New Records of Antarctic Tardigrada with Comments on Interpopulation Variability of the Paramacrobiotus fairbanksi Schill, Förster, Dandekar and Wolf, 2010

: Studies on Antarctic tardigrades started at the beginning of the twentieth century and have progressed very slowly and ca. 75 tardigrade species are known from this region. Paramacrobiotus fairbanksi was described from USA based on genetic markers and later reported from Italy, Poland, and Spain. The “everything is everywhere” hypothesis suggests that microscopic organisms have speciﬁc features which help them to inhabit most of environments and due to this they can be considered cosmopolitan. In the present paper, we report eight tardigrade taxa from Antarctic, including the ﬁrst report of Pam. fairbanksi from Southern Hemisphere, which could suggest that the “everything is everywhere” hypothesis could be true, at least for some tardigrade species. Moreover, we also genetically and morphologically compare a few di ﬀ erent populations of Pam. fairbanksi . The p-distances between COI haplotypes of all sequenced Pam. fairbanksi populations from Antarctica, Italy, Spain, USA and Poland ranged from 0.002% to 0.005%. In the case of COI polymorphism analyses, only one haplotype was observed in populations from Antarctica, USA and Poland, two haplotypes were found in population from Spain, and six haplotypes were observed in population from Italy. We also found some statistically signiﬁcant morphometrical di ﬀ erences between the populations of Pam. fairbanksi from di ﬀ erent regions and designed a new speciﬁc primers for Paramacrobiotus taxa.


Introduction
Antarctic is the coldest and most extreme region on the planet with the lack of surface liquid freshwater. Continental Antarctic with an area of ca. 14 million square kilometers has only ca. 0.3% ice-free lands. The entire continent is divided into three main units, i.e., East Antarctica, West Antarctica, and the Antarctic Peninsula [1]. Our knowledge of micro-invertebrates in this region is patchy and fragmented, because most studies have focused on easily accessible regions that are close to research stations [2]. Many of Antarctic invertebrates are endemic, some of them have a continental distribution, others are restricted only to maritime Antarctic and only a few can be characterized as pan-Antarctic species such as eutardigrade Acutuncus antarcticus (Richters [3]) e.g., [4,5].
According to the "everything is everywhere" hypothesis it is assumed that microorganisms and small invertebrates have cosmopolitan ranges. This hypothesis was introduced at the beginning of the 20th century [31,32]. Due to the presence of certain features in microscopic organisms that are not found in larger animals, they can inhabit most environments and are considered to be cosmopolitan species. Among the most frequently mentioned adaptations are the possibility of quite ease passive dispersion (through wind, sea currents, water, other animals, etc.), the presence of very resistant spore stages (which include, for example, cysts, eggs, or cryptobiotic individuals) that help to survive extreme conditions and maintain in any environment, as well as the presence of asexual or parthenogenetic reproduction that allows a rapid increase in the number of individuals e.g., [32][33][34][35][36]. However, the cosmopolitanism of microscopic organisms is increasingly discussed and undermined also within tardigrades e.g., [37][38][39]. Gąsiorek et al. [40] provided a first Australasian, molecularly verified record of a heterotardigrade species Echiniscus testudo (Doyère [41]) originally described from Holarctic (Europe, France) which could suggest agreement with the hypothesis "everything is everywhere" for tardigrades.
In this paper we report eight tardigrade taxa from Antarctic including the first report of Pam. fairbanksi from the Southern Hemisphere which is another strong argument suggesting that "everything is everywhere" hypothesis could be true, at least for some tardigrade species. Moreover, we also genetically and morphologically compare a few different populations of Pam. fairbanksi, and designed a new specific primers for Paramacrobiotus taxa.

Sample Processing
Eleven moss-lichen and algae samples from soil, rocks, and freshwater sediments were collected during the 5th (2012-2013), 7th (2014-2015) and 9th (2016-2017) Belarusian Antarctic Expeditions (BAE) near the Belarusian Antarctic Station "Vechernia" and Russian Station "Progress" (for details see Table 1). The samples were packed in paper envelopes, dried at a room temperature and delivered to the laboratory at the Faculty of Biology, Adam Mickiewicz University, Poznań, Poland. Tardigrades were extracted from the samples and studied following the protocol of Stec et al. [42].

