From Genetic Diversity to Economic Valorization: The Case of the ‘Zidi’ Fig Cultivar (Ficus carica L.) in the Region of Kesra, North-West Tunisia

Abstract

The use of fresh and dried figs has increased markedly in Tunisia in recent years, leading to a rise in public subsidies to support the further development of fig cultivation, especially in well-adapted production areas and with the adoption of innovative agronomic practices. This study aimed to carry out a comprehensive morphological and genetic characterization of fig germplasm from the Kesra region in north-western Tunisia, which is widely recognized for its long-standing tradition of fig production, in particular of the ‘Zidi’ cultivar. Field surveys and prospections enabled the identification of 26 distinct local fig cultivars, which were morphologically characterized and genotyped using 12 microsatellite (SSR) markers. All cultivars showed a specific allelic profile, including ‘Zidi’. The results provided valuable information for the conservation and management of Tunisian fig genetic resources. Moreover, the data will support the valorization of local fig production in the area through the establishment of the Protected Designation of Origin (PDO) ‘Fig of Kesra’, thereby contributing to the sustainable development of traditional orchards and the preservation of local agrobiodiversity.

1. Introduction

The fig tree (Ficus carica L.) (2n = 26) belongs to the mulberry family (Moraceae) within the genus Ficus, which includes more than 800 species [1]. The geographical origin of the fig remains debated: while [2] considered it native to the southwestern Mediterranean region, [3] identified Transcaucasia as the centre of origin and diversity, based on the presence of numerous transitional forms between wild and cultivated figs in the southern Caucasus. Botanically, fig is a gynodioecious species. In female trees, fruit set typically depends on pollination (caprification) by the wasp Blastophaga psenes L., which transfers pollen from male fig trees (caprifigs). Caprification has been practiced for centuries and traditionally involves placing male figs containing pollinating wasps onto female trees. Based on their reproductive system, fig cultivars are classified into four main types: Smyrna, San Pedro, common fig, and caprifig [4]. Smyrna figs require caprification and produce large, sweet fruits, whereas common figs set fruit without pollination. San Pedro figs produce two crops per year: the breba crop develops without pollination, while the main crop requires pollination. Their fruits are typically large and juicy, and have thin skin. Caprifigs do not produce edible fruits but play a crucial role as hosts for pollinating wasps. Their fruits are small and firm, contain numerous seeds, and are commonly grown near Smyrna or San Pedro figs to ensure effective pollination.

The fig tree is highly adaptable, thriving in hot, dry conditions, prolonged drought and poor soils, while preferring full sun and well-drained substrates. Long-lived and often centuries old, figs grow wild across the Mediterranean basin and play a key ecological role in maintaining ecosystem stability in semi-arid and arid environments [5].

Currently, Turkey, Egypt, Morocco, and Algeria are the leading fig-producing countries in the Mediterranean region [6]. However, Spain, Syria, Afghanistan, Brazil, and the United States also contribute substantially to global fig production. In Tunisia, fig cultivation covers approximately 37,774 ha under diverse environmental conditions, with an average annual production of 27,350 t reported in 2018. In particular, the mountainous Kesra El Olia region, located in the Tunisian Dorsal at approximately 1100 m above sea level, are renowned for both fig diversity and fruit quality. The Kesra region is well-defined, characterized by steep slopes, highly diversified soils, and a continental climate with cold winters and hot summers. Bioclimatic conditions range from subhumid to semi-arid, with average annual rainfall of 400–500 mm and relative humidity varying from approximately 40% during dry periods to 80% in wetter seasons. Despite these constraints, the region benefits from abundant water resources which support orchard irrigation, and allows a highly diversified polyculture system comprising fruit trees (olive, plum, pear, apple, cherry), vegetables, and horticultural crops. More than 20 locally named fig cultivars have been identified in the area, reflecting effective in situ conservation of local genetic resources. Among them, the cultivar ‘Zidi’ is the most widespread and highly appreciated, showing strong potential for commercial exploitation and representing a promising candidate for a future Protected Designation of Origin (PDO), ‘Fig of Kesra.’ PDO certification requires that products be produced, processed, and prepared entirely within a defined geographical area using recognized traditional know-how, with a demonstrated link between product quality and the local environment, and strict compliance with an approved production specification. However, the effective valorization of local germplasm requires more in-depth knowledge of fig genetic resources, overcoming current limitations in varietal identification, which still relies largely on morphological descriptors and only sporadically on molecular markers [7,8,9,10]. Simple sequence repeats (SSRs) are widely recognized as effective tools for crop true-to-type identification and traceability [11,12,13] and could play a key role in supporting the sustainable conservation and utilization of fig genetic resources [14,15,16,17,18,19].

The aim of this research was to characterize both morphologically and genetically the local fig cultivars of the Kesra region to support their preservation for sustainable use, with particular emphasis on the most widespread and valued ‘Zidi’ cultivar.

