Spectral transmission of the principal-eye corneas of jumping spiders: implications for ultraviolet vision

Ultraviolet (UV; wavelengths <400 nm) vision is widespread in animals and often plays an important role in interspecific and intraspecific communication (Silberglied, 1979; Tovée, 1995). UV sensitivity is also known in jumping spiders (Araneae: Salticidae), a group renowned for their unique and complex eyes that support exceptional spatial acuity (Land, 1969; Land, 1985; Williams and McIntyre, 1980; Blest et al., 1990; Harland et al., 1999; Land and Nilsson, 2002; Harland et al., 2012). Like most spiders, salticids have four pairs of eyes: one pair of principal eyes and three pairs of secondary eyes. The large, forward-facing principal eyes are primarily responsible for good spatial acuity (see Land, 1985; Blest et al., 1990) and allow salticids to identify conspecifics, prey and predators from a distance, even when denied access to information from other sensory modalities (Jackson and Blest, 1982).

It is also known that a few species of salticids can distinguish colours (Homann, 1928; Nakamura and Yamashita, 2000) and have UV photoreceptors in their eyes (DeVoe, 1975; Yamashita and Tateda, 1976; Blest et al., 1981; Peaslee and Wilson, 1989). Electrophysiological experiments, including experiments based on intracellular recordings from individual photoreceptor cells in the principal-eye retina, have revealed that four species of salticids have UV-sensitive photoreceptor cells in the retina, with peak sensitivity in the range of 330–380 nm: Phidippus regius (DeVoe, 1975), Menemerus fulvus (formerly confuses) (Yamashita and Tateda, 1976), Plexippus validus (Blest et al., 1981) and Maevia inclemens (Peaslee and Wilson, 1989). The existence of opsin gene Rh3 also suggests the likelihood of UV sensitivity in Hasarius adansoni and Plexippus paykulli (Koyanagi et al., 2008). However, whether UV sensitivity is widespread in salticids remains unclear at present.

Many salticids are spectacularly colourful (Jackson, 1982), and some species have strikingly iridescent markings (Lim and Li, 2006b; Li et al., 2008b; Tay and Li, 2010). Iridescence is of particular interest because we now know that it often adds UV colouration to salticids (Land et al., 2007; Lim and Li, 2006b; Lim and Li, 2007; Lim et al., 2007; Li et al., 2008b). Behavioural experiments (Lim and Li 2006a; Li et al., 2008a; Lim et al., 2007; Li et al., 2008b) have confirmed that two species of salticids (Cosmophasis umbratica and Phintella vittata) can discriminate between signals in both the UV and the human-visible (i.e. 400–700 nm) wavelength ranges. These species also use UV-based signals in male–male competition and female mate choice (Lim and Li, 2006a; Li et al., 2008a; Lim et al., 2007; Lim et al., 2008). Portia labiata, a spider-eating salticid that is anything but ornate, also makes use of UV vision, but in the context of predation rather than courtship and intraspecific aggression. The orb webs built by some of the spiders preyed upon by P. labiata include densely woven patterns of silk called web decorations (Herberstein et al., 2000). In some species, silk decorations reflect UV light, which in turn attracts UV-oriented insects that are preyed upon by the resident spider (Herberstein et al., 2000; Bruce et al., 2005). UV-reflecting web decorations also attract P. labiata, which then invades the web and preys on the resident spider (Seah and Li, 2001; Li and Lim, 2005; Zou et al., 2011). Therefore, perhaps many salticids, especially those with strikingly iridescent patterns, have eyes that are sensitive to UV light.

One requirement for vision based on UV sensitivity is a retina that contains photoreceptors that are sensitive to UV. However, there is another more basic requirement for UV vision: the lenses, corneas and other ocular media (e.g. retinal oil droplets in birds) of the eyes must also allow UV light to reach the retinas. Thus, any consideration of UV vision in salticids must take into account the transmission of their ocular media. Like other spiders, the main type of ocular media in salticids is the cornea (Land and Nilsson, 2002). However, no studies have been conducted to examine the transmission properties of salticid corneas.

