Chia (Salvia hispanica L.), a functional ‘superfood’: new insights into its botanical, genetic and nutraceutical characteristics

Abstract

Background

Chia (Salvia hispanica L.) seeds have become increasingly popular among health-conscious consumers owing to their high content of ω-3 fatty acids, which provide various health benefits. Comprehensive chemical analyses of the fatty acids and proteins in chia seeds have been conducted, revealing their functional properties. Recent studies have confirmed the high ω-3 content of chia seed oil and have hinted at additional functional characteristics.

Scope

This review article aims to provide an overview of the botanical, morphological and biochemical features of chia plants, seeds and seed mucilage. Additionally, we discuss the recent developments in genetic and molecular research on chia, including the latest transcriptomic and functional studies that examine the genes responsible for chia fatty acid biosynthesis. In recent years, research on chia seeds has shifted its focus from studying the physicochemical characteristics and chemical composition of seeds to understanding the metabolic pathways and molecular mechanisms that contribute to their nutritional benefits. This has led to a growing interest in various pharmaceutical, nutraceutical and agricultural applications of chia. In this context, we discuss the latest research on chia and the questions that remain unanswered, and we identify areas that require further exploration.

Conclusions

Nutraceutical compounds associated with significant health benefits, including ω-3 polyunsaturated fatty acids, proteins and phenolic compounds with antioxidant activity, have been measured in high quantities in chia seeds. However, comprehensive investigations through both in vitro experiments and in vivo animal and controlled human trials are expected to provide greater clarity on the medicinal, antimicrobial and antifungal effects of chia seeds. The recently published genome of chia and gene-editing technologies, such as CRISPR, facilitate functional studies deciphering molecular mechanisms of biosynthesis and metabolic pathways in this crop. This necessitates development of stable transformation protocols and creation of a publicly available lipid database, mutant collection and large-scale transcriptomic datasets for chia.

INTRODUCTION

Chia (Salvia hispanica L.) seeds are among the most promising gluten-free foods for human health owing to their perceived beneficial properties. Recently, the European Food Safety Authority (EFSA) Panel on Nutrition, Novel Foods, and Food Allergens (NDA) has assessed and approved the incorporation of chia seeds and defatted powder as a safe food ingredient into different food products, such as multigrain bread, cereals, cookies, pasta, nutritional bars and crackers, owing to their high content of ω-3 polyunsaturated fatty acids (PUFAs) (Turck et al., 2023). The addition of chia seeds as a ‘superfood’ supplement to the daily human diet has been proposed to prevent and improve degenerative diseases such as type 2 diabetes, hypertension, dyslipidaemia, liver inflammation and cardiovascular disorders (Fernández‐Martínez et al., 2019; Martino et al., 2020; Oliva et al., 2021; Joubert et al., 2022; Juangco et al., 2022). Moreover, chia seeds are thought to have memory-boosting and antidepressant properties owing to their high content of ω-3 fatty acids (FAs), particularly α-linolenic acid (ALA 18:3n-3), an ω-3 PUFA known to enhance cognitive function and brain abilities (Onneken, 2018; Schreyer et al., 2020; El-Feky et al., 2022).

Being one of the richest botanical sources of ω-3 FAs, chia seeds are considered a sustainable alternative to current sources of ω-3, such as fish oil, for inclusion in human and animal diets (Santos et al., 2020; Ofori‐Mensah et al., 2022). The increasing demand for long-chain ω-3 PUFAs, such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) found in fish oils, is causing a strain on fisheries and leading to a decline in fish populations and species diversity. As a result, it is crucial to explore alternative and sustainable sources of ω-3 fatty acids, such as chia seeds, to use as a nutritional supplement. Incorporating plant-based sources of ω-3s, such as chia seeds, could potentially reduce the negative impacts of fishing on the environment (Lenihan-Geels et al., 2013; Santos et al., 2020). Chia seeds are an excellent source of key nutrients, including ω-3 fatty acids, antioxidants, dietary fibre, carbohydrates and proteins, all of which are essential components of a healthy diet (Timilsena et al., 2017; Chim-Chi et al., 2018; Alcântara et al., 2019; Mihafu et al., 2020; Mohamed et al., 2020; Quintal‐Bojórquez et al., 2021; Aljumayi et al., 2022).

Chia seeds were a significant food source for the Aztec people in pre-Columbian times, and its flour (Chianpinolli) has been used in beverages and tortillas since then. Archaeological evidence reveals that chia seed oil was used for artwork, varnishes and cosmetics (Cahill, 2003). The traditions of the Aztecs were transferred into the 18th and 19th centuries when chia seeds were used in lemonades to make a popular drink known as ‘chia fresca’. All parts of the chia plant have been used in traditional medicines; in Mexico and Central America, chia seeds, for example, were used as an ingredient for medicinal infusions to treat skin injuries and gastrointestinal and respiratory disorders (Cahill, 2003).

BOTANICAL AND MORPHOLOGICAL CHARACTERISTICS OF THE CHIA PLANT, SEEDS AND SEED MUCILAGE

Chia (Salvia hispanica L.) is an annual herbaceous plant belonging to the mint family (Lamiaceae), which originated from Central and Southern Mexico and Guatemala (Fig. 1; Ayerza and Coates, 2005). Chia reproduces naturally by self-pollination and is highly photosensitive, i.e. it requires short-day conditions to initiate flowering (Bochicchio et al., 2015). Chia is known as a heliophilous plant; as such, it is attracted to sunlight and can tolerate high levels of direct sun; however, it can also adapt to partly shaded environments. Chia can also adapt to barren land and tolerate moderate drought (Baginsky et al., 2016; de Paiva et al., 2016; Lovelli et al., 2019). Chia is grown commercially in various countries, including the USA, Argentina, Bolivia, Ecuador, Peru, Brazil, Colombia, Guatemala, Mexico, Nicaragua, Chile, Italy, Australia, Germany and Kenya (de Mello et al., 2015; Ayerza, 2019; Grimes et al., 2019; Ishak et al., 2020; Cabrera-Santos et al., 2021; Costa et al., 2021; de Falco et al., 2021; Koefender et al., 2021; Hassani et al., 2022; Njoka et al., 2022; Harisha et al., 2023).

Fig. 1.

Chia plant (Salvia hispanica L.). (A) Radicle (primary root) emerging from the seed (3–4 days after seed imbibition is completed). (B) Cotyledons unfold completely (hypocotyl) (3 days after germination). (C) seedling (two leaves unfolded). (D) young chia seedling (four leaves unfolded). (E) Chia plant at mid-vegetative stage. (F) Emergence of verticillaster. (G) Inflorescences with purple flowers. (H) Mature chia plant.

The phenological life cycle of chia was recently defined using the Biologische Bundesanstalt, Bundessortenamt and Chemical (BBCH) system (Brandán et al., 2019). Different studies have reported a germination rate of 5–7 days for chia seeds, with initial signs of germination observed within 2 days of sowing when temperatures are optimal (16–26 °C) (Fig. 1A, B; de Paiva et al., 2016, 2018; Nadtochii et al., 2019; Sorana et al., 2019; Cabrera-Santos et al., 2021). Typically, chia plants grow to a height of 1.5–2.3 m over a period of 120–150 days, producing several branches with an average of 15 subsections (Fig. 1H; Grimes et al., 2018; Win et al., 2018; Brandán et al., 2019). A chia plant has green, oval-shaped leaves with serrated edges paired on a node at opposite sides of the stem (Fig. 1C–E). The leaves have long petioles, arc-serrated edges and an acuminate apex. Chia verticillasters are composed of axillary (Fig. 1F) and terminal racemose inflorescences (80–140 florets each) with oval green bracts (Figs 1G and 2A). The bell-shaped calyxes of chia flowers protect four ovaries, each containing a single seed (Fig. 2A; Win et al., 2018). At a later stage, the calyx will slowly dry out, and seeds will become mature until the plant completes senescence (Fig. 2B). Chia plants produce many small, oval-shaped, flattened seeds of various colours, including black, white and beige, with dark spots, with the black and white coloured genotypes being the most common commercially produced ones (Fig. 2C, D; Peláez et al., 2019).

Fig. 2.

Chia inflorescence and seeds. (A) Chia inflorescences consist of several (~80–140) hermaphrodite florets. (B) Floral dehiscence; at this stage, the calyx will slowly dry out, and seeds will become mature until the plant completes senescence. (C, D) Mature black and white chia seeds with an average size ranging from 1.87 to 2.15 mm in length, 1.21 to 1.40 mm in width and 0.77 to 0.88 mm in thickness.

These seeds consist of three layers, including an epidermal layer (exotestal, which classifies it as a nutlet) that is divided into two internal layers, a subepidermal layer and the core endosperm, which composes ~80 % of the seed (Fig. 3A, B; Cowley et al., 2020; Ang et al., 2023). Chia produces myxospermous seeds, which release a hydrophilic mucilage when they come into contact with water. Upon hydration, small filaments develop, and several volcano-shaped structures expand on the seed surface and form a transparent capsule around the seed (Fig. 3C, D; Ang et al., 2023). The chia seed mucilage is firmly attached to the seed. It comprises a branched inner layer (~10 µm thick) situated close to the volcano-shaped structures and a cloudy and homogeneous outer layer visible to the naked eye (Ang et al., 2023). The seed mucilage is an irregular network of elongated microfibrils with small pores, forming a sponge-like structure that extrudes from the epidermal cells of the seeds (Cowley et al., 2020; Ang et al., 2023).

Fig. 3.

The structure of a chia seed and its seed mucilage. (A) A schematic diagram of the chia seed structure (co, cotyledon; emb, embryo; endo, endosperm; ex, exocarp; m, mucilage; radi, radicle). (B) The pericarp layers of a dry seed. (C) A chia seed with mucilage. (D) A hydrated chia seed coat. For images B and D, chia nutlets (seeds plus pericarp), both dry and hydrated, were embedded in resin for sectioning. After embedding in resin, 0.5-µm-thick sections were cut using an Ultramicrotome R (Leica Microsystems) and stained with Toluidine Blue. Images were obtained using a Leica DM6000 optical microscope equipped with a Leica DMC4500 camera. The exocarp (ex), mesocarp (mc), multiple layers of sclereid cells (sc) and endocarp (ec) can be observed in the pericarp of the dry chia seed. The endosperm (endo) is adjacent to the pericarp. In the hydrated seed, lightly stained mucilage (m) has been extruded from the pericarp cells. The exocarp has been separated from the mesocarp with non-mucilaginous cells (nc). The mesocarp forms a layer beneath the mucilage, and a layer of highly refractive cells (x) sits between the mesocarp layer and the sclereid cells.

INFLUENCE OF ENVIRONMENTAL FACTORS AND OTHER VARIABLES ON GROWTH AND YIELD OF CHIA

Oil seed crops require a balanced resource allocation for optimal growth and development. However, environmental factors can impact crop yield and nutritional composition (Grimes et al., 2018; Ebrahimian et al., 2019; Solomou and Sfougaris, 2021; Brandán et al., 2022). Numerous studies have demonstrated that different environmental and agronomic conditions, such as light, temperature, photoperiod, abiotic stresses and fertilization regimes, can significantly affect the germination, growth and yield of chia seeds. Here, we discuss these studies and delve into the environmental and agronomic factors that impact chia growth.

