Filaments of euryhaline Ulva species were collected in an intertidal area of Oslo fjord (59°43′18.2″ N, 10°43′39.1″ E) in August and October 2013. On site, the algae were exposed to extreme salinity fluctuation due to persistent freshwater outflow from a nearby eutrophic stream (Årungselva), seasonal rainfall, and tidal exchange. Fjord seawater salinity was 29 parts per thousand (ppt).

Ulva filaments collected in August 2013 were examined for the presence of oomycetes. Single threads were carefully separated and immersed in Petri dishes containing sterile Milli-Q (MQ) water. After 4 days, aseptate hyphae with appressoria formation were observed on one of the filaments. Part of the filament was stained with cotton blue for better observation of the hyphal growth (Fig. 1A). The other part of the filament was plated onto semi-selective medium for oomycetes CMA P₅ARP (cornmeal agar amended with 5 mg L⁻¹ pimaricin, 250 mg L⁻¹ ampicillin, 10 mg L⁻¹ rifampicin, and 100 mg L⁻¹ pentachloronitrobenzene) (Ferguson and Jeffers 1999). The plate was incubated at room temperature and mycelial growth was transferred onto V8 juice agar (V8A), containing 200 mL V8 juice (Campbell Soup Co., Camden, NJ, USA), 800 mL MQ H₂O, 3 g CaCO₃, and 15 g agar. The isolate was maintained on pieces of V8A submerged in vials with MQ water at 5 and 15°C. The isolate was deposited in the culture collection of NIBIO at Ås, Norway, with accession No. 231095 and in the Westerdijk Fungal Biodiversity Institute collection, Netherland, with accession No. CBS 139622.

Total genomic DNA was extracted from Pythium mycelia grown on V8A or from algal filaments using the Qiagen DNeasy Plant Mini Kit (Hilden, Germany), and DNA segments were amplified by PCR using Taq polymerase and designated primers. For Pythium, the ITS was amplified and sequenced in both directions using primers ITS5 and ITS4 (White et al. 1990), and part of the COI gene using the oomycete-specific primers OomCoxILevup and Fm85mod (Robideau et al. 2011). A fragment of the ribulose-bisphosphate carboxylase large subunit (rbcL) gene was targeted using primers rbcLa-F and rbcLa-R for the Ulva samples (Kuzmina et al. 2012). The amplicons were sequenced by GATC Biotech, Germany, and the raw data were assembled and analyzed using the Lasergene Sequence Analysis Software (DNASTAR Inc., Madison, WI, USA). The program BLASTN (National Center for Biotechnology Information, Bethesda, MD, USA) was used to search existing DNA databases for sequences showing similarity to the obtained sequences for the three genes. The best hits obtained from the blast searches from the sequences of Pythium and Ulva were downloaded and used in the phylogenetic analyses. We also included representative sequences from Pythium specimens (Robideau et al. 2011, Uzuhashi et al. 2015). In the case of Ulva, representative sequences of Ulva were selected from a previous analysis of Kang et al. (2014). The sequences for the three gene sets (ITS, COI, and rbcL) were aligned and manually inspected in AliView (Larsson 2014). Phylogenetic relationships were inferred on each separate aligned matrix with maximum likelihood as implemented in RAxML ver. 8.2 (Stamatakis 2014) using the GTRCAT model, 1,000 bootstrap and default settings. Pythium undulatum and Umbraulva olivascens (= Umbraulva dangeardii M. J. Wynne & G. Furnari) were used as outgroups for the phylogenetic analyses. Obtained trees were edited with FigTree (Rambaut and Drummond 2016). All new sequences generated in this study are deposited in the GenBank database under accession Nos. MK644144, MK644145, MK644146, and MK644147.

The Pythium isolate was subsequently grown on V8A and on potato carrot agar (PCA) (Van der Plaats-Niterink 1981) for morphological observations and growth studies. The maximum, minimum, and optimum temperatures for growth as well as the radial growth per day at different temperatures were determined on PCA as described by De Cock and Lévesque (2004) with small modifications. Agar plates were inoculated in the center and placed at room temperature. After 24 h, the plates were incubated at 1, 5, 10, 15, 20, 25, 30, 35, 37, and 40°C. Two PCA plates were used for each temperature. Radial growth was measured at two perpendicular diameters every 24 h for 3 days. After 3 days, the plates without mycelial growth were placed at room temperature for 1 week to check if growth resumed.

