Inhibitory properties of crude microalgal extracts on the in vitro replication of cyprinid herpesvirus 3

Extract preparation with the basic protocol

In this work, ASE was chosen for the extraction of potentially bioactive substances from microalgal biomass under defined conditions. Due to reports classifying EtOH as a safe and environmentally friendly solvent suitable for the extraction of natural value products from microalgae^36,45, 95% EtOH was chosen. Considering a possible thermolability of target molecules, the extraction process was initially performed at room temperature (RT) (basic extraction). As summarized in Table 1, the obtained yields varied between 2.8 and 22% depending on the biomass. There was no clear correlation between extraction yield and taxonomic affiliation of the organisms, since both the best and the worst extraction were achieved with species from the division of chlorophyta.

Effects of the basic extracts on virus replication and cellular vitality

To obtain a first assessment of possible inhibitory effects of the basic crude extracts on viral replication, common carp brain (CCB) cells were infected with CyHV-3 along with different extract concentrations and were incubated for 72 h at 25 °C without removing the viral supernatant after inoculation. In this set-up, the samples were only analyzed by characterizing the CyHV-3 DNA copy number, always in comparison to the replication control (without extract addition). With this fast-forward test, viral replication could be monitored, but not the quantity of infectious viral particles. Therefore, samples showing a possible inhibiting effect on viral DNA replication were additionally analyzed regarding their viral titer using the 50% tissue culture infection dose assay (TCID₅₀). To exclude that a reduced viral replication was merely a result of damaged host cells, the cytotoxic influence of the extract on CCB cells was examined using the same extract concentrations.

Since the yields of the basic extracts differed by up to a factor of 9, the maximum concentrations that could be used for the first effectiveness evaluation were very different, as shown in Fig. 1. Furthermore, in case of H. pluvialis and A. platensis, an additional dilution factor must be considered because of the maximal applicable solvent concentration and an increased solvent use due to poor solubility of these basic extracts. We therefore examined whether the extract concentrations were sufficient to determine half maximal inhibitory concentration (IC₅₀), half maximal effective concentration (EC₅₀) and selectivity index (SI), since these values allow for an evaluation of the strength of the antiviral effect in comparison to the cytotoxic effect. In general, the highest possible SI should be aimed for. Acyclovir, which is often used against herpes viruses, for example, can achieve an SI well over 100 for HSV-1^46,47. The IC₅₀, and EC₅₀ values were determined on the basis of the corresponding dose–response curves (Fig. S1, Supporting Material) and are summarized in Table 2. Samples treated with the basic extracts of A. platensis and H. pluvialis showed no or only minor differences in viral DNA concentration compared to the replication control (Fig. 1a,b). The maximum concentrations of 0.6 mg_de mL⁻¹ and 2.2 mg_de mL⁻¹, respectively, were not sufficient to affect the cell vitality or viral replication. Therefore, neither EC₅₀ nor IC₅₀ could be determined. In contrast, using C. kessleri, N. punctiforme, S. obliquus and C. reinhardtii extracts both cell viability and replication of CyHV-3 were affected in a dose-dependent manner (Fig. 1c–f, S1). Depending on the biomass used, these effects showed up differently. Due to the strongest cytotoxic effect, only a low SI could be determined for the extract of S. obliquus (Table 2). The calculated SI for C. kessleri was also comparatively low, since both inhibitory and cytotoxic effects occurred only at high concentrations resulting in high IC₅₀ and EC₅₀ values. The extracts of C. reinhardtii and N. punctiforme showed a comparable antiviral activity to the extract from S. obliquus, but at the same time an approximately threefold lower cytotoxicity, which resulted in higher SI values of 4.3 and 4.1, respectively (Table 2). A more detailed analysis of the antiviral activity at extract concentrations with only moderate effects on the cell viability (Table 3) also showed that the extracts from C. reinhardtii and N. punctiforme were the most promising ones, which led to a reduction of viral DNA by 0.7 and 1.6 orders of magnitude, respectively and virus titer below the limit of detection (Fig. 1e,f). However, in comparison to N. punctiforme, the viral titer was reduced stronger (2.2 orders of magnitude in comparison to 1.4) when using the extract of C. reinhardtii.

