Assessment of Candidate Reference Genes for Gene Expression Studies Using RT-qPCR in Colletotrichum fructicola from Litchi

Litchi (Litchi chinensis Sonn.) is a tropical fruit originating from southern China that is currently cultivated in subtropical and tropical regions worldwide. Litchi anthracnose, caused by Colletotrichum fructicola, a dominant species of Colletotrichum spp., is an important disease of litchi that damages the fruits in fields and in post-harvest storage. Real-time quantitative PCR (RT-qPCR) is a common technique with which to detect the expression of and function of target genes quickly and precisely, and stable reference genes are crucial. However, there is no comprehensive information on suitable reference genes of C. fructicola present. Here, we designed eight candidate genes (GAPDH, α-tubulin, 18S, β-tubulin, EF1a, TATA, RPS5, and EF3) using RefFinder software (programs: geNorm, ΔCt, BestKeeper, and NormFinder) to investigate their reliability in the detection of C. fructicola under five different treatments (fungal development stage, temperature, UV, culture medium, and fungicide). The results showed the optimal reference genes under different conditions: EF1a and α-tubulin for developmental stage; α-tubulin and β-tubulin for temperature; α-tubulin and RPS5 for UV treatment; RPS5 and α-tubulin for culture medium; α-tubulin, GAPDH, and TATA for fungicide treatments. The corresponding expression patterns of HSP70 (Heat shock protein 70) were significantly different when the most and the least stable reference genes were selected when treated under different conditions. Our study provides the first detailed list of optimal reference genes for the analysis of gene expression in C. fructicola via RT-qPCR, which should be useful for future functional studies of target genes in C. fructicola.


Introduction
Litchi (Litchi chinensis Sonn.), a native fruit originating from southern China, has been extensively cultivated in various subtropical and tropical regions for its delicious taste and medicinal value [1].Litchi anthracnose is an important disease, which can cause flower and fruit drop, as well as fruit rot, thereby severely impacting the litchi industry.Colletotrichum fructicola is the dominant pathogenic fungus in Colletotrichum spp.causing litchi anthracnose, accounting for more than 70% of cases (our unpublished data).Colletotrichum is a widely distributed fungal genus that can cause a variety of plant diseases [2].Recognized as one of the top 10 plant pathogenic fungi worldwide in the "2016 Global Plant Status Assessment Report", Colletotrichum significantly affects plants globally [3].Statistics show that fruits exported from the tropical, subtropical, and Mediterranean regions are seriously affected by Colletotrichum spp., resulting in economic losses exceeding 80% without effective control measures [4].Given the impact of Colletotrichum spp. on the economic benefits of fruits, researchers have conducted extensive research on various aspects of its biology, including fungal-plant interaction, genomics and genetics, pathogen infection, cell biology, fungicide resistance, and virulence factors [2,5,6].At the same

Inoculation of C. fructicola
The isolate GDZJ, isolated from litchi fruits with typical symptoms of anthracnose, was identified molecularly as C. fructicola.Litchi fruit was obtained from Zhanjiang City, Guangdong Province, China in 2019.The isolate was incubated on potato dextrose agar (PDA), consisting of 200 g of potato, 20 g of dextrose, 20 g of agar, and 1 L of distilled water, sloped at 4 • C before this experiment.

