The identification of suitable internal reference genes in quinoa seeds subjected to abscisic acid and gibberellin treatment

Abstract:

Quinoa has been recognized as the sole “comprehensive nutritional crop”; however, it is susceptible to pre-harvest sprouting (PHS). While quantitative reverse transcription polymerase chain reaction (RT-qPCR) has been extensively employed for gene expression level detection, the selection of suitable reference genes is imperative to ensure precise gene expression quantification across diverse conditions. This study aims to identify stable reference genes in quinoa seeds under ABA and GA, in order to provide a basis for subsequent research on PHS. Seeds were subjected to different concentrations of ABA and GA (10 μM, 50 μM, 100 μM, and 200 μM). The most suitable treatment concentration was determined based on seed viability. Here, MON1, GAPDH, EIF3, EF1α, ACT, TUB1, and TUB6 were selected as candidate genes. The suitability of these reference genes under different conditions was assessed using various methods including Ct values, geNorm, NormFinder, BestKeeper, Delta Ct, and RefFinder. Based on the results obtained from the hormone experiments, it was observed that the application of 100 μM ABA and 200 μM GA yielded the most advantageous outcomes. Additionally, the most appropriate reference genes for different treatments are ACT and TUB1 (H₂O treatment), EIF3 and MON1 (ABA, GA treatment and also for the combined data set of the three groups). However, GAPDH exhibited the least stability across all treatments. In summary, ACT is recommended as the reference gene for natural quinoa germination, while EIF3 and MON1 should be used for ABA and GA treatments.

Index terms:
ABA, Chenopodium quinoa Willd.; GA, reference gene, RT-qPCR

Resumo:

A quinoa foi reconhecida como a única “cultura nutricional abrangente”; no entanto, é suscetível à germinação na pré-colheita (BHS). Embora a reação em cadeia da polimerase com transcrição reversa quantitativa (RT-qPCR) tenha sido amplamente empregada para detecção do nível de expressão gênica, a seleção de genes de referência adequados é essencial para garantir a quantificação precisa da expressão gênica em diversas condições. Este estudo tem como objetivo identificar genes de referência estáveis em sementes de quinoa tratadas com ABA e GA, a fim de fornecer uma base para pesquisas subsequentes em BPC. As sementes foram submetidas a diferentes concentrações de ABA e GA (10 μM, 50 μM, 100 μM e 200 μM). A concentração de tratamento mais adequada foi determinada com base na viabilidade das sementes. MON1, GAPDH, EIF3, EF1α, ACT, TUB1 e TUB6 foram selecionados como genes candidatos. A adequação destes genes de referência sob diferentes condições foi avaliada utilizando vários métodos, incluindo valores Ct, geNorm, NormFinder, BestKeeper, Delta Ct e RefFinder. Com base nos resultados obtidos nos experimentos com hormônios, observou-se que a aplicação de 100 μM de ABA e 200 μM de GA produziu os resultados mais vantajosos. Além disso, os genes de referência mais apropriados para diferentes tratamentos são ACT e TUB1 (tratamento com H₂O), EIF3 e MON1 (tratamento com ABA, GA e para o conjunto de dados combinados dos três grupos). No entanto, o GAPDH exibiu a menor estabilidade em todos os tratamentos. Em resumo, o ACT é recomendado como gene de referência para a germinação natural da quinoa, enquanto o EIF3 e o MON1 devem ser utilizados para os tratamentos com ABA e GA.

Termos para indexação:
ABA, Chenopodium quinoa Willd.; GA, gene de referência, RT-qPCR

INTRODUCTION

Chenopodium quinoa Willd., a member of the Amaranthaceae family indigenous to South America, possesses a rich composition of proteins, vitamins, minerals, dietary fiber, plant sterols, phenolic compounds, and essential amino acids that align with human nutritional requirements (Dakhili et al., 2019). Quinoa is susceptible to pre-harvest sprouting (PHS), resulting in substantial losses. The global economic losses attributed to PHS are estimated to be reach one billion dollars annually (Tai et al., 2021). Breaking dormancy at an inappropriate time may affect plant morphogenesis or lead to the occurrence of PHS (Kashiwakura et al., 2016). Dormancy and germination processes are influenced by a multitude of factors, including temperature, air humidity, soil moisture content, exogenous chemicals, genetic factors, reactive oxygen species levels, seed maturity, and hormone levels (Barrero et al. 2020; Wang et al. 2020; Sohn et al. 2021; Pelissari et al. 2022; Rabieyan et al. 2022; Pinto et al. 2023). Among these factors, the hormones Abscisic acid (ABA) and Gibberellin (GA) exert a particularly significant impact (Barrero et al., 2020; Wang, et al., 2020; Sohn et al., 2021; Pelissari et al., 2022; Pinto et al., 2023; Rabieyan et al., 2022). The synthesis and metabolism of ABA and GA involve various multiple enzyme-catalyzed processes. Enzymes such as zeaxanthin epoxidase, 9-cis-epoxycarotenoid dioxygenase, and ABA-aldehyde oxidase participate in ABA synthesis, while ABA 8’-hydroxylase (ABA8ox) is responsible for the degradation of ABA into Phaseic acid (Seo and Koshiba 2002; Minami et al. 2018; Perreau et al. 2020; Wang et al. 2021; Jia et al. 2022). GA is produced through a sequence of enzymatic reactions involving precursor substances. These reactions are catalyzed by GA 20-oxidases and GA 3-oxidases, resulting in the formation of biologically active GA forms. Conversely, GA can be deactivated by the activity of GA2 oxidases (GA2ox) (Lor and Olszewski, 2015).

Examining the expression levels of genes associated with PHS is an imperative approach to facilitate the resolution of PHS. Quantitative Reverse Transcription Polymerase Chain Reaction (RT-qPCR) has been extensively utilized in the domains of medical and scientific research, for the purpose of detecting gene expression levels and investigating gene transcription, regulation, and validation (Ma et al., 2020; Wu et al., 2021).In order to ensure the utmost reliability of experimental outcomes, it is imperative to implement stringent control measures for crucial experimental variables, including sample quality control, experimental design, statistical analysis, and the selection and validation of reference genes, owing to the heightened sensitivity and specificity of the methodology (Škiljaica et al., 2022). In the context of RT-qPCR, the accurate acquisition of results heavily relies on the meticulous selection of suitable reference genes for data normalization (Deng et al., 2016).

In theory, reference genes are characterized as genes whose expression remains unaltered by variations in tissue types, experimental conditions, or environmental factors (Zhu et al., 2021). Nevertheless, an increasing body of scholarly investigation suggests that reference genes do not consistently demonstrate stable expression across all circumstances, and the arbitrary selection of reference genes may yield unreliable experimental data (Liu et al., 2022). In studies involving multiple species such as Brassica rapa L. (Ma et al., 2020); Momordica charantia (Wang et al., 2019); Stellera chamaejasme (Liu et al., 2018); celery (Feng et al., 2019); Fragaria chiloensis (Gaete-Eastman et al., 2022); Bromus sterilis (Sen et al., 2021), and barley (Cai et al., 2018; Walling et al., 2018), the results of identifying reference genes indicate that for the same species under different conditions or different species under the same conditions, the identification of reference genes may different (Škiljaica et al., 2022). Hence, the selection of the most suitable housekeeping gene as the internal reference gene for RT-qPCR analysis in various experimental treatments holds significant importance in obtaining accurate gene expression results.

In this study, the selection of seven candidate genes was informed by literature reports. Notably, Alpha tubulin -1 (TUB1) and Elongation factor 1-alpha (EF1α) demonstrated acceptable stability when subjected to NaCl and hormone treatment in quinoa, as observed by Zhu (Zhu et al., 2021). Additionally, the investigation of saponins in quinoa leaves involved the utilization of Vacuolar fusion protein MON1 (MON1) as an internal reference gene for qPCR analysis, as reported by Fiallos-Jurado et al. (2016). In the study conducted by (Wang et al., 2022b) it was found that Actin-1 (ACT) emerged as the most stable reference gene when subjecting Neoscytalidium dimidiatum under different temperatures. Similarly, Wang et al. (2022a) observed that Glyceraldehyde-3-phosphate dehydrogenase A (GAPDH) demonstrated the highest stability in the in vitro proliferation of satellite cells derived from bovine skeletal muscle. Furthermore, the selection of internal reference genes, such as Beta tubulin -6 (TUB6) and Eukaryotic translation initiation factor 3 (EIF3), is influenced by the variation in species and conditions (Shi et al., 2012; Taki and Zhang, 2013; Bevitori et al., 2014). To assess the stability of these internal reference genes, GeNorm (Vandesompele et al., 2002), NormFinder (Wang et al., 2022a), BestKeeper (Pfaffl et al., 2004; Wang et al., 2022a), RefFinder (Taki and Zhang, 2013), Cycle threshold (Ct) value, and Delta Ct were employed (Ruduś and Kępczyński, 2018).

