Cyanogenic glycosides (cyanoglycosides, CGs) are secondary metabolites of predominantly plant origin and account for nearly 90% of the broader group of plant toxins known as cyanogens (Nampoothiri, 2017). Chemically, CGs are α-hydroxynitrile glucoside consisting of two main components: a sugar moiety - most commonly glucose - and an aglycone, the non-sugar part of the molecule that contains the cyanogenic group (CN). These components are linked through a glycosidic bond. Glycosylation plays a crucial role in determining the stability, solubility, and biological activity of CGs, including their potential antitumor properties (Mosayyebi et al., 2020). It also influences the interaction between the aglycone and cellular structures, such as receptors and proteins, thereby affecting a compound’s biological function (Pelley, 2012). The aglycone can vary in its chemical structure, most commonly appearing as aliphatic, cyclic, aromatic, or heterocyclic compounds. This part of the molecule largely determines the toxicity of CGs. Natural cyanogenic glycosides display considerable structural diversity in both their sugar and aglycone components (Vetter, 2017). Some naturally occurring CGs exist as stereoisomers, for example: (R)-lotaustralin/(S)-epilotaustralin, (R)- prunasin/(S)- sambunigrin, and (2R)-taxyphyllin/(2S)-dhurrin (Yulvianti and Zidorn, 2021). The general structure of CGs is illustrated in Figure 1 , with the structures of the most significant compounds shown in Figure 2 . The chemical diversity of plant CGs are described in more detail in article Yulvianti and Zidorn (2021).

Trivial names of CGs are usually derived from the Latin names of the plants from which they were first isolated (e.g. almond amygdalin - Prunus amygdalus). However, several isolated CGs do not have trivial names.

Currently, 112 distinct CGs are known from the plant kingdom (Yulvianti and Zidorn, 2021). For plants they are important as protection against being consumed by animals and also as protection against various microorganisms (Zagrobelny et al., 2018). But actually, this protection is not provided by CG itself, but rather by the toxic hydrogen cyanide (HCN) released from stored CGs, cyanolipids, or cyanohydrins (Lechtenberg, 2011). This process occurs in an acidic environment (at low pH) or under the influence of hydrolytic enzymes with the formation of free HCN after the mechanical disruption of tissues. Cyanogenesis occurs in two phases: Phase 1 - cleavage of the carbohydrate component, Phase 2 - cleavage of the aglycone to aldehyde or ketone and HCN ( Figure 3 ).

While CGs are stored in vacuoles, β-glucosidases are localized in the apoplastic space, bound to cell walls in dicotyledonous plants, and in the cytoplasm and chloroplasts in monocotyledonous plants. Hydroxynitrile enzymes accumulate mainly in the cytoplasm and plasma membranes. When plant tissue is disrupted, CGs and enzymes come into contact, and the CGs degrade into cyanohydrins, HCN, and ketones. The different compartmentalization of CGs and enzymes helps prevent excessive HCN production and its toxicity in plants (Vetter, 2017). Yet the cause of the typical bitter odor in the mechanical disruption of seeds containing CGs is not HCN, but the released benzaldehyde (Griffin, 1974; Moertel et al., 1982). CGs are also a re-mobilizable reservoir of reduced nitrogen, and increase plant tolerance by reducing oxidative stress and may support seedling development (Sanchez-Perez et al., 2009; Pičmanová et al., 2015). Moreover, free cyanide, including that released from the CGs, may act as a signaling molecule (Siegień and Bogatek, 2006).

Cyanogenic glycosides (CGs) are primarily derived from aliphatic amino acids (L-valine, L-isoleucine, L-leucine) and aromatic amino acids (L-phenylalanine, L-tyrosine). However, certain CGs - such as deidaclin, gynocardin, acalyphin, cycasin, and ranunculin - are synthesized from non-proteinogenic precursors (Nyirenda, 2020). While cyanogenic ferns and gymnosperm species predominantly produce aromatic CGs, angiosperms are known to synthesize both aliphatic and aromatic forms (Bak et al., 2006). To date, amino acid-derived cyanogenic glucoside pathways have been elucidated in various plant species. Despite species-specific variations, three conserved enzymatic steps have been identified across all CG biosynthetic pathways ( Figure 4 ): 1. Amino acid hydroxylation – the conversion of α-amino acids to aldoximes via N-hydroxylated derivatives, mediated by membrane-bound enzymes from the cytochrome P450 (CYP) family. In gymnosperms and angiosperms, this is functionally conserved as the enzyme CYP79. 2. Cyanohydrin formation – the transformation of aldoximes into unstable cyanohydrins via further P450 cytochrome enzymes. In angiosperms, several more or less specific CYPs involved in this pathway have been characterized (CYP71, CYP706, CYP736). 3. Glycosylation - the attachment of a glucose unit, which stabilizes the cyanohydrins into cyanogenic glucosides. This step is catalyzed by the enzyme UDP-glucosyltransferase (in angiosperms, UGT85 and UGT94 have been characterized).

