresponding gene structures of all genes after submission. It should be noted that the gene structure at this time only includes the structure of the CDS region but not the untranslation region (UTR), which needs to be modified. All cDNA sequence files were uploaded to CDS (FASTA) input data, and gDNA sequence files were uploaded to genomic sequence (FASTA) in the same way. At this point, the gene structure does not distinguish between CDS and UTR, but all the gene structure is mapped. By comparing the two images generated on the website, the specific structure in the UTR was modified and the complete gene structure was finally obtained (Fig. S3). The NCBI blast function was used to compare gene sequences. By comparing the gDNA sequence with the CDS and cDNA sequence, the gDNA position corresponding to each exon fragment (including UTR and CDS) was obtained (Fig. S3). According to the gene structure of isoforms, the different areas between each isoform were found. Drawing the structure of pre-mRNA and marking the splicing site for AS event analysis. NCBI blast results can also quickly find the junction site and the sequence of the surrounding region, which is conducive to the subsequent design and selection of specific primers. Based on these results, we find the location of each AS event and choose a suitable primer design method according to the different AS events. To comprehensively identify each type of AS event by PCR, we introduce a QuantAS-based primer design method here by designing isoform-specific primers at exon–exon junction, alternatively spliced exons or retained introns as shown in Fig. 1. In a scenario where a gene has two isoforms AS1/AS2 and intron retention occurred in AS2, a reverse primer encompassing exon 2/3 splice junctions was used to detect AS1, and another primer encompassing 3′ end of intron and 5′ end of exon 3 was used to detect AS2. These two reverse primers correspond exclusively to AS1 and AS2, respectively. It is worth noting that in cases such as AS2, it is recommended to design primers on retained introns, and the proportion of primer sequences on both sides of the splice junction should be taken into account when designing primers on the splice junction to prevent the possibility of primer mismatch to AS1. The three primers used in this paper aim to reduce the cost and experimental complexity, but the experimental design can be changed according to the specific situation of the sequence in the experiment. More specific primer pairs can increase the accuracy of the results (primers were synthesized by Tsingke Biotechnology Co., Ltd, Beijing, China). Different AS events, the design of specific primers, and their corresponding isoforms are presented in Fig. 2(a). The areas that can be selected by primers are marked with magenta lines so that users can intuitively see where the primers can be designed. The plasmid standard was constructed through gene synthesis (Sangon Biotech (Shanghai) Co., Ltd, Shanghai, China), and the complete fragment including specific primers is part of the plasmid. The plasmid was required to be of high purity and devoid of any contaminants, preferably A260/280 = 1.8. The plasmid was dissolved in ddH2O, and the concentration was measured by Thermo ScientificTM Qubit Fluorometer or Thermo ScientificTM Nanodrop Fluorospectrometer (Thermo Fisher Scientific, Shanghai, China), and the unit is ng μl−1. For the standard curve, 10-fold serial dilutions were made starting from a plasmid concentration. The generation of identical duplicates at high CT values was significantly improved by diluting the last concentration solution. The dynamic range of the standard curve spanned at least five orders of magnitude. Droplets from the tube wall may be briefly shaken and centrifuged for collection, which is repeated several times while scaling up the dilution volume so that the standard can be adequately diluted. It is important to note that the accuracy of standard dilution is closely related to the amplification efficiency, so it is important to ensure volume differences with each pipette. Tips: for accurate calculation of copy number, users need to construct the standard curve in each experiment to ensure that the reaction conditions of the sample and the standard are consistent for splicing isoform identification. Since the expression of each isoform in the organism is not fixed, there are often some isoforms with low expression. Therefore, it is recommended to increase the dilution level when diluting the standard sample and ensure that the CT value of the sample will eventually fall into the range of the drawn standard curve utilizing a preliminary experiment. It is important to note that the standard curve may only be used for interpolation and cannot extrapolate the quantity of the unknown sample because the analysis may not be linear beyond the scope covered by the standards tested. The mortar and pestle were soaked in 0.1% DEPC solution, completely wrapped with aluminum foil, and baked at 300°C for > 2 h. EP tubes and other consumables were soaked in 0.1% DEPC solution and sterilized at high temperature and high pressure. All the items were dried before use. The samples were ground into powder in liquid nitrogen and stored