Salt Sensitive Tet-Off-Like Systems to Knockdown Primordial Germ Cell Genes for Repressible Transgenic Sterilization in Channel Catﬁsh, Ictalurus Punctatus

marine drugs

Article Salt Sensitive Tet-Off-Like Systems to Knockdown Primordial Germ Cell Genes for Repressible Transgenic Sterilization in Channel Catﬁsh, Ictalurus punctatus

Hanbo Li 1,†, Baofeng Su 1,2,†, Guyu Qin 1, Zhi Ye 1, Ahmed Alsaquﬁ 1,3, Dayan A. Perera 1,4, Mei Shang 1,2, Ramjie Odin 1, Khoi Vo 1, David Drescher 1, Dalton Robinson 1, Dan Zhang 1, Nermeen Abass 1,5 and Rex A. Dunham 1,*

1 School of Fisheries, Aquaculture and Aquatic Sciences, Auburn University, Auburn, AL 36849, USA; [email protected] (H.L.); [email protected] (B.S.); [email protected] (G.Q.); [email protected] (Z.Y.); [email protected] (A.A.); [email protected] (D.A.P.); [email protected] (M.S.); [email protected] (R.O.); [email protected] (K.V.); [email protected] (D.D.); [email protected] (D.R.); [email protected] (D.Z.); [email protected] (N.A.) 2 National and Local United Engineering Laboratory for Freshwater Fish Breeding, Heilongjiang River Fisheries Research Institute, Chinese Academy of Fisheries Sciences, Harbin 150070, China 3 Department of Aquaculture and Animal production, King Faisal University, Al Ahsa 31982, Saudi Arabia 4 Research and Development Corporation, Gus R. Douglass Institute, West Virginia State University, WV 25112, USA 5 Department of Agricultural Botany, Faculty of Agriculture Saba-Basha, Alexandria University, Alexandria City, P.O. Box 2153, Egypt * Correspondence: [email protected]; Tel.: +334-8442345 † These authors contributed equally to this work.

Academic Editors: Allen Place, Rosemary Jagus, Joy Watts and Paul Long Received: 1 April 2017; Accepted: 25 May 2017; Published: 31 May 2017

Abstract: Repressible knockdown approaches were investigated for transgenic sterilization in channel catfish, Ictalurus punctatus. Two primordial germ cell (PGC) marker genes, nanos and dead end, were targeted for knockdown, and an off-target gene, vasa, was monitored. Two potentially salt sensitive repressible promoters, zebrafish adenylosuccinate synthase 2 (ADSS) and zebrafish racemase (Rm), were each coupled with four knockdown strategies: ds-sh RNA targeting the 50 end (N1) or 30 end (N2) of channel catfish nanos, full-length cDNA sequence of channel catfish nanos for overexpression (cDNA) and ds-sh RNA targeting channel catfish dead end (DND). Each construct had an untreated group and treated group with sodium chloride as the repressor compound. Spawning rates of full-sibling P1 fish exposed or not exposed to the constructs as treated and untreated embryos were 93% and 59%, respectively, indicating potential sterilization of fish and repression of the constructs. Although the mRNA expression data of PGC marker genes were inconsistent in P1 fish, most F1 individuals were able to downregulate the target genes in untreated groups and repress the knockdown process in treated groups. The results indicate that repressible transgenic sterilization is feasible for reproductive control of fish, but more data from F2 or F3 are needed for evaluation.

Keywords: transgenic sterilization; repressible by sodium chloride; PGC migration; RNAi

Mar. Drugs 2017, 15, 155; doi:10.3390/md15060155 www.mdpi.com/journal/marinedrugs Mar. Drugs 2017, 15, 155 2 of 23

1. Introduction Since the commercialization of AquaBounty Technologies’s AquAdvantage® Salmon [1], the control of escapement of transgenic offspring has received as much attention as the advantages of transgenic fish. A major concern regarding the use of transgenic fish is potential ecological impact from escapees or released individuals on natural populations [2,3]. All transgenic organisms should be adequately confined both physically and genetically to minimize environmental risk. Physical confinement cannot guarantee that transgenic fish will never establish in the wild since the possibility of theft, human error and catastrophic events could compromise the physical confinement, leading to escape of transgenic animals from aquaculture systems [4]. A genetic confinement system that requires human intervention to allow brood stock production and fertility overcomes these weaknesses. Obviously, the use of both and redundant systems further decreases the risk of establishment of transgenes. If used outside their native geographical range, all species are potentially invasive. Different permutations of transgenic sterilization can potentially be used to prevent spread of invasive species. The embryonic precursors of the gametes are primordial germ cells (PGCs), their origin is distant from the developmental site of the gonads, and they must migrate to the embryo’s genital ridge [5]. PGCs will eventually die without migration, but the death has no effect on somatic development. As a result, disrupting the migration of PGCs is a potentially viable approach to complete transgenic sterilization. In zebrafish (Danio rerio), a number of genetic markers are associated with PGC migration, such as vasa, nanos (nos), askopos, dead end and dazl. Knock out of those markers genes can prevent or disrupt the migration of PGCs [6–10]. Prevention of PGC migration invariably produces sterile fish [8–12]. Nanos is essential for proper migration and survival of PGCs in zebrafish [5,6]. Although the formation of PGCs does not require nanos activity, PGCs with nanos deficiency demonstrate abnormal conditions such as lack of colonization in the gonad, premature activation of germ cell markers, abnormal morphology and expression of mRNAs that are normally expressed in the soma. Vasa is an RNA binding protein with an RNA dependent helicase. Long-term and continual expression in the germ line makes vasa a unique PGC marker [5,13,14]. As a PGC marker, the role and functions of vasa are not as conserved as other markers as it may have evolved to have different functions in different species, which may explain its longer duration of expression compared to other PGC markers [9,15]. In zebrafish, vasa encodes an RNA helicase, and is essential for the assembly of the germplasm and the migration of PGCs [16]. Another function of vasa is to overcome the repressive effect of nanos translational control element, an evolutionarily conserved dual stem-loop structure in the 30 UTR, which acts independently of the localization signal to repress translation of nanos mRNA [17]. Dead end protein is a RNA-binding factor that positively regulates gene expression by prohibiting micro RNA (miRNA)-mediated gene suppression [18]. In zebrafish, miR-430 is a miRNA responsible for the downregulation of several hundred targeted mRNAs in soma cells including nanos. By binding to U-rich sequence elements in 30 UTR of those mRNAs and inhibiting the activity of miR-430, dead end protein protects those mRNAs from degradation in the germ cells., thus, relieving miRNA repression in germline cells by blocking the accessibility of target mRNAs to miRNAs. As the first factor to play a specific role in the initiation of PGC mobility, knockdown of dead end terminates the initial migration step of PGCs, dorsal movement within the deep blastoderm [12]. RNAi is frequently used in molecular biotechnology for targeted gene knockdown. Different strategies such as double-stranded RNA (dsRNA), short hairpin RNA (shRNA), miRNA and small interfering RNA (siRNA) can be applied to achieve the goal [19]. When exogenous RNA, which is synthesized with a sequence complementary to a gene of interest, is expressed in vivo and the RNAi pathway activated, the target gene is silenced [20]. Unlike the use of dsRNA and shRNA for RNAi, overexpression of a target gene via full-length cDNA is another option to inhibit target genes. This process should trigger a negative feedback loop or have undesired malicious effects, leading to disruption of the stoichiometry of the protein complex and loss of function. In the case of the negative Mar. Drugs 2017, 15, 155 3 of 23 feedback loop, the excess nanos cDNA would be detected and the response would be the interruption of nanos mRNA transcription or protein translation. The effect of the response is the downregulation of nanos expression or degradation of nanos protein, resulting in nanos knockdown. Function of the bone morphogenetic protein 2 (BMP-2) gene in zebrafish, common carp and channel catfish has been successfully disrupted using the cDNA overexpression approach [21]. Bernick et al. [22] overexpressed Unc-45b in zebrafish embryos, which resulted in defective thick filament organization. Overexpression could complement loss-of-function screens, and has dominant effects [23]. Modified Tet-off system has been applied to produce repressible transgenic sterile fish and oysters [21], although these systems have yet to be shown 100% effective or commercially feasible. Theoretically, the Tet-off system should be an effective repressible sterility system for preventing the establishment of feral populations of transgenic animals or exotic species [24]; however, the antibiotic repressors used in the system, tetracycline and doxycycline, are expensive when used in large-scale applications. In addition, accumulation or release of large amounts of antibiotics may result in antibiotic resistance of pathogens in the aquaculture environment, as well as the presence of antibiotics in drinking water. Additionally, the Tet-off transgenes contain small viral sequences that might be negatively perceived by consumers. Environmentally friendly and efficient Tet-off-like systems are needed, and promoters that can be repressed by non-antibiotic chemicals are preferred in the new systems. Sodium chloride based Tet-off like systems could be such a system. Channel catfish embryos can tolerate 6 ppt of salinity in hatchery solutions [25], sodium chloride is commonly used in catfish culture to control disease and address water quality problems such as high nitrite [26]. It is inexpensive, easy to apply and when used in small quantities would cause no environmental damage. Thus, salt sensitive promoters could be viable replacements in Tet-off-like system, such as zebrafish adenylosuccinate synthase 2 (adss2) and racemace. These promoters were salt sensitive and can be repressed by 4 ppt NaCl in the laboratory [27], and thus, may be very appropriate for repressible sterilization in freshwater fish. The knockdown constructs driven by a salt-sensitive promoter could be introduced into the fish. The PGC knockdown construct expresses, prevents PGC migration, and thus, sterilizes embryos not exposed to salt. These would be the production fish that are sterile and cannot reproduce if they were to escape into the natural environment. A portion of the embryos would be treated with salt, which should repress the knockdown construct and allow normal sexual development for the development of brood stock. If these fish escape, they can only produce sterile offspring, thus the transgene cannot become established in the receiving population and is present for one generation or less. The objectives of the this study were to use transgenic technology to prevent PGC migration and gamete formation, leading to sterility in channel catfish, and then repress this process to produce fertile brood stock that generate sterile offspring. Specifically, zebrafish adss2 and racemase promoters were evaluated to control the expression of ds-shRNA or cDNA constructs to knockdown targeted PGC migration related genes nanos and dead end, and their repression evaluated by dosing with sodium chloride, towards the ultimate objective of successful repressible transgenic sterilization. Spawning success was measured in treated and untreated P1 brood stock, and PGC marker gene expression was determined in salt treated and untreated P1 and F1 generations.

2. Results

2.1. Expression of Nanos, Vasa and Dead End in Channel Catﬁsh Embryos At 48 and 120 hours post fertilization (hpf), nanos expression was declining relatively slowly, −1.75X and −4.21X, respectively, relative to expression levels present at 24 hpf. The level of vasa was 7.36X lower at 48 hpf than at 24 hpf, and was dramatically lower, −61.11X at 120 hpf compared to 24 hpf. Dead end was downregulated 3.35X at 48 hpf and 12.91X at 120 hpf compared to 24 hpf (Figure1). Mar. Drugs 2017, 15, 155 4 of 23 Mar. Drugs 2017, 15, x 4 of 25

nanos vasa dnd 0.00 ‐1.75 ‐4.21 ‐3.35 ‐10.00 ‐7.36 ‐12.91 ‐20.00 expression

gene ‐30.00 ‐ 48 Hour

‐40.00 120 Hour marker

PGC ‐50.00

‐60.00 Relative ‐61.11 ‐70.00 Marker genes

FigureFigure 1. Relative1. Relative expression expression of primordialof primordial germ germ cell (PGC)cell (PGC) marker marker genes, genes,nanos ,nanosvasa,, andvasadead, end (dndand) in dead non-transgenic end (dnd) in non channel‐transgenic catﬁsh channel (Ictalurus catfish punctatus (Ictalurus) embryos.punctatus) Theembryos. samples The weresamples analyzed were at 24, 48analyzed and120 h at post 24, 48 fertilization and120 h post (hpf). fertilization Relative nanos(hpf)., Relativevasa and nanosdnd,genes vasa and expression dnd genes at expression 48 hpf and at 120 hpf were48 hpf expressed and 120 ashpf fold were change expressed over as 24fold hpf change sample over as 24 normalized hpf sample as to normalized change in to the change expression in the of 18 s control.expression All the of red18 s fold control. values All numbers the red fold shown values were numbers signiﬁcant shown at thewere level significant of p < 0.05 at the using level Pairwise Fixedof Reallocationp < 0.05 using Randomization Pairwise Fixed Test (PFRR)Reallocation from RESTRandomization (Relative ExpressionTest (PFRR) Software from REST Tool). (Relative Expression Software Tool).

