Impact of time of harvest and drying method on antimicrobial activity of Saccharina latissima against two Staphylococcus aureus strains

ABSTRACT Antibiotic resistance is one of the greatest public health threats of our time, and the bacterium Staphylococcus aureus, of which there are numerous drug-resistant and drug-sensitive strains, is a pathogen of worldwide concern. Scientists are turning their focus to underexplored marine ecosystems to identify novel antibacterial agents effective against S. aureus. Here, we report inhibition of S. aureus strains Newman and USA300 by extracts from Saccharina latissima (sugar kelp), grown and harvested in the Western Gulf of Maine, USA. We examined how time of harvest throughout the growing season as well as the drying method pre-extraction affected the antimicrobial activity of the kelp extracts. Optimal antimicrobial activity was observed at the beginning of April (203 days since sporing), when increased water pH and higher salinity levels were also observed. Oven-dried crude extracts showed greater inhibition against S. aureus Newman, whereas freeze-dried crude extracts demonstrated greater inhibition against S. aureus USA300. Overall, our data indicate that cultivated S. latissima from the Western Gulf of Maine possesses significant value-added antimicrobial activity and identify early spring as an optimal harvest time to harness antimicrobial activity.

The cultivation and harvesting of the edible brown macroalga, S. latissima, commonly known as sugar kelp, is a rapidly growing sector of the New England aquaculture and food industries (Grebe, Byron, Gelais, Kotowicz, & Olson, 2019;Piconi, Veidenheimer, & Chase, 2020). Identification and optimization of any potential antimicrobial properties would add value to this important commodity. Little is known of potential antimicrobial properties of S. latissima harvested from coastal New England, or whether exogenous factors or post-harvesting processing conditions might influence its antimicrobial activity. Therefore, our objectives were to examine the effect of time of harvest and post-harvest drying method on antimicrobial activity of S. latissima harvested from the southern coast of Maine, USA. Specifically, we harvested on eight dates during the kelp growing season (March, April and May 2019) and subjected the kelp to either oven-drying or freeze-drying prior to chemical extraction. Our results provide insight into the optimal time to harvest farmed sugar kelp and the optimal drying method prior to extraction to ensure and preserve the highest degree of antimicrobial activity.

Materials and methods
Saccharina latissima cultivation S. latissima "seed" was produced using methods described by Redmond, Green, Yarish, Kim, and Neefus (2014). In short, wild reproductive S. latissima was collected and stressed via desiccation in the laboratory to induce zoospore release on 11 September 2019 (termed "sporing"). The meiospores were allowed to settle on thin, nylon line for 24 hours. Then the line was transferred to aquaria maintained at optimal growth conditions and the gametophyte stages produced juvenile sporophytes. On the 5th of December, the kelp sporophytes (1-2 mm in length) were outplanted on a 60 m longline located near Wood Island, Saco Bay, ME, USA ( Figure 1). Spacer buoys were used to maintain the longline at 2 m below the water surface.

Saccharina latissima sampling and drying
S. latissima was harvested eight times between March and May 2019 (176-259 days since sporing). Whole thalli were cut from the longline, placed into plastic bags, transported in a cooler at 8°C, and refrigerated until time of processing (no longer than 12 h). To prepare them for drying, the fronds were gently cleaned with tap water to remove epiphytes and then randomly sorted into one of two drying groups with equal mass. The first group was oven-dried in a 60°C in a Thermo Scientific™ Heratherm™ Advanced Protocol Oven for 5 days. The second group was frozen overnight at -20°C, then placed in a LABCONCO™ FreeZone freeze dryer at -58°C for 48 h. Each group was kept in the respective drying unit until 90% water loss was achieved for both groups. Water loss was determined by calculating the change in mass of the algae before and after drying. The dried algae were then ground to flakes using a mortar and pestle. Individual 4 g portions of dried, ground algae were immediately stored in glass scintillation vials at -80°C until further use. All samples were stored under argon with parafilm seal and protected from light with aluminium foil until extraction.

