Introduction

Potato (Solanum tuberosum L.) belonging to Solanaceae is the world’s third most important staple food crop after wheat and rice. It is heterozygous, autotetraploid having diverse and adaptable tuber which act as storage organ as well as a vegetative propagation system. This crop has emerged from South America’s Andean mountains and currently has ~ 5000 varieties worldwide (Zaheer and Akhtar 2016). Potatoes are produced in about 125 countries and their total production is > 374 million tonnes globally (www.cipotato.org). In recent times, potato production has increased rapidly in developing countries compared to developed economies. China was the largest potato-producing country (91.81 million tonnes), followed by India (50.19 million tonnes), Russia (22.07 million tonnes), Ukraine (20.26 million tonnes) and the USA (19.18 million tonnes) in the year 2019 (www.fao.org/faostat). Both China and India collectively account for almost one-third of global potato production.

Potatoes are a rich source of various essential nutrients including carbohydrates, protein, vitamin C, vitamin B6, magnesium, potassium, iron and dietary fibers. The predominant type of carbohydrate in potatoes is starch and when fermented partially or wholly by the microflora in the large intestine, it is referred to as ‘resistant starch’. This resistant starch has health-promoting properties such as the production of short-chain fatty acids, lowering glucose and insulin responses, and acting as pre-biotics by supporting the growth of beneficial gut bacteria (Zhao et al. 2018). Owing to global health issues, the United Nations was instrumental in increasing awareness on the contribution of potatoes vis-à-vis global food security and its favorable nutrient-to-price ratio by declaring the year 2008 as ‘International Year of the Potato’. Being a versatile crop, it can be processed in a variety of forms including french fries, chips, flakes, mashed and canned potatoes. It has also been widely used by the pharmaceutical, textile, wood, and paper industries (Clasen et al. 2016; Fritsch et al. 2017).

The cultivated potato has a narrow genetic base due to limited germplasm introductions, thereby leading to acute inbreeding depression and susceptibility to many devastating pests and pathogens (Xu et al. 2011). Therefore, there is an urgent need to facilitate advances in breeding by the generation of novel functional genomics resources and understanding the genetic traits of economic importance for the successful breeding of potato cultivars. Also, with changing climate scenarios and alleviation of hidden hunger, generation of nutritionally rich potatoes along with improved yields are required prerequisites but unfortunately, various biotic and abiotic stresses have hindered such interventions. In recent years, the potato genome sequencing consortium has predicted 844 Mb genome sequence and 39,031 protein-coding genes in doubled-monoploid potato (Xu et al. 2011), which has created a platform for functional genomics research in potato. In the last decade, transcriptomic profiling has been proven a valuable and high throughput platform for the identification of novel candidate genes underlying specific traits of interest. Several gene expression studies have been done on the potato for detection of genes involved in tuber development, starch metabolism, biotic and abiotic stress resistance, etc. (Ferreira et al. 2010; Siddappa et al. 2014; Tiwari et al. 2015; Singh et al. 2015; Jeevalatha et al. 2017). However, functional validation of the predicted genes is the main challenge in the post-sequencing era. Functional genomics tools provide researchers with the ability to apply high-throughput techniques to determine the function and interaction of a diverse range of genes. The most effective way to identify functionally relevant gene/s regulating a mechanism is through the development of functional mutants by either loss of function or gain of function approaches.

In potato, various loss of function techniques such as virus-induced gene silencing (VIGS), RNA-mediated interference (RNAi), spray induced gene silencing (SIGS), targeting induced local lesions in genomes (TILLING), zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs) have been utilized in investigating the function of genes and characterization of relevant, candidate genes for important agronomical traits, resistance against biotic and abiotic stress and other biological processes. These approaches are useful for high-throughput screening, where gene function is disrupted or abolished and morphological variations are analyzed, resulting in unraveling the specific role of genes on a genome-wide scale. However, these gene disruption technologies are not always effective in investigating the function of redundant genes in a crop like potato, as loss of function of single loci may not result in a morphological variation. Also, genes involved in the growth and development of plants and some required only under particular environmental/stress conditions are difficult to analyze using the loss of function approaches (Wang et al. 2013). Further, gain of function approaches like activation tagging enhances gene expression and aid in the identification of gene functions. However, simultaneous expression of more than one gene may complicate the interpretation of gene function. Recently developed gene-editing technologies such as clustered regularly interspaced short palindromic repeats (CRISPR) –associated protein (Cas) confers both gain of function and loss of function by enabling targeted insertion, replacement, or disruption of genes in plants. But, this technique has yet to be fully understood in clonally propagated polyploids like potato. Here, in this article, we provide an insight into the different gain of function and loss of function tools applied in functional genomics of potato, which will facilitate their application in the development of elite cultivars with improved traits along with biotic and abiotic resistant varieties. We have also discussed the advantages and disadvantages of these techniques for planning molecular engineering studies in this versatile crop. Application of these functional genomic tools in potato and pros and cons of these techniques have been summarized in Table 1 and Fig. 1.

