Regulatory and retrograde signaling networks in the chlorophyll biosynthetic pathway

ABSTRACT

Plants, algae and photosynthetic bacteria convert light into chemical energy by means of photosynthesis, thus providing food and energy for most organisms on Earth. Photosynthetic pigments, including chlorophylls (Chls) and carotenoids, are essential components that absorb the light energy necessary to drive electron transport in photosynthesis. The biosynthesis of Chl shares several steps in common with the biosynthesis of other tetrapyrroles, including siroheme, heme and phycobilins. Given that many tetrapyrrole precursors possess photo‐oxidative properties that are deleterious to macromolecules and can lead to cell death, tetrapyrrole biosynthesis (TBS) requires stringent regulation under various developmental and environmental conditions. Thanks to decades of research on model plants and algae, we now have a deeper understanding of the regulatory mechanisms that underlie Chl synthesis, including (i) the many factors that control the activity and stability of TBS enzymes, (ii) the transcriptional and post‐translational regulation of the TBS pathway, and (iii) the complex roles of tetrapyrrole‐mediated retrograde signaling from chloroplasts to the cytoplasm and the nucleus. Based on these new findings, Chls and their derivatives will find broad applications in synthetic biology and agriculture in the future.

This review summarizes the regulatory mechanisms that underlie chlorophyll synthesis, including (1) factors that control the activity and stability of tetrapyrrole biosynthesis enzymes, (2) transcriptional and post‐translational regulation of the tetrapyrrole biosynthesis pathway, and (3) the complex roles of tetrapyrrole‐mediated retrograde signaling from chloroplasts to the cytoplasm and the nucleus.

INTRODUCTION

The dynamic character of chlorophyll (Chl) metabolism, including its biosynthesis and degradation, plays important roles in photosynthetic organisms. This is not only reflected in nature and agriculture, but also finds applications in modern biotechnology and medicine (Kräutler, 2008; Solymosi and Mysliwa‐Kurdziel, 2017; Simkin et al., 2022; Martins et al., 2023; Sun et al., 2024). Apart from carotenoids and phycobilins, Chls and bacteriochlorophylls (BChls) are the most prominent and essential photosynthetic pigments found in plants, algae, cyanobacteria and other oxygenic and anaerobic photosynthetic bacteria. They harvest light energy, transfer excitation energy to reaction centers (RCs), and the so‐called special pair enables the charge separation that drives the photosynthetic electron transport chain (Grimm, 2019; Bryant et al., 2020; Willows et al., 2023). In addition, Chl homeostasis protects photosynthetic protein complexes from photo‐oxidation, mediates detoxification, and enables appropriate responses to stress (Dogra et al., 2018; Liu et al., 2024). Furthermore, Chl degradation in perennial plants, which is manifested in the splendid color changes of the leaves in the fall, is also vital for the replacement of damaged proteins of the photosystems and ensures the ripening of fruits and the abscission of other sink organs, such as wheat ears (Kuai et al., 2017; Li et al., 2017; Koyama, 2018; Kanojia et al., 2021; Kapoor et al., 2022).

Chls and BChls absorb light energy and transfer it to the RCs of photosynthetic protein complexes. Among these, chlorophyll a (Chl a), bacteriochlorophyll a (BChl a), and BChl g are the most common species that participate in the light‐driven charge‐separation process in the RCs (Chen et al., 2020, 2023a; Suga and Shen, 2020; Xu et al., 2021a). In addition to these, there are several other types of Chls and BChls, such as Chls b, c, d, and f (see Figure 1) and the BChls b, c, d, e, and g. The structure, function and regulation of the BChls in anoxygenic phototrophs are not discussed in this review, and we refer the readers to the recent literature on these topics (see, e.g., Chew and Bryant, 2007; Zappa et al., 2010; Saer and Blankenship, 2017; Bryant et al., 2020; Yang et al., 2021).

Figure 1.

The aerobic chlorophyll biosynthesis pathway

Overall reactions from Protoporphyrin IX (PPIX) to Chlorophyll (Chl) a and other Chls. The chemical changes in the molecular structures are marked in red at each step. 2OG, 2‐oxoglutarate; ADP, adenosine diphosphate; ATP, adenosine triphosphate; NADP⁺, nicotinamide adenine dinucleotide phosphate; NADPH, the reduced form of NADP⁺; Pi, inorganic phosphate; SAH, S‐adenosyl‐l‐homocysteine; SAM, S‐adenosyl‐l‐methionine. *The enzymatic steps for which non‐homologous isozymes are known in different phototrophs. ^†ChlF‐containing PSII (super‐rogue PSII), instead of an isolated ChlF subunit, catalyzes the formation of Chl f (Trinugroho et al., 2020).

