**1. Introduction**

It is well documented for many groups of organisms that exposure to solar radiation might cause severe direct and indirect negative e ffects [1,2]. Planktonic species are exposed, to a di fferent extent, to solar ultraviolet radiation (UVR) which penetrates the water column. In subalpine lakes (i.e., located below the treeline), UVR is attenuated within the uppermost meters because chromophoric dissolved organic matter and phytoplankton absorb these short wavelengths [3,4]. To escape high UVR levels at the lake surface, some organisms such as zooplankton perform diel vertical migrations [5]. This adaptation is unknown for ciliates, and the species assemblage is believed to move along the water column and over the season according to food availability or water temperature [6–8]. Characteristically, during summer/autumn, mixotrophic ciliates that live in symbiosis with algal endosymbionts can be detected in the epilimnion of temperate lakes [6,9,10]. Such a mutualistic relationship between green algae and a ciliate host has di fferent advantages for both partners, namely, the ciliate receives nutrients from its partners and the algae are transported into sunlit areas, ensuring a positive photosynthetic balance. In addition, the algae receive shelter from chloroviruses [11–13]. Another putative advantage of the algal–ciliate relationship, but less known, is photoprotection against UVR [14–16].

The short wavelengths of the sunlight spectrum in the ultraviolet range (280–400 nm) potentially cause damage to the DNA and other cell targets. In particular, the absorption of ultraviolet-B (UV-B; 280–315 nm) and of ultraviolet-A (UV-A; 315–400 nm) radiation by DNA can damage its structure and can cause both mutagenic and lethal e ffects [17]. Consequently, organisms have evolved a variety of response mechanisms to prevent or repair damage from UVR, including physical avoidance by regulating their position in the water column, accumulation or synthesis of sunscreens (e.g., carotenoids and mycosporine-like amino acids), or repair of DNA damage (e.g., photoenzymatic repair and nucleotide excision repair [2,18–20]). Thus, UVR may not only have negative e ffects on organisms, but longer UV-A wavelengths and photosynthetically active radiation (PAR) can upregulate photoenzymatic repair (PER), where DNA damage is repaired with the enzyme photolyase. Alternatively, organisms may have a nucleotide excision repair mechanism ("dark repair"), where the damaged part of the DNA is removed and resynthesized [17,21]. PER and dark repair appear to be widespread among taxonomically diverse organisms and have been also identified in protists [17,22–24]. When all of these mechanisms are ine fficient, diverse negative e ffects are observed. In protists, damage by UVR is known to lead to reduced motility and retarded division as well as a reduction in growth rates [25–29]. However, from the few studies available, for example, on ciliates, it is known that both damage by UVR and the presence of photoprotective mechanisms are species-specific [10,14–16,22,23,30–35]. For example, in laboratory experiments, *Glaucoma* sp. and *Parauronema acutum* recovered under photoreactive radiation or in the dark, whereas a *Cyclidium* species did not [22,23]. In some mixotrophic ciliates, two photoprotective mechanisms provided by the algal symbionts have been identified, namely, the synthesis of sunscreen compounds (i.e., mycosporine-like amino acids (MAAs)) and the self-shading e ffect given by the formation of dense algal layers inside the cell that prevent UVR from reaching the sensitive nuclear material [14–16]. Generally, MAAs are detected in a variety of organisms and they e fficiently absorb UVR in the UV-B and the UV-A ranges between 309 and 362 nm [36,37]. These secondary metabolites are considered e fficient sunscreen compounds, and some of them also have antioxidant capacities as UVR is a source of oxidative stress [37]. In contrast to pigments, MAAs are colorless, water-soluble compounds and they are probably evenly distributed within the cytoplasm of an organism [38]. The biochemical pathways of their synthesis are still not ye<sup>t</sup> fully identified in all taxa [39,40]. Therefore, it remains unclear if ciliates themselves are able to produce sunscreens or if they can only acquire them from an algal partner or can only extract them from their diet [14,34]. The internal shading of the nuclear material in ciliates through dense symbiotic algal layers seems to be an additional e ffective photoprotective mechanism [10,15,16]. This is important because not all ciliate species contain MAAs, but a well-directed internal allocation of algal symbionts regulates the photoprotective e fficiency in non-planktonic ciliates such as *Paramecium bursaria* [15,16].

