3.1. Chemical Composition of Volatile Compounds
Probably the least studied aspect of propolis is the composition of its volatile organic compounds (VOCs), since a common approach to extracting them from the raw material is maceration followed by evaporation of the solvent, during which the most volatile components are inevitably lost. A number of studies have used labour-intensive techniques of sorption concentration [
42] or steam stripping [
43,
44] to determine the composition of VOCs. A much more efficient and less time-consuming approach is to use the HS-SPME/GC-MS technique, which combines the stages of concentration and sample preparation, which has proven itself well in the study of VOCs of European [
36,
45] and Brazilian [
46] honeybee propolis.
As a result of the determination of volatile substances by this method, 174 peaks were registered on the chromatograms of all four samples of propolis from SLBs. Typical VOC chromatograms are shown in
Figure 1. Each of the chromatograms contains 57 to 105 peaks of organic C
2–C
15 compounds of various classes. Registered components are divided into nine groups, which are shown in
Table 2 together with the main representatives of each of them. The complete composition of the volatiles, along with some of the analytical parameters used to identify them, is shown in
Table S1 in the Supplementary Data.
Although the VOCs of each of the four types of propolis contain representatives of all nine groups, their individual composition is quite specific: only 16 compounds (less than 10%) were common for all samples. Nine of them belonged to terpenes characteristic of plant essential oils (C10H16 monoterpenes α- and β-pinenes, myrcene, 3-carene and limonene, as well as C15H24 sesquiterpenes α-copaene, β-caryophyllene, α-humulene and δ-cadinene). Other common components were ethanol, acetic acid and their esters, isopentanol, 2-ethylhexanol and its acetate, and the aromatic hydrocarbons styrene and p-cymene. However, one can note the pairwise similarity of the VOC composition of propolis samples 1 and 4, as well as samples 2 and 3. VOCs of propolis 1 and 4 are characterized by a relatively low content of terpene compounds but a high content of C2–C9 alcohols and their esters, while terpenoids prevail in the secretions of propolis 2 and 3.
3.2. Extractive Compounds
Substances extracted with diethyl ether from SLB propolis are mainly relatively low-polarity compounds with one or two functional (carboxyl and/or alcohol) groups. The chromatograms of all four samples showed peaks of 287 compounds, of which 210 were identified based on mass spectrometric and chromatographic information. The most representative groups were formed by diterpenes and triterpenes (60 and 62 compounds, respectively), which also accounted for the largest part of the total ion current (TIC) of the chromatograms. The third-largest group (26 compounds) with a much smaller contribution to the TIC was formed by phenol derivatives, long-chain (C
15–C
19) alkyl- and alkenylphenols, resorcinols and salicylates. On the chromatograms of
T. clavipes propolis, 20 sesquiterpene compounds were registered; however, most of them belonged to minor components, the individual concentration of which did not exceed 0.1% TIC. Typical chromatograms of ether extracts are shown in
Figure 2. The group composition of these extracts is given in
Table 3, but the full composition is shown in
Table S2 in the Supplementary Data.
Of the total identified compounds, only six were present in all samples: diterpenes, abietic, dehydroabietic, isopimaric and 13-
epi-cupressic acids, and pentacyclic triterpene alcohols, α- and β-amyrins. Thus, the composition of the diethyl ether extracts of the propolis of the four species of SLBs that we are interested in is also highly specific, but as in the case of VOCs, there is a pairwise similarity of its group composition. This similarity is clearly demonstrated by the dendrogram in
Figure 3, built according to the data on the group composition of the components given in
Table 2 and
Table 3. The main constituents of samples 2 and 3 were diterpenoids (mainly diterpene acids), which were contained in propolis 1 and 4 only in small amounts. However, the latter were characterized by a high content of triterpenoids. In addition, phenolic lipids, alkylphenols, alkylresorcinols and alkylsalicylates were found only in them.
It is worth mentioning that a pairwise similarity (samples 1 and 4 vs. 2 and 3) was also observed in the composition of pot honey of the same four SLB species from the same region of Argentina and collected simultaneously with propolis samples 1–4 (our unpublished data). All this together testifies to the peculiarities of the preferences of different species of bees for collecting nectar and plant resins (as well as the coincidence of their collecting preferences). Indeed, the similarity of food preferences of
C. postica (1) and
T. fiebrigi (4) bees was shown previously on the basis of palynological analysis of honey prepared by them: plant nectar of the families Fabaceae, Arecaceae and Euphorbiaceae was the main food resource of these bees [
47,
48]. However, the source of resin for the manufacture of propolis may be plants from other families.