Microscopy, Morphometrics, and Morphological Nomenclature
Specimens for light microscopy were mounted on microscope slides in a small drop of Hoyer's medium, prepared according to Ramazzotti and Maucci [10] as in the English translation by Beasley [43], and secured with a cover slip. Then, the slides were placed in an incubator and dried for two days at ca. 60 • C. Dried slides were sealed with a transparent nail polish and examined under an Olympus BX41.
All measurements are given in micrometers [µm]. Structures were measured only if their orientation was suitable. Body length was measured from the anterior extremity to the end of the body, excluding the hind legs. The types of bucco-pharyngeal apparatuses and claws were classified according to Pilato and Binda [44]. Buccal tubes, claws, and eggs were measured according to Kaczmarek and Michalczyk [45]. Macroplacoid length sequence was given according to Kaczmarek et al. [46]. The pt ratio is the ratio of the length of a given structure to the length of the buccal tube, expressed as a percentage [47]. The pt values are always provided in italics.
The map and figures were assembled in Corel Photo-Paint 2017.

Comparative Material
Species were identified based on original descriptions [3,15,25,51,52] and later partial redescriptions [53,54]. The type material (paratypes) of Dastychius improvisus (Dastych [52]) and Mesobiotus blocki (Dastych [52]) from the collection of Zoologisches Museum of Hamburg (Germany) were examined. Specimens of Pam. fairbanksi from type locality and additional material from Italy from the Roberto Guidetii collection (University of Modena and Reggio Emilia, Modena, Italy) were also used in analysis, altogether with specimens of Pam. fairbanksi from the laboratory culture conducted at the Department of Animal Taxonomy Gąsiorek) were cultured in small Petri dishes in spring water mixed with distilled water (3:1) with the rotifers and nematodes added as a food ad libitum. Cultures were kept in an environmental chamber with a temperature at 18 • C and in the darkness.

DNA Extraction and Genotyping
Four specimens and 31 eggs of Pam. fairbanksi were identified (partially based on [25]) in vivo using light microscopy (LM) prior to DNA extraction for genotyping analysis. Eggs were divided into three samples. Ten eggs were placed in two of them and eleven eggs in the last sample. Genomic DNA of adult individuals was extracted using a Chelex®100 resin (Bio-Rad, Warsaw, Poland) extraction Diversity 2020, 12, 108 4 of 22 method described by Casquet et al. [55] with a modified protocol included in Stec et al. [42]. In turn, genomic DNA from eggs was extracted using silica membranes from the commercial Genomic Mini kit (A&A Biotechnology, Gdynia, Poland).
All the polymerase chain reactions were carried out in 20 µL volume containing 0.8× JumpStart Taq ReadyMix (1 U of JumpStart Taq DNA polymerase, 4 mM Tris-HCl (pH 8.3), 20 mM KCl, 0.6 mM MgCl 2 , 0.08 mM of dNTP; Sigma-Aldrich, Darmstadt, Germany), 0.4 µM of proper forward and reverse primers and ca. 1 ng of DNA. The PCR reactions were performed in a BiometraTProfessional thermocycler. The PCR cycling profile to amplify the 18S rRNA gene fragment was as follows: initial denaturation at 95 • C for 5 min followed by 40 cycles of 95 • C for 30 s, 55 • C for 40 s, and 72 • C for 1 min 30 s, and ending with 72 • C for 4 min. The 28S rRNA gene fragment was amplified under conditions described by Mironov et al. [57]. The PCR protocol for the amplification of the COI gene fragment using HCO2198 and LCO1490 primers was as follows: initial denaturation at 94 • C for 5 min followed by 35 cycles of 94 • C for 30 s, 45 • C for 1 min, and 72 • C for 1 min, and ending with 72 • C for 5 min. In turn, amplification of the COI gene fragment using the newly designed degenerate primers was performed under the following conditions: initial denaturation at 95 • C for 5 min followed by 44 cycles of 95 • C for 30 s, 51.5 • C for 40 s, and 72 • C for 1 min, and ending with 72 • C for 5 min.
Amplified PCR products were separated in 1% agarose gel in a 1× SB buffer and visualized under UV light using Midori Green Advance DNA Stain (Genetics). Exonuclease I (20 U/µL, Thermo Scientific) and alkaline phosphatase FastAP (1 U/µL, Thermo Scientific, Waltham, MA, USA) were applied to clean the PCR products. Sequencing directly in both directions was performed using the BigDyeTM terminator cycle sequencing method and ABI Prism 3130xl genetic analyzer (Life Technologies, Carlsbad, CA, USA).

