Nondestructive Larval Genotyping of Danio rerio for Mitochondrial and Nuclear DNA Genetics

Zebrafish handling and husbandry

All adult zebrafish and embryos were maintained according to the guidelines established by the Mayo Clinic Institutional Animal Care and Use Committee (IACUC number: A34513-13-R16) and The University of Texas at Austin Institutional Animal Care and Use Committee (IACUC number: AUP-2023–00200).

ProK-based larval genotyping

WT Mayo Recessive Free zebrafish and mutant transgenic lines, mt-tl1, ¹⁴ were used to generate larvae for nondestructive genotyping. ²² Three days post-fertilization (dpf) larvae were rinsed three times in embryo media followed by three washes in tricaine buffer (30 mM Tris-HCl in 0.4% tricaine solution). Using a p200 pipette tip trimmed to accommodate larval width, individual larvae were aspirated in 40 μL of ProK genotyping buffer (25 μg/mL ProK [catalog no.: EO0491, Thermo Fisher Scientific, USA] in tricaine buffer) and placed in 96-well (catalog no.: 3455, Thermo Fisher Scientific, USA) and 48-well plates (catalog no.: CC7672-7548, CytoOne). For 48-well plates, droplets were dispensed centrally within each well to minimize contact between larvae and well walls during agitation. ProK concentration and incubation time can be adjusted based on larval age and condition to optimize survival rates. The plate was incubated in a heated shaker (catalog no.: VORTEMP 56, Labnet International, Inc., USA) at 37°C with 500 rpm shaking for 20 min. After incubation, 30 μL of the liquid around the fish was carefully pipetted (avoiding larval aspiration) into 15 μL of lysis buffer (10 mM Tris-HCl, pH 8.0, 50 mM KCl, 0.3% Tween-20, 0.3% NP-40, 1 mM EDTA) in 0.2 mL strip tubes and pipetted up and down ∼10 times. The solution was then incubated at 98°C for 5 min to extract DNA. Following media collection, wells were quickly replenished with embryo media using a Pasteur pipette, and larvae were transferred to 24-well plates containing fresh embryo media.

To prevent tail damage during the shaking procedure, plates can be pretreated with a protein-rich blocking solution. If extensive tail damage is observed within shaken larvae, plates can be incubated with 5% nonfat dry milk (NFDM) solution prepared in embryo media for 5 min, followed by three washes with embryo media before genotyping (Fig. 1A).

OPEN IN VIEWER

FIG. 1.

Nondestructive enzyme genotyping workflow and validation of survival and sensitivity. (A) Schematic of nondestructive enzyme genotyping protocol. (B) Schematic of the zebrafish mitochondrial genome showing genes targeted for nondestructive genotyping (highlighted in red boxes). (C) Box-and-whisker plot comparing survival percentages of larvae 24 h post-genotyping in 96-well, 48-well, and unshaken control groups. Whiskers represent maximum and minimum values. p Values were determined by Student’s t-test (*p < 0.05; **p < 0.01; ***p < 0.001). (N = 100 technical replicates across five experiments). (D) Sensitivity of the protocol is represented by the percentage of successfully amplified samples across four mitochondrial genes (N = 60 technical replicates across three experiments). No statistically significant difference was observed between the well formats with respect to amplicon sensitivity. Error bars are represented as standard error of the mean. Figure 1A and B was created using Biorender.com.

Survival assay

Larvae collected after enzymatic genotyping were transferred into individual wells of a 24-well plate. Unshaken age-matched zebrafish larvae obtained from the same clutch were used as controls. Live counts for larvae from each group were recorded at 24 h post-genotyping, and morphological defects were examined pre- and post-procedure.

