Characterization of genetic diversity of cactus species ( Opuntia spp . ) in Morocco by morphological traits and molecular markers

The genetic diversity within and among 124 accessions of Opuntia spp. collected from different regions of Morocco was assessed using morphological descriptors and molecular markers. Based on 10 morphological traits, the accessions were separated into 3 main clusters; each cluster was containing accessions from different regions and species. Polymerase chain reaction (PCR) was then performed on 22 accessions from different regions and species, with 10 inter-simple sequence repeat (ISSR) primers and one random amplified polymorphic DNA (RAPD) primer. ISSR primers produced 66 bands overall, 64 (96.9 %) of which were polymorphic while 6 bands were generated by the RAPD marker, all polymorphic. The polymorphic information content (PIC) values ranged from 0.62 to 0.97, with an average of 0.82. The dendrogram of genetic differences generated using the unweighted pair-group method using arithmetic averages (UPGMA) method showed 7 different clusters at a similarity of 0.76, which was confirmed by the principal component analysis (PCA). The main conclusion of our work is the high genetic similarity between Opuntia ficus indica and Opuntia megacantha species in Morocco. Our results will be useful for plant breeding and genetic resource conservation programs. arTiClE hisTorY Received: 27 July 2017 Accepted: 5 September 2017


introduction
The cactus pear (Opuntia spp.) is a plant genus native to arid and semi-arid regions of Mexico 1 .It belongs to the family Cactaceae, which is consisting of about1500 species 30 and it is widely distributed throughout temperate, subtropical and cold regions (Feugang et al. 2006).In Morocco, the cactus pear is one of the most important crop plants in the semi-arid zones.It covers an area of about 120000 ha 5 and it is represented mainly by the species Opuntia ficus indica.The accessions of this species are classified into three distinct populations: (i) the Christians' nopal with prickly cladodes used as a field fence; (ii) the Muslims' nopal with inermis cladodes used as a green fodder for cattle and (iii) the Moses' nopal with large inermis cladodes that produce big pears 20 .The cactus pear population is present throughout the country, from Eastern to Southern Morocco 18 , with predominance in Guelmim-Sidi Ifni and Haouz-El Kelâa des Sraghnas regions 5 .Increasing the cactus pear area of cultivation is one of the strategies of the Moroccan Ministry of Agriculture.The cactus pear is a tree with high economic and ecologic value and various uses.It is cultivated for human consumption since it is characterized by important nutritional components 42 .The cactus pear has also many beneficial effects on health, including anti-cancer effect, anti-viral effect and anti-diabetic-effect among others 23 .On the other hand, the cactus pear may be utilized as forage due to its nutritional characteristics; along this line, Andrade-Montemayor et al. (2011) 4 reported that cactus cladodes could be used for goats feeding in semi-arid regions.The cactus pear was also successfully used for feeding Santa Inês lambs 13 and dromedary camel 15 .Furthermore, it was reported that some cactus constituents could be extracted and used as food additives or for pharmaceutical and cosmetic purposes 32 .The process of agricultural intensification and modern agricultural practices has led to genetic diversity degradation and gene pools loss  . Detemining the level of genetic diversity among and within plant populations is primordial for genetic resources conservation and breeding programs [24][25][26][27][28] .The use of morphological descriptors is of great help for genetic diversity assessment 14 .In the recent years, molecular markers have been proven powerful to determine the genetic diversity among and within species 12 and could be used as complementary methods to morphological descriptors 14 .Along this line, inter-simple sequence repeat (ISSR) and random amplified polymorphic DNA (RAPD) markers have been widely used to assess the genetic diversity in many species.Both the techniques are characterized by their simplicity, rapidity, ability to show high polymorphism with no prior knowledge of the genome studied [36][37][38][39][40][41] . In th recent years, several studies have been carried out on the genetic diversity of the cactus pear.For example, Valadez-Moctezuma et al. (2014) 39 studied the genetic diversity within and among Mexican Opuntia species using RAPD and ISSR markers.In Italy, Labra et al. (2003)  27 evaluated the genetic diversity in Opuntia species with cpSSR and AFLP markers. Zohlami et al. (2007)  44 have assessed the genetic diversity of Opuntia ficus indica L. in Tunisia using RAPD markers.More recently, Bendhifi et al. (2013)  6 published another work on the genetic diversity of the same species in Tunisia using morphological descriptors and RAPD markers.Despite the high importance of the cactus pear in Morocco, its genetic diversity has been scarcely studied.Therefore, the aim of this investigation was to assess the genetic diversity within and among Opuntia spp.from different regions of Morocco, using morphological descriptors as well as ISSR and RAPD markers.