Photoperiod, light and temperature

Plants have the ability to respond to photoperiodism, which means that they require a specific daylength to complete the vegetative stage and transition into the reproductive phase of their developmental cycle. The photoperiod is particularly important for chia plants because they are highly photosensitive, and it plays a crucial role in their growth and yield (de Paiva et al., 2016; Grimes et al., 2018; Rodríguez-Abello et al., 2018). After thousands of years post-domestication, chia is still mainly cultivated in Central America owing to the lack of adaptation to photoperiods required for flower induction, making it difficult to cultivate in other parts of the world. For instance, in North America the flowering is induced very late, which prevents the seeds from ripening before winter starts. Jamboonsri et al. (2012) established a germplasm bank for chia with genetic diversity in photoperiodic floral induction using chemical mutagenesis. This pioneering work has now made it possible to grow chia successfully in areas with days lasting ≤15 h (Jamboonsri et al., 2012).

An investigation was conducted to study the impact of different photoperiods (ranging from 11 to 15 h) on the growth and yield of Brazilian chia genotypes during different seasons. The results showed that the chia planted in early spring (during short days) exhibited the maximum growth, whereas the chia planted in early summer (during long days) produced the highest grain yield (Goergen et al., 2019). However, for chia cultivated in south-eastern Mexico, planting in late winter resulted in greater plant height, increased number of leaves and inflorescences, longer mean flowering time and greater seed yield in comparison to planting in early spring (Rodríguez-Abello et al., 2018). The study also suggested that a decrease in daylength reduced the growth rate during the vegetative stage, causing a delay in the beginning of the reproductive phase (Rodríguez-Abello et al., 2018). However, the study did not take into account the genetic variations within the cultivars. There is currently limited literature on the optimal photoperiod for chia growth owing to specific geographical locations, light regimes and genotypes that impact flowering time.

The effect of the photothermal environment on the duration of vegetative stages was determined for two Argentinian chia genotypes, including white and mixed-colour seed varieties (Brandán et al., 2020). It was observed that shorter daylengths resulted in shorter preflowering sub-stages, which is in line with previous research by Rodriguez-Abello et al. (2018). Interestingly, the verticillaster dry weight at the flowering stage was significantly increased in response to longer sub-stage periods, leading to an increased seed yield in this study (Brandán et al., 2020).

In contrast, seed germination appears unaffected by light conditions (constant light, darkness or alternating dark–light) or temperature (20–30 °C) (de Paiva et al., 2016), and light exposure does not significantly affect the germination rate of chia seeds (Nadtochii et al., 2019). However, the study by Nadtochii et al. (2019) revealed that the optimal germination temperature for chia seeds was 25 °C (determined by seed germination energy, the germination of seeds, the speed of germination and the seedling vigour), with lower (20 °C) and higher (30 °C) temperatures hindering seed growth. To determine the effects of different temperatures (15–35 °C), light regimes and germination media (paper, sand and paper roll) on the physiological potential of chia seeds, Sorana et al. (2019) suggested a standard germination test. This test revealed that germination on paper at 25 °C was ideal, which is consistent with the findings by Nadtochii et al. (2019).

Together, these studies have demonstrated that the growth and development of chia plants are influenced by environmental factors and that temperature variations outside the normal range negatively impact the growth of chia. Given the importance of producing nutrient-rich foods, such as chia, these findings call for further investigation to improve plant productivity.

Abiotic stresses

Recently, the impact of abiotic stresses on the growth and physiology of chia has been investigated. Lovelli et al. (2019) investigated the impact of irrigation on the water conductance, CO₂ concentrations, transpiration and photosynthesis of two chia genotypes in a controlled glasshouse. The genotypes included a black-seeded, short-day-flowering genotype from Peru and a long-day-flowering mutant (G8). The drought conditions caused a reduction in the photosynthetic rate, transpiration rate, water-use efficiency, intercellular CO₂ content, dry mass and leaf area of both genotypes. Compared with the G8 mutant, the short-day-flowering genotype demonstrated a lower photosynthetic rate and intercellular CO₂ concentration but showed increased water-use efficiency, biomass and leaf water potential. This variation is attributable to the earlier induction of flowering in the G8 genotype, which leads to smaller leaves and reduced vegetative growth (Lovelli et al., 2019). This suggests a link between photosensitivity and response to abiotic stresses.

Salinity stress was found to have negative effects on the early stages of growth and development in germinating chia seeds and seedlings (Izadi et al., 2022). Moreover, irrigation with saline water negatively affected chia plant growth, flowering rate and seed yield (Moghith et al., 2020). Applying silicon in form of potassium silicate (K₂SiO₃) through foliar spraying might improve vegetative tissue growth, flowering rate and seed yield while mitigating salinity stress (Moghith et al., 2020). The application of silicon also increased the rigidity of the leaf epidermis and the nitrogen, phosphorous and potassium content of mature leaves (Moghith et al., 2020). However, how the silicon treatment ameliorates the negative effects of salt stress remains to be determined.

Fertilization

In Europe, considerable efforts were made to determine the impact of fertilization on the growth and yield of chia. In their first attempt to cultivate chia commercially in Europe, Bochicchio et al. (2015) observed no significant impact on the growth and yield of a short-day Peruvian cultivar with the application of nitrogen fertilizer. However, increasing the planting density resulted in higher grain yields per hectare despite a reduced yield per plant (Bochicchio et al., 2015). These findings are consistent with a study in Greece, in which the application of organic fertilizers (sheep manure) and commercial fertilizers had no effect on growth, biomass and vegetative tissue content (Bilalis et al., 2016). However, organic fertilizers increased the total protein content of vegetative tissues and the content of neutral detergent fibre and acid detergent fibre. These are important factors in determining feedstock quality for ruminants at different growth stages of the animal. The study found that increasing the sowing density resulted in higher biomass but did not affect total protein content in vegetative tissues (Bilalis et al., 2016). Both studies suggest that vegetative chia tissues can be used as an alternative ruminant forage.

A study conducted by Souza and Chaves (2017) analysed the impact of nitrogen, potassium and phosphorus on the growth of a Brazilian chia cultivar. In contrast to the studies by Bochicchio et al. (2015) and Bilalis et al. (2016), Souza and Chaves (2017) found that an individual application of these nutrients improved plant height, leaf number, inflorescence number and dry weight. Although a later study recommended that the optimal nitrogen:phosphorus:potassium ratio for chia growth in glasshouse conditions is 125:100:40, further research is needed to understand fully the reported discrepancies and determine ideal fertilizer concentrations for commercial chia production. Moreover, although chia can adapt to suboptimal soil conditions (Souza and Chaves, 2017), how soil composition can impact chia growth and yield has not yet been examined.

THE CHEMICAL COMPOSITION OF CHIA SEED AND ITS MUCILAGE

The chemical composition of chia seeds gained attention in the 1990s (Ayerza, 1995; Coates and Ayerza, 1996). Current knowledge of the chemical composition of chia is based primarily on studies that examined its fatty acid and lipid contents, protein composition, phenolic constituents and antioxidant activity. These studies will be discussed in the following sections.

Proximate composition

Pseudocereals, such as buckwheat, teff, amaranth and chia seeds, are more nutritious than staple cereals, such as wheat, rice and corn. They contain fewer carbohydrates but more proteins and lipids. Staple cereals mainly store starch in the endosperm, whereas many pseudocereals do not contain starch, and instead their endosperm and embryo are rich in lipids and proteins (Bekkering and Tian, 2019).

On average, chia seeds contain 35 % carbohydrates, 31 % fat, 25 % protein, 5 % water and 4 % inorganic matter (Table 1). Gas chromatography–mass spectrometry (GC-MS) analysis revealed that the majority of the polar compounds in mature chia seeds are carbohydrates, with sucrose and methylgalactose being the primary disaccharides and monosaccharides, respectively (de Falco et al., 2018a). Other carbohydrates detected in chia seeds include glucose, fructose, galactose, raffinose, gluconic acid, arabinose and mannose (de Falco et al., 2018b). The chia seeds contain higher total carbohydrates and inorganic contents in comparison to perilla (Perilla frutescens), at 23–26 % and 3 %, respectively, and flax seeds (Linum usitatissimum), at 28–30 % and 3 %, respectively (Table 1; Sargi et al., 2013). The fibre content of chia seeds (28–35 %) is also higher than that of flaxseed (23 %), barley (17 %), soybean (15 %), corn (13 %), wheat (13 %), sesame seed (8–12 %) and sunflower seed (9 %) (Ayerza, 2013; Porras‐Loaiza et al., 2014; Romankiewicz et al., 2017; Han et al., 2022).

Table 1.

A summary of the proximate composition of chia seeds from various studies published since 2010.

Carbohydrate (%)	Fat (%)	Protein (%)	Ash (%)	Moisture (%)	Study
37	32	25	6	6	Olivos-Lugo et al. (2010)
n/r	30–34	16–26	n/r	n/r	Ayerza and Catoes (2011)
33	32	22	4	7	Martinez et al. (2012)
n/r	35	n/r	n/r	n/r	Ciftci et al. (2012)
23–26	32–34	19	n/d	n/d	Ayerza (2013)
45	22	22	4	8	Sargi et al. (2013)
36	32	23	4	4	Sandoval-Oliveros and Paredes-Lopez (2013)
37	30	25	4	6	da Silva Marineli et al. (2014)
28–55	21–33	18–22	4–5	5–8	Porras-Loaiza et al. (2014)
n/r	29–32	n/r	n/r	n/r	Amato et al. (2015)
30–33	33–35	21–23	4–5	6–7	Barreto et al. (2016)
n/r	28–36	18–25	n/r	n/r	Ayerza (2016)
23	35	18	4	6	Imran et al (2016)
37	28	23	5	3	Scapin et al. (2016)
43	31	26	4	9	Romnakiewicz et al. (2017)
n/r	23–28	n/r	n/r	n/r	Grave et al. (2019)
n/r	33–34	17–22	5	6	Tuncil and Celik (2019)
Average (s.d.)
35 (6.7)	31 (3.4)	23 (2.4)	4 (0.6)	6 (1.6)

n/r refers to not recorded and n/d refers to not detected.

Multiple studies have measured the total protein content of chia, which, at 25 %, is equivalent to that of pea (Pisum sativum) (25 %), but higher than that of lentil (Lens culinaris) (23 %), chickpea (Cicer arietinum) (21 %), sunflower (Helianthus annuus) (21 %), sesame (Sesamum indicum) (18 %), oat (Avena sativa) (15 %), corn (Zea mays) (14 %), wheat (Triticum aestivum) (14 %), barley (Hordeum vulgare) (9 %) and rice (Oryza sativa) (8 %) (Table 1; Olivos-Lugo et al., 2010; Sargi et al., 2013; Porras‐Loaiza et al., 2014; Romankiewicz et al., 2017). Chia seeds have a lower lipid content (31 %) in comparison to perilla (40–42 %) and flax (38–45 %), but a higher content than soybean (24 %) and cotton seed (Gossypium hirsutum L.) (22–24 %) (Table 1; Olivos-Lugo et al., 2010; Ciftci et al., 2012; Sargi et al., 2013; Romankiewicz et al., 2017).