PCR amplification and restriction fragment length polymorphism for mitochondrial genetic loci

To investigate the sensitivity of the protocol for mtDNA genotyping, we prioritized protein-coding genes from each of the respiratory chain complex, transfer RNA, and ribosomal RNA genes. We targeted six mitochondrial genetic loci with distinct functional roles. Four of them were protein coding genes: mitochondrially encoded NADH:ubiquinone oxidoreductase core subunit 1 (mt-nd1), mitochondrially encoded cytochrome b (mt-cytb), mitochondrially encoded ATP synthase membrane subunit 6 (mt-atp6), and mitochondrially encoded cytochrome c oxidase subunit 1 (mt-co1). The remaining two were noncoding genes: mitochondrially encoded tRNA-Leu (mt-tl1) and mitochondrially encoded 12S rRNA (mt-rnr1) (Fig. 1B). PCR amplification was performed for each of the six genes using their respective primers mentioned in Supplementary Table S1. PCR consisted of 10 μL of 2X OneTaq master mix with standard buffer (NEB, USA), 1 μL of 10 μM forward primer, 1 μL of 10 μM reverse primer, and 8 μL of template. Samples were run in a thermal cycler with the following PCR conditions: (1) 95°C, 4 min; (2) 95°C, 30 s; (3) 60°C, 30 s; (4) 72°C, 30 s; (5) repeat steps 2–4 for 30 cycles; (6) 72°C, 4 min; and (7) 12°C, hold. PCR amplicons were run on 1% agarose gel and analyzed.

Five to eight microliters of the PCR amplicon (depending on the intensity of amplicon on the gel) was used as the substrate for restriction fragment length polymorphism (RFLP) in 20 μL of reaction solution (2 μL of the 10X corresponding buffer and 5U of restriction enzyme and remaining volume with nuclease-free water). PCR amplicons for genetic loci mt-nd1, mt-cytb, mt-tl1, and mt-rnr1 were surveyed for restriction digestion by SpeI, AvaI, SacI, and HindIII, respectively. A subset of the PCR amplicons was then submitted to Sanger sequencing at GENEWIZ (Azenta, USA) to confirm sequence specificity. Chromatogram files (.ab1) were then uploaded to the EditR ²³ web server to estimate the heteroplasmy in the mtDNA mutants.

PCR amplification of nuclear genetic loci

To investigate the sensitivity of the protocol for nuclear gene genotyping, we prioritized protein-coding nuclear genes with roles in mitochondrial function and signaling. The three nuclear loci were as follows: mitochondrial antiviral signaling (mavs), cytochrome c oxidase subunit 7B (cox7b), and leucine-rich pentatricopeptide repeat-containing (lrpprc). PCR amplification was performed for each of these genes using their respective primers listed in Supplementary Table S1. PCRs consisted of 10 μL of 2X OneTaq master mix with standard buffer (NEB, USA), 1 μL of 10 μM forward primer, 1 μL of 10 μM reverse primer, and 8 μL of template DNA. PCR amplification was carried out under the following conditions: (1) 95°C, 4 min; (2) 95°C, 30 s; (3) annealing at 58°C for cox7b and lrpprc, or 54°C for mavs, 30 s; (4) 72°C, 30 s for mavs and lrpprc; and 2 min for cox7b; (5) repeat steps 2–4 for 34 cycles; (6) 72°C, 4 min; (7) hold at 12°C. PCR amplicons were run on 1% agarose gel and analyzed. A subset of PCR products was submitted to Sanger sequencing at GENEWIZ (Azenta, USA) to confirm sequence specificity.

RNA extraction and sample preparation

To assess for any differential transcriptional signatures in the ProK-shaken larvae to those unshaken, we performed whole-body RNA sequencing for the two groups. Adult zebrafish were incrossed and nondestructive genotyping was performed on larvae aged 3 dpf. Larvae were subjected to protocol for the nondestructive enzymatic ProK conditions and were then allowed to recover in embryo water until 6 dpf. Six larvae were pooled, and RNA isolation was performed using the Qiagen RNeasy Mini kit as per the manufacturer’s instructions. An additional DNAse treatment step was performed before the final wash and elution. RNA concentration was measured using a spectrophotometer.