In order to enhance the precision and replicability of the experiment, the present study adhered to the guidelines outlined in the Minimum Information for Publication of Quantitative Real-Time PCR Experiments (MIQE) (Bustin et al., 2009). Presently, the extent of research concerning reference genes in quinoa is considerably limited, primarily centered around the assessment of candidate genes in various tissues or plants subjected to hormone treatments. Nevertheless, this investigation distinctively emphasizes the identification of internal reference genes in quinoa seeds throughout the process of natural germination and exogenous hormone treatments. Consequently, this research is anticipated to make a valuable contribution to the exploration of quinoa germination dormancy, PHS, and forthcoming systematic molecular biology studies in quinoa.

MATERIAL AND METHODS

Materials and exogenous hormone treatment

The experimental material utilized in this study was the H1 cultivar of quinoa, which was obtained from the Key Laboratory of Coarse Cereal Processing, Ministry of Agriculture and Rural Affairs P.R. China. The quinoa samples were stored at a temperature of 4 ^oC. In order to examine the impact of exogenous hormone treatment on seed germination, precise measurements of 0.0661 g and 0.0866 g of ABA and GA, respectively, were obtained from storage at -20 ^oC. Afterwards, the samples were placed into individual containers containing a small amount of anhydrous ethanol and mixed until complete dissolution of ABA/GA occurred. The resulting solutions were then diluted with ultrapure water to a final volume of 250 mL, resulting in a stock solution of 1 mM ABA and GA. Subsequently, ABA and GA solutions were prepared by serially diluting the stock solution to concentrations of 200 μM, 100 μM, 50 μM, and 10 μM. Plump quinoa seeds, chosen based on their visual characteristics, were immersed in 70 % ethanol for a duration of 20 s. The seeds were then rinsed three times with water and any excess surface moisture was absorbed using filter paper. A set of 100 seeds were subsequently positioned on a circular glass Petri dish, which possessed a radius of 4.5 cm, and was equipped with germination filter paper. Each Petri dish was filled with 6mL of a hormone solution that had been prepared with different concentrations. In the control group, an equivalent volume of H₂O was introduced. The germination filter paper and seeds were diligently maintained in a moist state. Each treatment was replicated three times biologically. The petri dishes were subsequently placed in a constant temperature chamber (Jiangnan, Zhejiang, China) with a light/dark cycle of 16 h light and 8 h darkness, maintaining a temperature of 25 °C during the day and 22 °C during the night, a relative humidity of 50%, and a light intensity of 40% (Tang et al., 2022). After 48 h, the seeds were washed three times with water. Fresh filter paper was inserted into the petri dishes, and 6 mL of ultrapure water was added to each plate. The plates were then incubated for an additional 24 h under the same conditions.

The germination standard was operationally defined as the point at which the quinoa sprout attains 50 % of the seed width (Tang et al., 2022). The initial germination stage spans from 0 to 48 h, during which data is collected at 4-hour intervals. Subsequently, in the second stage, spanning from 48 to 72 h, data is collected at 12-hour intervals. At each data collection point, the germinated seeds are carefully extracted from the culture dish. The calculation formulas for germination rate (GR) (Metwally et al., 2022), germination potential (GP) (Wang et al., 2023), germination index (GI) (Aloui et al., 2014; Chen et al., 2018), mean germination time (MGT) (Ullah et al., 2022; Mohanty et al., 2023), and peak germination (PG) (Arnolds et al., 2015) are provided below.

$GR\left(\%\right)=\left(G_{n1}/N\right)\times 100$

$GP\left(\%\right)=\left(G_{n2}/N\right)\times 100$

$GI=\sum {(G}_t/T_t$

$MGT=(\sum \left(Gt*Tt\right))/G_{n1}$

$PG=max(Gt/Tt)$

In the context of formulas, G_n1 represents the count of seeds that have germinated after a 72-hour period, while G_n2 represents the count of seeds that have germinated within a 12 to 16-hour timeframe. N denotes the total number of seeds, G_t signifies the count of seeds that have germinated between the time intervals t and t-4 hour, and Tt represents the duration in t hours. The maximum quotient of G_t/T_t is denoted as PG. The determination of the optimal treatment concentration was made by considering the outcomes of the conducted experiments.

A total of 0.5 g of quinoa seeds were accurately measured and subsequently treated with 6 mL of ABA/GA/H₂O in separate petri dishes. The concentrations of ABA/GA utilized in the treatment were determined based on the concentrations established in the preceding experiment. The experimental conditions were upheld as previously outlined. Each treatment was replicated three times using distinct biological samples. Sampling was conducted at 4 h, 12 h, 20 h, 28 h, 36 h, and 44 h. The seeds’ surface moisture was absorbed using filter paper, followed by their wrapping in aluminum foil and subsequent freezing in liquid nitrogen. The frozen seeds were then stored at a temperature of -80 °C to facilitate the subsequent extraction of RNA.

RNA extraction, first strand cDNA synthesis and primer design

The material was removed from the freezer at a temperature of -80 °C, pulverized in a mortar with the addition of liquid nitrogen, and a quantity of 50 mg of the resulting powder was utilized for the extraction of RNA. The extraction of RNA from all samples was performed using a plant tissue extraction kit (Tiangen, Beijing, China). Following this, the integrity and concentration of the RNA were evaluated using the Scan Drop100 high-throughput protein concentration meter (Analytik jena AG, Jena, Germany) by measuring the absorbance at wavelengths of 260 nm and 280 nm. The A260 / A280 ratio, which is typically between 1.8 and 2.2, is commonly used as an indicator of RNA purity and concentration. Furthermore, a 50 M TAE buffer (pH=8.0-8.6, Sangon Biotech, Shanghai, China) was diluted to 1 M TAE for the purpose of gel electrophoresis of nucleic acids. The gel electrophoresis was performed using a 1.00 % agarose gel in 1 M TAE buffer, and GoldviewTM dye (ZOMANBIO, Beijing, China) was utilized as the staining agent. The electrophoretic results were observed using a gel imaging system (BIO RAD, California, U.S.A) to validate the RNA quality based on established experimental criteria. Following the confirmation of RNA quality, the first-strand cDNA synthesis was conducted using the FastKing RT Kit (with gDNase, Tiangen, Beijing, China). The gDNA removal system was prepared by combining 2 μL of 5× gDNA Buffer, 1 μg RNA, and RNase-Free ddH₂O to achieve a final volume of 10 μL. The resulting mixture was thoroughly mixed and incubated at 42 °C for 3 m, followed by immediate placement on ice. For the reverse transcription system, a total volume of 10 μL was prepared by combining 2 μL of 10King RT Buffer, 1 μL of FastKing RT Enzyme Mix, and 2 μL of FQ-RT Primer Mix, with the remaining volume supplemented with RNase-Free ddH₂O. The resulting mixture was combined with the gDNA removal system, and the subsequent reaction was carried out at a temperature of 42 °C for a duration of 15 minutes. This was followed by a denaturation step at 95 °C for 3 m, resulting in the synthesis of the first chain cDNA.

Following a comprehensive literature review, a total of seven potential reference genes, namely TUB1, TUB6, EIF3, EFIα, GAPDH, MON1, and ACT, were chosen for the purpose of validation. The sequences of the target genes that underwent successful validation were sourced from the Gene Database of the National Center for Biotechnology Information (NCBI) available at https://www.ncbi.nlm.nih.gov. The protein sequences were employed as templates for conducting BLAST searches to ascertain homologous sequences in quinoa. The homologous sequences that were obtained through downloading were chosen based on alignment metrics, including total score, query cover, E-value, and percent identity. Subsequently, primers were devised to span exonic regions, with lengths varying from 80 to 200 bp and GC falling within the range of content between 40% to 60%. The design of these primers was accomplished through the utilization of NCBI-primer-Blast and Primer Premier 5.0.

RT-qPCR assay

The samples were acquired from three biological replicates. The double-stranded cDNA synthesis process was monitored using the M5 Hiper SYBR Premix EsTaq with Tli RNaseH (Mei5 Bio, Beijing, China). The RT-qPCR was performed on white 96-well plates using the qTOWER3 G instrument (Analytik jena AG, Jena, Germany). Each reaction was prepared using a 20 µL system, consisting of 1 µL of diluted cDNA (with an approximate concentration of 80 ng.µL^-1), 0.4 µL of each forward and reverse primer (100 µM), and 10 µL of 2*M5 Hiper SYBR Premix EsTaq (containing Tli RNaseH). The total volume was adjusted to 20 µL by adding additional ddH₂O. The RT-qPCR program comprised a pre-denaturation step at 95 °C for 30 s, followed by a reaction stage at 95 °C for 5 s, and 60 °C for 20 s, repeated for a total of 40 cycles. Melting curves were generated by gradually increasing the temperature from 60 °C to 95 °C at a rate of 5 °C per second to confirm the specificity of the primers used for amplification. Additionally, no template controls (NTC) were included in the experimental process, and no amplification was observed in the NTC, indicating the absence of primer dimer formation.

Amplification efficiency and stability analysis

A fivefold dilution process was applied to the cDNA samples to produce templates with diverse concentrations. The amplification efficiency was subsequently determined using the formula $E\left(\%\right)=(10^(-1/slope)-1)*100$, as described by (Wang et al., 2022a) and (Migocka and Papierniak, 2011), with a qualification of 90-110 % amplification efficiency. Additionally, the primer amplification efficiency was further validated in this study using LinRegPCR, as outlined by (Borges et al., 2012).

