Multiple,Isoforms,of,Olive,Flounder,(Paralichthys,olivaceus)Pax3a,and,Pax3b,Display,Differential,Regulations,on,Myogenic,Differentiation

JIAO Shuang, TAN Xungang, YOU Feng, and PANG Qiuxiang

Multiple Isoforms of Olive Flounder ()Pax3a and Pax3b Display Differential Regulations on Myogenic Differentiation

JIAO Shuang1), 2), 3), 4), *, TAN Xungang3), 4), *, YOU Feng3), 4), and PANG Qiuxiang1), 2)

1),,,255000,2),,,255000,3),,,,266071,4),,266237,

Paired box 3(Pax3) is a critical upstream regulator of the onset of myogenesis. We have previously identified two spliced isoforms of(and) and three spliced isoforms of(,, and) in olive flounder, but their roles in myogenesis are unknown. In this study, we investigated their cellular localization, transcriptional activity ongene regulation, and roles in myogenesis. Different Pax3aandPax3b isoforms revealed various subcellular localizations, which were re- lated to their corresponding protein structures. Pax3a-1, Pax3a-2, and Pax3b-1 promoted the transcriptional activity ofto dif- ferent degrees, whereas Pax3b-2 and Pax3b-3 had a slight inhibitory or no effect. The pairwise interaction analysis demonstrated the synergistic effect of Pax3b-1 and Pax3b-3 ontranscriptional activity. The overexpression of differentandisoforms differentially altered the spatial expression patterns ofand differentially regulated the expression levels of their target genes (,, and)in zebrafish embryonic myogenesis. In addition, the different flounderpromoter-driven/isoform expression vectors were successfully introduced into the skeletal muscles of juvenile flounder by electroporation. How- ever, none of them could change the mRNA expression levels of,,,,, andin the electroporated muscles. These results suggest that different Pax3a and Pax3b isoforms may precisely and collaboratively regulate embryonic myo- genesis, but their roles in juvenile myogenesis are uncertain.

Pax3a; Pax3b; isoforms; myod; myogenesis

()genesencode a family of transcrip- tion factors that play key roles in various stages of embryo- nic development and in adult organogenesis (Paixao-Cor- tes., 2015). All Pax proteins contain a highly conserved 128-amino acid DNA-binding domain, that is, the paired domain (PD). They may also have an additional DNA-bind- ing homeodomain (HD) and/or an octapeptide motif (OP) (Blake and Ziman, 2014). A transactivation domain (TAD) at the C-terminus of Pax mediates transcriptional regula- tion (Wang., 2008).genes are grouped into four subfamilies based on their structural and expression do- mains (Mayran., 2015). The Pax3/7 subfamily regu- lates myogenesis (Buckingham and Relaix, 2015) and neu- rogenesis (Koblar., 1999).

During vertebrate evolution,andgenes arose by duplication from a unique ancestralgene (Pai- xao-Cortes., 2013). Furthermore, as the result of addi- tional whole genome duplication in the fish lineage(Ven- katesh, 2003), two(and) and two(and) genes appeared in numerous teleost fishes, such as zebrafish () (Minchin and Hug- hes, 2008), Japanese pufferfish () (Akol- kar., 2016), and olive flounder () (Jiao., 2015a, 2015b). In addition, several alter- native splicing variants ofandhave been iden- tified in fishes, mice, and humans (Seo., 1998; Barber., 1999; Jiao., 2015a, 2015b;Imbriano and Mo- linari, 2018). The existence of multiple isoforms is likely to increase the functional diversity of Pax3/7 proteins or allow the fine regulation of gene expression through mul- tiple mechanisms.

Pax3 and Pax7 are key regulators of myogenesis. Pax3 is first expressed in the presomitic mesoderm (Williams and Ordahl, 1994) and becomes restricted to myogenic pro- genitors of the dermomyotome (Goulding., 1991), which develops into the hypaxial body and limb muscle during development (Tremblay., 1998; Relaix., 2005;Bajard., 2006). Pax7 is expressed later and co-expressed in the central domain of the dermomyotome.transcription is down-regulated in fetal muscles when Pax7 becomes the predominant Pax factor in myogenic proge- nitors during fetal development. By contrast, Pax7 is dis- pensable for fetal myogenesis, but it predominates in myo- genesis during postnatal growth and muscle regeneration in the adult. Therefore, in the context of myogenesis, Pax3 and Pax7 play overlapping yet nonredundant roles(Chang and Rudnicki, 2014). Similar to Pax3 and Pax7 proteins, myogenic regulatory factors (MRFs) are important genes controlling myogenesis, including Myod, Myf5, Myoge- nin, and Mrf4 (Asfour., 2018). Pax3 and Pax7 direct- ly bind and transactivate thepromoter in myoblasts (Hu., 2008). Pax3 directly regulatesexpression by binding the distal enhancer element ofgene dur- ing embryonic myogenesis (Bajard., 2006); Pax7 di- rectly regulates5 in myoblasts derived from satellite cells (McKinnell., 2008). In addition, Pax3 directly activates the receptor tyrosine kinase coding gene, which is involved in the migration of myogenic precursor cells (Epstein., 1996);which is involved in the self-renewal versus differentiation of muscle progenitors(Lagha., 2008a); and, which is expressed on sites of skeletal muscle formation(Lagha., 2013).

Olive flounder () is a commercial-ly important flatfish species cultured in East Asia. Fish mus- cle, the main edible part, is one of the most economically important traits affecting the profitability of fish produc- tion. Thus, the study on the regulatory mechanism of ske- letal muscle of flounders has been attracting the attention of researchers. We have previously reported the isolation of two spliced isoforms of(and) and three spliced isoforms of,, and) in olive flounder (Jiao., 2015b). In the present study, we investigated their cellular localization, transcriptional activity on the regulation ofgene, androles in myogenesis. The results indicated that different Pax3a and Pax3b isoforms may precisely and collaborative- ly regulate early embryonic myogenesis, but their roles in juvenile myogenesis are uncertain. This research is the first study to analyze the roles of multiple Pax3a and Pax3b iso- forms in myogenesis in fish, which is helpful for further research in the future.