2.2. P1 Spawning 2.2. P1 Spawning TheThe overall overall spawning spawning rates rates were were significantly significantly different different ( p(p< < 0.05) 0.05) for for pooled pooled untreated untreated andand treated groups,treated 59% groups, and 59% 93% and (Table 93%1 ),(Table respectively. 1), respectively. P1 fish P1 are fish always are always mosaic mosaic by by tissue tissue and and by by cells cells within tissues.within Whentissues. exogenous When exogenous DNA DNA is introduced is introduced to fishto fish embryos, embryos, it it replicatesreplicates and and is ispresent present in in the cytoplasmthe cytoplasm for a for long a long period period during during development development before before it itis isdegraded. degraded. Therefore, Therefore, transgenes transgenes that that influenceinfluence embryonic embryonic traitstraits can can have have a phenotypic a phenotypic effect effect on the on F1 the through F1 through exposure exposure alone and alone do not and do notnecessarily necessarily need need to tobe beintegrated integrated for forproof proof of principle. of principle. We have We haveunpublished unpublished data, as data, well asas well as publishedpublished data data [21 [21],], indicating indicating phenotypic phenotypic change change in in embryonic embryonic traits traits by byexposure exposure (electroporating) (electroporating) of of the the embryos to transgenes. Transfection rates are typically 20%–50% [28,29] in P1. the the embryos to transgenes. Transfection rates are typically 20–50% [28,29] in P . Fish from three, ADSSN1, ADSSN2 and RMDND, of the eight constructs in the untreated1 group had spawning rates lower than 50%, while females from five of the eight constructs in the treated Table 1. Spawning rates of P female channel catfish (Ictalurus punctatus) electroporated with constructs group had 100% spawning rates.1 There were significant differences (p < 0.05) between treated and designed to disrupt primordial germ cell migration and then either untreated or treated with sodium untreated groups, for ADSSN2 construct as well as in pooled groups of all constructs. chloride expected to repress the expression of the transgene. Construct components were: ADSS,

zebrafishTable 1.adenylosuccinate Spawning rates Synthaseof P1 female 2 (adss2) channelpromoter; catfish RM,(Ictalurus zebrafish punctatusracemase) electroporatedpromoter; with N1, 5’ end ds-shRNAconstructs targeting designed channel to disrupt catfish primordialnanos germgene; cell N2, migration 3’ end ds-shRNA and then either targeting untreated channel or treated catfish nanos gene;with cDNA, sodium full-length chloride expected cDNA to sequencerepress the of expression channel of catfish the transgene.nanos gene; Construct DND, components ds-shRNA were: targeting channelADSS, catfishzebrafishdead adenylosuccinate end gene. Treated Synthase embryos 2 (adss2) promoter; were repressed RM, zebrafish with 4racemase ppt sodium promoter; chloride N1, 5’ (NaCl). * Theend difference ds‐shRNA betweentargeting untreatedchannel catfish and nanos treated gene; spawning N2, 3’ end rates ds‐shRNA of fish targeting possessng channel that construct catfish was significantnanos gene; (p

ADSSDND 6 8 75 5 5 100 RMN1 7 9 78 8 8 100 RMN2 6 8 75 8 10 80 Mar. Drugs 2017, 15, 155 5 of 23

Table 1. Cont.

Untreated Treated Constructs Spawned Total % Spawned Total % RMcDNA 2 3 67 3 3 100 RMDND 2 5 40 2 3 67 Total ** 34 58 59 38 41 93

Fish from three, ADSSN1, ADSSN2 and RMDND, of the eight constructs in the untreated group had spawning rates lower than 50%, while females from five of the eight constructs in the treated group had 100% spawning rates. There were significant differences (p < 0.05) between treated and untreated groups, for ADSSN2 construct as well as in pooled groups of all constructs. When pooled by promoters, spawning rates in untreated groups and treated groups from the ADSS system had significant differences (p < 0.05) between transgenic treated and untreated groups. The ratio of treated spawning rate/untreated spawning rate was 1.94 (Table2).

Table 2. Spawning rates of P1 female channel catfish (Ictalurus punctatus) electroporated with constructs designed to disrupt primordial germ cell migration and then either untreated or treated with sodium chloride expected to repress the expression of the transgene. Construct components were: ADSS, zebrafish adenylosuccinate Synthase 2 (adss2) promoter; RM, zebrafish racemase promoter; N1, 5’ end ds-shRNA targeting channel catfish nanos gene; N2, 3’ end ds-shRNA targeting channel catfish nanos gene; cDNA, full-length cDNA sequence of channel catfish nanos gene; DND, ds-shRNA targeting channel catfish dead end gene. Asterisk (*) behind the construct name means the difference untreated and treated spawning rates of fish possessed that construct is significant (p < 0.05, Fisher’s Exact Test). Treated embryos were repressed with 4 ppt sodium chloride (NaCl). Ratio T/U: treated spawning rate/untreated spawning rate.

Untreated Treated Constructs Ratio of T/U Spawned Total % Spawned Total % ADSS * 17 33 52 17 17 100 1.94 Rm 17 25 68 21 24 88 1.29 N1 8 13 62 8 8 100 1.61 N2 * 8 16 50 12 14 86 1.72 cDNA 10 16 63 11 11 100 1.59 DND 8 13 62 7 8 82 1.32

When pooling by knockdown strategy, spawning rates for untreated and treated groups of N2 had the best repression/knockdown efﬁciency compared to other strategies, with the efﬁciency ratio of 1.72 (Table2).

2.3. Real-Time Quantitative PCR Results for the P1 and F1 Generation Transgenic Channel Catﬁsh

2.3.1. ADSS System

Nanos of ADSSN1 was not signiﬁcantly downregulated in P1 (Figure2), for both treated and untreated groups. Treated and untreated groups showed similar off-target patterns of decreasing expression of dead end, from 6.82X to 3.48X for salt treated and 4.48X to 1.62X for the untreated. Differences between the salt treated and untreated group were minimal and inconsistent. There were no data generated for ADSSN1 in the F1 generation as no females were available to spawn in the treated group and the embryos from the untreated group were not transgenic (Table2). Mar. Drugs 2017, 15, 155 6 of 23 Mar. Drugs 2017, 15, x 6 of 25

24 48 120 nanos vasa dnd nanos vasa dnd nanos vasa dnd 3.00 2.00 1.00

expression 0.00

‐1.00 AdssN1‐u gene ‐ ‐2.00 AdssN1‐t ‐3.00 ‐2.71 marker

‐4.00 ‐3.48 PGC ‐5.00 ‐4.48 ‐4.25 ‐6.00

Relative ‐7.00 ‐6.82 ‐8.00 Marker genes‐hours post fertilization

Figure 2. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end Figure(dnd) in 2. the Relative treated expression and untreated of primordial groups exposed germ cell to zebrafish (PGC) markeradenylosuccinate genes, nanos, synthase vasa, 2and(adss2 dead) gene end (dnd) in the treated and untreated groups exposed to zebrafish adenylosuccinate synthase 2 (adss2) gene (ADSS) promoter fused with 5’ end shRNAi targeting channel catfish nanos gene (N1) as P1 ADSSN1 (ADSS)transgenic promoter channel fused catfish with (Ictalurus 5’ end shRNAi punctatus targeting) embryos. channelu represents catfish nanos untreated gene (N1) group, as Pt 1represents ADSSN1 transgenictreated group. channel The treatedcatfish group(Ictalurus was punctatus incubated) embryos. in 4 ppt sodium u represents chloride. untreated The samples group, were t represents analyzed treatedat 24, 48 group. and 120 The h treated post fertilization group was (hpf),incubated respectively. in 4 ppt sodium Relative chloride.nanos, vasa, The andsamplesdead were end (dnd) analyzedgene atexpression 24, 48 and was 120 expressed h post fertilization as fold change (hpf), over respectively. non-transgenic Relative control nanos, samples vasa, and at thedead same end (dnd) time point.gene expressionAll the red was fold expressed values numbers as fold shown change were over significantly non‐transgenic different control compared samples to at non-transgenic the same time controlpoint. Allat thethe levelred fold of p values< 0.05 usingnumbers Pairwise shown Fixed were Reallocation significantly Randomization different compared Test (PFRR)to non‐ fromtransgenic REST control(Relative at Expressionthe level of Software p < 0.05 Tool). using Red Pairwise X on the Fixed SE barReallocation indicated theRandomization relative expression Test (PFRR) between from the RESTtreated (Relative and untreated Expression groups Software was significantly Tool). Red different X on the from SE eachbar indicated other using the PFRR. relative expression between the treated and untreated groups was significantly different from each other using PFRR.

ADSSN2 knocked down nanos in the P1 treated group from 1.98X to 2.91X at three time points ADSSN2 knocked down nanos in the P1 treated group from 1.98X to 2.91X at three time points (Figure3a). Untreated ADSSN2 had variable effects on nanos at different time points as there was (Figure 3a). Untreated ADSSN2 had variable effects on nanos at different time points as there was downregulation for nanos (2.32X) at 24 hpf, and upregulation for nanos (6.40X) at 120 hpf. Similar to downregulation for nanos (2.32X) at 24 hpf, and upregulation for nanos (6.40X) at 120 hpf. Similar to ADSSN1, a strong off-target effect was observed in the treated group for knockdown dead end. ADSSN1, a strong off‐target effect was observed in the treated group for knockdown dead end. F1 ADSSN2 treatments yielded results as hypothesized. Nanos was significantly downregulated F1 ADSSN2 treatments yielded results as hypothesized. Nanos was significantly downregulated for 16.75X at 48 hpf and 40.31X at 120 hpf (Figure3b), in untreated group. Most time points from the for 16.75X at 48 hpf and 40.31X at 120 hpf (Figure 3b), in untreated group. Most time points from the treated group had similar expression patterns as the non-transgenic control, except the off-target dead treated group had similar expression patterns as the non‐transgenic control, except the off‐target dead end at 48 hpf, which was significantly upregulated for 11.87X. end at 48 hpf, which was significantly upregulated for 11.87X. Both ADSScDNA treated and untreated P1 strongly upregulated the expression of nanos gene at three time points (Figure4a). In the treated group, upregulation of nanos varied from 167.78X to 747.39X at the three time points, and the peak value was at 48 hpf. In the untreated group, the upregulation of nanos was from 32.51X (24 hpf) to 296.27X (120 hpf). Dead end was down regulated in both treated and untreated groups from 2X to 10X at three time points. Vasa and dead end was downregulated 8.17X and 10.01X at 120 hpf in the treated group. For the F1 generation, ADSScDNA downregulated nanos at three time points in the untreated groups (Figure4b), especially at 48 hpf (9.53X). Fish with this construct were consistently downregulated for vasa. At 24 hpf, nanos, vasa and dead end were downregulated in a similar fashion in both treated and untreated groups. For P1 ADSSDND exposed channel catfish, dead end was downregulated from 2.98X to 5.04X, at three time points in treated groups, and at 24 hpf and 48 hpf in untreated group (Figure5a). At 120 hpf, dead end was significantly up regulated in the untreated group. Strong off-target effects

Mar. Drugs 2017, 15, 155 7 of 23 were observed: vasa expression was knocked down at 48 and 120 hpf in the untreated group, from Mar.3.02X Drugs to 7.46X. 2017, 15, x 7 of 25

