Microtopography of the eye surface of the crab Carcinus maenas: an atomic force microscope study suggesting a possible antifouling potential

Abstract

Marine biofouling causes problems for technologies based on the sea, including ships, power plants and marine sensors. Several antifouling techniques have been applied to marine sensors, but most of these methodologies are environmentally unfriendly or ineffective. Bioinspiration, seeking guidance from natural solutions, is a promising approach to antifouling. Here, the eye of the green crab Carcinus maenas was regarded as a marine sensor model and its surface characterized by means of atomic force microscopy. Engineered surface micro- and nanotopography is a new mechanism found to limit biofouling, promising an effective solution with much reduced environmental impact. Besides giving a new insight into the morphology of C. maenas eye and its characterization, our study indicates that the eye surface probably has antifouling/fouling-release potential. Furthermore, the topographical features of the surface may influence the wettability properties of the structure and its interaction with organic molecules. Results indicate that the eye surface micro- and nanotopography may lead to bioinspired solutions to antifouling protection.

1. Introduction

Fouling is an undesirable process in which a surface becomes encrusted with material from the surrounding environment. In the case of biofouling, that material consists of organisms (micro- and macro-organisms) and their by-products (e.g. extracellular polysaccharides and metabolites) [1]. Biofouling poses a serious problem for living organisms and man-made underwater structures (e.g. ship hulls, aquaculture nets, moorings and sensors). More sophisticated optical, electrochemical and biosensors are needed to monitor the marine environment [2]. Sensor performance in sea water is degraded by biofouling, often quickly. Attachment of biofouling communities on sensors’ optical windows obstructs the light path, causing scattering and/or uneven absorption of different wavelengths, especially when fouling is dominated by photosynthetic organisms [3]. The recent introduction of International Maritime Organization pollution regulations has led the scientific community to seek new environmentally and economically acceptable antifouling methods.

No proposed solutions to marine biofouling problems meet the main specifications for protection of sensors against attachment of micro- and macro-organisms. Valuable antifouling strategies should:

• — not affect the measurement or the environment;

• — not consume much energy (to enhance endurance); and

• — be reliable under aggressive conditions (sea water corrosion, sediments, hydrostatic pressure) [4].

A promising approach involves imitating natural solutions (biomimicry) and improving them for the ideal engineering solution (bioinspiration) [5,6]. In the sea, every metazoan has sensors and their performance maintenance is crucial. Organisms use mechanical, chemical, physical and behavioural techniques to protect themselves against fouling. Among the physical methods used, surface micro- and nanotopography has received much recent attention [7–18].

The microtopography of decapod carapaces has a role (even if only temporary) in preventing attachment of organisms [18], but no studies of other crustacean structures have been recorded. Eyes are sensors that have to be constantly kept clean despite often being exposed to severe external conditions. Here, the eye of the green crab Carcinus maenas is regarded as an optical sensor model and its surface characterized by atomic force microscopy (AFM). Crab antifouling has been studied to some extent [4,9], but not eye micro- and nanotopography.

Crabs use various antifouling strategies to keep surfaces clean. Although eyes are largely kept clean by brush-like structures mounted on various appendages, the peculiar surface of the eyes may also be a key factor in their antifouling strategy, which is very effective, because scanning electron microscopy studies show that eyes normally show negligible levels of fouling by micro-organisms [19].

Carcinus maenas is a widely distributed epibenthic crab that inhabits hard and soft intertidal shallow habitats of European coasts and estuaries. It periodically moults (ecdysis) and during its life cycle is characterized by a change in shell colour, from bright green to dark red, owing to increasingly long intermoult phase durations, during which red astaxanthin is deposited in the shell. Red and green morphs are characterized by different levels of deterioration of the outer surfaces [19] as well as physiological and behavioural differences [20]. Here, both phenotypic morphs were compared. Also, because of the deterioration of all external surfaces due to intermoult abrasion caused by biotic and abiotic influences, and because of the decalcification process that takes place during the moulting cycle [19], specimens before and after moulting were investigated.