2.1 Ethics Statement

Experiments involving olive flounder and zebrafish were approved by the Animal Care and Use Committee of the Shandong University of Technology and the Institute of Oceanology, Chinese Academy of Sciences.

2.2 Experimental Animals

Zebrafish () were maintained at 28.5℃±0.5℃ in a recirculating aquatic system at a photoperiod of 14h:10h light:dark cycle and fed twice daily. Embryos were obtained by natural crossing. Fertilized eggs were raised in embryo medium at 28.5℃ and staged in accor- dance with a standard method (Kimmel., 1995). The olive flounders used in this study were obtained from a local fish farm in Jiaonan, Qingdao, China. A total of 70 juvenile fish with a mean mass of 102.00g±18.54g were randomly selected and cultured in a 1m3aerated seawater tank and fed with commercial food particles twice a day. The fish were acclimated for one week before the start of electroporation experiments.

2.3 Construction of Plasmids

Each isoform of olive flounderandhas been cloned into pEasy-T3 (TransGen Biotech, Beijing, Chi- na) previously (Jiao., 2015b). The corresponding plas- mids were used as templates to construct the following plasmids.

For subcellular localization analysis, the entire open read- ing frame of each isoform without a stop codon was am- plified by polymerase chain reaction (PCR). The primers are listed in Table 1. Restriction sites ofI andI with protective bases were added to 5’ forward primers and reverse primers, respectively. The Kozak sequence (GCCACC) was added immediately upstream of the start codon of the forward primers. 2×MasterMix (CWBIO,Ltd., Beijing, China) was used. The PCR was performed as follows: initial denaturation at 94℃ for 2min; 35 cy- cles at 94℃ for 30s, 55℃ for 30s, and 72℃ for 2min; a final extension at 72℃ for 5min. The amplified sequences were subcloned into the modified pCS2+ enhanced green fluorescent protein (EGFP) vector (pCS2+NeGFP). The EGFP was fused to the C-terminus of each isoform. All plas- mids were verified by DNA sequencing.

For the luciferase assay, the encoding regions of each isoform with stop codons were amplified by PCR. The pri- mers are listed in Table 1. The restriction sites and Kozak sequence were the same as those in the preparation of pCS2 +NeGFP-related plasmids, and the same DNA polymerase and PCR procedure were applied. The amplified sequences were subcloned into the pCS2+Flag vector. The Flag was fused to the N-terminus of each isoform. All plasmids were verified by DNA sequencing.

For the construction of plasmids used for electropora- tion, the/isoform--SV40 polyA fragments were digested withI andI (Takara, Dalian, Chi- na) from pCS2-NeGFP vectors and cloned into MyoDP- GFP vector (Zhang., 2008)to construct the MyoDP- Pax3a/3b isoform-EGFP plasmids. In these plasmids, the/isoform-fragments were under the control of a muscle-specific promoter. The recombinant plasmids were extracted using an Endo-free Plasmid Kit (CWBIO, Beijing, China) in accordance with the manufacturer’s pro- tocol. The quality and quantity of plasmids were detected by gel electrophoresis and optical density readings with Na- nodrop 2000, respectively (Thermo, USA).

Table 1 Primers for PCR analyses

2.4 Subcellular Localization

The subcellular localization studies were performed fol- lowing a published procedure (Jiao., 2011). Briefly, human cervical carcinoma (HeLa) cells were seeded into sterile coverslips in six-well culture plates, and 2µg plas- mid DNA was transfected into cells. One day after trans- fection, the cells were washed briefly with PBS, stained with 4’,6-diamidino-2-phenylindole (DAPI), and photo- graphed under a fluorescence microscope (Nikon Eclipse 80i; Nikon, Tokyo, Japan).

2.5 Luciferase Reporter Assay

Human embryonic kidney 293T (HEK 293T) cells were cultured in high-glucose Dulbecco’s Modified Eagle Me- dium (DMEM）supplemented with 10% fetal bovine serumin a humidified air atmosphere containing 5% CO2. Be- fore transfection, the cells were seeded in 24-well plates. The pGL3-mouse MyoD promoter vector was provided by Dr. Ping Hu (University of California) (Blake and Ziman, 2014). 300ng pGL3-mouse MyoD promoter vector and 40ng pRL-TK vector were cotransfected with pCS2-Flag, pCS2-Flag-Pax3a-1, pCS2-Flag-Pax3a-2, pCS2-Flag-Pax3b- 1, pCS2-Flag-Pax3b-2, pCS2-Flag-Pax3b-3, or an equiva- lent mixture of two plasmids using polyethylenimine (PEI) reagent. The cells were incubated for 4h with PEI. At 24h after transfection, the cells were washed with PBS and lysed in 100µL luciferase lysis buffer. The lysates were analyzedusing a dual-luciferase reporter assay system (Promega, WI, USA) in accordance with the manufacturer’s instruc- tions. The results were expressed as fold changes over the pCS2-Flag control vector and pRL-TK transfected group. Transfection efficiency was normalized by Renilla luci- ferase activity.

2.6 Overexpression Experiments in Zebrafish Embryos

To examine the possible regulation ofexpression by different Pax3a/3b isoforms, we synthesized capped mRNAs using a commercial kit and linearized plasmid DNA as a template (mMESSAGE mMACHINETMSP6 Transcrip- tion kit; Ambion, Inc.). The synthesized mRNA (500pg per embryo) was microinjected into zebrafish embryos at the 1–2 cell stage as reported previously (Jiao., 2011).ThemRNA injected embryos were used as controls. Af- ter injection, the embryos were placed in embryo-rearing medium and kept at 28.5℃±0.5℃.

2.7 Whole-Mount in situ Hybridization

Single-colorhybridization was performed with di- goxigenin-labeledantisense riboprobe as described previously (Jiao., 2015b). Images were captured with a Leica DFC420C camera mounted on a Leica DMLB2 mi- croscope (Leica, Wetzlar, Germany).