Plant Material
In the following investigation, we used 124   1).The accessions are planted in the experimental site called Ain Nzagh of National Institute of Agronomic Research-Regional Research Centre of Settat, (INRA-CRRA Settat, Morocco).Originally, the accessions were collected from different geographical sites of Morocco as shown in Table 1 and Figure 1 then planted in the experimental site in June 2011.All the accessions have undergone the same culture edaphoclimatic conditions.

Morphological analysis
In this study, 10 quantitative morphological traits including plant height, plant diameter, average number of cladodes per plant, cladode height, cladode width, index of cladode shape (calculated as the ration of one-year-old cladodes length divided by their width), cladode thickness, mean distance between areoles, average number of spines per areole and mean length of the longest spine per areole.Data were recorded from 4 randomly chosen plants (2 years old) for each accession.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.All statistical analyses were performed with SPSS v 16.0 software (IBM, Chicago, IL, USA).

Molecular analysis dNa Extraction
In this study, we used 22 ecotypes from different geographic origins in Morocco, and which represent seven species of Opuntia identified from the morphological traits; for example, for O. ficus indica, we used 8 accessions, each one was collected from a different region.Total genomic DNA was extracted according to the protocol of Doyle and Doyle (1990) 17 with some slight modifications, using 1 g of lyophilized and ground cladode tissue.The DNA concentration was then diluted to 20 ng/µL with TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8) using a biophotometer (Eppendorf).

issr analysis
Fourteen ISSR primers (Table 2) purchased either from Operon Technologies Inc. (Alameda, CA, USA) or Qiagen (Valencia, CA, USA) were tested in this study; and 2.5 µL of each primer was added to a PCR mixture containing 3 µL DNA template, 12.3 µL sterile distilled water, 3.5 µL (25 mM) MgCl2, 1 µL(10 mM) dNTPs, 0.2 µL(5u/µL) Taq polymerase and 2.5 µL Taq buffer (All purchased from Promega, Madison, WI, USA).DNA amplifications were performed using a DNA thermal cycler (Techne) with the following parameters: initial denaturation at 94°C for 5 min, 45 cycles of 15 s of denaturation at 94 °C, 30 s of annealing at the specific temperature of each primer (Table 2), and 60 s of extension at 72°C; then a final extension at 72 °C for 7 min.

rapd analysis
For RAPD analysis, 34 RAPD primers (Table 2) purchased from The University of British Columbia (UBC; Vancouver, Canada) were tested.The PCR mixture was containing the same above-mentioned components for ISSR primers, but with substituting ISSR primers with 0.4 µM of each RAPD primer.Amplification conditions were as follows: initial denaturation at 94°C for 5 min followed by 30 cycles of denaturation at 94 °C for 60 s, annealing at 36 °C for 60 s, and extension at 72 °C for 90 s; with a final extension at 72 °C for 15 min.

issr and rapd Gel Electrophoresis
After amplification, 5 µL bromophenol blue (Sigma, St. Louis, MO, USA) was added to each amplification product.Afterwards, 8 µL of each sample was electrophoresed at 90 V, 5 W for 1 h on 1.8 % agarose gel (Promega, Madison, WI, USA) in 100 mL TBE buffer (90 mL sterile distilled water, 10 mL TBE buffer 10X; Promega, Madison, WI, USA) supplemented with 6.6 µL ethidium bromide (Sigma, St. Louis, MO, USA).Molecular weight of the amplified bands was estimated by comparison with lambda DNA/EcoR I and Hind III markers (Promega, Madison, WI, USA) as standard.The samples were then visualized and photographed with a gel documentation system (Fisher Scientific).

statistical analysis of Molecular data
For each gel, we recorded the presence (1)  or absence (0) of bands in individual lanes; consequently, a binary data matrix was created.For each marker, we calculated the polymorphic where pi and pj are the frequencies of the ith and jth alleles and n is the total number of such alleles.
The Dice's coefficients were calculated and a dendrogram was generated based on the similarity matrix and the SAHN module using unweighted pair-group method of arithmetic analysis (UPGMA).Afterwards, a principal component analysis (PCA) was performed.All these analyses were conducted using NTSYSpc v 2.1 software 34 .

