ABSTRACT

The traditional method of hybrid identification and genetic diversity evaluation based on differences in range of expressions of morphological and agronomical characters has been routinely employed in the assessment of oil palm germplasm and breeding populations at the Nigerian Institute of Oil Palm Research (NIFOR). Considering the limitations of this conventional method and the advantages of molecular markers in complementing conventional breeding methods, this study was conducted to determine the legitimacy of NIFOR oil palm progenies and to assess the genetic variations and relationships existing among the breeding populations using microsatellite (SSR) markers. Ten microsatellite markers were used to screen 226 oil palm samples which included 215 samples from NIFOR and 11 other samples belonging to the Malaysian Palm Oil Board (MPOB) advanced breeding lines and germplasm materials. Results obtained revealed that out of 200 F1 oil palm progeny derived from 11 of the 15 parents evaluated, 57% (114) were true-to-type and 43%

(86)    were contaminated. Almost all the contamination or illegitimacy detected was due to pollination errors. High genetic diversity was detected among the 15 NIFOR parents with the NIFOR tenera parents recording the highest number of alleles (5), rare alleles (17), and gene diversity (0.650) when compared to the Deli dura NIFOR and NIFOR dura parents. With

reference to the MPOB breeding materials, the various oil palm sources showed significant and large value of genetic differentiation (FST= 0.177, P = 0.001) due to variations within the sources of parental materials. Rogers’ dissimilarity coefficient matrix displayed two main clusters, one separating MPOB Madagascar accessions from the rest of the samples. Principal co-ordinates analysis (PCoA) showed that the NIFOR breeding parents clustered closely with the MPOB Nigeria and Angola derived materials indicating a common origin of mainland genotypes. A comparative assessment of molecular and morphological methods of describing genetic relationships in the NIFOR oil palm progenies showed that SSR analysis recorded the highest level of polymorphism (100%) in all the progenies. Although the correlation between agronomic and genetic distance matrices was very low and insignificant (r = 0.2989), both matrices discriminated the progenies effectively into two groups as per their agronomic performance and pedigree or origin, respectively.

1

INTRODUCTION

The oil palm, Elaeis guineensis Jacq is a diploid (2n = 32) monocotyledonous and perennial crop of the humid tropics. Fossil, archaeological, historic, and linguistic evidence indicate that oil palm originated in Africa (Hartley, 1977 and1988; Corley and Tinker, 2003). Fossil pollen similar to oil palm was extracted from Miocene sediments of Nigeria (Zeven, 1964), while Raynaud et al. (1996) reported oil palm pollen found in lake sediments of south-east Cameroon (Hartley, 1977; Corley and Tinker, 2003). Sowunmi (1999) discovered oil palm nut shells in a rainforest site and speculated that an increase in late Holocene times (5000 years ago) connotes the beginning of oil palm cultivation and its importance in the subsistence economy of Africa. Historical records of African origin of oil palm were traced to the major landmarks of Portuguese and English exploration and trade in Africa (Rees 1965). The short and direct translation of the West African vernacular names of oil palm is strong linguistic evidence supporting a West African origin of oil palm (Zeven, 1965; Hartley, 1988). A high concentration of natural/semi-natural groves estimated at about 2.1 million hectares occur in Nigeria and studies have shown that the Nigerian groves have the highest level of polymorphism with respect to the number of alleles, indicating that Nigeria is likely to be the centre of distribution for oil palm (Maizura et al., 2001; Rajanaidu, 2002; Maizura et al., 2006; Bakoumé et al., 2015).