2.8 RNA Isolation and Quantitative Real-Time PCR (qRT-PCR)

The total RNA was extracted using Trizol Reagent (In- vitrogen, USA). The quantity and purity of the RNAs were checked by Nanodrop 2000 and electrophoresis in 1% aga- rose gel. The first-strand cDNA synthesis was carried out by using PrimeScript™ RT reagent Kit with gDNA Eraser (Takara, Dalian, China). The() in zebrafish andmRNA in olive flounder were used as internal references for normalization. The primers used in qRT-PCR are shown in Table 2. The qRT-PCR was con- ducted using TB Green Premix Ex TaqTMII (Tli RNaseH Plus) (Takara, Dalian, China) following the manufacturer’s instructions. The amplification procedure consisted of an initial denaturation step at 95℃ for 30s, 40 cycles of dena- turation at 95℃ for 5s, annealing and extension at 60℃ for 30s, followed by melting curve analysis. Four to six samples in each group were analyzed in triplicate, and the relative gene expression levels were calculated using the 2−ΔΔCtmethod.

2.9 Electroporation in the Muscle of Olive Flounder

The olive flounder were randomly divided into seven groups, with six fish in each group. The fish were firstan- esthetized with 200mgL−1MS-222 (Sigma, St. Louis, MO,USA) in sea water and given an intramuscular injection of a solution of 30µg plasmid DNA (30µL) at the 9th myo- tome of the ventral side in the dorsal part using a 50µL microinjector. After injection, tweezer-type electrodes were used to electroporate DNA into the fish muscle. Electro- poration was carried out using the Super Electroporator NEPA 21 (NEPA GENE, Japan). Electroporation was con-ducted under optimal conditions (voltage: 50 V; pulse length:30ms; pulse interval: 50ms; number of pulses: 3; decay rate: 10%). The fish were returned to seven tanks (50L, six fishper tank) by groups immediately after electroporation. Af- ter 16 days, the electroporated fish were sacrificed after ana- esthetization with 200mgL−1MS222, and the electropo- rated myotomes were dissected. The sampled muscle tis- sues were used for histological evaluation and qRT-PCR analysis.

Table 2 Primer sequences used in qRT-PCR

2.10 Muscle Histology

For histological analysis, the samples were fixed in Da- vidson’s, dehydrated in an ascending gradient of ethanol, cleared in xylene, and embedded in paraffin wax. Sections (5µm) were stained with hematoxylin-eosin (HE). The sec- tions were observed and photographed under a Nikon EC- LIPSE 80i microscope equipped with a digital camera (Ni- kon, Melville, NY, USA).

2.11 Statistics

All data were presented as means±standard error of the mean (SEM). Statistical significance among experimental groups was determined by one-way analysis of variance, followed by Tukey’s multiple comparison test (GraphPad Software version 5.01, San Diego, CA), and the significance level was set at<0.05.

3.1 Differential Subcellular Localizations of Flounder Pax3a and Pax3b Isoforms

The putative nuclear localization signals (NLS) in floun- der Pax3a and Pax3b isoforms were predicted by the on- line server PredictProtein. Two putative NLSs, L(215) KR- KQRR(221) and R(270)RARWRKQ(277), were found at both ends of the HD domain (Fig.1A). Pax3a-1, Pax3a-2, and Pax3b-1 had the above two putative NLSs. However, Pax3b-2 and Pax3b-3 lacked NLSs. We introduced diffe- rent pCS2-Pax3a/3b isoform-NeGFP vectors in HeLa cells to determine their cellular localization. As shown in Fig.1B, the EGFP signals in pCS2-NeGFP, pCS2-Pax3b-2-NeGFP, and pCS2-Pax3b-3-NeGFP-transfected cells were similar- ly observed in the cytoplasm and nucleus (Figs.1B a, e, and f). By comparison, in the pCS2-Pax3a-1-NeGFP and Pax3a-2-NeGFP-transfected cells, EGFP signals were only ob- served in the nucleus (Fig.1B b–c). However, in the pCS2- Pax3b-1-NeGFP-transfected cells, the EGFP signals were mainly expressed in the nucleus and weakly in the cyto- plasm (Figs.1B d). We also introduced the above vectors in flounder skeletal muscle cells (Peng., 2016) and ob- tained similar results (data not shown).

3.2 Differential Regulation of myod Transcriptional Activity by Flounder Pax3a and Pax3b Isoforms

Luciferase activity assay was used to measure the ef- fects of flounder Pax3a/3b isoforms on thepromoter activity. As shown in Fig.2, Pax3a-1, Pax3a-2, and Pax3b-1 all elicited significant increases in luciferase activity (<0.05). For Pax3a-1 and Pax3a-2, no significant differenceswere observed among the relative luciferase activities when the expression vectors increased from 200ng to 800ng (>0.05). The effect of Pax3b-1 was concentration-dependent (<0.05). By comparison, Pax3b-2 caused significant de- creases in luciferase activity at 200 and 400ng, and these inhibitory effects were abolished at 800ng (<0.05). How- ever, with the cotransfection amount of the Pax3b-3 expres- sion vector increasing from 0ng to 800ng, no significant difference was observed among the relative luciferase acti- vities (>0.05). Comparison at the same concentration of 400ng showed that Pax3a-1, Pax3a-2, and Pax3b-1 signifi- cantly increased thetranscriptional activity, with the order of Pax3a-1>Pax3a-2>Pax3b-1 (<0.05), whereasPax3b-2 and Pax3b-3 had no effects.

3.3 Mutual Effects Between Every Two Flounder Pax3a and Pax3b Isoforms on myod Transcriptional Activity

As shown in Fig.3, Pax3a-2, Pax3b-1, Pax3b-2, and Pax3b-3 all can suppress the Pax3a-1-induced luciferase activity at different degrees (<0.05). Pax3b-1, Pax3b-2,and Pax3b-3 all reduced the Pax3a-2-induced luciferaseactivity at the same degree (<0.05). However, Pax3b-2 did not affect the activity of Pax3b-1 on luciferase acti- vity, and Pax3b-3 did not affect that of Pax3b-2. Interest- ingly, Pax3b-1 and Pax3b-3 had a synergistic promoting effect ontranscriptional activity (<0.05).

