Translating the combinations of the widest possible sources of heritable variations efficiently into crop varieties whose increased yields, improved nutritional quality attributes and enhanced adaptations to abiotic and biotic stresses exceed those of the prior gains of the 20th century Green Revolution cannot be attained with a business-as-usual mindset. The current yield-centric breeding practices, of oftentimes weak breeding programs, whose objectives are largely conceived solely by the plant breeders, must evolve into participatory, multidisciplinary, and demand-driven programs that, underpinned by nurturing policy environments, make use of the most suitable scientific and technological tools to harness the potentials of PGRFA. Plant-breeding activities must perforce be re-oriented in order to have a reasonable chance of succeeding in the development of the envisaged portfolio of ‘smart’ crop varieties. We discuss some of the specific attributes that must characterize the result-oriented crop improvement programs of the 21st century.
Participatory plant breeding
Factoring in the perspectives of the growers and other stakeholders such as consumers, extensionists, vendors, industry, and rural cooperatives in the crop improvement endeavor of developing new varieties is known as Participatory Plant Breeding (PPB; ). The need for this paradigm in plant breeding is probably greatest in developing countries relative to the industrialized countries where market forces determine agricultural research and development (R&D) themes including plant-breeding objectives. By having farmers and other end-users involved in the development of varieties, feedback mechanisms are enhanced hence improving the relevance of the breeding activities to the needs of the growers. Farmers’ participation in plant breeding can be categorized under the three stages of design, testing, and diffusion . During the design stage, breeding goals are set and variability to be used created while at the testing stage, the breeding materials are evaluated and narrowed down to the few promising ones. The diffusion stage encompasses activities spanning varietal release, on-farm trials under farmer management and the identification of the mechanisms for the dissemination of the seeds and planting materials of the improved varieties.
Farmers, as the custodians of PGRFA, have over the several millennia of selecting from, improving, and exchanging local genetic diversity contributed immensely to the diversity of plants we grow. With the upsurge in the ready availability of modern crop varieties bred in research institutes, the roles of farmers in ensuring diversity and adding value to PGRFA have waned significantly. One effect of this shift is the precariously narrow genetic base of the modern crop varieties. The obvious threat that this poses to food security calls for the systematic re-integration of farmers’ knowledge and perspectives in the developing of modern crop varieties. PPB is a veritable and validated means for ensuring this. The International Treaty, through its Article 9, also requires of contracting parties the safeguarding of farmer’s rights to access and benefit from PGRFA. Those rights are not safeguarded when crop varieties that do not meet their food security and nutritional needs and/or do not enhance the resilience of their farming systems are all that are available to them.
In general, PPB facilitates the rapid and enthusiastic adoption of crop varieties . The related Participatory Varietal Selection (PVS) is a means for involving these stakeholders in breeding when elite materials are already available to select from and is relatively more rapid and cost-effective than the more resource-intensive PPB . Ashby  identified the impact pathways for PPB and PVS and concluded that their characteristic of producing more acceptable varieties and hence increasing adoption was the most compelling incentive for plant breeders to adopt this paradigm. Indeed, a CGIAR-wide review of plant breeding had recommended that PPB constitute ‘an organic part of each center’s breeding program’ .
Novel plant-breeding techniques
The incredible advances in biotechnology demonstrably hold great promise for crop improvement . For instance, molecular breeding, the integration of molecular biology techniques in plant breeding , through enhanced efficiencies, has great potentials for changing permanently the science and art of plant breeding. Molecular breeding encompasses both the use of distinguishing molecular profiles to select breeding materials and the applications of recombinant deoxyribonucleic acid (DNA) methods, that is genetic transformation, to add value to PGRFA. There are also a number of other emerging molecular biology-based techniques that hold promise for enhancing the efficiency levels of plant breeding activities. We provide some overview of the use of these technologies and techniques in developing novel crop varieties.
