**2. Genetic factors involved in maize haploid induction**

The inheritance of haploid induction rate (HIR) has been extensively investigated during the last two decades. *In-vivo* haploid induction in maize is accomplished using three approaches (**Figure 2**). The first method uses a mutation originating from 'Stock 6' to induce maternal haploid induction, and it is commonly employed in commercial maize breeding programs [3]. On chromosome 1, a pollen-specific phospholipase-A gene known as *Matrilineal* (*MTL*) was discovered with a 4-bp insertion in the final exon of the gene, causing a flame shift mutation and premature stop codon [4–6]. When coupled with the *MTL* gene in homozygous recessive form (*mtl/mtl*), a second mutant gene called *ZmDMP* (encoding DUF679 domain membrane protein) on chromosome 9 increases the haploid induction rate [7]. As a result, mutations in the *MTL* and *ZmDMP* genes are required for maize maternal haploid induction. The second strategy, on the other hand, entails a mutation in the *indeterminate gametophyte1* (*ig1*) gene, which was identified in Wisconsin-23 (W23) inbred for paternal haploid induction [8]. It was previously recognized that ig1 on chromosome 3 encodes a lateral organ boundary (LOB)-domain protein, which is part of a wide family of transcription factors important for plant lateral organ development [9]. Because the underlying determinants for both procedures (maternal and paternal) were different, the mechanisms of haploid induction in both ways differed greatly in the commercial breeding programs. Aside from these two key changes, induction of haploids in maize is influenced by several variables, including

*Accelerated Generation of Elite Inbreds in Maize Using Doubled Haploid Technology DOI: http://dx.doi.org/10.5772/intechopen.105824*

#### **Figure 2.**

*Production of haploids using in-vivo induction system. A: Maternal haploid induction using* mtl *mutant; B: Paternal haploid induction using* ig1 *mutant; C: Both maternal and paternal haploid induction using* cenh3 *mutant.*

the maternal genetic background of the donor germplasms and the environment in which the induction crosses are made [1, 10, 11]. Another approach for generating both maternal and paternal haploids from a single inducer is *CENH3*-based haploidization [12]. *CENH3* gene codes a histone H3 protein variant found in centromeric nucleosomes that have two primary domains. In CENH3 protein, the N terminal tail domain has little in common with regular histone H3, but the C terminal histone fold region has a lot in common with normal histones [13]. At the homozygous stage, a mutation in *CENH3* is fatal because chromosomes fail to segregate to poles during cell division due to the lack of functional centromeres [14].

### **3. Haploids identification after induction crosses**

During an induction cross, haploids appear at a frequency of ~10% depending on the HIR, while the remaining 90% of seeds are diploid with no utility [3]. As a result, distinguishing haploids from diploid offspring at the seed, seedling, or mature plant stage is critical. Reducing the number of progenies would be helpful since it would lower the cost of developing DH lines. Different morphological and molecular indicators can be included in the inducer genotypes utilized in the DH development process [15, 16]. The dominant genetic marker produced in the seed or seedling stage can be included in mother haploid inducers, allowing haploids formed from induction crosses to be differentiated [1]. In most maize breeding programs across the world, haploid inducer with *R1-nj* (*Navajo*) allele is commonly utilized for haploid identification [1, 15]. The purple color of the scutellum of the embryo and the aleurone layer of the endosperm in diploid seeds influence the *Navajo* phenotype. Anthocyanin, on the other hand, is exclusively found in the endosperm of haploid seeds, not in the embryo. *R1-nj* marker makes it easier to distinguish haploids from

diploids based on visual inspection due to the different colors of the embryo [17, 18]. The existence of a dominant inhibitor allele *C1-I* in tropical elite inbred lines inhibits anthocyanin production on seeds, which is a fundamental restriction of the *R1-nj* marker-based method [16]. When identifying haploids exclusively based on the *R1-nj* marker is challenging, the *Pl1* gene is employed as an alternative [19]. Pl1 gene induces light-independent anthocyanin synthesis in seedling roots, allowing haploids and diploids to be identified that were previously misclassified by the *R1-nj* marker. The *Pl1* gene, on the other hand, can frequently lead to misinterpretation due to the formation of red roots in seedlings after exposure to sunlight [19]. Furthermore, in the adult stage, recessive phenotypic mutations such as *liguleless* can be utilized to identify haploids [1].

Because of the numerous drawbacks of phenotypic morphological markers, multiple attempts have been made to use genetic markers based on the xenia effect of high oil content for haploid identification [20]. The use of a haploid inducer with a high oil content would be advantageous since the high oil marker is not genotype-dependent, allowing it to be applied to practically all genotypes, including landraces and wild cousins like teosinte [1]. As a result, the genes that cause high oil content may be targeted in order to create inducers with high oil content. The effectiveness of the oilbased identification technique, on the other hand, is dependent on a large difference in oil content between source germplasm and inducer, since a little difference would result in a higher number of false positives and false negatives [21, 22]. Automating the process of haploid identification would be a cost-effective and practical solution since it would considerably cut the cost of wages for those participating in the haploid identification process [23]. Several mechanical approaches have been altered based on *R1-nj* marker expression on embryo and endosperm employing multispectral, hyperspectral, and fluorescence imaging techniques (**Figure 3**). In this case, an imagining-based automated approach powered by machine learning and deep learning understanding might

**Figure 3.**

*Development of automation system based on hyperspectral spectroradiometer using a machine learning algorithm.*

*Accelerated Generation of Elite Inbreds in Maize Using Doubled Haploid Technology DOI: http://dx.doi.org/10.5772/intechopen.105824*

be a feasible option since it decreases the time and effort required to identify haploids [23, 24]. As a result, numerous approaches for haploid identification employing *in-vivo* HI are available. The main task is to create an HI with a mix of appropriate markers that can solve all of the problems associated with the identifying procedure. No marker method can give a universal strategy that is applicable to all germplasm [1]. As a result, developing an inducer with a proper marker system for the desired breeding program is critical before starting any DH programs. As a result, it is often recommended to create a set of HI with distinct marker systems appropriate for different types of source germplasms.
