Doubled Haploids in Maize: Development.

The slides deals with the Maize doubled haploid technology, its development, its use and the challenges associated with this technology.

1. Doubled Haploids in Maize: Development, Deployment and Challenges By Himakara Datta Mandalapu , Id no: 2021608010 II Ph.D. (GPB), Department of Plant Breeding and Genetics GPB 692- Doctoral Seminar (0+1)

2. Chairperson: Dr. S. Sivakumar , Professor (PBG) and Head, Department of Millets, CPBG, TNAU, Coimbatore Member 1 Dr. N. Kumari Vinodhana , Assistant Professor (PBG) Dept. of Millets, CPBG, TNAU, Coimbatore Member 2 Dr. N. Senthil , Dean, School of Post Graduate Studies, TNAU, Coimbatore Member 3 Dr. A. Senthil , Professor (Crop Physiology) and Head, Dept. of Crop Physiology, TNAU, Coimbatore Member 4 Dr. D. Uma , Professor (Biochemistry) and Head, Dept. of Biohemistry , TNAU, Coimbatore ADVISORY COMMITTEE

3. Doubled Haploids When haploid (n) cells undergo chromosomal doubling, they generate a genotype known as a Doubled Haploid (DH) Efficient tool to obtain complete homozygosity within a heterozygous progeny in single step Discovered in 1921 by A. D. Bergner in Datura stramonium L . and reported by Blakeslee et al., in 1922. Been adapted in plant breeding programs for many decades Being practiced in crop species like maize, wheat, barley, rice, oat, brassica, tomato, sorghum, millet, cotton .

4. History Maize haploids were first reported by Randolph in 1932. Use of colchicine to artificial doubling – Blakeslee and Avery (1937) DH use in maize breeding – Chase (1952) Anther culture – Guha and Maheshwari , 1966. Niizeki and Oono (1968) – Production of haploids in tobacco Wide crossing in Barley – Kasha and Kao (1970)

5. Techniques for Producing Doubled Haploids Techniques 1 Androgenesis Anther or pollen culture Preferred because of their cellular totipotency . 2 Gynogenesis Used where androgenesis and wide hybridization are not possible or effectively functional. In vitro culture of unfertilized female gametes

6. 3 Wide hybridisation Crossing between species from same genus or two different genera Bulbosum method in barley (crosses with H.bulbosum ) Crosses b/w Oat and maize for oat haploids Crosses between Wheat and maize for wheat haploids 4 Parthenogenesis Induced in cucurbits and some citrus species Use of irradiated pollen Obtaining the haploid plants and chromosome doubling PsASGR -BABY BOOM like gene in Pennisetum squamulatum – HIR -35% in pearl millet (Conner et al., 2015). 5 Miscellaneous Sparse pollination Alien cytoplasm – Aegilops caudate in wheat ( Kihara and Tsunewaki , 1962) Centromere-mediated genome elimination – CENH3 locus in A. thaliana

7. Doubled Haploids in Maize Breeding Chase pioneered the study of maize monoploids Developed a hybrid, DeKalb 640, which is a double cross with three DH lines and a station inbred Relied on spontaneous induction and doubling Not commercially feasible Coe (1959) - In vivo haploid induction – Stock 6 Line of maize with high haploid frequency – 2.3% - Coe Developed marker systems such as R1-nj , C1-I K.R. Sarkar – Genetic analysis of the origin of maternal haploids in maize, genetic analysis of the R1-nj marker

8. Why DH in Maize Breeding? Efficient and precise selection Predictable and precise Simplified logistics Reduced line maintenance efforts Allows accelerated development Access to the germplasm present within either the female or the male parental lines of hybrid cultivars Time-saving Application in genomics High genetic gain per year

9. Production of maize DH lines Involves four steps Induction of haploids Identification of haploids at seed or seedling stage Chromosome doubling Selfing the fertile doubled haploids to produce DH seeds

