Abstract

Macadamia is a nut tree native to Australian rainforest. Due to the small wild populations and the economic interest, studies assessing genetic information are fundamental for use and conservation of this species. Therefore, the present study aimed to assess the structure and genetic diversity of 28 cultivars of macadamia originated from United States, Australia and Brazil, introduced or developed by the Agronomic Institute’s breeding program, using 29 new microsatellite loci. The microsatellite loci showed high genetic diversity (3.65 alleles per locus on average). Our results also suggested the existence of no genetic structure between cultivars, regardless of their geographic origin. The modified Rogers’ distances between cultivars ranged from 0.227 to 0.671. Despite the lack of information about genealogy, the cultivars from this Brazilian germplasm showed moderate genetic diversity, so they can be used as parents in future crosses.

Keywords:
Microsatellite; nut tree; cross-amplification; molecular marker; polymorphism

INTRODUCTION

Macadamia integrifolia Maiden & Betche (Proteaceae) is a nut tree native to Australia and occurs wildly in the southeast of Queensland and northern New South Wales States (Topp et al. 2019Topp BL, Nock CJ, Hardner CM, Alam M, O’Connor KM2019 Macadamia (Macadamia spp.) breeding. In Al-Khayri JM, Jain S and Johnson D (eds) Advances in plant breeding strategies: nut and beverage crops. Springer International Publishing, Cham, p. 221-251). Currently, only 1% of the Australian rainforest is preserved and restricted into fragments (Sobierajski 2019Sobierajski GR2019 Botânica, morfologia e exigências edafoclimáticas. In Pimentel L and Borém A (eds) Macadâmia: do plantio à colheita. UFV, Viçosa, p. 20-35). This scenario emphasizes the importance of well-characterized ex situ germplasms (Busanello et al. 2020Busanello C, Venske E, Stafen CF, Pedrolo AM, Luz VK, Pedron T, Paniz FP, Batista BL, Magalhães Júnior AM, Oliveira AC, Pegoraro C2020 Is the genetic variability of elite rice in southern Brazil really disappearing? Crop Breeding and Applied Biotechnology 20:e262620214, Guan et al. 2020Guan C, Zhang Y, Zhang P, Chachar S, Wang R, Du X, Yang Y2020 Germplasm conservation, molecular identity and morphological characterization of persimmon (Diospyros kaki Thunb.) in the NFGP of China. Scientia Horticulturae 272:109490, Carena 2021Carena M2021 Germplasm enhancement and cultivar development: the need for sustainable breeding. Crop Breeding and Applied Biotechnology 21:e385621). The genetic variation existing in germplasms can provide genes for resistance to pests, diseases and adaptation to climate changes (Top.. et al. 2019Topp BL, Nock CJ, Hardner CM, Alam M, O’Connor KM2019 Macadamia (Macadamia spp.) breeding. In Al-Khayri JM, Jain S and Johnson D (eds) Advances in plant breeding strategies: nut and beverage crops. Springer International Publishing, Cham, p. 221-251, Carena 2021Carena M2021 Germplasm enhancement and cultivar development: the need for sustainable breeding. Crop Breeding and Applied Biotechnology 21:e385621)..