2. Material and methods

2.1. Crab collection and maintenance

Crabs were trapped during spring/summer at Monkstown (51°50′ 59.06″ N, 8°19′ 51.55″ W, Cork Harbour, Cork, Ireland). Thirty C. maenas (carapace width 40–60 mm) were used: 10 of the green morph, 10 of the red phenotypic morph and 10 ‘peelers’ (crabs about to moult). Crab colour was determined using the Munsell colour system [21,22]. Specimens with endopodites of third maxillipeds of hues from 5.0 Y to 10.0 GY were considered green, those of hues 5.0 R to 10.0 YR were considered red. Both morphs were kept in communal 40 l tanks (10 animals per tank) in a recirculating system with natural sea water at environmental temperature (about 15°C). Green/red crabs were fed weekly on fish and mussel flesh. Peeler crabs were kept in individual aerated 2 l tanks with natural sea water renewed daily at 20°C. Peelers were fed daily to induce moulting, within 2–7 days. After moult, exuviae (moulted integuments) were collected, and newly moulted soft crabs left in sea water for two weeks before analysis to allow sufficient shell hardening to permit AFM study. All crabs were killed with clove oil [23].

2.2. Atomic force microscopy: morphological analysis

Specimens’ eyes (green crabs, red crabs, exuviae and newly moulted crabs) were removed, fixed and treated with ultrasound and anionic surfactant (Tween 80) prior to AFM. Measurements were performed with a Veeco multimode V microscope; maximum travels were 125 µm on X,Y and 5.5 µm on Z; silicon tips (cantilever, nominal spring constant 2 N m⁻¹, nominal resonant frequency 70 kHz, tip radius less than 10 nm) in-air in tapping mode were used; the scan rate was between 0.3 and 1.0 Hz.

For image analysis, two images each 60 × 60 μm and two images each 10 × 10 μm were randomly taken for each crab eye. Different scan sizes were necessary to characterize the topography at different dimensional scales, one given by features of the order of micrometres and one given by features of the order of nanometres. Images (10 × 10 μm) were taken randomly inside the hexagon, avoiding channels between ommatidia (for eye morphology elucidation, see below). Analysis of the general morphology of the eye surface was performed on 60 × 60 μm image sections of 20 green crabs, using Nanoscope v. 7.20 (Veeco software). The newly moulted, red and exuviae groups were not considered in the morphological analysis because they could have been incompletely formed or characterized by too much surface deterioration.

2.3. Atomic force microscopy: eye surface characterization

To perform surface analysis, the same procedure used for morphology investigation was adopted: 10 crab eyes for each group were considered (green crabs, red crabs, exuviae and newly moulted crabs) and two images each 60 × 60 μm and two images each 10 × 10 μm were randomly taken for each crab eye. Open source software (ImageJ), using a plugin developed for this study, was used to characterize the surface roughness as root mean-squared (r.m.s.) of different crab groups’ eye surfaces. The plugin allowed calculation of surface roughness values (SRVs) at increasing cut-off lengths, thus enabling identification of the length at which the r.m.s. reaches a saturation value (correlation length) and remains constant [24]. A cut-off length higher than the correlation length was arbitrarily chosen for both image size scales (30 μm for 60 × 60 μm and 5 μm for 10 × 10 μm images), and all values above the selected cut-off length were used to produce an average r.m.s. value. Specific software developed for this study (based on LabView) was used to calculate: fractal dimension (F_d), texture aspect ratio (A_r), waviness (W_v), skewness of roughness (Sk_r) and skewness of waviness (Sk_w). All these parameters are important to evaluate the antifouling and fouling-release potential of a natural surface and their definitions are given in table 1. From table 2, it is possible to observe the type of correlation of the parameters considered with antifouling and fouling-release properties, whether positive (the higher the value the stronger the effect considered), or negative (the lower the value the stronger the effect considered) [15].

Table 1.

Definitions of surface parameters considered in the study (adapted from Scardino et al. [15]).