RNA sequencing and analyses

RNA library preparations and sequencing reactions were performed at GENEWIZ (Azenta USA). For transcriptome sequencing, RNA was isolated by pooling six individual larvae from each condition. RNA quantification was performed using the Qubit 2.0 Fluorometer (Thermo Fisher Scientific, USA), and RNA integrity was assessed using the Agilent TapeStation (Agilent Technologies, USA). RNA sequencing libraries were constructed following the NEBNext Ultra II RNA Library Prep Kit for Illumina protocol (NEB, USA). Briefly, mRNA was enriched using oligo(dT) beads, fragmented for 15 min at 94°C, and then reverse-transcribed into cDNA. The cDNA was end-repaired, adenylated at the 3′ ends, and ligated with universal adapters, followed by index addition and PCR-based library amplification. The sequencing library was validated on the Agilent TapeStation (Agilent Technologies, USA) and quantified using both the Qubit 2.0 Fluorometer and quantitative PCR (KAPA Biosystems, USA). The sequencing libraries were then multiplexed, clustered onto a flow cell, and loaded onto the Illumina HiSeq X instrument according to the manufacturer’s instructions. Sequencing was performed using a 2 × 150 bp paired-end (PE) configuration. Image analysis and base calling were carried out with HiSeq Control Software. The raw sequence data (.bcl files) generated from the Illumina HiSeq X platform were converted into fastq files and demultiplexed using Illumina’s bcl2fastq 2.20 software with one mismatch allowed for index sequence identification. The raw fastq data were uploaded to the Basepair cloud bioinformatic platform (https://www.basepairtech.com/). Filtered reads were aligned to the zebrafish reference genome (Zv11 version) using STAR. Raw and normalized gene and transcript counts were calculated. Differential expression analysis was performed using DESeq2, ²⁴ with a threshold of p adjusted <0.05 for identifying differentially expressed genes between the genotyped and unshaken control larvae.

Survival rates of enzyme-genotyped zebrafish larvae support high-throughput screening

Nondestructive enzyme genotyping of zebrafish larvae demonstrates high survival rates and compatibility with high-throughput screening. Larvae shaken in 96-well plates without milk pretreatment had a median survival rate of 75%, while 48-well larvae had a median of 94%, compared with 100% in unshaken controls (Fig. 1C). Most larval deaths occurred within 24 h post-shaking, primarily due to tail fin damage, with no observed differences in survival between undamaged shaken and unshaken fish after this period. Key survival-enhancing steps included rapidly replacing embryo water after DNA collection and promptly removing larvae from the heated shaker. These parameters can be further adapted to accommodate different larval ages and sensitive phenotypes. Importantly, no morphological changes were observed in surviving genotyped larvae compared with unshaken controls, suggesting the method’s minimal physiological impact.

Mitigating characteristic tail fin damage among ProK-genotyped larvae via 48-well plates and milk solution pretreatment

Our initial attempts using 96-well plates for ProK genotyping resulted in a subset of larvae exhibiting fatal damage, including partial-to-complete tail fin tearing. We attributed this damage pattern to tail fin adhesion to the well walls during heated shaking, exacerbated by ProK-induced weakening of fin structural integrity. We determined that this damage pattern likely resulted from tail fin adhesion to well walls during heated shaking, exacerbated by ProK-induced weakening of fin structural integrity. Two effective mitigation strategies were thus developed. First, utilization of 48-well plates enabled placement of larvae in centered droplets without wall contact, significantly increasing survival and reducing tail fin damage. However, the efficacy of this approach may vary depending on the strength of the hydrophobic coating applied by different manufacturers to maintain protective droplet formation. Alternatively, pretreating 96-well plates with 5% NFDM solution created a protective proteinaceous film that reduced larval tail fin adhesion during shaking (Fig. 1A). We observed decreased tail fin damage and improved survival rates among genotyped larvae in the 96-well plate treated with 5% NFDM in embryo media (Supplementary Fig. S1).

