Insects

We used three A. pisum strains, namely, ApL, KNb, and 09003A, originally collected from Sapporo, Japan. These strains were gifts from Dr. Shin-ichi Akimoto (Hokkaido University). The characteristics of each strain are described in a previous study by Matsuda et al. [26]. Parthenogenetic aphid clones of the stock populations were maintained on vetch seedlings (Vicia faba) in growth chambers under long-day photoperiod conditions (16 h light:8 h dark cycle) at 16 °C.

Overview of the CRISPR/Cas9-mediated knockout workflow in the pea aphid

We developed a non-homologous end-joining (NHEJ)-mediated knockout method for pea aphids using the CRISPR/Cas9 system. The workflow is illustrated in Fig 1. First, asexual reproduction was induced to obtain eggs for the CRISPR/Cas9 injection. Aphids that were maintained regularly in the laboratory during asexual viviparous reproduction were transitioned to sexual reproduction by shifting to short-day conditions for 1.5–2 months. To reduce inbreeding depression [26], we outbred two strains: ApL (female) × 0903A (male) and ApL (male) × KNb.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. An outline of the CRISPR/Cas9-mediated knockout workflow in the pea aphid.

Sexual reproduction is induced by rearing aphids under a short-day photoperiod (8 h light/ 16 h dark) at 15 °C, while the laboratory stock is maintained in a parthenogenetic, viviparous state under a long-day photoperiod (16 h light/ 8 h dark) at 16 °C. To mitigate inbreeding depression, two outcrossed combinations are used: ApL (female) × 0903A (male) and ApL (male) × KNb. A duplex of tracrRNA and target-specific crRNA (both components of the Alt-R CRISPR-Cas9 system by IDT) is incubated with Cas9 protein to form the CRISPR-Cas9 ribonucleoprotein (RNP) complex. The RNP complex was injected into the mid-to-posterior region of 0–24 h AEL (after egg laying) eggs, carefully avoiding disruption of the bacterial symbiont cluster located at the posterior pole.

https://doi.org/10.1371/journal.pgen.1011557.g001

Oviparous females that mated with males were allowed to lay eggs for 24 h, and the RNP complex was injected into the eggs (0–24 h after egg laying (AEL)). This injection time window was chosen because primordial germ cells (PGCs) are cellularized at approximately 24–28 h AEL during embryogenesis [27] to facilitate efficient germline mutagenesis. Our injection targeted a region anteriorly adjacent to the cluster of endosymbiotic bacteria at the posterior end of the egg, where the germ plasm is localized and PGCs are formed. The germ plasm is invisible, but the symbiont cluster is visible in green. Although laid eggs initially exhibited an ivory color, healthy eggs turned black at 2–3 d AEL, which indicates survival without damage from the microinjection. A small subset of eggs was collected and subjected to amplicon sequencing to estimate the mutation efficiency in each batch of experiments.

Injected eggs were incubated at 16°C for 14 days and subsequently transferred to 4°C for 45 days in a moist chamber (see Diapause and hatching section below for details) to mimic natural overwintering diapause. Under natural winter conditions, pea aphids require a 100-day diapause period before hatching in spring [10]; we shortened the diapause period to 59 days to expedite mutant generation. At the end of the diapause period, eggs were transferred to 16°C waiting for hatching, which usually takes 0–10 days. Phenotypes were inspected in the hatched larvae (G0, founder generation; a.k.a. crispant).

The G0 larvae were grown to adulthood, and the fundatrices produced their offspring (G1) via viviparous parthenogenesis. Each G1 individual produced genetically identical clonal offspring, leading to the establishment of a stable line. Genotyping was performed using the RGEN-RFLP method, followed by Sanger sequencing or Illumina Amplicon-seq [28]. Mutant lines, confirmed by genotyping, were maintained as stable lines for further analyses.

Induction of sexual generation and egg production

First-instar nymphs produced by viviparous aphids were reared on vetch seedlings at 15°C under short-day conditions (8 h light:16 h dark cycle) to induce sexual reproduction in asexual mode. The viviparous nymphs (F1) were grown to adulthood (approximately 2 weeks), transitioning into sexual-reproducing asexual females called sexuparae. Under short-day conditions, sexuparae produced males and oviparous females. To obtain fertilized eggs at 0–24 h AEL, males and oviparous females from different strains were placed together for 24 h in a plastic case containing vetch seedlings at a male-to-female ratio of 1:3 or 1:4. Eggs deposited on the seedlings were collected and injected as described below.

Preparation of Cas9 RNP

Alt-R CRISPR-Cas9 crRNAs and custom tracrRNAs were purchased from Integrated DNA Technologies Inc. (IDT, IA, USA). The sequences of custom tracrRNAs are as follows. Lac2-4976AB:/AltR1/rUrUrG rGrCrA rCrGrG rArArU rCrCrA rCrCrA rGrArG rUrUrU rUrArG rArGrC rUrArU rGrCrU/AltR2/ (target sequence: TTGGCACGGAATCCACCAGA), and Lac2-4976BN:/AltR1/rGrCrU rGrUrG rArArU rCrGrU rArUrG rUrUrG rCrCrG rUrUrU rUrArG rArGrC rUrArU rGrCrU/AltR2/ (target sequence: GCTGTGAATCGTATGTTGCC). The tracrRNAs and crRNAs were annealed according to the manufacturer’s instructions and stored at a concentration of 10 µM in a freezer until use. The annealed tracrRNA/crRNA (final concentration: 30 nM) was mixed with the Cas9 nuclease (Alt-R S.p. Cas9 Nuclease V3, IDT; cat no. 1081058) at a final concentration of approximately 30 nM and incubated at 25°C for 10 min, resulting in the formation of Cas9:gRNA complex at a 1:1 molar ratio.

DNA templates for the in vitro transcription (IVT) sgRNA were generated using a PCR-based method [29]. PCR reactions were set up with a 100 µM forward primer (CRISPRF-[target]), 100 µM reverse primer (sgRNAR), and PrimeSTAR HS premix (TaKaRa, Japan; cat no. R040A) in a 50 µl reaction volume. The cycling conditions were as follows: initial denaturation at 95°C for 10 s, followed by 35 cycles of 95°C for 10 s, 60°C for 5 s, and 72°C for 30 s. The oligos used were: sgRNAR: 5′-AAAAGCACCGACTCGGTGCCACTTTTTCAAGTTGATAACGGACTAGCCTTATTTTAACTTGCTATTTCTAGCTCTAAAAC-3′; CRISPRF_Lac2–92: 5′-GAAATTAATACGACTCACTATAGGG CTGTGAATCGTATGTTGCC GTTTTAGAGCTAGAAATAGC-3′; and CRISPRF_Lac2–155: 5′-GAAATTAATACGACTCACTATAGGG ACGTCCTGAACCACATGCA GTTTTAGAGCTAGAAATAGC-3′. PCR products were purified using a QIAquick PCR Purification Kit (Qiagen, Hilden, Germany). The purified products were used as DNA templates for sgRNA synthesis using the MEGAshortscript T7 Transcription Kit (Thermo; cat# AM1354) according to the manufacturer’s instructions. The sgRNAs were purified using the RNeasy Mini Kit (Qiagen) with a modified protocol that increased the ethanol volume from 250 µL to 700 µL to enable binding of smaller RNA fragments to the column. RNP complexes were prepared as described above using the synthesized sgRNAs.