Roughness (r.m.s. root mean-squared) is the square root of the arithmetic mean of the departure of the surface from the mean line in the roughness profile. Values close to 0 are smoother with fewer peaks and troughs

Waviness is the average departure of the waviness profile from the mean line of the surface. Values close to 0 are smoother with fewer peaks and troughs

Skewness in the roughness profile (Skr) and waviness profile (Skw) is a measure of the symmetry of the amplitude distribution curve about the mean line. Skewness indicates whether the surface consists of mainly peaks, valleys or an equal combination of both. Positive values have a higher proportion of peaks than valleys

Texture aspect ratio (Ar) is used to identify texture pattern, isotropy or anisotropy. Anisotropy indicates the randomness or directionality of surface texture. Values close to 0 are highly directional whereas values closer to 1 have random roughness

Fractal dimension (Fd) quantifies whether irregular shapes have a common characteristic feature in that they are self-similar upon magnification. Values close to 3 are more fractal and values closer to 2 are less complex

Table 2.

Bioinspired design: recommended surface parameter values based on the fouling resistance and fouling-release properties of mollusc shells. ↓, negatively correlated with fouling resistance or fouling release; ↑, positively correlated with fouling resistance or fouling release (adapted from Scardino et al. [15]).

fouling resistance	fouling release
fractal dimension (Fd) ↓	waviness (Wv) ↑
skewness of roughness (Skr) ↑
skewness of waviness (Skw) ↑
texture aspect ratio (Ar) ↑
roughness (r.m.s.) ↓

2.4. Statistical analysis

Eye surface parameters were calculated for each organism by averaging the two independent images on the same eye, and the different crab groups were compared with one-way ANOVA or a Kruskal–Wallis test (the latter if conditions of normality and homogeneity of variance were not satisfied) was used at the significance level of 0.05 (software: SPSS Inc, Chertsey, UK).

3. Results

3.1. Morphological analysis

General morphological analysis of the crab eye surface was performed on 20 green morphs. These showed the typical arthropod compound eye structure with hexagonal ommatidia. Distances between sides (hexagon width), the channel depth and channel width were taken (black lines in figure 1). For analysis, the hexagons were considered as regular. The average length of each side of the hexagons was about 20 μm (19.94 ± 0.23 μm calculated as hexagon width/√3); they were interspersed with channels 1.5 μm wide with an average depth of about 400 nm (figure 1). Tip size (tip height = 14 μm) could have underestimated the channel depth by not reaching completely to the bottom of the channel.

Figure 1.

Morphological analysis: (a) transverse section of a Carcinus maenas ommatidium; (b) two adjacent ommatidia; (c) a channel between two ommatidia. Black bars indicate the measurement taken from the images: (a) hexagon width = 34.5 ± 0.4 µm; (b) channel depth = 420 ± 40 nm; (c) channel depth = 1.45 ± 0.04 µm Data are the mean of 20 crabs ± s.e.

3.2. Eye surface characterization

Analysis of the parameters characterizing the eye surface at both the 60 × 60 μm and 10 × 10 μm scales (figures 2 and 3) displayed almost no significant difference within the size scale groups. Results of the eye surface characterization in the four crab groups regarding the two dimensional scales are shown in tables 3 and 4. Crab groups are presented in the tables in order of intermoult phase duration. The exuviae group is the last group as it represents the end of the intermoult phase when the decalcification process and deterioration both affect the crab outer surface [19].

Figure 2.

Three-dimensional AFM image of a Carcinus maenas eye surface at the 60 × 60 μm scale.

Figure 3.

Three-dimensional AFM image of a Carcinus maenas eye surface at the 10 × 10 μm scale.

Table 3.