The stability of candidate genes was calculated by methods, including Ct values, geNorm, NormFinder, BestKeeper, RefFinder, and Delta Ct. The analysis of Ct values involved the plotting of all values to determine the maximum, minimum, percentiles, and medians, thereby providing an initial assessment of the dispersion of reference gene expression. It is important to note that the NormFinder, BestKeeper, RefFinder, and Delta Ct utilized in this study are R packages for Windows, while geNorm is a software. Each of these approaches employs unique calculation methodologies.

Before performing geNorm analysis, the Ct values were normalized using the 2^-ΔCt method, where ΔCt = (Ct - Ct _min) (Wang et al., 2019). GeNorm (Vandesompele et al., 2002) evaluates the stability of potential reference genes by comparing their expression levels to those of other candidates, utilizing the standard deviation of the logarithmic ratios of the disparities between them. The stability of candidate genes is assessed by the M value, where a smaller M value signifies greater stability. The impact of including an additional reference gene on the results is evaluated using the paired variation value Vn/(Vn+1) in geNorm, where a Vn/(Vn+1) value exceeding 0.15 suggests the requirement for more reference genes. Conversely, an Vn/(Vn+1) value below 0.15 suggests a negligible impact from the addition of another reference gene. NormFinder was employed to compute the S value for candidate genes, where a smaller S value indicates higher stability of the gene. The algorithms employed by BestKeeper rely on Cross Pinot (CP), a method that identifies optimal reference genes by conducting pairwise correlation analysis and calculating the geometric mean of sample pairs (Pfaffl et al., 2004). BestKeeper further computes various statistical measures, including the Geometric Mean (GM), Arithmetic Mean (AM), Minimal value (Min), Maximal value (Max), Standard Deviation (SD), and Coefficient of Variance (CV), based on the sample data. The stability of candidate genes increases as the coefficient of variation (CV) and standard deviation (SD) decrease. The Delta Ct method, as proposed by Hu et al. (2014), identifies reference genes with stable expression by comparing the relative expression levels of genes in the samples. However, it should be noted that there is not uniform consistency among all treatment groups in terms of the results obtained from different tools. To address this issue, RefFinder employs a weighting system for the calculation methods utilized by various tools mentioned. This enables the provision of an overall ranking of candidate reference gene stability.

Based on the aforementioned analysis findings, the two most stable candidates and the least stable candidate gene were chosen as reference genes. ABA8ox1 (AUR62030408-RA) and GA2ox1 (AUR62002752-RA), which are metabolic genes associated with ABA and GA, were designated as target genes. The 2^-ΔΔCt method was employed to quantify their expression levels. The RT-qPCR system and program used were consistent with those described in section ‘RT-qPCR assay’. To verify the dependability of the reference gene selection outcomes, the expression levels were assessed separately under different treatments.

Statistical analysis

Graphs were created using GraphPad Prism 8 and Origin 2023. Mean, standard deviation, and one-way ANOVA were analyzed using IBM SPSS Statistics 26. Amplification efficiency was calculated using Analytik Jena software, in conjunction with Excel 2016 and LinRegPCR.

RESULTS AND DISCUSSION

The germination rates under various treatments were observed and recorded in Figure 1 within a 72-hour time frame. At the 48-hour mark, the CK group exhibited an average germination rate of 98% ± 1%. In the ABA group, the germination rates at concentrations of 10 μM, 50 μM, 100 μM, and 200 μM were 98 % ± 1%, 63% ± 4%, 18% ± 3%, and 10% ± 2%, respectively. The germination rates of the GA group at concentrations of 10 μM, 50 μM, 100 μM, and 200 μM were determined to be 95% ± 1%, 98% ± 1%, 98% ± 2%, and 96% ± 0%, respectively. Comparatively, the germination rates of the ABA group at concentrations of 50 μM, 100 μM, and 200 μM were observed to be lower than those of both the CK and GA groups after 48 h. However, no statistically significant difference was found between the germination rates of the GA and CK groups. Following the completion of the initial 48-hour stage, the seeds were subsequently transferred into petri dishes containing water. During the subsequent 24-hour period, the groups treated with ABA at concentrations of 50, 100, and 200 μM, which had previously displayed lower germination rates in the first stage, demonstrated a rapid attainment of complete germination. Table 1 presents the average values, standard deviations, and results of a one-way ANOVA for the seed vigor-related indicators PG, GP, GI, and MGT in the CK, ABA, and GA groups. In the ABA group, significant differences were observed in all indicators at 50 μM, 100 μM, and 200 μM compared to the CK group. Conversely, in the GA group, no significant differences were found at 10 μM and 50 μM compared to the CK group. However, significant differences were observed in PG and GP at 100 μM and 200 μM, while MGT and GI did not show significant differences.

Figure 1
The statistical analysis conducted in this study examined the germination rates of seeds subjected to treatments involving H₂O, ABA, and GA within a 72-hour period. The control group, denoted as CK, was treated with H₂O. ABA treatments were categorized as I to IV, representing concentrations of 10 μM, 50 μM, 100 μM, and 200 μM, respectively. Similarly, GA treatments were categorized as V to VIII, representing concentrations of 10 μM, 50 μM, 100 μM, and 200 μM, respectively.

Table 1
Statistics of seed germination vitality indicators.

	PG	GP	GI	MGT
CK	1.875±0.165 c	0.300±0.026 c	5.458±0.050 ab	19.317±0.524 ef
I	1.612±0.099 c	0.257±0.015 c	4.704±0.109 c	22.211±0.349 d
II	0.483±0.060 e	0.067±0.012 d	2.742±0.210 d	41.019±1.509 c
III	1.161±0.042 d	0.040±0.010 de	2.021±0.099 e	53.947±1.610 b
IV	1.244±0.051 d	0.007±0.006 e	1.817±0.182 e	58.192±0.96 9a
V	1.646±0.036 c	0.263±0.006 c	5.185±0.077 b	20.855±0.895 de
VI	1.771±0.130 c	0.283±0.021 c	5.371±0.223 ab	19.843±0.765 ef
VII	2.250±0.225 b	0.360±0.036 b	5.536±0.188 a	18.859±0.525 f
VIII	2.583±0.253 a	0.413±0.040 a	5.337±0.093 ab	19.252±0.552 ef

• Perform significance analysis based on the P ≤ 0.05 level. PG: Peak germination, GP: germination potential, GI: germination index, MGT: Mean germination time; From I to IV represents ABA 10 µM, 50 µM, 100 µM, 200 µM; From V to VIII represents GA 10 µM, 50 µM, 100 µM, 200 µM.

After treatment with four concentrations of hormone (10 μM, 50 μM, 100 μM, and 200 μM) for 24 h, there was no significant effect observed on the germination of quinoa seeds when treated with 10 μM ABA (Figure 2A).

Figure 2
Quinoa seeds were subjected to various concentrations of ABA and GA. The germination state at 48 hours was denoted by (A), while the group treated with H₂O was represented by “ck”. Columns a, b, c, and d corresponded to treatments with concentrations of 10 μM, 50 μM, 100 μM, and 200 μM ABA, respectively. Columns e, f, g, and h referred to GA treatments with concentrations of 10 μM, 50 μM, 100 μM, and 200 μM, respectively. (B) depicts images of seedlings subjected to various treatments. The seedling states on the 5th day of cultivation with water and varying concentrations of GA are denoted by ck, e, f, g, and h, while the states following the transfer of ABA to water culture for 3 days are represented by a, b, c, and d.

When seeds were subjected to 50 μM ABA treatment, shoot elongation exhibited a decrease compared to the CK group. Furthermore, at concentrations of 100 μM and 200 μM, germination was suppressed. Conversely, from the perspective of shoot length, no significant impact on seed germination was observed from treatment with 10 μM, 50 μM, and 100 μM GA. Notably, at a GA concentration of 200 μM, seed shoot elongation exhibited a discernible increase. The sustained germination process indicates the potential for successful seedling development (Figure 2B). In summation, for subsequent treatments, GA was administered at a concentration of 200 μM, while ABA was applied at a concentration of 100 μM.

Following the implementation of 1% agarose gel electrophoresis on each of the experimental RNA samples, the outcomes were visually presented in Figure 3. The assessment of RNA and cDNA purity and concentration was performed using Scan Drop. The study revealed that the A260/A280 ratio of the RNA samples varied between 1.82 and 2.21, while the purity of the cDNA ranged from 1.78 to 1.85 (Table 2). Detailed descriptions of the candidate gene primers were provided, specifying a maximum primer length of 197 bp and a minimum primer length of 121bp (Table 3). The melting curves of all candidate genes exhibited a unimodal peak shape, as illustrated in Figure 4. The amplification efficiency of the primers, as presented in Table 3, fell within the acceptable range of 90% to 110%, except for the EF1α primer, which was excluded due to its deviation from the defined range. The efficiencies of the remaining primers ranged from 94.17% to 110.17%, and their amplification curves demonstrated a strong linear relationship, as evidenced by the R² values ranging from 0.9853 to 0.9996. The amplification efficiency during revalidation using LinRegPCR varied between 102.8% and 110.3% as indicated in Table 3.