Microinjection into eggs

The injection needles were prepared from glass capillaries with filaments (1 × 90 mm, Narishige GD-1) using a needle puller (Narishige PC-10). The glass capillaries were treated with Sigmacote (SIGMA SL2) and dried completely in a fume hood. Next, the needles were pulled with a needle puller using the following settings: step = 1, heater level = 60.0, number of weights = 25 g × 2 + 100 g × 2). Before injection, the needle tips were gently scratched by touching the edge of the glass slide, which allowed the needle to open.

The eggs were attached to a microscope glass slide (Matsunami) using double-sided adhesive tape (NICHIBAN/NW-R15), with the posterior end oriented toward the edge of the glass slide. The posterior pole was distinguished by the green color of the bacterial symbiont cluster. Microinjections were performed using a stereoscopic microscope (S8 APO; Leica, Wetzlar, Germany) equipped with a micromanipulator (MMO-220A and MN-4; Narishige, Tokyo, Japan) coupled to a FemtoJet express system (Eppendorf) with the following settings: injection pressure (Pi) = 900 hPa and compensation pressure (Pc) = 700 hPa. The RNP complexes, prepared as described in the Preparation of Cas9 RNP section, were injected into a region anteriorly adjacent to the symbiont cluster at the posterior end of the egg, where germ plasm is localized and PGCs form. Although the germ plasm was not directly visible, the symbiont cluster served as a marker because of its green coloration. The volume of RNP injected into the aphid eggs was estimated to be 0.01 – 0.05 μL.

Diapause and hatching

After RNP injection, eggs were incubated on the slide glass in a moist chamber at 25°C for 1 h (This step is optional). The slides were then transferred to a moist chamber and incubated at 16°C for 2 weeks, followed by 45 days at 4°C to simulate diapause conditions.

The moist chamber was prepared as follows: a Ziploc container was used as the chamber base. A Kim Towel was moistened with tap water, wrung out to remove excess water, and placed at the bottom of the container. The glass slides containing injected eggs were arranged in a slide holder and placed on top of the damp Kim Towel. The container was then sealed with its lid to maintain a high-humidity environment throughout incubation. Under these conditions, the relative humidity was maintained to be nearly 100%. The chamber was regularly monitored to ensure the high humidity and to prevent desiccation and fungal infection. After diapause, the glass slide was transferred to a growth chamber at 16°C to mimic spring conditions and induce hatching.

Formalin fumigation was used to minimize fungal infections during extended diapause. Before RNP injection, eggs aligned on glass slides were placed in a Ziploc case containing Kimwipe paper moistened with formalin in an open 50-mL Falcon tube. The eggs were left in this setup for 5–10 min to ensure sterilization.

In vitro Cas9 digestion

The newly designed tracrRNAs and crRNAs were evaluated through in vitro Cas9 digestion before their application in insects. The protocol provided by NEB (https://www.neb.com/ja-jp/protocols/2014/05/01/in-vitro-digestion-of-dna-with-cas9-nuclease-s-pyogenes-m0386) was followed, with minor modifications. Briefly, the reaction was assembled at room temperature as follows: 20 µl of nuclease-free water, 3 µl of NEBuffer r3.1, 3 µl of 300 nM sgRNA (final concentration: 30 nM), and 1 µl of 1 µM Cas9 Nuclease (final concentration: ~ 30 nM) (IDT), totaling 27 µl. The reaction components were pre-incubated for 10 min at 25°C. Next, 3 µl of 30 nM substrate DNA (final concentration: 3 nM) was added. PCR products amplified from the pea aphid genome using primers ApLac2_F1 (5′-AGCTTGTCAAGTGTGCACAC-3′) and ApLac2_R306 (5′-CTGTTGGTGTCGAATTGGTATCTG-3′) served as substrate DNA. The reaction mixture was then incubated at 37°C for 15 min. To stop the reactions, 1 µl of Proteinase K was added to each sample, followed by a 10-min incubation at room temperature. Digested DNA fragments were purified using the MinElute Reaction Clean-up Kit (QIAGEN) and eluted in 10 µl of Buffer EB. Purified DNA fragments were subsequently analyzed using agarose gel electrophoresis.

Genotyping through Amplicon-seq

Genomic DNA was extracted from the eggs or whole bodies of the aphids using either the DNeasy Blood and Tissue Kit (Qiagen) or a simplified method. For the simplified method, samples were homogenized using a Nippi-sterilized Biomasher II (Funakoshi Co., Ltd., Tokyo, Japan) in Lysis Buffer A (10 mM Tris-HCl, pH 8.0; 0.1 M EDTA; 25 mM NaCl). The homogenate was treated with 400 µg/mL Proteinase K at 37°C for 1 h, followed by heat inactivation at 98°C for 2 min. Amplicon sequencing libraries were prepared according to the Illumina “16S Metagenomic Sequencing Library Preparation” (https://support.illumina.com/documents/documentation/chemistry_documentation/16s/16s-metagenomic-library-prep-guide-15044223-b.pdf). In the first round of PCR, the target region containing the gRNA-targeting site was amplified from individual genomic DNA using KAPA HiFi (Roche Diagnostics, Basel, Switzerland) with a primer pair containing barcode and overhang adaptor sequences. For the Lac2 locus, we used the following primers: ApLac2_CT561F_IL: 5′-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGTACTTTGTGGAGCCACTGTCAG-3′ and ApLac2_CT810R_IL: 5′-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGAACGTGTTTCCCTCGTGAATC-3′. The PCR products were purified using AMPure XP (Beckman Coulter, Pasadena, CA, USA). A second round of PCR was performed using indexed primer sets to construct sequencing libraries using a Nextera XT Index Kit (Illumina, San Diego, CA, USA). The final libraries were purified using Ampure XP beads and sequenced on an Illumina MiSeq system in 150 × 150 paired-end format. Library construction and sequencing were conducted at the National Institute of Basic Biology (NIBB).

Amplicon-seq data analysis

MiSeq reads were preprocessed using Cutadapt (ver. 1.13) to trim the adapter sequences and low-quality regions using the following parameters: -q 20 -O 8-minimum length, 25. The cleaned reads were mapped to the A. pisum reference genome (version Acyr_2.0; GCA_000142985.2) [8] using bowtie2 [30] with the end-to-end option. Alignments were converted to BAM format and visualized using IGV [31], focusing on the target sites. Indel calling was performed using mpileup in SAMtools [32] with the following parameters: -x -Q 13 -A-no-BAQ -d 1000000. The output was parsed with a custom Ruby script to generate summary statistics and was further analyzed using a custom R script for data visualization. The Shell, Ruby, and R scripts used in this analysis are available in our GitHub repository (https://github.com/shigenobulab/24shuji-lac2crispr-paper).

Genotyping through RGEN-RFLP

RGEN-RFLP was performed as described previously [33], with some modifications. Briefly, 1 µM RNA duplex and 1 µM Cas9 nuclease (IDT) were mixed in 1 × PBS and incubated at room temperature for 5 min to create the RNP complex. Next, 1.5 µl of the RNP complex and 4 ng of PCR-amplified products from the pea aphid genome (primer pairs: ApLac2_F1 and ApLac2_R306) were incubated at 37°C for 60 min in a 15 µl reaction containing NEB buffer 3. After the cleavage reaction, 0.5 µl of RNase A (100 mg/ml, #1007885; QIAGEN) was added, and the reaction mixture was incubated at 37°C for 30 min to degrade the RNA. Reactions were then stopped by heating at 65°C for 10 min. The products were purified using the MinElute Reaction Clean-up Kit (QIAGEN) and eluted in 10 µl of Buffer EB. The purified DNA fragments were analyzed using a TapeStation (Agilent).