Carcinus maenas eye surface parameters (mean values ± s.e.) of the four crab groups (newly moulted, green crabs, red crabs, exuviae) at the 60 × 60 μm dimensional scale. r.m.s., roughness; W_v, waviness; F_d, fractal dimension; A_r, texture aspect ratio; Sk_r, skewness of roughness; Sk_w, skewness of waviness. F_d data in green crabs group and r.m.s. data in exuviae group (in bold) displayed significantly higher values than in all of the other crab groups (one-way ANOVA, p < 0.05).

parameter	newly moulted	green crabs	red crabs	exuviae
r.m.s. (nm)	140.06 ± 10.44	161.03 ± 10.44	137.25 ± 11.51	209.47 ± 20.54
Wv (nm)	19.60 ± 2.10	23.66 ± 3.37	18.77 ± 1.92	32.71 ± 4.58
Fd	2.58 ± 0.05	2.76 ± 0.03	2.58 ± 0.03	2.63 ± 0.04
Ar	0.53 ± 0.03	0.48 ± 0.03	0.52 ± 0.04	0.48 ± 0.04
Skr	−0.15 ± 0.05	0.11 ± 0.10	−0.02 ± 0.07	−0.02 ± 0.12
Skw	−0.04 ± 0.05	−0.04 ± 0.05	−0.09 ± 0.05	−0.03 ± 0.05

Table 4.

Carcinus maenas eye surface parameters (mean values ± s.e.) of the four crab groups (newly moulted, green crabs, red crabs, exuviae) at the 10 × 10 μm dimensional scale. r.m.s., roughness; W_v, waviness; F_d, fractal dimension; A_r, texture aspect ratio; Sk_r, skewness of roughness; Sk_w, skewness of waviness.

parameter	newly moulted	green crabs	red crabs	exuviae
r.m.s. (nm)	17.41 ± 1.40	16.83 ± 1.13	18.72 ± 1.48	17.59 ± 2.48
Wv (nm)	4.93 ± 1.32	4.04 + 1.41	3.59 ± 1.24	3.46 + 0.67
Fd	2.30 ± 0.05	2.31 ± 0.06	2.18 ± 0.04	2.16 ± 0.07
Ar	0.48 ± 0.05	0.59 ± 0.03	0.45 ± 0.04	0.51 ± 0.04
Skr	0.57 ± 0.13	0.77 ± 0.13	0.55 ± 0.14	0.64 ± 0.13
Skw	0.01 ± 0.08	−0.15 ± 0.10	−0.10 ± 0.07	−0.12 ± 0.10

The analysis of images of the eye surface at the 60 × 60 μm scale revealed significant differences among the different crab groups in just two out of the six investigated parameters: roughness (r.m.s.) and fractal dimension (F_d). R.m.s. values of exuviae (the moulted integument) were significantly higher than in all other crab groups (one-way ANOVA, p < 0.05). Waviness did not show any significant difference (one-way ANOVA, p = 0.05), even though the exuviae group seemed to be characterized by higher values than those of other categories (table 3). Fractal dimension (F_d) analysis revealed a highly significant difference between green crabs and all other groups (one-way ANOVA, p < 0.01), whereas texture aspect ratio (A_r) and skewness of roughness and waviness (Sk_r, Sk_w) did not show any significant differences among the tested crab groups (one-way ANOVA, p_Ar = 0.628; Kruskal–Wallis test, p_Skr = 0.362, p_Skw = 0.892). Green crabs clearly have a higher F_d than all other crab eye categories at the 60 × 60 μm scale, reflecting a greater physical complexity (figure 4).

Figure 4.

(a) Roughness values (r.m.s.) and (b) fractal dimension values (F_d) obtained from images at the 60 × 60 μm scale by ImageJ software in four crab (Carcinus maenas) groups: newly moulted, green crabs, red crabs, exuviae. r.m.s. values were taken from the cut-off length of 30 × 30 μm. Data are means of 20 observations (20 crabs) ± s.e. * = significantly different from other means (one-way ANOVA, p < 0.05); ** = highly significantly different from other means (one-way ANOVA, p < 0.01).

At the 10 × 10 μm scale (images were taken randomly inside the hexagon, avoiding the channels between ommatidia), no parameters differed significantly among the four tested crab groups, revealing a conservative feature of topography at the nanoscale (one-way ANOVA: p_r.m.s. = 0.645, p_Wv = 0.818, p_Fd = 0.144, p_Ar = 0.103; Kruskal–Wallis test: p_Skr = 0.468, p_Skw = 0.829; table 4).