Despite the Australian origin of macadamia, its first breeding program was established by the Hawaii Agricultural Experiment Station (HAES), Hawaii/United States (Aradhya et al. 1998Aradhya MK, Yee LK, Zee FT, Manshardt RM1998 Genetic variability in macadamia. Genetic Resources and Crop Evolution 45:19-32, Hardner et al. 2019Hardner CM, Silva JC, Williams E, Meyers N, McConchie C2019 Breeding new cultivars for the Australian macadamia industry. HortScience 54:621-628, Nock et al. 2019Nock CJ, Hardner CM, Montenegro JD, Termizi AAA, Hayashi S, Playford J, Edwards D, Batley J2019 Wild origins of macadamia domestication identified through intraspecific chloroplast genome sequencing. Frontiers in Plant Science 10:1-15, Topp et al. 2019Topp BL, Nock CJ, Hardner CM, Alam M, O’Connor KM2019 Macadamia (Macadamia spp.) breeding. In Al-Khayri JM, Jain S and Johnson D (eds) Advances in plant breeding strategies: nut and beverage crops. Springer International Publishing, Cham, p. 221-251). Since the Hawaiian germplasm was established from seeds of a small number of trees, the narrow genetic base may lead to limitations for cultivar development (Aradhya et al. 1998, Steiger et al. 2003Steiger DL, Moore PH, Zee F, Liu Z, Ming R2003 Genetic relationship of macadamia cultivars and species reveled by AFLP markers. Euphytica 132:269-277). This fact highlights the importance of germplasm characterization worldwide. Such characterization may shed light on the relationship between cultivars and wild individuals (Aradhya et al. 1998Aradhya MK, Yee LK, Zee FT, Manshardt RM1998 Genetic variability in macadamia. Genetic Resources and Crop Evolution 45:19-32, Steiger et al. 2003Steiger DL, Moore PH, Zee F, Liu Z, Ming R2003 Genetic relationship of macadamia cultivars and species reveled by AFLP markers. Euphytica 132:269-277).

Several genetic marker techniques have been used as part of strategies for conservation and genetic breeding of macadamia, e.g., Amplified Fragment Length Polymorphisms - AFLP (Steiger et al. 2003Steiger DL, Moore PH, Zee F, Liu Z, Ming R2003 Genetic relationship of macadamia cultivars and species reveled by AFLP markers. Euphytica 132:269-277), and Randomly amplified DNA fingerprinting - RAF (Peace et al. 2005Peace CP, Allan P, Vithanage V, Turnbull CGN, Carroll BJ2005 Genetic relationships among macadamia varieties grow in South Africa as assessed by RAF markers. South African Journal of Plant and Soil 22:71-75). The common point of these studies is the use of non-specific genetic markers, genotyping HAES and Australian cultivars, and wild species. Aradhya et al. (1998Aradhya MK, Yee LK, Zee FT, Manshardt RM1998 Genetic variability in macadamia. Genetic Resources and Crop Evolution 45:19-32) also carried out a molecular characterization study, but they considered only four cultivars released by the Agronomic Institute of Campinas (IAC) in Brazil. Schmidt et al. (2006Schmidt AL, Scott L, Lowe AJ2006 Isolation and characterization of microsatellite loci from macadamia. Molecular Ecology Notes 6:1060-1063) and Nock et al. (2014Nock CJ, Elphinstone MS, Ablett G, Kawamata A, Hancock W, Hardner CM, King GJ2014 Whole genome shotgun sequences for microsatellite discovery and application in cultivated and wild macadamia (Proteaceae). Applications in Plant Science 2:1300089) developed microsatellite (SSR) specific markers for macadamia. These last two studies considered no cultivar from IAC germplasm.

In Brazil, seeds of macadamia were introduced in 1931 from cultivars developed by HAES. The only macadamia breeding program in Brazil was established by the IAC in 1940 (Toledo Piza et al. 2019Toledo Piza PLB, Piza IM, Almeida Neto JTP2019 A cultura. In Pimentel L and Borém A (eds) Macadâmia: do plantio à colheita. UFV, Viçosa , p. 9-19). Up to this date, this program released 15 cultivars with high productivity and large nut size. Unfortunately, there is a lack of information on the pedigree and genetic diversity of these cultivars. Despite this drawback, Brazil ranked 7^th in the world standings with production of 6,000 tons of nut in shell in 2015, corresponding to 3.75% of world production (Toledo Piza et al. 2019Toledo Piza PLB, Piza IM, Almeida Neto JTP2019 A cultura. In Pimentel L and Borém A (eds) Macadâmia: do plantio à colheita. UFV, Viçosa , p. 9-19). The major producing regions are situated in Southeastern Brazil. However, the agro-climatic zoning for macadamia in Brazil indicates that regions situated in central-west and south Brazil may also be regarded as fit zones for establishment of new plantations (Sobierajski 2019Sobierajski GR2019 Botânica, morfologia e exigências edafoclimáticas. In Pimentel L and Borém A (eds) Macadâmia: do plantio à colheita. UFV, Viçosa, p. 20-35).