24 48 120 nanos vasa dnd nanos vasa dnd nanos vasa dnd 10.00

8.00 6.40 6.00 expression

4.00 2.83 2.00 AdssN2‐u gene ‐ 0.00 AdssN2‐t ‐2.00 marker

‐4.00 PGC ‐3.91 ‐6.00 ‐5.86 ‐8.00 Relative ‐10.00 ‐8.42 Marker genes‐hours post fertilization

(a) 24 48 120 nanos vasa dnd nanos vasa dnd nanos vasa dnd 20.00 11.87 x 10.00 expression

0.00 AdssN2‐u gene

‐ ‐10.00 AdssN2‐t ‐20.00 x marker ‐16.75 ‐30.00 PGC

‐40.00 x Relative ‐50.00 ‐40.31 Marker genes‐hours post fertilization

(b)

Figure 3. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end Figure 3. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end (dnd) in the treated and untreated groups exposed to zebrafish adenylosuccinate synthase 2 (adss2) gene (dnd) in the treated and untreated groups exposed to zebrafish adenylosuccinate synthase 2 (adss2) gene (ADSS) promoter fused with 30′ end shRNAi targeting channel catfish nanos gene (N2) as P1 (a) and F1 (ADSS) promoter fused with 3 end shRNAi targeting channel catfish nanos gene (N2) as P1 (a) and F1 (b) ADSSN2 transgenic channel catfish catfish ( Ictalurus punctatus)) embryos. uu representsrepresents untreated untreated group, group, tt representsrepresents treatedtreated group. group. The The treated treated group group was was incubated incubated in 4in ppt 4 ppt sodium sodium chloride. chloride. The samplesThe samples were wereanalyzed analyzed at 24, at 48 24, and 48 120 and h post120 h fertilization post fertilization (hpf), respectively. (hpf), respectively. Relative Relativenanos, vasa nanos, and, vasadead, and end (deaddnd) endgene (dnd expression) gene expression was expressed was expressed as fold change as fold over change non-transgenic over non‐transgenic control samples control at samples the same at time the samepoint. time All the point. red foldAll the values red numbersfold values shown numbers were significantlyshown were different significantly compared different to non-transgenic compared to noncontrol‐transgenic at the level control of p

Both ADSScDNA treated and untreated P1 strongly upregulated the expression of nanos gene at three time points (Figure 4a). In the treated group, upregulation of nanos varied from 167.78X to 747.39X at the three time points, and the peak value was at 48 hpf. In the untreated group, the upregulation of nanos was from 32.51X (24 hpf) to 296.27X (120 hpf). Dead end was down regulated in

Mar. Drugs 2017, 15, x 8 of 25

both treated and untreated groups from 2X to 10X at three time points. Vasa and dead end was Mar. Drugs 2017, 15, 155 8 of 23 downregulated 8.17X and 10.01X at 120 hpf in the treated group.

800.00 747.39

700.00

600.00 expression 500.00 337.79 gene ‐ 400.00 x 296.27 300.00 marker

AdsscDNA‐u 200.00 167.78 106.07 PGC AdsscDNA‐t 100.00 32.51 0.00

Relative ‐8.17 ‐10.01 ‐100.00 nanos vasa dnd nanos vasa dnd nanos vasa dnd 24 48 120 Marker genes‐hours post fertilization

(a) 24 48 120 nanos vasa dnd nanos vasa dnd nanos vasa dnd 5.00

0.00

expression ‐5.00 ‐4.00 AdsscDNA‐u

gene ‐7.46

‐ ‐10.00 x AdsscDNA‐t ‐15.00 ‐9.53 marker

‐20.00 PGC

‐25.00

Relative x ‐30.00 ‐25.76 Marker genes‐hours post fertilization

(b)

Figure 4. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end Figure 4. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end (dnd) in the treated and untreated groups exposed to zebrafish adenylosuccinate synthase 2 (adss2) gene (dnd) in the treated and untreated groups exposed to zebrafish adenylosuccinate synthase 2 (adss2) gene (ADSS) promoter fused with full‐length cDNA of channel catfish nanos gene (cDNA) as P1 (a) and F1 (ADSS) promoter fused with full-length cDNA of channel catfish nanos gene (cDNA) as P (a) and F (b) ADSScDNA transgenic channel catfish (Ictalurus punctatus) embryos. u represents untreated1 group, 1 (b) ADSScDNA transgenic channel catfish (Ictalurus punctatus) embryos. u represents untreated group, t represents treated group. The treated group was incubated in 4 ppt sodium chloride. The samples t represents treated group. The treated group was incubated in 4 ppt sodium chloride. The samples were analyzed at 24, 48 and 120 h post fertilization (hpf), respectively. Relative nanos, vasa, and dead were analyzed at 24, 48 and 120 h post fertilization (hpf), respectively. Relative nanos, vasa, and dead end (dnd) gene expression was expressed as fold change over non‐transgenic control samples at the end (dnd) same timegene point. expression All the red was fold expressed values numbers as fold shown change were over significantly non-transgenic different control compared samples to the at the samenon time‐transgenic point. control All the at red the fold level values of p < numbers 0.05 using shown Pairwise were Fixed significantly Reallocation different Randomization compared Test to the non-transgenic(PFRR) from REST control (Relative at the levelExpression of p < 0.05Software using Tool). Pairwise Red FixedX on the Reallocation SE bar indicated Randomization the relative Test (PFRR)expression from between REST (Relative the treated Expression and untreated Software groups Tool). was Red significantly X on the SE different bar indicated from each the other relative expressionusing PFRR. between the treated and untreated groups was significantly different from each other using PFRR.

For the F1 generation, ADSScDNA downregulated nanos at three time points in the untreated groups (Figure 4b), especially at 48 hpf (9.53X). Fish with this construct were consistently The untreated F1 ADSSDND group was signiﬁcantly downregulated for dead end at 48 hpf (13.36X) and 120 hpf (18.63X) (Figure5b). Signiﬁcant off-target downregulation was observed for nanos and vasa in the untreated group at all three time points, especially at 24 hpf (8.61X for nanos and 9.53X for vasa), 48 hpf (9.82X for vasa), and 120 hpf (11.28X for nanos). Mar. Drugs 2017, 15, x 9 of 25

downregulated for vasa. At 24 hpf, nanos, vasa and dead end were downregulated in a similar fashion in both treated and untreated groups. For P1 ADSSDND exposed channel catfish, dead end was downregulated from 2.98X to 5.04X, at three time points in treated groups, and at 24 hpf and 48 hpf in untreated group (Figure 5a). At 120 hpf, dead end was significantly up regulated in the untreated group. Strong off‐target effects Mar. Drugswere 2017observed:, 15, 155 vasa expression was knocked down at 48 and 120 hpf in the untreated group, 9 of 23 from 3.02X to 7.46X.

24 48 120 nanos vasa dnd nanos vasa dnd nanos vasa dnd 4.00

2.00

0.00 expression

AdssDND‐u gene ‐2.00 ‐ AdssDND‐t ‐4.00 ‐2.98 ‐2.95 ‐3.02 marker ‐4.19 ‐4.37 ‐4.28 x

PGC ‐6.00 ‐5.04 ‐4.57

‐8.00 ‐7.46 Relative

‐10.00 Marker genes‐hours post fertilization

(a) 24 48 120 nanos vasa dnd nanos vasa dnd nanos vasa dnd 5.00

0.00 expression

‐5.00 AdssDND‐u gene ‐ AdssDND‐t

marker ‐10.00 x

‐8.61 x x

PGC ‐9.53

‐9.82 x ‐11.28 ‐15.00 x ‐13.36 Relative

‐20.00 x Marker genes‐hours post fertilization ‐18.63

(b)

Figure 5. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end Figure 5. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end (dnd) in the treated and untreated groups exposed to zebrafish adenylosuccinate synthase 2 (adss2) gene (dnd) in the treated and untreated groups exposed to zebrafish adenylosuccinate synthase 2 (adss2) gene (ADSS) promoter fused with shRNAi targeting channel catfish dead end gene (DND) as P1 (a) and F1 (ADSS)(b) promoterADSSDND fused transgenic with shRNAichannel catfish targeting (Ictalurus channel punctatus catfish) deadembryos. end gene u represents (DND) asthe P 1untreated(a) and F 1 (b) ADSSDNDgroup, t transgenic represents channeltreated group. catfish The (Ictalurus treated punctatusgroup was) embryos.incubated inu represents4 ppt sodium the chloride. untreated The group, t representssamples treatedwere analyzed group. at The 24, treated48 and 120 group h post was fertilization incubated (hpf), in 4 respectively. ppt sodium Relative chloride. nanos The, vasa samples, wereand analyzed dead end at (dnd 24,) gene 48 and expression 120 h post was fertilization expressed as fold (hpf), change respectively. over non‐ Relativetransgenicnanos control, vasa samples, and dead end (dndat the) gene same expressiontime point. All was the expressed red fold values as fold numbers change shown over were non-transgenic significantly different control samplescompared at the sameto time the non point.‐transgenic All the control red fold at valuesthe level numbers of p < 0.05 shown using Pairwise were significantly Fixed Reallocation different Randomization compared to the non-transgenic control at the level of p < 0.05 using Pairwise Fixed Reallocation Randomization Test (PFRR) from REST (Relative Expression Software Tool). Red X on the SE bar indicated the relative expression between the treated and untreated groups was significantly different from each other using PFRR.

2.3.2. RM System

Nanos was slightly downregulated in P1 embryos electroporated with RMN1 construct at 24 hpf and 48 hpf in both treated and untreated groups (Figure6a). At 120 hpf, nanos was upregulated in treated and untreated groups, 3.68X and 2.58X, respectively. The off-target dead end was downregulated at three time points in both groups, from 2.11X to 6.22X. Mar. Drugs 2017, 15, x 10 of 25

Test (PFRR) from REST (Relative Expression Software Tool). Red X on the SE bar indicated the relative expression between the treated and untreated groups was significantly different from each other using PFRR.

The untreated F1 ADSSDND group was significantly downregulated for dead end at 48 hpf (13.36X) and 120 hpf (18.63X) (Figure 5b). Significant off‐target downregulation was observed for nanos and vasa in the untreated group at all three time points, especially at 24 hpf (8.61X for nanos and 9.53X for vasa), 48 hpf (9.82X for vasa), and 120 hpf (11.28X for nanos).

2.3.2. RM System

Nanos was slightly downregulated in P1 embryos electroporated with RMN1 construct at 24 hpf and 48 hpf in both treated and untreated groups (Figure 6a). At 120 hpf, nanos was upregulated in Mar. Drugstreated2017, 15and, 155 untreated groups, 3.68X and 2.58X, respectively. The off‐target dead end was 10 of 23 downregulated at three time points in both groups, from 2.11X to 6.22X.

24 48 120 nanos vasa dnd nanos vasa dnd nanos vasa dnd 6.00 3.68 4.00

2.00 expression

RMN1‐u gene 0.00 ‐ RMN1‐t ‐2.00 marker

PGC ‐4.00

‐3.77 ‐4.25 ‐4.48 ‐6.00 ‐4.86

Relative ‐6.22 ‐8.00 Marker genes‐hours post fertilization

(a) 24 48 120 nanos vasa dnd nanos vasa dnd nanos vasa dnd 10.00 5.87 5.00

0.00 expression ‐5.00 ‐4.91 ‐5.11 RMN1‐u gene

‐ ‐10.00 ‐9.49 RMN1‐t ‐15.00 marker ‐16.30 ‐20.00 PGC

‐25.00

‐30.00 Relative x ‐35.00 ‐31.55 Marker genes‐hours post fertilization

(b)

Figure 6. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end (dnd) in the treated and untreated groups exposed to zebrafish racemase gene (RM) promoter fused 0 with 5 end shRNAi targeting channel catfish nanos gene (N1) as P1 (a) and F1 (b) RMN1 transgenic channel catfish (Ictalurus punctatus) embryos. u represents untreated group, t represents treated group. The treated group was incubated in 4 ppt sodium chloride. The samples were analyzed at 24, 48 and 120 h post fertilization (hpf), respectively. Relative nanos, vasa, and dead end (dnd) gene expression was expressed as fold change over non-transgenic control samples at the same time point. All the red fold values numbers shown were significantly different compared to the non-transgenic control at the level of p < 0.05 using Pairwise Fixed Reallocation Randomization Test (PFRR) from REST (Relative Expression Software Tool). Red X on the SE bar indicated the relative expression between the treated and untreated groups was significantly different from each other using PFRR.