at −80°C. One hundred milligram of sample powder was added to the EP tube, and 1 ml TRIzon Reagent (CW0580S; CoWin Biotech, Beijing, China) was added. The mixture was shaken mixed and allowed to stand for 5 min. After adding 200 μl chloroform, the liquid was quickly shaken and mixed, incubated for 10 min, and centrifuged at 13 500 g at 4°C for 15 min. After centrifugation, the liquid was divided into three layers and RNA was concentrated in the supernatant. The supernatant was collected (being careful not to suck into the middle layer), a new EP tube was added, and isopropyl alcohol was added, fully absorbed, and mixed. The samples were incubated for 10 min and then centrifuged at 4°C for 10 min. The supernatant was adsorbed and discarded, freshly prepared 75% ethanol (RNase-free water was used for preparation) was added to the tube, and the precipitate was gently tapped to ensure that the alcohol fully contacted the precipitate. Then, the supernatant was adsorbed and discarded after centrifugation at 4°C for 5 min. After a short centrifugation, the liquid was removed as much as possible, and the EP tube was dried on a super-clean platform for 10 min. Care was taken not to over-dry the RNA sample, otherwise, the precipitate would be difficult to dissolve. Add 50 μl RNase-free water and mix gently. The RNA concentration was determined by UV spectrophotometry, and RNA integrity was examined by agarose gel electrophoresis. Finally, cDNA was obtained by reverse transcription with Evo M-MLV RT Premix for qPCR (AG11706; Accurate Biotechnology (Hunan)Co., Ltd, ChangSha, China). Experiments were conducted in Thermo ScientificTM StepOnePlus Real-Time PCR System (Thermo Fisher Scientific) according to the manufacturer's instructions. All reactions were performed in 20 μl reaction volumes in 8-Strip PCR tubes with domed lids. The 2× SYBR Green Pro Taq HS Premix (AG11701; Accurate Biotechnology (Hunan)Co., Ltd) and 20 μM ROX passive reference dye were used. After full mixing, they were evenly divided into each well (Table S1). The cDNA template was mixed with water to reduce error and then added to the point sample well. After capping the tube or sealing the film, the bubbles are removed by centrifugation at high speed. Thermal cycling consisted of an initial denaturation at 95°C for 30 s followed by 40 cycles of denaturation at 95°C for 5 s, and annealing and extension at 60°C for 30 s. Melting curve analysis was carried out from 60°C to 95°C with 0.3°C increments. Threshold cycle (CT) values were determined by automated threshold analysis. PCR efficiencies were determined from dilutions of DNA and calculated from the slopes of the standard curves according to the equation. The standard curve mode was selected for equipment options, the standard curve was set, and the two-step method was used for the experiment. (If the amplification efficiency consistently fails to meet the expected standard, a three-step method may be considered. Thermal cycling consisted of an initial denaturation at 95°C for 30 s followed by 40 cycles of denaturation at 95°C for 5 s, annealing at 60°C for 30 s, and extension at 72°C for 30 s). After completing the qPCR program, the original data were exported for subsequent processing and analysis. Under normal circumstances, amplifying curves corresponding to standard curves should have consistent spacing and plateau signal intensity. The standard curve showed high repeatability, R2 > 0.980, and the amplification efficiency was between 90% and 110%. The melt curve required that the curve peaks of the same gene should be consistent, the signal curve should be smooth, and the fluorescence signal intensity should be kept at the same level as shown in Fig. S4(b) (In qPCR, many factors are often affected, such as primer quality, primer secondary structure, amplification inhibitors, and RNA quality. Due to the influence of different interference factors and equipment parameters, the standard curve amplification efficiency and PCR reaction may be affected to different degrees. Detailed analysis and solutions to experimental problems are shown in Table 1). Experiments were conducted in Thermo ScientificTM StepOnePlus (Thermo Fisher Scientific) according to the manufacturer's instructions. All reactions were performed in 20 μl reaction volumes in 8-Strip PCR tubes with domed lids. The 2× Pro Taq HS Probe Premix (AG11704; Accurate Biotechnology (Hunan)Co., Ltd) and 20 μM Rox passive reference dye were used (Table S2). After thorough mixing, they were evenly divided into each well. It is worth noting that in multiplex PCRs, primers and their corresponding probes need to be added simultaneously in one reaction. To ensure that different types of fluorescent signals can be detected by the instrument, the fluorescent label of probes should be determined according to the commissioning equipment parameters. Thermal cycling consisted of an initial denaturation at 95°C for 30 s followed by 40 cycles of denaturation at 95°C for 5 s, and annealing and extension at 60°C for 30 s. Threshold cycle (CT) values were determined by automated threshold analysis. PCR efficiency (E) was determined from dilutions of DNA and calculated from the slopes of the standard curves according to the