Land plants make use of Chls a and b, while aquatic photosynthetic organisms, such as algae and cyanobacteria, employ a wider range of Chl variants (a, b, c, d, and f). The diversity of Chl types in marine organisms allows them to efficiently harvest light energy across a broader spectrum, which is advantageous in an environment in which light availability and quality can vary significantly with depth and turbidity (Croce and van Amerongen, 2014; Kume et al., 2018). Chlorophylls a, b, and c were first isolated and described in the nineteenth century (Stokes, 1864; Sorby, 1873). Chlorophyll a is considered the most abundant variant of Chls (Björn et al., 2009). Chlorophyll b ,with a formyl group on ring B, is the second most abundant Chl and is found in the light‐harvesting systerm, but not in the RCs (Melkozernov and Blankenship, 2006). Chlorophyll a, with absorption peaks at 430 and 662 nm, is present in all photosynthetic organisms, while Chl b exhibits absorption maxima at 457 and 646 nm. Besides plants, some cyanobacteria such as Prochloron also utilize Chl b in their photosystems (Scheer, 2006). Chlorophyll c is widely distributed in chromophyte algae (Zapata et al., 2006; Green, 2011; Büchel, 2020). Relative to Chls a and b, Chl c absorbs more blue light (around 446 nm) and less red light. It has therefore been proposed to facilitate energy transfer from blue‐green‐absorbing carotenoids to Chl a in the light‐harvesting apparatus (Zapata et al., 2006). Most Chl c species are not esterified to a hydrophobic phytol chain, but contain a free carboxyl group, and it has been suggested that the lack of the phytyl chain allows for a stronger binding of fucoxanthin to the light‐harvesting (LHC) proteins (Wang et al., 2019). Chlorophyll d is a far‐red‐absorbing primary photosynthetic pigment of cyanobacteria, with a maximal absorbance peak at around 687 nm in methanol; in Acaryochloris marina that harbors Chl d as the major Chl, its in vivo absorption peaks at around 716 nm (Miyashita et al., 1996, 1997; Kühl et al., 2005). Chlorophyll f is the most recently discovered Chl, and was first isolated from stromatolites in 2010, before it became apparent that it is, in fact, more widely distributed in cyanobacteria (Chen et al., 2010; Gan et al., 2014a, 2014b). It is by far the most red‐shifted Chl, with a Qy peak of 707 nm. In vivo, Chl f can absorb photons with a wavelength of up to 790 nm (Chen, 2014). Chl d has been established to be present in the RC, whereas recent studies have identified only antenna sites for Chl f in photosystems (Gisriel et al., 2020; Kato et al., 2020; Gisriel, 2024; Shen et al., 2024).

Once plastids had formed by endosymbiosis, their subsequent development depended on both nucleus‐encoded and chloroplast‐encoded gene products (Richly et al., 2003; Waters and Langdale, 2009). More than 90% of the chloroplast‐localized proteins are now encoded in the nucleus, synthesized in the cytosol as precursor proteins, and ultimately imported into the chloroplasts (Kim et al., 2022). In land plants (like Arabidopsis thaliana), all proteins that take part in tetrapyrrole biosynthesis (TBS) are encoded in the nucleus (Tanaka et al., 2011). Expression, import and redistribution of TBS proteins within the chloroplast are strictly regulated to ensure the correct synthesis of Chls (Kobayashi and Masuda, 2016; Wang et al., 2022; Gao et al., 2023). In addition, chloroplasts generate retrograde signals that control nuclear gene expression in response to developmental cues and/or environmental perturbations (Chi et al., 2013; Chan et al., 2016; Jan et al., 2022). This review discusses our current understanding of the metabolism, regulation, and signaling pathways involved in the Chl biosynthetic pathway in eukaryotes, especially land plants.

BIOSYNTHESIS AND BREAKDOWN

Overview of Chl biosynthesis and breakdown in plants

In plants, Chl is synthesized via the magnesium (Mg) branch of TBS (Yaronskaya and Grimm, 2006). The common precursor of tetrapyrroles is 5‐aminolevulinic acid (ALA), which is synthesized from glutamyl‐tRNA^Glu by the actions of glutamyl‐tRNA reductase (GluTR) and glutamate 1‐semialdehyde aminotransferase (GSA) (Ilag et al., 1994). Protoporphyrin IX (PPIX), the common precursor of Chls and heme, is derived from ALA via a series of six enzymatic reactions (Warren and Scott, 1990). The biosynthesis of Chl is initiated when Mg²⁺ is inserted into the PPIX instead of the ferrous iron (Fe²⁺) required for heme biosynthesis. This step is catalyzed by the enzyme Mg‐chelatase (MgCh), which consists of the subunits CHLI (acting as an ATPase), CHLD and CHLH (the substrate‐binding, catalytic subunit, also referred to as “genomes uncoupled 5” (GUN5)), and is the first committed enzyme in Chl biosynthesis (Shen et al., 2006; Tanaka and Tanaka, 2007). MgCh‐mediated formation of Mg‐protoporphyrin IX (MgPPIX) is enhanced by the positive regulator genomes uncoupled 4 (GUN4), a porphyrin‐binding protein (Larkin et al., 2003; Davison et al., 2005; Peter and Grimm, 2009; Chen et al., 2015; Richter et al., 2023). The subsequent reaction catalyzes the conversion of MgPPIX into MgPPIX monomethyl ester (MgPME), which requires magnesium‐protoporphyrin methyl transferase (CHLM) and the methyl donor, S‐adenosyl‐l‐methionine (SAM) (Gibson et al., 1963; Shepherd et al., 2003). The enzyme Mg‐ProtoIX monomethylester (oxidative) ring cyclase (known as CHL27 in Arabidopsis and CRD1/CTH1 in Chlamydomonas) then catalyzes the formation of 3,8‐divinyl protochlorophyllide (DV Pchlide a), and the resulting fifth ring (E ring) is accompanied by the transition to a green color. This structural change in turn creates an absorption peak at 630 nm. This is a crucial step in the eventual formation of Chl a, which exhibits strong absorption at 665 nm in methanol (Moseley et al., 2000; Tottey et al., 2003). DV Pchlide is then reduced to chlorophyllide (Chlide) at the C17=C18 double bond by the action of protochlorophyllide oxidoreductase (POR) (Apel et al., 1980). Subsequently, the 8‐vinyl group on the B ring is reduced by the NADPH‐dependent 3,8‐divinyl (Chlide) 8‐vinyl reductase (DVR) (Nagata et al., 2005). Finally, the Chl synthase (CHLG) esterifies Chlide with activated C20 isoprenoid alcohol molecules, normally phytyl pyrophosphate (PhyPP), leading to Chl a directly, or geranylgeranyl pyrophosphate (GGPP), leading to GG‐Chl a. Geranylgeranyl reductase (GGR/CHLP) can reduce GGPP to generate PhyPP, and can also reduce GG‐Chl a to ultimately form Chl a (Addlesee et al., 1996; Keller et al., 1998). Chlorophyll a oxygenase (CAO) accepts Chlide a and catalyzes its conversion to Chlide b (Tanaka et al., 1998; Oster et al., 2000) (Figure 1). Rather than being a one‐way process, Chls a and b can be interconverted via the “Chl cycle” (Ito et al., 1996; Tanaka and Tanaka, 2011). In the reverse pathway, Chl b is converted into Chl a through two consecutive enzymatic steps, which are catalyzed by Chl b reductase (CBR) (Kusaba et al., 2007; Sato et al., 2009) and 7‐hydroxymethyl Chl a reductase (HCAR) (Meguro et al., 2011), respectively.

