**The Role of Vegetable Genetic Resources in Nutrition Security and Vegetable Breeding**

#### **Andreas W. Ebert**

World Vegetable Center, 60 Yi-Min Liao, Shanhua, Tainan 74151, Taiwan; ebert.andreas6@gmail.com

Received: 8 May 2020; Accepted: 9 June 2020; Published: 11 June 2020

**Abstract:** Malnutrition, comprising undernutrition, micronutrient deficiency, and overnutrition, is more widespread than hunger per se and affects most nations around the globe. The diversity and the quality of food produced and consumed are decisive factors when addressing the triple burden of malnutrition. In this context, fruit, vegetables, and nuts are increasingly moving into the focus of the nutrition community. Agricultural policies and investments in agriculture are predominantly focused on staple food production, neglecting the economic and nutritional potential of fruit and vegetables. While global vegetables are well represented in genebanks around the globe, this is much less the case for traditional vegetables. Collecting efforts in hotspots of vegetable diversity in Africa and Asia are required to conserve this germplasm before it is being replaced by modern varieties. Home gardens, community seedbanks, and variety introduction through vegetable seed kits are ways how genebanks can link with the farming community to strengthen the informal seed sector. This in turn may result in more diverse production systems and increased consumption of fruit and vegetables. In the formal seed sector, vegetable breeders need access to a wide diversity of genetic resources, predominantly farmers' varieties, landraces, and crop wild relatives. Genomics-assisted breeding is increasingly facilitating the introgression of favorable genes and quantitative trait loci (QTLs) with complex inheritance patterns from wild species into cultigens. This will lead to wider use of crop wild relatives in the development of resilient cultivars.

**Keywords:** malnutrition; food security; vegetables; genetic resources; ex situ conservation; home gardens; community seedbanks; variety introduction; vegetable breeding

#### **1. Introduction**

The Green Revolution had positive impacts on poverty reduction and lower food prices [1]. These effects were mostly driven by crop germplasm improvement programs of the CGIAR centers, resulting in impressive yield increases per hectare. The CGIAR is a global agricultural research partnership for a food secure future that was founded in the 1970s. The new, improved varieties were taken up by national agricultural agencies for adoption and broad-scale dissemination among the farming communities. From 1960 to 2000, yields across all developing countries increased 208% for wheat, 109% for rice, 157% for maize, 78% for potatoes, and 36% for cassava [2]. Estimates of the impact of crop germplasm improvement alone indicate average productivity gains, across all regions of the world, of 1.0% per annum for wheat, 0.8% for rice, 0.7% for maize, 0.6% for millets, and 0.5% for sorghum [3].

The historic success of the Green Revolution in terms of yield gains, together with lower food prices, ensured adequate quantities of staple cereal grain, thereby drastically reducing the problem of famine. However, after years of steady decline, the trend in world hunger reverted in 2015 and is now slowly increasing. About 820 million people or almost 11% of the global population suffered from chronic undernourishment in 2018 [4]. This underscores the immense challenge of achieving the Zero Hunger target of the 2030 Agenda for Sustainable Development adopted by the United Nations (UN) in 2015, which comprises 17 Sustainable Development Goals (SDGs). As stated by the United Nations

Development Program [5], the SDGs were adopted as a "call for global action to end poverty, protect the planet, and ensure that all people enjoy peace and prosperity by 2030". SDG 2—Zero Hunger calls for the eradication of hunger and all forms of malnutrition, with targets for doubling agricultural productivity and incomes of small-scale food producers (SDG 2.3), ensuring sustainable food systems (SDG 2.4), and maintaining genetic diversity (SDG 2.5).

Global demand for agricultural crops is predicted to increase until 2050 due to population growth, greater per capita purchasing power which translates into higher meat and dairy consumption, and biofuel use [6–9]. Analyses undertaken by Tilmann et al. [8] revealed that per capita demand for crop calories or protein is, since 1960, similarly dependent on per capita real income within and among seven economic groups. Those groups ranged from highest (group A) to lowest (group G) national average per capita inflation-adjusted gross domestic product (GDP). By 2050, global crop production should be 60% to 110% higher than that of 2005/2007 to feed the expected world population of more than nine billion [8–10]. Depenbusch and Klasen [11] estimated a 61% increase in calorie requirements between 2010 and 2100 if weight per age–sex group would remain stable. Considering the current trend toward increased human height and weight at a given height, expressed in terms of the average body mass index (BMI), this might add another 18.7 percentage points to the projected 61% increase in future national and global caloric requirements. Thus, an additional significant increase in food production would be required to satisfy demand.

Malnutrition, comprising undernutrition, micronutrient deficiency, and overnutrition, is more widespread than hunger per se and affects most nations around the globe. According to the 2018 Global Nutrition Report [12], children under five years of age face multiple burdens: stunting (low height for a child's weight) affects 150.8 million, wasting (low weight for a child's height) affects 50.5 million, and overweight (BMI at or above the 85th percentile and below the 95th percentile for children and teens of the same age and sex [13]) affects 38.3 million. Each year, 20 million babies are born of low birth weight (less than 2500 g) and are 20 times more likely to die in infancy than heavier babies. In 2016, 131 million children 5–9 years old, 207 million adolescents, and two billion adults were overweight [4]. About one-third of overweight adolescents and adults, and 44% of overweight children aged 5–9 were obese. Thus, overweight and obesity are at an alarming rate globally. Obesity is defined as a BMI at or above the 95th percentile for children and teens of the same age and sex [13]. In adults, BMI values of 30.0 or higher fall within the obese range.

Malnutrition is the single most important risk factor for disease. Diet-related diseases such as diabetes, cardiovascular disease, hypertension, stroke, cancer, and obesity are escalating at a global level. In 2016, an estimated 40.5 million (71%) of the 56.9 million deaths worldwide were caused by non-communicable diseases (NCDs) [14]. Approximately 32.2 million NCD deaths (80%) were attributable to cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes, whereas the remaining 8.3 million (20%) were caused by other NCDs. These figures illustrate the seriousness of diet-related diseases for the healthcare sector. NCDs now pose a greater risk to morbidity and mortality than unsafe sex, alcohol, and drug and tobacco use combined [15].

To address SDG 2, it becomes obvious that there is a clear need to reorient agricultural production from a mere increase in food quantity toward delivery of more diverse and more nutritious food produced in a sustainable manner [15,16]. As incomes rise and food consumption patterns change, overnutrition from imbalanced diets also becomes a matter for concern, both in developed and in developing countries. The diversity and quality of food produced and consumed is a decisive factor when addressing the triple burden of malnutrition, i.e., undernutrition, micronutrient deficiency, and overnutrition.

Although fruit and vegetables are usually mentioned jointly when addressing malnutrition, this paper focuses mainly on the role of vegetables for nutrition security, their global production, and the current ex situ conservation of both global and traditional vegetables. The need for additional conservation efforts in hotspots of vegetable diversity to safeguard valuable germplasm for current and future breeding efforts is addressed. Ways in which genebanks can link with the farming communities to strengthen the informal seed sector and to introduce genetic diversity at the farm and household level are briefly described. In the formal seed sector, vegetable breeders need access to a wide diversity of genetic resources, predominantly farmers' varieties, landraces, and crop wild relatives (CWR). The use of conventional and modern breeding tools to make efficient use of CWR for the development of resilient cultivars is briefly discussed.

#### **2. The Importance of Vegetables for Nutrition Security**

The rise of modern agriculture is generally considered to have led to a decline in the number of plant species upon which humans depend for food [17,18]. This decline particularly affected the wild, semidomesticated, and cultivated traditional vegetables and fruits, spices, and other food plants that supplemented staple crops with micronutrients, essential for a healthy diet. Those crops/species also strengthened food security historically during failures of the main crops [19]. More recent analyses of food consumption trends, regionally and globally, revealed increases in the diversity of plants contributing to diets locally in developing countries [20]. These changes are driven by rising income, urbanization, trade liberalization, transnational food corporations, and food industry marketing and retailing through supermarkets. Consumers are now offered more diversity and convenience but at the same time cheaper, less healthy, and highly processed food with high content of fats and sugars has been made more easily accessible and affordable and is, therefore, in high demand [21]. These developments led to a Westernization transition of local diets with a preference for energy-dense foods, such as animal products, plant oils, and sugars over cereals, pulses, and vegetables, with a general preference for more global crop plants over locally produced traditional crops [22,23].

A recent study examined changes in the richness, abundance, and composition of crop commodities in the food supplies of 142 countries (comprising 98% of the global population from 1961 to 2009 [24]. During this 50-year period, national food supplies expanded quantitatively in terms of food calories, protein, fat, and weight, and significant quantities of food consumed were derived from energy-dense food sources, known to contribute to malnutrition. Because of these global food consumption trends, national food supplies became more similar in composition globally, thanks to a constant flow of a range of truly global cereal and oil crops, while the supply of local traditional crops declined. The growing reliance on the supply of those global food crops is leading to stronger interdependence among countries in their food supplies, underlying genetic resources, and nutritional priorities.

It is increasingly recognized that the global food system must shift its focus from food quantity toward dietary quality and health and environmental outcomes [15,16]. Fruit, vegetables, and nuts are increasingly entering into the focus of the nutrition community for their potential in combating the triple burden of malnutrition [25]. Traditional, underutilized crops, especially those which are locally available and culturally acceptable, are ideally placed to play a much greater role in contributing to improved nutrition and health [26], in line with the strategy proposed by the British Royal Society: "The preferred strategy to eliminate hidden hunger will always involve strategies to increase the diversity of diet with increased access to fruit and vegetables" [27]. The World Health Organization (WHO) recommends a population-wide daily intake of 400 g of edible fruit and vegetables for the prevention of NCDs, as well as for the prevention and alleviation of several micronutrient deficiencies [28]. This WHO recommendation translates to roughly five portions per day. It should be noted that potatoes, sweet potatoes, cassava, and other starchy roots are not considered as fruits or vegetables.

People able to enjoy more diverse diets, in general, also have better nutrition and health. A recent study analyzing data of a health survey in Great Britain revealed that there is a robust inverse association between fruit and vegetable consumption and mortality [29]. A study of Helen Keller International's integrated household food production program targeting women in Burkina Faso found significant improvements in several child nutrition indicators, including reductions in anemia, wasting, and diarrhea among young children [30]. The underlying agriculture production activities of this integrated program included input distribution (seeds, saplings, chicks, and gardening tools) and agricultural training. Household food production activities focused primarily on micronutrient-rich vegetables and fruits, eggs, and chicks.

Increasing production and consumption of fruit and vegetables is an obvious pathway to improve dietary diversity and enhance nutrition security, especially in the case of diets that are dominated by high-energy foods with low levels of micronutrients [29]. Such a move, coupled with a transition to diets higher in plant-based protein, will also help protect valuable habitats such as the Amazon rainforest and help meet the SDGs [31]. However, several studies suggested that current and projected fruit and vegetable production levels will fail to meet healthy consumption levels [31,32]. Following age-specific recommendations, only 40 countries representing 36% of the global population had adequate availability of fruit and vegetables by 2015 [33]. In many food-insecure countries in sub-Saharan Africa and South Asia, average fruit and vegetable consumption is well below WHO-recommended levels, with 10 countries not even meeting 30% of the recommended intake levels [34].

Similarly, projections to 2050 indicate that between 0.8 and 1.9 billion people living in countries in sub-Saharan Africa might need to contend with average fruit and vegetable availability below 400 g/person per day [33]. This estimate does not yet take food waste into consideration. Under a high-food-waste scenario, these authors predict that 139 countries representing 5.6 billion people will not be able to provide enough fruit and vegetables to the population by 2050—an increase of 1.5 billion people compared with a no-waste scenario. Reducing waste in vegetable and fruit value chains may also reduce the negative impact of their production on the environment and could keep consumer prices more accessible [35].

#### *2.1. The Commodity Group "Vegetables"*

Vegetables comprise a wide range of genera and species and are an important component of a healthy diet. They ensure nutrition security through the provision of vitamins, antioxidants, minerals, fiber, amino acids, and other health-promoting compounds [36], while enhancing diversity, flavor, and taste of many otherwise bland staple dishes. Due to their wide range of uses, it is difficult to assign crops to the commodity group "vegetables". Some legume crops, mainly grown for their dry seeds, are important vegetable crops when harvested and consumed at the immature stage as seeds, green pods, and/or leaves, as is common in Asia and Africa. This applies, for example, to crops like vegetable soybean (*Glycine*), mungbean (*Vigna radiata*), Azuki beans (*Vigna angularis*), cowpea (*Vigna unguiculata*), yard-long bean (*Vigna unguiculata* subsp. *sesquipedalis*), black gram (*Vigna mungo*), common bean (*Phaseolus*), winged bean (*Psophocarpus*), and garden pea (*Pisum*). Although the starchy root of cassava (*Manihot esculenta*) constitutes the primary use of this crop, cassava leaves are a common leafy vegetable in many African countries; hence, this crop also falls partially under vegetables.

The most dominant vegetables, also called global vegetables as they are grown in many countries around the globe, are tomatoes, cucurbits (pumpkins, squashes, cucumbers, and gherkins), alliums (onion, garlic, shallot), and chilies (sweet and hot pepper; *Capsicum* spp.). Other major vegetable crops based on farmgate value of global production, but not always truly global vegetables are spinach, brassicas (cabbages, broccoli, rape), vegetable legumes, eggplants, lettuce and chicory, carrots and turnips, and asparagus [37]. Production statistics usually do not list indigenous or traditional vegetables as these are often produced in home or family gardens or collected from the wild for family consumption, and they are, in general, only offered in local markets. The term "indigenous vegetables" primarily refers to plants grown in their centers of origin or diversity [38], but also encompasses plant

species introduced from other geographical areas that adapted well, naturalized, and evolved in the new environment [39]. Indigenous vegetables are often more nutrient-dense than global vegetables [40], require low levels of external inputs, and cope well with abiotic and biotic stresses if grown on a small scale and in mixed cropping systems as is the case in their centers of origin. However, data on nutritional profiles of indigenous vegetables in raw and cooked forms are scarce. Among the traditional vegetables with high potential to be mainstreamed into urban markets and consumer diets are leafy amaranth (*Amaranthus* spp.), African eggplant (*Solanum macrocarpon*), jute mallow (*Corchorus olitorius*), roselle (*Hibiscus sabdari*ff*a*), spider plant (*Cleome gynandra*), Ethiopian mustard (*Brassica carinata*), okra (*Abelmoschus esculentus*), and vegetable cowpea (*Vigna unguiculata*) [41].

#### *2.2. Global Vegetable Production*

The statistics of the Food and Agriculture Organization of the United Nations (FAO) cover 25 primary vegetable products (http://www.fao.org/faostat/en/#data/QC). Global primary vegetable production reached 1.09 billion tons in 2018, about 37% of global cereal production (2.96 billion tons) [42]. Asia is by far the largest producer of primary vegetables, responsible for three-quarters of global production (Figure 1). During the past 10 years (2008–2018), there was a 24% increase in global commercial vegetable production, mainly attributable to a significant production increase in Africa (32%) and Asia (28.3%) (Figure 2). The estimated farmgate value of annual global vegetable production reached 543 billion United States dollars (US\$) in 2012–2013, about 65% of all food cereals combined, estimated at 837 billion US\$ [37].

**Figure 1.** Comparison of global production of primary vegetables in 2008 and 2018, by major regions (Source: Statistics Division of FAO). Available online: http://www.fao.org/faostat/en/#data/QC/ (accessed on 14 March 2020).

**Figure 2.** Percentage production increase of primary vegetables from 2008 to 2018 (calculated from production figures retrieved from Statistics Division of FAO). Available online: http://www.fao.org/ faostat/en/#data/QC/ (accessed on 14 March 2020).

#### *2.3. Research and Development E*ff*orts Focusing on Vegetables*

Although the farmgate value of annual global vegetable production is impressive when compared to global cereal production, the potential of vegetables to generate positive economic and nutritional impacts is still limited due to a relatively low level of support provided by national governments and development agencies to public sector vegetable research and development (R&D) [43]. Agricultural policies and investments in agriculture by the public and private sectors are still mainly focused on staple food production (cereals, roots, and tubers) that provide the primary source of calories, especially for low-income consumers. Fruit and vegetables, rich in micronutrients and vitamins, are not given appropriate support [44,45]. About 45% of private sector agricultural research and development investments into biotechnology and the crop seed industry are dedicated to a single crop—maize—and this investment is mainly taking place in industrialized countries [46].

With a few exceptions, private and public sector investments in vegetable research are mainly focusing on the development of hybrid cultivars of predominantly global vegetables, such as tomatoes, chilies, lettuce, cucumbers, and onions, while indigenous or traditional vegetables are being neglected. Data for 70 countries compiled by Schreinemachers et al. [37] clearly indicate the underinvestment in vegetable research and development compared to cereals. While there are 4–5 publicly funded cereal researchers per one million inhabitants in all country groups, low- and lower–middle-income countries, on average, have only one researcher working on fruit or vegetables [38]. When research investment in fruit and vegetables is placed in relation to production value of the different commodity groups, investment in fruit and vegetables is about the same for all country groups. However, higher-income countries allocate much higher research investment resources per dollar of cereal output than per dollar of vegetable output.