The Preparation and Collection of Samples
The candidate reference genes of C. fructicola were assessed with respect to the stage of development, temperature, culture medium, presence of different fungicides, and UV treatment.Each C. fructicola developmental stage was sampled; mycelial plugs obtained from the edge of fresh colonies which were inoculated for 5 days were placed into PDB and onto PDA plates, and then ungerminated conidia were harvested after incubation in PDB at 28 • C for 5 days in darkness, shaking at 150 rpm; mycelia were harvested after incubation for 3 and 7 days on PDA under a photoperiod of 12 h.For the temperature treatment, mycelia plugs were incubated on PDA plates with a temperature of 25 • C for 5 days, followed by 15, 25, and 35 • C for 2 days.For different culture media treatments, mycelial plugs taken from the edge of fresh colonies were placed onto the following media: PDA, AEA (with 20 mL of glycerol, 5 g of yeast extract, 0.25 g of MgSO 4 •7H 2 O, 6 g of NaNO 3 , 1 g of KCl, 1.5 g of KH 2 PO 4 , 20 g of agar, and 1 L of distilled water), Czapek's medium (6 g of NaNO 3 , 1 g of K 2 HPO 4 , 0.5 g of MgSO 4 •7H 2 O, 0.5 g of KCl, 0.01 g of FeSO 4 , 30 g of sucrose, 20 g of agar, and 1 L of distilled water), and YBA (10 g of peptone, 10 g of yeast extract, 0.5 g of sodium acetate, 20 g of agar, and1 L of distilled water).After 5 days of incubation in a biological incubator at 28 • C with a photoperiod of 12 h, mycelia of each treatment were collected.For different fungicide treatments, the effective concentrations for the 50% inhibition (EC 50 ) of prochloraz, azoxystrobin, pyraclostrobin, thiophanate-methyl, chlorothalonil, and mancozeb in C. fructicola were estimated based on mycelial linear growth rates.Then, mycelia were collected after being treated with different fungicides for 5 days by transferring 5 mm agar plugs of actively growing mycelia to the center of PDA plates, to which were added prochloraz, azoxystrobin, pyraclostrobin, thiophanate-methyl, chlorothalonil, and mancozeb at final concentrations of the respective EC 50 values.For UV irradiation treatments, 5-day-old mycelia cultured on PDA plates were exposed to UV radiation at distance of 30 cm from a stable UV lamp (30 W; 254 nm) for 30, 90, and 150 s.After irradiation, the mycelia were immediately placed in a dark environment for 1 h and then collected.All treatments were repeated three times.

The Extraction of Total RNA and the Synthesis of cDNA
All-In-One DNA/RNA Mini-Preps Kit (Sangon, Shanghai, China) was used to extract total RNA from different samples of C. fructicola.NanoDrop One spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) was used to determine RNA concentration and purity.RNA integrity was evaluated via 1% agarose gel electrophoresis.As directed by the manufacturer, 1 g of RNA from each sample was used to synthesis cDNA using PrimeScript TM RT Reagent Kit (Takara, Kyoto, Japan).The concentration of cDNA was determined and diluted to 100 ng/µL before the RT-qPCR.

The Design of Primers and the Cloning of Genes
Eight candidate genes (GAPDH, α-tubulin, 18S, β-tubulin, EF1a, TATA, RPS5, and EF3) have been reported as suitable reference genes for qPCR and were used in this study [32][33][34][35].According to recently sequenced transcriptomes for C. fructicola, specific primers (Table 1) were designed with Primer Premier 5 Tool.All primers met the following parameters: the amplicon length was between 75 and 250 bases; the length of the primers was 20 ± 2 bases; the annealing temperature (Tm) was between 55 • C and 60 • C, and the GC content was at about 50%.PCR was performed with a 25 µL reaction mixture containing LA Taq DNA polymerase (Takara, Japan).In accordance with previous studies, PCR settings were employed [36].Each PCR product was purified and inserted into the vector pMD18-T (Takara, Japan) for sequencing.To confirm all reference genes' sequences, the National Center of Biotechnology Information (NCBI) database was used.

RT-qPCR
The RT-qPCR reaction was performed utilizing SYBR Green PCR Mix (2×) Kit (Takara, Japan) in the CFX96 real-time PCR system (Bio-Rad, Hercules, CA, USA).In a 50 µL reaction mix, 2.5 µL of each primer was mixed with 25 µL of SYBR Green Premix (Takara, Japan), 2.5 µL of diluted cDNA (about 100 ng/µL), and 17.5 µL of RNase-free water.In the following step, three technical replicates were prepared, each containing approximately 15 µL of solution.The RT-qPCR program started with a denaturation step at 95 • C for 3 min, and after being subjected to a temperature of 95 • C for 10 s and 55 • C for 30 s, the process continued for 40 cycles.A comprehensive analysis of the melting curve took place following the completion of the PCR run, spanning temperatures from 55 to 95 • C, increasing the temperature gradually by 0.5 • C every 10 s.Three technical replicates of the samples were employed in this study.Based on linear regression models, slope analysis was used to calculate the RT-qPCR for each gene.The diluted conidia cDNA samples (5 −1 , 5 −2 , 5 −3 , 5 −4 , and 5 −5 ) were utilized to construct standard curves for the purpose of determining gene amplification efficiency (E).

