Freshwater Cyanobacterial Toxins, Cyanopeptides and Neurodegenerative Diseases

1. Introduction

Cyanobacteria are photosynthetic gram-negative bacterial common inhabitants of diverse aquatic freshwater, marine, and terrestrial environments. They are ancient prokaryotic life forms on Earth having photosynthesis which contributed oxygen to our atmosphere over 3.5 billion years ago, and can survive in different, sometimes frequently changing environmental conditions. Cyanotoxin production is considered to be an ancient trade exceeding 2.5 billion years of age [1].

The ability of cyanobacteria to produce cyanotoxins and their ubiquity in freshwater ecosystems with increasing demands upon water resources require better detection and a more comprehensive understanding of the cyanobacterial distribution and its impact on animals and humans. Although cyanobacterial cell numbers change seasonally, the toxins can persist in water for several months, extending low-dose exposure [2]. Aquatic organisms (e.g., grazers/herbivores, fish, mammals) may consume aquatic plants with high concentrations of cyanotoxins, and trophic transfer may happen to higher-order organisms though, for some toxins, biodilution, and not biomagnification may be a predominant process in the food webs [3]. The detectable presence of cyanobacterial toxins in animal tissues has been found to be associated with mass mortalities of animals, including cows, dogs, and sea mammals [4,5,6,7,8,9,10,11,12], birds [13], and some human cases [14,15]. The terrestrial vertebrates affected by cyanotoxins are more diverse than was thought before [16,17]. Moreover, cyanotoxins demonstrate sublethal effects, including growth inhibition in zooplankton and aquatic plants, macroinvertebrates, and aquatic plants [18].

The predicted climate changes favor increased water temperatures, anthropogenic nutrient loadings, and freshwater cyanobacterial frequency, duration, and size of algal blooms [19,20,21]. The cyanobacterial abundance has increased disproportionally relative to other phytoplankton since 1945 [22]. Though our knowledge about cyanotoxins mostly comes from temperate and tropic areas, Arctic regions undergo the most pronounced and rapid climate changes, and high-latitude lakes support cyanobacteria blooming and cyanotoxins production [23]. Moreover, the occurrence and intensity of near-surface phytoplankton harmful algal blooms (HABs) have been increasing across the world [24,25] due to the eutrophication or nutrient enrichment of water bodies [26,27]. The potential presence of low doses of cyanobacterial toxins in drinking water is likely to be a continuing problem.

2. Cyanobacterial Toxins

Cyanobacteria produce a variety of toxins. Traditionally, they are divided based on functional properties into main groups: hepatotoxins, neurotoxins, dermatotoxins, and cytotoxins. Lipopolysaccharides (components of the cyanobacteria cell wall), due to their toxic effects, are classified as a separate group called – endotoxins [21,28]. The toxins found in cyanoHABs include microcystins and nodularins, and neurotoxins such as anatoxins (anatoxin-a, homoanatoxin-a, guanitoxin), ciguatoxins, saxitoxins, ß-methylamino-L-alanine (BMAA) and its isomers (2,4-diaminobutyric acid (2,4-DAB) and N-(2-aminoethyl)-glycine (AEG) [21]. More than 80 cyanobacteria species are known to be toxigenic, and assays for the detection and toxicity of cyanotoxins continue to develop.

The major source of cyanotoxins for humans is drinking water. However, guidelines for water quality monitoring are limited to microcystins [29] and very few neurotoxins. Furthermore, cyanobacteria may be present in fields used for livestock grazing or can be fed to livestock directly [30,31,32,33], which aggregates in cow’s milk [34,35] or bird eggs and increase human exposure [36]. There is an urgent need to detect other cyanobacterial toxins in drinking water and food and to understand how they are involved in the pathogenesis of chronic diseases such as neurodegenerative disorders.

Many cyanobacterial toxins still have to be discovered. For major toxins groups, new variants can be found and characterized. Thus, a structural variant of anatoxin-a, dihydro-anatoxin-a has been recently identified in many samples of benthic cyanobacteria, even exceeding the abundance of anatoxin-a [37]. Vacuolar spongiform myelopathy has recently been linked to aetokthonotoxin (AETX) from epiphytic cyanobacterium Aetokhonos hydrillicola that is growing in man-made water bodies of the southeastern United States [38]. This finding warrants further research into the potential toxins produced by epiphytic and benthic cyanobacteria species.

2.1. Microcystins (MC) Family

A full structural chemical analysis of MCs was achieved in the 1980s through a combination of spectroscopy, nuclear magnetic resonance, mass spectrometry, and amino acid analysis and demonstrated that the chemical structure of microcystins consists of a cyclic heptapeptide with two variable and five relatively conservative amino acids biosynthesized non-ribosomally via an MC synthetase gene cluster [39]. A universal nomenclature system was suggested based on the positions of amino acid residues 2 and 4 (denoted as X and Z in the original structure of microcystin i,e, microcystin -XZ) [40]. By 2019 the identification of at least 279 different MC congeners was reported in the literature [39].

Between cyanobacterial toxins, MCs are the most diverse group and the best described, though MC-LR and MC-RR—are the only two widely researched. Minority MC congeners demonstrate different toxicokinetic and toxicodynamic features [41]. Variations in vivo toxicity between MC congeners can be attributed to the differences in their uptake by organic anion transporting polypeptides (OATP) transport versus serine/threonine protein phosphatases (PP) 1 and 2a inhibition [42]. The toxicity of MCs depends on variations in their chemical structure and ranges over six orders of magnitude [43].

MC toxicity affects not only the liver but also the brain [44] and other organs. Multiple neurotoxic effects of MC-LR were demonstrated using multiple biological models, including birds, fishes, and mammals [45,46,47]. For example, using murine brain cell line as a model, congener-dependent pronounced neurodegenerative effects were identified (MC-LF>>MC-LW>MC-LR) [48].

2.2. BMAA (β-N-methylamino-l-alanine) and Isomers

BMAA is a non-proteogenic amino acid produced by all known groups of free-living and symbiotic cyanobacteria [49]. The isomeric forms of BMAA, such as 2,4-diaminobutyric acid (2,4-DAB) and N-(2-aminoethyl) glycine (AEG), can also be found in different species of cyanobacteria, including Anabaena, Leptolyngbya sp., Oscillatoria sp., Merismopedia sp., and Microcystis aeruginosa [50]. The isomers are detected in nature along with BMAA but are less studied. BMAA isomers are neurotoxic [51]. Recent research using larval zebrafish as a biological model identified 2,4 DAB as a more potent neurotoxin than AEG and BMAA [52].

