Photosynthetic, Respirational, and Growth Responses of Six Benthic Diatoms from the Antarctic Peninsula as Functions of Salinity and Temperature Variations

Temperature and salinity are some of the most influential abiotic parameters shaping biota in aquatic ecosystems. In recent decades, climate change has had a crucial impact on both factors—especially around the Antarctic Peninsula—with increasing air and water temperature leading to glacial melting and the accompanying freshwater increase in coastal areas. Antarctic soft and hard bottoms are typically inhabited by microphytobenthic communities, which are often dominated by benthic diatoms. Their physiology and primary production are assumed to be negatively affected by increased temperatures and lower salinity. In this study, six representative benthic diatom strains were isolated from different aquatic habitats at King George Island, Antarctic Peninsula, and comprehensively identified based on molecular markers and morphological traits. Photosynthesis, respiration, and growth response patterns were investigated as functions of varying light availability, temperature, and salinity. Photosynthesis–irradiance curve measurements pointed to low light requirements, as light-saturated photosynthesis was reached at <70 µmol photons m−2 s−1. The marine isolates exhibited the highest effective quantum yield between 25 and 45 SA (absolute salinity), but also tolerance to lower and higher salinities at 1 SA and 55 SA, respectively, and in a few cases even <100 SA. In contrast, the limnic isolates showed the highest effective quantum yield at salinities ranging from 1 SA to 20 SA. Almost all isolates exhibited high effective quantum yields between 1.5 °C and 25 °C, pointing to a broad temperature tolerance, which was supported by measurements of the short-term temperature-dependent photosynthesis. All studied Antarctic benthic diatoms showed activity patterns over a broader environmental range than they usually experience in situ. Therefore, it is likely that their high ecophysiological plasticity represents an important trait to cope with climate change in the Antarctic Peninsula.


Introduction
Global warming is unequivocal, as is now evident from observations of increases in average global air and ocean temperatures, leading to the widespread melting of snow and ice in the polar regions, and rising average global sea levels [1]. However, the effects are quite different in Antarctica and the Arctic. While global warming is already strongly affecting the whole Arctic region [2], in Antarctica, thus far, mainly the Antarctic Peninsula has gotten warmer, where the air temperature and near-surface sea temperature have risen by 3 • C and 1 • C, respectively, in the past 50 years. This has resulted in a significant retreat Sediment surface samples taken in January/February 2020 from four study sites ( Figure 1 19 ) on King George Island were used for benthic diatom isolation. All isolates were established from samples collected in Potter Cove, which is separated into an inner part with a maximum depth of 50 m, and an outer part connected to the open ocean, with maximum depths of 100-200 m [37]. addition, we expected euryhaline and stenohaline responses for the marine and freshwater isolates, respectively.

Study Site
Sediment surface samples taken in January/February 2020 from four study sites ( Figure 1, Table 1) near the Argentinian research station Carlini Base (S 62°14′17.45″, W 58°40′2.19″) on King George Island were used for benthic diatom isolation. All isolates were established from samples collected in Potter Cove, which is separated into an inner part with a maximum depth of 50 m, and an outer part connected to the open ocean, with maximum depths of 100-200 m [37].
The marine culture N. criophiliforma originated from sample location APC06 (S 62 • 14 30.55 , W 58 • 40 54.96 ), which was an intertidal rock pool. Due to its location in the intertidal zone, abiotic parameters such as temperature and salinity strongly varied. C. gerlachei was isolated from a sample at the inner part of Potter Cove (APC14, S 62 • 13 43.61 , W 58 • 39 49.36 ), at 15 m depth, from a biofilm on top of a sediment. Unfortunately, oxidized material from the strain N. criophiliforma had low quality, and species identification on this isolate alone was not possible. A genetically identical species was isolated from brackish water at sample location APC12, and material of this strain was used for identification. Navicula concordia from a sample location at the outer cove (S 62 • 14 16.50 , W 58 • 42 44.20 ), at 5 m depth, originated from a biofilm on top of stones. According to Hernández et al. [38], the minimum water temperature was measured at −1.69 • C in the inner part of Potter Cove and −1.4 • C at the outer part, while the maximum temperature was 2.89 • C and 1.98 • C, respectively. Furthermore, the salinity of the outer cove is stable at ca. 33.5 S A , while the salinity in the inner cove can drop down to 29.6 S A .
The limnic isolates (N. annewillemsiana, Planothidium sp., and P. papilio) were established from biofilms on top of stones in a freshwater drinking reservoir (S 62 • 14 16.30 , W 58 • 39 44.10 ). During sampling, no measurements of pH, temperature, or conductivity were taken, due to malfunctioning instruments.
The diatom cells were isolated from aliquots of environmental samples to establish clonal cultures. Under an inverse light microscope (100-400× magnification, Olympus, Japan), single cells were transferred using a microcapillary glass pipette onto microwell plates containing culture medium (Guillard's f/2 medium [39,40] or Walne's medium [41]; 34 S A for marine samples and 1 S A for freshwater samples). All samples and isolated diatom cells from Antarctica were maintained at 5-7 • C. Irradiation was provided by white-light LEDs at 5000 K under a 16:8 h light:dark cycle, with several dark phases during the day to prevent photo-oxidative stress. After successful establishment of clonal cultures, they were separated into subsamples for DNA extraction, morphological analysis, and ecophysiological experiments. For the latter, diatom cultures were cultivated in sterile filtered Baltic Sea water, enriched with Guillard's f/2 medium [39,40] and sodium metasilicate (Na 2 SiO 3 • 5 H 2 O; 10 g 100 mL −1 ) to a final concentration of 0.6 mM (further referred to as cultivation medium). Salinity of 33 S A for the marine cultures was adjusted by using artificial sea salt (hw-Marinemix ® professional, Wiegandt GmbH, Germany), while 1 S A for the limnic cultures was achieved by dilution with deionized water.