Due to the significant reductions of viral DNA copies, an inhibiting influence on the replication of CyHV-3 by the crude extracts of C. kessleri, S. obliquus, N. punctiforme and C. reinhardtii can be assumed, despite some cytotoxic effects of the latter (Table 2). These results are in good agreement with the already reported antiviral effects of N. punctiforme and S. obliquus extracts against enveloped DNA viruses^21,48,49. In the mentioned studies polysaccharides (e.g. nostoflan), short-chain free fatty acids or glycolipids such as SQDG were identified as active compounds. Furthermore, it is believed that these compounds are also the reason for the antiviral effects of C. kessleri and S. obliquus extracts^50,51,52,53. Although, the active substances and their mechanism of action have not yet been investigated in more detail in this work, our data shows that even the crude extracts, produced under basic extraction conditions and not yet purified, have an antiviral effect against CyHV-3.

Extract preparation with an enhanced protocol

As already mentioned, the extraction conditions play a major role in the extraction efficiency and the substances extracted, determined by the used biomass. Depending on the selected solvent and the extraction temperature, different substances can be dissolved from the biomass to different extent. We therefore started a second extraction run to test how modified extraction conditions might affect and improve the yield and the virus inhibiting effect of the extracts. In this second set-up, a focus was put on C. reinhardtii and N. punctiforme after the most promising effects were observed for the corresponding basic extracts. The extraction of A. platensis and H. pluvialis was also re-examined to test whether other extraction conditions would result in higher applicable concentrations and perhaps other bioactive effects, as these two microalgae has shown antiviral properties against various viruses before^32,54. Especially A. platensis is known to be a source of extra- and intracellular polysaccharides that were identified in several studies as potent antiviral substances against e.g. HSV and also CyHV-3^15,49,55. Thus, it was considered that the parameters chosen for the basic extraction might have not been suitable to extract these metabolites. Therefore, the water content of the extraction solvent was increased to 20% (v/v) to facilitate the extraction of more polar compounds. Additionally, the influence of an elevated extraction temperature (100 °C) for the isolation of anti-CyHV-3 bioactives was evaluated in this second experimental set-up called enhanced extraction.

For all four species, the extraction yield could be raised by factors between 1.3 and 5.7 by applying the second protocol as can be inferred from Table 1. These results suggest that indeed more metabolites were extracted from the applied biomass when using an increased water content in the solvent and a higher extraction temperature, which is in good agreement with previously published data^56,57. Moreover, the improved extraction procedure enabled us to perform further studies on the antiviral properties of the selected species with higher concentrations of the extracts.

