Sulfur compounds: From plants to humans and their role in chronic disease prevention

Abstract

Sulfur is essential for the health of plants and is an indispensable dietary component for human health and disease prevention. Its incorporation into our food supply is heavily reliant upon the uptake of sulfur into plant tissue and our subsequent intake. Dietary requirements for sulfur are largely calculated based upon requirements for the sulfur-containing amino acids (SAA), cysteine and methionine, to meet the demands for synthesis of proteins, enzymes, co-enzymes, vitamins, and hormones. SAA are found in abundance in animal sources and are relatively low in plants. However, some plants, particularly cruciferous and allium vegetables, produce many protective sulfur-containing secondary metabolites, such as glucosinolates and cysteine sulfoxides. The variety and quantity of these sulfur-containing metabolites are extensive and their effects on human health are wide-reaching. Many benefits appear to be related to sulfur’s role in redox biochemistry, protecting against uncontrolled oxidative stress and inflammation; features consistent within cardiometabolic dysfunction and many chronic metabolic diseases of aging. This narrative explores the origins and importance of sulfur, its incorporation into our food supply and dietary sources. It also explores the overarching potential of sulfur for human health, particularly around the amelioration of oxidative stress and chronic inflammation, and subsequent chronic disease prevention.

Keywords:

• Sulfur-containing secondary metabolites
• cruciferous vegetables
• allium vegetables
• glucosinolates
• cysteine sulfoxides
• cardiometabolic

Introduction

Sulfur is essential for the successful growth of plants, and many biologically important pathways in both animals and humans. Within plants, a key role of sulfur is to provide defense against pathogens and facilitate healthy plant growth (Nwachukwu, Slusarenko, and Gruhlke 2012). Within humans, sulfur is widely acknowledged as a component of two principal sulfur-containing amino acids (SAA); cysteine and methionine; and subsequent protein recycling (Ingenbleek and Kimura 2013). Sulfur is also necessary for the synthesis of key compounds with central roles in metabolic health and homeostasis (Ingenbleek and Kimura 2013). Such examples include the production of S-adenosylmethionine (SAM), $\gamma$-glutamyl-cysteine-glycine, lipoic acid, various hormones, some vitamins (thiamine, biotin), enzymes, co-enzymes (coenzyme A), other amino acids (cystine, taurine, homocysteine) and varying cofactors (Ingenbleek and Kimura 2013). Furthermore, sulfur-containing compounds are an integral component of many antioxidants important for intracellular defense mechanisms, such as glutathione (GSH), lipoic acid, and N-acetyl cysteine (Colovic et al. 2018). However, apart from some marine microorganisms, e.g., algae and microbacteria, animals are unable to utilize sulfur in its organic form. Sulfur must first be sequestered from the environment and become ‘fixed’ into an organic, living, carbon-based structure, i.e., plant tissue (Francioso et al. 2020). Animals, including humans, then rely on the consumption of sulfur-containing plant tissue for their sulfur intake (Maruyama-Nakashita 2017). This intake can be divided into either SAA (predominantly, although not exclusively, animal proteins) or sulfur-containing secondary metabolites (predominantly plant-based sources). Importantly, higher-order animals (i.e., humans) can synthesize additional sulfur-based compounds following consumption (Francioso et al. 2020). To date, dietary recommendations for sulfur are largely based upon the well-established proteinogenic roles of SAA. Yet the overarching importance of plant-based, non-SAA compounds for human health and disease prevention is gaining traction. These plant-derived sulfur-containing secondary metabolites play key roles in antioxidant and anti-inflammatory pathways yet are overlooked within our dietary recommendations.

Oxidative stressors come in several forms; as by-products of normal internal metabolism, or from external environmental and xenobiotic influences (Mukwevho, Ferreira, and Ayeleso 2014). When given inadequate antioxidant defenses to maintain homeostasis against stressors, a hyper-oxidative state can occur, resulting in elevated oxidative stress, inflammation, protein and lipid peroxidation, as well as cellular membrane and DNA damage (Mukwevho, Ferreira, and Ayeleso 2014). Uncontrolled oxidative stress, activation of innate immunity and the ensuing pro-inflammatory response contributes to accelerated aging in humans, with the concomitant low-grade chronic inflammation aptly coined ‘inflammaging’ (Livshits and Kalinkovich 2019; Liguori et al. 2018; Franceschi et al. 2018). Elevated oxidative stress and inflammation have been implicated in the pathogenesis of many lifestyle-related chronic cardiometabolic diseases, such as atherosclerosis-related cardiovascular disease (CVD), vascular disorders, hypertension and chronic kidney disease, dyslipidemia, poor glucose regulation, cognitive decline and some cancers (da Costa et al. 2019; Ahmed et al. 2017; Ferrucci and Fabbri 2018; Ruhee et al. 2020). Furthermore, chronic metabolic-related diseases are the largest contributor to death globally (Raghupathi and Raghupathi 2018; World Health Organisation 2018). In humans, dietary intake of sulfur has wide-reaching activity across multiple biological pathways, including antioxidant defense pathways making it essential for our health and wellbeing (Mukwevho, Ferreira, and Ayeleso 2014). Despite this, an important, yet often overlooked source of sulfur is found within sulfur-rich vegetables, such as cruciferous (e.g., broccoli, cauliflower, cabbage) and allium (e.g., garlic, onion, leek) vegetables, in the form of sulfur-containing secondary metabolites. There have been numerous health benefits associated with higher consumption of sulfur-rich vegetables, including cardiovascular, glycemic, neuroprotective, anticancer, anti-microbial, anti-inflammatory, and antioxidant (Petropoulos, Di Gioia, and Ntatsi 2017). This narrative provides insight into the origins of sulfur within our environment, its incorporation into our food supply, dietary sources, and their overarching importance in human health. We particularly focus upon sulfur’s potential in ameliorating chronic pro-oxidation and hyper-inflammatory pathways and subsequent, chronic disease prevention.

The origins of sulfur

Environmental sources

Sulfur is widely found in various forms in nature; as free sulfur, inorganic sulfides and sulfates, sulfur oxides, sulfurous gases as well as bound within organosulfur compounds in plant tissues and living organisms (Hawkesford 2007; De Kok et al. 2007; Stefels 2007; Tausz 2007). The largest source of inorganic sulfur is found within our marine (aquatic) environments (which include wetlands, salt marshes and estuaries). A diverse number of water-based microorganisms such as phytoplankton, algae and bacteria, release sulfur gases, such as hydrogen sulfide (H₂S) into the atmosphere (Francioso et al. 2020; De Kok et al. 2007; Stefels 2007). Another substantial source of sulfur is of anthropogenic origin, emitted during the industrial processing of minerals, petroleum oils and coal (De Kok et al. 2007). An especially high terrestrial source comes from volcanic and geothermic emissions spilling into land reservoirs (Francioso et al. 2020; Hawkesford 2007). The continual recycling of sulfur between the terrestrial, aquatic and atmospheric environments are collectively referred to as our global sulfur cycle (Hawkesford 2007). This global sulfur cycle continuously feeds sulfur into plant tissue in its reduced form, i.e., as sulfate, with very sophisticated adaptive mechanisms regulating its uptake, which in turn are strongly influenced by demand, availability, and environmental conditions.

Metabolism in plants

Sulfur is essential for the formation of proteins, vitamins, enzymes, and defensive compounds in plants, as well as the fixation of, and optimal utilization of nitrogen to enhance plant yield (Maruyama-Nakashita 2017; Hawkesford 2007). Consequently, sulfur is considered an essential nutrient for the growth and health of all plants. Sulfur deficiency results in stunted plant growth, reduction in crop yield and lowered plant defense (Künstler et al. 2020; Hawkesford 2007; Maruyama-Nakashita 2017). Therefore, the application of sulfur-rich fertilizers, during periods of plant growth is a common agricultural practice (Petropoulos, Di Gioia, and Ntatsi 2017; Künstler et al. 2020).

The amount of sulfur taken up by a plant is primarily influenced by the amount readily available within the soil, as well as the actions of sulfur-sensing transporters (Maruyama-Nakashita 2017; Gigolashvili and Kopriva 2014). The assimilation of sulfur, as well as its formation into varying sulfur-based compounds, is influenced not only by abiotic environmental conditions (e.g., water, temperature, salinity, nutrient supply), but also by the need for resistance against biotic stressors (Künstler et al. 2020; Nguyen et al. 2020).

Sulfate to cysteine

Sulfate is primarily taken up through the plant root system from the rhizosphere by way of sulfur-sensing sulfate transporters located within root epidermal cells and this process is influenced by several regulatory proteins and enzymes inside plant cells (Maruyama-Nakashita 2017). The uptake and assimilation into plant tissue is tightly regulated to maintain the supply and demand for sulfur (Künstler et al. 2020). Whilst some sulfate is stored in root cell vacuoles, the majority is transported into the xylem, whose primary role is to passively transport water and dissolved minerals to the shoots of growing plants, where a sulfur reservoir is maintained through a series of reducing reactions to trap and incorporate sulfur into a variety of cellular metabolites (Hawkesford 2007). Much of this reducing activity occurs within the adenosine triphosphate (ATP) and reductant rich chloroplasts found in the photosynthetic tissues of plants (Hawkesford 2007). After activation by ATP sulfurylase to form adenosine-5′-phosphosulfate (APS), APS reductase (APR) reduces the APS to sulfite, and lastly ferridoxin-dependent sulfite reductase (SiR) catalyzes the six-electron reduction to sulfide (Maruyama-Nakashita 2017; Jez 2019; Khan et al. 2010). Sulfide, in the presence of O-acetyl-L-serine, is acted upon by O-acetylserine(thiol)lyase (OASTL) in the cytosol, chloroplasts and mitochondria of plant cells to form cysteine (Romero et al. 2014). Cysteine is the first organic form of sulfur available in plant metabolism (Maruyama-Nakashita 2017). Cysteine is then incorporated into thousands of different proteins in plants through protein synthesis in the cytosol, chloroplast and mitochondrion and is also the critical intermediary step for assimilation of sulfur into many beneficial sulfur-containing compounds (Hawkesford 2007). A diagrammatic overview indicating sulfur metabolism in plants is illustrated in Figure 1.

Figure 1. Overview of sulfur metabolism in plants; adapted from (Jez 2019; Maruyama-Nakashita 2017).

Cysteine to sulfur-based compounds

Plant metabolism is divided into (1) primary metabolism, required for plant survival and widely conserved in the plant kingdom, and (2) secondary metabolism that is less conserved and often specific to different plant species. Primary sulfur metabolism involves the initial steps of converting sulfur through to cysteine, methionine and GSH (Maruyama-Nakashita 2017). Secondary sulfur metabolism typically refers to the subsequent production of sulfur-containing secondary metabolites (Künstler et al. 2020). The sulfur-containing secondary metabolites, sometimes referred to as ‘defensive compounds’ can act as either a phytoanticipin; an antimicrobial compound already present in anticipation of pathogenic exposure, or as a phytoalexin; a compound synthesized in response to ward off stress and improve defense mechanisms (Nwachukwu and Slusarenko 2014; Nwachukwu, Slusarenko, and Gruhlke 2012). Sulfur-containing secondary metabolites can include glucosinolates, thiosulfinates (e.g., allicin, derived from cysteine sulfoxides), reactive sulfur species (e.g., H₂S and sodium sulfate), and antimicrobial peptides (e.g., defensins and thionins) (Künstler et al. 2020).