Table 1 Applications of different functional analysis techniques in potato
Full size table
Fig. 1
figure 1

Advantages and disadvantages of different loss and gain of function techniques

Full size image

VIGS

VIGS is a rapid and high throughput technique for functional analysis of genes through transient post-transcriptional gene silencing (PTGS). In this method, a short cDNA sequence from the gene of interest is cloned into a VIGS vector, which is further transfected into the host plant via Agrobacterium tumefaciens or virus sap inoculation or RNA transcript inoculation or DNA bombardment method. The VIGS vectors are obtained by modifying the plant viruses by removing pathogenicity-related genes and cloning the cDNAs of viral genomes into binary vectors under strong promoters like CaMV35S along with multiple cloning sites to enable appropriate insertion of target gene fragments (Voinnet 2001; Ramegowda et al. 2014). After the delivery of recombinant virus into plant cells, the transgene along with viral RNA is amplified by an endogenous or a viral RNA-dependent RNA polymerase creating a double-stranded RNA (dsRNA) molecule. This dsRNA is degraded by plant Dicer-like (DCL) enzymes into small interfering RNA (siRNA) molecules resulting in activation of PTGS, thus leading to the generation of siRNA homologous to the target gene which finally results in the silencing of the endogenous plant gene and a loss of function phenotype thereby, enabling the validation of gene function (Senthil-Kumar and Mysore 2014). In potatoes, PVX and TRV viral vectors have been found suitable for VIGS-based silencing (Brigneti et al. 2004; Faivre-Rampant et al. 2004). Brigneti et al. (2004) have tested the utility of the VIGS system to assess the function of candidate resistance (R) genes in potatoes and their wild relative. They have silenced known resistance genes R1 and Rx in S. tuberosum and RB in S. bulbocastanum and obtained susceptible phenotypes in detached leaf tests and explicated that the VIGS system is an effective method of rapidly assessing the gene function in potatoes. Du et al. (2013) identified lipoxygenase and suberization-associated anionic peroxidase genes that provide resistance to P. infestans in potatoes using TRV vector for VIGS and detached leaf assays. Salaria et al. (2020) have elucidated the role of cycling DOF factor (CDF) 1.2 allele in earliness and tuberization of Indian potato cultivars, Kufri Surya using real-time expression and VIGS technology. Recently, genes StSSH2 (succinic semialdehyde reductase isoform 2), StWTF (WRKY transcription factor), StUGT (UDP glucosyltransferase), StBHP (Bel1 homeotic protein) and StFLTP (FLOWERING LOCUS T protein) were evaluated using cDNA microarray studies and found to regulate and function as transcriptional factors, hormonal signaling, cellular homeostasis, and mobile tuberization signals in different potato cultivars. The further validation using VIGS revealed their role in tuber signaling and heat tolerance in potatoes (Tomar et al. 2021). It was also highlighted that the VIGS is an effective high-throughput technique for potato functional genomics.

The main advantages of VIGS include its low cost and quick performance by identifying a loss of function phenotype for a particular gene within a single generation and also permitting large-scale screening of genes for functional analysis in polyploid crops. In vegetatively propagating crops like potato, this technique allows the generation of genotypically identical silenced plants. It does not require stable transformants therefore, it is particularly useful for those potato cultivars which are difficult to transform (Burch-Smith et al. 2004; Singh et al. 2018). Also, there is no need to screen large populations to detect the function of a specific gene; only a single plant is enough to follow phenotype with targeted silencing. Therefore, it is extensively used as a powerful tool for decoding the functional relevance of the genes in potatoes (Becker and Lange 2010). However, this technique has certain limitations as the phenotypes obtained are not heritable; hence, could not be used for genetic engineering. Further, it may not result in complete knockdown of gene function due to the tetraploid nature of potato and even a low level of gene expression could produce functional protein and phenotype in the silenced plant. Many potato cultivars used in laboratory experimental setups are not well amenable to available VIGS systems. In addition, the levels of silencing could also vary between different potato varieties and experiments depending on the construct and the growth conditions (Burch-Smith et al. 2004; Gilchrist and Haughn 2010; Senthil-Kumar and Mysore 2011).

Stable RNA-mediated interference -RNAi

RNAi is a natural cellular process that downregulates the gene expression of the target gene by promoting the degradation of mRNA. It is a PTGS process and refers to a multi-step process, including the introduction of a DNA construct into a cell that produces dsRNA complementary to the gene of interest. This dsRNA is cleaved into ~ 21 to 25 nucleotides long siRNAs with 3′ two-nucleotide overhangs by Dicer or DCL enzymes. The siRNA contains a passenger (sense) strand and a guide (antisense) strand wherein the guide strand activates and incorporates into RNA-induced silencing complex (RISC) while the passenger strand is degraded by subsequent cellular events in the cytoplasm. The guide strand of the siRNA–RISC complex then binds to the target mRNA sequence fully complementary to it. The endonuclease Argonaute (AGO), the main catalytic component of the RISC complex then cleaves the target mRNA and thus, prevents the translation of the target transcript (Majumdar et al. 2017).

The microRNAs (miRNAs) are another 21 nucleotides long ssRNAs with the mode of action similar to siRNAs i.e., inhibiting the gene expression in a post-transcriptional manner. However, the biogenesis and gene silencing effects of miRNAs and siRNAs are different. miRNAs are produced by DCL1 from specific hairpin precursor transcripts and non-specifically target the expression of multiple mRNAs, whereas, the siRNAs inhibit the expression of one specific target mRNA in homology dependent manner (Lam et al. 2015; Moin et al. 2017). Moreover, gene silencing technologies like hpRNA and artificial miRNA (amiRNA) have been used to artificially induce RNA silencing in plants which play role in gene function analysis and genetic engineering for crop improvement. The hpRNA construct encompasses a sense and an antisense sequence of a segment of target gene mRNA as inverted repeats between a plant promoter and terminator. These inverted repeats are separated by a non-complementary spacer region which provides stability to the construct and enhances the silencing efficacy. The sense and antisense sequences in the transcribed RNA are complementary to each other and form a hairpin RNA structure which is then processed by DCL4 to generate 21-nucleotides siRNAs. These siRNAs guide RISCs to repress the expression of the target gene (Guo et al. 2016). The amiRNA construct is prepared by replacing the endogenous miRNA and miRNA* sequences in miRNA precursor with carefully designed amiRNA and amiRNA* sequences using overlap PCR while maintaining the stem-loop structure of the original miRNA precursor. The miRNA strand of an amiRNA construct consists of the sequence complementary to target mRNA, while the sequence of the miRNA* strand is designed to maintain the miRNA:miRNA* duplex structure of the native miRNA precursor (Guo et al. 2016). Thus, amiRNA transgenes use the natural miRNA pathway to silence genes. However, while preparing the amiRNA design, careful sequence selection is required for efficient AGO binding (Guo et al. 2016).

