In planta genetic transformation to produce CRISPRed high-oleic peanut

Compared to its normal-oleic counterpart, high-oleic peanut has better keeping quality and much more health benefits. Breeding high-oleic peanut through conventional means is a tedious process that typically takes several years. Genome editing, however, may shorten the duration. This study aimed to test the effectiveness of the node injection method coupled with CRISPR/Cas9 technology in inducing FAD2B mutations and high-oleic phenotype in peanut. Huayu 23, a popular normal-oleic runner type peanut cultivar having dysfunctional FAD2A and functional FAD2B, was transformed with CRISPR/Cas9 construct targeting FAD2B, resulting in two T1 seeds with over 80% oleic acid and a 442 A insertion in FAD2B. The high-oleic phenotype in T2 seeds was inheritable from the T1 generation. As a genotype-independent, simple and easy method for peanut genetic transformation, node injection has great potential in functional analysis of genes and peanut varietal improvement. This method is of reference value to other seed plant species.


Introduction
Rich in culinary oil and highly digestible protein, the cultivated peanut (Arachis hypogaea L.) occupies an important position in human and animal nutrition, and rural economy (Li et al. 2023).Fatty acid profile is an indicator of its quality.Oleic and linoleic acids together constitute about 80% of total fatty acids in peanut seeds.As compared to linoleic acid, oleic acid is less prone to oxidation.Increase in oleic acid and decrease in linoleic acid in peanut seeds may result in extended shelf life of peanut produce and much more health benefits (Nkuna et al. 2021;Zhao et al. 2022).Hence, high oleate has become one of the main breeding objectives of peanut.
The cultivated peanut originated from polyploidization after hybridization of two diploid wild species (Bertioli et al. 2019).In the tetraploid cultivated peanut, FAD2A and FAD2B on the A and B sub-genome respectively, control the conversion of oleic acid to linoleic acid; expression of the high-oleic phenotype (at least 70% oleic acid content) in peanut cultigen, therefore, requires inactivation of both genes (Nawade et al. 2018).Natural, chemical, and physical peanut mutants with high oleate have been reported and used in hybridization and backcross to develop higholeic peanut cultivars (Wang et al. 2021(Wang et al. , 2022;;Han et al. 2022).In contrast to the lengthy process of conventional breeding, genome editing may be a much faster alternative.Using peanut germs with a cotyledon attached for transformation, Wen et al. (2018) demonstrated that TAL-ENs-mediated targeted mutagenesis of FAD2 in peanut cv Yueyou 7 raised oleic acid content from 43-60%~80%, while decreasing linoleic acid content from 35.5% to lower than 20%.Yuan et al. (2019) induced FAD2B mutations in peanut protoplasts and hairy roots using CRISPR/Cas9 based genome editing.Zhang et al. (2021) bombarded the embryonic calli of peanut cv Luhua 11 with a CRISPR/ Cas9 gene editing vector targeting FAD2, and regenerated plants were obtained, but the fatty acid profiles of the descendants were not reported.Tang et al. ( 2022) produced genetically edited Huayu 23 lines with decreased saturated fatty acid and increased unsaturated fatty acid through knockout of FATB by CRISPR/Cas9 system.To minimize the negative effects on other plant tissues while improving the fatty acid profile of peanut seeds, Neelakandan et al. (2022a) performed CRISPR/Cas9 genome editing on FAD2 cis-regulatory motifs in peanut using the Agrobacterium-mediated calyx tube injection method, and obtained T 1 seeds with elevated oleate content, two of which had 66.27% and 68.43% oleate, respectively.Neelakandan et al. (2022b) reported successful base editing in peanut FAD2 with CRISPR/nCas9 and hairy root transformation.The industry accepted standard for high oleic peanut is > 74% oleic acid (Davis et al. 2021).However, in the above mentioned genome editing studies, only Wen et al. (2018) obtained peanut seeds with an oleic acid content of 74% or more.
Suitable transformation procedures may facilitate the application of genome editing tools.Previously, the node injection method, an easy-to-follow in planta peanut transformation protocol, was developed at our laboratory (Wang et al. 2013).Using this method, SCTF-1, the soybean C 2 H 2 -type zinc finger protein gene (Song et al. 2012), was transferred to chill susceptible peanut.RT-PCR and Southern blot analysis confirmed the transgenic events.SCTF-1 transgenic peanut plants showed good tolerance to chill stress (V.M. Vacu, unpublished data).Likewise, transfer of AhCYP, a Ralstonia solanacearum infection responsive gene from Rihua 1, a Virginia type peanut cultivar with verified high resistance to BW both in field and at laboratory (Ding et al. 2012), to Huayu 40, a susceptible peanut cultivar, following the same protocol, enhanced BW resistance of the recipient (Wu et al. 2019).
The present study aimed to test the effectiveness of the node injection method in inducing FAD2B mutations and high-oleic phenotype in peanut using CRISPR/Cas9 technology.