Effects of the enhanced extracts on virus replication and cellular vitality

Applying the enhanced extracts CyHV-3 infection experiments were performed again using the above-explained procedures, but now with a deviation affecting the incubation post infection. It cannot be ruled out that virus particles, whose penetration has been initially prevented by the extract, subsequently infect cells anyway after prolonged exposure to the cells because of a possible weakening extract effect due to for example oxidizing impacts. In such a case, higher infectious titers would be measured than induced by a fresh extract. Therefore, based on previous studies^32,55,58 and since we assumed that 1 h is sufficient for the virus to enter the cells, further experiments were carried out without constant virus exposure. Thus, to remove unabsorbed viruses after the infection phase the infectious supernatant of all samples and also the supernatant of the replication controls was discarded after inoculation. Subsequently, fresh cell culture medium with the appropriate extract concentrations was added to the cells. As a result, using the enhanced extracts, a dose-dependent inhibition of viral replication could now be observed for the extracts of all four species (Fig. 2). Although, no effects on the CyHV-3 replication were seen in the previous experiments for A. platensis and H. pluvialis, now inhibition was registered when using the enhanced extraction procedure. Applying the A. platensis extract at high concentrations, a reduction in CyHV-3 DNA copy numbers occurred up to approximately 5 orders of magnitude (Fig. 2a). Hence, to verify the data, the infectious viral particles were analyzed using TCID₅₀. For higher extract concentrations viral titers lay even below the limit of detection of the assay, corresponding to a reduction of over 5 orders of magnitude. However, also cytotoxic effects of the A. platensis enhanced extract on the CCB cells were observed. Thus, the EC₅₀ and IC₅₀ values calculated from these results led to a relatively low SI (Table 4). As described above for A. platensis, also for H. pluvialis higher extract concentrations could be applied for the antiviral studies due to the improved extraction procedure. Consequently, inhibiting effects were also revealed for this alga. In this case, significant reductions of viral DNA of up to 4 orders of magnitude (99.99%) occurred, while the viral titer was no longer detectable (reduction of over 4 orders of magnitude) (Fig. 2d). It was noticeable that no significant signs of damage to the CCB cells were detectable via the applied cytotoxicity assay even at such elevated extract concentrations. Up to concentrations of 7.4 mg_de mL⁻¹ the viability of CCB cells was actually exceeding the growth control and decreased to a minimum of yet 90% at the highest extract concentration used here. These high viabilities at elevated extract concentrations led to an increased EC₅₀ value, resulting in a comparably high SI of 60 (Table 4). Such normalized cell viabilities of above 100% were already found when examining the influence of cytotoxic compounds by the MTT assay, which measures the metabolic activity of the cells^59,60. This phenomenon could be explained by the hypothesis of hormesis, constituting a positive biological response to low concentrations of potentially toxic substances^61,62. Overall, using enhanced A. platensis and H. pluvialis extracts and the improved procedure, viral replication was significantly inhibited also without cytotoxic effects (Table 5). For both algae the dose-dependent inhibitory effect on CyHV-3 replication is in good agreement with the above-mentioned data indicating antiviral effects of A. platensis extracts and isolated compounds against herpesviruses such as HSV-1, HCMV and human herpesvirus 6 (HHV6), that were supposed to inhibit viral adsorption and entry into the target cells^25,49. Also, for extracts of H. pluvialis an antiviral effect against HSV-1 was previously reported, Santoyo et al. identified polysaccharides and short-chain free fatty acids as possible antiviral substances³². It was again suggested that the mentioned bioactive substances can hinder the virus to penetrate the cell or even damage the lipid envelope of virions and therefore inactivate the virus^53,63,64. The extracts obtained in this work may also contain such bioactives responsible for the antiviral effects of the crude extracts; however, this must be investigated further.

Moreover, the data presented for these two species showed in which manner the selection of the experimental conditions and analysis methods could affect the outcome of such screening studies. If only the preliminary experiments with A. platensis and H. pluvialis were considered, the inhibitory activity of their extracts on CyHV-3 would remain unrecognized, due to the lower tested extract concentrations and the lack of effects on the viral DNA. However, when varying the experimental conditions and extending the analysis to the viral titer, very promising results regarding an antiviral effect against the targeted virus could be registered.

Furthermore, the influence of the changed extraction conditions was examined for N. punctiforme and C. reinhardtii. As for the preliminary experiments, a dose-dependent reduction of replicated viral DNA could be detected for both enhanced microalgae extracts (Fig. 2b,c). Moreover, a steadily decreasing titer of infectious viral particles was recognizable with increasing extract concentrations. Even at concentrations where replication of viral DNA was not yet affected, a reduced titer was found. Yet, it is noticeable that the IC₅₀ values of the enhanced extracts could not be improved for both organisms.

Quite the contrary, in the case of C. reinhardtii it was even slightly worse (Tables 2 and 4). Thus, the enhanced extraction led to higher concentrations and biomass yields, but to a lower inhibiting effect. This underpins the fact that with crude extracts an increased extraction efficiency does not necessarily correspond to a stronger effect. Nevertheless, whereas the SI values of the basic and enhanced extracts from N. punctiforme were comparable, the SI of the enhanced C. reinhardtii extract was improved by a factor of 1.5 since a reduced cytotoxic effect overcompensated the slight loss of antiviral activity. Table 5 shows that, among all considered microalgae in this study, the greatest inhibitory effect without damaging the CCB cells could be reached with the C. reinhardtii extract resulting in a reduction of viral DNA by 3.4 (> 99.9%) and viral titer by over 4 (> 99.99%) orders of magnitude. Although no analysis of the specific extract composition of C. reinhardtii was performed in this work, there is some data in the peer-reviewed literature on bioactive compounds contained in extracts of this species. In addition to carbohydrates such as sulphated polysaccharides, (glyco)proteins and (glyco)lipids were previously isolated from C. reinhardtii depending on the selected extraction conditions^{51,65,66,67,68}. These bioactive substances could be able either to hinder virus-cell adsorption and membrane fusion or damage the lipid envelope of virions and therefore inactivate the virus, as observed for such compounds originating from other species. Though the first examinations of the total protein and lipid content of the selected extracts were performed in this work (Table S4 and S5, Supporting Material), no conclusions about active fractions could be drawn from these data. Thus, to identify the bioactive substances that are responsible for the antiviral effect of the crude extract of C. reinhardtii, further investigations including fractionation, antiviral activity and content analysis of the extracts are necessary.