In untreated F1 RMN1 channel catﬁsh embryos, nanos was downregulated from 3.54X to 31.55X at three points, while in the treated group, the expression of nanos was similar to the non-transgenic control, except at 120 hpf (Figure6b). In addition, off-target downregulation occurred for vasa at 24 hpf (16.30X) in the untreated group and 9.49X in the treated group. At 24 hpf and 48 hpf, untreated P1 embryos exposed to RMN2 were downregulated for nanos, 2.32X and 2.07X (Figure7a). The salt treatment repressed downregulation of nanos at 24 and 48 hpf, but not at 120 hpf (2.10X). Off-target downregulation, 1.85X to 5.39X, was observed for dead end in both groups at three time points. The differences in expression between salt treated and untreated embryos Mar. Drugs 2017, 15, x 11 of 25

Figure 6. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end (dnd) in the treated and untreated groups exposed to zebrafish racemase gene (RM) promoter fused with 5′ end shRNAi targeting channel catfish nanos gene (N1) as P1 (a) and F1 (b) RMN1 transgenic channel catfish (Ictalurus punctatus) embryos. u represents untreated group, t represents treated group. The treated group was incubated in 4 ppt sodium chloride. The samples were analyzed at 24, 48 and 120 h post fertilization (hpf), respectively. Relative nanos, vasa, and dead end (dnd) gene expression was expressed as fold change over non‐transgenic control samples at the same time point. All the red fold values numbers shown were significantly different compared to the non‐transgenic control at the level of p < 0.05 using Pairwise Fixed Reallocation Randomization Test (PFRR) from REST (Relative Expression Software Tool). Red X on the SE bar indicated the relative expression between the treated and untreated groups was significantly different from each other using PFRR.

In untreated F1 RMN1 channel catfish embryos, nanos was downregulated from 3.54X to 31.55X at three points, while in the treated group, the expression of nanos was similar to the non‐transgenic control, except at 120 hpf (Figure 6b). In addition, off‐target downregulation occurred for vasa at 24 hpf (16.30X) in the untreated group and 9.49X in the treated group. At 24 hpf and 48 hpf, untreated P1 embryos exposed to RMN2 were downregulated for nanos, Mar. Drugs 2017 15 2.32X and, , 2.07X 155 (Figure 7a). The salt treatment repressed downregulation of nanos at 24 and 48 hpf, 11 of 23 but not at 120 hpf (2.10X). Off‐target downregulation, 1.85X to 5.39X, was observed for dead end in both groups at three time points. The differences in expression between salt treated and untreated at the sameembryos time at the point same were time minimal.point were minimal. At 48 hpf, At 48vasa hpf,was vasa also was also signiﬁcantly significantly downregulated, downregulated, 3.19X, in the untreated3.19X, in the group. untreated group.

24 48 120 nanos vasa dnd nanos vasa dnd nanos vasa dnd 3.00 2.00 1.00 expression 0.00 RMN2‐u gene

‐ ‐1.00 RMN2‐t ‐2.00 marker ‐2.07 ‐2.10 ‐3.00 ‐2.32

PGC ‐2.85 ‐3.19 ‐4.00 ‐3.91 ‐4.31 ‐5.00 Relative ‐6.00 ‐5.39 Marker genes‐hours post fertilization Mar. Drugs 2017, 15, x 12 of 25 (a) 10.00

0.00

‐10.00 ‐4.90 ‐4.86 expression X x ‐10.36 x ‐6.98 ‐6.92

gene ‐20.00 ‐ ‐12.40

‐30.00

marker RMN2‐u

RMN2‐t PGC ‐40.00

‐50.00

Relative x ‐60.00 ‐55.58 nanos vasa dnd nanos vasa dnd nanos vasa dnd 24 48 120 Marker genes‐hours post fertilization

(b)

Figure 7. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end Figure 7. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end (dnd) in the treated and untreated groups exposed to zebrafish racemase gene (RM) promoter fused (dnd) in the treated and untreated groups exposed to zebrafish racemase gene (RM) promoter fused with 3′ end shRNAi targeting channel catfish nanos gene (N2) as P1 (a) and F1 (b) RMN2 transgenic 0 with 3 channelend shRNAi catfish ( targetingIctalurus punctatus channel) embryos. catfish unanos representsgene untreated (N2) as group, P1 (a) t andrepresents F1 (b treated) RMN2 group. transgenic channelThe catfish treated (Ictalurus group was punctatus incubated) embryos. in 4 ppt sodiumu represents chloride. untreatedThe samples group, were analyzedt represents at 24, treated48 and group. The treated120 h grouppost fertilization was incubated (hpf), respectively. in 4 ppt sodium Relative nanos chloride., vasa, Theand dead samples end (dnd were) gene analyzed expression at was 24, 48 and 120 h postexpressed fertilization as fold change (hpf), over respectively. non‐transgenic Relative control nanossamples, vasa at the, andsamedead time endpoint.(dnd All )the gene red fold expression values numbers shown were significantly different compared to the non‐transgenic control at the was expressed as fold change over non-transgenic control samples at the same time point. All the red level of p < 0.05 using Pairwise Fixed Reallocation Randomization Test (PFRR) from REST (Relative fold valuesExpression numbers Software shown Tool). were Red significantly X on the SE bar different indicated compared the relative toexpression the non-transgenic between the treated control at the level ofandp < untreated 0.05 using groups Pairwise was significantly Fixed Reallocation different from Randomization each other using Test PFRR. (PFRR) from REST (Relative Expression Software Tool). Red X on the SE bar indicated the relative expression between the treated and untreatedIn the case groups of F1 wasRMN2 significantly channel catfish different embryos, from nanos each was other significantly using PFRR. downregulated at three time points in the untreated group (4.90X to 55.58X) (Figure 7b). For salt treated groups, downregulation of nanos was successfully repressed. Off‐target effects occurred as vasa and dead end Inalso the had case decreased of F1 RMN2 expression, channel at 24 hpf catfish and 48 embryos, hpf for vasananos, 48 hpfwas and significantly 120 hpf for dead downregulated end in the at three timeuntreated points group. in the untreated group (4.90X to 55.58X) (Figure7b). For salt treated groups, downregulationUpregulation of nanos of wasnanos successfully in P1 RMcDNA repressed. electroporated Off-target embryos effects was significant, occurred 29.44X as vasa (120and hpf)dead end also hadto decreased510.83X (48 hpf) expression, in both groups at 24 (Figure hpf and 8a). 48 In the hpf untreated for vasa group,, 48 hpf the and upregulation 120 hpf of for nanosdead was end in the continuously increasing from 32.51X (24 hpf) to 384.38X (120 hpf), and in the treated group, untreated group. the upregulation started at 275.90X at 24 hpf, reached its peak of 510.83X at 48 hpf, and then decreased to 29.44X at 120 hpf.

Mar. Drugs 2017, 15, 155 12 of 23

Upregulation of nanos in P1 RMcDNA electroporated embryos was signiﬁcant, 29.44X (120 hpf) to 510.83X (48 hpf) in both groups (Figure8a). In the untreated group, the upregulation of nanos was continuously increasing from 32.51X (24 hpf) to 384.38X (120 hpf), and in the treated group, the upregulation started at 275.90X at 24 hpf, reached its peak of 510.83X at 48 hpf, and then decreased to 29.44XMar. Drugs at 120 2017 hpf., 15, x 13 of 25

600.00 510.83 500.00 384.38 365.82 x

expression 400.00 275.90 x gene

‐ 300.00

200.00 RMcDNA‐u marker

100.00 RMcDNA‐t PGC

32.51 29.44

0.00

Relative ‐8.76 ‐100.00 nanos vasa dnd nanos vasa dnd nanos vasa dnd 24 48 120 Marker genes‐hours post fertilization

(a) 16.00 13.47 14.00 11.44 11.72 X 12.00 9.11

expression X 9.23

10.00 8.00 gene ‐ 6.00 3.64 3.75 x 3.56 4.00 marker RMcDNA‐u

2.00 RMcDNA‐t PGC

0.00 ‐2.00 Relative ‐4.00 nanos vasa dnd nanos vasa dnd nanos vasa dnd 24 48 120 Marker genes‐hours post fertilization

(b)

Figure 8. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end Figure 8. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end (dnd) in the treated and untreated groups exposed to zebrafish racemase gene (ADSS) promoter fused (dnd) in the treated and untreated groups exposed to zebrafish racemase gene (ADSS) promoter fused with full‐length cDNA of channel catfish nanos gene (cDNA) as P1 (a) and F1 (b) RMcDNA transgenic with full-length cDNA of channel catfish nanos gene (cDNA) as P (a) and F (b) RMcDNA transgenic channel catfish (Ictalurus punctatus) embryos. u represents untreated1 group, t represents1 treated group. channel catfish (Ictalurus punctatus) embryos. u represents untreated group, t represents treated group. The treated group was incubated in 4 ppt sodium chloride. The samples were analyzed at 24, 48 and The treated group was incubated in 4 ppt sodium chloride. The samples were analyzed at 24, 48 and 120 h post fertilization (hpf), respectively. Relative nanos, vasa, and dead end (dnd) gene expression was 120 h post fertilization (hpf), respectively. Relative nanos, vasa, and dead end (dnd) gene expression expressed as fold change over non‐transgenic control samples at the same time point. All the red fold wasvalues expressed numbers as fold shown change were over significantly non-transgenic different control compared samples to the at thenon same‐transgenic time point. control All at the the red foldlevel values of p numbers< 0.05 using shown Pairwise were Fixed significantly Reallocation different Randomization compared toTest the (PFRR) non-transgenic from REST control (Relative at the levelExpression of p < 0.05 Software using Tool). Pairwise Red Fixed X on the Reallocation SE bar indicated Randomization the relative Test expression (PFRR) between from REST the treated (Relative Expressionand untreated Software groups Tool). was Redsignificantly X on the different SE bar indicated from each the other relative using expression PFRR. between the treated and untreated groups was significantly different from each other using PFRR.

For F1 embryos in the untreated RMcDNA group, nanos was significantly upregulated at 48 hpf and 120 hpf, 13.47X and 11.72X, respectively (Figure 8b). In the salt treated group, the upregulation of nanos was largely repressed compared to untreated at 48 hpf (downregulation of 3.56X), but not at 120 hpf (9.23X). Strong off‐target effects for vasa were apparent at 24 hpf (upregulation of 9.11X) and 120 hpf (upregulation of 11.44X) for the untreated group.

Mar. Drugs 2017, 15, 155 13 of 23

For F1 embryos in the untreated RMcDNA group, nanos was signiﬁcantly upregulated at 48 hpf and 120 hpf, 13.47X and 11.72X, respectively (Figure8b). In the salt treated group, the upregulation of nanos was largely repressed compared to untreated at 48 hpf (downregulation of 3.56X), but not at 120 hpf (9.23X). Strong off-target effects for vasa were apparent at 24 hpf (upregulation of 9.11X) and 120 hpfMar. (upregulation Drugs 2017, 15, x of 11.44X) for the untreated group. 14 of 25

Dead endDeadwas end downregulated,was downregulated, 1.71X 1.71X to to 3.42X, 3.42X, for for RMDND RMDND in both in both P1 group P1 groupss at three at time three points time points except inexcept treated in treated group group at 24 at hpf24 hpf (upregulated (upregulated 164.19X) (Figure (Figure 9a).9 a).Additionally, Additionally, vasa andvasa nanosand werenanos were signiﬁcantlysignificantly knocked knocked down down at 120at 120 hpf hpf in in the the untreateduntreated group group (4.98X (4.98X and and 4.37X). 4.37X).

