Comprehensive multivariate correlations between climatic effect, metabolite-profile, antioxidant capacity and antibacterial activity of Brazilian red propolis metabolites during seasonal study

The standardization of apiceutical products like as propolis extracts has been widely debated worldwide and variations in the propolis chemical composition are still very relevant topics for use-standardized of different propolis-type as medication by much of the world’s population. The present manuscript discuss important issues related to the climate effect and variations in propolis metabolite-profiling changes, antioxidant capacity and variations of the antibacterial activity of the Brazilian red propolis metabolites using comprehensive multivariate correlations. It was observed the increasing of guttiferones concentrations during the intense drought period and drastic decreasing in rainy period. The climate variation induced the high concentration of flavonoids in rainy period with pronounced dropped in some rainy months. The Pearson´s analysis demonstrated correlation between IC50 from DPPH and guttiferones and flavonoids concentrations. The PCA-X and Hotelling T2 test showed outliers during the months with lowest concentrations of formononetin and isoliquiritigenin was observed in antibacterial tests. The PLS-DA, OPLS-DA and VIP analysis demonstrate guttiferone E, guttiferone B, liquiritigenin, naringenin are considered important substances responsible by anti-staphylococcal activity in red propolis composition during the rainy season and drought period, but a synergistic effect with other flavonoids and isoflavonoids are not ruled out.

on-line DAD (200-600nm) and UV at 280 nm analysis. The MS detection range was from 100 to 1200 m/z set at 30,000 resolutions and the scanning was performed under ESI negative polarity mode with capillary temperature was 250°C. The mobile phase (0.1% of formic acid in H2O: 0.1% of formic acid in Acetonitrile; A:B) in flow rate of 300µL/min and ACE® C18 columns (100 x 4.6mm; 3µm) from (Hichrom, Reading UK) was used in separation of the phenolic compounds from red propolis. The gradient elution was programmed as follows: 0-6 min linear gradient 30% to 45% of B, 6-14 min linear gradient 45% to 75% of B, 14-20 min linear gradient 75% to 100% of B, 20-51 min at 100% of B for elution of the guttiferones and cleaning of the column, 51-54 min decreasing in B to 45%, 54-55 linear gradient 30% of B, 55-60min isocratic condition with 30% of B to re-equilibration of the column for next run. The injection volume was 10 µL.

S.1.3 Chromatographic Methods validation S.1.3.1 LC-UV-DAD
In LC-UV-DAD analysis the analytical standards of flavonoids were weighed (2.00 mg) and solubilized in 10 mL of a solvent system (methanol:H2O; 7:3 v:v) homogenized and filtered in PTFE syringe filters of 0.22 µm. These stock solutions were diluted for the concentrations of 0. 15, 0.25, 0.50, 1.00, 5.00, 7.50, 10.00, 12.50, 15.0 µg/mL for establishment of calibration curve during three different days. The concentrations of 5.00, 10.00 and 15.00 µg/mL were used as Quality Control Samples (QCS´s) during the intra-day and inter day precision and accuracy tests during three different days. The validation assays following ICH guidelines requirements (ICH, 2005) 90 . All stock solutions were stored in the refrigerator at the temperature of 2 ºC.
The suitability of the method was performed by injection of all standards in mixture at the concentration of 10 µg/mL prior to the work solution of crude extract of red propolis and always in triplicate. It was necessary to match the peaks of the identifiable compounds and avoid uncertain in retention time variation in this complex matrix. Robustness and parallelism were performed using aliquots of 250 µL, 500 µL and 750 µL of the analytical standards of quercetin and chrysin (200 µg/mL), which were added to get a concentrations of 5.00, 10.00 and 15.00 µg/mL of the quercetin and chrysin in final solution of the crude extract (250 µg/mL) of red propolis. This procedure was used for determination of inter-day precision and accuracy by standard addition in three different days and during the seasonal study. These tests were important establish the linearity of the dilution outliers of the calibration curve.
Crude extracts (100 mg) of the seasonality samples were weighed and solubilized in (10 mL) ethanol transferred to a volumetric flask and then an aliquot of 250 µL was diluted with methanol (10 mL) in volumetric flask to obtain work solutions (250 µg/mL) of the red propolis crude extract, which were directly injected in LC-UV-DAD. Aliquots of 250 µL, 500µL and 750 µL of the quercetin and chrysin analytical standards (200 µg/mL) were added to the final dilution of red propolis extract to obtain concentrations of 5.00, 10.00 and 15.00 µg/mL of quercetin and chrysin. Final concentration (250 µg/mL) of the crude extracts of red propolis also were directly injected in chromatographer during the seasonal study using LC-UV-DAD. These preparations were performed in all validation and seasonal samples to evaluate the suitability of the method during validation study and to serve as reference peaks during running of the seasonal samples. All seasonal samples were prepared freshly prior to be injected in LC-UV-DAD.

