HEXOKINASE1 and glucose-6-phosphate fuel plant growth and development Free

As sessile organisms, plants have developed sophisticated biosynthetic mechanisms to create key energetic intermediates, structural compounds and diverse families of secondary metabolites. Plants fuel these pathways through the production of simple sugars, such as glucose, through photosynthesis and the Calvin cycle. These sugars are then modified and broken down by families of enzymes in subsequent metabolic steps. Hexokinases (HXKs) are a conserved family of glycolytic enzymes and sugar sensors that catalyze the phosphorylation of hexoses, such as glucose and fructose (Meyerhof, 1927). This catalytic activity is crucial for the initiation of glycolysis and other biosynthetic pathways. In 1953, Saltman described the first plant hexokinase in Triticum aestivum (Saltman, 1953). The HXK family in Arabidopsis thaliana is comprised of six members, three being catalytically active (HXK1, HXK2, HXK3) and three being catalytically inactive related proteins, termed Hexokinase-Like (HKL1, HKL2, HKL3). However, only HXK1 has been studied extensively in Arabidopsis. Similarly, HXK families have been identified in a variety of other plant species, but full characterization has been limited (Aguilera-Alvarado and Sanchez-Nieto, 2017).

In Arabidopsis, the first HXK1 mutants were identified in a screen for glucose insensitive (gin) mutants that develop normally on media containing 6% glucose, a ‘high glucose’ condition that causes chlorosis (the insufficient production of chlorophyll) and growth defects in wild-type plants (Fig. 1A) (Jang et al., 1997; Moore et al., 2003). The gin2-1 and gin2-2 mutants were found to contain nonsense and missense mutants within HXK1, respectively. Gin2-1 and gin2-2 had no phenotype compared with wild-type when grown on non-glucose media, demonstrating that HXK1 function is specifically linked to high glucose conditions (Moore et al., 2003). Although both gin2-1 and gin2-2 mutant plants displayed improved growth on 6% glucose relative to wild-type plants, the gin2-1 allele was chosen for further examination due to several factors. When compared with gin2-1 plants, the gin2-2 allele displayed a weaker reduction of HXK1 transcripts and retained some growth restriction and chlorosis in high glucose conditions (Moore et al., 2003). Furthermore, gin2-1 plants had highly reduced catalytic activity (hexose phosphorylation) of HXK1 and levels of phosphorylated hexoses, as determined by spectrophotometric assays (Moore et al., 2003). Subsequently, the gin2-1 allele was used for complementation experiments that showed that glucose sensitivity could be restored by expressing HXK1 that still lacked the catalytic activity. This led to the discovery that HXK1 could act as a glucose sensor independently of its catalytic activity. In Arabidopsis, and now other plant species, it has been shown that the glucose-sensing activity of HXK1 acts within signaling pathways that repress the expression of genes involved in photosynthesis, such as the CHLOROPHYLL A/B BINDING (CAB; also known as LHCB2, AT3G27690) and CARBONIC ANHYDRASE (CAA; CA2, AT5G14740) in the presence of high glucose concentrations, explaining the chlorosis phenotype of mutant plants grown on 6% glucose media (Cho et al., 2009; Kim et al., 2013, 2016; Moore et al., 2003; Veramendi et al., 2002; Wu et al., 2023).

Recent studies of HXKs have largely been limited to their role as glucose sensors; however, the catalytic activity of these proteins likely plays a key role in plant development. Although HXK1 is capable of phosphorylating other hexoses, glucose-6-phosphate (G6P) is the primary output of HXK1 activity and an important precursor for many branches of plant metabolism (Fig. 2). For example, to drive plant growth and energy production, G6P may proceed through the catalytic steps of glycolysis and the tricarboxylic acid (TCA) cycle. This process produces electron carriers needed for ATP production during oxidative phosphorylation and generates precursors for lipid, amino acid and hormone synthesis. Lipids serve as important energy storages molecules and are components of cellular membranes and plant cuticle. These lipid-based structures provide structural integrity and environmental protection for the plant. In addition, amino acids are required for the synthesis of proteins, which are necessary for all aspects of plant development, such as cell division, growth and photosynthesis. Currently, little is known about how HXK1 catalytic activity impacts these pathways and plant development. In this Spotlight, we discuss the crucial role of HXK1 in development in Arabidopsis, highlight selected biosynthetic pathways impacted by HXK1 catalytic activity and examine the different roles of HXK1-dependent signaling and catalysis.