Assessment of Gene Expression Stability
By using the website RefFinder (http://blooge.cn/RefFinder/(accessed on 23 August 2023)), which offers four automatic programs for analysis, geNorm [37], NormFinder [38], Best-Keeper [39], and the ∆Ct method [40], each reference gene's stability was assessed.In the geNorm program, pairwise variation (Vn/Vn+1) was computed for the standardization factors NFn and NFn+1.If the calculated Vn/Vn+1 value was equal to or less than 0.15, it indicated that the optimal number of reference genes for the RT-qPCR analysis was n.
Absolute quantification was used to determine the copy number of HSP70 under different temperature treatments.The RT-qPCR primer containing the HSP70 gene fragment was designed, (forward primer: 5 ′ -CCCCTCTTTCCCTCGGTAT-3 ′ ; reverse primer: 5 ′ -GCTGGTTGTCGGAGAAGGTA-3 ′ ).The PCR product was purified and inserted into the vector pMD18-T (Takara, Japan) and then the recombinant plasmid DNA was extracted.The plasmid copy numbers were calculated using the following formula: Plasmid copy number (copy number/µL) = [6.02× 1023 × plasmid concentration (ng/µL) × 10 − 9]/[plasmid length × 660].Plasmid DNA was diluted in a 5-fold gradient for the preparation of calibration curves.For better accuracy, each sample in the dilution series was run in triplicate.
To test the stability of selected reference genes, HSP70 from C. fructicola was used as the target gene.Target gene expression was validated and quantified under different treatment conditions.Gene expression was determined by using the forward primer (5 ′ -3 ′ ) ACCTTCTCCACCGTCATCTG and reverse primer (5 ′ -3 ′ ) ATTGTGAGATTGCC-GACGA. The quantification of HSP70 expression in various treatments was determined by normalizing it to that of the two or three genes with the highest stability and the lowest stability.

Statistical Analysis
Using the 2 −∆∆Ct method, HSP70's relative expression pattern was calculated [41].SPSS software (v18.0)(SPSS Inc., Chicago, IL, USA) was used to conduct a one-way analysis of variance for all treatments in order to detect significant differences in HSP70 expression levels.

Detection of Primer Specificity and Efficiency
According to the agarose gel electrophoresis of the PCR products, all the primers amplified a single fragment of the eight selected reference genes from the DNA template (Figure 1); the size of each DNA fragment is shown in Table 1.
vector pMD18-T (Takara, Japan) and then the recombinant plasmid DNA The plasmid copy numbers were calculated using the following formula: number (copy number/µL) = [6.02× 1023 × plasmid concentration (ng/µL) mid length × 660].Plasmid DNA was diluted in a 5-fold gradient for the calibration curves.For better accuracy, each sample in the dilution series w licate.
To test the stability of selected reference genes, HSP70 from C. fructico the target gene.Target gene expression was validated and quantified under ment conditions.Gene expression was determined by using the forward ACCTTCTCCACCGTCATCTG and reverse primer (5′-3′) ATTGTGAGATT The quantification of HSP70 expression in various treatments was determin izing it to that of the two or three genes with the highest stability and the lo

Statistical Analysis
Using the 2 −∆∆Ct method, HSP70 s relative expression pattern was calcul software (v18.0)(SPSS Inc., Chicago, IL, USA) was used to conduct a one-w variance for all treatments in order to detect significant differences in HSP levels.

Detection of Primer Specificity and Efficiency
According to the agarose gel electrophoresis of the PCR products, a amplified a single fragment of the eight selected reference genes from the D (Figure 1); the size of each DNA fragment is shown in Table 1.All eight reference genes were expressed in C. fructicola, and the melti ysis indicated that a single peak was produced by each pair of primers, wi cific bands (Figure 2).All eight reference genes were expressed in C. fructicola, and the melting curve analysis indicated that a single peak was produced by each pair of primers, with no non-specific bands (Figure 2).
We constructed a standard curve (Figure 3) encompassing all R 2 of primers that exhibited a correlation coefficient exceeding 0.98, while maintaining an amplification efficiency (E) value ranging from 89.63% to 122.27% (Table 1).The equations for linear regression are presented in Table 1.We constructed a standard curve (Figure 3) encompassing all R 2 of primers that exhibited a correlation coefficient exceeding 0.98, while maintaining an amplification efficiency (E) value ranging from 89.63% to 122.27% (Table 1).The equations for linear regression are presented in Table 1.