The increasingly available rapid, efficient, high throughput, and cost-effective molecular biology tools for identifying the sources, and tracing the inheritance, of desired traits are revolutionizing the management of PGRFA in general and plant breeding in particular. Advances in molecular biology, including the ever cheaper sequencing of whole genomes, have resulted in the availability of significant amounts of information on, and hence tools for assaying, the totality of an individual’s genetic make-up, that is the genome; this is known as genomics. The related proteomics (the study of proteins) and metabolomics (the study of metabolites), made possible by an ever growing volume of publicly accessible DNA, gene, and protein sequence information, are also novel ways for investigating the heredity of traits. Equally significant, advances in bioinformatics and computational molecular biology which are facilitated greatly by the novel sophisticated and powerful information technology platforms for storing and analyzing the huge volumes of data generated through these molecular biology strategies, permit the making of valid inferences in the molecular characterization of germplasm, assessments of genetic diversity and for the selections of breeding materials.
The ability to use appropriate molecular approaches in identifying genome segments that discriminate between individuals (that is molecular markers) and to apply statistical algorithms in identifying precisely where these ‘landmarks’ are located on the genome has changed plant breeding permanently and will be key in developing the ‘smart’ crops of the 21st century. Molecular markers are now demonstrably the tools of choice for tracing the inheritance of target regions of genomes in breeding materials, a plant breeding methodology known as marker-assisted (or -aided) selection (MAS).
MAS entails the use of environment-neutral molecular markers to trace the inheritance of genes, and hence the trait(s) they control, in a breeding program with or without phenotypic selection . The utility of MAS is greatest for genes whose effects are difficult, time-consuming, or otherwise expensive to evaluate in a population. This may be on account of the phenotypic effects being evident only at maturity, low heritabilities, the absence of the particular stress factor being bred for or as a result of confounding environmental influences on the trait.
The use of MAS is relatively straightforward in breeding for qualitative monogenic traits with clear-cut differences between phenotypes, such as disease resistance in plants, as the genetic mapping of the associated marker results in the mapping of the trait also and vice versa. For quantitative traits, the validation of the trait-marker association through large-scale field experimentations and statistical methods in order to more precisely identify the target genome segments, that is quantitative trait loci (QTL), is additionally required [76, 77]. In general, once the marker-trait association has been verifiably established, the transmission of trait genes from parent to offspring is monitored by querying segregating materials for closely linked markers using suitably designed marker-assisted backcrossing, for instance. The utility of MAS in breeding for polygenic traits can also be derived in gene pyramiding, that is the accumulation of two or more genes, say for disease and pest resistance, which seems feasible only with this method .
It has been demonstrated that consistently, MAS, either as a standalone strategy or in combination with phenotyping, significantly reduces the number of generations for evaluating segregating breeding materials and generally increases efficiency levels [2, 74, 75, 78–93]. Indeed, it has been demonstrated that MAS permits a seven-fold increase in data handling and ultimately halves the time required for breeding a new crop variety . Nonetheless, the cost-benefit analysis for adopting MAS relative to phenotypic selection is always a critical consideration that must be borne in mind in devising breeding strategies especially for developing countries.
Already routinely applied in the private sector breeding companies, such as the multinational companies, Monsanto ; Pioneer Hi-Bred  and Syngenta , MAS is yet to take hold in public crop improvement programs mostly on account of high set-up costs and intellectual property rights (IPR) restrictions. This implies that public sector plant breeding is clearly missing out on this singularly promising opportunity to innovate. Thro et al.  captured the immense expectations riding on the investments in plant genomics in relation to crop improvement in characterizing plant breeding as the ‘translator’ of knowledge into improved crop varieties. Public sector plant breeding is yet to assume this ‘translator’ role in the new dispensation of crop improvement that must be ‘knowledge-intensive’.
An encouraging trend, though, is the progressive decline in the cost and the concomitant improvement in the high throughput applicability of molecular biology assays and equipment. It is logical to assume that at some point in the near future, set-up costs would be generally affordable and routine assays sufficiently efficient  as to permit wide adoption of MAS in the public sector. The continued successful use of MAS in the private sector is providing the much needed validation and proof of concept for this paradigm. This is critically important as capacity for this breeding methodology will be critical in handling the large populations of new breeding materials to be produced from pre-breeding activities using non-adapted genetic resources, for instance. The Integrated Breeding Platform (IBP) of the Generation Challenge Program of the CGIAR  is an example of multi-stakeholder efforts to extend the use of MAS to developing elite varieties of food security crops in developing countries.