2. Materials and Methods

2.1. Plant Material

With the support of the local authorities, a comprehensive field prospection was carried out in 2022 in the Kesra region in the governorate of Siliana, north-west Tunisia, at an altitude of 1100 m above sea level (Figure 1). Farms involved in fig tree cultivation were identified and interviews with farmers were conducted to gather information on orchard management practices, cultivated varieties, production levels, and uses of fig fruits. Following the field survey, a total of 26 samples representative of the different types and cultivars grown in the area were collected (

Table 1. List of fig accessions from Kesra region (Tunisia) under investigation.

Cultivar	Type
Bargoug	Smyrna
Bedhenjel	Smyrna
Bezoul khadem	Smyrna
Bidh bhim	Smyrna
Bither	San Pedro
Bouabda	Smyrna
Dchich wa Assal	Smyrna
Dorghami	Smyrna
Harchi	Smyrna
Hemri	Common
Khedhri	Common
Khedhri Magloub	Smyrna
Neb Jmel	Smyrna
Sefri	Smyrna
Sefri Kares	Smyrna
Sekni	Smyrna
Slatni	Smyrna
Souidi	Common
Souidi Jwayed	Smyrna
Tiri	Smyrna
Tiri Boulanegue	Smyrna
Zergui	Common
Zidi	Smyrna
Zidi Hamam Kesra	Smyrna
Zidi Midenya	Smyrna
Zidi Nafessa	Smyrna

). Sampling was designed to capture the diversity present in the area while ensuring that each accession corresponded to a distinct, locally recognized cultivar.

2.2. Morphological Characterization

For morphological characterization, 20 fully developed leaves per sample were collected in late spring, while 26 fully ripened fruits were harvested between August and September. A total of seven descriptors—five quantitative and two qualitative—were assessed in accordance with the standardized descriptors established by the IPGRI-CIHEAM catalogue, including ‘leaf shape’ and ‘fruit shape’ (Supplementary Figures S1 and S2) [20]. Quantitative traits were measured in millimetres or grams and subsequently converted into categorical classes to facilitate comparison and analysis (

Table 2. List of morphological traits of leaves and fruit observed in Kesra fig accessions. * Blade shape classified from A to H according to IPGRI catalogue.

Organs	Trait	Trait Code	Descriptor No.	Description
				0	1	2	3	4	5	6	7	8	9	10	11	12
Leaf	Blade shape	BS	IPGRI 7.3.2	From A to H
	Lobe numbers	NL	IPGRI 7.3.3	Absent	Three	Five	Seven	More than seven
	Petiole length [mm]	PL	IPGRI 7.3.18		Short (<50)	Medium (50–80)	Long (>80)
	Petiole diameter [mm]	PD	-		Narrow (1–5)	Medium (6–10)	Large (11–15)	Very large (>15)
	Lobe length [mm]	LL	-		Short (<200)	Medium (200–350)	Long (>350)
	Lobe width [mm]	WL	-		Narrow (<200)	Medium (200–300)	Long (>300)
Fruit	Fruit shape [I = width/length]	FS	IPGRI 7.4.1		Oblong (I < 0.9)	Globose (I =0.9–1.1)	Oblate (I > 1.1)
	Skin Colour fruit	CF	IPGRI 7.4.2		Violet-black	Light green	Yellow	Yellow with red flush	Red	Purple	Dark purple	Reddish purple	Violet	Brown	Black	Green with red flush
	Fruit weight [g]	FW	IPGRI 7.4.5		Very light (10–30)	Light (30–50)	Medium (50–70)	Heavy (70–90)	Very heavy (>90)
	Fruit diameter [mm]	FD	IPGRI 7.4.6		Narrow (1–20)	Medium (20–40)	Large (40–60)	Very large (>60)
	Fruit length [mm]	FL	IPGRI 7.4.7		Short (29–46)	Medium (29–54)	Long (54–75)	Very long (>75)
	Ostiole form	OF	IPGRI 7.4.10		Closed	Semi-closed	Opened
	Ostiole diameter [mm]	OD	IPGRI 7.4.11		Narrow (<1)	Medium (1–3)	Large (4–5)	Very large (> 5)
	Skin cracks	SCr	IPGRI 7.4.21		Throughout	Many	Few	Without
	Skin thickness [mm]	ST	IPGRI 7.4.23		Very thin (<0.5)	Thin (0.5–1)	Medium (1–1.5)	Thick (>1.5)

).