Several studies have reported an abundance of mineral nutrients, including phosphorus, potassium, calcium, zinc, magnesium, copper and iron, in chia seeds (Table 2; Ixtaina et al., 2011; Barreto et al., 2016; Santana et al., 2020). Consuming 15 g of chia seeds per day can provide a significant portion of recommended daily intake of essential micronutrients for adults (Table 2; Santana et al., 2020).

Table 2.

The percentage of essential micronutrients recommended daily intake present in 15 g of chia seeds (Ixtaina et al., 2011; Barreto et al., 2016; Santana et al., 2020).

Nutrients	Milligrams in 100 g chia seeds	Recommended daily intake for 19–30 years (mg day−1)		Percentage of recommended daily intake in 15 g chia seeds
		Female	Male	Female (minimum)	Male (minimum)
Phosphorus	765–780	1000	1000	11.5–11.7	11.5–11.7
Potassium	612–635	2800	3800	3.3–3.4	2.4–2.5
Calcium	531–567	1000	1000	8.0–8.5	8.0–8.5
Zinc	4.6–7.7	8	14	8.6–14.4	4.9–8.2
Magnesium	3.0–8.6	310	400	0.1–0.4	0.1–0.3
Copper	0.01–12.4	1.2	1.7	0.1–155	0.1–109.4
Iron	0.03–12.9	18	8	0.0–10.7	0.1–24.2

Fatty acid composition and lipid profile

Chia seeds are being studied extensively as an alternative to current sources of the long-chain fatty acids (LCFAs) EPA and DHA, such as fish and ocean krill. The ω-3 FAs in chia seeds could act as precursors for endogenous LCFA production in the human body (Crawford et al., 2000; Cowley et al., 2021). These seeds have shown promising results in reducing the levels of triglycerides in human blood, which, in turn, improves cardiovascular and cognitive functions. In a recent study, liquid chromatography–mass spectrometry was used to determine the lipid profile of chia seeds (Zare et al., 2019). The endosperm of chia seeds contains several major lipid classes, including triglycerides (TGs), diacylglycerols (DGs) and phosphatidylcholines (PCs). These classes are abundant in ω-3 PUFAs. Other lipid classes present in the endosperm of chia seeds include phosphatidylethanolamines (PEs), phosphatidylinositol (PIs), lysophosphatidylcholines (LPC), lysophosphatidylethanolamine (LPE), phosphatidylglycerols (PGs) and phosphatidylserines (PSs) (Zare et al., 2019).

Using liquid chromatography–mass spectrometry and GC-MS techniques, Ciftci et al. (2012) compared the lipid and FA profiles of chia, flax and perilla seeds and demonstrated that chia seeds contained the most TGs (51 %), followed by perilla (43 %) and flax (41 %) seeds (Table 3). Chia seeds also contained the highest amount of PUFAs, while the ω-3 FA content was similar amongst the three seed types. However, the ω-6:ω-3 ratio was higher in chia seeds (0.30) compared with flax (0.27) and perilla (0.22) seeds (Ciftci et al., 2012; Sargi et al., 2013). Moreover, the presence of four groups of phytosterols, including β-sitosterol (50 %), stigmasterol (30 %), campesterol (11 %) and Δ5-avenasterol (9 %), was reported in chia (Ciftci et al., 2012). Phytosterols, similar to cholesterol, are present in plant membranes. They can reduce the absorption of low-density lipoprotein cholesterol, potentially reducing the risk of hypercholesterolaemia in humans (Gao et al., 2023).

Table 3.

Summary of fatty acid analyses of chia seeds undertaken since 2010 in addition to two key references from 1995 and 1996.

Method	Location of cultivation	Palmitic acid (%)	Stearic acid (%)	Oleic acid (%)	Linoleic acid (%)	α-Linolenic acid (%)	ω-6:ω-3	Study
GC	Argentina	6–7	3–4	7–8	20–21	61–63	0.31–0.33	Ayerza (1995)
GC	Argentina	6–8	2–3	6–7	17–20	63–68	0.25–0.32	Coates and Ayerza (1996)
GC	Italy	7	3	6	19	64	0.29	Peiretti and Gai (2009)
GC	Ecuador	4–8	3–5	6–9	16–20	60–67	0.23–0.33	Ayerza (2010)
GC	Argentina, Ecuador & Bolivia	6–8	3–4	7–9	17–23	57–65	0.26–0.39	Ayerza and Catoes (2011)
GC/NMR	Argentina	5–7	3–4	5–6	17–20	64–69	0.24–0.31	Ixtaina et al. (2011)
GC	Argentina	7	3	7	22	60	0.37	Martinez et al. (2012)
GC	Prue	7	3	11	20	60	0.33	Ciftci et al. (2012)
GC	Ecuador	6	4	7	17–18	63–64	0.27–0.29	Ayerza (2013)
GC	Brazil	6	2	6	17	54	0.32	Sargi et al. (2013)
GC	Chile	7	3	7	18	63	0.29	da Silva Marineli et al. (2014)
GC	Mexico	6–7	3–5	7–9	19–20	60–63	0.31–0.32	Porras-Loaiza et al. (2014)
GC	Peru, Australia & Italy	10–12	1–3	5–7	19–21	58–63	0.32–0.35	Amato et al. (2015)
GC	Brazil	6	2	5	20	66	0.31	de Mello et al. (2015)
GC	Ecuador & Bolivia	6–8	3–4	7–9	18–21	58–64	0.28–0.36	Ayerza (2016)
GC	n/r	7	3	8	12	61	0.2	Imran et al. (2016)
GC-MS	Bolivia	3–4	6–7	6	18	64–66	0.27–0.29	Castejon et al. (2017)
GC	n/r	5	2	4	17	73	0.23	Romnakiewicz et al. (2017)
GC	France	8	3	8	20	60	0.33	Grave et al. (2019)
		Average (s.d.)
		7 (1.4)	3 (1.0)	7 (1.5)	19 (2.0)	62 (3.7)	0.30 (0.04)

Chia seeds contain three important C18 FAs: oleic acid (OA; C18:1n-9, ω-9), linoleic acid (LA; C18:2n-6, ω-6) and α-linolenic acid (ALA; C18:3n-3, ω-3). On average, saturated fatty acids (SFAs) and PUFAs make up 10 % and 81 % of the total FA content, respectively (Table 3). The major monounsaturated fatty acid in chia seeds is oleic acid (7 %). In addition to this, small amounts of palmitoleic acid (C16:1n-7, ω-7) and paullinic acid (C20:1n-7, ω-7) have also been detected. Some saturated fatty acids, such as myristic acid (C₁₄H₂₈O₂), behenic acid (C₂₂H₄₄O₂) and lignoceric acid (C₂₄H₄₈O₂), have also been found in trace amounts in chia seeds (Ayerza, 2010, 2016; Ciftci et al., 2012). Over the past decade, researchers have studied the FA and lipid compositions of chia seeds extensively. However, many unanswered questions remain regarding how chia produces higher levels of ω-3 PUFAs in comparison to other oilseed crops.

Ecological and agronomic factors affecting FA synthesis

The FA profile of chia seeds can be influenced by environmental factors (Ayerza, 2010, 2016; Ayerza and Coates, 2011; Porras‐Loaiza et al., 2014). One study involved two types of domesticated genotypes [white (Tzotzo1) and spotted black (Iztac)] cultivated in five different locations across Ecuador (Ayerza, 2010). These locations varied in elevation, latitude, longitude, mean temperature, annual rainfall and soil type. Combined data from both genotypes revealed that high elevation positively impacts the total ω-3 FA content by significantly altering oil content and FA profiles between environments. However, when the results for each genotype were considered independently, the environment did not affect the chemical composition of the seeds (Ayerza, 2010).

Ayerza and Coates (2011) found significant differences in lipid and protein content and in FA profiles of three chia varieties cultivated in Argentina, Bolivia and Ecuador. Seeds from plants grown at higher elevations were lower in protein content but contained more ω-3 and lower levels of ω-6, ω-9 and SFAs, including palmitic acid (PA) and stearic acid (SA). Therefore, the location where chia seeds are grown should be considered, because it can affect their FA composition (Ayerza and Coates, 2011). In addition, although nitrogen fertilization did not change the total lipid and FA composition of chia seeds (Amato et al., 2015; de Falco et al., 2017), different irrigation regimes (i.e. no irrigation vs. non-limiting water supply) alter FA composition; for example, for the G8 genotype, it was observed that unlimited water supply increased the ω-3, ω-6 and ω-9 content (de Falco et al., 2018b).

Genotype and age

Studies conducted by Ayerza in 2010 and 2013 have suggested that the total lipid content, the FA profile and the ratio of ω-6:ω-3 in chia seeds are relatively similar across different genotypes. However, some studies have reported significant differences in the FA profile and ω-6:ω-3 ratio among different genotypes (Ixtaina et al., 2011; de Falco et al., 2017, 2018a, 2018b). The inconsistent findings from previous studies make it challenging to determine whether the genotype has an impact on the FA composition of chia seeds. Therefore, future research should consider population genetics, allelic association experiments and gene–environment interaction studies to determine the genetic basis of this phenomenon.

The effect of seed maturation on lipid content was investigated for a commercial chia cultivar (Oruro^©) bred for Europe (Gravé et al., 2019). During the maturation of the Oruro^© seeds, there was an increase in the ω-3 content and a decrease in the content of palmitic acid, stearic acid, ω-9 and ω-6 FAs. The ω6:ω3 ratio was highest (0.42) at 17 days after flowering and remained stable at 0.33–0.35 until 24 days after flowering (Gravé et al., 2019). The FA profiles of vegetative tissues and seeds of chia changed as the plant aged and transitioned to later developmental stages. The FA profiles of both vegetative tissues and seeds were examined to determine their potential as animal feed (Peiretti and Gai, 2009). In mature plants, the ω-3 FA content in vegetative tissues decreased while an increase in ω-6 and ω-9 FA contents was observed (Peiretti and Gai, 2009). Further exploration is needed to determine whether chia vegetative tissues can be used as ruminant feed to supplement milk and meat with ω-9 FAs despite the increase in ω-6 and ω-9 FAs in mature vegetative chia tissue.

Extraction methods

The extraction methods used for obtaining oil from chia seeds significantly impact the oil composition and the bioavailability of nutrients. These factors ultimately impact the commercial production and shelf-life of the oil. For instance, when chia seeds are pressed, the oil yield is lower (20–27 %) than when using a solvent extraction method (25–34 %) (Ixtaina et al., 2011). A study on the screw pressing extraction method found that the oil yield decreases with higher seed moisture and pressing speed, whereas temperature during pressing has no significant effect (Martínez et al., 2012). It was found that altering seed moisture, pressing speed and restriction dies can negatively affect the peroxide index, which measures oil oxidation. However, other oil quality measures remained unaffected, including acidity, total tocopherol content and antioxidant activity (Martínez et al., 2012).