3.4 Overexpression of Flounder Pax3a and Pax3b Isoforms Differentially Disrupted the Spatial Expression Patterns of myod During Zebrafish Embryonic Myogenesis

Due to the hard and elastic chorion and osmotic pres- sure of fertilized flounder eggs, microinjection was not car- ried out easily. Zebrafish embryos are amenable to gene- tic manipulations. The protein encoded by flounderandgenes share 84.62% and 71.88% similarity with their corresponding zebrafish homologues, respective- ly (Jiao., 2015b). Therefore, the effect of ectopic ex- pression of flounder Pax3a and Pax3b isoforms on embry- onic myogenesis was analyzed in zebrafish throughexpression, a marker of paraxial mesoderm, which lies bi- laterally along the anterior-posterior axis. The ectopic ex- pression of different isoforms caused differential changes inspatial expression patterns (Fig.4). Based on their severity and similarity, we grouped the spatial expression patterns ofinto eight categories. Type 1 embryos had normalexpression patterns, whereas types 2–8 em- bryos displayed differential abnormalexpression pat-terns. In type 2 embryos, the shape ofsignals be- came shorter but was still bilateral in the paraxial meso- derm. In type 3 embryos, thewas predominantly ex- pressed on one side of the paraxial mesoderm with weak- er signals on the other side. However, in type 4 embryos,was only expressed on one side. In type 5 embryos, the shape ofsignals was curved. In type 6 embryos, the shape ofsignals became shorter and outward. In type 7 embryos, the shape ofsignals on one side was outward, and that on the other side almost disappeared. In type 8 embryos, the shape ofsignals was irregularly twisted. In-injected groups, 60.66% of embryos displayed normalexpression patterns, with 6.56%, 21.31%, and 11.48% of embryos having types 3, 4, and 5 abnormalities, respectively. By contrast, in the-in- jected group, 26.45% of embryos had normalex- pression patterns. However, 3.31%, 46.28%, 22.31%, and 1.65% of embryos displayed types 3, 4, 6, and 8 abnorma- lities, respectively. In the-injected group, 62.71% of embryos had normalexpression patterns, and 8.47%, 11.86%, and 16.95% showed types 4, 6, and 7 abnormali- ties, respectively. In,, or equal mixture ofand-injected groups, 92.19%, 85.58%, and 84.42% of embryos displayed normalexpression pat- terns, respectively. A total of 7.81%, 14.42%, and 15.58% of these embryos showed types 2, 7, and 7 abnormalities, respectively. Collectively, the overexpression of different isoforms variably affected thenormal expression. These results suggested that different Pax3aand Pax3b iso-forms may coordinately regulate theexpression in early myogenesis.

Fig.1 Structural analysis of flounder Pax3a and Pax3b isoforms and their cellular localization.(A) Protein structure, pre- dicted molecular weight, and predicted NLS (Nuclear localization sequence) of olive flounder Pax3a and Pax3b isoforms. The predicted NLS is in red letters, and the intact HD sequence is underlined. N, amino terminal; C, carboxy terminal; PD, paired box domain; OP, octapeptide; HD, homeodomain; TAD, transactivation domain; M.W., molecular weight; Da, dal- ton. (B) Cellular localization of olive flounder Pax3a and Pax3b isoforms. HeLa cells transiently transfected with pCS2- NeGFP (a), pCS2-Pax3a-1-NeGFP (b), pCS2-Pax3a-2-NeGFP (c), pCS2-Pax3b-1-NeGFP (d), pCS2-Pax3b-2-NeGFP (e), or pCS2-Pax3b-3-NeGFP (f) plasmids for 24h were stained by 4’, 6-diamidino-2-phenylindole (DAPI). The merged views of EGFP and DAPI were shown. Scale bar, 20μm.

Fig.2 Different isoforms of flounder pax3a and pax3b had different myod transcriptional activity. HEK293T cells were transfected with the indicated plasmid together with the pGL3-mouse MyoD promoter. pRL-SV40, a renilla luciferase plasmid, was cotransfected as an internal control. The results were normalized and expressed as the fold increase over the pCS2 group. (A–E) Dose-dependent effects of flounder Pax3a-1 (A), Pax3a-2 (B), Pax3b-1 (C), Pax3b-2 (D), and Pax3b-3 (E). (F) The effects of different flounder Pax3a and Pax3b isoforms at the same plasmid DNA dose used for transfection (400ng of the indicated plasmids with 400ng of the pCS2 plasmid). Data shown are means±SEM (n=3–6). Different letters indicate significant differences (P<0.05).

Fig.3 Pairwise interaction analysis of flounder Pax3a and Pax3b isoforms on myod transcriptional activity. HEK293T cells were cotransfected with equal amounts of two different plasmids (200ng per plasmid per well for 24-well plates) together with constant pGL3-mouse MyoD promoter and pRL-SV40. The results were normalized and expressed as the fold in- crease over the group transfected with 400ng pCS2 plasmid. Data shown are means±SEM (n=3). Different letters indi- cate significant differences (P<0.05).

3.5 Effects of Flounder Pax3a and Pax3b Isoforms on myod, myogenin,and c-met mRNA Expression in Zebrafish Embryos

We next used qRT-PCR to compare the relative expres- sion levels of the MRFsand(Figs.5A and 5B, respectively). The ectopic expressions ofandresulted in significant increases inandmRNA levels, whereas the expressions of any other isoforms, including,,, and an equal mixture ofand, had no effect onandexpression. We also compared the relative expres- sion levels of, a target gene regulated by Pax3 (Ep- stein., 1996). However, our qRT-PCR data did not show any changes inmRNA expression level (Fig.5C).