Glutathione is a primary sulfur-containing tripeptide (glycine, glutamate and cysteine), synthesized by the action of $\gamma$-glutamylcysteine synthetase (aka glutamate-cysteine ligase) and GSH synthetase (Jez 2019). In a normal physiological non-stressed state, glutathione exists 90% in its non-oxidized/reduced form (GSH) and 10% in its oxidized form (glutathione disulfide [GSSG]) (Künstler et al. 2020; Zechmann 2020). GSH has a well-recognized role as a central regulator for plant redox signaling, mediating plant defenses, mitigating the damage of ROS and activating defensive genes in plants (Jez 2019; Künstler et al. 2020; Zechmann 2020). Cellular GSH is distributed throughout the plant cells in both the cytosol and major organelles (Zechmann 2020). GSH is readily oxidized to GSSG when its cysteine residues are exposed to electrophilic substances (e.g., ROS, free radicals). GSH-transporters can re-distribute oxidized GSH (i.e., GSSG) to various subcellular components of the cells in response to attack, thereby triggering defensive compounds such as salicylic acid, jasmonic acid, auxins, abscisic acid and ethylenes within the plant tissue (Künstler et al. 2020; Zechmann 2020). Aside from being a major plant defensive compound, GSH is also required for the synthesis of many glucosinolates in plants; a process involving conjugation of cysteine with an amino acid, (e.g., methionine, tryptophan), in the presence of 3-phosphoadenosine-5′-phosphosulfate (PAPS) and GSH. (Maruyama-Nakashita 2017; Mugford et al. 2011).

Glucosinolates

Glucosinolates are sulfur-containing non-reactive phytoanticipins found in the plant order Brassicales which include a number of major vegetable crop plants (e.g., broccoli, cabbage, cauliflower, etc.). Glucosinolates are constructed from a sulfonated oxime group that is linked to a thioglucose group and a side chain of an amino-acid (Esteve 2020). These molecules are classified into three categories, aliphatic, benzenic or indolic, according to the amino acids that provide the side chain. The side chain for the aliphatic glucosinolates are derived from alanine, valine, leucine, isoleucine, or methionine, the benzenic (also commonly referred to as ‘aromatic’) are derived from phenylalanine or tyrosine, whilst the indolic are derived from tryptophan (Mitreiter and Gigolashvili 2021; Nguyen et al. 2020). In their native state, glucosinolates are stable and are not normally biologically reactive. They only fulfill significant functions when they are activated following stress (pathogenic or otherwise) through hydrolysis of their glucose residue by myrosinases (also known as thioglucosides) to form isothiocyanates, thiocyanates, nitriles and epithionitriles within plant tissues (Künstler et al. 2020). In plants, glucosinolates and myrosinases are kept spatially separate, either in different compartments of the same cell or in different cell types (Kliebenstein, Kroymann, and Mitchell-Olds 2005). They typically only come together to form isothiocyanates, thiocyanates, nitriles and epithionitriles when cells are broken through herbivory or disease. Glucosinolates are well recognized for their protective roles against oxidative stress and pathogen resistance in plants following their synthesis, storage, release, and conversion to defense compounds through hydrolysis (Künstler et al. 2020; Zechmann 2020).

At one time it was thought glucosinolate production was not regulated in plants, largely due to the wide diversity of levels and types in many different growth conditions and following plant treatments (Mitreiter and Gigolashvili 2021). However, it is now known that this complexity arises from control systems that are indeed highly complex, containing multiple layers which are interconnected to other parts of plant metabolism. As a fortunate accident of history, Arabidopsis thaliana, the first plant to have its genome sequenced and the ‘lab rat’ of plant science, is itself from the Brassicaceae plant family and produces over 40 glucosinolates (Sønderby et al. 2010). This has resulted in extensive investigation into the mechanism of control of glucosinolate synthesis. A strong level of control of glucosinolate production is necessary as the maximal rate can use up to 15% of the energy from photosynthesis at any one time (Bekaert et al. 2012). Fitness benefits from glucosinolate production are seen during intra-specific competition between high and low glucosinolate varieties when they are challenged by herbivores (Züst et al. 2011). Research in Arabidopsis has shown that a series of myelobastosis (MYB) and basic helix-loop-helix (bHLH) transcription factors form a transcriptional complex that regulates glucosinolate production in plants and makes it responsive to a wide variety of biotic and abiotic stresses (Mitreiter and Gigolashvili 2021). This transcriptional hub is also highly responsive to an array of plant hormones, enabling the accumulation of glucosinolates to be finely tuned to the potential need for their deployment as defense compounds. The most extensive work linking glucosinolates with hormone levels has been performed in the context of the jasmonate (JA) signaling pathway. When plants are attacked such as in the presence of pathogens or herbivores (Schweizer et al. 2013), mechanical damage or even physical touch through rainfall (Van Moerkercke et al. 2019), it is perceived by plant pattern recognition receptors which induce JA synthesis, thereby directly activating MYB transcriptional factors and subsequent glucosinolate production pathways (Mitreiter and Gigolashvili 2021).

S-Alk(en)yl cysteine sulfoxides

S-alk(en)yl cysteine sulfoxides (ASCOs) are found predominantly within the allium family of vegetables (i.e., garlic, onion, leeks) and their thiosulfinate derivatives are behind much of the aroma and flavor compounds of these vegetables (Yamazaki et al. 2010). Around ten ASCOs have been identified to date in various alliums (including homoisoalliin, butiin, cycloalliin, deoxyalliin, ethiin and S-ethyl-L-cysteine sulfoxide) (Petropoulos, Di Gioia, and Ntatsi 2017), although the dominant cysteine sulfoxides are alliin, methiin, isoalliin and, to a lesser extent propiin (Horníčková et al. 2010; Rektorisova et al. 2020; Rose et al. 2019).

The majority of sulfur found in alliums is present within $\gamma$-glutamyl S-alk(en)yl cysteines (GSAkCs) and $\gamma$-glutamyl S-alk(en)yl cysteine sulfoxides, in addition to S-alk(en)ylglutathione, various selenomethionine and GSH conjugates, each acting as precursory intermediates in ASCO production (Kodera et al. 2020; Yoshimoto and Saito 2019). Although the synthesis of ASCOs is not yet fully understood, it is assumed that GSAkCs undergo deglutamylation and S-oxygenation to yield their relevant ASCO (Yoshimoto and Saito 2019). For example, $\gamma$-glutamyl S-allylcysteine converts to S-allylcysteine before becoming the ASCO, S-allyl cysteine sulfoxide (i.e., alliin) (Kodera et al. 2020). A similar process occurs for the remaining ASCOs (i.e., $\gamma$-glutamyl-S-methylcysteine becomes S-methylcysteine sulfoxide [methiin]; $\gamma$-glutamyl-S-1-propenylcysteine becomes S-propenyl cysteine sulfoxide [isoalliin] and $\gamma$-glutamyl-S-1-propylcysteine becomes S-propylcysteine sulfoxide [propiin]) (Yoshimoto and Saito 2019; Rektorisova et al. 2020). Alternatively, the production of methiin has been found to result from the deglyclation (removal of glycine) from S-methylglutathione, before its deglutamylation and S-oxygenation process, described above (Yoshimoto and Saito 2019; Lancaster and Shaw 1989). Whilst more research into understanding pathways of cysteine sulfoxide synthesis is needed, it is evident that significant increases in overall ASCOs occur with both maturity and aging of alliums (Kodera et al. 2020; Yoshimoto and Saito 2019). Collectively, the ASCOs make up over half of the total sulfur content within the mature garlic bulb (Rektorisova et al. 2020). Like glucosinolates, these compounds are relatively inert, until upon tissue destruction (i.e., slicing, chewing), they are acted upon by cysteine sulfoxide lyases (e.g., alliinase) stored within separated vacuoles in the plant, resulting in sulfenic and pyruvic acids, and ammonia (Rektorisova et al. 2020). Such examples of the multiple downstream sulfur-containing derivatives are thiosulfinates, various sulfides, and ajoenes (Rektorisova et al. 2020; Yoshimoto and Saito 2019).

Other sulfur-based compounds

Despite this review focusing primarily on glucosinolates and ASCOs, there are other sulfur-containing secondary metabolites synthesized from plants that are of research interest (e.g., H₂S, sulfur dioxide [SO2], antimicrobial peptides) (Künstler et al. 2020). Emerging evidence suggests that H₂S is able to directly trigger plant defense and adaptive mechanisms following exposure to pathogens or harmful environmental stimuli (Vojtovič, Luhová, and Petřivalský 2021). Similarly, SO₂ has been found to advantageously regulate various defensive signaling pathways within plants, whilst defensins and thionins (i.e., anti-microbial peptides) have been shown to provide broad antiviral, antifungal and antibacterial protection (Künstler et al. 2020).

The diverse potential of plant-based sulfur-containing metabolites

These sulfur-containing metabolites activate multiple detoxification reactions in plant tissue following exposure to stress, regulating reactive oxygen species, enhancing antioxidant defense and detoxification pathways, thereby providing additional protection against both pathogenic and environmental stressors as required (Künstler et al. 2020; Gullner et al. 2017; Hasanuzzaman et al. 2017). It is worth noting that some sulfur-containing metabolites can be argued as having both phytoanticipant and phytoalexin characteristics, with more research into understanding the complexity behind sulfur chemistry and its metabolites required. The type and quantity of different sulfur-containing metabolites produced within plants are highly dependent on sulfur availability although also unique to each plant species (Vojtovič, Luhová, and Petřivalský 2021).

The considerable potential of many sulfur derivatives is largely due to sulfur being classed as a chalcogen, known for having biologically diverse characteristics including the ability to provide unique redox potential and atom exchange reactions (Francioso et al. 2020). Therefore, these characteristics equate to sulfur derivates being highly reactive, performing “a complex labyrinth of interacting signaling and control pathways” (Giles et al. 2017, 1). As research has enabled greater understanding, it has been suggested that sulfur is possibly more important than reactive oxygen species (ROS) or reactive nitrogen species (RNS) in our antioxidant defense systems (Giles et al. 2017). The presence of sulfur provides unique versatility and diversity of activity to simple carbon, hydrogen, nitrogen and oxygen compounds from plants (Francioso et al. 2020). This is enabled by a diverse set of enzymes that hydroxylate, methoxylate, oxidize, desaturate or benzoylate sulfur-rich compounds (Halkier and Gershenzon 2006), coupled to a wide series of chemical reactions that occur during their interconversions. As a result, after sulfur has been incorporated into plant tissue, a vast number of sulfur-containing compounds can be formed, well beyond the primary metabolic products—cysteine, methionine, and GSH. Up to and including the year of 2012, it was recognized that there may be as many as 550 organosulfur compounds in plants (Iranshahi 2012) and over 130 different glucosinolates alone reported in plants (Blažević et al. 2020). With emerging analytical techniques, many of these sulfur-containing compounds and their downstream metabolites are being isolated and subsequently being increasingly investigated for their health potential in humans.

