Three species of terrestrial snails (A. fulica, C. aspersum, and C. nemoralis) and three terrestrial slugs (A. ater, A. hortensis, and L. flavus) were included in this study. Apart from the snail A. fulica, which is native to Africa, all other species are found in Europe. A. hortensis, A. ater, C. nemoralis, C. aspersum, and L. flavus were collected in Hillsborough Park, Hillsborough, Sheffield (53.4080° N, 1.5015° W). A. fulica snails were purchased as juveniles and reared in house. All species were kept in plastic containers (39 × 48 × 20 cm) with ∼6 cm of vermiculite layering the bottom of the box at 22°C ± 1°C and high humidity. Animals were fed ad libitum twice weekly with cucumber, lettuce, and sweet potatoes.

For each species, three specimens were removed from the containment area, cleaned using type II water, and placed in an empty and clean plastic container. Animals were then allowed to move freely across a clean sheet of glass for 5 min to avoid collecting adhesive mucus. After that, locomotive mucus was collected from the glass surface using two razor blades (cleaned using ethanol and then type II water), kept in a 2.0-ml polypropylene-graduated centrifuge tube with a cap, stored at room temperature, and subjected to analysis on the day of collection.

TGA tests were conducted using an MX-50 (A&D Instruments, United Kingdom) moisture content analyser. An alumina crucible (9.5 mm diameter and 14 mm high, Almath Crucibles Ltd., United Kingdom) with 1 ml of fresh native mucus collected as described previously was used for all experiments. The heating rate was 1°C/min, from 25°C to 120°C, with a data interval of 15 s. All experiments were repeated three times per species, with separate samples collected for each test from the same group of individuals at the same time.

With concentrations obtained from TGA analysis, dilutions were prepared for all species, and samples were set to the same concentration of ∼1 mg/ml. Type II water was used as a solvent, and native mucus was collected as described in Materials and Methods. Then, all samples were analysed at room temperature (22°C ± 1°C) using a UV-300 Spectronic Unicam spectrometer (Thermo, United Kingdom) in 1 cm path-length polystyrene cuvettes. Type II water was used as a blank, and mucus samples were manipulated using a 1,000-μL micro-pipette. Scans were performed from 200 to 500 nm at a scan speed of 240 nm/min and 300 steps in total.

We deliberately chose to use offset absorbance wavelength values to determine protein concentration in order to analyse the peak maximum in the UV range because A. fulica mucus shows a shift to lower wavenumbers of its protein absorption peaks, most likely due to differences in phenylalanine (Phe) chromophores. Therefore, concentrations can be estimated for each sample, using maximum absorbance at 215, 225–230, and 260 nm, as follows (Walker, 2002; Liu et al., 2009): C o n c e n t r a t i o n   m g / m l = 0.183   ×   A 230 n m   –   0.075   ×   A 260 n m , (1) C o n c e n t r a t i o n   µ g / m l = 144   ×   A 215 n m   –   A 225 n m . (2)

The use of Eqs 1, 2 depends on the presence of peaks at 215 nm with absorbance values <2.0 (Walker, 2002). To corroborate the use of Eqs 1, 2 for each species based on maximum absorption peaks, deconvolution of UV-vis spectra and curve fitting were performed between 200 and 350 nm, and a 10-point multipeak Gaussian fitting (every 10 nm) was performed using Origin software, v2020 (Supplementary Figure S1). Eq. 1 was used for C. aspersum, C. nemoralis, A. ater, A. hortensis, and L. flavus (snails and slugs native from the United Kingdom), whereas Eq. 2 was used for A. fulica.

Gastropod locomotive mucus was collected as described in Materials and Methods. Samples were weighed, placed in 2-ml polypropylene-graduated centrifuge tubes, and diluted with type I water to 5.92 mg/ml, based on the lowest dry weight measured previously by TGA. Aliquots of 20 µL were taken and mixed with an equal volume of a solution containing sodium dodecyl sulphate and β-mercaptoethanol, according to the method of Laemmli (Walker, 2002). Aliquots of 20 µL of the resulting solutions were resolved on 4%–20% Tris-glycine gel under reducing conditions, using a mini gel tank Invitrogen (Thermo Fisher Scientific, United States) and power supply PS 250 (Hybaid Ltd., United Kingdom), over 100 min at 120 V. Protein bands were visualised by staining with 0.25% Coomassie Brilliant Blue R-250 (VWR Chemicals, United States) and imaged using a Scanjet G2710 scanner (HP Inc., United States). A HiMark Pre-Stained Protein Standard (Thermo Fisher Scientific, United States) was used as a ladder.