A key point to macadamia production is the development of cultivars well-adapted to local conditions. For instance, new macadamia cultivars have increased producers’ profits in Australia (Topp et al. 2017Topp B, Hardner C, Alam M, Akinsanmi O, O'Connor K, Mai T, Russell D2017 Opportunities and challenges in macadamia breeding. In International Macadamia Research Symposium. USDA, Big Island, p. 18, Kern et al. 2022Kern S, Santos B, Topp B, Cave R, Bignell G, Mulo S, Hardner C2022 Using choice analysis of growers’ preferences to prioritize breeding traits in horticultural tree crops: a macadamia case study. Scientia Horticulturae 294:110766). Therefore, it is essential to characterize the germplasm from a molecular-genetic point of view. Considering the genetic diversity as a key-factor for assessing differences among individuals of the same species (He and Li 2020He T, Li C2020 Harness the power of genomic selection and the potential of germplasm in crop breeding for global food security in the era with rapid climate change. Crop Journal 8:688-700, Carena 2021Carena M2021 Germplasm enhancement and cultivar development: the need for sustainable breeding. Crop Breeding and Applied Biotechnology 21:e385621), the hypotheses of the present study are that the Brazilian macadamia germplasm shows genetic diversity among cultivars, which may be used in the IAC’s macadamia breeding program. Aiming at generating new information on macadamia germplasm in Brazil, we developed twenty-nine M. integrifolia specific SSR loci and assessed the genetic diversity and structure of 28 cultivars of macadamia. Additionally, these loci were tested for cross-amplifications in Grevillea robusta and G. banksii. These two species were added to this study because they are from the same botanical family as Macadamia sp., are economic relevant and there is no specific SSR marker developed for G. banksii.

MATERIAL AND METHODS

Genomic DNA from macadamia cultivar IAC 9-20, which has high yield and nut quality, was extracted from fresh leaves using Cetyltrimethylammonium bromide (CTAB) extraction method (Inglis et al. 2018Inglis PW, Pappas MCR, Resende LV, Grattapaglia D2018 Fast and inexpensive protocols for consistent extraction of high-quality DNA and RNA from challenging plant and fungal samples for high throughput SNP genotyping and sequencing applications. PLoS One 13:e0206085). This DNA was used to construct the enriched genomic library according to the methodology described in Billotte et al. (1999Billotte N, Lagoda PJR, Risterucci AM, Baurens FC1999 Microsatellite enriched libraries: applied methodology for the development of SSR markers in tropical crops. Fruits 54:277-288). This methodology includes the following stages: DNA digestion with the RsaI enzyme; Rsa21 and Rsa25 adapters linkage; amplification, purification and selection of fragments containing microsatellite regions; cloning of fragments by pGEM-T vector in Escherichia coli (Promega, WI, USA); and sequencing using SP6 and T7 primers promoters. Sequences from the recombinant colonies were assembled and edited using Seqman software (DNA Star 2000)DNA Star2000 Laser Gene: expert sequence analysis software. Available at <Available at https://www.dnastar.com/ >. Accessed on January 26, 2012.
https://www.dnastar.com/... . The Simple Sequence Repeat Identification Tool (SSRIT) was used to identify SSRs present in all non-redundant sequences (Temnykh et al. 2001Temnykh S, DeClerck G, Lukashova A, Lipovich L, Catinhour S, McCouch S2001 Computational and experimental analysis of microsatellites in rice (Oryza sativa L.): frequency, length variation, transposon associations and genetic marker potential. Genome Resources 11:1441-1452). The Primer Select software (DNA Star 2000) was used to design the primer pairs. The DNA sequences were submitted to GenBank (Table 1), from National Center for Biotechnology Information - NCBI (http://www.ncbi.nlm.nih.gov).

