Tunisian inland water microflora as a source of phycobiliproteins and biological activity with beneficial effects on human health

The microalgae strains investigated in this work were collected in eight Tunisian inland water bodies (river, lagoon, dam reservoir, spring) located in the north, center, and south of the country (Table 1).

All samples were taken to the laboratory where each strain was subjected to capillary isolation in liquid medium, using an inverted microscope (Leica microsystems, Wetzlar, Germany). A single individual was isolated (a single cell, colony, or filament) in 300 μl of the appropriate culture medium, and growth of each isolate was achieved by gradually increasing the culture volume from the initial isolation to the final batch. All strains were cultivated in 2 l laboratory flasks, using a monospecific batch culture system under sterile conditions. Growing was carried out at 25 ± 1°C temperature in a thermostatically controlled room, and using BG11 (Rippka et al. 1979) or CONWAY (Blancheton 1985) media, depending on the origin of each strain (Table 1). Culture media were sterilized by autoclaving at 120°C for 20 min before use. Growth was conducted in a 16:8 h light:dark regime under illumination of cool white fluorescent tubes with a light intensity of approximately 25 μmol m⁻² s⁻¹. Cells were harvested only in the exponential growth phase by centrifugation (3000 rpm; Thermo IEC CL31R Multispeed centrifuge; Massachusetts – USA), then lyophilized using a Christ Alpha 2-4 LD Plus freeze-dryer (Harz, Germany) after a cultivation period between 18 and 25 days. Samples were then stored at −20°C until analysis.

For screening of the investigated biological activity, the lyophilized biomass of each strain was ground using a mortar and pestle for a few minutes and extracted adding 1:10 w/v of solvent mixture (dichloromethane–methanol, 1:1, v/v) with the assistance of an ultrasonic bath (Branson 2800, Branson Ultrasonics Corporation, Danbury, USA) (110 w; 40 kHz) for 15 min at room temperature, and then filtered. The filtrate was subsequently collected and the remaining microalgal material was ground in a mortar, mixed with the same extraction buffer and filtered again until the completion of three successive extraction cycles for maximum extraction yield. The solvents were then evaporated using a Rotavapor (KNF, Freiburg, Germany) and the obtained extracts were stored at −80°C between each analysis.

Antioxidant activity of microalgal extracts was assessed using 1,1 Diphenyl 2-Picryl Hydrazyl (DPPH) assay for free radical scavenging activity according to Cheel et al. (2007). Samples of microalgae extracts were prepared in methanol at a final concentration ranging from 62.5 μg ml⁻¹ to 2 mg ml⁻¹. They were subsequently mixed with 2 ml of (60 μM) DPPH solution prepared in methanol, then incubated in the dark for 30 min at room temperature. Absorbance measurements were taken against a blank at 517 nm using a UV/Vis spectrophotometer and butylated hydroxytoluene (BHT) as a standard. Free radical scavenging activity was determined according to the following equation: (2) %inhibition=(AB−ASAB)×100 \% inhibition = \left({{{{A_B} - {A_S}} \over {{A_B}}}} \right) \times 100 where A_B = absorbance of blank (containing all reagents except the sample) and A_S = absorbance of a sample.

IC₅₀ was expressed as the concentration of extract inhibiting 50% of the DPPH free radical scavenging activity.

Antileishmanial activity was evaluated in 96-well plates on promastigotes of Leishmania major (gLC94) and Leishmania infantum (LVMon). In their stationary growth phase, promastigotes were maintained on RPMI-1640 medium (Gibco), containing 100 μg of streptomycin ml⁻¹ and 100 U penicillin ml⁻¹ (Gibco), supplemented with 10% of fetal bovine serum FBS (Gibco) and incubated at 27°C in a humidified atmosphere with 5% CO₂. Promastigote culture (2 × 10⁵ parasites ml⁻¹) was added to each well. Two-fold serial dilutions of microalgae extracts were added to a final concentration, ranging from 7.81 μg ml⁻¹ to 1 mg ml⁻¹, and plates were incubated at 27°C for 72 h (Essid et al. 2015). The parasite viability was assessed after addition of 10 μl of 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) (Sigma, St. Louis, MO, USA) to 10 mg ml⁻¹. After 4 h of incubation in the dark and in the presence of MTT, the medium was removed and formazan crystals were solubilized by adding 100 μl of pure DMSO. The relative amount of formazan blue produced by viable parasites was determined spectrophotometrically using an ELISA plate reader (Synergy HT, Bio-TEK) at 560 nm. Amphotericin B (98% purity, from Sigma-Aldrich, USA) was used as a positive control for all tests. Antipromastigote activity was expressed as the IC₅₀ value, defined as the concentration of a compound that inhibits the growth of promastigotes by 50%. Three replicates were performed, and the average was calculated from IC₅₀ values obtained for each separate experiment (Essid et al. 2015).