Expression Levels of Candidate Reference Genes
The analysis of the Ct values of the eight selected candidate genes under five experimental conditions showed that Ct values for eight candidate genes ranged from 12.2 to 30.95 (Figure 4).In addition, 18S exhibited the lowest Ct values, indicating that it was the gene with the highest expression across all experimental scenarios.Except for individual genes under specific circumstances, Ct values for the remaining seven candidate reference genes were mostly around 23.

Expression Levels of Candidate Reference Genes
The analysis of the Ct values of the eight selected candidate genes under five experimental conditions showed that Ct values for eight candidate genes ranged from 12.2 to 30.95 (Figure 4).In addition, 18S exhibited the lowest Ct values, indicating that it was the gene with the highest expression across all experimental scenarios.Except for individual genes under specific circumstances, Ct values for the remaining seven candidate reference genes were mostly around 23.

The Stability of the Eight Reference Genes under Specific Experimental Conditions
Under the different experimental conditions, the ΔCt, bestkeeping, geNorm, and NormFinder tools were used to assess the stability of the eight candidate reference genes (Table 2).

The Stability of the Eight Reference Genes under Specific Experimental Conditions
Under the different experimental conditions, the ∆Ct, bestkeeping, geNorm, and NormFinder tools were used to assess the stability of the eight candidate reference genes (Table 2).geNorm.To assess the minimum quantity number of candidate genes for optimal normalization, the pairwise variation was computed with geNorm (Figure 5).The results showed that initial V-value (V2/3) was less than 0.15 in four experimental conditions (developmental stage, temperature, UV and culture medium), which indicated that the quantity of two reference genes was adequate for optimal normalization under various experimental conditions.For fungicide experimental settings, the initial V-value (V3/4) was less than 0.15.β-tubulin and EF1a; GAPDH and EF3; α-tubulin and EF1a; EF1a and RPS5; and α-tubulin, β-tubulin, and TATA were considered to be optimal for the development stage, UV treatment, temperature, culture medium, and fungicide treatment, respectively (Table 2).than 0.15.β-tubulin and EF1a; GAPDH and EF3; α-tubulin and EF1a; EF1a and RPS5; and α-tubulin, β-tubulin, and TATA were considered to be optimal for the development stage, UV treatment, temperature, culture medium, and fungicide treatment, respectively (Table 2).NormFinder.NormFinder analysis identified the most stable reference gene for each condition: EF1a for fungal development stage, α-tubulin for temperature and culture medium, RPS5 for UV treatments, and GAPDH for fungicide comparisons.
BestKeeper.In the fungal developmental stages, the standard deviation values for RPS5 and TATA exceeded 1, excluding them from being utilized as reference genes.Except for 18S, none of the other genes could be used to research temperature variations.The genes GAPDH, β-tubulin, and EF3 should not be used to research UV treatment variations.Moreover, it is not advisable to consider β-tubulin and EF3 reference genes when comparing culture mediums.Additionally, GAPDH, EF3, RPS5, β-tubulin, and 18S are unsuitable as reference genes for fungicide differences.The best gene for developmental stage was GAPDH, and that for culture medium comparisons and fungicide was EF1a.However, in comparisons of temperature and UV treatment, 18S was determined to be the most stable gene.
ΔCt method.When considering fungal development stages, EF1a was the best reference gene, while α-tubulin was the best gene for temperature, UV treatment, and culture medium-related experiments.For fungicide treatment, TATA was most stable reference gene.NormFinder.NormFinder analysis identified the most stable reference gene for each condition: EF1a for fungal development stage, α-tubulin for temperature and culture medium, RPS5 for UV treatments, and GAPDH for fungicide comparisons.
BestKeeper.In the fungal developmental stages, the standard deviation values for RPS5 and TATA exceeded 1, excluding them from being utilized as reference genes.Except for 18S, none of the other genes could be used to research temperature variations.The genes GAPDH, β-tubulin, and EF3 should not be used to research UV treatment variations.Moreover, it is not advisable to consider β-tubulin and EF3 reference genes when comparing culture mediums.Additionally, GAPDH, EF3, RPS5, β-tubulin, and 18S are unsuitable as reference genes for fungicide differences.The best gene for developmental stage was GAPDH, and that for culture medium comparisons and fungicide was EF1a.However, in comparisons of temperature and UV treatment, 18S was determined to be the most stable gene.
∆Ct method.When considering fungal development stages, EF1a was the best reference gene, while α-tubulin was the best gene for temperature, UV treatment, and culture medium-related experiments.For fungicide treatment, TATA was most stable reference gene.