Figure 3
RNA agarose gel electrophoresis was performed on samples subjected to various treatments. Lanes 1-6 contained RNA extracted from samples treated with ABA for 4 h, 12 h, 20 h, 28 h, 36 h and 44 h, respectively.; Lanes 7-12 represented RNA obtained from samples treated with GA for the same time intervals. Lanes 13-18 contained RNA collected from samples treated with H₂O for 4 h, 12 h, 20 h, 28 h, 36 h, and 44 h, respectively. Lane 19 contained RNA from samples collected at 0 h. A 2000 bp marker was used for reference.

Table 2
Purity and concentration of RNA /cDNA.

Treatment	Time	RNA		cDNA
		A260/A280	Concentration ng.µL-1	A260/A280	Concentration ng.µL-1
	0h	1.93	165.69	1.81	923.67
GA	4 h	1.82	66.55	1.8	964.32
	12 h	1.97	216.35	1.83	1005.96
	20 h	1.95	325.37	1.83	1029.58
	28 h	2.21	346.33	1.81	882.77
	36 h	2.17	723.87	1.85	1043.86
	44 h	2.13	537.51	1.79	852.25
ABA	4 h	2.13	76.43	1.82	960.73
	12 h	2.01	210.92	1.81	983.06
	20 h	1.96	364.58	1.8	968.48
	28 h	2.16	612.26	1.84	975.35
	36 h	2.17	539.78	1.8	890.02
	44 h	2.15	344.11	1.78	887.63
H2O	4 h	1.88	52.63	1.81	964.96
	12 h	2.03	208.75	1.78	936.43
	20 h	2.02	338.04	1.84	1037.34
	28 h	2.15	442.07	1.85	998.2
	36 h	2.15	537.98	1.84	974.3
	44 h	2.16	538.86	1.83	1004.76

Table 3
Candidate reference genes characteristics and primer sequences information.

Gene name	Description	Gene ID	Amplicon length (bp)	Primer sequences	Slope	Intercept	R2	Efficiency
				forward /reverse(5´to3´)				Analytik Jena	LinRegPCR
ACT	Actin-1	110715281	173	AGCTTCTTGACGAAATGGGTG	-3.38	28.31	0.9996	97.63%	110.30%
				ACAACTCCTCACCTTCTCATGG
TUB6	Beta tubulin -6	110711758	157	GGAAAAATGAGAGAAATCCTTCACA	-3.35	29.79	0.9853	98.84%	105.60%
				AGTAGACATTGACGCGCTCC
TUB1	Alpha tubulin -1	110686159	144	GTGCTCCTGTATGGTGCCAA	-3.1	27.09	0.9986	110.17%	102.80%
EIF3	Eukaryotic translation initiation factor 3	110711181	121	AGCGGACCAACAAACAGAAAAG	-3.45	30.34	0.9914	94.92%	103.75%
				AGATCGATGTTGAGGCGAGC
GAPDH	Glyceraldehyde-3-phosphate dehydrogenase A	110711214	197	GGAAAAATGAGAGAAATCCTTCACA	-3.19	31.33	0.9961	105.82%	108.62%
				TCTTCCACAAGGCCAAGGAA
MON1	Vacuolar fusion protein MON1	110720838	147	TGCGAAGCCAGCGTAGTTTA	-3.47	31.27	0.9906	94.17%	103.85%
				GCTGGCTGCAGCTCAATATAC

Figure 4
The melting curves of candidate internal reference genes. The temperature indicated by the dashed line is the gene Tm value.

The Ct value is used to reflect the transcription level of a gene in mRNA, with a lower Ct value indicating a higher transcription level (Tang et al., 2017). In Figure 5, the maximum Ct value for GAPDH among all samples was 30.47, while the minimum Ct value for TUB1 was 17.49. The Interquartile range (IQR), analysis revealed that, as for GAPDH, the three treatments exhibited a relatively large dispersion, ranging from 19.11 to 30.47. In contrast, MON1 and EIF3 showed a more concentrated distribution, with Ct values ranging from 21.83 to 27.02 and 19.97 to 25.6, respectively.

Figure 5
The expression parameters of candidate genes are depicted using a box plot, where the box represents the interquartile range. The lower and upper boundaries of the box represent the 25th and 75th percentiles, respectively, while the line inside the box indicates the median. Specifically, (a) represents the Ct values of all treatments, (b) corresponds to the H₂O treatment group, (c) represents the ABA treatment group, and (d) refers to the GA treatment group.

According to the boxplot analysis, the H2O treatment group, the medians of TUB1, ACT, MON1, and GAPDH are positioned towards the lower end, while the median of TUB6 falls in the middle. The average Ct value of TUB1 in this group was the lowest, measuring at 20.92. In contrast, the ABA and GA treatment groups displayed the smallest IQR and the lowest dispersion for MON1 and EIF3. TUB6 demonstrated the lowest average Ct value in both hormone treatment groups, with a Ct value of 20.88 in the ABA treatment group and 20.51 in the GA treatment group.

The stability of six candidate genes was assessed using geNorm analysis (Table 4). In the H₂O group, ACT and TUB6 exhibited the highest stability (M = 0.075), while GAPDH demonstrated the lowest stability (M = 0.225). Conversely, in the ABA treatment group, TUB6 and ACT were identified as the most stable genes (M = 0.068), whereas in the GA treatment group, ACT and GAPDH displayed the highest stability (M = 0.072), with TUB6 being the least stable. When considering all treatments as a single group, geNorm analysis revealed that TUB1 and ACT exhibited high stability, as indicated by an M value of 0.097, whereas GAPDH demonstrated the lowest stability with an M value of 0.213. The determination of the appropriate number of reference genes was based on the paired variation values (Figure 6), which were 0.0295 (H₂O), 0.0276 (ABA), 0.0494 (GA), and 0.0446 (all groups) for V2/3.

Table 4
Analysis results and stability ranking by geNorm, NormFinder, Delta Ct, RefFinder.

Treatment	Rank	geNorm		NormFinder		Delta Ct		RefFinder
		Gene	mean M	Gene	S value	Gene	Stability	Gene	Stability
H2O	1	TUB6	0.075	TUB1	0.21	ACT	1.18	ACT	1.68
	2	ACT	0.075	ACT	0.43	TUB1	1.19	TUB1	1.78
	3	EIF3	0.088	MON1	0.51	TUB6	1.2	MON1	2.63
	4	MON1	0.097	TUB6	0.68	MON1	1.21	TUB6	3.22
	5	TUB1	0.103	EIF3	1.19	EIF3	1.49	EIF3	3.98
	6	GAPDH	0.225	GAPDH	2.51	GAPDH	2.58	GAPDH	6
ABA	1	TUB6	0.068	TUB6	0.28	TUB6	0.83	MON1	1.68
	2	ACT	0.068	MON1	0.56	MON1	0.91	EIF3	1.73
	3	TUB1	0.082	EIF3	0.67	EIF3	0.96	TUB6	1.86
	4	MON1	0.091	TUB1	0.67	TUB1	0.98	ACT	4.16
	5	EIF3	0.098	ACT	0.71	ACT	1.01	TUB1	4.47
	6	GAPDH	0.154	GAPDH	1.34	GAPDH	1.47	GAPDH	6
GA	1	ACT	0.072	EIF3	0.65	EIF3	1.15	EIF3	1.19
	2	GAPDH	0.072	TUB1	0.66	TUB1	1.19	MON1	1.73
	3	TUB1	0.126	MON1	0.83	MON1	1.23	TUB1	2.78
	4	EIF3	0.157	ACT	0.86	ACT	1.27	ACT	4
	5	MON1	0.189	TUB6	1	TUB6	1.35	TUB6	4.4
	6	TUB6	0.227	GAPDH	1.57	GAPDH	1.76	GAPDH	6
All treatments	1	TUB1	0.097	TUB1	0.51	MON1	1.11	MON1	1.19
	2	ACT	0.097	MON1	0.58	ACT	1.13	EIF3	2.51
	3	EIF3	0.127	ACT	0.62	TUB1	1.14	ACT	2.91
	4	MON1	0.149	EIF3	0.75	TUB6	1.17	TUB1	2.94
	5	TUB6	0.164	TUB6	0.76	EIF3	1.17	TUB6	3.94
	6	GAPDH	0.213	GAPDH	1.98	GAPDH	2.08	GAPDH	6

Figure 6
The geNorm method was employed to determine the pairwise variation (V) of candidate reference genes.

These findings demonstrate that employing the top-ranked two candidate genes as internal reference genes resulted in unaffected experimental normalization when a third or more internal reference genes were added. Consequently, it is recommended that the utilization of two internal reference genes is adequate, obviating the necessity of introducing additional internal reference genes.

