Pyridine (Mallinckrodt Baker, Inc., Phillipsburg, NJ, USA); L-Butanol (Mallinckrodt Baker, Inc., Phillipsburg, NJ, USA); Sephacryl S-300-H (Sigma-Aldrich Co., St. Louis, MO); Standard Acemannan from Verapol Premium® (donated by Natural Aloe of Costa Rica S.A., 4 km Liberia-Guanacaste). H₂SO₄ (Mallinckrodt Baker, Inc., Phillipsburg, NJ, USA); 3,5-Dinitrosalicylic acid (Sigma Co., St. Louis, MO); Silica gel 60 G (Merck, Street 5, No.7 Industrial Park Alce Blanco, 53370, Mexico).

Freshly cut Aloe vera leaves were obtained from “Sabilerosdel Mayab” plantations located in Yucatan, Mexico. The leaves were harvested in two different seasons, the first in the months of August-September (rainy season) and the second in March-April (dry season), to evaluate the effect of water on the concentration of the polysaccharides.

The leaves of Aloe vera were washed and disinfected in chlorinated water for 20 min and allowed to dry. Two cross sections of 15 cm and 5 cm were performed at the apex and the base respectively, and allowed to drain to remove the bitterness. The side edges were cut to loosen the epidermis on both sides. The fillet obtained was homogenized with a blender to obtain a juice.

The composition of total carbohydrates was assessed using a colorimetric assay with phenol red (5% w/v in water) and concentrated sulfuric acid. The absorbance reading was 490 nm [7] .

The analysis of reducing sugars was performed following the DNS method [8] .

One liter of Aloe vera gel was filtered through four layers of gauze and then with Whatmanfilter paper no. 1 to remove coarse solids. The filtrate was passed through a membrane of 0.2 μm in order to remove smaller impurities.

Using an ultrafiltration device Amicon^® Ultra-0.5 (Millipore, Sigma-Aldrich Co., St. Louis, MO), we obtained two fractions of A. vera gel with a nominal molecular weight limit (NMWL) of 100 kDa. One fraction contained components smaller than 100 kDa (concentrate), and the other compounds larger than 100 kDa (filtrate).

The 100 kDa fraction was used to extract polysaccharides by size-exclusion chromatography (SEC). We used a chromatographic column of 95 cm of height and 1 cm of diameter, and a Sephacryl S-300 (Sigma-Aldrich Co., St. Louis, MO). The elution was performed with sterile bi-deionized water, and 80 fractions of 1 mL were collected. We used molecular weight markers (dextran blue) to estimate the size of the polysaccharides contained in the fractions. The 80 fractions were evaluated with thin-layer chromatography (TLC), using n-butanol: pyridine. water as solvent (6:4:3). The development was done with H₂SO₄ and H₂O (2:1), with subsequent heating at 105˚C.

To determine the content of polysaccharides in the Aloe extract, we quantified acemannan using a colorimetric assay [9] .

Samples of Aloe juice collected during times of drought and rain filtered with filter paper showed values of total sugars of 5.944 mg/mL and 2.988 mg/mL respectively, while the values for samples filtered with a 0.22 µm membrane were 4.348 mg/mL and 1,839 mg/mL, respectively (Figure 1(a)).

Similarly, it was observed that the content of reducing sugars in samples of Aloe juice collected during the periods evaluated was 2.71 mg/mL and 1.659 mg/mL in samples filtered with paper, and 2.695 mg/mL and 1.437 mg/mL in samples filtered with a 0.22 µm membrane during times of drought and rain, respectively (Figure 1(b)).

The dry season led to an increase in the concentration of total and reducing sugars in both fractions. Studies found that Aloe vera plants underwent osmotic adjustment due to the increase in total sugars, proline and other components when subjected to different amounts of water [10] . Likewise, previous studies have indicated concluded that the presence of these compounds can vary according to the different stages of growth [11] . It is possible that the increase of sugars is related to the amount of water during the rainy season.

The fractions were analyzed by thin-layer chromatography (TLC): concentrate (≤100 kDa) and filtrate (>100 kDa); we used glucose, galactose, arabinose and acemannan standards (Figure 2). Table 2 shows the retardation

factor (Rf) obtained for each applied sample. The Rf for glucose was 0.475, 0.425 for galactose and 0.5 for arabinose; these values are similar to those reported by previous studies [12] . The acemannan standard was revealed at the application point; two compounds were observed in the concentrate fraction, one at the point of application, and another with an Rf of 0.437. Two compounds were also observed in the filtrate fraction, one with an Rf of 0.437 and another with an Rf of 0.625. None of the Rf coincides with the reference patterns, indicating that these are different compounds.

The analysis by TLC of the fractions of the dry season obtained by size exclusion chromatography (SEC) yielded the record of a compound at the origin of fractions 28, 30 and 32, and another analysis yielded a fainter record of a compound in fraction 34, all at the application point of the fraction (Figure 3(a)). In the sample of the rainy season, a compound was observed at the origin of fractions 28 to 36 (faint spot) (Figure 3(b)), similar

to the standard of acemannan in both seasons. Similar results were reported in TLC chromatograms of the mucilaginous polysaccharide Basellaalba Linn, where the compound was found near the origin [13] .

The fractions with compounds larger than 200 kDa were grouped together, according to the results (Figure 4). The type and molecular size of the polysaccharides isolated from Aloe gel seem to be very diverse, as described by several authors [14] [15] . The heterogeneity of the molecular weight of the polysaccharides may be due to the techniques used to isolate them, or to their degradation by the activity of the endogenous enzymes [12] . Similarly, previous studies, collected polysaccharides of between 21 and 109 kDa from elutions of 20 to 60 mL in the neutral separation (distilled water) of carbohydrates present in Aloe arborescens [16] .

Figure 5 shows the concentration of acemannan in the fractions obtained from the size exclusion column in the two seasons evaluated. The fraction of low molecular weight showed 26.939 ppm and 9.364 ppm of acemannan in the dry and rainy seasons, respectively. As expected, the shortage of rain increased the concentration of acemannan by nearly twice (Figure 5(A)).

In contrast, the high molecular weight fraction showed concentrations of acemannan of 106.03 ppm and 99.97 ppm for the dry and rainy seasons, respectively (Figure 5(B)). The increase in the concentration of acemannan was not significant (6.06%) during the dry season, indicating that there is a higher concentration of acemannan in the fraction of molecular weight greater than 200 kDa, which is consistent with the reports of some authors that the average molecular weight of the native polysaccharides such as acemannan fluctuates around two million daltons or higher [14] [15] .

Recent research estimated the presence of acemannan as a phytomarker using the colorimetric method with congo red in commercial formulations of creams containing Aloe vera. The developed spectrophotometric method accurately measured the acemannan sample and it was found that the method was sensitive and accurate for routine analysis of the quality of Aloe vera in cosmetic formulations [17] .