All solvents and reagents used for the extraction procedures, antioxidant activity, and antileishmanial assay were purchased from Sigma-Aldrich (GmbH, Steinheim, Germany). Amphotericin B was purchased from Sigma (98% purity, Sigma–Aldrich, USA).

Assays were conducted in triplicate and data were expressed as the mean ± standard error of three independent measurements. To assess the significance of linear correlations between the antioxidant activity and total phenolic and flavonoid amounts in each microalgal extract, Pearson’s correlation analysis was carried out using r-Project software. Differences were considered significant at p < 0.05. The inhibitory concentration was calculated from the GraphPad Prism dose– response curve (statistical programmer) obtained by plotting the percent inhibition against concentrations.

The antioxidant activity of microalgal extracts against DPPH free radicals are shown in Table 3. The higher the DPPH radical scavenging activity of a compound, the lower its IC₅₀ value (inversely proportional), since the IC₅₀ represents the amount of antioxidants needed to reduce the concentration of a free radical by 50%.

The lowest IC₅₀ values were found in Chroococcus sp. and Leptolyngbya sp1. (212 and 263 μg ml⁻¹, respectively), thus indicating their better ability to reduce DPPH radical scavenging activity compared to the other strains investigated. The other strains tested showed higher inhibitory concentrations, ranging from 680 μg ml⁻¹ (Dunaliella sp.) to 1 mg ml⁻¹ (Arthrospira platensis). The results also indicated that certain strains exhibited no detectable antioxidant activity at the maximum extract concentration (Planktothrix agardhii, Limnotrix sp., Spirulina sp., and Cylindrospermopsis raciborskii).

The antibacterial activity of microalgae extracts was studied against human pathogenic bacteria (Table 4). Most strains of microalgae showed no detectable antibacterial activity against all species of bacteria tested, except for Dunaliella sp. (Duna-GV), which showed moderate inhibitory activity against methicillin-resistant Staphylococcus aureus (MRSA). This antibacterial capacity was materialized by an inhibition zone of 16 mm, corresponding to a minimum inhibitory concentration (MIC) of 375 μg ml⁻¹.

The antileishmanial capacity of microalgae extracts was tested against the promastigotes L. major (gLC94) and L. infantum (LVMon) (Table 5). The unicellular green algae Dunaliella sp. (Duna-GV) exhibited the best antileishmanial activity against both leishmaniasis pathogens. The crude extract of Duna-GV reduced the viability of L. infantum and, to a lesser extent, that of L. major (IC₅₀ = 151 and 284 μg ml⁻¹, respectively). It was followed by the Oscillatoriales cyanobacterium Limnothrix sp. that showed interesting IC₅₀ values against the two pathogens (188 and 377 μg ml⁻¹ for L. infantum and L. major, respectively).

