Selection of reference genes for quantitative real-time PCR analysis in exogenous hormone-treated Lycoris aurea Authors: Wei Zhao, Ying Tian, Yi Wang, Yun Wu, Ziming Ren, Kehu Li Published: 2024 The bulbs of Lycoris aurea Herb can produce unique alkaloids, which have been shown to have antiviral, anti-tumor, anti-malarial, and immunostimulatory properties and are effective in the symptomatic treatment of Alzheimer’s disease [1]. As a consequence, the demand for Lycoris bulbs has dramatically increased, especially for L. aurea, which is native to southern China and is highly important for increasing bulb yield [2]. Previous research in this area reported that the use of exogenous hormones is a way to optimize the artificial reproduction of their bulbs [3]. Therefore, studying the molecular mechanism of bulb growth and development is highly important. Quantitative real-time PCR (qRT-PCR) is widely used in gene expression analysis to explore gene functions that regulate plant growth and development because of its high specificity, accuracy, sensitivity and efficiency, and selecting reference genes correctly is crucial for obtaining proper results and interpretations via qRT-PCR analysis [4]. The selection and validation of reference genes of some Lycoris species in different floral development stages and tissues have been carried out, but under some abnormal conditions, including abiotic stress and hormone treatments, have not been assessed [ 5, 6]. In this study, we aimed to identify suitable reference genes for L. aurea under seven hormone treatments. A total of seven candidate reference genes were selected for investigation. The expressions of these candidate genes were measured by qRT-PCR. The expression stability of candidate genes was comprehensively evaluated by four different statistical algorithms, including geNorm, BestKeeper, NormFinder and RefFinder [ 7– 10]. The bulbs of L. aurea with similar diameters growing in our greenhouse were selected. For hormone treatments, the bulbs were exposed to 25 mg/L 6-BA (6-benzylaminopurine), 200 mg/L ethrel, 1 mg/L 2,4-EBR (2,4-epi-brassinolide), 100 mg/L MeJA (methyl jasmonate), 100 mg/L GA (gibberellic acid), 0.384 mg/L (1 μM) SL (strigolactone) or 100 mg/L ABA (abscisic acid) for 0, 4, 18, or 48 h. 6-BA (containing NaOH), ethrel, 2,4-EBR (containing absolute ethanol), MeJA (containing absolute ethanol), GA (containing absolute ethanol), SL (containing propanol) and ABA (containing absolute ethanol) were all dissolved in distilled water. Moreover, the bulbs were placed in a plant growth chamber (RXZ-308B; Ningbo Jiangnan Instrument Factoy, Ningbo, China) with a mean temperature of 25±5°C, a relative humidity of 95.0%, and a day/night cycle of 16/8 h. Then, the bulbs were sampled at different time points (0, 4, 18, and 48 h after treatment) for each treatment, with three biological replicates, and snap-frozen in liquid nitrogen. All the samples were subsequently stored at –80°C for subsequent experiments. Total RNA was extracted from the samples using the MolPure ® Plant RNA kit (Yeasen Biotechnology, Shanghai, China). K5600C (KAIAO Technology Development Co., Ltd., Beijing, China) and 1% agarose gel electrophoresis were used to measure the concentration and quality of the RNA. The first strand of cDNA was synthesized using Hifair ® III 1st Strand cDNA Synthesis SuperMix for qPCR (gDNA digester plus) (Yeasen Biotechnology). Seven common housekeeping genes, including activating enhancer binding protein 4 ( AP4), ubiquitin 4 ( UBQ4), β-tubulin ( TUB2), eukaryotic translation initiation factor ( EIF), ubiquitin conjugating enzyme E2 ( UBCE), TATA box binding protein-associated factor ( TBP), and UBC core domain-containing protein ( UBC24), were selected as candidate reference genes, and all specific primers were designed with Primer Premier 5.0 software ( Table 1). All cDNA samples were mixed equally as a template and diluted in 4 gradients 3 times for each gradient. The specificity of the primers was visualized by 1% agarose gel electrophoresis of the PCR products. The standard curves of each gene were drawn from the results of the qRT-PCR reaction. The amplification efficiency (E) and correlation coefficient (R 2) of each primer were calculated. Table 1 Primers and amplification characteristics of candidate reference genes Gene name Gene ID Primer sequence (5′→3′) Product (bp) Tm (°C) E (%) R2 AP4 Unigene48898_All F: GACCAACTGGAGAGAGTGCTTGAG 120 61.6 110.1 0.999 R: CGTGTTGCCTGCCTCTATACCTAC 61.3 UBQ4 Unigene47709_All F: TAGTCTTGCGTCTTCGTGGA 198 57.4 96.7 1.000 R: AATCAGCAAGGGTCCTCCCA 57.3 TUB2 Unigene7296_All F: GAGGCGGAGAATTGTGATTGCTTG 150 61.6 98.3 0.997 R: GGACGGGAAGACGGAGAAAGTG 60.0 EIF Unigene95331_All F: CCTAGTTTGGCTGAGTCTTCTT 116 57.0 108.6 0.993 R: GACCACTGCCGATAGTCATATT 56.0 UBCE Unigene45273_All F: ATGGCAAGGTGGATGAGAAAGGG 113 59.9 95.4 0.998 R: CATTGACTGCCGTTGAGATGTGAG 61.2 TBP Unigene54738_All F: CCTAGACCTGGAGAGGTTATTTG 115 56.8 108.9 0.994 R: CTCCCTCTAGTCCTCCTGATATT 57.4 UBC24 Unigene96_All F: GAGGCTGGCTACGACAAACT 127 58.0 95.3 0.999 R: AATGCTTAGGAGGCTTGCGT 58.8 Table 1 Primers and amplification characteristics of candidate reference genes Gene name Gene ID Primer sequence (5′→3′) Product (bp) Tm (°C) E (%) R2 AP4 Unigene48898_All F: GACCAACTGGAGAGAGTGCTTGAG 120 61.6 110.1 0.999 R: CGTGTTGCCTGCCTCTATACCTAC 61.3 UBQ4 Unigene47709_All F: TAGTCTTGCGTCTTCGTGGA 198 57.4 96.7 1.000 R: AATCAGCAAGGGTCCTCCCA 57.3 TUB2 Unigene7296_All F: GAGGCGGAGAATTGTGATTGCTTG 150 61.6 98.3 0.997 R: GGACGGGAAGACGGAGAAAGTG 60.0 EIF Unigene95331_All F: CCTAGTTTGGCTGAGTCTTCTT 116 57.0 108.6 0.993 R: GACCACTGCCGATAGTCATATT 56.0 UBCE Unigene45273_All F: ATGGCAAGGTGGATGAGAAAGGG 113 59.9 95.4 0.998 R: CATTGACTGCCGTTGAGATGTGAG 61.2 TBP Unigene54738_All F: CCTAGACCTGGAGAGGTTATTTG 115 56.8 108.9 0.994 R: CTCCCTCTAGTCCTCCTGATATT 57.4 UBC24 Unigene96_All F: GAGGCTGGCTACGACAAACT 127 58.0 95.3 0.999 R: AATGCTTAGGAGGCTTGCGT 58.8 The qRT-PCR experiments of the reference genes were performed with Hieff ® qPCR SYBR Green Master Mix (No Rox) (Yeasen Biotechnology) on the Real-Time system from Bio-Rad Laboratories (Hercules, USA). The