Tolerance to Anhydrobiotic Conditions Among Two Coexisting Tardigrade Species Differing in Life Strategies

© 2021 Academia Sinica, Taiwan Open Access Tolerance to Anhydrobiotic Conditions Among Two Coexisting Tardigrade Species Differing in Life Strategies Milena Roszkowska1,2 , Bartłomiej Gołdyn3, Daria Wojciechowska2,4 , Jakub Z. Kosicki5, Edyta Fiałkowska6, Hanna Kmita2, and Łukasz Kaczmarek1,* 1Department of Animal Taxonomy and Ecology, Faculty of Biology, Adam Mickiewicz University, Poznań, Uniwersytetu Poznańskiego 6, 61-614 Poznań, Poland. *Correspondence: Tel: +48606384687. E-mail: [email protected] (Kaczmarek) 2Department of Bioenergetics, Institute of Molecular Biology and Biotechnology, Faculty of Biology, Adam Mickiewicz University, Poznań, Uniwersytetu Poznańskiego 6, 61-614 Poznań, Poland. E-mail: [email protected] (Roszkowska); [email protected] (Wojciechowska); [email protected] (Kmita) 3Department of General Zoology, Faculty of Biology, Adam Mickiewicz University, Poznań, Uniwersytetu Poznańskiego 6, 61-614 Poznań, Poland. E-mail: [email protected] (Gołdyn) 4Department of Macromolecular Physics, Faculty of Physics, Adam Mickiewicz University, Poznań, Uniwersytetu Poznańskiego 2, 61-614 Poznań, Poland. 5Department of Avian Biology and Ecology, Faculty of Biology, Adam Mickiewicz University, Poznań, Uniwersytetu Poznańskiego 6, 61-614 Poznań, Poland. E-mail: [email protected] (Kosicki) 6Institute of Environmental Sciences, Jagiellonian University, Gronostajowa 7, 30-387 Kraków, Poland. E-mail: [email protected] (Fiałkowska) Received 20 October 2020 / Accepted 6 October 2021 / Published 7 December 2021 Communicated by Benny K.K. Chan Water availability is one of the most important factors for terrestrial life. Terrestrial habitats may periodically become dry, which can be overcome by an organism’s capability to undergo anhydrobiosis. In animals, this phenomenon has been reported for invertebrates, with tardigrades being the best-known. However, different tardigrade species appear to significantly differ in their anhydrobiotic abilities. While several studies have addressed this issue, established experimental protocols for tardigrade dehydration differ both within and among species, leading to ambiguous results. Therefore, we apply unified conditions to estimate intraand interspecies differences in anhydrobiosis ability reflected by the return to active life. We analysed Milnesium inceptum and Ramazzottius subanomalus representing predatory and herbivorous species, respectively, and often co-occur in the same habitat. The results indicated that the carnivorous Mil. inceptum displays better anhydrobiosis survivability than the herbivorous Ram. subanomalus. This tendency to some degree coincides with the time of “waking up” since Mil. inceptum showed first movements and full activity of any first individual later than Ram. subanomalus. The movements of all individuals were however observed to be faster for Mil. inceptum. Differences between the experimental groups varying in anhydrobiosis length were also observed: the longer tun state duration, the more time was necessary to return to activity. Key words: Anhydrobiosis, Cryptobiosis, Milnesium inceptum, Ramazzottius subanomalus, Recovery. BACKGROUND Cryptobiosis is the state of an organism characterised by no visible signs of life (Keilin 1959) and is commonly described as an adaptation to fluctuating environmental conditions that allow organisms to survive when the environment becomes unsuitable for active life. The phenomenon originated independently Citation: Roszkowska M, Gołdyn B, Wojciechowska D, Kosicki JZ, Fiałkowska E, Kmita H, Kaczmarek Ł. 2021. Tolerance to anhydrobiotic conditions among two coexisting tardigrade species differing in life strategies. Zool Stud 60:74. doi:10.6620/ZS.2021.60-74. Zoological Studies 60:74 (2021) doi:10.6620/ZS.2021.60-74 1 © 2021 Academia Sinica, Taiwan several times in diverse phylogenetic lines comprising unicellular (e.g., bacteria and protists) as well as multicellular (e.g., lichens, liverworts, vascular plants, and invertebrate animals) organisms. From the wide spectrum of cryptobiosis types determined by a given extreme environmental factor (e.g., Rebecchi et al. 2007; Guidetti et al. 2011a; Møbjerg et al. 2011), the