Transcription factors of the basic helix-loop-helix (bHLH) type play a key role in the regulation of CGs biosynthesis (Harun and Mohamed-Hussein, 2024). The plasticity of CYP gene expression, combined with their catalytic versatility, has made them key drivers of evolutionary innovation in plant secondary metabolism, allowing plants to colonize new environments and co-evolve with herbivores and pathogens (Bak et al., 2006; Xu et al., 2015).

In recent decades, significant progress has been made in the study of CG biosynthetic pathways and their regulation, which has enabled a deeper understanding of plant adaptation mechanisms and their evolutionary processes. This topic has been explored in more detail in studies by Forslund et al. (2004); Bak et al. (2006); Morant et al. (2007); Sun et al. (2018); Thodberg et al. (2020); Yulvianti and Zidorn (2021); Boter and Diaz (2023); Harun and Mohamed-Hussein (2024).

Cyanogenesis was first described in white clover (Trifolium repens) (Mirande, 1912), and it soon became evident that this species is polymorphic in terms of cyanogenesis – that is, both cyanogenic and acyanogenic plants occur within the same population (Armstrong et al., 1913). It was shown that this form of ecological adaptation results from polymorphism (the presence or absence) of genes responsible for both the synthesis of CGs (Ac) and the synthesis of β-glucosidases, enzymes that break down CGs (Li) (Hughes, 1991). Plants that carry at least one dominant (functional) allele at both genes (Ac and Li) are cyanogenic, while the occurrence of two nonfunctional alleles (ac and li) at either gene confers the acyanogenic phenotype. The Ac gene corresponds to the gene encoding cytochrome P450 from the CYP79D protein subgroup (specifically CYP79D15). CYP79D orthologs catalyze the first step in the biosynthesis of cyanogenic glycosides ( Figure 4 ) (Olsen et al., 2008, 2013).

This chemical defense polymorphism is among the most long-studied and best-documented examples of adaptive polymorphism in plants. More cyanogenic plants are found in warmer and more humid regions with higher herbivore activity. However, since cyanogenesis is quite energetically costly, cyanogenic plants exhibit slower growth and reproduction in these areas. This represents a classic example of an evolutionary trade-off between defense and growth. It should, however, be noted that not all cyanogenic plants exhibit adaptive polymorphism. In many species, cyanogenesis is genetically fixed - either all individuals are cyanogenic, or none are. Adaptive polymorphism, as thoroughly documented in Trifolium repens, represents a specific evolutionary phenomenon that occurs in only certain species where selective pressures maintain both the presence and absence of cyanogenic expression within the same population (Olsen et al., 2008).

CG synthesis is relatively widespread in the plant kingdom. More than 3000 plant species belonging to 130 families are cyanogenic (Yadav et al., 2023), including ferns, gymnosperms and angiosperms. In the agricultural context, the main sources of CGs are seeds and by-products of crops such as flax (Linus usitatissimum), apricot (Prunus armeniaca), bitter almond (Prunus dulcis), sorghum (Sorghum vulgare), wheat (Triticum aestivum), barley (Hordeum vulgare), oat (Avena sativa), cassava (Manihot esculenta) and apple (Malus pumila) (Hegnauer, 1986; Jones, 1998). In general, CGs exhibit a highly specialized distribution, with a given type of CG typically occurring in only one or two plant families. Furthermore, individual plant species generally produce only one or two types of CGs, reflecting their metabolic specialization and ecological adaptations (Bolarinwa et al., 2014a; Süli et al., 2017). Amygdalin and prunasin, for example, are predominantly found in plants of the Rosaceae family (e.g. Prunus spp., Malus spp.), where it functions as a chemical defense against herbivores. Linamarin and lotaustralin are characteristic of tropical and subtropical plants from the Fabaceae and Euphorbiaceae families (e.g., Phaseolus lunatus, Manihot esculenta), primarily serving to protect these plants from insect herbivores and microbial pathogens. A representative cyanogenic glycoside of the Poaceae family is dhurrin, which is especially abundant in young leaves of Sorghum bicolor, where it enhances the plant’s resistance to herbivores during early developmental stages.