In potato, the RNAi approach has been used extensively for imparting resistance and identifying genes responsible for defense against pathogens, insects, and viruses that cause significant economic losses (Table 1). Missiou et al. (2004) have developed highly resistant transgenic potato lines against 3 strains of potato virus Y (PVY) by expressing hpRNA derived from the 3′ terminal part of the viral coat protein gene of PVY. Bhaskar et al. (2009) have demonstrated a double agroinfiltration method for RNAi-based silencing construct which could be used for screening candidate genes involved in late blight resistance pathway mediated by resistance gene RB. In another study, Eschen-Lippold et al. (2012) have assessed the role of vesicle fusion processes and callose deposition in secretory defense responses of potato against Phytophthora infestans. They have generated transgenic plants expressing RNAi constructs targeted against plasma membrane-localized syntaxin-related 1 (StSYR1) gene of potato which reduced the growth of P. infestans in potato. They have enhanced the resistance against late blight by down-regulating the expression of the syntaxin gene in potatoes. Cytological examination revealed that the infection caused by P. infestans was coincided with aberrant callose deposition and decreased papilla formation, signifying an involvement of syntaxins in secretory defense responses in potatoes. Moreover, oomycetes and fungal plant pathogens secrete various effector proteins into host cells which modulate host innate immunity and establish infection. Avr3a is an essential gene responsible for the virulence of P. infestans, the causal agent of late blight in potatoes. Sanju et al. (2015) have done the silencing of P. infestans Avr3a effector gene in popular cultivars of potato for development of resistant varieties using hpRNA construct approach. They observed that siRNA targeted against single effector gene Avr3a conferred partial resistance to P. infestans and indicated the need of targeting cumulative effect of effector genes to achieve complete resistance in potato. Similarly, Thakur et al. (2015) have used amiRNA for silencing P. infestans single effector Avr3a gene causing pathogen death or loss of virulence and imparted resistance against late blight in potato. Jahan et al. (2015) also designed and introduced an hpRNA construct containing the GFP marker gene in potatoes. They found hp-PiGPB1 targeting the G protein β-subunit (PiGPB1) important for pathogenicity resulting in most restricted disease progress. Hameed et al. (2017) have designed the expression cassette to generate dsRNAs having a hairpin loop configuration and developed transgenic potato lines expressing fused viral coat protein-coding sequences from PVX, PVY and potato virus S (PVS). They obtained nearly 100% resistance against 3 RNA viruses viz. PVX, PVY and PVS infection in transgenic lines compared to untransformed controls exhibiting severe viral disease symptoms. Tomar et al. (2018) targeted replication-associated protein gene (AC1) of ToLCNDV-Potato virus by post-transcriptional gene silencing using hairpin loop construct to confer resistance against apical leaf curl disease in potato. Recently, Hussain et al. (2019) have silenced the highly specific molting-associated Ecdysone receptor (EcR) gene of Colorado potato beetle (CPB) using RNAi. They have transformed the potato plants with Agrobacterium harboring RNAi constructs with EcR gene of CPB and when these larvae fed on transgenic plants, reduced EcR transcripts were observed which indicated the functionality of dsRNA in silencing EcR gene expression. Recently, Plantenga et al. (2019) have investigated the relation between tuberization and flower bud development by silencing the tuberization signal gene SELF PRUNING 6A (StSP6A) in potato plants. They showed that the tuberization signal StSP6A prevents flower bud development and reduction of this signal improves the development of flower buds. In addition, the expression of dsRNAs with synthetic promoters can control the spatial and temporal gene silencing in transgenic plants (Liu and Stewart 2016). A synthetic promoter contains a core promoter along with synthetic motifs and provides strength and specificity in transgene expression and thus, is ideal for gene function and advanced crop engineering research in potatoes. Li et al. (2013) have used a synthetic promoter, pCL to obtain an antisense expression of the acid vacuolar invertase (StvacINV1) gene from S. tuberosum. At low temperatures, specific regulation in expression and activity of target gene was observed in transgenic tubers, which prevented the cold-induced sweetening in potato (Li et al. 2013).

