Morphological, Phytochemical and Molecular Characteristics of NSIC-Registered Varieties of Garlic (Allium sativum L.) in the Philippines

Garlic (Allium sativum L.) is a bulbous flowering plant belonging to the Amaryllidaceae family that is native to the Central Asian region extending to Northern Iran. It is cultivated throughout the world with Asia having 90% of the world’s production, the bulk of which is coming from China with more than 20 MT of annual production in 2021 (Food and Agriculture Organization of the United Nations, 2022). Garlic is commonly used as a spice and ingredient in most Filipinos dishes (Lopez and Anit, 1994) and for traditional medicine in some indigenous communities to treat dog bites, phlegm and measles (Tantengco et al., 2018).  Currently, the Philippines has seven NSIC-registered varieties of garlic, namely–Ilocos White, Ilocos Pink, Batanes Red, Mexican, Bang-ar, Ilocos Tan Bolters and MMSU Gem (Bureau of Agriculture and Fisheries Standards, 2021). Despite the availability of these newly registered varieties, there is a dearth of information regarding their morphological, phytochemical and molecular characteristics.

Characterizing germplasm materials contributes to the efficient and effective use of genotypes for future crop improvement. It also allows the breeders to avoid duplication in collections (Dalir and Safarnejad, 2017). Morphological characterization is the first step in classifying and assessing plant genetic resources for it allows for identifying and selecting desirable traits (Malek et al., 2014). Likewise, phytochemical characterization is also an effective tool for differentiating among medicinal plant germplasm (Nsuala et al., 2017). Since garlic has vital use as food and medicine, the World Health Organization (WHO) emphasized the evaluation of the effectiveness of drugs found in the plant (Ayyanar et al., 2008). Several techniques have been adopted to detect the presence of chemical compounds in plants of which high-performance thin layer chromatography (HPTLC) method is commonly  preferred (Akabari et al., 2015). HPTLC remains one of the most flexible, reliable and cost-efficient separation techniques for analyzing botanicals and herbal drugs (Attimarad, 2011). On the other hand, molecular markers are essential in breeding programs and germplasm conservation for determining phylogenetic relationships, identifying species and gene mapping (Amom et al., 2017). Farooq (2002) emphasized the role of DNA fingerprinting in germplasm management to ensure genetically diverse and uniform germplasm.

Therefore, characterizing the different garlic varieties using these tools would provide baseline information important to the conservation and improvement of the crop. Thus, this study aimed to characterize the NSIC-registered garlic varieties along with two landraces for their morphological characteristics and phytochemical properties and to develop DNA fingerprints.

Germplasm acquisition
 
Physiologically mature and well-developed bulbs from average-sized to big cloves of the six NSIC-registered varieties and two additional local accessions were obtained from the Department of Agriculture-Ilocos Integrated Agricultural Research Center (DA-ILIARC) in the Philippines. The planting materials were protected from diseases and mechanical damage for good growth and development. Bulbs were prepared first by splitting the clove from each other with the thick protective scale intact. 
 
Crop establishment
 
The experiment was conducted from February 2020 to June 2021 at the Crop Breeding and Genetic Resources, Institute of Crop Science, College of Agriculture and Food Science (CAFS), University of the Philippines, Los Baños (UPLB). Plots measuring 1.5 x 0.3 m were prepared following two passes of plough and harrow at seven-day intervals. The different garlic varieties were assigned randomly to each plot containing 24 hills with a planting distance of 15 x 15 cm. Cloves were sown upright up to 2/3 of the length at 3 cm planting depth. Plastic tunnels were installed to protect the plants from excessive rain. Weeding was done twice a week or as the need arose either manually for those within the bed, or using a grass cutter when between plots. Irrigation was applied 1-2 days before planting to ensure sufficient moisture until 70-85 days after planting. Mature bulbs were harvested once 75% of the leaves have turned yellow or approximately 100-120 days after planting.
 
Morphological characterization and evaluation
 
The eight garlic accessions were characterized using the Allium spp. descriptor lists developed by the International Plant Genetic Resources Institute, European Cooperative Programme for Crop Genetic Resources Networks (ECPGR) and Asian Vegetable Research and Development Center (IPGRI, 2001). Morphological characterization focused only on the vegetative stage using only the bulb and leaf descriptors.