As we discuss below, plants containing HXK1 mutant alleles and transgenics that overexpress HXK1 display a wide array of developmental defects (Fig. 1A-D). These studies suggest that HXK1 expression and activity must be tightly controlled to promote canonical plant development. In addition, these experiments suggest that both HXK1 signaling and catalytic activity regulate plant development. However, further work is required to fully disentangle these different functions of HXK1 and explain the connections to plant development.

In addition to having reduced sensitivity to high glucose conditions, gin2-1 plants display a variety of other developmental phenotypes, such as reduced growth and delayed senescence, attributed to loss of HXK1 signaling (Fig. 1C) (Moore et al., 2003). To examine gin2-1 plants at later developmental stages, high light intensities (300 μmol/m²/s) were used to mimic high glucose conditions. This approach is effective because plants use light to produce sugars through photosynthesis and the Calvin cycle; therefore, high light intensities result in increased internal glucose production within plants (Fernandez et al., 2017; Proietti et al., 2023). Thus, high light intensities can be experienced as ‘high-energy’ conditions by plants, similar to high exogenous glucose concentrations. Under high light intensities, gin2-1 mutants have reduced cell expansion that results in a severe growth defect; specifically, gin2-1 mutant plants display reduced root growth, delayed senescence, and reduced flower and silique production when grown under these high light intensities (Fig. 1C). Expressing catalytically inactive HXK1 mutants (HXK1-S177A or HXK1-G104D) in the gin2-1 background rescued the growth defects observed in high light intensities and the glucose insensitivity seen high glucose conditions, confirming that HXK1 signaling can promote plant development independent of its catalytic activity (Fig. 1A) (Moore et al., 2003). Furthermore, recent work reveals that HXKs from Oryza sativa, Nicotiana tabacum, Solanum tuberosum and Fragaria pentaphylla can function as glucose sensors by partially or fully restoring glucose sensitivity in the gin2-1 mutant in Arabidopsis (Cho et al., 2009; Kim et al., 2013, 2016; Veramendi et al., 2002; Wu et al., 2023).

Subsequently, T-DNA insertion alleles of HXK1 have been used in various studies to further characterize the function of the protein. Insertion nomenclature and phenotypes have been summarized elsewhere (Castro et al., 2020). Generally, T-DNA insertion lines with undetected HXK1 transcript and protein display normal growth in high glucose conditions and growth defects of varying severity (Aki et al., 2007; Hsu et al., 2014; Huang et al., 2015; Reda, 2013, 2015; Rottmann et al., 2018; Van Dingenen et al., 2019). Although these phenotypes resemble those of gin2-1 plants, many of these mutant alleles display additional phenotypes in standard or low light intensities (Fig. 1B,D). Because HXK1 signaling is not expected to be active in these ‘low-energy’ conditions, these developmental phenotypes cannot be fully attributed to HXK1 glucose sensing and signaling defects. To address this knowledge gap, further experiments are needed to determine whether catalytically inactive HXK1 mutants can rescue the defects induced by the HXK1 T-DNA insertion alleles. Overall, these studies indicate that the glucose-sensing activity of HXK1 is crucial for the regulation of plant growth, senescence, flowering and seed production in high glucose concentrations and high light intensities; however, further work is required to determine whether this is true for all HXK1 mutant alleles.