Recombinant DNA technology, involving the use of molecules containing DNA sequences derived from more than one source to create novel genetic variation, has become an important crop improvement option. This is known as genetic modification (or transformation) with the new variants referred to as transgenics or simply genetically modified organisms (GMOs). The procedures involve the incorporation of exogenous DNA or ribonucleic acid (RNA) sequences, using either biolistics or vectors, into the genome of the recipient organism which, as a result, expresses novel and agronomically useful traits. Though transgenic varieties of only four crops, maize, soybean, canola, and cotton, harboring two transformation events, that is herbicide tolerance and insect resistance or their combinations, have been grown commercially since the first approvals in 1996, James  estimated that there had been a 94-fold increase in hectarage in the 16 years of the commercialization of genetically modified (GM) crops (from 1.7 million hectares in 1996 to 160 million hectares in 2011). Grown in 29 countries (19 developing and 10 industrial), the author estimated the value of the GMO seed market at US$13.2 billion in 2011 while the produce for GM maize, soybean, and cotton were valued in excess of US$160 billion for the same year.
In spite of the low numbers of commercial GM crops and the transformation events that confer the modified agronomic traits, four and two, respectively, the development and deployment of GM crops signal a trend in crop improvement that can no longer be ignored. This is more so as approvals for the importation of GM crops and release to the environment had been approved in 31 other countries . Tester and Langridge  pointed out that, though the major contributions to crop improvement for this decade will be non-GM, the production and evaluation of GM crops remained an actively researched theme with only political and bioethical considerations (both driven mostly by public negative perceptions for the technology) constituting the main hindrances to wider access to the technology by growers in more countries.
Technically, the drawbacks to more widespread development of GM varieties include the lack of efficient genotype-independent regeneration systems for most crops. Also, the lingering technical difficulties with the stacking of transformation events severely limits the utility of genetic transformation in breeding for polygenic straits such as resistance to the abiotic stresses, for example salinity and drought, being caused by climate change and variations. However, the successful stacking of genes conferring insect resistance and herbicide tolerance  is indicative of progress in addressing this constraint. Also, research efforts must target the increasing of the range of agronomic traits being improved through this method; the two transformation events in commercial varieties are simply inadequate for GM technology to become a dominant crop improvement method.
Probably the most limiting of all factors, however, is the associated intellectual property rights (IPR) protections that restrict access to the technology. Such IPR regimes have made GMOs remain the exclusive preserve of multinational plant breeding and seed companies in developed countries that effectively use patents to restrict access to several technologies relevant to the R&D efforts for the production of the transgenic crops. These constraints must be addressed in order that this technology be used fully in realizing its possible contributions to the development of the ‘smart’ crop varieties of this century. With GMO crops currently grown in developing countries, for example about 60 million hectares in South America in 2011 and with millions of small holder farmers cultivating transgenic cotton in both India and China [100–102], it is plausible to expect that the IPR regimes will be changing in the future. Another hindrance to wider adoption of the GM technology is the absence of biosafety regulatory frameworks as specified by the Cartagena Protocol on Biosafety to the Convention on Biological Diversity  in many countries.
Efforts to address the constraints that impede both the use of the GM technology in R&D and the cultivation of GMOs have been significant as well. For instance, the African Agricultural Technology Foundation (AATF; ), based in Nairobi, Kenya, is acquiring and deploying proprietary agricultural technologies in sub-Saharan Africa. In one instance, AATF obtained ‘a royalty-free, nonexclusive license to Monsanto technology, a Bacillus thuringiensis (Bt) gene (cry-1Ab)’ which is being used in the development of cowpea varieties with resistance to the cowpea pod borer . Similarly, the US-based Public Sector Intellectual Property Resource for Agriculture (PIPRA; ), assists ‘foundations, not-for-profit organizations, universities, international aid agencies, and governments’ in dealing with IPR issues in order to enable access to proprietary technologies. Also, Cambia, an Australian private, non-profit research institute, publishes relevant patents, white papers, and provides tutorials as means ‘to provide technical solutions that empower local innovators to develop new agricultural solutions’ . The activities of these organizations underscore the seriousness of the impediments that IPR protections pose for innovations in agriculture and the countervailing efforts to extend the reach of the technologies and applications especially into the public goods and commons R&D domains.