reaction system (20 μL) included 10 μL of 1× Hieff ® qPCR SYBR Green Master Mix (No Rox), 2 μL of cDNA, 0.4 μL of each primer (10 μM) and 7.2 μL of ddH 2O. The reaction program was as follows: 95°C for 5 min, 40 cycles of 95°C for 10 s and 60°C for 30 s, and a melting curve was generated at 60–95°C. Each qRT-PCR was performed in triplicate. The stability of the seven candidate reference genes was evaluated by four statistical software programs, including geNorm, BestKeeper, NormFinder and RefFinder [ 7– 10]. For the geNorm and NormFinder analyses, according to the formula 2 –ΔCt (ΔCt = each corresponding Ct value–the minimum Ct value), all of the raw cycle threshold (Ct) values were transformed into relative expression levels, which were then imported into two software programs (geNorm and NormFinder) to evaluate gene expression stability. The stability of reference gene expression within each treatment, as well as across all treatments (all sample subsets), was compared. The analysis of the qRT-PCR melting curves revealed that each applied pair of primers generated a single amplification peak ( Figure 1).Furthermore, the amplification efficiencies (E) of all the candidate reference genes ranged from 95.3% to 110.1%, and the linear R 2 values (regression coefficients) varied from 0.9993 to 1 ( Table 1). These results indicated the high specificity and suitability of the seven reference genes for qRT-PCR. Figure 1 Melting curves of seven candidate reference genes The lines inside the box represent the median values; the lower/upper boundaries of the box indicate the first/third quartiles, respectively. The whiskers represent the minimum and maximum values. The Ct refers to the number of amplification cycles at which the fluorescent signal of the amplified product reaches the set fluorescence threshold during qRT-PCR amplification process. A lower Ct value reflects a higher expression level. According to the experimental results, the Ct values of all the candidate reference genes ranged from 22.37 to 29.88. The mean Ct values were mostly between 23.47 and 29.18 across all the tested samples. UBQ4, which presented the lowest average Ct value of 23.47 ± 0.98 (mean ± SD), was the most abundantly expressed gene, whereas with a mean Ct value of 29.18 ± 0.54, and TBP presented the lowest expression level ( Figure 2A). Figure 2 Distribution of Ct values (A), pairwise variation (V n/V n +1) (B) and average expression stability values (M) (C) of seven candidate reference genes For geNorm, the Ct values cannot be imported into the software to assess the best reference genes directly; rather, they need to be transformed to relative expression levels via a formula and then calculated. GeNorm judges the stability of gene expression on the basis of the M value. The more stable the gene expression was, the lower the M value was, but the M value was suggested to be less than 1.5. The M values of all the samples are shown in Figure 2C and Table 2. UBQ4 was the most unstable gene across the 5 treatments, including the SL (M = 1.049), GA (M = 1.216), MeJ (M = 0.815), 2,4-EBR (M = 0.752), and ABA (M = 0.950) treatments and all the sample subsets (M = 1.120), whereas EIF was the worst gene in the other two sets. In the SL, ethrel, and 2,4-EB treatments and all sample subsets, UBCE and TBP had the lowest M values of 0.398, 0.423, 0.270 and 0.652, respectively. TUB2 and UBCE (M = 0.569) presented the greatest stability in the GA treatment, whereas UBC24 and TBP (M = 0.648) presented the greatest stability in the 6-BA treatment. AP4 and TBP (M = 0.539) were identified as excellent reference genes in the MeJ treatment, whereas UBCE and UBC24 (M = 0.609) were the best in the ABA treatment. By the geNorm algorithm, the pairwise variation (V n/V n+1) between the reference genes can also be calculated to determine the appropriate number of reference genes for accurate normalization. The criterion was that if the V n/V n+1 values were greater than the threshold of 0.15, then the optimal number of reference genes was No. (n+1). As shown in Figure 2B, the pairwise variation (V 2/V 3) of all the sets was greater than 1.5, and except the 2,4-EBR treatment, the pairwise variation (V 3/V 4) of the other treatments was greater than 1.5. This finding indicates that accurate normalization of tested gene expression can be achieved via three reference genes in the 2,4-EBR; however, the V3/4 values in other sample sets were greater than 0.15, suggesting that more reference genes are needed for accurate normalization of gene expression. Table 2 The expression stability of the seven candidate reference genes in different hormones treatments calculated by the geNorm, NormFinder, Bestkeeper and RefFinder Software Rank SL GA 6-BA MeJ Ethrel 2,4-EBR ABA All samples geNorm Gene M Gene M Gene M Gene M Gene M Gene M Gene M Gene M 1 UBCE 0.398 TUB2 0.569 UBC24 0.648 AP4 0.539 UBCE 0.423 UBCE 0.270 UBCE 0.609 UBCE 0.652 2 TBP 0.398 UBCE 0.569 TBP 0.648 TBP 0.539 TBP 0.423 TBP 0.270 UBC24 0.609 TBP 0.652 3 EIF 0.571 TBP 0.624 UBQ4 0.698 TUB2 0.563 TUB2 0.616 UBC24 0.432 AP4 0.660 AP4 0.765 4 UBC24 0.656 AP4 0.783 AP4 0.759 EIF 0.628 UBQ4 0.724 AP4 0.459 EIF 0.708 UBC24 0.860 5 AP4 0.752 UBC24 0.882 TUB2 0.800 UBCE 0.662 UBC24 0.844 TUB2 0.554 TBP 0.745 TUB2 0.968 6 TUB2 0.903 EIF 1.082 UBCE 0.848 UBC24 0.746 AP4 0.915 EIF 0.665 TUB2 0.778 EIF 1.050 7 UBQ4 1.049 UBQ4 1.216 EIF 0.889 UBQ4 0.815 EIF 1.062 UBQ4 0.752 UBQ4 0.950 UBQ4 1.120 NormFinder Gene SV Gene SV Gene SV Gene SV Gene SV Gene SV Gene SV Gene SV 1 TBP 0.356 UBCE 0.191 TUB2 0.422 AP4 0.203 UBCE 0.227 UBCE 0.135 AP4 0.518 UBCE 0.220 2 UBCE 0.387 TUB2 0.373 UBC24 0.490 TBP 0.372 TBP 0.326 TBP 0.326 UBC24 0.519 TBP 0.392 3 UBC24 0.639 TBP 0.629 UBQ4 0.630 TUB2 0.445 TUB2 0.608 TUB2 0.460 EIF 0.522 UBC24 0.647 4 UBQ4 0.660 AP4 0.888 TBP 0.659 EIF 0.600 UBC24 0.769 UBC24 0.504 UBCE 0.550 EIF 0.899 5 EIF 0.683 UBC24 1.001 UBCE 0.682 UBCE 0.647 UBQ4 0.772 AP4 0.586 TUB2 0.612 TUB2 1.493 6 AP4 0.835 EIF 1.231 EIF 0.814 UBC24 0.739 AP4 0.829 EIF 0.756 TBP 0.666 AP4 1.908 7 TUB2 0.915 UBQ4 1.368 AP4 1.085 UBQ4 0.882 EIF 1.457 UBQ4 0.860 UBQ4 1.221 UBQ4 