Evaluation of the Selected Reference Genes
In order to determine the copy number of HSP70 under different temperature treatments, a standard curve of the HSP70 gene was made (Figure 7A) and the copy number of HSP70 under different temperature treatments was determined from the standard curve.Based on the results, we learnt that HSP70 has a higher copy number at both low and high temperatures as compared to that under 25 • C conditions (Figure 7B).Similarly, when we calculated the relative expression of HSP70 using the most stable reference gene, this was consistent with the above results (Figure 7C).However, when the least stable reference gene was utilized, inconsistent results were obtained (Figure 7D).
genes under different (A) developmental stage, (B) temperatures, (C) durations of UV treatment, (D) culture media, and (E) fungicides.A lower Geomean value means a more stable expression.

Evaluation of the Selected Reference Genes
In order to determine the copy number of HSP70 under different temperature treatments, a standard curve of the HSP70 gene was made (Figure 7A) and the copy number of HSP70 under different temperature treatments was determined from the standard curve.Based on the results, we learnt that HSP70 has a higher copy number at both low and high temperatures as compared to that under 25 °C conditions (Figure 7B).Similarly, when we calculated the relative expression of HSP70 using the most stable reference gene, this was consistent with the above results (Figure 7C).However, when the least stable reference gene was utilized, inconsistent results were obtained (Figure 7D).To assess and contrast the stability of chosen reference genes, we focused on the expression among all the different treatments with HSP70.However, under different development stages, UV treatments, culture media, and fungicides, the expression pattern and relative expression of the HSP70 gene was different between the most stable and the least stable reference genes (Figure 8A-D).To assess and contrast the stability of chosen reference genes, we focused on the expression among all the different treatments with HSP70.However, under different development stages, UV treatments, culture media, and fungicides, the expression pattern and relative expression of the HSP70 gene was different between the most stable and the least stable reference genes (Figure 8A-D).under UV treatment times (A) was normalized to that of the most stable (α-tubulin and RPS5) and least stable (TATA and EF3) reference genes.CK means that the UV treatment time was 0. The relative expression of HSP70 for culture media (B) was normalized to that of the best genes (RPS5 and α-tubulin) and worst genes (β-tubulin and EF3), respectively.The relative expression of HSP70 in developmental stages (C) was normalized to that of the best genes (EF1a and α-tubulin) and worst genes (18S and RPS5), respectively.The relative expression of HSP70 treated using different fungicides (D) was normalized to that of the best genes (α-tubulin, TATA and GAPDH) and worst genes (EF3, RPS5, and 18S).Control means that fungi were cultured without fungicides.The values are means ± standard error.Significant differences normalized to the most stable are indicated by different lowercase letters (p < 0.05).Significant differences normalized to the least stable are indicated by different uppercase letters (p < 0.05).