Fig.4 Effects of overexpression of flounder Pax3a and Pax3b isoforms on myod expression during zebrafish embryonic myogenesis detected by in situ hybridization. (A) Classification of the spatial expression patterns of myod caused by forced expression of different flounder Pax3a and Pax3b isoforms. Dorsal views are shown with the animal pore to the right. Scale bar: 100µm. The egfp, pax3a-1, pax3a-2, pax3b-1, pax3b-2, pax3b-3, or equal mixture of pax3b-1 and pax3b-3mRNA was injected into 1–2 cell stage embryos (500pg per embryo), respectively. The embryos were raised to 10.5h post fertilization (hpf) and performed in situ hybridization with a myod probe. (B) The phenotypes in each microinjection group were scored based on the categories shown in A. Percentage of embryos in each category is shown. The total embryo number is given on the top.

Fig.5 Effects of forced expression of flounder Pax3a and Pax3b isoforms on myod, myogenin, and c-met expression dur- ing embryonic myogenesis detected by qRT-PCR. The egfp, pax3a-1, pax3a-2, pax3b-1, pax3b-2, pax3b-3, or equal mix- ture of pax3b-1 and pax3b-3 mRNA was injected into 1–2 cell stage embryos (500pg per embryo). The embryos were raised to 28hpf. RNA was isolated and analyzed by qRT-PCR. The relative myod (A), myogenin (B), and c-met (C) mRNA le- vels were shown. 18S ribosomal RNA was used as an internal control. Data are shown as means±SEM (n=4). Different letters indicate significant differences (P<0.05).

3.6 Effects of Flounder Pax3a and Pax3b Isoforms on the Expression of Muscle-Related Genes in Juvenile Olive Flounder

First, different isoforms of the fusion protein of Pax3aand Pax3b-GFP were confirmed to be successfully expres- sed in the skeletal muscle of flounder subjected to injec- tion and electroporation through fluorescent microscopy, and no muscle histological changes were observed by HE staining analysis (Fig.6). Next, to study the effects of dif- ferent isoforms of Pax3a and Pax3bon flounder muscle de- velopment, we used qRT-PCR to analyze the levels of se- veral muscle-related genes (Fig.7). The results showed thatnone of the Pax3a and Pax3b isoforms could change the mRNA levels of(), a negative regulator of muscle growth (Fig.7A). These isoforms also could not change the mRNA levels of myogenic factors,, and(Figs.7B–7D, respectively). The mRNA le- vels of duplicatedgenes, namely,and, which are markers for satellite cells, were not changed ei- ther (Figs.7E and 7F, respectively).

As a transcription factor, Pax3 needs to enter the nucleus and activate the transcription of specific genes. In olive flounder, two duplicatedgenes (and) en- coding five isoforms (,,,, and) were isolated previously (Jiao., 2015b).Hence, the first step was to determine if these multiple iso- forms could enter the nucleus. The NLS prediction results suggested that Pax3a-1, Pax3a-2, and Pax3b-1 had two pu- tative NLSs at both ends of the HD domain. By contrast, Pax3b-2 and Pax3b-3 lacked NLSs because they were both truncated before the NLSs sites. When we transfected them into human or fish cells, all of them were detected in the nucleus. Moreover, Pax3b-1, Pax3b-2, and Pax3b-3 were also detected in the cytoplasm. Protein synthesis occurs in the cytoplasm, and transcription factors must transit the nu- clear pore complexes (NPC) in the nuclear membrane to enter the nucleus (Hampoelz., 2019). Molecules smal- ler than 40kDa may pass the NPC by passive diffusion, whereas larger molecules require specific NLSs to be tar- geted specifically to the nucleusactive transport (Ef- tekharzadeh., 2018). The predicted molecular masses of flounder Pax3a-1, Pax3a-2, Pax3b-1, Pax3b-2, and Pax3b-3 are 54.05, 51.27, 54.01, 22.51, and 19.73kDa, respective- ly. Accordingly, Pax3a-1, Pax3a-2, and Pax3b-1 may enter the nucleus by active transport because they not only have NLSs but are also larger than 40kDa. Pax3b-2 and Pax3b- 3 may diffuse passively between the nucleus and cytoplasm given their lack of NLSs and their size, which is smaller than 40kDa.

Fig.6 The successfully electroporated flounder Pax3a and Pax3b isoforms had no apparent effects on myofiber morpho- logy in juvenile flounder. The EGFP(A–A’’), Pax3a-1-EGFP (B–B’’), Pax3a-2-EGFP(C–C’’), Pax3b-1-EGFP(D–D’’),Pax3b-2-EGFP (E–E’’), Pax3b-3-EGFP(F–F’’), or an equal mixture of Pax3b-1-EGFP and Pax3b-3-EGFP (G–G’’) plasmids was electroporated into the trunk skeletal muscle of juvenile flounder (30μg per fish). The GFP signals (A, B, C, D, E, F, and G), HE staining results (A’, B’, C’, D’, E’, F’, and G’), and the corresponding merged (A’’, B’’, C’’, D’’, E’’, F’’, and G’’) pictures are shown. Scale bars: 20µm.

Fig.7 The effects of flounder Pax3a and Pax3b isoforms on muscle-related gene expression in juvenile flounder detected by qRT-PCR. The EGFP, Pax3a-1-EGFP, Pax3a-2-EGFP, Pax3b-1-EGFP, Pax3b-2-EGFP, Pax3b-3-EGFP, or equal mixture of Pax3b-1-EGFP and Pax3b-3-EGFP plasmids was successfully electroporatedinto the trunk skeletal muscle of juvenile olive flounder (30μg per fish). The mstn (A), myf5 (B), myod (C), myogenin (D), pax7a (E), and pax7b (F) mRNA levels in flounder muscle, as normalized to that of β-actin, are shown. Expression was determined by qRT-PCR. Data are shown as means±SEM, n=6.