The ripening of fruits, the senescence of plants, and stress responses in algae are all accompanied by Chl catabolism (Jiang et al., 2011; Guo et al., 2021; Dong et al., 2023; Chen et al., 2023b). Chlorophyll turnover is mediated by two pathways, which include several branches. The Chl salvage pathway involves Chl dephytylation by the enzyme chlorophyll dephytylase (CLD1), before Chlide is recycled by Chl synthase (Lin and Charng, 2021). In citrus fruits and in Chenopodium album, chlorophyllase (CLH) shows dephytylation activity in vitro (Tsuchiya et al., 1997; Jacob‐Wilk et al., 1999). Although loss‐of‐function mutants of Arabidopsis CLH genes show no defects in Chl catabolism during senescence (Schenk et al., 2007), it has been demonstrated that CLH plays a role in Chl dephytylation during the repair process of Photosystem II (PSII) in Arabidopsis (Tian et al., 2021). CLH has dephytylation activities on Chl a, Chl b, and pheophytin a (Phein a) (Lin and Charng, 2021). In contrast, CLD isoforms, which have been characterized by genetic studies in Arabidopsis and Synechocystis, are involved in the dephytylation of Chl, specifically under heat stress conditions in plants. Notably, CLD proteins cannot use Phein a as a substrate (Lin et al., 2016; Takatani et al., 2022). Moreover, an alpha/beta hydrolase with esterase activity (VTE7) has also been proposed to hydrolyze Chl (Albert et al., 2022). The phytol derived from Chl breakdown was shown to be subsequently salvaged for tocopherol biosynthesis. Activation of phytol is catalyzed by successive phosphorylation mediated by the kinases VTE5 and VTE6 (Valentin et al., 2006; Almeida et al., 2016).

In addition to the Chl salvage cycle, a second Chl degradation pathway operates in senescing leaves as well as stressed green leaves. Here, Mg‐dechelatase catalyzes the removal of the central Mg²⁺ from the Chl structure, thus triggering the formation of Phein a. Knockout of the Mg‐dechelatase‐encoding gene results in a stay‐green (SGR) phenotype (Ren et al., 2010; Shimoda et al., 2016). In continuation of this Chl catabolic pathway, pheophytinase (PPH), another homolog of CLD1, has been shown to function as a dephytylase of Phein in Arabidopsis (Schelbert et al., 2009). The enzyme pheophorbide a oxygenase (PAO) cleaves the porphyrin ring of Pheide a to generate an oxidized red Chl catabolite (RCC), which is subsequently converted into a primary fluorescent Chl catabolite (pFCC) by RCC reductase (Ginsburg et al., 1994; Hörtensteiner et al., 2000). Finally, pFCC is modified, transported into the vacuole, and isomerized to yield a non‐fluorescent product (Kreuz et al., 1996).

Isozymes in Chl biosynthesis

Although this review emphasizes Chl metabolism in plants, it is worth mentioning that several steps in Chl biosynthesis are catalyzed by evolutionarily unrelated isoenzymes across different phototrophic lineages. These instances have been extensively summarized in recent literature (Raymond and Blankenship, 2004; Fujita et al., 2015; Li and Bridwell‐Rabb, 2019). In the following section, we will discuss three representative cases (labeled with asterisks in Figure 1) to illustrate the reasons behind the utilization of unique enzymes and the potential evolutionary trajectory of these enzymes from their ancestral forms.

Oxygen has significantly influenced the evolution of the Chl biosynthetic pathway. Initially, photosynthesis evolved in anaerobic environments, but the emergence of oxygenic phototrophs eventually led to an oxygen‐rich atmosphere. This shift in environmental conditions prompted a transition from oxygen‐independent (or even oxygen‐sensitive) enzymes to their oxygen‐tolerant (or even oxygen‐requiring) counterparts. A prime example of this transition is the formation of the E ring from MgPME toward DV Pchlide a. In this step, the oxygen‐dependent cyclase, which was initially named AcsF (for “aerobic cyclization system Fe‐containing subunit”) in Rubrivivax gelatinosus, has homologs in cyanobacteria (CycI/II /ChlAI/ChlAII) (Minamizaki et al., 2008; Peter et al., 2009), green algae (CRDI/CTHI) (Moseley et al., 2000, 2002) and higher plants (CHL27) (Tottey et al., 2003). An oxygen‐independent cyclase called BchE has also been described in anaerobic phototrophic organisms (Bollivar et al., 1994; Yamanashi et al., 2015). The oxygen‐dependent cyclases function as monooxygenases that use molecular oxygen to form an oxo group at the C13 position of the tetrapyrrole. In reconstituted systems, AcsF/CHL27 requires ferredoxin, which is provided by NADPH and ferredoxin:NADP⁺ reductase, indicating that the cyclase is linked to the photosynthetic electron transport chain (Herbst et al., 2018). BchE is derived from the radical SAM protein family, and catalyzes this cyclization reaction via a cobalamin‐dependent step, in which an oxygen atom is incorporated that is derived from water (Gough et al., 2000).