BMAA has been shown to contribute to protein misfolding, enzyme inhibition, and neuroinflammation [53]. BMAA toxic effects were found to be related to the misincorporation of serine in multiple human proteins [54,55] which can lead to the formation of inclusion bodies in neurons [56]. Since some serine sites, such as tau serine residue 422 have been identified of key importance in neuropathologies [57], misincorporation in such sites can be particularly detrimental. Even a low level of misincorporation rate (1 per 10,000 codons) can lead to neurodegeneration in a rodent model [58]. Several groups found in their in vitro and in vivo studies that BMAA leads to the overexpression TDP-43 (TAR DNA-binding protein 43) encoded by the TARDBP gene [59,60,61].

Most animal models’ BMAA studies were concentrated on investigating BMAA effects in the brain and other organs. Using mice Xie and co-authors reported that less than 1% of total BMAA from adult mice plasma was taken in the brain [62], i.e., BBB is not easily permeable for BMAA. The mechanism of neurotoxicity may involve a direct action on the NMDA receptor, activation of glutamate receptor 5, and induction of oxidative stress.

Recently, Han and co-authors [63] found that BMAA can serve as a substrate for human alanyltRNA synthetase (AlaRS), avoiding the intrinsic editing activity of AlaRS, acting as a competitive inhibitor, and comprising the editing ability of AlaRS. Terminally differentiated cells, such as neurons, are particularly susceptible to mistranslation and accumulation and forming of misfolded and aggregated proteins [58,64].

2.3. Other Cyanobacterial Neurotoxins

Traditional neurotoxins from cyanobacteria with acute effects include alkaloid or organophosphorus compounds such as (a) anatoxin-a and homologs which affect nicotinic acetylcholine alkaloid toxins and muscarinic acetylcholine receptors [65,66]; (b) saxitoxins can be produced by both dinoflagellates and by cyanobacteria from several genera including Aphanizomenon, Cylindrospermopsis and Dolichospermum; (c) guanitoxins which are similar in structure to organophosphates and able to irreversibly inhibit acetylcholinesterase [67]; their presence was also found in desert assemblages [68,69]. Cylindrospermopsin toxin- another frequent finding during fish kills is a highly biologically active alkaloid consisting of a tricyclic guanidine moiety combined with hydroxymethyluracil [70] interferes with cellular metabolism and causes hepatotoxic and genotoxic effects, as well as neurotoxic effects [71]. There are an increasing number of research on different groups of cyanotoxins from cyanobacteria genera—Nostacales (Nostoc, Hapalosiphon, Fischerella, Anabaenopsis, Aphanizomenon, Gloeotrichia, Cylindrospermopsis, Scytonema, Raphidiopsis, Cuspidothrix, Nodularia, Stigonema, Calothrix, Cylindrospermum, and Desmonostoc species) and others [72].

It is known that cyanobacteria are capable of producing different toxins which can be present during the same HABs [73]. It is not clear, however, how the environmental factors regulate the abundance of different MC congeners and isoforms of other toxins in a bloom [39]. In natural lake cyanobacteria, species composition can be complex, and cyanotoxins can be represented by a structural variety of toxins and cyanopeptides. Thus, BMAA can co-occur with its isomers (2-DAB and AEG) and have synergistic neurotoxic effects, as demonstrated in in vitro cell line experiments [74]. Moreover, the joint presence of MC-LR and BMAA leads to their interaction in vivo and to the neurotoxic effect enhancement [75]. The development of methods allowing for the assessment of multiple toxins during algal blooms is needed [76].

2.4. Cyanopeptides

The majority of cyanobacterial secondary metabolites are peptides or include peptidic substructures. Cyanopeptides are non-ribosomal peptides rich with posttranslational modifications and non-proteinogenic amino acids and consist of linear, cyclic or multicyclic molecules with basic, depsipeptidic, or lipopeptidic structures. More than 500 cyanopeptides range from app. 300 to 2000 Da have been structurally identified by 2019 [77]. Natural selection did not minimize the pool of peptides but favors the production of a wide array of different peptide structures. The biosynthetic non-ribosomal pathways of peptide synthesis are evolutionarily ancient and precede synthetic pathways of higher plants and animals. During HABs, cyanobacteria produce a tremendous amount of diverse cyanopeptides; however, their ecological significance is unclear. They can happen at surface waters in the same nanomolar concentration as MCs, exhibit toxicity towards grazers in the same micromolar range as MCs, and their production is synchronized with Microcystis sp. While the abundance of MCs can be monitored successfully, more studies on cyanopeptides appearance and persistence during blooms [77,78] and their potential for chronic toxicity are needed. The bioactive compounds produced by cyanobacteria are not limited by peptides and also include alkaloids, cyclophanes, terpenes, lactones, etc. Cyanobacterial compounds have a broad bioactive spectrum, with many acting as serine protease inhibitors, trypsin and chymotrypsin inhibitors, and anti-cancer compounds, capable of modulating infectious diseases [79].

2.5. Chronic Effects of Cyanobacterial Toxins

The epidemiological studies of human health impacts of chronic cyanobacterial toxins exposure are nascent. They have been associated with neurodegenerative diseases, including Alzheimer’s disease (AD), Parkinson’s disease (PD), and amyotrophic lateral sclerosis (ALS) [80,81,82,83,84]. Clusters of ALS and ALS-like diseases have been reported in relation to cyanobacteria in Guam, France, Japan, New Hampshire, and Wisconsin (summarized in Table 1). The slow onset of neurodegenerative diseases (time distance between exposure and possible outcome) and problems with the assessment of environmental exposure interfere with our understanding of the role and significance of cyanotoxins in neurodegenerative diseases.