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)

RNA was purified using NucleoSpin RNA XS (Takara Bio Inc., Shiga, JP). Laccase2 expression was evaluated through RT-qPCR using the One Step TB Green PrimeScript PLUS RT-PCR kit (Perfect Real Time) (Takara Bio) and the Roche LightCycler 96 real-time PCR instrument (Roche Applied Science, Penzberg, DE). RT-qPCR was initiated with a single round of reverse transcription at 42 °C for 5 min, followed by denaturation at 95 °C for 10 s. Subsequently, 45 cycles of PCR reactions were performed as follows: denaturation at 95 °C for 5 s, followed by annealing and extension at 60 °C for 20 s. Finally, a 3-step melting curve analysis was applied, starting with 95 °C for 10 s, followed by 65 °C for 60 s and 97 °C for 1 s. We used the ApLac2_F659 (5′-GCGAAGGAGACAAGGTCGTCA-3′) and ApLac2_R861 (5′-GGAATGGGCGTGCCAGAAATG-3′) primers to amplify Lac2, whereas the RpL7_F (5′-GCGCGCCGAGGCTTAT-3′) and RpL7_R (5′-CCGGATTTCTTTGCATTTCTTG-3′) primers [34] were used to amplify the Ribosomal protein L7 gene, serving as a reference for normalization.

DIPA-CRISPR – Adult injection in oviparous aphids

Adult injections were administered to evaluate the baseline and optimized DIPA-CRISPR conditions in oviparous aphids. Initially, experiments were conducted to establish the effectiveness of Cas9 RNPs under standard conditions. Adult female aphids (KNb strain) were randomly selected and anesthetized on ice. The Cas9 RNP solution was injected into the hemocoel of each aphid using a FemtoJet 4i (Eppendorf) and a finely pulled glass capillary needle until the cauda visibly swelled. The injection solution, targeting the Lac2 locus (5′-GCTGTGAATCGTATGTTGCC-3′) (Lac2-4976BN), was prepared by mixing 1.8 µL of 10 µM crRNA/tracrRNA duplex and 0.3 µL of 10 µg/µL Cas9 protein, resulting in final concentrations of 8.57 µM and 1.43 µg/µL, respectively. After injection, the females recovered at 16°C and were placed in plastic cases (cat# 02270c; Sanplatec) with two seedlings each. Males (ApL strain) were added at a 2:1 female-to-male ratio to ensure mating.

We conducted two sets of experiments to systematically evaluate the effectiveness of the Cas9 RNP injections in oviparous aphids. In the first set, we injected nine females and collected their eggs over one week following the injection, with the day after the injection designated as day 1. No eggs were collected on day 1. We collected eight eggs for genomic DNA extraction from days 2–7. In the second set of experiments, we injected six females and collected eggs at 2-day intervals to monitor temporal changes in indel rates. Specifically, six, seven, and twelve eggs were collected on Days 2–3, 4–5, and 6–7, respectively. The eggs were pooled separately for genomic DNA extraction. Pooled genomic DNA samples from each interval were subjected to amplicon sequencing to assess indel rates using the same amplicon-seq protocol previously described for egg microinjection analysis.

To optimize DIPA-CRISPR in oviparous aphids, we investigated three variables: mating time relative to the injection, age of the adult female at the time of injection, and the injection site. The mating timing was varied by allowing some females to mate before injection and others to mate immediately after injection. The adult females were divided into age groups from days 0 (day of adult emergence) to 8. Each group underwent the same injection procedure described in this section. We tested injection sites at the base of the third thoracic leg and on the dorsal or ventral sides of the abdomen, excluding the midline. Eggs were collected at 2-day intervals for eight days post-injection, individually processed for genomic DNA extraction, and subjected to amplicon sequencing to determine indel rates and assess genome editing efficiency. More than ten females were injected under each condition.

As a control, we injected Cas9 RNPs targeting the BCR3 locus (5′-GCAACCACTGTAACATACGG-3′) using the same procedure as for the Lac2-targeted experiments. BCR3 is specifically expressed in bacteriocytes, specialized cells involved in endosymbiosis with Buchnera [35]. The BCR3-targeting sgRNA exhibits high cutting efficiency, comparable to that of the Lac2-targeting sgRNA, and successfully generated knockout lines via egg microinjection. Results from these control experiments were compared with those from the Lac2-targeted experiments to evaluate the specificity and effectiveness of DIPA-CRISPR in oviparous aphids. The same PCR primers were used for amplicon sequencing as those in the Lac2-targeted experiments to quantify potential off-target cleavage and background noise at the Lac2 locus under identical conditions. This control experiment was conducted using four biological replicates.

Immunohistochemistry

Ovaries were dissected from oviparous females 24 h post-Cas9 RNP injection and fixed in 4% paraformaldehyde in 1 × PBS for 30 min. The ovaries were washed twice with PAT3 (0.5% Triton X-100/0.5% BSA in 1 × PBS) and blocked with 3% goat serum in PAT3 (3% NGS/PAT3) for 3 h. Samples were then incubated overnight at 4°C with a His-tagged mouse monoclonal antibody (cat#70796; Novagen) diluted to 1:1000 in 3% NGS/PAT3. After three washes with PAT3, the ovaries were incubated overnight at 4°C with Alexa Fluor 647 goat anti-mouse IgG (cat#A21236; Thermo Fisher Scientific) diluted to 1:1000 in 3% NGS/PAT3. Next, the ovaries were washed three times with PAT3, counterstained with DAPI (1:1000 in 1 × PBS), and washed three more times with PAT3. Finally, the samples were mounted in Vectashield Antifade Mounting Medium (cat#H-1000; Vector Laboratories) and analyzed using a Zeiss LSM 980 with an Airyscan 2 laser scanning confocal microscope. 3D image data were reconstructed from Z-stack images using ZEN software (Carl Zeiss Microscopy, Germany).

Scanning electron microscopy (SEM)

SEM images of the eggshell surfaces were acquired using a Keyence VHX-D510 digital microscope. Egg samples collected within one month of laying were observed without pretreatment or coating. The eggs were then placed on a microscope stage and imaged under low-vacuum conditions to reduce the risk of sample charging. The microscope settings, including magnification and lighting, were optimized to capture detailed surface structures while maintaining sample integrity.

Atomic force microscopy (AFM)

Young’s modulus measurements were conducted using an AFM (CellHesion 200; Bruker, JPK Instruments AG, Germany) integrated with an Axio Zoom V.16 system (Zeiss). The Rtespa-150 cantilever (Bruker) (tip radius = 8 nm, length = 125 μm, and width = 35 μm) was used for the measurements. The aphid eggs were immobilized on glass slides using a manually applied layer of epoxy adhesive (Cemedine High Super 30; CEMEDINE Co., Ltd., Tokyo, Japan). Force curves of each eggshell were recorded using AFM, and the Young’s modulus was calculated using the JPK data processing software (Bruker). The measurement parameters were as follows: setpoint = 30 nN, extension speed = 5 μm/s, sample rate = 1000 Hz, and measured spring constant range = 2.74–2.87 N/m.

Statistical analysis

All statistical analyses were performed using R version 4.1.1. Data visualization was conducted with the ggplot2 package (ver. 3.5.1) in R. To assess differences in eggshell hardness, one-tailed Welch’s t-tests were applied to the dataset of Young’s modulus values.