4. Discussion

Characteristics of the C. maenas eye surface are compatible with it being exploitable for bioinspired marine sensor antifouling. Three key elements prevent biofouling attachment to surfaces: (i) cells/organisms must remain above topographical features and not be able to settle between them; (ii) they must not be able to contact and settle their entire structure on a single feature; and (iii) if they bridge two features, then they must not be able to contact the intervening floor [25].

Crab eye morphology shows deep narrow channels (approx. 1.45 µm width) that tend to prevent the attachment of algal zoospores [26], because the settling organisms cannot reach the floors of the channels. On the other hand, at the microscopic level, the wide surfaces of the ommatidia apparently represent an attractive substratum for the attachment of invertebrate larvae/algal zoospores. While it is usually assumed that mechanical cleaning mechanisms keep these surfaces clean, our detailed investigation of their micro- and nanotopographies indicates that these may be compatible with prevention of biofilm formation that influences the subsequent attachment of other biofoulers [27–32]. The r.m.s. value recorded at the 10 × 10 µm scale is similar to roughness values already known to have anti-microfouling and microfouling-release properties by virtue of providing a reduced number of ‘attachment points’ for potential settlers. Details of attachment point theory and its relevance to natural and micro-engineered surfaces are available elsewhere [33–36]. This SRV (15 nm) has been suggested for the design of the surface of the optical components for marine sensors, and the properties given by this SRV seem to be independent of the material used [37].

Almost all of the analysed surface parameters at the 60 × 60 µm size scale and all at the 10 × 10 µm showed constant values amongst all of the crab groups, identifying a conservative feature of the eye surface. The difference recorded in r.m.s. values for the exuviae at the 60 × 60 µm size scale can be explained by the extreme modification of the crab outer surface owing to the time-dependent deterioration and the decalcification of the surface observed in the exuviae occurring during the moulting process [19].

Differences in F_d values recorded for the green crab group at the 60 × 60 µm scale are probably due to the greater complexity of green crabs’ eye surfaces. It seems that the integument in newly moulted crabs is incompletely developed because of partial calcification and it is flattened both in red crabs and exuviae groups by time-dependent deterioration and decalcification, thus producing lower F_d values.

Image analysis revealed surface parameters similar to those already demonstrated for the external shell surfaces of some bivalve molluscs, already shown to have antifouling and/or fouling-release properties [15]. For example, negative skewness values, indicating the predominance of valleys over peaks on the surface, and the fractal dimension are in the same range as those of shells of Mytilus galloprovincialis at the 60 × 60 µm scale. At the 10 × 10 µm scale, fractal dimension values were similar to those recorded for the outer shell surfaces of Tellina inflata, known to have valuable antifouling properties [15]. This suggests the crab eye nanostructured surface is a useful model for the prevention of biofouling. Values obtained for waviness, roughness, fractal dimension and aspect ratio, even if not directly comparable with values displayed by other natural surfaces with antifouling or fouling-release properties, were fundamental to the characterization of the surface and thus represent a valuable starting point for possible fabrication of artificial materials (or epoxy resins) with surfaces similar to the crab eye surface. This would be the next step for full evaluation of the antifouling and fouling-release potential of crab eye micro- and nanotopography.

5. Conclusions

Optical windows used in cameras and light detectors, membranes used in chemical sensors and diaphragms used in microphones are presently limited to short deployments in marine environments, because the formation of the biofouling community (biofilm) degrades their performance. Our study, besides giving new insight into the morphology of the C. maenas eye, indicates an antifouling/fouling-release potential for the surface, especially at the scale of 10 × 10 µm. The two different levels of roughness characterizing the surface may influence the wettability properties of the structure and its interaction with organic molecules [37–44]. Materials inspired by crab eye surface topography could be used synergistically with other techniques to prevent fouling-induced degradation of marine sensors.