Comparative Molecular Analysis
BLAST (Basic Local Alignment Search Tool; [60]) searches at NCBI were performed to verify the identity and homology of the amplified nuclear and mitochondrial barcode sequences. All sequences obtained in our study and described by Guidetti et al. [28] as originated from Pam. fairbanksi, were then aligned using the ClustalW Multiple Alignment tool [61] implemented in BioEdit v. 7.2.5 [62]. Alignment sequences were trimmed to 822 and 574 bp for 18S rRNA and COI barcodes (only obtained from eggs), respectively. Calculation for the uncorrected pairwise distances (p-distances) was performed for COI sequences using the MEGA X.
The COI sequences were translated into amino acid sequences using the EMBOSS-TRANSEQ application [63,64] to check for indels and internal stop codons. The COI haplotypes were retrieved using DnaSP v5.10.01 software [65].

Statistical Analysis
Statistical significance of the differences in morphometrics between the studied populations of Pam. fairbanksi was analyzed using the analysis of variance (ANOVA) test with Bonferroni correction for multiple comparisons. Measurements of the body and buccal tube length (BL and BTL) were used as dependent and populations as grouping variable. Shapiro test was used to check for normality of distribution in residuals resulting from these models.
Apart from the BL and BTL measurements, we also analyzed the overall differences in morphometrics of other measured traits: stylet support insertion points (SSIP), external width of buccal tube (BTEW), and placoids (M1, macroplacoid 1; M2, macroplacoid 2; M3, macroplacoid 3; Mi, microplacoid; MR, macroplacoid row; PR, placoid row). We performed principal component analysis (PCA) on these measurements (the PCA input data matrix was tardigrade specimens × characters length), and then used the resulting first component as the dependent variable in ANOVA, with populations as grouping variable and Bonferroni correction. Later, to compare the differences in proportions of the same characters, the same analyses were performed for the pt ratios of exact characters. Measurements of 44 tardigrade specimens were used in the analyses, including 16 Antarctic, 15 Polish, 9 Alaskan, and 4 Italian. In case some of the measurements were missing they were replaced by mean values.
ANOVA was also used to test for the differences in egg morphology between the populations studied. To do so, for each obtained egg its bare diameter (EBD), full diameter (EFD), and processes height (PH) were measured and used as dependent variable in the models, again with population as the variable determining compared groups and Bonferroni correction. Measurements of 49 tardigrade eggs were used in the analyses, including 18 Alaskan, 16 Antarctic, and 15 Polish. In case some of the measurements were missing they were replaced by mean values. All the analyses were performed in R 3.4.4. We considered p < 0.05 as a level determining statistical significance; in case of post hoc tests only statistically significant results are presented.

Morphometric Comparison of Different Populations of the Pam. fairbanksi
Significant differences in BL between the studied populations (Tables 3-10
The pt of BL also differed between the populations (df = 3, F = 5.371, p = 0.0034, n = 44), with specimens from Alaska having significantly lower pt of BL than those from Antarctic (p = 0.0029, Figure 8). Similar differences (p = 0.0021) between the two populations were found when first component of PCA for pt of all the measurements was analyzed (df = 3, F = 6.332, p = 0.0013, n = 44), with additional significant differences between Alaskan and Italian populations (p = 0.019, Figure 9). ANOVA performed for pt of SSIP showed no significant differences between the populations (df = 3, F = 1.196, p = 0.324, n = 44). Table 3. Measurements (in µm) and pt values of selected morphological structures of individuals of the Antarctic population of Paramacrobiotus fairbanksi Schill, Förster, Dandekar and Wolf [25] mounted in Hoyer's medium (N, number of specimens/structures measured; RANGE refers to the smallest and the largest structure among all measured specimens; SD, standard deviation; pt, ratio of the length of a given structure to the length of the buccal tube expressed as a percentage).

Genetic Comparison of Different Populations of the Pam. fairbanksi
All DNA barcode sequences of Pam. fairbanksi from Antarctic were obtained from DNA extracted from the eggs. In DNA extracted from adult individuals, signs of degradation were observed; however, newly designed degenerate primers to amplify COI marker allowed confirmation of molecular species identification. The COI sequences (GenBank accession numbers: MN964281-MN964282) were 639 bp long (using HCO2198 and LCO1490 primers and 392 bp long using designed The ANOVA performed on EBD measurements of eggs (df = 2, F = 18.72, p < 0.0001, n = 49) showed significant differences between all the populations, where the eggs in Polish population were significantly smaller than these from Alaska (p = 0.0197) and Antarctica (p < 0.0001) and the eggs from Alaskan population clearly smaller than those from Antarctic (p = 0.0033) ( Figure 10). Analysis on EFD values, however, showed no differences between measurements of Polish and Alaskan eggs (p = 1, n = 49, Figure 11), but eggs in Antarctic population were significantly bigger than these collected in Alaska (p = 0.0008) and Poland (p = 0.0002), the model was significant at p < 0.0001 (df = 2, F= 11.72). The differences between the studied populations were also visible in the size of PH (df = 2, F = 18.91, p < 0.0001, n = 44). Egg processes were longest in Polish population in comparison to both Antarctic (p = 0.0009) and Alaskan (p < 0.0001) populations. There was no significant difference between the length of egg processes in Alaskan and Antarctic populations (p = 0.1252, n = 49, Figure 12).