most prevalent form is anhydrobiosis (e.g., Wright 2001; Watanabe 2006; Rebecchi 2013; Kaczmarek et al. 2019). Anhydrobiosis denotes dehydration tolerance, which refers to the ability of an anhydrobiotic organism, to survive dehydration leading to the near-complete loss of body water. The function of this process as an adaptation strategy appears to be understandable in habitats of unstable conditions like mosses, lichens, small temporary ponds or puddles, cryoconite holes, or tree holes. In the case of multicellular organisms anhydrobiosis may occur during particular stages of development but rarely concerns adulthood (e.g., Watanabe 2006; Rebecchi et al. 2007; Guidetti et al. 2011a; Møbjerg et al. 2011; Rebecchi 2013; Wharton 2015). A dried organism may remain in this state for a long period of time (see e.g., Roszkowska et al. 2020) and may subsequently rehydrate, and resume active life when water once again becomes available. This is probably a costly strategy, because organisms need to use energy from storage cells (Reuner et al. 2010; Czerneková et al. 2018). What is more, the longer the specimen is in the dried state, the longer it takes to return to active life. Moreover, there is a potentially lethal outcome when the dehydration duration threshold is surpassed (Ramazzotti and Maucci 1983; Wright 1989; Westh and Ramløv 1991; Horikawa and Higashi 2004; Rebecchi et al. 2006 2009a; Schill and Hengherr 2018). Tardigrades (as well as nematodes and rotifers) are among the animals best known for their anhydrobiotic capacity and the rare ability to undergo anhydrobiosis at each life stage (e.g., Clegg 2001; Wright 2001; Watanabe 2006; Mali et al. 2010; Møbjerg et al. 2011; Wełnicz et al. 2011; Rebecchi 2013; Schill and Hengherr 2018; Kaczmarek et al. 2019). In the case of tardigrades, the anhydrobiosis process can be divided into three stages: dehydration, “tun” and rehydration. Notably, certain tardigrade species differ in their ability to undergo anhydrobiosis. It is known that aquatic tardigrades are in general less capable of anhydrobiosis than limno-terrestrial ones (e.g., Wright 1989; Rebecchi et al. 2007; Hygum et al. 2016) and the ability of limno-terrestrial tardigrades to enter anhydrobiosis is suggested to be crucial for the colonisation of habitats exposed to shorter and longer periods of drying (Guidetti et al. 2011b). Differences were also observed for species from the same habitat but representing different classes, i.e., Heterotardigrada and Eutardigrada (Rebecchi et al. 2006). Moreover, dehydration conditions and the duration of the tun stage are important for the rate and efficiency of returning to an active state, though the optimal conditions (e.g., air humidity and temperature) of dehydration and the tun stage duration may differ between species (e.g., Wright 1989; Rebecchi et al. 2007) as well as within the same species (e.g., Horikawa and Higashi 2004; Rebecchi et al. 2009a; Kondo et al. 2015). To date, experiments on anhydrobiosis performed under controlled conditions have been conducted for several species, namely Acutuncus antarcticus (Richters, 1904) (Giovannini et al. 2018), Bertolanius volubilis (Durante Pasa and Maucci, 1975) (Altiero et al. 2012), Hypsibius exemplaris Gąsiorek, Stec, Morek and Michalczyk, 2018 (formerly Hys. dujardini (Doyère, 1840)) (Horikawa et al. 2013; Kondo et al. 2015; Arakawa et al. 2016; Yoshida et al. 2017; Kuzmic et al. 2018), Milnesium tardigradum Doyère, 1840 (Hengherr et al. 2008; Mali et al. 2010; Schokraie et al. 2012; Wang et al. 2014), Paramacrobiotus richtersi (Murray, 1911) (Rebecchi et al. 2009b; Altiero et al. 2011), Ramazzottius oberhaeuseri (Doyère, 1840) (Jönsson et al. 2001; Faurby et al. 2008), Ram. varieornatus Bertolani and Kinchin, 1993 (Horikawa et al. 2013; Yoshida et al. 2017) and Richtersius coronifer (Richters, 1903) (Jönsson et al. 2001; Jönsson and Rebecchi 2002; Faurby et al. 2008; Halberg et al. 2013; Czerneková and Jönsson 2016). However, results of these various experiments on anhydrobiosis under controlled conditions are incompatible. This inconsistency is likely the result of the experimental protocols, where particular species differed in dehydration procedure details (e.g., humidity level and the method of its application) even