equation. ddPCR was performed using DQ-24 (Sniper, Beijing, China) according to the manufacturer's recommendations. The reactions were incubated at room temperature for 20 min for template digestion. Prepared droplets were placed in PCR plates, and PCR was processed according to the following program: 95°C for 10 min, then 45 cycles of 95°C for 30 s, and 57°C for 1 min with a 2°C ramp rate. After PCR is completed, the photograph is taken for different fluorescence channels, and the number of positive droplets will be counted. According to the Poisson distribution, the copy number is corrected to obtain a more accurate copy number. In the dPCR droplet plot, droplets above the threshold line are positive, and those below are negative. The droplet of the two partitions should be centralized otherwise the specificity is poor. However, the dPCR instrument only detects the fluorescence signal at the end of the amplification reaction without any effect on the amplification efficiency. The reaction unit was only judged as positive or negative according to the presence or absence of a fluorescence signal (Fig. S4b). Calculation and processing of data in absolute quantitative qPCR is a crucial part. To show the data processing process of qPCR to users, we chose PtU1C (gene structure is shown in Fig. S5a, and primers are shown in Table S4) as an example. The main point is the CT value corresponding to each test well, which represents the expression of the sample. We diluted the standard by nine orders of magnitude (the dilution ratio is 10) and named it S0–S9 according to the concentration from low to high (S0 was the lowest solution and S9 was the stock standard solution). Note: The order of magnitude difference per unit in each calculation step needs to be focused on (the complete calculation steps are shown in Fig. S6b). Gene structure assembly plays a guiding role in the analysis of AS event types, which is essential for the accurate performance of QuantAS. We selected a gene, PtLuc7-rl (Potri.003G109000) of the woody model plant poplar as an example, which has a typical feature caused by intron retention (IR; Fig. S4a). These two isoforms are generated by typical IR events, resulting in a truncated protein (the data quality control of qPCR and dPCR are elaborated in Fig. S4b, and the corresponding primers are listed in Table S3). In most studies on gene functions, isoforms are usually not distinguished for verification of gene expression. The long isoform AS1 is assumed to perform gene functions by default. This often ignores the functionality of isoforms. We selected PtU1-70 K (Potri.007G026900) as our identification target, which shows an altered isoform ratio in response to stress (Fig. 2f). U1-70K is the central component of U1 small nuclear ribonucleoprotein (U1 snRNP) responsible for the recognition of the 5′ splice site during splicing, which is likely to function in response to osmotic stress in Arabidopsis thaliana (Fan et al., 2021). The change in AS2 was considerably higher than that in AS1 in response to lead(II) stress (P < 0.05) in roots, implying that AS2 may play a dominant role in coping with this stress. Isoforms may be involved in gene expression or even play a dominant role in the organism. In response to external stimuli, AS1 may not be intensively induced, but other isoforms change in large quantities to direct the regulation of cellular activities. In identifying splice isoforms, it is difficult to determine the copy number of the isoform by one pair of specific primers. This is commonly due to different transcripts with differences in length only. In this case, the primer design cannot completely follow the previous strategy and the corresponding copy number needs to be obtained using a different approach (Fig. 3a). It is easy to calculate the specific copy numbers of AS1, AS2, and AS3 through these equations. In the case of genes with complex structures containing multiple AS events among isoforms, the copy number of each gene can also be calculated by this MLE. According to the calculation principle of MLE, we used a poplar gene PtRBM25 (Potri.010G094700) as a candidate case that has multiple isoforms generated by AS, which is complex in structure with a variety of AS events including IR, alternative 3′ splice site (3′AE) and alternative 5′ splice site (5′AE) (Fig. S7a). Although the identification of most isoforms with complex structures can be solved using MLE, there are still some more complex cases that need careful consideration (Fig. 3b). For instance, the sum of fragment A for the three isoforms. Due to the lack of elements, the unknown isoform could not be obtained by the elimination method. By analogy, the computation of isoforms with inclusion relationships needs further work. PtPRP40a (Potri.014G013100) was one of the genes whose individual isoforms could not be calculated by MLE (Fig. S7b). One of the reasons for this is the inclusion relation among isoforms in this gene. Compared with semi-quantitative RT-PCR, QuantAS offers several advantages. First, QuantAS allows for the observation of normal amplification signals and enables the assessment of isoform expression levels (Fig. 2b). Then, it can distinguish isoforms with small size differences