Phylogenetic analysis

The protein sequences were retrieved from the OrthoDB database (https://www.orthodb.org/) [36]. Orthologous Laccase genes were searched at both the Insecta and Arthropoda taxonomic levels using the text query “laccase.” The search yielded 27 full-length amino acid sequences at the Insecta level and 55 sequences at the Arthropoda level from 14 species: Acyrthosiphon pisum, Myzus persicae, Aphis gossypii, Rhopalosiphum maidis, Bemisia tabaci, Zootermopsis nevadensis, Drosophila melanogaster, Anopheles gambiae, Culex pipiens, Tribolium castaneum, Apis mellifera, Bombyx mori, Papilio polytes, and Manduca sexta. At the Arthropoda level, in addition to Laccase genes, multicopper oxidase-related proteins (MCORPs) were also retrieved. To identify MCORPs, we used the amino acid sequence previously reported as an MCORP in T. castaneum (NCBI accession: KJ500311) [37] as a query in OrthoDB. This search identified the T. castaneum sequence 7070_0:002c8d, which is identical to 070_0:002c8d, as the corresponding MCORP. From the same 14 species listed above, 16 additional full-length MCORP amino acid sequences were retrieved. The L-ascorbate oxidase gene from Hypsibius exemplaris (a tardigrade species) was used as an outgroup. A complete list of sequences is shown in S1 Table.

Multiple sequence alignments of 28 and 56 sequences from the Insecta and Arthropoda taxonomic levels, respectively, were performed using MUSCLE with default parameters. Phylogenetic trees were constructed using the maximum likelihood (ML) method with the LG model (G + I) for amino acid substitutions and partial deletions (95% site coverage cutoff). The best-fit model was selected using the MEGA model selection tool in MEGA11 [38]. Bootstrap analysis with 1,000 replicates was used to assess branch support for the Insecta dataset, and 500 replicates were used for the Arthropoda dataset. Phylogenetic trees were generated and visualized using MEGA11 software.

Motif analysis and phylogenetic alignment of Lac2 from A. pisum

Domain analysis of the full-length Lac2 amino acid sequence from A. pisum was performed using InterPro (version 101.0) and three Cu-oxidase domains (accession numbers PF00394, PF07731, and PF07732), consistent with previous findings in other insect species [39]. Signal peptide predictions were performed using SignalP (version 6.0). Multiple alignments of Lac2 sequences were performed using MUSCLE with default parameters. The Lac2 sequences were extracted from a larger dataset used for constructing the ML tree (see “Phylogenetic analysis” in Materials and Methods section). A cysteine-rich region was identified based on this alignment and previous findings [40]. The multicopper oxidase signature regions were further extracted using PROSITE (accession numbers PS00079 and PS00080).

Identification and phylogenetic position of Lac2 in the pea aphid

Lac2 encodes phenol oxidases and is critical for cuticle pigmentation and sclerotization in various insects, such as Tribolium castaneum [2], western corn rootworm [41], crickets [42,43], and stinkbugs [44]. A candidate A. pisum Lac2 gene, identified via a BLAST search using T. castaneum Lac2 (Gene ID: 641461) as a query, was annotated as laccase-5 (NCBI Gene ID: 100164049). We used the OrthoDB (v12.0) ortholog database to verify its ortholog and correct annotation [36]. The orthogroup “Group 1603365at50557” at Insecta level included 787 genes in 707 species, representing the Lac2 orthogroup. This group contains a single A. pisum gene (GeneID: 100164049), along with well-characterized Lac2 homologs from T. castaneum (GeneID: 641461) [2] and M. sexta (GeneID: 115442328) [40]. Thus, we confirmed that the A. pisum gene model (NCBI Gene ID: 100164049; XP_001950788.1; UniProt: A0A8R2A7M0) is a Lac2 ortholog. The database suggests that other aphid species, such as the corn aphid (R. maidis), the cotton aphid (A. gossypi), the green peach aphid (M. persicae), also contain a single copy of Lac2 in each genome. Lac2 gene is located on autosome 3 and consists of 12 exons in A. pisum (Fig 2A).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. The structure and phylogenetic context of Lac2 in A. pisum.

(A) The Lac2 locus in A. pisum is located on chromosome 3, as determined based on NCBI Gene annotation (LOC100164049). Exons are depicted in dark green for protein-coding regions and in light green for UTRs. A close-up view of exon 4 highlights the target sites for CRISPR/Cas9-mediated gene knockout, with the designed sgRNAs and corresponding PAM sequences indicated. sgRNAs 4976BN and 4976AB were synthesized using the Alt-R CRISPR-Cas9 system (IDT), while Lac2-92 and Lac2-155 were prepared via in vitro transcription. Notably, 4976BN and Lac2-92 target the same genomic position. The position of the Cu-oxidase_3 domain (PF07732) within Lac2 is also marked. (B) Domain architecture of the A. pisum Lac2 protein. The key domains are labeled, and boundaries are highlighted. Below the domain structure, cysteine-rich regions (CR) and multicopper oxidase signatures (MO) from Lac2 orthologs in multiple insect species are shown. Arrowheads below the sequences indicate the cysteine residues conserved across all Lac2 proteins. A signal peptide (SP) is detected at the N-terminus. The sequence accession numbers are in parentheses. (C) Molecular phylogenetic tree of the full-length sequences of laccase2 (MCO2) and MCO1. A tree was constructed using the maximum likelihood (ML) method and the LG model. Branch lengths were proportional to sequence divergence, and the tree was rooted in L-ascorbate oxidase from Hypsibius exemplaris, a species of tardigrade. Bootstrap values (1,000 replicates) greater than 50% are shown at the nodes. The scale bar represents the number of substitutions per site. In total, 28 sequences were analyzed, with in-group sequences obtained from OrthoDB (https://www.orthodb.org/); their accession numbers are listed.

https://doi.org/10.1371/journal.pgen.1011557.g002

Multiple alignments of the amino acid sequences of Lac2 orthologs from 13 insect species demonstrated evolutionary conservation of this gene across insects (Fig 2B). To further characterize the domains and motifs in A. pisum Lac2, we conducted a domain search using InterPro, which identified three well-defined Cu-oxidase domains (Pfam: PF00394, PF07731, and PF07732). SignalP analysis predicted a signal peptide at the N-terminus, indicating that Lac2 was likely to be secreted into the extracellular space, which is consistent with its expected role in the cuticle where phenol oxidase activity is required for sclerotization and melanization. Furthermore, a cysteine-rich region associated with the carbohydrate-binding domain [40] and a highly conserved multicopper oxidase signature sequence [Prosites PS00079 and PS00080] were identified (Fig 2B). These conserved features across Lac2 sequences from various insect species underscore the functional importance of these motifs in the enzymatic activities related to melanization and cuticle formation. Structural conservation also suggests that aphid Lac2 retains molecular and enzymatic functions similar to those of Lac2 orthologs in other insects.