Another well known case involving isozymes is the reduction of the tetrapyrrolic D‐ring by POR, a crucial step in the Chl biosynthetic pathway (Figure 1). There are two distinct ways to achieve this reduction, which require either a light‐dependent or a dark‐operative POR (abbreviated as LPOR or DPOR) (Heyes et al., 2006). Dark POR consists of the subunits ChlL, ChlN, and ChlB, which show similarities to the subunits of the nitrogenases. In recent years, structural biology has played an important role in elucidating their catalytic mechanisms. One striking example is the visualization of an extensive hydrogen‐bonding network involving the active‐site residues of Arg38, Lys42, and Asp63 in LPOR, which reveals the molecular mechanism of light‐driven Pchlide reduction (Zhang et al., 2019). The overall structure of LPOR is similar to that of other members of the SDR family, and exhibits a typical dinucleotide‐binding Rossman fold (Tonfack et al., 2011). From an evolutionary perspective, DPOR first appeared in anoxygenic phototrophs. It is sensitive to oxygen and can be inactivated during daylight by the O₂ produced by PSII. Following the transition to oxygenic phototrophs, the oxygen‐insensitive LPOR evolved to become the primary active form in the presence of light (Chernomor et al., 2021). During the primary endosymbiotic event, the LPOR gene(s) was transferred to the eukaryotic nucleus; while the gene(s) encoding the DPOR subunits remains in the plastid genome (Vedalankar and Tripathy, 2019). Anaerobic photosynthetic bacteria contain DPOR, whereas cyanobacteria, green algae, red algae, and most gymnosperms possess both LPOR and DPOR. Diatoms and angiosperms (flowering plants) only have LPOR (Chen, 2014; Vedalankar and Tripathy, 2019).

Most Chls have an ethyl group at the C8 position, which is generated by the reduction of a vinyl group. Studies on this step have underscored the significance of enzyme promiscuity in the evolution of substrate or reaction alterations. In this context, there are three evolutionarily unrelated 8‐vinyl reductases, one of which is the NADPH‐dependent 8‐vinyl reductase (N‐DVR). This enzyme is found in plants, some marine cyanobacteria, and anoxygenic photosynthetic bacteria (Ito and Tanaka, 2014). A second enzyme, ferredoxin‐dependent 8‐vinyl reductase (F‐DVR), is predominantly found in freshwater cyanobacteria and anoxygenic photosynthetic bacteria. Most photosynthetic organisms contain either N‐DVR or F‐DVR, but diatoms retain both enzymes. N‐DVR exhibits a higher substrate specificity for 3,8‐divinyl Chlide. In contrast, cyanobacterial F‐DVR possesses a broad substrate specificity, and can convert 3,8‐divinyl Pchlide, 3,8‐divinyl Chlide, and 3,8‐divinyl Chl a into their ethyl derivatives. Notably, it was proposed that HCAR, involved in the Chl cycle, specifically the conversion from Chl b to Chl a, may have evolved from F‐DVR. This hypothesis stems from the demonstration of a marginal HCAR activity in cyanobacterial F‐DVR when tested in vitro (Ito and Tanaka, 2014). A third 8‐vinyl reductase has been discovered in Rhodobacter capsulatus. Its activity is attributed to an alternative activity of a chlorophyllide a oxidoreductase (COR), which serves to reduce the C8‐vinyl group during BChl biosynthesis. Chlorophyllide a oxidoreductase is another nitrogenase‐like enzyme that is closely related to DPOR and was initially reported for its activity to reduce the C7=C8 double bond of Chlide a to form 3‐vinyl BChlide a (Nomata et al., 2006; Tsukatani et al., 2013).

The transition from an anoxic to an oxic world had profound consequences for the development of early life forms. At the molecular level, one major change involved the evolution of enzymes that are either tolerant to oxygen or dependent upon it (Raymond and Blankenship, 2004). The development of aerobic, light‐dependent enzymes and isozymes depended on the use of different cofactors, and ultimately resulted in a novel form of biochemistry driven by the evolution of oxygenic photosynthesis. In addition, it has been observed that diatoms apparently lack a homolog of the AcsF protein or the BchE protein (Nymark et al., 2013). Recent discussions have also focused on the possible existence of so far unidentified Chl c synthase isoenzymes in brown algae (Jiang et al., 2023b; Jinkerson et al., 2024), which suggests that some isoenzymes in photosynthetic eukaryotes still remain to be discovered.

Biosynthesis of Chl c, Chl d, and Chl f

In photosynthetic organisms, various light‐harvesting antenna complexes have been characterized that are able to exploit different wavelengths and intensities of light in different environments. Light‐harvesting systems are composed of core and peripheral antenna complexes, and Chl b, Chl c, and Chl f are regarded as accessory pigments in the antenna systems of photosynthetic organisms. Generally speaking, these pigments serve to absorb energy and transfer it to the photosynthetic RC, although they do not take part in the charge‐separation (photo‐oxidation) step of the photosynthetic electron transport chain. Conversely, both Chl d and Chl a are capable of functioning not only within the light‐harvesting complexes, but also as part of the special pair in the photosynthetic RC (Allakhverdiev et al., 2016).

Chl c is commonly found in members of the Chromista group (Yoon et al., 2002; Zapata et al., 2006), which encompasses a wide range of eukaryotic organisms, including various algal lineages. Within the Chromista groups, certain algae have obtained plastids via secondary endosymbiosis. This has resulted in combinations of characteristics derived from different origins, including diverse Chl biosynthesis pathways that may have originated from either the host's nuclear genome, the nuclear genome of the engulfed alga, the chloroplast genome of the engulfed alga, or even via horizontal gene transfer (HGT) from other species (Muñoz‐Gómez et al., 2021). Indeed, a recent study has identified a 2‐oxoglutarate (2‐OG)‐dependent dioxygenase protein (CHLC) in diatoms that converts Chl precursors into Chl c (Jiang et al., 2023b). Interestingly, brown algae also possess Chl c, but seem to lack CHLC‐like genes in their genomes (Jiang et al., 2023b; Jinkerson et al., 2024), suggesting an alternative evolutionary pathway distinct from that seen in diatoms.