when two species were studied simultaneously. This makes it difficult to identify general patterns. Moreover, the duration of the tun stage and the estimation of anhydrobiosis success were usually limited to calculations of survival ratio over a given period of time (most often after one or 24 hours). Therefore, it seems reasonable that an attempt to find a general rehydration pattern should be based on a comparison of different species with various life strategies from the same habitat (e.g., occupying different ecological microniches and trophic levels) kept under the same controlled conditions. Thus, we established a long-term experiment (up to 240 days) focused on determining the anhydrobiotic capabilities of two tardigrade species representing Eutardigrada under standardised laboratory conditions. The species included in the present study are Mil. inceptum Morek, Suzuki, Schill, Georgiev, Yankova, page 2 of 13Zoological Studies 60:74 (2021) © 2021 Academia Sinica, Taiwan Marley and Michalczyk, 2019 and Ram. subanomalus (Biserov, 1985) belonging to the families Milnesiidae and Hybsibiidae, respectively, and differing in some aspects of biology. Milnesium inceptum is an obligatory predatory species that feeds on rotifers, nematodes and other tardigrades, and lays smooth eggs in old exuviae, whereas Ram. subanomalus is a herbivorous species that feeds on algae and lays ornamented eggs directly in the environment. However, both species inhabit the same habitat (i.e., frequently drying mosses growing on concrete walls) and often co-occur. Considering the differences in feeding behaviour between these two species, we decided to check if these species would adopt different anhydrobiotic strategies. Namely, if either the predator species (Mil. inceptum) or the prey species (Ram. subanomalus) may recover to full activity faster in order to hunt more effectively or to avoid being eaten, respectively. MATERIALS AND METHODS Species and sample processing Milnesium inceptum (Apochela) and Ram. subanomalus (Parachela) were extracted from the same moss sample collected from a concrete wall in Poznań, Poland (52°24'15"N, 16°53'18"E; 87 m asl). The moss sample was placed into plastic beaker containing 250 ml of tap water. After 18 hours, the water-saturated moss was strongly shaken with tweezers and all plant particles were removed. Water with tardigrades was then poured into a 250 ml plastic cylinder. After 30 minutes, the upper portion of water (ca. 200 ml) was decanted and the remaining 50 ml was poured into Petri dishes. Tardigrades were then extracted under a stereomicroscope (Olympus SZ). Specimens were later segregated and only fully active, adult specimens of medium body length were selected for the anhydrobiosis experiment. Genus abbreviations follow Perry et al. (2019). Anhydrobiosis experiment The experiment was performed in Petri dishes 3.5 cm in diameter (experimental dishes hereafter) lined at the bottom with white filter paper. Each experimental dish was filled with 450 μl of distilled water and selected individuals (extracted from the environmental sample) were transferred to the dishes (seven specimens per dish) using an automatic pipette. The experimental dishes were then closed with a cover of vented Petri dish (ensuring air exchange) and placed into an environmental chamber (PolLab, Q-Cell 140, https://www.pol-lab.eu/en/), where they were desiccated and stored under controlled conditions of 40–50% humidity and 20°C. The drying process took 72 hours (this amount of time was necessary to form correctly looking tuns) and that point was considered the beginning of the experiment. To test anhydrobiotic capabilities, specimens of both species were rehydrated after seven different periods of anhydrobiosis (0-, 7-, 14-, 30-, 60-, 120-, 240-days; each constituting a separate experimental group). This means that in the experimental group designated as “0”, specimens were rehydrated directly after dehydration (i.e., 72 hours after transferring tardigrades to experimental dishes), whereas specimens from the experimental group “7” were rehydrated after 7 days of anhydrobiosis, and so forth. Rehydration was conducted by adding 3 ml of distilled water to each experimental dish (Fig. 1). Specimens were then transferred to small glass