Emerging biotechnology techniques of relevance to plant breeding
The integration of biotechnologies into crop improvement is a very dynamic field of endeavor that is changing continually. A snapshot of the status of emerging technologies is provided by Lusser et al.  in response to a request by the European Commission ‘to provide information on the state of adoption and possible economic impact of new plant breeding techniques’. The authors identified eight new such techniques and concluded that the new varieties ensuing from these techniques might be released within 3 years. These new techniques and their features are:
Zinc finger nuclease (ZFN): Single mutations or short indels are generated or new genes are introduced into pre-determined target sites of the genome
Oligonucleotide directed mutagenesis (ODM): Targeted mutations of one or a few nucleotides are induced
Cisgenesis and intragenesis: GMOs are produced by the insertion of hereditary materials derived from the species itself or from a cross-compatible species and are contiguous and unchanged (cisgenesis) or the inserted DNA may be a new combination of DNA fragments but must still be from the species itself or from a cross-compatible species
RNA-dependent DNA methylation (RdDM): Still being refined, modified gene expressions are epigenetic with the new phenotypes inherited only over a few generations
Grafting (on GM rootstock): Desired improvements are achieved by the grafting of non-transgenic scions onto GM rootstock
Reverse breeding: A combination of recombinant DNA techniques and cell biology procedures is used to generate suitable transgene-free homozygous parental lines rapidly for reconstituting elite heterozygous genotypes
Agro-infiltration: Used mostly in research settings, for example to study plant-pathogen interaction in living tissues, to select parental lines or to evaluate the efficacy of transgenes, a liquid suspension of Agrobacterium sp. containing the desired gene(s) is used to infiltrate plant tissues, mostly leaves, so that the genes are locally and transiently expressed at high levels
Synthetic genomics: Large functional DNA molecules that are synthesized without any natural templates are used for constructing viable minimal genomes which can serve as platforms for the biochemical production of chemicals such as biofuels and pharmaceuticals
Lusser et al.  concluded that ODM, cisgenesis/intragenesis, and agro-infiltration were the most commonly used techniques with the crops developed using them having reached the commercial development phase. On the other hand, the ZFN technology, RdDM, grafting on GM rootstocks, and reverse breeding were the less used techniques in breeding. The authors further projected that the first commercial products derived from these technologies that will be released for production would be herbicide resistant oilseed rape and maize using ODM and fungal resistant potatoes, drought tolerant maize, scab resistant apples, and potatoes with reduced amylose content developed using cisgenesis and/or intragenesis.
The clearly identified needs for the further fine-tuning of technical impediments to the routine adoptions and use of these new techniques notwithstanding, it would appear that policy regulations that are expensive to comply with and public perceptions, rather than the ability to innovate, are holding back the unleashing of the incredible advances of science and technology in crop improvement. Considering that Blakeney  opined that ‘the right to patent agricultural innovations is increasingly located within a political context’, it is plausible that the magnitude of the worsening threats to global food security may ultimately serve as the critical inducement for policy-makers, interest groups, and leaders of thought and industries to unravel the thorny issues that constrain the scope of the integration of biotechnology into crop improvement.
High throughput phenotypic evaluations
The selections of few promising individuals out of large populations of segregating materials can be a very daunting task. With MAS, the volume of assays that can be carried out and data points generated per unit time has increased substantially. For the workflow to be wholly efficient, the assessments of the phenotypes must also keep pace with high throughput molecular assays. Indeed, for molecular data used in breeding to be reliable, the corresponding phenotypic data for which inferences are made, must also be accurate . Phenomics, the study of phenomes - the sum total of an individual’s phenotype is the term that describes the novel high throughput measurements of the physical and chemical attributes of an organism. Somewhat imprecisely named in this seeming analogy to genomics, it is defined by Houle et al.  as ‘the acquisition of high-dimensional phenotypic data on an organism-wide scale’. High throughput imaging of parts of a living plant, for example roots and leaves, using thermal infra-red, near infra-red, fluorescence, and even magnetic resonance imaging permit non-destructive physiological, morphological, and biochemical assays as means for dissecting complex traits such as drought and salinity tolerances into their component traits [112, 113]. Though significant technical challenges, such as data management, still require addressing, phenomics facilities are increasingly being set up with a number of them providing high throughput phenotyping services to requestors. These new facilities include the High Resolution Plant Phenomics Centre in Canberra and the Plant Accelerator in Adelaide, both in Australia ; LemnaTec in Wuerselen  and Jülich Plant Phenotyping Centre in Jülich  both in Germany; and Ecotron  and Ecophysiology Laboratory of Plant Under Environmental Stress (LEPSE; ) both in in Montpellier, France. In Canada, there is the The Biotron Experimental Climate Change Research Centre in London, Ontario . The high set-up costs and technical know-how may impede the access of developing countries to such platforms for some considerable time.