2.3. Molecular Characterization

For genomic DNA extraction, young leaves were collected from each cultivar, immediately frozen and maintained at −20 °C until processing. Samples were lyophilized and finely ground using a TissueLyser (Thermo Fisher Scientific, Waltham, MA, USA), and DNA was extracted following the protocol described by [21]. DNA quality and concentration were assessed by electrophoresis on 0.8% agarose gel and by spectrophotometric analysis using a NanoDrop TM ND2000c (Thermo Fisher Scientific) spectrophotometer. DNA concentrations were standardized to 50 ng/μL using 0.1 X TE buffer (10 mM Tris–HCl pH 8.0 and 1 mM EDTA), and samples were stored at −20 °C until further use. Genotyping of the 26 samples was performed using a set of 12 SSR markers selected for their high informativeness in assessing fig genetic variability and for their ability to produce polymorphic, clear, and reproducible amplification profiles [22] (Supplementary Table S1). PCR reactions were performed in a final volume of 12.5 µL, containing 1× Dream Taq buffer, 0.15 mM dNTP, 0.25 μM primer mix, 0.3 U Dream Taq, and 50 ng genomic DNA. All PCR reagents were provided by Thermo Fisher Scientific. PCR products were prepared as described in [7] and separated using a SeqStudio Flex Genetic Analyzer (Applied Biosystems, Foster City, CA, USA) with GeneScan 600 LIZ as an internal size standard (Applied Biosystems). Alleles sizes were determined using GeneMapper v.5.0 software (Applied Biosystems).

2.4. Genetic Diversity

SSR markers exhibiting clear and distinct molecular patterns were used to estimate genetic diversity parameters using GenAlEx v.6.5 software [23]. The following indices were calculated: number of alleles (Na), effective number of alleles (Ne), Shannon’s information index (I), observed (Ho) and expected (He) heterozygosity, and fixation index (F). GenAlEx v. 6.5 was also used to perform Lynch and Ritland pairwise relatedness (LRM) analysis [24] in order to evaluate allelic similarity among genotypes and to identify potential cases of synonymy within the collection. Following [25], an identification key (Id) for unambiguous identification of the cultivar ‘Zidi’ was developed, and the parameters ‘confusion probability’ (Cj) and ‘discriminating power’ (Dj = 1 − Cj) were estimated for each SSR locus, based on the frequencies of SSR allele patterns observed across the fig collection.

2.5. Genetic Relationships of Tunisian Samples with International Varieties

To investigate the genetic relationships between Tunisian fig cultivars and other Mediterranean germplasms, SSR profiles of 44 varieties—including Apulian genotypes and Italian reference varieties retrieved from the UNIFG database—were analysed (Supplementary Table S2). Principal coordinate analysis (PCoA) was performed with GenAlEx based on Nei’s pairwise unbiased genetic distance matrix. In addition, an unweighted neighbour-joining tree was constructed with DARWIN v. 6.0.010 software (http://darwin.cirad.fr, accessed on 22 April 2026) [26] and the robustness of the branches was evaluated by bootstrap analysis with 1000 replicates [27]. The resulting dendrogram was visualized using FigTree software -v1.4.3 (http://tree.bio.ed.ac.uk/software/figtree/, accessed on 22 April 2026).

Population genetic structure was inferred using the Bayesian clustering approach implemented in STRUCTURE v. 2.3.4 software [28]. This method assigned accessions to a predefined number of populations (K) using a Markov Chain Monte Carlo (MCMC) algorithm. Ten independent runs were performed for each K value ranging from 1 to 10, with 100,000 MCMC iterations and 10,000 burn-in periods. The most likely number of clusters was determined using Structure Harvester software [29] based on the ad hoc statistic ΔK test [30]. Accessions with a membership coefficient (qi) greater than 0.5 were assigned to defined populations, whereas those with lower values were considered to have admixed ancestry [31].

3. Results

3.1. Morphological Diversity of Kesra Fig Accessions

The fig accessions exhibited considerable variation in leaf and fruit morphological traits (

Table 3. Morphological traits data observed in the Kesra fig accessions. (A) Leaf trait; (B) fruit traits. For blade shape, see Supplementary Figure S1.