Sulfur intake in humans

Sulfur intake in humans can be divided into two categories, the SAA and the sulfur-containing secondary metabolites.

Sulfur-containing amino acids

The four SAA include cysteine, methionine, homocysteine, and taurine, with only the former two being proteinogenic (Brosnan and Brosnan 2006). Both methionine and cysteine are found in abundance in animal products, and to a lesser extent, plant proteins (Ingenbleek and Kimura 2013). Therefore, animal meats, such as lamb, beef, chicken, seafood, as well as eggs and dairy, are considered highly reliable sources of SAA, whilst those consuming ‘health-conscious’ diets free of red meat or animal proteins tend to have lower (albeit acceptable) levels of sulfur than their meat-eating counterparts (Nimni, Han, and Cordoba 2007). Methionine is an essential SAA and relies on dietary intake, whilst cysteine is considered semi-essential (able to be synthesized from an adequate supply of methionine) (Brosnan and Brosnan 2006). It is estimated that humans require a daily SAA intake (predominantly as methionine) of approximately 15 mg/kg body weight (i.e., ∼1,050 mg/day for a 70 kg human) (World Health Organisation 2007).

Briefly, cysteine is the initial step of sulfur incorporation into organic tissue (see Figure 1) which, in conjunction with methionine and other amino acids, assists in production of glucosinolates and GSH (Maruyama-Nakashita 2017). GSH is the most abundant non-protein thiol in the human body, with a major role in tightly regulating cellular redox status (Franco et al. 2007; Teskey et al. 2018; G. Wu et al. 2004). GSH is a significant reservoir for cysteine, accounting for an estimated 65% of mammalian whole-body cysteine flux, indicating its importance in regulating our antioxidant defensive system (Franco et al. 2007; G. Wu et al. 2004). Both GSH and its oxidized counterpart (GSSG), together act as a major redox couple, heavily involved in mediating detoxification, thiol disulfide reactions important for cellular signaling, and activation of defensive pathways (Jez 2019; Kalinina and Gavriliuk 2020; Zechmann 2020; Franco et al. 2007). In similarity with plants, GSH in mammals is largely found in its reduced form, shifting toward its oxidized (and more toxic) GSSG form upon exposure to stress (i.e., exposure to ROS/RNS) (Kalinina and Gavriliuk 2020). This in turn triggers feedback recycling of GSSG back to GSH (via glutathione reductase), thereby increasing Nrf2 translocation and subsequent de novo synthesis of GSH as well as enhanced transcription of defensive genes (Teskey et al. 2018; G. Wu et al. 2004).

As with many sulfur-based compounds, GSH is widely recognized as an important redox buffer in many chronic conditions such as cancer, cardio-metabolic dysfunction, diabetes, acquired immunodeficiency syndrome (AIDS), neurodegenerative, hepatic dysfunction as well as pathologies associated with aging (Franco et al. 2007; Kalinina and Gavriliuk 2020; Matuz-Mares et al. 2021; Teskey et al. 2018). More recently, GSH has been explored for its potential to mitigate the severe oxidative stress associated with deleterious COVID-19 pathogenesis and/or mortality (Khanfar and Qaroot 2020; Kumar et al. 2021; Polonikov 2020).

Methionine, through its activation to S-adenosylmethionine has an important role with the latter acting as a methyl donor for wide-reaching methylation reactions within almost every cell in the human body (Brosnan and Brosnan 2006). The demethylation of methionine produces homocysteine, important for the recycling of both methionine and cysteine. Provided there is adequate supply of B-vitamins (i.e., folate, B6, B12) as well as choline and betaine, the transulfuration pathway recycles homocysteine back to cysteine, or alternatively to methionine via the transmethylation pathway (Dong, Sinha, and Richie 2018).

Unique to cysteine is its readiness to form disulfide bonds, making it crucial in protein structure and in the formation of protein-folding pathways (Brosnan and Brosnan 2006). Furthermore, these disulfide bonds act as a ‘gateway’ conveniently located near the surface of cells for the convenient export of proteins to other areas of the body. In contrast, methionine displays hydrophobic tendencies, therefore, is mostly found contained within the protective interior hydrophobic core of proteins (Brosnan and Brosnan 2006). However, occasionally methionine residues can be found upon the exposed outer cell layer and, when reducing enzymes are impaired and these residues accumulate, the cell becomes vulnerable to oxidation and resultant inflammation (Brosnan and Brosnan 2006). It is this potential to trigger metabolic changes that has led to studies investigating the link between high intake of SAA from methionine-rich animal products with chronic inflammation associated with cardiometabolic risk and other chronic diseases of aging (Dong et al. 2020; Dong, Sinha, and Richie 2018). Recent results from the Third National Examination and Nutritional Health Survey (NHANES III) study found that SAA intake among American adults was in fact 2.5 times higher than the recommended ‘estimated average requirement’. Among the 11,576 participants, consuming higher intakes of SAA (positively associated with animal protein intake, i.e., meats) was associated with significant increases in cardiometabolic diseases, even after adjusting for confounders and known risk factors such as cholesterol and blood glucose (Dong et al. 2020). However, most importantly, the above estimation and associated effects are for SAA alone and overlook requirements for other non-SAA sources of sulfur intake, such as the sulfur-containing secondary metabolites.

Sulfur-containing secondary metabolites from plants

To date, there are no recommendations for an optimal total daily sulfur intake, nor are comprehensive databases readily available to quantify non-SAA sulfur-based intake from foods. Nonetheless, in addition to animal meats, many other food groups contribute to sulfur intake, including, wheat and starchy grains, dairy, and vegetables (Doleman et al. 2017). A recent laboratory-based study analyzed sulfur content within a variety of commonly consumed foods (e.g., fruits, vegetables, meats, fish, pastas, rice, and eggs), and then calculated daily intakes across 41 one-day food diaries self-reported from 32 healthy individuals. They estimated that ∼90% of total sulfur intake in the human population may be derived from non-SAA sources (Doleman et al. 2017), in the form of secondary metabolites, for which a recommended intake has not yet been established. Doleman et al. (2017) found that cruciferous and allium vegetables contribute ∼42% of total sulfur intake with the majority of this (∼95%) from cruciferous vegetables.

Cruciferous vegetables include broccoli, cabbage, Brussels sprouts, cauliflower, kale, rapeseed, radish, turnip, watercress, and mustard, and belong to the plant family Brassicaceae (Nguyen et al. 2020). The Alliaceae genus of vegetables includes the alliums, such as garlic, onion, leeks, chives, scallion (spring onion) and shallots (Petropoulos, Di Gioia, and Ntatsi 2017). It is the presence of sulfur compounds that produce the distinctive aroma and flavor of these vegetables when sliced (Bell et al. 2018; Yamazaki et al. 2010).

Sulfur-rich cruciferous and allium vegetables are recognized for their antioxidant, anti-inflammatory, hepatoprotective, antimicrobial, hypoglycemic, cardiovascular, anticarcinogenic, neuro and renal protective potential (Shang et al. 2019; Petropoulos, Di Gioia, and Ntatsi 2017). Observational studies suggest that consuming higher intakes of cruciferous and allium vegetables are inversely associated with atherosclerotic vascular disease mortality, the incidence of myocardial infarction and cerebrovascular disease (Blekkenhorst et al. 2017; Blekkenhorst et al. 2018), type-2 diabetes mellitus (T2DM) (Jia et al. 2016) and cancer risk (Aune et al. 2017). The increasingly recognized role of plant-based sulfur-containing secondary metabolites within these vegetables and their importance for human health and disease prevention is gaining traction.

Oxidative stress, inflammation and chronic disease

Reactive oxygen species (ROS) refers to the presence of reactive forms of oxygen (both radical and non-radical derivatives), largely formed through normal cellular metabolism or in response to environmental or xenobiotic stressors (Mukwevho, Ferreira, and Ayeleso 2014). All metabolic processes undergo a continual state of redox reactions kept in check by an appropriately coordinated immune, antioxidant, and inflammatory response (Ahmed et al. 2017). When exposed to a potential threat, ROS is generated through the activation of mast cells, monocytes, lymphocytes and macrophages recruited to the area of concern (Ahmed et al. 2017). Without adequate immune, inflammatory, and antioxidant defenses to maintain homeostasis, a pro-inflammatory state of hyper-oxidation can occur resulting in oxidative stress, hyper-inflammation, and DNA damage. Uncontrolled, this process is accompanied by marked increases in inflammatory mediators such as cytokines, chemokines and prostaglandins, which in turn, trigger further signaling mediators and transcription factors exacerbating the inflammation, disrupting homeostasis, and potentially becoming chronic in nature (Ahmed et al. 2017).

Central to much of the regulation of our inflammatory pathways and antioxidant defense are the activation or suppression of two key cytoplasmic transcription factors; nuclear factor-ĸB (NF-ĸB) and erythroid-2 related factor (Nrf2) (T. Liu et al. 2017; Ahmed et al. 2017). The activation of NF-ĸB triggers the production of pro-inflammatory cytokines, chemokines and adhesion molecules (T. Liu et al. 2017). In addition, NF-ĸB is responsible for the switching on of M1 macrophages (from the favored anti-inflammatory M2 macrophages) and in doing so, induces genes that further encode for tumor necrosis factor-α (TNF-α), several interleukins (IL) such as IL-1, IL-6, and IL-12, and cyclooxygenase (T. Liu et al. 2017). Furthermore, elevated TNF-α has been implicated in initiating transcription of NF-ĸB, accelerating ROS production, thrombosis, endothelial dysfunction and remodeling, and reducing nitric oxide (NO) bioavailability (H. Zhang et al. 2009). Detrimental microvascular and macrovascular changes, accelerated aging, and the development of diabetes have been reported in response to high TNF-α in both animal and human studies (H. Zhang et al. 2009).