10. Haploid Induction Can be done using both In vitro and in vivo methods In vitro methods: little promise for reliable production of large number of inbreds In vivo method – Use of haploid inducer lines (Coe,1959) Based on their adaptability Temperate haploid inducers HIR of 6-15% Founder – Stock 6 Lines: KMS, ZMS – Derivatives of Stock 6 WS14 – W23 ig x Stock 6 Poorly adapted for tropical climate Tropically adapted inducer lines (TAILS) Developed by CIMMYT in partnership with University of Hohenheim, Germany HIR of 8 – 10% 1 st generation TAILS – 6-9% HIR 2 nd generation TAILS – 9-14% HIR

11. Paternal inducer lines Carry the haploid genome of a male parent Haploid inducer line is used as a female Indeterminate gametophyte1 ( ig1 ) mutant gene Spontaneous mutation observed in Wisconsin-23 (W23) inbred line . ( K ermicle , 1994) HIR – 3% Paternal haploids contain the cytoplasm of the female inducer and haploid genome of the pollen donor Wild type ig1 is located on chromosome 3 Encodes LATERAL ORGAN BOUNDARIES (LOB)-domain protein

12. Paternal inducer lines Functions to switch from proliferation to differentiation in the maize embryo sac ig1 mutants contain a retrotransposon insertion within the second exon of gene, upstream of encoded LOB domain. Exact mechanism that leads to haploid induction is unknown Low frequency of haploids and changes in the constitution of cytoplasm from the donor genotype – Not attractive Ig1/ig1 genetic stock wit CMS cytoplasm can be developed Can be useful for conversion of inbreds into their CMS form

13. Maternal inducer lines Inducer line used as the pollen parent Haploids carry the genome of the female parent Exact sequence of event underlying haploid induction are not clearly understood Haploid induction trait of Stock 6 inducer – dominant and controlled by few major nuclear genes Early mapping revealed QTL on chromosome 1 – named as gynogenesis inducer1 ( ggi1 ) Further studies revealed eight QTLs on six chromosomes qtl for hir 1 ( qhir1 ) on chromosome 1 – 66% of variation for HIR ( Prigge et al., 2012) qhir 8 on chromosome 9 – 20% of variation for HIR Fine mapping of qhir1 revealed a region of 243 kb in length (Dong et al., 2013)

14. Maternal inducer lines – Contd.. GWAS study divided qhir1 into two region – qhir11 and qhir12 qhir 11 has a significant effect on haploid induction Phospolipase A gene within qhir11 is responsible for haploid induction and was found by three different groups independently Named the gene as MATRILINEAL (MTL ) ( Kelliher et al., 2017) / NOT LIKE DAD (NLD ) (Gilles et al., 2017) / ZmPLA1 (Liu et al., 2017) A 4-bp insertion in the fourth exon of MTL/PLA1/NLD is detected in inducer lines Accompanied with side effects such as kernel abortion and segregation distortion. Another main QTL found in maize qhir8 ( zmDMP gene) has the potential to double the haploid induction rate caused by qhir1

15. KMS ( Korichnevy Marker Saratovsky ) and ZMS, both derived from Stock6 ( Tyrnov & Zavalishina , 1984) WS14, derived by crossing W23ig and Stock6 KEMS ( Krasnador Embryo Marker Synthetic), MHI ( Moldovian Haploid Inducer) derived by KMS × RWS (Russian inducer KEMS + WS14), progeny of KEMS × WS14 hybridization ( Röber et al., 2005) UH400 derived from KEMS at the University of Hohenheim, Germany (Chang & Coe, 2009) CAUHOI developed by crossing Stock6 and Beijing High Oil Population at China Agricultural University (Li et al., 2009) PHI1, PHI2, PHI3, and PHI4 ( Procera Haploid Inducer) derived from Stock6 ( Rotarenco et al., 2010) Tropical Inducers TAIL series: TAIL7, TAIL8, TAIL9 TAIL8 x TAIL9 Second generation TAILS (CIM2GTAILs) 2GTAIL006, 2GTAIL007, 2GTAIL009, 2GTAIL102, 2GTAIL104 , 2GTAIL105, 2GTAIL109 , 2GTAIL114 2GTAIL009 × 2GTAIL006 L ist of Maize haploid inducers