Acquisition and Identification of Morphometric Data
In order to obtain clean diatom frustules, material harvested from the unialgal cultures was treated with 35% hydrogen peroxide at room temperature to oxidize the organic material, and then washed with distilled water. To prepare permanent slides for light microscopy (LM) analyses, the cleaned material (frustules and valves) was dispersed on glass coverslips, dried at room temperature, and embedded with the high-refraction-index mounting medium Naphrax ® (Morphisto GmbH, Offenbach, Germany).
Observations were conducted with a Zeiss Axioplan Microscope equipped with differential interference contrast (DIC), using a Zeiss 100 × Plan Apochromat objective, and microphotographs were taken with an AXIOAM MRc camera. Aliquots of cleaned sample material for scanning electron microscopy (SEM) observations were mounted on stubs and observed under a Hitachi FE 8010 scanning electron microscope operated at 1.0 kV.

DNA Extraction, Amplification, and Sequencing
Culture material was transferred to 1.5 mL tubes. DNA was isolated using the NucleoSpin ® Plant II Mini Kit (Macherey and Nagel, Düren, Germany), following the manufacturer's instructions. DNA fragment size and concentrations were evaluated via gel electrophoresis (1.5% agarose gel) and NanoDrop ® (PeqLab Biotechnology LLC; Erlangen, Germany), respectively. DNA samples were stored at −20 • C until further use, and finally deposited in the Berlin collection of the DNA Bank Network [42].
Amplification was conducted by polymerase chain reaction (PCR) as described by Zimmermann et al. [43] for the V4 region of 18 S. For the strain C. gerlachei, the whole 18 S gene was amplified as described by Jahn et al. [44]. The protein-coding plastid gene rbcL was amplified as described by Abarca et al. [45]. PCR products were visualized in a 1.5% agarose gel and cleaned with MSB Spin PCRapace ® (Invitek Molecular GmbH; Berlin, Germany), following the manufacturer's instructions. The samples were normalized to a total DNA content > 100 ng/µL using NanoDrop (PeqLab Biotechnology) for further sequencing. Sanger sequencing of the PCR products was conducted bidirectionally by Starseq ® (GENterprise LLC; Mainz, Germany).

Data Curation
Vouchers and DNA of all strains were deposited in the collections at Botanischer Garten und Botanisches Museum Berlin, Freie Universität Berlin (B). DNA samples were stored in the Berlin DNA bank, and are available via the Genome Biodiversity Network (GGBN) [46]. All sequences were submitted to the European Nucleotide Archive (ENA, http://www.ebi.ac.uk/ena/) using the software tool annonex2embl (accessed on 11 July 2022) [47] and can be retrieved from ENA under the study number PRJEB54671. All cultures are available from the authors at the culture collection of the Department of Applied Ecology and Phycology, University of Rostock.

Photosynthetic Efficiency
The photosynthetic potential of the six Antarctic benthic diatom strains as a function of salinity and temperature was measured using the pulse amplitude modulation (PAM) approach (PAM-2500, Heinz Walz GmbH, Effeltrich, Germany). The effective photochemical quantum yield of photosystem II in light-adapted cells, Y(II), was calculated (Equation (1)) by measurement of Fm (maximum chlorophyll fluorescence yield) and F (base fluorescence): Equation (1): Calculation of the effective photochemical quantum yield of photosystem II (Y(II)) Intensity of the measured light and gain were adjusted to reach F t (continuous base fluorescence) values between 0.15 and 0.2. Measurements were excluded from calculation when biomass did not surpass the F t value of 0.15 at the highest measured light intensity and gain.
All cultures were kept under culture conditions before transfer into the respective test media, with three replicates of. Two drops of thin diatom culture suspension were applied on a 25 mm glass-fiber filter (GF/6, Whatman, Little Chalfont, UK) and incubated in 2 mL of the respective treatment medium. To avoid nutrient deficiency, 1 mL of the medium was replaced with fresh medium every day. A radiator block was used during the measurements to avoid excessive temperature stress in the laboratory. Different salinity treatments were performed using sterile, filtered deionized water and artificial sea salt, with the addition of cultivation media. For the salinity treatments, marine isolates were incubated for three days under cultivation conditions at 1, 5,10,15,25,35,45,55,65,75,85, and 100 S A . Limnic isolates were exposed to salinities of 1, 5,10,20,30,35,40,45,55, and 65 S A . The isolates were incubated for three days prior to PAM measurements.
For the temperature treatments, experimental media were based on sterile, filtered deionized water and artificial sea salt (1 S A for limnic cultures and 35 S A for marine cultures), with the addition of cultivation media. All isolates were incubated for five days (t 5 ) at average temperatures of 1.5, 5, 7, 10, 15, 20, and 25 • C. Temperatures were achieved using a temperature organ. For 1.5 • C, an ice bath was used, with an exchange of ice every 12 h. The isolates were incubated for three days prior to PAM measurements.
PAM measurements were performed every 24 h, starting at day 0 (t 0 ), immediately after the transfer of the diatom cells onto the filter, until t 3 and t 5 .

