The cactus (Opuntia species) has a large area of distribution according to their wide ability to withstand arid and semi-arid conditions. These species can grow in less fertile soil. The genus Opuntia is economically the most important as a great number of species that produces edible fruits, like as Opuntia ficus-indica (L.) Mill., O. megacantha Salm-Dyck, O. stricta Haw., O. dilennei (Ker Gawl.) Haw. O. schummannii Weber, O. robusta Wendl. and O. albicarpa Scheinvar1. Opuntia species is dicotyledonous angiosperm, belongs to the family of Cactaceae which is made up of a group of approximately 1,600 species in 130 genera subdivided in the three subfamilies Pereskioideae, Opuntioideae and Cactoideae2. Opuntia species flowers are frequently hermaphrodite, but cross-pollination occurs most often3-5. Bees are the major pollinators of the prickly pears1. The Cactaceae are one of the most interesting families due to their extensive set of peculiar adaptations to water scarcity, which allow them to be perennial and evergreen despite the sometimes extreme dry conditions of their environment6.
Tunisia is one of the most important genetic sources for Opuntia species in the world and provides an important source of variation for plant breeding. However, as it’s the case with other species used in fruit production, our country has few species of its own native Opuntia species.
Opuntia species covers an area of about 600,000 ha in Tunisia, distributed mainly in west-central regions on the plains of Kasserine7. This species was cultivated in Tunisia as fruit tree on limited surfaces8. The species has a typical physiology with the photosynthetic process called Crassulacean Acid Metabolism (CAM); stomata close during the day in order to maintain hydration of tissues. The importance of Opuntia genus as a forage plant in the 19th century was the outcome of the need to feed livestock in the arid zones, where the dry seasons are very long. Opuntia is an excellent feed for livestock9.
The molecular techniques are very effective for the characterization of plant genetic resources, however, the morphological characterization should always be considered as a useful tool for the use in collections and description studies10. Opuntia genus identification and descriptor evaluation currently is a priority for all scientists involved in economic and agronomic use of this plant as a crop.
The descriptor list developed by Chessa and Nieddu11 is very helpful in identifying the many forms and varieties within the commercial fruit types on the basis of external morphological characters, such as overall plant architecture, pad morphology, spine characters and fruit characters. Lastly, the isozyme work of Chessa et al.12 and the RAPD data of Wang et al.13 have begun to identify unique proteins and DNA fragments to clones that are nearly identical morphologically14.
Description of the morphological characteristics is the usual methodology accepted from a legal point of view for patenting and registration of varieties15.
The aim of this study was to investigate genotypic variation among 45 ecotypes (11 species of Opuntia) selected from 9 areas cross the world using morphologic descriptor to increase fruit production and improve cattle feeding.
MATERIALS AND METHODS
Plant materials and site description: The study included 45 ecotypes collected from 9 different geographic origins in the world and which represent 8 species of Opuntia genus (Table 1). These are planted in the experimental site of National Institute of Research in Rural Engineering, Water and Forests (INRGREF, Tunisia). All the ecotypes have undergone the same culture edaphoclimatic conditions.
The study area is located in Kairouan (Hendi Zitoun), the central region of Tunisia (Lower semi-arid bioclimate, Latitude is 35°40'41", Longitude is 10°05'46"and altitude is 68 m a.s.l.). Annual rainfall in the area averages is 163 mm, the mean annual temperature is 21.47°C.
Evaluation methods: During harvest season of 2017 and 2018, 20 fruits from each ecotype randomly collected and 10, 1 year cladodes per each ecotype sampled to be evaluated according to the Opuntia descriptor established by the International Union for the Protection of New Varieties of Plants (UPOV16). The 45 ecotypes evaluated based on 5 qualitative traits and 16 quantitative traits. The study was carried out at forest ecology laboratory from Jan, 2017-October, 2018.
The following cladode and fruit quantitative traits were measured using a digital caliper: Cladode length (Lo_Ra), cladode width (Lar_Ra), cladode thickness (ER_Ra) the mean distance between areoles (Me_Ra), fruit length (Lo_Fr) and fruit diameter (Da-Fr). However, cladode fresh weight (PF_Ra), cladode dry weight (PS_Ra), water reserve (Te_Ea), fruit flesh weight (Po_Ch), fruit peel weight (Po_Pe), fruit fresh weight (PF_Fr) and weight of seeds (Po_Pp) were measured using an electronic balance. The average number of areoles (Nb_Ar) in each cladode and number of seeds in each fruit (Nb_Pe) were also counted and taken into consideration.