Rheological measurements (frequency sweep and temperature ramp tests) were performed using an AR-2000 (TA Instruments, Delaware, United States) rheometer equipped with a Peltier temperature-controlled bottom plate and cone-plate upper geometry (diameter 40 mm, cone angle 2°, and 55 μm truncation) or a cross-hatched parallel plates for validation of non-slip conditions (20 mm diameter, 0.7 mm gap). Native mucus, sufficient to completely fill the geometry gap, was placed onto the bottom plate, and the geometry was lowered to the truncation gap at the slowest speed to avoid any undue shearing of the sample. An isolated environmental chamber using a water trap and moistened tissue was used to avoid the native mucus drying out and minimise temperature variations during the tests. Oscillatory frequency sweep tests were performed from 0.6 to 62.8 rad/s (0.1–10 Hz) at 10% strain at 25°C (chosen to be within the linear viscoelastic region of the materials; Supplementary Figure S2). Single-frequency oscillatory tests were then conducted at 0.63 rad/s 10% strain between 15°C and 80°C using a ramp temperature of 5°C/min.

Statistical analysis was performed using Student’s t-test and p < 0.05 significance level (Origin software, v2020, United States).

TGA, UV-vis, and SDS-PAGE were performed to determine the compositional properties of gastropod locomotive mucus. In addition, total solids and protein concentrations were determined using TGA and UV-vis, respectively, to gain an initial and broad appreciation of composition (Table 1).

Our findings agree with previous studies where pedal mucus in terrestrial snails and slugs has been shown to consist of approximately 91%–98% water, with the remaining mass corresponding to inorganic material and high-molecular-weight organic compounds, mucus polymers, such as proteins and carbohydrates (Denny, 1973, 1984; Hausdorf, 2001; Smith, 2002). For all species tested here, water composition in gastropod mucus lay between 97.4% ± 0.013% in the slug A. hortensis and 99.4% ± 0.005% in the snail A. fulica. The snails C. aspersum and C. nemoralis have the closest percentages of water, 98.62% ± 0.043% and 98.79% ± 0.014, respectively, and notably, they are from the same family Helicidae. Simply inverting the TGA data from % water to solids concentration, the lowest total solids value corresponds to the African snail (A. fulica) with 5.97 mg/ml, whereas the highest value to the slug A. hortensis is 25.15 mg/ml.

UV-vis was used to determine the total protein concentration to determine the constituents of the total solids proportion of mucus. From Figure 1A, concerning the protein-to-total solids ratio, it is apparent that proteins only account for <5% of the total solids present. The remaining solids are assumed to be primarily carbohydrates, as previous studies have found <1% of mucus dry mass to be inorganic (Liudmyla et al., 2022).

Comparing species, if we first investigate the three slugs, we immediately identify a lower solids-to-protein ratio in the slug L. flavus from the Limacidae family, compared to the other two slugs, A. ater and A. hortensis, both of the Arionidae family. In terms of the three snails, a similar pattern is observed, which overall suggests that members of the same family show similar solid-to-protein ratios. With a value of 0.007, the African snail A. fulica has the lowest ratio, whereas the highest one is for the slug A. hortensis, with 0.042, which represents a significant difference (p = 0.0030) and one we attribute to the amount of aforementioned carbohydrates present in the mucus samples.

Upon closer inspection of UV-vis spectra, it is possible to observe some finer details. Figure 1B depicts a consistent main band at 225–230 nm associated with the presence of proteins for all species. However, there appears to be a shift in this band in the snail A. fulica to 215 nm (assigned to π   → π * bonds), which is thought to represent a proportional increase in the aromatic amino acids phenylalanine (Phe), tryptophan (Trp), or tyrosine (Tyr). This species may have a markedly different amino acid composition compared to the other snails and slugs (Crammer and Neuberger, 1943; Schmid, 2001; Lesniak et al., 2013; Antosiewicz and Shugar, 2016; Radicioni et al., 2016; Bansil and Turner, 2018; Petrou and Crouzier, 2018; Butnarasu et al., 2019). Furthermore, in the slug A. ater, there is a shoulder in the band at approximately 260 nm, which is thought to represent an increase in the essential amino acid phenylalanine in the mucus (Moran et al., 1983). This is unusual given the relative metabolic expense of excreting this amino acid, leading to the assumption that its presence has endowed a specific selective advantage for the species (Moran et al., 1983; Pakay et al., 2002; Yoo et al., 2013; Watz and Nyqvist, 2022).