Although various bioactive compounds and effects have been described in literature for extracts of C. reinhardtii, no inhibitory impact on the replication of viruses have been reported so far apart from the results obtained in this work. Again, this suggests that in studies regarding possible bioactive metabolites, both the choice of experimental conditions and the methods of analysis play an important role. Therefore, it is possible that C. reinhardtii was not yet considered as an interesting candidate as a source of antiviral compounds because, for example, other extraction methods or only one-sided analyses were used for evaluation and the corresponding data was not published. Hence, even if antiviral effects of extracts of certain species on selected viruses cannot be directly determined, these species should not be excluded from further investigations prematurely.

From the point of view of a future application and further studies, especially non-cytotoxic extract concentrations that are showing inhibitory effects on the viral replication are of great interest. With this in mind, the C. reinhardtii extract inhibited viral replication the most with unaffected cell viability (Table 5). On the other hand, the enhanced extract of H. pluvialis showed the lowest cytotoxicity and a high inhibitory influence on CyHV-3 which led to a relatively high SI of 60, which is significantly higher than the SI of 6.4 reached with C. reinhardtii (Table 4). However, keeping in mind, that crude extracts were used in this study, the latter might be optimized in further research for example by the identification, isolation and purification of the biologically active compounds. In general, biological extracts contain a variety of compounds, meaning that it is always unclear which biological effects are caused by which molecules. Thus, it has always to be analyzed whether a measured cytotoxic activity is a feature of the same molecule as the antiviral activity in order to avoid excluding active compounds from screenings only due to high cytotoxic effects of their crude extracts⁶⁹. Therefore, the isolation of the active antiviral or inhibitory agent or rather substance class might help to eliminate substances which affect the host cells cytotoxically and therefore improve the SI accordingly.

Since the SI of antiviral extracts and substances depends not only on the species, extraction method and the degree of purification, but also on the virus and the host cells used, cross-study comparisons must be done with caution and are better suited for comparing orders of magnitudes than concrete values. In a similar study performed with hot water extracts from A. platensis, the replication of HSV-1 and HSV-2 was inhibited by 50% at a concentration of 0.3 mg mL⁻¹ showing IC₅₀ values in the same order of magnitude and an SI of 26^49,58. However, when using purified products such as sulphated polysaccharides and SQDG, significantly lower IC₅₀ values were achieved where replication of HSV and HCMV was inhibited by 50% at concentrations in the range of μg mL^−121,27. Moreover, exopolysaccharides extracted and purified from A. platensis resulted in an IC₅₀ of less than 2.1 μg mL⁻¹ but also in a low SI of 3 for CyHV-3⁷⁰. Using H. pluvialis extracts, 99 μg mL⁻¹ of the polysaccharide-rich fraction reduced virus infectivity by 50% resulting in an SI of 19³².

Summarizing, using the enhanced extraction protocol for the four selected species—A. platensis, N. punctiforme, C. reinhardtii and H. pluvialis—antiviral effects were observed after application of the extracts in the range of lower mg_de mL⁻¹, although so far only unpurified crude extracts have been applied. A substantial reduction in virus replication (based on both viral DNA and titer) could be proven for all of them even at still high cell viabilities. We could demonstrate that ASE enables the extraction of bioactive substances and also that the application of elevated temperature for the isolation process did not negatively affect the antiviral properties of the extracts tested here. On the contrary, elevated temperature and water content in the extraction solvent led to a higher extraction yield and enabled the identification of antiviral effects even for extracts of the species which, based on the preliminary experiments, seemed not to have such properties. Within the investigated species, the clearest antiviral effects were registered for the C. reinhardtii and H. pluvialis extracts with the highest SI values calculated in this study. To the best of our knowledge, this is the first report of antiviral properties of extracts of C. reinhardtii. Furthermore, our work shows that the quantification of the actual infectious virus titer in addition to the viral DNA copy number is of great importance in order to assess the antiviral properties of such extracts. While in some cases viral DNA could still be produced by the extract treated and CyHV-3 infected cells, no infectious viral particles could be detected via titration assay. Moreover, a titer reduction under extract application was often more pronounced in comparison to the reduction of viral DNA. Thus, if only viral DNA was monitored the antiviral properties of some extracts might have remained unnoticed.