Light Irradiance Curves (P-I Curves)
Photosynthetic activity as a function of light availability was measured as described by Prelle et al. [48] in a self-constructed P-I (photosynthesis-irradiance) box. Three (n = 3) airtight oxygen electrode chambers (DW1, Hansatech Instruments, King's Lynn, UK), tempered at 8 • C, were filled with 3 mL of thin algal-log-phase suspension, with the addition of 30 µL of sodium bicarbonate (NaHCO 3 , final concentration 2 mM) to avoid carbon deficit during the measurements. Oxygen concentration measurements along 10 increasing photon flux density levels, ranging from 3.6 to~1670 µmol photons m −2 s −1 of photosynthetically active radiation (PAR), were undertaken using oxygen dipping probe DP sensors (PreSens Precision Sensing GmbH, Regensburg, Germany) linked to fiber-optic oxygen transmitters via optical fibers (Oxy 4 mini meter, PreSens Precision Sensing GmbH, Regensburg, Germany). Measurements started with a 30-min respirational phase, followed by a series of 10-min photosynthetic phases for each increasing light level.
Chlorophyll a concentration per chamber was measured after each final measurement by the extraction of 3 mL of the algal suspension using 96% ethanol (v/v). Chlorophyll a was measured spectrophotometrically at 665 nm and 750 nm (Shimadzu UV-2401 PC, Kyoto, Japan) [49], and calculated according to Equation (2): Equation (2): Chlorophyll a calculation, where v is the extraction volume (mL), d is the cell length (cm), and V is the volume of filtered suspension (mL).
Due to photoinhibition in some of the diatom strains, the mathematical photosynthesis model of Walsby [50] was used for fitting and calculation of the maximum rates of net primary production (NPP max ), respiration (R), light utilization coefficient (α), photoinhibition coefficient (β), light saturation point (I k ), and light compensation point (I c ).

Temperature-Dependent Photosynthesis and Respiration
Photosynthetic and respirational rates of the six Antarctic diatom strains in response to temperatures between 5 • C and 40 • C were measured using the same P-I box as for the P-I curves, following the approach of Karsten et al. [51]. In contrast to the P-I curves, light was switched off during the respirational phase and set to photosynthesis-saturated 342.2 ± 40 µmol photons m −2 s −1 PAR during the photosynthetic phase. Starting at 5 • C, a 20-min dark incubation phase was followed by a 10-min respirational phase and a 10-min photosynthetic phase. Afterwards, the temperature was increased by 5 • C, and the process was repeated until reaching 40 • C. Oxygen concentration measurements were also normalized to the total Chlorophyll a concentration [49].

Growth Rates
Growth rates of the marine diatom strain C. gerlachei and the limnic diatom strain P. papilio in response to salinity and temperature were determined as described by Karsten et al. [52], Gustavs et al. [53], and Prelle et al. [54]. Measurement of the in vivo fluorescence of chlorophyll a was used as a proxy for biomass. Using an MFMS fluorimeter (Hansatech Instruments, King's Lynn, UK), Chlorophyll a was excited by blue light emission and detected as relative units by an amplified photodiode that was separated from scattered excitation light. This method is particularly suitable for benthic diatoms, as chlorophyll a fluorescence units correlate well with chlorophyll a and cell carbon concentrations, as well as cell numbers, in diatoms [52,53]. Both diatom cultures were cultivated in disposable Petri dishes (n = 3) with cover lids, in a volume of 15 mL of the respective treatment medium, and measured every 24 h for 8 days. To avoid the measurement of potential initial shock reactions of the isolates, 1 mL of log-phase algae suspension was incubated in 14.5 mL of the respective trial medium for four days under experimental conditions.
Growth as a function of salinity was tested by exposure to salinities of 1, 5, 25, 35, 45, 65, 85, and 100 S A for the marine species C. gerlachei, and 1, 5, 10, 20, and 30 S A for the limnic species P. papilio. Salinities were adjusted using artificial sea salt (hw-Marinemix ® professional) dissolved in deionized water. All cultures were enriched with f/2 and metasilicate, and kept at 8-9 • C under standard cultivation conditions. Furthermore, growth in response to five temperatures (5,8,15,20, and 30 • C) at salinities of 1 S A (C. gerlachei) and 35 S A (P. papilio), with added f/2 and metasilicate, was investigated. Treatments at 5 • C were kept in a wine storage refrigerator with added LEDs; treatments at 8, 15, and 20 • C were kept in climate chambers; and treatments at 30 • C were carried out in a tempered water bath, with all reflecting light settings similar to cultivation conditions. After measurements, the growth rates of the logarithmic phase were calculated using the following equation: N = N 0 × e (µ × dt) , where N is the fluorescence on the measuring day, dt is the difference in time (days) between the measuring day and the starting day, and µ is the growth rate) [53].

Statistical Analysis
Our statistical analysis was similar to that of Prelle et al. [54], as Microsoft Office Excel (2016) was used for the calculation-partially by application of the solver function, by minimizing the sum of normalized squared deviations for the fitting of the model of Walsby [50]-and creation of figures. R (Version: 4.0.2) was used for the calculation of significance levels using one-way ANOVA followed by a post hoc Tukey's honestly significant differences test (critical p-value < 0.05), as well as for the fitting of the model of Yan and Hunt [55] for the temperature-dependent photosynthesis. Confidence intervals were calculated using the library nlstools in R.

