Nature, Source and Function of Pigments in Tardigrades: In Vivo Raman Imaging of Carotenoids in Echiniscus blumi

Tardigrades are microscopic aquatic animals with remarkable abilities to withstand harsh physical conditions such as dehydration or exposure to harmful highly energetic radiation. The mechanisms responsible for such robustness are presently little known, but protection against oxidative stresses is thought to play a role. Despite the fact that many tardigrade species are variously pigmented, scarce information is available about this characteristic. By applying Raman micro-spectroscopy on living specimens, pigments in the tardigrade Echiniscus blumi are identified as carotenoids, and their distribution within the animal body is visualized. The dietary origin of these pigments is demonstrated, as well as their presence in the eggs and in eye-spots of these animals, together with their absence in the outer layer of the animal (i.e., cuticle and epidermis). Using in-vivo semi-quantitative Raman micro-spectroscopy, a decrease in carotenoid content is detected after inducing oxidative stress, demonstrating that this approach can be used for studying the role of carotenoids in oxidative stress-related processes in tardigrades. This approach could be thus used in further investigations to test several hypotheses concerning the function of these carotenoids in tardigrades as photo-protective pigments against ionizing radiations or as antioxidants defending these organisms against the oxidative stress occurring during desiccation processes.


Introduction
Tardigrades, or water bears, are microscopic aquatic animals (100-1000 mm in length), most of which belong to the group of desiccation-tolerant multicellular organisms [1,2]. Although tardigrades colonized a wide range of ecosystems and habitats all over the world, the highest number of species is found in terrestrial ecosystems where they often constitute the major component of the fauna in lichens and mosses. These habitats provide a ''sponge-like'' environment featuring a myriad of small pockets of water, in which tardigrades live. Such habitats, however, are subject to occasional periods of desiccation or freezing: as mosses and lichens lose water, tardigrades desiccate or freeze consequently, entering in physiological states called ''anhydrobiosis'' and ''cryobiosis'', respectively. In such states, grouped under the general name of ''cryptobiosis'', metabolic activity is suspended because of absence of liquid water, until environmental conditions return favourable [1,2]. Tardigrades can persist in the anhydrobiotic state even for years, during which they are able to withstand several physical and chemical extremes [1][2][3][4][5], including exposure to vacuum and both solar and cosmic ionizing radiation as found in low Earth orbit space [3,4,6,7]. Remarkably, even in their hydrated active state, tardigrades show an unexpectedly high survival rate upon freezing (even with temperatures as low as 2196uC) [8,9] or exposure to high levels of ionizing radiations [3,5]. The biochemical and physiological mechanisms underlying the extraordinary resistance of both desiccated and hydrated tardigrades to such harsh conditions are currently little known and constitute an intriguing challenge for biologists and biochemists.
Regardless of the wide range of colonised environments, tardigrades do not exhibit a great variation in internal organ structure and organization. Their body, invariably constituted by an anterior region (or head region) followed by four trunk segments with a pair of legs each, is covered by a chitinous cuticle that must be shed as the organism grows and by a monolayer of flat epidermal cells [10]. All tardigrades have muscular, nervous, reproductive and digestive systems, and their body cavity, which is rich in free-floating storage cells, allows efficient nutrition and gas exchange with no need for circulatory or respiratory systems [10]. However, in spite of this uniformity in body structure, tardigrades exhibit a variety of pigmentations: several tardigrade species are translucent or whitish, but many other are colored, sometimes markedly, with a variety of shades of red-orange, brown, green and yellow pigments [10,11]. These pigments can be uniformly or unevenly distributed in the body cavity, or may be located in storage cells or in the epidermis cells. Sometimes, the animal pigmentation may be caused by gut content [10,11]. Moreover, dark or red pigmented eye-spots may be present on the head region in close association with the outer dorso-lateral lobe of the brain, depending on the genus or species [10,12]. Each eye-spot consists of a single pigment cup cell, filled with spherical electrondense pigmented granules, one or two microvillous sensory cells, and ciliary sensory cells [12].
Within tardigrades, the genus Echiniscus is the most rich in the number of species, and most of them are brightly coloured [13]. Despite the occurrence of pigments is well documented for many Echiniscus species and many other genera belonging to the family Echiniscidae, their chemical nature as well as their function remains unknown, especially in relation to their resistance to harsh physical and chemical conditions. Some hypotheses were formulated about these pigments [14,15], but to the best of our knowledge no direct and conclusive experimental proof about their structure, source or function has been reported. In order to resolve these open questions, approaches involving techniques capable of performing in-vivo chemical analysis on single tardigrades are highly desirable. The low sample quantity available and the in-vivo requirement rule out most of the traditional analytical methods such as separation-based, electrochemical or mass spectrometry techniques. Raman microspectroscopy, however, has all the characteristics required for such a task.
Raman spectroscopy is a non-destructive, semi-quantitative analytical technique based on the scattering of laser radiation by vibrating molecules that has proved to be an ideal tool for studying living cells and biological tissues [16,17]. Information about the biochemical species present in a sample can be simply obtained by illuminating it with a low-power laser and then analysing the scattered light. Since it originates from molecular vibrations, a Raman spectrum closely depends on molecular structure and is thus unique to each molecule, in principle allowing the identification of chemical species on a spectral basis [18]. The level of structural detail conveyed by Raman spectroscopy ranges from the identification of the complete molecular structure to a general classification of the molecular class of the analyte of interest, depending on its spectral characteristics as well as on the chemical complexity of the sample. Moreover, the intensity of the Raman scattering can be exploited as a ''contrast mechanism'' to image the spatial distribution of the main biochemical constituents of a biological sample (i.e. Raman mapping or imaging).
In this paper we unambiguously identify the pigments in the heterotardigrade Echiniscus blumi Richters 1903 as carotenoids, demonstrate their food origin and set-up a method to study their anti-oxidant function in these animals by in-vivo spectroscopic and imaging techniques based on Raman scattering. Using this semiquantitative methodology based on Raman imaging, changes in carotenoid content are monitored in living tardigrades undergoing oxidative stress.