In our initial phylogenetic analysis, we retrieved 27 Laccase-related sequences from OrthoDB at the Insecta taxonomic level. Most insects, including the four aphid species analyzed, possessed two Laccase genes that clustered into two distinct clades corresponding to MCO1 (Lac1) and Lac2. Clustering showed strong support with high sequence conservation within these clades among aphids (bootstrap value = 100) (Fig 2C). Next, to expand the phylogenetic context of our analysis, we retrieved 71 MCO-related sequences at the broader Arthropoda level from OrthoDB, including not only Laccase genes but also MCORP sequences. Representative MCORPs were identified using the previously reported MCORP protein from Tribolium castaneum (NCBI accession: KJ500311 [37]) as a query. The L-ascorbate oxidase gene from the tardigrade Hypsibius exemplaris was used as an outgroup. The phylogenetic tree (S1 Fig) showed that the four aphid species consistently retain three distinct MCO gene groups: MCO1, Lac2, and MCORP. In contrast, other insect lineages such as Culex pipiens exhibit extensive lineage-specific gene expansions, with up to seven MCO genes spanning multiple clades. In aphids, however, MCO-related sequences were confined to their respective clades without evidence of recent gene duplications. This conserved MCO-related gene repertoire in aphids suggests a lack of order-level diversification and highlights the evolutionary stability and likely functional significance of these genes within the aphid lineage.

Expression of Lac2 during egg pigmentation and post-embryonic development in A. pisum

Eggs laid by sexual A. pisum females undergo a distinct color transition, initially appearing cream before gradually turning dull green and eventually darkening to black in several days [11] (Fig 3A). As Lac2 encodes phenol oxidase, an enzyme involved in pigmentation in other insects, we hypothesized that it is critical for pigmentation during pea aphid egg development. To explore this hypothesis, we measured the temporal expression levels of Lac2 transcripts in wild-type aphid embryos using qRT-PCR (Fig 3A).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Expression of Lac2 mRNA.

(A) Temporal expression pattern of the Lac2 transcript in early egg development, as analyzed using quantitative RT-PCR. The expression levels (Y-axis) are presented as values normalized to the house-keeping gene rp49. Data are plotted as means (black circles) and standard deviation (black lines). Individual data points are plotted as orange dots. The accompanying images depict eggs at five developmental stages (<24 h, 30– 40 h, 40– 50 h, 50– 60 h, and >70 h after AEL (after egg laying), illustrating the progressive pigmentation of the eggs. (B) Lac2 transcript expression levels across various morphs and developmental stages, as analyzed using qRT-PCR. On the x-axis, “early” represents 0–72 h after egg laying, corresponding to data in (A); “middle” represents 1-week AEL; and “late” represents eggs at 1.5–2.0 months of AEL following diapause. “Fundatrix L1” indicates the L1 larva that hatches from overwintering eggs. For viviparous females, samples are further categorized as nymphs (L1–L4 stages mixed), molting, adult, and winged morphs. Winged viviparous morphs were induced under crowded conditions. In all cases, whole bodies or whole eggs were homogenized for qRT-PCR quantification. Expression levels are normalized to the housekeeping gene rp49.

https://doi.org/10.1371/journal.pgen.1011557.g003

Our analyses revealed a sharp increase in Lac2 expression coinciding with the onset of egg pigmentation. The results showed a low baseline expression of Lac2 before 20 h after egg laying (AEL), coinciding with the early cream-colored stage of the egg. Lac2 expression began to increase sharply at approximately 30–40 h AEL, corresponding to the onset of green pigmentation in the eggs. It peaked between 40 and 50 h AEL, when the eggs displayed a dull green color. Thereafter, Lac2 expression remained elevated but gradually declined as the eggs darkened to black by 70 h AEL. The timing of Lac2 expression closely mirrored the progression of pigmentation, suggesting its involvement in the enzymatic processes that drive egg coloration. The peak in Lac2 expression coincided with the most pronounced changes in pigmentation, further demonstrating its role in the early developmental stages of egg pigmentation.

We further examined the later stages of embryogenesis and post-embryonic development across various morphs, including viviparous females, oviparous females, fundatrix, and males, using qRT-PCR (Fig 3B). After an initial peak in Lac2 expression during early egg development, expression was almost undetectable in one-week-old eggs. However, we observed a resurgence in Lac2 expression in some eggs during the late diapause period before hatching (Fig 3B). Among the distinct morphs, males in the L3 and L4 nymph stages showed relatively high Lac2 expression, whereas the fundatrix and oviparous females showed low Lac2 expression. In viviparous nymphs, Lac2 expression was elevated in aphids undergoing molting or in the pre- and post-molting stages, which is consistent with its reported role in cuticle formation in other insects.

Optimization of CRISPR/Cas9-mediated knockout protocol in the pea aphid

To elucidate the role of Lac2 in pea aphids, we knocked out the Lac2 locus via the non-homologous end joining (NHEJ) pathway using CRISPR/Cas9-mediated genome editing. We tailored a CRISPR/Cas9-mediated knockout workflow to pea aphids, as outlined in Fig 1 to mediate the challenges associated with applying this method in aphids. While the full procedural details are provided in the Materials and Methods section, a summary of the key challenges and our solutions is provided below.

Complex life cycle: To ensure a constant supply of eggs, we established a workflow using multiple incubators with varying conditions that mimicked summer (16 h light: 8h dark cycle) and winter (8 h light: 16 h dark cycle), allowing the transfer of aphids to induce sexual reproduction.

Inbreeding depression: Our preliminary tests revealed a severe inbreeding depression in pea aphids [26]. Therefore, we collected nine aphid lines from various host plants (alfalfa Medicago sativa, red clover Trifolium pratense, Vicia cracca, and Melilotus spp.) and identified pairs that exhibited high hatching rates and fertility. Based on these results, we selected two outbred combinations, ApL (female) × 0903A (male) and ApL (male) × KNb, for use in subsequent experiments [26].

Bacterial endosymbiosis: Aphids harbor Buchnera aphidicola, an obligatory symbiont that is essential for their survival and is localized at the posterior pole of the egg. To avoid damaging these symbionts, we targeted our CRISPR/Cas9 injections to a region adjacent to the cluster of Buchnera at the posterior end of the egg, where the germ plasm is located and primordial germ cells (PGCs) are formed.

High nuclease activity: The internal environment of aphids is rich in RNases [45], which may inhibit efficient CRISPR/Cas9 genome editing. We used the Alt-R CRISPR system (developed by IDT) to mediate this problem. This commercial tracrRNA:crRNA system includes 2′-O-methyl and phosphorothioate modifications on the ends of tracrRNA, as well as proprietary chemical modifications to crRNA. This increases resistance to nucleases and reduces the innate immune response (https://sg.idtdna.com/pages/products/crispr-genome-editing/alt-r-crispr-cas9-system). A comparison of chemically synthesized gRNAs (Alt-R) and gRNAs synthesized via IVT revealed a gRNAs (Alt-R)-induced significant improvement in indel efficiency (Fig 4A; see IVT gRNA-mediated CRISPR/Cas9 genome editing in S1 Text).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Targeted disruption of Lac2 in A. pisum (crispant analyses).

(A) The indel percentages of eggs injected with Cas9 and gRNA, as calculated based on amplicon sequencing of individual eggs. See Fig 2A for the target position of each sgRNA. Notably, although 4976BN and Lac2-92 target the same genomic site, 4976BN induced a significant high indel frequency. (B) Quantification of amplicon-seq reads containing indels (insertions and/or deletions) at the target site. This plot summarizes the data of 39 eggs injected with Lac2-4976BN/Cas9. Data are plotted as means (circles) and standard deviations (lines). (C) Lac2 gRNA (Lac2-4976BN)/Cas9-injected eggs (crispants). Egg photos were captured after diapause (71–77 days after injection). Illumina reads containing indels are indicated at the bottom of each egg image and correspond to the peak values in Fig 2D. (D) Indel distribution profiles from amplicon-seq of the three Lac2 crispant eggs shown in (C). The percentage of Illumina reads containing indels is plotted relative to the target Cas9 cut site (position 0). Green, blue, and red lines correspond to the first, second, and third egg from the left in panel (C), respectively.

https://doi.org/10.1371/journal.pgen.1011557.g004

Diapause: Fertilized eggs must undergo a 3-month diapause in a winter-like environment before hatching [10], which presents a significant challenge for maintaining healthy eggs in the laboratory. Therefore, we developed a moist chamber system to maintain appropriate humidity levels and prevent desiccation. In addition, formalin was administered to prevent fungal infections.