2.500 Bestkeeper Gene SD CV Gene SD CV Gene SD CV Gene SD CV Gene SD CV Gene SD CV Gene SD CV Gene SD CV 1 AP4 0.716 2.556 UBCE 0.437 1.580 TUB2 0.590 2.588 UBQ4 0.745 3.329 UBCE 0.923 3.230 AP4 1.111 3.817 TBP 1.031 3.454 TBP 0.727 2.493 2 UBC24 0.815 2.922 TUB2 0.489 1.901 UBC24 0.630 2.356 AP4 0.917 3.230 TBP 1.030 3.482 UBQ4 1.276 5.151 EIF 1.115 3.732 UBCE 0.765 2.700 3 UBQ4 0.856 3.696 TBP 0.563 1.982 TBP 0.688 2.349 TUB2 0.953 3.535 AP4 1.086 3.710 UBC24 1.299 4.629 AP4 1.129 3.851 UBC24 0.884 3.219 4 UBCE 1.064 3.754 AP4 0.811 2.862 UBQ4 0.779 2.734 TBP 1.053 3.673 UBC24 1.160 4.272 TUB2 1.339 4.807 UBCE 1.337 4.617 EIF 1.030 3.551 5 TBP 1.085 3.758 EIF 0.893 3.024 UBCE 0.795 2.851 UBC24 1.114 4.157 TUB2 1.165 4.233 EIF 1.348 4.576 TUB2 1.340 4.656 AP4 1.531 5.479 6 TUB2 1.153 4.440 UBC24 1.077 3.973 AP4 0.856 3.696 UBCE 1.180 4.204 UBQ4 1.286 5.584 UBCE 1.349 4.660 UBC24 1.344 4.731 TUB2 1.599 6.030 7 EIF 1.163 4.109 UBQ4 1.171 5.033 EIF 0.996 3.463 EIF 1.324 4.687 EIF 1.739 6.049 TBP 1.370 4.630 UBQ4 1.635 6.570 UBQ4 1.655 6.816 RefFinder Gene RV Gene RV Gene RV Gene RV Gene RV Gene RV Gene RV Gene RV 1 UBCE 2.450 UBCE 1.570 TUB2 1.860 AP4 1.320 UBCE 1.190 UBCE 1.630 AP4 1.320 UBCE 1.570 2 TBP 2.550 TUB2 2.280 UBC24 2.110 TBP 3.410 TBP 2.000 TBP 2.380 TBP 2.740 TBP 1.570 3 UBC24 4.160 TBP 4.230 UBQ4 5.690 TUB2 4.000 TUB2 3.830 AP4 3.660 EIF 2.910 UBC24 5.000 4 AP4 4.880 AP4 6.240 TBP 5.730 UBQ4 5.200 UBQ4 6.240 UBC24 3.940 UBC24 3.740 EIF 6.480 5 UBQ4 5.630 UBC24 7.970 UBCE 7.480 UBCE 6.450 UBC24 6.960 TUB2 4.950 UBCE 4.470 TUB2 8.240 6 EIF 6.770 EIF 8.740 EIF 9.240 EIF 6.470 AP4 7.330 UBQ4 6.690 TUB2 6.000 AP4 8.740 7 TUB2 9.740 UBQ4 10.000 AP4 9.740 UBC24 7.440 EIF 10.000 EIF 8.130 UBQ4 10.000 UBQ4 10.000 Table 2 The expression stability of the seven candidate reference genes in different hormones treatments calculated by the geNorm, NormFinder, Bestkeeper and RefFinder Software Rank SL GA 6-BA MeJ Ethrel 2,4-EBR ABA All samples geNorm Gene M Gene M Gene M Gene M Gene M Gene M Gene M Gene M 1 UBCE 0.398 TUB2 0.569 UBC24 0.648 AP4 0.539 UBCE 0.423 UBCE 0.270 UBCE 0.609 UBCE 0.652 2 TBP 0.398 UBCE 0.569 TBP 0.648 TBP 0.539 TBP 0.423 TBP 0.270 UBC24 0.609 TBP 0.652 3 EIF 0.571 TBP 0.624 UBQ4 0.698 TUB2 0.563 TUB2 0.616 UBC24 0.432 AP4 0.660 AP4 0.765 4 UBC24 0.656 AP4 0.783 AP4 0.759 EIF 0.628 UBQ4 0.724 AP4 0.459 EIF 0.708 UBC24 0.860 5 AP4 0.752 UBC24 0.882 TUB2 0.800 UBCE 0.662 UBC24 0.844 TUB2 0.554 TBP 0.745 TUB2 0.968 6 TUB2 0.903 EIF 1.082 UBCE 0.848 UBC24 0.746 AP4 0.915 EIF 0.665 TUB2 0.778 EIF 1.050 7 UBQ4 1.049 UBQ4 1.216 EIF 0.889 UBQ4 0.815 EIF 1.062 UBQ4 0.752 UBQ4 0.950 UBQ4 1.120 NormFinder Gene SV Gene SV Gene SV Gene SV Gene SV Gene SV Gene SV Gene SV 1 TBP 0.356 UBCE 0.191 TUB2 0.422 AP4 0.203 UBCE 0.227 UBCE 0.135 AP4 0.518 UBCE 0.220 2 UBCE 0.387 TUB2 0.373 UBC24 0.490 TBP 0.372 TBP 0.326 TBP 0.326 UBC24 0.519 TBP 0.392 3 UBC24 0.639 TBP 0.629 UBQ4 0.630 TUB2 0.445 TUB2 0.608 TUB2 0.460 EIF 0.522 UBC24 0.647 4 UBQ4 0.660 AP4 0.888 TBP 0.659 EIF 0.600 UBC24 0.769 UBC24 0.504 UBCE 0.550 EIF 0.899 5 