(A) LEAF												(B) FRUIT
Accession	Blade Shape	Lobe Numbers		Lobe Lenght [mm]		Llobe Widht [mm]		Petiole Length [mm]		Petiole Diameter [mm]		Fruit Shape	Skin Color	Weight [g]	Length [mm]	Diameter [mm]	Ostiole Diameter [mm]	Skin Thickness [mm]	Skin Cracks	Ostiole Form
		Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD
Bargoug	B	4.9	0.9	204.8	30.9	190.6	24.6	82.7	19.5	4.7	1.0	Oblate	Yellow	Light	Short	Large	Large	Medium	Few	Semi-closed
Bedhenjel	E	3.5	0.8	165.7	26.2	160.6	16.5	67.5	30.8	4.5	0.4	Oblate	Yellow with red flush	Medium	Short	Large	Very large	Medium	Few	Semi-closed
Bezoul Khadem	E	3.1	0.4	237.6	60.7	202.9	33.3	97.4	16.3	4.9	0.8	Oblate	Brown	Medium	Short	Large	Very large	Medium	Without	Semi-closed
Bidh Bhim	A	5.3	1.5	219.7	29.7	208.0	29.7	104.8	22.0	6.9	11.0	Oblate	Red	Light	Short	Large	Very large	Medium	Few	Semi-closed
Bither	A	5.2	1.0	214.6	31.5	193.8	22.9	70.4	11.8	4.8	0.5	Oblate	Brown	Light	Short	Medium	Very large	Medium	Without	Opened
Bouabda	B	4.6	0.8	179.3	21.5	169.1	21.8	77.0	14.7	4.4	0.8	Oblate	Yellow with red flush	Light	Short	Large	Very large	Medium	Without	Opened
Dchich W Assal	B	4.0	1.0	214.8	22.3	187.1	21.6	91.5	14.5	3.5	0.5	Oblate	Yellow with red flush	Light	Short	Large	Very large	Medium	Few	Semi-closed
Dorghami	E	3.6	0.9	245.8	23.8	230.7	20.7	82.4	6.8	5.5	0.6	Oblate	Yellow with red flush	Light	Short	Large	Large	Medium	Few	Semi-closed
Harchi	C	4.9	0.8	201.7	24.4	179.1	14.2	89.3	19.6	5.7	6.6	Oblate	Brown	Light	Short	Medium	Large	Medium	Without	Semi-closed
Hemri	B	4.6	0.8	214.6	23.0	206.3	28.8	114.0	21.1	5.1	5.6	Oblate	Yellow with red flush	Light	Short	Large	Very large	Medium	Without	Opened
Khedhri	E	3.1	0.3	191.9	17.1	176.2	14.7	86.1	9.3	4.6	0.9	Oblate	Violet	Light	Short	Medium	Very large	Medium	Without	Semi-closed
Khedhri Magloub	E	4.1	1.2	210.9	20.9	191.1	29.8	87.5	21.3	5.4	0.7	Oblate	Violet	Light	Short	Medium	Very large	Medium	Without	Semi-closed
Neb Jmal	B	4.7	0.8	270.9	45.6	229.5	38.6	182.2	230.0	6.2	1.3	Oblate	Violet	Very light	Short	Medium	Very large	Medium	Without	Semi-closed
Sefri	E	3.9	0.9	217.1	30.0	203.3	29.2	77.2	13.3	4.9	1.0	Oblate	Violet	Light	Short	Medium	Very large	Medium	Without	Semi-closed
Sefri Gares	E	3.6	1.0	254.6	31.8	204.9	21.6	79.5	11.0	4.7	0.6	Oblate	Dark purple	Light	Short	Medium	Large	Medium	Few	Semi-closed
Sekni	E	4.1	0.8	205.5	20.6	193.2	19.4	92.3	21.1	4.4	0.7	Oblate	Brown	Medium	Short	Large	Very large	Medium	Few	Semi-closed
Slatni	E	3.3	1.1	218.8	23.5	190.2	21.1	77.2	18.7	5.4	0.7	Oblate	Violet	Light	Short	Large	Very large	Medium	Without	Semi-closed
Souidi	B	4.8	0.6	203.2	24.3	186.2	27.8	92.4	13.3	3.2	0.9	Oblate	Light green	Very light	Short	Medium	Large	Thin	Without	Semi-closed
Souidi Jwayed	A	4.8	0.5	177.8	17.4	202.0	132.0	83.5	24.9	4.4	0.5	Oblate	Violet-black	Very light	Short	Medium	Large	Thick	Few	Closed
Tiri	B	4.7	0.7	236.0	25.1	206.2	17.7	119.5	14.6	4.9	0.8	Oblate	Yellow with red flush	Light	Short	Large	Very large	Medium	Without	Semi-closed
Tiri Boulanegue	A	5.2	0.8	226.8	24.6	200.8	31.1	123.5	25.8	4.0	0.6	Oblate	Red	Very light	Short	Medium	Very large	Medium	Without	Semi-closed
Zergui	A	5.0	0.0	215.1	19.7	197.7	20.8	72.3	17.2	4.5	0.8	Oblate	Violet-black	Light	Short	Medium	Large	Medium	Without	Semi-closed
Zidi	E	4.2	0.9	306.1	50.6	281.4	39.6	110.8	15.8	7.2	3.1	Oblate	Violet-black	Medium	Long	Large	Very large	Medium	Without	Opened
Zidi Hamem	E	3.9	1.2	267.9	43.3	245.5	58.3	88.1	15.0	6.3	1.2	Oblate	Violet-black	Heavy	Short	Large	Very large	Medium	Few	Opened
Zidi Midenya	E	3.3	1.6	225.7	38.0	201.2	28.4	74.3	16.2	14.7	4.1	Oblate	Violet-black	Medium	Short	Large	Very large	Medium	Without	Opened
Zidi Nafesa	E	3.6	1.1	267.6	28.9	250.9	22.5	87.6	15.3	6.6	1.7	Oblate	Violet-black	Medium	Short	Large	Very large	Medium	Few	Opened

, Figure 2). Leaves were distributed in four out of the eight different classes (Table 3A, Figure 2 and Figure S1). Type A, characterized by a calcarate base and linear lobe, included the accessions ‘Bidh Bhim’, ‘Bither’, ‘Souidi’, ‘Tiri Boulanegue’ and ‘Zergui’; Type B, defined by a cordate base and spatulate lobes, comprised the accessions ‘Bargoug’, ‘Bouabda’, ‘Dchich Wa Assal’, ‘Hemri’, ‘Neb Jmal’, ‘Souidi Jwayed’ and ‘Tiri’. Type C, characterized by a calcarate base and lyrate lobes, was represented by the single accession ‘Harchi’. Type E, defined by a cordate base and three lobes, included most of the accessions, including ‘Zidi’, ‘Zidi Hammem Kesra, ‘Zidi’ Midenya’ and ‘Zidi Nafessa’.