180.00 164.19 160.00 140.00 120.00 100.00 expression

80.00 gene ‐ 60.00 RMDND‐u 40.00 RMDND‐t marker 20.00 PGC 0.00 ‐20.00 x ‐3.29 ‐4.37 ‐4.98 ‐3.42 ‐3.34 nanos vasa dnd nanos vasa dnd nanos vasa dnd Relative 24 48 120 Marker genes‐hours post fertilization

(a) 24 48 120 nanos vasa dnd nanos vasa dnd nanos vasa dnd 4.00

2.00

0.00 expression

‐2.00 gene

‐ RMDND‐u ‐4.00 ‐4.21 ‐4.15 RMDND‐t ‐6.00 marker ‐5.35 ‐5.27

PGC ‐8.00 ‐7.06

‐10.00 ‐8.67

Relative ‐12.00 ‐11.75 ‐14.00 Marker genes‐hours post fertilization (b)

Figure 9. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end Figure 9. Relative expression of primordial germ cell (PGC) marker genes, nanos, vasa, and dead end (dnd) in the treated and untreated groups exposed to zebrafish racemase gene (RM) promoter fused dnd racemase ( ) inwith the shRNAi treated targeting and untreated channel catfish groups dead exposed end gene to (DND) zebrafish as P1 (a) and F1 (bgene) RMDND (RM) transgenic promoter fused with shRNAichannel targetingcatfish (Ictalurus channel punctatus catfish) embryos.dead end u representsgene (DND) untreated as P group,1 (a) and t represents F1 (b) RMDND treated group. transgenic channelThe catfish treated (Ictalurus group was punctatus incubated) embryos.in 4 ppt sodiumu represents chloride. untreatedThe samples group, were analyzedt represents at 24, 48 treated and group. The treated120 h post group fertilization was incubated (hpf), respectively. in 4 ppt sodium Relative nanos chloride., vasa, and The dead samples end (dnd were) gene analyzed expression at was 24, 48 and 120 h postexpressed fertilization as fold change (hpf), over respectively. non‐transgenic Relative control samplesnanos, vasaat the, andsame deadtime point. end ( dndAll the) gene red fold expression values numbers shown were significantly different compared to the non‐transgenic control at the was expressed as fold change over non-transgenic control samples at the same time point. All the red level of p < 0.05 using Pairwise Fixed Reallocation Randomization Test (PFRR) from REST (Relative fold valuesExpression numbers Software shown Tool). were Red significantly X on the SE bar different indicated compared the relative toexpression the non-transgenic between the treated control at the level ofandp < untreated 0.05 using groups Pairwise was significantly Fixed Reallocation different from Randomization each other using TestPFRR. (PFRR) from REST (Relative Expression Software Tool). Red X on the SE bar indicated the relative expression between the treated and untreatedThe target groups gene wasdead significantlyend was significantly different downregulated, from each other 5.27X using to PFRR.11.75X, in the F1 untreated RMDND group at three time points (Figure 9b). In the treated group, downregulation of dead end was

Mar. Drugs 2017, 15, 155 14 of 23

The target gene dead end was signiﬁcantly downregulated, 5.27X to 11.75X, in the F1 untreated RMDND group at three time points (Figure9b). In the treated group, downregulation of dead end was rapidly reduced at each time point compared to the untreated group to a level of −1.60X to −2.74X. Nanos was signiﬁcantly downregulated in both groups at 48 hpf and 120 hpf, indicative of a strong off-target effect or gene interaction.

3. Discussion Knockdown constructs using shRNAi and overexpression of cDNA that targeted the PGC genes, nanos and dead end, which play important roles in PGC migration, were electroporated into channel catfish to achieve transgenic sterilization. Subsequently, F1 individuals were produced containing the knockdown constructs. These constructs were designed to be repressed by application of salt to accomplish repressible transgenic sterilization. Gene expression of PGC marker genes during the early development of P1 electroporated individuals, thus exposed to the constructs during putative PGC migration, that were either treated with 4 ppt sodium chloride or untreated indicated that, in some cases, the constructs were able to downregulate the targets or overexpress the target genes, but the repression was not very effective or in some cases, actually enhanced down or upregulation. The results were different for F1 transgenic individuals, as sterilization constructs were downregulating the key PGC genes, and that the repression was generally successful. However, the cDNA overexpression constructs did not work as expected with the ADSS driven construct reducing mRNA levels, but the RM driven construct dramatically increasing the mRNA levels. Many complementary off-target effects on the related PGC genes were observed. Spawning data on P1 treated and untreated brood stock exposed to the transgenes as embryos supported the hypothesis that the repressible transgenic system based on sodium chloride sensitive promoters is workable as spawning rates of the sodium chloride treated fish was higher than the untreated. In the case of control channel catfish embryos that had not been exposed to the transgenes, levels of nanos mRNA dropped naturally from 24 to 120 hpf, dead end dropped slightly faster than nanos, and vasa was dramatically lower, at 120 hpf compared to 24 hpf. Either the expression of mRNA of those three PGC migration marker genes was downregulated naturally or there was maternally derived mRNA of those marker genes in the early phase of embryonic development, which was being degraded later in embryogenesis. However, maternally derived nanos mRNA in zebrafish degrades rapidly before gastrulation [6], and Xu et al. [31] have shown that during embryogenesis in Asian seabass (Lates calcarifer), the vasa transcript is abundant in early stages, and persists at a reduced and detectable level in late stages. This supports the premise that the decreases in the nanos, vasa, and dead end mRNA levels in channel catfish were due to natural and transgenic downregulation, and not just degradation. These observations on embryonic expression levels of nanos, vasa, and dead end mRNA might have implications for different knockdown strategies to prevent migration of PGCs and their potential success in channel catfish and other fish species [32]. Understanding the mRNA regulation pattern of the target genes would help us better explain the knockdown results at different time points. A more detailed study of the expression level of these three mRNAs in embryonic development of channel catfish might be beneficial for identifying the transgenic sterilization constructs with the best probability of success and further refinement of the repressible transgenic sterilization systems. Previous data using copper sensitive promoters in zebrafish and channel catfish for repressible sterilization on P1 embryos have shown that they were able to drive transcription in channel catfish and common carp embryos that had been electroporated with the knockdown constructs and likely had large copy numbers in their cytoplasm [32,33]. The expression data of copper systems on the PGC targets of the F1 transgenic embryos that had integrated copies of the repressible transgenic sterilization constructs was in general agreement with the results from the P1 (unpublished data). The salt and copper systems both show good potential for repressible transgenic sterilization, and both present viable options for further development of this concept. Mar. Drugs 2017, 15, 155 15 of 23

The knockdown/overexpression and repression efficiency were quite different between P1 and F1 samples. Based on previous unpublished research from our laboratory, we predicted that the embryonic exposure to the constructs in the P1 generation would be indicative of results in the F1, but they did not correlate well. The P1 expression data did show that the constructs were capable of downregulation, but the repression appeared problematic. The downregulation in the F1 transgenic embryos was often greater than their parental embryos that were exposed to the foreign constructs via electroporation, and the repression often more efficient in F1 embryos. Another difference between P1 and F1 generations was that in ADSScDNA and RMcDNA, nanos was dramatically upregulated in the parental generation, but the upregulation was repressed or reversed in F1. There are several potential explanations for the differing results observed between the P1 and F1.P1 were likely a mixture of both transgenic and non-transgenic embryos as integration rates would likely between 30% and 70% [28]. We expect that exposure to the constructs should cause sterilization if they are expressing, however, the percentage of embryos for which the electroporation was successful and how many molecules penetrated the developing embryos is unknown. Expression of positive or exposed embryos could be partially masked by those that are negative or for those that the electroporation was not successful. These results also show that although data on P1 for embryonic traits generated from exposure to transgenes has some value and relationship to what might be obtained in integrated F1 individuals, the relationship may not be perfectly predictive. Results for P1 and F1 common carp exposed or containing a shRNAi targeting aromatase were contradictory [34]. However, in this case, usefulness of the P1 data may have been more compromised, as the trait in question, gender, is determined several days after hatching. Strong off-target effects for related PGC genes were common in both P1 and F1 generations. The off-target effects might be caused by mismatch of siRNA [35,36]. In addition, there might be strong genetic interaction effects among nanos, vasa and dead end (Table3). Dead end protein acting to counteract the inhibitory function of several miRNAs, thereby allowing the expression of PGC specific proteins such as nanos and tdrd7 [18,37]. In Drosophila, the vasa gene encodes an ATP-dependent RNA helicase of the DEAD-box family and was required for promoting translation of at least two known mRNAs, nanos and gurken [38]. Dazl stimulates vasa translation by binding to the 30 UTR of vasa mRNA in vivo, Dazl knockout mice have reduced levels of vasa protein [39]. In mouse, nanos2 post-transcriptionally represses Dazl mRNA expression in germ cells [40], therefore indirectly regulate the expression of vasa. In human, dnd1 have strong interaction score with dazl by prediction [41]. Consistent with this observation, ADSSDND targeted dead end, but also downregulated nanos and vasa, at all three time points; knockdown of nanos in RMN2 also downregulate vasa and dead end at different time points. Although there was no evidence of nanos affecting the expression of dead end, it is possible there are undiscovered direct or indirect interactions between them. If the off-target effects were restricted to PGC migration loci, they may contribute to sterility. In F1 generation, knockdown and repression in most constructs were much closer to what was expected compared to the results from their parents. In untreated groups, the knockdown constructs were able to significantly downregulate the mRNA expression of the corresponding target gene. In treated groups, sodium chloride treatment repressed expression of knockdown constructs and the mRNA levels of the target genes were near normal. The following constructs, ADSSN2, ADSSDND, RMN2, RMDND, had the best patterns of downregulation with effective repression, followed by RMN1 (not at all timepoints) and ADSScDNA (strong off-target effects with repression). These results are even more encouraging, considering we usually obtain an inheritance rate of 20–50% for the F1 generation. Thus, one-half or less of the embryos in the pooled positive samples are likely transgenic. Significantly different expression levels, downregulation, were obtained even though there were non-transgenic individuals mixed in with the transgenic embryos. In some cases, knockdown for the target gene was inconsistent at different embryonic stages evaluated. It is not known exactly when the critical period of PGC marker gene knockdown is that ensures the PGC does not reach the genital ridge or dies. Based on our results, it appears it is not Mar. Drugs 2017, 15, 155 16 of 23 necessary to knockdown the PGC gene expression during all of embryogenesis to achieve sterilization since in some cases knockdown was only achieved during one time period, yet P1 still had a reduced rate of reproduction. However, further research is necessary to confirm these apparent results, and to further refine the repressible sterilization system.

Table 3. Interaction between nanos, vasa and dead end. From the current experiment, assuming interaction rather than off-target knockdown, knockdown of dead end resulted in downregulation of nanos and vasa, knockdown of nanos resulted in downregulation of vasa and dead end. Dead end can regulate nanos by inhibiting the activity of miR-430 and protect nanos from degradation [18,37]. Dead end might be involved in regulation of dazl, thus indirectly regulates the expression of vasa [39,41]. Vasa is required for promoting the translation of nanos [38]. Nanos regulates vasa by suppressing the expression of dazl [39,40]. The potential mechanism of nanos regulating the expression of dead end is still unknown.