Oil palm is cultivated for the oil that is extracted from the mesocarp and the kernel. It produces more than five times oil/year/hectare of any annual oil crop (Basri-Wahid et al., 2005). Palm oil is the most valuable natural oil in the diets of Nigerians both as crude red palm oil and as refined oil (olein). It has played a very significant role in the socio-economic and political life of the people of Nigeria. With the recognition of the economic potential of the crop, both as dietary and industrial fats, concerted efforts towards the genetic improvement of the plant and management practices started at the turn of the last century in Nigeria (Okwuagwu and Ataga, 1992). Today, this genetic improvement strategy using traditional breeding methods has been very successful. The most spectacular achievements are the development of E. guineensis hybrids with one or a combination of the following attributes (i) early bearing, (ii) short stem, (iii) drought tolerance, and (iv) high and stable yield (Okwuagwu and Ataga, 1985; Okwuagwu, 1989; Okwuagwu et al., 2001 and 2005). In spite of these compelling results, oil palm yield potential is yet to be fully realized coupled with the slow and expensive procedure involved in the breeding and selection programme at NIFOR. Should this trend continue, genetic gain expected through conventional breeding

2

would not be able to keep pace with the increasing domestic demand for palm oil, let alone the competition from other vegetable oils and fats. Against this backdrop, the Federal Government of Nigeria has made a firm commitment to restore Nigeria’s agriculture notably oil palm agriculture to its past eminent position in the economy. A memorandum of understanding was then signed with the Nigerian Institute for Oil Palm Research (NIFOR) on the implementation of the oil palm transformation value chain. Subsequently, NIFOR is expected to annually produce 9 million improved tenera planting materials which will be delivered to recommended nursery operators in the various oil palm growing states of the country for distribution to farmers. This is in a bid to meet the growing domestic demand for palm oil, create employment for the youths and possibly re-enter the international market for vegetable oil and fats (Obibuzor, Personal Comm.).

To accommodate the anticipated increase in the demand for palm oil by the year 2020, the area under cultivation will need to increase from the current 470,000 hectares in 2013 (Oil World 2014) with additional 200,000 hectares. Presently, the commercial variety (tenera) yields 20 -25 mt of fresh fruit bunch ha-1 yr-1 and 3 – 3.5 t oil ha-1 yr-1 (Okwuagwu et al., 2001). The tenera shows great variations in yield with the best yielding about 40% more than the average. Steady breeding progress has been made, with yield being doubled between 1950 (2.5 – 5.0 mt of fresh fruit bunch ha-1 yr-1) and year 2000 (20 – 25 mt of fresh fruit bunch ha-1 yr-1). Accordingly, future breeding progress will rely increasingly on family selection and progeny testing, which require a high degree of legitimacy of both the parents and the progenies.

A notable threat to the oil palm industry in Nigeria is the conformity of planting materials. Oil palm is naturally out-breeding and controlled pollination is not always effective. Anomalous genetic segregation and contamination with unexpected fruit forms do occur from time to time (Corley, 2005). It is therefore crucial to select true hybrids/progenies in a crossing programme for breeding and production of planting materials. This situation is further aggravated by the impurity of planting materials as a result of high patronage of illegal sprouted seeds/seedling producers who pose as NIFOR agents to give credence to their illegal dealing. It has become very necessary to safeguard the industry by using markers by which NIFOR elite tenera hybrids could be characterized and distinguished at the seedling stage.

The traditional method of hybrid identification based on morphological traits is influenced by environmental factors (Murphy et al., 1996; Chakravarthi and Naravaneni, 2006) and most importantly lack means to identify hybrids at the seed or seedling stage. In

3

fact, oil palm has to be grown several years (3 to 4 years) before producing fresh fruit bunches to confirm authenticity of the hybrid. A reliable method for hybrid identification in oil palm at an early stage is therefore essential.

Another feature of oil palm breeding programmes is the very narrow genetic base of some of the ancestral populations (Rosenquist, 1986; Corley, 2005). This can lead to reduction in productivity and increased susceptibility to diseases and pests, thus threatening the profitability of the crop. Also, the reduction in genetic variability and increase in homozygosity is often followed by inbreeding depression due to the limited number of parent palms for subsequent selection cycle. Consequently, selection/genetic progress in inbred populations become slow and limited than would be expected based on heritability and selection intensity. Genetic diversity evaluation based on agro-morphological information which has been routinely used in the assessment of oil palm germplasm and breeding populations in NIFOR is no longer sufficient. Agro-morphological markers are often unreliable and ambiguous due to (i) long juvenile phase, (ii) confounding effects of developmental stage of the plant, (iii) low polymorphism due to predominating one environment-one phenotype relation, (iv) long term field evaluation, and (v) vulnerability to environmental effects. An additional and objective measure of genetic variation which would allow the use of germplasm materials with precision to widen the genetic base of breeding populations is necessary. Until recently, the only information linking different breeding populations was pedigree, inadequate in many cases and sometimes incorrect (Mayes et al., 2000).