Overarching policy environment for the PGRFA management continuum
The benefits of value addition to PGRFA, that is improved crop varieties that meet the needs of the growers, can be derived sustainably, especially for the most at-risk food insecure countries in the developing world, only with the comprehensive strengthening of, and forging of linkages between, the three components of the PGRFA value chain: (1) conservation; (2) plant breeding; and (3) the delivery of high quality seeds and planting materials to growers. This is the ‘PGRFA continuum’ , the seamless dovetailing of the three components, as distinct from targeting the strengthening of any of the three in isolation. Based on the cohesion in this value chain - that characterizes the activities of private sector commercial breeding companies and the PGRFA management of some emerging countries such as Brazil, China, and India  - it is logical to conclude that the real value of crop germplasm lies in its use in plant breeding. Pragmatically also, the efforts invested in breeding come to naught if there is no effective delivery system for the seeds and planting materials underscoring therefore the need to interlock all three components.
The successful implementation of the Second GPA  also envisages the adoption of this continuum approach. The 18 priority activities (Box 1) of the GPA provide a most practical template for countries for concerted interventions at the three components of the PGRFA value chain. These PAs are subdivided into four main themes: in-situ conservation and management; ex-situ conservation; sustainable use; and building sustainable institutional and human capacities.
The sustainable use of PGRFA encompasses activities relating to direct utilization of PGRFA by farmers and to their uses in crop improvement. The International Treaty, especially in its Article 6, equally requires of contracting parties not only to conserve their genetic resources but to use them (for value addition) and to deliver the improved varieties efficiently. FAO  opined that ‘any weakness in this continuum truncates the value chain and effectively scuttles all the efforts to grow the most suitable crop varieties’. It is in this vein that FAO and partners are working with developing countries to articulate National PGRFA Strategies for institutionalizing the continuum approach to managing PGRFA . The strategy identifies priority crops and relevant stakeholders; prescribes time-bound action plans across the continuum and enunciates governance mechanisms and means for monitoring implementation. Nurturing policy environments, especially those that enable countries adopt the continuum approach to the management of PGRFA, are critically important for reaping the most sustainable benefits from PGRFA, namely, the improved crop varieties. FAO’s normative activities provide support for the implementations of the International Treaty and the Second GPA and for developing the necessary policies, and legislations as means for attaining this goal.
The reorientation of crop improvement in order to be responsive to the drivers of food insecurity, especially in developing and emerging economies, will require a wider range of partnerships beyond the traditional National Agricultural Research and Extension Systems (NARES). FAO  reported the prevailing trend whereby the private sector (multinational and local commercial plant breeding and seed companies) is increasingly developing and deploying elite crop varieties especially in instances where markets, favorable policy regimes, and legal frameworks that spur investments are in place. In tandem, public investment in crop breeding programs is contracting implying therefore that the breeding and dissemination of elite varieties of crops that fall outside of the business remit of the private sector could, as is increasingly the case, be neglected to the detriment of food security. Equally important is the role of non-governmental organizations and myriad civil society actors in the provision of agricultural extension services in developing countries. These bourgeoning dynamics must influence the articulation of policies and the building of collaborations and wide-ranging partnerships. For such partnerships to succeed, local knowledge must be integrated just as relevant private and public sector entities including the NARES, centers of the CGIAR, and regional R&D networks are assembled. The safeguarding of intellectual property rights, including plant variety protection, and the respect of patents are means for attracting private sector investments. Public-private partnerships, for example the ongoing joint activities between Syngenta and public African NARES [122, 123], are particularly important for technology transfer, a critical vehicle for increasing the access of developing countries to novel biotechnologies that impact on crop improvement, for instance. On the other hand, public sector investments in food security must be ensured as the private sector, especially in developing countries, do not cater for all crops that are important for food security. Partnerships must also be cross-sectoral, for instance between ministries responsible for the environment, science and technology, commerce, education, and the ministry of agriculture. This ensures access to the full spectrum of PGRFA that may be needed for value addition while also ensuring a means for delivering the planting materials efficiently to the growers in gainful manners.