RNAi offers many advantages, primarily being its usage for discovering or validating gene functions along with genetic engineering studies due to its heritable expression in potatoes. Additionally, since silencing is sequence-specific; therefore, screening of large populations is not required and transcripts of multiple genes from a family could be silenced by a single construct, therefore, the utility of this technique is increasing for gene discovery and development of management strategies against various pathogens in this polyploid crop. Genetically modified potato plants using RNAi have already been approved at the commercial level and appropriate issues for risk assessment have been defined for this crop (Arpaia et al. 2020). Other advantages of RNAi in potatoes include its partial loss of function characteristic, thus producing several phenotypes of differing severity which could aid in the analysis of essential genes whose inactivation results in lethality or extremely severe pleiotropic phenotypes. RNAi-induced phenotypes are dominant and observed in the T1 generation, therefore, enables the easy screening of transformants in potatoes. (Small 2007; Eamens et al. 2008; McGinnis 2010). However, there are certain disadvantages of RNAi as some genes are difficult to silence by exogenous RNA due to its complex double-stranded structure. Also, there are chances of binding the siRNAs to off-target genes that have sufficient sequence homology to the target gene. The regulatory concerns increase, if off-target binding happens in non-target organisms, therefore, clarity regarding the selection of effective siRNA target sequences is required to pinpoint potential off-target genes (Gebremichael et al. 2021). The RNAi constructs are normally delivered as transgenes, via stable plant transformations and hence, they are required to undergo genetically modified organisms (GMO) regulatory compliance policies to get approval for commercial use (Arpaia et al. 2020). Also, silencing of undesired RNA targets may lead to modified phenotypes, which may be unfavorable for gene function screening experiments. Lastly, the silencing effect could vary depending on the gene copy number in the genome which may result in partial silencing of the target gene (Wang et al. 2005; Small 2007; Gilchrist and Haughn 2010).

SIGS

SIGS is a powerful, fast, and environmentally friendly method for efficient crop protection and validation of gene function in plants. In this technique, RNAi based in vitro synthesized dsRNA molecules, targeting the virulence-related genes of pathogens molecules are topically applied onto the plants to silence the targeted genes of pathogens and inhibit the infection caused by them (Wang and Jin 2017). These dsRNAs can be introduced into the vascular tissues of the plants via trunk injection, seed dressing, soil drench, or foliar application. This technique can potentially be developed against a wide range of pathogens or pests that have RNAi machinery. The dsRNA molecules applied on the plant surfaces act on the pathogen cells via two pathways. In the first pathway, plants take up the external dsRNAs which systemically spread in the plants and then transfer to the pathogen cells. These dsRNAs are processed into sRNAs by either the plant DCL proteins or pathogen DCL proteins. In the second pathway, pathogen cells directly take up the dsRNA and cleave it into sRNA by their DCL proteins (Wang and Jin 2017). In potato, Petek et al. (2020) have designed RNAi-based dsRNA to silence the CPB mesh gene (dsMESH) and validated its insecticidal action on beetle survival and development. They have found that spraying with insecticidal dsRNA has reduced larval growth in laboratory trials and is found to be a highly efficient strategy for controlling CPB. Recently, Sundaresha et al. (2021) have developed the RNAi-based dsRNA molecules targeting the infection and sporulation causing genes of P. infestans. The external application of this dsRNA is capable of effectively reducing the late blight infection in potatoes.

SIGS has various advantages, the most important is that it is environment friendly, because the dsRNA molecules are made up of nucleotides and will not release harmful residues in the environment. It is useful for those potato cultivars which are difficult to be genetically modified through transgenes or gene editing, thereby evading the problems in creating GMOs. This technique can be used for the identification and validation of target pathogen genes essential for disease development in potatoes. Further, unlike other structure-based antimicrobial sprays, the RNA sprays are based on target sequences therefore, do not persuade resistant or tolerant mutated pathogen strains (Vetukuri et al. 2021). However, there are certain limitations in this technique like the short-term stability of dsRNA in the environment. Effective SIGS is dependent on the efficient dsRNA uptake by the pathogens for disease control and gene function elucidation in potatoes. Also, the efficacy of topical application of dsRNAs is affected by the length and concentration of dsRNAs (Dubrovina et al. 2019). Nano clay-based formulations have been reported to enhance the delivery and boost the action of RNAi as SIGS for inhibiting the late blight progression in potatoes (Sundaresha et al. 2021). But, the nano clay or nanoparticle-based formulations are expensive and difficult to synthesize (Vetukuri et al. 2021). Further research in these areas is required for the efficient application and delivery of dsRNAs for RNAi-based plant disease control and gene function validation in potatoes.

Mutagenesis and TILLING

In mutagenesis, heritable changes are induced in the genetic information of an organism by chemical, physical and biological agents. Mainly, induced mutagenesis and insertion mutagenesis are employed in different mutation breeding programs. Induced mutagenesis is mostly instigated by radiations (gamma rays, X-rays, fast neutrons, etc.) or chemical mutagens (ethyl methane sulphonate, methyl methanesulphonate, ethyl nitrosourea, sodium azide, etc.) (Oladosu et al. 2016). Bombardment with gamma rays produces point mutations and small deletions, whereas fast neutron bombardment causes translocations, chromosome losses, and large deletions. Chemical mutagens are easy to use and tend to cause single base-pair changes or single-nucleotide polymorphisms (SNPs). These induced mutations are distributed randomly in the genome. They provide a high degree of saturation in the mutant population and warrant easy identification of gene function at the genome level (Gilchrist and Haughn 2010). There are some reports on mutation induction in potatoes and the development of a large number of mutants having different traits such as increased micro tuber yield upon γ-irradiation (Li et al. 2005), modified starch synthesis along with identification of new alleles based on the ethyl methane sulphonate mutagenesis (Muth et al. 2008), modified textural and histological properties (Nayak et al. 2007). On the other hand, in insertional mutagenesis, DNA elements that are randomly inserted within chromosomes such as T-DNAs, transposons, and retrotransposons are used as mutagens to create disruptions in target genes of interest along with the introduction of new genes or activation of endogenous genes in the plant genome. Since the sequence of the inserted element is known, the gene in which the insertion has occurred could be recovered using Thermal Asymmetric Interlaced PCR (TAIL-PCR) technique (Radhamony et al. 2005; Tadele 2016). Therefore, insertional mutagenesis is used as one of the most important tools in plant functional genomics to identify gene function. Duangpan et al. (2013) developed an insertional mutagenesis system in diploid wild potato species S. chacoense using the Tnt1 retrotransposon and obtained various phenotypes related to plant stature and leaf morphology during mutation screening of 38 families derived from Tnt1-containing lines. They demonstrated that this system could be exploited to tag every gene in the potato genome. Hence, insertion mutant libraries provide access to alleles for gene functional studies. Also, the phenotypic analysis of mutants signifies the role of a gene in biochemical, cellular, tissue, and organ characteristics, thereby, acting as a tool for studying the genetics behind the biological traits (O’Malley et al. 2015).