High sensitivity of amplification across diverse mitochondrial genetic loci

Nondestructive enzyme genotyping demonstrates high amplification sensitivity across diverse mitochondrial genes encoding critical cellular machinery. Amplification sensitivity was assessed for mt-nd1, mt-cytb, mt-rnr1, and mt-tl1 by calculating the percentage of successfully amplified samples across these genes. For each gene–well combination, we conducted three experimental trials, with each trial containing 20 larvae. Thus, we analyzed a total of 60 larvae for each gene–well combination, resulting in 120 total samples per gene across both formats. Consistently high amplification rates were observed across all four genes in both formats. All experimental sets exhibited over 80% sensitivity (Fig. 1D), demonstrating robust performance of this nondestructive genotyping method across diverse mitochondrial genes.

Specificity of mitochondrial DNA amplification verified by PCR and RFLP analysis

We validated the specificity of our nondestructive enzyme genotyping method through comprehensive molecular analysis of four mitochondrial genes. PCR amplification and RFLP analysis were performed on WT zebrafish larval DNA samples from 96-well and 48-well plates, with all samples referenced against Quick-Load Purple 50 bp DNA Ladder (NEB, USA). For mt-nd1, we observed a 350 bp PCR amplicon with SpeI restriction digest yielding 236 bp and 114 bp fragments (Fig. 2A, B). The mt-cytb gene generated a 460 bp PCR amplicon with AvaI restriction digest yielding 317 bp and 143 bp fragments (Fig. 2D, E). Amplification of mt-rnr1 showed a 395 bp PCR amplicon with HindIII restriction digest producing 253 bp and 142 bp fragments (Fig. 2G, H). The mt-tl1 gene produced a 363 bp PCR product (Fig. 3A, B) with SacI restriction digest creating 186 bp and 177 bp fragments. PCR amplification of mt-atp6 produced a 369 bp amplicon (Supplementary Fig. S2A), while mt-co1 generated a 494 bp PCR product (Supplementary Fig. S2B). Sanger sequencing confirmed the presence of restriction enzyme sites for the first four genes (Fig. 2C, F, I, 3E), and all samples selected for gel visualization represented the median amplification range within their respective cohorts.

OPEN IN VIEWER

FIG. 2.

Specificity of mitochondrial DNA amplification. (A) Gel electrophoresis of the mitochondrial gene mt-nd1, showing amplification of the 350 bp band. (B) Screening of mt-nd1 amplicon by restriction fragment length polymorphism (RFLP). PCR amplicons from wild-type larvae show the presence of uncut product (350 bp) and expected 236 and 114 bp digested bands. UD = undigested PCR product. (C) Representative chromatogram of the mt-nd1 amplicon depicting the restriction enzyme recognition site. (D) Agarose gel depicting amplification of the mitochondrial gene mt-cytb, with the expected 460 bp fragment. (E) RFLP analysis of the mt-cytb PCR product, showing the intact 460 bp band and the predicted 317 and 143 bp digestion fragments. UD = undigested PCR product. (F) Representative electropherogram of the mt-cytb amplicon, with the restriction enzyme recognition site indicated. (G) Visualization of mt-rnr1 amplification, showing a discrete 395 bp band. (H) Restriction digest of the mt-rnr1 PCR product, displaying both the undigested 395 bp fragment and the expected 253 and 142 bp cleavage products. UD = undigested PCR product. (I) Sanger sequencing trace of the mt-rnr1 amplicon, highlighting the enzyme cut site.

OPEN IN VIEWER

FIG. 3.

Screening for mitochondrial DNA (mtDNA) mutants using nondestructive genotyping in mt-tl1 locus. (A) PCR amplification of the mt-tl1 loci from wild-type larvae resolved on a 1% agarose gel to confirm product integrity. (B) Schematic of the SacI recognition site within the amplicon (363 bp) and the expected fragment sizes (186 and 177 bp) for both unedited and edited alleles. (C) PCR products from mt-tl1 mutants are shown on an agarose gel, demonstrating successful amplification. (D) restriction fragment length polymorphism (RFLP) analysis reveals C-to-T conversion at the target cytosine leading to loss of SacI recognition site: control larvae yield the expected 186 bp and 177 bp fragments, whereas edited samples display a single undigested band (each lane represents individual sample DNA; UD = undigested PCR product). (E) Sanger sequencing chromatograms from control and mutant larvae. Asterisk (*) denotes the site of edit with corresponding editing percentage (C-to-T). Chromatograms and editing table plot were obtained using EditR.