In response to a stressor, Nrf2, which is generally sequestered to Kelch-like ECH-associated protein (Keap-1) becomes dissociated. As a result, Nrf2 is then free to translocate to the cell nucleus where it triggers the expression of numerous genes as part of our sophisticated antioxidant Nrf2/Keap1/antioxidant response element (ARE) defense system (Ahmed et al. 2017). Nrf2 aids in the transcription of hundreds of cytoprotective genes involved in inhibiting pro-inflammatory cytokines, chemokines, and cell adhesion molecules (CAMs), cyclooxygenase-2 (COX-2), matrix metalloproteinases (MMPs) and inducible nitric oxide synthase (iNOS), thus inducing relevant detoxification pathways and overall cellular defense mechanisms (Ahmed et al. 2017). Adhesion molecules, such as vascular cell and intercellular adhesion molecules (VCAM-1 and ICAM-1, respectively) appear early in vascular remodeling (Blankenberg, Barbaux, and Tiret 2003) augmented by pro-inflammatory changes and inadequate anti-oxidant capability. Furthermore, Nrf2 is considered an emerging mediator of cardiometabolic disease risk (da Costa et al. 2019).

There is significant crosstalk between Nrf2 and NF-ĸβ, both competing for the same transcription coactivator (cAMP response element-binding [CREB] protein) for their counteracting roles in inflammatory responses (Bellezza et al. 2018). As such, this makes them key targets for chronic disease management (Ahmed et al. 2017). Inflammation has been strongly associated with increased cardiovascular risk (Arnold et al. 2021; Tuñón et al. 2018). Therefore, reducing heart and vascular inflammation is strongly targeted by many pharmaceuticals, immune-modulatory, and lipid-lowering therapies in an attempt to lower cardiovascular risk and mortality (Ridker 2018; Roman, Hernandez, and White 2020; Kottoor and Arora 2018; Moreira et al. 2015).

Plants, such as brassica and alliums, increase their sulfur-containing metabolites in response to both abiotic stress as well as pathogen exposure (Del Carmen Martínez-Ballesta, Moreno, and Carvajal 2013; Nguyen et al. 2020). Whilst many of the sulfur-containing plant metabolites act as primary defenders against disease and poor growth within plant tissue itself, it appears that many of these benefits may be extended to humans following their consumption (Fakhri et al. 2020). In fact, an observational study using data from the Shanghai Women’s Health cohort reported reduced levels of pro-inflammatory biomarkers (TNF-α, IL-1β and IL-6) in women consuming a higher intake of cruciferous vegetables (Y. Jiang et al. 2014). These vegetables and their sulfur-containing secondary metabolites may be a relatively cost-effective adjunct to lowering chronic cardiometabolic disease burden.

Sulfur-containing secondary metabolites—protection against chronic disease

The predominant plant-based sulfur-containing secondary metabolites within the cruciferous and allium vegetables include the glucosinolates and S-alk(en)yl cysteine sulfoxides, respectively. It should be restated that significant variations in metabolite quantity occur within these vegetables, influenced by not only defensive mechanisms described above, but by the naturally-occurring unequal distribution in different sections, i.e., leaves, roots, seeds (J. Chang, Wang, et al. 2019; Nguyen et al. 2020), stage of growth, and storage conditions post harvesting (Horníčková et al. 2010; Rektorisova et al. 2020). Furthermore, levels alter considerably according to dietary preparations (i.e., cooking techniques) (Baenas et al. 2019). For example, the estimated isothiocyanate bioavailability for cooked vegetables was found to be 2 to 6-fold lower when compared to raw and thoroughly chewed cruciferous vegetables (Vermeulen et al. 2006). Such growing and cooking variations result in wide compositional changes, details of which are well beyond the scope of this review. A detailed food composition database collectively compiling these compounds and the impacts of different growing conditions and cooking methods is needed. Common dietary sulfur-containing secondary metabolites found within cruciferous and allium vegetables are summarized below (Table 1).

Table 1. Common dietary sulfur-containing metabolites and their main derivatives within cruciferous and allium vegetables.

Major secondary metabolite/s	Post-hydrolysis derivative	Common food sources	References
Glucosinolates:
Glucoraphanin ^AL	Sulforaphane (ITC)	Broccoli sprouts, broccoli, cabbage, cauliflower, rocket, kale	(Farag and Motaal 2010; Bell et al. 2018; Vermeulen et al. 2006; J. Chang, Wang, et al. 2019; Liang et al. 2018)
Glucoerucin ^AL	Erucin (ITC)	Rocket, cabbage, radish, cauliflower, turnip, swede (rutabaga)	(J. Chang, Wang, et al. 2019; Vermeulen et al. 2006; Bell et al. 2018; Mithen 2007; Hwang et al. 2019; Liang et al. 2018)
Sinigrin ^AL	Allyl isothiocyanate (ITC) Allyl thiocyanate (TC)	Mustard seeds, wasabi, cabbage, cauliflower, broccoli, Brussels sprouts	(Vermeulen et al. 2006; Bell et al. 2018; Mithen 2007; Mazumder, Dwivedi, and du Plessis 2016; Liang et al. 2018)
Glucobrassicin ^IND	Indole-3-carbinol (I3C) (IND) 3,3-diindolymethane (DIM)	Broccoli, Chinese cabbage, green cabbage, red cabbage	(J. Chang, Wang, et al. 2019; Vermeulen et al. 2006; Bell et al. 2018; Hwang et al. 2019)
Glucoiberin ^AL	Iberin (ITC)	Chinese kale, broccoli, Brussels sprouts, cabbage, cauliflower	(J. Chang, Wang, et al. 2019; Vermeulen et al. 2006; Mithen 2007; Liang et al. 2018)
Glucobrassicanapin ^AL	4-pentanyl isothiocyanate (ITC)	Broccoli, Chinese cabbage, turnip, bok choi, pak choi, choy sum	(Vermeulen et al. 2006; Paul et al. 2019; Mithen 2007; Liang et al. 2018; Bell et al. 2018; Nugrahedi et al. 2017)
Gluconapin ^AL	3-butenyl isothiocyanate (ITC)	Chinese cabbage, Brussels sprouts, broccoli, kale, bok choi, brown mustard, kale, turnip	(J. Chang, Wang, et al. 2019; Vermeulen et al. 2006; Bell et al. 2018; Paul et al. 2019; Mithen 2007; Liang et al. 2018)
Gluconasturtiin ^ARO	Phenethyl isothiocyanate (ITC)	Radish, watercress, broccoli, cabbage, cauliflower, Brussels sprouts, turnip	(Paul et al. 2019; Mithen 2007; Liang et al. 2018)
Glucotropaeolin ^ARO	Benzyl isothiocyanate (ITC)	Broccoli, cabbage, watercress, nasturtium	(Bell et al. 2018; Dinh et al. 2021)
S-alk(en)yl cysteine Sulfoxides:
Alliin	Allicin (diallyl thiosulfinate) (TS)	Garlic	(Bernaert et al. 2012; P. Rose et al. 2005)
Methiin	Methanic acid, MMTSI, MMTSO (TS)	Garlic, onion, leek, kale, Brussels sprouts, cauliflower broccoli	(Bernaert et al. 2012; P. Rose et al. 2005; Coode-Bate et al. 2019)
Isoalliin	Propenyl-thiosulfinates (TS) Thiopropanal-S-oxide (LF)	Onion, leek, shallots	(Bernaert et al. 2012; P. Rose et al. 2005), Thomas 94

(ITC) isothiocyanate; (TC) thiocyanate; (IND) indole phytoalexin; TS thiosulfinate; MMTSI S-methyl methanethiosulfinate; MMTSO S-methyl methanethiosulfonate; LF lachrymatory factor; ^AL aliphatic glucosinolate; ^ARO aromatic glucosinolate; ^IND indole glucosinolate, classifications as per (Nguyen et al. 2020).

Glucosinolates (B-thioglucoside-N-hydroxysulfates)

Glucosinolate derivatives have been identified as having significant pharmacological potential, including antimicrobial, antioxidant, anti-cancer, and anti-inflammatory properties as well as anti-hypercholesterolemic, antihyperglycemic, anti-platelet, immunomodulatory, cardiac and neuroprotective effects (Russo et al. 2018; Maina et al. 2020; Mazumder, Dwivedi, and du Plessis 2016; Petropoulos, Di Gioia, and Ntatsi 2017; Dinkova-Kostova and Kostov 2012). Of the estimated >130 glucosinolates identified to date, the majority are found within cruciferous vegetables (Nguyen et al. 2020). A recent laboratory-based study analyzed total glucosinolate concentration across 8 brassicas (cruciferous) vegetables to be, from highest to lowest, cabbage, broccoli, cauliflower, mustard, Chinese cabbage, young radish, and kale (Hwang et al. 2019).

To recap, all glucosinolates contain both a thioglucose and sulfonate oxime group, in addition to one of three possible amino acid side chain structures (Esteve 2020). The side chains differentiate glucosinolates as being either aliphatic, benzenic (or aromatic) or indole, dependent on their main amino acid precursors. The aliphatic are the most prevalent glucosinolates, derived mostly from methionine (but also valine, alanine, leucine or isoleucine), whilst the aromatics are derived from tyrosine or phenylalanine, and the less-common indole glucosinolates are tryptophan-based (Fahey, Zalcmann, and Talalay 2001; Esteve 2020; Mitreiter and Gigolashvili 2021). The vast majority of glucosinolates are aliphatic, as shown in Table 1 (Nguyen et al. 2020). Considered biologically inactive water-soluble compounds, the glucosinolates undergo hydrolysis via the thioglucosidase myrosinase (made accessible by tissue maceration, i.e., chewing or slicing) or alternatively by thioglucosidases located within the intestinal flora (Mithen and Ho 2018; Tian et al. 2018; Narbad and Rossiter 2018). Subsequently, many downstream, biologically active derivatives are produced (Bell et al. 2018; Nguyen et al. 2020). These are predominantly the isothiocyanates and indoles, however, in the presence of specifier proteins, an alternative route favors the formation of thiocyanates, as well as nitriles and epithionitriles, with the potential for additional protective benefits (Eisenschmidt-Bönn et al. 2019; Nguyen et al. 2020). Despite evidence (mostly in animals) that thiocyanates inhibit iodine uptake, recent research suggests minimal adverse risks to thyroid health in humans consuming regular cruciferous intake (Felker, Bunch, and Leung 2016; Lietzow 2021). Allyl-, benzyl- and 4-methylsulfinylbutyl-glucosinolates each have the ability to form thiocyanates in the presence of these non-heme iron specifier proteins found in plants (Wittstock and Burow 2007). However, to date, sinigrin is the only glucosinolate known to form both isothiocyanates (i.e., allyl isothiocyanates) and thiocyanates (i.e., allyl thiocyanate) following hydrolysis (Nguyen et al. 2020). Collectively, it is these biologically active derivatives that are responsible for the sulfurous aroma and flavor of these vegetables, and for their biological activity (Z. Liu et al. 2021).

The post-hydrolysis derivatives that have gained the most notoriety for potential health benefits include isothiocyanates and indoles. Among the most studied metabolites and their derivatives are glucoraphanin, sinigrin, and gluconasturtin (isothiocyanates) and glucobrassican (indolic isothiocyanates) (Esteve 2020). Most studies have used in vitro and in vivo animal models. Some of their noted antioxidant, anti-inflammatory and cardiometabolic effects are tabled below (see Table 2).