Flounder Pax3a-1, Pax3a-2, and Pax3b-1 could variably activatetranscriptional activity. However, Pax3b-2 and Pax3b-3 showed an inhibitory or no effect. The full- length Pax3 protein contains a PD, an OP, an HD, and a TAD. The PD and HD domains are mainly involved in DNA binding, and the OP domain is involved in protein- protein interactions (Mayran., 2015). The TAD domain is a proline-, threonine-, and serine-rich region at the car- boxy terminus of Pax3 protein and mediates transcriptionalregulation (Niu., 2018). Flounder Pax3a-1 and Pax3b- 1 contain PD, OP, HD, and TAD domains. However, Pax3a- 2 lacks 24 amino acids in the PD domain, Pax3b-2 lacks the whole HD and TAD domains, and Pax3b-3 lacks both of them (Jiao., 2015b). Thus, flounder Pax3a-1 and Pax3b-1 can bind to DNA and activatetranscrip- tional activity. The results of the experiments confirmed the inference. Flounder Pax3a-2 has a partial deletion in the PD domain, but it can bind DNA and transcriptionally ac- tivatewith the aid of HD and TAD domains. By con- trast, flounder Pax3b-2 and Pax3b-3 have no transcriptional activity due to the absence of the TAD domain. Thus, the HD/TAD domain has a transcriptional activation function. In this study, flounder Pax3b-2 showed an inhibitory effect ontranscriptional activity in a certain range of do- sage. This unexpected result may be explained by the spe- culation that Pax3b-2 binding to DNA may inhibit the ba- sal level oftranscriptional activity. Or, it may bethat the Pax3b-2 competitively binds to DNA with othertranscription factors withtranscriptional activity. Thetranscriptional function of Pax proteins is likely to be fine- ly tuned by variations in the combinatorial and relative activity between activation and repression functions, and these domains can be further modified through alternative splicing (Chi and Epstein, 2002). In this study, the mutual effects between every two flounder Pax3a/3b isoforms ontranscriptional activity were tested, and most of themwere easy to understand. Pax3b-1 and Pax3b-3 displayed a synergistic promoting effect ontranscriptional acti- vity. The mutual effects may tightly regulatetran- scription, not to mention there are at least five Pax3aand Pax3b isoforms. The multiple isoforms may competitively bind to DNA to fine-tune the expression of downstream genes under the right conditions.

Pax3 is a key upstream regulator of the onset of embry- onic myogenesis, controlling progenitor cell survival, beha- vior, and entry into the myogenic program (Lagha., 2010;Buckingham, 2017). In C2C12 cells, Pax3 directly binds the sequences that regulate the expression of(Hu., 2008). In addition, we revealed the different roles of flounder Pax3a/3b isoforms in the transcriptional acti- vation ofgene. Nevertheless, whether the dif- ferent flounder Pax3a/3b isoforms are related to early em- bryonic myogenesis is unknown. We observed that the over- expression of different isoforms caused differential changes inexpression patterns in early myogenesis. Notably, the ectopic expression of flounder Pax3a-2 displayed the highest levels of abnormalexpression with an un- known mechanism. MyoD and Myf5 directed proliferat- ing myogenic progenitor cells toward a myogenic lineage, whereas myogenin and MRF4 controlled the differentiation and fusion of myoblasts to form myofibers (Buckingham, 2017). The ectopic expression of different flounder Pax3a/ 3b isoforms differentially regulatedandmRNA levels in embryonic myogenesis. However, the changes inmRNA levels did not correspond to the normal or abnormal spatialod patterns, possibly indica- ting the tight regulation ofexpression by different isoforms. We also tested the mRNA level of, which is a direct target of Pax3 and controls the delamination and migration of muscle progenitor cells (Buckingham., 2003). We showed that the ectopic expression of different flounder Pax3a/3b isoforms could not change themRNA levels. Perhaps, theinduced by endogenous zebrafish Pax3a and Pax3b has reached its maximum thres-hold. Thus, the overexpression of flounder Pax3a/3b iso-form had no effects. However, the mechanism is still un- known. These findings suggest that the existence of multiple flounder Pax3a/3bisoforms contributes to the fine re- gulation of embryonic myogenesis.

We also tested whether the ectopic expressions of diffe- rent flounder Pax3a/3b isoforms could affect the expressionof muscle-related genes in juvenile flounders. Neither changes in,,,,, andmRNA levels nor changes in muscle histology were de- tected. Pax3 plays a major role during embryonic myoge- nesis, whereas Pax7 predominates during postnatal growthand muscle regeneration in the adult (Buckingham and Re-laix, 2015). Pax3 cannot substitute for Pax7 in the post- natal skeletal muscle (Lagha., 2008b). Pax3 can also induce the differentiation of juvenile skeletal muscle stem cells without the transcriptional upregulation of canonical MRFs,,, and(Young and Wa-gers, 2010). These earlier studies raised at least three pos- sibilities for interpreting our results: 1) Pax7, not Pax3, plays a key role in juvenile muscle growth; therefore, ec- topic flounder/isoforms failed to induce expres- sion of any of the above factors; 2) ectopic flounder Pax3a/3b isoforms may induce muscle satellite cell differentiation without regulating any of the tested factors; 3) a small mi- nority of cells respond to ectopic flounder Pax3a/3b isoform expression by regulating the MRFs and differentiating. We roughly estimated 20%–30% of the GFP signals in the field of vision of the muscle histological section (data not shown). In this study, we studied the changes of the mRNAs of,,,,, andgenes, but we did not study the changes of their corresponding proteins due to the lack of appropriate antibodies. The abun- dance of these proteins may be affected by the ectopic ex- pressions of different flounder Pax3a/3b isoforms in juve- nile flounders. This is only the preliminary study of floun- der Pax3a/3b isoforms in juvenile flounders, and further studies are needed to better characterize their roles and un- derlying mechanisms.

In summary, different flounder Pax3aand Pax3b iso- forms showed differential subcellular localizations corres- ponding to their distinct protein structures. They display- ed differential and mutual effects ontranscriptional activity. The ectopic isoforms differentially changed the spa- tial expression patterns ofand differentially regulatedthe expression levels of their target genes (,, and)in zebrafish embryonic myogenesis. However, none of them could change the mRNA expression levels of,,,,, andin the electro- porated muscles of juvenile flounder. This research may be the first study to investigate the functions of different Pax3 isoforms that coordinately regulate muscle develop- ment in fish.

This study was supported by the National Natural Sci- ence Foundation of China (Nos. 31972774, 31502146, 316 72636), and the Key Research and Development Program of Shandong Province, China (No. 2019GHY112007).