Investments in fruit and vegetable production should primarily target regions where projected supply is insufficient, such as sub-Saharan Africa, parts of Asia, and the Pacific. As fruit and vegetable production is mostly in the hands of small-scale producers in these regions, investments in this sector would have substantial potential to increase incomes and food security in these regions, particularly in Africa [33], thus contributing to the attainment of SDG 1—no poverty and SDG 2—zero hunger.

More research is also needed with respect to how nutrient profiles of global and traditional vegetables change in relation to cultivar choice, growing conditions, processing, and cooking methods. In addition, it is essential to study the extent of nutrient uptake and absorption by the human body. A study of the factors determining micronutrient bioaccessibility in leafy vegetables revealed that the pectin content of the leaves impaired carotenoid bioaccessibility [47]. Leafy vegetables rich in condensed tannins, such as drumstick tree (*Moringa oleifera*), had exceptionally low content of pectin and were characterized by high micronutrient bioaccessibility. Therefore, selection and development of cultivars with high micronutrient and low pectin content is a good approach to improve absorption of micronutrients by the human gut.

Fruit and vegetables are also quite perishable, with some estimates suggesting that these commodities contribute more than 40% of total food losses and waste [48]. Investments in R&D of vegetables and fruit value chains, focusing on new and improved processing, storage, and distribution technologies, could lead to a significant reduction of food losses and waste. Moreover, value addition through the development of drying technologies and novel extraction methods transforming fruit and vegetable waste (such as pomace, peels, and seeds) into new dehydrated and nutraceutical by-products could lead to a further reduction of food waste [49].

In general, it is important that the research and development community not only focuses on the farm segment of the food system but increasingly widens the understanding and supports the transformation of the entire system. Research strategies must include vegetable processing, the logistics of storage and transport to urban markets, and wholesale of inputs and outputs in the food system [21]. R&D investment efforts in these off-farm segments which account for 40–70% of value added and costs of food merit increased attention and could lead to lower marketing margins in value chains, thus improving efficiency.

As low consumption is also a problem where fruit and vegetable availability is not a constraint, it is important to address necessary changes in consumer behavior, apart from increasing production and availability of such crops. The provision of quantified food-based dietary guidelines is indicated to give the population some orientation, and such guidelines are missing in many countries, especially in Africa [50]. Apart from behavior change communication aimed at encouraging people to adopt healthier food choices that include fruit and vegetables, other key approaches are vegetable home gardens (see below) and school gardens. The latter simultaneously strengthen demand and supply and create awareness of the importance of the inclusion of fruit and vegetables in household and school meals to improve nutrition and health [37].

#### *2.4. Diversifying Production Systems and the Role of Home and Household Gardens for Nutrition Security*

The growing demand for fruit and vegetables can be met through diversification of staple crop systems by including fruit and vegetable crops or through intensification of specialized fruit and vegetable systems, especially in peri-urban zones [51]. Integrating small-scale farmers with formal and informal market outlets was shown to strongly encourage those farmers to adopt diversified vegetable production and other sustainable intensification practices in Kenya, as these result in economic benefits to farmers [52].

In view of global climate change, diversifying agricultural and horticultural production will help increase resilience of farming systems for both biotic and abiotic stresses. By the end of the 21st century, crop calorific yields in single cropping systems in sub-Saharan Africa are predicted to reach only 40–55% of the crop calorific yields obtained in sequential cropping systems [53]. Multispecies cropping systems constitute practical applications of ecological principles based on biodiversity, plant interactions, and other natural regulation mechanisms [54]. Such systems offer potential advantages in productivity, yield stability, ecological sustainability, and resilience to disruptions caused by climate change and other natural events but are sometimes considered harder to manage effectively than monocultures. A wider use of neglected and undervalued fruit and vegetable crops and semi-domesticated species, either intercropped with main staples in cereal-based systems or as stand-alone crops, would provide multiple options to build temporal and spatial heterogeneity into uniform cropping systems. Such an approach will enhance resilience to biotic and abiotic stress factors which are exacerbated by global climate change and will ultimately lead to a more sustainable supply chain of diverse and nutritious food [55]. Pilot studies conducted in Kenya and Vietnam revealed that diversification of smallholder farms with underutilized and traditional vegetable crops such as African nightshade (*Solanum americanum*), cowpea (*Vigna unguiculata*) crotalaria (*Crotalaria brevidens*), French beans (*Phaseolus vulgaris*), groundnuts (*Arachis hypogea*), kale (*Brassica oleracea* var. *acephala*), pumpkin (*Cucurbita maxima*), purple amaranth (*Amaranthus blitum*), spider plant (*Cleome gynandra*), mustard greens (*Brassica juncea*), orange-fleshed sweet potato (*Ipomoea batatas*), and water spinach (*Ipomoea aquatica*) could cover vitamin A requirements of 10–31 extra people per hectare and enhance income by 25% to 185% [56]. However, a trade-off of these diversification interventions was reduced leisure time.

Apart from such specialized, intensive production systems, home or household garden interventions have a proven effect on nutritional outcomes. Widely practiced forms of gardening consist of the mixed cropping of different species of fruits, vegetables, herbs, spices, and other useful plants as a supplementary source of food and income, and such gardens are common in Asia [57]. A well-planned and well-managed garden is expected to supply the household with a diverse range and year-round harvest of fruit and vegetables for household consumption with possible surplus sold on local markets. Such interventions potentially contribute to eight of the 17 Sustainable Development Goals [58]. The key benefits of home gardening were summarized by Landon-Lane [59] as follows:


Home gardens are typically targeted at women as they are, in general, in control of meal choice and preparation, and this may lead to women's empowerment as shown in Bangladesh [60]. Experiments with home gardens in Hyderabad, southern India, including about two dozen vegetable species, showed that a small area of 6 m × 6 m can provide much of the vitamin A and C requirement for a family of four during the entire year [34]. Apart from the provision of essential vitamins, many of the vegetable crops included in home garden kits are known to be naturally nutrient-dense [40,61,62].

#### **3. Vegetable Genetic Resources Conservation and Linkages with the Farming Community**

As genetic erosion continues in situ and on-farm due to expansion of human settlements, climate change, and replacement of landraces by high-yielding hybrid cultivars, additional collecting and conservation efforts are mandatory with a major focus on crop wild relatives and poorly represented landraces of major and minor vegetable groups. Special attention is needed to conserve the genetic diversity of indigenous and underutilized vegetable crops which, in general, are poorly conserved.

#### *3.1. Ex Situ Conservation of Vegetable Genetic Resources and Collecting Needs*

The Second Report on the State of the World's Plant Genetic Resources for Food and Agriculture (SoWPGR-2) indicates that about 7.4 million accessions are currently maintained ex situ, globally, 1.4 million more than were reported in the first SoW report [63]. Information in the World Information and Early Warning System (WIEWS) on the Plant Genetic Resources for Food and Agriculture (PGRFA) database indicates that about 45% of all the accessions in the world's genebanks are cereals, followed by food legumes with 15% of global holdings [63]. About one million accessions of crops used fully or partially as vegetables are conserved ex situ [64]. In a narrow sense of crops exclusively used as vegetables, about 518,000 accessions of vegetables representing 7% of the globally held 7.4 million accessions of PGRFA are maintained ex situ [63]. Among vegetable commodities, tomatoes (84,289 accessions), chilies (73,572), brassicas (25,566), cucurbits, excluding melons and cantaloupe (39,583), alliums (29,898), okra (22,428), and eggplant (21,616) are well represented in ex situ collections at the global level [64].

The World Vegetable Center (WorldVeg) with its headquarters in Taiwan holds about 63,500 accessions of vegetable germplasm comprising 170 genera and 456 species from 158 countries of origin [41], including some of the world's largest vegetable collections held by a single institution, such as chilies, tomato, and eggplant, as well as about 12,000 accessions of indigenous vegetables [40] (https://avrdc.org/our-work/managing-germplasm/). The WorldVeg germplasm collection can either be searched in its own database AVGRIS (http://seed.worldveg.org/search/passport) containing 71,889 passport records, or in the global database Genesys (https://www.genesys-pgr.org/) with 59,954 accession records.

WorldVeg works with Kew Royal Botanic Gardens, United Kingdom (UK) and partners in the Global Crop Wild Relative project to rescue the diversity of eggplant wild relatives from Africa. Through the project, WorldVeg obtained the seeds of 217 accessions of 18 species of eggplant wild relatives [41]. These accessions are currently being multiplied and characterized by WorldVeg genebank staff. Wild relatives of cucurbits, vegetable cowpea, and okra are other priority species in need to be collected in Africa and Asia through partnerships with local players to safeguard landraces and wild relatives of vegetable crops in both continents.

WorldVeg collaborates with National Plant Genetic Resources Centers in Eastern and Southern Africa to identify hotspots of vegetable diversity and to upload vegetable biodiversity data to public databases. In total, 126 high-potential traditional African vegetables relevant for people's diets in different regions of Africa were identified in collaboration with the World Agroforestry Center and the Inland Norway University of Applied Sciences [41]. Hotspots of vegetable diversity were identified in the coastal regions of West Africa, in Cameroon, South Sudan, Ethiopia, Tanzania, Madagascar, and Eswatini (formerly known as Swaziland). This vegetable diversity is poorly represented in genebanks and requires urgent conservation action.

The large group of gourds—bitter gourd (*Momordica charantia*), sponge gourd (*Lu*ff*a aegyptiaca* and *L. acutangula*), bottle gourd (*Lagenaria siceraria*), wax/ash gourd *(Benincasa hispida*), ivy gourd (*Coccinia grandis*), snake gourd (*Trichosanthes cucumerina*), and spiny gourd (*Momordica dioica*)—requires additional collecting with a major focus on landraces from centers of origin and diversity, many of which are threatened by the introduction of hybrid cultivars by seed companies, a process which recently started in many countries in South and Southeast Asia. Collecting led by national genebanks should focus primarily on Bangladesh, Myanmar, Vietnam, and India, countries rich in landrace diversity of gourds and other underutilized traditional crops. Such efforts could be undertaken in the context of FAO's Regional Initiative on the Zero Hunger Challenge for Asia and the Pacific, which is targeting the hidden treasures embodied in neglected and underutilized species (NUS), calling them Future Smart Food [65]. These foods are considered smart as they can bolster dietary diversification, improve micronutrient intake, require fewer inputs such as chemical fertilizers, enhance soil health, and often provide resilience to climate change and adverse farming conditions. Based on national NUS scoping studies, eight countries in South and Southeast Asia already prioritized up to six promising NUS as candidates for future smart food.

#### *3.2. Linking Genebanks with the Farming Communities*

To combat malnutrition, there is a clear need for supportive policies to advance ex situ and on-farm/in situ conservation and documentation of underutilized traditional crops and to forge stronger links between genebanks and the farming communities to strengthen the production and consumption of a wide range of diverse vegetables and fruit trees. In formal seed systems, genebanks usually supply genetic diversity primarily to plant breeders and research organizations who act as intermediaries between them and the farming communities based on a multi-step system of selection, breeding, testing, marketing, and adoption [66]. Today, genebanks are serving an expanded range of actors and institutions, beyond the conventional R&D set-up as outlined in the revised international Genebank Standards published in 2014: "Genebanks should promote the availability of genetic resources for uses including research, breeding, education, farming, and repatriation" [67] (p. 54). Research undertaken by Westengen et al. [68] showed that farmers, farmer organizations, and non-governmental organizations (NGOs) comprise a considerable user group of genebank material, having received about the same percentage (8%) of seed samples distributed by international genebanks in 2015 as distributed to the commercial seed sector. They identified six potential direct genebank–farmer linkages [68]: (1) reintroduction, (2) emergency seed interventions, (3) community seed banks, (4) participatory plant breeding, (5) variety introduction, and (6) integrative seed system approaches. In addition to the common role of genebanks to provide diverse crop germplasm to breeders and, thus, feeding the formal seed sector, such alternative and complementary strategies would strengthen the informal seed sector by enhancing farmers' access to crop diversity. Two of these genebank–farmer linkages, i.e., community seed banks and variety introduction, are described below.

#### 3.2.1. Community Seedbanks for Locally Important Crop Diversity

Community seedbanks (CSB) can be described as structures in which organized groups of farmers are responsible for the different stages of the management of seed or vegetative propagules, such as selection, conservation, multiplication, exchange, and improvement [69]. CSBs were successfully established all over the world in the last few decades, often supported and funded by NGOs. The main functions of CSBs include (a) conserving and reintroducing predominantly local germplasm, (b) providing easy access to seeds for members of the community, and (c) enhancing seed and food sovereignty [70]. If responsibly managed and supported, such CSBs can support the local community effectively and serve as a reliable seed source of locally important germplasm. Even in Europe, community seed banks are being established at a rapid pace, with at least 130 initiatives reported in 2017 [71].

Key elements in ensuring sustainability for community seedbanks include capacity development to ensure management quality, sustainable mechanisms, i.e., voluntary (in-kind) contributions by farmers to reduce dependence on external funding, enabling legal and political framework conditions to ensure a safe legal basis for operation, enabling social structures among those involved, satisfactory physical infrastructure, and systematic planning processes with effective operational mechanisms.

Community seedbanks and on-farm conservation efforts of crop diversity are only successful if they are fully embedded within the farmers' livelihood strategies and support the production of nutritious food, the generation of income, and other benefits such as ecosystem services, and socio-cultural and religious practices. Successful examples were documented in a community-based agrobiodiversity management project across Latin America and Southeast Asia [24].

A project on community-based seed production of traditional vegetables in the Philippines provided technical support to farmers for the conservation and multiplication of traditional vegetables to ensure the availability of high-quality seed of promising lines for home and school gardens and commercial production [72]. The project introduced several varieties of five traditional vegetable crops from the WorldVeg genebank: jute mallow (*Corchorus olitorius*; six varieties); cucumber (*Cucumis sativus*; two varieties); bottle gourd (*Lagenaria siceraria*; four varieties); eggplant (*Solanum melongena*; six varieties); ridged gourd (*Lu*ff*a acutangula*; three varieties). These crops were complemented with farmer-saved seed of other popular local traditional vegetable crops, such as Lima bean (*Phaseolus lunatus*), butterfly pea (*Clitoria ternatea*), and vegetable hummingbird (*Sesbania grandiflora*). The project made a significant contribution to halting the on-going threat of genetic erosion of local landraces and semi-wild vegetable crops. It also empowered farmers, especially women, to save, use, exchange, and sell their seeds to sustain the diversity of crops grown on-farm and promote greater diversity in diets.

To create the necessary legal foundation for CSBs, it is recommended that national governments integrate farmers' rights to save, use, exchange, and sell farmer-saved seeds in national seed legislations [73]. In this way, CSBs represent a form to implement farmers' rights. CSBs also play an intermediary role between genebanks and local seed systems as they multiply the relatively small seed samples usually distributed by genebanks to seed lots that are big enough for the needs of farmers [68].

#### 3.2.2. Variety Introduction of Agricultural Crops in General and Vegetable Crops in Particular

Variety introduction efforts are sometimes triggered by the need to provide farmers with a range of crop species and varieties to help them cope with environmental limitations in a changing climate by matching varieties to diverse production conditions and weather extremes [74]. To strengthen food security in sub-Saharan Africa (SSA), van Etten [74] proposed a crowd-sourcing approach to seed-based innovation, starting with the distribution of many small seed packets of different crops and varieties consisting of farmer varieties, landraces, and modern cultivars. For their distribution, existing networks such as retail stores, NGOs, and churches could be used. After having grown out the seed samples, farmers should provide basic agronomic data on the different varieties using their mobile phones. Farmers could also be requested to provide feedback on the performance of their own varieties compared to the newly introduced varieties. If local varieties prove to be superior to the introduced varieties, they could, if farmers agree, be included in the next evaluation round. This would then lead to a constant exchange of diverse varieties among the farming communities.

Such an approach was initiated in 2009 by Bioversity International through a project called Seeds for Needs (S4N) with project sites in 13 countries in Africa, Asia, and Latin America [68]. The project used crowdsourcing to evaluate both landraces and advanced cultivars and showed that landraces have great potential by offering valuable options to farmers having to deal with climate-related risks [75].

In Spain, the National Center for Plant Genetic Resources (CRF) of the National Institute for Agricultural and Food Research and Technology (INIA) recently launched an initiative to engage farmers, farmer associations, and relevant companies in the primary evaluation of germplasm that is conserved by the national ex situ collection network [76]. This initiative is linked to the First Action Plan of the National Program for the Sustainable Conservation and Use of Genetic Resources for Food and Agriculture (Order APA/63/2019). CRF will make selected varieties of local crop germplasm available to the farming community for cultivation under diverse growing conditions. In return, farmers will share information on the performance of the varieties with CRF, providing data on yield, incidence of diseases, insect pests and weeds, organoleptic quality, and other relevant data. This initiative is expected to strengthen on-farm conservation of valuable crop germplasm and to enhance agrobiodiversity, in line with SDG 2 (Zero Hunger).