Species Identification
Five of the Antarctic isolates were identified to the species level, and one to the genus level. Table 2 lists the taxa, with information on the morphology and respective accession numbers of the marker genes rbcL and 18 SV4/18 S, while Figures 2-4 depict the LM and SEM images.
APC14 D296_001 was identified as Chamaepinnularia gerlachei Van de Vijver and Sterken ( Figure 2L-T; valves of strain D294_006 are depicted as well, since this strain was used as a supplement for identification). This species was first published in the work of Van de Vijver et al. [56], from dry soil samples collected at James Ross Island, near the northeastern extremity of the Antarctic Peninsula, and has been observed only in maritime Antarctica thus far [57][58][59].
Navicula concordia ( Figure 3A-K) was identified as N. concordia C. Riaux-Gobin and A. Witkowski, and APC06 D288_003 as Navicula criophiliforma A. Witkowski, C. Riaux-Gobin, and G. Daniszewska-Kowalczyk (Figure 2A-K). Both were first published in the work of Witkowski et al. [60], from the Kerguelen Islands coastal area, in the Southern Ocean. Recently, N. criophiliforma was reported from Livingston Island, north of the Antarctic Peninsula [61]. This species formed auxospores during cultivation, leading to high variance in the dimensions of the valves.
APC18 D300_012 ( Figure 3L-U) was identified as Nitzschia annewillemsiana Hamsher, Kopalová, Kociolek, Zidarova, and Van de Vijver. It was first published in the work of Hamsher et al. [62], from freshwater and wet soil samples from James Ross Island and the South Shetland Islands, and has been only reported from this area to date [59]. Table 2. List of strains established from Antarctic marine and freshwater samples collected at Carlini Station, King George Island, Potter Cove, in austral summer 2020 (January/February), with scientific name, information on dimensions of the valves, striae density, and sequenced marker genes. RV: raphe valve, SV: sternum valve.

Strain
Scientific Name  Figure 4M-Z). It was first described as Navicula papilio by Kellogg et al. [63], but this species has been reported several times from maritime Antarctica under different synonyms [32,57,59,64].

Marine/ Freshwater
APC18 D300_015 was identified to the genus level as Planothidium sp. ( Figure 4A-L). There was a high morphological resemblance to P. frequentissimum (Lange-Bertalot) Lange-Bertalot. However, molecular data showed differences in both marker genes compared to P. frequentissimum strains from GenBank. There were 4 base-pair differences in 18 SV4 and 20 in the rbcL gene compared to the P. frequentissimum strain PF1 (Accession numbers: KJ658409 and KJ658392). In comparison to the strain D06_139 (Accession numbers: KY650786 and KX650815), 11 bp differences were found in 18 SV4, and 19 in rbcL.

Photosynthetic Potential
The photosynthetic potential of all six diatom strains exhibited wide tolerance ranges between the tested salinities from 1 S A to 100 S A after three days of incubation ( Figure 5, Table S1). The overall highest and lowest optimal quantum yields were measured for N. concordia, with 0.595 at 45 S A and 0.033 at 5 S A , respectively. The three marine species N. criophiliforma, C. gerlachei, and N. concordia exhibited typical tolerance curve patterns, with significant optima at 25-35 S A , 25-45 S A and 5-45 S A , respectively (p < 0.05, Figure 5). The tolerance range of N. criophiliforma was narrower compared to both other marine species after calculation of the range of the highest effective quantum yield at the 80th percentile and above, between the 20th and 80th percentiles, and below the 20th percentile ( Figure 6A). This taxon exhibited high effective quantum yields (upper 80th percentile) at only two experimental salinities, while the other isolates covered 4-6 salinities ( Figure 6A). Nevertheless, all marine species exhibited a moderate effective quantum yield (between the 20th and 80th percentiles), ranging from 10 to 55/75 S A and up to 100 S A . In contrast, the two limnic species Planothidium sp. and P. papilio showed the highest significant optima at 10-20 S A and 1-10 S A , respectively (p < 0.05, Figure 5). Salinities higher than 10/20 S A resulted in a decreasing effective quantum yield up to 40/55 S A . Due to low biomass, Planothidium sp. was only tested in two salinities, of which the highest effective quantum yield was measured at 10 S A .
Genes 2022, 13, 1264 13 of SA. Due to low biomass, Planothidium sp. was only tested in two salinities, of which highest effective quantum yield was measured at 10 SA.  The photosynthetic potential of the six diatom strains after five days of incubation also exhibited broad tolerances for the investigated temperature range of 1.5 to 25 • C (Figure 7, Table S1). The highest overall effective quantum yield was found for N. concordia, with 0.585 at 15 • C, while the lowest was for Planothidium sp., with 0.123 at 25 • C. Between 1.5 and 25 • C, only small significant deviations of the effective quantum yield were found for all three marine taxa, as well as for P. papilio, while for N. annewillemsiana and Planothidium sp. The highest effective quantum yield was at 15 to 25 • C (p < 0.05, Figure 7). With the exception of N. criophiliforma at 25 • C and C. gerlachei at 1 • C and 25 • C, all species exhibited moderate photosynthetic potential between 1.5 • C and 25 • C. In comparison to t 5 , significance levels of t 0 between each temperature treatment of the marine species were not as distinct as for the limnic species (Figure 7). 14 o Figure 6. Effects of (A) salinity and (B) temperature on the effective quantum yield of photosys II (Fv/Fm) of six benthic diatom strains from Antarctica. Dark blue symbols represent the rang the highest effective quantum yield at the 80th percentile and above, medium blue symbols between the 20th and 80th percentiles, light blue symbols represent the 20th percentile and be and white symbols were not tested. Data represent mean values (n = 6).
The photosynthetic potential of the six diatom strains after five days of incuba also exhibited broad tolerances for the investigated temperature range of 1.5 to 25 (Figure 7, Table S1). The highest overall effective quantum yield was found for N. cordia, with 0.585 at 15 °C, while the lowest was for Planothidium sp., with 0.123 at 25 Between 1.5 and 25 °C, only small significant deviations of the effective quantum y