Opuntia ecotypes surveyed their codes and their origins
Fruit were collected at full maturity. Maturity was determined on the basis of the color characteristics of each ecotype. pH values of juice (pH_Fr) were measured using pH meter. Total Soluble solids of fruit (Ts_Fr) were determined with the refractometrical method using a Zeiss hand refractometer. Fruit flesh firmness (Fe_Fr) was recorded by a digital penetrometer (Turoni, Forlì, Italy) on two opposite cheeks.
Moreover, a series of qualitative fruit and cladode parameters were evaluated using the Opuntia descriptor (UPOV16), fruit shape in longitudinal section (Fo_Fr), depression of receptacle scar (Ci_Fr), length of stalk (LPF), juice color (Co_Ju) and shape of the cladode (Fo_Ra).
Statistical analysis: Analysis of morphological traits were carried out using SAS software package (SAS Institute Inc., Cary, NC, USA Version 9.2). The mean values for each parameter of a given genotype were used to perform statistical analysis of morphological traits. Morphological data were subjected to analysis of variance (ANOVA) and means were separated with the Student-Newman-Keuls test at a 0.05 probability level. A proximity matrix was generated using Squared Euclidean distance then clustering of accessions was performed using Ward’s method. The PCA was performed using SPSS software statistics, the dendrogram was constructed using the Unweighted Pair Group Method with Arithmetic means (UPGMA).
Quantitative characteristics of fruits and cladodes: The cladode size parameters showed noteworthy difference among the elevens species of Opuntia (Table 2). The analysis of variance showed that most parameters were significant (p<0.01) between the 45 ecotypes.
Cladode width encompassed an extensive range between the largest O. ficus-indica (Pr45) with 28.68 cm and the smallest one (Pr43) with 9.6 cm, while, Cladode length ranged between 45.02 and 21.44 cm for, respectively these two latter. Cladode thickness varied between 7.29 and 25.74 mm for O. ficus-indica (Pr2) from Ethiopia and O. Crassa (Pr35) from Algeria respectively. The two O. ficus-indica (Pr15 and Pr16) exhibited the highest cladode fresh weight. However, the O. ficus-indica (Pr43) which is also Tunisian origin registered the light cladode fresh weight. The O. ficus-indica (Pr44) registered the heavy cladode dray weight (124 g). Nevertheless, O. ficus-indica (Pr2) is the lightest. The Mexicain O. crassa (Pr11) and the Tunisian O. ficus-indica (Pr45) have a low number of areole (1.5 areoles/cladode). Water reserve varied widely between 71.9 and 94.14 g for Pr11 and Pr23, respectively. The highest value for the mean distance between areoles was determined in the Tunisian O. ficus-indica (Pr45) with an average of 6.32 cm and the lowest also in the Tunisian one (Pr44) with an average of 2.22 cm (Fig. 1).
Cladodes of 45 studied ecotypes
Cladode quantitative traits of 45 Opuntia ecotypes
Different small letters in the same column indicate significantly different values within ecotypes at α<0.05, -: Set of successive letters between the two mentioned letters
The heaviest fruit was found in O. ficus-indica ‘Pr28' from Sicily with an average of 125.8 g. In contrast, the mixture of species such as Pr1, Pr35, Pr37, Pr43 and Pr44 had significantly the lightest fruits (Table 3). They do not differ significantly in properties, but do differ significantly in relationship with all other genotypes.
The seeds weight of the studied Opuntia ecotypes ranged from 3.78-0.47 g. The lightest value was recorded for Pr45 and Pr35 and the lightest one for Pr33. Seeds number varied widely between 290.5 (at Pr38) and 23.66 (at Pr35). Weight of peel ranged between 85.99 g in Pr9 and 7.13 in Pr35, while the weight of flesh fruit varied from 62.42-2.62 g being the highest in Pr28. This latter genotype showed the heaviest fruit fresh weight. However the Pr1 showed the lightest fruit fresh weight. The Pr37, Pr40, Pr43 and Pr44 did not differ significantly from the Pr1. With regard to the firmness at fruit maturity, the Pr37 was firm. Nevertheless, Pr2 and Pr43 have the soft fruit with an average of 8.72 and 8.82N, respectively.
Fruit quantitative traits of 45 Opuntia ecotypes
Different small letters in the same column indicate significantly different values within cultivars at α<0.05, -: Set of successive letters between the two mentioned letters
Fruit diameter was between 50.31 and 20.56 mm, the lowest values were recorded by (Pr1, Pr35, Pr37, Pr43 and Pr44). Fruit length varied from 9.46 cm for the Tunisian O. ficus-indica (Pr45) to 3.29 cm for the Algerian O. crassa (Pr35). Based on fruit size properties, O. ficus-indica such as Pr1, Pr3 and Pr44 were the Opuntia genotypes with the smallest fruits, while Pr28 was the genotype with the largest ones (Fig. 2).