dishes in which they were observed for 10 hours (at room temperature, 24°C) under a stereomicroscope. Thereafter, the dishes were placed in environmental chamber (40–50% humidity and 20°C) overnight. The next morning, they were observed every 30 minutes until 24 hours following rehydration. During the observation period, the moment of the first movement and recovery to full activity were recorded. Specimens with observed full activity were removed from dishes and no longer observed. The first movement was defined as any visible sign of movement in the claws, legs, body, buccal tube, etc. The full activity was defined as coordinated movements of the body and legs (i.e., the start of crawling). All observations were completed after 24 hours and all not fully active specimens or those without any signs of movement (non-moving, that were interpreted as being dead) were counted. As previously stated, we prepared seven experimental groups differing in the duration of the anhydrobiotic period for both tardigrade species. For each group, we performed 10 repetitions (i.e., we prepared 10 experimental dishes with seven specimens in each of them). This means that one experimental group contained 70 tardigrade specimens of each species. In total, we used 70 specimens multiplied by each of the seven experimental groups for one tardigrade species (i.e., a total of 490 individuals of each species). We tested differences between the experimental groups and species in relation to six measures of activity: (i) the number of non-moving (dead) tardigrades (NM); (ii) the number of individuals that did not reach full activity until the end of the observation (NFA); (iii) the time required for the first movement of any first individual (FM) and the first movement of all individuals (FMA); and (iv) the time required for return page 3 of 13Zoological Studies 60:74 (2021) © 2021 Academia Sinica, Taiwan to full activity of the first (FA) and all individuals (FAA). Statistical analysis The distributions of all the recorded variables was far from normality, so we used nonparametric test to analyse all the data. To check for the presence of significant differences between Mil. inceptum and Ram. subanomalus in the overall ability to return to active life following the tun stage, observed values of the six measures of activity (NM, NFA, FM, FMA, FA, FAA) were compared between the species using Exact Wilcoxon-Mann-Whitney Test. Then, each species was tested for the influence of the time spent in anhydrobiosis (i.e., in the tun stage) on the values of these indices by comparing them between the experimental groups using Approximative KruskalWallis Test with 100,000 Monte Carlo replicates (Hájek et al. 1999; Hothorn et al. 2008). Dunn test for multiple comparisons adjusted with Holm’s correction was used as a post-hoc analysis to check for the differences between the particular pairs of experimental groups. Correlations between the measures of activity within each species as well as correlations between the values of indices and time spent in anhydrobiosis were calculated using Asymptotic Spearman Correlation Test with Holm’s correction. Relations between the duration of anhydrobiosis and measures of tardigrade activity were illustrated on graphs by applying locally estimated scatterplot smoothing (LOESS) using Local Polynomial Regression Fitting function (span = 0.6; degree = 1). All calculations were performed in R 4.0.2 (R Core Team 2020) under RStudio 1.3.1056 using the 'coin' package (Hothorn et al. 2008) and graphs were produced using the ggplot2 package (Wickham 2016). We considered p = 0.05 as the threshold determining statistical significance. The raw data used for all the calculations are presented in table S1. RESULTS Overall ability to return to active life The mean number of individuals without any sign of movement (NM) at the end of the experiment per each replicate sample for Ram. subanomalus was 0.80 (95% CL: 0.58–1.01), whereas the mean NM for Mil. inceptum was 0.26 (95% CL: 0.06–0.45). This translates to the mean percentage of surviving individuals being 88.6% in Ram. subanomalus and 96.3% in Mil. inceptum. The difference in anhydrobiosis survival between these species was significant (Z = -4.7877, p < 0.0001). Thus, under the tested