Plant material for genome editing and its cultivation
Huayu 23, a normal-oleic peanut cultivar of runner market type widely accepted by growers and food processors in China, was used in this study.As expected, Sanger sequencing of its FAD2A and FAD2B and subsequent sequence alignment revealed that this cultivar had a mutated FAD2A (448 G > A) and a wild type FAD2B.Peanut for genome editing was sown under polyethylene film mulching in an isolated region in SPRI Laixi Experimental Station on May 5, 2021.Agronomic practices were followed as routine (Wan 2003).

Transformation of Agrobacterium and node injection transformation of peanut
Genome editing construct was transformed into Agrobacterium tumefaciens strain GV3101 chemically competent cells (Veidi Biotech, Shanghai) according to the attached user's guide.Preparation of Agrobacterium for node injection was based on Pan et al. ( 2020) with some modifications.Positive single clones were verified by bacterial suspension PCR using the Cas9-F/Cas9-R primer pair.100 µl of freshly prepared bacterial suspension were cultured in 10 ml of YEB liquid medium at 28 °C with agitation (250 rpm) until OD 600 reached 0.6-0.8(about 12 h).The cultures were centrifuged at 6000 rpm for 1 min to collect the bacterial cells.Equal volume of infection solution containing 100 µmol/L acetosyringone (BBI Lifesciences, Hongkong), 10 mmol/L MES (Sangon Biotech, Shanghai) and 10 mmol/L MgCl 2 •6H 2 O was added to the pellets.Freshly prepared resuspended bacterial pellets were used for injection.Node injection procedure was essentially the same as that in our previous report except for the plant age (Wang et al. 2013) (Fig. 2).Volume for each injection was approximately 5 µl (Wang et al. 2013).Injection was done between 6:00-8:00 a.m. on July 17, 2021, and the positions injected were marked with threads (Fig. 2).Inverted U-shaped metal wires were used to facilitate the entry of pegs into the soil from higher nodes (Fig. 2).

Fatty acid analysis of resultant T 1 seeds
Pods were harvested when matured (Sept.17, 2021).These pods were sun-dried and hand shelled.Oleic and linoleic acid contents of the individual single seeds were predicted with NIRS (Wang et al. 2014).T 1 generation seeds of transgenic lines with at least 74% oleic acid along with the untreated control Huayu 23 were further analyzed for fatty acids by gas-chromatography using cotyledonary slices following the protocol of Yang et al. (2012).

Cultivation of T 1 plants
To obtain descendants as soon as possible, one T 1 seed was sown in a pot in the winter of 2021, but unfortunately it died and did not set any pods.The remaining T 1 seed along with a Huayu 23 seed was sown under a film mulch on May 23, 2022.Both grew to maturity and were harvested on September 20, 2022.

Fatty acid profiling of the T 2 seeds harvested from T 1 plant
Oleic and linoleic acid contents of the T 2 seeds from the T 1 plant both as bulk seed sample and as individual single seeds were determined by NIRS (Wang et al. 2014(Wang et al. , 2021)).Huayu 23 CK was also analyzed for main fatty acid content.

Amplification of the bar gene
As expected, the 2 high-oleic peanut seeds produced a band with size equal to the positive control Cas9 empty vector,    whereas the untransformed Huayu 23 (negative control) yielded no band (Fig. 5), verifying that the 2 high-oleic peanut seeds were transformants.

T 1 plants grown from high-oleic T 1 seeds
The two high-oleic T 1 seeds and the untransformed control were sown in soil in the isolated areas.Both T 1 seeds developed into plants, but one of them (Huayu 23-7-2) died prior to flowering.The remaining T 1 plant (Huayu 23-7-1) grew normally and set a total of 73 seeds.