Fruits were diverse in colour, ranging from violet/black shades (30.8%) to yellow with a red flush (23.1%) and uniform yellow (15.4%) (Table 3B). They were predominantly oblate and lightweight (57.7%), with a medium to large diameter. Pulp colour varied from brown (15.4%) to red (7.7%), dark purple (3.8%), and light green (3.8%). Most fruits had medium skin thickness with generally absent cracking (61.5%), while the ostiole was mostly very large in diameter (65.4%) but semi-closed (57.7%). ‘Zidi’ varieties were overall different, being characterized by large, violet-black fruits with thick skin and large opened ostioles. Only Bither, Bouabda and Hemri had an open ostiole, a trait shared with the four ‘Zidi’ accessions.

3.2. Genetic Diversity of Kesra Fig Accessions

The SSR markers generated clear and reproducible PCR banding patterns across all samples, enabling reliable discrimination between homozygous (single allele) and heterozygous (two alleles) genotypes (

Table 4. Genetic diversity indices revealed in the 26 Tunisian fig accessions analysed with 12 SSR markers: number of samples (N), allele range, number of alleles (Na), effective number of alleles (Ne), Shannon’s diversity index (I), observed heterozygosity (Ho), expected heterozygosity (He), fixation index (F), and PIC values.

Locus	N	Allele Range	Na	Ne	I	Ho	He	F	PIC
UFFc1	26	90–112	6	3.10	1.394	0.810	0.686	−0.178	0.640
UFFc2	26	212–220	3	2.28	0.950	0.610	0.561	−0.096	0.500
UFFc3	24	172–180	5	2.06	0.964	0.580	0.516	−0.131	0.460
UFFc4	24	210–214	3	2.08	0.780	0.500	0.52	0.038	0.410
UFFc5	25	108–124	2	1.85	0.653	0.560	0.461	−0.215	0.410
UFFc7	26	282–292	3	2.15	0.829	0.540	0.536	−0.006	0.430
UFFc8	26	124–130	2	1.30	0.395	0.270	0.233	−0.156	0.210
UFFc9	23	170–202	6	4.56	1.619	0.950	0.781	−0.225	0.770
UFFc10	26	128–140	2	1.16	0.271	0.150	0.142	−0.083	0.130
UFFc11	25	244–252	3	1.92	0.826	0.400	0.479	0.165	0.420
UFFc12	26	174–182	3	2.17	0.896	0.580	0.541	−0.067	0.470
UFFc14	26	84–102	3	1.86	0.750	0.500	0.462	−0.082	0.390
Total	303
Mean	25.25		3.42	2.22	0.860	0.540	0.49	−0.09	0.440

). Overall, 303 alleles were detected, with the number of alleles per locus (Na) ranging from two (at loci UFFc5, UFFc8 and UFFc10) to six (at loci UFFc1 and UFFc9). The effective number of alleles (Ne) varies from 1.16 at locus UFFc10 to 4.56 at UFFc9.

Locus UFFc10 exhibited the lowest polymorphic information content (PIC) (0.13), which indicates a reduced discriminatory capacity among the analysed genotypes. In contrast, loci UFFc1, UFFc2, and UFFc9 showed PIC values higher than 0.5, a threshold generally considered indicative of highly informative markers, resulting in a strong ability to distinguish the analysed individuals [32]. The collection expressed a mean Shannon’s information index (I) of 0.86, and an average observed heterozygosity (Ho) of 0.54, slightly higher than the average expected heterozygosity (He = 0.49). The resulting fixation index (F), close to zero (Table 4), indicated a minor deviation from Hardy–Weinberg equilibrium, suggesting possible hybridization or migration events generating variation that was subsequently selected and maintained by farmers.

All the accessions displayed distinct genetic profiles, with the exception of the four ‘Zidi’ samples collected from different locations within the Kesra region, which were genetically identical (LRM = 0.50) (Supplementary Table S3), and the accessions ‘Khedhri’ and ‘Khedhri Magloub’ which showed a high degree of genetic similarity (LRM = 0.44). These results were confirmed by the neighbour-joining dendrogram, in which the ‘Zidi’ cultivar clustered with the accessions ‘Bither’, ‘Bargoug’, ‘Harchi’, ‘Bargoug’ and ‘Bidh bhim’ within Cluster A, comprising the majority of the Kesra varieties. In contrast, the varieties ‘Tiri’, ‘Tiri Boulanegue’, ‘Souidi Yaied’, ‘Neb Jmal’ and ‘Bedhenjel’ formed a distinct Cluster B (Figure 3).

