The experimental assays in the present study were conducted on mature female inflorescences collected from nine date palm cultivars (seven located and two introduced) growing in continental and coastal regions. The different cultivars tested and their origins are presented in Table 1. DNm1, DNm2 (new date palm strains), and HR are three new, interesting and rare date palm genotypes. For each variety, there existed only one tree with no offshoots, and the possibility of their multiplication via the traditional offshoot method was, therefore, not feasible. Their cloning through the in vitro culture of vegetative tissues taken from the mother plant apex also presented the risk of losing uniquely available genotypes. DNm1 was the first new strain of Deglet Noor variety and was characterised by a large fruit size (the fruit size was bigger than the normal one by 30%); DNm2 was the second new strain and was characterised by a small seed size (1/30 of the total fruit size) compared to the normal seed size (1/10 of the total fruit size). In addition to maintaining all the characteristics of the Deglet Noor variety, the new characteristics (large fruit size and small seed size) added to DNm1 and DNm2 genotypes would secure them a higher economic position, particularly when propagated at a large scale since they met the international market requirements. The HR variety was issued from seed germination and had a good taste.

All the female inflorescences used in this study were collected in March immediately after the total emergence of the spathe, whose removal could be done easily without inflicting undesirable impacts on the subsequent growth of the mother plant. At this state, the lengths of the spathes and their relative inflorescences ranged from 33 to 40 cm and 26 to 37 cm, respectively.

This indicated that the cultured inflorescences were totally mature and belonged to the last stage of floral ontogeny [27,28].

The spathe external surfaces were rinsed with 70% (v/v) ethanol, and the inflorescences were removed and surface sterilized with 0.01% (w/v) mercuric chloride (HgCl₂) for 1 h. They were then rinsed three times with sterile distilled water under aseptic conditions. The inflorescence axes were sliced into 2 cm segments, each carrying two to three flowers, and were cultured by two segment groups on the callus induction media. The latter consisted of basal Murashige and Skoog (MS) [29] salts supplemented with KH₂PO₄ (120 mg/l), myo-inositol (100 mg/l), glycine (2 mg/l), L-glutamine (100 mg/l), adenine (25 mg/l), nicotinic acid (0.5 mg/l), HCL-pyrodoxin (0.5 mg/l), sucrose (50 g/l), and agar (8 g/l). Several 2,4-dichlorophenoxy acetic acid (2,4-D) concentrations (0.1, 1 and 10 mg/l) were assayed (Table 4). All the cultures were maintained in darkness at 28 °C for 8 to 10 months. In some cases, however, this period must be extended up to 12 months to obtain embryogenic or organogenic calli. The latter were transferred to the culture room at 26 ± 2 °C and 16 hour day-light condition, and subcultured at 3 months intervals.

The embryogenic calli obtained were incubated in a half strength MS liquid medium supplemented with activated charcoal (0.3 g/l) and 2,4-D (1 mg/l) [23]. They were then kept on a rotary shaker set at 120 rpm. The organogenic calli were transferred on a hormone-free medium (MS) or on a medium supplemented with 0.5 mg/l concentrations of BAP (6-benzylaminopurine) and α-naphthaleneacetic acid (NAA) to regenerate shoots which were finally transferred on a medium supplemented with 2 mg/l of Indole-3-butyric acid (IBA) for root formation.

The mature female flowers were fixed in Svaloff Navashine solution (chromic acid 0.5%, glacial acetic acid 5%, formaldehyde 15%, and ethanol 5%) for 48 h. The samples were then extensively washed with running water and dehydrated in ethanol solution series (from 50 to 100%). They were subsequently immersed in xylene-ethanol baths, followed by paraffin inclusion. Several 10-μm thick serial sections were prepared using a microtome, stained with Regaud ferric hematoxylin solution (Regaud hematoxylin solution 10%, glycerol 10%) [28], and then analyzed under a light microscope.

A X² test [30] was performed to analyze the statistical significance of the differences observed for the genotype and 2,4-D concentration effects on the percentages of callus formation. The different responses were determined once at the end of the culture period and the calculations were based on the initial number of explants.