EIF 0.683 UBC24 1.001 UBCE 0.682 UBCE 0.647 UBQ4 0.772 AP4 0.586 TUB2 0.612 TUB2 1.493 6 AP4 0.835 EIF 1.231 EIF 0.814 UBC24 0.739 AP4 0.829 EIF 0.756 TBP 0.666 AP4 1.908 7 TUB2 0.915 UBQ4 1.368 AP4 1.085 UBQ4 0.882 EIF 1.457 UBQ4 0.860 UBQ4 1.221 UBQ4 2.500 Bestkeeper Gene SD CV Gene SD CV Gene SD CV Gene SD CV Gene SD CV Gene SD CV Gene SD CV Gene SD CV 1 AP4 0.716 2.556 UBCE 0.437 1.580 TUB2 0.590 2.588 UBQ4 0.745 3.329 UBCE 0.923 3.230 AP4 1.111 3.817 TBP 1.031 3.454 TBP 0.727 2.493 2 UBC24 0.815 2.922 TUB2 0.489 1.901 UBC24 0.630 2.356 AP4 0.917 3.230 TBP 1.030 3.482 UBQ4 1.276 5.151 EIF 1.115 3.732 UBCE 0.765 2.700 3 UBQ4 0.856 3.696 TBP 0.563 1.982 TBP 0.688 2.349 TUB2 0.953 3.535 AP4 1.086 3.710 UBC24 1.299 4.629 AP4 1.129 3.851 UBC24 0.884 3.219 4 UBCE 1.064 3.754 AP4 0.811 2.862 UBQ4 0.779 2.734 TBP 1.053 3.673 UBC24 1.160 4.272 TUB2 1.339 4.807 UBCE 1.337 4.617 EIF 1.030 3.551 5 TBP 1.085 3.758 EIF 0.893 3.024 UBCE 0.795 2.851 UBC24 1.114 4.157 TUB2 1.165 4.233 EIF 1.348 4.576 TUB2 1.340 4.656 AP4 1.531 5.479 6 TUB2 1.153 4.440 UBC24 1.077 3.973 AP4 0.856 3.696 UBCE 1.180 4.204 UBQ4 1.286 5.584 UBCE 1.349 4.660 UBC24 1.344 4.731 TUB2 1.599 6.030 7 EIF 1.163 4.109 UBQ4 1.171 5.033 EIF 0.996 3.463 EIF 1.324 4.687 EIF 1.739 6.049 TBP 1.370 4.630 UBQ4 1.635 6.570 UBQ4 1.655 6.816 RefFinder Gene RV Gene RV Gene RV Gene RV Gene RV Gene RV Gene RV Gene RV 1 UBCE 2.450 UBCE 1.570 TUB2 1.860 AP4 1.320 UBCE 1.190 UBCE 1.630 AP4 1.320 UBCE 1.570 2 TBP 2.550 TUB2 2.280 UBC24 2.110 TBP 3.410 TBP 2.000 TBP 2.380 TBP 2.740 TBP 1.570 3 UBC24 4.160 TBP 4.230 UBQ4 5.690 TUB2 4.000 TUB2 3.830 AP4 3.660 EIF 2.910 UBC24 5.000 4 AP4 4.880 AP4 6.240 TBP 5.730 UBQ4 5.200 UBQ4 6.240 UBC24 3.940 UBC24 3.740 EIF 6.480 5 UBQ4 5.630 UBC24 7.970 UBCE 7.480 UBCE 6.450 UBC24 6.960 TUB2 4.950 UBCE 4.470 TUB2 8.240 6 EIF 6.770 EIF 8.740 EIF 9.240 EIF 6.470 AP4 7.330 UBQ4 6.690 TUB2 6.000 AP4 8.740 7 TUB2 9.740 UBQ4 10.000 AP4 9.740 UBC24 7.440 EIF 10.000 EIF 8.130 UBQ4 10.000 UBQ4 10.000 The NormFinder algorithm was used to analyze the expression stability of the candidate genes further. All SV values were calculated via the NormFinder program ( Table 2). According to the SV values, TBP (0.356) was the most stable gene in the SL subset, whereas TUB2 (0.915) was the least stable gene. Among the subset of 6-BA, TUB2 (0.422) presented the highest stability, whereas AP4 (1.085) was the most unstable gene. AP4 (0.203/0.518) was the best in the subset of MeJ and ABA, whereas UBQ4 (0.882/1.221) was the worst. In addition, UBCE (0.191/0.227/0.135/0.220) was the most stably expressed gene in the sets of GA, Ethrel, 2,4-EBR and all samples subset, whereas UBQ4 and EIF were the most unstable genes in the five sets. BestKeeper is also a program for analyzing the stability of gene expression by comparing the values of standard deviation (SD) and coefficient of variation (CV). Generally, the more stable the gene expression is, the lower the SD and CV values. The stability ranking results of the seven candidate reference genes are shown in Table 2. AP4 was the most