results and discussion Morphological Traits to assess Genetic Variability Genetic Variability Within opuntia Ficus indica
Our investigation showed that Opuntia ficus indica is present in all the studied areas.However, our results demonstrated clear differences among the 78 accessions with regard to morphological descriptors.For example, the plant height was ranging from 21.33 cm in Ofi-531 (from Eastern Rif) to 48.25 cm in Ofi-762 (from Agadir) and the plant diameter was ranging from 16.43 cm in Ofi-275 (from Ouardigha) to 71.38 in Ofi-163 (from Rhamna).In some cases,    accessions from the same geographical site exhibited significantly different measures of morphological traits; for example, in accessions from Rhamna, cladode length was ranging from 16.30 cm in Ofi-141 to 29.13 cm in Ofi-163 (Table 1).The O. ficus indica accessions showed similarity coefficients ranging from 0.00 to 1.00 (proximity matrix not shown).The lowest similarity coefficient was noticed between accessions Ofi-212 and Ofi-262 from Ouardigha and between accessions Ofi-771 and Ofi-772 from Agadir while the highest similarity coefficient was observed between accessions Ofi-163 (from Rhamna) and Ofi-275 (from Ouardigha).In some cases, accessions from different regions show very low similarity coefficient (e.g.Ofi-C3 from Jamaât Riah and Ofi-561 from Eastern Rif displayed a similarity coefficient of 0.001).The dendrogram presented in Figure 2 suggested 3 major clusters.The first cluster contains 30 accessions (38.46 % of the total accessions); the second cluster contains 5 accessions (6.41 % of the total accessions) and the third cluster is comprised of 43 accessions (55.13% of the total accessions).Surprisingly, the three clusters contain accessions collected from different geographical sites.

Genetic Variability Within opuntia Megacantha
The plant height of O. megacantha accessions ranged from 24.67 cm to 40.33 cm, plant diameter from 21.65 cm to 65.38 cm, the average number of cladodes per plant from 1.00 to 4.25, cladode height from 17.30 cm to 37.00 cm, cladode width from 9.40 cm to 14.67 cm, the index of cladode shape from 0.33 to 0.70, cladode thickness from 0.53 cm to 1.40 cm, the mean distance between areoles from 2.05 cm to 3.15 cm, the average number of spines per areole from 3.00 to 5.00 and the mean length of the longest spine per areole from 1.70 cm to 3.80 cm.Here again, some accessions from the same region displayed significantly different measures of morphological traits (Table 1).The range of similarity coefficients was from 0.00 to 1.00 (proximity matrix not shown).The lowest similarity was between Om-B1 and Om-A3, both from Jamaât Riah, while the highest similarity was between Om-K1 and Om-K3, from Jamaât Riah too.The dendrogram (Figure 3) shows 3 clusters, the first cluster consisted of 11 accessions (28.95 % of the total accessions); the second cluster contains 20 accessions (52.63 % of the total accessions) whereas the third cluster comprised 7 accessions (18.42 % of the total accessions).Interestingly, accessions collected from different regions could be found within the same cluster.4), and accessions from different species or geographic origins could be gathered into the same cluster.This shows that the clusters obtained do not fit with the species or geographical sites evaluated.

Molecular Markers to assess Genetic diversity
After screening 14 ISSR and 34 RAPD primers, 10 ISSR and 1 RAPD primers were selected for producing unambiguous, reproducible and intense bands.The ISSR primers generated a total of 66 bands, 64 (96.9 %) of which were polymorphic.The number of bands observed per locus varied from 4 to 9, with an average of 6.6.The RAPD primer (UBC-571) gave 6 bands, all polymorphic (Table 2).
Based on the PIC values, the markers were highly informative.Indeed, the PIC values varied from 0.62 to 0.97, with an average of 0.82.The highest PIC value was observed in primer ISSR-8 while the lowest value was observed in primer ISSR-29 (Table 2).The dendrogram displayed 7 clusters at a similarity of 0.76 (Figure 5).O. inermis was separated from all the other species.This was confirmed by PCA (Figure 6), which results were similar to those obtained by the dendrogram.