Preconsumption treatments play a vital role in determining the nutritional value of food products in conjunction with the extraction methods. Pretreatments, such as autoclaving (32 %), microwave roasting (31 %), oven drying (23 %) or water boiling (3 %), cause a significant decrease in the total lipid content extracted from chia seeds. These can also lower the levels of ω-3 and ω-6 FAs in the resultant oil and negatively affect its colour and flavour during storage (Imran et al., 2016). This suggests that harsh pretreatment conditions, including extreme temperatures, can reduce the nutritional value and commercial quality of chia seed oil. In a recent study, we investigated whether soaking chia seeds in water for 24 h or in 0.1 m hydrochloric acid (pH 2) under constant shaking at 300 rpm for 2 h affects the extractability of PUFAs and lipids. The amount of ω-3 and ω-6 FAs, in addition to lipids from the TG and DG families, increased when seeds were soaked in water for 24 h, resulting in a lower ω-6:ω-3 ratio. This ratio decreased from 0.50 to 0.32 (Zare et al., 2019). Moreover, the concentration of ω-3, ω-6 and ω-9 FAs extracted from pretreated chia seeds increased in comparison to untreated samples. These results suggest that precooking methods, such as grinding or overnight soaking, can make the beneficial nutrients of chia seeds more accessible to consumers (Zare et al., 2019).

There is currently no evidence to support the idea that consuming untreated chia seeds can provide the required amount of PUFAs in the human digestive system. It has been speculated that owing to the small size and hard outer shell of the chia seed, the human digestive system is unable to extract PUFAs from untreated chia seeds (Zare et al., 2019). However, recent studies suggest that pretreating or precooking the seeds might improve the bioavailability of PUFAs (Zare et al., 2019; Santana et al., 2020).

Essential fatty acids, such as ω-6 and ω-3, are likely to be beneficial for human health owing to their preventive effects against cardiovascular and inflammatory disorders. Maintaining a balanced ratio of ω-6 to ω-3 in the diet is crucial for optimal effectiveness. A ratio of <1 is recommended by Kaliannan et al. (2019), but the average Western diet has a much higher ratio of 15–17 (Simopoulos, 2016; Kaliannan et al., 2019). The ω-6:ω-3 ratio in chia seed (0.30) is lower than in olive (Olea europaea) (7.69), soybean (Glycine max) (6.67), walnut (Juglans) (5.00) and canola (Brassica napus) (2.22), suggesting that chia seeds could be a good option to help reduce the ω-6 to ω-3 ratio in our diet. The demand for alternative sources of health benefits associated with ω-3 PUFAs has fuelled research into improving our understanding of the molecular mechanisms underlying their biosynthesis and oil production in chia. This knowledge, combined with new bioengineering approaches driven by techniques, such as CRISPR (clustered regularly interspaced short palindromic repeats), could allow for the synthesis of beneficial ω-3 PUFAs in ideal ratios in chia and other oilseeds.

Protein and amino acid composition

The high protein content of chia seeds, along with their high in vitro digestibility (79 %), in comparison to maize (67 %), sorghum (59 %), rice (59 %) or wheat (53 %) (Sandoval-Oliveros and Paredes-López, 2013), suggests that chia seed is an excellent source of protein to be included in the human diet. The essential storage proteins in chia seeds, including prolamins, albumins, glutelin and globulins, are highly thermostable, making chia seeds ideal for use in high-temperature-processed foods, such as cakes and breads (Olivos-Lugo et al., 2010; Sandoval-Oliveros and Paredes-López, 2013; Peláez et al., 2019).

Chia seed proteins contain 42–43 % of essential amino acids (EAs), including histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan and valine. This is comparable to the contents in soybean (42 %) and cotton seed (39 %) (Olivos-Lugo et al., 2010). These nine essential EAs cannot be synthesized by humans and animals and thus must be obtained through a balanced diet to maintain a healthy metabolism. EAs offer a variety of health benefits; they help to maintain neuronal and cognitive functions, balance the levels of serotonin (a neurotransmitter that regulates appetite, mood and sleep) and assist in protein, hormone and enzyme biosynthesis and metabolism, all of which are crucial for human energy metabolism.

Essential amino acids also facilitate the absorption of calcium and zinc, which are crucial micronutrients for the development of muscle tissues and neural networks (Arumugam et al., 2021). Chia seeds contain various non-essential amino acids, including glutamic acid, arginine, aspartic acid, alanine, γ-aminobutyric acid, proline and N-acetylglutamic acid (Olivos-Lugo et al., 2010; Sandoval-Oliveros and Paredes-López, 2013; Urbizo-Reyes et al., 2019). The mechanisms of EA biosynthesis and its interactions with regulatory transcriptional and post-translational components in chia are not yet well understood. This emphasizes the requirement for more comprehensive and reliable genomic and transcriptomic datasets of chia. Such datasets will accelerate the study of the genetics behind EA synthesis.

Antioxidant properties and phenolic content

Chia seeds are classified as a functional food owing to their high phenolic compound content, ranging from 0.53 to 1.64 mg gallic acid (GAE) g⁻¹ (da Silva Marineli et al., 2014; Martínez-Cruz and Paredes-López, 2014; Beltrán-Orozco et al., 2020), in comparison to the range of 0.85–1.40 mg GAE g⁻¹ in wheat, barley and corn (Horvat et al., 2020). Numerous studies have examined the phenolic content of chia seeds and investigated the link between phenolic compounds and antioxidant activity. Functional foods are known for their high total antioxidant content, which is believed to reduce the risk of degenerative diseases, such as cancer, in humans by decreasing free radicals (Das Gupta and Suh, 2016).

Chia seeds are rich in phenolic compounds known for their potent antioxidant properties. These compounds include rosmarinic acid, protocatechuic acid, myricetin, kaempferol, quercetin, caffeic acid, genistein, daidzein, ferulic acid, gallic acid, chlorogenic acid, citric acid, quinic acid, isocitric acid, secoisolariciresinol diglucoside, 5-caffeoylquinic acid, choline, tanshinone I and 15,16-dihydrotanshinone (Ixtaina et al., 2011; Martínez-Cruz and Paredes-López, 2014; Oliveira-Alves et al., 2017; da Silva et al., 2020; Mitrović et al., 2021). Chia leaves contain phenolic compounds in addition to seeds. These compounds include hydroxycinnamic acid derivatives and two uncommon flavonoids, acetyl orientin and acetyl vitexin. These compounds are known for their potential antidepressant properties, making chia leaves a promising source of valuable nutraceuticals (Can et al., 2013; Amato et al., 2015).

Chia seed oil contains significant tocopherols, ranging from 238 to 895 mg kg⁻¹. Although this amount is similar to that found in peanut oil (399 mg kg⁻¹), it is lower than that of soybean (1798 mg kg⁻¹), sunflower (634 mg kg⁻¹) and flaxseed oil (588 mg kg⁻¹). Tocopherols are organic compounds with antioxidant properties that might help to lower the risk of human cancer or other degenerative diseases (Das Gupta and Suh, 2016). These lipid-soluble compounds are the main types of vitamin E and consist of four forms: α-, β-, δ- and γ-tocopherols (Das Gupta and Suh, 2016). All four groups of tocopherols are present in chia seeds and have a positive correlation with ω-3 FA content, but γ-tocopherol is the most abundant (Ixtaina et al., 2011; Ciftci et al., 2012; Amato et al., 2015; Castejon et al., 2017; Gravé et al., 2019).

Pigments, such as chlorophyll and carotenoids, responsible for the green, red, orange and yellow colours of plants, also act as antioxidants for both plants and humans (Kabir et al., 2022). These pigments have been found in significant amounts in chia seeds (Ixtaina et al., 2011; Amato et al., 2015), highlighting the potential of chia seeds as a functional food. Beta-carotene is found in various plant species and is most abundant in chia seeds (0.53–1.21 mg kg⁻¹). Although this is lower than rapeseed, which has 1.7 mg kg⁻¹, it is similar to the amount found in maize (0.9 mg kg⁻¹) and flaxseed (0.7 mg kg⁻¹). In comparison, chia seeds have a higher concentration of beta-carotene than sunflower (0.1 mg kg⁻¹) and soybean (0.3 mg kg⁻¹) (Ixtaina et al., 2011).

The Trolox equivalent antioxidant capacity (TEAC) of chia (2.56 mmol g⁻¹) as measured by the 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) radical cation assay (ABTS^.+ assay) is lower than that of brown perilla (4.06 mmol g⁻¹), white perilla (3.32 mmol g⁻¹) and flax seeds (3.38–3.70 mmol g⁻¹) (Sargi et al., 2013). However, screening the antioxidant capacity of several Salvia species, including S. candidissima (49.7 %), S. caespitosa (41.3 %), S. hypargeia (34.6 %), S. sclarea (23.4 %) and S. euphratica (20.7 %), suggests that with 68.83 %, chia has a higher antioxidant potential than other Salvia species (Tepe et al., 2006; Martínez-Cruz and Paredes-López, 2014). The amount of phenolics and flavonoids extracted from chia seeds varies depending on the extraction method. For instance, a method using higher solvent concentrations extracts more compounds, leading to higher antioxidant activity. However, extraction temperatures do not significantly impact the antioxidant properties of chia seed extracts (Scapin et al., 2016). Methanol extraction yields more free phenolic compounds with higher antioxidant activity in comparison to ethanol extraction (Mitrović et al., 2021).

Various agronomic factors can affect the phenolic composition and antioxidant potential of chia seeds. Research has shown that although the irrigation system has no significant effect, the application of nitrogen fertilizers negatively impacts the phenolic compound composition in chia seeds (Amato et al., 2015; de Falco et al., 2018b). Interestingly, studies on the relationship between genotypes, phenolic content and antioxidant activity have produced inconclusive results. For instance, De Falco et al. (2018b) did not detect any differences in antioxidant activity between chia genotypes. However, Tunçil and Çelik (2019) reported that white chia seeds had a higher antioxidant activity (232 µmol troloxequivalent (TE) g⁻¹) than black chia seeds (197 µmol TE g⁻¹). In another study, seeds of long-day-flowering genotypes were found to have higher antioxidant activity and polyphenol content than seeds of commercial short-day-flowering varieties (de Falco et al., 2018a). These findings suggest that the cultivation of chia genotypes in different climates could produce seeds with varying degrees of antioxidant nutritional value.