stably expressed gene under the SL and 2,4-EBR treatments. TUB2 and UBQ4 were the appropriate reference genes for 6-BA and MeJ, respectively. In particular, UBCE expression showed the highest stability in the GA set and ethrel. Among the sets of ABA genes and all sample subsets, TBP was the best gene. In view of the ranking results, UBQ4 and TBP were the most unstable genes for GA and 2,4-EBR. UBQ4 was ranked last under ABA treatment and in all sample subsets. In addition, EIF was the gene with the most unstable expression in all the other sets. Because of the differences in the results that the algorithms of geNorm, NormFinder and BestKeeper may generate, the results obtained using the three software programs might be inconsistent. Hence, a comprehensive ranking of the stability of reference genes was required to be roundly evaluated and obtained by RefFinder. UBCE was the most suitable reference gene for normalization under the SL, GA, ethrel, and 2,4-EBR treatments and for all sample subsets, whereas for MeJ and ABA, AP4 was the most stable reference gene ( Table 2). Additionally, comprehensive ranking calculations revealed that TUB2 was the most stable gene under the 6-BA treatment. In particular, UBQ4 was the least stable reference gene across the three treatments (GA, ABA and all sample subsets). TUB2 and AP4 were the most unstable reference genes in the SL set and the 6-BA treatment, respectively. In the set of MeJ genes, UBC24 was the most unstable reference gene. In addition, UBCE ranked among the three most stable genes in the five sets, and it can be regarded as a relatively stable reference gene in the bulbs of L. aurea. For quantifying gene expression, selecting suitable reference genes is an important prerequisite by qRT-PCR. Here, we verified a set of candidate reference genes for the normalization of gene expression using qRT-PCR in L. aurea under different hormone treatments. The expression stability of the seven candidates was evaluated via four common professional programs, namely, geNorm, NormFinder, BestKeeper and RefFinder. For the study of gene expression under SL, ethrel, and 2,4-EBR treatments and all sample subsets, we recommend UBCE and TBP for normalization of the qRT-PCR data. For gene expression studies under GA treatment, UBCE and TUB2 were the two most suitable reference genes. In addition, we identified the stable reference genes TUB2 and UBC24 for the 6-BA treatment and AP4 and TBP for the MeJ treatment. These results will contribute to further quantitative analysis of gene expression related to exogenous hormone treatment in L. aurea and indicate that it is important to identify specific reference genes for specific conditions in L. aurea. COMPETING INTERESTS The authors declare that they have no conflict of interest. Funding Statement This work was supported by the grants from the Guizhou Provincial Science and Technology Plan Project (Qiankehe Foundation-ZK[2023]General 123), the Guizhou University Natural Science Project (2020--23), and Guizhou University Seed Program (2020--26). References