The World Vegetable Center came to similar findings with variety introduction in Central Asia and the Caucasus (CAC). Starting from 2005, national research institutes in the CAC region received over several years a total of 2103 breeding lines and landraces of different vegetable crops from the World Vegetable Center [77]. About 45% of materials sent by WorldVeg were genebank accessions, mostly landraces and farmers' varieties [78]. The institutes undertook selection and adaptation trials with the materials received, and these efforts led to the registration and official release of 91 vegetable varieties as of 2017. Of these 91 new cultivars, 32 (35%) were developed from WorldVeg genebank accessions, 57 came from breeding lines, and only two resulted from actual crosses made using WorldVeg breeding lines. Meanwhile, another 10 varieties are already submitted for registration and subsequent release. The success of research institutes in releasing genebank accessions as varieties, without further cross-breeding, can be attributed to at least four factors [77]: (1) a relatively large number of accessions were tested for their performance under local conditions; (2) extensive selection trials were conducted to purify lines and stabilize traits by selecting for disease resistance and yield performance for at least three years; (3) disease pressure is relatively low in the CAC region due to favorable local climatic conditions (cold winters, hot and dry summers); (4) none of the institutes had well-functioning breeding pipelines at the start of this variety introduction process. Selection trials were, therefore, an effective way to create new cultivars.

The WorldVeg genebank of predominantly traditional vegetables in Arusha, Tanzania used a similar approach of variety introduction into East Africa via vegetable seed kit distribution. Between 2013 and 2017, this genebank distributed more than 42,500 seed kits totaling 183,193 seed samples to smallholder farmers in Tanzania, Kenya, and Uganda [79]. The seed kits usually comprised about four seed samples of different vegetable crops/varieties, enough to grow out in a home garden. About 32% of the seed distributed came from promising accessions of the genebank, while 68% were WorldVeg breeding lines. All materials distributed, including breeding lines, were maintained by the genebank. The WorldVeg genebank in Arusha used five important criteria when composing the seed kits [80]:


There is a clear trend in some East African countries (Kenya, Tanzania, Uganda) that seed companies become interested in multiplying and selling traditional African vegetable crops due to high demand from the farming community and consumers. To enhance vegetable diversity grown by smallholder farmers across SSA, the WorldVeg genebank in Arusha strongly engages with partners of the formal, as well as the informal, seed sectors.

#### **4. Vegetable Genetic Resources as Building Blocks for Vegetable Breeding**

Successful breeding, in general, requires access to a wide diversity of plant genetic resources, predominantly farmers' varieties, landraces, and crop wild relatives (CWR), the building blocks of intra- and interspecific crop diversity. Such plant genetic resources represent a treasure trove of genes for vegetable and legume crop improvement [80], enabling the delivery of more nutritious quality food in sufficient quantity for the world population.

Vegetable breeding should address the needs of both growers and consumers. Vegetable growers appreciate cultivars with high yield, uniformity, good market acceptance, multiple disease and pest resistance, and abiotic stress tolerance. Consumers like to buy vegetables with good appearance, shelf life, quality, taste, and nutritional value. Increased phytonutrient density in vegetable crops could help overcome micronutrient malnutrition and improve human health. The tomato breeding program at the World Vegetable Center includes enhanced phytonutrient content as a breeding objective, and it developed high-yielding and multiple disease-resistant lines with increased content of beta-carotene, lycopene, flavonoids, or anthocyanin in different fruit types [81]. WorldVeg breeders used the *Beta* allele from the wild species *Solanum hirsutum* which shifts tomato carotenoid from lycopene almost entirely to beta-carotene and results in orange fruit color. So-called "golden tomatoes" developed by WorldVeg breeders through conventional breeding techniques contain 3–6 times more provitamin A carotenoids than standard tomatoes, and one golden fresh market tomato can provide a person's full daily vitamin A requirements. WorldVeg *Beta* breeding lines in fresh market and cherry tomato fruit types were officially released as cultivars in Mali, Taiwan, and Bangladesh [81]. However, adoption is low so far as consumers are unaware or reluctant to accept the unfamiliar orange fruit color. Many major genes affecting nutrient content in tomato are known, and there is significant scope to enhance phytonutrient content in this crop through conventional breeding [81], but few breeding programs

pursue better nutrition in their breeding objectives due to lack of market incentives and not strongly enough articulated consumer demands.

Sustainable intensification of horticultural crops requires the development of new varieties with stable yields under climate change scenarios and adaptive capacity to diverse agro-ecosystems. The narrow genetic base of many vegetable crop cultivars is a major challenge for breeders aiming to develop improved varieties with multiple disease and insect pest resistance and tolerance to abiotic stresses such as heat, salinity, and drought, as well as increased input-use efficiency. Those breeding objectives require making use of interspecific crop diversity. Zhang et al. [82] were able to develop interspecific bridge lines among *Cucurbita pepo*, *C. moschata*, and *C. maxima*. With the development of these lines, it was possible to overcome the crossing barriers of interspecific hybridization and to eliminate the sexual obstacles of subsequent generations. This important breakthrough created a platform for breeders to transfer favorable traits among these species freely without the introgression of undesired characters from a wild species during the breeding process.

In contrast to public breeding programs such as those implemented by WorldVeg, for example, private seed companies primarily focus on the development of hybrid cultivars by exploiting heterosis effects and building multiple biotic stress resistance factors, as well as tolerance against abiotic stresses into a single commercial cultivar. This process ensures that growers must purchase fresh seeds for each growing cycle and the control of the parents prevents other seed companies from reproducing the hybrids. At the global level, the share of hybrid seed production is growing at a fast pace of 8–10% per annum in most major vegetable crops [83]. The global vegetable seed market was valued at US\$ 9.163 billion in 2018 and is projected to increase at a compound annual growth rate of 9.4% during the period from 2019 to 2024 [84]. North America is among the largest markets for vegetable seed production and consumption, followed by the Asia-Pacific region and Europe. Tomato, cabbage, sweet pepper, and lettuce are key players in the global seed market with a share of more than 30%.

Often, private seed companies make use of public breeding products when developing hybrid cultivars. A typical case is chili pepper (*Capsicum* spp.) variety development in India. Hybrid cultivars account for about 25% of the total area under chili pepper cultivation, about 25% of the area is under improved open-pollinated varieties (OPVs), and the remaining 50% area is still grown with local landraces [85]. The current chili pepper hybrid seed market in India is estimated at about 50 t per year, with an estimated turnover worth 16 million US\$, and this was made possible thanks to the use of WorldVeg cytoplasmic male sterility (CMS) breeding lines which reduce the cost of hybrid seed production by 50% compared to conventional hybrid seed development using manual emasculation. Conservative estimates suggest that hybrids involving WorldVeg germplasm and improved breeding lines as one of the parents were cultivated on more than 30,000 ha during 2012–2013 in different regions of India [85].

The potential of wild species as a source of genetic variation to bring about crop improvement was recognized early in the 20th century but is not yet widely used in crop breeding. An exception is perhaps the model crop tomato, which, in terms of production volume, is the most important vegetable crop grown worldwide. For this crop, an enormous amount of biotic and abiotic stress tolerance traits were already studied in the pool of wild relatives and extensively used in tomato breeding. Virtually all significant resistance genes to tomato diseases were sourced from wild relatives. An overview of such disease resistance genes introduced from wild species into cultivated tomato was provided by Ebert and Schafleitner [86]. However, it is essential to strengthen similar research in other major and minor vegetable crops as well.

Initially, molecular breeding complemented conventional breeding methods through marker-assisted selection (MAS) or marker-assisted backcrossing (MABC) [83]. Molecular markers intricately linked to the trait of interest can be detected and used in gene pyramiding, thus facilitating introgression of desirable, mostly monogenic traits from exotic germplasm into elite cultivars. Relatively little work was done with respect to traits that are governed by quantitative trait loci (QTLs). Traits such

as yield, quality, and stress response show complex inheritance patterns that result from the segregation of numerous interacting QTLs, the expression of which is modified by the environment [87].

New breeding approaches such as "introgressiomics" allow the creation of highly diverse plant materials and populations carrying introgressions of genome segments from mostly wild crop relatives into the genetic background of crops [88]. Introgressiomics can lead to the development of chromosome substitution lines (CLs), introgression lines (ILs), and multi-parent advanced inter-cross (MAGIC) populations. Such materials can be directly used in breeding pipelines and will facilitate the development of new resilient cultivars.

ILs contain the full genome of a given crop, except for a small chromosomal segment of a wild donor parent [87]. ILs are obtained through repeated backcrossing of the hybrid to the recurrent parent. Molecular markers help tracking the introgressed fragments, thus supporting the selection of beneficial materials for subsequent backcross cycles. A final step in the development of ILs is selfing or obtaining doubled haploids to fix the introgressed fragment in a homozygous state [89]. A further advantage of ILs is the ability to intercross favorable traits that are present in different ILs for the pyramiding of desirable alleles such as yield QTLs in tomato [90].

As early as 1995, Eshed and Zamir [91] developed a novel tomato population consisting of 50 ILs originating from a cross between the drought-tolerant, green-fruited wild tomato species *Lycopersicon pennellii*, and the elite tomato inbred line M82. Each of the lines contains a single homozygous restriction fragment length polymorphism-defined *L. pennellii* chromosome segment and, together, the lines provide complete coverage of the genome and a set of lines nearly isogenic to M82. These nearly isogenic lines of the IL population provide increased sensitivity for QTL mapping compared to whole-genome segregating populations, and they were extensively used during the last two decades to map QTLs for diverse tomato traits [92].

Only recently, the first eggplant introgression line population was developed using the drought-tolerant wild species *S. incanum* as a donor parent [93]. Sixty-eight candidate genes involved in drought tolerance were identified in the set of 25 fixed ILs. Apart from drought tolerance, *S. incanum* is also known to have a high content of bioactive phenolic compounds [94], as well as resistance to some diseases [95]. Open-field and screenhouse evaluations of the eggplant IL population mentioned above revealed that desirable traits such as lack of prickles and yield did not undergo considerable changes in most ILs despite the introgression of relatively large fragments from the wild exotic parent [96]. Ten stable QTLs distributed across seven chromosomes were detected, and three of the fruit-related QTLs appeared to be syntenic to other ones previously reported in eggplant populations. The other seven stable QTLs are new ones demonstrating that eggplant ILs are highly relevant for eggplant breeding under different environments and climatic conditions.

Genomics-assisted breeding is increasingly facilitating the introgression of favorable genes and QTLs with complex inheritance patterns from wild species into cultigens. The long-term conservation of genetic resources of landraces and crop wild relatives, their full characterization and evaluation, and their availability and accessibility will be instrumental for their successful use in public and private breeding programs. Such mobilization of the biodiversity available in the wider crop genepools will allow breeders to develop varieties that bear multiple disease and insect resistance and are able to adapt to rapidly changing environmental conditions, thus boosting agricultural production and ensuring food and nutrition security.

Moving such germplasm, once identified, across borders also requires robust phytosanitary capacity and practices and appropriate distribution capacities of genebanks, especially in the case of crops with recalcitrant seeds or those that are predominantly vegetatively propagated. Effective use of crop wild relatives and landraces also requires strong research capacity and participatory pre-breeding approaches. To enhance adoption, farmers should be actively involved in the definition of breeding objectives and the selection process for the development of new varieties, either through conventional or molecular breeding methods.

In contrast to the common vegetables grown globally, traditional vegetables, especially those which originated in Africa and the Asia-Pacific region receive much less attention from the research, conservation, and breeding community, although they have the potential to play a much greater role in more nutrition-oriented agriculture [97]. Neglect by research and breeders applies to many underutilized fruit and vegetable species. These include those crops with perennial growth forms such as trees, e.g., drumstick tree (*Moringa oleifera* [55]), and shrubs with edible leaves which are suited for agroforestry systems that may have an increasing role to play in sustainable vegetable production systems in developing countries under climate change scenarios.

Compared to major staple food crops, relatively little investment was made in breeding traditional, underutilized crop varieties [98]. Hence, the latter typically do not meet modern standards for uniformity and other characteristics required in the marketplace, and they tend to be less competitive than globally grown and traded crop cultivars. Farmers' varieties, landraces, and CWR were hitherto increasingly valued and exploited for genes that provide increased biotic resistance, tolerance to abiotic stress, yield, and quality [99–101]. However, use of agricultural biodiversity should not be restricted to exploiting valuable genes for use in breeding programs if our aim is to create more robust and resilient production systems. Underutilized food sources, including fruit and vegetables, legume crops, and root and tuber crops, have the potential to make a substantial contribution to food and nutrition security, protect against internal and external market disruptions and climate uncertainties, and lead to better ecosystem functions and services, thus enhancing crop and farming sustainability [102].

#### **5. Conclusions**

Vegetable crops are key sources of essential micronutrients required for good health. They contribute variety, flavor, taste, and nutritional quality to human diets. Increasing production and consumption of vegetables constitutes a direct and affordable way to deliver better health and overcome malnutrition. Vegetable production has the potential to generate more income and employment than any other segment of the agricultural economy. Vegetables can be grown on small areas of land, close to the consumers in urban and peri-urban settings, and they do not necessarily need advanced technologies to grow them. To realize those benefits, governments and donors need to give more weight and support to the ex situ, on-farm, and in situ conservation of genetic resources (farmers' varieties, landraces, and CWR) of global, as well as traditional, vegetables. The effective utilization of those genetic resources in breeding programs and the testing and deployment of newly developed varieties with tolerance to abiotic stresses and resistance against multiple diseases and insect pests in farmers' fields will ultimately benefit the farming community and consumers. By doing so, a significant reduction or, even better, a complete elimination of the obvious and persistent gap between WHO-recommended and actual intake levels of (fruit and) vegetables would make a significant contribution to the achievement of Sustainable Development Goals related to food and nutrition security and good health, in particular SDG 2.3 aiming at doubling agricultural productivity and incomes of small-scale food producers, SDG 2.4 ensuring sustainable food systems, and SDG 2.5 maintaining genetic diversity.

**Funding:** This research received no external funding.

**Acknowledgments:** The author wishes to thank Johannes M.M. Engels for his critical review and comments on a previous version of this manuscript.

**Conflicts of Interest:** The author declares no conflict of interest.

#### **References**


© 2020 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### **Macro- and Micronutrients from Traditional Food Plants Could Improve Nutrition and Reduce Non-Communicable Diseases of Islanders on Atolls in the South Pacific**

**Graham Lyons 1,\*, Geo**ff **Dean 2, Routan Tongaiaba 3, Siosiua Halavatau 4, Kabuati Nakabuta 3, Matio Lonalona <sup>5</sup> and Gibson Susumu <sup>6</sup>**


Received: 1 June 2020; Accepted: 21 July 2020; Published: 24 July 2020

**Abstract:** Pacific Islanders have paid dearly for abandoning traditional diets, with diabetes and other non-communicable diseases (NCD) widespread. Starchy root crops like sweet potato, taro, and cassava are difficult to grow on the potassium-deficient soils of atolls, and high energy, low nutrient imported foods and drinks are popular. Nutritious, leafy food plants adapted to alkaline, salty, coral soils could form part of a food system strategy to reduce NCD rates. This project targeted four atolls south of Tarawa, Kiribati, and was later extended to Tuvalu. Mineral levels in diverse, local leafy food plants were compared to reveal genotype–environment interactions. Food plants varied in ability to accumulate minerals in leaves and in tolerance of mineral-deficient soils. Awareness activities which included agriculture, health, and education officers targeted atoll communities. Agriculture staff grew planting material in nurseries and provided it to farmers. Rejuvenation of abandoned giant swamp taro pits to form diversified nutritious food gardens was encouraged. Factsheets promoted the most suitable species from 24 analyzed, with multiple samples of each. These included *Cnidoscolus aconitifolius* (chaya), *Pseuderanthemum whartonianum* (ofenga), *Polyscias scutellaria* (hedge panax), and *Portulaca oleracea* (purslane). The promoted plants have been shown in other studies to have anti-NCD effects. Inclusion of the findings in school curricula and practical application in the form of demonstration school food gardens, as well as increased uptake by farmers, are needed. Further research is needed on bioavailability of minerals in plants containing phytates and tannins.

**Keywords:** atolls; leafy vegetables; non-communicable diseases (NCD); nutrition security; mineral nutrients; natural biofortification

#### **1. Introduction**

#### *1.1. Epidemic of Non-Communicable Diseases (NCDs)*

Since the 1940s the consumption of high-energy, low-nutrient foods, including white flour, sugar and polished rice by Pacific Islanders, combined with reduced exercise, has resulted in alarming rates of obesity, heart disease, diabetes, and certain cancers [1,2]. Indeed, around 70% of deaths in Pacific Island countries (PICs) are due to non-communicable diseases (NCDs) [3–5]. Apart from the tragic personal cost, premature death and disability undermines national economic productivity. These diseases occurred at very low rates when traditional diets and lifestyles predominated [6]. In addition, many PICs are affected by the "double burden" of NCDs and under-nutrition; for example, high rates of iron-deficiency anemia in Papua New Guinea, Fiji, Solomon Islands, and Tuvalu [7,8]. Pacific Islanders have paid dearly for forsaking traditional diets. A recent study emphasizes the need, during nutrition transitions, for public health initiatives to promote traditional diets high in vegetables, fruits, and lean protein and agricultural initiatives to promote farm diversity [9].