We analysed our samples using SDS-PAGE to further compare these species and determine the molecular weight of proteins in their locomotive mucus (Figure 2).

Overall, most bands observed are unreported to date, and there appears to be little consistency across species. However, in general, proteins with molecular weights ranging from 41 to 377 kDa were seen and could be separated into 2–9 main bands (Table 2). Where assignments of protein bands were possible based on previous research, they are named within the table (Ulagesan and Kim, 2018).

Lower-molecular-weight proteins identified include haemocyanin at 78 kDa, normally present in the haemolymph (Huang et al., 2008) and reported for C. aspersum, which well ties our findings (Ballard et al., 2021). Achacin at 96 kDa, which is the glycoprotein associated with the antimicrobial activity of the African snail mucus and only seen in our A. fulica sample (Ogawa et al., 1999; Cilia and Fratini, 2018), and the metalloproteinase-like ADAM family of proteins between 66–67 and 52–51 kDa, which play a key role in protein degradation (Edwards et al., 2009; Ballard et al., 2021), which were evident in A. ater and C. aspersum. Interestingly, collagen (∼250 kDa) was observed for C. aspersum and C. nemoralis and is most likely to be variants II, III, and XI, which are mostly related to locomotive mucus. However, it could be collagens IV, IX, and X, which have been found in the locomotive, adhesive, and defensive mucus (Sripriya and Kumar, 2015; Ballard et al., 2021; Tachapuripunya et al., 2021; Cerullo et al., 2022).

Finally, a ∼350-kDa high-molecular-weight protein band was present in all our samples. Previously, this has been attributed to lectins (proteins that bind to carbohydrates) (Ito et al., 2011) but could readily be an unidentified mucus protein (McDermott et al., 2021). We believe that this protein could be a primary contributor to the rheological properties of mucus, given its length and possible carbohydrate-binding interactions (Fung, Gallego Lazo, and Smith, 2019).

To relate mucus composition to function (i.e., flow properties and stability), we conducted a range of rheological tests. First, a rheological “fingerprint” was captured for each species where linear viscoelastic moduli were measured across a range of timescales of deformation (obtaining the storage modulus G′ and loss modulus G′′, Figure 3) (detailed information is included in Supplementary Figure S3). Before venturing into a discussion of these results, it is important to note that we have accounted for experimental artefacts that can arise in rheological measurements when undertaking analysis on low-viscosity samples as a result of instrument and sample inertia and low-torque limits (Ewoldt et al., 2015) (Supplementary Figure S4 and Figure 3).

Our analysis indicates that all mucus samples behave as viscoelastic gels over the range of angular frequencies tested, with G′ > G′′ (Peters et al., 2021), a finding similar to mammalian mucus (Innes et al., 2009; Schuster et al., 2013; Murgia et al., 2016; Huck et al., 2019; Peters et al., 2021). The results from the previous two studies analysing mucus from slugs L. maximus and A. columbianus are included in Figure 3 for further reference (Denny and Gosline, 1980; Ewoldt et al., 2007). Both pioneering works reported modulus values within the overall range we report for our species, although these studies collected and tested their mucus samples under conditions different from ours (see Methods). For example, Ewoldt et al. (2007) tested slug mucus using a 2 cm plate with sandpaper, 200 µm gap, and at 22°C. Denny and Gosline (1980) collected slug mucus using a glass rod, and tests were performed at temperatures between 21°C and 24°C. In summary, we note that the overall spread of rheological properties reported for locomotive mucus to date is considerable, with a two-order-of-magnitude variation in modulus, but this is not unsurprising given the infancy of this field (Holland et al., 2006).