Table 2. Antioxidant and anti-inflammatory activity of the main sulfur-containing derivatives (sourced from the hydrolysis of glucosinolates and S-alk(en)yl cysteine sulfoxides) found in cruciferous and allium vegetables.

Secondary metabolite	Post-hydrolysis derivative given	Study/Model	Effect	Reference
GLUCOSINOLATES
Glucoraphanin	Sulforaphane	BALB-/c mice with LPS-induced lung inflammation	↑ Nrf2 ↓ NF-ĸβ	(Qi et al. 2016)
	Sulforaphane	Mouse cell model Alzheimer’s Disease, (in vitro)	↑ Nrf2 ↓ inflammatory cytokines ↓ ROS	(Zhao, Zhang, and Chang 2018)
	Sulforaphane	Mice with T2DM-induced nephropathy	↑ Nrf2	(Z. Li et al. 2020)
	Sulforaphane	Mice with T2DM-induced cardiomyopathy	↑ Nrf2 ↑ AMPK activation	(Y. Sun et al. 2020)
	Sulforaphane	Dystrophin deficient C57BL/10ScSn-Dmdmdx/NJU mice	↓ NF-ĸB in muscle cells, ↑ Nrf2 ↓ TNF-α, IL-1β, IL-6	(C.-C. Sun et al. 2015)
	Sulforaphane	Bronchial epithelial cells exposed to SARS-CoV-2 (in vitro)	↓ IL-6 and IL-8	(Gasparello et al. 2021)
	Sulforaphane	T2DM patients	↑ glucose control ↓ gluconeogenesis	(Axelsson et al. 2017)
	Sulforaphane	C57BL/6J T2DM mice fed high-fat diet	↑ Nrf2 ↓ aortic wall thickness, fibrosis, inflammation, oxidative stress	(Y. Wang et al. 2014)
	Sulforaphane	LPS-induced RAW 264.7 murine macrophages (in vitro)	↑ Nrf2 ↓ inflammatory cytokines, ↓ iNOS and NO expression	(Ruhee, Ma, and Suzuki 2019)
	Sulforaphane	Male C57BL/6J mice induced with neuropathic pain	↓ inflammatory cytokines,	(C. Wang and Wang 2017)
Glucoerucin	Erucin	LPS-stimulated RAW 264.7 mice macrophages and skin	↓ NF-ĸβ ↓TNF-α, IL-6, IL-1β, NO and COX-2	(Cho, Lee, and Park 2013)
	Erucin	Streptozotocin-diabetes-induced mice with diabetic neuropathic pain	↓ nociceptive pain and pain-related behavior	(Lucarini et al. 2019)
	Erucin	Erucin-treated human platelets (in vitro) and a bolus dose given to C57BL/6 mice induced with thrombosis	↓ NF-ĸβ, ↓ thromboxanes, TGF-1β, CCL5, IL-1β (in vitro) ↓ thrombosis formation (in vivo—mice)	(Fuentes et al. 2014)
	Erucin	LPS-induced RAW 264.7 macrophages (in vitro)	↓ TLR4, TNF-α, COX-2, IL-8 and IL-1β, ↑ IL-10 ↓ inflammatory mediators	(Gugliandolo et al. 2018)
Sinigrin	Allyl ITC	LPS-induced inflammation in macrophages (in vitro)	↓ NF-ĸβ, ↓ NO, COX-2, PGE2, TNF-α, ↓ IL-6, IL-Iβ, IL-18	(H. W. Lee et al. 2017)
	Allyl ITC	TNF-α-induced inflammation, in vascular smooth muscle cells (in vitro)	↓ NF-ĸβ, ↓ VCAM-1	(H. W. Lee and Lee 2015)
	Allyl ITC	ApoE-deficient mice, fed high cholesterol	↓ atherosclerosis ↓ NF-ĸβ, ↓ VCAM-1, ICAM-1, ↓ oxLDL, CHL, HDL, LDL, TG, ↓ IL-6, ↓ TNF-α	(Jang et al. 2017; Jang and Pyo 2015)
Glucobrassican	I3C (and DIM)	Doxorubicin-induced toxicity in Swiss Albino mice	↑ Nrf2 ↓ NF-ĸβ	(Hajra et al. 2018)
	I3C	Sprague-Dawley cirrhotic rats	↓ NF-ĸβ	(T. Chang, Ho, et al. 2019)
	I3C	LPS-induced RAW 264.7 cells; LPS-induced lung injury mice	↓ NO, IL-6, IL-1β	(J. Jiang et al. 2013)
Glucoiberin	Iberin	A2780 and OVCAR-3 (ovarian cancer) cells (in vitro) and BALB/c nude mice	↓ cell growth, ↑ cell cycle arrest & apoptosis, mediated ROS production	(Gong et al. 2021)
	Iberin (plus iberverin & cheirolin)	NIH3T3 fibroblast cells (in vitro)	↑ Nrf2 ↑ HO-1 and $\gamma$GCS	(Ernst et al. 2013)
	Iberin (plus alyssin)	Hepatocellular carcinoma cell HepG2 (in vitro)	↑ apoptosis	(Pocasap, Weerapreeyakul, and Thumanu 2019)
	Iberin	Human neuroblastoma cells (in vitro)	↓ cell growth, ↑ cytotoxicity, ↑ apoptosis (via ↑ caspase and PARP activation)	(Jadhav et al. 2007)
Gluconasturtiin	Phenethyl-ITC (PEITC)	High-fat induced hypercholesterolemic C57BL/6 mice, fed PEITC for 13-wks	↓ NF-ĸβ, ↓ TC, LDL, TG, ↓ HOMA-IR and BW ↓LOX-1, COX-2, p-mTOR, PPAR$\gamma$ and ↑-AMPK	(Gwon et al. 2020; Gwon and Yun 2021)
	PEITC	Human THP-1 derived macrophages exposed to oxidized-LDL and LPS to mimic foam cell formation (in vitro)	↓ NF-ĸβ, ↓ foam cell formation and lipid accumulation	(Im, Gwon, and Yun 2021)
Glucotropaeolin	PEITC and BITC	High-fat-induced hypercholesterolemic C57BL/6 mice, fed PEITC/BITC for 18-wks	↓ TC, TG and BW ↓ lipogenic/adipogenic transcription (e.g., via PPAR$\gamma$, LXRα) expression	(Chuang et al. 2019)
Gluconapin	3-butenyl ITC	SLC mice fed corn oil to induce hypercholesterolemia	↓ postprandial hypertriglyceridemia	(Washida et al. 2010)
ALLIUMS
Alliin	–	LPS-stimulated inflammation in 3T3-L1 adipocytes (in vitro)	↓ pro-inflammatory gene expression, IL-6, MCP-1, Egr-1	(Quintero-Fabián et al. 2013)
	–	Dextran sulfate sodium-induced colitic mice	↓ NF-ĸβ, MAPK, NO ↓ iNOS, TNF-α, IL-6	(Shi et al. 2017)
	–	LPS-stimulated RAW264.7 macrophages treated w/wo alliin (in vitro)	↓ iNOS, IL-1β, IL-6, TNF-α, and NO	(Shi et al. 2017)
	–	8-wk old C57BL/6 mice randomized to control or high-fat diet w/wo alliin	↓ hyperglycemia, ↓ adipocyte hypertrophy, adipocytokines, ↓ IL-6	(Sánchez-Sánchez et al. 2020)
	–	MDA-TBA-induced oxidative damage (in vitro) and NBT-induced generation of superoxide	↑ hydroxyl and superoxide scavenging activity	(Chung 2006)
	Allicin	High-glucose/hypoxia-induced endothelial cell injury in murine aortic-endothelial cells (in vitro)	↓ NF-ĸβ ↓ ROS production	(S.L. Wang et al. 2015)
	allicin	LPS-induced oxidative stress in human vein endothelial cells (in vitro)	↑ Nrf2, ↓ lipid peroxidation, ↓ LPS-induced apoptosis	(M. Zhang et al. 2017)
	DATS	LPS-induced inflammation in BV-2 microglial cells (in vitro)	↓ NF-ĸβ, ↓ NO, PGE2, iNOS, COX-2	(H.H. Lee et al. 2018)
	DAS	LPS and d-galactosamine-induced acute liver injury in mice	↓ MDA ↓ TNF-α, IL-1β and MCP-1 ↑ GSH, SOD and CAT (liver)	(M. Li et al. 2018)
	DAS	Malathion (insecticide)-induced oxidative stress in cerebral, hepatic, renal tissue of rats	↓ expression of inflammatory cytokines (NF-kB, IL-1B, TNF-α) ↓ ROS via suppression of CYP-2E1 hepatic enzyme	(Abdel-Daim et al. 2020)
	DADS	Liver microsomes from male Wistar rats (in vitro)	↑ hydroxyl scavenging activity ↓ lipid peroxidation	(Chung 2006)
Methiin (SMCSO)	–	Alloxan-induced diabetic Sprague-Dawley rats	↓ lipid peroxidation	(Kumari and Augusti 2002)
	–	Streptozotocin-induced diabetic rats	↓ NF-ĸβ	(Castro et al. 2021)
	–	Streptozotocin-induced diabetic rats	↓ hyperglycemia, TG and VLDL ↑ IL-10, ↑ SOD and CAT	(Lemos et al. 2021)
Isoalliin (deoxy)	–	SHRats and male Wistar Kyoto rats (control) given S-propenylcysteine compared to SAC, (Metabolomic studies)	↓ hypertension	(Matsutomo et al. 2017, 2019)
	–	SHRats and Wistar Kyoto rats (control) given S-propenylcysteine	↓ hypertension	(Ushijima et al. 2018)
	–	Human vascular endothelial cells stimulated with TNF-α, w/wo S-propenylcysteine (in vitro)	↓ TNF-α-induced vascular endothelial hyperpermeability	(Kunimura et al. 2021)

LPS lipopolysaccharide; Nrf2 nuclear factor-erythroid factor 2-related factor; NF-ĸβ nuclear factor kappa-light-chain-enhancer of activated B cells; ROS reactive oxygen species; T2DM type-2 diabetes mellitus; AMPK adenosine monophosphate-activated protein kinase; TNF-α tumor necrosis factors alpha; IL interleukin; iNOS inducible nitric oxide synthase; NO nitric oxide; COX-2 cycoloxygenase-2; TGF-1β transforming growth factor beta 1; CCL5 C-C motif chemokine ligand 5; TLR4 toll-like receptor-4; PGE2 prostaglandin E₂; VCAM-1 vascular adhesion molecule-1; ICAM Intercellular adhesion molecule-1; oxLDL oxidized low-density lipoprotein, CHL cholesterol; HDL high-density lipoprotein; LDL low-density lipoprotein; TG triglycerides; HO-1 heme oxygenase; $\gamma$GCS $\gamma$-glutamylcysteine synthetase; PARP poly(ADP-ribose) polymerase; HOMO-IR Homeostatic model assessment of insulin resistance; BW body weight; LOX-1 lectin-like oxidized low-density lipoprotein receptor-1; p-mTOR phosphorylated-mammalian target of rapamycin; PPAR$\gamma$ peroxisome proliferator-activated receptor gamma; LXRα liver X receptor; MCP-1 monocyte chemoattractant protein-1; Egr-1 early growth response 1; MAPK mitogen-activated protein kinase; MDA malondialdehyde; GSH glutathione; SOD superoxide dismutase; CAT catalase; I3C indole 3-carbinol; DIM 3,3-diindolymethane; MDA-TBA malondialdehyde/thiobarbituric acid; VLDL very-low-density lipoprotein; SAC S-allylcysteine.