In addition, microspectrophotometry (MSP), the analysis of retinal potentials using electrophysiological methods (electroretinography) and behavioural experiments based on visual discrimination tasks have been used to assess the UV sensitivity of animals, including salticids (e.g. DeVoe, 1975; Yamashita and Tateda, 1976; Blest et al., 1981; Lim and Li, 2006a; Lim et al., 2008; Li et al., 2008a). However, working with multiple species using these three methods is quite a difficult task. With more than 5000 salticid species (Platnick, 2012), measuring the transmission characteristics of their ocular media may be an effective first step in ascertaining the potential for UV vision (Siebeck and Marshall, 2000: Siebeck and Marshall, 2001; Siebeck and Marshall, 2007) because the ocular media can act as a filter that limits the wavelengths that reach the retina in animal eyes (Douglas and McGuigan, 1989; Siebeck and Marshall, 2000; Siebeck and Marshall, 2001). Moreover, previous studies of ocular media transmission have mainly focused on vertebrates, particularly on fish (Douglas and McGuigan, 1989; Thorpe et al., 1993; Douglas and Marshall, 1999; Siebeck and Marshall, 2000; Siebeck and Marshall, 2001; Losey et al., 2003; Siebeck and Marshall, 2007; Eckes et al., 2008). Ocular media transmittance in invertebrates is poorly understood (Douglas and Marshall, 1999).

Salticids occur worldwide, in a variety of habitats ranging from tropical rainforests to temperate grasslands, and are able to survive in a wide range of light environments. The range of wavelengths to which a salticid is sensitive may be tuned to the specific optical habitat in which it lives (Lythgoe, 1972; Lythgoe, 1979), which probably enhances its ability to detect mates, prey and predators within its specific habitat. As UV light varies in time and place (Endler, 1993; Flint and Caldwell, 1998), it is instructive to relate the spectral properties of salticid ocular media to the various light habitats, such as forests and open bush, and geographic locations such as tropical and temperate places.

In this study, we measured the spectral transmission properties of the principal eye corneas of 128 salticid species collected from forest and open bush, and from subtropical and tropical habitats, to test two hypotheses: (1) within the Salticidae, it is widespread for the principal-eye cornea to allow UV transmission; and (2) the transmission characteristics of the principal-eye cornea vary with the optical habitats in which salticids live.

A total of 547 juvenile and adult spiders belonging to 128 species of salticids in 13 subfamilies were collected from China (Hainan, Hubei, Hunan and Yunnan Provinces) and Singapore (Table 1). The spiders were transported to the laboratory at Hubei University (Wuhan, Hubei Province, China) and maintained following a protocol similar to that used in earlier salticid studies (Lim and Li, 2004; Lim and Li, 2006a; Lim and Li, 2006b; Li et al., 2008b). They were kept individually in plastic cylindrical cages (60×80 mm, diameter × height:). All spiders were provided ad libitum with a diet of fruit flies (Drosophila melanogaster) once a week, and housed in the laboratory under controlled environmental conditions (12 h:12 h light:dark cycle, lights on 08:00 h, 25±1°C, 70–90% relative humidity). Measurements were only obtained from living specimens.

The spectral transmission (280–700 nm) of the principal-eye corneas was measured with an Ocean Optics USB2000 spectrometer (Ocean Optics, Dunedin, FL, USA). Before taking the measurements, we anaesthetised each spider with CO₂ gas and placed it in 0.9% saline in a small Petri dish (9×2 cm). We then cut the principal eye from the cephalothorax, isolated the entire cornea from the other parts of the eye under a stereomicroscope as quickly as possible to avoid artefacts from tissue degradation, and measured the spectral transmission within 1 h (see Douglas and McGuigan, 1989). For the measurements, we mounted the isolated cornea directly on the pinhole of a quartz fibre-optic probe (50 μm diameter; Ocean Optics), held by a modified microscope stand (Fig. 1). To ensure that the lens was properly mounted on the pinhole, a portable magnifying glass was used to monitor the procedure (Fig. 1). Light from a quartz halogen bulb and pulsed xenon light source (DH-2000; Ocean Optics) was fed through the illuminating fibre-optic probe, directed through the cornea mounted on the pinhole, and into the quartz fibre-optic probe coupled to a USB2000 spectrometer (Fig. 1) (Siebeck and Marshall, 2001; Siebeck and Marshall, 2007). Five transmission readings were taken and averaged for each cornea.

The absolute transmission measurements could vary if the cornea's position changed, even slightly, within the holder (Douglas and McGuigan, 1989). However, the relative spectral transmission and the T₅₀ cut-off value, which represents the wavelength at which 50% of the maximum transmission is reached, would not be affected. Therefore, to eliminate any artefactual variation in absolute transmission and also to characterise eye corneal transmission, we calculated the T₅₀ cut-off wavelength for the principal-eye corneas as described in earlier studies (Partridge, 1989; Douglas and McGuigan, 1989). In doing so, we first normalised the transmission curve, making each spectral scan (1219 scans between 280 and 700 nm) equal 100% transmission at 700 nm. We then identified the 50% maximum transmission values for each of these scans, and used 30 points to either side of these values (approximately a range of 22 nm) to calculate, by linear regression, the equation of the tangent to the normalised transmission curve at that point. Finally, we used this line to obtain a value for the T₅₀ cut-off wavelength, the wavelength corresponding to 50% of the maximum transmission at 700 nm, which is close to a maximal transmission value (see also Douglas and MacGuigan, 1989). For convenience, we used the T₅₀ cut-off value and the T₅₀ cut-off wavelength interchangeably.