Twenty-eight genotypes of macadamia from the IAC germplasm and producers in Dois Córregos, São Paulo State, Brazil, were assessed with regard to their SSR polymorphisms. The DNA sampling was composed of 11 cultivars developed by the Agronomic Institute, five cultivars developed by Hawaii Agricultural Experiment Station, four cultivars developed by the Hidden Valley Plantation (Australia), and two producers’ selections (Aloha and Flor Roxa - Supplemental material 1). These cultivars were chosen because of their high acceptance by the Brazilian producers or because they show features of interest for breeding program. In addition, six genotypes showing morphological divergence from those of the IAC germplasm were included to confirm their genetic origin. The cross amplification was tested in two species of silver oak (Grevillea robusta and G. banksii), collected from the University of Campinas (UNICAMP) campus, for all polymorphic loci. The DNA extraction of all samples was carried out using Cetyltrimethylammonium bromide (CTAB) method (Inglis et al. 2018Inglis PW, Pappas MCR, Resende LV, Grattapaglia D2018 Fast and inexpensive protocols for consistent extraction of high-quality DNA and RNA from challenging plant and fungal samples for high throughput SNP genotyping and sequencing applications. PLoS One 13:e0206085).

PCR amplifications were performed in 25 µL total volume containing 10 ng of genomic DNA, 0.8 µM forward and reverse primers, 100 µM of each dNTP, 1.5 mM MgCl₂, 10 mM Tris-HCl, 50 mM KCl, and 0.5 U Taq DNA Polymerase (Invitrogen, CA, USA). The PTC-200 thermal cycler (MJ Research, Waltham, MA/USA) was used for PCR amplifications with touchdown cycling program (temperature ranging from 45 to 60 °C). Amplification products were genotyped using electrophoresis on 6% denaturing polyacrylamide gels, 1x TBE buffer, and visualized using silver staining 0.2% (Creste et al. 2001Creste S, Tulmann Neto A, Figueira A2001 Detection of single sequence repeat polymorphisms in denaturing polyacrylamide sequencing gels by silver staining. Plant Molecular Biology Reporter 19:299-306).

The SSR loci characterization was based on the following genetic parameters: expected heterozygosity (He), observed heterozygosity (Ho), and polymorphism information content (PIC; as described in Faria et al. 2022Faria SV, Zuffo LT, Rezende WM, Caixeta DG, Pereira HD, Azevedo CF, DeLima RO2022 Phenotypic and molecular characterization of a set of tropical maize inbred lines from a public breeding program in Brazil. BMC Genomics 23:54-70). These parameters were calculated using FSTAT program (Goudet 1995Goudet J1995 FSTAT (Version 1.2): a computer program to calculate F-Statistics. Journal of Heredity 86:485-486). The fixation index (f) and the tests of Hard-Weinberg Equilibrium (HWE) were performed using the Tools for Population Genetic Analysis (TFPGA; Miller 1997Miller MP1997 Tools for Population Genetic Analysis (TFPGA) 1.3.: a windows program for the analysis of allozyme and molecular population genetic data. Available at <Available at http://bioweb.usu.edumpmbio >. Accessed on January 26, 2012.
http://bioweb.usu.edumpmbio... ). The linkage disequilibrium (LD) was analyzed using the GDA software (Lewis and Zaykin 2002Lewis P, Zaykin D2002 Genetic Data Analysis (GDA): computer program for the analysis of allelic data. Available at <Available at http://phylogeny.uconn.edu/software/ >. Accessed on January 26, 2012.
http://phylogeny.uconn.edu/software/... ). The presence of null alleles was tested using the Micro-checker software (Oosterhout et al. 2004Oosterhout C, Hutchinson B, Wills D, Shipley P2004 Micro-Checker: Software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes 4:535-538).