Oleic and linoleic acid contents and main agronomic characters of T 1 plant
All the seeds from the T 1 plant Huayu 23-7-1 were firstly used as a bulk seed sample in NIRS.The T 1 plant had an oleic acid content of 79.07%, as against 44.52% in Huayu 23.Sixty-eight well-developed T 2 seeds were then analyzed with NIRS for individual single seeds, and all were found to be high-oleic (Table 3).Similarly, forty-five Huayu 23 seeds were also used as individual single seed samples in NIRS analysis and as expected, all were normal-oleic (Table 3).
Compared to Huayu 23, the T 1 plant was shorter, set slightly bigger pods (Fig. 6; Table 4), and had much higher pod weight and seed weight (Table 4).

Discussion
Both the high-oleic mutant seeds had an F435 type FAD2B mutation (442 A insertion).It has been well documented that the 442 A insertion in FAD2B could cause the dysfunction in FAD2B due to the early appearance of a stop codon in the coding region caused by the frame shift mutation (Yu et al. 2008).The mutated FAD2B encoded a truncated oleate desaturase with the loss of a conserved histidine box, which was vital to enzyme activity.Yeast expression analysis also demonstrated that the 442 A insertion in FAD2B resulted in a loss-of-function FAD2B (Yu et al. 2008).Decreased transcription levels of FAD2B in F435 type high-oleic peanut were also observed (Jung et al. 2000).Since the genomeedited peanut seeds obtained in our study showed the same mutation in FAD2B as that reported by Yu et al. (2008) and Jung et al. (2000), it is inferable that the high-oleic phenotype is a consequence of the 442 A insertion in FAD2B.This was also consistent with the observation that FAD2B    transcript levels in high-oleic peanut were significantly lower than in its normal-oleic counterpart (Yu et al. 2008).
It seems that biallelic genome editing in this report is questionable.In fact, in a separate study, using a novel technology and with the help of NIRS for bulk seed samples, we were able to obtain high-oleic chemical mutant single peanut plants as early as in M1 from a popular normal-oleic Spanish market type cultivar with wild type FAD2A and wild type FAD2B (C.T. Wang, unpublished data).In other words, it appeared that in these high-oleic mutants, not only were both FAD2A and FAD2B mutated, but that these mutations were homozygous (C.T. Wang, unpublished data).This is very similar to the unexpectedly early appearance of the gene edited high-oleic mutants in this study.These may be ascribed to targeting earlier cells rather than postzygotic cells/tissues.In the present study, it is speculated that prior to the formation of the floral organs, the primordial cells that would differentiate to form male and female gametes were transformed.Even though the editing of the FAD2B gene was not biallelic at this stage, gamete fusion would give the chance to produce some zygotes with homozygous FAD2B mutation in T 1 generation.
The possibility that the mutant FAD2B allele was in the background genotype of the cultivar used for transformation can be fully excluded.To be on the safe side, special care was taken to use breeder seeds, and the normal-oleic phenotype and FAD2A/FAD2B genotype of Huayu 23 used for transformation were ascertained prior to the experiment.
In this study, evidence from fatty acid profile of the T 1 and T 2 seeds, FAD2B genotyping, and amplification of the bar gene all supported that genome editing of Huayu 23 was successful.Two peanut mutant T 1 seeds with over 80% oleic acid were generated via CRISPR/Cas9 genome editing technology following the node injection method developed by Wang et al. (2013).The high-oleic phenotype was expressed in T 2 seeds, clearly verifying the usefulness of the peanut transformation protocol.One of the advantages of the genome editing technology based on the present transformation scheme is that the high-oleic phenotype can be expressed as early as in T 1 seeds, which will further shorten the breeding process (Wen et al. 2018, Neelakandan et al. 2022a).
It was noted that T 1 plants and non-transgenic control differed in plant height, length of cotyledonary branches, number of effective branches, pod weight, and kernel weight.The T 1 plant and its parent were cultivated apart to avoid possible mechanical and biological mixing.In addition to environmental factors, genetic factors such as the position effect of integration sites and/or the influence on non-target genes may bring about these changes.Observations on their descendants planted in close proximity will clarify whether these differences are caused by the environment.
The rationale behind the node injection method is that most of the peanut seeds set on the first (cotyledonary branches) and second pairs of branches, that the possibility of harvesting sound mature kernels from the lower nodes was high, and that peanut cells which will develop into reproductive cells, or "primordial" reproductive cells, can be transformed (Wang et al. 2013).Generally, only the first and second nodes counting from the intersection of the main stem and cotyledonary branches were injected at 30 days after sowing (Wang et al. 2013).However, in this study, when everything was ready, it was too late (63 days after sowing), and only higher nodes could be injected.That is the reason why only a small number of seeds were harvested.Failing to edit in the anticipated target site of FAD2B may be due to the small population and/or the unsuitable oligos designed for genome editing vector construction, as the website for oligo design had no peanut genome option.Anyway, two CRISPRed high-oleic peanut seeds were identified from the 4 resultant seeds.Changes in oleic and linoleic acid contents in T 1 and T 2 seeds and the FAD2B sequences (442 A) of the two peanut T 1 transformants demonstrated that the high-oleic phenotype was inheritable.In addition to high-oleic acid phenotype, high and stable productivity is a prerequisite for a peanut cultivar to be accepted by growers.If future commercialization is anticipated, it will still be necessary to assess the overall performance of the derived lines in due course.There is still a need to scale up the genome editing experiments and try out some new techniques in future studies.For example, as a more recent and more efficient technique, prime editing is more versatile than base editing in generating nearly all types of edits.It is a promising genome editing tool, as indicated by the study of Biswas et al. (2022) .
In peanut, pods develop from pegs.One or multiple peg(s) is/are born at each node with peg(s), depending on cultivar (Nigam et al. 1990).In the case of multiple pegs, one injection may result in more than one pods, and in the meantime, the "primordial" reproductive cells at different developmental stages may increase the chances of being transformed.In this regard, the node injection method is advantageous over flower injection.A maximum of one pod can be harvested from a single flower.Our earlier study indicated that the node injection method was genotype independent (Wang et al. 2013), where no tissue culture procedure was needed, expanding its scope of use.Nevertheless, in-depth developmental studies on reproductive cells/tissues of the nodes may help optimize peanut transformation efficiency of this method.
Since the peanut node injection transformation method is easy to implement, peanut is an oilseed crop with oleosins, conducive to the separation and purification of expression products, and more importantly, peanut seeds can be eaten raw, the node injection method may thus facilitate peanut molecular pharming, for example, the production of edible vaccines.It is anticipated that the method, coupled with genome editing technology where necessary, will find wide utility in areas such as functional analysis of candidate peanut genes and breeding genome edited peanut cultivars for improved safety quality, ideotype, and other valuable traits, as long as they are controlled by oligoenes.We believe that this node injection transformation method is not only useful to peanut, but also of some reference to other seed plant species.