To facilitate the discrimination of the ‘Zidi’ cultivar from other local accessions, three highly informative SSR loci with high discrimination power (Dj)—UFFc1, UFFc2, and UFFc11—were selected, allowing the development of a molecular identification key specific to the ‘Zidi’ (Figure 4).

3.3. Genetic Relationships with Italian Varieties

A comparative genetic analysis was subsequently performed including the Kesra fig accessions and 44 Italian genotypes, among which the reference cultivars ‘Marangiana bianca’, ‘Morettina’, ‘Patona’, ‘Nero di Terlizzi’, and ‘Dottato’ were included. This analysis confirmed the genetic uniqueness of Kesra accessions (LRM < 0.50) and revealed several cases of synonymy among Italian accessions, namely ‘Processotto pinto’/’Processotto’; ‘Columone bianco’/’Fiorone verde’; ‘Zingarello nero Boci’/’Zingarello grigio’; and ‘Nero di Terlizzi’/’Terlizzi grosso’. A very high level of genetic similarity was also observed between the Tunisian cultivar ‘Slatni’ and the Italian cultivar ‘Sanbenedetto’ (LRM = 0.45). The Tunisian and Italian collections exhibited numbers of effective alleles of 3.8 and 5.1, respectively, and the Shannon information index (I) values were 0.955 for the Tunisian collection and 1.073 for the Italian collection (

Table 5. Genetic diversity indices Na, Ne, I, Ho, He, F and private alleles obtained for Tunisian and Italian fig genotypes obtained with SSR markers.

Accessions	Locus	N	Na	Ne	I	Ho	He	F	Private Alleles
Tunisian									Locus	Allele	Freq
	Mean	23	3.8	2.5	0.955	0.584	0.523	−0.109	UFFc3	170	0.024
	SE		0.6	0.4	0.174	0.091	0.076	0.022	UFFc4	214	0.548
									UFFc9	170	0.167
Italian									UFFc1	100	0.045
									UFFc2	199	0.012
									UFFc2	206	0.023
									UFFc3	152	0.011
									UFFc3	160	0.011
	Mean	44	5.1	2.6	1.073	0.562	0.564	0.026	UFFc3	174	0.011
	SE		0.5	0.4	0.147	0.091	0.067	0.075	UFFc4	208	0.068
									UFFc5	109	0.045
									UFFc5	130	0.011
									UFFc8	108	0.012
									UFFc8	168	0.012
									UFFc8	174	0.012
									UFFc9	180	0.080

). The fixation index (F) was close to zero in both collections, which also displayed the presence of private alleles, reflecting germplasm-specific genetic signatures, in particular in the Italian varieties (Table 5).

Principal coordinate analysis (PCoA) revealed a partial separation between Tunisian and Italian accessions, with Tunisian genotypes mainly distributed in the third and fourth quadrants, whereas Italian cultivars were more broadly dispersed across all four quadrants (Figure 5). The first two principal components explained 25% of the total genetic variability. Several Tunisian accessions, namely ‘Bargoug’, ‘Souidi’, ‘Sekni’, ‘Bither’, ‘Harchi’, ‘Bezhoul Khadem’, ‘Sefri’ and ‘Sefri Kares’, including ‘Zidi’, were interspersed among Italian genotypes, suggesting a shared genetic background or historical germplasm exchange between the two gene pools. Notably, ‘Zidi’ exhibited a strong genetic relatedness with the Italian cultivar ‘Carlo Magno’.

These results were supported by the unweighted neighbour-joining dendrogram (Figure 6). Cluster A comprises mainly Italian varieties together with the Tunisian accessions Hemri, Slatni, Bither, Bezhoul Khadem, and Sefri Kares, whereas the majority of Tunisian accessions, including ‘Zidi’, were grouped in Cluster B, in close proximity to the Italian cultivars Terlizzi Grosso, Nero di Terlizzi, Borsamele Bianco, and Carlomagno.

Structure analysis through the Evanno method [30] assigned individuals to the same genetic cluster when the inferred membership coefficient exceeded 0.60 and showed that the best number of K was three, allowing the identification of three ancestral lineages (Figure 7A). Cluster q1 included 18 Italian accessions, Cluster q2 included 12 Tunisian varieties and the Italian cultivar ‘Processotto’, and Cluster q3 encompassed both Italian accessions and Tunisian ‘Bither’, ‘Bezoul Khadem’ and ‘Sefri Kares’ and ‘Slatni’. ‘Zidi’ showed an admixed genetic profile, together with ‘Souidi’, ‘Hemri’, ‘Bhid Bhim’, ‘Bouabda’, ‘Khedhri’ and ‘Harchi’. When the two collections were analysed separately, contrasting patterns of genetic structure emerged, with the three ancestral components contributing differently to the genetic structure in genotypes from each country (Figure 7B). Specifically, the Tunisian collection was characterized by a predominance of ancestry q2, whereas the Italian collection displayed prevalent contributions from the ancestries q1 and q3.