With this optimized protocol, we targeted to knock out Lac2 in the pea aphid to elucidate its function.

CRISPR-induced mutations at the Laccase2 locus

Germline transmission of mutant alleles.

We investigated whether somatic mutagenesis detected in adults reared from injected embryos (G0 animals) resulted in the transmission of stable mutant alleles to their offspring (G1 animals) through the germline. We focused on Lac2-4976BN crRNAs because this crRNA showed a higher somatic mutation rate in the crispant analysis (Fig 4A), with an expectation of a better chance of obtaining germline mutants. Embryos (n = 532) were injected with a mixture of Cas9, crRNA, and tracrRNA; 264 eggs survived. Seventy-one eggs hatched after 2-month embryonic diapause in a cold, moist chamber. Of these, 15 fundatrices grew into adulthood to give birth to the G1 offspring. Each G1 female produced a clonal G2 offspring. We genotyped 45 G1 families by analyzing G2 individuals (genetically identical to G1) using the RGEN-RFLP method, followed by amplicon sequencing. We obtained six stable mutant lines that were classified into unique four lines: 7 bp deletion (mut-1D in Fig 5A), 7 bp deletion (mut-4F), 3 bp deletion (mut-2B), and 5 bp insertion (mut-3E). These mutations were detected at the targeted cleavage site in each line (Fig 5A). All the aphids in these mutant lines carried heterozygous mutations and displayed no apparent lethal defects. The heterozygous lines were maintained in the laboratory as viviparous females that reproduced clonally.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. Germline transmission of Lac2 mutant alleles, generation of homozygous G1 mutants, and egg phenotypes.

(A) Genotypes of Lac2 G1 offspring analyzed by amplicon sequencing Targeted guide sequences are highlighted in blue. PAM sequences are underlined. Deletions are indicated by dashes (-) and insertions are indicated by lowercase letters. (B) Generation of a homozygous mutation by self-crossing with a stable line carrying a heterozygous Lac2 deletion corresponding to mut-4 in (A). The gel image shows the RGEN-RFLP results. Red and blue arrowheads indicate the undigested and digested PCR fragments, respectively. In this experiment, three eggs were found to possess biallelic mutants and are marked with *. Photographs of egg color phenotypes are shown at the bottom. (C) Example of a homozygous Lac2 mutant egg (left) compared with an uninjected control egg (right). A developed red eye, visible through the eggshell, is indicated by an arrowhead. (D) Differential interference contrast (DIC) images of cross-sections of unpigmented homozygous Lac2 mutant eggs (right) compared to black eggs (left) prepared from early embryos (12–14 days AEL). Both types of eggs were generated by inbreeding the Lac2 mut-3E line. (E) Fluorescence in situ hybridization images were overlaid with DIC images of cross-sectioned late embryos (100 d AEL). Both unpigmented and pigmented eggs were generated by inbreeding with the Lac2 mut-4F line. DAPI signals indicating DNA are colored in cyan, and Buchnera symbiont localization was visualized using a Cy5-Apis2a probe colored in magenta. (F) Box plot of Young’s modulus (MPa), as measured using atomic force microscopy (AFM). Each point represents an individual egg with the genotype color-coded: homozygous Lac2 mutants (green), heterozygous Lac2 mutants (red), and wild type (blue). Two Lac2 mutant lines (E and F), distinguished by shape, were used in this experiment: squares and circles represent lines E and F, respectibvely. Statistical significance was assessed using one-tailed Welch’s t-test, with the p-value (p = 0.0028) indicated between the two boxplots.

https://doi.org/10.1371/journal.pgen.1011557.g005

Phenotypes of heterozygous Lac2 mutants

All four heterozygous Lac2 mutants were viable and stably maintained in the laboratory for parthenogenetic viviparous reproduction. Although no noticeable differences were observed between the wild-type and heterozygous mutants, distinct morphological changes were observed in the adult male morphs of the heterozygous Lac2 mutants. The dorsal surface of the adult male thorax appeared dark in wild-type pea aphids (Fig 6A, left). In contrast, heterozygous Lac2 mutant males exhibited lighter pigmentation in the dorsal thorax (Fig 6A, right). Additionally, the legs of heterozygous mutant males showed reduced pigmentation compared to those of wild-type males (Fig 6A). Phenotypic differences were also observed between the wings. Wild-type males had straight and flat wings, whereas the wings of heterozygous Lac2 mutant males were noticeably wrinkled (Fig 6A). The high expression of the Lac2 transcript in this morph (Fig 3B) may underlie the phenotypic changes observed in male heterozygous mutants.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. Phenotypes of heterozygous Lac2 mutants.

(A) Adult males of a wild type (WT; KNb strain) and heterozygous Lac2 mutant (mut-4F; labeled as “het”). (B) Winged viviparous females of wild-type (KNb) and heterozygous Lac2 mutants from the mut-4F and mut-1D lines were induced from wingless morph aphids under crowded conditions. Black arrowheads indicate the thorax, where the tanning level of the cuticle surface is remarkably different between the wild-type and mutants. Grey arrows indicate representative legs showing remarkable different pigmentation levels between the wild-type and the heterozygous mutant.

https://doi.org/10.1371/journal.pgen.1011557.g006

The established Lac2 mutant lines were maintained in the laboratory as wingless viviparous females, and no noticeable morphological differences were observed compared with wild-type individuals. However, the color phenotype became evident when we induced a winged viviparous female morph from heterozygous mutant lines under crowded conditions. Similar to the males, heterozygous Lac2 mutant females exhibited lighter pigmentation in the dorsal thorax than the wild-type (Fig 6B). This reduction in pigmentation was consistent across the independent mutant lines tested (mut-4F, mut-1D, and mut-3E), suggesting that Lac2 contributes to cuticle tanning in a dose-dependent manner. Unlike in males, the legs of heterozygous females did not show a noticeable reduction in pigmentation. Additionally, winged heterozygous mutants induced from the mut-4F or mut-1D lines exhibited malformed wings with a reduced size and wrinkled appearance (Figs 6B and S2A), which was similar to the male phenotype. These findings suggest that Lac2 is involved in exoskeletal pigmentation and wing formation in winged viviparous female morphs.

Egg phenotypes of homozygous Lac2 mutants

Lac2 mutant eggs failed to hatch while developing up to eye formation stage.

After the successful generation of homozygous Lac2 mutants from the stable line Lac2 mut-4F through inbreeding, we generated homozygous mutants from the mut-1D, mut-2B, mut-3E, and mut-4F lines to assess reproducibility across mutant lines (Table 1). Consistent with the initial mut-4 experiments, we obtained two distinct egg color phenotypes: black and unpigmented (Table 1). Genotyping confirmed that all unpigmented eggs were homozygous mutants. After genotyping, the remaining eggs were subjected to diapause under winter conditions to monitor hatching. None of the unpigmented eggs hatched across all four mutant lines (80 eggs in total), whereas the black eggs showed a hatching rate of 7.6–47.3% (582 eggs) (Table 1). These results demonstrate that unpigmented eggs are non-viable, indicating that Lac2 is essential for embryogenesis in overwintering eggs.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Egg phenotypes of Lac2 mutants.

https://doi.org/10.1371/journal.pgen.1011557.t001

Well-developed red eyes were observed through transparent eggshells in 30% of the unpigmented eggs (42 of 132) (Fig 5C and Table 1). Because eye development occurs during the late stages of embryogenesis in oviparous aphids [10], this suggests that Lac2 is required for developmental processes in the late embryogenesis and/or hatching stages.