that may be easily concealed in agarose gel but remain distinguished in QuantAS by fluorescence signals in the amplification curve (Fig. 2c). The TM peaks of the melt curve show two different peaks, ensuring the specificity of the products and excluding the interference of nonspecific amplification (Fig. 2d). QuantAS also effectively reduces PCR amplification biases through specific primer or probe design coupled with absolute quantification. In this study, we designed different primers and TaqMan probes for QuantAS, relative quantification, and semi-quantitative RT-PCR in isoform verification to estimate PCR amplification biases. Absolute quantitative qPCR results were in good agreement with dPCR results, which are not affected by amplification bias due to their single-molecule fluorescence detection mechanism. Meanwhile, the changing trend of isoforms in different tissues and treatments is consistent. The results indicate that QuantAS employing dPCR or TaqMan probes in qPCR can effectively decrease amplification biases due to its high specificity (Fig. 2e). Additionally, QuantAS can also be used for simultaneously detecting various isoforms in one single reaction by employing different probes. We provide an example of the poplar gene PtU1C (Potri.003G058400), which has two isoforms, to demonstrate the applicability of QuantAS in this scenario (Fig. S5). Next-generation sequencing is also a technique for AS analysis; however, it becomes challenging to obtain reliable information on spliced isoforms from the RNA-Seq data alone, especially for lowly expressed isoforms. Furthermore, the data obtained from RNA-Seq also require further validation (Fig. 2e; Zhu et al., 2023). The use of a single pair of common primers may not be sufficient to amplify different isoforms accurately in RT-PCR, leading to challenges in discerning the expression of isoforms. This issue arises due to the competitive binding of primers to the template. QuantAS effectively addresses this problem by assessing the expression levels of various isoforms and the specificity of the amplification reaction through amplification and melting curve analysis. We provide some examples of AS isoforms that are easily detected by RT-PCR for comparison of QuantAS and RT-PCR to demonstrate that QuantAS can completely replace RT-PCR for identifying and validating AS events (Fig. S8). Taken together, QuantAS mainly takes the exon-exon junction or exon-intron junction as the anchor point of primer design and designs specific primers within the design range according to different AS events. QuantAS combines qPCR and dPCR to quantify the changing state of isoforms and is superior to semi-quantitative RT-PCR in terms of accuracy, compatibility, and specificity. Meanwhile, through CDS-based isoform classification, QuantAS allows for the simultaneous detection of multiple isoforms in a reaction using different labeled probes, reducing the workload of AS identification. QuantAS provides a convenient tool for advancing research on splice isoform function. Its key features include absolute quantitative technology and specific primer design for identifying AS events. The classification of isoform structures is based on CDS functional protein analysis, and specific primers are selected according to splicing site locations. Quantitative analysis of isoforms using intron or exon–exon junction primers solves the difficulty of identifying diverse AS events. Combining qPCR and dPCR complements each other, ensuring amplification specificity while improving quantitative accuracy. Meanwhile, a copy number calculation method called MLE is proposed, which can be used to calculate the copy number of isoforms with complex structures. It compensates for the quantitative error caused by amplification efficiency and inhibitors of qPCR by a single-molecule fluorescence quantitative technique. On the contrary, the melt curve of QuantAS effectively distinguishes amplified target fragment from nonspecific products and products with similar structures, greatly improving the accuracy and sensitivity during the qualitative and quantitative analysis. QuantAS provides a new approach to dissecting the AS mechanism of functional genes. Analyzing the effect of isoform expression patterns on overall expression levels based on different isoform copy numbers facilitates further functional studies. Compared with relative quantitative and semi-quantitative RT-PCR, an absolute quantitative technique combined with a TaqMan probe for isoform identification can substantially reduce the impact of amplification biases on quantitative results. Moreover, through multiplex PCR with different labels for probes, QuantAS enables simultaneous identification of multiple isoforms in one single reaction without interference, reducing operational errors and enhancing efficiency. In general, QuantAS provides an accurate quantitative method for isoform quantification, reduces systematic errors in experiments, and demonstrates high compatibility and specificity, while other methods, such as HiFENS, based on Fluorescence in situ hybridization (FISH) technique for endogenous splicing isoforms detection, which still relies on isoform-specific regions (Shilo et al., 2022). Both methods are