Discussion
Our study focused on the identification of the most suitable reference genes for C. fructicola in different development periods, temperatures, different durations of UV treatment, culture media, and fungicides via the evaluation of the stability of eight candidate genes, including GAPDH, α-tubulin, 18S, β-tubulin, EF1a, TATA, RPS5, and EF3.The stability of the eight genes was evaluated in five experimental conditions using RefFinder software including geNorm, ∆Ct, BestKeeper and NormFinder programs.Considering its important role in autophagy, a conserved mechanism in all stages of fungus development, HSP70, was used as a reporter gene.The stability of potential reference genes of C. fructicola causing litchi anthracnose has been systematically evaluated for the very first time.
RT-qPCR has emerged as a pivotal method through which to analyze gene transcript profiles owing to its sensitivity, accuracy, and reproducibility [19,42,43].To accurately determine the expression pattern of a target gene using RT-qPCR, an appropriate reference gene must be selected.The expression of reference genes is expected to remain stable and consistent regardless of time or different experimental conditions [19,42,43].Without a primary investigation of their stability, using previously identified reference genes for the normalization of target gene expression under specific experimental conditions could under UV treatment times (A) was normalized to that of the most stable (α-tubulin and RPS5) and least stable (TATA and EF3) reference genes.CK means that the UV treatment time was 0. The relative expression of HSP70 for culture media (B) was normalized to that of the best genes (RPS5 and α-tubulin) and worst genes (β-tubulin and EF3), respectively.The relative expression of HSP70 in developmental stages (C) was normalized to that of the best genes (EF1a and α-tubulin) and worst genes (18S and RPS5), respectively.The relative expression of HSP70 treated using different fungicides (D) was normalized to that of the best genes (α-tubulin, TATA and GAPDH) and worst genes (EF3, RPS5, and 18S).Control means that fungi were cultured without fungicides.The values are means ± standard error.Significant differences normalized to the most stable are indicated by different lowercase letters (p < 0.05).Significant differences normalized to the least stable are indicated by different uppercase letters (p < 0.05).