Our hypothesis in this study is that, in the UVR-flooded zone of a lake, mixotrophic ciliates are well-adapted to this natural stress factor. Thus, we tested three representative species of the planktonic ciliate assemblage, i.e., *Pelagodileptus trachelioides*, *Stokesia vernalis,* and *Vorticella chlorellata,* for their photoprotective and recovery strategies. In experiments with freshly collected individuals, we ran a series of laboratory experiments under artificial radiation. We first assessed the ciliates' general survival under exposure to the full solar radiation spectrum and to PAR only against a dark control (experiment 1). Second and only for *P. trachelioides*, we identified the wavelengths responsible for potential damage and/or mortality and exposed the ciliates to UVR by the exclusion of certain UV-B and UV-A wavelengths with a set of long-pass cuto ff filters (experiment 2). Third, we tested the availability of recovery strategies from UVR-induced impairments of the ciliates, including dark repair (all species) and PER (*P. trachelioides*) (experiment 3). Finally, we tested the survival of these ciliate species under extended UV exposure (experiment 1).

#### **2. Materials and Methods**

## *2.1. Ciliate Sampling*

Ciliates were collected in Piburgersee (PIB), an oligo-mesotrophic lake located in the Austrian Central Alps (47◦11 N 10◦53 E). The lake is situated at 913 m above sea level and has a maximum depth of 24.6 m. The lake is usually ice-covered from December through March/April. In PIB, UV-B

radiation is completely attenuated after a depth of 3 m and UV-A is attenuated after a depth of ca. 7m[10]. More information on UV transparency in PIB is given elsewhere [3].

From a previous study focused on PIB, we knew when the mixotrophic ciliate assemblage prevailed and the test species could be found [10]. Living ciliates for the experiments were collected by net tows in the uppermost 10 m of the water column with a 10-μm plankton net (Uwitec, Mondsee, Austria) on 31 August and 26 September 2011 for *P. trachelioides*, *S. vernalis*, and *V. chlorellata*, on 24 September 2012 for *P. trachelioides*, and on 1 and 17 October 2013 for *P. trachelioides*, *S. vernalis*, and *V. chlorellata*. Predatory zooplankton was excluded using a 250-μm plankton net before the water was poured into 1-L plastic bottles. After being transported to the laboratory, the ciliate samples were kept at ambient lake water temperatures that were measured along a depth profile with a thermometer attached inside a 5-L Schindler-Patalas sampler (16–17 ◦C on average in depths of 0–10 m).

#### *2.2. Ciliate Handling Prior to Experiments*

We screened the concentrated plankton using a stereomicroscope (Olympus SZ 40, Vienna, Austria), and individuals were identified morphologically (Olympus BX50 microscope, Vienna, Austria) under differential interference contrast following the key literature of [41]. The three mixotrophic ciliate species under study could not be successfully kept and enriched in earlier long-term cultures. Consequently, the experiments were carried out with freshly collected specimens. For acclimation, single ciliates were transferred with drawn glass pipettes and placed into 12-well plates (Bio-One, Greiner, Kremsmünster, Austria) containing 0.2-μm-filtered lake water (Minisart, Sartorius, Vienna, Austria). The experiments started the day after.

## *2.3. Experimental Design*

#### 2.3.1. General Experimental Setup

All experiments were run in a temperature-controlled walk-in chamber at 16–17 ◦C in well plates without lids. As an irradiation source, we used four A-340 Q-Panel lamps (8.60 W m<sup>−</sup><sup>2</sup> UV-A and 2.47 W m<sup>−</sup><sup>2</sup> UV-B; Q-Lab, Saarbrücken, Germany) and two Osram Cool White lamps NL-T8 36W/640-1 (72 μmol m<sup>−</sup><sup>2</sup> s<sup>−</sup><sup>1</sup> PAR; Osram, Vienna, Austria). This setup maintained over 4 h is able to simulate the daily UVR dose received at the lake's surface (air–water interface) in June at this latitude [29]. The ciliates were exposed (1) to the full spectrum of the two types of lamps (i.e., UV treatment), (2) to photosynthetically active radiation only (i.e., PAR treatment) by excluding the UVR with an Ultraphan-395 foil (UV-Opak, Digefra, Munich, Germany), and (3) to the dark (i.e., control) by wrapping the vessels with two layers of aluminum foil. The lamp spectrum weighted for the DNA Setlow action spectrum is given in [29].