Chlorophyll d is the primary Chl in the cyanobacterium Acaryochloris marina. Chlorophyll d features a formyl group substitution on the C3 position, like that found on Chl a (Loughlin et al., 2013). Using isotopic labeling methods, Schliep et al. proposed that the oxygen in the C3 formyl group of Chl d results from an oxygenase‐type reaction (Schliep et al., 2010). However, it is noteworthy that the “Chl d synthase” protein remains unknown in identity.

Chl f was only recently discovered in cyanobacteria, and differs from Chl a in having a formyl group on ring A, instead of the methyl group found in Chl a. This pigment enables oxygenic photosynthesis to take place in far‐red light (700–800 nm) (Chen et al., 2010). Chlorophyll f synthesis was found to depend on a paralog of PsbA (D1), named ChlF, in cyanobacteria that exhibit far‐red light photo‐acclimation (FaRLiP) (Gan et al., 2014b; Ho et al., 2016; MacGregor‐Chatwin et al., 2022). Moreover, it has been found that ChlF can substitute for D1 structurally, and form modified PSII complexes that are capable of producing Chl f in cyanobacteria without FaRLiP. Intriguingly, further research has demonstrated that the mutation of just two amino acid residues (Met127 and Gly128) in D1 is sufficient to convert oxygen‐evolving PSII into a Chl f synthase (Trinugroho et al., 2020).

FaRLiP cyanobacteria typically utilize not only Chl f, but also a low amount of Chl d to use far‐red light energy (Gan et al., 2014b; Nürnberg et al., 2018; Elias et al., 2024b). However, the Chl d biosynthesis pathway in these organisms is also unknown, and it is still unclear whether the synthesis of Chl d is catalyzed by the same enzymes or different ones in these cyanobacteria compared with Acaryochloris marina.

TRANSCRIPTIONAL REGULATION OF THE TBS PATHWAY

All of the TBS genes in angiosperms (like A. thaliana), except those that code for DPOR subunits (Tanaka et al., 2011), are encoded in the nucleus. In angiosperms, Chl biosynthesis is suppressed in the dark, as the conversion of protochlorophyllide (Pchlide) to chlorophyllide a (Chlide a) by LPOR requires light and the accumulation of photoreactive Pchlide must be avoided (Santel and Apel, 1981; Zhang et al., 2019). Light, phytohormones and the status of the chloroplasts are all involved in regulating TBS gene expression (Hills et al., 2015; Müller and Munné‐Bosch, 2021). Meanwhile, several families of transcription factors, including Golden‐like1/2 (GLK1/GLK2), GATA nitrate‐inducible carbon metabolism‐involved (GNC) and cytokinin‐induced GATA1/GNC‐like (CGA1/GNL), function as key players downstream of light and phytohormone signaling to regulate chloroplast development and Chl biosynthesis (Kobayashi and Masuda, 2016; Cackett et al., 2021; Schwechheimer et al., 2022) (Table 1). Thus, an elaborate and complex transcriptional regulatory network controls Chl biosynthesis (Figure 2).

Table 1.

Summary of the key TBS transcriptional regulators discussed in the text

Transcription factor or regulator	Targets	Cis‐elements	Interacting factors	Model of regulation	References
PIFs	GUN5, CAO, PORC	G‐box	DELLA	Repression	Cheminant et al. (2011); Toledo‐Ortiz et al. (2014)
	PORA, PORB	Indirect		Induction	Leivar et al., 2009; Shin et al., 2009; Cheminant et al., 2011
	HEMA1, URO1, URO2, PPO1, CHLI1, CHLI2, GUN4, CHLM, CHL27, DVR, PBGD, GSA1,GSA2	Indirect		Repression	Leivar et al., 2009; Shin et al., 2009
RVE1	PORA	EE‐box		Induction	Xu et al., 2015
	PORB, PORC, GSA2, GUN5, CRD1	Indirect		Induction	Xu et al., 2015
HDA15	GUN5	G‐box	PIF3	Repression	Liu et al., 2013
	CHLD	Indirect		Repression	Liu et al., 2013
BRM	PORC	G‐box	PIF1	Repression	Zhang et al., 2017
	PORA, PORB	Indirect		Repression	Zhang et al., 2017
HY5	URO2, PPO1, GUN5, GUN4, CHL27, DVR, PORC, CAO, CHLP, GUN2	G‐box		Induction	Lee et al., 2007; Toledo‐Ortiz et al., 2014
	HEMA1, PORB	Indirect		Induction	Zhang et al., 2024
FHY3	HEMB1	FBS motif	PIF1	Induction	Tang et al., 2012
	HEMA1	Indirect		Induction	McCormac and Terry, 2002
GLK1/GLK2	HEMA1, GUN5, GUN4, CHLM, CHL27, CAO, PORA, PORB, PORC, DVR	CCAATC		Induction	Waters et al., 2009
	CHLI1, CHLI2, CHLD	Indirect		Induction	Bastakis et al., 2018
GNC	URO1, CHLM, LCAA/YCF54	GATA motif		Induction	Xu et al., 2017
	GUN4, GUN5, CHLI1, CHLI2, CHLD, CHL27, DVR, PORB, PORC, GUN2, GUN6, DVR, GSA2, HEMA1, HEMA2,CHLP	Indirect		Induction	Bastakis et al., 2018
GNL	GUN5, CHLD, GUN4, DVR, GUN2, CHLI2 HEMA1, CHLI1, CHL27, CAO	GATA motif		Induction	Xu et al., 2017; Bastakis et al., 2018
		Indirect		Induction	Bastakis et al., 2018
EIN3	PORA, PORB	EBS‐box		Induction	Zhong et al., 2009
	HEMA2, GSA2, CPO1, PBGD, CHLG, CPO3, HO4, ALAD2, HO3	Indirect		Induction	Zhong et al., 2009
	UROS, HEMA3, CAO, CPO2, FC1	Indirect		Repression	Zhong et al., 2009
BZR1	HEMA2, HEMB1, UROS, PPO1, PPO2, CPO2	The motif is undefined	PIF4	Repression	Sun et al., 2010; Oh et al., 2012; Wang et al., 2020b
	GUN4, CHLH, DVR, URO2, CHLG, CHLI1, CHL27, LCAA, CHLM, PORA, PORB, PBGD, URO1, GSA1, CHLP, CPO1, FC2	Indirect		Repression	Sun et al., 2010; Oh et al., 2012; Wang et al., 2020b
DELLA	PORA, PORB, PORC	Indirect	PIF1	Induction	Cheminant et al., 2011
	CHLM, CAO, GUN4, CHL27, GUN5	Indirect		Induction	Cheminant et al., 2011
ARR10/12	HEMA1	AGATATG		Induction	Cortleven et al., 2016
SCL27	PORC	G(A/G)(A/T)AA(A/T)GT	DELLA	Repression	Ma et al., 2014

Note: The direct target genes were retrieved from previous studies or chromatin immunoprecipitation sequencing (ChIP‐seq) data. The indirect targets were derived from qPCR or RNA‐seq data described in previous studies.

Figure 2.

The transcriptional regulatory network that controls tetrapyrrole biosynthesis (TBS) levels in response to light and hormonal signals

Light and hormones co‐regulate TBS gene expression. GLK1/GLK2 (Golden‐Like 1/2), GNC (GATA nitrate‐inducible carbon metabolism‐involved) and GNL (cytokinin‐induced GATA1/GNC‐like) directly activate TBS genes in response to light and hormones. The relative levels of GLK1/2, GNC and GNL are strictly regulated at both the transcriptional and protein levels by light and hormonal signaling factors. These include negative interactions with phytochrome‐interacting factors (PIFs), RVE1 (Reveille1), BPG4 (BRZ‐insensitive‐pale green (BPG)), and BZR1 (Brassinazole Resistant 1) in the dark, and positive contacts with HY5 (Elongated Hypocotyl 5), ARFs (Auxin Response Factors), and ARR10/12 (B‐type Arabidopsis response regulators) in the light. Under dark conditions, COP1 and PIFs play key roles in inhibiting the transcription of TBS genes. Conversely, PIFs recruit EIN3 (Ethylene insensitive 3), DELLA (the negative regulator of GA), SCL27 (miR171‐targeted Scarecrow‐like protein) and some chromatin‐remodeling elements, such as HDA15 (Histone Deacetylase 15) and BRM (an SWI2/SNF2 chromatin‐remodeling ATPase), to inhibit the transcription of most TBS genes, but activate the expression of PORA and PORB. Conversely, COP1 inhibits the transcriptional activation of TBS genes by HY5 by promoting its degradation. Upon illumination, the photoreceptors phytochrome (PHY) and cryptochrome (CRY) trigger the activity of HY5 by inhibiting COP1. HY5 then activates the transcription of GLK1/GLK2, GNC, and GNL and most TBS genes; FHY3 (Far‐red elongated Hypocotyl 3) directly activates the transcription of HEMB1. In addition, the auxin factors (ARFs) and the CK factor ARR10/12 induce the transcription of GLK1/GLK2, GNC, and GNL to promote chlorophyll (Chl) synthesis. The direct targets of these factors are shown in Figure 2, and other indirect targets are listed in Table 1.

Light‐induced transcriptional regulation

Light‐regulated transcription of genes involved in Chl synthesis is suppressed in the dark and stimulated in the presence of light. In Arabidopsis, light is perceived by photoreceptors, including phytochromes (phyA–phyE), cryptochromes (CRY1/CRY2), UVB‐resistance 8 (UVR8), phototropins (phot1/phot2) and the LOV‐domain‐containing F‐box protein (ZTL/FKF/LPK2) (Paik and Huq, 2019). phyB/phyA and CRY1/CRY2 play key roles in regulating chloroplast development, owing to their ability to sense red/far‐red light and blue light, respectively (McCormac and Terry, 2002; Tepperman et al., 2006; Stephenson and Terry, 2008; Legris et al., 2019). In darkness, the basic helix–loop–helix (bHLH) transcription factors known as phytochrome‐interacting factors (PIFs) constitutively accumulate to inhibit photomorphogenesis (Leivar and Quail, 2011; Pham et al., 2018). CONSTITUTIVE PHOTOMORPHOGENIC 1 (COP1), a RING‐type E3 ubiquitin ligase, forms multiple complexes with SUPPRESSOR of PHYTOCHROME A (SPA), and functions as another key negative regulator of photomorphogenesis by degrading ELONGATED HYPOCOTYL 5 (HY5) (Deng et al., 1992; Osterlund et al., 2000; Lau and Deng, 2012; Pham et al., 2018). HY5 is a pivotal transcription factor in the regulation of light‐associated events including photomorphogenesis and chloroplast development (Bae and Choi, 2008; Xiao et al., 2022; Zhang et al., 2024). In the presence of light, the activated phyB/phyA physically interact with PIFs and lead to their phosphorylation, ubiquitination, and ultimately to their degradation by the 26S proteasome (Bauer et al., 2004; Shen et al., 2005; Paik and Huq, 2019). Together, phyB/phyA and CRY1/CRY2 also inhibit COP1/SPA activity by directly interacting with SPAs, or by promoting the degradation of SPA2, which results in the accumulation of HY5 (Balcerowicz et al., 2011; Lian et al., 2011; Lu et al., 2015; Sheerin et al., 2015).