Phenotypic diversity was estimated using the standardized Shannon Weaver’s diversity index (H’) and following the diversity index criteria (H’ ≥0.67= high; 0.34 ≤ H′≤0.66 = intermediate; 0.01≤H′≤0.33= low) of Eticha et al., (2006). On the other hand, qualitative data were scored using presence (+) and absence (-). A dendrogram was generated based on the unweighted pair group method of the arithmetic mean (UPGMA) computed through Gower’s distance using XLSTAT software.
 
Phytochemical screening and alliin content determination
 
For each accession, 1.0 gram of fresh cloves was ground in 10 ml methanol and water solution having a ratio of 8:2, which was then filtered and subjected to further analysis. The derivation of reagents was performed with 0.6 g of ninhydrin dissolved in 190 ml isopropanol and 10 ml of acetic acid. Chromatographic conditions used HPTLC silica gel 60 F254 (Merck) plates at 10 x 10 cm or 20 x 10 cm. Plates were pre-washed with methanol and then dried at 120°C for 20 minutes. The mobile phases used n-butanol, water, acetic acid and formic acid with a ratio of (28:8:9:2). Sample applications were done using the standard, 3 uL and 5 uL of test solution, applied in 8 mm bands with 2 mm as minimum part and 8 mm from the lower edge of the plate. The developing solvent was done at 10 x 10 cm or 20 x 20 cm Twin through the chamber. The unsaturated 5 mL developing solvent was done by developing a distance of 60 mm from the lower edge of the plate. After drying for 5 minutes, the plates were immersed in ninhydrin reagent for 1 second, then heated at 120°C until the band’s colors were stable for 2-5 minutes. Examination of the results was done under white light. One-way Analysis of Variance (ANOVA) followed by post hoc analysis (Tukey’s highly significance difference test) was used to determine significant differences in alliin content.

The different Allium accessions were screened for the presence of alkaloids, phenols, terpenoids, saponins and tannins using alkali, acetate, Salkowski, froth and FeCl3 tests, respectively. The presence of specific phytochemicals was indicated by color changes (intense yellow for flavonoids, greenish-gray for tannins), formation of precipitates (white for phenols), interface (reddish-brown for terpenoids) and foam (saponins) (Mamta and Jyoti, 2012).
 
Molecular characterization
 
DNA extraction and quality check
 
Mature, healthy and well-developed leaf samples were collected and stored at -20°C before processing. DNA was extracted following the modified protocol of Doyle and Doyle (1987) and quantified using Epoch Microplate Spectrophotometer (Biotek, USA). The quality was evaluated by running the samples in agarose gel electrophoresis. Bands were visualized by staining the agarose gels with GelRed™ (Biotium, USA) and viewed using the UVIdoc Documentation System.
 
PCR Amplification using SSR markers
 
DNA sequences were amplified using the three microsatellite primer pairs (Table 1) through polymerase chain reaction (PCR). The final volume of the PCR mixture was 20 µl containing 1X PCR buffer, 1 U/µl Taq Polymerase, 1.5 mM MgCl2, 0.2 mM dNTPs, 0.2 µM each of forward and reverse primers and 520 ng/µl of DNA. PCR reaction began with an initial denaturation at 94°C for 3 minutes and was followed by 35 cycles of denaturation at 94°C for 30 seconds, primer annealing at each annealing temperature (Table 1) for 45 seconds and elongation at 72°C for 1 minute. This concluded with a final elongation step at 72°C for 10 minutes.


 
Molecular marker scoring and analyses of banding patterns
 
The PCR product was resolved on 1.4% and 1.7% agarose gel by running at 100 V for 30 minutes. A 50-bp DNA ladder was used as a molecular size standard. The gels were stained with GelRed and subsequently photographed using a gel photo-documentation system bands were scored as present (1) and absent (0) for each SSR marker. DNA fingerprints were visualized through the elucidation of their banding patterns. Polymorphism information content (PIC) value was calculated for each primer following the method of Botstein et al., (1980).

Morphological characteristics
 
Qualitative descriptors for both clove and bulb had low diversity, H’= 0.22 while quantitative descriptors had intermediate diversity, H’= 0.51 (Fig 1; Table 2).





For clove descriptors, both qualitative and quantitative traits showed intermediate diversity, H’= 0.38, except for the number of cloves per bulb, having low diversity (0.23). On the other hand, no diversity was found in all of the qualitative bulb descriptors but bulb skin color with intermediate diversity (0.34) while both the quantitative descriptors, bulb height and weight, exhibited intermediate diversity (0.65) (Fig 1; Table 3).