NormFinder was employed to calculate the S value for six candidate genes (Table 4). The stability ranking assigned by NormFinder for various liquid treatments of quinoa seeds exhibited dissimilarity. Among the quinoa grain samples treated with H₂O, the stability ranking of the candidate genes, in descending order, was as follows: TUB1, ACT, MON1, TUB6, EIF3, and GAPDH, with TUB1 exhibiting the lowest S value of 0.21. In the ABA-treated group, the candidate genes exhibited a stability ranking of TUB6, MON1, EIF3, TUB1, ACT, and GAPDH, with S values ranging from 0.28 to 1.34. Notably, TUB6 displayed the lowest S value of 0.28. Conversely, in the GA-treated group, the stability ranking of the candidate genes was EIF3, TUB1, MON1, ACT, TUB6, and GAPDH, with EIF3 exhibiting an S value of 0.65. In terms of the comprehensive analysis findings across the three treatment groups, the stability ranking is as follows: TUB1, MON1, ACT, EIF3, TUB6, GAPDH. The highest-ranked candidate gene, TUB1, exhibits an S value of 0.51. NormFinder consistently identified GAPDH as the least stable candidate gene across all groups. In three conditions, the gene GAPDH consistently exhibited the lowest level of stability.

The data from four sets were subjected to analysis using BestKeeper. The sample sizes for the H₂O, ABA, and GA groups were 21, while the fourth group, encompassing all treatments, had a sample size of 57. Table 5 presents the GM, AM, Min, Max, SD, and CV. The results of the study revealed that MON1 and EIF3 were identified as the most stable genes across all four categories, as indicated in Table 5. The CV values, accompanied by the SD values, for MON1 and EIF3 in the H₂O, ABA, GA treatments, and for all treatments collectively were as follows: 5.79±1.39, 6.38±1.46; 4.88±1.18, 5.2±1.17; 6.09±1.47, 7.04±1.56; 5.55±1.33, 6.45±1.45; respectively. Notably, based on the analysis conducted using BestKeeper, MON1 emerged as the most stable candidate gene for both H₂O and GA treatments, as well as for all treatments combined. EIF3 was determined to be the most stable gene within the ABA group. Conversely, GAPDH was found to be the least dependable gene across all groups. Research by Wang et al. (2022a) has shown that in BestKeeper analysis, SD values less than 1 indicate higher stability. However, in this study, all SD values were less than 1, and this observation may be due to the sample size.

Table 5
Stability of candidate gene results by BestKeeper.

Treatment	Rank	Gene name	geo Mean	AR Mean	Min	Max	Standard deviation (SD)	Coefficient of variation (CV)
H2O	1	MON1	24.02	24.08	21.83	27.02	1.39	5.79
	2	EIF3	22.85	22.91	20.28	25.6	1.46	6.38
	3	TUB6	20.68	20.77	18.19	23.58	1.71	8.25
	4	ACT	21.93	22.05	19.29	25.81	2.03	9.22
	5	TUB1	20.79	20.92	18.25	24.67	2.06	9.86
	6	GAPDH	24.44	24.73	19.11	30.47	3.4	13.74
ABA	1	EIF3	22.53	22.57	20.85	25.08	1.17	5.2
	2	MON1	24.09	24.13	22.36	26.41	1.18	4.88
	3	ACT	22.22	22.28	20.34	25.47	1.45	6.51
	4	TUB6	20.78	20.88	18.47	24.5	1.73	8.28
	5	TUB1	21.56	21.67	19.39	25.66	1.91	8.82
	6	GAPDH	25.69	25.82	22.12	29.76	2.35	9.12
GA	1	MON1	24.04	24.09	22.02	25.74	1.47	6.09
	2	EIF3	22.06	22.12	19.97	24.22	1.56	7.04
	3	TUB6	20.4	20.51	17.56	23.93	1.81	8.81
	4	ACT	21.35	21.46	18.56	25.47	1.87	8.72
	5	TUB1	20.61	20.76	17.49	24.67	2.11	10.18
	6	GAPDH	24.81	25.01	20.51	28.41	2.97	11.87
All treatments	1	MON1	23.91	23.96	21.83	27.02	1.33	5.55
	2	EIF3	22.37	22.43	19.97	25.6	1.45	6.45
	3	ACT	21.48	21.57	18.56	25.81	1.52	7.04
	4	TUB6	20.35	20.44	17.56	24.5	1.61	7.89
	5	TUB1	20.65	20.77	17.49	25.66	1.76	8.49
	6	GAPDH	24.67	24.88	19.11	30.47	2.94	11.82

The Delta Ct calculation results were presented in Table 4. Among the H₂O-treated samples, ACT and TUB1 emerged as the most stable candidate reference genes, exhibiting stability values of 1.18 and 1.19, respectively. In the case of ABA treatment, TUB6 and MON1 were relatively stable candidate genes, displaying corresponding stability values of 0.83 and 0.91. Similarly, in the GA treatment, EIF3 and TUB1 were identified as the most stable candidate genes, with stability values of 1.15 and 1.19, respectively. When the three data treatments were consolidated into a single group, the Delta Ct calculation outcomes revealed that MON1 and ACT exhibited the highest levels of stability, with values of 1.11 and 1.13, respectively. In all groups, the Delta Ct method consistently indicated that GAPDH was the least stable reference gene candidate.

RefFinder utilizes geNorm, NormFinder, BestKeeper, and Delta Ct calculations to assign distinct weights and re-evaluate the stability, resulting in a revised ranking. The stability rankings of potential reference genes in the H₂O treatment group were determined to be ACT > TUB1 > MON1 > TUB6 > EIF3 > GAPDH. Similarly, in the ABA-treated group, the stability rankings were found to be MON1 > EIF3 > TUB6 > ACT > TUB1 > GAPDH, while in the GA-treated group, the stability rankings were determined to be EIF3 > MON1 > TUB1 > ACT > TUB6 > GAPDH, as presented in Table 4. RefFinder analysis yielded a stability ranking of MON1 > EIF3 > ACT > TUB1 > TUB6 > GAPDH when the three treatments were examined collectively. The findings of Zhu (Zhu et al., 2021) demonstrated that TUB1 and EF1α were identified as the most acceptable for Long Li N° 1 quinoa under salt stress and ABA treatment. Conversely, GAPDH exhibited the highest level of instability when subjected to salt stress, 200 μM ABA, and other treatments, using various tissues as RNA templates. However, our findings indicate that the MON1 and EIF3 genes exhibited the highest stability in the H1 variety when subjected to hormone treatments. Similarly, the GAPDH gene displayed low stability in the Long Li N° 1 quinoa variety. These variations in gene stability could be attributed to the inherent differences in cultivars and treated tissues.In all conditions, the GAPDH consistently exhibited the lowest level of stability, with the exception of the geNorm analysis of samples treated with GA, where it ranked second in terms of stability. These results align with previous studies(Ruduś and Kępczyński 2018; Sudhakar-Reddy et al. 2018), wherein diverse stability analysis methods yielded disparate outcomes.

Based on the findings of the study, the choice of candidate reference genes for quinoa seeds is influenced by the specific treatments applied. Specifically, under H₂O treatment, the genes ACT and TUB1 were identified as the most stable. Conversely, in the 100 μM ABA and 200 μM GA treatments, the candidate genes EIF3 and MON1 exhibited similar levels of stability. Furthermore, when the data from all three treatments were integrated and analyzed, EIF3 and MON1 emerged as the top two genes in terms of stability. GAPDH emerged as the least stable candidate genes across all groups. In order to ensure the credibility of the results, reference genes ACT and TUB1, which exhibited high stability in the presence of H₂O treatment, were chosen alongside GAPDH, which demonstrated the lowest stability, to ascertain the relative expression levels of the target genes ABA8ox1 (TGCAGACAAAGTTAAAAAGTATGGT/AAATTTAGCTGCATCCGGGC) and GA2ox1 (GTTGGTGACTCTTTGCAGGTG/TGTCAGCCAAAACCCTGTGT). Furthermore, the expression levels of the target genes were determined using EIF3 and MON1, which are characterized by high stability, and GAPDH, a gene with low stability, as reference genes under ABA treatment.

The validation outcome of the reference genes, as depicted in Figure 7, demonstrated that the relative expression levels of the target genes remained relatively consistent when ACT, TUB1, and ACT+TUB1 were employed as reference genes under the H₂O group. However, significant fluctuations were observed when the least stable candidate gene, GAPDH, was utilized as a reference gene. In the H₂O-treated samples of ABA8ox1, the control group at 0 h was established. The utilization of ACT, TUB1, and ACT+TUB1 as reference genes revealed a pattern of small-scale fluctuation in expression (Figure 7a). The gene expression levels ranged from 0.599 as the lowest value to 1.725 as the highest value across different time intervals. Conversely, when GAPDH was employed as a reference gene, the expression levels exhibited a wider distribution, ranging from 0.280 to 7.568. In Figure 7b, when employing the more stable genes as reference genes for GA2ox1, the expression levels exhibited a pattern of first decreasing, then fluctuating within a small range, value ranging from 0.142 to 0.480, and all were observed to be down-regulated. Conversely, when GAPDH was utilized as the reference gene, the gene expression levels ranged from 0.0735 to 3.053, displaying a contrasting trend compared to when ACT, TUB1, and ACT+TUB1 were employed as reference genes, wherein an initial increase was followed by a subsequent decrease. Hence, the utilization of unstable genes as reference genes during the process of quinoa seed germination in the absence of external substances may lead to an overestimation of the actual expression levels.