4. Discussion

The fig is a perennial species whose fruits are consumed fresh, dried or processed into various products for the food, health, and cosmetic industries. Despite centuries of cultivation, the species has received limited breeding inputs. Most varieties originate from the selection of seed-derived plants and have been maintained by vegetative propagation across different agroecological environments, forming groups of locally adapted genotypes shaped by both natural and human selection [33].

In Tunisia, fig is a traditional crop which represents a promising resource for the sustainable development of marginal rural areas, particularly in the context of increasing drought frequency, rising temperatures, and growing competition from imported food products [12,18]. Key priorities for local producers include large-scale valorization of local varieties, expansion of cultivated areas, yield improvement, rejuvenation of ageing orchards, and enhancement of fruit quality to develop competitive products suitable for export. In the Kesra region, fig cultivation is a major economic asset, with the predominant cultivar being ‘‘Zidi’, a highly valued and attractive cultivar known for its large size, intense colour, sweetness, and richness in magnesium, calcium, zinc and phenolic compounds with strong antioxidant activity, supporting its potential use in the food, cosmetic, and pharmaceutical sectors [34,35]. To support fig cultivation in the Kesra region and preserve its sustainable use, a program of characterization of fig germplasm in the area was carried out, with particular emphasis on the ‘Zidi’ cultivar.

Although morphological traits are strongly influenced by environmental conditions, phenotypic characterization remains a fundamental step in valorization programs [36,37], as fruit attributes such as taste, colour, and weight are key determinants of market success [38,39,40]. Morphological characterization of Kesra accessions revealed substantial biological diversity, with a wide range of well-differentiated high-quality fig types shaped by genetic and environmental factors. Leaf morphology was distributed among four classes, with the ‘Zidi’ accession exhibiting a distinctive large trilobate leaf with deep sinuses, a trait shared with accessions ‘Bezoul Khadem’, ‘Bedhenjel’, ‘Dorghami’, ‘Khedhri’, ‘Sefri’, ‘Sefri Gares’,‘Sekni’, and’ Slatni’. ‘Zidi’ also exhibited distinct fruit traits, producing large, heavy, oblate violet-black fruits with thick skin and wide opened ostioles. In contrast, Kesra accessions generally yielded smaller and lighter oblate fruits. Moreover, the four ‘Zidi’ samples collected from different locations were genetically identical, confirming the genetic stability of this cultivar. Altogether, these findings highlight the strong commercial potential of ‘Zidi’, as larger fruits are typically preferred by consumers and are often associated with superior flavour quality, whereas smaller fruits are usually destined for processing or canning [41,42]. Tunisian varieties displayed additional favourable commercial traits, including the absence of skin cracking, observed in 61.5% of the samples, and the presence of a semi-closed ostiole, observed in 65.4% of the accessions. An open ostiole was detected in ‘Zidi’, ‘Bither’, ‘Bouabda’ and ‘Hemri’. A small or closed ostiole is generally preferred for commercial fig production, since it limits the entry of rainwater, insects, and pathogens, thereby reducing fruit rot and disease incidence and improving overall fruit quality [43]. However, a wide opening facilitates the entry of the fig wasp required for pollination and fruit development and thus is considered a good trait in Smyrna and San Pedro fig types which are not self-fertile [44].

The distinctive phenotype of ‘Zidi’, along with its wide distribution and strong association with the Kesra territory, makes it a suitable candidate for protection under a Protected Designation of Origin (PDO) scheme. This would enable producers to achieve higher market prices and greater economic returns [45]. Although not in the EU, Tunisia is currently eligible to participate in EU quality certification schemes by submitting applications directly to the European Commission [46].

Within this framework, genetic characterization of figs from Kesra was undertaken as a key step in the certification process using 12 SSR markers, which are widely recognized for their efficiency in varietal authentication and traceability studies [11,37]. In vegetatively propagated fruit tree crops such as fig, the exchange of clonal material across regions—often under different local names—frequently leads to cases of synonymy and misnaming. This represents a major constraint for accurate genetic identification within quality schemes such as PDO [37].