The link between BMAA and neurodegenerative diseases is yet to be further elucidated. Several studies have reported the presence of BMAA post-mortem in the brain tissues of patients who die from ALS/PD [89,109]. However, in ALS/Alzheimer’s disease [110], and other studies were not able to identify BMAA presence [111,112]. The ALS/PD neurodegenerative disorder, formerly hyperendemic in Guam-USA, Kii-Japan, and Papua-Indonesia associated with several cycad food toxins, including BMAA [113], has now been identified in both aquatic and terrestrial eco-systems in North America [82,83], The Baltic Sea [114], France [101], Sweden [115], Peru [116], and Qatar [117]; and is produced by several different cyanobacteria [49], diatoms [118], and dinoflagellates [119].

Residential exposure to environmental pollutants may play an essential role in the etiology of ALS, which is supported by non-random distribution by addresses of ALS patients [120].

2.6. Stability of Cyanotoxins

Many cyanotoxins possessing cyclic peptide structure are resistant to chemical degradation [121], highly stable, and may persist in aquatic ecosystems for weeks and months [2,122]. Thus, MCs can be retained in mussels (Mytilus californians) for up to eight weeks [123]. The high stability of MCs, cylindrospermopsin [124,125], and other cyanotoxins over a wide range of pH and temperature might have significant consequences for aquatic ecosystems and contribute to bioaccumulation of toxins to higher levels of food chains. These peptides are synthesized non-ribosomally and may contain non-proteinogenic amino acids [126,127,128].

Toxins undergoing attenuation via photodegradation may vary depending on the type of toxin, HABs timing, and environmental conditions [129]. Though cyanotoxins are resistant to chemical degradation, recent advances in molecular microbial communities research have found that toxic cyanoHABs favor the specific members of bacterioplankton with degrading abilities towards cyanotoxins [130,131]. The strains of the bacterial genera Sphingomonas (majority of MC-degrading bacteria), Rhodococcus, Brevibacterium, Burkholderia, Mycobacterium, Pseudomonas, Novosphyngobium and others can degrade MCs in time scale from hours to days [132,133,134]. The genomes of some MCs-degrading bacteria are sequenced [135,136], and mlr gene cluster have been implicated in playing a prominent role in the sequential hydrolysis of MCs peptide bonds [137,138].

Similar to MCs, cylindrospermopsin toxin can also be degraded by bacteria isolated from cyanobacterial blooms (Bacillus sp., Aeromonas sp.) [139,140] and by some probiotic bacteria [141]. Not only bacteria but fungi demonstrate algicidal activities such as Trichoderma citrinoviride degrading MCs [142]. Furthermore, Mohamed and co-authors [143] summarized data on six fungal species with biodegrading activities against MCs. Between zooplankton grazers, metazoans, such as Daphnia and Cyclops, are also susceptible to cyanotoxins [144,145]. Protozoa, on the other hand, are highly resistant to cyanotoxins and demonstrate great potential in controlling harmful cyanobacteria and improving phytoplankton composition in eutrophic waters [146,147].

2.7. Data gaps in Cyanotoxins Analytical Methods

Analytical techniques (selectivity and sensitivity, fraction analysis, quality control) play a critical role in assessing the cyanotoxins effects [148,149]. There are significant data gaps in analytical methods, including (a) the absence of all the relevant standards [150]; (b) the need for validated methodologies for cyanotoxins outside the water samples: (c) the need for standardization of cyanotoxins in multi-center monitoring programs [151]; (d) the need for new technologies allowing simultaneous identification of as many toxins as possible; (e) and the need to improve robustness and a detection limit of detection methods Another critical challenge is analyze cyanotoxins faster and feasible in situ [152]. Unsuitable analytical methods may partly explain the lack of consensus over the widespread presence of some cyanotoxins (BMAA) in aquatic ecosystems [153].

Recently, high-resolution mass spectrometry (HRMS/MS) has become more available for researchers. Current methods rely on defined cyanotoxins and cyanopeptides targets and are generally inappropriate for detecting and identifying of emerging novel compounds. The recent approach of the non-targeted analysis of pollutants and toxins in water focus on a comprehensive workflow for the acquisition and treatment of the data generated after liquid chromatography coupled with high-resolution mass spectrometry (LC-HRMS) analysis [154,155,156,157]. So-called suspect screening identifies novel compounds, including cyanotoxins and cyanopeptides based on the exact mass (m/z), presence of one or more charge states (z = 1, 2, etc.), expected isotope pattern and common adducts) as well as secondary fragmentation (MS2) even without reference chemicals [158,159,160,161].

Since 1957, when the first MC was detected, many methods have been developed to analyze environmental samples for cyanotoxins [162]. Nowadays, the number of studied cyanotoxins and analytical methods for their qualitative and quantitative detections have increased. Detection approaches vary in terms of accuracy, sensitivity, and specificity. The most commonly used techniques for cyanotoxins detection are enzyme-linked immunosorbent assays (ELISA), protein phosphatase inhibition assay (PPIA), molecular assays - polymerase chain reaction (PCR), and quantitative real-time PCR (qPCR) for toxins producing genotypes for cyanobacteria identification, liquid chromatography (LC) and high-performance liquid chromatography (HPLC) combined with different detectors. Among these, liquid chromatography-mass spectrometry (LC-MS) takes a special place because it identifies the target cyanotoxins with high accuracy at a significantly low detection level [163,164]. Moreover, despite the structural diversity of cyanotoxins, LC-MS allows for determining groups of toxins simultaneously. That is one extra advantage of using this detection technique integrated with bioassays and molecular assays in complex environmental samples for complete water quality assessment [165,166].

Table 2. Chemical detection methods for cyanotoxins in water samples.