Measurement of water temperature, salinity and pH in situ
Water temperature at the farm site was recorded using Hobo Pendant Temperature/Light 8 K Data Loggers (Part 210 #: UA-002-08) suspended from the longline. At each kelp sampling event, water samples were also collected for salinity and pH measurements. Salinity was quantified using a Cole-Parmer RSA-BR90A Refractometer (0-90%). A HACH benchtop metre (model #: PW172KB0703F01) calibrated to certified standards was used to measure pH.

Generation of crude Saccharina latissima extracts
Dried, ground algae (4 g) was transferred to glass test tubes and 20 ml of dichloromethane was added, adapting procedures of Deveau et al. (2016). Three replicates of the extraction were run simultaneously. The triplicate algae-dichloromethane samples were incubated in a 40-50°C water bath for 5 min and then individually agitated with a vortex mixer for 20 s to facilitate compound extraction. After repeating the heating and mixing cycle twice, the samples were placed in an ice bath for 3 min followed by a 3 min centrifugation at 400 rpm and 25°C to remove particulates. Supernatants were transferred to individual clean round bottom flasks. Dichloromethane (20 ml) was added twice as the procedure was repeated for three cycles in total. The dichloromethane was evaporated from each flask using a Büchi rotary evaporator at 45°C under vacuum. The round bottom flasks were placed on a vacuum pump for 24 h. The resulting extract concentrates were weighed and percentage recovery calculated, before being resuspended in methanol, a solvent with less volatility than dichloromethane that is compatible with antimicrobial assay experiments. Methanolic extracts were brought to a final concentration of 10 mg ml -1 for use in disc diffusion assays. Extract solutions were stored at -80°C until use.

Bacterial strains and growth conditions
The methicillin-sensitive Staphylococcus aureus (MSSA) strain Newman was sourced from the American Type Culture Collection (ATCC 25904). The methicillinresistant S. aureus (MRSA) USA300 strain was obtained from the Network on Antimicrobial Resistance in Staphylococcus aureus (NARSA NR-46070). All bacterial strains were sub-cultured from -80°C stocks in Mueller Hinton (MH) broth at 37°C in a shaking incubator (Deveau et al., 2016). Prior to performing disc diffusion assays, bacterial colonies were isolated on MH agar plates and cultures were evaluated for purity.

Disc diffusion antimicrobial activity assays
Disc diffusion assays were performed using the guidelines of the Clinical Laboratory Standards Institute (CLSI) and as previously described (Deveau et al., 2016). Filter paper discs (7 mm) were infused with individual methanolic algal extracts (15 μl of 10 mg ml -1 extract to achieve 150 μg extract per disc), vancomycin (30 μg per disc), ampicillin (10 μg per disc), chloramphenicol (5 μg per disc) and methanol (15 μl per disc, vehicle control). Discs were applied to the surface of MH agar plates after the plates were inoculated with 100 μl of each bacterial culture adjusted to a 0.5 McFarland standard (approximately 10 8 CFU ml -1 ). Plates were incubated at 37°C for 24 h and then assessed for zones of growth inhibition around treated discs. When zones of inhibition were present, zone diameters were measured and corrected for disc diameter (7 mm). For all studies reported, disc diffusion zones of inhibition are the average of triplicate experiments.

Statistical analysis
Differences in the zone of inhibition were assessed across independent variables (time points and drying method) using linear mixed-effects models. A suite of all models incorporating all possible parameter combinations was ranked using Akaike's Information Criterion (AIC). Pairwise comparisons were then conducted for the top model (ΔAIC > 2). All error bars represent standard error. Effects of environmental factors (water salinity and water pH) on zone of inhibition were assessed across independent variables (time-points and drying methods) using linear regression models. We set α at 0.05 throughout. All statistical analyses were conducted in R v2.1 (Rstudio Team, 2016).