Overall, mutagenesis is a useful tool for linking sequence information to the biological function of genes and elucidation of novel variations for trait improvement in potatoes. These mutagenesis methods are low cost, heritable in nature, could result in complete knock out of the gene function, and lead to easy PCR-based detection of the mutation site. The major benefit of induced mutation methods is the non-requirement of the tedious transformation procedure. The point mutations generated by chemical mutagens could also produce a gain of function phenotypes if altered protein activity is obtained (Oladosu et al. 2016; Penna and Jain 2017; Kolakar et al. 2018). However, mutations are randomly distributed in the whole genome and special care is required while handling different mutagens (Kutscher and Shaham 2014). For a polyploid crop like potato, the insertions may end up in one or some of the four homologous chromosomes, leaving other genes intact, thus leading to unexpected phenotypes. Also, a very large mutagenized population is required to identify gene function making the process difficult and labor-intensive.

TILLING is a reverse genetics technique that allows high throughput detection of induced mutations in populations of physically/chemically mutagenized individuals (Tadele 2016). In this technique, seeds/vegetative propagules of plants are treated with chemical/physical mutagens to obtain M1 plants which are further self-fertilized to generate M2 plants. DNA is isolated from M2 plants and pooled for the identification of mutations. PCR is performed to target the desired locus which is end-labeled using fluorescently labeled forward and reverse primers containing the IRDye 700 and IRDye 800, respectively. PCR products are denatured followed by slow annealing to form homoduplexes and heteroduplexes comprising of wild type and mutant amplicons, which are further cleaved by single-strand specific nucleases of S1 nuclease family such as CEL I which cleaves to the 3' side of mismatches in heteroduplexes. The most common method for mutation detection is through size-fractionation of cleaved products representing mutations by denaturing polyacrylamide gel electrophoresis which is visualized by fluorescence using the LI-COR DNA analyzer (McCallum 2000; Elias 2009; Fondong 2016). EcoTILLING is an extension of TILLING, which identifies polymorphism (SNPs or small insertions/deletions [INDELS]) for the gene of interest among natural populations (Tadele 2016). It follows a similar protocol to the TILLING in which mismatches formed by hybridization of different genotypes in a test panel are cleaved with CEL I. This technique is useful in search of variations in disease resistance genes in various plant species (Fondong 2016).

Sequencing-based TILLING approaches have also been employed in crops like rice and wheat which are highly sensitive, requiring specific methods for the discovery of rare mutations in populations (Tsai et al. 2011). In potato, Elias et al. (2009) standardized the TILLING and EcoTILLING based protocol for primer design, enzymatic mismatch cleavage, and fluorescence detection of polymorphism in gamma-irradiated potato cultivars, which enabled rapid germplasm characterization and mutation identification in high-throughput reverse genetic screens of large germplasm in a short time frame.

This stable technique has importance in tetraploid potato as it produces different kinds of mutations including missense, nonsense, and silent leading to complete/partial loss of function or gain of function which could play role in the identification of gene function. Since exogenous DNA is not introduced into the plant, therefore it does not require the lengthy procedure of regeneration to form a transgenic plant thereby, aiding in its usage in recalcitrant cultivars of potato and exemption from regulatory restrictions applied on transgenic products. Also, TILLING can be utilized for a particular gene of interest; hence, mutations that are difficult to be known by forward genetics could be identified using this technique (Fondong 2016; Tadele 2016). However, the requirement of prior knowledge of sequence for locus-specific amplification and complex detection of mutations close to simple sequence repeats due to polymerase slippage aroused mutations (Fondong 2016) are some of the major limitations of TILLING.

Genome editing

Sequence editing either by deleting or modifying the genes individually and then studying the subsequent mutant phenotypes could address the challenges of understanding the function of genes. For precise DNA manipulations, new genome editing systems have emerged in recent times. These can induce double-stranded breaks (DSBs) at specific sites in the genome which is repaired naturally using DNA repair mechanism by non-homologous end-joining or homologous recombination, thereby ensuring the gene mutation at the target site. Genome editing is a versatile tool that could be employed for overexpression as well as gene knockout studies. This system is facilitated by protein-guided nucleases, such as ZFNs, TALENs, or special RNA/DNA-guided nucleases, including RNA-dependent DNA cleavage systems like CRISPR/Cas system (Zhang et al. 2017).