Although HXK1 glucose sensing and signaling is clearly important for plant development, recent studies are beginning to investigate importance of HXK1 catalytic activity. Recent work has reported that the T-DNA insertion allele hxk1-4 displays a severe growth defect and decreased leaf cell number and cell size compared with wild-type, without the need for high-light or high-glucose conditions (Fig. 1B) (Van Dingenen et al., 2019). HXK1 glucose sensing and signaling is expected to be most active during high energy conditions (high glucose or high light); thus, other aspects of HXK1 function, such as HXK1 catalytic activity, may be important for growth promotion (Moore et al., 2003; Van Dingenen et al., 2019). To address this uncertainty, future work should determine whether catalytically inactive HXK1 mutants can rescue the defects observed in hxk1-4. This experiment will help characterize the role of HXK1 catalytic activity and glucose sensing in growth regulation, as has been described for gin1-2 (Moore et al., 2003). Previously, it has also been reported that both gin2-1 and the T-DNA insertion allele hxk1-3 have reduced hypocotyl elongation in low light intensities (less than 10-15 µmol/m²/s) compared with wild type (Fig. 1D) (Lincoln et al., 2023 preprint; Moore et al., 2003). Because these conditions would not generate high-glucose conditions required for HXK1 signaling, the authors of a recent study argue that HXK1 enzymatic activity is required for hypocotyl growth (Lincoln et al., 2023 preprint). Strikingly, they observed that G6P treatment rescued the hypocotyl defects observed in low light, confirming that HXK1 catalytic activity promotes hypocotyl elongation (Fig. 1D). Overall, these emerging studies indicate that HXK1 catalytic activity impacts plant development more than previously thought.

Overexpression (OX) of HXK1 in various plant species supports the involvement of HXK1 in a variety of developmental pathways and suggests that HXK1 glucose phosphorylation activity may play a key role. Initial studies in Arabidopsis revealed that HXK1-OX plants were hypersensitive to high glucose concentrations and this was attributed to overactive HXK1 glucose signaling (Fig. 1A) (Jang et al., 1997). However, these experiments were limited to sterile media and it was therefore difficult to investigate the consequences of HXK1-OX at later developmental stages (Kelly et al., 2012). Subsequent experiments have used various HXK1-OX lines to examine the effects of overactive HXK1 catalytic activity by directly inhibiting HXK1 glucose signaling or overexpressing catalytically inactive HXK1 mutants (Dai et al., 1999; Zheng et al., 2021). Interestingly, overexpression of Arabidopsis HXK1 in tomato (Lycopersicon esculentum) raises G6P and fructose-6-phosphate (F6P) levels, blocks vegetative growth and induces premature senescence (Dai et al., 1999). In addition, the growth inhibition of HXK1-OX tomato plants could be rescued in the presence of mannoheptulose, a competitive inhibitor of HXKs. Mannoheptulose is a glucose analog that is unable to be phosphorylated by HXK1. Thus, treatment with this chemical activates HXK1-dependent glucose sensing and signaling while blocking the catalytic activity of the enzyme. The rescued growth in the presence of mannoheptulose indicates that elevated HXK1 hexose phosphorylation activity may be responsible for the observed phenotypes of Arabidopsis HXK1-OX in tomato. Finally, Kelly and colleagues overcame challenges in transgenic selection to identify Arabidopsis plants overexpressing HXK1 with strong developmental phenotypes in mature plants (Kelly et al., 2012). These HXK1-OX plants displayed accelerated senescence, limited photosynthetic rates and growth defects in the absence of exogenous sugars (Fig. 1B) (Kelly et al., 2012). More recently, overexpression studies in rice (O. sativa) revealed that HXK1 is a positive regulator of leaf senescence and reactive oxygen species accumulation (Zheng et al., 2021). In addition, rice plants overexpressing catalytically inactive HXK1 appeared to be phenotypically wild-type, supporting the importance of HXK1 catalytic activity in senescence regulation.