However, the defensive potential of CGs is also manifested in the process of plant adaptation to various abiotic stressors, such as drought, excessive moisture, mineral imbalance, frost, trampling, and herbicide exposure (Bolarinwa et al., 2014a). Moreover, the degree of HCN induction appears to differ depending on whether the stress is chronic or acute (Wheeler et al., 1990; Woodrow et al., 2002). In stressed plants, where photosynthetic rate is reduced, CGs may also provide a ready source of nitrogen, remobilized when the stress is alleviated (O’Donnell et al., 2013; Schmidt et al., 2018). Under stress conditions, they also reduce oxidative stress and regulate the transport of carbon and nitrogen in plants (Conn, 1980; Rosati et al., 2019). Younger plants contain CGs much more than older ones (Dreyer et al., 1981). Some plants are not completely cyanogenic, others are not cyanogenic throughout the growing season. Cereal leaves are cyanogenic for example, but the grains are not. Papaya and mango leaves are also cyanogenic, but the fruits are not. Drought, frost, and the use of nitrates and herbicides can increase their amount and thus their toxicity to animals (Busk and Møller, 2002). Seasonal changes in the cyanide content of some species have also been reported (Robakowski et al., 2016; Bartnik and Facey, 2017). The amount of the most significant CGs in plants (expressed as the equivalent amount of HCN) is given in Table 1 . The absolute amounts of individual CGs are listed in Tables 2 – 9 . However, the reported CG levels in plant tissues also depend on the method of extraction and determination, as well as on the genotype, plant age, soil condition, fertilizer application, climatic conditions, and other factors (Bolarinwa et al., 2014a; Tahir et al., 2024).

Some specialized herbivores (mainly insects) preferentially feed on cyanogenic plants and use them as protection against predators. Several arthropod species (e.g., Diplopoda, Chilopoda, Insecta) can even synthesize CGs de novo. The unique plant-insect interaction based on CG is extensively discussed in the study by Zagrobelny et al. (2018).

CGs are considered antinutrients that reduce the quality of feed and food, causing various health issues in animals, including humans. It is recommended that such plants be treated prior to consumption to minimize HCN content. Different types of processing methods are used to reduce CG content in plants. The most important processing methods include drying, grinding, dipping, peeling, ultrasound-assisted detoxification, autoclaving, soaking, boiling and fermentation (Bolarinwa et al., 2014a). The latter has proven to be highly effective, for example, in reducing CG content in bamboo shoots (Chongtham et al., 2022). In the process of acid fermentation of certain CGs, the bacteria Lactobacillus plantarum, Bacillus subtilis, Bacillus licheniformis, and Bacillus sonorensis proved to be effective (Abban et al., 2013; Menon et al., 2015). Sun drying after retting reduces cyanide content by 98.6%. Boiling/cooking can reduce free cyanide content by 96% within 15 minutes. After heating for 25 minutes, bound cyanide is reduced by 55% (Nampoothiri, 2017; Chongtham et al., 2022). A reduction of cyanides by 93% was also achieved by applying sodium bicarbonate (5 mL of a 0.4% NaHCO₃ solution) to 1 g of cassava leaves (Latif et al., 2019). Conserved stone fruit must be peeled because cyanides also occur in the resulting infusion up to 33 mg·kg^-1 HCN. However, the processing methods applied are not always sufficiently effective, and a certain amount of CG remains in plant products, thus posing potential health risks. Tables 2 - 9 also show varying amounts of CG in differently processed products. The issue of reducing cyanide content in plants and processed products is further explored by Rawat et al. (2015); Bolarinwa et al. (2016); Tahir et al. (2024) and others. Studies have also been developed to estimate the risks associated with the daily intake of CGs in food (Schrenk et al., 2019; Park et al., 2024).