In darkness, the mRNA levels of TBS genes in the pifq mutant (a quadruple mutant in which four genes PIF1, PIF3, PIF4, and PIF5 are disrupted) are significantly increased, suggesting that PIFs play an important role in inhibiting Chl synthesis (Huq et al., 2004; Leivar et al., 2009; Shin et al., 2009). When considering the mode of action of PIFs in the regulation of Chl synthesis, the following findings should be taken into account: (i) PIFs directly or indirectly inhibit the expression of key TBS genes such as HEMA1, CHLH, GUN4, and others, and promote the expression of PORA in the dark (Cheminant et al., 2011; Liu et al., 2013; Toledo‐Ortiz et al., 2014). (ii) Phytochrome‐interacting factors collaborate with COP1 to promote the degradation of HY5, thus weakening the activation of TBS genes. Furthermore, PIFs themselves antagonistically interact with HY5 to regulate target genes presumably in the same G‐box‐containing region (Chen et al., 2013; Toledo‐Ortiz et al., 2014; Zhu et al., 2015). (iii) Phytochrome‐interacting factors regulate TBS gene expression at the chromatin level by interacting with histone deacetylase 15/19 (HDA15/HDA19) and an SWI2/SNF2 chromatin‐remodeling ATPase known as the BRAHMA (BRM) complex (Liu et al., 2013; Zhang et al., 2017; Guo et al., 2023). Combined with COP1, PIFs have been shown to modulate the expression of CHLH and CAB (encoding Chl a/b‐binding proteins) by mediating the rapid repositioning of these gene loci within the nucleus (Feng et al., 2014). (iv) Phytochrome‐interacting factors function as a key hub that integrates light‐dependent and hormonal signals by interacting with factors such as ethylene insensitive 3 (EIN3) (see Integrated transcriptional regulation by light and hormones). In addition to PIFs, the transcription factor REVEILLE1 (RVE1) directly binds to the PORA promoter to promote its expression, but inhibits the expression of GSA2, CHLH and CRD1 in the dark (Xu et al., 2015).

Upon illumination, the flow of Chl precursors resumes as PIFs are degraded and positive factors, like HY5, accumulate. Chromatin immunoprecipitation sequencing (ChIP‐seq) data have shown that many TBS genes, such as URO2 and CHLH, are directly targeted by HY5 (Lee et al., 2007). HY5 directly binds to the G‐boxes in the promoters of CHLH and PORC, and triggers the expression of HEMA1 to enhance Chl synthesis (Toledo‐Ortiz et al., 2014). HY5 activity is also modified by hormone signals, and HY5 itself controls hormone biosynthesis and signaling (Gangappa and Botto, 2016) (see Integrated transcriptional regulation by light and hormones). In addition to HY5, the transcription factor FAR‐RED ELONGATED HYPOCOTYL 3 (FHY3) binds directly to the promoter of the HEMB1 gene that codes for 5‐aminolevulinic acid dehydratase, and that of HEMA1, to regulate Chl synthesis (McCormac and Terry, 2002; Ouyang et al., 2011; Tang et al., 2012; Wang et al., 2016). Two enhancers of polycomb‐like proteins, EPL1A and EPL1B (EPL1A/B, known as subunits of the NuA4‐type histone acetyltransferase complex in A. thaliana) were found to function redundantly in Chl biosynthesis. The EPL1A/B‐dependent transcription and H4K5Ac are enriched for genes involved in TBS (Zhou et al., 2022). It was recently reported that the CRY2–SPA1 complex undergoes light‐induced liquid–liquid phase separation (LLPS), which results in a condensate that harbors the METTL16‐type m⁶A writer Fiona 1 (FIO1) (Jiang et al., 2023a). FIO1‐specific m⁶A methylation, in turn, enables the translation of mRNAs encoding at least six Chl homeostasis regulators (CHRs), whose functions are described in the following: (i) The cysteine synthase isomer CYSC1 (AT3G61440) is involved in β‐cyanoalanine biosynthesis and cyanide detoxification. (ii) TRR14 (AT4G10300), an RmlC‐like member of the cupin superfamily, participates in the regulation of photorespiration and osmotic stress responses. (iii) The plastid metabolite transporter SAMT1 (AT4G39460) mediates the transport of S‐adenosylmethionine (SAM) into chloroplasts. (iv) The tryptophan synthase beta‐subunit TSB1 (AT5G54810) is required for the synthesis of phytohormones, such as auxin. (v) Zeaxanthin epoxidase ABA1 (AT5G67030) is required for the synthesis of phytohormones, such abscisic acid. (vi) The chloroplast compartmentation component REC1 (AT1G01320) codes for the protein REDUCED CHLOROPLAST COVERAGE 1 (Jiang et al., 2023a).

Integrated transcriptional regulation by light and hormones

GLK1 and GLK2 were first identified in maize, and they belong to the Golden2, ARR‐B, and PSR1 (GARP) transcription factor family (Langdale and Kidner, 1994; Hall et al., 1998). Most TBS genes, including HEMA1, CHLH, GUN4, CAO, PORA, PORB, and PORC, are directly targeted by GLK1/GLK2 (Waters et al., 2009). The pale‐green phenotype of the glk1 glk2 double mutant is associated with smaller chloroplasts, but the glk1 glk2 mutant shows no obvious defects in leaf development, which suggests that both proteins act specifically upon chloroplast development (Fitter et al., 2002; Waters et al., 2008, 2009). GNC and GNL belong to the GATA transcription factor family. Chromatin immunoprecipitation sequencing data have shown that URO1, CHLM, and LCAA are targeted by GNC, while GNL interacts with CHLH, CHLD, GUN4, DVR, and GUN2 (Xu et al., 2017; Bastakis et al., 2018). The gnc gnl double mutant synthesizes less Chl than the wild‐type plant, while overexpression of either GNC or GNL in Arabidopsis or rice leads to increased Chl content, suggesting that GNC and GNL positively regulate Chl synthesis (Richter et al., 2010; Chiang et al., 2012; Hudson et al., 2013; Kobayashi et al., 2013; Lee et al., 2021). In summary, GLK1/GLK2, GNC and GNL play key roles in regulating Chl biosynthesis and chloroplast development (Schreiber, 2004; Zubo et al., 2018).