Ultimately, HXK1 overexpression in a variety of plant species suggests that HXK1 catalytic activity is a crucial driver of plant senescence and growth. However, it remains unclear how HXK1 and G6P overproduction could create these developmental consequences. G6P may be converted to a vast array of chemical compounds, so further work is needed to determine whether the overproduction of any of these downstream products results in premature senescence and growth defects (Fig. 2). Therefore, further metabolomic experiments are required to determine whether overactive HXK1 catalytic activity alters metabolite levels. In addition, several hormones, such as salicylic acid and ethylene, are known to promote senescence and are biosynthetically downstream of HXK1 and G6P (Zhang et al., 2021). Thus, we recommend examining the phenotype of HXK1-OX plants when other downstream enzymes, such as ethylene and salicylic acid biosynthesis enzymes, are mutated. These experiments will determine what downstream pathways and enzymes contribute to the phenotypes observed in HXK1-OX plants.

HXK1 catalyzes the phosphorylation of glucose to produce G6P. This is a key gateway step for many biosynthetic pathways that are crucial for plant growth and fitness (Fig. 2). Glucose may progress through glycolysis and the TCA cycle to produce energetic intermediates, such as NAD(P)H, and important families of compounds, including amino acids, fatty acids and starch (Plaxton, 1996; Wiskich and Dry, 1985). Additional biosynthetic pathways may produce hormones, pigments and cell wall polymers that ensure plants can survive challenging environments, grow at ideal times and maintain structural integrity (Endler and Persson, 2011; Smith et al., 2017; Tanaka et al., 2008; Zhong et al., 2019). Although these biosynthetic routes are highly complex, recent studies indicate that HXK1 catalytic activity plays a key role in promoting metabolic homeostasis and, thus, plant development.

Recently, a metabolomic study examined the levels of various metabolites of gin2-1 and wild-type (Landsberg erecta, Ler) plants grown in high light stress (1100-1500 µmol/m²/s) or control conditions (120 µmol/m²/s) (Küstner et al., 2019). Although these high light intensities can act as a ‘high energy’ condition, they also may be perceived as environmental stress. Under control conditions, the authors reported differences in amino acid and carbonic acid levels between gin2-1 and Ler. However, differences in metabolite levels are more striking when gin2-1 and Ler plants are grown under high light intensities. Under high light stress, fructose and proline are enriched in Ler, whereas raffinose, citric acid, aspartic acid, tryptophan and oxaloacetate were enriched in gin2-1. Because these metabolic differences occur specifically in high light stress, these changes could contribute to the growth defects seen in gin2-1 plants grown under high light intensities (Fig. 1C). Specifically, proline is involved in maintaining redox balance and is induced in response to environmental stresses (high salt or water) (Küstner et al., 2019; Szabados and Savouré, 2010). The lack of proline induction in gin2-1 plants may contribute to the light-sensitive response of these mutant plants (Fig. 1C) (Küstner et al., 2019). Further metabolomic experiments indicate that the levels of pyruvate and TCA cycle intermediates are disrupted in gin2-1 plants grown in control conditions (Fürtauer et al., 2019). These datasets suggest that HXK1 regulates the level of several amino acids and TCA cycle intermediates (Fürtauer et al., 2019; Küstner et al., 2019). These molecules are often important precursors for hormones, such as brassinosteroid, jasmonate, abscisic acid, strigolactone, gibberellin, ethylene and cytokinins (Fig. 2). These hormones are crucial regulators of plant growth, senescence and stress responses; however, it is currently unclear whether HXK1 catalytic activity directly impacts the abundance of these important compounds (Smith et al., 2017). Ultimately, none of these metabolic changes have been conclusively linked to the developmental phenotypes observed in HXK1 mutant and HXK1-OX plants, so further experimental work is needed.