Neoformations of different types were observed at various levels of the female flower, the most abundant of which was callus proliferation at the basal level of the flower (Fig. 1), notably in sepal and petal armpits (Fig. 2) which makes them more predisposed to this type of evolution than any of the other tissues of the mature flower at an entirely differentiated state.

Nevertheless, some proliferations can occasionally occur at various organs of the flower (sepals, petals, carpels) to contribute to the amplification process.

In fact, the histological examination of the female flowers at their final developmental phase, where they were totally differentiated and ready for pollination (Fig. 3), revealed the presence of three welded sepals, three largely-recovered free petals, six tiny ring-shaped staminodes located at the sepal and petal armpits, and three separated ovule-harboring carpels. This analysis also showed that these flowers were at the final stages of floral development (stages 7 and 8) [27,28].

The calli emerged distinctly in the armpit of one or several sepals or petals (Fig. 4) but also in the total central zone of the flower located between the carpels (Fig. 5).

A limited number of vegetative buds or embryos also appeared early from the flower tissues without an apparent manifestation of callus structure (Fig. 6). The careful examination of these structures confirmed the transit through the callus phase with a swift neoformation of buds and/or somatic embryos from undifferentiated tissues. This type of evolution occurred more frequently on media containing low 2,4-D concentrations (0.1 mg/l) where these structures remained for a sufficiently long period of time exceeding of more than 2 months.

While low 2,4-D concentration (0.1 mg/l) was noted to result in low percentages of callus formations for the DN and Mkn cultivars with no potential for plant regeneration, higher 2,4-D doses of 1 and 10 mg/l were observed to be much more efficient in the induction of high percentages of callus for most of the varieties tested, with a relatively important capacity of plant regeneration (Table 3).

In fact, when compared to the percentages obtained for callus formation using the medium supplemented with 2,4-D at a concentration of 1 mg/l, those obtained with 2,4-D at 10 mg/l were noted to be significantly higher at P < 0.0001 for the DN and Mkn varieties and at P < 0.001 for the Blo and Mkr varieties. Those concentrations were also noted to exert similarly significant effects in terms of the differences observed for the percentages of embryogenic/organogenic callus formations, which were statistically significant at P < 0.0001 for DN and at P < 0.0004 for Mkn and Blo varieties.

In addition to the effects attributed to 2,4-D concentration, a second factor that was equally implied in tissue response related to the source plant genotype of the inflorescences (Tables 3 and 4). In fact, several differences in terms of response were observed between the cultures maintained in the same conditions for the DNS, Ba, DN, HR, Blo, Gon, DNG and Mkn cultivars cultivated on MS + 2,4-D (1 mg/l) and the ones for the Mkn, Blo, DN and Mkr cultivars cultured on MS + 2,4-D (10 mg/l). A X² contingency test revealed that these differences were statistically significant (P < 0.001).

The data presented in Table 3 showed that for the initiation phase, the culture of explants on media containing 2,4-D at a concentration of 10 mg/l had a beneficial effect for it enhanced the percentages of induced calli and their future acquisition of new embryogenic or organogenic capacities. Several varieties were noted to show considerable flexibility even at 1 mg/l of 2,4-D, which was the case for the DN and Mkn varieties whose floral organs revealed high potencies that had never been reported before in the literature, namely the appearance of callus between the carpels and at petal axils, the transformation of the carpel into callus, and even the direct induction of the shoot.

Other varieties, such as Mkr, did not show any multiplication response only when 2,4-D was used at a concentration of 10 mg/l. The recalcitrance of this cultivar was confirmed by its very low reactivity and need for a longer period of culture on a 2,4-D rich medium (10 mg/l) (Table 3). In fact, Mkr was the sole variety among all the other varieties tested that did not show any reactivity to 2,4-D at 1 mg/l. Moreover, when 2,4-D was used at 10 mg/l, the Mkr cultivar showed the lowest percentage of callus formation of only 10.64 against the 35, 37.91 and 21.11% for the DN, Mkn, and Blo varieties, respectively (Table 3). The callus initiation process underwent by this variety was also observed to be very late (after 6 month) and to extend over a very long period (10 months to 1.5 years).