Total Soluble solids (Ts_Fr) are between 15.33 °Brix for the Morrocain O. ficus-indica (Pr25) and 4.8 °Brix for the Algerian O. tomentosa (Pr37). The O. ficus-indica such as Pr31, Pr42 and Pr45 did not differ significantly from Pr37 (Fig. 3). The pH of fruit juice varied between 5.4 in O. laevis (Pr7) to 2.96 equally for O. crassa (Pr35) and O. ficus-indica (Pr45) (Fig. 4).
Qualitative characteristics of fruits and cladodes: The fruit trait such as length of receptacle scar (Ci_Fr) showed noteworthy differences among genotypes. This parameter varied between slightly to strongly depressed. It can be seen that the main differences among the 8 studied species were related to the juice color (Fig. 5) and fruit shape (Fig. 2).
Fruits of 45 studied ecotypes
Total soluble solids for fruit juice of 45 Opuntia ecotypes
Qualitative traits of 45 Opuntia ecotypes
Generally, the Opuntia species had fruits with broad elliptic shape and orange to purple juice color (Table 4). Regarding the shape of cladode (FR), a noteworthy degree of variability among genotypes was observed (Fig. 1). The shape of cladode varied from medium-narrow elliptic to oblong (Table 4).
Correlation among variables: Simple correlation coefficient analysis showed the existence of significant positive and negative correlations among quantitative characteristics (Table 5).
Principal component analysis (PCA) was used to examine the variation of Opuntia species based on quantitative traits.
The first three axes accounted for 58.62% of the variability among 45 ecotypes (Table 6). The first PC axis (PC1) accounted for 5.74% of the variation, whereas the second (PC2) and the third axes (PC3) accounted for 2.67 and 2.13%, respectively. The first one was mainly related to cladode thickness, cladode fresh weight, fruit fresh weight, fruit diameter, seeds number, seeds weight, peel weight, flesh weight, fruit length, total soluble solid and pH of fruit juice.
pH of the 45 Opuntia ecotypes fruit juice
Juice color of the studied Opuntia ecotypes
Dendrogram for cluster analysis
showing the 23 morphometric parameters relationships between the 45 ecotypes studied in hendi Zitoun (central region of Tunisia)
→: Cutting level
Correlation among quantitative traits
*Significant at α<0.05, **Significant at α<0.01, Ns: Not significant
The second principal component (PC2) was highly correlated to width of cladode and mean distance between areoles. The third component (PC3) was determined by cladode length, cladode dry weight, areole number per cladode, cladode water reserve and fruit firmness. The PCA showed that 18 phenotypic traits are the discriminating characters for the 45 Opuntia genotypes in Tunisia.
The different Opuntia ecotypes are identified, based on the similarity of their morphological characteristics and their hierarchical clustering are shown in Fig. 6. The populations were grouped into five clusters by cluster analysis. These 5 groups can be considered as distinct germplasm pools. Genotypes in the same group have the greatest similarity.
Principal component analysis (PCA) for the 45 ecotypes studied in Hindi Zitoun central region of Tunisia using quantitative traits
The Opuntia ecotypes collected from Central Region of Tunisia and assessed by using morphological traits, showed a wide variation. The cladode dimension varied independently on the species which is in contradiction of the results found by Feugang et al.17. The O. ficus-indica (Pr45) with the greatest cladode length and width and O. crassa (Pr35) with the highest thickness will be selected in the objective to increase fruit production and improve cattle feeding in the arid and semi-arid zones of the country, which largely depend on Opuntia plant during droughts.
The cladodes with the highest dimensions (length, width, thickness, fresh weight and water content) are responsible for a large amount of mucilage which seems to be the most important limiting factor for new consumers using as forage. Opuntia species can provide a continuous, valuable supply of fresh fodder during the dry season, given its succulent non-deciduous vegetative structure, a feature rarely found in other forage species.
The presence of spines on the cladode is a serious impediment to widespread utilization of Opuntia genus. The Mexicain O. crassa (Pr11) and the Tunisian O. ficus-indica (Pr45) possess the low number of areole per cladode (1.5 areoles/cladode) and can be classified as spineless species. However, the Tunisian O. ficus-indica (Pr44) was the spiny species and had the greatest number of areoles/cladode. This latter, is more aggressive and better adapted to spread.