The genetic structure of IAC macadamia germplasm was investigated by Structure software (Pritchard et al. 2000Pritchard JK, Stephens Mand Donnelly P2000 Inference of population structure using multilocus genotype data. Genetics 155:945-959), considering the probability of each genotype being allocated in K clusters. The parameters established to run the software were: 20 independent iterations for each K value; K ranging from 1 to 10; and 10,000 Markov Chain Monte Carlo (MCMC; 10,000 burn-in) steps. The selected model of ancestry was the admixture, which assumes that the genomes may have a proportion of more than one population (Grünwald et al. 2017Grünwald NJ, Everhart SE, Knaus BJ, Kamvar ZN2017 Best practices for population genetic analyses. Phytopathology 107:1000-1010). The model of allele frequency adopted in this study was correlated among populations. The estimate of the most-likely number of K subpopulations was obtained by ΔK (Evanno et al. 2005Evanno G, Regnaut S, Goudet J2005 Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14:2611-2620), from Structure Selector software (Li and Liu 2017Li YL, Liu JX2017 Structure Selector: A web-based software to select and visualize the optimal number of clusters using multiple methods. Molecular Ecology Resources 18:176-177).

The genetic distances between pairs of macadamia cultivars and Grevillea sp. were estimated by the modified Rogers’ distances, and the unweighted pair group method, with arithmetic mean (UPGMA) as agglomeration method (Silva 2016Silva AR2016 Métodos de análise multivariada em R. FEALQ, Piracicaba, 167p), was used to construct the dendrogram. To verify the accuracy of these estimates, 1,000 bootstraps per locus were performed. The cut-off was based on the K-means algorithm from ‘factoextra’ R package (Kassambara and Mundt 2019Kassambara A, Mundt F2019 Package ‘factoextra’. Available at <Available at https://cran.r-project.org/web/packages/factoextra/factoextra.pdf >. Accessed on May 5, 2019.
https://cran.r-project.org/web/packages/... ).

The Principal Component Analysis (PCA) was applied to describe similarities among the cultivars of the macadamia germplasm and Grevillea sp. The modified Rogers’ distance, bootstraps, dendrogram and PCA were obtained using the ‘cluster’ (Maechler et al. 2019Maechler M, Rousseeuw P, Struyf A, Hubert M, Hornik K, Studer M, Roudier P, Gonzalez J, Kozlowski K, Schubert E, Murphy K2019 Package ‘cluster’. Available at <Available at https://cran.r-project.org/web/packages/cluster/cluster.pdf >. Accessed on May 5, 2019.
https://cran.r-project.org/web/packages/... ), ‘MASS’ (Ripley et al. 2019Ripley B, Venables B, Bates DM, Hornik K, Gebhardt A, Firth D2019 ‘MASS’ Package. Available at <Available at https://cran.r-project.org/web/packages/MASS/MASS.pdf >. Accessed on May 31, 2019.
https://cran.r-project.org/web/packages/... ) and ‘ggplot2’ (Wickham et al. 2019Wickham H, Chang W, Henry L, Pedersen TL, Takahashi K, Wilke K, Woo K2019 Package "ggplot2". Available at <Available at https://cran.r-project.org/web/packages/ggplot2/ggplot2.pdf >. Accessed on May 31, 2019
https://cran.r-project.org/web/packages/... ) packages in R-software environment (R Core Team 2019R Core Team2019 R: A language and environment for statistical computing. Available at <Available at http://www.r-project.org >. Accessed on May 31, 2019.
http://www.r-project.org... ).