Effect of harvest date on antimicrobial activity of crude Saccharina latissima extracts
An average calculated percentage recovery of 0.5-3.9% (Supplementary table S1) was achieved for each triplicate extraction, across harvest dates. There was minimal difference in the recovery values for individual harvest dates, but the date of S. latissima harvest had a significant impact on antimicrobial activity of the crude methanolic kelp extracts (p < 0.05). When observing the zones of inhibition of the crude S. latissima extracts for the first two harvest dates of the Both water salinity and pH were greatest (31 psu; pH 8) on the harvest date of the 2nd of April 2019 (Figure 3). We observed a significant relationship between salinity and antimicrobial activity against both S. aureus Newman (y = 0.071x + 0.140, p < 0.05; Supplementary figure S2) and USA300 (y = 0.116x -1.26, p < 0.05; Supplementary figure S3). The maximum antimicrobial activity of the crude methanolic kelp extracts for the season occurred in April when salinity was high (Figure 3). The decrease in antimicrobial activity between the 17th and 29th of April against both strains of S. aureus (Figure 2a, b) correlate with the drop in water salinity and pH (Figure 3). We also observed a significant relationship between water pH and antimicrobial activity of algal extracts against both S. aureus Newman (y = 5.273x-39.19, p < 0.05; Supplementary figure S4) and USA300 (y = 5.852x -44.0, p < 0.05; Supplementary figure S5). On the harvest date of the 17th of April, when a decrease in salinity was observed, there was also a decrease in water pH, which correlated with the decrease in antimicrobial activity (Figures 2, 3). The salinity of the seawater increased as the growing season continued, whereas the pH continued to decrease. However, on the last harvest date of the season (28th of May) increased antimicrobial activity was observed but the pH of the seawater was comparable to the pH observed on the 2nd of April harvest date, and water salinity was markedly higher on 28th of May in comparison to the harvest dates between 17th and 29th of April. The increase in both pH and salinity on the last harvest date corresponds to the increase in antimicrobial activity on the 28th of May 2019 (259 days post sporing). Water temperature was measured as 2.3, 3.1, 3.9, 4.3, 5.0, 6.3, 6.4 and 9.9°C for the eight sequential harvest dates. No significant relationship between the water temperature and antimicrobial activity was observed (p > 0.05).

Effect of drying method on antimicrobial activity of crude Saccharina Latissima extracts
Crude methanolic extracts from kelp collected on the first two harvest dates (6 March, 2019-176 days post sporing, and 13 March, 2019-183 days post sporing) exhibited no zones of bacterial inhibition and therefore no antimicrobial activity against either strain of S. aureus, regardless of drying method (Figure 2; Supplementary table S1 and figure S1). However, for both drying methods, an increase in antimicrobial activity was observed on the 25th of March 2019 (195 days post sporing) against both S. aureus strains. Importantly, both oven and freeze-dried kelp extracts showed the largest zones of bacterial inhibition (against both S. aureus Newman and USA300) and reached the maximum activity for 2019 growing season on the harvest date of the 2nd of April, 2019 (203 days post sporing). On the 2nd of April, the oven-dried extracts possessed 1.2-times more antimicrobial activity than the freeze-dried extracts against S. aureus Newman (Figure 2a), whereas the freeze-dried extracts possessed 1.3-times more antimicrobial activity than the oven-dried extracts against S. aureus USA300 (Figure 2b). There was a decrease in activity for both drying methods for the remainder of the harvest dates, save the 28th of May (259 days post sporing) when a significant increase in antimicrobial activity against S. aureus was seen in both drying methods (p < 0.05). Antimicrobial activity against S. aureus USA300 in ovendried extracts had two times greater antimicrobial activity than freeze-dried extracts (p < 0.05). Staphylococcus aureus Newman (a) and USA300 (b). A disc diffusion assay was used to test the antimicrobial activity of crude extracts prepared from S. latissima harvested on different dates of the growing season and dried via freeze drying (−58°C for 48 hours) or oven drying (60°C for 5 days). Filter paper discs were saturated with extracts (150 µg/disc) or methanol (vehicle control) and applied to MH agar plates immediately after inoculation with S. aureus USA300. Data are presented as a diameter of the zone of inhibition (mm) surrounding extract-saturated discs, which is indicative of bacterial growth inhibition. Zone sizes are corrected for diameter of disc as well as the vehicle control (methanol, always less than 1 mm/disc). Different letters (a, b, c, d, e, f) represent significant differences using pairwise comparisons (p < 0.05). NZ represents no zone of inhibition measured. Error bars indicate standard error.