Egg surface structure and susceptibility to fungal infection.

To investigate the role of Lac2 in eggshell surface structure, we conducted scanning electron microscopy (SEM) on ten wild-type eggs and five Lac2-deficient eggs, all collected within one month after laying (Fig 7A). SEM images of wild-type eggs revealed a smooth and continuous surface with compact and well-organized structural elements, which likely contributed to the protective and waterproofing functions of the eggshell (Fig 7B and 7C). In contrast, 60% of Lac2-deficient eggs displayed noticeable dents (Fig 7D), suggesting reduced hardness compared with that of wild-type eggs. Additionally, all Lac2-deficient eggs exhibited a distinct phenotype characterized by vein-like raised structures covering the entire surface (Fig 7D and 7E). These raised structures were associated with fungal growth, as conidial chains (bead-like fungal spores) were observed in some Lac2-deficient eggs (Fig 7E and 7F). Fungi were not detected in any wild-type eggs. These findings suggest that the absence of Lac2 compromises the structural integrity and barrier function of the eggshell, making it more susceptible to microbial colonization and fungal infection.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 7. SEM analysis of the eggshell surface in wild-type and Lac2-deficient eggs.

(A) Comparison of fungal growth rates between wild-type and Lac2-deficient (Δ7/Δ7) eggs. (B–E) SEM images of the eggshell surface for wild-type (B, C) and Lac2-deficient eggs (D, E). Lateral views are shown in (B) and (D), whereas posterior views are shown in (C) and (E). The dented surface areas of Lac2 mutant eggs are highlighted with a single white outline bracket (D). (F) Magnified view of the area indicated by the dashed line in (D), showing a chain of conidia on the fungal structure.

https://doi.org/10.1371/journal.pgen.1011557.g007

Development of DIPA-CRISPR for oviparous females in the pea aphid

Recent studies have induced genome editing via NHEJ in fertilized eggs by directly injecting RNPs into the body cavity of several insects, such as cockroaches, red flour beetles, mosquitoes, and whiteflies [25,46,47]. This method, termed DIPA-CRISPR, offers a promising and straightforward alternative to egg injection, and significantly simplifies the injection process. Thus, we further explored the feasibility of DIPA-CRISPR in adult oviparous females of pea aphids.

We injected Cas9 RNPs targeting Lac2 (sgRNA: Lac2-4976BN) into the body cavities of oviparous female aphids and examined indel mutation rates in the eggs they laid. Nine adult females were randomly selected, injected with Cas9 RNPs, and mated with males. Genomic DNA extracted from eight eggs was analyzed using the Illumina MiSeq platform. Amplicon sequencing revealed indels in 14.6% of the sequences (Fig 8A). As a control, Cas9 RNPs targeting the BCR3 locus were injected into 14 oviparous females. Among the 64 fertilized eggs obtained, four were randomly selected for amplicon sequencing. BCR3 is an aphid-specific gene unrelated to Lac2, and the corresponding sgRNA has been well characterized. No apparent phenotypic changes were observed in the laid eggs, and amplicon sequencing showed an indel rate close to 0% (0.018%), a level consistent with background PCR or sequencing errors rather than off-target activity [48]. We further evaluated the correlation between egg-laying timing and the induced indel rate; we injected six oviparous females and collected eggs at 2-day intervals, with day 1 defined as the day after the injection. The indel rates were 41.8%, 25.2%, and 21.4% for days 2–3 (6 eggs), 4–5 (7 eggs), and 6–7 (12 eggs), respectively (Fig 8A). These results indicate that genome-edited fertilized eggs can be obtained at least one week post-injection.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 8. DIPA-CRISPR for oviparous aphid females and its application in Lac2 knockout.

(A) Optimization of DIPA-CRISPR: indel rates in fertilized eggs from two DIPA-CRISPR experimental sets of oviparous aphids. The first set shows the indel rates for eggs collected on days 2–7 post-injection. The second set shows the indel rates for eggs collected at two-day intervals: days 2–3, days 4–5, and days 6–7 post-injection. Genomic DNA from each collection was pooled and analyzed using amplicon sequencing. (B) Comparison of pre- and post-injection RNP: Indel rates in fertilized eggs from pre- and post-injection mated females. (C) Indel rates in fertilized eggs from females injected at different days post-emergence, collected over days 2–7 post injection. (D) Indel rates in fertilized eggs from females injected in the thorax or abdomen, collected at two-day intervals. (E) Established workflow of DIPA-CRISPR for oviparous aphid females. (F) Cas9 RNP uptake: Confocal images showing Cas9 signals at the posterior pole of vitellogenic oocytes, as indicated by a white arrowhead.

https://doi.org/10.1371/journal.pgen.1011557.g008

We investigated the effects of mating timing, optimal timing for Cas9 RNPs injection, and effective injection sites on the body to optimize DIPA-CRISPR in oviparous aphids. We examined the effect of mating on genome editing efficiency by injecting Cas9 RNPs into pre- and post-mating females. Premating females were injected and mated, after which they laid eggs. Amplicon sequencing showed no significant changes in indel rates between pre- and post-mating females, suggesting that mating did not affect the efficiency of mutation induction in fertilized eggs (Fig 8B). We also evaluated the optimal injection timing and target body parts for higher knockout efficiency (S2 Text). Our results indicated that the timing of injection did not significantly affect the knockout efficiency (Fig 8C). However, injections targeting the abdomen resulted in higher indel rates than those targeting the thorax (Fig 8D).

In summary, we successfully established a DIPA-CRISPR method for genome editing in oviparous female pea aphids (Fig 8E). The optimized conditions included injecting Cas9 RNPs into pre-mating females, targeting the abdomen, and administering injections within the first eight days post-adult emergence (Fig 8E).

Mosaic knockout phenotype of DIPA-CRISPR-mediated Lac2 mutation

We obtained two mosaic Lac2 mutant eggs from Cas9 RNP-injected mothers using the newly established DIPA-CRISPR method for oviparous aphid females (Fig 9A). Both eggs exhibited a mosaic pattern of black pigmentation loss. Because Lac2 is the dominant gene in pigmentation, the absence of black pigmentation in these regions indicates a biallelic KO, suggesting that both Lac2 alleles were disrupted in these areas. The indel rates in the two eggs were 94.2% and 60.4%, respectively. These were associated with a marked reduction in black pigmentation on the egg surface (Fig 9A). These results suggest that biallelic KO mutations were induced early during embryogenesis rather than during oogenesis.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 9. Mosaic Lac2 mutant phenotypes obtained through DIPA-CRISPR.