Regulator Gene Responder Gene Nanos Vasa Dead End Vasa is required for Dead end protects nanos Nanos promoting from being degraded by translation of nanos miR-430 in germ cells Dead end potentially Nanos suppresses dazl, dazl Vasa regulates dazl, dazl regulates vasa expression regulates vasa expression Nanos potentially regulates No known Dead end dead end, mechanism unknown regulation found

With respect to spawning, ADSSN2 was the best system (treated P1 with higher spawning rate than untreated). In general, the repression was quite effective, as the treated brood stock had 93% spawning rate for all treatments pooled compared to 59% spawning rate for the untreated brood stock. To meet the goal of eliminating environmental risk and preventing introgression of the transgene into wild populations, the sterilization likely need to be 100% effective, and would likely be a requirement of regulatory agencies. Biologically, an effectiveness of less than 100% would likely still achieve the goal of the transgenes not establishing unless they had a selective advantage. This rate of sterilization in the P1 was not achieved, which was not surprising because of the expected mosaicism in P1 transgenic fish and the fact that electroporation may not have been successful for each embryo. The observed spawning differences between the treated and untreated P1 brood stock indicate that this salt-based sterilization and repression system has good potential. F1 spawning data will be needed to determine if this system can be 100% effective. Based upon preliminary, unpublished results with related constructs, we predict 100% sterility of the F1 with the salt-sensitive constructs. Four ppt sodium chloride was used as the repressor. Future research is needed regarding the minimum doses needed to repress the knockdown constructs. This is especially important as if the promoters are too sensitive and can be repressed at low, but naturally found levels of sodium chloride, escaped transgenic channel catfish could establish in the wild, negating the usefulness of the sterilization constructs evaluated in the current study. Channel catfish and blue catfish, I. furcatus, have spawned on rare occasion at salinity levels as high as 2 ppt [42]. For the vast majority of the geographic range of channel catfish, the salinity is less than 100 ppm. Likely, ADSS and RM promoters need to not be sensitive to 2.5 ppt sodium chloride and lower for this transgenic sterilization system to be commercially applicable. Otherwise, escaped catfish transgenic for performance traits stacked with the salt sterilization constructs could spawn if they encountered partially saline water in estuarine environments that repressed the construct. If these constructs were selectively advantageous, the transgenes could become established in the natural environment with possible positive, negative or neutral impacts on the ecosystem. If these promoters are repressed at ≤2.5 ppt, it might be possible to alter them to be less sensitive to salt to make them usable in the entire geographic range of channel catfish to prevent establishment of transgenic channel catfish. Mar. Drugs 2017, 15, 155 17 of 23

4. Materials and Methods All investigations and experimental studies on animals were conducted according to Institutional Animal Care and Use Committees (IACUC) and Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) protocols and guidelines.

4.1. Construction of Plasmids FRMwg plasmid [30] was used as the vector for all the transgene constructs in this experiment, and it had three functional components: insulator, ocean pout (Zoarces americanus) terminator and boundary element. Between these components, there were two salt sensitized promoters that activate the expression of knockdown constructs: a 577 bp zebraﬁsh adss2 (ADSS) promoter and a 565 bp zebraﬁshMar.racemase Drugs 2017, (RM)15, x promoter. 19 of 25

Channel catfish, Ictalurus punctatus: deadend ds-shRNA design Oligo1. 5’–GTTCCAGAGCGTAGCGCCGCTCTACGAGTTCCGAGCGGCGCTACGCTCTGGAAC–3’ Oligo2. 3’–CAAGGTCTCGCATCGCGGCGAGATGCTCAAGGCTCGCCGCGATGCGAGACCTTG–5’ Channel catfish, Ictalurus punctatus: nanos ds-shRNA design N1 targeting 5’ end Oligo1. 5’-CTCTCTATCCCGATGACTCGCGATGGTGTTTTCGCGAGTCATCGGGATAGAGAG-3’ Oligo2. 3’-GAGAGATAGGGCTACTGAGCGCTACCACAAAAGCGCTCAGTAGCCCTATCTCTC-5’ N2 targeting 3’ end Oligo3. 5’-CCGAAAATCTGAACCCCACTCTCACACTCGCTAGAGTGGGGTTCAGATTTTCGG-3’ Oligo4. 3’-GGCTTTTAGACTTGGGGTGAGAGTGTGAGCGATCTCACCCCAAGTCTAAAAGCC-5’ Channel catfish, Ictalurus punctatus: nanos full length cDNA GTGTTTACAGTAGCCTAGCTACGACTCTTTTTTAATAATTATTGTTTTAGTAAACAAACAAACAAAAT TAATGGCATGTGCGCGCTGATTGGTTCCCGCAGCAGCACCCGCCTTTCACGCGCCACACGCGTGTGC GCTCCAGACTCGCCTACTTAAGGGGGGTTTCTTGGTTGTTTTAAGTTTTCACCCCCCTCTCTATCCCGA TGACTCGCGATGGTGTTTTCTCTGCTGCGCTACATCCTGGCGGCTCACGGCTCCATGGAGCCCGAGGA CCAGCGTCACTTCCAGCCGTGGCGGGATTACTTGGGTCTCGCGGACACGGTGCGTGCGATGCGGGAA ACCTCTGGTCCCGAACCTCACCAGGTCCAGGAGTTCGGGTTTACGCCCGAGATTTCTCATCACCGGG GCCGTAATGCTGCAATGCTCCAGACTTTCCCCGAGCCTGAGATGTCCGTGTTCAGAGGAAGACCGAA AACGCAGATGGGGCGACTGAAGTGCGGTCGTACCGGGACCACAGGGCCTCCACCTCCATCACCTCA AGGTTCTCGCTCTGGGGAAAAGTTCTGCAGTTTCTGCAAACATAACGGCGAGTCTGAACAAGTGTTC ACGTCGCACTGCCTGAGGAGCCGCGGCGGAGCAGTGGCGTGTCCTTACCTGCGCCGCTACGTGTGTC CGCAGTGCGGGGCGACGGGCGACCGCGCGCACACCAAACGCTTCTGCCCGCTCGTGGACAGCACGT ACGCCTCCGTGTACACCCGGCCTGCCCGTAAATGACCGAAAATCTGAACCCCACTCTCACACTCGCT ACCAAACTGTAGGTTATTCTTTTTTTTTTTTTTTTTACTTGGGAAGGTGAACAAGAAGCTTTAGACAA AAGCTGCACAGGTACGTCAGCGGTCCTTAAAGTCCGGTACTGTACCTGGAATGCTTTTATATGTAGG CTTCAACTATATTTTCAAAAGGTACTAAAGATGTACCAGTTATGATTCAATTTCAGAGAAAAGCTCA AGTACAGTTGGTGCTTTTTATCTGAGAGTGGCTGTAGAAAGTTGTAAGTCCTTTTTAAAAAAAAAAA AAAAAAAATCAGCATTATATTTTTAATGTCTGCATTACTGTGCTTATTATTATGGCTTAGAGCTGTCG GGTTTAGTTGTTTGAAACTCCGGAATGACCTGCCCTGGGTTTACAGCTGTAACACCTGGAACGCTGTG GGTGTCAAGAGTTTTGCTTTACTAACTTTGTGTGCACTTTGTGTATGCACTTGTGTTGTGTGTTTATTT TGATTGGTGTGTTTTGTTTTGAAGCTGATTTCTCTAACGAGCTTGTGCTCAGGCCTCTCTGGCTCATCA CAGGTGCAGCATGTTACAGGTGCGGGTCTATGCAGGGCTTCATGATGGGACCGTGGATCTCCGACCT GCTATTTTTCTGCTCCATTTTATTGTCCATTCGAAGAACTTCTGACGTGTTGTGACTTTTTAAAGTGTT TTAGACCATTTGGGATTTGAGTTAATATACTTTATATGCATGTAACAAGCCTCAGTGCTGCATTTGTT TTTATATATTATATAAGACGTGTTGGACTGTTTTGGTACGAATGACCTCGTCGATGCCTCTGAATCTT CTGCAATTCTGTAAGTTTCAATTTCTAATATATTTAAAGTGTGAAAAAAAAAAAAAAA

FigureFigure 10. Primordial 10. Primordial germ germ cell cell knockdown knockdown sequences for for channel channel catfish, catﬁsh, IctalurusIctalurus punctatus punctatus. Loops . Loops were inwere grey in shade,grey shade, stem stem sequences sequences are are underlined. underlined. Channel Channel catfish catﬁsh nanosnanos full‐lengthfull-length cDNA was cDNA was sequencedsequenced by Su by et Su al. et [ 32al. ].[32].

These constructs were designed and built as Tet‐off‐like systems (Figure 11). The constructs were synthesized at Genscript Biotech Corporation (Piscataway, NJ, USA).

Mar. Drugs 2017, 15, 155 18 of 23

Channel catfish full-length nanos cDNA and dead end sequences were known (GenBank: KM874264.1 and KM874265.1). Based on these sequences, three RNAi and one cDNA sequence were ligated into the knockdown constructs targeting the marker genes (Figure 10). The knockdown strategies included: (1) a ds-sh RNA targeting the 50 end of channel catfish nanos gene (N1); (2) a ds-sh RNA targeting the 30 end of channel catfish nanos gene (N2); (3) a full length cDNA sequence of channel catfish nanos gene to overexpress nanos (cDNA); and (4) a ds-sh RNA targeting channel catfish dead end gene (DND). Except for the nanos cDNA sequence, all other constructs have a short hairpin structures and double Mar.stranded Drugs 2017 RNAs., 15, x 20 of 25 These constructs were designed and built as Tet-off-like systems (Figure 11). The constructs were synthesized at Genscript Biotech Corporation (Piscataway, NJ, USA).

