Propolis and insects Propolis

Raw propolis provided by private beekeeper Dr. Oliver Schwarz (Stuttgart, Germany) (Figure 1A) was harvested and homogenised as described in [1]. For homogenisation, propolis chunks were mixed, frozen finely ground and subsequently stored at −20 °C (Figure 1B). The pulverizing procedure was based on a method that is used to produce propolis extract [21]. To prevent contamination, propolis was only handled wearing gloves cleaned with ethanol (Rotipuran®, ≥99.8%, p.a., Carl Roth GmbH & Co. KG, Karlsruhe, Germany).

Figure 1: Bee propolis. (A) Raw propolis as collected from the hive. (B) Homogenised propolis powder. (C) Cone-shaped propolis sample used for adhesion tests. Scale bar: 1 cm (A, B), 500 µm (C). Figure 1 was reproduced from [1] (© 2021 Saccardi, Schiebl, Weber, Schwarz, Gorb and Kovalev, published by Frontiers, distributed under the terms of the Creative Commons Attribution 4.0 International License, https://creativecommons.org/licenses/by/4.0).

Insects

Adult worker bees (Apis mellifera) were collected in gardens in Kiel (Germany) in July 2019 and immediately used for experiments. They are hereinafter referred to as „nectar collectors“. In October 2018, bees were taken directly from the hive (Stuttgart, Germany). They are further named „hive bees“. They were kept in a cage and provided with water and honey until they were prepared for experiments. The remaining bees were frozen and stored at −20 °C for three months to be used in further experiments.

Honeybees (Apis mellifera), returning to the hive with resin attached to their pollen baskets, were caught and subsequently frozen in September and October 2018. These resin-collecting bees will be henceforth referred to as “propolis bees”.

Imaging and structural studies

Bee mandibles were prepared and subsequently examined with binoculars, a scanning electron microscope (SEM), and a confocal 3D laser scanning microscope in order to identify anatomy and surface structure.

Anatomy of the honeybee mandible

Mandibles of all collected bees were prepared under binoculars by carefully separating them from the insect’s head with a scalpel (Figure 2). General morphology, structures, and contamination of every prepared mandible were studied with a binocular microscope (Leica M205 A, Leica Microsystems GmbH, Wetzlar, Germany) equipped with a camera (Leica DFC420) prior to further experiments. For the following anatomical studies, mandibles of nectar collector and hive bees were then washed with acetone and water. They were air-dried, mounted on holders (inside up), sputter-coated with a 10 nm thick layer of gold–palladium and studied using a SEM (Hitachi S-4800, Hitachi High-Technologies Corp., Tokyo, Japan) at 3 kV accelerating voltage. Images of the spoon-shaped mandible tip were taken systematically and later assembled into one high resolution image. Higher magnified pictures were taken in characteristic areas of the mandible surface. For one additional experiment, instead of air-drying, four washed mandibles were dried using a critical-point-drier (Leica EM CPD300) and subsequently studied in the SEM.

Figure 2: Head and mandible of a worker bee. (A) Frontal view of a honeybee head with an arrow pointing to the outside of the left mandible. Scale bar: 1 mm. (B) Medial surface of the left mandible of a worker bee. Proximal and distal ends are marked. Scale bar: 200 µm.

Surface structures on bee mandibles

Surface structures on mandibles were studied in the SEM as described above. Additionally, the medial surface profile of the mandible was studied on fresh, chloroform-washed specimens fixed to a glass slide using a drop of a melted mixture of 50% paraffin wax and 50% colophony. The mandibles were examined with a confocal 3D laser scanning microscope (Keyence VK-X250; Keyence Corporation, Osaka, Japan). The MultiFileAnalyser software (Version 1.2.6.106, Keyence Corporation, Osaka, Japan) was used to measure profiles and the structural dimensions in different areas of the mandible.

Surface structures on propolis bee mandibles

Freshly defrosted and prepared mandibles of propolis bees were studied in the cryo-SEM in the frozen state so the resin contaminations did not dry out. They were placed on holders (medial surface up), rapidly frozen on the table in the preparation chamber at −140 °C, sputter coated with gold palladium (3 nm), and studied at −120 °C with a cryoSEM (Hitachi S-4800, Hitachi High-Technologies Corp., Tokyo, Japan) equipped with a cryopreparation system (Gatan ALTO 2500, Gatan, Inc., Abingdon, UK) at 3 kV accelerating voltage.

Image processing

SEM images were processed using Gimp, version 2.10.14. All adjustments were applied to the whole image. Color levels, contrast, and brightness were adjusted, and scale bars and other labels were added. Profiles measured with the confocal 3D laser scanning microscope were digitized using WebPlotDigitizer (version 4.2, https://automeris.io/WebPlotDigitizer, Ankit Rohatgi, San Francisco, USA).