The potential of chia seed extracts as an antioxidant might not necessarily lead to effective digestion and absorption in the human body. Pellegrini et al. (2018) investigated the bioaccessibility, recovery rate and antioxidant capacity of flavonoids and phenolic compounds using simulated in vitro gastrointestinal digestion. Both normal and defatted chia seeds showed similar recovery rates of phenolic compounds after each digestion phase. However, oral digestion negatively impacted the recovery rate, whereas the gastric stage restored it, and intestinal digestion resulted in an increased recovery rate (Pellegrini et al., 2018). The recovery rate of flavonoids in normal and defatted chia seeds was found to differ only during the intestinal digestion stage. Bioaccessibility of flavonoids was comparatively lower than than that of other phenolic compounds in untreated chia seeds, while defatted seeds showed similar trends for both. The antioxidant activity of differently treated chia seeds showed a similar pattern to phenolic compounds, i.e. reduction after the oral phase, recovery during the gastric phase, and increase after the intestinal stage. However, defatted chia seeds exhibited greater antioxidant capacity than non-treated seeds in all stages of digestion (Pellegrini et al., 2018).

In addition to antioxidant activity observed in seed extracts, other tissues, such as leaves and stems, exhibited antimicrobial and antifungal activity. When tested in vitro with phytopathogenic bacteria and fungi, chia stem and leaf oil inhibited bacterial growth and mycelium expansion (Elshafie et al., 2018). The essential oil extracted from aerial parts of chia contains 60 sesquiterpenes that display antimicrobial and antifungal activity. These sesquiterpenes include sesquiterpene hydrocarbons like (Z)-caryophyllene, α-humulene, δ-amorphene, γ-gurjunene, oxygenated sesquiterpenes such as α-eudesmol, caryophyllene oxide, spathulenol and monoterpenes (<1 %). Among these, caryophyllene makes up 30 % of essential oils and is known for its antioxidant and anti-inflammatory effects. Although most studies focused on the fatty acid composition of chia, these observations present a promising opportunity for future research in other nutraceutical compounds and active biomarkers. Such research could potentially lead to new discoveries with significant human health benefits and applications in the pharmaceutical industry.

Chemical composition and properties of chia seed mucilage

Mucilage is a hydrophilic polysaccharide that some plant seeds release upon imbibition (Cowley et al., 2021). The composition of the seed mucilage varies between plant species and is associated with various biological functions (Geneve et al., 2017; Souza et al., 2020). The main components of chia seed mucilage are carbohydrates (primarily C1–C6 sugars), which make up 74–94 % of the total mucilage. The remainder contains small amounts of moisture (4–11 %), protein (3–4 %), lipid (1–2 %) and inorganic matter (1–10 %) (Salgado-Cruz et al., 2013; Timilsena et al., 2016; de Campo et al., 2017; Di Marsico et al., 2018; Lazaro et al., 2018). Chia seed mucilage contains heteropolysaccharides made up of xylose (38 %), glucose (20 %), glucuronic acid (19 %), arabinose (10 %), galactose (6 %) and galacturonic acid (5 %), but other sugar residues, including (1→4)-α-ᴅ-glucopyranosyl, (1→4)-β-ᴅ-xylopyranosyl and 4-O-methyl-α-ᴅ-glucopyranosyluronic acid, have also been detected (Timilsena et al., 2016).

Interestingly, a new trisaccharide called planteose has been discovered in chia seed mucilage. Planteose is made up of α-ᴅ-galactopyranosyl-(1→6)-β-ᴅ-fructofuranosyl-(2→)-α-ᴅ-glucopyranoside (Xing et al., 2017). The high levels of planteose found in chia seed mucilage and a proposed high-yield purification method by Xing et al. (2017) make it a promising compound for future research. This includes studying its functions and potential prebiotic properties. Although recent discoveries have shed some light on the composition of chia seed mucilage, significant discrepancies exist between the type and level of chemicals reported by various research groups (Ang et al., 2023). Therefore, further studies are necessary to understand the molecular mechanisms behind mucilage formation and extrusion in chia seeds.

Biological functions of chia seed mucilage

Seed mucilage serves various purposes, such as retaining water for germination, protecting against environmental and biological stresses, helping seeds adhere to moist soils (antitelechory) and facilitating seed dispersal through animal hosts (zoochory) (Phan and Burton, 2018). However, the specific biological functions of seed mucilage vary among different species and have only recently been studied in relationship to chia seeds.

Removing the mucilage from chia seeds led to a decrease in germination rate from 93 to 68 % and also caused the radicle length to reduce by half after 26 h of imbibition. However, the final germination rate after 48 h remained the same. Mucilage removal also resulted in lower germination rates on salt media, preventing mucilage expansion (Geneve et al., 2017). Surprisingly, drought stress did not affect the germination rates of seeds without mucilage (Geneve et al., 2017). This contradicts the assumption that mucilage helps in retaining water (Phan and Burton, 2018). However, when subjected to salinity stress induced by sodium chloride (NaCl) or sodium carbonate (Na₂CO₃), the germination of seeds without mucilage was significantly reduced (≤48 %). This suggests that mucilage could act as a buffer zone by excluding or diluting salt during germination. It was found that exposing chia seeds to cold treatment does not affect their germination rate, regardless of whether or not they have mucilage (Geneve et al., 2017). When chia seeds were inoculated with Colletotrichum graminicola, a hemibiotrophic fungus economically important to plant health and disease prevention, their germination rate was reduced to 61–64 %. However, the presence of mucilage did not affect this rate. Interestingly, fungal growth was observed more around seeds that had mucilage, suggesting that mucilage might play a role in the interactions between seeds and microbes, like root mucilage in the rhizosphere (Geneve et al., 2017).

Other studies have also revealed that seed mucilage increases the soil microbial community. For example, enhanced beneficial fungal–bacterial interactions increased drought tolerance and seedling growth in Artemisia sphaerocephala (Hu et al., 2019a, b). Chia seed mucilage has been suggested to act as a soil stabilizer by enhancing its aggregate stability and by regulating oxygen and CO₂ levels. This, in turn, affects seed germination, dormancy and viability, making it a potential candidate for agronomic applications (Di Marsico et al., 2018). However, the biological functions of chia seed mucilage are not yet fully understood and could provide an untapped opportunity for discovering new insights with significant economic importance.

Commercial and industrial applications of chia seed mucilage

Several studies have highlighted the potential of chia seed mucilage as a hydrocolloid for use in food products (e.g. chia seed pudding). Owing to its fibre size and chemical composition, it has been suggested as an emulsifier and stabilizer in food products or as a pharmaceutical agent for the controlled release of drugs (Salgado-Cruz et al., 2013; Timilsena et al., 2016; Urbizo-Reyes et al., 2019).

The water-holding capacity of chia seed mucilage is a key characteristic for food applications. Chia seeds can absorb ≤27 times their weight in water. However, this value can vary depending on the temperature, pH and ionic strength of the solution. The maximum absorption occurs at pH 9, 80 °C and low ionic strength (Muñoz et al., 2012). The extracted chia seed mucilage has a water-holding capacity ranging from 23 to 35 g g⁻¹ (water/mucilage), four times higher than Arabic gum and comparable to guar gum (25 g g⁻¹) (Timilsena et al., 2016; Lazaro et al., 2018). The high thermal stability of chia seed mucilage (224–300 °C) makes it suitable for baking at high temperatures, resulting in reduced fat content and calorific value without impacting the protein content of the processed food (Timilsena et al., 2016; Punia and Dhull, 2019).

Chia seed mucilage has been proposed as an anionic polyelectrolyte in processing functional foods or in coacervation technology for encapsulation of bioactive ingredients (Timilsena et al., 2016; de Campo et al., 2017; Us‐Medina et al., 2017). De Campo and colleagues have used chia mucilage to encapsulate chia oil by creating stable oil nanoparticles. This process resulted in a high encapsulation efficiency of 83 %, while the nanoparticle stability was maintained ≤300 °C. The study found that using chia seed mucilage at high temperatures promotes its use in food processing. Additionally, the encapsulated oil displayed better oxidative stability, specifically for PUFAs, in comparison to non-encapsulated oil (de Campo et al., 2017). In an effort to improve the encapsulation efficiency of chia oil using chia seed mucilage, a study by Us-Medina et al. (2017) used a 50:50 ratio of mucilage to alginate and obtained similar results to a previous study conducted by de Campo et al. (2017). Both studies concluded that chia seed mucilage is a suitable replacement for synthetic polymers in the encapsulation of bioactive oils.

The mucilage of chia seeds possesses unique physical properties and chemical characteristics that suggest a wide range of possible applications in the food, medicinal and agricultural industries. However, further studies are needed to develop efficient methods for extracting the mucilage, taking into account factors such as temperature and the ratio of seeds to water. Additionally, antimicrobial and antifungal properties of chia seed mucilage need to be investigated further and validated. This could potentially lead to the development of new products aimed at promoting human health and well-being. The use of chia seed mucilage as a soil moisturizer is also worth exploring, because it could improve crop growth and yield while reducing soil erosion and degradation. It is important to investigate both laboratory and field applications to maximize crop production in the future (Di Marsico et al., 2018).

THE GENETIC AND MOLECULAR PROPERTIES OF CHIA

Little research has been conducted on the genetics of chia plants despite significant efforts to analyse the chemical composition of chia seeds. Haque and Ghoshal (1980) conducted research that established the chromosome number of chia to be 2n = 12. This was confirmed by Estilai et al. (1990) in a study that focused on the meiotic behaviour of the cells and autosomal chromosomes in the root tips of chia plants. Moreover, it was found that chia has six chromosome pairs, including one metacentric pair, one submetacentric pair, three sub-telocentric pairs and one telocentric pair (Estilai et al., 1990; Palma-Rojas et al., 2017), and its genome size was estimated to be ~460 Mbp (gDNA = 0.93 ± 0.016 pg), using flow cytometry (Maynard and Ruter, 2022). The chloroplast genome of chia was estimated to be 0.15 Mbp in size, with a GC content of 38 %. It comprises 114 unique genes, including 80 protein-coding genes, 30 tRNAs and four rRNAs (Zhang et al., 2020).

The genome of chia has recently been assembled and sequenced in studies conducted by Wang et al. (2022), Gupta et al. (2023) and Zare et al. (2024). These studies aimed to uncover the genetic mechanisms responsible for variations in chia seed coat colour (Wang et al., 2022), oil biosynthesis, duplication events, types and synteny blocks (Zare et al., 2024). The analysis by Wang et al. (2022) resulted in a chromosome-level genome assembly (347.6 Mbp) that covers 77 % of the estimated genome size. The authors suggested that a single gene controls the black and white colour of the chia seed coat (Wang et al., 2022), which supports the findings of Cahill and Provance (2002), who found that a single recessive gene (scc) is responsible for the white colour of the seed coat (Cahill and Provance, 2002). However, there are insufficient data to identify the candidate gene responsible for the variation in chia seed colour and its pigmentation patterns (Wang et al., 2022). Wang et al. (2022) proposed that the high production of ω-3 FAs in chia seeds was linked to an increase in the number of fatty acid biosynthesis 2 (FAB2) gene copies. However, this hypothesis lacked support from a thorough evolutionary analysis.

