Intercomparison of Two Fluorescent Dyes to Visualize Parasitic Fungi (Chytridiomycota) on Phytoplankton

Fungal microparasites (here chytrids) are widely distributed and yet, they are often overlooked in aquatic environments. To facilitate the detection of microparasites, we revisited the applicability of two fungal cell wall markers, Calcofluor White (CFW) and wheat germ agglutinin (WGA), for the direct visualization of chytrid infections on phytoplankton in laboratory-maintained isolates and field-sampled communities. Using a comprehensive set of chytrid–phytoplankton model pathosystems, we verified the staining pattern on diverse morphological structures of chytrids via fluorescence microscopy. Empty sporangia were stained most effectively, followed by encysted zoospores and im-/mature sporangia, while the staining success was more variable for rhizoids, stalks, and resting spores. In a few instances, the staining was unsuccessful (mostly with WGA), presumably due to insufficient cell fixation, gelatinous cell coatings, and multilayered cell walls. CFW and WGA staining could be done in Utermöhl chambers or on polycarbonate filters, but CFW staining on filters seemed less advisable due to high background fluorescence. To visualize chytrids, 1 µg dye mL−1 was sufficient (but 5 µg mL−1 are recommended). Using a dual CFW–WGA staining protocol, we detected multiple, mostly undescribed chytrids in two natural systems (freshwater and coastal), while falsely positive or negative stained cells were well detectable. As a proof-of-concept, we moreover conducted imaging flow cytometry, as a potential high-throughput technology for quantifying chytrid infections. Our guidelines and recommendations are expected to facilitate the detection of chytrid epidemics and to unveil their ecological and economical imprint in natural and engineered aquatic systems. Supplementary Information The online version contains supplementary material available at 10.1007/s00248-021-01893-7.


Material and methods
Text S1. Chemicals the solution should be prepared under a fume hood add 100 mL of dH2O to a glass beaker stir and heat to 60°C, while stirring, add 10 g of paraformaldehyde powder cover the beaker with parafilm and maintain at 60°C until PFA powder is dissolved (optional step) -raise the pH for better dissolution by adding 2N NaOH (dropwise, ca. 1 drop per 100 mL) -after the addition of NaOH, the solution should clear within a couple of minutes (some fine particles may remain) -do not heat above 70°C -filter the solution through 0.45 μm membrane filter to remove any remaining particles, and aliquot into ca. 10 mL and store at -20°C

Other chemicals
Sodium thiosulfate anhydrous (Na2S2O3, Merck 1.06512, solution for de-staining from Lugol) -Stock solution (30 g L -1 ) o dissolve 1.5 g Na2S2O3 in 50 mL of deionized water o filter solution through a 0.2 µm membrane filter o storage at RT for months -20-40 µL of Na2S2O3 clears 1 mL Lugol-preserved sample (10 µL Lugol mL -1 ) Culture medium (modified CHU-10) add chemicals as listed in Table S1, except for vitamins and Fe-EDTA adjust to pH 6.4 with HCl solution or NaOH solution autoclave (121°C, 30 min) -add vitamins and Fe-EDTA after autoclaving and cooling down to avoid heat damage of vitamins storage: o CHU-10 medium can be stored at RT for several months o CHU-10 medium + vitamins/Fe-EDTA can be stored at RT and should preferably be used within 1-2 months

Equipment
-Inverted epifluorescence microscope (a Nikon Eclipse Ti2-U, Nikon, Japan was used in this study), equipped with an Utermöhl chamber holder Procedure 1. Sample fixation fix water sample with Lugol´s solution (e.g., final conc. 10 µl mL -1 ) -we recommend the use of neutral or alkaline Lugol´s solution as acidified Lugol´s solution reduces the pH of the sample outside the recommended pH range of 6.5-8.5 given for Alexa Fluor® dyes. Although Alexa Fluor® dyes are pH insensitive across the range of 4-10 [1], we observed a different WGA-Alexa Fluor™ 488 staining behavior for acidic and neutral/alkaline Lugolpreserved samples (Supplementary Figure S5). The pH in the preserved samples was 4 (acidic Lugol) and ca. 7 (neutral/alkaline Lugol). -store samples overnight at RT in the dark or 4°C for long-term storage we recommend an overnight fixation of samples before staining, to ensure an effective WGA binding Optional: Before staining, fixed phytoplankton cells can be concentrated by sedimentation, as follows.
homogenize the sample by gentle shaking, and transfer the desired volume to a measuring cylinder or equivalent (e.g., 50 mL centrifugation tube) -let the sample undisturbed for at least 24 h at RT, to allow the cells to settle to the bottom carefully siphon off 90% of the top water layer (any turbulence and cell resuspension should be avoided), to reach a 10x enrichment factor of cells in the sample 2. WGA and CFW dual-staining transfer 1-2 mL of the sample into a 2 mL microcentrifuge tube add 20-40 µl of Na2S2O3 (30 g L -1 stock solution) and gently shake manually to "clear" the Lugol (sample should become transparent) -add WGA-Alexa Fluor™ 488 Conjugate and CFW (final concentration of both stains 5 µg mL -1 ) and mix gently. The order of stain addition does not influence the staining results.