Akolkar, D. B., Asaduzzaman, M., Kinoshita, S., Asakawa, S., and Watabe, S., 2016. Characterization ofandgenesand their expression patterns during different development and growth stages of Japanese pufferfish., 575: 21-28, DOI: 10.1016/j.gene.2015.08.031.

Asfour, H. A., Allouh, M. Z., and Said, R. S., 2018. Myogenic regulatory factors: The orchestrators of myogenesis after 30 years of discovery., 243: 118-128, DOI: 10.1177/1535370217749494.

Bajard, L., Relaix, F., Lagha, M., Rocancourt, D., Daubas, P., and Buckingham, M. E., 2006. A novel genetic hierarchy func-tions during hypaxial myogenesis: Pax3 directly activates, 20: 2450-2464, DOI: 10.1101/gad.382806.

Barber, T. D., Barber, M. C., Cloutier, T. E., and Friedman, T. B., 1999.gene structure, alternative splicing and evolution., 237: 311-319, DOI: 10.1016/s0378-1119(99)00339-x.

Blake, J. A., and Ziman, M. R., 2014. Pax genes: Regulators oflineage specification and progenitor cell maintenance., 141: 737-751, DOI: 10.1242/dev.091785.

Buckingham, M., 2017. Gene regulatory networks and cell line- ages that underlie the formation of skeletal muscle., 114: 5830-5837, DOI: 10.1073/pnas.1610605114.

Buckingham, M., and Relaix, F., 2015. PAX3 and PAX7 as up- stream regulators of myogenesis., 44: 115-125, DOI: 10.1016/j.semcdb.2015.09.017.

Buckingham, M., Bajard, L., Chang, T., Daubas, P., Hadchouel, J., Meilhac, S.,., 2003. The formation of skeletal muscle: From somite to limb., 202: 59-68, DOI: 10.1046/j.1469-7580.2003.00139.x.

Chang, N. C., and Rudnicki, M. A., 2014. Satellite cells: The ar- chitects of skeletal muscle., 107: 161-181, DOI: 10.1016/B978-0-12-416022-4.00006-8.

Chi, N., and Epstein, J. A., 2002. Getting your Pax straight: Pax proteins in development and disease., 18: 41-47, DOI: 10.1016/s0168-9525(01)02594-x.

Eftekharzadeh, B., Daigle, J. G., Kapinos, L. E., Coyne, A., Schi- antarelli, J., Carlomagno, Y.,., 2018. Tau protein disrupts nucleocytoplasmic transport in Alzheimer’s disease., 99: e927, DOI: 10.1016/j.neuron.2018.07.039.

Epstein, J. A., Shapiro, D. N., Cheng, J., Lam, P. Y., and Maas, R. L., 1996. Pax3 modulates expression of the c-Met receptor during limb muscle development., 93: 4213-4218, DOI: 10.1073/pnas.93.9.4213.

Figueiredo, M. A., Mareco, E. A., Silva, M. D., and Marins, L. F., 2012. Muscle-specific growth hormone receptor (GHR) over- expression induces hyperplasia but not hypertrophy in trans- genic zebrafish., 21: 457-469, DOI: 10.1007/s11248-011-9546-2.

Goulding, M. D., Chalepakis, G., Deutsch, U., Erselius, J. R., and Gruss, P., 1991. Pax-3, a novel murine DNA binding protein expressed during early neurogenesis., 10: 1135-1147, DOI: 10.1002/j.1460-2075.1991.tb08054.x.

Hampoelz, B., Andres-Pons, A., Kastritis, P., and Beck, M., 2019. Structure and assembly of the nuclear pore complex., 48: 515-536, DOI: 10.1146/annurev-biophys-052118-115308.

Hu, P., Geles, K. G., Paik, J. H., DePinho, R. A., and Tjian, R., 2008. Codependent activators direct myoblast-specific MyoD transcription., 15: 534-546, DOI: 10.1016/j.devcel.2008.08.018.

Imbriano, C., and Molinari, S., 2018. Alternative splicing of tran-scription factors genes in muscle physiology and pathology., 9: 107, DOI: 10.3390/genes9020107.

Jiao, S., Dai, W., Lu, L., Liu, Y., Zhou, J., Li, Y.,., 2011. Theconserved clusterin gene is expressed in the developing choroid plexus under the regulation of notch but not IGF signaling in zebrafish., 152: 1860-1871, DOI: 10.1210/en.2010-1183.

Jiao, S., Tan, X., Li, M., Sui, Y., Du, S. J., and You, F., 2015a. The duplicated() genes differentially transcribed during Japanese flounder () embryogenesis., 189: 62-68, DOI: 10.1016/j.cbpb.2015.08.003.

Jiao, S., Tan, X., Wang, Q., Li, M., and Du, S. J., 2015b. The olive flounder ()homologues are highly conserved, encode multiple isoforms and show unique expression patterns., 180: 7-15, DOI: 10.1016/j.cbpb.2014.10.002.

Kimmel, C. B., Ballard, W. W., Kimmel, S. R., Ullmann, B., and Schilling, T. F., 1995. Stages of embryonic development of the zebrafish., 203: 253-310, DOI: 10.1002/aja.1002030302.

Koblar, S. A., Murphy, M., Barrett, G. L., Underhill, A., Gros, P., and Bartlett, P. F., 1999.regulates neurogenesis in neu- ral crest-derived precursor cells., 56: 518-530, DOI: 10.1002/(SICI)1097-4547(19990601)56:5<518::AID-JNR7>3.0.CO;2-8.

Lagha, M., Kormish, J. D., Rocancourt, D., Manceau, M., Ep- stein, J. A., Zaret, K. S.,., 2008a. Pax3 regulation of FGF signaling affects the progression of embryonic progenitor cells into the myogenic program., 22: 1828-1837, DOI: 10.1101/gad.477908.

Lagha, M., Mayeuf-Louchart, A., Chang, T., Montarras, D., Ro- cancourt, D., Zalc, A.,., 2013.is a Pax3 target gene, expressed at sites of skeletal muscle formation., 8: e63143, DOI: 10.1371/journal.pone.0063143.