Figure 7
Validation results of stability of internal reference genes. (a), (b) represents the relative expression levels of ABA8ox1 and GA2ox1 in H₂O treatment, and (c), (d) refers to the relative expression levels of ABA8ox1 and GA2ox1 in ABA treatment, respectively.

Following the application of ABA treatment on quinoa seeds, the reference genes EIF3, MON1, and EIF3+MON1 were employed, resulting in expression levels of ABA8ox1 ranging from 1.491 to 24.533 when compared to the control group. Similarly, the expression levels of GA2ox1 ranged from 0.891 to 5.169 (Figures 7c and d). When GAPDH was used as a reference gene, the expression level of ABA8ox1 ranged from 0.522 to 3.971, while the minimum and maximum expression levels of GA2ox1 were 0.0761 and 2.707, as depicted in Figure 7 c and d, respectively. When utilizing genes with high stability as reference genes, the ABA8ox1 trend in the ABA treatment group exhibited a gradual increase from 0 to 44 h. Conversely, the utilization of the less stable GAPDH gene displayed a distinct trend. Similarly, when employing EIF3, MON1, and EIF3+MON1 as reference genes, the expression trend of GA2ox1 remained consistent from 0 to 44 h, while diverging from that of GAPDH. During the validation process of reference gene stability, it was observed that when either MON1 or EIF3 was used as the sole reference gene in the hormone-treated group, the expressing trend remained consistent. Nonetheless, there exist numerical differences in the gene expression levels. Hence, employing a combination of two genes as reference genes is deemed to be a more advantageous approach.

This study represents the initial exploration of appropriate internal reference genes for grain germination in quinoa, employing diverse exogenous hormone treatments. In essence, this research offers significant contributions to the understanding of seed dormancy and PHS, also paves the way for future investigations in the fields of molecular biology, genetics, and quinoa breeding.

CONCLUSIONS

Quinoa seeds were subjected to varying concentrations of ABA and GA, specifically 200 μM, 100 μM, 50 μM, and 10 μM. Notably, the concentrations of 100 μM ABA and 200 μM GA exhibited the most favorable outcomes.

Consequently, our findings recommend ACT as the internal reference gene for H₂O treatment, while the combination of MON1 and EIF3 is proposed for hormone treatment, as well as for experiments involving concurrent H₂O and ABA or GA treatments.

Improper selection of reference genes for quinoa seeds subjected to water treatment may lead to an overestimation of the expression level of the target gene. Conversely, when exposed to ABA and GA hormone treatments, such selection may result in an underestimation of the expression level of the target gene.

ACKNOWLEDGMENTS

The author expresses gratitude to the Sichuan Province Science and Technology Program (Grant 2022YFQ0041) and the Project of Sichuan Provincial Administration of Traditional Chinese Medicine (2023MS273) for their generous financial assistance.