conditions, the survival rate of Mil. inceptum was significantly higher than that of Ram. subanomalus. In both species the duration of anhydrobiosis (i.e., the tun stage) had no significant effect on the number of NM individuals (Ram. subanomalus: chi2 = 10.379, p = 0.105, Fig. 2B; Mil. inceptum: chi2 = 2.644, p-value = 0.8402, Fig. 2A). As a consequence, in both species there was no significant difference in any pair of experimental groups with respect to NM (see Table S2 for the details). The only value that was significantly correlated with NM was the number of not fully active (NFA) individuals (Ram. subanomalus: Z = 7.5677; p < 0.0001 and Mil. inceptum: Z = 6.3548; p < 0.0001; Table 1). The mean number of NFA individuals of Ram. Fig. 1. A simple schematic illustration of the experiments. page 4 of 13Zoological Studies 60:74 (2021) © 2021 Academia Sinica, Taiwan Table 1. Results of Asymptotic Spearman Correlation Test with Holm’s correction checking for relationships between the values of particular activity measures recorded during the presented experiments. Upper diagonal: p-values with Holm’s correction; lower diagonal: values of the Z statistic Ram. subanomalus FM FMA FA FAA NM NFA FM < 0.0001 < 0.0001 0.0036 1.0000 1.0000 FMA 5.2362 < 0.0001 < 0.0001 1.0000 1.0000 FA 5.1043 5.0566 < 0.0001 0.7968 0.7785 FAA 3.5381 5.0345 5.0785 1.0000 1.0000 NM -0.1343 -0.0868 1.5812 1.2780 < 0.0001 NFA 0.3155 0.9958 1.6580 1.0748 7.5677 Mil. inceptum FM FMA FA FAA NM NFA FM < 0.0001 < 0.0001 < 0.0001 1.0000 1.0000 FMA 5.6210 < 0.0001 < 0.0001 1.0000 1.0000 FA 6.7596 5.8638 < 0.0001 1.0000 1.0000 FAA 4.5807 5.6882 5.0422 1.0000 1.0000 NM 0.3162 0.2851 -0.0070 -0.7153 < 0.0001 NFA 0.3779 0.1117 0.5211 -0.3851 6.3547 FMfirst movement of any first individual, FMA - first movement of all individuals, FA - full activity of the first, FAA - full activity of all individuals, NM - non-moving (dead) tardigrades, NFA - not fully active/individuals that did not reach full activity. Fig. 2. Differences in the number of non-moving (NM) and not fully active (NFA) individuals between experimental groups representing increasing duration of the tun stage for Milnesium inceptum (A, C) and Ramazzottius subanomalus (B, D). The number of replicate samples for each group n = 10. page 5 of 13Zoological Studies 60:74 (2021) © 2021 Academia Sinica, Taiwan subanomalus was 1.04 (95% CL: 0.78–1.30), while that of Mil. inceptum was 0.29 (95% CL: 0.14–0.43), with the mean percentage of fully active (FA) individuals being 85.1 and 95.9, respectively. Again, the difference between the species was statistically significant and Mil. inceptum was characterized by a significantly lower ratio of NFA individuals (Z = -4.9422, p < 0.0001). However, there were no significant differences between the experimental groups in both species (Ram. subanomalus: chi2 = 6.7554, p = 0.3509, Fig. 2D; Mil. inceptum: chi2 = 5.2496, p-value = 0.5275, Fig. 2C); as well as between species (comparisons between the pairs of experimental groups: p > 0.4 in all the cases) (Table S2). This denotes that duration of anhydrobiosis (i.e., the tun stage) had no significant effect on the number of NFA individuals. The values of NFA specimens were not significantly correlated with any measured indices of the activity except for NM (see above and Table 1). The timing of the first movement As shown in table 2, the shortest time of the first movement of any first individual (FM) ranged from 2 to 24 minutes for Mil. inceptum (the experimental groups “0” and “240”, respectively) and from 2 to 18 minutes for Ram. subanomalus (the experimental groups “7” and “240”, respectively). According to the Wilcoxon test the difference between the species was significant at Z = 2.1214, p = 0.0337 and Mil. inceptum showed FM significantly later than Ram. subanomalus. Moreover, the time to FM differed significantly between the experimental groups for both species (Mil. inceptum: chi2 = 53.602, p < 0.0001, Fig. 3A; Ram. subanomalus: chi2 = 56.776, p < 0.0001, Fig. 3B). As indicated by the applied post-hoc analysis, for Mil. inceptum such differences were significant when the experimental group “240” was compared with the experimental groups “0”, “7” and “14”; the experimental group “120” with the experimental groups “0” and “7”; and the experimental groups “30” and “60” with the experimental group “0”. In Ram. subanomalus the differences were significant in the case of the following comparisons of the experimental groups: “240” vs “0”; “7”, “14” and “30”; “120” vs “0” and “7”; “7” vs “14” and “60” (See Table S2 for details). The shortest time required for all individuals to display the first movements (FMA) was identical for both species, i.e., 7 minutes and was recorded in the experimental group “0” for Mil. inceptum and in the experimental group “7” for Ram. subanomalus. The longest time to FMA for Mil. inceptum was observed in the experimental group “30” (127 minutes) and for Ram. subanomalus in the experimental group “120” (453 minutes) (Table 2). The differences in FMA Table 2. Descriptive statistics of the first movement and full activity time (EG’ experimental groups) Mil. inceptum Ram. subanomalus Mean (± SD) and range time of: EG FM (min.) FMA (min.) FM (min.) FMA (min.) 0 4.5 (± 1.0) (2–6) 11.3 (± 4.7) (7–20) 4.3 (± 0.9) (3–6) 25.6 (± 23.9) (9–81) 7 5.3 (± 1.3) (3–7) 22.4 (± 14.4) (8–56) 3.4 (± 0.6) (2–4) 21.1 (± 23.5) (7–84) 14 5.8 (± 1.2) (4–8) 11.1 (± 3.4) (8–18) 6.1 (± 1.1) (5–8) 53.3 (± 65.3) (12–226) 30 7.5 (± 1.3) (5–9) 37.5 (± 33.7) (12–127) 5.3 (± 0.8) (4–6) 25.7 (± 29.8) (9–108) 60 8.0 (± 1.1) (6–10) 28.2 (± 15.3) (16–70) 6.7 (± 1.7) (5–10) 42.2 (± 40.8) (17–152) 120 8.4 (± 1.5) (7–12) 19.5 (± 4.8) (13–29) 9.3 (± 2.1) (7–14) 153.4 (± 129.6) (20–453) 240 18.9 (± 3.6) (13–24) 59.7 (± 29.3) (25–121) 12.7 (± 2.4) (10–18) 132.3 (± 103.9) (44–331) Mean (± SD) and range time of: EG FA (min.) FAA (min.) FA (min.) FAA (min.) 0 10.3 (± 3.6) (7–20) 11.3 (± 4.7) (7–20) 13.2 (± 5.3) (6–25) 110.7 (± 254.7) (16–835) 7 14.3 (± 3.9) (9–21) 350.6 (± 60.2) (31–1126) 15.2 (± 17.2) (8–64) 135.3 (± 307.0) (15–1003) 14 16.1 (± 4.2) (12–25) 121.2 (± 277.9) (23–912) 17.5 (± 6.4) (12–33) 296.7 (± 380.0) (23–1089) 30 22.2 (± 5.6) (16–30) 174.8 (± 263.5) (27–903) 12.2 (± 3.2) (9–18) 106.6 (± 137.8) (12–373) 60 19.2 (± 4.7) (13–27) 82.4 (± 34.2) (43–142) 13.9 (± 1.9) (11–17) 67.4 (± 51.5) (14–195) 120 20.4 (± 4.9) (13–29) 68.2 (± 29.1) (33–112) 18.9 (± 4.9) (15–31) 225.9 (± 228.7) (37 –849) 240 41.5 (± 10.9) (28–57) 374.9 (± 373.0) (69–1046) 27.0 (± 6.4) (18–36) 138.5 (± 59.7) (77–271) FM - first movement of any first individual, FMA - first movement of all individuals, FA - full activity of the first, FAA - full activity of all individuals, NM - non-moving (dead) tardigrades, NFA - not fully active/individuals that did not reach full activity. page 6 of 13Zoological Studies 60:74 (2021) © 2021 Academia Sinica, Taiwan were significant when the species were compared (Z = -2.5813, p = 0.0096) and Mil. inceptum showed significantly faster FMA. Moreover, based on the applied tests, differences between FMA were also significant when the particular experimental groups within each species were analysed (Mil. inceptum: chi2 = 42.689, p < 0.0001, Fig. 3C, Ram. subanomalus: chi2 = 33.476, p < 0.0001, Fig. 3D). According to the Dunn test, in Mil. inceptum such differences were significant when the experimental group “240” was compared with experimental groups “0”, “7”, “14” and “120”; experimental group “14” with the experimental groups “30” and “60” days; and the experimental groups “30” and “60” with the experimental group “0”. In Ram. subanomalus the differences were significant in cases of the following comparisons of the experimental groups: “240” vs “0”, “7” and “30”, “120” vs “0”, “7” and “30” (See Table S2 for details). The timing of the return to full activity The shortest time to return to full activity of any first individual (FA) ranged from 7 to 57 minutes for Mil. inceptum (the experimental groups “0” and Fig. 3. A time from the start of rehydration to the first movement (FM) of any first individual and all individuals (FMA) in the experimental groups representing increasing duration of the tun stage for Milnesium inceptum (A, C) and Ramazzottius subanomalus (B, D). The number of replicate samples for each group n = 10. page 7 of 13Zoological Studies 60:74 (2021) © 2021 Academia Sinica, Taiwan “240”, respectively) and from 6 to 64 minutes for Ram. subanomalus (the experimental groups “0” and “7”, respectively) (Table 2). Despite the similarities, Wilcoxon