Glucoraphanin

Glucoraphanin (4-methylsulfinylbutyl glucosinolate) and its derivative sulforaphane, is arguably the most well-researched organosulfur compound (Yagishita et al. 2019). Sulforaphane is an aliphatic isothiocyanate first identified within broccoli as a major inducer of cytoprotective enzymes in mice in 1992 (Y. Zhang et al. 1992). Sulforaphane has since been identified as having protective antioxidant, immunomodulatory and anti-inflammatory effects, with several reviews reporting its potential for cancer, cardiac, vascular, neurological, and glycemic benefits (Bai et al. 2015; Kim 2021; Dinkova-Kostova and Kostov 2012; Houghton 2019; Ramirez et al. 2017). Various rodent models have identified positive cardiometabolic changes when administered sulforaphane, including hypertension, atherosclerosis, glucose management and for mitigating diabetic complications, largely due to activation of Nrf2 and associated downstream effects (Bai et al. 2015). Kim (2021) presented the multi-faceted neuroprotective effects of sulforaphane in various animal and human studies, its ability to induce phase II detoxification enzymes, activate Nrf2 transcription, reduce oxidative stress; and its subsequent potential in ameliorating the pathophysiological changes associated with cognitive decline, Alzheimer’s disease and other neurological-based diseases (Kim 2021). A potential nephroprotective role has also been reviewed, through sulforaphane’s Nrf2 inducing activity and potential for mitigating the hypertensive impacts of chronic kidney disease (Liebman and Le 2021). Evidence also suggests sulforaphane is a powerful inducer of cytoprotective enzymes involved in phase II detoxification, in particular NAD(P)H:(quinone-acceptor) oxidoreductase 1 (NQO1) and glutathione S-transferases (GSTs), thus making it prolific in anti-cancer research (Russo et al. 2018). Sulforaphane has been shown to upregulate Nrf2 activity, inhibiting lipopolysaccharide (LPS)-induced prostaglandins, COX-2, MMPs and iNOS in mice (Qi et al. 2016). Similarly, increased Nrf2 activity, and subsequently lowered inflammatory cytokines, lipid peroxidation markers and ROS, alongside elevated superoxide dismutase (SOD) has been seen in an in-vitro Alzheimer’s disease mice model (Zhao, Zhang, and Chang 2018). Sulforaphane has elicited cardioprotective effects against inflammatory and oxidative damage, activating Nrf2 activity, mediating glucose signaling and preventing both type-2-diabetic-induced cardiomyopathy (Y. Sun et al. 2020) and nephropathy (Z. Li et al. 2020) in mice. Whilst more research is needed, sulforaphane has improved glycemic control (Axelsson et al. 2017), insulin resistance (Bahadoran, Tohidi, et al. 2012), triglycerides, and lipid peroxidation biomarkers in T2DM (Bahadoran, Mirmiran, et al. 2012). A randomized controlled trial of 137 high-risk CVD participants randomly assigned to receive either a high dose glucoraphanin broccoli (i.e., high sulforaphane) for 12-weeks had significantly lowered low-density-lipoprotein (LDL) compared with those given standard broccoli (Armah et al. 2015). A recent study in 40 healthy, although overweight humans, also reported sulforaphane-rich broccoli sprouts were shown to decrease multiple inflammatory markers; IL-6, C-reactive protein, and IL-1β (López-Chillón et al. 2019). More recently, in vitro evidence demonstrated sulforaphane’s potential in reducing the inflammatory cytokine storm (both IL-6 and IL-8) associated with the SARS-CoV-2 spike protein and associated lung injury (Gasparello et al. 2021). The popularity and breadth of research into glucoraphanin and sulforaphane has led to its commercialization and use in clinical trials (Fahey and Kensler 2021; Houghton 2019) as well as the breeding of new glucoraphanin-rich broccoli cultivars (Bell and Wagstaff 2017).

Glucoerucin

Glucoerucin (4-methylthiobutyl glucosinolate) and its derivative, erucin (4-[methylthio]butyl isothiocyanate) have been heralded as a ‘new promising anticancer agent from cruciferous vegetables’ (Melchini and Traka 2010). This is largely due to glucoerucin’s ability to regulate detoxification, mitigate oxidant stress and ROS-associated pathways, and improve overall cellular defense mechanisms (Melchini and Traka 2010). Due to erucin’s ability to interconvert with sulforaphane in both animals and humans, it shares many features with sulforaphane, particularly around antioxidant potential (Clarke et al. 2011). Erucin has been shown to reduce multiple inflammatory markers (IL-6, IL-1β, prostaglandin E2 and COX-2) following an LPS-induced inflammatory state in murine RAW 264.7 cells (Cho, Lee, and Park 2013). A significant reduction in iNOS production, TNF-α and NF-ĸβ signaling also occurred within the erucin-treated cells (Cho, Lee, and Park 2013). Erucin has been shown to reduce both the pain and neuro-inflammation associated with diabetes-induced neuropathic pain in mice (Lucarini et al. 2019). An extract of erucin displayed antithrombotic and antiplatelet activity in vitro, lowering NF-ĸβ expression along with several inflammatory mediators (including IL-1β and thromboxanes) within human platelet cells, and showed antithrombotic activity when administered to mice (Fuentes et al. 2014). An in vitro study also reported a dampening of LPS-induced neuro-inflammation, with a reduction in pro-inflammatory cytokines, elimination of COX-2 and improved inflammasome expression in cells pretreated with an erucin extract (Gugliandolo et al. 2018).

Sinigrin

Sinigrin (2-propenyl-glucosinolate) is an aliphatic glucosinolate found in broccoli, Brussels sprouts, and mustard seeds (Mazumder, Dwivedi, and du Plessis 2016). Studies suggest sinigrin has anti-cancer, antimicrobial, antioxidant and anti-inflammatory benefits, wound healing properties and, like many glucosinolates, acts as a natural pesticide (Mazumder, Dwivedi, and du Plessis 2016). Upon hydrolysis, its derivative is allyl isothiocyanate (Allyl ITC). Sinigrin has been shown to reduce various atherosclerotic markers in four murine-based studies (H. W. Lee et al. 2017; H. W. Lee and Lee 2015; Jang et al. 2017; Jang and Pyo 2015). ApoE-deficient mice fed a high cholesterol diet were found to have a reduction in VCAM-1 within smooth muscle cells in those treated with sinigrin, bought about by reduced expression of both TNF-α and NF-ĸβ (H. W. Lee and Lee 2015; H. W. Lee et al. 2017). Similar results were found by Jang and Pyo (2015) whereby sinigrin reduced TNF-α and subsequently lowered both VCAM-1 and NF-ĸβ expression. Various protective cardiometabolic markers were evident from administration of sinigrin including lowered cholesterol fractions and triglycerides, as well as reduced oxidized LDL, lowered IL-6, and subsequent reduction in proatherogenic risk (Jang et al. 2017; Jang and Pyo 2015).

Glucobrassicin

Following hydrolysis, glucobrassicin forms the highly unstable indole 3-carbinol (I3C), and its condensation product, 3,3-diindolymethane (DIM). The majority of research into glucobrassicin and its derivatives explore the anti-cancer, proapoptotic potential across several cancer types, including the inhibition of NF-ĸβ and iNOS, modulation of ROS, reduced TNF-α and inflammatory downstream derivates (Maruthanila, Poornima, and Mirunalini 2014). Like many glucosinolates, DIM has been identified to impact detoxification pathways, enhancing conjugation and excretion of potential carcinogens (Williams 2021). This derivative is recognized to enhance both phase I and phase II detoxification pathways, facilitating the modification of potentially harmful metabolites, and their conjugation and excretion from the body (Ampofo et al. 2018). Within phase I, it particularly enhances cytochrome P450 enzymes favoring estrogen hydroxylation, thereby explaining its research as a potential therapeutic agent for estrogen-driven cancers. DIM has been shown to induce multiple transcription targets for Nrf2-induced phase II and antioxidant signaling (e.g., GSTs, NQO1, HO-1, SOD) (Ampofo et al. 2018; Fuentes et al. 2014; Williams 2021). Both I3C and DIM have been shown to increase the expression of Nrf2 and reduce NF-ĸβ activity in animal studies (Hajra et al. 2018; T. Chang, Ho, et al. 2019). Furthermore, I3C has been identified as having potential anti-thrombotic and anti-inflammatory activity, with evidence that it lowers inflammatory cytokines and ROS production, as well as enhances microvascular integrity (Ampofo et al. 2018). Administration of glucobrassican has been shown to improve cardiac output and vascular resistance in hypertensive cirrhotic rats, subsequently lowering high blood pressure (T. Chang, Ho, et al. 2019). I3C was also shown to almost completely blunt ethanol-induced liver inflammation, lipid peroxidation and levels of H₂O₂ in ethanol-fed mice, indicating antioxidant, anti-inflammatory and anti-apoptotic effects following administration (Choi, Abdelmegeed, and Song 2018).

Glucoiberin

Glucoiberin (3-methylsulfinylpropyl glucosinolate) and its naturally produced hydrolysis derivative, iberin [3-(methylsulfinyl)-propyl isothiocyanate) has been reported to promote xenobiotic phase II detoxification (e.g., quinone reductase and GSTs) suggesting a cytoprotective potential in animal studies (Munday and Munday 2004). In combination with other sulfoxide analogues (sulforaphane and erucin), iberin was demonstrated to upregulate thioredoxin reductase 1 expression within human breast cancer MCF-7 cells, in vitro (W. Wang et al. 2005). Furthermore, in very similar potency to sulforaphane, iberin was found to increase Nrf2-translocation and subsequent activity in vitro, inducing detoxification enzyme heme oxygenase 1 (HO-1) and $\gamma$-glutamylcysteine synthetase, thereby enhancing antioxidant defenses (Ernst et al. 2013). Previous studies have also reported chemo-preventive, anti-cancer, and apoptotic potential in several in vitro models (Gong et al. 2021; Pocasap, Weerapreeyakul, and Thumanu 2019; Jadhav et al. 2007).