Since the composition of propolis depends on the plants that the bees visit to collect the resin and it contains chemical markers of these plants [
34,
35,
49,
50], it is possible to make assumptions about the botanical origin of the studied SLB propolis sample, based on previously obtained information about propolis from other regions and even other bee species.
The most likely plant precursor of propolis samples 2 and 3, the main components of which are diterpene acids, also called resin acids, are the resinous secretions of coniferous trees. The high content of these compounds in propolis is typical for the Mediterranean region [
51], whose flora includes various species of cypresses and pines. In the Neotropics of the Western Hemisphere, coniferous trees of the Pinaceae and Araucariaceae families grow, the resinous secretions of which contain diterpene acids, but their individual composition is different. Bankova et al. [
52,
53] found labdane-type diterpenoids in Brazilian propolis and concluded that their source is most likely local species of
Araucaria. However, it is impossible to exclude plants of the family Pinaceae from the list of possible sources of diterpenes: Marcucci et al. [
54] found dehydroabietic and abietic acids in honeybee propolis from beehives in a natural pine forest in the state of São Paulo (Brazil). Considering the fact that near the places where meliponines are localized in the state of Misiones (Argentina), from which samples 1 and 4 were taken, there are industrial slash pine (
Pinus elliottii) plantations, and we are inclined to think that these trees serve as a source of diterpenoids in them. This assumption is also supported by the qualitative composition of propolis terpenoids: the ether extracts contained all eight resin acids (abietic, dehydroabietic, isopimaric, levopimaric, neoabietic, palustral, pimaric and sandaracopimaric), and the VOCs contained all eight monoterpenes (α-pinene, β-pinene, camphene, myrcene, 3-carene, limonene, terpinolene and β-phellandrene) characteristic of resins of all pine species [
55].
Triterpenoids and lipids with a phenolic core, the main components of samples 1 and 4, have been found in propolis from tropical regions, both from families of honeybees [
56,
57] and some species of SLBs [
15,
21,
50]. The resins of
Mangifera indica, belonging to the Anacardiaceae family, have been named as a plant source of these lipids. Reliable evidence in favour of this is the presence in propolis of such triterpenes as mangiferolic, isomangiferolic and mangiferonic acids (
Table 3). It cannot be ruled out that other resiniferous representatives of the Anacardiaceae family serve as a source of phenolic lipids found in propolis from South America [
29].
Propolis samples from
S. postica bees from Barra do Corda, Maranhao State and Rio Grande do Norte State (northeast Brazil) have been reported to contain phenolic lipids, but their main components are flavonols, such as quercetin methyl esters and methoxychalcones [
21]. As their alleged precursor, the secretions on the tops of the shoots of
Mimosa tenuiflora are considered. However, these compounds were completely absent in Argentine propolis sample 1 from the same SLB species. A complete discrepancy in the composition of propolis collected by
T. fiebrigi bees in Brazil and Argentina was also observed: phenylpropenoids absent from Argentinian propolis were found in Brazilian propolis by Campos et al. [
9], but triterpenoids and phenolic lipids, the main components of sample 4 (
Table 3), were completely absent in Brazilian propolis. It is likely that SLBs do not show strong selectivity in resin collection and use different resources for the production of propolis, provided by local vegetation in the equatorial regions of Brazil and in the subtropics of Argentina.
A feature of the chemical composition of the studied propolis is the absence of flavonoids and phenolcarboxylic or cinnamic acids and their derivatives, which is attributed to the biological activity and medicinal properties of this bee product [
58,
59]. However, this does not mean that it is deprived of such activity and properties, and this was demonstrated previously [
11] using the example of propolis from two SLBs,
T. fiebrigi and
S. jujuyensis. Alcoholic extracts of both types of propolis, practically devoid of flavonoids (their content was 0.08%), nevertheless exhibited antimicrobial, antioxidant and antinociceptive activities. It is of interest to further study the medicinal properties of propolis with a ‘non-flavonoid’ composition. For this purpose, we determined the anticancer and antimicrobial activities of extracts from the studied propolis samples.