Lagha, M., Sato, T., Bajard, L., Daubas, P., Esner, M., Montarras, D.,., 2008b. Regulation of skeletal muscle stem cell be- havior by Pax3 and Pax7., 73: 307-315, DOI: 10.1101/sqb.2008.73.006.

Lagha, M., Sato, T., Regnault, B., Cumano, A., Zuniga, A., Licht, J.,., 2010. Transcriptome analyses based on genetic screens for Pax3 myogenic targets in the mouse embryo., 11: 696, DOI: 10.1186/1471-2164-11-696.

Li, M., Tan, X., Sui, Y., Jiao, S., Wu, Z., Wang, L.,., 2017. The stimulatory effect of neuropeptide Y on growth hormone expression, food intake, and growth in olive flounder ()., 43: 11-18, DOI: 10.1007/s10695-016-0263-x.

Lister, J. A., Capper, A., Zeng, Z., Mathers, M. E., Richardson, J., Paranthaman, K.,., 2014. A conditional zebrafish MITF mutation reveals MITF levels are critical for melanoma pro- motion. regression., 134: 133-140, DOI: 10.1038/jid.2013.293.

Mayran, A., Pelletier, A., and Drouin, J., 2015. Pax factors in tran-scription and epigenetic remodelling., 44: 135-144, DOI: 10.1016/j.semcdb.2015.07.007.

McKinnell, I. W., Ishibashi, J., Le Grand, F., Punch, V. G., Ad- dicks, G. C., Greenblatt, J. F.,., 2008. Pax7 activates myo-genic genes by recruitment of a histone methyltransferase com-plex., 10: 77-84, DOI: 10.1038/ncb1671.

Minchin, J. E., and Hughes, S. M., 2008. Sequential actions of Pax3 and Pax7 drive xanthophore development in zebrafish neural crest., 317: 508-522, DOI: 10.1016/j.ydbio.2008.02.058.

Niu, Z., Li, J., Tang, F., Sun, J., Wang, X., Jiang, L.,., 2018. Identification and functional analysis of a novel mutation in the PAX3 gene associated with Waardenburg syndrome type I., 642: 362-366, DOI: 10.1016/j.gene.2017.11.035.

Paixao-Cortes, V. R., Salzano, F. M., and Bortolini, M. C., 2013. Evolutionary history of chordategenes: Dynamics of change in a complex gene family., 8: e73560, DOI: 10.1371/journal.pone.0073560.

Paixao-Cortes, V. R., Salzano, F. M., and Bortolini, M. C., 2015. Origins and evolvability of thefamily., 44: 64-74, DOI: 10.1016/j.semcdb.2015.08.014.

Peng, L. M., Zheng, Y., You, F., Wu, Z. H., Tan, X., Jiao, S.,., 2016. Comparison of growth characteristics between skeletal muscle satellite cell lines from diploid and triploid olive floun-der, 4: e1519, DOI: 10.7717/peerj.1519.

Relaix, F., Rocancourt, D., Mansouri, A., and Buckingham, M., 2005. A Pax3/Pax7-dependent population of skeletal muscle progenitor cells., 435: 948-953, DOI: 10.1038/nature03594.

Schnapp, E., Pistocchi, A. S., Karampetsou, E., Foglia, E., Lamia,C. L., Cotelli, F.,., 2009. Induced early expression ofbut notrescues myogenesis in the/double-morphant zebrafish embryo., 122: 481-488, DOI: 10.1242/jcs.038356.

Seo, H. C., Saetre, B. O., Havik, B., Ellingsen, S., and Fjose, A., 1998. The zebrafishandhomologues are highly con-served, encode multiple isoforms and show dynamic segment-like expression in the developing brain., 70: 49-63, DOI: 10.1016/s0925-4773(97)00175-5.

Tremblay, P., Dietrich, S., Mericskay, M., Schubert, F. R., Li, Z., and Paulin, D., 1998. A crucial role forin the develop- ment of the hypaxial musculature and the long-range migra- tion of muscle precursors., 203: 49-61, DOI: 10.1006/dbio.1998.9041.

Venkatesh, B., 2003. Evolution and diversity of fish genomes., 13: 588-592, DOI: 10.1016/j.gde.2003.09.001.

Wang, Q., Fang, W. H., Krupinski, J., Kumar, S., Slevin, M., and Kumar, P., 2008.genes in embryogenesis and oncogenesis., 12: 2281-2294, DOI: 10.1111/j.1582-4934.2008.00427.x.

Williams, B. A., and Ordahl, C. P., 1994. Pax-3 expression in seg-mental mesoderm marks early stages in myogenic cell speci- fication., 120: 785-796.

Young, A. P., and Wagers, A. J., 2010. Pax3 induces differentia- tion of juvenile skeletal muscle stem cells without transcrip- tional upregulation of canonical myogenic regulatory factors., 123: 2632-2639, DOI: 10.1242/jcs.061606.

Zhang, Y. Q., Tan, X. G., Xu, P., Sun, W., Xu, Y. L., and Zhang, P. J., 2010. Quantitative comparison of the expression of myo-genic regulatory factors in flounder () embryos and adult tissues., 28: 248-253, DOI: 10.1007/s00343-010-9286-9.

Zhang, Y., Tan, X., Zhang, P. J., Sun, W., Xu, P., and Xu, Y. L., 2008. Characterization of promoters from flounder () muscle-regulatory genes,and., 32: 31-35 (in Chinese with English abstract).

Zhong, Q. W., Zhang, Q. Q., Chen, Y. J., Sun, Y. Y., Qi, J., Wang, Z. G.,, 2008. The isolation and characterization of myo- statin gene in Japanese flounder (): Ubi- quitous tissue expression and developmental specific regula- tion., 280: 247-255, DOI: 10.1016/j.aquaculture.2008.04.015.

February 17, 2021;

April 1, 2021;

October 12, 2021

© Ocean University of China, Science Press and Springer-Verlag GmbH Germany 2022

. E-mail: jiaoshuang@sdut.edu.cn

E-mail: tanx@qdio.ac.cn

(Edited by Qiu Yantao)