The principal-eye corneas of the 128 salticid species we examined had uniformly high transmission at long wavelengths which declined at shorter wavelengths (Fig. 2A). The spectral position of the T₅₀ cut-off was species-specific, and the T₅₀ cut-off wavelength ranged from 291 nm (Mendoza elongate) to 334 nm (Phintella sp. 3; Table 1).

Based on the slope and shape of the curve, three distinct classes (Classes I, II and III) of corneal transmission curves could be distinguished (Fig. 2A). The Class I curve had a single sharp cut-off and a steep slope (Fig. 2A, black curve), and its T₅₀ values ranged from 291 nm (M. elongate) to 317 nm (Thorelliola ensifera). The Class II curve had a single, less-steep slope and a gradual onset of the cut-off (Fig. 2A, red curve), and the T₅₀ values ranged between 295 nm (Pseudicius vulpes and Siler semiglaucus) and 334 nm (Phintella sp. 3). The Class III curve had an intermediate step, which appeared as a shoulder on the descending part of the transmission curve (Fig. 2A, green curve). Its 25 and 75% maximal transmission positioned at 298 and 372 nm, respectively, and the range between the T₂₅ and T₇₅ values was larger than 70 nm (Table 1; Plexippoides regius). The Class II transmission curve (76.6% of 128 species) was more common than the Class I (22.7%) and Class III (0.7%) curves (Fig. 2B).

The frequency distribution of the corneal transmission T₅₀ values across all measured species except one (Plexippoides regius, which did not have a T₅₀ value) appeared to be bimodal, with one pronounced peak located at 300 nm (97 species cut off between 295 and 305 nm) and a small peak at around 320 nm (13 species cut off between 315 and 325 nm; Fig. 3).

The principal-eye corneal diameter varied with species, ranging from 0.30 to 0.99 mm (Table 1). However, the corneal transmission T₅₀ cut-off values showed no significant changes with corneal diameter across 127 species (Fig. 4), or in any species in which at least 10 individuals were examined.

Habitat was divided into four types based on the optic environment and geographical region: tropical forest (TF), tropical bush (TB), subtropical forest (SF) and subtropical bush (SB). There was a significant difference in the corneal transmission T₅₀ values of salticids living in different habitats (Kruskal–Wallis test: H=14.226, d.f.=3, N=127, P=0.003; Fig. 5). Post hoc paired comparisons showed that the corneal transmission T₅₀ cut-off wavelengths were higher for subtropical forest salticids than for subtropical bush and tropical (forest and bush) salticids, and tropical forest salticids had higher cut-off wavelengths than those of salticids living in tropical and subtropical bush. However, no significant difference was found between the corneal transmission T₅₀ values of salticid species living in subtropical and tropical bush.

Our study shows that UV radiation readily passes through the principal-eye corneas of all measured salticid species, despite there being substantial interspecific variation in transmission properties. This supports the hypothesis that it is common for the principal-eye corneas of salticids to allow UV transmission. These findings, when considered in conjunction with earlier studies that established for a few species that UV has a role in predation (Seah and Li, 2001; Li and Lim, 2005; Zou et al., 2011) and intraspecific communication (Lim and Li, 2006a; Lim et al., 2007; Lim et al., 2008; Li et al., 2008a), suggest that UV-sensitive receptors are also common in salticids.

Previous studies have documented considerable age-related and interspecific variations in transmission of lenses and corneas in vertebrates, particularly in fish (Douglas and McGuigan, 1989; Thorpe and Douglas, 1993; Losey et al., 2000; Nelson et al., 2003; Siebeck and Marshall, 2001; Siebeck and Marshall, 2007). Age is known to influence the transmission properties of the lens and cornea, as the diameter of lenses and corneas generally increases with age (Thorpe and Douglas, 1993; Eckes et al., 2008). However, the diameter of the principal-eye cornea in salticids does not seem to explain the species distribution of the corneal transmission, as we found no significant effects of the principal-eye cornea diameter on the corneal transmission T₅₀ cut-off wavelengths across or within species (Table 1, Fig. 4). Perhaps change in the UV transmission properties is not dependent on the increase in the cornea diameter alone. Changes in the UV-blocking pigment type and/or concentration of the cornea may also contribute to the variation in the corneal transmission in salticids (Douglas, 1989; Thorpe and Douglas, 1993; Losey et al., 2000; Nelson et al., 2003). It is also worth noting that the phylogeny may mask the significant relationship between the principal-eye cornea diameter and the corneal transmission T₅₀ cut-off wavelengths (Siebeck and Marshall, 2007). Therefore, a phylogenetic framework should be used to test such a relationship.