16. Mechanism behind maternal haploid induction

17. Factors affecting the HIR Differential rates of HIR maybe attributed to Genetic background of the maternal parent Adaptation potential of germplasm Characteristics and potential of inducer lines Rainfall patterns GCA and SCA effects Meteorological conditions and Severity of biotic stresses

18. Identification of Haploids Chromosome counting Efficient and accurate Requires skill and time Flow Cytometry Quantified by fluorescence intensity emitted from fluorescence stained nuclei Most efficient based on accuracy and number of samples studied per day Costly Molecular markers: Microsatellite markers/SSRs SSR markers are used to distinguish between homozygous haploid and diploids in maize BNLG1223 primer was used by Couto et al., (2013) and validated by others Second most important step Direct strategies include conventional cytogenetic methods such as chromosome counting, flow cytometry, and molecular markers

19. Phenotypic markers Most common markers: R1-nj (R1-Navajo) , radicle length, radicle colour , plant vigor, pollen viability, and other seedling traits R-Navajo marker Used to differentiate haploids from diploids by producing endosperm and embryo pigmentation Preferred because of its xenia expression Dominant allele Pigmentation in crown region of kernel endosperm along with other dominant genes of anthocyanin pathway such as ( A1, A2, Bz1, C1 and C2) Practical limitations of R1-nj marker: Can be inhibited due to dominant anthocyanin inhibitor genes like C1-I Common in temperate flint lines Tropical breeding populations (F1 or F2) show segregation for the marker – only a proportion of haploids can be identified High number of false positives and false negatives Masking of the R1-nj expression in germplasm that expresses purple or red anthocyanin coloration on the pericarp

20. Additional color markers that are expressed in root and stem. Pl1 (Purple1) - Sunlight independent purple pigmentation in plant tissues B1 (Booster1) – Sunlight dependent purple pigmentation in most of the above-ground plant tissues Integration of Pl1 and B1 with R1-nj leads to coloration in seedling root or stem color. Diploids – purple roots and stems Haploids – No coloration Limitations of B1 and Pl1 system Many source population contain B1 and Pl1 genes, hence haploids may also express coloration Expression of these genes is affected by plant growth conditions like sunlight and temperature

21. Chromosome doubling of haploids Haploids are generally male sterile Restoring fertility can be achieved by artificial chromosomal doubling Artificial chromosome doubling can be achieved by treating the haploids with chemicals with anti-mitotic activity. E.g. Colchicine Colchicine binds to β -tubulin and prevents the formation of microtubule Standard protocol involves immersing the seedling in 0.04 – 0.06% colchicine solution with 0.5% Dimethyl Sulfoxide (DMSO) for 8-12 hrs.

22. Growing the haploid seeds in germination paper Cutting the root and shoot tips Colchicine tank Immerse the seedling in colchicine solution for 8-12 hrs Plant the seedlings in germination trays

23. D 1 cobs Selfing of D plants Transplanting

24. Application of Doubled Haploids in Maize Breeding Inbred Line development Improving genetic gains in recurrent selection Identification of QTLs Marker assisted breeding Unlocking new genetic variation Accelerated crop breeding Genomic selection Genetic enrichment Forward breeding

25. Inducer variety types Vast majority of available inducers are inbred lines Exceptions are ZMK1 inducer population, the German hybrid RWS/RWK-76, and a tropically-adapted hybrid inducer developed by CIMMYT Types of inducers : Inbred inducers Hybrid inducers Synthetic inducers

26. Inbred inducers Breed true and are uniform Simplified logistics Best choice if discrimination based on oil content is required Disadvantages: Inbreeding depression High susceptibility to diseases Less pollen production Weak performance in isolation fields

27. Hybrid inducers High pollen yield, disease tolerance and good performance in isolation fields Both parents must be fixed for the same marker traits Requires continuous production of both parental and hybrid seeds Development of parents with same set of marker traits is time consuming and challenging Synthetic inducers Combines the advantages of inbreds and hybrids Easier to maintain compared to hybrid inducers Longer pollen shedding periods but low pollen yield Need to be renewed periodically Fixation of markers in multiple parents is more complicated

28. Comparison of the three inducer variety types

29. Strategy for Inducer development Select a set of suitable inbreds Backcrossing - if the objective is to improve adaptation or agronomic traits