(A) Mosaic analysis of Lac2 knockout fertilized eggs. Indel rate and phenotypic analysis of two mosaic eggs with indel rates of 94.2% and 60.4%, respectively. Significant inhibition of black pigmentation is shown. (B) G0 mosaic nymphs (first-instar stem mothers) with Lac2 knockout. The nymph exhibited inhibited pigmentation in the cuticle, specifically in the area to the left of the white dotted line and in the compound eyes indicated by a red arrowhead. Additionally, the nymphs exhibited abnormal locomotion owing to incomplete embryonic molting. The black arrow indicates an attached old cuticle due to incomplete embryo molting.

https://doi.org/10.1371/journal.pgen.1011557.g009

One egg laid by a mother treated with DIPA-CRISPR targeting the Lac2 locus successfully hatched after diapause, yielding a G0 mosaic mutant fundatrix nymph. However, this G0 nymph died shortly after hatching and did not undergo the first molt. The G0 nymphs exhibited mosaic inhibition of pigmentation in both the cuticle and compound eyes, where red pigmentation was disrupted (Fig 9B). The G0 mutant nymph also displayed a thin membrane that partially adhered to the body surface (Fig 9B) and exhibited abnormal locomotion (S1 Movie). Given the known role of Lac2 in cuticle formation and hardening in other insects, it is likely that the knockout led to defective cuticle formation, resulting in incomplete embryonic molting in G0 nymph. The relatively high expression of Lac2 in late diapause eggs of (1.5–2 months AEL) (Fig 3A) may account for the defects observed in embryonic molting in the Lac2 knockout mutants generated via DIPA-CRISPR.

Cas9 RNP incorporation into oocytes in DIPA-CRISPR

To examine how injected Cas9 RNPs are incorporated into developing oocytes, we performed immunohistochemistry using an anti-His tag monoclonal antibody to detect Cas9 localization. Because Cas9 was injected as a part of the Cas9 RNP complex, the detected Cas9 signals indicate Cas9 RNP localization. Ovarioles were collected 24 h after the Cas9 RNP injection into oviparous females. Our immunohistochemical analysis revealed a distinct localization of Cas9 RNPs within the developing aphid oocytes. Cas9 signals were predominantly detected in large spherical follicular cells at the posterior poles of vitellogenic oocytes. These posterior signals were consistent and clear, suggesting a robust uptake mechanism at this site (Fig 8F).

Z-stack imaging revealed that the signal at the posterior pole comprised multiple signals (S3A Fig), suggesting that they were clustered. This clustering of Cas9 RNPs implies that they may bind to specific cellular structures or protein complexes, providing insights into the mechanisms underlying RNP uptake and intracellular transport pathways. Additionally, Cas9 signals were detected along the lateral regions of the oocytes in follicular cells (S3B Fig). Although these lateral signals were less distinct and varied in intensity, they indicated that Cas9 RNPs may be taken up through pathways other than those at the posterior pole. The presence of these signals suggested the possibility of multiple entry points or mechanisms for Cas9 RNP incorporation into oocytes.

Functions of Lac2 in the pea aphid

Our CRISPR/Cas9-mediated disruption of Lac2 revealed its essential roles in multiple developmental stages and morphs of the pea aphid. Homozygous knockouts resulted in pigment-less, soft-shelled eggs that were embryonically lethal (Fig 5C–5F), while G0 and heterozygous mutants exhibited phenotypes such as reduced cuticle pigmentation, malformed wings, and molting defects (Figs 4C,6,9A and 9B). These findings demonstrated the multiple critical roles of Lac2 in different developmental stages and morphs of the pea aphid.

Essential roles of Lac2 in overwintering eggs and environmental adaptation

The Lac2 enzyme plays a crucial role in sclerotization and pigmentation of insect cuticles [2,6,39,44,49–51]. Tanning caused by Lac2 is also essential for other structures such as eggshells, chorions, oothecae, and butterfly pupa [52–54]. In this study, eggs injected with Lac2 gRNA/Cas9 showed reduced pigmentation (Fig 4C), and loss-of-function Lac2 mutants generated by crossing stable heterozygous Lac2 mutants exhibited a complete lack of eggshell pigmentation (Fig 5B–5E), with none of the eggs hatching (Table 1). The expression pattern of Lac2 mRNA in aphid eggs coincided with the onset of eggshell pigmentation (Fig 3A), indicating that Lac2 is required for the pigmentation of pea aphid eggs. In addition, the unpigmented eggshells of homozygous Lac2 mutants were significantly less stiff than pigmented eggs (Fig 5F), highlighting the critical role of Lac2 in the hardening of aphid eggshells. Similar phenotypes have been reported in other insects such as the mosquito (Aedes albopictus), in which Lac2 is essential for eggshell sclerotization and pigmentation [54]. The colorless eggshell phenotype in mosquitoes closely resembles that observed in pea aphids in this study.

The failure of unpigmented aphid eggs to hatch, and their consequent death (Table 1), indicate the essential role of Lac2 in embryonic development in pea aphids. The visibility of red eyes through colorless eggshells suggests that the embryos developed to a late stage, as eye formation occurs during the final stages of aphid embryogenesis [10]. Cryosectioning of late-stage diapause eggs confirmed the presence of developed embryos (Fig 5E). However, the embryos did not hatch. The exact cause of mortality remains unclear. One possibility is that Lac2 knockout embryos lack sufficiently hardened cuticles for successful hatching, owing to defects in sclerotization. Alternatively, the loss of eggshell protective function may lead to accumulated damage during prolonged diapause. The damage can be a physical, fungal, or bacterial pathogenic infection. Indeed, we observed a higher incidence of fungal infection on the surface of Lac2 mutant eggs than in wild-type eggs (Fig 7). These observations indicate that Lac2 plays essential roles in the survival and development of overwintering eggs through five key functions: eggshell pigmentation, eggshell hardening, late-stage embryonic development, hatching, and protection against fungal infections. These functions represent ecological and evolutionary adaptations in the complex life cycle of pea aphids, including overwintering diapause and cyclic parthenogenesis.

Lac2 roles against fungal infection

Pathogenic fungi pose a major threat to overwintering aphid survival. In the bird cherry-oat aphid (Rhopalosiphum padi), approximately 60% of overwintering eggs in the wild die, primarily owing to fungal infections [55]. Similarly, fungal infections rapidly kill eggs without ant grooming in hinoki cypress-feeding aphids (Stomaphis hirukawai) [56]. These examples show the importance of strategies that protect aphid eggs from fungi, whether through mutualistic relationships with ants or intrinsic defenses within the eggshell. Our study demonstrated that Lac2 inhibits fungal development in pea aphid eggshells (Fig 7). Extensive fungal growth was observed on the surface of aphid eggs in the absence of Lac2, suggesting that the enzyme is essential for maintaining eggshell integrity and preventing fungal colonization.

Lac2 catalyzes the oxidation of phenolic compounds, consequently producing quinones that facilitate the crosslinking of chitin and cuticular proteins [4]. This cross-linking is crucial for creating a hardened cohesive cuticle that acts as a barrier against environmental factors such as moisture and microbial invasion [57,58]. This crosslinking was likely disrupted in Lac2-deficient eggs, leading to vein-like fungal growth patterns (Fig 7D–7F). These fungal structures suggest that the weakened and uneven eggshell surfaces provide sites for fungal spores to adhere and proliferate.

Although our study showed that Lac2 is critical for eggshell integrity and fungal resistance, SEM analysis revealed no obvious surface irregularities in Lac2-deficient eggs at the magnifications used. Thus, higher magnification and resolution may be required to detect microstructural changes. Surface depressions or cracks can create niches for fungal spores and promote fungal growth [57,59]. Therefore, future studies using high-resolution imaging are required to further elucidate the microstructural impact of Lac2 deficiency.

CRISPR-Cas9-mediated genome editing in the pea aphid