Figure 11. Primordial germ cell knockdown constructs for the sterilization of channel catfish, FigureIctalurus 11. Primordial punctatus .(germa) ADSS:cell knockdown The zebrafish constructsadenylosuccinate for the sterilization synthase of 2 channelpromoter catfish, is salt Ictalurus sensitive punctatusand was. ( represseda) ADSS: byThe sodium zebrafish chloride adenylosuccinate higher at 4 ppt.synthase (b) RM: 2 promoter The zebrafish is saltracemase sensitivepromoter and was is salt repressedsensitive by and sodium was repressedchloride higher by sodium at 4 ppt. chloride (b) RM: higher The zebrafish at 4 ppt. Insulator, racemase promoter ocean pout is salt terminator sensitive and andboundary was repressed element by weresodium from chloride FRMwg higher plasmid at 4 ppt. (GenBank: Insulator, AF170915.1). ocean pout terminator and boundary element were from FRMwg plasmid (GenBank: AF170915.1). 4.2. P1 Fish 4.2. P1 FishPlasmids with the transgene of interest were cloned into Invitrogen Top 10 E. coli cells, according toPlasmids the recommended with the transgene protocols of by interest Invitrogen were cloned (user guide into Invitrogen 280126, Invitrogen, Top 10 E. coli Carlsbad, cells, according CA, USA). to Afterthe recommended the appropriate protocols period by of cloningInvitrogen cell (user culture, guide plasmid 280126, DNA Invitrogen, was extracted Carlsbad, using CA, the USA). Qiagen Aftermaxi-prep the appropriate kit (Cat No./ID: period of 12163) cloning and cell linearized culture, with plasmid SfiI (20,000DNA was units/mL, extracted BioLabs, using the Ipswich, Qiagen MA, maxiUSA).‐prep Plasmids kit (Cat were No./ID: then 12163) purified and using linearized the phenol-chloroform-ethanol with SfiI (20,000 units/mL, method BioLabs, and quantifiedIpswich, MA, using USA).Thermo Plasmids scientific were Nanodrop2000 then purified® (Thermousing the Fisher phenol Scientific,‐chloroform Waltham,‐ethanol MA, method USA). and Transgenes quantified were usingtransferred Thermo intoscientific one-cell, Nanodrop2000 fertilized eggs® (Thermo using Fisher electroporation Scientific, withWaltham, a Baekon MA, 2000USA) macromolecule. Transgenes weretransfer transferred system (Baekon,into one Inc.,‐cell, Saratoga,fertilized CA, eggs USA) using to generateelectroporation the P1 brood with stocka Baekon in the current2000 macromoleculeexperiment. Electroporation transfer system parameters (Baekon, Inc., were Saratoga, set at 6 kV, CA, 27 USA) pulses, to 0.8generate s burst, the 4 cycles, P1 brood 160 stockµs [28 in,32 ]. theNon-contact current experiment. mode of Electroporation electroporation parameters with the electrode were set at 1–2 6 kV, mm 27 above pulses, the 0.8 buffer s burst, was 4 cycles, applied. 160This μs [28,32]. procedure Non generated‐contact mode the P 1ofbrood electroporation stock used with in the the current electrode experiment. 1–2 mm above the buffer was applied. This procedure generated the P1 brood stock used in the current experiment. 4.3. Artificial Spawning and Embryo Culture 4.3. ArtificialDuring Spawning catfish spawningand Embryo seasons Culture when water temperature was 25 ◦C or higher, gravid females andDuring well developed catfish spawning males were seasons injected when or water implanted temperature with luteinizing was 25 °C hormone or higher, releasing gravid females hormone −1 andanalogue well developed (LHRHa) males to induce were injected the ovulation or implanted of eggs with and spermluteinizing with hormone a priming releasing dose of hormone 30 µg·kg −1 analoguefollowed (LHRHa) by a resolving to induce dose the of ovulation 150 µg·kg of doseeggs 12and h latersperm [43 with,44]. a When priming females dose began of 30 μ tog ovulate∙kg−1 −1 followedthey were by a anesthetized resolving dose in 200 of mg150· μLg∙kgtricaine−1 dose methanesulfonate 12 h later [43,44]. (Finquel:When females MS-222) began with to equal ovulate parts theyof NaHCOwere anesthetized3 to maintain in 200 neutral mg∙ pHL−1 tricaine for 10 min methanesulfonate and then rinsed (Finquel: with freshwater MS‐222) [ 45with]. Eggs equal were parts then of handNaHCO stripped3 to maintain into a metal neutral pan pH greased for 10 withmin and vegetable then rinsed shortening. with freshwater [45]. Eggs were then hand strippedP1 males into and a metal females pan putatively greased with transgenic vegetable with shortening. the same transgene constructs were used to produceP1 males F1 offspring.and females Since putatively P1 transgenic transgenic fish arewith always the same mosaic transgene and usually constructs mosaic were by tissueused to [28 ], produce F1 offspring. Since P1 transgenic fish are always mosaic and usually mosaic by tissue [28], “transgenic” brood stock were identified based on the presence of the transgene in their F1 progeny. Thus, the initial incubation and treatments for the F1 were blind in regards to transgenic and non‐ transgenic groups. Testes were collected from euthanized males with well‐developed secondary characteristics. Sperm were collected and then diluted with 0.9% saline at a 1:9 sperm: saline ratio. For fertilization, 2–3 mL diluted sperm solution was added to about 300 eggs then mixed. Enough pond water to submerse the eggs was added for gamete activation and fertilization. After 5 min, the pans with embryos were transferred to a trough with flow‐through water and calcium for 1 h for water hardening. Embryos for treated groups (T) were then transferred into aerated tubs with 5 L modified Holtfreter’s solution (NaCl 4 g, NaHCO3 0.2 g, KCl 0.05 g, MgSO4

Mar. Drugs 2017, 15, 155 19 of 23

“transgenic” brood stock were identified based on the presence of the transgene in their F1 progeny. Thus, the initial incubation and treatments for the F1 were blind in regards to transgenic and non-transgenic groups. Testes were collected from euthanized males with well-developed secondary characteristics. Sperm were collected and then diluted with 0.9% saline at a 1:9 sperm: saline ratio. For fertilization, 2–3 mL diluted sperm solution was added to about 300 eggs then mixed. Enough pond water to submerse the eggs was added for gamete activation and fertilization. After 5 min, the pans with embryos were transferred to a trough with flow-through water and calcium for 1 h for water hardening. Embryos for treated groups (T) were then transferred into aerated tubs with 5 L modified Holtfreter’s solution (NaCl 4 g, NaHCO3 0.2 g, KCl 0.05 g, MgSO4 333 µL (300 g in 500 mL), CaCl2 333 µL (150 g in 500 mL), pH: 7–7.5 in 1.0 L Dechlorinated water) [46]. Embryos from untreated groups (U) were transferred into flow through troughs fitted with aeration and paddlewheels for incubation. The repressor treatment was 4 ppt sodium chloride, and, in this case, incubation was in tubs holding approximately 100 embryos and the treatment was continuous for the first 6 days of embryonic development. Incubation solutions were changed every 24 h.

4.4. Embryo Sample Collection and RNA Extraction Triplicate samples of embryos were collected at 24, 48 and 120 h post-fertilization (hpf) for both T and U groups. A total of 10–15 embryos were collected per replicate, paper towel dried, placed into 1.5 mL Eppendorf tube and immediately frozen in liquid nitrogen. Samples were then stored in −80 ◦C for future RNA or DNA extraction. Non-transgenic samples were also collected as a control. Samples were ground into powder and dissolved in Invitrogen TRIzol® reagent (catalog # 15596-018), and RNA was then extracted following the manufacturer’s instructions. The quality and concentration of all the RNA samples were checked by both gel electrophoresis and with a Thermo scientiﬁc ® Nanodrop2000 . All extracted samples had an A260/280 ratio greater than 1.8, and were diluted to around 500 ng·µL−1.

4.5. PCR PCR was used to identify transgenic positive and negative F1 families. Embryos were sampled from all families at 24 hpf, and DNA was extracted using QIAGEN® buffer set, following the procedures of the QIAGEN® genomic DNA handbook (2012), then puriﬁed and diluted to work as PCR template. Three sets of 3 pooled embryos were tested for each family for possession of transgenes. Eight pairs of primers were designed to amplify speciﬁc segments of the FRMwg plasmid bone structure sequence corresponding to the two different promoter sequences and the four different knockdown sequences (Table4). PCR products were gel electrophoresed to identify positive families.

4.6. Real-Time PCR RNAs were reverse transcribed into cDNAs by iScript Synthesis Kit (Bio-Rad, catalog #: 170-8891). Each reaction consisted of a total volume of 10 µL containing 4.0 µL iScript reaction mix, 1.0 µL iScript reverse transcriptase, 500 ng RNA template, and water to reach the 10.0 µL volume. The reaction followed the protocol: 5 min at 25 ◦C, 30 min at 42 ◦C, 5 min at 85 ◦C. cDNAs were then diluted to 200 ng·µL−1. qPCR was conducted with a C1000 Thermo-Cycler (Bio-Rad, USA) by using the SsoFastTM EvaGreen® supermix kit (Bio-Rad, catalog: #172-5201). Reactions were performed in a 10 µL total reaction volume (9.0 µL mix and 1.0 µL cDNA). The mix contained 1.0 µL of each primer (5 µmol·µL−1), 4.0 µL SsoFast EvaGreen supermix, and 3.0 µL RNase/DNase-free water. The same cycling conditions were used for all the tested samples: (1) denaturation, 95 ◦C for 30 s; (2) 40 cycles of 95 ◦C for 5 s, and 57 ◦C for 15 s; followed by (3) melting curve analysis, 5 s at 65 ◦C, then up to 95 ◦C with a 0.1 ◦C temperature increase every second. The mRNA levels of all the samples were normalized to the levels of the non-transgenic control sample of the same time point. Ribosome 18 s mRNA was used as a Mar. Drugs 2017, 15, 155 20 of 23 reference gene. Crossing-point (Ct) values were exported into Excel sheet from Bio-Rad CRX Manager (Version 1.6.541.1028, 2008). Su et al. [32] designed primers used for qPCR for the current experiment.

Table 4. List of primers used for PCR identification of transgenic channel catfish (Ictalurus punctatus) for 8 primordial germ cell knockdown constructs. Construct components were: ADSS, zebrafish adenylosuccinate synthase 2 (adss2) promoter; RM, zebrafish racemase promoter; N1, 50 end ds-shRNA targeting channel catfish nanos gene; N2, 30 end ds-shRNA targeting channel catfish nanos gene; cDNA, full-length cDNA sequence of channel catfish nanos gene; DND, ds-shRNA targeting channel catfish dead end gene.

Primers Sequence (50–30) Forward primer for insulator ACACTGTCCTTGGTATCAGCA Reverse primer for insulator ACACCTCCTTGATCCTGTGC Forward primer for ocean pout terminator TGCCCAAATTTCCTGCCTGA Reverse primer for ocean pout terminator TGCCTGTGAGGGTGACAAAA Forward primer for adss2 promoter CCGCCAAAATACGACTG Reverse primer for adss2 promoter TAAGCGAGGGAACAACG Forward primer for racemase promoter TACAACACCCACAATCCC Reverse primer for racemase promoter CACCGTAACAAAGAAACTCC Forward primer for N1 ATGACTCGCGATGGTGTTTT Reverse primer for N1 CATACTGGGGTGTTTGCTGA Forward primer for N2 CACTCGCTAGAGTGGGGTTC Reverse primer for N2 TCATACTGGGGTGTTTGCTG Forward primer for cDNA GAACCCCACTCTCACACTCG Reverse primer for cDNA AGACCCGCACCTGTAACATG Forward primer for DND TAGCGCCGCTCTACGAGT Reverse primer for DND TCATACTGGGGTGTTTGCTG

4.7. Spawning Evaluation

From 2012 to 2015, 99 P1 females electroporated with (exposed to) the salt sensitive sterilization constructs as one-cell embryos were injected with LHRHa to induce spawning or were evaluated for spawning readiness. A total of 58 were from the untreated group and 41 were from the treated group (Table1). The spawning rates were recorded by transgenic treated and untreated, 2 promoter and 4 knockdown strategies, resulting in 16 groups in total. The records were then summarized and analyzed to reveal the spawning rate differences between knockdown strategies, between promoters and between treated and untreated groups.

4.8. Statistical Analysis Fisher’s Exact Test and Fisher’s Multi-treatment Exact Test were used to compare rates of sexual maturity and spawning between and among the different treatment groups. REST (Relative Expression Software Tool) [47], using Pairwise Fixed Reallocation Randomization Test, was applied to compare the RNA expression level of the three PGC migration marker genes (nanos, vasa and dead end) among the treated group, untreated group and non-transgenic group and 18S was used as reference gene for this evaluation. Group means were used to express differences in expression between untreated, non-transgenic and treated samples. Statistical signiﬁcance was performed by randomization test with 2000 randomizations per test (p < 0.05).