When assessing the harmful effects of substances involved in cyanogenesis, the focus is mostly on the effects of released HCN; other components (intact glycosides and their hydrolysis products) do not appear to be serious in terms of acute toxicity. HCN is extremely toxic to animals, including humans. The lethal HCN dosage in most animal species is in the range of 2 mg·kg^-1 to 2.5 mg·kg^-1, with the exception of pandas (Clarke et al., 1981; Panter, 2018). The acute oral lethal dose of HCN for humans is reported to be 0.5 – 3.5 mg·kg^-1 of body weight (Halstrom and Moiler, 1945). The permissible limit of cyanogen content in food is 500 mg·kg^-1 (Food and Agriculture Organization, 2005).

HCN toxicity in animals, including humans, is due to blocking the release of energy from ATP (adenosine triphosphate) by inhibiting cytochrome oxidase activity in the respiratory chain ( Figure 4 ). Hence the tissues and cells of the organisms are unable to utilize the oxygen that is transported by the blood, which can lead to internal suffocation (Gracia and Shepherd, 2004). The most important laboratory finding in cyanide poisoning is metabolic acidosis with dramatically increased lactate concentration (Baud et al., 2002) ( Figure 4 ). The effects of HCN on the ability of the thyroid gland to store and process iodine are also documented (Erdogan, 2003). Clinical signs of acute poisoning include rapid breathing, decreased blood pressure and rapid pulse, dizziness, convulsions, vomiting, and blue discoloration of the skin due to lack of oxygen. As cellular hypoxia worsens, consciousness progresses to coma. Symptoms appear within seconds to minutes.

Cyanide detoxification in plants and animals is a critical biochemical process that helps mitigate the toxic effects of cyanogenic compounds. Both plants and animals have evolved mechanisms to detoxify or tolerate cyanide to survive in environments where these compounds are prevalent. The primary mechanism for cyanide detoxification in most plants is the β-cyanoalanine pathway. In this process, cyanide reacts with the amino acid L-cysteine to form β-cyanoalanine, catalyzed by the enzyme β-cyanoalanine synthase (CAS). This reaction occurs mainly in the mitochondria and is the major route by which plants detoxify endogenous cyanide. β-cyanoalanine can be further converted into asparagine, aspartate, and ammonia by β-cyanoalanine hydratase or nitrilase, integrating the cyanide-derived nitrogen into the plant’s nitrogen metabolism (Velišek and Hajšlova, 2009). The β-cyanoalanine synthase pathway is described in more detail by Machingura et al. (2016).

The most significant detoxification system in animals is the rhodanese enzyme system, which converts cyanide into thiocyanate, which is much less toxic and can be safely excreted through the urine (Gracia and Shepherd, 2004) ( Figure 5 ). A further manner of detoxification is the binding of cyanide to hydroxocobalamin (vitamin B12), resulting in the formation of nontoxic cyanocobalamin.

The ability of an animal to tolerate certain doses of HCN also depends on the animal species, body weight, digestion rate, type of food, and the animal’s ability to detoxify the released HCN. The lethal dose for sheep is 2.4, cattle 2.0, mice 3.7, cats 2.0, 0, rats 0.5 – 10.0 and dogs 1.5 mg·kg^-1 body weight (Jones, 1998). Ruminants are more sensitive to HCN poisoning because the enzymes that facilitate the release of HCN are destroyed by gastric HCl in these animals. Of this group, goats appear to be the most susceptible to cyanide (Patel et al., 2014). The specifics of CG poisoning in ruminants are described in detail by Gensa (2019). In non-ruminants, CGs are partially cleaved, and HCN is released only by the action of the colonic microflora where the pH is more suitable for the action of glycosides. But the hydrolysis is not complete, some glycosides are absorbed in their original form. In ruminants, many bacteria found in the rumen can hydrolyze CGs, with the degree of effectiveness depending on glycoside type and feed ration. The composition of gut microbiota also plays a significant role in the tolerance of mammals to the content of secondary metabolites in their diet. The gut microbiome of the giant panda and red panda contains a higher proportion of Pseudomonas bacteria compared to other mammals. Their microbiome is thus enriched with genes that encode the enzymes involved in the potential degradation or detoxification of HCN (Zhu et al., 2018). This is likely an evolutionary adaptation, that is not unique in the context of the plant kingdom, and can also be observed in some animals or microorganisms (Panter, 2018). Lemurs and gorillas also possess the unique ability to utilize high cyanide content in their diet without any acute or chronic harmful effects (Ballhorn et al., 2016). The metabolism of CGs in animals is described in more detail in the article by Cressey and Reeve (2019). However, estimating the health risks related to consuming CG-containing plants is often not straightforward, as sometimes the entire plant, including seeds, is consumed, while in other cases only the fruit or other parts are eaten. However, chronic intoxication can occur if plants containing CGs are part of the daily diet and consumed in larger quantities (Hartanti and Cahiani, 2020). More detailed data on the daily intake of CGs in selected foods is documented by Park et al. (2024).