Conclusions
Altogether, using the node injection transformation method and a CRISPR/Cas9 construct targeting FAD2B, two high-oleic T 1 peanut mutant seeds with 442 A insertion in FAD2B were generated from normal-oleic cultivar Huayu 23 already having dysfunctional FAD2A.Amplification of the bar gene verified that the 2 high-oleic peanut seeds were true transformants.One of the T 1 seed developed into a healthy plant.NIRS analysis demonstrated that the high-oleic phenotype was expressed in T 2 seed generation.It is noteworthy that in this study the homozygous FAD2B mutation appeared in T 1 seed generation.This may be explained by the transforming of cells that would generate male and female gametes, and the editing of FAD2B at very early developmental stages.Therefore, the outcome of this study is better than that reported in other studies.

Table 1
Chemical quality of single T 1 peanut seeds and Huayu 23 predicted by near infra-red spectroscopy

Table 2
Fatty acids (%) in single peanut seeds determined by gas chromatographyIn each column of the fatty acid content, figures followed by the same letter were not significantly different at 0.01 level.Fatty acid content was expressed as mean ± SE

Table 3
Chemical quality of bulk and single T 2 peanut seeds and Huayu 23 predicted by NIRS Fatty acid content and O/L were expressed as mean ± SE, where appropriate

Table 4
Main agronomic characters of gene-edited Huayu 23 T 1 plant and Huayu 23