№	Cyanotoxins	Detection Techniques	Sensivity		Reference
			LOD	LOQ
1.	MC-LR and 2 congeners	UHPLC-MS/MS	0.02–0.04 ng/mL	–	[167]
2.	MC-LR and 11 congeners	UHPLC-MS/MS	–	0.2 µg/L	[168]
3.	MC-LR and 4 congeners	LC-MS/MS	0.005–0.0817 µg/L	0.005– 0.0817µg/L	[169]
	Nodularin		0.0048 µg/L	0.0048 µg/L
	Anatoxin-a		0.0001 µg/L	0.0004 µg/L
	Cylindrospermopsin		0.0001 µg/L	0.0004 µg/L
4.	BMAA	UHPLC-MS/MS	0.02 pg/µL	0.05 pg/µL	[170]
	2,4-DAB		0.04 pg/µL	0.13 pg/µL
5.	MC-LR and 7 congeners	LC-MS/MS	–	0.04–0.5 µg/L	[171]
	Anatoxin-a		–	0.02 µg/L
	Cylindrospermopsin (and deoxyCYN)		–	0.01 –0.02 µg/L
	Saxitoxins (4 congeners), GTX (5 congeners), decarbamoylgonyautoxin, N-sulfogonyautoxins-1 and -2		–	0.1–2 µg/L
6.	MC-LR and 11 congeners	HPLC-MS/MS	0.01±0.01–0.19±0.2 μg/L	0.04±0.04–0.64±0.65 μg/L	[172]
	Nodularin		0.04 ± 0.02 μg/L	0.13 ± 0.06 μg/L
7.	MC-LR and 2 congeners	HPLC-UV/PDA	3–4 μg/L	9–13 μg/L	[173]
8.	MC-LR and 2 congeners	HPLC-HRMS	0.002 μg/L	–	[174]
9.	Anatoxin-a,	HILIC-MS/MS	0.004 ng/mL	0.01 ng/mL	[175]
	Cylindrospermopsin		0.07 ng/mL	0.23 ng/mL
	Saxitoxin		0.01 ng/mL	0.04 ng/mL
	MC-LR and 4 congeners	RPLC- MS/MS	0.02–0.08 ng/mL	0.07–0.28 ng/mL
	Nodularin		0.05 ng/mL	0.18 ng/mL
10.	MC-LR and 5 congeners	UHPLC-MS/MS	–	0.046–0.052 µg/L	[176]
	Nodularin		–	0.049 µg/L
	Cylindrospermopsin		–	0.052 µg/L
11.	BMAA	UHPLC-HRMS	5 ng/L	10 ng/L	[177]
	DAB		3 ng/L	5 ng/L
	AEG		2 ng/L	5 ng/L
	BAMA		5 ng/L	10 ng/L
12.	MC-LR and 5 congeners	HPLC-MS/MS	0.0003–0.0009 µg/L	–	[178]
	Cylindrospermopsin		0.0005 µg/L	–
	Saxitoxin, dcSTX		0.0009–0.0013 µg/L	–
13.	BMAA	LC-MS/MS	10 ng/L	–	[179]
14.	MC-LR and 5 congeners	LC-MS/MS	0.04–0.8 μg/L	0.1–2.3 μg/L	[180]
	Nodularin		0.3 μg/L	0.9 μg/L
	Anatoxin-a		0.27 μg/L	0.81 μg/L
	Cylindrospermopsin		0.14 μg/L	0.4 μg/L
15.	MC-LR and 2 congeners	HPLC-DAD	0.08–0.15 µg/l	–	[181]
16.	MC-LR and 11 congeners	LC-MS/MS	0.001–0.007 μg/L	0.003–0.020 μg/L	[182]
	Nodularin		0.002 μg/L	0.006 μg/L
	Anatoxin-a		0.001 μg/L	0.003 μg/L
	Cylindrospermopsin		0.001 μg/L	0.003 μg/L
17.	MCs	UPLC-MS/MS	0.005 µg/L	–	[183]
	Anatoxin-a		0.02 µg/L	–
	Cylindrospermopsin		0.02 µg/L	–
	Saxitoxin		0.8 µg/L	–
	BMAA		0.03 µg/L	–
18.	BMAA	LC-MS/MS	0.030 μg/L	0.096 μg/L	[184]
19.	MC-LR and 11 congeners	on-line SPE – UHPLC-HRMS	5–37 ng/L	15–130 ng/L	[185]
	Anatoxin-a		15–18 ng/L	50–60 ng/L
	Homoanatoxin-a		11–12 ng/L	30–40 ng/L
	Cylindrospermopsin		41–53 ng/L	130–170 ng/L
20.	MC-LR and 1 congener	HPLC-DAD	0.2–0.3 μg/L	–	[186]
21.	Anatoxin-a	UHPLC-MS/MS	1.1 ng/L	2.5 ng/L	[187]
	Cylindrospermopsin		10.9 ng/L	21.7 ng/L
	Saxitoxins (4 congeners)		3.5–9 ng/L	7.1–26.9 ng/L
	GTX (7 congeners)		18.5–54.5 ng/L	42.2–227.6 ng/L
22.	MC-LR and 7 congeners	UHPLC-MS/MS	0.1 µg/L	0.5 µg/L	[164]
	Nodularin		0.1 µg/L	0.5 µg/L
23.	MC-LR and 2 congeners	UHPLC-MS/MS	0.1 µg/L	24 µg/L	[188]
24.	Cylindrospermopsin	UHPLC-MS/MS	0.029 μg/L	0.091 μg/L	[189]
25.	Saxitoxins (4 congeners)	on-line SPE–HILIC-HRMS	0.72 –3.9 ng/L	2.4–13 ng/L	[190]
26.	MC-LR and 1 congener	tandem-SPE-HILIC-MS/MS	0.0012–0.0021 μg/L	0.004–0.007 μg/L	[191]
	Nodularin		0.0021 μg/L	0.007 μg/L
	Anatoxin-a		0.03 μg/L	0.1 μg/L
	Cylindrospermopsin		0.0012 μg/L	0.004 μg/L
	BMAA		0.015 μg/L	0.05 μg/L
	DAB		0.009 μg/L	0.03 μg/L
	AEG		0.006 μg/L	0.02 μg/L
27.	BMAA	LC-MS/MS	2.8 ng/mL	–	[192]
	DAB		1.7 ng/mL	–
28.	BMAA	on-line SPE-UHPLC-HRMS	10 ng/L	–	[153]
	BAMA		10 ng/L	–
	DAB		10 ng/L	–
	AEG		5 ng/L	–
29.	BMA	UHPLC-MS/MS	–	2.5 µg/L	[193]
	AEG		–	2.5 µg/L
	DABA		–	5 µg/L
30.	MC-LR and 7 congeners	UHPLC-MS/MS (ESI)	0.02–0.2 µg/L	0.05–0.5 µg/L	[194]

LOD—limit of detection; LOQ—limit of quantification; UHPLC—ultra high-performance liquid chromatography; HILIC-MS/MS—hydrophilic interaction liquid chromatography-tandem mass spectrometry; RPLC-MS/MS—reverse phase chromatography tandem mass spectrometry; UV/PDA—ultraviolet/photodiode array detection; DAD—diode array detector; ESI—electrospray ionization; SPE—solid phase extraction; BAMA—β-amino-N-methylalanine; GTX—gonyautoxins; dcSTX—decarbamoylsaxitoxin.