Discussion
In an effort to determine an optimal time of harvest and drying method for preserving and extracting S. latissima antimicrobial compounds, we tested the antimicrobial activity of crude methanolic S. latissima extracts harvested at eight different dates throughout the growing season and dried the algae either in an oven (60°C) or freeze drier (-58°C). Other studies have shown that differences in the chemical composition and antimicrobial activity of algal extracts can be seen depending on when the algae are harvested during their growing season (Ehrig & Alban, 2015;Liu et al., 2018;Overland et al., 2018;Schiener et al., 2015). Previous reports have also shown that the method by which the algae is dried can impact algal chemical composition (Ehrig & Alban, 2015;Liu et al., 2018;Milledge et al., 2014;Overland et al., 2019;Patarra, Paiva, Neto, Lima, & Baptista, 2011;Schiener et al., 2015). Even though lipid and phenolic content were not directly measured in this study, secondary metabolites are common in macroalgae and not only possess antimicrobial activity, but also to be affected by various exogenous factors (Abou Zeid et al., 2014;Beaulieu et al., 2015;Cox et al., 2010;Deyab & Abou-Dobara, 2013;Kadam et al., 2015;Perez et al., 2016;Rodrigues et al., 2015;Shannon & Abu-Ghannam, 2016).

Effect of date of harvest
In this study, data from Figure 2 (and Supplementary figure S1 and table S2) indicate that crude methanolic S. latissima extracts possessed the lowest antimicrobial activity in early March and late April, the highest activity in early April, and moderate antimicrobial activity at the end of May. Other research indicates that lipid and fatty acid content in algae is highest in the winter and spring when the algae are relatively young, and the water is relatively cold (El Maghraby & Fakhry, 2015;Nomura et al., 2012;Sappati et al., 2019;Schiener et al., 2015) while lower lipid and fatty acid contents are observed in the summer when the algae are more mature, and the water is relatively warmer (El Maghraby & Fakhry, 2015;Nomura et al., 2012;Sappati et al., 2019;Schiener et al., 2015). With the water temperatures in Saco Bay averaging 3.3°C in March, 5.2°C in April, and 8.2°C in May, our data suggest that a change in the chemical composition of polar extracts in S. latissima also occurs during March, April and May when water is colder and individuals are younger. These trends could help explain the increase in antimicrobial activity in early April, suggesting that the young algae in colder water (4.3°C ± 0.2) potentially contain biologically active compounds with greater antimicrobial activity in comparison to the older algae (259 days post sporing) in warmer water (9.9°C ± 0.4). However, our study shows no significant relationship between water temperature and antimicrobial activity of crude methanolic sugar kelp extracts (p > 0.05). Salinity, as well as temperature, within Saco Bay fluctuates from season to season, especially in the spring (April) due to snow melt. Ehrig & Alban (2015) showed that S. latissima growing in higher salinity environments produced biologically active compounds with . Salinity and pH measurements at 2 metres below sea level at same depth as longline. At each harvest date, salinity was measured using a Cole-Parmer RSA-BR90A Refractometer (0-90%), and pH was measured using a HACH benchtop metre (model #: PW172KB0703F01) calibrated to certified standards.
increased activity. Furthermore, Floreto, Hirata, Ando, & Yamasaki (1993) found that algae contain higher concentrations of lipids and fatty acids when exposed to higher salinity environments. Water pH is related to salinity, such that when seawater drops in salinity, pH usually decreases as well. These observations suggest that water salinity and pH may impact the types of compounds produced by S. latissima from season to season. These data and our experimental results suggest that when the salinity is high in Saco Bay, S. latissima contains higher amounts of bioactive compounds that are potent against S. aureus Newman and USA300. Therefore, the types of biologically active compounds that can be extracted from Maine sugar kelp may also vary by growth season.