REFERENCES

• ALOUI, H.; SOUGUIRC, M.; HANNACHI, C. Determination of an optimal priming duration and concentration protocol for pepper seeds (Capsicum annuum L). Acta Agriculturae Slovenica, v.103, n.2, p.213-221,2014. https://doi.org/10.14720/aas.2014.103.2.6
  » https://doi.org/https://doi.org/10.14720/aas.2014.103.2.6
• ARNOLDS, J.L.; MUSIL, C.F.; REBELOG, A.G.; KRüGER, H. Experimental climate warming enforces seed dormancy in South African Proteaceae but seedling drought resilience exceeds summer drought periods. Oecologia, v.177, p.1103-1116,2015. https://doi.org/10.1007/s00442-014-3173-6
  » https://doi.org/https://doi.org/10.1007/s00442-014-3173-6
• BARRERO, J.M.; PORFIRIO, L.; HUGHES, T.; CHEN, J.; DILLON, S.; GUBLERJ, F.; RAL, F. Evaluation of the impact of heat on wheat dormancy, late maturity α-amylase and grain size under controlled conditions in diverse germplasm. Scientific Reports, v.10, n.1, p.1-11,2020. https://doi.org/10.1038/s41598-020-73707-8
  » https://doi.org/https://doi.org/10.1038/s41598-020-73707-8
• BEVITORI, R.; OLIVEIRA, M.B.; GROSSI-DE-SÁ, M.F.; LANNA, A.C.; SILVEIRA, R.D.; PETROFEZA, S. Selection of optimized candidate reference genes for qRT-PCR normalization in rice (Oryza sativa L) during Magnaporthe oryzae infection and drought. Genetics and Molecular Research, v.13, n.4, p.9795-9805,2014. https://doi.org/10.4238/2014.November.27.7
  » https://doi.org/https://doi.org/10.4238/2014.November.27.7
• BORGES, A.; TSAID, S.M.; CALDAS, G. Validation of reference genes for RT-qPCR normalization in common bean during biotic and abiotic stresses. Plant Cell Reports, v.31, n.5, p.827-838,2012. https://doi.org/10.1007/s00299-011-1204-x
  » https://doi.org/https://doi.org/10.1007/s00299-011-1204-x
• BUSTIN, S.A.; BENES, V.; GARSON, J.A.; HELLEMANS, J.; HUGGETT, J.; KUBISTA, M.; MUELLER, R.; NOLAN, T.; PFAFFL, M.W.; SHIPLEY, G.L.; VANDESOMPELEC, J.; WITTWER, T. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clinical Chemistry, v.55, n.4, p.611-622,2009. https://doi.org/10.1373/clinchem.2008.112797
  » https://doi.org/https://doi.org/10.1373/clinchem.2008.112797
• CAI, J.; P. LI; X. LUO; T. CHANG; J. LI; Y. ZHAOY. XU. Selection of appropriate reference genes for the detection of rhythmic gene expression via quantitative real-time PCR in Tibetan hulless barley PLoS One, v.13, n.1, p.1-19,2018. https://doi.org/10.1371/journal.pone.0190559
  » https://doi.org/https://doi.org/10.1371/journal.pone.0190559
• CHEN, L.; TAN, G.J.T.; PANG, X.; YUAN, W.; LAI. S.;. YANG, H. Energy Regulated Nutritive and Antioxidant Properties during the Germination and Sprouting of Broccoli Sprouts (Brassica oleracea var. italica). Journal of Agricultural and Food Chemistry, v.66, n.27, p.6975-6985,2018. https://doi.org/10.1021/acs.jafc.8b00466
  » https://doi.org/https://doi.org/10.1021/acs.jafc.8b00466
• DAKHILI, S.; ABDOLALIZADEH, L.; HOSSEINI, S.M.; SHOJAEE-ALIABADIL, S. MIRMOGHTADAIE. Quinoa protein: Composition, structure and functional properties. Food Chemistry, v.299, p.1-10,2019. https://doi.org/10.1016/j.foodchem.2019.125161
  » https://doi.org/https://doi.org/10.1016/j.foodchem.2019.125161
• DENG, L.T.; WU, Y.L.; LI, J.C.; OUYANG, K.X.; DING, M.M.; ZHANG, J.J.; LI, S.Q.; LIN, M.F.; CHEN, H.B.; HUX, X.S.; CHEN, Y. Screening reliable reference genes for RT-qPCR analysis of gene expression in Moringa oleifera PLoS One , v.11, n.8, 2016. https://doi.org/10.1371/journal.pone.0159458
  » https://doi.org/https://doi.org/10.1371/journal.pone.0159458
• FENG, K.; LIU, J.X.; XING, G.M.; SUN, S.; LI, S.; DUAN, A.Q.; WANG, F.; LI, M.Y.; XUA, Z.S.; XIONG, S. Selection of appropriate reference genes for RT-qPCR analysis under abiotic stress and hormone treatment in celery. PeerJ, v.7, p.1-19,2019. https://doi.org/10.7717/peerj.7925
  » https://doi.org/https://doi.org/10.7717/peerj.7925
• FIALLOS-JURADO, J.; POLLIER, J.; MOSES, T.; ARENDT, P.; BARRIGA-MEDINA, N.; MORILLO, E.; ARAHANA, V.; TORRES, M.L.; GOOSSENSA, A. . LEON-REYES, A. Saponin determination, expression analysis and functional characterization of saponin biosynthetic genes in Chenopodium quinoa leaves. Plant Science, v.250, p.188-197,2016. https://doi.org/10.1016/j.plantsci.2016.05.015
  » https://doi.org/https://doi.org/10.1016/j.plantsci.2016.05.015
• GAETE-EASTMAN, C.; MATTUS-ARAYA, E.; HERRERAM, R.; MOYA-LEÓN, A. Evaluation of reference genes for transcript normalization in Fragaria chiloensis fruit and vegetative tissues. Physiology and Molecular Biology of Plants, v.28, n.8, p.1535-1544,2022. https://doi.org/10.1007/s12298-022-01227-y
  » https://doi.org/https://doi.org/10.1007/s12298-022-01227-y
• HU, Q.; GUO, H.; GAO, Y.; TANG, R.; LI, D. Reference gene selection for real-time RT-PCR normalization in rice field eel (Monopterus albus) during gonad development. Fish Physiology and Biochemistry, v.40, n.6, p.1721-1730,2014. https://doi.org/10.1007/s10695-014-9962-3
  » https://doi.org/https://doi.org/10.1007/s10695-014-9962-3
• JIA, K.P.; MI, J.; ALI, S.; OHYANAGI, H.; MORENO, J.C.; ABLAZOV, A.; BALAKRISHNA, A.; BERQDAR, L.; FIORE, A.; DIRETTO, G.; MARTíNEZ, C.; LERA, A.R.; GOJOBORIS, T.. AL-BABILI, S. An alternative, zeaxanthin epoxidase-independent abscisic acid biosynthetic pathway in plants. Molecular Plant, v.15, n.1, p.151-166,2022. https://doi.org/10.1016/j.molp.2021.09.008
  » https://doi.org/https://doi.org/10.1016/j.molp.2021.09.008
• KASHIWAKURA, Y.; KOBAYASHI, D.; JIKUMARU, Y.; TAKEBAYASHI, Y.; NAMBARA, E.; SEO, M.; KAMIYA, Y.; KUSHIRO, T.; KAWAKAMI, N. Highly sprouting-tolerant wheat grain exhibits extreme dormancy and cold Imbibition-resistant accumulation of abscisic acid. Plant Cell Physiology, v.57, n.4, p.715-732,2016. https://doi.org/10.1093/pcp/pcw051
  » https://doi.org/https://doi.org/10.1093/pcp/pcw051
• LIU, X.; GUAN, H.; SONG, M.; FU, Y.; HAN, X.; LEI, M.; REN, J.; GUO, B.; HE, W.; WEI, Y. Reference gene selection for qRT-PCR assays in Stellera chamaejasme subjected to abiotic stresses and hormone treatments based on transcriptome datasets. PeerJ , v.6, p.1-21,2018. https://doi.org/10.7717/peerj.4535
  » https://doi.org/https://doi.org/10.7717/peerj.4535
• LIU, Z.; XIAO, J.; XIA, Y.; WU, Q.; ZHAO, C.; LI, D. Selection and validation of reference genes for RT-qPCR-based analyses of Anastatus japonicus Ashmead (Hymenoptera: Helicopteridae) Frontiers in Physiology, v.13, p.1-13,2022. https://doi.org/10.3389/fphys.2022.1046204
  » https://doi.org/https://doi.org/10.3389/fphys.2022.1046204
• LOR, V.S.; OLSZEWSKI, N.E. GA signalling and cross-talk with other signalling pathways. Essays in Biochemistry, v.58, p.49-60,2015. https://doi.org/10.1042/bse0580049
  » https://doi.org/https://doi.org/10.1042/bse0580049
• MA, L.; WU, J.; QI, W.; COULTER, J.A.; FANG, Y.; LI, X.; LIU, L.; JIN, J.; NIU, Z.; J. YUE, J.;SUN, W. Screening and verification of reference genes for analysis of gene expression in winter rapeseed (Brassica rapa L.) under abiotic stress. PLoS One , v.15, n.9, 2020. https://doi.org/10.1371/journal.pone.0236577
  » https://doi.org/https://doi.org/10.1371/journal.pone.0236577
• METWALLY, R.A.; ABDELHAMEED, R.E.; SOLIMANA, S.A.; AL-BADWY, H. Potential use of beneficial fungal microorganisms and C-phycocyanin extract for enhancing seed germination, seedling growth and biochemical traits of Solanum lycopersicum L BMC Microbiology, v.22, n.1, p.1-17,2022. https://doi.org/10.1186/s12866-022-02509-x
  » https://doi.org/https://doi.org/10.1186/s12866-022-02509-x
• MIGOCKA, M.; PAPIERNIAK, A. Identification of suitable reference genes for studying gene expression in cucumber plants subjected to abiotic stress and growth regulators. Molecular Breeding, v.28, n.3, p.343-357,2011. https://link.springer.com/article/10.1007/s11032-010-9487-0
  » https://link.springer.com/article/10.1007/s11032-010-9487-0
• MINAMI, A.; YANO, K.; GAMUYAO, R.; NAGAI, K.; KUROHA, T.; AYANO, M.; NAKAMORI, M.; KOIKE, M.; KONDO, Y.; NIIMI, Y.; KUWATA, K.; SUZUKI, T.; HIGASHIYAMA, T.; TAKEBAYASHI, Y.; KOJIMA, M.; SAKAKIBARA, H.; TOYODA, A.; FUJIYAMA, A.; KURATA, N.; ASHIKARIS, M.; REUSCHER, S. Time-course transcriptomics analysis reveals key responses of submerged deepwater rice to flooding. Plant Physiology, v.176, n.4, p.3081-3102,2018. https://doi.org/10.1104/pp.17.00858
  » https://doi.org/https://doi.org/10.1104/pp.17.00858
• MOHANTY, S.P.; NAYAK, D.K.; SANGHAMITRA, P.; BARIK, S.R.; PANDIT, E.; BEHERA, A.; PANI, D.R.; MOHAPATRA, S.; RAJ, K.R.R.; PRADHAN, K.C.; SAHOO, C.R.; MOHANTY, M.R.; BEHERA, C.; PANDA, A.K.; JENA, B.K.; BEHERA, L.; DASHS, P.K.; PRADHAN, K. Mapping the genomic regions controlling germination rate and early seedling growth parameters in rice. Genes, v.14, n.4, p.1-27,2023. https://doi.org/10.3390/genes14040902
  » https://doi.org/https://doi.org/10.3390/genes14040902
• PELISSARI, F.; PEREIRA, W.V.S.; PEREIRA, T.M.; SANTOS, H.O.; FARIA, J.M.R.; JOSE, A.C. Effect of PEG and ABA on desiccation tolerance and storage of Magnolia ovata (A.St.-Hil.) Spreng. seeds. Journal of Seed Science, v.44, n.e202244002, 2022. http://dx.doi.org/10.1590/2317-1545v44256114
  » https://doi.org/http://dx.doi.org/10.1590/2317-1545v44256114