CMA plates were prepared by using sterile MQ water or saline solutions at 5, 10, 15, 20, 25, 30, and 35 ppt. The saline solutions were prepared by dissolving Sea Salt (Sigma, St. Louis, MO, USA) in sterile MQ water. The plates were inoculated at the border and incubated at 20°C in the dark. The radial growth along a dish diameter was measured every 24 h for 4 days. Four Petri dishes were use per each salinity.

Sporulation at different salinities was studied in grass water cultures. The isolated Pythium was grown on PCA plates for 2 days. Thereafter, 1–2 cm pieces of autoclaved lawn grass were placed on top of the growing Pythium colonies. After 24 h, the grass pieces infected with Pythium were placed in 9 mm Petri dishes containing ca. 20 mL of sterile MQ water or saline solutions at 15 and 30 ppt. The saline solutions were prepared by dissolving Sea Salt (Sigma) in sterile MQ water. The plates with the infected grass pieces immersed in different saline concentrations were incubated at room temperature (22–23°C). The presence or absence of vesicles and zoospores was checked after 2 and 5 h. Observations were made with an invert microscope that allows a quick check of the dishes without disturbance of the samples or the need of slides preparations. The dishes were periodically observed for formation of sexual stages for 2 weeks. The experiment was performed twice.

The pathogenicity of the isolated Pythium was tested on a filamentous Ulva species collected in October 2013. The Ulva filaments were carefully examined to make sure that they were not infected with oomycetes. Two pathogenicity tests were performed. In the first test, Ulva filaments were placed on the growing margin of a Pythium colony on PCA. The day after, the infected filaments were transferred to MQ water and two different saline concentrations (15 and 30 ppt). Non-inoculated filaments were placed in MQ water as controls. The filaments were periodically observed for 2 weeks, using an inverted microscope, in order to detect mycelium development and sporulation. To obtain photos, slide preparations were made for use in an upright microscope. After 2 weeks the filaments were transferred onto CMA P₅ARP to check the survival of the Pythium sp. at the different saline concentrations. In the second test, the algal filaments were infected with zoospores. Small Petri dishes (5 cm diameter) were perforated with small orifices in the angle formed by the bottom and the wall of the dishes. These dishes were placed inside larger dishes (9 cm diameter) filled with fresh water. Grass pieces inoculated with Pythium, as described above, were placed inside the smaller dish and algal filaments in the larger dishes. There was no direct contact between the algal filaments and the grass pieces. The Pythium infecting the grass pieces produced zoospores that could move from the small dishes to the larger ones though the orifices. The filaments were observed periodically for 2 weeks as in the former assay.

All data except radial growth under different salinities were analyzed qualitatively. Radial growth data were tested for homogeneity of variance (Levene Statistics) and normality (Kolmogorov-Smirnov test). Thereafter, the statistical significance of differences in radial growth as affected by salinity treatment was tested using analysis of variance (ANOVA, p = 0.05). This was followed by Tukey honestly significant difference (HSD) test (p < 0.05) to find means that are significantly different from each other. Statistical analyses were done using the SPSS 11.5 software (SPSS Inc., Chicago, IL, USA).

The Pythium-infected Ulva filaments collected in August 2013 (accession No. MK644145) had a 99.7% sequence identity to the rbcL gene of Ulva linza Linnaeus (accession No. KP233764.1). The Ulva filaments collected in October 2013 (accession No. MK644146) and used in the pathogenicity test had 100% sequence identity to the rbcL gene of several voucher specimens of U. intestinalis. The Ulva sample where Pythium was isolated from belong to the LPP (linza-procera-prolifera) complex (Shimada et al. 2008). Ulva linza and U. prolifera are currently accepted taxonomically while U. procera (K. Ahlner) H. S. Hayden, Blomster, Maggs, P. C. Silva, Stanshope & Waaland is currently regarded as a synonym of U. prolifera (Guiry and Guiry 2019). This group is taxonomically not well resolved and thus the Ulva sample was not conclusively identified. On the other hand, the Ulva used for pathogenicity tests clustered with U. intestinalis samples with a 100% bootstrap values (Fig. 2). Morphologically, U. intestinalis and U. compressa Linnaeus have distinguishing characteristics that separate them from other filamentous Ulva species. Moreover, sequences of rbcL can clearly distinguish these two species (Hayden et al. 2003).