Table 2. Chemical detection methods for cyanotoxins in water samples.

№	Cyanotoxins	Detection Techniques	Sensivity		Reference
			LOD	LOQ
1.	MC-LR and 2 congeners	UHPLC-MS/MS	0.02–0.04 ng/mL	–	[167]
2.	MC-LR and 11 congeners	UHPLC-MS/MS	–	0.2 µg/L	[168]
3.	MC-LR and 4 congeners	LC-MS/MS	0.005–0.0817 µg/L	0.005– 0.0817µg/L	[169]
	Nodularin		0.0048 µg/L	0.0048 µg/L
	Anatoxin-a		0.0001 µg/L	0.0004 µg/L
	Cylindrospermopsin		0.0001 µg/L	0.0004 µg/L
4.	BMAA	UHPLC-MS/MS	0.02 pg/µL	0.05 pg/µL	[170]
	2,4-DAB		0.04 pg/µL	0.13 pg/µL
5.	MC-LR and 7 congeners	LC-MS/MS	–	0.04–0.5 µg/L	[171]
	Anatoxin-a		–	0.02 µg/L
	Cylindrospermopsin (and deoxyCYN)		–	0.01 –0.02 µg/L
	Saxitoxins (4 congeners), GTX (5 congeners), decarbamoylgonyautoxin, N-sulfogonyautoxins-1 and -2		–	0.1–2 µg/L
6.	MC-LR and 11 congeners	HPLC-MS/MS	0.01±0.01–0.19±0.2 μg/L	0.04±0.04–0.64±0.65 μg/L	[172]
	Nodularin		0.04 ± 0.02 μg/L	0.13 ± 0.06 μg/L
7.	MC-LR and 2 congeners	HPLC-UV/PDA	3–4 μg/L	9–13 μg/L	[173]
8.	MC-LR and 2 congeners	HPLC-HRMS	0.002 μg/L	–	[174]
9.	Anatoxin-a,	HILIC-MS/MS	0.004 ng/mL	0.01 ng/mL	[175]
	Cylindrospermopsin		0.07 ng/mL	0.23 ng/mL
	Saxitoxin		0.01 ng/mL	0.04 ng/mL
	MC-LR and 4 congeners	RPLC- MS/MS	0.02–0.08 ng/mL	0.07–0.28 ng/mL
	Nodularin		0.05 ng/mL	0.18 ng/mL
10.	MC-LR and 5 congeners	UHPLC-MS/MS	–	0.046–0.052 µg/L	[176]
	Nodularin		–	0.049 µg/L
	Cylindrospermopsin		–	0.052 µg/L
11.	BMAA	UHPLC-HRMS	5 ng/L	10 ng/L	[177]
	DAB		3 ng/L	5 ng/L
	AEG		2 ng/L	5 ng/L
	BAMA		5 ng/L	10 ng/L
12.	MC-LR and 5 congeners	HPLC-MS/MS	0.0003–0.0009 µg/L	–	[178]
	Cylindrospermopsin		0.0005 µg/L	–
	Saxitoxin, dcSTX		0.0009–0.0013 µg/L	–
13.	BMAA	LC-MS/MS	10 ng/L	–	[179]
14.	MC-LR and 5 congeners	LC-MS/MS	0.04–0.8 μg/L	0.1–2.3 μg/L	[180]
	Nodularin		0.3 μg/L	0.9 μg/L
	Anatoxin-a		0.27 μg/L	0.81 μg/L
	Cylindrospermopsin		0.14 μg/L	0.4 μg/L
15.	MC-LR and 2 congeners	HPLC-DAD	0.08–0.15 µg/l	–	[181]
16.	MC-LR and 11 congeners	LC-MS/MS	0.001–0.007 μg/L	0.003–0.020 μg/L	[182]
	Nodularin		0.002 μg/L	0.006 μg/L
	Anatoxin-a		0.001 μg/L	0.003 μg/L
	Cylindrospermopsin		0.001 μg/L	0.003 μg/L
17.	MCs	UPLC-MS/MS	0.005 µg/L	–	[183]
	Anatoxin-a		0.02 µg/L	–
	Cylindrospermopsin		0.02 µg/L	–
	Saxitoxin		0.8 µg/L	–
	BMAA		0.03 µg/L	–
18.	BMAA	LC-MS/MS	0.030 μg/L	0.096 μg/L	[184]
19.	MC-LR and 11 congeners	on-line SPE – UHPLC-HRMS	5–37 ng/L	15–130 ng/L	[185]
	Anatoxin-a		15–18 ng/L	50–60 ng/L
	Homoanatoxin-a		11–12 ng/L	30–40 ng/L
	Cylindrospermopsin		41–53 ng/L	130–170 ng/L
20.	MC-LR and 1 congener	HPLC-DAD	0.2–0.3 μg/L	–	[186]
21.	Anatoxin-a	UHPLC-MS/MS	1.1 ng/L	2.5 ng/L	[187]
	Cylindrospermopsin		10.9 ng/L	21.7 ng/L
	Saxitoxins (4 congeners)		3.5–9 ng/L	7.1–26.9 ng/L
	GTX (7 congeners)		18.5–54.5 ng/L	42.2–227.6 ng/L
22.	MC-LR and 7 congeners	UHPLC-MS/MS	0.1 µg/L	0.5 µg/L	[164]
	Nodularin		0.1 µg/L	0.5 µg/L
23.	MC-LR and 2 congeners	UHPLC-MS/MS	0.1 µg/L	24 µg/L	[188]
24.	Cylindrospermopsin	UHPLC-MS/MS	0.029 μg/L	0.091 μg/L	[189]
25.	Saxitoxins (4 congeners)	on-line SPE–HILIC-HRMS	0.72 –3.9 ng/L	2.4–13 ng/L	[190]
26.	MC-LR and 1 congener	tandem-SPE-HILIC-MS/MS	0.0012–0.0021 μg/L	0.004–0.007 μg/L	[191]
	Nodularin		0.0021 μg/L	0.007 μg/L
	Anatoxin-a		0.03 μg/L	0.1 μg/L
	Cylindrospermopsin		0.0012 μg/L	0.004 μg/L
	BMAA		0.015 μg/L	0.05 μg/L
	DAB		0.009 μg/L	0.03 μg/L
	AEG		0.006 μg/L	0.02 μg/L
27.	BMAA	LC-MS/MS	2.8 ng/mL	–	[192]
	DAB		1.7 ng/mL	–
28.	BMAA	on-line SPE-UHPLC-HRMS	10 ng/L	–	[153]
	BAMA		10 ng/L	–
	DAB		10 ng/L	–
	AEG		5 ng/L	–
29.	BMA	UHPLC-MS/MS	–	2.5 µg/L	[193]
	AEG		–	2.5 µg/L
	DABA		–	5 µg/L
30.	MC-LR and 7 congeners	UHPLC-MS/MS (ESI)	0.02–0.2 µg/L	0.05–0.5 µg/L	[194]