ZFNs

ZFNs are the first generation of engineered endonucleases that were developed following the discovery of the natural type IIS restriction enzyme, FokI having separate DNA binding and cleavage domain (Li et al. 1992). It provided the opportunity to engineer vertically any DNA binding protein into a nuclease (Kim and Chandrasegaran 1994) and identification of functional Cys2-His2 zinc finger (ZF) domain which was composed of 30 amino-acid residues that fold into a unique ββα configuration (Pavletich and Pabo 1991; Kim et al. 1996; Pabo et al. 2001). ZFN monomer mainly consists of two different functional domains viz. synthetic ZF Cys2-His2 domain at N-terminal, which recognizes 3–4 base pairs of DNA and binds by inserting the α-helix into the major groove of the double helix, and a nonspecific FokI endonuclease domain at C-terminal region, which cleave the DNA at a precise location. A critical property of FokI is that it dimerizes to cleave DNA so that two adjacent ZFN pairs orient themselves with appropriate spacing at the target site. Selection and linking of zinc fingers in a sequence allows recognition of 18- or 24-base target sequence, therefore, enhancing sequence specificity and reducing off-site cleavage to manipulate genomic sites of interest. The advantages of ZFNs include their efficiency, high specificity, and minimal non-target effects compared to other techniques. These have been successfully applied to gene modification in different plants such as Arabidopsis (Lloyd et al. 2005; Osakabe et al. 2010), tobacco (Townsend et al. 2009), and maize (Shukla et al. 2009). However, attaining efficient ZFNs is an extensive and time-consuming process and its efficiency varies with each construct (Xiong et al. 2015).

TALENs

Recently, TALENs have rapidly appeared as an alternative to ZFNs for genome editing. TALENs are based on the fusion of sequence-specific DNA binding TALE proteins to a non-sequence-specific FokI nuclease generating a targeted double-stranded break (Christian et al. 2010; Li et al. 2011; Miller et al. 2011). TALE proteins recognize and activate specific plant promoters through a central repeat domain embracing repeating units of 34 amino acid residues which binds to one nucleotide in the target nucleotide sequence. Each repeat is identical apart from the repeat variable residues at positions 12 and 13 which confer the recognition specificity (Boch et al. 2009). TALENs are easier to generate compared to ZFNs, therefore have been successfully used in various plant species. In potato, Sawai et al. (2014) confirmed the use of TALENs for the first time by knocking out the 4 alleles of sterol side chain reductase 2 (SSR2) to reduce the anti-nutritional sterol glycoalkaloid production in tubers. Further, the Acetolactate synthase1 (ALS1) coding gene was targeted in tetraploid potato by transient TALENs expression in protoplasts and successful regeneration of ALS1 knock-out lines were observed (Nicolia et al. 2015). Clasen et al. (2016) enhanced cold storage and processing traits in potato variety, ‘Ranger Russet’ by targeting Vacuolar invertase (VInv) using TALENs via protoplast transformation and regeneration. Tubers obtained from VInv-knockout plants showed undetectable levels of reducing sugars and their processed chips were lightly colored with reduced levels of acrylamide. Studies related to high-frequency T-DNA integration and effective delivery of TALENs via agroinfiltration for targeting genes encoding starch branching enzyme and acid invertase have been conducted in different potato cultivars (Forsyth et al. 2016; Ma et al. 2017). Recently, Yasumoto et al. 2019 have developed a highly active Platinum TALEN expression vector construction system and targeted SSR2 gene, which was found suitable for controlling the levels of toxic metabolites in potatoes.

The main advantage of TALENs includes its design and construction in a short period and large numbers. Also, there are fewer constraints on site selection for TALENs. However, this technique is expensive and its efficiency varies with each construct. Another limitation is its large size which makes it difficult to deliver and express into cells (Boettcher and McManus 2015; Unniyampurath et al. 2016; Malzahn et al. 2017).

CRISPR

More recently, a new genome-editing tool CRISPR/Cas system has been developed as an easy, specific, and efficient alternative to ZFNs and TALENs for inducing targeted genome editing. CRISPR is an RNA-dependent DNA cleavage system that has been found to function as an adaptive defense mechanism in bacteria and archaea that recognizes and cleaves the complementary DNA sequences present in invading viruses, phages, and plasmids (Wiedenheft et al. 2012; Malzahn et al. 2017). There are mainly three types of CRISPR-Cas systems namely Type I, Type II, and Type III based on the target and composition of the Cas genes. DNA editing can be performed by all three types of CRISPR/Cas systems. Type II and III systems can target DNA as well as RNA, while only DNA is targeted by the type I system. The type II system has a single Cas9 protein and two different RNA subunits in complex. Type I and III systems have multiple Cas proteins in complex with a single RNA (Makarova et al. 2011; Unniyampurath et al. 2016).

Type II CRISPR-Cas system is most prominent and widely adapted for eukaryotic gene editing, in which a single and very large protein Cas9 is sufficient for recognizing and cleaving the target DNA (Unniyampurath et al. 2016). This bacterial CRISPR/Cas system cleaves foreign DNA by involving 3 components: a Cas9 nuclease, CRISPR RNA (crRNA), and trans-activating crRNA (tracrRNA). crRNA and tracrRNA form a complex by base pairing which activates and guide Cas9 to the target sites complementary to the 20-nucleotide sequence of crRNA. Cas9 is a DNA endonuclease with two nuclease domains, HNH and RuvC which introduce DSB in target DNA. The HNH domain of Cas9 cleaves the complementary strand of crRNA while the RuvC domain cuts the identical strand of double-stranded DNA. For efficient cleavage, Cas9 protein also needs a protospacer adjacent motif (PAM) sequence downstream of the target DNA, which usually has the sequence 5′-NGG-3′ (Soda et al. 2018). CRISPR/Cas immune system of bacteria can identify between self and non-self-sequences during interference via PAM recognition. CRISPR/Cas system has developed as a gene-editing tool from this biological phenomenon by combining the targeting specificity of crRNA and structural properties of tracrRNA into a chimeric single guide RNA (sgRNA), thus reducing the system from three components to Cas9 endonuclease and a sgRNA (Doudna and Charpentier (2014). Therefore, by changing 20 nucleotides in sgRNA, the target DNA sequence can be reprogrammed easily. After the generation of DSB, DNA repair mechanisms like non-homologous end-joining (NHEJ) or homology-directed repair (HDR) are initiated. In most cases, DSB is repaired by NHEJ and generates mismatches and gene INDELs, leading to gene knockout. When an oligo template with regions of homology is present to the sequence near DSB, HDR induces specific gene replacement or foreign DNA knock-ins (Liu et al. 2017).