G6P can also be converted to 6-phosphogluconolactone by glucose-6-phosphate dehydrogenase (G6PDH) at the start of the pentose phosphate pathway (PPP) (Fig. 2). The PPP produces ribose-5-phosphate, used in the synthesis of nucleotides, and erythrose-4-phosphate, used in the synthesis of aromatic amino acids, pigments (flavonoids), anti-herbivory compounds (glucosinolates) and hormones (auxin and salicylic acid). Pigments, such as flavonols and anthocyanins, provide protection from UV radiation, help recruit pollinators and serve as important signaling molecules. Interestingly, HXK1 glucose signaling activity has been shown to promote anthocyanin production in apple (Malus domestica) (Hu et al., 2016). The authors demonstrated that HXK1 phosphorylates bHLH3, an anthocyanin-associated transcription factor, to promote pigment synthesis. Furthermore, auxin and salicylic acid hormones are crucial regulators of plant growth, senescence and stress responses. In addition, the PPP is an important producer of NAD(P)H and is involved in maintaining active pools of important antioxidants, such as ascorbate and glutathione. Ultimately, the PPP produces a diversity of small molecules and ensures plant fitness during environmental stress; however, more work is needed to connect these pathways to HXK1 catalytic activity and G6P levels.

G6P may also be funneled into the synthesis of various structural and energy storage polymers (Fig. 2). Plant cell walls are primarily composed of pectins, cellulose and hemicellulose polymers, and these structures determine cell morphology and provide structural support. For example, mutations changing the composition or abundance of pectins can result in severe cell adhesion defects or changes in cell and organ morphology (Krupková et al., 2007; Mouille et al., 2007; Saffer, 2018; Saffer et al., 2017). These cell wall polymers are synthesized from a variety of sugars, including glucose and G6P. In addition, G6P may be converted to glucose-1-phosphate by phosphoglucomutase (PGM) to initiate starch production in plants (Börnke and Sonnewald, 2011). Starch is a crucial energy storage polymer, and its breakdown and synthesis are tightly regulated to ensure adequate sucrose availability (Börnke and Sonnewald, 2011; Streb and Zeeman, 2012).

G6P is also directly used in the production of trehalose-6-phosphate (T6P), a crucial signaling disaccharide (Fig. 2). T6P levels are positively correlated with overall metabolic status and energy availability within the plant. T6P can promote plant flowering, and T6P synthesis mutants display delayed flowering attributed to the decrease in T6P signaling (Baena-González and Lunn, 2020; Fichtner and Lunn, 2021; Wahl et al., 2013). T6P signal transduction involves the direct inhibition of Snf1-related protein kinases (SnRKs) (Zacharaki et al., 2022; Zhai et al., 2018) and is connected to the regulation of plant flowering (Baena-González and Lunn, 2020; Baena-González et al., 2007). Interestingly, gin2-1 plants display delayed flowering in high light intensities (Fig. 1C), mimicking the phenotypes of T6P synthesis mutants. However, it is unclear whether T6P levels and signaling are altered in gin2-1 plants. Thus further work is needed to determine whether HXK1 catalytic activity and G6P levels are key modulators of T6P homeostasis and plant flowering.

Previously, we described the growth defects and premature senescence of HXK1-OX plants (Fig. 1B). Although the cause of these phenotypes has been largely unexplored, studies investigating myo-inositol (MI) synthesis suggest that overactive HXK1 catalytic activity and disordered metabolism may promote senescence. The synthesis of MI and its derivatives relies on HXK1 catalytic activity to produce G6P as a precursor (Fig. 2). MI itself is a crucial precursor for cell wall and membrane biosynthesis, as well as important signaling molecules such as inositol 1,4,5-trisphosphate (Chen and Xiong, 2010; Donahue et al., 2010; Ma et al., 2016; Meng et al., 2009). 1L-MYO-INOSITOL-1-PHOSPHATE SYNTHASE (MIPS1) is required for MI synthesis and is a key regulator of plant vegetative growth and development; the mips1 mutant displays stunted vegetative growth and lesions when grown in long-day growth conditions in light intensities above the photosynthetic compensation point (16 h day and 8 h night with light intensity greater than 40 µmol/m²/s) (Meng et al., 2009; Wang et al., 2023 preprint). Recent work revealed that MIPS1 activity and plant growth is controlled by the length of the photosynthetic period and a metabolic day-length measurement system (Wang et al., 2023 preprint). Previously, HXK1 was found to be a regulator of MI homeostasis and programmed cell death and the hxk1 mutant was identified when screening for mutations that suppress the mips1 lesion phenotype (Bruggeman et al., 2015). HXK1 catalytic activity, independent of glucose sensing and signaling, is necessary for the formation of mips1-derived lesions; MIPS2 activity is increased in the double mutant background and can restore MI levels (Bruggeman et al., 2015). However, it is unclear how HXK1 regulates MIPS2 activity and how HXK1 catalytic activity promotes lesion formation, but disordered metabolism and altered glucose levels have been offered as potential explanations (Bruggeman et al., 2015).