• PERREAU, F.; FREY, A.; EFFROY-CUZZI, D.; SAVANE, P.; BERGER, A.; GISSOTA, L.; MARION-POLL, A. ABSCISIC ACID-DEFICIENT4 has an essential function in both cis-violaxanthin and cis-neoxanthin synthesis. Plant Physiology , v.184, n.3, p.1303-1316,2020. https://doi.org/10.1104/pp.20.00947
  » https://doi.org/https://doi.org/10.1104/pp.20.00947
• PFAFFL, M.W.; TICHOPAD, A.; PRGOMETT, C.; NEUVIANS, P. Determination of stable housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper--Excel-based tool using pair-wise correlations. Biotechnology Letters, v.26, n.6, p.509-515,2004. https://doi.org/10.1023/B:BILE.0000019559.84305.47
  » https://doi.org/https://doi.org/10.1023/B:BILE.0000019559.84305.47
• PINTO, D.B.B.; FERREIRA, E.; HENNING, F.A.; AMARAL, H.F.; HUNGRIA, M.; NOGUEIRA, M.A. Recovery of Bradyrhizobium cells and effects on the physiological quality of soybean seeds sown in dry soil. Journal of Seed Science , v.45, n.e202345001, 2023. https://doi.org/10.1590/2317-1545v45259694
  » https://doi.org/https://doi.org/10.1590/2317-1545v45259694
• RABIEYAN, E.; BIHAMTA, M.R.; MOGHADDAM, M.E.; MOHAMMADIH, V.; ALIPOUR, H. Genome-wide association mapping and genomic prediction for pre-harvest sprouting resistance, low α-amylase and seed color in Iranian bread wheat. BMC Plant Biology, v.22, n.1, p.1-23,2022. https://doi.org/10.1186/s12870-022-03628-3
  » https://doi.org/https://doi.org/10.1186/s12870-022-03628-3
• RUDUŚ, I.; KĘPCZYŃSKI, J. Reference gene selection for molecular studies of dormancy in wild oat (Avena fatua L.) caryopses by RT-qPCR method. PLoS One , v.13, n.2, p.1-21,2018. https://doi.org/10.1371/journal.pone.0192343
  » https://doi.org/https://doi.org/10.1371/journal.pone.0192343
• SEN, M.K.; HAMOUZOVá, K; KOŠNAROVá, P.; ROY, A.; SOUKUP, J.S. Identification of the most suitable reference gene for gene expression studies with development and abiotic stress response in Bromus sterilis Scientific Reports , v.11, n.1, p.1-10,2021. https://doi.org/10.1038/s41598-021-92780-1
  » https://doi.org/https://doi.org/10.1038/s41598-021-92780-1
• SEO, M.; KOSHIBA, T. Complex regulation of ABA biosynthesis in plants. Trends in Plant Science , v.7, n.1, p.8-41,2002. https://doi.org/10.1016/s1360-1385(01)02187-2
  » https://doi.org/https://doi.org/10.1016/s1360-1385(01)02187-2
• SHI, J.; LIU, M.; SHI, J.; ZHENG, G.; WANG, Y.; WANG, J.; CHEN, Y.; LU, C.; YIN, W. Reference gene selection for qPCR in Ammopiptanthus mongolicus under abiotic stresses and expression analysis of seven ROS-scavenging enzyme genes. Plant Cell Reports , v.31, n.7, p.1245-1254,2012. https://doi.org/10.1007/s00299-012-1245-9
  » https://doi.org/https://doi.org/10.1007/s00299-012-1245-9
• ŠKILJAICA, A.; JAGIĆ, M.; VUK, T.; D. LELJAK LEVANIĆ, N.; BAUER, N.; MARKULIN, L. Evaluation of reference genes for RT-qPCR gene expression analysis in Arabidopsis thaliana exposed to elevated temperatures. Plant Biology, v.24, n.2, p.367-379,2022. https://doi.org/10.1111/plb.13382
  » https://doi.org/https://doi.org/10.1111/plb.13382
• SOHN, S.I.; PANDIAN, S.; KUMAR, T.S.; ZOCLANCLOUNON, Y.A.B.; MUTHURAMALINGAM, P.; SHILPHA, J.; SATISH, L.; RAMESH, M. Seed dormancy and pre-harvest sprouting in rice-an updated overview. International Journal of Molecular Sciences, v.22, n.21, p.1-22,2021. https://doi.org/10.3390/ijms222111804
  » https://doi.org/https://doi.org/10.3390/ijms222111804
• SUDHAKAR-REDDY, P.; DHAWARE, M.G.; SRINIVAS REDDY, D.; PRADEEP REDDY, B.; DIVYA, K.; SHARMA, K.K.; BHATNAGAR-MATHUR, P. Comprehensive evaluation of candidate reference genes for real-time quantitative PCR (RT-qPCR) data normalization in nutri-cereal finger millet [Eleusine Coracana (L.)]. PLoS One , v.13, n.10, p.1-17,2018. https://doi.org/10.1371/journal.pone.0205668
  » https://doi.org/https://doi.org/10.1371/journal.pone.0205668
• TAI, L.; WANG, H.J.; XU, X.J.; SUN, W.H.; JU, L.; LIU, W.T.; LI, W.Q.; SUNK, J.; CHEN, M. Pre-harvest sprouting in cereals: genetic and biochemical mechanisms. Journal of Experimental Botany, v.72, n.8, p.2857-2876,2021. https://doi.org/10.1093/jxb/erab024
  » https://doi.org/https://doi.org/10.1093/jxb/erab024
• TAKI, F.A.; ZHANG, B. Determination of reliable reference genes for multi-generational gene expression analysis on C. elegans exposed to abused drug nicotine. Psychopharmacology, v.230, n.1, p.77-88,2013. https://doi.org/10.1007/s00213-013-3139-0
  » https://doi.org/https://doi.org/10.1007/s00213-013-3139-0
• TANG, W.; GUO, H.; YIN, J.; DING, X.; XU, X.; WANG, T.; YANG, C.; XIONG, W.; ZHONG, S.; TAO, Q.; SUN, J. Germination ecology of Chenopodium album L. and implications for weed management. PLoS One , v.17, n.10, p.1-23,2022. https://doi.org/10.1371/journal.pone.0276176
  » https://doi.org/https://doi.org/10.1371/journal.pone.0276176
• TANG, X.; ZHANG, N.; SI, H.; CALDERóN-URREA. A. Selection and validation of reference genes for RT-qPCR analysis in potato under abiotic stress. Plant Methods, v.13, p.85-93,2017. https://doi.org/10.1186/s13007-017-0238-7
  » https://doi.org/https://doi.org/10.1186/s13007-017-0238-7
• ULLAH, A.; SADAF, S.; ULLAH, S.; ALSHAYA, H.; OKLA, M.K.; ALWASEL, Y.A.; TARIQ, A. Using halothermal time model to describe barley (Hordeumvulgare L.) seed germination response to water potential and temperature. Life, v.12, n.2, p.1-15,2022. https://doi.org/10.3390/life12020209
  » https://doi.org/https://doi.org/10.3390/life12020209
• VANDESOMPELE, J.; PRETER, K.; PATTYN, F.; POPPE, B.; VAN ROY, N.; PAEPEF. SPELEMAN, A. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biology, v.3, n.7, p.1-12,2002. https://doi.org/10.1186/gb-2002-3-7-research0034
  » https://doi.org/https://doi.org/10.1186/gb-2002-3-7-research0034
• WALLING, J.G.; ZALAPAM, L.A.; VINJE, A. Evaluation and selection of internal reference genes from two- and six-row U.S. malting barley varieties throughout micromalting for use in RT-qPCR. PLoS One , v.13, n.5, p.1-11,2018. https://doi.org/10.1371/journal.pone.0196966
  » https://doi.org/https://doi.org/10.1371/journal.pone.0196966
• WANG, G.H.; LIANG, C.C.; LI, B.Z.; DU, X.Z.; ZHANG, W.Z.; CHENGL, G.; ZAN, S. Screening and validation of reference genes for qRT-PCR of bovine skeletal muscle-derived satellite cells. Scientific Reports , v.12, n.1, p.5653, 2022a. https://doi.org/10.1038/s41598-022-09476-3
  » https://doi.org/https://doi.org/10.1038/s41598-022-09476-3
• WANG, M.; WANG, Z.; WEI, S.; XIE, J.; HUANG, J.; LI, D.; HU, W.; LI, H.; TANG, H. Identification of RT-qPCR reference genes suitable for gene function studies in the pitaya canker disease pathogen Neoscytalidium dimidiatum Scientific Reports , v.12, n.1, p.1-9, 2022b. https://doi.org/10.1038/s41598-022-27041-w
  » https://doi.org/https://doi.org/10.1038/s41598-022-27041-w
• WANG, P.; LU, S.; ZHANG, X.; HYDEN, B.; QIN, L.; LIU, L.; BAI, Y.; HAN, Y.; WEN, Z.; XU, J.; CAO, H.; CHEN, H. Double NCED isozymes control ABA biosynthesis for ripening and senescent regulation in peach fruits. Plant Science , v.304, p.110739,2021. https://doi.org/10.1016/j.plantsci.2020.110739
  » https://doi.org/https://doi.org/10.1016/j.plantsci.2020.110739
• WANG, Q.; LIN, Q.; WU, T.; DUAN, E.; HUANG, Y.; YANG, C.; MOU, C.; LAN, J.; ZHOU, C.; XIE, K.; LIU, X.; ZHANG, X.; GUO, X.; WANG, J.; JIANG, L.; WAN, J. OsDOG1L-3 regulates seed dormancy through the abscisic acid pathway in rice. Plant Science , v.298, p.1-11,2020. https://doi.org/10.1016/j.plantsci.2020.110570
  » https://doi.org/https://doi.org/10.1016/j.plantsci.2020.110570
• WANG, R.; ZHOU, Z.; XIONG, M.; DU, M.; LIN, X.; LIU, C.; LU, M.; LIU, Z.; CHANG, Y.; LIU, E. Mining salt tolerance SNP Loci and prediction of candidate genes in the rice bud stage by genome-wide association analysis. Plants, v.12, n.11, p.1-21,2023. https://doi.org/10.3390/plants12112163
  » https://doi.org/https://doi.org/10.3390/plants12112163
• WANG, Z.; XU, J.; LIU, Y.; CHEN, J.; LIN, H.; HUANG, Y.; BIAN, X.; ZHAO, Y. Selection and validation of appropriate reference genes for real-time quantitative PCR analysis in Momordica charantia Phytochemistry, v.164, p.1-11,2019. https://doi.org/10.1016/j.phytochem.2019.04.010
  » https://doi.org/https://doi.org/10.1016/j.phytochem.2019.04.010
• WU, Y.; ZHANG, C.; YANG, H.; LYU, L.; LI, W.; WU, W. Selection and validation of candidate reference genes for gene expression analysis by RT-qPCR in Rubus International Journal of Molecular Sciences , v.22, n.19, 2021. https://doi.org/10.3390/ijms221910533
  » https://doi.org/https://doi.org/10.3390/ijms221910533
• WANG, Z.; XU, J.; LIU; Y.; CHEN, J.; LIN, H.; HUANG, Y.; BIAN, X.; ZHAO, Y. Selection and validation of appropriate reference genes for real-time quantitative PCR analysis in Momordica charantia Phytochemistry , p.1-11,2019. https://doi.org/10.1016/j.phytochem.2019.04.010
  » https://doi.org/https://doi.org/10.1016/j.phytochem.2019.04.010
• ZHU, X.; WANG, B.; WANG, X.; WEI, X. Screening of stable internal reference gene of Quinoa under hormone treatment and abiotic stress. Physiology and Molecular Biology of Plants , v.27, n.11, p.2459-2470, 2021. https://doi.org/10.1007/s12298-021-01094-z
  » https://doi.org/https://doi.org/10.1007/s12298-021-01094-z

Publication Dates

• Publication in this collection
  10 Nov 2023
• Date of issue
  2023

History

• Received
  08 May 2023
• Accepted
  11 Oct 2023