Tardigrade Collection
Animals and exuvia with eggs of the moss-dwelling and anhydrobiotic heterotardigrade Echiniscus blumi Richters 1903 were extracted from the moss Grimmia orbicularis Bruch ex Wilson growing on a rocky outcrop at Sasso Tignoso (Modena, Italy; N44u12.8379, E10u33.2379, 1382 m a.s.l.). In order to extract tardigrades, moss was maintained in water for about 30 min. After soaking, specimens were separated using sieves, collected under a stereomicroscope and maintained over 4 h in water (natural mineral water/distilled water; 1:1) at 20uC before any experiments. No specific permits were required for the described field studies, which did not involve endangered or protected species.

Raman Spectroscopy and Imaging Measurements
All Raman and photoluminescence spectra and maps were acquired using an InVia micro-Raman spectrometer (Renishaw plc, Wotton-under-Edge, UK) equipped an argon ion laser emitting 50 mW at 514.5 nm (Modu-Laser, Centerville, UT) and a motorized stage (Prior, Cambridge, UK). The laser was focused on samples via a 606objective (N.A. 1.00), its power being below 1 mW at the sample and kept constant during each measurement. Raman maps were collected in the fast-imaging Streamline TM mode, with acquisition times varying from approximately 3 to 5 min (depending on the map size), and with lateral resolutions (i.e. step sizes) of 1.1 mm (Fig. 1C), 4.3 mm (Figs. 1B, 1E, 2A, and 3A') or 5.4 mm ( Fig. 4 and Figs S3 ad S4 in SI). Data pre-processing (cosmic rays removal, baseline correction, normalization) and analysis, as well as image production, was performed with hyperSpec package (http://hyperspec.r-forge.r-project.org) for R [19].
For Raman measurements, tardigrades were singularly put in a drop of water between a CaF 2 slide (Crystal GmbH, Berlin, Germany) and a 15615 mm, 0.5 mm thick glass coverslip, in such a way that the animals were slightly compressed and unable to move. After collection of Raman data, each specimen was put again in water to check its viability on the basis of active movements of its body. All the data reported were collected from tardigrades which satisfied the viability criterion before and after the Raman measurements.