Light-Dependent Photosynthesis
The photosynthetic and respirational rates of all six diatom strains exhibited speciesspecific responses towards increasing photon fluence rates, resulting in different P-I parameters ( Figure 8, Table 3). The overall highest NPP max was for the marine species N. criophiliforma, with 202.3 ± 45.4 µmol O 2 mg −1 Chl a h −1 , which was at least twice as high as that of the remaining isolates. Respiration rates varied among the isolates, between −47 ± 8.9 µmol O 2 mg −1 Chl a h −1 (N. criophiliforma) and −10.5 ± 3.1 µmol O 2 mg −1 Chl a h −1 (N. concordia) (Figure 8, Table 3). All isolates had low light compensation points (I c ), varying significantly between 5.8 ± 1 µmol photons m −2 s −1 (N. concordia) and 17.5 ± 3 µmol photons m −2 s −1 (Planothidium sp.) (p < 0.05, Table 3). The light saturation points (I k ) for all six isolates ranged between 64 ± 11.5 µmol photons m −2 s −1 (N. criophiliforma) and 16.3 ± 3.9 µmol photons m −2 s −1 (N. annewillemsiana). Photoinhibition was detected in almost all isolates except for Planothidium sp. and P. papilio. The highest photoinhibition was found in N. criophiliforma, with −0.03 ± 0.02, which was, however, not significant between C. gerlachei, N. concordia, and N. annewillemsiana (p < 0.05, Table 3).  Table 3. Parameters of respective P-I curves (Figure 8) of six benthic diatom species (n = 3) kept at 8 °C. Different lowercase letters represent significance levels among all means as calculated by one-way ANOVA (Tukey's test, p < 0.05). NPPmax represents the maximal oxygen production rate, α is the initial slope of production in the light-limited range, β is the terminal slope of production in extensive light range (photoinhibition), Ik is the light saturation point, and Ic is the light compensation point.    Table 3. Parameters of respective P-I curves ( Figure 8) of six benthic diatom species (n = 3) kept at 8 • C. Different lowercase letters represent significance levels among all means as calculated by one-way ANOVA (Tukey's test, p < 0.05). NPP max represents the maximal oxygen production rate, α is the initial slope of production in the light-limited range, β is the terminal slope of production in extensive light range (photoinhibition), I k is the light saturation point, and I c is the light compensation point.

Temperature-Dependent Photosynthesis and Respiration
Photosynthetic and respirational responses under increasing temperatures from 5 to 40 • C resulted in individual response patterns (Figure 9, Table S1). Photosynthesis and respiration rates typically rose with increasing temperature and decreased after reaching the optimal temperature. The overall highest photosynthesis and respiration rates were measured for N. criophiliforma at 15 • C and 30 • C, respectively (Figure 9). In general, positive net photosynthetic rates ranged between 5 • C and 25/30 • C, with varying optima for each strain (from 5 • C to 20 • C). At temperatures > 25/30 • C, photosynthesis was inhibited, and only respirational oxygen consumption could be measured (Figure 9). Respirational rates could be detected over the entire temperature range from 5 to 40 • C, with optima between 20 and 35 • C (Figure 9). Fitting of the measured data using the model of Yan and Hunt [55] revealed maximum photosynthetic rates of the marine isolates between 11.1 and 15.7 • C, and a positive net photosynthesis up to 32.5 and 35.6 • C, respectively ( Table 4). The optimal temperature for the limnic species was slightly lower-between 3.0 and 12.

Temperature-Dependent Photosynthesis and Respiration
Photosynthetic and respirational responses under increasing temperatures from 5 to 40 °C resulted in individual response patterns (Figure 9, Table S1). Photosynthesis and respiration rates typically rose with increasing temperature and decreased after reaching the optimal temperature. The overall highest photosynthesis and respiration rates were measured for N. criophiliforma at 15 °C and 30 °C, respectively (Figure 9). In general, positive net photosynthetic rates ranged between 5 °C and 25/30 °C, with varying optima for each strain (from 5 °C to 20 °C). At temperatures > 25/30 °C, photosynthesis was inhibited, and only respirational oxygen consumption could be measured (Figure 9). Respirational rates could be detected over the entire temperature range from 5 to 40 °C, with optima between 20 and 35 °C (Figure 9). Fitting of the measured data using the model of Yan and Hunt [55] revealed maximum photosynthetic rates of the marine isolates between 11.1 and 15.7 °C, and a positive net photosynthesis up to 32.5 and 35.6 °C, respectively ( Table 4). The optimal temperature for the limnic species was slightly lower-between 3.0 and 12.