Genetic Comparison of Different Populations of the Pam. fairbanksi
All DNA barcode sequences of Pam. fairbanksi from Antarctic were obtained from DNA extracted from the eggs. In DNA extracted from adult individuals, signs of degradation were observed; however, newly designed degenerate primers to amplify COI marker allowed confirmation of molecular species identification. The COI sequences (GenBank accession numbers: MN964281-MN964282) were 639 bp long (using HCO2198 and LCO1490 primers and 392 bp long using designed ParCOIF and ParCOIR primers) and represented only one haplotype. No insertions, deletions, or stop codons were identified. The translation was successfully carried out with the invertebrate mitochondrial codon table and the -2th reading frame. The p-distances between COI haplotypes of all sequenced Pam. fairbanksi populations from Antarctica, Italy, Spain, USA (Alaska), and Poland ranged from 0.002% to 0.005% (an average distance of 0.003%). In case of COI polymorphism analyses, only one haplotype was observed in populations from Antarctica, USA (Alaska), and Poland, two haplotypes were found in population from Spain, and six haplotypes were observed in population from Italy (for details see Table 11). Table 11. Estimates of evolutionary divergence between COI haplotypes (H1-H10 *) based on p-distances (below the diagonal). Standard error estimates are shown above the diagonal and were obtained by a bootstrap procedure (500 replicates). The 18S rRNA sequences (GenBank accession numbers: MN960302-MN960304) were 1717 to 1725 bp long and one nucleotide substitution was found. Comparing with available in GenBank sequences of Pam. fairbanksi (from Italy and Alaska, alignment sequences were trimmed to 822 bp, see [28]) also showed one nucleotide substitution.

H1
The 28S rRNA sequences (GenBank accession numbers: MN960306-MN960307) were 796 to 800 bp long and one nucleotide substitution was found. We obtained 28S rRNA sequences for Pam. fairbanksi for the very first time, so that no sequences of this species were available for comparison.

Taxa
Most of species found in the present study are endemic to the Antarctic region. Barbaria pseudowendti, Das. improvisus, and Mes. blocki are known only from Enderby Land, East Antarctica [12]. Acutuncus antarcticus is a pan-Antarctic species, however, as it is a parthenogenetic species some populations are isolated and show quite large genetic distance which can suggest a process of speciation [5].
Hebesuncus mollispinus has quite recently been described as a species from Charcot Island, close to Antarctic Peninsula (West Antarctica) [15]. It was also reported from Alamode and Terra Firma Islands (close to Antarctic Peninsula), South Georgia (sub-Antarctica), and Argentinian Andes [15,66]. The present report is the first in East Antarctica. Such geographical range of this species suggests their wide distribution in the Antarctic region and also inhabiting glacier regions in South America.
The taxonomic status of Mil. quadrifidum is unclear, especially that their original description is very scanty [51] and lacks a detailed description of animal morphology. In the past it was synonymized with nominal Mil. tar. tardigradum Doyère [41], but recently Morek et al. [67] proposed it as a valid species being the only member of the genus Milnesium with four points on all secondary branches of claws. However, the same authors also suggested that this species needs a detailed redescription. Only recently, Kaczmarek et al. [68] described another species with four points on secondary branches of claws IV, i.e., Mil. wrightae Kaczmarek, Grobys, Kulpa, Bartylak, Kmita, Kepel, Kepel and Roszkowska [68]. Dastych [52] reported Mil. tar. tardigradum from King George Island and Enderby Land with many claw aberrations (i.e., different number of points on secondary branches of claws, even in the same animal). As the number of points on claws is rather constant (but may change during ontogenesis) and species-specific character [67,69,70], it should be considered that specimens of Mil. tar. tardigradum reported by Dastych [52] could possibly belong to a different species. In this situation it is impossible to determine a detailed morphology of Mil. quadrifidum or their distribution. The specimens found in the present study correspond to the original description of Mil. quadrifidum, but taking into consideration all the taxonomic problems addressed for this species, their presence in Antarctica should be confirmed after detailed redescription of this species from the type locality.
Paramacrobiotus fairbanksi was described from Alaska (USA) and later reported from Italy, Poland and Spain [25][26][27][28], and now we have provided the next record from Antarctica. This species perfectly fits the model taxon for the "everything is everywhere" hypothesis, i.e., is microscopic, easy to disperse, is able to cryptobiosis and it is parthenogenetic. However, before the final conclusion that Pam. fairbanksi is a true cosmopolitan species, its presence should also be confirmed on other continents.
Due to the lack of eggs, determination of the specimens of the genus Mesobiotus from two samples was not possible. It is of course highly probable that these specimens belong to Mes. blocki which was also found in the present study.