RESULTS AND DISCUSSION

Eighty-two clones containing microsatellite sequences were considered suitable for primer design. Thus, 51 primer pairs were designed, of which 29 were polymorphic (Table 1). Twenty-five microsatellite loci were cross-amplified in G. robusta and G. banksii. The number of macadamia’s alleles per locus ranged from two to nine, with an average value of 3.65 alleles per locus (Table 2). Schmidt et al. (2006Schmidt AL, Scott L, Lowe AJ2006 Isolation and characterization of microsatellite loci from macadamia. Molecular Ecology Notes 6:1060-1063) developed 33 SSR loci for macadamia and obtained from one to 15 alleles per locus, while five loci were monomorphic. Nock et al. (2014Nock CJ, Elphinstone MS, Ablett G, Kawamata A, Hancock W, Hardner CM, King GJ2014 Whole genome shotgun sequences for microsatellite discovery and application in cultivated and wild macadamia (Proteaceae). Applications in Plant Science 2:1300089) developed 12 SSR loci for macadamia, which showed between two and nine alleles per locus, with average values equal to 5.92 (M. integrifolia and hybrids) and 4.92 (M. tetraphylla) alleles. This study obtained allele frequencies ranging from 0.017 to 0.983. 106 alleles were detected, of which 25 were classified as rare alleles (frequency ≤ 0.05), representing 23.6% of total alleles. The loci MInt011 and MInt065 showed the largest number of alleles (nine), of which five and three were classified as rare, respectively. Only one locus (MInt093) showed allele frequency over 0.95, indicating a high risk of fixation.

The values for He and Ho ranged from 0.034 to 0.836 and 0.034 to 0.967, respectively (Table 3). The average values for He and Ho were 0.46 and 0.43, respectively, showing small reduction of heterozygosity when the Hard-Weinberg Equilibrium (HWE) is expected. Of the 29 loci developed in this study, only two showed significant deviations of HWE after Bonferroni correction (p < 0.005): MInt011 and MInt035. Despite the selection pressure resulting from the breeding program, this result suggests that, in terms of mean of the loci, the cultivars meet the Hardy-Weinberg Equilibrium. The recent macadamia domestication may explain this result, since the current cultivars are genetically close to their wild relatives (Topp et al. 2019Topp BL, Nock CJ, Hardner CM, Alam M, O’Connor KM2019 Macadamia (Macadamia spp.) breeding. In Al-Khayri JM, Jain S and Johnson D (eds) Advances in plant breeding strategies: nut and beverage crops. Springer International Publishing, Cham, p. 221-251). Schmidt et al. (2006Schmidt AL, Scott L, Lowe AJ2006 Isolation and characterization of microsatellite loci from macadamia. Molecular Ecology Notes 6:1060-1063) found He and Ho values ranging from 0.027 to 0.875 and 0.027 to 0.882, respectively, with 20 loci showing significant HWE deviations. Nock et al. (2014Nock CJ, Elphinstone MS, Ablett G, Kawamata A, Hancock W, Hardner CM, King GJ2014 Whole genome shotgun sequences for microsatellite discovery and application in cultivated and wild macadamia (Proteaceae). Applications in Plant Science 2:1300089) found a mean He of 0.626 (from 0.165 to 0.8 loci) and mean Ho of 0.571 (from 0.091 to 0.909) for M. integrifolia and hybrids, whereas for M. tetraphylla, they found mean He of 0.632 (from 0.133 to 0.847) and Ho of 0.573 (from 0.143 to 0.857).

The average PIC value among loci was 0.40, with values ranging from 0.03 to 0.80 (Table 3). Schmidt et al. (2006Schmidt AL, Scott L, Lowe AJ2006 Isolation and characterization of microsatellite loci from macadamia. Molecular Ecology Notes 6:1060-1063) obtained an average value for PIC equal to 0.480 (from 0.026 to 0.848). The loci MInt009, MInt015, MInt035 and MInt079 were significant for the presence of null alleles. Null alleles may occur when mutations in the prime annealing site prevent the efficiency of at least one primer and this leads to a failure of amplification during the PCR reaction (Rico et al. 2017Rico C, Cuesta JA, Drake P, Macpherson E, Bernatchez L, Marie AD2017 Null alleles are ubiquitous at microsatellite loci in the wedge clam (Donax trunculus). PeerJ 5:e3188). The presence of null alleles may be the result of underestimation of heterozygotes (Rico et al. 2017Rico C, Cuesta JA, Drake P, Macpherson E, Bernatchez L, Marie AD2017 Null alleles are ubiquitous at microsatellite loci in the wedge clam (Donax trunculus). PeerJ 5:e3188). Schmidt et al. (2006Schmidt AL, Scott L, Lowe AJ2006 Isolation and characterization of microsatellite loci from macadamia. Molecular Ecology Notes 6:1060-1063) showed evidence of the presence of null alleles in locus MinµS53, which had no PCR products in 48% of the samples. These authors recommended that this locus, despite the evidence of null alleles, should be used as codominant marker in genome mapping if the parents were contrasting for presence/absence of mark. No significant linkage disequilibrium was detected among all loci, as indicated by the chi-squared test (p < 0.001).