It is also possible to extract a complex viscosity (resistance to flow) from these fingerprinting tests (Figure 4). Akin to many polymeric materials, gastropod mucus clearly exhibits non-Newtonian shear thinning fluid properties (Perez-vilar and Hill, 1999), with viscosity dependent on the deformation rate. This was also the case when synthetic mucus was tested and used as a viscoelastic biofluid (Aghakhani et al., 2022). Our results show that A. fulica exhibits the lowest overall complex viscosity, whereas A. hortensis represents the highest. To corroborate if slippage occurred in our measurements due to the use of a smooth cone plate geometry, we tested C. aspersum locomotive mucus using a 20 mm cross-hatched geometry (Ewoldt et al., 2015). Results indicated values in similar ranges when comparing cone-plate geometry measurements with those performed under a rough surface (Supplementary Figure S5).

Figure 5 attempts to establish whether this rheological diversity is related to the total solids or protein content in gastropod mucus. There is a positive relationship between solid concentration and viscosity (Figure 5A). However, the same is seen for protein concentration (Figure 5B), indicating that proteins are most likely linked to mucus rheology.

If previous assumptions and observations on different kinds of mucus are valid, and mucus polymers and their protein constituents (e.g., lectins and their carbohydrate-binding ability) are the main factors governing the rheological behaviour of mucus (Bansil and Turner, 2006; Hill et al., 2014), then an examination of viscosity as a function of temperature could add valuable information. The hypothesis is that thermal transitions, such as protein glass transition and, ultimately, denaturation, disrupt overall gel strength and subsequent flow properties. Figure 6 shows the average complex viscosity versus temperature. As a precaution, Lauger et al. (2002) described the importance of obtaining reliable rheological data when performing tests as a function of temperature. Therefore, we complemented our temperature ramp tests with experiments performed at a lower heating rate, using a cross-hatched geometry and the same environmentally isolated chamber to account for any temperature gradients that may arise from our tests. Reassuringly, our findings showed no major differences compared with the data in Figure 6 (Supplementary Figure S6).

Figure 6 shows that mucus across all species retains up to 80% of its original viscosity from 15°C to 30°C presumably to maintain consistency in flow behaviour during the natural habitat temperature ranges these animals encounter. Between 45°C and 48°C, a 50% reduction in viscosity is observed for A. hortensis, A. ater, C. aspersum, and L. flavus. A. fulica is an exception; compared to the other European species, it continues a steady decline in viscosity until 60°C before quickly dropping. Once viscosity has fallen for all species, potentially as a result of the change in protein conformation, it appears to plateau before beginning to rise again slightly. This lower plateau represents the minimum level of viscosity and, thus, minimal mucus polymer molecular interactions in the mucus. We attribute the subsequent rise to one of two scenarios: the unlikely possibility that the samples are slightly drying out and forming a skin at the edges of the gap, or there may be protein structures undergoing denaturation, leading to dehydration and aggregation, resulting in the reformation of a newer, phase-separated, and stronger gel structure, consistent with recent FTIR observations (Barajas-Ledesma and Holland, 2023).

There is a clear difference between A. fulica and the other five species. Mucus from this snail exhibits a higher stability in viscosity when the temperature increases, and the drastic viscosity drop occurs at 15°C higher than the other samples. Here, it is also important to remember that Figure 6 depicts relative changes based on the original viscosity of the mucus, and A. fulica already has the lowest viscosity tested (Figure 4). Hence, this variation could result from A. fulica having the highest water content of all mucus tested and the lowest ratio of proteins to total solids. In this instance, the other mucus polymer components, such as carbohydrates, could contribute more to the locomotive mucus viscosity of A. fulica and hence the different thermal response or the smaller amount of proteins present could have a higher thermal tolerance by comparison (Towns, 1995; Petrou and Crouzier, 2018).

This can be rationalised through an evolutionary context, as whilst terrestrial snails and slugs stem from the same common ancestor in the Stylommatophora clade (Wade et al., 2006), it is highly likely that specific adaptations of mucus to certain habitats have arisen. This is evidenced in a recent FTIR study of mucus biodiversity, showing that species identification can be made based on the unique chemical fingerprint of locomotive mucus alone (Barajas-Ledesma and Holland, 2023). In our results, A. fulica has a more stable mucus under thermal stress, implying its mucus polymers maintain their function over a wider range of temperatures, which is consistent with the observation that this African species is native to a region with higher average temperature compared to the other five European species included in this study (Albuquerque et al., 2008). Furthermore, FTIR spectra of A. fulica indicates that its Amide III region differs from the other species tested, suggesting the specialisation of the proteins present in the mucus (Barajas-Ledesma and Holland, 2023).