Page 32 - 《水产学报》2026年第2期

2 期 水 产 学 报 50 卷

Tab. 2 Summary of five candidate reference genes
gene primer sequence (5′→3′) primer length/bp PCR efficiency/% R 2 accession no. references
ef-1α F:ACAAGATGGACACAGCCGAATA 22 98.890 0.986 PV135310 this study
R:CCAGACACAGGCACAAAAGGTA 22
ef-1γ F:GCAAAAACCCCACCAAAGCCGA 22 95.600 0.997 PV684489 [47]
R:AAGTGTTCCCAAAAGTACGGCA 22
gapdh F:TGGTGTGTTCACTACCCTGG 19 99.870 0.978 PV135311 this study
R:GTCATTGTTCACGCCCATC 18
18S F:ACTCAACACGGGAAAACTCACC 22 105.450 0.994 PV135238 [47]
R:TTATCGGAATCAACCAGACGGA 22
β-actin F:GCCAGTTGCTCGTTACAG 18 92.670 0.998 PV684488 [24]
R:GCCAACAATAGATGGGAAT 19

1.4 Quantitative real-time PCR (qPCR) metric mean of their weights.
Furthermore, geNorm employs a pairwise vari-
The mRNA expression levels of reference genes
ation analysis to ascertain the optimal number of refer-
were analyzed by quantitative real-time PCR (qPCR)
[2]
ence genes . This method calculates the geometric
carried out CFX96™ real-time PCR System (Bio-Rad,
mean of selected genes exhibiting stable expression to
USA). A 10 μL reaction was prepared using Universal
ensure accurate normalization. A pairwise variation
SYBR Green qPCR Supermix reagent (Sangon,
value (V ) below 0.15 indicates that the inclusion
China), containing 5 μL Universal SYBR Green qPCR n/n+1
Supermix, 0.25 μL each of forward and reverse of an additional control gene (n+1) would not signific-
antly enhance the normalization factor.
primers, 3.5 μL DNase/RNase-free water, 1 μL cDNA
(250 ng/μL). The qPCR cycling protocol was as fol- 1.6 Bioinformatic screening and initial valida-
lows: pre-denaturation at 95 ℃ for 2 min, amplifica- tion
tion 40 cycles including denaturation at 95 ℃ for 5 sec
An in-depth bioinformatic mining of the pre-
and annealing at 60 ℃ for 30 sec. The qPCR of all tis-
existing gill and kidney transcriptome datasets of male
sues was carried out in same reaction system and pro-
S. japonica (n = 3) from our research group was con-
cedures for each reference gene to obtain cycle
ducted for further validation. Here, four target genes,
threshold (C ) value. Each tissue was sampled from
t
namely cysteine-rich with EGF-like domain protein 2
five individuals, and each individual contributed three
isoform X1 gene (creld2), CD109 antigen gene
biological repeats.
(cd109), aminoacylase-1-like gene (acy1), and inos-
1.5 Assessment of reference gene stability and
itol oxygenase gene (miox) were selected. The primers
optimal reference gene number
are listed in Tab. 3.
The online-tool RefFinder (https://www.ciidir
1.7 Statistical analysis
[14]
sinaloa.com.mx/RefFinder-master/) , which harbors
Statistical analysis was performed using Graph-
[2]
[11]
four algorithms, geNorm , NorFinder ,

[12]
BestKeeper , and ΔC t [13] , was used to assess the sta- Tab. 3 Primers used for qPCR validation of four genes
bility of the candidate reference genes. Raw C values primers sequences (5'→3') source
t
were the input data for all algorithms, while the out- creld2-F-qPCR GTGAAAAAGACCCGTGTAG this study
put included different stability values (geNorm: M, creld2-R-qPCR ATTACACTTTCCACTGCCA
NormFinder: ρ, BestKeeper: mean SD of mean and cd109-F-qPCR ACGTCCATACATCCCCTCAT
CV, ΔC : Mean SD of mean C ) and stability rank cd109-R-qPCR GTCAGCGTCTCAATACCCG
t
t
order. In each program, a gene that is more stable has acy1-F-qPCR CAGACGGTATTTACGAATCAAG
a lower value and a lower rank. Meanwhile, depend- acy1-R-qPCR TCAACAGTCATCACAAACACAG
miox-F-qPCR TGGATGTAGACCAGCCGATTC
ing on the rankings from each program, RefFinder
miox-R-qPCR GCCCGTGTGCCAAGGATAG
offers a comprehensive ranking by calculating the geo-
https://www.china-fishery.cn 中国水产学会主办 sponsored by China Society of Fisheries
4