According to the results obtained, the Cylindrospermopsis strain Cylind-NB occupies an exceptional position among all the strains studied, being by far the most productive of total phycobiliproteins, with a total amount exceeding 3000 μg ml⁻¹. More interestingly, the PC level in this Nostocales cyanobacterium is more than ten times higher than the two other constituents of phycobiliprotein combined (2873 μg ml⁻¹ of PC against 277 μg ml⁻¹ for APC and PE, combined). Compared to the results obtained by other authors, the Cylindrospermopsis strain tested in the present study appears to be a very rich source of PC. For example, Horvath et al. (2013) reported a PC amount of 5 μg ml⁻¹ in a Cylindrospermopsis raciborskii strain isolated from surface water samples collected from Lake Balaton (Hungary), using the same extraction method (repeated cycles of freezing and thawing combined with sonication). Aside from the strain effect, which is the main factor explaining the huge difference observed between the two isolates, other factors may also have intervened, notably the high light intensity applied by Horvath et al. (2013), i.e. 40 μmol m⁻² s⁻¹, compared to the present work (25 μmol m⁻² s⁻¹). In fact, the negative effect of light intensities above 25 μmol m⁻² s⁻¹ on the cellular amount of PC in cyanobacterial strains has been reported in several studies such as Takano et al. (1995) and De Oliveira et al. (2014). These studies demonstrated that the level of incident light is one of the factors that most affects the cyanobacteria metabolism, and that lower light intensities were more advantageous in terms of total yields of phycobiliproteins produced by cyanobacteria isolates of the genera Nostoc and Synechococcus. It has been suggested that this phenomenon is probably due to a strategy of preventing photo-oxidative damage caused by the production of free radicals in cells. Apart from the Cylindrospermopsis genius, lower amounts of PC (ranging from 15 to 264 μg ml⁻¹) have also been reported by Basheva et al. (2018) in 18 cyanobacterial strains belonging mainly to the genera Microcoleus, Nostoc, Phormidesmis, and Leptolyngbya. These values were obtained under optimal culture conditions, in particular the culture mode (extensive culture), culture medium (liquid Z-medium), pH (8.5), and temperature (24 ± 1°C). On the other hand, a high PC level, quite close to our result, was obtained by Bachchhav et al. (2016) in an Indian strain of Arthrospira platensis under a mixotrophic regime and yellow LED illumination. The authors recorded a PC amount of 2500 μg ml⁻¹ produced by A. platensis, thus proving its significant potential for use in the biotechnological production of this pigment.

To the best of our knowledge, this work is the first to highlight the capacity of the cyanobacterium Cylindrospermopsis raciborskii to produce huge PC amounts, making it an effective alternative natural source for the biological production of one of the most valuable microalgal accessory pigments. In fact, many potent bioactivities, with extremely beneficial effects on human health, have been attributed to PC, such as the positive influence on the immune system, especially the defense mechanisms against infectious diseases. For instance, Kawamura et al. (2004) found that cyanobacterial PC enhances the biological defense activity against infectious diseases through sustaining functions of the mucosal immune system, and reduces allergic inflammation by suppressing antigen-specific IgE antibodies. It was concluded that the PC-rich cyanobacterial biomass or its processed products are not only useful dietary supplements, but they also strengthen the defense mechanisms against infectious diseases, food allergies, and other inflammatory diseases.

In addition to the previously described beneficial effects, some epidemiological studies have demonstrated the efficacy of PC as an inhibitor of cancer cell proliferation in vivo and in vitro. For example, Liu et al. (2000) reported that PC of cyanobacterial origin significantly inhibited the growth of chronic myeloid leukemia/chronic human blast crisis K562 cells in a dose-dependent manner. Tumor cells were arrested in the G1 phase and blocked to progress through the S phase, suggesting that the inhibitory effect of PC against the growth of K562 cells may have occurred through pathways other than apoptosis, and that the altered expression pattern of the c-myc protein may be involved in such inhibition.

Moreover, the neuroprotective role of PC against global cerebral ischemia/reperfusion damage in gerbils has been reported (Penton-Rol et al. 2011). PC was found to exhibit a potent protective effect against neuronal cell death in the hippocampus, and significantly improve locomotor behavior and survival in gerbils after just one week of reperfusion. In another study, Rimbau et al. (2001) found strong evidence for a remarkable protective role of PC against cell death caused by 24 h withdrawal of potassium and serum (K/S) in rat granular cerebellar cell cultures. These findings highlight some promising perspectives on the use of our Cylindrospermopsis raciborskii strain as a reliable source of potent and non-toxic natural drugs for the treatment of neuronal damage induced by oxidative stress in neurodegenerative disorders, ischemic strokes, Alzheimer’s disease and Parkinson’s disease.