Morphometric Comparison of Different Populations of the Pam. fairbanksi
It is clearly visible in Tables 3-10 and Figures 1-9 and supported by the statistical analyses that some small morphometric differences are present between the populations of Pam. fairbanksi from different regions. However, measurements of all structures generally overlap which rather clearly allows the identification of this species even without genetic markers. This suggests that, based on the morphology, all specimens, even collected in different regions, should be correctly classified if a proper number of specimens is examined. Nonetheless, specimens from the Polish population are the smallest and lay the smallest eggs, equipped, however, with the longest egg processes. All observations and measurements of the Polish and Alaskan population of Pam. fairbanksi are based on cultured specimens, whereas in other cases specimens were extracted from the wild. In general, in both cultured populations, the length of the body is the smallest which could be caused by the well-known phenomenon when cultured populations of various animal species are dwarfing as a result of suboptimal conditions, high culture densities, and inbreeding. In addition, the measurements of other morphological structures of adults in Polish population are also the smallest, whereas, in contrast, in Alaska they are similar to those from Antarctic and Italy. Therefore, the observed dissimilarities could well be a sign of conserved differences in the morphology of the populations as a result of ongoing speciation.
It is also visible that the eggs of the Antarctic population are clearly bigger and have rather small egg processes. At present, we suggest that the harsh conditions in Antarctica favor laying larger eggs, while in cultures the eggs are smaller because of the lack of such selective pressure. In contrast, it is quite strange that the number of processes on the egg circumference in the Alaskan population is higher (10 to 20 processes, based on data from the original description) than in other populations (11 to 12 processes, based on the own observations). However, this could be an artefact caused by differences in the methodology used. Summarizing, at present we should consider this species as effectively morphologically defined with small interpopulation differences.

Genetic Comparison of Different Populations of the Pam. fairbanksi
Paramacrobiotus fairbanksi has a wide distribution and until now was observed in Italia, Spain [26], Poland [27], and Alaska [25]. From each locality, COI sequences of Pams fairbanksi are available. In the case of nucleotide markers, the 18S rRNA sequences have been obtained from Italia, Alaska, and presented in our study from Antarctica. In turn, the 28S rRNA sequences have been obtained only from our population from Antarctica. As examined in this study, Pam. fairbanksi population from Antarctica is the most similar to the population from Poland, with the same COI haplotype (Figure 13). Molecular analysis based on 18S rRNA sequences indicated no differences between our population from Antarctica and populations from Italy, whereas those based on COI sequences revealed low genetic distances, i.e., 0.002 to 0.003. The genetic distances, based on COI markers among Pam. fairbanksi from Antarctica and Alaska was 0.003, and one nucleotide substitution was indicated in 18S rRNA sequences. On the basis of this molecular data for populations from very distant regions, with physical barriers to gene flow, we showed the low evolutionary rate of Pam. fairbanksi.
Paramacrobiotus fairbanksi has a wide distribution and until now was observed in Italia, Spain [26], Poland [27], and Alaska [25]. From each locality, COI sequences of Pams fairbanksi are available. In the case of nucleotide markers, the 18S rRNA sequences have been obtained from Italia, Alaska, and presented in our study from Antarctica. In turn, the 28S rRNA sequences have been obtained only from our population from Antarctica. As examined in this study, Pam. fairbanksi population from Antarctica is the most similar to the population from Poland, with the same COI haplotype ( Figure  10). Molecular analysis based on 18S rRNA sequences indicated no differences between our population from Antarctica and populations from Italy, whereas those based on COI sequences revealed low genetic distances, i.e., 0.002 to 0.003. The genetic distances, based on COI markers among Pam. fairbanksi from Antarctica and Alaska was 0.003, and one nucleotide substitution was indicated in 18S rRNA sequences. On the basis of this molecular data for populations from very distant regions, with physical barriers to gene flow, we showed the low evolutionary rate of Pam. fairbanksi.