HXK1 catalytic activity and G6P production are clearly at the heart of plant metabolism. However, the connections between these biosynthetic pathways and HXK1 developmental phenotypes is not always obvious. Study of MI and MIPS1 suggests that overactive HXK1 catalytic activity induces premature leaf senescence and disrupts plant metabolism. As HXK1 is important for many areas of plant development (Fig. 1), further work is needed to connect these to the diverse metabolic networks HXK1 is involved in.

It is crucial for us to understand the steps and outputs of metabolic systems and we are beginning to learn that these pathways may become harmful – and even deadly – to plant fitness when they are grossly disrupted. Alterations to HXK1 and other metabolic enzymes can lead to cell death and a variety of other developmental defects (Fig. 1) (Dai et al., 1999; Jang et al., 1997; Kelly et al., 2012; Lincoln et al., 2023 preprint; Moore et al., 2003; Van Dingenen et al., 2019; Zheng et al., 2021). However, we are only just beginning to understand how these metabolic disruptions impact plant fitness. Although HXK1-OX Arabidopsis plants display accelerated senescence (Fig. 1B), there has been little study of the factors that contribute to this phenotype (Kelly et al., 2012). Therefore we recommend a thorough investigation of metabolic pathways derived from G6P and senescence-associated plant hormones in HXK1-OX plants. To our knowledge, HXK1 has not been overexpressed in different biosynthesis mutant backgrounds, and we believe these genetic experiments will clarify which pathways and metabolites are involved in HXK1-associated senescence. Here, we have described that HXK1 has clear developmental significance and G6P may be directed to a wide variety of metabolic fates. Therefore, we suggest that HXK1 and G6P are key regulators of plant development and that diverse metabolic pathways are involved in supporting plant growth. However, there are still gaps in our understanding of plant metabolism and development, so we believe it important to consider how metabolic pathways may fuel hormone production, gene expression and plant growth.

In addition, HXK1 uniquely functions in both glucose sensing and signaling, and as a catalytic enzyme in metabolism, so more thorough exploration is needed to distinguish between these diverging functions. Emerging studies suggest that these dual functions of HXK1 impact distinct areas of plant development (Fig. 3). Furthermore, other glycolytic enzymes have characterized non-enzymatic functions in humans and other mammals; however, this phenomenon is not well studied in plants (Rodríguez-Saavedra et al., 2021). Because these enzymes are crucial elements of plant biosynthetic pathways, we suggest that it is increasingly important to distinguish between their proposed signaling and catalytic activities. Furthermore, it is currently unclear precisely what glucose concentrations or light intensities induce HXK1 signaling pathways. This gap of knowledge makes it difficult to disentangle the effects of HXK1 signaling and catalytic activity. Therefore, we recommend examining CAB and CAA gene expression levels when gin2-1 plants are grown on a variety of glucose concentrations and light intensities. These investigations will help us understand and predict plant molecular responses to environmental conditions and connect these changes to crucial developmental outcomes.

We apologize to colleagues whose work was not cited due to space limitations.