Throughout all experiments, the ciliates were checked consecutively by eye under the stereomicroscope at 30–90 min intervals to record changes in numbers, shape, and movement. The number of individuals in the experimental wells was intentionally kept low, namely, five per well (*P. trachelioides* and *S. vernalis*) to be able to recognize any possible sublethal effects. As the epiplanktonic *V. chlorellata* was attached to colonies of *Botryococcus braunii* and it was not possible to successfully detach an adequate number from the algae, experiments were made to test only changes in shape and movement.

This general setup accompanied all three specific experiments (see below).

#### 2.3.2. Experiment 1 (exp 1) to Test the Ciliates' Overall Resistance to UVR

Here, we tested the species-specific response to UVR and PAR, both against a dark control. Individual ciliates were monitored over an irradiation period of 4 h (i.e., the natural daily dose at the surface for this geographical location) and of 7.5 h (i.e., an increased dose). This experiment was repeated twice each with *V. chlorellata* and *S. vernalis* (in 2011 and 2013) and five times with *P. trachelioides* (in 2011, 2012, and 2013) as well as in four replicates containing five individuals each (except for *V. chlorellata*; see explanation above). To test for variability under the di fferent exposure conditions, in 2011, we once quantified the MAAs for *S. vernalis* and *P. trachelioides* following the protocol of [14]. For *V. chlorellata*, MAAs could not be determined because the ciliates could not be separated from their algal attachment sites (that probably also contained MAAs).

2.3.3. Experiment 2 (exp 2) to Identify the UVR Wavelength-Specific Response of *P. trachelioides*

To identify changes in the survival, movement, and shape of *P. trachelioides* at specific wavelengths, we exposed the ciliates under a series of long-pass cuto ff quartz-glass filters (Andover Corporation, Salem, NH). The filters let UV-B and UV-A pass above specific wavelengths, i.e., 280, 295, 305, 320, 335, 345, and 360 nm. Due to their size of 50 × 50 mm and their 3 mm thickness, one filter covered four wells of a 12-well plate at once (i.e., four replicates at five individuals each). To keep o ffside irradiation during experiments, the rest of each plate was covered with black foil. Exp 2 was carried out three times with individuals from 2011 (once) and 2012 (twice) and ran over 7.5 h in total. We calculated the transmission of each cuto ff filter from the measured UV spectrum and integrated the dose rate over time for the UV-B and UV-A wavelength-specific ranges (Table S1).

2.3.4. Experiment 3 (exp 3) to Test for the Ciliates' Recovery Potential by "Dark Repair" (All Species) and PER (*P. trachelioides*)

In exp 3, we aimed to identify the recovery processes for curing the sublethal e ffects caused by previous exposure (exp 1 and exp 2). We recorded survival, behavioral, and morphological changes and symbiont dislocation.

To test for dark repair, survivors from exp 1 and exp 2 were kept at ambient temperatures in the dark for 12 h, and subsequently, they were maintained under light/dark conditions (16:8 h; 80 mmol m<sup>−</sup><sup>2</sup> s<sup>−</sup><sup>1</sup> PAR, 0.10 W m<sup>−</sup><sup>2</sup> UV-A). The dark repair approach subsequently followed exp 1 and exp 2 with any of the three species involved.

To test for PER, individuals of *P. trachelioides* were exposed to UVR (two parallel sets at once) and one dark control similar to exp 1 with the following modifications: after 4 h of exposure, one UV set was covered with Ultraphan-395 foil to cut o ff the UVR wavelengths (hereafter, UVR\_4h) and the second set was covered after 6 h (hereafter, UVR\_6h). After the respective UV treatments (i.e., UVR\_4 h or UVR\_6 h) the ciliates were exposed to photo-repairing light, i.e., 3.5 and 5.5 h, accordingly. The PER exp 3 was done twice (i.e., in 2011 in triplicates with nine individuals each and in 2012 in 12 replicates with five individuals each).