Once controlled, the multiplication technique based on mature female flowers was eventually selected for further application as a multiplication strategy and to avoid problems related to somaclonal variation, the medium for callus induction was used with 2,4-D at a concentration 1 mg/l (Table 4). In fact, 2,4-D at a concentration of 0.1 mg/l proved inefficient in terms of plant regeneration. At a concentration of 0.5 mg/l, on the other hand, it induced the formation of new callus and plants at relatively low rates when compared to those attained at 1 and 10 mg/l concentrations, especially for the DN and Mkn cultivars.

The findings demonstrated that maintaining the mature flowers on the same culture medium for a long period (3 months) before their transfer to another one, was beneficial and sufficient to achieve good new tissue evolutions specifically before the first subculture (Table 5). While this process was observed to limit the necrotic tissue rates, it was later noted to ensure a better evolution for the derived callus; otherwise the explants were noted to shrivel. In fact, previous research indicated that this period must be shortened only when the explants produce high amounts of phenolic compound oxidants that are highly toxic to the plant tissues [31,32].

The first key factor of the technique presented in the current study, which is based on completely differentiated flowers, related to the attainment of the callus. In fact, the return to the meristematic level is a highly sophisticated process that depends heavily on a complex of various physico-chemical factors, among which the hormones used are particularly important. In this context, the 2,4-D hormone has often been described as a potent auxin that has the ability to stimulate differentiated tissues to reach meristematic status without causing tissue damage or mutations and that should be used at relatively high concentrations [14,33].

3.3.1 Hyper-hydrated calli

3.3.2 Hydrated calli

This type of calli consisted of hydrated and partially separated cell groups that started to appear at various rates 2 to 3 months after the first culture, which was attributable to the 2,4-D hormone effect. The percentage generated for this type of calli revealed that it was the most abundant type among all the cultivars investigated, except for the DN and Blo varieties (42–62%) (Fig. 7a) (Tables 3 and 4). These calli showed normal growth (water content: 87.63%) and were able to regenerate plants.

3.3.3 Nodular calli

This type of calli consisted of nodules of different sizes (1–4 mm), usually yellowish (Fig. 8A) and attached or separated by undefined borders. They started to appear at the third or fourth month of culture and characterized by regular surfaces and hard consistencies. They were noted to develop within the petal tissues and around the point at which the flowers and their floral axes were joined. These calli showed slow growth rates that were maintained throughout the subsequent cultures, under the same conditions or with changes in hormone level or culture period. A few percentages of this type of calli were noted to differentiate a vegetative bud (organogenic tissues) (Fig. 8B) or small granules on their surfaces. These granules later formed embryogenic callus (Fig. 8C) (Tables 3 and 4). In some cultivars (Gon and Mkn), the nodules were able to proliferate and develop into stems within about 3 months of culture, which was the case for more than 6% of all the flowers cultured on appropriate culture media (MS basal medium supplemented with BAP and NAA at 1 mg/l concentration each).

The flower carpels, on the other hand, were rarely responsive. Overall, they were noted to follow one of two developmental patterns. In the first pattern, the carpels underwent an increase in size (5–10 times) but remained regular and hard. They did not show signs of development towards organogenic growth or callogenesis processes and remained at this state for a long period before an eventual necrosis. In the second developmental pattern, unlike the other flower parts, the carpels were noted to show an exceptionally significant increase in size. They also underwent a clear delay followed by a development of small granules at their external surface, which occasionally led to the formation of tissues that were able to develop into embryogenic calli (Fig. 9). This distinctive type of evolution followed by the carpel was observed starting from the 6th month using Mkn, Gon, and DNm2 varieties at a percentage of 1.64, 5.29 and 7.35, respectively. The results also indicated that the compact large-size nodules usually formed roots.

Longitudinal and transverse section analyses, carried out on flowers taken at different levels of callus formation process, revealed that the petals (Fig. 10) and flower bases (Fig. 11) were the most reactive organs, showing numerous granules in the tissues especially around the nutrition vessels. Those tissues were those that would later develop the abovementioned types of calli. The results indicated that different growth forms would later lead to the formation of shoots or somatic embryos, were discerned at the sepal and petal armpits.