The results of the Structure software suggested the existence of no genetic structure among the cultivars. Although the Evanno et al. (2005Evanno G, Regnaut S, Goudet J2005 Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14:2611-2620) method showed the highest ∆K for K = 2 (Figure 1A), a more detailed analysis of the estimated average and standard deviation of the likelihood of the distinct models (K = 1to K = 10) indicated that the model with K = 1 is also a likely one (absence of structure). This latter model showed the second highest likelihood and lowest standard deviation among the 20 repetitions (Figure 1B). As pointed out by Soares et al. (2020Soares SD, Bandeira LF, Ribeiro SB, Telles MPC, Silva JA, Borges CT, Coelho ASG, Novaes E2020 Genetic diversity in populations of African mahogany (Khaya grandioliola C. DC.) introduced in Brazil. Genetics and Molecular Biology 43:e20180162), the Evanno et al. (2005Evanno G, Regnaut S, Goudet J2005 Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14:2611-2620) ∆K method may not be able to assess the lack of genetic structure. As described below, this lack of genetic structure between the cultivars is also suggested by all other methods used in this study.

Figure 1
Genetic structure of IAC’s macadamia germplasm: (A) ΔK values for the different number of clusters (K), calculated according to Evanno et al. (2005Evanno G, Regnaut S, Goudet J2005 Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14:2611-2620); (B) mean and standard deviation of the likelihoods of the different models obtained with K ranging from 1 to 10, indicating K = 1 as the most likely model.

The modified Rogers’ distances ranged from 0.227 (between ‘HAES 741’ and ‘HAES 660’) to 0.671 (between ‘IAC Campinas B a’ and the species G. banksii; Figure 2). The dendrogram classified the genotypes into two groups: an external formed by Grevillea species and another corresponding to all macadamia cultivars. One may expect that cultivars originated from distinct geographic regions would form different clusters, belonging to distinct subpopulations. However, this expectation could not be observed in Figure 2. According to the modified Rogers’ distances, all macadamia cultivars formed a single cluster, apart from the Grevillea species. In other words, all macadamia cultivars, regardless of their geographic origin, did not show genetic divergence capable of placing them into distinct subpopulations. This result is in line with Aradhya et al. (1998Aradhya MK, Yee LK, Zee FT, Manshardt RM1998 Genetic variability in macadamia. Genetic Resources and Crop Evolution 45:19-32), who observed that cultivars released by IAC were dispersed among Hawaiian, Californian and Australian cultivars. The results depicted in Figure 2 also indicate that the genotypes ‘IAC 4-20 a’ and ‘IAC 4-20 b’ showed a relatively high genetic divergence with respect to ‘IAC 4-20’ cultivar, which is from the IAC macadamia germplasm. This may suggest a nomenclature mistake. A similar feature is also observed in ‘IAC 9-20’ and ‘IAC Campinas B’ and their corresponding genotypes. Several authors have described historical (Hardner 2016Hardner CM2016 Macadamia domestication in Hawaii. Genetic Resources and Crop Evolution 63:1411-1430), phenotypical (Alam et al. 2019Alam M, Hardner CM, Nock C, O'Connor K, Topp B2019 Historical and molecular evidence of genetic identity of macadamia cultivars HAES741 and HAES660. HortScience 54:616-620) and molecular genetic (Aradhya et al. 1998, Steiger et al. 2003Steiger DL, Moore PH, Zee F, Liu Z, Ming R2003 Genetic relationship of macadamia cultivars and species reveled by AFLP markers. Euphytica 132:269-277, Peace et al. 2005Peace CP, Allan P, Vithanage V, Turnbull CGN, Carroll BJ2005 Genetic relationships among macadamia varieties grow in South Africa as assessed by RAF markers. South African Journal of Plant and Soil 22:71-75, Alam et al. 2019) similarity between ‘HAES 660’ and ‘HAES 741’. This similarity between these two cultivars indicate that they were obtained from an autogamous full-sib progeny (Alam et al. 2019Alam M, Hardner CM, Nock C, O'Connor K, Topp B2019 Historical and molecular evidence of genetic identity of macadamia cultivars HAES741 and HAES660. HortScience 54:616-620). The results of the PCA (Figure 3) agree with those depicted in Figure 2. As the cluster analysis, the PCA also indicated that G. robusta and G. banksii formed an external group and the macadamia cultivars formed a single cloud of points regardless of their geographic origin.