When comparing our results with previous studies, it can be observed that, in general, the levels of antioxidant activity against DPPH radicals obtained in the present work remain relatively less pronounced. IC₅₀ values ranging from 67.5 to 119.6 μg ml⁻¹ and from 30.72 to 102.47 μg ml⁻¹ were reported by Ijaz & Hasnain (2016) and Babic et al. (2016), respectively, for the antioxidant activity of crud extracts of several cyanobacterial strains against DPPH radicals. Similarly, Safafar et al. (2015) recorded 10-fold higher inhibitory activity for Anabaenopsis and for Arthrospira (IC₅₀ = 90 and 100 μg ml⁻¹, respectively) than that recorded for the same strains in our study. On the other hand, the most active isolate against DPPH radicals characterized in the present work (Chroococcus sp.) showed much higher antioxidant activity (6 times higher) than that found by Singh et al. (2017) for the same species (IC₅₀ = 1260 Vs 212 μg ml⁻¹).

The particularly interesting antioxidant activity recorded in Chroococcus sp. and, to a lesser extent, Leptolyngbya sp1. may suggest that they are good candidates to be considered as new natural sources of effective and non-toxic antioxidants, with a number of extremely beneficial health effects. In fact, many authors have reported the importance of antioxidant and radical scavenging activity of cyanobacterial origin among the most interesting therapeutic and nutritional functions that are particularly sought after by the food and pharmaceutical industry in recent decades (Hossain et al. 2016). Nevertheless, extracts of these cyanobacterial strains remain remarkably less active compared to the synthetic standard BHT used as a positive control in the present work (IC₅₀ = 17.34 μg ml⁻¹).

This finding appears to be consistent with previous studies, e.g. Blagojevic et al. (2018), where an extremely low IC₅₀ value was reported for BHT (about 10 μg ml⁻¹), reflecting much stronger DPPH scavenging efficiency compared to 10 microalgal isolates essayed in their study.

Nevertheless, it should be noted that such extraordinary antioxidant efficiency of BHT among other synthetic commercial antioxidants, especially butylated hydroxyanisole (BHA), tertbutyl hydroquinone (TBHQ), or propyl gallate (PG), is not without a significant cost. In fact, the use of BHT and BHA as food additives to neutralize or delay lipid peroxidation in the food industry has been shown to be associated with several serious health risks. Some epidemiological studies focusing on the medium and long-term health effects of BHT and BHA on animal models have proved that these products are characterized by high toxicity and carcinogenic or mutagenic effects (Lanigan & Yamarik 2002; Agyei et al. 2015).

These adverse health effects attributed to BHT and BHA and other commonly used synthetic antioxidants have put more pressure on nutraceutical, pharmaceutical, and food industries, and have made the need to identify new sources of natural antioxidants as safe alternatives even more appealing and urgent.

Regarding total polyphenols, very few studies have measured the polyphenol content in microalgae compared to other cellular constituents (Goiris et al. 2012). Our results showed TPP levels varying between undetectable values and a maximum of 8.9 mg EqGA/g DW (Chroococcus sp.). These levels are within the range reported by Safafar et al. (2015), who assessed the TPP content in different microalgal species belonging to several taxonomic classes (between 4 and 8 mg EqGA/g DW). However, Goiris et al. (2012) and Hossain et al. (2016) reported lower levels, not exceeding 4 mg EqGA/g DW, in extracts from more than 35 industrial microalgal strains. The observed differences can be explained by extrinsic factors such as culture conditions (photoperiod and nutrient concentration) and culture method (autotrophic or mixotrophic) as well as genetic and physiological factors, especially strain intrinsic characteristics, its origin, and even the stage of development at harvest time.