In addition to the health benefits of traditional diets, local food crop (including wild food) biodiversity strengthens the resilience of food systems to climate events through increasing crop species richness, thus improving food and nutrition security [10–14]. It is also economically advantageous. Growing foods such as leafy greens, breadfruit, pumpkin, and bananas to improve nutrition helps to reduce trade deficits associated with high consumption of imported foods in the Pacific. In Kiribati and Tuvalu (see Figure 1), imported food comprises about 65% of food eaten [15].

**Figure 1.** Map of part of the south-western Pacific Ocean, featuring Kiribati, Tuvalu, and Fiji, and (inset) the Southern Gilbert Islands. This project focused on Abemama, Tabiteuea North, Nonouti, and Beru, and (for value-chain activity) Abaiang, just north of Tarawa.

#### *1.2. The Special Case of Atolls*

Kiribati (population 114,000) and Tuvalu (11,000) are small Pacific nations where around half the people live on the main atolls of South Tarawa and Funafuti, respectively.

Atoll soils are formed almost entirely from coral (predominantly calcium carbonate with some magnesium). They are coarse textured with no clay, so water flows straight through them. Moreover, droughts are common in this part of the world. The soil is often salty, highly alkaline (pH (H2O) 8.6–9.2), and low in nutrients such as potassium, iron, and manganese, and, unless well composted, low in organic matter [16]. Furthermore, inorganic fertilizers and chemical pesticides are prohibited on many atolls as they could pollute valuable underground fresh water. Improving soil health through targeted composting, along with growing and eating nutritious crops on atolls should lead to improved diet, nutrition, and health.

#### *1.3. Why Leafy Plants?*

On many atolls, in particular those of the Southern Gilbert Islands (part of Kiribati, see Figure 1), which often experience drought, starchy root crops can be difficult to grow, resulting in low tuber/storage root yield. This is associated with potassium deficiency and lack of sufficient water during the weeks after planting. Potassium is needed to ensure adequate storage root initiation, and the high potassium content of tubers/storage roots depletes soil potassium with each harvest [17,18].

On the other hand, hardy leafy food plants can yield well under these conditions. Many different types of leafy vegetables and leaves of other plants/food crops are grown and eaten in the Pacific region (e.g., edible ferns, kangkong, amaranth, drumstick, and leaves of starchy root crops like taro, sweet potato, and cassava) [10,19–23]. When available, indigenous vegetables are usually inexpensive and thus affordable to most people in both urban and rural areas. Despite this, they are often overlooked and regarded as "low status foods" [1,2]. However, they are important for human health, being nutritious and rich in protein, minerals, vitamins (e.g., A, B, C, K), beneficial phytocompounds, and fiber [10,22–32]. A study in Africa found that "orphan" (unimproved) leafy vegetables were popular with farmers if they were full-season varieties with high leaf yield, and resistant to pests, diseases, and abiotic stress (e.g., drought, heat, salinity). Retailers and consumers valued good appearance, long shelf-life, affordability, and high nutritional value [33].

Iron provides an example of an important micronutrient found in leafy vegetables. Lack of iron can cause iron-deficiency anemia, common in women, inducing fatigue and weakness, and in children, affecting growth, energy levels, and learning ability. Purslane, pumpkin leaves, kangkong, yellow beach pea, and chaya are all good sources of iron [23,29].

Phytocompounds such as flavonoids, anthocyanins, polyphenols, and carotenoids are beneficial to humans as antioxidants and anti-inflammatory agents in reducing the risk of diabetes, heart disease, and cancers. Examples include glucosinolates in drumstick leaves and anthocyanins in purple sweet potato leaves. Certain carotenoids, notably beta- and alpha-carotene, are converted to vitamin A when eaten, especially if consumed with some oil (e.g., coconut cream) [34]. Others, notably lutein (which is usually abundant in leafy vegetables) and zeaxanthin are important for eye health and can reduce the risk of cataracts [35]. Importantly, given the current NCD pandemic, there is growing evidence for specific activity of certain plants against diabetes and cardiovascular disease, e.g., drumstick [36–38] and chaya [39,40].

#### *1.4. Project* O*bjective and Strategy*

The objective of this project (2014–2019) was to support and enhance an awareness program aimed at increasing production of nutritious leafy plants to reduce rates of NCDs in Kiribati and Tuvalu. This was achieved through:

• A survey of mineral nutrients in local leafy food plants collected in Kiribati and Tuvalu.


#### **2. Results and Discussion**

#### *2.1. G x E Study*

During the 2014 scoping study for the current project, we found 11 of the 12 leafy vegetables featured in the earlier factsheets growing on South Tarawa and Funafuti. This was surprising in view of the almost universal inhospitable coralline atoll soils, compared with, for example, soils of Solomon Islands. Most were growing in gardens and hedges close to homes; however, they were usually used for animal feed or as ornamentals. Clearly, raising awareness is an important program component, which includes school food gardens and curriculum development, farmer field schools, village workshops, and media promotion. See further discussion on this below.

Plant production is limited mostly by soil plant-available mineral content. Comparing mineral concentrations in genetically diverse plants provides insights into the plant–environment interactions that control mineral nutrient accumulation [41]. This not only enables identification of nutritious food plants for humans and animals, but also can lead to improvement in sustainable yield.

Mineral concentrations in leaves varied widely with species, with five-fold variations common between species grown on the same soil (Tables 1 and 2). Variation was less marked between sites (environment effect) for most minerals than is usually found, due to the relatively uniform coralline soils. For example, most minerals in leaves of chaya (*Cnidoscolus aconitifolius*) varied by no more than two-and-a-half-fold across seven sites, the exceptions being manganese and zinc (Table 3).


**Table 1.** Concentrations of macro- and micronutrients in leaves (dry weight basis) of different food plant species grown together on the same soil type at Vaiaku, Funafuti, Tuvalu in August 2014.

Notes: Concentrations are on a dry weight (DW) basis throughout the manuscript. N % × 4.4 provides an estimate of crude protein %. Ca was uniformly high (range 1.61–2.20%). S was moderate in six species (0.21–0.38%) but high in drumstick tree (1.13%). Note that the data in Tables 1–3 are analyses conducted on single representative sub-samples of pooled samples for each species.


**Table 2.** Concentrations of macro- and micronutrients in leaves (dry weight basis) of different food plant species grown together on the same soil type at Tanaea, South Tarawa, Kiribati in August 2014.

**Table 3.** Variation in selected minerals in leaves of *Cnidoscolus aconitifolius* (chaya) growing at seven locations in Tuvalu (sites 1 and 2) and Kiribati (sites 3–7) from 2014 to 2017. This study illustrates variation due mostly to differences in plant-available levels of these nutrients in soil. Most minerals (Zn, Mg, N) varied by less than three-fold.


#### *2.2. Natural Biofortifiers with Variability in Micronutrient E*ffi*ciency*

Table 4 features the leafy plants which consistently accumulated the highest levels of minerals. These species could be described as natural biofortifiers of the corresponding nutrient. Leaves of pumpkin (*Cucurbita pepo*), purslane (*Portulaca oleracea*), and chilli (*Capsicum frutescens*) contained relatively high concentrations of most of the minerals, and thus, at least with respect to minerals, could be regarded as the most nutritious overall. Other studies also report the high nutritive value of these plants [42–44].

Species such as hedge panax and birdsnest fern were not observed with leaf chlorosis during this study, regardless of high soil pH. They are not exceptional Fe accumulators, e.g., birdsnest fern collected on Papaelise Island, Funafuti contained only 13 mg/kg DW of Fe, but looked healthy, and this compares with cassava, with a critical level for Fe of around 50 mg/kg [45]. Nevertheless, it is likely that plants such as birdsnest fern are efficient for Fe [46], and probably also for other nutrients, e.g., Mn, Cu, K in short supply in coralline soils. These plants seem to be able to function normally, especially with respect to photosynthesis, even when the nutrient is present at low plant-available levels in the soil. This is a different trait (involving different genes) to the ability to take up and accumulate high levels of a nutrient. Birdsnest fern is also an exceptional accumulator of K. Most cassava varieties, on the other hand, suffered from chlorosis, stunted growth, and lack of sizeable storage roots on the southern atolls. However, cassava and purslane are adept at extracting Zn from the soil and accumulating it in leaves. **Table 4.** Selected mineral nutrients and the leafy vegetable species found (using opportunistic GxE analysis) in this study to be the most effective accumulators of these minerals in leaves. Samples were collected from various locations in Kiribati and Tuvalu. The values in brackets are representative concentrations of the relevant mineral for each species in this region.


Notes: Selenium is a micronutrient for humans and animals but not for higher plants; μg = micrograms; N % × 4.4 provides an estimate of crude protein % in leaves.

#### *2.3. Factsheets*

In addition to the introductory factsheet, 12 species factsheets were produced, which feature the most atoll suitable nutritious leafy vegetables identified during the project. Several of these species have also been recognized for their nutritional value in other studies (Bailey, 1992; French, 2010; SPC, 2012). The featured plants are *Amaranthus viridis* (amaranth), *Cnidoscolus aconitifolius* (chaya, tree spinach), *Moringa oleifera* (drumstick tree), *Polyscias scutellaria* (hedge panax), *Pseuderanthemum whartonianum* (ofenga, Carruthers' falseface), *Vigna marina* (yellow beach pea, beach cowpea), *Ipomoea aquatica* (kangkong), *Cucurbita pepo* (pumpkin), *Sechium edule* (choko), *Abelmoschus manihot* (bele, aibika, slippery cabbage), *Capsicum frutescens* (chilli), *Portulaca oleracea* (purslane, pigweed). Factsheet 13 discusses nutritional aspects of composting methods suitable for atolls.

Although not featured in the factsheets due to budgetary constraints (and their overall mineral levels were a little below the selected species) other nutritious leafy vegetables included *Asplenium nidus* (birdsnest fern), *Pisonia grandis* (big lettuce tree), and the leaves of these starchy root crops: *Ipomoea batatas* (sweet potato), *Manihot esculenta* (cassava), *Xanthosoma sagittifolium* (cocoyam), and *Colocasia esculenta* (taro). The nutritional value of these species has been noted in earlier studies [10,19,22,28].

Yellow beach pea was included more for its importance as a well-adapted legume on atolls than for its eating quality. It is an efficient N-fixer, with extensive root nodulation observed whenever sampled in Kiribati and Tuvalu, and is salt- and drought-tolerant. It grows better on strongly alkaline soils than *Mucuna*, *Pueraria, Centrosema, Gliricidia, Erythrina,* and *Sesbania*, can smother weeds, and its relatively high N, Fe, and Zn content make it a valuable green manure and compost component. Its seed pods are good to eat and highly nutritious when green, although all but the youngest of its leaves are chewy due to their high fiber content. A similar creeping legume is the Mauna Loa bean (*Canavalia cathartica*), which also thrives near beaches of Tuvalu and Kiribati, and has purple flowers, larger pods, and is a more vigorous tree-climber than *Vigna marina.*

#### *2.4. Medicinal E*ff*ects*

Although this project focused on the food/nutritional value of leafy green vegetables, traditionally in many countries they are also used for specific medical applications. For example, chaya (which originated in Mexico and Mesoamerica) protects the heart, liver, and kidneys from toxin damage [47,48]; drumstick (India and Pakistan) has anti-bacterial effects [49,50]; and bele (Papua New Guinea and

Solomon Islands) is used for bone repair and treating osteoporosis [51]. Hedge panax, drumstick, chaya, bele, and purslane are galactogogues that can stimulate lactation [10,52–54]. Indeed, purslane's generic name Portulaca means "to carry milk". This plant is so ubiquitous and prolific globally that it is usually regarded as a weed. It is renowned for its high n-3 fatty acid content [14], and in this study was found to be the best accumulator of Mg, Fe and Zn of all the plants analyzed. Pumpkin thrives in composted atoll soils and is already grown widely in both countries. Ofenga, in particular the red-leaf form, is better known in Kiribati for its embalming ability than as food.

High-protein species drumstick and chaya were fostered in Kiribati and Tuvalu under the Pacific Regional Agricultural Programme (PRAP) in the 1980s, and have adapted well to harsh atoll conditions [21]. Drumstick is high in *b*-carotene, sulphur, and selenium [14]. This species had the highest *b*-carotene level (427 mg/kg) of all plants analyzed in the earlier ACIAR project in the Pacific and northern Australia, and its protein is considered to be high in quality, with a similar pattern of essential amino acids as soybean [1]. It regularly accumulates around 12 times the concentration of selenium and around four times the concentration of sulphur compared to most other plants grown on the same soil. At the Vaiaku, Funafuti site (Table 1), drumstick leaves contained 25 times the Se concentration of the mean of the other plants growing there. Similar differences have also been observed in Africa [55]. This trait would be especially valuable in Sub-Saharan Africa, where these minerals are deficient in many soils [56–58]. Their deficiency is considered by some researchers to increase risk of HIV/AIDS [59].

Chaya is also renowned for its nutritional and medicinal effects. Like drumstick, it is an excellent source of high-quality plant protein and carotenoids and is renowned for its liver- and kidney-health enhancing effects [48].

Bele is not as climate- or insect-resilient as chaya, but has been included as its flavor is highly regarded, it is noted for health properties, including high levels of the important carotenoids lutein and *b*-carotene [1], and it grows well on composted soils with sufficient rainfall. It is the most popular leafy vegetable in Solomon Islands and Papua New Guinea.

Especially important, given the high NCD (particularly diabetes) rates in the Pacific and northern Australia, are the anti-diabetes and anti-cardiovascular disease effects of most of the plants featured in the factsheets, demonstrated in scientific studies. Studies with evidence for this include the following: drumstick [36–38,50,60], amaranth [61–63], bele [64–66], chilli [67–69], purslane [70–72], kangkong [73– 75], ofenga [76–78], hedge panax [79,80], chaya [39,40], pumpkin [81,82], and choko [83,84]. Their inclusion in the diet in sufficient quantity is likely to reduce the risk of diabetes and cardiovascular disease, not only by reducing glycemic load when they are included with high-carbohydrate meals, but also because of specific anti-diabetes effects.

#### *2.5. Mineral Deficiencies of Atoll Food Plants*

The plant leaf mineral data revealed that the most common mineral deficiencies in both countries were K and Mn. For example, 51% of the plant samples had K < 15,000 mg/kg, and 46% had Mn < 15 mg/kg. Phosphorus was mostly in the adequate range of 2500–4000 mg/kg, with 23% at marginal levels. Copper was marginal (<4 mg/kg) in 19% of leaf samples. Nitrogen deficiency was rare, and 33% of leaf samples had >4% N, testament to effective long-term composting, along with the inclusion of legumes, cassava, chaya, and drumstick (which are all inherently high in N) in the sample collection. Sites which had been composted for several years were higher in N (both nitrate and ammonium), available P, K, Mg, B, Cu, Fe, Mn, and Zn.

Iron deficiency, which is usually associated with alkaline soils, was not widespread, with most leaf samples > 30 mg/kg. Critical Fe levels are species-specific; as noted earlier, hedge panax can function normally with less Fe than can cassava. Likewise, most plants had sufficient Zn and B, with levels mostly in the range 30–70 mg/kg, and only 11% of leaf samples were <20 mg/kg in either.

Leaf Na levels were, unsurprisingly, relatively high, mostly >2000 mg/kg in Kiribati and >5000 mg/kg in Tuvalu, but symptoms of Na, Cl, or NaCl toxicity were not observed, even in cocoyam with leaf Na of 31,000 mg/kg on South Tarawa. High soil and plant levels of Ca and Mg counteract Na toxicity [85,86].

#### *2.6. Giant Swamp Taro Food Garden*

The value of home gardens comprising diverse, nutritious, traditional food crops to supplement the diet of subsistence households is well documented [1,11,87–90]. The cultivation of the giant swamp taro (*Cyrtosperma merkusii*, called babai in Kiribati and pulaka in Tuvalu) is traditional on atolls. Pits are dug by hand down to the water table, which in many Kiribati atolls is only 1–2 m below the surface. Many of these pits are now neglected but they provide a strong connection to both culture and fresh ground water.

In an adaptation of this system, kangkong can be grown in the water with the swamp taro. Hence the drought-tolerance requirement is waived for this species. Its vigorous growth means that, unless harvested regularly, it can intrude upon the giant swamp taro, and may need to be grown in a separate pit. The other crops can be grown on constructed terraces formed from the pit walls, and drumstick, ofenga, hedge panax, and beach cowpea are planted around the pit at ground level. Other crops, such as bananas, pawpaw, sweet potato, and annual vegetables can be included as well (Figure 2). These babai food gardens (BFG) (or pulaka food gardens (PFG) in Tuvalu) represent a mini food system that can, once established, contribute significantly to a family's nutrition. For example, 100 g fresh weight of purslane leaves/stem will provide around 20% of an individual's daily Zn requirement. The size of the garden can be as small as 40 m<sup>2</sup> or as large as 0.3 ha. Even in crowded places, such as Betio on South Tarawa, there is usually room to plant a drumstick tree or two, which would soon provide a sustainable daily supply of leaves for a family.

On Nonouti, which has perhaps the toughest environment of the four atolls in the study, especially for drought frequency and duration, 27 BFG were commenced during the project, and at July 2019, 12 appeared well established and sustainable, a reasonable success rate for a new concept. Participating households usually have a BFG as well as food plants growing nearby on the land surface and around the house. These include sweet potato, ofenga, chaya, lemon grass, chilli, Brazilian spinach, pumpkin, banana, pawpaw, coconut, and breadfruit.