In the other genome study, Zare et al. (2024) investigated the evolution of oil biosynthesis genes in chia using a high-quality, near-complete chromosome-level reference genome assembly (RefSeq: GCF_023119035.1, 321.5 Mbp). An analysis of 32 922 protein-coding genes across six chia chromosomes indicated that the chia genome underwent a whole-genome duplication event, which is typical in Salvia species. Furthermore, the chia genome also underwent a γ whole-genome triplication event, common to all angiosperms (Zare et al., 2024). According to the orthogroup inference and 4-fold degenerative transversion rates, it has been suggested that the high abundance of ω-3 FAs in chia seeds is not the result of whole-genome duplication. Instead, the evolutionary expansion of certain lipid biosynthesis gene families, particularly the (stearoyl-acyl carrier protein desaturase) SAD gene family (which has 13 gene copies), is believed to be the main driver behind this phenomenon (Zare et al., 2024).

Authors have identified 11 Sh-SAD genes in the telomeric region of chromosome 1 in the reference genome of chia (Zare et al., 2024). Among these genes is a tandem array of six Sh-SAD genes, of which five genes are unique to chia. These chia-specific genes and their particular arrangement in the tandem array could potentially significantly impact oil biosynthesis and ω-3 FA accumulation in chia seeds. This hypothesis is supported by the co-expression of four tandem genes (ShSAD2, ShSAD3, ShSAD4 and ShSAD6) with five other ShSAD genes (ShSAD1, ShSAD7, ShSAD11, ShSAD12 and ShSAD13) in reproductive tissues. Overactivity and co-expression of ShSAD genes in these tissues produce an abundance of C18:1-ACP, which serves as a substrate for the desaturation of FAs in both the plastid and endoplasmic reticulum (ER) (Zare et al., 2024). Their identification suggests an avenue for detailed investigation into their mechanisms, regulation and potential manipulation. As a result, these genes could be explored as candidates for further study in molecular genetics and genetic engineering approaches.

Transcript profiles of chia seed and vegetative tissues

The de novo assembly of transcripts from five developmental stages of white chia seeds identified 76 014 transcripts (Sreedhar et al., 2015). A total transcript length of 66.9 Mbp was determined, and 27 122 unique genes were identified. Among the 5596 reported single sequence repeats, 50.4 % were found to be trinucleotide motifs, while 35.4 % were dinucleotide motifs, with AG/CT and GA/TC being the most prevalent repeat motifs (Sreedhar et al., 2015). Sreedhar et al. (2015) identified candidate genes that are linked to various metabolic processes, such as glycerophospholipid metabolism, inositol phosphate metabolism, phosphatidylinositol signalling, glycosylphosphatidylinositol-anchor biosynthesis and glycerolipid metabolism. Notably, the transcriptome of developing chia seeds had a high number of transcripts related to ALA, AL and TG biosynthesis. These transcripts were highly abundant in both the Kennedy and soluble TG biosynthesis pathways (Sreedhar et al., 2015).

This transcript analysis was later expanded by other studies conducting comparative transcript profiling of chia seeds and vegetative tissues at different stages of development (Peláez et al., 2019; Wimberley et al., 2020; Gupta et al., 2021). Peláez et al. (2019) investigated the transcriptome of four domesticated (Black, White, Spotted and Xonotli) and four wild (Cualac, Michoacán, Oaxaca and Guerrero) Mexican chia varieties. A reference transcriptome with a total size of 171 Mbp was generated, and a total of 146 951 transcripts were assembled, where the shortest and longest contigs were 201 and 66 772 bp, respectively, with an N50 contig length (N50) of 1949 bp. The annotation of the reference transcriptome led to the identification of 69 873 annotated transcripts, of which 12 583 were mapped to unique identifiers representing >10 000 different known genes (Peláez et al., 2019). The authors found that Arabidopsis and chia seeds share similarities in various pathways related to functional mechanisms, such as the elongation of FAs, wax biosynthesis, phospholipid (PC) signalling, TG biosynthesis, PC synthesis and editing, and FA biosynthesis. The gene ontology (GO) annotations reported by Sreedhar et al. (2015) were similar to the Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, which showed overexpression of gene families involved in various biosynthetic pathways, such as glycerophospholipid, glycerolipid, FA metabolism and the biosynthesis of secondary metabolites in Mexican chia varieties.

Only 13 genes were found to be present exclusively in wild populations and 18 in domesticated varieties, confirming a high similarity between wild and domesticated chia varieties (Peláez et al., 2019). This similarity is because chia has not been bred extensively and therefore has not experienced the genetic divergence between wild and domesticated varieties as is typical in breeding. The RNA-dependent RNA polymerase 1 (RDR1), which plays an important role in plant defence and virus resistance (Prakash et al., 2020), is an example of a unique gene present only in domesticated Mexican chia varieties (Peláez et al., 2019). Although very few genes were variety specific, 1034 transcripts differed in abundance between wild and cultivated varieties. Domesticated varieties showed a higher expression of acyl-lipid-related genes, which were classified in the GO groups such as ‘Very long-chain FA metabolic processes’, ‘FA biosynthetic process’ and ‘FA metabolic process’ (Peláez et al., 2019). This indicates that chia has adapted to breeding for seeds that have higher oil content. In a recent study, Gupta et al. (2021) identified the primary enzymes responsible for seed synthesis and elongation of FAs. Interestingly, transcripts encoding the acetyl-CoA carboxylase (ACCase), responsible for producing malonyl-CoA from acetyl-CoA early in FA biosynthesis, were found only in vegetative tissues. This suggests that ACCase mainly regulates carbon flow to FA biosynthesis in the chloroplasts of chia leaves (Fan et al., 2019; Gupta et al., 2021).

In contrast, the wild varieties have shown higher expression of Reduced Oleate Desaturation 1 (ROD1) and other differentially expressed genes involved in the synthesis of TG, FA degradation and β-oxidation [e.g. multifunctional protein (MFP), 3,5-delta2,4-dienoyl-CoA isomerase (DCI), transcripts encoding steroleosin, caleosin, triacylglycerol and monoacylglycerol lipase] and wax biosynthesis (e.g. ribonuclease CER7) (Peláez et al., 2019). The genetic variation in lipid biosynthesis between wild and cultivated chia varieties could be the result of natural selection or domestication. This phenomenon could be confirmed by measuring changes in amino acid substitution rates in the coding regions of the genes involved in chia lipid biosynthesis.

Peláez et al. (2019) also investigated the impact of environmental factors, such as elevation, rainfall and annual average temperature, on lipid metabolism genes in plants grown in different regions of Mexico. The study found that plants cultivated in different locations in Mexico but with similar latitudes had >5265 differentially expressed transcripts, indicating a clear response to diverse environments at the transcript level (Peláez et al., 2019). The study identified 2243 transcripts related to lipid metabolism, including 18 genes related to FA elongation and wax biosynthesis, 15 genes linked to TG biosynthesis, and 11 genes related to TG and FA degradation. Overall, the expression of genes associated with TG and FA synthesis/degradation differed between domesticated and wild-type chia seeds and in different growth conditions. The authors conducted a genome-wide analysis of single nucleotide polymorphisms, concluding that wild-type varieties exhibit higher genetic diversity, confirming findings made by Cahill (2003).

Wimberley et al. (2020) and Gupta et al. (2021) conducted studies to explore the gene expression patterns in various chia tissues at different stages of development. Wimberley et al. (2020) investigated the gene expression of secondary metabolites and terpenoid biosynthesis in the leaves and roots of the Pinta cultivar. The results revealed that defence response and lignin metabolism transcripts are more prevalent in roots, whereas photosynthetic processes are enriched in leaves, as expected. The GO annotation revealed significant enrichment of vitamin biosynthesis in leaves, including the biosynthesis of riboflavin (Vitamin B2), thiamine (Vitamin B1), pyridoxine (Vitamin B6), ubiquinone and other terpenoid-quinone metabolic pathways (Wimberley et al., 2020).

The study conducted by Gupta et al. (2021) on chia leaves found a significant variation in gene expression levels between the early and late stages of leaf development. In mature leaves, there was a notable abundance of transcripts for wall-associated receptor kinases (WAKs), leucine-rich receptor-like kinases (LRR-RLKs), root hair defective-binding protein (RHD3), cytochrome P450s, glycosyltransferases and WRKY transcription factors (TFs) (Gupta et al., 2021). During the early stages of leaf development, bHLH domain-containing TFs (SPEECHLESS) and growth regulating factors (GRF2 and GRF5) were found to be highly expressed, indicating their crucial roles in regulating cell proliferation in young leaves (Gupta et al., 2021; Zhang et al., 2021). Similar studies on other plant species suggest that GRFs are crucial for leaf growth and development, whereas WAKs and LRR-RLKs are important for abiotic stress and disease resistance (Wu et al., 2020; Zhang et al., 2021). Moreover, during the initial stages of chia development, the gene expression profiles of cotyledons and shoots showed noticeable differences between leaves and seeds. Specifically, certain genes were found to encode photosystem I and II proteins, TFs from the leucine zipper family, calcineurin-like phosphoesterases domain-containing proteins and aquaporins (Gupta et al., 2021).

Upon further analysis of the differentially expressed genes, it was observed that certain transcripts in chia leaf and root tissues are evolutionarily related to the terpene synthase (TPS) gene family of tomato (Solanum lycopersicum) (Wimberley et al., 2020). The TPS enzymes catalyse the synthesis of terpenes, which are crucial for plant defence against both biotic and abiotic stresses and for attracting pollinators (Zhou et al., 2020). Wimberley et al. (2020) investigated the impact of abiotic stress and hormone treatments on the chia leaf transcriptome to explore the biological function of genes involved in TPS biosynthesis. Although hormone treatments using gibberellic acid and indole-3-acetic acid did not have any effect on the expression of TPS genes, cold treatment affected only one TPS gene (TPS-g). However, heat treatment had a mixed impact on the expression of different members of the TPS gene family. For instance, TPS-a expression was decreased, whereas TPS-b expression was significantly elevated (Wimberley et al., 2020). Terpenes possess anti-plasmodial and other medicinal properties, including antioxidant, anti-inflammatory, antiseptic, diuretic and digestive activity similar to antimalarial drugs (Braich et al., 2019). The chia leaves and flowers contain terpene-rich oils and substances that could potentially be used as an antimicrobial phytopharmaceutical. Despite the numerous benefits of chia, only a few studies have been conducted on the biosynthesis of terpenoids in this plant. As a result, there is limited knowledge on how terpenes are produced in chia. Therefore, it is essential for future research to focus on understanding the enzymes responsible for terpene production in chia; for example, by exploring interspecific gene variation between multiple Salvia spp. that are known to produce a variety of terpenes (Schmiderer et al., 2023).

Multiple studies have been conducted to investigate the gene expression in different tissues of chia. However, comparing the qualitative and quantitative data of these studies is often difficult owing to the use of different chia populations, genotypes and treatments. To address this issue, a set of seven internal reference genes has been proposed to normalize the transcripts precisely and to be used as controls across various developmental stages and in different treatments. The reference genes mentioned are derived from already published datasets. These genes can be helpful in overcoming the challenges associated with comparing gene expression data from different studies (Gopalam et al., 2017). The reference genes include glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and cyclophilin (CYP), which can be used as controls for cold and heat stress treatments; clathrin adaptor complex (CAC) and serine/threonine-protein phosphatase 2A (PP2A), which can be used in studies of reproductive development, such as flower formation; and eukaryotic translation initiation factor 3E (ETIF3E), alpha tubulin (α-TUB) and PP2A, which are suitable for studies of the vegetative stages of development (Gopalam et al., 2017).