Investigation of propolis adhesion on mandibles Insect preparation for adhesion tests

After insects for experiments were caught, they were placed and stored in the freezer at −20 °C for a minimum of 15 min and up to many months. The mandibles were prepared as described above (Figure 2). Without further treatment a mandible was then glued, inside facing up, to a glass slide using a drop of a melted mixture of 50% paraffin wax and 50% colophony to avoid unwanted movement during adhesion experiments. The spoon-shaped, distal part of the mandible was oriented using a binocular microscope so that the flat area near the sharp edge of the mandible was parallel to the glass surface. Adhesion experiments on mandibles were carried out immediately after fixing the mandible to the slide to avoid material desiccation.

Test method

Adhesion experiments were performed in a manner similar to that in [1]. Just before each adhesion experiment, a small amount of homogenised propolis powder was defrosted and kneaded into a homogeneous mass. Cone-shaped propolis samples with a spherical tip were subsequently formed by hand wearing ethanol-cleaned gloves (Figure 1C). The topography of the sample was analysed using a 3D optical profilometer (Keyence VR 3100; Keyence Corporation, Osaka, Japan). The profile of the sample was measured along five lines arranged in a star shape going through the highest point of the tip. To estimate the radius at the sample tip, a circle was fitted to the sample profiles in the five orientations (Figure 3D). The radii of the circles were measured and then averaged.

Figure 3: Adhesion experiments. (A) Experimental set-up for adhesion testing with the Basalt-01 mechanical tester (Tetra GmbH). (B) Propolis contact in the presence of fluid (mineral oil) and without fluid. (C) Schematic of a bee mandible. Adhesion measurements were conducted along the arrow, the circle represents the size of the contact area. Scale bar: 500 µm. (D) 3D profile of a propolis sample. The inscribed circle was used to estimate the tip radius. The small subimage depicts the sample topography with darker areas being higher than lighter areas. (E) Typical force–distance curve obtained from adhesion experiments. BM, bee mandible (optional); FL, fluid; FOS, fibre-optic sensor; GC, glass capillary; GS, glass slide or other substrate material; MM, micro-manipulators; MR, mirror; MS, metal spring; PS, propolis sample. Figure 3A was adapted and Figure 3B, D, and E were reproduced from [1] (© 2021 Saccardi, Schiebl, Weber, Schwarz, Gorb and Kovalev, published by Frontiers, distributed under the terms of the Creative Commons Attribution 4.0 International License, https://creativecommons.org/licenses/by/4.0).

The effective elastic modulus and pull-off force of propolis were measured with a microforce measurement device (Basalt-01; Tetra GmbH, Ilmenau, Germany) [22-24]. The device mainly consists of micromanipulators as a platform holding the substrate material, a metal spring (springs with spring constants of 618 N/m and 539 N/m were used) and a fibre-optic sensor (Figure 3A). The piezo drive moves the spring down to load and up to unload the sample. A shortened glass capillary (5 µL micropipet Blaubrand® IntraEND, Brand GmbH & Co. KG, Wertheim, Germany) was attached to the metal spring with cyanoacrylate glue (5925 Ergo® Elastomer–Kisling GmbH, Bad Mergentheim, Germany). The freshly formed, cone-shaped sample of propolis was then mounted on the tip of the capillary without any additional glue. The propolis sample was brought into contact with the substrate and retracted from the surface as soon as the load force reached 5 mN. The load was chosen to resemble the force applied by bees when handling propolis. As no studies exist on mandibular forces and pressures of honeybees, pressures measured at the tip of mandibles of predacious coleoptera [25] where used as a reference point. Tip pressures were calculated as suggested by [25]:

(1)

where Fa is the applied force and A is the contact area obtained from the contact radius.

With each propolis sample, a set of ten single measurements was performed, each on a different spot on the spoon-shaped tip of the bee mandible, starting at the sharp edge and moving towards the hairy edge (Figure 3C). Another measurement with an extended contact time of 60 s was performed on the flat area of the bee mandible. Afterwards, six reference measurements were carried out on a smooth, clean glass surface (standard microscopy slides (soda lime glass); Carl Roth GmbH & Co. KG, Karlsruhe, Germany) with the same propolis sample.

Experiments were carried out at room temperature (24.00 ± 0.53 °C) and a relative humidity of 36.80% ± 9.0%. For every condition, 50–150 single tests were performed on different spots of the mandible (N = 5–15 propolis samples and mandibles, n = 10 measurements on mandible per sample). After the adhesion experiments, the substrate material was examined under a binocular microscope (Leica M205 A) in order to find possible propolis residues/prints in the contact area. Adhesion tests on fresh mandibles were performed in July and November. As a reference, propolis adhesion on bee mandibles was compared to propolis adhesion measured on glass under dry and fluid conditions [1]. For the reference experiments in fluid conditions, a drop of oil (mineral oil, light, Sigma-Aldrich, St. Louis, USA) was placed on the glass surface as described in [1] and shown in Figure 3B. Oil was chosen because we assumed that the mandibles surface might be oily.

Evaluation of a potential coating of the mandibles

As previously suggested, a fluid layer covering the mandible surface could be a strategy to reduce adhesion [16]. Therefore, fresh, untreated bee mandibles were studied in the frozen state, as cryo-SEM has been reported to be a successful method for visualizing biological fluids such as lipids and water-based solutions [26]. Cryo-SEM was performed as described in section “Surface structures on propolis bee mandibles”.