Gluconasturtiin

Gluconasturtiin is another commonly consumed aromatic glucosinolate and is the precursor to phenethyl isothiocyanate (PEITC) (Dayalan Naidu et al. 2018). Previous authors have reviewed the anticarcinogenic potential of PEITC (Gupta et al. 2014; Ramirez et al. 2017). Dayalan et al has reviewed PEITC’s ability to enhance transcription of Nrf2 and heat-shock proteins, and inhibit phase I detoxification enzymes, thus improving inflammation and oxidant support within in vitro models (Dayalan Naidu et al. 2018). It has also been found that PEITC, in combination with benzyl isothiocyanate (BITC) from gluconapin, over 18-weeks significantly reduced obesity and cholesterol in C57BL/6 mice induced with cardiometabolic dysfunction via a high-fat diet (Chuang et al. 2019). Mice administered both PEITC and BITC had reduced expression of lipogenic and adipogenic transcription proteins when compared to control mice (Chuang et al. 2019); proteins involved in inhibiting inflammation and pro-inflammatory cytokines (Bilotta et al. 2020; Martin 2010). Another study involving high-fat fed C57BL/6 mice found oral PEITC over 13 weeks resulted in decreased NF-ĸβ expression, along with a significantly blunted rise in cholesterol (lowered atherogenic index), weight gain and improved glucose control. Furthermore, the PEITC-fed group had significantly less narrowing of blood vessels and hepatic adipocyte accumulation than the control group (Gwon et al. 2020). A later study reported that these improved cardiometabolic parameters were likely a result of PEITCs ability to influence the regulation of mTOR/AMPK/PPAR$\gamma$ expression involved in lipid, energy, and glucose metabolism and in lowering inflammation (Gwon and Yun 2021). A recent in vitro study using human THP-1 derived macrophages induced with LPS and oxidized LDL, also reported a reduction in vascular inflammation, formation of foam cells and their accumulation of lipids and overall atherosclerotic risk when administered PEITC. The researchers also reported the PEITC-treated cells had blunted activation of NF-ĸβ and LOX-2 among others, indicating its potential in lowering inflammatory drivers (Im, Gwon, and Yun 2021).

Other glucosinolates

Glucotropaeolin forms into BITC, which, in addition to mitigating anti-inflammatory cytokines in combination with PEITC described above (Chuang et al. 2019), has also been shown to have anticarcinogenic potential across several animal and in vitro studies (Dinh et al. 2021). Gluconapin (3-butenylglucosinolate) is known to produce the metabolite, 3-butenyl isothiocyanate. Mice fed corn-oil and given a bolus dose of gluconapin had a significant blunting in post-prandial hypertriglyceridemia, when compared to control (Washida et al. 2010). Furthermore, 3-butenyl isothiocyanate has been shown to have cytotoxic potential across multiple human cancer cell lines in vitro (Arora et al. 2016). Collectively, many of the glucosinolate post-hydrolysis derivatives act as potential sulforaphane analogues, yet research exploring many of these and their metabolites in isolation is sparse.

S-alk(en)yl cysteine sulfoxides (ASCOs)

The overall collective pharmacological, anti-inflammatory, antioxidant, anti-carcinogenic and cardiometabolic potential of ASCO-rich allium vegetables upon health have been reported in recent reviews (Shang et al. 2019; El-Saber Batiha et al. 2020; Farhat 2021; Ansary et al. 2020; Galavi, Hosseinzadeh, and Marjan Razavi 2021). As already discussed, like the glucosinolates, the ASCOs are stored in separate vacuoles within the plant tissues, only to be released upon exposure to cysteine sulfoxide lyases bought about by tissue destruction (i.e., slicing, chewing) and by the actions of the gut microbiota following ingestion (Rektorisova et al. 2020). Some of the benefits mediated by the main three ASCOs individually, are described below.

Alliin (S-allylcysteine sulfoxide)

Alliin is the major sulfur-containing cysteine sulfoxide found within garlic. Its formation into allicin (diallyl thiosulfinate) is responsible for the strong aroma associated with garlic and is the most researched ASCO investigating garlic’s potential benefits. Research identifies allicin, and its predecessor alliin, as having antioxidant and anti-inflammatory activity, with cardiovascular, hypolipidemic, hypoglycemic and neuroprotective potential (Quintero-Fabián et al. 2013; Chan et al. 2013; Zhai et al. 2018). Alliin has been shown to reduce several markers of inflammation (iNOS expression, IL-6, TNF-α and NF-ĸβ) in both in vivo (mice) and in vitro studies (Shi et al. 2017). It has also been shown to decrease IL-6, improve insulin sensitivity and glycemic control in mice (Sánchez-Sánchez et al. 2020). S. Wang et al. (2015) reported allicin mitigated the oxidant-mediated damage and cellular dysfunction associated with diabetic-like conditions in vitro. Within this study, murine aortic endothelial cells exposed to allicin had significantly lowered ROS, 8-hydroxydeoxyguanosine, NF-ĸβ and hypoxia-inducible factor-1α (HIF-1α), thereby potentially offering protection against microvascular and atherosclerotic dysfunction (S.L. Wang et al. 2015). Research suggests that allicin increases superoxide scavenging activity, with its derivatives increasing hydroxyl scavenging and ameliorating lipid peroxidation in vitro (Chung 2006). Administration of allicin was also found to decrease the expression of Nrf2 in human endothelial cells exposed to LPS-induced oxidative stress, as evidenced by lowered ROS production, lipid peroxidation and reduced inflammatory response in blood vessels (M. Zhang et al. 2017). Allicin has many downstream metabolites that have been studied in isolation including dithiins, ajoenes, and thiosulfur lipid allyl sulfides such as diallyl sulfide (DAS), diallyl disulfide (DADS) and diallyl trisulfide (DATS) (El-Saber Batiha et al. 2020). Several in vitro and animal-based in vivo studies have explored the effects of DAS, DADS and DATS derivatives in reducing expression of inflammatory cytokines, regulating carcinogenic activity, GSH production, improving detoxification and overall antioxidant mechanisms (Abdel-Daim et al. 2020; Puccinelli and Stan 2017; M. Li et al. 2018; Suman and Shukla 2016; H.H. Lee et al. 2018). Administration of DAS was found to lower expression of NF-ĸβ, IL-1-β, and TNF-α and ameliorate reduction in GSH, SOD and other markers of ROS within cerebral, hepatic and renal tissues of rats when exposed to malathion-induced oxidative stress (Abdel-Daim et al. 2020). DAS was also found to blunt the oxidative damage of LPS/d-galactosamine-induced liver injury in rats in a dose dependent manner, with lowered malondialdehyde, TNF-α, IL-1-β, and MCP-1 levels evident in both blood and liver (M. Li et al. 2018). A previous review has reported on the antioxidant and anti-inflammatory activity of DAS, including its preventive role across several chronic metabolic diseases (Suman and Shukla 2016). Within an in vitro model, LPS-stimulated microglial cells exposed to DATS had reduced NF-ĸβ expression along with lowered NO and inflammatory chemokines (H.H. Lee et al. 2018).

Methiin (S-methyl cysteine sulfoxide, aka SMCSO)

Another ASCO found in relative abundance within allium vegetables is methiin, most often referred to as S-methyl cysteine sulfoxide (SMCSO). Interestingly, SMCSO is found in greater quantities in cruciferous vegetables (1–4% dry weight) than the glucosinolates (0.1–0.6%) (Mae, Ohira, and Fujiwara 1971; Marks et al. 1992). Yet due to its link to causing hemolytic anemia in cattle fed large quantities of SMCSO-rich cruciferous vegetables (Smith, Earl, and Matheson 1974), this metabolite has received comparatively little attention until recently. Giving glucoraphanin-rich broccoli to men prior to prostate biopsies, found SMCSO accumulates within human prostate tissue and takes considerably longer than glucosinolates to be excreted (Coode-Bate et al. 2019). Further research is required to determine if this accumulation has any significance, or indeed biological activity. As such, less is understood on the complexity of SMCSO derivatives, although it is evident that cysteine sulfoxide lyases, within the plant tissue and human microbiota, lead to the release of ammonia, pyruvate and the highly-reactive yet transient metabolite, methanesulfenic acid (Edmands et al. 2013). Several metabolites are then produced, most predominantly being S-methyl methanethiosulfinate (MMTSI, aka, dimethyl disulfide sulfoxide) and S-methyl methanethiosulfonate (MMTSO, aka, dimethyl disulfide sulfone) (Edmands et al. 2013). Both MMTSI and MMTSO appear to have protective roles in plant tissue, particularly against pathogens (Edmands et al. 2013). Several in vitro and in vivo animal (rat and mice) studies also indicate anti-carcinogenic activity when administered allium-derived thiosulfinates, mostly as MMTSI and MMTSO (Edmands et al. 2013). Nonetheless, whether these derivatives are produced, or have similar effects within humans is still lacking. More broadly, SMCSO has been shown to inhibit both glucose-6-phosphatase and HMG-CoA reductase, thereby lowering both hyperglycemia and hypercholesterolemia in diabetes-induced rats (Kumari, Mathew, and Augusti 1995). Furthermore, the amelioration of blood glucose when treated with SMCSO was not only dose-dependent, but also comparable to the effects of insulin (Kumari, Mathew, and Augusti 1995). SMCSO was found to improve the antioxidant potential in both hepatic and cardiac tissues of diabetic rats, evidenced by reduced lipid peroxidation markers; malondialdehyde, hydroperoxides and conjugated dienes (Kumari and Augusti 2002). More recently, streptozotocin-induced diabetic rats fed SMCSO were found to have reduced NF-ĸβ levels within their duodenal tissue when compared to control (Castro et al. 2021). Lemos et al. (2021) reported a higher IL-10, increased hepatic SOD and catalase activity in streptozotocin-induced diabetic rats, when administered SMCSO, providing additional evidence for conceivable antioxidant benefits. Furthermore, the SMCSO-treated rats also had reduced hyperglycemia, triglycerides, and VLDL-cholesterol over their control counterparts (Lemos et al. 2021). Whilst further evidence is needed, it does suggest a potential regulatory role of this metabolite in inflammatory and/or antioxidant pathways. To date, SMCSO has not been studied in isolation in human trials.