3.3. Anticancer Activity of Ether Extracts
To determine the effect of four propolis ether extracts on the viability of A375, C32, SCC-25, AGS and DLD-1 cells or fibroblasts, the cells were treated with extracts at concentrations of 1.5, 3.1, 6.2, 12.5, 25, 50, 100, 200, 400 and 800 μg mL
−1 for 24 or 48 h and their viability was assessed using MTT assay (
Figure S1). The purpose of the experiment was to establish the dependence of cell survival on the concentration of propolis extract and on the duration of exposure. Based on the dose–response curves, the IC
50 values were calculated (
Table 4).
At the 24 h time point, IC50 values of all propolis extracts in A375 cells were above 50 μg mL−1. After 48 h of treatment, the cytotoxic effect was stronger and IC50 values of extracts 1, 2, 3 and 4 were 29.4 ± 2.2, 16.8 ± 1.5, 43.9 ± 3.6 and 36.1 ± 2.4 μg mL−1, respectively. At the 24 h time point, C32 cells were more susceptible to the cytotoxic action of propolis extracts than were A375 cells. Prolongation of the incubation time to 48 h led to a decrease in the IC50 value of extract 2 to 17.3 ± 1.3 μg mL−1, whereas the viability of C32 cells treated with the other three extracts reduced slightly.
After 24 h of treatment, the viability of SCC-25 cells was highly suppressed by extracts 1, 2 and 4. The cytotoxic action of extract 3 was still high (IC50: 42.9 ± 3.5 μg mL−1) and similar to that observed in C32 cells. However, prolonged incubation of SCC-25 cells with extracts resulted in a decrease in IC50 values only by about 10–20%.
The viability of AGS cells after 24 h of incubation was most strongly inhibited by extract 2 (IC50: 38.2 ± 2.4 μg mL−1). For the other three extracts, IC50 values were slightly higher than 50 μg mL−1. At the 48 h time point, IC50 values of all propolis extracts decreased significantly and the lowest value (12.6 ± 0.9 μg mL−1) was found for extract 2.
Treatment of DLD-1 cells for 24 h with propolis extracts resulted in a relatively slight decrease in cell viability (IC50 > 50 μg mL−1). However, incubation of cells for 48 h significantly suppressed the proliferation of DLD-1 cells, and the IC50 values of extracts 1, 2, 3 and 4 were 31.9 ± 2.0, 20.2 ± 1.5, 32.9 ± 2.4 and 44.3 ± 3.6 μg mL−1, respectively. At the 24 h time point, the IC50 values of propolis extracts in normal skin fibroblasts were above 50 μg mL−1. At the 48 h time point, the IC50 value of extract 3 was still above 50 μg mL−1, while in the case of the remaining three extracts, it decreased to 34–43 μg mL−1.
Thus, the data from the performed experiments indicate that all tested propolis extracts inhibit the growth of cancer cells in both a dose- and a time-dependent manner. For all cancer cell lines tested, propolis extract 2 collected by T. clavipes showed the highest cytotoxicity.
It is noteworthy that propolis extract 3 is characterized by higher IC50 values than those of extract 2, which is close to it in composition: the main components of both are diterpene acids. At present, it is difficult to explain this discrepancy. However, it can be assumed that in the case of propolis extract 2, synergy with triterpenes is manifested, the relative total content of which in it is much higher: 12.54% versus 0.39% in extract 3. In particular, the content of α- and β-amyrins, triterpene alcohols with well-documented anticancer activity, in the former is much higher. However, this assumption about the presence of synergy needs to be tested.
3.4. Antibacterial Activity
To evaluate the antibacterial activity of propolis ether extracts, we determined the MIC, as well as the MBC and MFC. The test results are shown in
Table 5 along with our earlier data on MIC values for samples of European honeybee propolis, mainly ‘poplar’ and mixed ‘birch’ and ‘aspen’ types. As can be seen from the data presented, all tested extracts inhibited the growth of test cultures, although to varying degrees. Gram-negative bacteria were found to be the least sensitive, consistent with numerous previously reported data.
