**1. Introduction**

*Lindera megaphylla* is a predominant, broad-leaved, and aromatic evergreen tree species belonging to the Lauraceae family and is widely distributed in the subtropical and warmtemperate zones of China. *L. megaphylla* has not only ecological and ornamental value, but also medicinal and therapeutic value as a source of an essential oils, spices, and drugs [1–3]. For example, d-dicentrine, an aporphine alkaloid, is isolated from the root of *L. megaphylla* and has potential antitumor activity [4,5]. These trees are rich in terpenoids, alkaloids, and flavonoids, many of which could be used to make pesticides or industrial feedstocks, while its volatile compounds, mainly terpenoids, have strong bactericidal ability. These trees can also help to improve air quality [6], a property that could be improved through molecular breeding. To synthesize antibacterial compounds, it is necessary to first explore the related regulatory genes and to analyze their functions. To date, studies of *L. megaphylla* have largely focused on the cultivation of seedlings, various kinds of biotic and abiotic stress responses and the analysis of volatile substances and their potential applications [7–12], while few studies have focused on the molecular biology of *L. megaphylla* due to a lack of genomic information. Gene expression analysis circumvents that lack of a sequenced genome to explore the molecular mechanisms underlying transcriptional regulation of phenotype. Transcriptome datasets derived from different tissues and differently aged

**Citation:** Liu, H.; Liu, J.; Chen, P.; Zhang, X.; Wang, K.; Lu, J.; Li, Y. Selection and Validation of Optimal RT-qPCR Reference Genes for the Normalization of Gene Expression under Different Experimental Conditions in *Lindera megaphylla*. *Plants* **2023**, *12*, 2185. https:// doi.org/10.3390/plants12112185

Academic Editors: Aiping Song and Yu Chen

Received: 12 April 2023 Revised: 18 May 2023 Accepted: 29 May 2023 Published: 31 May 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

leaves of *L. megaphylla* have been obtained (unpublished data), which will greatly promote functional genetic studies in this species.

Real-time reverse transcriptase quantitative polymerase chain reaction (RT-qPCR) is an important tool for analyzing gene expression because of its high throughput, sensitivity and precision [13,14]. However, RT-qPCR data are affected by many factors, such as extraction protocols, the purity and integrity of the extracted RNA, the efficiency of reverse transcription and PCR amplification, and primer specificity [15,16]. Therefore, a stably expressed reference gene is essential to avoid unnecessary errors generated through confounding factors and to increase the accuracy of the RT-qPCR data analysis. In the pre-genomic era, traditional reference genes were chosen based on their known or suspected housekeeping roles in basic cellular processes, cell structure, or primary metabolism, including genes encoding actin (*ACT*), 18S rRNA (*18S*) and glyceraldehyde-3 phosphate dehydrogenase (*GAPDH*) [17–20]. Unfortunately, these housekeeping genes are not stably expressed in all tissues, under different experimental conditions, or between different species [21,22]. In addition, a growing number of studies have revealed that some novel genes can also be used as internal references and are better than some traditional housekeeping genes [23–26]. Thus, it is necessary to systematically select the most appropriate reference genes to ensure the accuracy of RT-qPCR analysis for specific conditions and in specific materials. For this reason, optimal reference genes for transcript normalization must be determined through statistical algorithms such as delta cycle threshold (ΔCt) [27], geNorm [28], NormFinder [29], BestKeeper [30], and RefFinder [31] for each type of sample. These algorithms are widely used to assess the transcript stability of candidate reference genes in various species [32–35]. The use of reference genes in expression analysis has greatly facilitated our understanding of the important information related to gene functions and complex biological processes in plants, such as the signaling and metabolic pathways that underlie developmental and cellular processes [36–39].

Gene expression databases of model plant species such as *Arabidopsis thaliana* and tomato (*Lycopersicon esculentum*) are important resources for identifying and searching for genes of interest and their expression patterns (http://www.ebi.ac.uk/arrayexpress/; accessed on 5 June 2022; http://www.ncbi.nlm.nih.gov/geo/, accessed on 5 June 2022) [17]. Czechowski et al. selected, verified, and recommended 18 new reference genes that were superior to traditional reference genes in terms of expression stability across an extensive sample series or under a range of environmental conditions through the publicly available AtGenExpress database (http://web.uni-frankfurt.de/fb15/botanik/mcb/AFGN/atgenex. htm, accessed on 5 June 2022) and the author's own ATH1 database [17]. Orthologues of known genes in Arabidopsis can serve the same purposes as in other species. In addition, Lin et al. validated suitable reference genes for reliable normalization of data from *Litsea cubeba* [40]. Moreover, both *L. megaphylla* and *L. cubeba* belong to the Lauraceae family, and their evolutionary relationship is relatively close.

Based on these previous results, we first selected 40 genes that have been used as internal reference from published model plants such as Arabidopsis thaliana and other species in the Lauraceae family as candidate reference genes for *L. megaphylla*. Second, we screened homologous sequences in the transcriptome database of *L. megaphylla* and obtained 20 candidate genes based on expression multiples less than 1.5 and FPKM value > 50 in different tissues.

In this study, 14 candidate genes were identified with E-values between 91.035% and 107.169% and R2 values from 0.991 to 0.999, indicating that the primer pairs may be more accurate for standardized evaluation by RT-qPCR. The candidate genes included translationally controlled tumor protein (*TCTP*), *ACT7*, *GAPDH*, ubiquitin-conjugating enzyme E2 36/7 (*UB C36*, *UBC7*), elongation factor 2-like (*EF2*), peptidyl-prolyl cis-trans isomerase CYP20-2, chloroplastic (*CYP20-2*), polyubiquitin (*UBQ*), alpha-tubulin (*TUA*), ubiquitin-conjugating enzyme E2 28-like (*UBC28*), NADH dehydrogenase (*ubiquinone*), pentatricopeptide repeat-containing protein (PPR), eukaryotic initiation factor 4A-3-like (*EIF4A-3*), DEAD-box ATP-dependent RNA helicase 15 (*helicase-15*), and polyadenylatebinding protein 2-like (*PAB2*). These 14 candidate genes were then assessed for stability of expression under specific conditions, including different tissues of one-year-old seedlings (roots, stems, and leaves), tissues of 10-year-old trees (leaf buds, young stems, young seeds, young leaves, and mature leaves), 16 different leaf developmental stages and under different temperature stresses (cold and heat). Finally, the *NAC* and *ERF* genes, which are from an important family of transcription factors in plants, were used to verify the reliability of the selected reference genes in different samples. Our research identified the best reference genes for RT-qPCR analysis of *L. megaphylla* tissues under different conditions, laying a basis for further studies of the molecular mechanisms regulating gene expression in this important tree species.