Acknowledgments: This work was partially funded by the United States Department of Agriculture- Biotechnology Risk Assessment Grant Program (grant No. 2009-33522-05774). The FRMwg plasmid was supplied by Patrick Gibbs from the Rosenstiel School of Marine and Atmospheric Science, University of Miami, Florida. Author Contributions: Experimental design and project concept: R. D., B. S. and H.L. Molecular data collection and analysis: H. L. and B. S. Field work, brood stock preparation, spawning, hatching and growth experiments and data collection: H. L., B. S., G. Q., Z. Y., A. A., D. P., M. S., R. O., K. V., D. D., D. R., D. Z., N. A. Manuscript writing: H. L. and R. D. Mar. Drugs 2017, 15, 155 21 of 23

Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. AquaBounty Technologies. Available online: https://aquabounty.com/innovation/technology (accessed on 25 January 2017). 2. Devlin, R.H.; Sundström, L.F.; Leggatt, R.A. Assessing ecological and evolutionary consequences of growth-accelerated genetically engineered fishes. BioScience 2015, 65, 685–700. [CrossRef] 3. Rasmussen, R.S.; Morrissey, M.T. Biotechnology in aquaculture: Transgenics and polyploidy. Compr. Rev. Food Sci. Food Saf. 2007, 6, 2–16. [CrossRef] 4. Mair, G.C.; Nam, Y.K.; Solar, I. Environmental Risk Assessment of Genetically Modified Organisms, Volume 3. In Methodologies for Transgenic Fish; School of Biological Sciences, Flinders University: Adelaide, Australia, 2007. 5. Molyneaux, K.; Wylie, C. Primordial germ cell migration. Int. J. Dev. Biol. 2004, 48, 537–544. [CrossRef] [PubMed] 6. Köprunner, M.; Thisse, C.; Thisse, B.; Raz, E. A zebrafish nanos-related gene is essential for the development of primordial germ cells. Genes Dev. 2001, 15, 2877–2885. [PubMed] 7. Li, M.; Hong, N.; Xu, H.; Yi, M.; Li, C.; Gui, J.; Hong, Y. Medaka vasa is required for migration but not survival of primordial germ cells. Mech. Dev. 2009, 126, 366–381. [CrossRef][PubMed] 8. Richardson, B.E.; Lehmann, R. Mechanisms guiding primordial germ cell migration: Strategies from different organisms. Nat. Rev. Mol. Cell Biol. 2010, 11, 37–49. [CrossRef][PubMed] 9. Saito, T.; Fujimoto, T.; Maegawa, S.; Inoue, K.; Tanaka, M.; Arai, K.; Yamaha, E. Visualization of primordial germ cells in vivo using GFP-nos1 30UTR mRNA. Int. J. Dev. Biol. 2006, 50, 691–699. [CrossRef][PubMed] 10. Wong, T.T.; Collodi, P. Inducible sterilization of zebrafish by disruption of primordial germ cell migration. PLoS ONE 2013, 8, e68455. [CrossRef][PubMed] 11. Slanchev, K.; Stebler, J.; de la Cueva-Méndez, G.; Raz, E. Development without germ cells: The role of the germ line in zebrafish sex differentiation. Proc. Nat. Acad. Sci. USA 2005, 102, 4074–4079. [CrossRef] [PubMed] 12. Weidinger, G.; Stebler, J.; Slanchev, K.; Dumstrei, K.; Wise, C.; Lovell-Badge, R.; Thisse, C.; Thisse, C.; Raz, E. Dead end, a novel vertebrate germ plasm component, is required for zebrafish primordial germ cell migration and survival. Curr. Biol. 2003, 13, 1429–1434. [CrossRef] 13. Fan, L.; Moon, J.; Wong, T.T.; Crodian, J.; Collodi, P. Zebrafish primordial germ cell cultures derived from vasa::RFP transgenic embryos. Stem Cells Dev. 2008, 17, 585–597. [CrossRef][PubMed] 14. Yoon, C.; Kawakami, K.; Hopkins, N. Zebrafish vasa homologue RNA is localized to the cleavage planes of 2- and 4-cell-stage embryos and is expressed in the primordial germ cells. Development 1997, 124, 3157–3165. [PubMed] 15. Herpin, A.; Rohr, S.; Riedel, D.; Kluever, N.; Raz, E.; Schartl, M. Specification of primordial germ cells in medaka (Oryzias latipes). BMC Dev. Biol. 2007, 7, 3. [CrossRef][PubMed] 16. Braat, A.K.; van de Water, S.; Korving, J.; Zivkovic, D. A zebrafish vasa morphant abolishes vasa protein but does not affect the establishment of the germline. Genesis 2001, 30, 183–185. [CrossRef][PubMed] 17. Gavis, E.R.; Lunsford, L.; Bergsten, S.E.; Lehmann, R. A conserved 90 nucleotide element mediates translational repression of nanos RNA. Development 1996, 122, 2791–2800. [PubMed] 18. Kedde, M.; Strasser, M.J.; Boldajipour, B.; Vrielink, J.A.; Slanchev, K.; le Sage, C.; Nagel, R.; Voorhoeve, P.M.; van Duijse, J.; Orom, U.A.; et al. RNA-binding protein Dnd1 inhibits microRNA access to target mRNA. Cell 2007, 131, 1273–1286. [CrossRef][PubMed] 19. Saurabh, S.; Vidyarthi, A.S.; Prasad, D. RNA interference: Concept to reality in crop improvement. Planta 2014, 239, 543–564. [CrossRef][PubMed] 20. Daneholt, B. Advanced information: RNA interference. The Nobel Prize in Physiology or Medicine 2006. Archived from the Original on 20 January 2007. 21. Thresher, R.; Grewe, P.; Patil, J.G.; Whyard, S.; Templeton, C.M.; Chaimongol, A.; Hardy, C.; Hinds, L.A.; Dunham, R.A. Development of repressible sterility to prevent the establishment of feral populations of exotic and genetically modified animals. Aquaculture 2009, 290, 104–109. [CrossRef] Mar. Drugs 2017, 15, 155 22 of 23

22. Bernick, E.P.; Zhang, P.J.; Du, S. Knockdown and overexpression of Unc-45b result in defective myofibril organization in skeletal muscles of zebrafish embryos. BMC Cell Biol. 2010, 11, 70. [CrossRef][PubMed] 23. Prelich, G. Gene Overexpression: Uses, mechanisms, and interpretation. Genetics 2012, 190, 841–854. [CrossRef][PubMed] 24. Thresher, R.; Hinds, L.A.; Hardy, C.; Whyard, S.; Vignarajan, S.; Grewe, P.; Patil, J. Repressible Sterility of Animals. U.S. Patent 20050071891, 31 March 2005. 25. Borode, A.O.; Balogun, A.M.; Omoyeni, B.A. Effect of salinity on embryonic development, hatchability, and growth of african catfish, Clarias gariepinus, eggs and larvae. J. Appl. Aquac. 2008, 12, 89–93. [CrossRef] 26. Swann, L.; Fitzgerald, S. The Use and Application of Salt in Aquaculture; Purdue University: West Lafayette, IN, USA, 1992. 27. Hoshijima, K.; Hirose, S. Expression of endocrine genes in zebrafish larvae in response to environmental salinity. J. Endocrinol. 2007, 193, 481–491. [CrossRef][PubMed] 28. Powers, D.A.; Cole, T.; Creech, K.; Chen, T.T.; Lin, C.M.; Kight, K.; Dunham, R.A. Electroporation: A method for transferring genes into the gametes of zebrafish, Danio rerio, channel catfish, Ictalurus punctatus, and common carp, Cyprinus carpio. Mol. Mar. Biol. Biotechnol. 1992, 1, 301–309. [PubMed] 29. Cheng, Q.; Su, B.F.; Qin, Z.K.; Weng, C.C.; Yin, F.; Zhou, Y.G.; Fobes, M.; Rex, D.A. Interaction of diet and the masou salmon ∆5-desaturase transgene on ∆6-desaturase and stearoyl-CoA desaturase gene expression and N-3 fatty acid level in common carp (Cyprinus carpio). Transgen. Res. 2014, 23, 729–742. [CrossRef][PubMed] 30. Gibbs, P.D.; Schmale, M.C. GFP as a genetic marker scorable throughout the life cycle of transgenic zebra fish. Mar. Biotechnol. 2000, 2, 107–125. [PubMed] 31. Xu, H.; Lim, M.; Dwarakanath, M.; Hong, Y. Vasa identifies germ cells and critical stages of oogenesis in the asian seabass. Int. J. Biol. Sci. 2004, 10, 225–235. [CrossRef][PubMed] 32. Su, B.F.; Shang, M.; Grewe, P.M.; Patil, J.G.; Peatman, E.; Perera, D.A.; Cheng, Q.; Li, C.; Weng, C.C.; Li, P.; et al. Suppression and restoration of primordial germ cell marker gene expression in channel catfish, Ictalurus punctatus, using knockdown constructs regulated by copper transport protein gene promoters: Potential for reversible transgenic sterilization. Theriogenology 2015, 84, 1499–1512. [CrossRef][PubMed] 33. Su, B.F.; Peatman, E.; Shang, M.; Thresher, R.; Grewe, P.; Patil, J.; Pinkert, C.A.; Irwin, M.H.; Li, C.; Perera, D.A.; et al. Expression and knockdown of primordial germ cell genes, vasa, nanos and dead end in common carp (Cyprinus carpio) embryos for transgenic sterilization and reduced sexual maturity. Aquaculture 2014, 420–421, 72–84. [CrossRef] 34. Zhang, D. Transgenic Disruption of Aromatase Using the Daughterless Construct to Alter Sex Ratio in Common Carp, Cyprinus carpio. Master’s Thesis, Auburn University, Auburn, AL, USA, 1 August 2016. 35. Saxena, S.; Jonsson, Z.O.; Dutta, A. Small RNAs with imperfect match to endogenous mRNA repress translation. J. Biol. Chem. 2003, 278, 44312–44319. [CrossRef][PubMed] 36. Scacheri, P.C.; Rozenblatt-Rosen, O.; Caplen, N.J.; Wolfsberg, T.G.; Umayam, L.; Lee, J.C.; Hughes, C.M.; Shanmugam, K.S.; Bhattacharjee, A.; Meyerson, M.; et al. Short interfering RNAs can induce unexpected and divergent changes in the levels of untargeted proteins in mammalian cells. Proc. Nat. Acad. Sci. USA 2004, 101, 1892–1897. [CrossRef][PubMed] 37. Slanchev, K.; Stebler, J.; Goudarzi, M.; Cojocaru, V.; Weidinger, G.; Raz, E. Control of Dead end localization and activity—Implications for the function of the protein in antagonizing miRNA function. Mech. Dev. 2009, 126, 270–277. [CrossRef][PubMed] 38. Raz, E. The function and regulation of vasa-like genes in germ-cell development. Gen. Biol. 2000, 1, 1017.1–1017.6. [CrossRef][PubMed] 39. Reynolds, N.; Collier, B.; Maratou, K.; Bingham, V.; Speed, R.M.; Taggart, M.; Semple, C.A.; Gray, N.K.; Cooke, H.J. Dazl binds in vivo to specific transcripts and can regulate the pre-meiotic translation of Mvh in germ cells. Hum. Mol. Genet. 2005, 14, 3899–3909. [CrossRef][PubMed] 40. Kato, Y.; Katsuki, T.; Kokubo, H.; Masuda, A.; Saga, Y. Dazl is a target RNA suppressed by mammalian NANOS2 in sexually differentiating male germ cells. Nat. Commun. 2016, 7, 11272. [CrossRef][PubMed] 41. PIPs: Human protein-protein interactions prediction database. Available online: http://www.compbio. dundee.ac.uk/www-pips/evidence.jsp?Prot=IPI00395811&Prot1=21244&Prot2=10958&Interactions= Prot&LR=0.025&modules=on (accessed on 2 May 2017). 42. Perry, W.G. Note on the spawning of blue and channel catfish in brackish water ponds. Prog. Fish Cultur. 1973, 35, 164–166. [CrossRef] Mar. Drugs 2017, 15, 155 23 of 23

43. Perera, D.A. Studies for Improvement of Reproductive Biotechnology for Production of Channel Catfish (Ictalurus punctatus) Female X Blue Catfish (Ictalurus furcatus) Male Hybrid Embryos; Auburn University: Auburn, AL, USA, 2012. 44. Su, B.F.; Perera, D.A.; Zohar, Y.; Abraham, E.; Stubblefield, J.; Fobes, M.; Beam, R.; Argue, B.; Ligeon, C.; Padi, J.; et al. Relative effectiveness of carp pituitary extract, luteininzing hormone releasing hormone analog (LHRHa) injections and LHRHa implants for producing hybrid catfish fry. Aquaculture 2013, 372–375, 133–136. [CrossRef] 45. Chatakondi, N.G.; Yant, D.R.; Dunham, R.A. Final Report-Spawning and Hatchery Technology to Improve Hybrid Catfish Fry Production; SBIR project No. 2002-00384; U.S. Department of Agriculture (USDA): Washington, DC, USA, 2005. 46. Armstrong, J.B.; Duhon, S.T.; Malacinski, G.M. Raising the Axolotl in Captivity. In Developmental Biology of the Axolotl; Armstrong, J.B., Malacinski, G.M., Eds.; Oxford University Press: New York, NY, USA, 1989; pp. 220–227. 47. Pfaffl, M.W.; Horgan, G.W.; Dempfle, L. Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res. 2002, 30, e36. [CrossRef][PubMed]

© 2017 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 (http://creativecommons.org/licenses/by/4.0/).