Characterization of key genes involved in PUFA biosynthesis in chia

In addition to transcriptomic studies, some research has explored the molecular mechanisms behind oil biosynthesis in chia by investigating genes involved in FA desaturation (Xue et al., 2017a, 2018, 2021; Gopalam et al., 2021). In plants, certain enzymes known as FA desaturases are responsible for synthesizing monounsaturated fatty acids and PUFAs by introducing double bonds into the acyl chains. This process is catalysed by two types of enzymes: stearoyl-acyl carrier protein (ACP) desaturases (SADs) and fatty acid desaturases (FAD) enzymes, including FAD2, FAD3, FAD4, FAD5, FAD6, FAD7 and FAD8 (Xue et al., 2018, 2023; Yang et al., 2022). The role of FADs is crucial in maintaining the integrity of cell membranes by introducing unsaturated double bonds into the carbon chains of fatty acids. This process modulates the physical properties of lipid membranes. The unsaturated FAs produced by FADs play a key role in regulating membrane fluidity and modulating membrane function (Prasertthai et al., 2022).

The regulation of FAD genes depends on specific promoters based on their function and cellular localization, and their transcription can be influenced by environmental factors, such as temperature, light, wounding, hormones and pathogen attack (Ahmadizadeh et al., 2020). For example, SAD and ω-6 FA desaturases (FAD2 and FAD6) are crucial enzymes for membrane lipid biosynthesis and are therefore expressed constitutively (Hajiahmadi et al., 2020). According to Zare et al. (2024), the Sh-SAD gene family in chia has expanded, probably as a response to various stresses that impact membrane lipids. However, certain membrane lipids require additional synthesis at a specific developmental stage. For instance, the expansion of cotyledons in Arabidopsis plants necessitates the incorporation of additional PUFAs into thylakoid membranes for photosynthesis (Wang et al., 2017).

The biosynthetic pathways of PUFAs in higher plants are well understood through the study of Arabidopsis mutants, which lack membrane-bound FA desaturase activity at different stages of lipid metabolism (Ohlrogge and Browse, 1995). The process of converting SA to OA is regulated by SAD genes, which encode a stearoyl-ACP desaturase enzyme, whilst the conversion of OA to LA is controlled by FAD2 in the ER (Xue et al., 2017a) and by FAD6 in the plastid (Xue et al., 2021). Further desaturation of LA to ALA is catalysed by FAD3 in the ER and by FAD7 and FAD8 in the chloroplast (Fig. 4; Xue et al., 2018). In chia, a unique tandem array of Sh-SAD genes alters the production of OA, which could plausibly explain the high content of ω-3 FAs in its seeds (Zare et al., 2024).

Fig. 4.

Lipid biosynthesis in oleaginous plants. The lipid biosynthesis begins with the de novo synthesis of fatty acids in plastids. Acetyl-CoA is formed through the dehydrogenation of pyruvate, which is carboxylated by acetyl-CoA carboxylase, forming malonyl-CoA. Subsequently, the malonyl-CoA:ACP malonyltransferase (MCAT) transfers the acyl-carrier proteins (ACP) to the malonyl moiety to provide the carbon donor for fatty acid biosynthesis through malonyl-ACP. Fatty acid biosynthesis involves condensation and elongation reactions, catalysed by 3-ketoacyl-ACP III (KASS III) and ketoacyl-ACP synthase I (KAS I). These reactions result in the production of C16:0 and C18:0 acyl-ACPs in plastids. The C16:0-ACP can be hydrolysed further by acyl-ACP thioesterase B (FATB) or elongated by ketoacyl-ACP synthase II (KASS II) to form C18:0-ACP, which can be desaturated to C18:1-ACP by stearoyl-ACP desaturase (SAD). Desaturation of C18:1-ACP in plastids occurs via oleoyl-ACP thioesterase (FATA) and plastidial desaturase enzymes FAD6 (ω6-desaturase) and FAD7/8 (ω3-desaturase). After fatty acid biosynthesis, free fatty acids move into the cytosol, where they are converted by long-chain acyl-CoA synthetase (LACS) into acyl-CoA. This provides building blocks for glycerolipid synthesis in the endoplasmic reticulum (ER)). The glycolysis pathway begins with the conversion of dihydroxyacetone-3-phosphate (DHAP) to glycerol-3-phosphate (G3P) by glycerol-3-phosphate dehydrogenase (GPDH) in the ER, followed by two acylation reactions at the sn-1 and sn-2 positions of G3P. These reactions are catalysed by glycerol-3-phosphate acyltransferase (GPAT) and lysophosphatidic acid acyltransferase (LPAAT) at the sn-1 and sn-2 positions. Phosphatidic acid (PA) undergoes further acylation to form diacylglycerol (DG). DG is then used for biosynthesis triacylglycerol (TG) or as a precursor to produce the membrane lipid phosphatidylcholines (PC). TGs are formed by the esterification of DG, which is catalysed by an enzyme called diacylglycerol: acyl-CoA acyltransferase (DGAT). TG is then transported from the ER to oil bodies for storage. Phosphatidylcholines (PCs) can be further desaturated into two types of fatty acids: C18:2-PC (ω-6) and C18-3PC (ω-3). Two microsomal fatty acid desaturase enzymes, Δ12 oleate desaturase (product of FAD2) and ω-3/Δ15 oleate desaturase (product of FAD3), respectively, are responsible for this conversion process. Abbreviations: DAG-CPT, diacylglycerol cholinephosphotransferase; EA, enoyl-ACP reductase; HAD, hydroxyacyl-ACP dehydrase; KAR, ketoacyl-ACP reductase; LPCAT, lysophosphatidylcholine:acyl-CoA acyltransferase; PDCT, phosphatidylcholine:diacylglycerol cholinephosphotransferase; PDH, pyruvate dehydrogenase; PLA2, phospholipase A2. The figure was created with BioRender.com.

A comparison of the evolutionary lineage of prokaryotic and plastidial Δ12 oleate desaturases revealed that they are the ancestors of the ω-3 desaturases (also known as Δ15 desaturases, when counting the location of the double bond from the carboxyl end of the carbon chain). This suggests that the Δ12 oleate desaturase (or ω-6 desaturase when counting the location of the second double bond from the methyl end of the carbon chain) produces LA as a substrate for the ω-3 desaturase by converting OA to LA in the chloroplast (López Alonso et al., 2003). The phylogenetic relationship of Δ15 (ω-3) and Δ12 oleate (ω-6) membrane-bound desaturases in chia and other commercially important plants is shown in Fig. 5. Although the lipid biosynthesis pathways and lipidome in Arabidopsis have been well studied (Li-Beisson et al., 2013; Kehelpannala et al., 2021), the genetic mechanisms that drive the abundant production of PUFAs and their accumulation in chia seeds are only beginning to be understood. Recent studies shed light on FA biosynthesis and the underlying genetic mechanisms that govern the process (Wang et al., 2022; Gupta et al., 2023; Xue et al., 2023; Zare et al., 2024).

Fig. 5.

Phylogenetic relationships of ω-3 (Δ15) and ω-6 (Δ12) membrane desaturase gene families in higher plants. Sequence alignment was performed with ClustalW, and the phylogenetic tree was reconstructed by the neighbour-joining (NJ) method with bootstrap analysis for 1000 repetitions using the MEGA v.10 (Kumar et al., 2018). The circular cladogram was annotated with iTOL v.5.7 (Letunic and Bork, 2021). Species included in this analysis are as follows: Arabidopsis thaliana (At-FAD); Arachis hypogea L. (Ah-FAD); Brassica juncea (Bj-FAD); Brassica napus (Bn-FAD); Brassica rapa (Br-FAD); Brachypodium distachyon (Bd-FAD); Camelina sativa (Cs-FAD); Carthamus tinctorius (Ct-FAD); Coffea arabica (Ca-FAD); Cynara cardunculus (Cc-FAD); Glycine max (Gm-FAD); Gossypium hirsutum (Gh-FAD); Helianthus annuus (Ha-FAD); Linum usitatissimum (Lu-FAD); Nicotiana tabacum (Nt-FAD); Olea europaea (Oe-FAD); Oryza sativa (Os-FAD2); Perilla frutescens (Pf-FAD); Persea americana (Pa-FAD); Ricinus communis (Rc-FAD); Salvia hispanica (Sh-FAD); Salvia splendens (Ss-FAD); Sesamum indicum (Si-FAD); Solanum lycopersicum (Sl-FAD); Vitis vinifera (Vv-FAD); Zea mays (Zm-FAD). Accession numbers are depicted in the tree and chia Sh-FADs are labelled in red.

ω-6 fatty acid desaturases (FAD2 and FAD6)

In one of the few studies conducted, it was confirmed that both isoforms of the chia Sh-FAD2 genes (Sh-FAD2-1 and Sh-FAD2-2) encode for a Δ12 oleate desaturase, which is the primary enzyme responsible for the conversion of ω-9 to ω-6 FAs (Xue et al., 2017a). The Sh-FAD2 proteins are homologous to the sesame Si-FAD2-1 (86 % on average for both isoforms of Sh-FADs) and Arabidopsis At-FAD2 (77 %). Both isoforms of Sh-FAD2 share 93 % amino acid sequence identity and have an equivalent length of 383 amino acids (Xue et al., 2017a). The FAD2 gene was first discovered in Arabidopsis by Okuley et al. (1994). Later studies have revealed that the gene contains four isomers, namely At-FAD2-1, At-FAD2-2, At-FAD2-3 and At-FAD2-4, which differ in their biological function, localization and expression pattern, reviewed by Dar et al. (2017). The FAD2 genes are primarily active in reproductive tissues, such as flowers, and at all developmental stages of seeds, with mid-maturation stages displaying a higher level of expression when compared with early or late stages. These genes are only weakly active in vegetative tissues, as observed in several plant species, such as cucumber (Dong et al., 2016), perilla (Xue et al., 2017b), sunflower (Celik Altunoglu et al., 2018), Camelina (Ahmadizadeh et al., 2020) and olive (Razeghi-Jahromi et al., 2022).

In cotton, FAD2-1 is a seed-specific desaturase that controls PUFA biosynthesis in immature seeds and developing flower buds (Liu et al., 2017). The FAD2-2 plays a significant role in controlling the synthesis of ALA in olive fruits and is expressed at various growth stages of cotton (Salimonti et al., 2020; Razeghi-Jahromi et al., 2022). Likewise, FAD2-3 and FAD2-4 are expressed constitutively throughout the cotton plant (Liu et al., 2017). In chia plants, Xue et al. (2017a) found that Sh-FAD2-1 is expressed in most organs, whereas Sh-FAD2-2 is expressed mainly in reproductive organs and young seeds. The Sh-FAD2-1/2 genes contain a predicted FADs domain and a PLN02505 conserved domain belonging to the membrane FADs-like protein domain superfamily (Cao et al., 2021). An analysis of the amino acid sequences of membrane-bound FADs from higher plants has revealed that they contain three conserved histidine box motifs: [HX_{(3 or 4)}H], [HX_{(2 or 3)}HH] and [HX_{(2 or 3)}HH] (Cao et al., 2021). The Sh-FAD2-1 and Sh-FAD2-2 proteins have six and four transmembrane helices, respectively. They also have three conserved histidine box motifs and eight histidine residues in their conserved domains (Xue et al., 2017a). These residues are crucial for FA desaturase activity because they act as di-iron ligands (Cao et al., 2021).