Additionally, visualization of the mandible cuticle and any additional surface layers was performed using cryo-SEM on fresh fractures of untreated bee mandibles. Mandibles were tightly clamped into a metal holder and frozen in the preparation chamber (−140 °C). The samples were then fractured within the preparation chamber by cutting off a part of the mandible tip using a cold scalpel blade mounted on a user-controlled handle. Fractured samples were then sputter coated and examined in a frozen state as described above.

Where possible, the contact angle of the substance coating the mandible surface was measured on SEM images of fresh fractures of untreated bee mandibles using Gimp, version 2.10.14. For this purpose, two measurement lines were drawn: one along the surface of the cuticle and one along the surface of the substance on top of the cuticle. The angle was then measured between the two lines.

Furthermore, different solvents were used to wash fresh mandibles in order to try removing any potential surface coating. Mandibles were washed in an ultrasonic bath with distilled water for 10 min or with either chloroform (Rotisolv® HPLC stabilised with 1% ethanol, Carl Roth GmbH & Co. KG), or acetone (Rotipuran® ≥99.8%, p. a. ACS ISO, Carl Roth GmbH & Co. KG) for 5 min followed by 5 min washing with distilled water. Distilled water was used as an agent for removing hydrophilic substances from the surface of bee mandibles. Chloroform and acetone are known to remove hydrophobic substances and the wax layer from mandibles [27]. The ultrasonic bath facilitates the dissolving procedure and particle removal from the surface.

To visualise the effect of washing, dry untreated mandibles and mandibles treated with the different solvents were examined using a SEM as described before (section “Anatomy of the honeybee mandible”). Mandibles washed with chloroform and water were studied in the cryo-SEM as described above for untreated mandibles.

Adhesion on washed bee mandibles

To test the effect of potential surface coatings on propolis, adhesion experiments with propolis were performed on mandibles washed with the different methods as described for untreated bee mandibles (section “Investigation of propolis adhesion on mandibles”).

Adhesion on resin replicas of bee mandibles

In order to test how shape and surface structures of the mandibles affect adhesion, replicas of bee mandibles were made using a two-step moulding method [26,28]. The method allows one to replicate the surface structure with nanometre precision. Replication substitutes chemical complex and heterogeneous biological surfaces with a well-studied epoxy surface. In this way the effect of surface chemistry on adhesion was excluded. First, washed mandibles fixed to glass slides, as described above (section “Investigation of propolis adhesion on mandibles”), were covered with a two-component dental wax (Affinis light body, ISO 4823, polyvinylsiloxane, Coltène Whaledent AG, Altstätten, Switzerland). After the wax was cured, the mandibles were removed from the mould, the cavity was filled with Spurr epoxy resin (Spurr’s low viscosity kit; Plano, Wetzlar, Germany, composition: vinyl cyclohexene dioxide 10 g, diglycidyl ether of polypropylene glycol 6 g, nonenyl succinic anhydride 26 g, and S-1 dimethylaminoethanol 0.4 g [29]) and covered with a smooth sheet of dental wax to create a level surface for the mandible replica to stand on. The resin mandibles were cured at 70 °C for 48 h, demoulded and fixed to a glass slide using double-sided adhesive tape. A binocular microscope (Leica M205 A) was used to visually check whether the features of the real mandibles, especially the surface structures, were successfully replicated in the artificial mandibles.

Adhesion experiments with propolis were carried out as described for real bee mandibles (section “Investigation of propolis adhesion on mandibles”). For mandible replicas, reference measurements were performed on the smooth resin substrate. 50–60 single tests were performed on different spots of the mandible replica (N = 5–6 propolis samples and mandibles, n = 10 measurements on mandible per sample).

Data analysis and statistics

Adhesion experiments were evaluated as described in [1] using Matlab, version R2015b. The unloading part of force–distance curves (Figure 3E) acquired from adhesion experiments was fitted according to the JKR theory [30]:

(2)

where a is the contact radius, Fa is the applied load, R is the tip radius, and E and ∆γ are the effective elastic modulus and the work of adhesion, respectively. The work of adhesion ∆γ is the energy per unit of area needed to separate two bodies in contact. It was chosen as a measure of adhesion because it is independent of the contact area. To characterize viscoelastic properties of propolis, a generalized Maxwell model was used [31]. The viscosity of the sample was estimated from experimental force curves using the following equation [32,33]:

(3)

where d is the displacement, t is the time under load, E∞/E1/E2 and η1/η2 are the Young’s moduli and viscosities of the static and the two dynamic components, and ν is the Poisson ratio assumed to be equal to 0.49 [33].

The data were statistically analysed using the software R, version 3.6.1. Data was tested for normal distribution and variance homogeneity using Kolmogorov–Smirnov and Levene's tests, respectively. The comparison of propolis adhesion under different conditions and on different substrates was performed with a one-way ANOVA and a pairwise multiple comparison procedure (Tukey test). An unpaired two-sample t-test was performed to compare the mean Young’s modulus of propolis at 24 and 26 °C. Correlation analysis of Young’s modulus and work of adhesion obtained from adhesion experiments was performed by calculating the Pearson correlation coefficient.