test showed that the difference in FA was statistically significant at Z = 2.4853, p = 0.0127. As in the case of FM, Mil. inceptum achieved FA significantly later than Ram. subanomalus. Moreover, the values of FA differed significantly between the experimental groups (Mil. inceptum: chi2 = 45.81, p < 0.0001, Fig. 4A; Ram. subanomalus chi2 = 36.458, p < 0.0001, Fig. 4B). According to the applied post-hoc analysis, in Mil. inceptum such differences were significant when comparing the experimental group “240” with the experimental groups “0”, “7”, “14” and “60” and the experimental groups “30”, “60” or “120” with the experimental group “0”. In Ram. subanomalus the differences were significant in cases of the following comparisons of the experimental groups: “240” vs “0”, “7”, “30” and ”60” and “120” vs “7” days (See Table S2 for details). The fastest return to full activity by all individuals (FAA) was observed after 7 minutes for Mil. inceptum in the experimental group “0” and after 12 minutes for Ram. subanomalus in the experimental group “30”. The longest time to FAA for Mil. inceptum was observed after 1126 minutes (the experimental group “7”) and after 1089 minutes for Ram. subanomalus (the experimental group “14”) (Table 2). The difference Fig. 4. A time from the start of rehydration to the full activity (FA) of any first individual and all individuals (FAA) in the experimental groups representing increasing duration of the tun stage for Milnesium inceptum (A, C) and Ramazzottius subanomalus (B, D). The number of replicate samples for each group n = 10. page 8 of 13Zoological Studies 60:74 (2021) © 2021 Academia Sinica, Taiwan in FAA between the species was not significant (Z = -0.23549, p-value = 0.8152); however, the particular experimental groups differed significantly both in Mil. inceptum (chi2 = 40.146, p < 0.0001, Fig. 4C) and Ram. subanomalus (chi2 = 20.958, p-value = 0.0007, Fig. 4D). According to the applied posthoc analysis, in Mil. inceptum such differences were significant when the experimental group “240” was compared with the experimental groups “0” and “14”, and the experimental groups “7”, “30”, “60” or “120” with the experimental group “0”. In Ram. subanomalus the differences were significant when timing of FAA was compared between the groups spending 120 vs 0 and 120 vs 7 days in anhydrobiosis (See Table S2 for details). All measures of timing of FM and FA were highly intercorrelated (p < 0.0001 in all but one case, see Table 1), but none of them were correlated with NM or NFA. Relation of activity indices with time spent in anhydrobiosis In both species, FM, FMA, FA and FAA were highly positively correlated with the time spent in anhydrobiosis. The longer anhydrobiosis lasted, the more time was needed to observe the FM (Mil. inceptum: 7.1643, p < 0.0001; Ram. subanomalus: Z = 7.000, p < 0.0001), FMA (repetitively: Z = 5.0326, p < 0.0001 and Z = 5.0956, p < 0.0001), FA (repetitively: Z = 6.2675, p < 0.0001 and Z = 4.7077, p < 0.0001) and FAA (repetitively: Z = 4.5642, p < 0.0001 and Z = 3.4128, p < 0.0001). However, as shown by the LOESS curves (Figs. 5–6), the relation was almost always far from linear—except for FM and FA in Ram. subanomalus. The FMA and FAA in that species showed slight declines after 120 days of anhydrobiosis (Fig. 5A–D). The indices of the first movements (FM, FMA) and full activity (FA, FAA) in Mil. inceptum flattened after ca. 30 days and showed another steep increase after 120 days (Fig. 6A–D). The NM and NFA were not correlated with the time spent in anhydrobiosis in either species (NM: Mil. inceptum: Z = -0.1032 p > 0.9999; Ram. subanomalus: Z = -1.2743, p = 0.4051; NFA: Mil. inceptum: Z = -0.1725, p > 0.9999; Ram. subanomalus: Z = -0.8293, p = 0.4070). DISCUSSION Here we present results of a long-term anhydrobiosis Fig. 5. Relationship between the measured activity indices and time spent during the tun stage for Ramazzottius subanomalus: (A) time to first movement of any first individuals (FM); (B) time to first movement of all individuals (FAA); (C) time to full activity of any first individual (FA); (D) time to full activity of all individuals (FAA). Curves ± SE were fitted with Local Polynomial Regression Fitting (LOESS). page 9 of 13Zoological Studies 60:74 (2021)