The list of test cultures includes pathogens for both honeybees and humans. It is well known that the most destructive brood disease of honeybees called American foulbrood (AFB), caused by the Gram-positive spore-forming bacterium
Paenibacillus larvae, is a serious problem for global beekeeping [
61]. Therefore, great efforts are being made to discover natural remedies that can replace the currently banned antibiotics and other synthetic drugs that have been widely used for a long time to combat this infection. These natural remedies include propolis, whose anti-AFB activity has been associated with phenols, such as flavonoids, phenolcarboxylic and hydroxycinnamic acids and their esters [
35,
60,
62], as well as triterpenoids [
63].
Comparison of the anti-AFB activity of propolis from SLBs that do not contain these components with the previously obtained MIC values [
60] shows that it is at a level typical for propolis of the ‘birch’ type (15.6–31.8 μg mL
−1) in the case of extracts 1 and 4 containing phenolic lipids and triterpenoids. Extracts 2 and 3, with a high content of diterpenoids, are characterized by MIC values approximately two times higher than those of the least potent ‘aspen’ type European propolis (62.4 μg mL
−1).
Ether extracts of Argentine propolis (1–4) also showed high activity against the tested human pathogens: the Gram-positive bacterium
S. aureus and the Gram-negative bacteria
E. coli and
P. aeruginosa, as well as the fungus
C. albicans. Particularly sensitive to the action of extracts 2 and 3 rich in diterpenoids were the bacteria
B. cereus and
B. subtilis, which can cause food poisoning in humans. Interestingly, the MIC values given in
Table 5 for the propolis extract 4 of
T. fiebrigi in relation to
S. aureus,
P. aeruginosa and
C. albicans turned out to be approximately an order of magnitude lower than in the case of Brazilian propolis of the same species of bees [
9], which did not contain phenolic lipids and triterpenoids (
Section 3.1).
3.5. Anti-Biofilm Action
The high activity of Argentinian propolis from SLBs against certain microorganisms may be due to various reasons, one of which may be the anti-biofilm-forming effect of its extracts [
11,
64]. In this study, the effect of four propolis extracts on the ability of bacterial cells to form biofilm was determined using the crystal violet staining method. Three Gram-positive strains (
P. larvae,
B. cereus and
B. subtilis) and one Gram-negative strain (
E. coli) were used for the experiment. The effect of propolis extracts at concentrations below the determined MIC values (
Table 5) on biofilm formation is shown in
Figure S2. All tested extracts in the range of concentrations from 1/32 MIC to ½ MIC statistically significantly (
p < 0.05) reduced the biofilm biomass of both Gram-positive and Gram-negative strains relative to control cells. Furthermore, extracts at the concentrations determined did not inhibit bacterial growth statistically significantly (data not shown), but sub-MIC concentrations reduced the ability of these strains to form biofilm. Thus, the action of extracts from SLBs as anti-biofilm agents was not associated with bacterial growth inhibition.
S. jujuyensis propolis also reduced the formation of
Staphylococcus aureus and
P. aeruginosa biofilm in a previous study, and this, according to the authors [
11], was also not associated with the inhibition of bacterial growth.
Based on the obtained data, minimum biofilm inhibitory concentration (MBIC
50) values were determined (
Table 6). The MBIC
50 values of extracts varied between 0.035 and 311.466 µg mL
−1. The results showed that all extracts inhibit biofilm formation but to varying degrees. The strongest activity for all extracts was observed against
B. cereus and
B. subtilis strains. In contrast, they were much less effective against
P. larvae and
E. coli strains.
In general, the lowest MBIC50 values (below 0.06 µg mL−1) against B. cereus and B. subtilis strains were observed for extracts 2 and 3. Sufficiently strong anti-biofilm activity against the same species was also observed in extracts 1 and 4. The higher MBIC50 values were obtained for the Gram-negative E. coli strain, but the activity of the extracts was arranged in the same order as for the Gram-positive bacteria B. cereus and B. subtilis: extracts 3 and 2 significantly more strongly inhibited the formation of biofilm in E. coli compared to extracts 1 and 4. On the contrary, extracts 1 and 4 showed stronger inhibition of biofilm formation in P. larvae.
Thus, the high anti-biofilm activity of all four types of SLB propolis was demonstrated. The Gram-negative bacterium
E. coli was the least sensitive, which is consistent with the literature data [
64]; however, even against that, the effect of extracts was manifested at concentrations less than 1 mg mL
−1. Also noteworthy is the fact that in this case, there is a pairwise similarity of the tested propolis samples, which is undoubtedly related to their chemical composition.