Spineless Opuntia species are thought to be the result of domestication and hybridization to facilitate their use as forage and human consummation. Zimmer18 reported that there is evidence that only spineless forms were introduced to South Africa more than 250 years ago and they reverted back to the original spiny form over a period of nearly 200 years. These clues suggest the existence of recessive genes associated with spininess and confirm the ability of the species of Opuntia to reproduce from seed.
The O. ficus-indica (Pr28) from Sicily recorded the highest fruit fresh weight (125.83 g) which explain the adaptation of this one to arid Tunisian climate. Nevertheless, the Tunisian O. ficus-indica (Pr1, Pr40, Pr43 and Pr44) and the two Algerian species O. crassa (Pr35) and O. tomentosa (Pr37) registered the lowest fruit fresh weight (Table 3). This large difference in fruit fresh weight can be under various factors such as the richness of the soil in mineral elements, climate conditions and genetic potential of the species. Reyes-Aguero and Valiente-Banuet4 showed that fruit weight in Opuntia is affected by the order of production of the flower bud and the number of fruits on the cladode. Thus, floral buds that sprout earlier usually become heavier fruits. Furthermore, the heaviest fruits are obtained from cladodes with only six fruits19,20.
The Tunisian O. ficus-indica (Pr38) contains 290 seeds/fruit. However, the O. crassa (Pr 35) contain the few number (23.66 seeds/fruit).The O. maxima registered an average of 164-161 seeds/fruit which is in accordance with these found by Vila and Gimeno21. Several authors have reported a great variation in the number of seeds, from 1-5 to more than 2000 seeds per fruit4,18. This variation is observed within and/or between species depending on factors such as the age and size of the plant and the number of flowers/plant17. The presence of a great number of normal seeds in the Opuntia fruit is considered an obstacle for broadening its commercialization22.
Flesh firmness, skin color and total soluble solids are the main parameters to assess the fruit maturity at harvesting. The O. ficus-indica (Pr38) is the firmest fruit (18.81 N) in comparison to the Ethiopian one (Pr2) with an average of 8.72 N. The O. ficus-indica (Pr38) characterized by high firmness can be stored for a long time. Firmness is a relevant property for consumer acceptability and quality control23. This fruit consists of a thick peel, covered with small thin spines, enclosing a sweet juicy pulp intermixed with many hard seeds. So, Firmness can be related to the thickness of peel. Result found by El-Gharras et al.24 in Morocco showed that the fruit juice of Opuntia genotypes contains about 11-16% total soluble solids which is higher than the values of fruit juice of this collection. A total soluble solid does not depend on environmental conditions while fruit size increases with moisture availability during fruit maturation25. In general, it may be concluded that the knowledge of pH and total soluble solids in fruit may prove to be a powerful tool in evaluating both fruit maturity and quality.
The difference detected in morphological parameters can be the result of genetic variability inherent to Opuntia species. This is confirmed by the studies that have been conducted by Arba26, where he noted that the difference in vegetative and fruit production in the Opuntia explains that it has significant genetic diversity, which varies from one variety to another and from one locality to another.
Great variability was detected in this study with juice color. A natural dye would be extracted to be used as commercial food colorant.
Therefore diversity in morphological traits among genotypes may be due to climatic or genetic differences. The morphological traits are probably affected by genetic and ecological factors. Indeed, it has been reported that altitude, soil drainage, temperature and precipitation affect the development and morphology of Opuntia plants27.
This gene pool is useful in breeding program to produce other genotypes when combining the best morphological traits of the fruit and cladode with resistance to climatic and edaphic factors. It is necessary to know where the sterility barriers to hybridization occur within the currently species, such as O. ficus-indica, O. maxima, O. Crassa, O. laevis, O. tomentosa and O. helvetica.
Opuntia ficus-indica (Pr28) has the longest stalk and the heaviest fresh fruit. However, the O. tomentosa (Pr37) has the longest stalk and the lightest fresh fruit weight. This finding can confirm the absence of link between fruit weight and stalk length.
Simple correlation coefficient analysis showed the existence of significant positive and negative correlations among characteristics (Table 5). The Opuntia fruit traits such as (D-Fr), (Po-Ch), (Po-Pe), (Po-Pe) and Nb-Pe revealed a strong dependence on cladode fresh weight (PF-Ra) with (r = 0.48, r = 0.26, r = 0.31, r = 0.30 and r = 0.35), respectively. This latter was strongly correlated (P 0.01) with fruit diameter (r = 0.48) and less with weight of fruit flesh (r = 0.26), seeds weight (r = 0.31), peel weight r = 0.3) and seed number (r = 0.35). Areole number/cladode was also affected negatively by fresh cladode weight (r = -0.39) and cladode width and thickness with (r=-0.65) and (r = -0.31), respectively. The majority variables related to fruit size were correlated positively with cladode fresh weight at p = 0.01 (Table 5), indicating role of cladode to increase fruit size, this can be attributed to the photosynthetic rate of the cladode.