The corneas classified as Class I probably contain no particular absorbing pigment and can thus transmit almost all light down to approximately 291–317 nm. The transmission spectra of corneas classified as Class II, however, suggest that in species with T₅₀ cut-off values longer than 320 nm the corneas may have one or more specific pigments that absorb shorter wavelengths (Table 1). However, the chemical composition of the corneal pigments has not yet been investigated. The lens pigments of UV-absorbable materials have been identified in fish: for instance, mycosporine compounds with absorbance maxima at 320, 330, 360 and 385 nm; 3-hydroxykynurenine with an absorbance maximum at 370 nm (Thorpe and Douglas, 1993); and carotenoid pigmentation with maxima at approximately 425, 440 and 480 nm (Muntz, 1973; Appleby and Muntz 1979). Nevertheless, further studies are needed to identify the UV-absorbable proteins in the corneas of salticids and to determine whether such pigments can account for the corneal transmission spectra observed here.

Interestingly, there is a substantial difference in the corneal transmission T₅₀ cut-off values among salticids living in different habitats. This implies that there may be a strong correlation between the light habitat and corneal transmission of salticid spiders. Whether in the tropics or subtropics, salticid species living in shade forests where there are lower levels of illumination have principal-eye corneas with higher T₅₀ cut-off wavelengths than salticid species living in open shrubs, where there are higher levels of sunlight. This appears to be in contrast to fish, many of which have ocular media transparent to UV wavelengths, yet they experience lower levels of illumination (Douglas and McGuigan, 1989). Perhaps these results are related to the nature of sunlight being different in different habitats. UV-B (280–315 nm) radiation can be easily blocked off, so almost no UV-B radiation would be available to be utilised in the forest environments (Endler, 1993). In adapting to such a light habitat, perhaps either UV-B visual pigments or the corneal transmission of extremely short wavelength radiation may have been lost through long-term evolution. We propose that divergent transmission of UV-B by the principal-eye cornea has evolved at least twice in salticids, based on the evolution of the ultrastructure of salticid principal eyes (Su et al., 2007). There are two possibilities: a species that only recently gained SWS1 (UV) opsin expression might maintain a UV-B transmitting cornea, or a species that gained a UV-B blocking cornea might not have lost SWS2B (violet) opsin expression immediately (Hofmann et al., 2010).

It has long been recognised that UV sensitivity comes at a cost because in many animals, high-energy solar UV, especially UV-B, has the potential to damage the ocular tissues of the eyes (Zigman, 1971; Waxler and Hitchins, 1986), including the retina (Paul and Gwynn-Jones, 2003). It seems obvious, therefore, that an animal that is habitually exposed to high levels of UV would benefit from a UV-blocking filter that prevents these wavelengths from reaching the retina. However, UV transmission in the principal-eye corneas of all measured salticid species implies that the majority of these species do not have such a UV-blocking filter. Although it is unclear whether UV radiation has the potential to damage the ocular tissues of salticid eyes, the results from our study suggest that seeing UV may confer an advantage that overrides the potential cost of UV damage.

We are very grateful to Tsyr-Huei Chiou, Hua Duan, N. Justin Marshall, Yingying Ren, Ulrike E. Siekeck and Shoupan Yin for their technical advice. We thank Wenjin Gan, Matjaz Gregoric, Matjaz Kuntner, Matjaz Jie Liao and Lingbing Wu for collecting spiders; Jie Liu for identifying some of the species; and Wenjin Gan, Jie Liu, Yu Peng, Yingying Ren and Shoupan Yin for their support and help. Valuable and generous assistance was provided by the Xishuangbanna Tropical Botanic Garden (XTBG) at the Chinese Academy of Sciences. Special thanks are extended to Jin Chen, Liming Li and Zhihu Li from the XTBG for their generous assistance, without which this study would not have been possible. We also thank CITES and the Agri-food & Veterinary Authority of Singapore for providing export permits.