CRISPR/Cas system is further modified by developing different variants of Cas protein to increase the efficiency of this system and target a wide range of genes. CRISPR-Cpf1 (CRISPR from Prevotella and Francisella1 or Cas12a) system is a substitute to CRISPR/Cas9 and an advanced as well as the more efficient version of a genome-editing tool (Alok et al. 2020). Cpf1 is also RNA guided nuclease and makes DSB specified by crRNA, but unlike Cas9, it introduces 5 base pairs staggered cut and recognizes thymidine-rich PAM, located at 5′ end of the displaced strand of the protospacer (Zetsche et al. 2015). CRISPR/Cpf1 requires only crRNA to make DSB at the target site and does not require tracrRNA. Therefore, it is a simple system and can be used in various biological approaches including multiplex gene targeting, transcription, epigenetic modulation, and base editing (Safari et al. 2019). One of the other variants of Cas9 is dCas9 i.e. CRISPR nuclease deficient Cas9. This is catalytically inactive or dead Cas9 which cannot cleave DNA but specifically and strongly binds to target sequence with the guidance of sgRNA. Fusing different transcription repressors or activators to the CRISPR/dCas9 system modifies the action of target genes by inhibiting or enhancing transcription (Adli 2018). Dimeric RNA-guided FokI nucleases (RFNs) are another alternative to the CRISPR/Cas9 system and are a fusion of catalytically inactive dCas9 protein with the FokI nuclease domain. Dimerization of two RFNs is required for efficient genome editing. These endonucleases introduce break in DNA only when two sgRNAs binds to complementary DNA sequences at a precise spacing and orientation, thereby reducing error and improving the specificities of genome editing (Khatodia et al. 2016). Apart from these variants, there is another alternative approach that directly delivers ribonucleoproteins (RNPs) complex consisting of the Cas9 protein and a targeting sgRNA, to plant cells via transfection or biolistic method. This CRISPR/Cas9 RNP method does not use any foreign DNA, therefore, help in raising marker-free or transgene-free plants (Soda et al. 2018).

Freshly, a new CRISPR-derived genome editing system called prime editing has been developed by fusing a reverse transcriptase (RT) to a catalytically inactive Cas9 endonuclease (CRISPR/dCas9 H840A) (Anzalone et al. 2019). It is a ‘search-and-replace’ genome editing technique that directly inserts a new sequence without a DNA template in the targeted genome site. This system contains a programmable prime editing guide RNA (pegRNA) to produce sgRNAs and RT templates of targeted edits (Anzalone et al. 2019). Prime editing system allows the introduction of all kinds of mutations including insertions, deletions, and 12 types of base conversions, without generating DSBs or requiring donor DNA templates (Anzalone et al. 2019). Therefore, this system holds great promises for functional genomics through precision editing and several traits of interest can be conferred by point mutations rather than gene loss-of-function.

In potato, Wang et al. (2015) and Butler et al. (2015) have successfully demonstrated the use of CRISPR-Cas for targeting the StIAA2 gene encoding an Aux/IAA protein and ALS1 gene, respectively in different potato cultivars. Andersson et al. (2017) altered the tuber starch quality in a tetraploid potato by knocking out all 4 alleles of granule-bound starch synthase (GBSS) using CRISPR-Cas9. Further, Andersson et al. (2018) developed amylopectin starch potatoes by eliminating the amylose formation by knocking out the GBSS enzyme responsible for amylose production via CRISPR‐Cas9 RNP technology. Johansen et al. (2019) have increased editing efficiencies in the GBSS gene at the protoplast level by replacing the Arabidopsis U6 promoter, thereby driving expression of the CRISPR component with endogenous potato U6 promotor in potato. Previously, Ye et al. (2018) also developed self-compatible diploid potatoes by knocking out the Stylar ribonuclease S-RNase gene responsible for self-incompatibility using the CRISPR–Cas9 system. Recently, González et al. (2020) have studied the application of the CRISPR/Cas9 system to induce mutations in the polyphenol oxidases (PPOs), which catalyze the conversion of phenolic substrates to quinones and lead to the formation of dark-colored precipitates in potatoes. They targeted the StPPO2 gene and showed that editing of this gene resulted in a lower PPO activity in the tuber with the subsequent reduction of the enzymatic browning. More recently, Kieu et al. (2021) have used the CRISPR-Cas9 mediated loss of gene function technique and showed that StDND1, StCHL1, and StDMR6-1 genes were involved in late blight susceptibility, which could be targeted for the breeding of new disease resistance cultivars in potatoes. In potato, prime editing has also been successfully applied for the introduction of simultaneous transition and transversion mutations in the ALS1 gene that encodes a key enzyme in the biosynthetic pathway for branched-chain amino acids (Veillet et al. 2020). Hence, these studies suggest that a CRISPR-Cas9 system is a powerful tool for genome editing in potatoes.