30. Selection steps in Inducer development

31. Traits important to haploid inducers

32. Maintenance of Inducers Requires frequent testing of HIR as natural selection acts strongly against this trait Contamination of stocks leads to reduction in HIR due to selective advantage of contaminant pollen Rouging of plants that do not show inducer characteristics In each cycle, 20 – 50 plants are selected and are crossed with a recessive tester and are selfed

33. Contd.. Test crosses can be used to determine HIR and selfed ears from plants with high HIR are forwarded Maintenance breeding requires selection of traits such as pollen production and plant vigor. To avoid loss of vigor due to selfing, sib mating can be employed

34. Challenges in Doubled - Haploid technology Low haploid fertility rates Maternal haploid induction is highly influenced by the nature and type of germplasm used HIR is influenced by climatic conditions Selection of haploid kernels after haploid induction Dependence on colchicine based chromosome doubling is a limiting factor in developing countries

35. Challenges in Doubled - Haploid technology Lower rates of chromosome doubling through colchicine treatment Establishment of a DH facility at an ideal location that is stress free and has sustainable resources Poor performance of induction crosses in under extreme weather conditions and poor agronomic conditions Low number of DH lines produced from single maize plant High cost of production

36. Status of DH technology in maize Adopted in many commercial maize breeding programs in Europe, North America and China Technology is made accessible by CIMMYT in collaboration with University of Hohenheim, Germany A maize doubled haploid facility is established in Kenya by CIMMYT in partnership with Kenya Agricultural and Livestock Research Organisation (KALRO) Development and release of improved maize hybrids with DH lines as parents in reported in Africa

37. . Case studies

38. Material and methods: Plant material – Five commercial single cross hybrids Generations under study: F 1 , F 2 Inducer: LI-ESALQ ((W23xStock6) x Tropical maize hybrid) Marker: R1-nj Methodology Haploid induction Seed selection based on anthocyanin expression Germination and chromosome doubling using colchicine Field experiment after chromosome doubling Studied false discovery rate: proportion of diploids present in the group selected as haploids Categorization of seeds: putative haploids, diploids and inhibitors

39. Parameters estimated: Haploid induction rate (HIR), inhibition seed rate (ISR) and diploid seed rate (DSR) Effectiveness of working steps was calculated for each generation as per se and relative percentage Genotyping: 7,430 SNPs Heterozygous loci remained in data and residual heterozygosity were considered as missing values Analysis of genome variation: DH of F 1 /F 2 generations were estimated using: Minor Allele Frequency: frequency at which the 2 nd most common allele occurs in a given population Polymorphism Information Content: ability to detect polymorphism in a population Nei’s gene diversity: probability of any two alleles at a locus, chosen at random from the population , are different to each other. Estimation of potential genetic variance Inbreeding effective population Response to selection

40. F 1 F 2

41. Number of individuals (N°), inbreeding effective population size (N e ), estimation of the potential genetic variance (E VG ), Nei’s genetic diversity (D G ), polymorphic information content (PIC), minor allele frequency (MAF), coefficient of inbreeding (F i ), and response to selection (RS). In parentheses are the maximum and minimum values.

42. PCA analysis grouped the germplasm into 4 groups Group 1: DKB390, Group 2: 30F53H and 2B587PW Group 3: STATUS VIPTERA Group 4: BM820 No separation of subgroups due to the F1 and F2 generations within each source germplasm. This indicated additional recombination in F 2 was not sufficient to create new subgroups

43. Conclusion For HIR, ISR, and DSR values, the best generation to induce haploids in tropical maize should be F2, given that the segregation of inhibitory alleles would enable greater haploid selection But F 2 needs additional one cycle of generation Low difference of HIR b/w F 1 and F 2 F2 showed lower values of N e , E VG , D G , MAF, PIC, and RS than the F 1 Doubled recombination in F2 DH lines was not sufficient to create new groups in population structure and kinship analyses, or increase the population parameter estimates when compared with F1 . Results showed that the induction of haploids must continue in the F1 generation, while F2 should be used in specific objectives of the breeding program . Genetic background of donor also affects the HIR