Genetic analysis assigned distinct genetic profiles to each accession, with no cases of synonymy detected among the Kesra cultivars. Among the loci analysed, UFFc1, UFFc2, and UFFc9 proved to be highly informative, displaying high polymorphism and strong discriminatory power, making them particularly effective tools for fig genotyping. The ‘Zidi’ cultivar exhibited a unique and consistent genetic profile shared among samples collected from different locations, indicating low intra-varietal diversity and suggesting the cultivar is a single clonal lineage, likely resulting from extensive vegetative propagation driven by its desirable traits. Within the Kesra germplasm, the combined analysis of morphological and molecular data clearly distinguished ‘Zidi’, with only a partial genetic similarity observed with the varieties ‘Harchi’, ‘Bargoug’, ‘Bither’ and’ Bidh Bhim’. The genetic and morphological distinctiveness of these Tunisian accessions constitutes a key element for varietal traceability and the protection of geographical origin, central elements to PDO regulations. In other fruit species, apple, grapevine, and olive genetic fingerprinting is already integrated into PDO/PGI protection processes [47]. Accordingly, ‘Zidi’ appears to be well-defined and easily identifiable using the highly discriminant SSR loci UFFc1, UFFc2, and UFFc11. This approach provides an effective tool for the identification of the cultivar and is fundamental for registration and certification processes, as well as for its long-term conservation and valorization.

The comparison of Kesra fig accessions and Italian genotypes, including Italian references ‘Dottato’, ‘Morettina’ and ‘Nero di Terlizzi’, provided insights into the genetic relationships between the two germplasms. Both germplasms displayed private alleles, suggesting that they experienced some level of independent selection pressures. Nevertheless, the phylogenetic analysis also revealed a significant intermixing between the Tunisian and Italian varieties. Specifically, the Tunisian accessions ‘Hemri’, ‘Slatni’, ‘Bither’, ‘Bezhoul Khadem’, and’ Sefri Kares’ clustered with several Apulian fig varieties, whereas ‘Bargoug’, ‘Souidi’, ‘Sekni’ and ‘Bidh Bhin’ were closely associated with the Italian reference cultivar ‘Dottato’, one of the most valued and widely grown cultivars in Italy, particularly in Calabria (Fichi di Cosenza PDO) and in Cilento (Fico Bianco del Cilento PDO). These results were further supported by PCoA, which showed that the first two principal components explained 25% of the total genetic variability, a value slightly higher than those reported in previous studies on Mediterranean fig germplasm [27,43]. Both the phylogenetic analysis and PCoA thus highlighted a pronounced intermixing between Tunisian and Italian varieties, consistently indicating a close relationship of the Tunisian varieties ‘Zidi’ and ‘Bouabda’ and the Italian cultivars ‘Nero di Terlizzi’, ‘Borsamele bianco’, and, in particular, ‘Carlomagno’, all known for their large fruit size and high sweetness. This pattern likely reflects the historical exchange of fig germplasm across the Mediterranean basin through the maritime routes that connected North Africa and Southern Europe over centuries. Such exchanges likely promoted the dissemination of elite plant material selected for desirable agronomic traits. Following their introduction into new environments, genotypes derived from the same ancestral variety may have accumulated somatic mutations, which were subsequently maintained through vegetative propagation and recognized as distinct local cultivars [33]. Similar findings have been reported in other studies which underlined the key role of human-mediated dispersal, vegetative propagation, and regional selection in shaping fig diversity [19].

Structure analysis based on the Evanno method identified K = 3 as the most likely number of ancestries, each contributing differently to the genetic structure of Italian and Tunisian genotypes. In the Tunisian collection, ancestry q2 was predominant, whereas the Italian accessions displayed contributions from ancestries q1 and q3, with the latter representing a secondary component that was consistently shared with admixed Tunisian accessions. These results are in agreement with previous studies reporting complex patterns of ancestry and gene flow among Mediterranean fig populations [31,48,49]. In particular, according to [33], in the Mediterranean fig germplasm it is possible to identify three main genetic pools, one for the Western Mediterranean (Spain/Tunisia/Italy) and another for the Eastern Mediterranean (Turkey/Caucasus), while the third gene pool would originate from the regions of Anatolia or Turkmenistan. Further and large-scale genetic studies on the genetic variability of F. carica across the Mediterranean region are needed for greatly enhancing our knowledge on domestication pathways and diversification of this iconic Mediterranean species.

5. Conclusions

Microsatellite markers are reliable tools for varietal authentication, the safeguarding of traditional cultivars, and the support of certification systems aimed at the valorization of local varieties [50]. They also represent instruments for sustainable germplasm management, thus preventing genetic erosion. In this study, microsatellite markers were used both to highlight the high richness and diversity of fig germplasm in the Kesra region and for the valorization of the ‘Zidi’ cultivar. This highly appreciated cultivar represents an excellent candidate for valorization through a PDO label, a recognition that could enhance its market value and provide a concrete economic incentive for local producers. The obtained results should raise awareness of the need to preserve the full spectrum of genetic resources, avoiding narrowing of the crop genetic base which may ultimately lead to genetic erosion [51]. Such an approach would contribute to sustainable rural development in the region and reinforce the importance of conserving these genetic resources which reflect a long history of local farmer-driven selection as part of the country’s cultural heritage. Although this study presents some limitations, including morphological characterization based on a single year and location and the lack of fruit-related genetic and biochemical analyses, it provides a valuable foundation for future research, encouraging closer collaboration among stakeholders.