Plants, being immobile, face various environmental challenges, such as abiotic and biotic stresses. To combat these stresses and maintain membrane fluidity, plants have developed a response mechanism involving the accumulation of unsaturated FAs through altered FAD activity. For instance, At-FAD2 regulates the lipid composition of intracellular membranes and ensures proper functioning of membrane proteins under salt stress (Nguyen et al., 2019). In the case of mechanically damaged olive fruits, Oe-FAD2 is responsible for controlling the levels of LA and PA in the mesocarp (Luisa Hernández et al., 2020). Likewise, in avocado plants infected with the fungus Colletotrichum gloeosporioides, FAD2 is responsible for controlling the lipid levels in the fruits (Wang et al., 2004). In chia, the expression of Sh-FAD2-1 and Sh-FAD2-2 under various abiotic stresses, such as cold, heat, salt and drought, and in response to hormones, such as methyl jasmonate (MeJA), salicylic acid (SA) and abscisic acid (ABA), has been studied (Xue et al., 2017a).

Both isoforms of Sh-FAD2 exhibited similar expression levels in response to cold (upregulation) and heat (downregulation). Upon treatment with MeJA and ABA, the Sh-FAD2 genes were upregulated, whereas they were downregulated after wounding. In the cases of SA, drought and salinity treatments, the expression of Sh-FAD2 genes was initially increased, followed by a continuous decrease over the time studied (Xue et al., 2017a). Xue et al. (2021) found that chia plants exposed to low temperatures (4 °C) exhibited increased expression of the plastid Sh-FAD6 gene, similar to the response observed for the Sh-FAD2 genes studied under cold treatment (Xue et al., 2017a). However, when exposed to higher temperatures (38 °C), the expression of Sh-FAD6 was downregulated (Xue et al., 2021), and the expression of Sh-FAD2 was primarily inhibited (Xue et al., 2017a).

ω-3 fatty acid desaturases (FAD3 and FAD7/8)

Three members of the ω-3 desaturase gene family were isolated from chia seedlings, including two distinct isoforms of Sh-FAD3-1/2, a single copy of Sh-FAD7a/b (two distinct alleles) and Sh-FAD8 (Xue et al., 2018). In silico analysis of the deduced amino acid sequences of chia ω-3 desaturases revealed the lack of a C-terminal ER retrieval signal (-KSKIN) in Sh-FAD isoforms (i.e. Sh-FAD3-1/2). However, an N-terminal chloroplast transit peptide in Sh-FAD7a/b and Sh-FAD8 was identified (Xue et al., 2018). Evolutionary studies and phylogenetic analysis have shown that Sh-FAD3-1 and Sh-FAD3-2 group together with ER-localized ω-3 desaturases, such as Oe-FAD3, Si-FAD3-1 and those from other species, such as Arabidopsis, sunflower and cocoa. This suggests that Sh-FAD3-1 and Sh-FAD3-2 are also localized in the ER. Conversely, Sh-FAD7a/b and Sh-FAD8 cluster with chloroplast-localized Si-FAD7/8 and Oe-FAD7, as in other species, such as wild peanut, chestnut and sunflower, indicating that these genes are also localized in the chloroplast (Xue et al., 2018). Chia ω-3 desaturases (i.e. Sh-FAD3-1/2, Sh-FAD7a/b and Sh-FAD8) are predicted to have an uncharacterized DUF3474 domain associated with the FA desaturase domain, in addition to three histidine boxes (HDCGH, HRTHH and HVI(V)HH) (Xue et al., 2018). These histidine boxes play a crucial role in Δ15 linoleate desaturase activity (Cao et al., 2021).

After comparing the homologous protein sequences, it was found that Sh-FAD3-1 and Sh-FAD3-2 are closely related to perilla Pf-FAD3a/b (78.4 and 80.3 %, respectively) and At-FAD3 (62.0 and 62.6 %, respectively). Additionally, Sh-FAD7a/b and Pf-FAD7a/b (79.8 and 80.1 %, respectively) and Sh-FAD8 and Pf-FAD8a/b (78.4 and 73.3 %, respectively) shared a high similarity. However, a lower similarity was observed between Sh-FAD7a/b and Sh-FAD8 with At-FAD7 and At-FAD8 (63 and 66 %, respectively) (Xue et al., 2018). This analysis was followed by heterologous expression of Sh-FAD3-1 and Sh-FAD3-2 in yeast and subsequent protein activity assays, which showed that Sh-FAD3-1 and Sh-FAD3-2 possess Δ15 linoleate desaturase activity and convert ω-6 to ω-3 FAs (Xue et al., 2018).

In higher plants, FAD3 genes are expressed in leaves and during all stages of seed development. However, certain FAD3 genes are reported to be seed specific, probably linked to the synthesis and accumulation of ALA in seed oils (Rani et al., 2018; Yeom et al., 2020). Transcripts of Sh-FAD7a/b are highly expressed in flowers and stems and, to a lesser extent, in mature seeds, whereas Sh-FAD8 is expressed ubiquitously in all organs except for mid- and late-stage seeds (Xue et al., 2018). Notably, FAD8 expression is induced by cold treatment, and FAD8 is not expressed at optimal growth temperatures (Soria-García et al., 2019). Chia ω-3 desaturase genes show disparate responses to cold and heat stress. When exposed to cold, the expression of Sh-FAD3-1 and Sh-FAD3-2 increased, whereas the expression levels of Sh-FAD7a/b and Sh-FAD8 remained unchanged. In contrast, drought stress initially led to an increase in the expression of all ω-3 desaturase genes, which was followed by a continuous decrease. However, under heat stress, only Sh-FAD3-1 and Sh-FAD8 displayed a similar pattern of expression (Xue et al., 2018). Chia ω-3 desaturase genes were significantly downregulated by salinity, wounding, ABA and MeJA treatments, except for a short-term peak of increased expression after MeJA application. This suggests that ω-3 desaturases might play a role in acclimatization to environmental stress (Xue et al., 2018).

Further research is needed to gain a better understanding of the biological role of chia ω-6 and ω-3 desaturases in the biosynthesis of PUFAs. Specifically, the interaction of the ER-localized Sh-FAD3-1 and Sh-FAD3-2 and the plastidial Sh-FAD7a/b and Sh-FAD8 and their role in determining the amount of ω-3 in membrane lipids is still not fully understood. Molecular approaches are needed to study the links between these functionally related genes in FA and lipid biosynthesis in chia. The findings from these studies could lead to a better understanding of the underlying molecular mechanisms of oil accumulation and the development of biotechnological tools to improve the ALA contents of oil crops.

FUTURE PERSPECTIVE

Over the past two decades, the scientific community has researched the chia plant and its seeds extensively, to determine their nutraceutical properties. Studies have shown that chia seeds contain a high amount of ω-3 PUFAs, proteins and phenolic compounds with antioxidant activity. These nutraceutical compounds have been associated with significant benefits for human health, leading researchers to shift their focus from analysing chemical compounds to uncovering the metabolic pathways and molecular mechanisms behind their biosynthesis. However, the lack of adequate genetic resources, tools and stable transformation protocols has significantly hindered progress in understanding molecular and genetic characteristics of chia.

Recently, the genome of chia has been sequenced successfully by researchers (Wang et al., 2022; Gupta et al., 2023; Zare et al., 2024). This sequence information, which includes a high-quality reference genome available at NCBI Datasets (Zare et al., 2024), will serve as a valuable genomic resource and accelerate the progress of molecular studies in chia. Recent advances in precise gene-editing technologies, such as CRISPR (Ali and Zhang, 2023), have significantly advanced our understanding of gene functions in various oil crops. Once adopted, these technologies provide an essential foundation for mutant studies in chia and will enable genetic manipulation for gene function studies. In the future, the establishment of these techniques, along with stable transformation protocols for chia, might allow for the creation of a chia mutant collection that can be used as a community resource for functional analysis of chia genes, as has been done for other plants, such as Arabidopsis, rice, maize, sorghum, Brachypodium and Camelina (Hayta et al., 2019; Hus et al., 2020; Rönspies et al., 2022; Thakur et al., 2022; Wang et al., 2023; Cai et al., 2024).

Chia producers face a significant challenge in increasing their crop yield while dealing with different and often difficult environmental conditions. It is essential to gain an understanding of how gene expression and regulation work to engineer crops that interact well with the environment and produce optimal yields. However, the regulation of chia genes at the transcriptional level is not well understood. Creating a publicly available web-based database, which includes large-scale transcriptomic comparisons in various conditions, in different tissues and using the tools available for model plant species, can collect such information effectively and accelerate the development of genomic breeding techniques. Developing an optimal lipid extraction technique and a comprehensive lipid database similar to the one created for Arabidopsis (Kehelpannala et al., 2020, 2021) can help with metabolomic/lipidomic studies on chia. This could provide additional insight into the production and enrichment of ω-3 FAs in chia seeds.

These tools and resources can be applied to exploit the genetic make-up of chia and translate its characteristics to other crops. The knowledge gained from studies in chia could be applied to other oilseed food crops to engineer them to produce higher quantities or specific types of unsaturated FAs or antioxidant compounds. Genetic tools will also aid in discovering the biological functions of chia seed mucilage, which could open new avenues for the commercial use and application of chia. Chia seed mucilage has shown potential for targeted plant-based drug delivery in pharmaceutical production and for modifying plant–microbe interactions for soil conditioning in commercial agriculture. Although the nutritional value of chia is well established, there is still ongoing debate regarding the medicinal, antimicrobial and antifungal effects of chia seed compounds. However, comprehensive investigations through both in vitro experiments and in vivo animal and controlled human trials are expected to provide greater clarity in the future.

FUNDING

This work was supported by the Research Training Program Scholarship, The Rosewood Foundation Grant, the Alfred Nicholas Fellowship, the Megan Klemm Postgraduate Research Scholarship and the Norma Hilda Schuster (née Swift) Scholarship from the University of Melbourne awarded to T.Z. and the Inaugural Botany Foundation Fellowship 2020 from the University of Melbourne Botany Foundation to B.E.

AUTHOR CONTRIBUTIONS

T.Z.: conceptualization; literature review; visualization; writing—original draft; funding acquisition. A.F.-L.: conceptualization; supervision; writing—review and editing. B.E.: conceptualization; supervision; writing—review and editing. U.R.: conceptualization; supervision; writing—review and editing; project administration; resources; funding acquisition.