LOD—limit of detection; LOQ—limit of quantification; UHPLC—ultra high-performance liquid chromatography; HILIC-MS/MS—hydrophilic interaction liquid chromatography-tandem mass spectrometry; RPLC-MS/MS—reverse phase chromatography tandem mass spectrometry; UV/PDA—ultraviolet/photodiode array detection; DAD—diode array detector; ESI—electrospray ionization; SPE—solid phase extraction; BAMA—β-amino-N-methylalanine; GTX—gonyautoxins; dcSTX—decarbamoylsaxitoxin.

As separation instruments, HPLC and UHPLC are usually used. UHPLC, is faster due to the higher pressure applied, and the online SPE-procedure provides reduced sample time processing [2,181]. HPLC-UV/PDA [173], HPLC-DAD [186] have less LOD values (3–4 μg/L, 0.2–0.3 μg/L, respectively) for MCs than LC-MS/MS where minimum LOD values vary within 0.0003–0.1 µg/L [164,178,188]. Concerning other cyanotoxins, MS detection techniques also provide relatively low values of LOD and LOQ. The LC-MS method requires expensive instruments and thorough sample preparation, which makes it a time-consuming procedure. That limits LC-MS techniques' application as ubiquitous [195]. Nevertheless, this method remains preferable for precise quantitative analysis of cyanotoxins in water samples.

3. Toxin Exposure Pathways

Major cyanotoxins exposure routes include ingestion through drinking water or dermal contact with recreational waters [196], also through food, and inhalation since cyanotoxins were identified in aerosols generated by HABs [197]. Historically, cyanoHABs were considered a public health threat to freshwater lakes, rivers, and reservoirs. However, freshwater-sourced MCs can accumulate in marine mollusks in concentrations 100-fold greater than in surrounding water [10,123].

3.1. Transport of Cyanotoxins in Freshwater and Marine Systems

Recent studies demonstrated that cyanotoxins could persist during transport into estuarine and marine waters and can directly affect marine ecosystems [10,123,198,199,200]. MCs and other toxins produced by freshwater cyanobacteria can enter the marine ecosystem via freshwater channels and outflows [12]. This changes HABs management approach, requiring monitoring of multiple toxins across the freshwater-to-marine continuum and including cyanotoxins in marine and estuarine monitoring [201].

3.2. Toxin exposure pathways: Oral (Drinking water)

When drinking water is impacted by cyanobacterial toxins resulting from HABs and not treated adequately to reduce the cyanotoxin levels, it can cause serious effects on the entire region [197].

The causes of cyanobacteria proliferation in urban environments are mainly the disposal of untreated domestic sewage in water reservoirs and surface runoff water from soils. In analyzing sewage disposal systems in the main cities of Kazakhstan—Almaty and Astana, the efficiency of biogenic compounds removal remains unsatisfactory, reaching only 30–40%. This eutrophication is due to the increase of nutrients, such as phosphorus and nitrogen, arising from human action, representing a serious risk to the health of living beings and drastically reducing water quality. To cope with this problem, the possibility of intensifying nitrogen and phosphorus removal using zeolite as a biofilm carrier in an activated sludge tank is examined [202].

3.3. Toxin Exposure Pathways: Oral (Food)

Food is an important source of cyanotoxin exposure [203]. Worryingly, exposure of crop plants to cyanotoxins through irrigation was already demonstrated [204,205,206]. For centuries some species of Nostoc—the symbiotic colonial cyanobacteria N. flagelliforme, N. commune, and N. sphaeroides – have been wild-harvested and consumed as a part of the traditional diet by indigenous people in different countries, including Peru, China, Ecquador, Mexico, Fiji, Filippines, Mongolia [209,210].

Chronic dietary exposure to BMAA present in the traditional Chamorro diet was associated with the formation of both β-amyloid deposits and neurofibrils tangles (NFT) found in brain tissues of Chamorros who died with ALS/ Parkinson's dementia complex (ALS/PD). BMAA occurs not only as a free amino acid at different levels of the trophic chain (cyanobacteria Nostoc sp., root symbioses, cycad seeds, flying foxes, and brain tissues of Chamorro people who passed away from ALS/PD) but can also be released by acid hydrolysis increasing in concentrations 10- to 240-fold [207]. Vervet monkeys fed for only six months with BMAA-dosed fruit developed β-amyloid deposits and NFT in the brain. Increasing the amount of L-serine in the vervets diet reduced the density of NFT and the risk of neurodegenerative pathological brain findings [56]. Recently, Downing and co-authors [208] revealed that human liver hepatocyte and intestinal epithelial cultures could not metabolize BMAA, demonstrating that BMAA detoxication is impossible and BMAA will likely accumulate in these cells [53].

3.4. Toxin Exposure Pathways: Air (Aerosolization)

The algae can be dispersed by air [211], and aerosol can be created from algae during HABs [212]. The increase in the salinity of freshwater streams is likely to influence the abundance and diversity of aerosolized bacteria [213]. The cyanotoxins may be transported in aerosols from lakes with high concentrations of toxigenic cyanobacteria [214,215,216,217]. Recent findings with rat models confirmed that BMAA exposure was insufficient in producing gross toxic effects; however, it still leaves the possibility of lifelong exposure via inhalation [218].