Test to Evaluate the Effects of H 2 O 2 on Tardigrade Survival
Ten different final concentration of H 2 O 2 (5, 10, 20, 25, 35, 45, 55, 65, 75 and 150 mM) were tested. For each H 2 O 2 concentration, five replicates each of 5 hydrated animals were immersed for 15 min in 300 ml of H 2 O 2 (30% w/w) in water at 20uC. Then, animals were transferred in water and their viability was verified after 1 h and 24 h from the end of the stress. As control, 5 replicates each with 5 tardigrades were maintained in water for 15 min at 20uC. The survival of E. blumi treated with H 2 O 2 is shown in SI (Fig. S2). The percentage survival of control tardigrades was 100%.

Measurements of Carotenoids Content in Tardigrade before and after H 2 O 2 Treatment
Oxidative stress was individually induced in 8 E. blumi specimens (i.e. 8 independent replicates: experimental group) upon exposure to 300 ml of H 2 O 2 25 mM for 15 min at 20uC. As control group, 8 specimens individually (i.e. 8 independent replicates) maintained in water for 15 min at 20uC were used. For each specimen in the experimental group, two Raman maps were collected, one before H 2 O 2 exposure and the other one 1 h after the end of H 2 O 2 exposure. For the control group, the two Raman maps were collected immediately before and 1 h after immersion in water. For both the experimental and control groups, the tardigrade viability was checked after the former Raman map collection, after H 2 O 2 exposure, after the latter Raman map collection, and 24 h after the end of treatment.
A first estimate of the total carotenoid content in E. blumi specimens was made upon comparing the intensity of the average Raman spectrum of the map of each animal before and after H 2 O 2 exposure. Spectra associated to background pixels were not taken into account for averaging. For a more accurate estimate, histograms were calculated and compared, showing the distribution of the integrated Raman intensity calculated for the n 1 band (1460-1570 cm 21 ). The total integrated Raman intensity for the same band was calculated by summing the integrated intensities from all the spectra in a map, and it was taken as a third estimate of the total carotenoid content.