Growth Rates
One marine and one limnic culture were exemplarily investigated for growth as a function of salinity and temperature ( Figure 10, Table S1). C. gerlachei exhibited a strong optimum at 15 • C, with growth rates of 0.84 µ d −1 , while the optimal growth temperature for P. papilio ranged between 5 and 15 • C, with similar growth rates around 0.4 µ d −1 (Figure 10). Both diatom strains were unable to grow at temperatures > 20 • C. Using the model of Yan and Hunt [55], the optimal growth temperature of >80% of the maximal growth ranged from 6.5 to 19.9 • C for C. gerlachei, and from 1.6 to 14.5 • C for P. papilio ( Table 4). The overall maximal growth rate for C. gerlachei was at 13.0 • C (0.44 µ d −1 ), and for P. papilio it was at 6.5 • C (0.30 µ d −1 ). Growth rates as a function of salinity for the marine species C. gerlachei were determined over a range from 1 to 65 S A . This species exhibited a broad salinity tolerance, as reflected in growth rates between 0 and 79.2 S A (0.2 to 0.4 µ d −1 ), with a distinct optimum at 6.5 S A (0.58 µ d −1 ) ( Table 4). In contrast, the limnic species P. papilio grew only over a range of 1 to 20 S A , with optima between 1 and 10 S A . The model calculation for salinity exhibited highest growth rate of 0.42 µ d −1 at 5.28 S A ( Table 4). The optimal growth range >80% growth rate for this isolate ranged between 0.9 and 13.7 S A .  Table 4). The optimal growth range >80% growth rate for this isolate ranged betwee 0.9 and 13.7 SA.

Discussion
All six marine and limnic benthic diatom species from the maritime Antarctic Pen insula exhibited broad tolerances towards light availability as well as euryhaline an eurythermal traits, far surpassing the environmental conditions of their respective hab tats. In general, Antarctic organisms are expected to be rather stenohaline and steno therm due to the long cold-water history of the Southern Ocean. However, maritim Antarctica is characterized by stronger seasonal and diurnal fluctuations in the abiot parameters compared to continental Antarctica; hence, broader ecophysiological tole ances of the inhabiting biota might be assumed. Nevertheless, an important aspect th should be considered is related to the fact that all six benthic diatom species were grow as clonal cultures for >1 year under controlled lab conditions before the experiments wer

Discussion
All six marine and limnic benthic diatom species from the maritime Antarctic Peninsula exhibited broad tolerances towards light availability as well as euryhaline and eurythermal traits, far surpassing the environmental conditions of their respective habitats. In general, Antarctic organisms are expected to be rather stenohaline and stenotherm due to the long cold-water history of the Southern Ocean. However, maritime Antarctica is characterized by stronger seasonal and diurnal fluctuations in the abiotic parameters compared to continental Antarctica; hence, broader ecophysiological tolerances of the inhabiting biota might be assumed. Nevertheless, an important aspect that should be considered is related to the fact that all six benthic diatom species were grown as clonal cultures for >1 year under controlled lab conditions before the experiments were undertaken. It might be possible that the measured ecophysiological response patterns do not always reflect the in situ responses. In addition, due to the cultivation procedure described we can not rule out that we selected for the most tolerant species while sensitive taxa were outcompeted. Although logistically challenging in Antarctica, more field experiments are urgently needed to better understand the real world.

Light
Photosynthesis-the driving force for the energy metabolism and, hence, essential for the viability and survival of benthic diatoms-is primarily dependent on light availability. All six diatom species exhibited taxon-specific response patterns over a wide range of photon fluence rates, with only slight photoinhibition. Overall, the marine isolates exhibited higher NPP max compared to the limnic ones. The highest photosynthetic rates were reached at low photon fluence rates, as reflected by low light compensation and light saturation points. All data clearly point to low light requirements for photosynthesis. In general, Antarctic diatoms are known for their fast growth in low-light conditions [65], because their photosynthesis seems to be especially shade-adapted [66]. The few data available on benthic diatoms from polar regions confirm a high photophysiological plasticity to acclimate to the prevailing, often very low light conditions [10,11,[21][22][23]67]. In addition, this wide photophysiological plasticity seems to be a rather general trait of many diatom species [24], as documented in species from Arctic Kongsfjorden [27], but also in numerous species from the shallow waters of the temperate Baltic Sea [48,54].
Particularly for benthic diatoms, low light adaptation is crucial, since Antarctic microphytobenthic communities experience a strong seasonally changing light climate, often with low average photon fluence rates. During winter periods, with ice cover and short daylight periods, little or no light reaches the benthic diatoms-especially if the ice is covered by snow [68]. During summer, incident light penetration can be also reduced due to increased turbidity, which is driven by suspended particulate matter from glacial meltwater and riverine discharge [69]. In addition, wind-and organism-induced resuspension of the sediment can lead to a decline in the light availability through burial of the diatom cells. However, due to their motility, raphid diatoms are able to escape unfavorable low-light conditions in the sediment [70]. Vertical migration of benthic diatoms has been identified as an important behavioral trait to control the short-term variability of photosynthesis-at least in temperate regions. Although published studies on the vertical migration of benthic diatoms in Antarctica and the Arctic under polar day and night conditions are lacking, a few reports also indicate motility in polar species [27]. After the sea ice breakup in spring, solar radiation penetrates the coastal water column, with strong attenuation of the short wavelengths due to the prevailing optical properties, which are influenced by particle load from glaciers and yellow substances originating from meltwater and terrestrial runoff [69,71]. At 10 m depth in the inner Potter Cove, PAR was measured between 10 and 200 µmol photons m −2 s −1 in the winter and summer, respectively [28]. An interesting aspect was the overall twofold-higher NPPmax exclusively in Navicula criophiliforma (about 200 µmol O 2 mg −1 Chl a h −1 ) compared to all other studied Antarctic benthic diatom species. At present, we can only speculate to explain these data, but the largest cell size of N. criophiliforma (<52 × 8.5 µm, Table 2, Figure 2) among all species leads to the highest cell volume and, hence, to more chloroplasts and pigments. Recent data on the green microalga Dunaliella teriolecta experimentally prove that the established package effect theory, which predicts that larger phytoplankton cells should show poorer photosynthetic performance because of reduced intracellular self-shading, is challenged [72]. The latter authors reported that larger cells of D. teriolecta showed substantially higher rates of oxygen production along with higher chlorophyll values compared to smaller cells.
All six species could not only cope well with low-light conditions, but also showed high photosynthetic rates up to 1600 µmol photons m −2 s −1 , with a minor-to-moderate degree of photoinhibition-especially in the marine strains. During the process of photoinhibition, diatoms are still able to perform photosynthesis without being completely inhibited. Excessive light is absorbed by the photosystems and harmlessly emitted via heat as a protective mechanism (non-photochemical quenching) for the photosynthetic apparatus [73]. Further exposure to excessive light, however, can lead to damage of the D1 protein, leading to a decrease in electron transfer [74]. All benthic diatom species exhibited low light requirements for photosynthesis combined with a pronounced photophysiological plasticity that also allowed broad tolerance to high-incident-light conditions.