Genotyping of mtDNA mutant

We demonstrated the utility of our nondestructive genotyping approach for gene editing by introducing a C-to-T base substitution in zebrafish mt‐tl1 mutants. ¹⁴ Three dpf mt-tl1 zebrafish mutants were genotyped nondestructively, enabling subsequent longitudinal studies (Fig. 3C). The engineered C-to-T transition at position m.C3744 (C10 on the forward strand) resulted in the loss of the SacI restriction site, thereby facilitating the identification of larvae with mtDNA variations. RFLP analysis demonstrated the loss of the SacI restriction site in mutant larvae, with digested products of 186 bp and 177 bp, and an undigested product of 363 bp. WT controls from both 96-well and 48-well cohorts showed only digested products of 186 bp and 177 bp (Fig. 3D). RFLP analysis was confirmed by Sanger sequencing where sequencing files (.ab1) were assessed using EditR ²³ software to quantify the level of editing in the mutant larvae. Sanger sequencing confirmed the C-to-T edit, revealing over 40% mtDNA mutant heteroplasmy levels at the target locus (Fig. 3E).

Amplification across different nuclear genetic loci

To expand the utility of the optimized protocol, we adopted nondestructive genotyping to amplify different nuclear genetic loci. We observed successful amplification of the nuclear genes mavs (Fig. 4A, B), cox7b (Fig. 4C, D), and lrpprc (Fig. 4E, F) by agarose gel electrophoresis (Fig. 4). PCR amplicons for mavs and lrpprc were referenced against Quick-Load Purple 50 bp DNA Ladder (NEB, USA), and for cox7b, amplicons were referenced against HyperLadder 1 kb (Meridian Bioscience, USA). Sanger sequencing confirmed that PCR products corresponded to the intended nuclear gene loci. This demonstrated that these optimizations can be used for nuclear DNA genotyping as well.

OPEN IN VIEWER

FIG. 4.

Specificity of nuclear DNA amplification. (A–F) Gel electrophoresis demonstrates successful PCR amplification for each nuclear locus: mavs (A) yields a distinct 367 bp product; cox7b (C) a 1367 bp product; and lrpprc (E) a 380 bp product, corresponding to expected amplicon size for each locus. (B, D, F) Representative Sanger sequencing chromatograms for the mavs (B), cox7b (D), and lrpprc (F) amplicons show high-quality, clearly resolved peaks, confirming the fidelity of PCR amplification and enabling downstream sequence analysis.

Transcriptomic signature in nondestructively genotyped zebrafish larvae

To examine the genome-wide transcriptional changes in zebrafish larvae following nondestructive genotyping, we performed strand-specific PE RNA sequencing to identify a candidate differential expression profile. Sequence data were processed using the DESeq2 ²⁴ pipeline on the Basepair cloud bioinformatic platform (https://www.basepairtech.com/). For both ProK-shaken larvae and WT controls, we generated between 34 and 39 million reads. Reads below a Phred score of Q30 were discarded, and the remaining high-quality reads were aligned to the Zv11 zebrafish reference genome, with approximately 78% mapping uniquely. These uniquely mapped reads were then assembled to quantify transcript levels across the genome. Upon applying the criteria of a log₂ fold change of ≥1 or ≤–1 with an adjusted p value <0.05 in the dataset (Supplementary Fig. S3), zero downregulated genes were observed. We identified seven annotated genes and six uncharacterized genes to be significantly upregulated in the 96-well genotyping treatment compared with unshaken controls (Supplementary Fig. S3 and Supplementary Data).