Interestingly, only the intensity of anthocyanin coloration at the pseudostem base registered high diversity, H’= 0.82, among all the qualitative descriptors while only the number of cloves per bulb recorded low diversity for quantitative descriptors. The low diversity observed for qualitative traits can be attributed to geographical relatedness as discussed previously in Egea et al., (2017) and Kumar et al., (2018) where accessions collected from neighboring locations tend to cluster together. Comparably, Trani et al., (2005) observed high genetic diversity for quantitative traits among garlic germplasm in Sao Paolo, Brazil. This is, however, contrary to the findings of Panthee et al., (2006) on the discriminating value of qualitative characters at the varietal level, as with the case of garlic accessions from greater Asia and Southeast Asia having highly similar quantitative traits for clove and bulb (Hirata et al., 2016). This only confirms that morphological traits alone do not always readily reflect genetic variations among germplasm collections.

Cluster analysis using both qualitative and quantitative traits revealed three clusters (Fig 2).



Mindoro White can be delineated by its light violet clove skin colour. Batanes White, Bang-ar and Ilocos Pink showed similar morphological characters having long bulb height and low intensity of anthocyanin coloration at pseudostem base. On the other hand, the rest of the accessions have relatively smaller bulbs exhibited in its bulb height and weight.
 
Alliin content determination
 
The eight garlic accessions were found to contain significantly different alliin contents (Table 4).



Among these, Bang-ar and Mindoro White showed the highest, 37.70±0.70 mg/g and lowest, 17.22±0.68 mg/g, alliin contents, respectively. The observed higher alliin content in the varieties grown in areas under Type I climate (with pronounced wet and dry seasons), compared with Mindoro, cultivated within Type 3 climate (without pronounced wet and dry seasons) corroborates the findings of Singh and Hiremath (2013) that alliin content varies among garlic varieties grown under differing climatic conditions. In addition, Huchette et al., (2005) confirmed that along with genotype influence, environmental conditions affect alliin accumulation in garlic. According to Rahman (2007), alliin is the main sulfur compound in raw and powdered garlic, averaging ~8 mg/g alliin per clove, with the highest attainable content of 20-25 mg/g in powder form. The alliin content of Philippine garlic varieties is comparable with those of the Indian garlics and higher than those from China (Siddiqui et al., 2016) and the different ecophysiological groups in Iran (Akbarpour et al., 2021) and Argentina (Gonzalez et al., 2009). Kim et al., (1994) reported that the degree of pungency is positively correlated with alliin content.
 
Phytochemical screening
 
Detection of phytochemicals revealed that only flavonoids and phenols are present in all the garlic varieties studied (Table 5).



Similarly, Arify et al., (2018) confirmed the presence of phenols and flavonoids and the absence of tannins in garlic accessions from India. Strati et al., (2018) regarded Allium species, especially garlic and leek, to be generally rich in phenols and flavonoids. Phenols are considered responsible for the pungency of garlic, while flavonoids influence the stability of Allium spp. when cooked and stored (Lanzotti, 2006).
 
Molecular analysis-SSR markers
 
Three primers were able to amplify five DNA fragments, three monomorphic and the other polymorphic. ASA-08 primer generated only one monomorphic band at 250 bp while ASA-10 produced two at 250 and 150 bp. On the other hand, two polymorphic bands were detected at 250 and 225 bp in the ASA-24 primer amplification profile (Fig 3).



This runs contrary to the results obtained by Kumar et al., (2018) where ASA-24 primer failed to amplify any DNA fragment and Anwar et al., (2020) where ASA-10 primer was found to be highly polymorphic. Furthermore, it is only in Batanes White that ASA-24 primer was not able to generate any DNA fragment at all amplicon sizes while MMSU Gem is the only accession that has one polymorphic band amplified by the same primer. Thus, only one primer, ASA-24, can be used to discriminate Batanes White and MMSU-Gem among the NSIC-registered varieties and landraces of garlic in the Philippines and could be used to delineate the registered varieties.

We were able to successfully characterize the six NSIC-registered varieties and two additional local accessions from the Philippines. Minimal differences were observed in the morphology among the different garlic accessions; however, distinctive differences were observed in their alliin contents and genotypic characteristics. The primer, ASA-24, showed polymorphic bands that can discriminate Batanes White and MMSU Gem from the other garlic accessions. The established characteristics could serve as an additional resource in identifying registered garlic varieties in the Philippines.

The authors would like to acknowledge the UP System Enhanced Creative Work and Research Grant (ECWRG-2020-2-25R), DOST-ASTHRDP and DMMMSU for their funding and support.