Temperature
Photosynthesis, respiration, and growth, along with their underlying enzymatic mechanisms, are strongly controlled by temperature. Therefore, reductions in photosynthetic and respirational activity, as well as in growth under saturated light conditions, are a consequence of inhibition of the most temperature-sensitive enzymes. Low temperatures slow down electron transport, thereby decreasing the ability to use photons for photochemically produced energy. High temperatures can influence the photorespiration activity of RuBisCO (ribulose-1,5-bisphosphat-carboxylase/-oxygenase) by removing its specificity towards CO 2 binding rather than that of O 2 , thereby increasing energy demand [75]. Similar to other studies using the same methodological approach on Baltic Sea benthic diatoms [48,54], the photosynthetic and respirational rates of Antarctic diatom strains seemed to also be decoupled from one another, with respiration always showing optima at higher temperatures compared to photosynthesis. For temperate diatoms, but also for terrestrial green algae, temperature requirements for respiration and photosynthesis differ, as explained by the higher dependency of photosynthesis on light, while temperaturedependent enzymatic activities mainly control respiration [76,77]. A more recent study partially confirmed that light-dependent photosynthetic reactions are indeed unaffected by temperature, while the carbon fixation reactions are driven by temperature [78]. Furthermore, respiratory and photosynthetic activities in diatoms are strongly coupled, which is mechanistically explained by tight physical interactions between mitochondria and chloroplasts [79]. Consequently, light stimulates respiration, resulting in an optimal ATP/NADPH ratio for subsequent carbon dioxide fixation by RuBisCO.
Another important aspect is the observation that the optimal temperature for photosynthesis ( Figure 9, 20 • C) was higher compared to that for growth ( Figure 10, <15 • C). These differences in both physiological processes can be explained by the exposure time to the stressor temperature. The time scale of stress is relevant, as algae may cope temporarily with strong temperature stress if acting only for hours to days, and may subsequently recover from damage under optimal conditions [80]. However, on a longer time scale (weeks), the algae experience progressively more impaired cellular processes until the upper temperature for survival is reached. Consequently, temperature optima for photosynthesis are often higher than those for growth, because both physiological processes are not directly coupled and, hence, photosynthesis does not necessarily match the temperature-growth pattern. In addition, growth is a more general physiological process that integrates all positive and negative influences of temperature on the whole metabolism [81]. The data shown clearly indicate broad temperature tolerance of photosynthesis and respiration in the Antarctic benthic diatom isolates, far exceeding in situ temperatures in their respective habitats. While the temperature tolerance of Antarctic phytoplankton-which usually do not survive temperatures > 8-9 • C, and which is consistent with the maximum in situ temperature-is recognized as stenotherm [82,83], benthic diatoms in shallow waters or in tidal pools during the polar day can be exposed to temperatures that are several times higher compared to the water column. For Potter Cove, where the investigated strains were sampled, the tides are semi-diurnal, and the temperature in some tidal pools may change from 2 to 14 • C within only 8 h [37].
As already mentioned in the introduction, all benthic diatoms from the Arctic that have so far been experimentally studied under controlled conditions typically exhibit eurythermal and psychrotolerant traits, while those from Antarctica show stenothermal and psychrophilic features [27,28]. These fundamental differences in the response patterns can be explained by the geologically 10-fold longer cold-water history of Antarctica compared to the Arctic, fostering adaptive and evolutionary processes in the inhabiting organisms, which finally led to many endemic marine organisms in Antarctica [32]. However, in sharp contrast to the data of Longhi et al. [28], all six benthic diatom species in the present study exhibited very similar ecophysiological response patterns, comparable not only to those from their Arctic counterparts, but also to those from temperate regions such as the Baltic Sea [48,54], hence pointing to eurythermal and psychrotolerant traits. The unexpectedly broad temperature tolerances are not easy to explain, but Potter Cove is one of the few places in Antarctica where long-term ecological observational data exist. Based on >20-year time series of sea surface temperature, data prove a temperature increase of 0.7 to 0.8 • C in the last two decades, accelerating biological activities and physicochemical processes in the shallow coastal waters of Potter Cove [84]. As a consequence, summer meltwater runoff from coastal ice sheets and from thawing of coastal permafrost areas causes freshening of the shallow water, along with increasing turbidity due to mobilization of lithogenic particles, so that benthic biota are strongly affected by such highly dynamic and new climate-sensitive environmental conditions [84]. Therefore, it is reasonable to assume that during the ca. 20-year time span between the study of Longhi et al. [28] and the data presented here, changes within the benthic diatom community took place, i.e., from more stenoecious (endemic) to euryoecious (non-endemic) taxa. It might also be possible that non-endemic benthic diatoms invaded the Antarctic Peninsula from sub-Antarctic islands and from South America-for example, as hull biofouling organisms-as shown for other benthic organisms, such as invertebrates [85]. However, comprehensive information on the biodiversity and biogeography of marine benthic diatoms in Antarctica is still lacking, while freshwater species are very well studied [34].