S.1.3.2 LC-Orbitrap-FTMS
In LC-Orbitrap-FTMS analysis, analytical standards of flavonoids or guttiferones (10.0 mg) were exactly weighed and solubilized in volumetric flask of 5 mL with methanol (stock solutions at 2000.0 µg/mL) and then the work solutions were prepared using a methanol. Aliquots of each standard were diluted for same volumetric flask in a pool of phenolic compounds ( a mixture of standards) to get concentrations of 10,000.0; 5,000.0; 2,000.0; 1,000.0; 500.0; 100.0; 50.0; 10.0 and 1.0 ng/mL and injected directly into the LC-Orbitrap-FTMS. Calibration curve, Accuracy, intra and inter-day precision were performed following ICH guidelines and USP requirements during a six different days and 1 ng/mL was determinate the limit of quantification.
The crude extracts (100 mg) of 36 samples of seasonal study were solubilized in ethanol (5 mL) to obtain stock solution of 20 mg/mL. Two work solutions were prepared at the concentration of 200 µg/mL (for flavonoids in low concentration) and 10 µg/mL (for phenolic compounds in high concentration) which were injected directly into the LC-Orbitrap-FTMS. The samples have been stable for a period of this study. The stock solutions of the standards and the samples of seasonality were stored in freezer (-20 ºC) and refrigerator (8 ºC), respectively, during a period of 30 days without degradation or decreasing in the area integration. All workday calibration curve was performed at the linear range of concentration and two replicates. The validation assay follows also follows the ICH guidelines 90 and FDA 91 requirements.
The determination of phenolic acids was standardized according the Folin-Ciocalteu method as follows: Gallic acid (100 mg) was weighed and transferred to volumetric flask and then solubilized in (10mL) distilled water to obtain a "stock solution" of 10,000 μg/mL. An (1mL) aliquot was removed of the stock solution and solubilized in water to a flask of 10 mL to obtain "work solution" of 1000 μg/mL. Aliquots of 20,25,40,50,60,70,75,80 and 100 μL were removed from work solution and transferred to (10mL) volumetric flasks and then was added 250 μL of Folin-Ciocalteu, 6 mL of distilled water and stirred for 1 minute, then an aliquot of 750 µL sodium carbonate (0.3 mg/mL) was added and stirred for 30 seconds. Distilled water was added to the volumetric flask to obtain concentrations 2.0, 2.5 4.0 5.0, 6.0, 7.0, 7.5, 8.0 and 10 µg/mL. The samples were stored in the place protected from light. Direct readings in UV were taken every 30 minutes over a period of 2 hours (30, 60, 90 and 120 minutes) the UV-VIS equipment at a wavelength of 750 nm.
The stock solutions (10 mg/mL) and (1 mg/mL) of the crude extract of propolis were prepared in the same conditions, but absolute ethanol was used to complete solubilisation. Aliquots of 250, 350, 500, 650 and 750 µL of the working solutions (1 mg/mL) were taken and transferred to the (10 mL) volumetric flasks and then was added 250 µL of Folin-Ciocalteu, 6 mL of distilled water and stirred for 1 minute, and then an aliquot of 750 µL of sodium carbonate (0.3 mg/mL) was added and stirred for 30 seconds. Distilled water was added to the volumetric flask to obtain concentrations 25, 35, 50, 65 and 75 µg/mL (working solutions). The samples were stored protected from light. The UV-Vis absorbance measurement were taken every 30 minutes over a period of 2 hours (30, 60, 90 and 120 minutes) the UV-VIS equipment was set at a wavelength of 750 nm. The assays were performed in triplicate and the total phenolic content was determinate. These methods were previously validated in the laboratory. Quality control samples of 35, 50 and 65 µg/mL were used for the precision and accuracy tests.
An alternative method also was standardized in laboratory using direct absorbance measurement using UV-Vis spectrophotometer set at the 280 nm wavelength. Determination was based on the concentration of catechin analytical standard using in an external calibration. Catechin analytical standard (20 mg) was weighed and transferred to a volumetric flask and solubilized in absolute ethanol (10 mL) to obtain a stock solution (2000 µg/mL), which was diluted to the concentrations of 1, 2, 4, 5, 6, 8, 10, 12, 15 and 20 µg/mL. Then, submitted to UV-Vis absorbance measurement at 280 nm. The red propolis extracts were previously submitted to desiccation in infrared oven at 105 °C during 15 minutes. Crude extracts (100 mg) of red propolis were exactly weighed and solubilized with ethanol 96 °GL in a volumetric flask (10 mL) to obtain a concentration of 10 mg/mL (Stock solution). The direct measurement were taken by UV spectrophotometry at the maximum wavelength (280 nm) after previous dilution of the sample to a concentration of 5, 10, 20, 30, 40, 50, 60 and 70 μg/mL (working solutions). The assays were performed in triplicate and the total flavonoid was determinate. These methods were previously validated in the laboratory. Quality control samples of 40, 50 and 60 μg/mL were used for the precision and accuracy tests (Supplementary dataset 1).