**Figure 2.** Layout of a babai/pulaka food garden. Other nutritious food crops can be substituted or added if desired.

#### *2.7. How to Eat these Nutritious Vegetables*

It is recommended to eat around three handfuls (around 100 g) of leafy vegetables each day. Some green leaves can be eaten uncooked, e.g., purslane, kangkong, and chilli, which preserves most vitamins, but it is usually preferable to cook them. It is important to wash the leaves in clean water first, to remove dirt and possible pathogenic microbes. Optimum cooking methods are steaming, simmering in a little water, baking, or stir frying in a little oil (ideally virgin coconut oil or coconut cream) for minimal time to limit nutrient loss. The cooking water should not be discarded, but instead used for soup. A simple method suitable for these vegetables is to chop them into small pieces (except drumstick, in which case strip the leaflets from the wiry petioles), simmer in water for 10–15 min, add coconut cream, and simmer for a further 10–15 min. Other ingredients can be added to enhance the flavor if desired. More elaborate recipes are included in several of the factsheets. A recipe book that includes most of the featured nutritious leafy plants was produced, written in Kiribati language, by the IFAD community awareness team.

The bioavailability of Fe, Zn, and other minerals will be reduced by the presence of phytate, tannins, and polyphenols, e.g., in drumstick, chaya, and purslane [91–94]. The effects of these so-called antinutrients can be reduced by various cooking methods. In chaya, for example, boiling significantly reduced phytates, oxalates, and tannins [94] and virtually eliminated cyanogenic glycosides [93].

#### *2.8. Awareness Program and Planting Material Provision*

In order to achieve impact, the project collaborated with multiple government ministries (Agriculture, Health, Education, Works), churches, NGOs, Island Councils, and communities on the target atolls. In Kiribati, about 1500 farmers attended information and training sessions on growing, handling, cooking, and preserving locally grown foods. Our findings agreed with an FAO study in Samoa which found that the main external factors which influenced people's decisions about food were availability, accessibility, cultural obligations, and family income [95].

Increasing awareness and generating interest must be met with availability of supply, whether planting material for home gardens or on a larger scale for farmers to produce for markets and tourism outlets [2]. Suitable planting material of the selected species was supplied via ALD nurseries on each atoll. The ALD Tanaea HQ, an infertile site with multi-micronutrient deficiencies, was transformed, using an improved watering system and composting, into an important germplasm source and model nutritious food garden. In Tuvalu, secondary bush was cleared, a water tank and irrigation system installed on Funafala island, Funafuti atoll and nutritious food crops grown to supply nearby Vaiaku, the main population center. This highlights the need for more resources to be devoted to conservation of diverse leafy food plants, starchy root crops, and fruits in the Pacific region [2,89].

The Pacific Community's Centre for Pacific Crops and Trees (CePaCT), Suva plays a key role in germplasm conservation and distribution. Conservation of traditional crops is especially important in countries such as Papua New Guinea and Solomon Islands, where the natural environment is threatened by logging, mining, and oil palm establishment [89].

Value chain research is essential: the producer needs to be convinced that production of green leafy vegetables is worth the effort. Strategies are needed to deliver health benefits to consumers and economic benefits to local horticultural producers and other value chain participants [13].

Improving nutrition is usually seen as the task of health agencies, but it is apparent that a cross-sectoral and multi-agency food system approach is needed [96]. The NCD pandemic can be addressed by increasing diversity on the farm and extending this diversity (of which nutritious leafy vegetables form an important part) to the diet. Involvement of children in promotion of nutritious local foods is integral; in many countries their importance in influencing lifestyle factors, especially diet, is becoming recognized. For example, schools can include food gardens featuring the most nutritious local plants, provide more nutrition education, and students can transfer knowledge back to their villages [1]. In the current study, around 1700 students (to date) have attended awareness and training programs. Further research is needed on how to optimize awareness and promotion. This is crucial

for Kiribati and Tuvalu, where leafy plants were not major components of traditional diets. Studies examining the bioavailability of minerals in these plants are also needed.

#### **3. Materials and Methods**

Adaptability of leafy food plants to tough atoll conditions was clear from observation on the four atolls and on South Tarawa and Funafuti, precluding the need for formal trials. A survey was conducted to identify the most nutritious leafy food plants, in terms of minerals and protein that grow in Kiribati and Tuvalu. Particular attention was paid to species that thrive in the atoll environment. Leaf tissue samples were collected in Kiribati and Tuvalu from 2014 to 2018 (*n* = 140), and with the inclusion of leaf mineral data from the previous Pacific-Northern Australia nutritious leafy vegetable project (ACIAR PC/2010/063) [1] (*n* = 274), a total of 414 samples (with 65 food plant species, of which 24 were found growing in Kiribati and Tuvalu, and also 50 species used for herbal medicines (18) or compost (32); usually multiple samples of the same species growing at different sites were analyzed) informed the factsheets produced during the current project. In Kiribati, samples were collected on the islands of South Tarawa, Abemama, Tabiteuea North, Nonouti, and Beru. In Tuvalu, samples were collected on Vaiaku, Funafala, and Papaelise.

As with the earlier project, an opportunistic genotype–environment (GxE) strategy was employed. This included sampling of single leafy vegetable species growing at different sites as well as sampling multiple species growing at the same site. Note that due to daily time and budgetary constraints, the data presented in Tables 1–3 (in Section 2) are for single samples. The locations reported in the tables were chosen to typify the leaf mineral concentrations found throughout the study for these species.

This enabled the effects of environment (mostly soil type) and genetics (plant species/variety) to be separated, thus allowing an assessment of the ability of each species/variety to take up and accumulate essential minerals in their leaves. The minerals studied were the macronutrients nitrogen, phosphorus, potassium, calcium, magnesium, and sulphur, in addition to sodium, often present in "macro concentrations" but required in micro amounts; along with the micronutrients iron, manganese, boron, copper, and zinc. All of these macro- and micronutrients, with the possible exception of boron, are required by humans and animals as well as by plants. A sub-sample was analyzed for selenium, an essential micronutrient for humans and animals, but which is not required by higher plants. The analyses also enabled detection of any mineral deficiencies in the plants sampled.

Each leaf tissue sample comprised around two handfuls of relatively young leaves: not the youngest or older leaves, but, e.g., in sweet potato, the 5th to 9th youngest leaves, sampled from several representative plants, avoiding plants with disease (e.g., virus, scab) symptoms. The exception was pumpkin where just the tips (up to 25 cm) were sampled. If the plants were dusty (e.g., if the plants were growing near a road), they were washed in clean water. Samples were dried in a microwave oven or perspex covered trays soon after collection and placed in labelled plastic ziplock bags. Soil and compost samples were also collected from numerous sites in Kiribati and Tuvalu, and these will inform a future article.

The samples were brought to Australia under a Federal Department of Agriculture permit and irradiated. They were then acid digested and N analyzed by the combustion method (using an Elementar instrument, and limit of detection (LOD) calculated as 10 x the standard deviation of the calibration blank). Protein % was estimated by multiplying nitrogen % by 4.4. The other minerals listed above were analyzed by inductively coupled plasma atomic emission spectrometry (ICPOES) (using a radial CIROS instrument, and LOD as above), while Se was analyzed using ICP mass spectrometry (ICPMS) (using ICPMS method-*7B2G*, and LOD as above). Appropriate quality control measures were applied, including regular duplicate samples and analyses of aluminum and titanium to detect dust/soil contamination, which inflates Fe concentration.

In the previous study, carotenoids were also analyzed, and to minimize enzymatic degradation, samples were dried rapidly in a microwave oven as soon as practicable after collection, usually the same night. The carotenoids *beta*-carotene (the major pro-vitamin A carotenoid), lutein (usually the most abundant carotenoid in leaves), and *alpha*-carotene were analyzed by size-exclusion high-performance liquid chromatography (HPLC) at the Mares laboratory, Waite Campus, University of Adelaide [1]. Budgetary constraints precluded carotenoid analyses in the current study; however, these methodological details are included as carotenoid data are included for several of the species featured in the factsheets, the main study output.

The criteria for atoll suitable leafy plants, as listed under Project objective above, were (1) highly nutritious, (2) taste good, (3) tolerant of alkalinity (i.e., soil pH (H2O) > 8.5), (4) tolerant of salt and drought (with the exception of *Ipomoea aquatica*, kangkong, which grows in fresh water), and (5) easy to grow, prepare, and cook.

For the factsheets, photographs, characteristics, uses, availability, propagation and growing methods, disease and pest threats, and advice on harvesting and storage were included for each species. Leaf mineral and carotenoid data (if the species was included in the previous factsheets) were presented in a table which included the featured leafy vegetable sampled at a representative site, compared with other leafy vegetables growing at the same site. English cabbage was also included, as a moderately nutritious yardstick, using average values of samples purchased at markets in the South Pacific.

Factsheets 1, 2, 4, 5, 6, 8, 9, 10, and 11 were adapted from those published during the ACIAR PC/2010/063 project, Feasibility study on increasing the consumption of nutritionally-rich leafy vegetables by indigenous communities in Samoa, Solomon Islands and Northern Australia (www. remoteindigenousgardens.net\T1\textgreater{}2013/08\T1\textgreater{}new-resources-fact-sheets) [23].

The factsheets (500 sets) were graphically designed, printed, and laminated to improve durability by SPC, Suva, Fiji and distributed in Kiribati and Tuvalu, and also published online: www.researchgate.net\T1\textgreater{}publication\T1\textgreater{}327261351\_Tackling\_NCDs\_ from\_the\_ground\_up\_Nutritious\_ leafy\_vegetables\_to\_improve\_nutrition\_security\_on\_atolls [29]. Later, they were translated into the Kiribati and Tuvalu languages. A recipe book featuring the selected plants was compiled and distributed to communities.

The project also trialed and promoted starchy root crops (taro, sweet potato, cassava) and non-leafy vegetables, including beans, tomato, cucumber, capsicum, eggplant, and watermelon, and these components are reported elsewhere.

**Author Contributions:** Conceptualization, G.L., G.D. and S.H.; Funding acquisition, G.D., S.H. and G.S.; Investigation, G.L., G.D., R.T., S.H., K.N. and M.L.; Project administration, R.T., S.H., K.N., M.L. and G.S.; Supervision, G.L., G.D., R.T., S.H., K.N., M.L. and G.S.; Visualization, G.L.; Writing—Original draft, G.L.; Writing—Review and editing, G.D., R.T., S.H., K.N., M.L. and G.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Australian Centre for International Agricultural Research (ACIAR), grant number SMCN-2014-089.

**Acknowledgments:** ACIAR Program Managers Robert Edis and James Quilty, and IFAD, our main collaborator. Rosalind Kiata for her journalistic expertise in the promotion campaign. Walter Wasile, Suva for the graphics. Mary Taylor for her assistance with the scoping study for this project in 2014. This article is dedicated to Lois Englberger, an inspirational leader in improving human nutrition in the Pacific region, who founded the *Go Local* movement from her base at The Island Food Community of Pohnpei.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

### **Assessment of the Nutritional and Medicinal Potential of Tubers from Hairy Stork's-Bill (***Erodium crassifolium* **L 'Hér), a Wild Plant Species Inhabiting Arid Southeast Mediterranean Regions**

**Shabtai Cohen 1, Hinanit Koltai 2, Gopinath Selvaraj 2, Moran Mazuz 2, Moran Segoli 1, Amnon Bustan <sup>1</sup> and Ofer Guy 1,\***


Received: 31 July 2020; Accepted: 19 August 2020; Published: 20 August 2020

**Abstract:** Emerging needs for diversifying human diet and to explore novel therapeutic procedures have led to increasing attempts to retrieve traditional nourishments and recruit beneficial wild plant species. Species of the genus *Erodium* (Geraniaceae) harbor medicinal indications and substances known from folklore and scientific research. Hairy stork's bill (*Erodium crassifolium* L'Hér), is a small hemicryptophyte that inhabits arid southeast Mediterranean regions. *E. crassifolium* is among the very few Geraniaceae species known to produce tubers. Traditional knowledge holds that the tubers are edible and used by Bedouin tribes. However, no scientific information was found regarding nutrition or medicinal properties of these tubers. The objectives of our project are to unravel potential nutritional and medicinal benefits of the tubers, conduct initial steps towards domestication and develop agricultural practices enhancing *E. crassifolium* tuber yield and quality. Tubers show high water content (90%), low caloric value (23 Kcal 100−<sup>1</sup> g) and considerable contents of minerals and vitamins. In addition, the tubers contain significant amounts of catechins and epigallocatechin, polyphenolic compounds known for their antioxidative, anti-inflammatory and antiproliferative activities. Furthermore, in vitro experiments demonstrated significant anti-inflammatory effects on human cell cultures. *E. crassifolium* is highly responsive to environmental changes; fertigation (700 mm) increased tuber yield by 10-fold, compared to simulated wild conditions (50–200 mm). These results indicate a significant potential of *E. crassifolium* becoming a valuable crop species. Therefore, there is a need for continued efforts in domestication, including ecotype selection, breeding, development of suitable agricultural practices and further exploration of its medicinal benefits.

**Keywords:** anti-inflammatory activity; antioxidants; catechin; domestication; *Erodium crassifolium*; underutilized species

#### **1. Introduction**

Human diet and therapy have featured valuable plants gathered from the wild since ancient times. The agricultural civilization developed in the past 12,000 years has been founded on the domestication of many useful species [1]. However, in the modern era, global economic considerations have significantly changed agricultural approaches and scales, which brought about overexploiting of land resources, consistent diminishing of natural biodiversity and negative modifications of human diet and health [2,3]. Globalization and urbanization have accelerated the unification of the current

human diet and have led to a further narrow hoard of available useful species [4]. The recent burst of noncommunicative diseases (NCD), including obesity, diabetes, cardiovascular disorders and cancer among populations of developing countries as well as lower socioeconomic classes of developed countries, is largely attributed to nutritional and health disorders derived from current human diets [5].

The emerging need for improving human diet has led to increasing interests in traditional nourishments, such as the Mediterranean diet [6] and in primordial therapeutic aids. These are usually based on diverse resources that change through locations, seasons and include numerous wild edible plant species. Attempts to retrieve old beneficial species and recruit them for new commercial use have a vast potential, considering that a very small portion of known plant species has ever been adequately studied for such purposes [7,8]. There is an increasing research activity focused on identifying and characterizing wild plant species with particular attributes to human diet and health [4,9–15]. Special attention is paid to edible plants as sources for essential mineral elements [16] and of antioxidative-active compounds [17]. In addition, there is a rising interest in plants encompassing anti-inflammatory activities [18].

Geraniaceae family includes seven genera and about 830 species distributed from temperate to tropic and arid climates. The largest genera are *Geranium* (430 species), *Pelargonium* (280 species) and *Erodium* (80 species). The family is known for the production of essential oils and ornamentals. Many Geraniaceae species are ascribed to have various medicinal values [19–23]. Additionally, Some of the species belonging to the genus *Erodium* have recognized medicinal indications from folklore and empirical data [24]. *Erodium* species are used to treat a variety of human ailments such as colds, coughs, diarrhea, hemorrhaging and are used to dress wounds [24–28].

Hairy stork's bill (HSB) (*Erodium crassifolium* L'Hér) is a Saharo-Arabian common perennial hemicryptophyte (i.e., buds are at or near the soil surface) that inhabits shrub–steppes of arid southeast Mediterranean regions. The species is distributed from northwest at Crete [29], through few Aegean Sea islands [30], the Libyan [31] and Egyptian [32,33] coasts, north Sinai Peninsula [32,34], Cyprus [35], the Negev Desert of Israel [36], until Edom mountains of Jordan and Saudi Arabian deserts, on southeast [37]. In Israel, HSB can be found in the Negev and Judean deserts, where the annual rainfall is 30–250 mm. The species is most abundant in the stony and arid loess soils and on slopes of limestone hills [36]. Vegetative phase in form of rosette (early season) and 30 cm stems and leaves (late season) starts with the first effective rains. Flowering time is February to May and the flowers are pink to purple with dark color base (Figure 1A,B) that attracts pollinating insects. The ripe ovary splits into five diaspores, each contains one seed covered by the ovary hairy wall with a sharply pointed tip (Figure 1C). The diaspores are carried away from the mother plant by strong winds and react to humidity changes by creating screw-like twists to penetrate the soil [36]. Among all Geraniaceae species, *E. crassifolium* is among the few known to produce tubers (Figure 1D). The tubers are formed on roots at depth of 5–20 cm and are typically small and spherical (1–2 cm in diameter). HSB tubers have a sweet taste and are best in late winter or early spring when they are whitish in color [36]. Traditional knowledge holds that the tubers are edible and Bedouin tribes are their primary consumers [33]. However, no information exists regarding nutritional value of these tubers. Furthermore, in spite of enduring claims associating some medicinal properties with HSB tubers, no supportive documented evidence have been found so far.

**Figure 1.** Hairy stork's-bill (*E. crassifolium*). (**A**) Shrub with flowers and seed pods; (**B**) flower; (**C**) seeds with typical hairy feather-like awns; (**D**) root tubers, connected to the root-shoot transition region (RSTR), the perennial plant organ.

The objective of the present study was to evaluate, for the first time, the nutritive and medicinal potential of *E. crassifolium* for becoming a useful new crop species. In addition to a moderate nutritive potential, we demonstrate here significant in vitro anti-inflammatory capacities in the tubers, assigned to well-known bioactive compounds. Furthermore, HSB displayed impressive tuber productivity when exposed to agricultural conditions, indicating a promising potential for domestication.

#### **2. Results and Discussion**

The domestication of plant species is a complex iterating process, which includes selection according to subjective preferences [38], such as color, palatability and size. In the modern era, however, with the increasing awareness to the potential nutrition benefits or hazards of a given food product, an established list of nutrition facts has become an essential step in the initial evaluation of a candidate species. Here, for the first time, a comprehensive nutrition facts list is presented for HSB tubers (Table 1). The very high water content of HSB tubers makes them highly valuable for nomads in the hot dry desert, supporting the knowledge about their traditional use by Bedouin tribes [33]. Tubers' caloric content is about half than in carrots, but they provide considerable amounts of essential minerals such as calcium, sulfur, magnesium and iron, as well as phosphorus and potassium. In addition, HSB tubers are a good source of vitamins A and C, harboring about a quarter and a third of their contents in carrot, respectively (Table 1).


**Table 1.** Nutritional profile of hairy stork's bill (HSB) tubers compared to that of carrots (adopted from USDA, 2020 [39]).

Compared to carrots, the attractiveness of the HSB tuber in the fresh product market is quite moderate; it is crunchy, but inadequately sweet with no flavor. HSB tubers improve when served cooked, but further culinary efforts are required. As for any wild species, a long course of selection and breeding would be necessary to bring HSB tubers to a status of a common food produce.

Enduring claims associating some medicinal properties with HSB tubers, so far with no supportive documented evidence, have encouraged our curiosity. Electromechanical analysis of the water-soluble extract from HSB tubers revealed significant reducing power, identifying at least six substances or groups of antioxidants (data not shown). Fractionation of the tubers' ethanolic extract (EE) into 11 fractions (Figure 2A) and subsequent in vitro evaluations of possible anti-inflammatory capacities revealed significant activity in fractions F3 and F4, as well as in the original EE and the pooled fractions, PF (Figure 2B).

**Figure 2.** *Cont*.

**Figure 2.** (**A**) HPLC profile of 70% ethanol extract (EE) of HSB tubers at 220 nm. Each fraction (F1–F11) was collected during 5 min out of the total 55 min of HPLC run; (**B**) levels of IL-8, an indicator of cell inflammatory status, in response to treatment with crude EE, EE fractions (F1–F11) and pooled fractions (PF) in in vitro trials using human cells. NT—non-treated control; TNF-α—an inflammation excitatory factor; DXM—dexamethasone (100 μM), positive control. Means of replicates were subjected to statistical analysis using Tukey–Kramer multiple comparison test. Different letters indicate significant differences between treatments; \*\*\*, \*\* and \*, indicate *p* < 0.001, 0.01 and 0.05, respectively. Bars indicate standard error.

Biochemical analyses showed that fraction F4 comprised mainly of epigallocatechin, trans- and cis-catechin and gallic acid (Table 2), all of which are known for their robust bioactive capacities, including anti-inflammatory activity [40–46]. Interestingly, the bioactivity of fraction F4 was significantly greater than that of each compound alone at its corresponding concentration (data not shown), indicating synergic relationships in the natural extract. Furthermore, HSB tubers' extract displayed significantly stronger anti-inflammatory capacity compared to extracts of green tea (*Camellia sinensis*) or turmeric (*Curcuma longa*) [47], well-known sources of antioxidative and bioactive compounds [40,48].


**Table 2.** Compounds identified by GC-MS from fraction 4 (F4) of 70% ethanol extract of *E. crassifolium* tubers. RT—retention time.

Screening several HSB ecotypes collected from diverse locations in Israeli deserts revealed that they all share a similar range of substantial anti-inflammatory capacity; however, there were considerable differences among the ecotypes (Figure 3), indicating that the natural diversity within the species may offer a promising potential for genetic enhancement of the tubers' bioactive capacities. Accordingly, in the recent few years, seeds were collected from the wild across the Negev Desert and parts of the Rift Valley in order to broaden this potential.

**Figure 3.** In vitro indications of the anti-inflammatory activity in the ethanolic extract from tubers of various *E. crassifolium* ecotypes, compared to untreated (NT) and stimulated (TNF-α) human cells. Bars indicate standard error. Different letters indicate significant differences between treatment at *p* < 0.05.

Beyond selection and breeding, optimization of the growth conditions through irrigation, soil fertilization and weeding are founding principles of agriculture [38,49,50] and positive responses of a given wild plant species to these manipulations are prerequisite to successful domestication.

In the wild, HSB is a classic opportunist desert shrub, the proliferation of which largely depends on the current water availability during a given growth season. In arid conditions, the occurrence and intensity of late autumn rain events determine the rate of seed germination and of young seedlings survival. Later on, during winter and early spring, the extents of canopy development, reproductive phase and the duration of the growing season are governed by the intermittent desert precipitation regime [51].

Simulation of various natural scenarios of the precipitation regime from November to May showed that in a relatively dry winter (50 mm), HSB exhibited low seed germination and seedlings survival rates. Conversely, many more plants persisted and grew in a rainy (200 mm) season (data not shown). Interestingly, the effects of water availability on tuber production per plant were very limited within the low precipitation range (Figure 4). However, tuber production and tuber weight increased significantly when water application increased to 700 mm, supplied consistently using drip irrigation, and no problems of germination or seedling persistence occurred. Moreover, plants produced significantly greater number (Figure 4B) of substantially larger tubers (Figure 4A). Adding fertilizer to the irrigated water had only a small influence on the number of tubers (Figure 4B), but tuber size increased by more than 50% (Figure 4A). Thus, application of a basic agricultural practice gave rise to a 10-fold increase in the mean tuber yield of an individual HSB plant (Figure 4C). Finally, the difference in plant and crop performance between wild and agricultural environments is unexpected (Figure 5); the greater germination and survival rates, fortified by enhanced tubers' growth and development have brought about the current yield potential of cultivated HSB to about 15 Mg ha<sup>−</sup>1.

**Figure 4.** (**A**) Mean HSB tuber weight, (**B**) number of tubers and (**C**) mean tuber yield as affected by increased season water application at simulated natural precipitation patterns (50 and 200 mm) and at intensive agricultural environments (700 mm, drip irrigation and drip fertigation \*); data were pooled from two different experiments conducted in the 2017–2018 season. Bars indicate standard deviation.

**Figure 5.** (**A**) Demonstration of an agricultural HSB cropping system at Ramat Negev with (**C**) representative tubers, compared to (**B**) the solitaire HSB phase in a wild niche, with (**D**) typical tubers' size, form and age.

#### **3. Materials and Methods**

#### *3.1. Ecosystem and Plant Material*

The HSB project took place at Ramat Negev Desert Agro-Research Center (RN-DARC), Israel (30◦58 N 34◦42 E), 305 m above sea level. Soil texture varies across short distances from sandy dunes to sandy-loam Loess—and consequently differ in water retention and cation exchange capacity. The natural life cycle of HSB occurs in the rainy season, germinating from November and dispersing seeds up to May. Mean annual precipitation is about 82 mm, but the amount of rain can substantially vary among years and locations, as well as the distribution and intensity of rain events. During the growing season, daily average temperatures decline from 17.8 ◦C in November to about 10 ◦C in January and then steadily rise to 23 ◦C in May (Figure S1).

Most of the trials with HSB employed ecotype RNDARC (Accession No. 309198), which was collected near the trial site several years ago. Additional seed sources were ecotype Revivim (Accession No. 309199), the seeds of which were collected yearly near Kibbutz Revivim and three other accessions (ecotypes 25525, 25493 and 26164), kindly received from the Israeli Gene Bank, ARO.

#### *3.2. Field Experiments*

Two separate field experiments were conducted from November 2017 to June 2018. The first examined HSB growth performance in response to distinct precipitation regimes, whereas the second experiment evaluated it under two different agricultural practices.

#### 3.2.1. Simulated Precipitation Regimes

Treatments represent four principal scenarios of rainy seasons in the Negev Desert of Israel, differing in the intensity and frequency of rain episodes, as follows: A. low and concentrated (50 mm, comprised of two events); (B) low and scattered (50 mm, spread in ten events); (C) high and concentrated (200 mm, in four events); and, D. high and scattered (200 mm, in ten events) precipitation. Seeds were sown in loess soil at a density of eight seeds m−2, in 6 m2 plots. The experiment was organized in a random block design with four replicates. Irrigation was executed using computer-controlled sprinklers, and no fertilizer or soil amendments were used. The amount of water was monitored from germination and included rain. Plants were counted weekly. At harvest in June, tuber number and weight were determined per each surviving plant.

#### 3.2.2. Agricultural Practice

Seeds were sown at late November 2017 in an open field on sandy soil beds. Water was supplied using two drip lines per bed (40 cm apart), with five 1.6 L h−<sup>1</sup> emitters per 1 m of drip line. Two seeds were sown on either side of each emitter, resulting in a density of 16 plants m−2. Fresh water (0.7 dS m−1) was supplied at 6 mm day−<sup>1</sup> until germination, which occurred about 2 weeks after sowing. After germination, irrigation was reduced to 4 mm day−<sup>1</sup> to accomplish 700 mm until late May. An unfertilized control was tested against a fertigated treatment, which was applied using a liquid fertilizer (Shefer 4:2:6, 35% ammonium and 65% nitrate, Israel Chemicals, Ltd.) at 1.5 L m<sup>−</sup>3. The experimental design was of random blocks with four replicates (plot size was 2.4 m2). Bloom began in April and continued until June. Tubers began forming at a very early stage of plant growth and continued to form and develop throughout the season. Tuber harvest took place in June, toward seeds ripening. At harvest, tuber yield was determined; tubers were cleaned and stored at −20 ◦C until further examinations.

#### *3.3. Evaluation of Tubers Nutrition Facts*

To assess HSB tuber nutrition facts (Table 1), samples (2 kg each) were sent to BactoChem, Ltd., Ness Ziona, Israel, an officially certified laboratory, tightly committed to AOAC protocols.

#### *3.4. Ethanolic Extract (EE)*

HSB tubers were removed from cold storage (−20 ◦C) and frozen in liquid nitrogen. The frozen tubers were homogenized using an electrical blender and weighed. For each 1 g of fresh material, 4 mL of 70% ethanol were added immediately to the crushed tubers and incubated overnight at 28 ◦C with shaking at 180 rpm, after which the samples were centrifuged in 50-mL tubes for 5 min at 2500 rpm in an Eppendorf 5810R centrifuge with a 26 cm rotor (1820 RCF). The supernatant was transferred to new tubes and the solvent was evaporated in vacuo overnight. The remaining water content was lyophilized to powder and stored at −20 ◦C. From each gram of tubers, approx. 60 mg of lyophilized extract was obtained.

Just before further analysis, the lyophilized material was weighed and dissolved in 100 μL of 70% ethanol and then 900 μL of double distilled water (DDW), to obtain a 60 mg mL−<sup>1</sup> sample, which was filtered through a 0.45-μm membrane.

#### *3.5. High-Performance Liquid Chromatography (HPLC) Analysis*

The filtered EE sample was separated using an Ultimate 3000 HPLC system coupled with a WPS-3000(T) Autosampler, HPG-3400 pump and DAD-300 detector. The separation was performed using a Purospher RP-18 endcapped column (250 mm × 4.6 mm I.D.; Merck KGaA, Darmstadt, Germany) with a guard column (4 mm × 4 mm I.D.). Solvent gradients were formed by varying the proportion of solvent A [water with 0.1% acetic acid (*v*/*v*)] to solvent B (methanol) with a flow rate of 1.0 mL min−1. Solvent B was maintained initially at 10% for 5 min and then increased to

100% in 25 min. This 100% of Solvent B was maintained for 10 min, then decreased to 10% in 10 min and equilibrated for 5 min (total run time 55 min). The compound peaks were detected at three different wavelengths: 220, 240 and 280 nm. The same program was used to obtain fractions in bulk using a preparative HPLC (11250 Infinity, Agilent Technologies) using reversed-phase C18 column (Kinetex 5u EVO C18-100A—250 × 21.2 mm). After collection, the fractions were lyophilized to powder. These lyophilized fractions were resuspended in 7% ethanol and checked for their effect on IL-8 levels, as described below. Further analyses were carried out to correlate the activity and peak profile for detecting the active compound peak(s).

#### *3.6. GC–MS Analysis*

Prior to GC–MS analysis, samples underwent derivatization; 200 μL of N,O-bis(trimethylsilyl) trifluoroacetamide (BSTFA, Sigma-Aldrich, T-6381, USA) containing 1% of trimethylchlorosilane (TMCS) was added to each completely dried extract and heated to 70 ◦C for 20 min.

GC-MS analyses were carried out using a HP7890 gas chromatograph coupled to a HP6973 mass spectrometer (electron multiplier potential 2 KV, filament current 0.35 mA, electron energy 70 eV and the spectra were recorded over the range *m*/*z* 45 to 1000). An Agilent 7683 autosampler was used for sample introduction. Helium was used as a carrier gas at a constant flow of 1.1 mL s–1. One μL of each sample was injected into the GC–MS using a 1:10 split ratio injection mode. An isothermal hold at 50 ◦C was kept for 2 min, followed by a heating gradient of 6 ◦C min−<sup>1</sup> to 300 ◦C, with the final temperature held for 4 min. A 30 m, 0.25 mm ID 5% cross-linked phenylmethyl siloxane capillary column (HP-5MS) with a 0.25-μm film thickness was used for the separation, and the injection port temperature was 220 ◦C. The MS interface temperature was 280 ◦C. Peak assignments were carried out with the aid of library spectra (NIST 14.0) and compared with published data and MS data obtained from the injection of standards [(-)-Epigallocatechin, 08,108; catechin, *U-*49,040; gallic acid, 91,215] purchased from Sigma-Aldrich, Switzerland or USA.

#### *3.7. Human Cell Culture*

HaCaT is a spontaneously transformed aneuploid immortal keratinocyte cell line from adult human skin [52], widely used in scientific research [53]. HaCaT (ATCC-HB-241) normal skin cells were grown at 37 ◦C in a humidified 5% CO2–95% air atmosphere. Cells were maintained in Dulbecco's modified Eagle's medium (DMEM, Biologic Industries, 01-055-1A, Beit-Haemek, Israel) with 10% fetal bovine serum (FBS, Biological Industries, 04-007-1A, Israel) and penicillin (100 units mL−1)—streptomycin (100 μg mL<sup>−</sup>1) solution (Biologic Industries, 03-031-1B, Beit-Haemek, Israel).

Determination of Interleukin 8 (IL-8) levels in HaCaT cells

HaCaT cells were seeded into 24-well plates at 50,000 cells per well in triplicate in 500μL of media and then incubated for 24 h at 37 ◦C in a humidified 5% CO2–95% air atmosphere. After incubation, cell excitation was performed with recombinant human tumor necrosis factor-α (TNF-α, PeproTech, 300-01A, Cranbury, NJ, USA). Cultures in each well were treated with a final concentration of 50 ng mL−<sup>1</sup> of TNFα and 50 μL plant extract. Three different controls were included in all experiments: (a) untreated cells, with neither TNF-α nor plant extracts, (b) cells treated with TNF-α alone, (c) cells treated with TNF-α and the solvent (7% ethanol). The supernatant was taken, and the level of IL-8 was measured 16 h posttreatment using the commercial Human CXCL8/IL-8 DuoSet ELISA kit (R&D Systems, DY208-05, McKinley Place MN, USA) according to the manufacturer's protocol. IL-8 is a common biomarker for inflammatory skin diseases [54,55]. The induction of detectable IL-8 levels requires a 16 h exposure to TNF-α [56]. Dexamethasone (Sigma-Aldrich, D4902, St. Louis, MO, USA) was used as a positive control [57,58].

#### **4. Conclusions**

Wild plant species may hold immense potential resources of nutritional value and therapeutic substances. In the present study, the tubers of *E. crassifolium*, an ignored desert shrub, were shown to harbor significant nutritional values and anti-inflammatory capacities. The catechins found in the tubers' ethanolic extract have well-established remedial effects in serious human ailments. Furthermore, indications exist suggesting that HSB tubers have a greater medicinal potential. While a demonstrated bioactive capacity is a prerequisite, high productivity is essential in realizing the potential of an underutilized species. The dramatic increase in HSB tubers yield in response to fertigation paves the way for this plant to become a potential industrial crop.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2223-7747/9/9/1069/s1, Figure S1: Mean monthly maximum, average, and minimum temperatures at RNDARC.

**Author Contributions:** Conceptualization, S.C., O.G. and H.K.; methodology, A.B., O.G., G.S. and M.S.; investigation, O.G., H.K., M.S., G.S. and M.M.; formal analysis, G.S., M.M., M.S. and O.G.; writing—Original draft preparation, A.B., H.K. and O.G.; writing—Review & editing, A.B., M.S., O.G. and S.C.; funding acquisition, O.G. and H.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by KKL-JNF, the Israeli Ministry of Agriculture and ICA-Israel.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Punica protopunica* **Balf., the Forgotten Sister of the Common Pomegranate (***Punica granatum* **L.): Features and Medicinal Properties—A Review**

**José Antonio Guerrero-Solano 1, Osmar Antonio Jaramillo-Morales 2,\*, Tania Jiménez-Cabrera 1, Thania Alejandra Urrutia-Hernández 3, Alejandro Chehue-Romero 1, Elena G. Olvera-Hernández <sup>1</sup> and Mirandeli Bautista 1,\***


Received: 1 September 2020; Accepted: 12 September 2020; Published: 16 September 2020

**Abstract:** *Punica protopunica* Balf. is one of only two species housed by the *Punica* genera. *Punica protopunica.* Balf., known as Socotran pomegranate, is an endemic, isolated species found only in Socotra archipelago in the northwestern Indian Ocean, and is considered to be the ancestor of pomegranate. This review stems from the fact that in many *Punica granatum* L. articles, *Punica protopunica* Balf. is mentioned, but just in an informative way, without mentioning their taxonomic and genetic relationship and their medicinal properties. It is there where the need arises to know more about this forgotten species: "the other pomegranate tree." A large part of the human population does not know of its existence, since only its "sister" has spread throughout the world. The present review deals with the taxonomy and origin of *Punica protopunica* Balf., the morphology of the tree, distribution, cultivation, vulnerability, and as well as its relationship with *Punica granatum* L. It also discusses its uses in traditional medicine, its antioxidant capacity, and the medicinal properties of this forgotten species.

**Keywords:** *Punica protopunica* Balf.; *Punica granatum* L.; *Punica* genera; *Lythraceae*

#### **1. Introduction**

Myrtales is an order within the classification of terrestrial green plants (Viridiplantae-Streptophyta) [1,2]. The Lythraceae family (from the order of the Myrtales) is composed of herbs, shrubs and trees that are mainly recognized for their flaky bark, crumpled petals from the bud (when emerging out of the rim of the calyx tube of the sepals), leaves oppositely paired, seeds with multi-layered outer integuments, and the fruit is usually a capsule [1–4]. The Lythraceae family comprises 31 genera including the *Punica* genera [5]. This is quite surprising as the *Punica* genera has previously been assigned to the monogenic *Punicaceae* family [6,7]. However, the results of numerous molecular and morphological investigations revealed the close relationship of the genera *Punica* with the Lythraceae family [1,8–13]. Previously it was considered a monogenic Punicaceae family that contains only one genera, *Punica* [14,15]. According to Pliny, the name *Punica* was given by the Romans, referring to the city of Cartago, in Tunis (Punic, Phoenician, Carthaginian), from where the best pomegranate (from Latin "*pome*" witch means apple and "*granate*" meaning many seeded) arrived in Europe. The genera *Punica* contains two species, *Punica granatum* L. and *Punica protopunica* Balf., [16]. Initially, *Punica granatum* L. was known as *Malum punicum*, the apple of Cartago, but later, Carl Linnaeus (1707–1778) chose the current name, with a specific epithet of *granatum*, which means granular [6]. On the other hand, *Punica protopunica* Balf. was first described by the Scottish botanist Isaac Bayley Balfour (1853–1922) during his arboreal and botanical expedition in 1880, and published in the Proceedings of the Royal Society of Edinburgh in 1882 [4,14–16]. *P. granatum* is native to the region that covers territories from a part of Iran to northern India [17,18]. Wild *P. granatum* L. types have their natural distribution in central Asia from Iran, Afghanistan, Turkmenistan, to northern India, and this region is considered the center of origin of pomegranate [6]. Later, the pomegranate was distributed to the Mediterranean, East Asia, America and South Africa, and this distribution originated the genetic diversity of the pomegranate [19], on the other hand, *P. protopunica* Balf. is endemic to the Socotra archipelago (located between the Arabian sea and the Guardafui channel in the Indian Ocean, off the coast of the Horn of Africa) [20]. In this context, the objective of this review was to compile the available information on the *P. protopunica* Balf. species (morphology, distribution, cultivation, vulnerability, uses), including its antioxidant capacity and the medicinal properties, to make it known and allow a wider use of this forgotten wild species.

#### **2. Results**

#### *2.1. Taxonomic Positioning and Distribution*

*P. protopunica* Balf. (taxonomic positioning in Table 1) is an endemic species, found only in the remote archipelago of Socotra, and is considered as one of the most important endemic species on the archipelago [21,22]. Socotra belongs to the Republic of Yemen, it is located at 12◦19'–12◦42' N and 53◦18'–54◦32' E, on the Arabian sea of the Indian Ocean [21,23]. Socotra archipelago, also known as the "Galápagos of the Indian ocean" is a group of four islands, Socotra being the most important and largest one. Socotra archipelago (isolated from the rest of the world) is an island with great biological diversity (approximately 900 plant species, 30% endemic) and it hosts unique fauna and flora [24]. People of Socotra use medicinal plants and it is known that this people has a deep respect for nature and its environment [24]. Socotra was included in 2008 to the select list of World Heritage by the United Nations Educational, Scientific and Cultural Organization (UNESCO) under the criteria of natural site [25]. Additionally, *P. protopunica* Balf. is considered from an independent evolutionary path (due to isolation from the rest of the world) [26], commonly considered as "the other pomegranate tree", it is an unknown species worldwide, in contrast to the sacred status granted by the name of its species, *protopunica*: prototype [27].



Adapted from [28,29].

*P. protopunica* Balf. is distributed in different regions of the island, but mainly grows in the central-western part of Socotra, in humid forest regions, in the Haggeher mountains whose slopes are characterized by being made of granite and having a maximum elevation of 1503 m. It also grows on Diksam, the island's limestone plateau, which rises precipitously 1520 m above sea level. The total area covered has been calculated to be 1/15 of the total area of Socotra (3796 km2) [27].

#### *2.2. Morphology*

Balfour, the discoverer of *P. protopunica* Balf., described it as "trees with branches, often thorny; elliptic leaves round sheath, oblique; below the oblong, obtuse flower bracts; obovate petals; joined carpels, horizontal basal tone center spiral. From Socotra, a new species that abounds and grows on the peninsula" [30]. Additionally, Balfour wrote: "In general habit, it is not unlike the pomegranate, but its leaves are larger and coarser, and it wants the delicate character of the foliage of that species. The flowers, too, are somewhat smaller, and their turbinate base is more angular. The fruit is very much less in size" [31]. *P. protopunica* Balf. has morphological differences compared to *P. granatum* L.; it has larger, narrower leaves, different foliage, continuous flowering, and smaller, pink (not red) flowers. The fruit of *P. protopunica* Balf. is round, pommel-shaped, with a maximum diameter of 3 cm and a characteristic yellow-greenish or red-brown color when ripe, is smaller, evergreen, with white seeds and less sweet than *Punica granatum* L. [20,30–33]. Table 2 shows morphological characteristics of the species.


**Table 2.** Morphological characteristics of *P. protopunica* Balf.

Adapted from [20,26,27,29,31].

#### *2.3. Cultivation*

A hardiness zone is a geographical area defined in the quality of a specific category of plant life capable of growing. The most widely used system is that of the United States Department of Agriculture (USDA), which includes 13 zones characterized by annual extreme minimum temperatures. The use of this system has spread throughout the world and has been adapted in other countries. Using this system, *P. protopunica* Balf. has USDA hardiness zone 7a through 11b; from −17.8 ◦C to +10 ◦C [20], which means that, emulating optimal conditions, it can be grown in other regions of the world.

*P. protopunica* Balf. can grow out of the island of Socotra, however certain requirements must be met that are necessary for its growth: (1) temperatures above 10 ◦C; (2) sunlight from 6 to 8 h per day and protection against the wind; (3) constant accumulation of humidity (1000 mm) with a percentage of 20% to 40%; and (4) alkaline soils with pH 7 and content of calcareous or rocky gravel. In cultivation, sowing, cutting and grafting can be methods of propagation. Botanist Alan Radcliffe-Smith from the Kew Royal Botanic Gardens Herbarium (1938–2007) successfully propagated *P. protopunica* Balf. using all three methods, although cutting and grafting did not result in fruit in the varieties [22,27].

#### *2.4. Vulnerability and Conservation of the Species*

As early as 1978, *P. protopunica* Balfwas considered a vulnerable species with fragmented populations. Later in 2004 [34] the red list of the International Union for Conservation of Nature (IUCN) specified that *P. protopunica* Balf. was a vulnerable species; It indicated that efforts should be made to protect *P. protopunica* Balf., since it is the only congener of *P. granatum* L., although, according to Miller [29,34], this information is outdated, since it has been shown that *P. protopunica* Balf. is widely found on the island of Socotra and is quite common in some regions. Its total area of occupation is approximately 100 km2, 2/15 of the total area of Socotra. Miller also reported that *P. protopunica* Balf. has a fragmented distribution with different subpopulations and regenerates actively, however, there are large areas in which the tree does not grow, except in some areas with remaining populations without regeneration being observed. The foliage of *P. protopunica* Balf. is of no interest to livestock and the tree is not cut down for fuelwood, even in dry spells. Wood is not important neither as firewood nor for construction.

Socotra was included on the World Heritage List in 2008 as a natural site, which has had a positive impact on the conservation of its species [35]. The inclusion was under the selection criteria number 10 that establishes: "to contain the most important and significant natural habitats for in-situ conservation of biological diversity, including those containing threatened species of outstanding universal value from the point of view of science or conservation". In its operational guidelines, UNESCO requests that in order to maintain this status, the assets and properties included on the list must be protected by well-established and delimited legislation and institutional regulations, to guarantee their safeguarding. Likewise, the states must demonstrate that actions are carried out to protect them at the national, regional and local levels, and must attach the appropriate texts to the nomination with a detailed explanation of how this protection operates [36]. So the archipelago species, including *P. protopunica* Balf., are now more protected than ever, however, UNESCO has detected some factors that affect the property, such as livestock farming/grazing of domesticated animals, management plan, uncontrolled developments including ground transport infrastructure: the road network, absence of biosecurity politics to eradicate the introduction of invasive species, extreme weather events (storms and cyclones) and industrial activities among others [37,38]. There is a development and conservation program for the archipelago called: Socotra Archipelago Conservation and Development Program (SCDP). This is an initiative of the Republic of Yemen to develop and conserve the archipelago's resources. The SCDP is supported by the United Nations Development Program (UNDP) and the governments of three European countries (Italy, Poland and the Netherlands), in conjunction with international donors and private non-profit organizations. The mission of the SCDP is to join the efforts of all the aforementioned organizations and countries for the human development of the population and conservation of the biodiversity of the Socotra archipelago with a sustainable approach [39].

#### *2.5. Relationship between P. protopunica Balf., and P. granatum L.*

In 1973, Shilkina reported that Socotran pomegranate tree wood contains fiber tracheids, (water-conducting cells). She found that tracheids are the only characteristic that differentiates *P. protopunica* Balf. from the rest of botanical and arboreal varieties of the order Myrtales. She stated that this unique characteristic is not shared by even *P. granatum* L. Therefore, she urged that *P. protopunica* Balf. be not only the only member of its species, but also the only species of a new genera. It appears

that, based on the anatomy of the xylem, *P. protopunica* Balf. was suggested as an ancestor of the domesticated species *P. granatum* [7]. On the other hand, in *P. protopunica* Balf., 2n = 14, therefore, the haploid number of chromosomes is n = 7, unlike n = 8 in *P. granatum* L.; this difference is considered as a primitive characteristic of *P. protopunica* Balf. from the evolutionary point of view, since n = 8 is a development factor [40].

Recently, Youssef et al., [33] analyzed genetic diversity and the relationship between *P. protopunica* Balf. and eleven accessions of *P. granatum* L. (from Egypt, México and Yemen), using the following molecular markers: (1) amplified sequence-related polymorphism (SRAP); (2) amplification polymorphism of the target region (TRAP); (3) amplified intron-directed polymorphism (ITAP); and (4) sequence analysis of the *pgWD40* gene (involved in anthocyanin biosynthesis). It was found that the relationship between accessions of *P. granatum*, grouped by regions, was approximately 90% similar, while, evidently, the degree of genetic variation was altered within of each region. However, these markers revealed the relationship between *P. protopunica* and *P. granatum* at 33% similarity. ITAP, TRAP and SRAP generated a total of 719 bands, of these, 193 were specific for *P. protopunica* Balf., and 234 bands were shared between both species. The pgWD40 gene analysis showed 100% identity between *P. granatum* L. accessions and 98% with *P. protopunica* Balf. Phylogenetic analysis of the WD40 sequences of species, including both species of the *Punica* genera confirmed the relationship between *P. protopunica* Balf. and *P. granatum* L., supporting the hypothesis that *P. protopunica* Balf. could be an ancestor of *P granatum* L.

Moreover, Muhammad et al., [41] studied the genetic association among the genotypes of the species of *Punica* (20 genotypes of *P. granatum* L., 20 of *P. protopunica* Balf.), based on morphological and biochemical characterization. The phylogenetic tree was constructed with the 40 genotype data matrix based on morphology to represent the similarity of the two species. The phylogenetic tree divided the two species in two lineages (Regions R-I and R-II). R-I holds the 20 genotypes of *P. granatum* L., while R-II enclosed all genotypes of *P. protopunica* Balf. The similarity indexes were performed for the genotype of 2 species that was 53.84% for *P. protopunica* Balf. and *P. granatum* L. In the biochemical characterization, the total seed protein profiling was carried out on slab gel electrophoresis; 10 bands were recorded in both species (molecular weight 15 KDa–180 KDa) intra locus contribution toward the genetic disagreement was 10% in *P. protopunica* Balf. and 50% in *P. granatum* L. Inter species locus contribution toward genetic diversity was 50%.

#### *2.6. Uses in Traditional Medicine*

Despite the great scientific advancement in medicine and pharmaceuticals, much of Yemenis actively practice traditional medicine, they use medicinal plants for their daily health care needs and have a long tradition in herbal medicine. The vegetation and flora of Socotra provide healers with "natural pharmacies" with a great variety of plants, to prepare phytomedicine, in order to alleviate a great variety of human and veterinary diseases [42]. Some authors have reported the use of *P. protopunica* Balf. fruit peel, seed and flower in traditional medicine; the extraction and consumption techniques are by decoction, boiling, infusion, maceration of ethanol and fresh juice. The above mentioned are helpful for treating diseases such as peptic ulcer, diarrhea, dysentery, sores and wounds, urinary infections, dry cough, digestive problems, skin disease, mouth and throat sore and jaundice. It is also used because of the anthelmintic and anti-diabetic properties [29,34,41,43–46].

#### *2.7. Bioactive Compounds of Pomegranate and Their Medicinal Properties*

#### 2.7.1. Phenolic Content and Antioxidant Activity

Muhammad et al., in 2019 [41], evaluated the antioxidant potential of methanolic extracts of *P. protopunica* Balf. and *P. granatum* L. species, both cultivated in Swat Valley, KP, Pakistan. They found high amounts of total phenols in both species, as well as flavonoids and antioxidant activity, with *P. prototunica* Balf. showing the highest flavonoid content. Antioxidant activity was

similar between species. The other study was led by Al-Huqail et al., 2018 [47]. They studied the antioxidant effect of the aqueous ethanolic extracts of the peel and seed coat of *P. granatum* L. and *P. protopunica* Balf., in vitro. The two extracts not only contained significantly different phenolic and total flavonoid contents but also different phytochemical constituents. Gas chromatography mass spectroscopy (GC-MS) analysis of the peel extracts revealed twenty-six compounds. The main ones were benzenepropanoic acid, 1H-pyrrole-2,5-dione, 1,2-benzenedicarboxylic acid, 1-(propylthio)-(CAS) ethanol (CAS) ethylalcohol methyl ester of 3-methoxypropionic and 2-propanol. *P. protopunica* Balf. seed coat extract showed the presence of 14 phytochemical constituents, the major constituents were Di-2 (2-ethylhexyl) phthalate, 1,2-benzenedicarboxylic acid, propanoic acid, 2-hydroxy-ethyl formic acid and benzoic acid. In the malondialdehyde method (MDA), hydrogen peroxide (H2O2) scavenging and DPPH· assays, the two seed coat extracts exhibited very high antioxidant activities, with higher activity observed for the *P. granatum* L. extract [47]. These differences in the antioxidant activity in the two species may be attributed to their different phytochemical constituents. The importance of the high concentration of phenolic compounds is that they protect cells from the damaging effect of free radicals, molecules responsible for altering biological systems, causing diseases or accelerating aging [48].

#### 2.7.2. Antimicrobial, Antiviral and Antiprotozoal Activity

Mothana and Lindequist [49] evaluated the antimicrobial effect of twenty-five medicinal plants of the island of Socotra, including the fruit and leaves extracts of *P. protopunica* Balf. (4 mg of the dried extract), on nine types of Gram-positive and Gram-negative bacteria: *Bacillus subtilis* (ATCC 6059), *Candida maltosa* (SBUG), *Escherichia coli* (ATCC 11229), *Micrococcus flavus* (SBUG 16), *Pseudomonas aeruginosa* (ATCC 27853), *Staphylococcus aureus* (ATCC 6538), multi-resistant strains *Staphylococcus aureus*, *Staphylococcus epidermidis* 847, and *Staphylococcus haemolyticus* 535 and against a species of yeast. The methanolic extract of *P. protopunica* Balf., was found to be one of the species with the highest antimicrobial activity, especially on Gram-positive bacteria including multi-resistant strains of *Staphylococcus*, but without activity on yeast.

The antiviral activity of the methanolic and aqueous extracts of twenty-five medicinal plants including *P. protopunica* Balf. fruit and leaves, were evaluated in two in vitro models (unreported concentrations) by Mothana et al., [50], one with MDCK cells with type A influenza virus and the other with Vero cells infected with herpes simplex virus type 1 (HSV-1). HSV-1 was more sensitive than type A influenza against the extracts evaluated. The half maximal inhibitory concentration (IC50) for *P. protopunica* Balf. was anti-influenza virus A = 75.7 μg/mL and anti-HSV-1 = 5.8 μg/mL. The species was not considered by the authors as a plant with potential for the development of antiviral drugs.

Additionally, Mothana et al., [45] evaluated the in vitro antiprotozoal activity of twenty Socotra plants including the methanolic extract of the fruit of *P. protopunica* Balf., (at 5 concentrations: 0.25, 1, 4, 16, and 64 μg/mL), *Plasmodium falciparum* erythrocytic schizonts were used to evaluate antiplasmodial activity. The antileishmanial activity was evaluated using a model of intracellular amastigotes of *Leishmania infantum*, and finally the antitripanosomal activity was evaluated using intracellular amastigotes of *Trypanosoma cruzi* and free trypomastigotes of *T. brucei*. The results indicated that there is selective activity of *P. protopunica* Balf., against *Plasmodium* (IC50 2.2 μg/mL), and a potential for relevant antileishmanial and antitrypanosomal activity was also found. In the same line of research, Barzinji et al. [44] investigated the antimalarial efficacy in vitro of methanolic and aqueous extracts of thirteen traditionally used plant species from Yemen, including *P. protopunica* Balf., in blood samples from positive malaria patients. The methanolic extract from *P. protopunica* Balf. (20 mg) was one of the three extracts with the highest antimalarial activity (IC 0.98 μg/mL), and also exhibited schizont maturation inhibition (SMI) of 31.25 μg/mL [49].

#### 2.7.3. Anticancer Activity

In 2007, Mothana et al., published the study "anticancer potential of Yemeni plants used in folk medicine". Twenty-four methanolic extracts of common plants in traditional medicine in Socotra and other Yemen states were evaluated. *P. protopunica* Balf., (leaf and fruit extracts) was included. They used a microtiter plate assay based on cell staining with violet crystal to evaluate the in vitro cytotoxic potency of the extracts at different concentrations, with 5 human cancer cell lines: two urinary bladder carcinomas (5637 and RT-112), two lung cancer line (A-427 and LCLC-103H) and a breast cancer line (MCF-7). The methanolic extracts of *P. protopunica* Balf. exhibited a moderate potency of toxicity (seventh place of the extracts evaluated) in all tumor cell lines with IC50 values of 16.5 to 37.6 μg/mL. The methanolic extract of *Dendrosicyos socotrana* had the greatest cytotoxic effect against all the cancer cell lines analyzed [51].

#### 2.7.4. Cytotoxicity

In some of the studies cited above, cytotoxicity tests were carried out. Cultured MRC-5 SV2 cells were used for the toxicity test for *P. protopunica* Balf. fruit and leaves extracts. Cell viability was evaluated fluorimetrically. Fluorescence was measured and cell viability results were compared versus control group (data were expressed as a percentage reduction in cell viability). The IC50 for *P. protopunica* Balf. was 29.5 ± 3.7 μg, which is interpreted as low toxicity by the authors [45]. A cytotoxicity assay on the proliferation of MDCK and Vero cells was also carried out using culture plates and incubated at 37 ◦C with 5% CO2. Confluent monolayers were incubated with dilutions of extracts (100, 50, 25, and 12.5 μg/mL) in culture medium for 3 days. The 50% inhibitory cellular concentration (ICC50) was measured by a dye absorption assay, with culture medium as a control. The ICC50 of the pomegranate was less than 10 μg, which was considered non-cytotoxic [50]. With the little information available, it cannot be stated that there is no toxicity of the tree and its components, however the trials point to that conclusion.

#### **3. Discussion**

Socotra Archipelago is known to be a section of what was once the Gondwana Supercontinent [52, 53]. The biodiversity and geology of the island of Socotra are living and tangible proof of the historical biogeography of the supercontinent of Gondwana [54]. In the early Cretaceous, Gondwana included the continental areas of Africa, South Asia, Polynesia and Central America, which later separated to create the continents as we know them today. Subsequently, the 4 islands of the Socotra archipelago, was isolated from the Indian Ocean approximately 20 million years ago, when the African and Arab plates separated, which resulted in the formation of the Gulf of Aden [52,55]. At some point in that transformation, the Socotra archipelago was left behind, isolating all the living matter contained in that piece of land. It is known that Socotra is not an island formed by volcanic lava but is a separate portion of a continent, so the endemic species are really old and "rare" in the eyes of modern humans. The dragon's blood tree *Dracaena cinnabari* Balf. and Socotran *Adenium obesum* tree are clear examples, that look like they came out of a science-fiction novel. This is how many of the species that have been home to the island for millions of years had a different evolution, with fewer changes than in other lands, hence the genetics of *P. protopunica* Balf.

The morphological differences between *P. protopunica* and *P. granatum* are evident, but the most notable are the size of the tree, which can be more than 7 m in natural conditions for *P. granatum* L. [41], as opposed to the 4.5 m (average height) for *P. protopunica* Balf. [20]. The flowers are orange to red in *P. granatum* L. and smaller bright pink in *P. protopunica* Balf. [27,56], and the color of the fruit peel in *P. granatum* L. can be yellow, reddish yellow, or different shades of green and red (starting from pink to crimson); purple is less common and there are even unique varieties, such as black pomegranate, which acquires its coloration from immaturity and remains so until overripe [40,57]. On the other hand, as we already mentioned, *P. protopunica* Balf., has a fruit that ripens from green to greenish yellow, which can have a dark pink hue and they are in continuous bloom [23,29].

As we have seen, *P. protopunica* Balf. is endemic to Socotra and the distribution is centered on the island, however one of the studies we reviewed used specimens of *P. protopunica* Balf. from Pakistan (it was introduced to this region) [41]. It has been reported that with special care, the species can be cultivated outside the region where it grows naturally [22], making it a viable option for the species to be cultivated in other regions of the world, and taking into account vulnerability reports in Socotra and thus preserve it in a better way. In addition, efforts have been made to collect, conserve and evaluate the germplasm of *Punica* species [58]. There are collections of germplasm of wild and domesticated varieties of pomegranate in gene banks and seed banks of Albania, China, Cyprus, Egypt, France, Germany, Greece, Hungary, India, Iran, Israel, Italy, Morocco, Portugal, Spain, Tunisia, Turkey, Turkmenistan, Ukraine, USA and Uzbekistan [59–61]. However the largest collection is located in St. Petersburg, Russia. Interestingly, there was a collection considered the largest, in Garrygala Turkmenistan, but Levin reported that it was destroyed when Turkmenistan separated from Russia [40].

The domestication process of *P. granatum* L. gave rise to fruits and plants with magnum seeds, some infertile seeds and fruits, as well as fruits and seeds of different shades of color [58]. Chandra et al. (2010) [19] gave us a detailed history of the pomegranate, explaining that this was from the first fruit crops to be domesticated and planted in the years 4000 and 3000 BC., being one of the oldest edible first fruits [19,62]. It is known to have been cultivated in Egypt and consumed in India (it was an important food in Indian royalty) so early that there is a Sanskrit word for pomegranate. There are also records of its consumption in China during the Han and Sung dynasties, carried from the Middle East by merchants. It was adopted and consumed regularly in medieval Europe and spread around the world in European conquests [60].

Due to the globalization of pomegranate cultivation, there are genetic variations within the same species. More than 500 varieties of *P. granatum* L. are known, although few varieties bring their cultivation to a commercial level of production (about 50) [63]. In contrast, *P. protopunica* Balf. has a smaller wild fruit, a lesser variety of colors, and an acidic flavor that makes it an inedible fruit. There is a great polarization when referring to the two species of the *Punica* genera, on the one hand, there is the common pomegranate, widespread throughout the world, reported as a super fruit in any recipe magazine article (*P. granatum* L.). The other is the Socotran pomegranate (*P. protopunica* Balf.), which is little known, hidden from the eyes of the world and only used by the people of Socotra for medicinal purposes, the one without admirers.

*P. protopunica* Balf. is one of the only two species of the *Punica* genera, being considered as the "sister" of *P. granatum* L. [14]. However, according to its origin (due to its independent evolutionary line and which seems to be an ancestor of the *Punica* genera), we think that it may be, beyond the taxonomic classification, the "grandmother" of *P. granatum* L.

*P. protopunica* Balf. has strong ties to *P. granatum* L., and a strong relationship with the flora of the adjacent continental areas of Arabia and Northeast Africa, tropical Africa, Madagascar, India, South Asia, Polynesia and Central America that, as already mentioned, were united in the Cretaceous. It is believed that at least since the late Cretaceous, much of Socotra was emerged, considering itself one of the longest isolated landmasses on earth. Its vascular plants have an endemicity index, quite similar to that of other islands such as the Canary Islands, and it is a refuge for interesting paleoendemisms of very ancient origin, including the case of *P. protopunica* Balf. [64].

Herbalism is one of the most used treatments in traditional Yemeni medicine, predominately in rural territories, herbalism is practiced by Yemen population to all kind of ailments [65]. The uses of *P. protopunica* Balf. in traditional medicine agree with the uses that have been reported in *P. granatum* L., although *P. granatum* L. has more uses and has had more effects attributed to it. *P. granatum* L. is traditionally used for diarrhea, stomatitis, ulcers, bleeding, enemas, vaginal discharge, inflammation of the pancreas, gallbladder diseases, dysentery and stomach disorders, antiparasitic (taenicide and others), antibacterial, inflammatory diseases, astringent, abortion, burns, pain, snakebite, bronchitis, cough, and nausea [66–70]. Traditional and alternative medicine has many followers in the Republic of Yemen, because access to occidental medicine is still restricted, and Yemen is a country with difficult access [71].

Pomegranate is a source of bioactive compounds, present both in the fruit (peel and arils), and in the leaves and bark [72]. Ozgen et al. [73] affirm that pomegranate is a fruit rich in phenolic antioxidants, specifically anthocyanins, but their content varies between varieties, or sub-species. The fruit has a large number of flavonoids, it is estimated that about 0.2% to 1% of the weight of the fruit represent this group of compounds, of which about 30% of all the anthocyanidins are in the peel [74]. Pomegranate juice has a high content of polyphenols, significant amounts of ellagic acid, caffeic acid, chlorogenic acid, coumaric acid, catechins, ferulic acid and a large list of anthocyanins [75]. The literature indicates that pomegranate contains 124 different compounds and that among these phytochemicals, high molecular weight polyphenols (such as ellagitannins and punicalagin) are likely to mediate most of the fruit's protective effects against harmful agents [75]. In the peel, almost 48 phenolic compounds have been identified [76]. Some authors identified and quantified phenolic compounds with more than 50 varieties of pomegranate fruits, using high-performance liquid chromatography with electrospray ionization and mass spectrometry (HPLC-DAD-ESI/MSn), they concluded that ellagitannins were the most abundant compounds in all the investigated samples (mesocarp, peel, arils, juices), and all the varieties had ellagitannins and anthocyanins [77,78]. All these data were taken from studies in *P. granatum* L. It can be hypothesized that these compounds are also present in *P. protopunica* Balf., however there is a large information gap in this regard. Hundreds of studies have been conducted in which the antioxidant capacity and total phenols of *P. granatum* L. are evaluated, however, for *P. protopunica* Balf., the scenario is different. It is curious that there are not enough published studies of the content of total phenols and antioxidant effect of *P. protopunica* Balf. Nor did we find studies of the nutrimental and chemical composition of the fruit. This may be due in the first place to the isolation of the species, and also due to the difficulty in obtaining government permits, as it originates from a declared a World Heritage Site. Perhaps there are unpublished works.

Many of the uses in traditional medicine have been explained and demonstrated with a great diversity of in vitro and in vivo studies. For example, in *P. granatum* L. hypoglycemic, hypocholesterolemic, hypotriglyceridemic, antihypertensive, anti-atherosclerotic, anti-inflammatory, against metabolic syndrome, various types of cancer, antimicrobial and antifungal, healing, among others effects were reported [79–94]. However, not enough reports of biological effects of the *P. protopunica* Balf., species were found. The antimicrobial, antiviral, antiparasitic and anticancer activity found in *P. protopunica* Balf. has also been found in *P. granatum* L., and many of the mechanisms of action reported by the authors are related with the high concentration of bioactive compounds of secondary metabolism of the fruit, specially ellagic acid, gallic acid, punicic acid and flavonoids as quercetin and kaempferol [45,77,82,85,87,94–102]. Finally, the consumption of *P. granatum* L., including the peel and edible parts, as well as its extracts, are considered safe in vitro and in vivo [103–109], and these results agree with the non-cytotoxicity results of *P. protopunica* Balf., however, more studies in *P. protopunica* Balf. are required to conclude that.

The data provided in this review lead us to suggest that this poorly studied species may have similar pharmacological effects like *P. granatum* L., because it belongs to the same genera and the probability that they share the same active compounds, although perhaps in different proportions. So its pharmacological properties can be better, or just the opposite.

#### **4. Materials and Methods**

#### *4.1. Search Criteria*

A search for information on the subject was made, using as inclusion criteria all the articles, books and official web pages published to date (June 2020). In this study, published peer-reviewed articles without language restrictions (unpublished data were not included) in repositories such as PubMed, ScienceDirect, Worldwide science, Springer link, Refseek, SciELO, Cochrane Library were considered as primary sources. Secondary sources of information included data from official web pages and books of taxonomic, botanical, cultivation and vulnerability information of the *P. protopunica* Balf., species with the keywords: *Punica protopunica* Balf., wild pomegranate, Socotra pomegranate, *Punica granatum*

L., *Punica* gender, Lythraceae, Myrtales, *Punica protopunica* taxonomy, health effects, biological effects, composition, chemistry, cultivation, distribution and morphology.

#### *4.2. Data Extraction*

The data used for this article were collected, classified, summarized, analyzed, compared, discussed and written. The conclusions were made accordingly. The data extracted from each study were extracted and reported using a thematic and subtopic analysis, the results were also compared with *P. granatum* L., as a reference species.

#### **5. Conclusions**

In summary, this review supports the idea that the *P. protopunica* Balf. species can be considered as a powerful source of different pharmacological activities. Taking into account that it has been widely reported that this species exhibits antiviral, antimicrobial, antiprotozoal, antioxidant and anticancer activities. It was shown that there is a relationship between *P. protopunica* and *P. granatum*. It is not entirely clear whether the species remains vulnerable, but there are efforts to conserve it. The presence of various bioactive compounds could justify their effect, however, it is necessary further studies to demonstrate its pharmacological actions and their possible adverse effects. Additionally, it is essential to consider clinical studies to establish adequate doses in humans to evaluate the bioactive compounds of *P. protopunica* Balf., whereas the use of traditional medicine and complementary or alternative medicine are increasingly used in developing and developed countries.

#### **6. Prospects**

There is a wide range of possibilities for studying this species. In biomedical sciences, more information is needed on the effect of *P. protopunica* Balf. on human health. In vitro and in vivo models in various pathologies are necessary. It is suggested to delve into the mechanisms of action of the species in each pathology studied. Studies of the chemical composition and evaluation of its bioactive compounds (tree and especially of the fruit) are necessary, and compare these results with the composition of *P. granatum* L. Clinical studies are considered necessary to corroborate the effects reported in traditional medicine, once it has been reported that there is no toxicity. It is suggested to use other compound extraction methods and the use of high, medium and low polarity solvents.

On the other hand, the scientific community has focused all its efforts on studying *P. granatum* L. and the immense variety of cultivars it presents. There are thousands of articles on *P. granatum* L., there are high-quality books and reviews, and in most of them, they only mention *P. protopunica* Balf. as a curiosity. This review represents an effort and a call for the scientific community and the population in general to know more about this wonderful species, and to the extent possible, studies in the area of biology, health and pharmacology can be carried out. By studying the benefits of this species, a culture of respect and care can be generated. It is recommended that if scientists wish to study this species, they cultivate it in its places of origin, to preserve the species and take care of the wonderful place that houses it, Socotra.

**Author Contributions:** Conceptualization, writing, original draft preparation and methodology, J.A.G.-S.; formal analysis, writing, T.J.-C. and T.A.U.-H.; investigation, supervision, A.C.-R.; investigation, conceptualization, E.G.O.-H.; supervision, review and editing, O.A.J.-M. and M.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** One of the authors (J.A.G.-S.) is grateful for a scholarship N.554424 provided by the National Council of Science and Technology of Mexico (CONACyT).

**Acknowledgments:** We acknowledge to PROFEXCE 2020-2021 program.

**Conflicts of Interest:** The authors declare that there are no conflict of interest in the publication of this article.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