Discussion
Our study focused on the identification of the most suitable reference genes for C. fructicola in different development periods, temperatures, different durations of UV treatment, culture media, and fungicides via the evaluation of the stability of eight candidate genes, including GAPDH, α-tubulin, 18S, β-tubulin, EF1a, TATA, RPS5, and EF3.The stability of the eight genes was evaluated in five experimental conditions using RefFinder software including geNorm, ∆Ct, BestKeeper and NormFinder programs.Considering its important role in autophagy, a conserved mechanism in all stages of fungus development, HSP70, was used as a reporter gene.The stability of potential reference genes of C. fructicola causing litchi anthracnose has been systematically evaluated for the very first time.
RT-qPCR has emerged as a pivotal method through which to analyze gene transcript profiles owing to its sensitivity, accuracy, and reproducibility [19,42,43].To accurately determine the expression pattern of a target gene using RT-qPCR, an appropriate reference gene must be selected.The expression of reference genes is expected to remain stable and consistent regardless of time or different experimental conditions [19,42,43].Without a primary investigation of their stability, using previously identified reference genes for the normalization of target gene expression under specific experimental conditions could result in significant issues and inaccuracies in the data and their interpretation [44].Hence, it is crucial to validate the chosen reference genes under specific experimental conditions in order to address this concern.
As a dominant disease on litchi, litchi anthracnose severely impacted the litchi industry.A comprehensive understanding of the pathogenesis and fungicide resistance mechanisms of pathogenic fungi that cause such disease requires the study and elucidation of the role of genes and variations in gene expression under different conditions [45][46][47].In the past few years, reports on the screening of reference genes in plant pathogenic fungi have been on the rise [17,40,48], but the reference genes of the litchi anthracnose pathogen have been largely neglected to date.Therefore, eight reference genes (GAPDH, α-tubulin, 18S, β-tubulin, EF1a, TATA, RPS5, and EF3) were selected and evaluated in different experimental treatments in this study.
Reference genes need consistent high-level expression under diverse conditions.Briefly, 18S is generally considered to have high expression in various experimental conditions [36,44], and the remaining seven candidate reference genes are also highly expressed under various conditions and have been commonly used as reference genes [28,36,43,44].Therefore, these eight genes have the potential to serve as reference genes in C. fructicola studies.After evaluating the Ct value in RT-qPCR, we initially examined the expression of the candidate reference genes in this research.The Ct values of the eight potential reference genes were approximately 23 for the majority of the tested samples.Additionally, the linear amplification of all candidate reference genes was exceptional, as demonstrated by R2 values surpassing 98%.The amplification efficiency met the basic requirements of 89.63-122.27%[28].
The appropriateness of reference genes for transcript normalization was determined using ∆Ct, geNorm, NormFinder, and BestKeeper programs during fungal development stage, temperature, UV, culture medium, and fungicide conditions in C. fructicola [48,49].Taken together, the rankings of the optimal reference genes derived from various conditions and software analyses were similar to the results of previous studies [44].Previous studies have highlighted the importance of selecting multiple reference genes for unbiased normalization in RT-qPCR analysis [28].The required number of reference genes for proper normalization may differ based on the specific experimental conditions [44].Figueiredo determined that three reference genes were sufficient for the entire dataset, while four reference genes were sufficient for Catimor 88 and two or three reference genes were sufficient for Caturra hypocotyls inoculated with Colletotrichum kahawae [43].However, utilizing an excessive number of reference genes may create practical challenges during the execution and operation of a study [24].Therefore, to ensure reliable normalization, it is optimal to choose a minimum of two highly accurate reference genes, and this has been confirmed in previous studies [48,50].In this study, we utilized geNorm program to perform pairwise analysis, confirming that for assessing gene expression levels via RT-qPCR during fungal development or under different temperature, UV, or culture medium conditions, at least two out of the eight genes could serve as validated reference genes.On the other hand, three reference genes were adequate for normalization under fungicide conditions.Based on RefFinder, EF1a and α-tubulin are the most stable genes of C. fructicola for developmental stage, α-tubulin and β-tubulin are the most stable for for different temperatures, α-tubulin and RPS5 are the most stable for different UV treatments, RPS5 and α-tubulin are the most stable for different culture media, and α-tubulin, TATA, and GAPDH are the most stable for different fungicides.
Tubulin (β-tubulin and α-tubulin) served as a reference gene in numerous research studies [44].For example, β-tubulin was used as a reference gene in the manipulation of temperature conditions [48,50], which produced results consistent with ours.Furthermore, the level of expression for a-tubulin remains consistent across various tissues and sexes of Coleomegilla maculata [51], and a-tubulin was identified as the most stable reference gene for diverse UV exposure, temperature, and fungicide treatment experiments in this study.It is widely acknowledged that ribosomal protein (RP), a highly conserved protein in all living organisms, plays a crucial role as a fundamental constituent of ribosomes.Previous research has consistently identified RP-encoding genes (like RPL and RPS) as optimal reference genes for studying gene expression.For example, in C. kahawae samples, RPL18 was revealed served as the optimal reference gene [24].Based on our results, the stability of RPS5 as a reference gene in experiments related to different culture mediums is unparalleled.However, it is the least stable reference gene for fungal development stage experiments.EF1a, involved in protein synthesis, is widely used for normalizing gene expression.For example, EF1a showed the highest level of stability among the reference genes in Aphis craccivora regardless of the developmental stages or temperature treatments [44].Our findings indicate that EF1a was consistently stable across various development stages.Overall, our results provide additional support for the need to validate the suitability of frequently utilized reference genes under distinct experimental circumstances.
In order to confirm the reliability of the potential reference genes found in our study, we evaluated the expression profile of the target gene HSP70 using a selection of the most consistent reference genes and the least consistent reference genes among all the different treatments.However, under different development stages, UV, culture media, and fungicides, the expression pattern and relative expression of the HSP70 gene were different between the use of the most stable and the least stable reference genes.The results illustrated the importance of selecting the appropriate reference gene when performing RT-qPCR analysis, since the utilization of unreliable reference genes led to imprecise or insignificant variations in transcript levels.Thus, in order to ensure accurate analyses of RT-qPCR data in C. fructicola, it is essential to utilize a dependable reference gene.

Conclusions
Eight candidate reference genes of C. fructicola were selected and their stability was evaluated under five experimental conditions.To our knowledge, this is the first report of a comparison and identification of RT-qPCR reference genes for C. fructicola in five distinct experimental conditions.The findings should facilitate future analyses of target gene expression profiles in C. fructicola, which should prove useful for functional studies of the target genes in C. fructicola from litchi.

Figure 4 .
Figure 4. Box and whisker plots of Ct values for eight candidate reference gene across five experimental conditions.(A) different developmental stages, (B) different temperatures, (C) different times treated by UV, (D) different culture media, and (E) different fungicides.All the independent experiments were repeated three times.Boxes represent lower and upper quartiles of cycle thresholds range with medians indicated, and whisker caps represent maximum and minimum values.About 75 ng of cDNA was added for each experiment.

Figure 4 .
Figure 4. Box and whisker plots of Ct values for eight candidate reference gene across five experimental conditions.(A) different developmental stages, (B) different temperatures, (C) different times treated by UV, (D) different culture media, and (E) different fungicides.All the independent experiments were repeated three times.Boxes represent lower and upper quartiles of cycle thresholds range with medians indicated, and whisker caps represent maximum and minimum values.About 75 ng of cDNA was added for each experiment.