The main advantages of this technique include its easy, fast and efficient performance. It can target a different gene by replacing a 20-bp complementary nucleotide sequence. It has the potential of editing multiple genes in one line through a single transformation (Cong et al. 2013; Gaj et al. 2013; Xiong et al. 2015). Further, it has an additional advantage of targeting methylated DNA. Also, the mutations created in the primary transgenic plants using this tool are stable in the next generation (Feng et al. 2014). However, there is a requirement for a PAM motif to target a sequence that limits the target. Moreover, there is the possibility of off-target effects which could be difficult to detect and require scanning the genome for mutations at sites with sequence similarity to the gRNA target sequence (Boettcher and McManus 2015; Unniyampurath et al. 2016; Malzahn et al. 2017). Due to the high heterozygosity in potatoes, it is important to select gRNAs from the most conservative regions of the gene. Editing can lead to the formation of heterozygous, biallelic, chimeric, or homozygous edited plants, and amongst these, the homozygous state is the most desirable, which is however not achieved frequently in polyploid crops like potato (Fizikova et al. 2021).

Activation tagging

Activation tagging is a powerful method to determine the function of a gene product by generating a gain of function mutants in plants that limits traditional screens for loss of function mutations. In this technique, T-DNA containing 4 tandem copies of multimerized transcriptional enhancers from the CaMV 35S gene is randomly integrated into the plant genome, which enhances the expression of either side of the flanking genes. The first T-DNA vector containing tetrameric CaMV 35S enhancer elements for activation tagging was constructed by Hayashi et al. (1992). This vector has been successfully utilized in Arabidopsis tissue culture cells for the identification of genes involved in cytokinin signaling (Kakimoto 1996). Different novel activation tagging vectors conferring resistance to kanamycin or herbicide glyphosate have been developed and employed to generate a large number of transformed plants (Weigel et al. 2000).

This system has been used to characterize several genes involved in important functions in different plant species including Arabidopsis, petunia, tomato, poplar and barley (Weigel et al. 2000; Zubko et al. 2002; Mathews et al. 2003; Ayliffe et al. 2007; Busov et al. 2011). In potato, a large number of activation tagged lines were generated by the Canadian Potato Genome Project which was screened for traits related to tuber health and tuber quality (Regan et al. 2006). Aulakh et al. (2014) characterized the potato activation tagged mutant AT615 and identified 1632 genes that were differentially expressed between mutant and wild type. Further, Aulakh et al. (2015) identified the T-DNA insertion and flanking genes on chromosome 12 and observed the cytosine methylation in wild-type, mutant and revertant phenotype plants.

There are various advantages of activation tagging technology. Firstly, it generates a dominant phenotype which makes the screening of phenotypes feasible in the T1 generation. Secondly, redundant genes are most likely to produce phenotype as only one allele of one locus of a redundant gene needs to be activated to observe gene activation. Thirdly, unlike conventional screening, the gain of function screens obtained using this technology could be easily identified for further experimentation. Also, problems related to ectopic expression could be overcome by using the 35S enhancers instead of using full constitutive promoters. However, this technique requires transformation, thereby making it tedious for the generation of mutant lines in some plant species. Moreover, simultaneous expression of more than one gene in activation tagged mutants may lead to complex phenotype resulting in difficulty to saturate genes (Ichikawa et al. 2003). Also, it is difficult to obtain a complete set of tagged mutants for the functional analysis of all genes in activation tagging (Tomoko et al. 2010).

Conclusion and perspectives

Potato has a major role in supporting global food security but its genome remains relatively unexplored. Therefore, the big challenge is to systematically assign biological functions to all predicted genes in the genome. Functional genomics provides a better understanding of deciphering new experimental opportunities for developing nutritionally rich potato varieties in combating hidden hunger, generating biotic and abiotic stress-resistant varieties and developing new breeding programs for its improvement. The general approach to identify functions of genes is based on the analyses in phenotypic variations between wild-type and its mutant. There are different methodologies available for functional analysis studies in plants and their execution depends on the advantages and disadvantages of the techniques with their ability to address different questions. VIGS and RNAi are low-cost methods that can create mutants for the targeted genes, which could thereby be used in studying the unknown function of genes. RNAi has the potential to target multiple genes for silencing using a single transformation construct aimed at gene discovery in potatoes; however, sequence-based off-target effects remain the most challenging issue in RNAi experiments. Therefore, in-depth in silico homology searches are required to avoid off-target effects while designing the construct. As a non-GMO alternative to host-induced gene silencing, SIGS is emerging as a new approach for the validation of gene functions in potatoes. The efficiency of SIGS is directly related to RNA uptake efficiency by the pathogen and hence, further optimization of delivery vectors and methods is required to establish the efficacy of this technique across a wide range of pathogens. Mutagenesis can lead to a complete knockout but a very large mutagenized population is needed to be screened for achieving genome saturation which again is difficult and time-consuming. All these techniques could miss phenotypes that are masked by functional redundancy between gene family members. TILLING also needs prior information of sequence for locus-specific amplification. Newly emerging genome editing techniques like ZFNs, TALENs, and CRISPR have efficiency, high specificity, and low cost compared to other techniques and exhibits the potential for knocking out a target gene in potato. Since potato is tetraploid with high genome heterozygosity, selection of gRNA is very crucial to avoid off-targets. Activation tagging in potatoes could create a large number of gain of function mutations which could be easily identified along with in-depth analysis of redundant genes. Hence, the aforementioned techniques could result in the production of functionally important mutants/genetic resources leading to identification of genes playing important role in biotic and abiotic stress resistance along with breeding healthy and nutritionally rich potatoes.