Carotenoid Nature and Localization of the Pigments
The typical average Raman spectrum of living E. blumi specimens ( Fig. 1) is completely dominated by characteristic carotenoid bands which are observed in spectra excited with a blue or green laser, such as the intense n 1 band at 1521 cm 21 arising from stretching vibrations of conjugated C = C bonds [20]. Bands due to other usual biochemical species such as proteins, sugars, lipids and nucleic acids are not observed because their Raman signal is overwhelmed by carotenoid bands, whose intensity is selectively enhanced by several orders of magnitude via the resonance Raman (RR) effect [20,21].
A Raman shift of 1521 cm 21 for the n 1 band, the relatively weak n 4 band at 962 cm 21 and the intensity pattern for the n 2 bands cluster (between 1150 and 1200 cm 21 ) suggest an average molecular structure for carotenoids consisting of an undistorted, linear main carbon chain having 11 conjugated C = C double bonds in an ''all-trans'' configuration [20,22,23].
A qualitative picture of the spatial distribution of pigments within E. blumi specimens can be visualized upon mapping the Raman intensity of the most intense carotenoid band (Fig. 1). Carotenoids are present inside the whole body cavity, but they appear to be absent in the outer layer of the animal (i.e. cuticle and epidermis) as well as in the claws and in the buccal-pharyngeal apparatus (which is located within the head region). In the tardigrade investigated in Fig. 1, the carotenoid concentration in the body cavity is higher than in the region corresponding to the gut (Fig. 1B), which appears as a dark pigmented area in the middle of the animal body (Fig. 1A). Not all the specimens investigated present such differences in carotenoid concentration between the gut and the body cavity (see Figs. S3 and S4 in SI). On the other hand, all the tardigrades examined in this study show an overall decrease of carotenoid concentration in the head region, with the exception of two small round areas corresponding to the tardigrade eye-spots, where the carotenoid concentration is higher (Fig. 1C). The same carotenoids observed in the body cavity appear to be present, although with a more homogeneous distribution, in the yolk of E. blumi eggs (Fig. 1E), whose average Raman spectrum (Fig. 1F, bottom) is identical to that of the animal (Fig. 1F, top). Conversely, the exuvium (i.e. the cuticle left by tardigrades during moulting and used as container for the laid eggs) does not show carotenoid Raman bands, further indicating that the cuticular structures of this species lack such pigments.
Upon a more detailed analysis, slight differences in band position and relative intensities appear between spectra from different parts of the tardigrade body (Fig. 2). Specifically, the ratio between the intensities of the n 1 and the n 2 bands in the spectra of carotenoids localized in the body cavity and in the eye-spots is higher than that of carotenoids in the gut ( Fig. 2A, B; pixels in red correspond to higher intensity ratio), indicating a variation in the relative composition of the molecular structure between the pigments located in those different parts of the tardigrade. As a further indication of such differences, there is an inverse correlation between the n 1 /n 2 intensity ratio and the position of the n 1 band (Fig. 2B'), so that in spectra with a higher ratio the n 1 band occurs at a slightly lower Raman shift, with a maximum difference of 2 cm 21 . Such differences in Raman shift of the n 1 band for carotenoids localized in the gut of tardigrade suggest that they have a lower number of conjugated C = C bonds with respect to the carotenoids observed in the body cavity.

Dietary Origin of the Pigments
Raman spectra collected from leaves of the moss G. orbicularis, which constitutes the habitat of E. blumi population considered in this paper, are identical to those acquired from the animal's gut (Fig. 3), indicating that the carotenoids observed in the tardigrade are the same as those present in the moss, thus revealing their dietary origin. As inferred from the average Raman intensities, carotenoids are much (i.e. approximately 20 times) less concentrated in moss leaves than in the tardigrade (Fig. S1). The presence of moss cellular material in the digestive system of this tardigrade is further confirmed by the occurrence of characteristic chlorophyll fluorescence bands in the photoluminescence emission spectra (Fig. 3A) collected from the gut of the same specimen investigated in Figs. 1 and 2. Upon mapping fluorescence intensity at 670 nm, chlorophyll appears to be specifically found in the tardigrade gut (Fig. 3A'), co-localized with those carotenoids having a lower n 1 /n 2 intensity ratio and a n 1 band at 1523 cm 21 (blue area in Fig. 2A), which are the same spectral features of moss carotenoids (Fig. 3B).

Semi-quantification of Oxidative Stress-related Effects on the Pigments
Experiments on E. blumi specimens showed that the percentage survival, at both 1 h and 24 h after inducing the oxidative stress, sharply decreased upon exposure to H 2 O 2 concentrations higher than 65 mM, reaching 0% at 150 mM (see Fig. S2 in SI).
According to these data we decided to induce oxidative stress upon exposing tardigrades to 25 mM H 2 O 2 (i.e. a concentration permitting high stress compatible with high survival) and then, by exploiting the correlation between the concentration of carotenoids and the intensity of their Raman signal [18,24,25], to perform a semi-quantitative analysis of these pigments in living tardigrades before and after H 2 O 2 exposure.  It should be noted that un-treated specimens, in both experimental and control groups, showed remarkable interindividual differences in total carotenoid content, as inferred from the total integrated Raman intensity.