Figure 2
Dendrogram based on modified Rogers’ distances showing the genetic relationship among 28 Macadamia integrifolia cultivars and the species Grevillea robusta and G. banksii (cut-off based on K-means algorithm; Kassambara and Mundt 2019Kassambara A, Mundt F2019 Package ‘factoextra’. Available at <Available at https://cran.r-project.org/web/packages/factoextra/factoextra.pdf >. Accessed on May 5, 2019.
https://cran.r-project.org/web/packages/... ).

Figure 3
Principal Component Analysis with the respective proportions of variation explained by the first three principal components. Cultivars: 1 - IAC 2-23; 2 - IAC 4-20; 3 - IAC 9-20; 4 - IAC 11-18; 5 - IAC 4-10; 6 - IAC 4-8; 7 - IAC 11-8; 8 - IAC 8-17; 9 - IAC 1-21; 10 - IAC 4-12 B; 11 - IAC 4-20 a; 12 - HAES 816; 13 - HAES 344; 14 - HAES 741; 15 - HAES 660; 16 - HAES 333; 17 - IAC Campinas B; 18 - IAC Campinas B a; 19 - IAC 4-20 b; 20 - IAC 9-20 a; 21 - IAC 9-20 b; 22 - IAC Campinas B b; 23 - Flor Roxa; 24 - A-4; 25 - A-16; 26 - A-29; 27 - A- 38; 28 - Aloha; 29 - Grevillea robusta; 30 - G. banksii.

CONCLUSIONS

The Agronomic Institute’s Macadamia Germplasm shows no genetic structure between cultivars. The 28 cultivars evaluated in this study were classified into single population regardless of their distinct geographic origin.

Despite the lack of information about the historical genealogy, the cultivars from the IAC Germplasm have moderate genetic diversity, so they can be used as parents for new crosses.

The microsatellite markers developed in this study show high polymorphic level among the loci, which may be suitable for genetic studies, including those addressing genetic conservation, determination of the degree of relatedness among individuals or groups of accessions, and breeding.

The cross-species amplification data suggest that microsatellites developed for M. integrifolia are useful for genetics studies of G. robusta and G. banksii.

ACKNOWLEDGMENTS

To São Paulo State Research Support Foundation (FAPESP, Brazil, Process 2011/06141-0) and to Agronomic Institute (project SGP 66) for the financial support. To Brazilian Agricultural Research Corporation (EMBRAPA, Brazil) and to National Council for Scientific and Technological Development (CNPq, Brazil, Process 307616/2019-3) for providing the grant for the first and the second author, respectively. Supplemental materials are available from corresponding author.

REFERENCES

Publication Dates

History