MC-LR exposure in the existing rodent models increases lung infiltration with granulocytes [219,220] and increases proinflammatory cytokine expression [221,222]. Recently, Breidenbach and co-authors [223] reported that human airway epithelium response to MC-LR is represented by proinflammatory phenotype, including chemokines.

The aerosolization of cyanobacteria was proposed as a risk factor for ALS [100]. Aerial link of exposure was investigated with ALS/PD. BMAA and its isomers (DAB and AEG) were measured in air filters around lake Mascoma [83]. Moreover, Facciponte and co-authors [224] found that humans routinely inhale aerosolized cyanobacteria. Using PCR, authors identified cyanobacteria at high frequencies in the upper respiratory tract (93.20%) and central airway (79.31%). They concluded that cyanobacteria exposure might be a prevalent and chronic phenomenon and not necessarily restricted to water bodies.

Autoradiographic imaging in mice showed a distinct localization of radioactivity in olfactory mucosa and bulb following intranasal instillation of radiolabelled BMAA, confirming a direct transfer of BMAA via olfactory pathways to mice brain circumventing the blood-brain barrier [225].

3.6. Natural Model of Toxin Exposure

The complexity of neurodegenerative diseases requires a deep understanding of the disease biology and makes it challenging to develop a model of cyanotoxin exposure close to neurodegenerative findings in humans due to the species-specific variations in the phosphorylation and cleavage of the tau protein [226]. Natural animal models should recapitulate two major features of human neurodegenerative diseases: Aβ deposition and NFT formation. Chronic low BMAA concentrations induce neurodegenerative changes in non-human primates [56,227]. BMAA can bioaccumulate in marine apex predators such as dolphins and sharks, and in humans [207,228,229]. It was detected in the brains of stranded dolphins with pathological hallmarks of AD at concentrations higher than those found post-mortem in individuals with ALS and AD [230]. Chronic low BMAA concentrations induce neurodegenerative changes in non-human primates [56,227]. There are increased numbers of β-amyloid+ and dystrophic neurites in the auditory cortex compared to the visual cortex and brainstem [230].

3.7. Cyanotoxins and Infections

BMAA can facilitate most of the mechanisms related to neurodegeneration [231]. Thus, Lobner and co-authors [232] demonstrated that BMAA at the concentration range of 10–100 µmol potentiates neurotoxicity induced by amyloid-β and NMDA.

STX doubled the quantity of ZIKV-induced neural cell death in progenitor areas of human brain organoids, while the chronic ingestion of water contaminated with STX before and during gestation caused brain abnormalities in offspring of ZIKV-infected immunocompetent C57BL/6J mice. These results raise a public health concern regarding the consequences of arbovirus outbreaks in areas with droughts and/or frequent freshwater cyanobacterial blooms [233].

The outbreak of Zika syndrome coincided with a major drought in the region between 2012 and 2016. Characteristic of dry seasons, the concentration of nutrients from untreated effluents and lower volume of water, and an increase in atmospheric temperature allow greater blooming of cyanobacteria. Consequently, the concentration of cyanotoxins, such as saxitoxins, increases. This led authors to formulate the hypothesis that cyanobacteria in the water supply would be a causal cofactor of zika-associated microcephaly.

4. Mechanisms of Brain Toxicity

Well-studied neurotoxins of algal origin are alkaloids saxitoxins (STXs) that have been identified in dinoflagellates and several cyanobacterial genera, including Anabaena, Aphanizomaenon, Planktothrix, Cylindrospermopsis, and Scytonema [234,235]. STXs are represented by more than 50 structural analogs commonly known as paralytic shellfish toxins (PSTs). They block the passage of sodium across a biological membrane and interfere with potassium and calcium-mediated ion channels [72].

While the pathophysiology of some toxins (STXs, anatoxins, etc.) are relatively well studied, others, such as ciguatera, are not clear. Recently, the neurotoxic effects of cyanopeptides attracted more attention [77,236,237,238]. Some cyanopeptides exhibit anti-proliferative effects on tubulin and microtubules essential for neurons. Thus, the anti-proliferative toxic cyclodepsipeptides cryptophycins are 100–1000 fold compared with paclitaxel and vinblastin [236]. Further research needs to improve analytical methods and assess potentially toxic cyanopeptides.

4.1. Neurodevelopmental Effects

The link between neurodegeneration and neonatal BMAA exposure, dose-dependent neuronal loss, beta-amyloid deposition, and behavioral deficits was recently demonstrated in a rat model [239]. Autoradiographic imaging confirmed transplacental research of radiolabelled BMAA and specific uptake in mouse fetal [240]. Furthermore, in neonatal rats, the free BMAA concentration was higher in the neonatal brain than in peripheral tissues such as the thymus, pancreas, and spleen, except for the liver. The level of protein-associated BMAA was significantly higher in the hippocampus than in other brain regions [241]. The BMAA exposure to neural stem cells decreased neurite outgrowth, and a number of neurites in neural stem cells (NSC) [242]. The authors conclude that BMAA acts as a developmental toxin. BMAA can negatively impact NSC homeostasis, increasing susceptibility to neurodegenerative disease later in life [242]. Perinatal exposure in mice, even with low doses of BMAA, leads to neurobehavioral disturbances during the postnatal period and adulthood [243].

4.2. Blood-Brain Barrier (BBB)

The blood-brain barrier and the blood-CSF barrier separate CNS from blood and include the endothelial lining of the brain capillaries associated with astrocytes, pericytes, and neurons. The pericytes and astrocytes are closely associated with the endothelial cells and are required for capillary maturation (pericytes) and the maintenance of the permeability-barrier functions (astrocytes). The basement membrane (contains laminin, proteoglycans, fibronectin, collagen IV, nidogen, and entactin) and is essential for blood-barrier differentiation. BBB separates neurons from the circulating blood and maintains the internal chemical composition of the brain "milieu" responsible for the proper functioning of neuronal circuits, neurogenesis, angiogenesis, synaptic transmission, etc. BBB breakdown due to disruption of the tight junctions may result in synaptic and neuronal dysfunction and contribute to neurodegenerative disorders such as ALS, Alzheimer's disease, Parkinson's disease, and multiple sclerosis [244].