A) LC-DAD-UV and LC-Orbitrap-FTMS
The values of area obtained from the chromatograms integrations in the samples of propolis red extracts (Propolis A, B and C of the 12 seasonal months) were substituted in the variable (Y), of equation 1 (Y = AX + B), to find the value of concentration (X) correspond to each flavonoids or guttiferones quantified in diluted samples of these experiments. The percentile concentration (%C) of flavonoids or guttiferones quantified in red propolis extracts was performed using equation 2 and expressed as percentage and follows: , corresponds to the concentration (µg/mL) of the red propolis extract analysed in the UV-VIS spectrophotometer, being 25, 35, 50, 65 and 75 µg/mL for the total phenolic acid content, and 40, 50 and 60 μg/mL for the total flavonoid assay. These concentrations were used as quality control samples in the assay of determination of total phenolic acid content or total flavonoid content, respectively. It was possible to calculate the final concentration of total phenol content or total flavonoid content in the initially weighed samples (100 mg) of the red propolis extract from the percentage concentration (%C) (Supplementary dataset 1) and then, the results were expressed graphically.
The method of determination of specific flavonoids using LC-DAD-UV showed linear in the range of 0.15 to 15.00 μg/mL with intra-day precision < 5.00%, inter-day precision < 8.00% and an accuracy < 13.00% to the analytical standards tested (liquiritigenin, daidzein, pinobanksin, luteolin, genistein, isoliquiritigenin, formononetin, pinocembrin and Biochanin A) and quercetin and chrysin were used as internal standard for this method.

S.2.2 Determination of total phenolic content and total flavonoids content
The methods of determination of total phenolic acids using Folin-Ciocalteu and UV-vis at 280 nm methods were previously validated using three quality control samples. The total phenolic acids method presented intermediate precision of 4.30% and an accuracy of 4.53%. The method was linear in the range of 2.00 to 10.00 μg/mL (y = 0.1059x + 0.066) for the gallic acid analytical standard. The UV-vis reading method to catechin analytical standard presented an intra-day precision of 5.50%, intermediate precision of 5.98% and an accuracy of 7.06%. The assay showed an intermediary precision of 2.64% and accuracy of 3.54% with a linearity range between 1.35 to 20.00 μg/mL (y = 0.0687x + 0.0069; R2 = 0.9998) for the catechin standard. The crude extract presented intermediate precision of 6.90% and an accuracy of 8.80% and a linearity in the range of 5.00 to 70.00 μg / mL (y = 0.01126x + 0.0059; R2 0.9998). The UV-vis method and method of determination of total phenol compounds met the regulatory requirements for validation of analytical and bioanalytical methods. The complexity of the red propolis crude extracts justifies this greater variability (7.06% and 8.80%, and therefore 3.80%) higher than the limits recommended by regulatory agencies 29 in relation to the tests with the analytical standards of gallic acid and catechin.
The propolis A (Ilha do Porto apiary), propolis B (Primavera apiary) and propolis C (Paripueira apiary) had an average concentration of 26 mg, 30 mg and 33 mg of catechin/100 mg of Red Propolis Extract, respectively. The seasonal variation of Propolis A presented higher concentrations of flavonoids for the July, Propolis A, B and C showed an interesting total phenols profile with an average concentration of 17.85, 23.53 and 18.23 mg of gallic acid/100mg of propolis extract during a year cycle. The higher concentration was observed during August, November and July, (23.58 mg, 27.64 mg and 24.1 mg of gallic acid/100 mg propolis extract, respectively to Propolis A, B and C), respectively; and a decrease in concentration were observed in October (propolis A), June (propolis B) and April (propolis C) (14.87, 17.62 and 12.90 mg of gallic acid/100 mg of propolis extract, respectively) ( Figure S2

Competing interests
The authors declare that there are no competing interest including financial or non-financial.

Availability data
The authors declare the availability of the research data when requested.   Table S3. Meteorological data rainfall intensity during the time series between 2009 and 2019.

SUPPLEMENTARY TABLES
Cumulative amount of rainfall (mm) per month

500
-      The NOAA agency considers El Niño conditions to be present when the Oceanic Niño Index is +0.5C or higher, indicating the east-central tropical Pacific is significantly warmer than usual. La Niña conditions exist when the Oceanic Niño Index is -0.5C or lower, indicating the region is cooler than usual. Figure S6. Analytical method correlation between Folin-Ciocalteu method, UV-vis for catechin determination, LC-DAD-UV method and LC-MS method. Propolis A (A and B), Propolis B (C and D) and Propolis C (E and F). Graphs A, C and E were used in Pearson´s correlation among analytical methods. Graphs B, D and F (LC-MS data only) were used in IC50 from DPPH method and PCA analysis with MIC tests using Staphylococcus aureus and Pseudomonas aeruginona strains.