Discussion
Massonneau and May hypothesized the carotenoid nature of the pigments in Echiniscus tardigrades on the basis of solubility tests and chemical reactions [15]. However, since their work in 1950, no other experimental proof has been reported to corroborate their proposition, which remained a well-grounded hypothesis. The Raman spectra reported in this paper definitely decide this issue, unambiguously identifying these pigments as carotenoids having on average a main carbon chain with 11 conjugated C = C bonds in an ''all-trans'' configuration. Raman spectroscopy alone, however, is hardly capable of unambiguously identifying individual types of carotenoids occurring in biological samples [20,26]. This difficulty mainly arises from the fact that carotenoids, in spite of the variety of terminal substituents characterising the individual carotenoid species, all share the same fundamental structure of conjugated C = C double bonds, which is responsible for the intense resonant Raman spectra. This holds true also for carotenoids present in E. blumi, for which no information other than the average number of conjugated C = C bonds and their conformation can be directly inferred from Raman data. The small intensity standard deviation for the Raman maps of E. blumi, however, yields information about the chemical homogeneity of carotenoid species present in this tardigrade species, suggesting the occurrence of carotenoids all having similar molecular structures. A more complete picture of the number and types of carotenoids present in E. blumi could perhaps be achieved by using several other wavelengths for Raman excitation (e.g. 413 nm and 488 nm) [27]. Although a multi-wavelength mapping analysis on the same animal could present some difficulties (the animal must be kept alive and still under the microscope for tens of minutes), this approach is definitely worth further investigation.
Since the Raman data presented clearly demonstrate the dietary origin of carotenoids in tardigrades, further information on the nature of the carotenoid pigments can be indirectly achieved from the analysis of the moss constituting the food of these animals. In particular, the identification of Grimmia moss as the source of carotenoids allows us to put forward specific hypotheses about the number and type of carotenoid species detected in the E. blumi specimens examined. Lutein, zeaxanthin, b-cryptoxanthin and bcarotene (Fig. S5 in SI) are the most commonly occurring carotenoids in mosses, and in particular they constitute up to ,90% of the total carotenoids in Grimmia species [28]. In general, the position of the n 1 band in Raman spectra of carotenoids correlates well with the number of the conjugated double C = C bonds in their structure [20,22,23]; the average carotenoid spectra from E. blumi, and in particular spectra from the animal body cavity, well match the spectra of b-carotene, zeaxanthin and bcryptoxanthin, all having 11 conjugated C = C bonds, but not that of lutein, which only has 10 [22,27]. Moreover, previously reported chemical reactivity data of Echiniscus pigments [15] suggest that they do not include oxygen in their structure, leaving b-carotene as the most likely candidate for the pigmentation in the tardigrade of the genus Echiniscus.
Interestingly, the differences between the spectra of carotenoids localized in the gut and those in the body cavity are consistent with the presence of moss cellular material in the tardigrade gut. Such spectral differences could be explained considering a selective uptake of specific moss carotenoids by the tardigrade, as it has been reported for other animals [29,30]. Indeed, Raman spectra from moss (both intact leaves or moss cellular material inside the tardigrade gut) must include spectral contributions from all its major carotenoids constituents, including lutein, whose n 1 band occurs at Raman shifts higher than 1521 cm 21 [22,27]. Thus, the slight up-shift in the n 1 band position for those spectra collected from the gut with respect to spectra from the body cavity could be attributed to lutein, which is likely to be present in the moss cellular material inside the gut but might be less abundant or absent in the other organs of the tardigrade.
In general, the dietary origin of pigments in E. blumi also well explains the observed inter-individual variability in the overall carotenoid content, which may depend both on the quantity of food ingested and the time passed since the last food intake, assuming that carotenoids inside the animals are consumed. Since carotenoids are much less concentrated in the moss than in the E. blumi body cavity, this tardigrade appears to get these pigments from moss and accumulate them in its body cavity.
In animals, carotenoids have a diversity of functions: pigmentation with signalling purposes, immunity response, vision and protection against oxidative damage, among others [31]. In coloured tardigrades, the function of pigments is disputed.