good, but QuantAS proves to be a better choice for validating AS, especially in nonmodel plants where established no efficient protoplast transfection systems are lacking. The CDS-based isoform structure classification method proposed focuses on different functional coding regions. By applying QuantAS to study all isoforms, a comprehensive understanding of their expression patterns and regulatory modes under different conditions can be achieved. In addition to facilitating gene function studies, QuantAS has potential applications in phylogenetic analysis and early screening of the disease by detection of abnormal AS events (K. L. Zhang et al., 2021; Y. Zhang et al., 2021). For example, QuantAS can delve into evolutionary relationships and phylogenetic processes by examining changes in all isoforms, providing insights into common ancestry of species based on conserved AS event types (Fig. 3c). We conducted phylogenetic tree construction using the poplar gene PtHSP70 and its homologs, revealing that related species often share the same AS events (Fig. 3d). This indicates that AS patterns may be evolutionarily conserved, and the conserved splice sites sequences or AS events can serve as potential predictors of gene function. Comparing conserved sequences at the splice sites for the same AS events, isoforms produced by genes with the same AS events in different species may either perform similar functions or undergo similar modulation. Therefore, using QuantAS to accurately identify conserved genes or conserved sequences, can aid in predicting isoform function. Some conserved sequences may be used as markers in phylogenetic analysis using QuantAS. QuantAS is a valuable tool for addressing the complexities of isoform identification arising from various splicing events. It proves to be versatile and applicable in numerous scenarios and diverse cascades. However, there are instances where the precise quantification of specific isoforms remains challenging. For example, certain isoforms fully overlap with other isoforms, rendering it impossible to design specific primers for their identification (as illustrated in Figs S7a, 7b). QuantAS can also be widely used to estimate the accuracy of spliced sequences in the present mainstream database. Notably, MLE for isoform detection/removal depends on the integrity of the underlying annotation. The existing methods can only verify the known isoforms, and there is no effective way to verify the undetected isoforms, which may have a certain impact on final quantitative results. In conclusion, QuantAS offers a universal method for the detection and quantification of isoforms in plants, allowing isoforms to be reclassified for different functional protein-coding regions and for precise AS identification to determine AS event types. QuantAS is also able to detect multiple isoforms simultaneously in a single reaction, thus decreasing the redundant identification and calculating the copy number of the complicated isoform according to MLE. Coupled with qPCR and dPCR techniques, it will allow rapid and precise screening of isoform changes that participate in physiological responses. The straightforward experimental design and procedures make QuantAS a valuable addition to the current tool chest for the study of AS, especially in the validation of splicing events identified in large-scale omics data. This work was supported by the Natural Science Foundation of Jiangsu Province (BK20221334 and SBK2020042924), the Jiangsu Agricultural Science and Technology Innovation Fund (CX (21) 2023), the Science Technology and Innovation Committee of Shenzhen (JCYJ20210324115408023), the Major Project of Natural Science Research in Colleges of Jiangsu Province (20KJA220001), and the Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX23_1115). None declared. Y-CS wrote the manuscript and performed the experiments. M-XC and K-LZ performed the data analysis. ASNR and F-LC provided a critical review of the manuscript. F-YZ generally supervised the research group and designed the research. All data supporting the findings are contained in this manuscript and Supporting Information. The transcript sequence set reads are available from the following BioProjects at NCBI database (PRJNA921955 and PRJNA936161). Fig. S1 Avenue for alternative splicing research by PCR. Fig. S2 Principal characteristics of different PCR-based techniques for AS detection. Fig. S3 Diagram of the steps of gene structure assembly. Fig. S4 Results illustration and comparison between qPCR and dPCR by absolute quantification approach. Fig. S5 Application of multiplex PCR in isoform detection. Fig. S6 qPCR data processing and copy number calculation formula. Fig. S7 Two typical cases for MLE. Fig. S8 Gene structures of examples and result comparison between QuantAS and RT-PCR. Table S1 qPCR (SYBR Green I) reaction system. Table S2 qPCR (TaqMan Probe) reaction system. Table S3 Luminescence information of common reporter and corresponding quencher. Table S4 Primers and probes for multi-isoform detection. Table S5 Copy number calculation parameters of poplar gene Potri.003G058400 in roots and leaves under lead(II) stress. Please note: Wiley is not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.