Hierarchical cluster analysis (Fig. 6) was carried out based on quantitative traits, allowed the assessment of similarity or dissimilarity and clarified inter and intra-specific relationships in studied Opuntia germplasm. There was diversity among genotypes, which were separated into 5 groups (Fig. 6). The first cluster (I) consisted of 11 O. ficus-indica from Tunisia (Pr12, Pr22, Pr24 and Pr45), Morocco (Pr5, Pr20 and Pr25), Mexico (Pr32) and Ethiopia (Pr2) and one O. halvetica from Tunisia (Pr21).This latter genotype, Pr25 and Pr20 were found to be very similar despite their belonging to two different species.
The second cluster (II) contained the majority of O. ficus-indica genotypes, O. maxima and O. laevis. O. maxima (P34 and Pr29) are distinguishable particularly by the low quantitative traits related to fruit and cladode. A series of homonyms were detected. Indeed, the O. ficus-indica species (Pr13, Pr27, Pr23, Pr4 and Pr3) and the O. laevis (Pr 8) were the closest, as, both species were characterized by high fruit firmness and water reserve. The O. ficus-indica (Pr 3 and Pr42) were probably homonyms and the O. ficus-indica (Pr31, Pr30 and Pr15) also can be considered as homonyms. The highest similarity was observed between the five O. ficus-indica genotypes (Pr17, Pr41, Pr36, Pr19 and Pr18) and the two O. maxima species (Pr34 and Pr29) identified as being synonymous with the smallest fruit. The two O. ficus-indica (Pr6) from Italy and Pr40 from Tunisia were easily distinguished from other genotypes. Then it’s only supposed that DNA fingerprinting could help to explain the morphological similarities among the closely linked genotype.
The third cluster (III) was divided in two sub-clusters, the first one with three similar couples (Pr16, Pr33), (Pr10, Pr14) and (Pr7, Pr9) and likely homonyms and the second with the Italian O. ficus-indica (Pr28).The Tunisian O. ficus-indica (Pr38) formed a separate group (IV) and displayed a distinct branching pattern indicates the native Tunisian originality.
The fifth group (V) contained three Opuntia species with high distribution which was divided into two sub clusters. The first contained one Mexican O. crassa (Pr11) and the second contained a mixture of different species from different geographical origin such as Pr1, Pr35, Pr37, Pr43 and Pr44. We distinguish that O. ficus-indica (Pr1) and (Pr43) from Tunisia were very closely related and they probably belong to the same genotypes with the smallest quantitative traits (Cladodes and fruits). However, O. carassa (Pr11) and (Pr35) From Mexico and Algeria respectively and O. tomentosa (Pr37) from Algeria showed few differences. This shows the adaptation of the introduced species and the modification of the morphological characters according to the Tunisian climatic conditions. While, Erre et al.27 found that soil and land characteristics were not correlated with morphometric parameters. It should also be noted that Opuntia ecotypes are distributed among the different groups independently of their locality (Fig. 6).
Dendrogram showed high variation between Opuntia genotypes indicating that studied germplasm can be considered in breeding programs as a good gene pool for different traits.
The analysis of morphologic traits can allow the investigation of the level of genetic diversity among Opuntia ecotypes from different regions of the world. Results of this study will be useful for conserving and managing genetic resources in Tunisia. These results showed large variations in the morphologic and pomological properties of 45 Opuntia ecotypes. However, this information may not be enough to characterize the cactus (Opuntia species) collected in the central area of Tunisia. Therefore, it could be necessary to characterize all the species based on biochemical and molecular techniques.
This study discover several morpho-chemical characteristics of Opuntia species like natural dye, number of seeds, °Brix… that can be beneficial for human health. This study will help the researcher to uncover the critical areas of commercial exploration in agricultural activity and this collection of 45 ecotypes can be considered in breeding programs as a good gene pool for different traits that many researchers were not able to explore. Thus a new theory on drawing up a descriptor peculiar to the genus Opuntia may be arrived at.
We thank the technicians of the forest ecology laboratory (LEF) for their help in practical parity and the anonymous reviewers for constructive comments on the manuscript. This work was funded by the National Research Institute for Rural Engineering Water and Forestry (INRGREF) in Tunisia.