Figure 5 .
Figure 5. Optimal reference gene numbers for effective normalization.Pairwise variation (V) in the candidate reference genes calculated via geNorm based on different comparisons: developmental stage, temperature, UV, culture medium, and fungicide.

Figure 5 .
Figure 5. Optimal reference gene numbers for effective normalization.Pairwise variation (V) in the candidate reference genes calculated via geNorm based on different comparisons: developmental stage, temperature, UV, culture medium, and fungicide.

Figure 6 .
Figure 6.Stability of the expression of the eight candidate reference genes in C. fructicola under different conditions analyzed with RefFinder.Stability of the expression of the eight candidate reference

Figure 6 .
Figure 6.Stability of the expression of the eight candidate reference genes in C. fructicola under different conditions analyzed with RefFinder.Stability of the expression of the eight candidate reference genes under different (A) developmental stage, (B) temperatures, (C) durations of UV treatment, (D) culture media, and (E) fungicides.A lower Geomean value means a more stable expression.

Figure 7 .
Figure 7.The expression of HSP70 under temperature treatment.Standard curves of HSP70 gene (A).The copy number of HSP70 under different temperature treatments (B).The relative abundance of HSP70 under different temperatures normalized to that of the best genes (α-tubulin and β-tubulin) (C) and worst genes (RPS5 and TATA) (D).The values are means ± standard error.Significant differences are indicated by different lowercase letters (p < 0.05).

Figure 7 .
Figure 7.The expression of HSP70 under temperature treatment.Standard curves of HSP70 gene (A).The copy number of HSP70 under different temperature treatments (B).The relative abundance of HSP70 under different temperatures normalized to that of the best genes (α-tubulin and β-tubulin) (C) and worst genes (RPS5 and TATA) (D).The values are means ± standard error.Significant differences are indicated by different lowercase letters (p < 0.05).

Figure 8 .
Figure 8.The relative expression of HSP70 in different treatments.The relative expression of HSP70under UV treatment times (A) was normalized to that of the most stable (α-tubulin and RPS5) and least stable (TATA and EF3) reference genes.CK means that the UV treatment time was 0. The relative expression of HSP70 for culture media (B) was normalized to that of the best genes (RPS5 and α-tubulin) and worst genes (β-tubulin and EF3), respectively.The relative expression of HSP70 in developmental stages (C) was normalized to that of the best genes (EF1a and α-tubulin) and worst genes (18S and RPS5), respectively.The relative expression of HSP70 treated using different fungicides (D) was normalized to that of the best genes (α-tubulin, TATA and GAPDH) and worst genes (EF3, RPS5, and 18S).Control means that fungi were cultured without fungicides.The values are means ± standard error.Significant differences normalized to the most stable are indicated by different lowercase letters (p < 0.05).Significant differences normalized to the least stable are indicated by different uppercase letters (p < 0.05).

Figure 8 .
Figure 8.The relative expression of HSP70 in different treatments.The relative expression of HSP70under UV treatment times (A) was normalized to that of the most stable (α-tubulin and RPS5) and least stable (TATA and EF3) reference genes.CK means that the UV treatment time was 0. The relative expression of HSP70 for culture media (B) was normalized to that of the best genes (RPS5 and α-tubulin) and worst genes (β-tubulin and EF3), respectively.The relative expression of HSP70 in developmental stages (C) was normalized to that of the best genes (EF1a and α-tubulin) and worst genes (18S and RPS5), respectively.The relative expression of HSP70 treated using different fungicides (D) was normalized to that of the best genes (α-tubulin, TATA and GAPDH) and worst genes (EF3, RPS5, and 18S).Control means that fungi were cultured without fungicides.The values are means ± standard error.Significant differences normalized to the most stable are indicated by different lowercase letters (p < 0.05).Significant differences normalized to the least stable are indicated by different uppercase letters (p < 0.05).

Table 1 .
Information about the candidate reference genes in a nutshell.

Table 2 .
The stability of gene expression assessed using geNorm, Normfinder, BestKeeper, and ∆Ct for eight candidate reference genes of C. fructicola under different experimental conditions.