The ITS sequence of the isolated Pythium (accession No. MK644147) has 98.9% sequence identity to isolates deposited in GenBank as Pyt. dissotocum (accession Nos. KM061705.1 and KM061706.1); however, these two isolates lack COI sequence in GenBank. The COI sequence of the isolated Pythium (accession No. MK644144) has 97.3% identity to Pyt. brachiatum isolate UZ00746 (KJ995594.1). Our phylogenetic analyses based on ITS and COI sequences suggest that the Norwegian isolate belongs to the clade B2 of Lévesque and De Cock (2004) (Fig. 3).

Pythium sp. colonies showed a chrysanthemum-rosette pattern on V8A and PCA (Fig. 1B). The radial growth per day at different temperatures is shown in Fig. 4. At 35°C, limited growth was observed that ceased after day 1; however, growth resumed when culture was transferred to room temperature. At lower temperature, growth ceased at 1°C, but growth resumed when the culture was transferred to room temperature. The main hypha was up to 7 μm in diameter on PCA. The Pythium sp. produced filamentous sporangia that was not easily distinguishable from hyphae. The sexual stage was not observed neither in PCA, V8A, nor in grass in water cultures.

Radial growth at different salinities is shown in Fig. 5. Analysis of variance showed that salinity significantly affected radial growth of Pythium sp. (p < 0.001). Posthoc Tukey’s HSD showed growth was highest at salinity 10 = 15, 10 > 20, 15 = 20 > 25 = 5 > 35 > zero. The sexual stage was not observed at any salinity.

The Pythium was observed to sporulate in fresh and saline waters, but sporulation was faster in freshwater than in saline water. Already after 2 hours of immersion in fresh water, sporulation was observed in the infected pieces of grass in both performed trials. After 2 hours in salinity 15 ppt, the Pythium sporulated in one of the trials and produced only vesicles in the other. In salinity 30 ppt, no sporulation was observed after 2 h; however, in one trial, vesicle formation was observed after 2 h. After 5 h, zoospore release was observed in all treatments.

When U. intestinalis filaments were placed in close proximity of the mycelium of a Pythium colony, they were readily infected. The hyphae that came in contact with the seaweed formed a large amount of appressoria on the surface of the filaments (Fig. 1C). When Pythium in Ulva filaments was cultured in fresh water, it grew very well. After 5 days, the Pythium had fully colonized the algal filaments (Fig. 1D), and after 2 weeks, the filaments became brownish and completed surrounded by Pythium hyphae (Fig. 1E). Conversely, when Pythium and Ulva filaments were incubated in salinity 15 and 30 ppt, no Pythium growth was observed, and after 2 weeks these algal filaments remained green. The non-infected Ulva filaments (controls) placed in fresh water also remained green. After 2 weeks, the infected filaments were plated on CMA P₅ARP and Pythium was re-isolated from freshwater and the two saline concentrations. The Pythium re-isolated from diseased brown filaments was identified by ITS sequencing, confirming that it was the same Pythium sp. that was used for inoculation. Sporulation was never observed in Ulva filaments during the 2-week assay period, neither in fresh water nor in any of the saline concentrations.

In the second test, zoospores produced by Pythium also readily infected the algal filaments. Chemotaxis was not observed; the zoospores seemed to randomly swim and invade the seaweeds. Appressoria formation and mycelial growth on Ulva intestinalis were observed 1 week after inoculation. After 2 weeks, the algal filaments became brownish and there was abundant mycelial growth around the filaments.