Because they absorb UV and violet-blue light, carotenoids can act as light filters protecting against this harmful high-energy solar radiation [31], which is an indirect cause of oxidative damage to DNA, lipid membranes, and other cell components, through inducing the generation of free radicals [31]. The carotenoids present in tardigrades might well serve this photo-protective function. The observation that most tardigrade pigmented species dwell in sun-exposed habitats, such as the glacier surface [11] or on mosses and lichens growing on rocks located at high altitude and latitude, well support this hypothesis.
In E. blumi, the same carotenoids found in the body cavity are present in the eye-spots. Although ultrastructural data about these visual organs in tardigrades are available, the chemical nature of their pigments was previously unknown [12]. With the unique exception of a carotenoid photoactive protein found in cyanobacteria [32], carotenoids are not light receptors, but are nonetheless present in the eyes of several animals, as well as in humans, affording protection of photoreceptor cells against damaging radiation or contributing to obtain information regarding direction of the light [31]. Also the occurrence of carotenoids in the yolk of E. blumi eggs is not surprising, since these molecules are frequently found in high concentrations in eggs and in reproductive tissues of both vertebrates and invertebrates [31]. For example, the yolk of the eggs deposited above waterline by apple snails are supplied with adequate amount of carotenoids with photo-protective and antioxidant functions, to allow the development of embryos under harsh conditions such as desiccation and solar radiation exposure [33]. In both E. blumi eyespots and eggs, carotenoids are likely to have photo-protective and antioxidant function as well. On the other hand, in view of the ultrastructural data available about tardigrades eye-spots and of the presence of the same carotenoids throughout the E. blumi body, we tend to exclude any role of these pigments as light receptors in this species.
Because of their well-known antioxidant functions, the carotenoids accumulated in the body cavity might help to protect the tardigrade from the dehydration-induced oxidative stress which occurs during the entering and remaining in anhydrobiosis [34,35]. Organisms undergoing dehydration are known to suffer oxidative damage to their lipids, proteins and DNA, and tolerance to drying is correlated with an increase in their antioxidant potential [34,36]. In E. blumi, the accumulation of free-radical scavengers such as carotenoids is likely to be one of the possible strategies, together with other mechanisms that include enzymes (e.g. peroxidases, catalases, superoxide dismutases) and antioxidants (e.g. glutathione) [36,37] to fight the oxidative stress involved with anhydrobiosis. We simulated oxidative stress due to dehydration by H 2 O 2 treatment, and then we tested if Raman micro-spectroscopy was able to measure variations in carotenoid content in living tardigrades. By measuring in vivo the radical quenching antioxidant action of carotenoids upon treatment with H 2 O 2 , Raman micro-spectroscopy shows how pigments are consumed in tardigrades exposed to oxidative stress. Such a method to monitor carotenoid variations could thus be used to further investigate in vivo the role of such molecules as antioxidants in tardigrades during real dehydration or other oxidative stressrelated processes such as exposure to solar radiations.

Conclusions
The method we propose, based on Raman imaging of living tardigrades, proves to be a reliable method for the investigation of the nature and function of pigments in tardigrades. It demonstrates that the pigments observed in the body (including eyes) and eggs of E. blumi are carotenoids (presumably b-carotene) which tardigrades obtain from the moss they feed on. The same spectroscopic method could be used to study other tardigrade species as well, extending our knowledge on the different kinds of pigments found in these animals in relation with their habitat and diet.
Moreover, we propose a role for carotenoids in tardigrades as scavengers for reactive oxygen species forming during exposition to solar radiations and/or dehydration processes. We tested this hypothesis in vivo by simulating the oxidative stress with a H 2 O 2 treatment and measuring the decrease in carotenoid content.
Considering the lack of methods to study directly the antioxidant function of carotenoids in vivo [31], pigmented tardigrades, investigated with Raman imaging, could be used as model organisms for this purpose. Figure S1 Average Raman spectra (black) together with the intensity standard deviation (grey) from a leaf of the moss G. orbicularis and from a tardigrade E. blumi. The averages and standard deviation were calculated out of a set of 300 spectra for each sample, collected in the same experimental conditions (i.e. acquisition times, microscope objective, laser power). (TIF)