Page 43 - 《水产学报》2026年第2期

2 期 李 双，等：跨组织/性别系统评估曼氏无针乌贼内参基因的稳定性 50 卷

and β-actin (Fig. 8-q), which robustly supported the actin, which is largely consistent with rankings of
outcomes of the tissue-specific quantification. algorithm-dependent quantifications.
In female S. japonica, ef-1α was ranked as the
2.7 Comprehensive ranking of reference genes
most stable gene in more tissues (seven), never
by RefFinder
occupying the lowest stability position in any tissue
RefFinder was applied to generate a comprehens-
(Fig. 9-b, Tab. 6). Following ef-1α, ef-1γ and 18S
ive stability ranking for the five candidate genes in
ranked consistently behind. Similarly, although the
each tissue of both male and female S. japonica, with
other two genes showed relatively high rankings in
the rank orders detailed in Tab. 5 (male) and Tab. 6
one or two tissues, their comprehensive rankings were
(female). For a concise overview of these rankings, the
lagging behind. Eventually, the comprehensive rank-
distribution of each gene’s ranking, from the most to
ing was slightly different from that of males, namely ef-
least stable position, was quantified for both sexes
1α > ef-1γ > 18S > gapdh > β-actin.
(Fig. 9).
2.8 The optimal number of reference genes
The analysis of male S. japonica indicated that ef-
1γ ranked as the most stable gene in eleven tissues Based on the average pairwise variation (V n/n+1 )
(Fig. 9-a, Tab. 5). ef-1α and 18S attained the highest value, the optimal number of reference genes required
stability ranking in three and one tissues, respectively, for accurate data normalization in each tissue was
with ef-1α demonstrating a relatively higher fre- estimated by geNorm (Fig. 10). A default value of
quency in the second-place ranking. In contrast, 0.15 was adopted as the threshold when performing
although gapdh and β-actin achieved a maximum the calculation. In this case, when the V n/n+1 value is
ranking of second in a limited number of specific tis- less than 0.15, adding one or more internal genes has
sues, their overall comprehensive performance no essential effect on data normalization. As shown in
remained suboptimal across the entire dataset. Con- Fig. 10-a, in males, the V value fell below 0.15 in the
2/3
sequently, the overall comprehensive rank order for optic lobe, heart, gill, branchial heart, liver, pancreas,
male was as follows: ef-1γ > ef-1α > 18S > gapdh > β- and white body, suggesting that two reference genes

Tab. 5 Comprehensive ranking in males evaluated by RefFinder
recommended comprehensive rank
male
1 2 3 4 5
brain ef-1α ef-1γ 18S gapdh β-actin
optic lobe ef-1γ gapdh ef-1α 18S β-actin
heart ef-1α ef-1γ β-actin gapdh 18S
gill ef-1γ ef-1α 18S gapdh β-actin
branchial heart 18S ef-1α ef-1γ gapdh β-actin
intestine ef-1γ 18S ef-1α gapdh β-actin
stomach ef-1γ ef-1α 18S β-actin gapdh
caecus ef-1γ ef-1α 18S gapdh β-actin
kidney ef-1α ef-1γ β-actin 18S gapdh
liver ef-1γ ef-1α β-actin 18S gapdh
pancreas ef-1γ β-actin 18S gapdh ef-1α
white body ef-1γ ef-1α 18S gapdh β-actin
skin ef-1γ gapdh 18S ef-1α β-actin
testis ef-1γ ef-1α 18S β-actin gapdh
spermatophore ef-1γ β-actin ef-1α gapdh 18S
all tissues ef-1γ ef-1α 18S gapdh β-actin

中国水产学会主办 sponsored by China Society of Fisheries https://www.china-fishery.cn
15