Berntzon and co-authors found BMAA in the CSF of a patient with ALS and some controls, though they did not confirm a prevalence of BMAA findings in ALS patients [109]. Significant amounts of BMAA were found in brain tissues of american ALS and Alzheimer’s disease patients, confirming the ability of BMAA to cross the BBB [110]. Alternative findings regarding AD patients were reported by Meneely and co-authors [112]. The other possible route of entry to the CNS is through the olfactory epithelium and the nasal passage or via the blood. The cyanobacterial neurotoxin BMAA can be directly transferred through olfactory pathways circumventing the BBB in mice and directly affecting olfactory neurons [225].

Microcystin-LR (MC-LR) has been confirmed to cause blood-brain barrier disruption and enter the brain tissue, resulting in non-negligible toxic effects. However, the neurotoxicity of MC-LR is mainly unknown. This study revealed that MC-LR disrupted the function of the ubiquitin-proteasome system in neurons, which inhibited the degradation of α-synuclein (α-syn), leading to its release from neurons for transport into microglia. α-Syn is the main component of Lewy bodies, which has been identified as one of the main pathological features of Parkinson’s disease (PD). In vitro, we observed that α-syn mediated by MC-LR activated HMC3 cells and polarized them towards M1 type. In addition, we confirmed that α-syn was transported into HMC3 cells through TLR4 receptors and activated the NLRP3 inflammasome, which in turn enhanced the maturation and release of IL-18 and IL-1β [245].

4.3. Glia

Microglial activation and neuroinflammation are common to many neurodegenerative diseases. Glial cells, including microglia, have long been suspected of playing a role in Alzheimer’s disease but only because of their ability to react to neuronal dysfunctions (e.g., amyloid and Tau aggregates). This neurocentric view, which considered glial cells as secondary, has been challenged recently by the results of genetic association studies identifying genetic loci associated with the risk of Alzheimer’s that are associated with genes preferentially or exclusively expressed in glial cells [246].

The research on cyanopeptides effects on glia is limited. Chiu and co-authors [247,248] demonstrated a gliotoxicity of BMAA using the olfactory ensheathing cell as in vitro model. A study conducted by Bubic and co-workers [249] showed that depsipeptide planktopeptin and anabaenopeptins impair the metabolic activities of normal human astrocytes via membrane perforation, oxidative stress, and changes in mitochondrial metabolism. Later, Mello and co-authors showed cytotoxic effects of BMAA and MC-LR on primary astrocytes isolated from mixed adult brain cell cultures [250], and Soto, with co-workers, demonstrated damaging BMAA effects on Muller’s glial cells [251]. Both glial cells and neurons can to uptake and accumulate BMAA, as demonstrated using a specific, polyclonal antibody against BMAA [252].

The role of dysfunctional astrocytes in the pathogenesis of ALS and other neurodegenerative diseases indicates that astrocytes may be targeted with strategies for their revival. These strategies may include direct intervention on astrocytes with modulatory medicines, exosomes and miRNA-based therapies, or their replacement.

5. Cyanotoxins, Cyanopeptides and Neurodegenerative Diseases

A central dogma of age-related neurodegenerative diseases claims that the accumulation and propagation of aggregated proteins cause neurodegeneration [253]. Recently, a mechanism that does not involve a specific neuropathogenic protein but is mediated by error-prone translation leading to stochastic near-cognate missense substitutions was suggested. Drummond and Wilke proposed in 2008 that tolerance to translation errors of certain proteins provides a new mechanism to explain their propensity to misfold pathologically. Mistranslation destabilizes the proteome by leading to misfolding and accumulation in the cells of potentially toxic protein aggregates [58,254]. The finding that translational error increases with age in some biological models (Drosophila) [255] may suggest the possibility that the rate of translation leading to aging-related proteostasis failure may be a key event in early ND diseases [256]. The mistranslating cells exhibit severely inhibited protein synthesis and formation of protein aggregates in the cellular ND model [257]. Aminoacyl-tRNA synthetases (AARSs) catalyze covalent binding tRNA with their cognate amino acids and are 2–3 orders of magnitude more selective than other amino acid-utilizing [258].

Hundreds of non-proteinogenic amino acids produced by cyanobacteria include BMAA and can, in principle, enter human protein synthesis through foods and drinking water. Earlier studies support the ability of BMAA to be incorporated into the proteins [54,259]. It has been suggested that BMAA is misincorporated at serine codons during protein synthesis [54]. However, recently Han and co-authors [63] demonstrated that BMAA is not a substrate for human seryl-tRNA synthetase (SerRS) but a substrate for human alanyl-tRNA synthetase (AlaRS) and can form BMAA-tRNAAla by escaping from the intrinsic AlaRS proofreading activity. Furthermore, BMAA inhibits the cognate amino acid activation, the editing functions of AlaRS, and the deacylation activity of HsAlaRS on Ser-tRNAAla [63]. The AlaRS possess canonical and non-canonical cellular functions and are predominantly linked to neurodegenerative disorders in human and mouse models [260]. Furthermore, using transcriptomic analysis, Wang and co-authors [261] confirmed that BMAA could alter the expression of major genes encoding components related to translation in prokaryotes (diazotrophic algae Anabaena). The ability of BMAA to affect protein hemostasis can be evolutionarily ancient and initially directed to inhibit the growth of neighboring microalgae. The inhibition of cell growth and progression in the cell cycle of eukaryotic cells was demonstrated in in vitro experiments [262]. The production of cyclic peptides, including non-proteinogenic amino acids, leads to the lysis of cyanobacteria and may be an effective control mechanism of cyanobacterial density during algal blooms [126].

6. Conclusions

The structural variety of cyanotoxins and cyanopeptides is produced during cyanobacterial blooms. Many structural aspects of key metabolites involved in the cyanotoxins pathways have yet to be elucidated. However, it is becoming clear that non-proteinogenic amino acids, free-existing or initially a part of cyanopeptides, may affect protein hemostasis and lead to mistranslation and misfolding of proteins in eukaryotic cells, building a link to neurodegenerative diseases development. There are many aspects pertaining to the regulation, role, and function of these compounds that also require the development of novel detection approaches. This knowledge may be harnessed to identify novel biomarkers for neurodegenerative diseases and new targets for interventions.