44. Materials and methods For QTL mapping: B73XMo17 DH population comprising of 250 lines (IBM Syn 10 DH) For GWAS : association panel of 310 inbreds were used The traits measured in this study included cob diameter (CD), cob weight (CW), ear diameter (ED), ear length (EL), ear row number (ERN ), ear weight (EW), 100-kernel weight (HKW), kernel number per row (KNR), and kernel weight per ear (KWE ). GWAS: 56,110 SNPs were analyzed QTL analysis: LOD threshold – 2.5

45. Trait Number of QTLs detected Range of PVE Cob diameter 13 (11 common) 3.85-10.09% Cob weight 10 (8 common) 4.00-7.96% Ear diameter 14 (11 common) 4.01-9.29% Ear length 18 (13 common) 3.82-8.04% Ear weight 11 (11 common) 4.23 to 7.39% Ear row number 41 (21 common QTLs) 3.36-12.13% Kernel number per row 12 (7 common) 3.82%-10.04% Hundred Kernel weight 18 (10 common) 3.97-9.04% Kernel weight/ear 10 (8 common) 3.34%-6.71%

46. GWAS loci Number of SNPs remained after filtering: 43,782 SNPs Significant SNPs detected – 138 Trait Number of significant SNPs detected Location of SNPs Cob diameter 25 1 , 2, 3,4,5,6,7 Cob weight 18 1,2,3,4,5,6,7,8,9 Ear diameter 16 1,2,3,4,5,6,7,8,9 Ear length 13 1,2,3,7,8,10 Ear weight 11 1,2,3,5,6,8,9,10 Ear row number 14 1,2,3,4,5,8,9 Kernel number per row 16 1,2,3,4,5,6,7,9,10 Hundred Kernel weight 21 1,2,3,4,5,6,7,8,9 Kernel weight/ear 7 2,5,6,7,8

47. Trait QTLs detected across multiple environments SNPs Cob diameter qCD2 SYN15262 Cob weight qCW8 PZE108063983 Ear diameter qED 10, qED6 (2) PZE103169263 Ear length qEL1-3, qEL5-3, qEL5-4, qEL3-2 SYN38130 and SYN21886) Ear weight Nil Nil Ear row number qERN4-2, qERN8-2 SYN18170 Kernel number per row qKNR1, qKNR8 PZE-104088869 and PZE107108370 Hundred Kernel weight qHKW3-2, qHKW5, qHKW10- 1, qHKW10-2, and qHKW10-3 Kernel weight/ear Nil PZE-106027614 QTLs and SNPs detected across multiple environments

48. Pleiotropic loci detected by GWAS and QTL QTLs that influence different traits and are fully/partially overlapped – pleiotropic QTL (p-QTL) SNPs associated with multiple traits – p-SNP 18 pQTLs were identified with strong linkage across different environments Three SNPs individually are related to multiple trait expression

49. Predicted candidate genes Eight SNPs were simultaneously detected by both QTL analysis and GWAS 52 gene models were identified flanking 300 kb of these SNPs 42 genes were annotated, 10 gene functions were unkown Cob diameter – 6 gene models Kernel number per row – 14 gene models Ear diameter – 13 gene models Ear length – 6 gene models Earn row number – 13 gene models

50. SNP SYN8062 is associated with SBP-transcription factor Validation of this SNP in IBM Syn10 DH population showed the ERN of lines with A-allele in SYN8062 was significantly larger than the lines with G-allele

51. Conclusion . Doubled-haploid technology has facilitated studies at the molecular and genomic level . Use of DH technology in breeding programs has increased genetic gains . Shortens the time needed for cultivar development, and also reduces the cost of breeding programs Combining DH with other technologies, such as MAS, induced mutagenesis, and transgenic technology, would accelerate crop improvement.

52. Future perspectives . Increased efficiency and reduced cost of DH production aids in seamless integration into breeding programs MAS of major loci and further phenotypic selection may enable development of inducers with high HIR Automation of haploid identification Inducers with multiple marker system for fool proof identification Optimization of chromosomal doubling protocols based on spontaneous doubling or non hazardous chemicals

53. THANK YOU