Microscope and filter sets
Microscopy analysis of dual-stained WGA-CFW samples can be performed on an inverted epifluorescence microscope equipped with ultraviolet (for CFW) and blue (for WGA-488) excitation filters (Table S2). Alternatively, WGA-fluorochrome conjugates with different excitation wavelengths can be purchased to match the available microscope filter sets.
UV excitation/long pass emission filters enable simultaneous visualization of CFW fluorescence and phytoplankton autofluorescence and can be combined with transmitted light observation (bright-field or DIC [differential inference contrast]). Likewise, blue excitation/long pass emission filters can be used for this purpose in combination with WGA-488. Such a set-up is convenient for identifying and enumerating phytoplankton species (transmitted light/autofluorescence) and attached chytrid sporangia (CFW/WGA) simultaneously within the same field of view, without the need to switch between fluorescent channels and transmitted light.

Objective lenses
Chytrid infections and sporangia associated with microphytoplankton (20-200 µm) are preferably evaluated with an objective lens of 20X/1.5 or 40X (300X or 400X magnification, respectively). Evaluation of chytrids associated with nanophytoplankton (2-20 µm) is preferably done with an objective lens of 60X-60X/1.5 (600X or 900X magnification, respectively). Criteria to identify parasitic chytrids (sporangia) on phytoplankton positive CFW and WGA staining of the chytrid structure, or positive staining with at least one of the stains stained sporangia have a distinctly outlined globose to ovoid shape and are physically attached to a phytoplankton cell (via stalks and/or rhizoids) -phytoplankton cell carrying the stained sporangia shows signs of chloroplast degradation, indicated by low red autofluorescence stained sporangia display very little or rather no Chl a autofluorescence

Text S3. Culturing the model pathosystems
The nine taxonomically different host-chytrid pathosystems ( Table 1 in the main text) were grown as batch cultures in CHU-10 medium in 100 mL Erlenmeyer flasks at constant temperature (17°C). The light-dark cycle was 16:8 h, providing 40 µmol photons m -2 s -1 during the 16-h light phase. The pathosystems were maintained by transferring 0.5-1 ml of the infected culture (infection prevalence 50-95%) to 60 ml of non-infected host cells at 7-14 d intervals [5,6]. The co-cultures were not exposed to continuous shaking, but cells were gently resuspended through manual shaking at least once per week. Table S3. Phytoplankton host and chytrid strains used in this study with available Genbank accession numbers for rbcL (ribulose-1,5-bisphosphate carboxylase) and rRNA genes (18S/28S -small/large subunits of eukaryotic ribosomal RNA, ITS -Internal transcribed spacer).

Text S4. Statistical analysis
Statistical differences between two (non-paired) samples were calculated using the Mann-Whitney test for non-normally distributed data, the Welsh-test for normally distributed data with non-equal variance, and the t-test for normally distributed data with equal variance. Normal distribution was verified using the Shapiro-test and data variance with the F-test. Statistical differences between multiple groups were determined with the Kruskal-Wallis test (if the non-normal distribution was given, with Bonferroni correction for p-value adjustment, also called Kruskal-Wallis H-test or one-way ANOVA on ranks). Data were considered statistically significant when the pvalue was <0.05. Data were processed and plotted in RStudio (3.6.1) and Origin 2021.
The arrows indicate the step-wise workflow. Samples were left without fixative, or preserved with Lugol or PFA. Non-fixed, live samples were stained and counted immediately (0 days), while Lugol-/PFA-preserved samples were analyzed after 1 day (1 d), 4 weeks (4 wk), or 6 months (6 mo). Samples that were stored for 1 d were stained with three different dye concentrations (1, 5, and 25 µg mL -1 ), to test the effect of different dye concentrations on the staining effectivity. Samples that were stored for either 1 d, 4 wk or 6 mo were stained with 5 µg mL -1 , to test the effect of different storage times. Lugol-preserved cells (chytrid sporangia) were counted in Utermoehl chambers, while PFA-preserved samples were counted on PC filters. The fluorescence intensity was analyzed for chlorophyll-autofluorescent host cells and WGA-stained sporangia, but not for CFW-stained sporangia.   Figure S3. Micrographs of Ceratium (dinoflagellate) with associated chytrids. Chytrid sporangia were not visible after CFW-staining (but cellulosic cell walls of Ceratium were CFW-stained), while WGA effectively visualized associated sporangia (sp) and rhizoids. DIC -differential interference contrast.  . Images display two mature sporangia (sp) and four host cells (host). Both cell types could be separated through image analyses in ImageJ, to allow for automatized enumeration of parasite and host cell abundances. The WGA-stained sporangia could be separated after subtracting the host's autofluorescence. For the enumeration of host cells, the bright field image was best applicable.