Isoalliin (S-propenyl cysteine sulfoxide, aka PeSCO)

Isoalliin is only found in small quantities in young garlic although increases considerably during the aging process (i.e., aged garlic) (Kodera et al. 2017); by as much as 6-fold after 8-weeks (Horníčková et al. 2010) and 54-fold after 6-months of storage (Rektorisova et al. 2020). The action of alliin-lyase acting upon isoalliin when cutting isoalliin-rich onions, triggers the release of 1-propenyl thiosulfinates and thiolpropanal-S-oxide (aka lachrymatory factor), the compound responsible for production of tears when chopping onions (Rose et al. 2019). However, new sulfur-based isoalliin derivatives, including thiolanes, have only been identified recently which, along with the previously known cepaenes, zwibelanes and bis(sulfine) derivatives, may open new investigations in coming years (Štefanová et al. 2019). Like methiin above, there is also limited information on the biological activity of this ASCO for health, when studied alone. In fact, only one study was found, which explored the physiological effects of isoalliin for its potential use in cachexic conditions (L. Wang et al. 2021). Based on the premise that onion has a historical use of stimulating dietary intake, and with isoalliin being the main ASCO; this study found that administering a bolus dose of isoalliin to C57BL/6 mice enhanced their dose-dependent responsiveness to the neuropeptide hormone ghrelin, stimulating appetite and subsequent intake over 24-hours (L. Wang et al. 2021). Despite this scarcity of research, its precursor deoxy form (S-propenylcysteine) has been shown to lower blood pressure in hypertensive rats, both over time, or as a single bolus and dose-dependently (Matsutomo et al. 2017; Matsutomo et al. 2019; Ushijima et al. 2018). Matsutomo et al (2017) measured metabolomic output in hypertensive rats following the administration of either S-propenylcysteine or S-allylcysteine (SAC), compared to control. Whilst the SAC-administered group had no significant change, an anti-hypertensive effect was observed in those given S-propenylcysteine. Furthermore, the S-propenylcysteine-group had significant alterations in synthesis of glycerophospholipids, fatty acids, and normalization of metabolites with antihypertensive, osmolytic (e.g., betaine) and NO-inducing properties (e.g., tryptophan) (Matsutomo et al. 2017). A subsequent investigation found that treatment with S-propenylcysteine produced not only an antihypertensive effect but coincided with a significant increase in plasma histidine levels (Matsutomo et al. 2019). Interestingly, histidine is known for its role in suppressing several markers of inflammation (i.e., cytokines, interleukins, TNF-α, ICAM-1 expression) (Hirasawa 2019). A subsequent in vitro study on human vascular endothelial cells has shown this deoxygenated form of isoalliin reduces the TNF-α-induced hyperpermeability of the vascular endothelial barrier; of key importance in maintaining vascular homeostasis and preventing vascular disease (Kunimura et al. 2021).

Hydrogen sulfide activity

Isothiocyanate-rich cruciferous vegetables are also known to act as H₂S donors, eliciting additional cardioprotective benefits (Martelli, Piragine, et al. 2020). More specifically, erucin has been shown to produce vasodilation in rat hearts, inhibit in vitro noradrenaline-induced vasoconstriction and thereby elicit an anti-hypertensive effect, owing to this initiation of H₂S (Citi et al. 2019; Martelli, Piragine, et al. 2020). Other cruciferous-derived isothiocyanates have shown to be effective H₂S donors and include benzyl isothiocyanate (derived from glucotropaeolin), allyl isothiocyanate (from sinigrin), sulforaphane (from glucoraphanin) and 4-hydroxybenzyl isothiocyanate (from sinalbin) (Citi et al. 2014).

Hydrogen sulfide (H₂S) is identified as the third most important gaseous signaling molecule in the human body after NO and carbon monoxide (CO), and is involved in regulating several pathophysiological pathways (Corvino et al. 2021). These include, but are not limited to, enhanced NO bioavailability (vasodilation), activation of Nrf2 (enhanced antioxidant and anti-inflammatory pathways) and reduced NF-ĸβ expression, thereby offering significant anti-inflammatory and antioxidant defense (Scammahorn et al. 2021). As described by Francioso et al. (2020, 3), H₂S plays a “crucial role in cellular redox homeostasis by modulating GSH concentration and Nrf2 transcription.” The release of H₂S from isothiocyanates is achieved via non-enzymic adducts formed with thiol groups (e.g., cysteine or GSH) and/or the interaction with H₂S-generating enzymes (Corvino et al. 2021; Hsu and Tain 2021; Magli et al. 2021). There are four tissue-specific H₂S-generating enzymes including cystathionine $\gamma$-lyase (CSE), cystathionine β-synthase (CBS), 3-mercaptopyruvate sulfurtransferase (3-MST) and cysteine amino transferase (CAT) (Citi et al. 2014; Corvino et al. 2021; Martelli, Citi, et al. 2020). CSE is most concentrated within the cardiovascular tissue, CBS within the central nervous system, hepatic, kidney, uterine and placental tissue, CAT and 3-MST within the vascular endothelium and brain, with 3-MST also concentrated within mitochondria; each generating the release, and thus benefits, of H₂S into their respective surrounding tissues (Corvino et al. 2021; Magli et al. 2021). We refer the reader to several excellent reviews into the emerging potential of H₂S in cardiometabolic protection and other diseases associated with oxidative stress (Corvino et al. 2021; Corsello, Komaravelli, and Casola 2018; D. Wu, Hu, and Zhu 2018; Kuschman, Palczewski, and Thomas 2021; Hsu and Tain 2021). Many, if not all, of the sulfur-containing secondary metabolites are potential sources of H₂S production. It is imperative that future studies on sulfur-containing secondary metabolites explore this concept further, ensuring efforts are made to differentiate the impacts derived from individual metabolites as much as possible.

Conclusions and future directions

There are several hundred sulfur-containing secondary metabolites found within the foods we regularly consume, especially abundant within cruciferous and allium vegetables. This review has provided an overview on sulfur uptake from plants, the defensive role within plants, their conversion into the main sulfur-containing secondary metabolites and primary dietary sources, being cruciferous and allium vegetables. We have also discussed the potential biological importance of these metabolites for long-term chronic disease prevention and potential cardiometabolic risk reduction. This review has broadly focused on their overall anti-inflammatory and antioxidant activity, with chronic disease prevention in mind. We recognize there is strong evidence to suggest that these metabolites also have antimicrobial and anticarcinogenic activity, although deemed beyond the scope of this review. The reader is therefore directed to further readings (Bhatwalkar et al. 2021; Barbieri et al. 2017; Krause et al. 2021; Melrose 2019; De Greef et al. 2021; Favela-González, Hernández-Almanza, and De la Fuente-Salcido 2020; Soundararajan and Kim 2018; Miękus et al. 2020). We also recognize that, due to the vast quantity of sulfur-containing secondary metabolites found within cruciferous and allium vegetables, only those for which the most research is available, have been mentioned.

Of course, humans eat wholefoods, and often the physiological effects of differing phytochemicals work in synergy, rather than isolation. The same may be evident for several of these sulfur-containing secondary metabolites. Whilst some have been isolated and researched extensively for their benefits (e.g., allicin and sulforaphane) and become commercialized into nutraceuticals, most have not been researched, or research is limited to a relatively small number of in vitro and/or in vivo animal studies. Furthermore, with the development of new vegetable cultivars (e.g., glucoraphanin/sulforaphane-rich Beneforte broccoli) (Bell and Wagstaff 2017), scientists and plant breeders need to ensure that this is not done at the expense of other equally important, but less studied sulfur-containing secondary metabolites.

It is important to remind the reader that whilst around 130 glucosinolates have been identified to date; an extensive list provided by Nguyen and colleagues (Nguyen et al. 2020); not all of these, have yet been explored for their contribution to the biological activity of these vegetables. For example, gluconapin (e.g., in Chinese cabbage) is commonly consumed (Mithen and Ho 2018), yet little research beyond its quantity in such vegetables is currently available. Similarly, little evidence was found on the biological effects of glucobrassicanapin, despite it being the predominant glucosinolate in Chinese cabbage and Pak choi (Nugrahedi et al. 2017). This same gap in the literature was evident for both the ASCOs, isoalliin and methiin. Methiin is particularly interesting as it contributes significantly more sulfur to cruciferous vegetables than do the glucosinolates and is also ubiquitously found across several alliums. Considering this, the lack of research into its biological potential, particularly in humans is surprising (Edmands et al. 2013). Therefore, more research and understanding of the individual influences of these metabolites are needed in future studies.

Extracting individual sulfur-containing metabolites and exploring their effects in human trials with certainty is enormously difficult, compounded by fluctuating quantities within different vegetables in response to defensive mechanisms during growth, environmental and storage conditions. This may largely account for the scarcity of data from human intervention studies regarding the physiological effects of consumption of these compounds. This undoubtedly makes translation into ‘real world’ eating and dietary intervention studies a challenge (Palliyaguru et al. 2018). Future studies involving sulfur-rich vegetables as a whole-food intervention must, as a minimum, ensure they measure and report the chemical analysis of their vegetable intake, to infer true meaning to their clinical results, particularly for determining which metabolites might be responsible. Fortunately, whilst the complete understanding of mammalian sulfur metabolism is far from being fully understood, the emergence of new measurement techniques, e.g., various-omics platforms and detection techniques (Yagishita et al. 2019), make future research into this area an exciting reality.

In conclusion, Clifford et al. (2021) argues that there is limited evidence that phytochemicals, such as sulforaphane or indeed, any other dietary metabolites, can modulate Nrf2 or other key pathways within humans. Whilst several recent reviews have outlined the promising benefits of glucosinolates, (particularly glucoraphanin, i.e., sulforaphane) upon human health (Houghton 2019; Marino et al. 2021; Yagishita et al. 2019), these reviews have also recognized significant gaps to be addressed moving forward. It is recognized that considerable heterogeneity across studies exists, and that more understanding is required to fully understand the underlying mechanisms of these metabolites and their pharmacodynamic effects upon various biomarkers in humans (Marino et al. 2021; Yagishita et al. 2019). Furthermore, research exploring the effects of the many other sulfur-based metabolites in commonly consumed foods (i.e., cruciferous and alliums) and their effects within human clinical trials are lacking. Future studies are needed to further explore sulfur-containing metabolites identified in plants both individually and collectively within whole-foods, to identify dose dependent responses upon various pathways for human health. Only then, can we formulate dietary recommendations for optimal intake of cruciferous and allium vegetables as a preventive adjunct against the increased prevalence of cardiometabolic and other chronic diseases.

Authors’ contribution statement

The authors’ responsibilities were as follows: CRH and LCB: were responsible for the concept of this paper; CRH: Writing and original drafting of manuscript; AS, LB, JRL, JMH, AHM, and LCB: provided a critical review and editing of the manuscript. All authors have read and agreed to the final published version of the manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This research received no additional nor specific funding. CRH is supported by an Edith Cowan University Research Excellence Fellowship Scholarship; AS is supported by an Edith Cowan University Strategic Research Fellowship; The salary of LB is supported by the School of Medical and Health Science at Edith Cowan University; JRL is supported by a National Heart Foundation Future Leadership Fellowship (ID: 102817); The salary of JMH is supported by a National Health and Medical Research Council of Australia Senior Research Fellowship, Australia (Grant number APP1116973); AHM is supported through the Australian Research Council (ARC) Center of Excellence in Plant Energy Biology and as an ARC Australian Laureate Fellow (CE140100008, FL200100057); LCB is supported by a National Health and Medical Research Council of Australia Emerging Leadership Investigator Grant (ID: 1172987) and a National Heart Foundation of Australia Post-Doctoral Research Fellowship (ID: 102498). Publication fees were supported by the Royal Perth Hospital Research Foundation (RPH RF Research Excellence Award, awarded to LCB 2020).