Effect of drying method
In our study, we observed greater inhibition of S. aureus Newman and USA300 by crude methanolic S. latissima extracts from marine algae harvested in the middle of our growing season, regardless of the drying method used. Our data suggest that the biologically active compounds present in S. latissima extracts when oven dried (60°C) have more of an effect against S. aureus Newman (1.2 times more activity than the freeze-dried extracts in best collection, 2nd of April). Conversely, the biologically active compounds present in S. latissima extracts when freeze-dried can have more of an effect against S. aureus USA300 (1.3 times more activity than the oven dried extracts, 2nd of April). The collections from the 2nd of April possessed significantly more antimicrobial activity against both strains of S. aureus for both drying methods than any other time point. These data suggest that regardless of the drying method on this harvest date, bioactive compounds have significant antimicrobial activity against each bacterial strain. Other research by Sappati et al. (2019) found that depending on whether S. latissima was freeze-dried, sundried, or dried at varying temperatures with varying humidity, different chemicals, bioactive compounds, and components of the kelp were preserved. These data together further suggest that drying method post-harvest has an overall effect on the chemical composition of S. latissima resulting in varying levels of biologically active compounds. Since drying methods have been shown to have differing effects on the chemical composition of S. latissima, this could therefore help us explain the differences seen in antimicrobial activity between oven dried algae and freezedried algae against different strains of S. aureus (Zubia et al., 2008;Olofsson et al., 2012;Sappati et al., 2019;this study). For the algal extracts tested against S. aureus Newman, the oven-dried methanolic extract possessed significantly more antimicrobial activity than the freeze-dried methanolic extracts, suggesting the biologically active compounds preserved during oven-drying effectively inhibit this bacterium (Cadieux, Vijayakumaran, Bernards, McGavin, & Heinrichs, 2014;Olofsson et al., 2012;Sappati et al., 2019;Zubia et al., 2008). Furthermore, for the algal extracts tested against S. aureus USA300, the freeze-dried methanolic extracts showed significantly more antimicrobial activity at the early April harvest date, suggesting that the compounds preserved from freeze-drying are more potent against this bacterium. While not explored in our work, others have found that enhanced antimicrobial activity of extracts may be attributed to synergistic antimicrobial effects achieved from interacting bioactive compounds (Cadieux et al., 2014). This increase in bioactivity observed at the end of May could ultimately be a result of higher concentrations of biologically active compounds in the crude extract during that time of year, and/or less masking of activity by other compounds during the transition period from spring to summer, as suggested by Zubia et al. (2008). However, the significant difference between the antimicrobial activity in both drying methods in early April compared to late May against both bacterial strains suggests that early spring is still the best collection time regardless of the drying method used (Cox et al., 2010;Shannon & Abu-Ghannam, 2016).
In conclusion, our results indicate that crude methanolic extracts from sugar kelp cultivated in southern coastal Maine possess greater antimicrobial activity in early spring (203 days post sporing) when water pH and salinity are high. The data do not indicate a correlation between antimicrobial activity and water temperature, but show that pH, salinity, and kelp age each contribute to the observed bacterial inhibition (Figures 2, 3). S. latissima extracts possess the greatest antimicrobial activity in early spring regardless of drying method. However, because extracts from the freeze-dried and oven-dried sugar kelp optimally inhibited different bacterial strains, we posit that these two drying methods may each preserve unique compounds and/or different quantities of biologically active compounds. Overall, this study suggests that antimicrobial activity of S. latissima extracts may be enhanced through careful selection of harvest windows and drying methods. Additional studies of bioactive properties of southern Maine S. latissima constituents will lead to a broader understanding of the medicinal value of a crop that is important to an evolving aquaculture industry (Cole et al., 2017) and promote the discovery of novel bioactives that inhibit pathogenic Staphylococci as well as other targets.