Statistical analysis revealed that there was no significant correlation between antioxidant activity and polyphenol content in the crude extracts, either in terms of PPT (r = 0.142; p = 0.782), or FT (r = 0.214; p = 0.661). The highest DPPH scavenging activity was observed in Leptolyngbya sp1., the extract of which contained the lowest concentration of PPT. On the other hand, Arthrospira platensis was the richest strain in FT and yet exhibited the lowest DPPH scavenging activity among all the investigated strains. This indicates that probably other compounds contribute strongly to the antioxidant effects observed for the strains tested in the present study. This could be attributed, for instance, to phenolic compounds that cannot be detected by the method adopted here or to other non-phenolic compounds. The later hypothesis seems to be the most consistent with a previous study (Cepoi et al. 2009) where the high antioxidant activity recorded in the biomass of Arthrospira platensis and Nostoc linckia was suggested to be related to considerable amounts of carotenoids and tocopherols, which were better extracted at higher concentrations of ethanol (55 and 70%). In general, these microorganisms are known for their ability to synthesize various phytochemical compounds such as tannins, terpenoids, UV screening compounds, steroids, and polyunsaturated fatty acids, thus forming a complex mixture of compounds in which antioxidants can act synergistically in a manner that is only partly understood.

The antibacterial activity assessment against several human pathogenic bacteria revealed a moderate antibacterial potential of the unicellular green alga Dunaliella sp. (Duna-GV) only against methicillin-resistant Staphylococcus aureus (MRSA), with a MIC value of 375 μg ml⁻¹, corresponding to an inhibition diameter of 16 mm. This growth inhibitory capacity of such a pathogen, considered the most virulent of the genus Staphylococcus, can be qualified as moderate on the basis of the classification criteria provided by Askun et al. (2013). Indeed, the antibacterial activity of plant extracts is considered to be high for MICs below 100 μg ml⁻¹, moderate between 100 and 625 μg ml⁻¹, and inactive for MICs above 800 μg ml⁻¹. The result is consistent with the previous study by Maadane et al. (2017) on several microalgal species, including Dunaliella sp. and Dunaliella salina, collected along the Moroccan coast, who concluded that the ethanolic extracts of Dunaliella strains exhibited moderate inhibitory activity, targeting in particular the pathogens P. aeruginosa and Escherichia coli.

According to previous research published by Maadane et al. (2015), ethanolic extracts of Dunaliella sp. contained significant amounts of carotenoids, up to 10.8 mg g⁻¹ DW, suggesting a possible role of these compounds in the total antimicrobial potential observed in this strain. The last hypothesis seems to be in agreement with Bhagavathy et al. (2011), where different carotenoids extracted from Chlorococcum humicola exhibited important antimicrobial activity, affecting both Gram positive and Gram negative bacteria.

The antileishmanial activity of microalgal extracts was evaluated against promastigote forms of the parasites L. infantum (LVMon) and L. major (gLC94). The results showed that the unicellular green algae Dunaliella sp. (Duna-GV) exhibited the highest inhibition against the two Leishmania pathogens, with IC₅₀ values of 151.2 μg ml⁻¹ and 284.8 μg ml⁻¹ for L. infantum and L. major, respectively.

At present, marine organisms are increasingly recognized by scientists as an interesting natural source of new bioactive products and a promising alternative in the therapy and control of leishmaniasis. However, very few reports have been published on the antileishmanial capacity of microalgal extracts (Pereira et al. 2015).

To the best of our knowledge, this is the first study to demonstrate the antileishmanial effect of the genus Dunaliella. It strongly suggests that this eukaryotic unicellular green microalga may represent a good candidate to be used in the treatment of cutaneous and visceral leishmaniasis caused by the pathogens L. major and L. infantum. This appears to be of particular interest for Tunisia, a country endemic to CL and VL, where CL caused by L. major continues to pose a major health threat and causes thousands of new cases annually, with alarming infection rates reaching up to 60% of the population of certain villages (WHO 2017).

In addition, the results revealed that most of the isolates tested exhibited a particularly targeted antileishmanial activity against L. infantum more than L. major parasites. This could also be of particular importance in the wider context of the Mediterranean region, where L. infantum is the sole causative agent of canine leishmaniasis as well as cutaneous and visceral forms of human leishmaniasis. They are described as endemic in 22 countries of the Mediterranean zone, where they are strongly considered as a serious public and veterinary health problem (Campino et al. 2006). Furthermore, in the Iberian Peninsula in Southwestern Europe, infections with L. infantum in humans are mainly associated with immunosuppressive diseases, in particular with co-infection with the AIDS virus, HIV. According to Campino & Maia (2010), no effective human or canine vaccine against this pathogen has been developed, and chemotherapy is the only way to control cases of infection.