The genetic improvement in oil palm bunch and oil yields has generally been achieved through conventional breeding in NIFOR. Unfortunately, the method is a slow (≥10 years) and expensive process fraught with the limitations of long generation interval of approximately 19 years for phenotypic evaluation of testcrosses and inter-crossing of the best palms to initiate a new cycle, large planting areas for evaluation of testcrosses and enormous manpower requirement to manage and conduct oil palm breeding trials (Soh et al., 1990; Rance et al., 2001). The application of DNA/molecular marker technology in the NIFOR oil palm breeding programme would not only reduce the time taken for conventional breeding but ensure greater precision in the production of planting materials (Mohan et al., 1997; Mayes et al., 2008).

Molecular markers provide an opportunity to characterize genotypes, and to determine genetic diversity of natural or breeding populations more precisely than agro-morphological markers (Brumlop and Finckh, 2010). Furthermore, they have been considered

4

an indispensable tool to complement reciprocal recurrent selection scheme; allowing monitoring of the genetic variability of the original and selected populations, identification of contaminants and selection of families to be recombined to maximize heterosis effect ( Pinto et al., 2003; Tardin et al., 2007).

Among the available molecular markers, microsatellites or simple sequence repeats (SSRs) are considered as ideal genetic marker for plant genetic and breeding studies due to (i) their high polymorphism, (ii) co-dominant inheritance, reproducibility and (iii) abundance throughout the genome when compared to restriction fragment length polymorphic DNA (RFLP), random amplified polymorphic DNA (RAPD), and amplified fragment length polymorphism (AFLP) (Powell et al., 1996; Gupta et al., 1999; Philips and Vasil, 2001; McCouch et al., 2002; Mohammadi and Prasanna, 2003; Schlotterer, 2004; Varshney et al. 2004; Bindu et al., 2004; Singh and Cheah, 2005; Feng et al., 2009). In addition, they are readily transferable, and easily assayed using polymerase chain reaction (PCR) (Peakall et al., 1998). Multiplexing PCR and multiloading PCR products on single gels also reduces the workload for studies requiring a large number of samples (Saghai-Maroof et al., 1994).

SSRs has proven its advantage and suitability in oil palm population genetics and breeding studies (Billotte et al., 2001; Billotte et al., 2005; Bakoumé et al., 2007; Singh et al., 2008; Arias et al., 2012; Bakoumé et al., 2015), varietal identification (Rajinder et al., 2007; Norziha et al., 2008; Thawaro and Te-chato, 2010; Bakoume et al., 2011; Hama-Ali et al., 2014), pedigree analysis, genome mapping and quantitative trait loci (QTL) detection for molecular marker-assisted selection (Billotte et al., 2010; Montoya et al., 2013; Ting et al., 2013).

Considerable amount of molecular markers studies on oil palm have already been carried out in Malaysia (MPOB), France (CIRAD), Colombia (CENIPALMA), to name a few. Unfortunately, not as much works have emerged from the proposed centre of distribution of the species i.e. Nigeria as well as from the West Africa oil palm belt. Essentially, authors have either attempted evaluation of within species genetic variability, molecular marker-trait association (QTLs) or establishment of phylogenetic relationship using populations of some African origin while others have tried to characterize the trait potentials of the varieties at molecular level.

Despite the importance of information on genetic diversity in improvement and efficient conservation of crops, no studies, to date, have been specifically conducted on the genetic diversity of the current NIFOR breeding populations using SSR. Therefore, the objectives of the present study were: