Crop Wild Relatives (CWR) Priority in Italy: Distribution, Ecology, In Situ and Ex Situ Conservation and Expected Actions

: The study presents an updated overview of the 14 non-endemic threatened crop wild relatives (CWR) in Italy: Aegilops biuncialis , Ae. uniaristata , Ae. ventricosa , Asparagus pastorianus , Beta macrocarpa , Brassica insularis , B. montana , Crambe hispanica subsp. hispanica , C. tataria subsp. , Ipomoea sagittata , Lathyrus amphicarpos , L. palustris , Vicia cusnae and V. serinica . Geographical distribution, ecology (with plant communities and habitat 92/43/EEC aspects), genetics (focused on gene pools), property, and in situ and ex situ conservation were analyzed. In addition, with the aim of their protection and valorization, speciﬁc actions are recommended.


Introduction
Crop wild relatives (CWRs) are wild species closely related to crops and are potential sources of important traits (such as pest or disease resistance), yield improvement and/or stability [1]. It must also be considered that they are a critical component of plant genetic resources for food and agriculture (PGRFA), although they have been neglected for conservation purposes [1], and in situ and ex situ conservation approaches should be deployed to ensure their availability for use [2].
In monetary terms, the CWRs have contributed significantly to the agricultural and horticultural industries, and to the world economy [3,4]. Pimentel et al. [5] estimated that wild relatives contribute approximately USD 20 billion toward increased crop yields per year in the United States, and USD 115 billion worldwide. Phillips and Meilleur [6] noted that losses of rare wild plants represent a substantial economic loss to agriculture, estimating that the endangered food crop relatives have a worth of about USD 10 billion annually in wholesale farm values. Although these studies show significant divergence, they highlight the major global economic value of CWR diversity to humanity.
Following the definition of Maxted et al. [7], the CWRs are taxa belonging to the same genus as the cultivated species. With this approach, about 80% of the European and Mediterranean flora species are CWRs and important from a socioeconomic point of view [8]. However, a genetic rather than a taxonomic approach suggests that only those species able to interbreed with cultivated species in relation to their "gene pool" should be considered CWRs. According to Harlan and de Wet [9], the gene pool represents a reservoir of diversity that can be tapped into by organisms to adapt to a changing environment, and breeders for crop improvement. Wild relatives of a given crop are thought to be in the same gene pool, and even when they appear to be taxonomically different, they can exchange genes with their related cultivated taxon. Unfortunately, not all wild relatives are equally ready to do this. For this reason, CWRs have been classified into three groups (GP1, GP2, GP3) based on the ability to exchange genes with the cultivated species to which they are naturally related [9]. The primary gene pool (GP1) includes species that can be directly crossed with the cultivated species to produce fertile breeds. For example, it is easier for Beta macrocarpa Guss. (GP1) to interbreed with cultivated chard (Beta vulgaris L.) because Table 1. Prioritized list of 14 (non-endemic) taxon group crop wild relatives and reasons for threat, adapted and updated from Landucci et al. [35], Magrini et al. [16] and Perrino and Perrino [1].  [15]. IS (ISTAT): taxa mentioned by the Italian Institute of Statistics (ISTAT) for cultivated areas and yield in the last 5 years before 2012 [20]. 1 (>Bilz et al. [24]; European Red List): EN, Endangered; LC, Least Concern; NT, Near Threatened; VU, Vulnerable. ‡ (Conti et al. [22]) 2, Italy; 3, Sicily; 4, Sardinia: EW, Extinct in the wild; CR, Critically Endangered; EN, Endangered; VU, Vulnerable; LR, Lower Risk; DD, Data deficient. 5 (Conti et al. [21]; National Red List): E, Endangered; V, Vulnerable; R, Rare. 6 (Rossi et al. [25]; Italian Red List of Policy Species and other threatened species): EN, Endangered; VU, Vulnerable; NT, Near Threatened. 7 (RED LIST OF THREATENED VASCULAR PLANTS IN ITALY: taxa included in 1 Rossi et al. [26] and in 2 Orsenigo et al. [32]): EN, Endangered; VU, Vulnerable, NT, Near Threatened. 8 (OTHER IUCN CARDS: taxa included in "Red List of Italian Vascular and Cryptogamic Flora cards" published since 2013 on Informatore Botanico Italiano become Italian Botanist, 1 Perrino and Wagensommer [28], 2 Perrino and Wagensommer [27], 3 Perrino and Wagensommer [29], 4 Santo et al. [31], 5 Perrino et al. [30], 6  For each of the 14 non-endemic wild relatives, three levels of attention were considered regarding ex situ conservation: (1) High Priority (HP) for taxa present in the Italian RIBES (Rete Italiana delle Banche del Germoplasma) seed banks with zero accessions; (2) Normal Priority (NP) for taxa present with fewer than five accessions (from 1 to 4); and (3) Zero Priority (ZP) for those species present with five or more accessions (from 5 to 140). Furthermore, for the same taxa, one level of attention for the in situ conservation (A) was considered, which included the native taxa related to a crop of worldwide and national importance for food and agriculture, which were included in (at least) the National and in the European Red Lists, and in the International Conventions and that need specific monitoring/protection measures (Table 1). No taxa belonging to the other two levels of attention for the in situ conservation were identified: (1) level (B), concerning the native taxa related to important crops, which on the basis of current knowledge, have no need of any immediate specific protection or monitoring measures; (2) level (C) that includes the native taxa related to important crops, neither endemic nor subendemic to Italy, which, on the basis of current knowledge, have no need of any immediate specific protection or monitoring measures [1,35]. For a better evaluation of in situ and ex situ conservation, vegetation and 92/43/EEC habitat data have been included (Table 2).
Finally, the 14 non-endemic wild relatives were evaluated considering their gene pools (GP1, GP2 and GP3), according to the concept of Harlan and de Wet [9], by consulting the checklist www.cwrdiversity.org/checklist/ (accessed on 2 November 2020) and Vincent et al. [36], and checking their in situ and ex situ conservation priorities (Table 3). Ex situ priority conservation. HP: taxa with high priority (zero accessions), NP: taxa with normal priority (1-4 accessions), ZP: taxa with no priority (5-140 accessions). Adapted and updated from Magrini et al. [16]. In situ priority conservation. A: native taxa related to a crop of worldwide and national importance for food and agriculture, which are included as Threatened (EW, CR, EN, VU) or Near Threatened in (at least) one of the following sources: IUCN (European Red List) [24], Regional Red List (national catalogue and catalogue for Sicily and Sardinia) [22], National Red List [21,23,[25][26][27][28][29][30][31]37]. These taxa need specific protection and/or monitoring measures. Vegetation type and/or Habitat 92/43 EEC (Italy). Code habitat [38]. Vegetation type (see reference in the text when discussing the relative species). Table 3. Crop wild relatives (8 out of 14) belonging to at least one gene pool and their conservation prioritization updated from Landucci et al. [35] and Perrino and Perrino [1].

Taxa Gene Pools (GP) Ex Situ Priority Conservation In Situ Priority Conservation
zero priority). In conclusion, all species had high in situ priority (A) and need monitoring and updating, and should be considered at risk (Table 2).

The 14 Taxon Group CWRs in the Light of the Gene Pool Concept
Plant breeders concentrate on wild relatives that may cross easily with crops; therefore, we have checked which ones of the 14 taxon group wild species belong to the three gene pools, foreseen by the Harlan and de Wet [9] concept. The results (Table 3) show that only 8 species out of the 14 belonged to one or two gene pools. In particular, two species, Beta macrocarpa and Crambe hispanica subsp. hispanica, shared only the primary gene pool (GP1); five species, Aegilops biuncialis, Ae. uniaristata, Ae. ventricosa, Brassica insularis (Policy species), and B. montana, shared the secondary and tertiary gene pools (GP2 and GP3); and one species, Lathyrus amphicarpos, belonged only to the secondary gene pool (GP2), and not GP3, as indicated by www.cwrdiversity.org/checklist/ (accessed on 20 November 2020). In conclusion, two species belonged only to GP1, one only to GP2, five shared GP2 and GP3, while for the other six taxa, at the moment, to the best of our knowledge, there is no information.

Aegilops biuncialis Vis., Aegilops uniaristata Vis. and Aegilops ventricosa Tausch
The genus Aegilops L. has been intensively studied due to its close relationship with cultivated wheats, and their vast genetic diversity represents a rich source of alleles of agronomic interest, which could be used to widen the wheat gene pool and improve tolerance to diseases, pests, drought, cold and other environmental stresses [39], and for improving micro-nutrient content (such as Fe and Zn) in wheat grains. About the last point, it should be noted that Zn deficiency affects 17.3% of the world population, mostly in Asia and Africa, leading to the death of over 400,000 children every year [40][41][42]. Wheat rich in micronutrients, i.e., bio-fortified wheat, can improve the lives of these people. It is difficult to find germplasm with high Zn and Fe content in the wheat gene pool [43], although some Ae. show three-to four-fold higher Zn and Fe grain content, such as Ae. ventricosa (genome DN) [44].
In Europe, wild wheat relatives of the Triticum-Aegilops complex grow in sympatry with cultivated bread wheat (Triticum aestivum L.), and spontaneous hybridization is known for most of the tetraploid Ae. species. The probability of gene transfer and gene retention in hybrid progenies is, however, higher when a gene is located on a shared genome, particularly on the D genome shared with Ae. cylindrica and Ae. ventricosa. Through optimized experimentation, some studies have shown to support the hybridization (experimental soil layout, flowering synchrony) that the cross-pollination between the cultivated wheat and its relatives occur at a significant level as for Ae. biuncialis [45].
The species belonging to this genus are mainly distributed in Southwest and Central Asia and throughout the Mediterranean Basin [48,49]. In Italy, their geographical distribution, ecology, vulnerability has been updated [50]. Among the priority CWRs, it is the most represented genus, with three species (Ae. biuncalis (genome UM), Ae. uniaristata (genome N) and Ae. ventricosa (genome DN)) all listed as threatened in the red lists (Table 1), with high in situ priority (A), and normal ex situ priority (NP) ( Table 2) and secondary and tertiary gene pools (GP2 and GP3) ( Table 3).
The flowering time of Ae. in Italy is from April to June, depending on the species and its eco-geographical location [50], and partially meets the flowering of the cultivated wheat that starts in May and ends in June [51], a phenological condition that would suggest in situ crossbreeding experimentation.

•
In situ and ex situ conservation to prevent the risk of extinction by increasing the number of individuals of existing wild populations.

•
In situ translocation to the edge of fields of cultivated ancestral wheat to verify and update the hybridization capacity, thanks to the comparable flowering periods.

•
Starting cultivation in cooperation with local farmers, especially of Ae. ventricosa that, unlike cultivated wheat varieties, has a higher quantity of microelements such as Fe and Zn. Then, verify the prospect of production and marketing of its flour and/or pasta as a natural alternative to conventional medicine, and helpful for people with Fe and Zn deficiencies. • Study plant communities of annual meadows to define their phytosociological framework.

Asparagus pastorianus Webb & Berthel
Several wild species of the genus Asparagus L. in the Italian Peninsula have long been the object of harvesting for food consumption, and in the case of A. officinalis L., also of ancient domestication and cultivation. Since the Middle Ages, the cultivated and wild species of this genus has always had an important place in the gastronomic culture. The young shoots of A. pastorianus are eaten in Morocco (vernacular name: sekoum), the stems and roots are used in popular medicine as aphrodisiacs [53,54] in the Canary Islands (vernacular names: "esparraguera de espinas" or "espina blanca") and to produce smoke, prepare infusion, decoction with white wine, as insect repellent, and diuretic slimming. The bioactive phytochemicals are glycosides and sapogenins [55].
A. pastorianus is a perennial shrub that grows in the garrigues near the sea, and has a south-western Mediterranean-Macaronesian distribution. In Italy, the species grows only in a restricted area of the Sicilian Region [17]. On the southern coast of Sicily, between Selinunte and the mouth of the Verdura River, on Pleistocene deposits consisting of a succession of calcarenites and sandy clays, grows a peculiar low shrubby plant community association characterized by A. pastorianus, described as Asparago pastoriani-Chamaeropetum humilis Raimondo & Bazan 2008 [56], included in the alliance Oleo-Ceratonion Br.-Bl. ex Guinochet & Drouineau 1944 em. Rivas-Martínez 1975.
The dispersion of seeds in the Canary Islands (Lanzarote, Fuerteventura, Gran Canaria, Tenerife, La Gomera) occurs through small mammals (e.g., squirrels) [57] and birds (e.g., shrikes and kestrels) [58], although there are no available data from the Sicilian population.
The chromosome number is 2 n = 40 [59] (material from Santa Lucía, Gran Canaria, cultivated in the Botanical Garden in Oslo). Although the gene pool is unknown, among the conservation priorities, A. pastorianus is one of the most important for conservation interest because it is listed in the red list with VU category in Europe [24] and NT category in Italy [32] (Table 1), resulting in a high in situ (A) and ex situ (HP) priority (Table 2).

•
Monitoring of the few known sites of coastal area in southern Sicily, well preserved and for which in situ conservation actions would be appropriate, because of the following potential threats: (a) policy developments that aim at tourist exploitation [56]; and (b) potential negative effect of mammals, especially rodents and lagomorphs on seed germination, as already observed in the Canary Islands [60].

•
Targeted actions for the collection of germplasm to ex situ conservation because the species has zero accessions in the RIBES seed banks.

•
Research activities to verify the gene pool through crossing with other species of the same genus and any differences with the populations of the Canary Islands and Morocco.

•
Verify the seed dispersal system in Sicily, as was done in the Canary Islands.
• Evaluate the enhancement of the Sicily populations for the production of their use for medicinal purposes.

Beta macrocarpa Guss
The genus Beta L. is divided into two sections: Beta and Corollinae [61]. The section Beta includes five taxa, B. vulgaris L. subsp. maritima (L.) Arcang. (the sea beet), which is considered as the wild ancestor of all cultivated beets, the different forms of cultivated beets (B. vulgaris L. subsp. vulgaris), discovered wild in Calabria [62], B. macrocarpa, which is an annual self-compatible plant thought to reproduce predominantly by autogamy, B. patula Aiton, which is endemic to two small islets of the Madeira Archipelago [63], and finally B. vulgaris L. subsp. adanensis (Pamukç.) Ford-Lloyd & J.T. Williams, that grows in some Eastern Mediterranean areas of Greece and Turkey [61,64,65]. Therefore, in the western Mediterranean area, only two species of the section Beta can be found in coastal and inland ruderal habitats: B. vulgaris subsp. maritima and B. macrocarpa [66]. B. macrocarpa is located in inland or coastal habitats in western and eastern Mediterranean areas [67] and in Italy it grows in uncultivated clayey soils [68] in Campania, Basilicata and Sicily, while its occurrence is doubtful in Sardinia and Trentino Alto Adige [17].
The chromosome number of B. macrocarpa is 2 n = 36 (from accessions of the Canary Islands) [69]. B. macrocarpa is closely related to B. vulgaris subsp. maritima by genetic structure. In particular, B. macrocarpa has a genotypic structure and a high level of genetic differentiation indicative for selfing (an extreme degree of inbreeding) [66]. In fact, the two species can spontaneously hybridize [69][70][71], sharing the primary gene pool (GP1) ( Table 3).
This species is listed in the European red list [24] with EN category (Table 1), shows high in situ priority (A) and high ex situ priority (HP) ( Table 2), and primary gene pool (GP1) ( Table 3). It is worth noting that the wild taxa of section Beta, except B. vulgaris subsp. maritima, are all listed in IUCN Red List, as VU (B. vulgaris subsp. adanensis) [72], EN (B. macrocarpa) [73] and CR (B. patula) [74]. For threatened therophytes, such as B. macrocarpa, it is important to consider that natural phenomena can cause considerable fluctuations in the number of individuals, and that therefore repeated counts in subsequent years are necessary for a correct estimate of the population size [75].
A specific study on herbaceous vegetation useful for animals on wetland environments in Tunisia showed a high concentration of minerals in B. macrocarpa, in particular the highest ones compared to all the other herbaceous species, locally sampled, on K (15.4 g kg −1 dry matter), Ca (31.2 g kg −1 dry matter), Mg (15.1 g kg −1 dry matter) and although high on average, the lowest NaCl (54.3 g kg −1 dry matter) content among the Chenopodiaceae family, in addition to a high concentration value of phenols (30.1 g kg −1 dry matter) and oxalate (64.6 g kg −1 dry matter) [76].

Brassica insularis Moris (Policy Species), Brassica montana Pourr
Wild taxa in B. oleracea L. play an important role to improve cultivated crops, but the genomic relationships between wild and cultivated forms have not been well clarified [77]. B. insularis and B. montana belong to B. sect. Brassica, which encloses the taxa with the same C genome (n = 9) of B. oleracea crops [78][79][80], and the crossing experiments have confirmed that they are closely related [81].
B. insularis is an endemic Mediterranean member of the B. oleracea group which occurs only in France (Corsica), Italy (Sardinia and Pantelleria), Tunisia (La Galite, Zembra and Zembretta) and Algeria (Kabylie) [31,82], while B. montana is widespread along the coasts of the northern Mediterranean Sea, from north-eastern Spain to south-western Italy [83]. In Italy, B. insularis grows only in two islands (Sicily and Sardinia), and its occurrence is doubtful in Tuscany, while B. montana has a fragmented distribution (Liguria (very common along the coast), Emilia Romagna, Tuscany, Marche, Latium, Campania, Basilicata and Calabria) [17], probably due to its relict origin [84].
Both taxa are listed in red lists (Table 1), with high in situ priority (A), and zero ex situ priority (ZP) ( Table 2) due to the 27 accessions present in the seed-banks of Sardinia and Perugia [16], and secondary and tertiary gene pools (GP2, GP3) ( Table 3). B. insularis is also listed in Annex II of the Habitat Directive 92/43/EEC [34] and under Appendix I of the Bern Convention [33].
B. insularis is a perennial rupestrian, xerophilous species that grows under the influence of wet marine flows with high soil salinity and marine aerosols, while it is less frequent in inland areas, on slopes, cliffs, and vertical walls, at altitudes from 0 to 1200 m a.s.l. [31,85], with a flowering period that extends from March to May, and with only a small proportion of individuals flowering in any given month [86].
B. montana grows in habitats influenced by human activities, for instance quarries, roadsides and building grounds [87], and the flowering period is from March to April Several studies testify that wild forms can be considered as potential resources to improve the current B. oleracea crops, especially when some favorable traits have been identified in wild types of B. oleracea such as resistance against Sclerotinia sclerotiorum [88], blackleg (Peronospora parasitica) [89,90], cabbage white fly (Aleyrodes proletella) [91], and cabbage root fly (Delia radicum) [92]. B. insularis showed seed sinigrin content, with unusual glucosinolate patterns, low progoitrin and high gluconasturtiin levels, and benzyl glucosinolates traces [93], while B. montana showed a high seed glucosinolate content that could be used for increasing the total content of specific glucosinolate profiles for improving biocidal and anticarcinogenic activity in cultivated Brassica [94].

•
All Italian taxa of genus Brassica may be used as genetic resources, with potential host valuable traits that could be transferred to the respective cultivated crops (cabbage, cauliflower, broccoli, etc.) [83], starting from the places with greater ecological affinities and closest to the known localities where the wild species grow.

•
In situ experiments in cooperation with local growers, thanks to their high potential agronomic value and high tolerance to drought, insects, and high content of glucosinolate [95,96].

•
In situ and ex situ crosses with cultivated B. oleracea, although some preliminary studies have shown low fertility values of crosses between B. montana and cultivated forms of B. oleracea [87]. The wild populations could be maintained with low onsite management because they grow on cliff sites and suffer especially due to the availability of nutrients. • Ex situ conservation of wild populations is necessary, especially to avoid species extinction or further genetic erosion after ecological changes [97], and can be realized by plant conservation in botanical gardens and seed-banks, the latter started by Gomez-Campo and Gustafsson in 1986 [82].
• Low levels of observed heterozygosity in natural populations of B. insularis document the importance of developing conservation guidelines appropriate for the populations of this species [82]. Geographical variation studies might be further investigated with physiological analyses [98].

•
Monitoring of known populations and field surveys to find new sites (especially for B. insularis). Despite the restriction on collecting B. insularis from the known sites, the cleaning of cliffs to create suitable climbing areas could be a problem. Preventing access to the B. insularis populations appears to be the most suitable conservation measure with the support of protection policies, which was how it was achieved for the Corsica populations [82]. • Ecological studies are needed to determine the role of grazing (especially by goats) on population maintenance [99].

•
Phytosociological studies in Italy, where it occurs, to evaluate vegetation, habitat, ecology, and biodiversity, especially for B. insularis for which there is a lack of data.

Crambe hispanica L. subsp. hispanica, Crambe tataria Sebeók subsp. tataria (Policy Species)
Many species of the genus Crambe L. are considered industrial crops [100]. For instance, C. tataria can be used for paper production when mixed with long fibrous materials [101], to obtain higher oil and erucic acid yield [102]; C. hispanica was used for the production of special lubricants, in industrial vulcanization processes, and in those that lead to erucamide from erucic acid [103][104][105], biodiesel, meal and husk for animal feed [106]. For this purpose, in 1975, C. hispanica seed samples were collected in Apulia (Gargano) and Sardinia by a team of breeders from California (U.S.A.) and agronomists from the Germoplasm Istitute of Bari (CNR) [107], for its cultivation as a new alternative crop to other industrial crops [108,109].
The genus Crambe has an extensive area of distribution that goes from the Macaronesian archipelagoes to the west of China and north of India, and from the Arctic Polar Circle on the Scandinavian Peninsula to 5 • latitude south in the northern Tanzania. It is well-represented in the Macaronesian, Euro-Siberian, Mediterranean, Sindico-Saharan, Irano-Turkish and Sudan-Zambezian (Ethiopia and Tanzania) regions [103] and includes more than 35 species [110,111]. Based mainly on the dimensions and shape of the proximal joint of the fruit, the genus is divided in three sections (Crambe, Dendrocrambe and Leptocrambe) that closely correspond to the geographical areas of distribution [112,113]. In Italy, only two species of Crambe are present [17], both considered CWRs [1]: C. hispanica subsp. hispanica (section Leptocrambe) and C. tataria subsp. tataria (section Crambe).
C. tataria subsp. tataria is endemic to the Pontic-Pannonian region, with a strong disjunction from its main distributional range in the Friuli-Venezia Giulia region, the only site in Italy [117].
C. hispanica subsp. hispanica is a sub-nitrophilic-synanthropic species, which in Italy grows on calcareous soils, on sandy soils of volcanic origin, and on brown soils, exclusively in habitats subject to anthropic disturbance and semi-rupestrian environments. It can be located at the edge of abandoned olive groves, near lake basins, along dry-stone walls, in the shade of isolated trees, often of Quercus trojana Webb ascribed to the Crambetum hispanicae Perrino, Tomaselli, Signorile, Angiulli, Silletti 2011 association [118], along the banks of rivers [119], in shrub vegetation dominated by Cytisus villosus Pourr. and Spartium junceum L., to margins of thermophilous woods, in uncultivated arid areas [120], and also in correlation with road bumps. In the Apulian populations, it grows on calcareous substrates, with a certain enrichment, never very intense, in soil nutrients, while in Sicily the species prefers humid environments and on Mt. Etna, where the best-preserved Sicilian populations occur, it grows on shallow and very humified soils [121]. The flowering time of C. hispanica subsp. hispanica in Italy is from March to April.
C. tataria subsp. tataria is reported on steppes and hills rich in clay and limestone from eastern Europe to the Caucasus [113]. In Italy, it grows on extensive deep beds of alluvial, calcareous gravel deposited by the rivers Cellina and Meduna that characterize the "magredi" landscape. The use of land for military purposes in sparsely inhabited areas has somehow helped in the preservation of "magredi" fragments, where C. tataria is one of the most typical elements of this characteristic grassland formation [117] referred to Centaureo dichroanthae-Globularietum cordifoliae Pignatti 1953 association [122] which is considered habitat 92/43/EEC "Eastern sub-mediterranean dry grasslands (Scorzoneretalia villosae)" (code 62A0). The flowering time of C. tataria subsp. tataria in Italy is from May to June.
C. hispanica subsp. hispanica and C. tataria subsp. tataria are both reported as threatened in Italy, while they are considered LC in the European Red List (Table 1). Both species have a high in situ priority (A). As for ex situ conservation, C. tataria subsp. tataria with one accession has normal priority (NP), and C. hispanica subsp. hispanica with five accessions has no priority (ZP) [16] (Table 2), although the number of accessions is still low. In addition, C. hispanica subsp. hispanica is important for its primary gene pool (GP1) ( Table 3). However, independently from the gene pool, C. tataria also needs widespread protection because it is rare throughout the global range and its habitats are often destroyed; in fact, the taxon is also reported in the Red Books of the USSR and Kazakhstan [124].

•
In situ and ex situ conservation to prevent the risk of extinction by increasing the number of individuals in wild populations. For C. hispanica subsp. hispanica, an ex situ conservation strategy is strongly needed to support the industrial purposes as an oil plant, through on farm conservation, while as for in situ conservation, research to learn more about the breeding system and the vertical pollen transfer is needed [121].

•
The man-made summer fires with the aim to clean the soil in the habitat of C. hispanica subsp. hispanica (Crambetum hispanicae), when not avoidable, must be targeted to the dry component of the plant and carried out from the end of July, after the period of seed dissemination (May-June), otherwise the plants could be irreversibly damaged.
As for the populations growing along the road, it is crucial that bodies responsible for road management be informed about the presence of threatened species. The species in question is linked to abandoned or extensively managed agricultural areas, which have a good naturalness (defined HNVF, high nature value farmland) due to the immediate proximity of shrubland or woodland vegetation. Therefore, one of the main threats to the survival of the species and its plant community is the current trend to use high impact agriculture, including chemical input, such as herbicides.

Ipomoea sagittata Poir
The genus Ipomoea L. has an amphi-Atlantic distribution, and probably it arrived in Europe only after contact with the new world [127], with 18 known species [128]. I. sagittata is known in the eastern Atlantic and the Mediterranean region from Algeria, the Balearic Islands, Corsica, Cyprus, Greece, Lebanon, Italy, Malta, Portugal, Sicily, Spain, Syria, Tunisia, and Turkey [128]. Austin [129] suggests that I. sagittata is native to the circum-Caribbean region of the Americas, and it probably arrived in Europe for the first time in Greece or France by sea with maritime trade and then it spread to other Mediterranean territories. Other authors [130] believe that seed dispersion had been mediated through ocean currents, and this would probably explain why I. sagittata grows in salt marshes, and could have made it to Europe in prehistoric times. Thus, we can conclude that due to the large disjunction in its distribution it is a controversial species, because even if introduced a long time ago, it probably is an exotic wild species [131]. As a result, its nativity needs to be re-evaluated.
In Italy, I. sagittata grows with other ten species of the same genus: I. imperati  [62]. The Italian distribution of I. sagittata includes the Latium, Apulia, Calabria, and Sicily regions [17].
I. sagittata is a rhizomatous geophyte flowering from June to September, typical of coastal marshes and wet brackish muds and banks [134], and is always very localized in Italy and threatened by the rarity and vulnerability of the environments in which it grows, as shown by its disappearance in historic sites such as those of the coast of Mondello (Palermo-Sicily), "Pantano del Taro" (Taranto-Apulia) and on the islet of "S. Nicolicchio" (Taranto-Apulia) for reasons related to human activity [134]. Fortunately, botanical explorations have made it possible to discover new stations in the Salento peninsula (Apulia region), such as those at "Le Cesine" [135] "Palude di Rauccio" (Lecce) [136], "Laghi Alimini" [137], "Torre Rinalda", Basins of Ugento, at "Punta Prosciutto" in the "Palude del Conte" [134], and "Torre Chianca" [138], in many localities in the Province of Trapani (Sicily), such as "Isola Grande dello Stagnone" [139], "Santa Ninfa" [140], "Petrosino" along the drainage canals [141], and near the halophilous reeds of "Lago Prato" in the province of Catanzaro (Calabria) [142]. It is confirmed along the southern edges of "Lago Fondi" and "Canale S. Anastasia" into the Regional Natural Park of "Monti Ausoni e Lago di Fondi" in the Lazio region [143].
The chromosome number is 2 n = 30 (from accessions of Pali district in India) [144]; it has been reported with a synonym (=I. sagittifolia Ker Gawl.) and with interesting information of 77% of pollen fertility.
This species is listed as VU in the European Red List and as EN in the Italian National Red List (Table 1), with high priority in situ (A) and ex situ (HP) conservation (Table 2), while there are no data about gene pool (Table 3). It is worth noting that despite its limited Italian distribution, the species grows in several wetland types of vegetation as companion species, i.e.  [146] all observed in the Salento peninsula, and Ranunculetum peltati Sauer 1947 at Anguillara (Trapani) in Sicily [147]. It is also considered a diagnostic taxon of the alliance Calystegion sepium Tüxen ex Oberdorfer 1957 nom. mut. propos. Rivas-Martínez, T.E. Díaz, Fernandez-Gonzales, Izco, Loidi, Lousã & Penas 2002, which encloses the nitrophilous tall-herb communities that develop in humid, periodically inundated, habitats and are subjected to long periods of drainage and occasionally with a moderate salinity [18]. This peculiar type of vegetation is reported in Annex I of the Habitat Directive 92/43/ECC as "Hydrophilous tall herb fringe communities of plains and of the montane to alpine levels" (code 6430) ( Table 2).
It has been observed that, in North America, I. sagittata can be harmful for agricultural purposes because it harbors an insect, "the weevil", that can infest crop potato, although found only in limited sections of the sweet potato-growing areas, mostly in the coastal and tide-marsh margins [148]. There are no recorded medicinal uses of I. sagittata in the Old World [129], and there are doubts that it is a real CWR, as indicated by Bilz et al. [24].

•
Clarify with ad hoc studies and researchers (including historical ones) if the taxon is exotic or native in Europe.  [149,150]. Heywood et al. [151] extended the secondary gene pool also to L. chrysanthus Boiss., L. gorgoni Parl., L. marmoratus Boiss. & Balansa ex Boiss. and L. pseudocicera Pamp., with which L. sativus can cross and produce ovules, and more remotely to L. amphicarpos, L. blepharicarpus Boiss., L. chloranthus Boiss. & Balansa, L. cicera, L. hierosolymitanus Boiss. and L. hirsutus L., with which L. sativus can cross to form pods. L. gorgoni and L. hirsutus are also reported in Italy [17]. The remaining species of the genus can be considered members of the tertiary gene pool (GP3) [152]. The results of electrophoretic comparative analysis of seed albumins and globulins showed L. sativus to be considerably different from the allied species. Consequently, exploitation of the germplasm resources in the breeding improvement programs of the grasspea should be concentrated on the primary gene pool (GP1), as suggested by Yunus and Jackson [150].
L. amphicarpos is a Mediterranean taxon which occurs in Algeria, the Balearic Islands, France (Corsica), Greece, Crete, Italy, Morocco, Portugal, and Spain, while L. palustris has a wider distribution, being a circumboreal taxon [155]. In Italy, L. amphicarpos grows only in the Sicily, Apulia, and Latium regions, while L. palustris has a fragmentary distribution in the northern regions and is absent in the Center-South of the Italian Peninsula [17]. L. amphicarpos grows in Latium on arid meadows and garrigues from 250 up to 600 m a.s.l. in the Ausoni chain on Mt. Leano [156], Mt. Cucca, M. Cavallo Bianco and M. Saiano [157], in the Natural Reserve "Pizzo Cane, Pizzo Trigna and Grotta Mazzamuto" (north-west of Sicily) [158], in other sites of Palermo municipality, at Monte Sparacio (Trapani) on Nebrodi mountains (Messina) in Sicily [159], and in the southern sector of the Daunia Mountains in Apulia [160].
Among wild species of genus Lathyrus, L. amphicarpos showed the best antioxidant activity results in seed methanolic extracts [161] and higher total saturated fatty acids. These data, combined with the benefits attributable to the secondary metabolites (polyphenol contents), suggests the use of the genus Lathyrus, and in particular of L. amphicarpos, in human and animal diets [162]. L. palustris also has favorable histological characteristics for use as a fodder crop [154,163].
L. amphicarpos is listed as NT in the European Red List and as LR in the Italian National Red List (Table 1), with high in situ (A) and high ex situ (HP) priority (Table 2), and as a secondary gene pool (GP2) ( Table 3) giving rise viable hybrids in crosses with L. sativus [150,164]; while L. palustris is listed as EN in the Italian Red List (Table 1), with high in situ (A) and ex situ priority (HP) ( Table 2), and with no information on gene pool (Table 3).
L. amphicarpos is an annual plant with an elongated flower axis, often without leaves, that flowers from March to April, from sea level to 600 m of altitude. The only ecological information in Italy comes from Sicily, where the species is found in limestone and stony ground with sparse vegetation of annual species located in degraded Ampelodesmos mauritanicus (Poir.) T. Durand & Schinz grasslands subjected to the action of fire and grazing, and in stations exposed to the action of atmospheric elements that cause soil erosion. In these habitats, the species spreads its slender roots among the stones, developing the phenomenon of amphicarpy, a typical adaption in acid habitats, subject to fire [159].
L. palustris is a perennial plant that flowers from June to August, from sea level to 800 m of altitude, for which little ecological data are available, especially about the vegetation in which it grows. In Alto Adige, it is observed in the humid, uncultivated grasslands of the Molinion caeruleae Koch 1926 alliance [165], attributable to the habitat 92/43/EEC "Molinia meadows on calcareous, peaty or clayey-silty soils (Molinion caeruleae)" (code 6410), although it is a diagnostic of the "Mediterranean high and humid herbaceous grasslands of Molinio-Holoschoenion" habitat (code 6420) [166]. However, there is a gap in phytosociological studies for this species.

•
In situ and ex situ conservation to prevent the risk of extinction by increasing the number of existing wild populations for both Lathyrus species. • Use in the human diet of L. amphicarpos due to good antioxidant activity present in their seeds and for the presence of high content of saturated fatty acids [162], but only after research and breeding with the aim to turn saturated fat to unsaturated fat acid content.

•
Monitoring the known populations of L. amphicarpos and field surveys to find new sites because of the earlier confusion with L. cicera, which is very similar in morphology, although much more widespread in Italy.

•
Specific ecological studies on L. palustris wetlands habitats, which could provide information on factors related to the maintenance of wetlands and their conservation, considering the potential benefit of use as grazing fodder, especially for buffalo [167].  [168], and can be considered geographic vicariants among themselves [169], although there are different readings, such as that of Davis [170], which consider V. serinica and V. variegata as a subspecies of V. canescens. There are several morphological differences between the species of the group that concern microcharacters located in different parts of the plant [169].
V. cusnae is reported in three circumscribed sites: two in Italy, in the National Park of the Tuscan-Emilian Apennines (Emilia Romagna Region), at M. Cusna [169] and at Rio Re at M. Prado [171]; and, thanks to Philippe Küpfer, in France in the Aurouze Massif [23]. V. serinica is reported in southern Italy and in northern Greece [172]. In Greece, it was collected by Gustavsson in several mountains at Sterea Ellas, by Aldén from Mount Kakarditsa in Pindhos [173], and by Strid and Papanicolaou [174] on Mt. Belles (Kerkini), north-east of the village of Ano Poroia. In Italy, it occurs only in a very confined area of Basilicata region, while it was reported by mistake in Campania [17]. In Basilicata, it grows in only four stations of Sirino-Papa Massif [175,176], in the municipality of Potenza (confirmed by E.V. Perrino, unpubl. data).
V. cusnae and V. serinica have both the same chromosomal number (2 n = 10), just like the other three species of the group [169]. This datum can be interpreted as schizoen-demisms, with puntiform distribution and geographically isolated [177], that make highly improbable genetic exchanges between the populations of these two species with the other conspecific populations.
V. cusnae is an alpine glareicole taxon that flowers from July to August on detrital soils of sedimentary rocks with southern exposure and in xerothermic conditions, from 1800 to 2100 m of altitude [23]. It reproduces mainly by vegetative parts, thanks to the presence of short underground stolons, which issue new close shoots [23], as was also observed for V. serinica at M. Sirino (observed by E.V. Perrino, unpubl. data), rather than by seed dispersal [178]. It covers large areas in which it is a dominant taxon, and it is assigned to the Thlaspion rotundifolii Jenny-Lips 1930 alliance [179] and habitat 92/43/EEC "Calcareous rocky slopes with chasmophytic vegetation" (code 8210).
V. serinica flowers in July on soils similar to those of V. cusnae, from 1500 to 1850 m of altitude, colonizing peculiar niches reserved for highly specialized species that are able to grow in extreme environmental conditions. The soil has a copious skeleton in the superficial horizons and a high sand content in all layers. The annual average precipitation is about 1400 mm, while the bioclimate is oceanic temperate of the humid supratemperate type [176]. The vegetation of V. serinica is referred to be Sideridenion italicae Biondi et al. 1995 [176], to be related to the priority habitat 92/43/EEC "Semi-natural dry grasslands and scrubland facies on calcareous substrates (Festuco-Brometalia) (*important orchid sites)" (code 6210*) [38].
V. cusnae and V. serinica are listed as VU [25] and EN [32], respectively, in the Italian Red List (Table 1), with high in situ (A) and ex situ priority (HP) ( Table 2). There are no available data on their gene pools (Table 3).

•
Research in situ with monitoring programs to better understand the reproductive biology and ecology of the species and the populations trends. • Evaluate ecological and genetic affinities between the different populations of both species.

•
Phytosociological studies to define the phytosociological association and discover why similar environments produce different types of vegetation and habitats. • Ex situ conservation for both species. For V. cusnae it is possible, because it is an orthodox species, which means that it tolerates seed drying with high levels of germination (80%) after scarification, at 21 • C [180]. The only germplasm accessions of V. cusnae are preserved at the Millennium Seed Bank of the Royal Botanic Gardens in Kew (U.K.) and those of V. serinica in the seed bank collections of the Institute of Biosciences and Bioresources (IBBR-CNR) of Bari, but both are absent in RIBES seedbanks. • Start crossbreeding studies with V. sativa L., whose seeds are consumed by birds and often used as forage, to test their gene pools and to check their taxonomy and systematics.

Conclusions
In Italy, according to the taxon group concept, there are 43 CWRs at risk of inadequate conservation either in situ or ex situ. However, disregarding the species endemic to Italy, the number of 43 falls to 14. Furthermore, according to the gene pool concept, which is more important from a plant breeding point of view, the number 14 falls to 8. For these latter species, this paper provides a picture as complete as possible about their geographical distribution, level of protection, ecology (including vegetation and habitat 92/43 EEC), properties, gene pools, and actions to avoid further genetic erosion, to improve in situ and ex situ conservation of the species and habitats, with the final goal of enhancing genetic resources management and their use both in plant breeding and to promote sustainable agriculture and environmental conservation through ad hoc research, suggested for each of the 14 CWRs considered at risk. Author Contributions: Conceptualization, methodology and investigation, E.V.P.; validation, formal analysis, and data curation, E.V.P. and R.P.W.; writing-original draft preparation, E.V.P.; writingreview and editing, E.V.P. and R.P.W. The authors have read and agreed to the published version of the manuscript.
Funding: This study was financially supported by the University of Bari "Aldo Moro".
Institutional Review Board Statement: Not applicable.