discussion
In the present study, the genetic diversity of 124 accessions of Opuntia spp. was evaluated based on 10 morphological descriptors.The use of morphological descriptors to determine genetic diversity within plant populations is simple, rapid and inexpensive 9 .Thus, such descriptors were used in many plant species such as date palm 19 , hexaploid oat 16 and indigenous rice 43 25 highlighted a clear differentiation between Opuntia species using descriptors based on fruit, flower, seed, spine and trunk characteristics.In our case, the 124 accessions showed significant differences in the measured parameters and were distributed into 3 main distinct clusters based on Ward's method, containing 56, 45 and 23 accessions, respectively.Interestingly, accessions classified in the same species could be located in different clusters and vice versa, even though some species contain spines while others do not.This may be explained by several factors including the environmental effect and evolutionary change  . Our rsults are in good agreement with Valadez-Moctezuma et al. ( 2014) 39 who supported the hypothesis about the existence of a smaller number of Opuntia species compared to those currently described, but with high intraspecific genetic variation; and with Labra et al. ( 2003) 27 who suggested that the presence of spines should not be considered as a parameter in Opuntia taxonomy.Our results showed the influence of geographical origins on the morphological traits within the same species.Indeed, in some cases, accessions of the same species but from different regions were gathered into different clusters.This might be due to the environment effect [37][38] .Our results confirm previous findings on Opuntia spp. in Sardinia and Corsica islands, which show that environmental factors such as elevation, soil drainage, temperature and rainfall affect plant morphology 21 . Asar as Opuntia ficus indica is concerned, the use of morphological or agro-morphological traits to assess genetic diversity within its accessions is very scarce.Here again, significant differences were observed between accessions.In our case, the 10 morphological traits showed significant differences between the 78 accessions.Remarkably, some accessions from the same geographical origin displayed significantly different measures of some traits.The use of morphological markers to assess genetic diversity within populations is simple and inexpensive. Hoever, such markers can be affected by the environment 3 .Therefore, in the recent years, molecular markers have been widely used to assess genetic diversity within plant populations.Indeed, it was reported that molecular markers cover a larger proportion of the genome than the morphological ones and are less influenced by the environment since genetic variability is evaluated based on genotype rather than phenotype [26][27][28][29][30][31][32][33][34][35][36][37][38] .In addition, molecular markers have been proven powerful to detect variability when it could not be possible with morphological markers; i.e. a given morphological marker can affect other traits and the usefulness of morphological markers is restricted by their limited number 3 .Along this line, molecular markers have been applied alone or in association with morphological descriptors to assess genetic variability in many plant species; e.g. in Phoenix dactylifera L. 35 , Olea europaea L. 8 , Citrus sinensis L. 29   to assess the genetic variability within 36 Tunisian accessions, and a considerable genetic diversity was detected.In Morocco, the genetic diversity within O. ficus indica accessions was evaluated using RAPD markers 18 , and the effect of the geographical origin was revealed.In our case, findings showed high genetic similarity among the studied accessions of O. ficus indica, even though they were collected from different regions of Morocco.

Conclusion
This paper is the first report to assess the genetic diversity within and among Opuntia species in Morocco using morphological descriptors and molecular markers.A low level of genetic diversity was observed among the accessions analyzed and a high similarity between O. ficus indica and O. megacantha accessions was highlighted.Our findings show also that ISSR markers are very efficient and informative in Opuntia species.Our results should be taken into consideration for plant breeding and genetic resource conservation programs.Further investigations are currently underway in order to determine the biochemical characteristics of cladode, flower, fruit and juice of these species, and to specify their molecular families to highlight the most appropriate use of each species.

Fig. 1 :
Fig. 1: Map of Morocco showing the geographic origin of the 124 accessions of Opuntia spp.used in this study.

Fig. 2 :
Fig. 2: dendrogram grouping 78 accessions of Opuntia ficus indica based on 10 morphological traits and Ward's method.The red stars indicate accessions used for molecular analysis.

Fig. 3 :
Fig. 3: dendrogram grouping 38 accessions of Opuntia megacantha based on 10 morphological traits and Ward's method.The red stars indicate accessions used for molecular analysis.

Fig. 4 :
Fig. 4: dendrogram grouping 124 accessions of Opuntia spp.based on 10 morphological traits and Ward's method.The red stars indicate accessions used for molecular analysis.

Fig. 6 :
Fig. 6: 3d graphical distribution based on principal components analysis (PCa)showing the distribution of the 22 opuntia spp.accessions.

Table 1 )
. Among the studied species, O. ficus indica, O. dillenii and O. inermis are spineless.The similarity coefficients were ranging from 0.00 to 1.00.The dendrogram shows three different clusters (Figure . To the best of our knowledge, this is the first work to assess genetic variability within and among Opuntia populations in Morocco using morphological descriptors.Studies on the genetic variability in Opuntia species using morphological descriptors are scarce.Peña-Valdivia et al.(2008) 31assessed genetic diversity within 55 accessions of Opuntia spp. in Mexico based on 65 morphological descriptors; and indicated that only few morphological traits showed significantly different measures within the studied populations.In a more recent study,Helsen et al. (2009) 39ylogenetic relationships of 15 Opuntia species from Argentina, Bolivia, Brazil, Paraguay, and Uruguay33.Our findings are in contrast with those of Valadez-Moctezuma et al. (2014)39, who indicated that the assessment of genetic diversity with RAPD markers was relatively more effective than with ISSR markers.In our study, the total number of polymorphic bands (64) and the PIC values (0.62-0.97) of ISSR markers were different from those found by Valadez-Moctezuma et al. (2014)39, in which the number of polymorphic bands was 70 and the PIC varied from 0.21 to 0.27.This difference in polymorphic bands number and PIC values could be explained by the different ISSR primers employed (different sequences), the difference among accessions and the high distance between the geographical sites (Mexico and Morocco).Regarding the species Opuntia ficus indica,Zoghlami etal.(2007) 44 used 8 RAPD primers (over 22 screened)