Salinity
In general, the photosynthetic potential of the six benthic diatom species exhibited broad tolerances, with habitat-typical salinity optima of 25 to 35 S A for the marine strains and 1 to 10 S A for the freshwater strains. Due to the topographic division within Potter Cove and the freshwater runoff, salinity in the inner part of the bay exhibits lower salinities > 29.6 S A , compared to the outer part, which has fully marine salinities [38]. Salinity stress is related to the toxic effects of Na + and Cl − , and often results in a decline in photosynthetic activity or a change in PS II efficiency [86][87][88][89]. This can lead to oxidative stress, consequently damaging lipid membranes, proteins, or nucleic acids [90], while also interfering with the photosynthetic and respiratory electron transport. The accompanying effect of changing cell volumes can also lead to the deactivation of the photosynthetic apparatus [88].
In the marine rock pools, benthic diatoms are typically exposed to strong tidal-induced salinity changes, as they are cut off from the main body of marine water. In the rock pools, salinity can increase as a result of strong evaporation due to insolation and wind, or decrease because of precipitation or glacial freshwater inflow [91]. The marine strain N. criophiliforma, sampled from a rock pool in Potter Cove, exhibited a wide euryhaline tolerance, with >20% photosynthetic potential between 10 and 55 S A , thereby able to cope well with this abiotic factor. The remaining two marine species were also euryhaline in terms of photosynthesis. Such broad tolerance ranges are typically found in diatoms living under and within the sea ice, and which can cope with salinities of up to 60-100 S A [87]. However, apart from sea-ice diatoms, there exist only fragmentary data on salinity responses in polar benthic diatoms. For the Arctic Nitzschia cf. aurariae, growth between 15 and 45 S A , with an optimum at 20 to 40 S A , was reported, and it was thus characterized as moderately euryhaline [92]. In contrast to polar benthic diatoms, their temperate counterparts are well studied in terms of a commonly wide salinity tolerance. Numerous benthic diatoms from the North Sea exhibited high growth rates between 2 and 45 S A [93], and between 10 and 40 S A [94], while a study from the Baltic Sea reported growth between 1 and 50 S A [95].
The underlying mechanisms of osmotic acclimation have not yet been studied in Antarctic benthic diatoms. In contrast, ice-associated diatoms trapped in the brine channels can experience salinities three times that of seawater. These algae typically synthesize and accumulate high concentrations of organic osmolytes and compatible solutes in response to hypersaline stress, such as proline, mannitol, glycine betaine, and/or dimethylsulfoniopropionate (DMSP) [96].

Conclusions
In conclusion, all six benthic diatom species isolated from the Antarctic Peninsula exhibited strong euryhaline and eurythermal traits far surpassing the environmental conditions of their respective habitats. Pronounced low light requirements and species-specific photophysiological plasticity with minor photoinhibition were present. With regard to the ongoing climate change-particularly in maritime Antarctica-the increasing water temperatures of Potter Cove, and the accompanying fluctuations in salinity and the light field, all of the isolates seemed to be well acclimated, as reflected in their eurythermal and euryhaline response patterns.
Author Contributions: L.R.P. and U.K. developed the idea and elaborated the concept. I.S. and D.J. provided the experimental data. K.S., N.A. and J.Z. provided the taxonomical data. O.S. isolated and established clonal cultures. L.R.P., K.S., J.Z., U.K. and I.S. wrote the manuscript, and L.R.P. provided all statistical data. All authors have read and agreed to the published version of the manuscript.
Funding: This study was conducted within the framework of the Research Training Group Baltic TRANSCOAST funded by the DFG (Deutsche Forschungsgemeinschaft) under grant number GRK 2000/1 (Subproject B2: Microphytobenthos). This is Baltic TRANSCOAST publication no. GRK2000/0056). Furthermore, this project was funded within the framework of the SPP 1158 Antarktisforschung by the DFG under the grant numbers KA899/38-1 and ZI 1628/2-1.