Influence of pre-treatment on the antiviral activity of the enhanced extracts

In order to obtain information about the virus replication phase affected by the inhibiting extracts, different pre-incubation set-ups were examined. Either CCB cells or CyHV-3 suspensions were pre-incubated for 1 h with the selected concentrations of extracts before inoculation. In addition, a combined mode was examined, where both, cells and virus suspension, were pre-incubated with the extracts. After this pre-incubation, the assays for antiviral activity were carried out as described above. Using this strategy, it should be possible to differentiate, whether the inhibiting agents influence either the cells or the viral particles or even both. For comparison, samples with no extract pre-incubation (corresponding to the previous set-up for the investigation of antiviral properties of the extracts) and the untreated replication controls were prepared along with the other samples.

The results of the pre-treatment experiment with the enhanced extracts from A. platensis, N. punctiforme, C. reinhardtii and H. pluvialis are illustrated in Fig. 3a–d. For all samples where one of the four extracts was applied, varying inhibitory effects on viral DNA and titer were observed. Depending on whether cells, the virus suspension, both or neither of them were pre-incubated with the corresponding extract, the reductions in viral DNA or titer occurred to different degrees. Thus, while the least effect on viral infection was always observed when no pre-incubation was applied, the pre-treatment of both cells and virus prior to the inoculation with respective extracts resulted in significantly lower viral DNA and titer for all four species. When pre-incubating the extracts of A. platensis and N. punctiforme with either cells or virus suspension, only a slight influence on the number of copies of viral DNA was registered. Simultaneously, the same use of these extracts caused a more pronounced reduction of infectious virus particles compared to viral DNA—as it has also been observed in previous experiments. The application of C. reinhardtii extract to either cells, virus or both led to a reduction of viral DNA and viral titer below the limit of detection. However, even stronger effects were observed for each scenario in this set-up when using the extract of H. pluvialis. All in all, a more pronounced interference with viral replication could be recognized when the cells were in contact with the extract before inoculation. Consequently, it can be assumed that the extract affects components of the cells like impairing viral adsorption to the cells.

However, the data obtained especially for the combined pre-incubation set-up suggest that the antiviral effect is not solely based on the inhibition of virus binding to the host cells, as viral replication was much more impeded in the combined mode in comparison to just treating the cells. Most likely, also following replication steps such as virus assembly or release can be affected by the extracts used in this work. Furthermore, since infectious virus particles were also not detected after the virus stock was pre-treated with the extracts of C. reinhardtii and H. pluvialis, some virucidal properties of the latter can be assumed. Thus, the inhibitory effects on the replication seem to be a combination of the interference of bioactive substances with the attachment of virus particles to the target cells, preventing viral entry as well as a possible activation of some cell defense mechanisms. Therefore, the production of viral proteins and thus in a consequence the assembly of mature virions could possibly impeded. Since a drop in the titer below the initially added values was observed for most extracts, they can also have virucidal effects, at least for the virus investigated here. Nevertheless, based only on the presented here data a clear designation of mode of action and an appointment of the active compounds of these extracts is not yet possible. Altogether, the pre-incubation data confirmed the inhibitory effects of the crude extracts of A. platensis, N. punctiforme, C. reinhardtii and H. pluvialis and forms a basis for identifying the bioactive substances and their mechanisms of action in further studies.

In summary, we showed that ethanolic extracts of six different chlorophyta and cyanobacteria obtained by ASE inhibited the in vitro replication of CyHV-3, suggesting bioactives with antiviral properties against the selected virus present in these crude extracts. Although further research regarding the identification and purification of the antiviral substances and investigation of their modes of action is necessary, the work presented here confirms the immense potential of phototrophic microorganisms when searching for antiviral agents. In particular, further examination of the more closely studied microalgal species A. platensis, H. pluvialis and C. reinhardtii, which are already well characterized and used in food and pharmaceutical industry, is very promising regarding their future use for treatment against CyHV-3.