Chronic cyanide toxicity causes several diseases, especially in tropical areas where the main food is plant-based. Growth retardation, goiter, and cretinism are relatively common diseases in developing countries where people consume food with very low iodine content (<100 μg/day) and high cyanide content (Odo et al., 2014; Bolarinwa et al., 2016). High and sustained intake of cyanogens in sublethal concentrations of manioc (cassava) flour in combination with the low intake of sulfur amino acids also causes konzo - a disease of the upper motor neurons characterized by irreversible but non-progressive symmetric spastic para/tetraparesis which mostly affects children and women in the tropics (Howlett et al., 1990). Konzo disease is common during periods of drought, or when there is a shortage of alternative foods due to unfavorable social or environmental factors (Nzwalo and Cliff, 2011). Practically identical to konzo disease is mantakassa (Howlett et al., 1990; Bruyn and Poser, 2003). Tropical ataxic neuropathy (TAN) is another health problem associated with the continual consumption of improperly processed cassava products, especially in Africa (e.g. Nigeria). TAN is used to describe several neurological syndromes attributed to toxicolutricative causes. Symptoms of TAN include tongue pain, optic atrophy, neurosensory deafness, and sensory gait ataxia (Bolarinwa et al., 2016). Diseases caused by CGs are described in more detail in the study by Nyirenda (2020).

The detection of CGs in food is important for public health protection, as improper food processing can release toxic cyanide that is highly harmful to humans. Additionally, the detection of these substances plays an important role in complying with food regulations, which set maximum allowable concentrations of CGs in various foods (Cressey and Reeve, 2019; Vetter, 2000). Many countries have already introduced regulations to reduce the risk of cyanide exposure from consuming foods that contain these compounds.

The detection of CGs depends on several factors (Cressey and Reeve, 2019; Tahir et al., 2024), such as:

CGs are quantified using direct and indirect methods of determination. The direct method targets CGs as the molecules of interest, while the indirect method focuses on the released HCN after hydrolysis (Azmi, 2019; Hartanti and Cahiani, 2020) ( Figure 6 ). One of the most well-known indirect methods of determination is the Guignard sodium picrate test (Hartanti and Cahiani, 2020).

Many reviews summarize this issue. Analytical methods for the determination of amygdalin are clearly presented by Popa et al. (2021) who highlight various analytical methods with detailed parameters. Zhao et al. (2024) focus on the issue in the comparison of HPLC/UPLC methods for the determination of CGs. Risk assessment of food safety associated with foods containing CGs was addressed by Cressey et al., with a focus on rural New Zealand (Cressey et al., 2022).

Cyanogenic glycosides represent a broad group of structurally differing compounds with various biochemical properties. Some organisms use cyanogenic acids as protection against predators. These compounds are also present in many plants, which in some countries form an important part of the diet for local populations. The harmful effects of CGs on the human body are fairly well researched, and there is a vast database of scientific studies on their toxic properties. The risks associated with the consumption of processed and unprocessed plant parts containing these substances can now be more accurately estimated. Although cyanide itself is extremely toxic and can cause severe poisoning, some plants containing CGs are the subject of intensive research, especially for their potential in therapeutic applications. Current studies are focusing on the synthesis of derivatives of these compounds that have enhanced anti-tumor effects, which opens up new opportunities for cancer treatment. However, it is essential that the risks associated with the release of cyanide, which remains highly toxic, are not overlooked in this research. As a result, much research is focused on developing technologies and methods that allow the breakdown of cyanide compounds to be controlled or minimized with the aim to avoid adverse health effects.