This is also evident during plant de‐etiolation in response to both light and hormone signaling. Expression of GLK1/GLK2, GNC and GNL is inhibited by PIFs in the dark (Richter et al., 2010; Pfeiffer et al., 2014; Song et al., 2014; Martín et al., 2016). HY5 interacts with GLKs and binds to GLK promoters to activate their expression in Arabidopsis (Zhang et al., 2024). Moreover, the activities of GLK1/GLK2, GNC, and GNL are also regulated by hormone signaling. For example, the expression of GNC and GNL is strongly induced by cytokinins (CKs) and plays a crucial role in CK signaling (Naito et al., 2007). However, the functions of hormones in Chl metabolism during leaf senescence and fruit ripening differ from those observed during de‐etiolation. This is particularly true in the case of ethylene (ET), as discussed in detail in previous reviews (Guo et al., 2021; Lei et al., 2023). Here, we focus on three groups of hormones that act in different ways to regulate Chl biosynthesis.

(1) The first set consists of brassinosteroids (BR), gibberellin (GA), and ET, which primarily inhibit the development of chloroplasts in the dark. BRs play a central role in this process. Defects in BR biosynthesis and BR signaling (as shown in the BR biosynthesis mutants det2 and dwf4, for example) lead to increased TBS gene expression and higher levels of Chl precursors in the dark (Chory et al., 1991; Clouse et al., 1996; Szekeres et al., 1996; Azpiroz et al., 1998; Li and He, 2016). Conversely, BR‐mediated inhibition of Chl synthesis is achieved in several ways. (i) The transcription factors BRASSINAZOLE RESISTANT 1 (BZR1) and BZR2/BES1 form homodimers to inhibit GATA2, GATA4, and GLK1/GLK2 expression (Luo et al., 2010; Yu et al., 2011). (ii) BZR1/BES1 interacts with PIF4 to inhibit the expression of GLK1/GLK2 and TBS genes (Oh et al., 2012; Wang et al., 2020c). (iii) COP1 promotes the degradation of the phosphorylated form of BZR1 (inactive form) in the dark, which in turn leads to the accumulation of active, non‐phosphorylated BZR1 (Kim et al., 2014). In addition, the COP1–SPA complex interacts directly with PIF3 to block BIN2‐mediated phosphorylation and degradation of PIF3 (Ling et al., 2017). (iv) Under illumination, HY5 inhibits BR signaling by interacting with BZR1 and inhibiting its activity, which in turn stabilizes the BIN2 kinase, a negative regulator of BRs (Li and He, 2016; Li et al., 2020b). (v) More recent studies have shown that BIN2 activates chloroplast development by phosphorylating GLK1 in the light, and BRZ‐INSENSITIVE‐PALE GREEN 4 (BPG4) interacts with GLK1 to inhibit its transcriptional activity in response to BR and light signals (Zhang et al., 2021a; Tachibana et al., 2024). Green light was shown to promote hypocotyl elongation via BR signaling (Hao et al., 2023). It will be of interest to investigate how green light regulates Chl biosynthesis.

The ET‐inducible transcription factor ETHYLENE INSENSITIVE 3 (EIN3) upregulates PORA and PORB by directly binding to their promoters (Zhong et al., 2009). EIN3 also interacts with PIF3 and interdependently regulates the expression of many TBS genes (Liu et al., 2017). COP1 degrades the EIN3‐BINDING F‐BOX 1/2 FACTOR (EBF1/2) in the dark, which enhances the stability of EIN3, while PHYB promotes the interaction between EBF1/2 and EIN3 to accelerate the degradation of EIN3 in the light (Zhong et al., 2014; Shi et al., 2016a; 2016b). GA promotes skotomorphogenesis in the dark by negatively regulating DELLA‐mediated repression of the activity and stability of PIFs, which leads to inhibition of the accumulation of Chl precursors (de Lucas et al., 2008; Feng et al., 2008; Li et al., 2016). DELLA also upregulates PORA and PORB expression in the dark, independently of PIFs (Cheminant et al., 2011). In addition, DELLA physically interacts with the miR171‐targeted Scarecrow‐like protein (SCL27) and inhibits its ability to bind to the PORC promoter (Ma et al., 2014). DELLA‐induced PORA/B expression in darkness makes an important contribution to the suppression of photobleaching upon subsequent light exposure.

(2) The second group of phytohormones includes CK and auxin, and mainly promotes Chl synthesis and photosynthesis in the light. CK accelerates Chl production by promoting the formation of ALA and the light‐dependent conversion of Pchlide into Chlide (Kusnetsov et al., 1998; Yaronskaya et al., 2006; Cortleven and Schmülling, 2015; Cortleven et al., 2016). B‐type Arabidopsis response regulators (B‐type ARRs), like ARR10 and ARR12, act as transcription factors and regulate early CK‐responsive genes, like HEMA1 (Argyros et al., 2008; Cortleven et al., 2016). Auxin participates in the regulation of almost all aspects of plant growth and development, including chloroplast development (Salazar‐Iribe and De‐la‐Peña, 2020). One positive effect of auxin on Chl biosynthesis has been reported in the shoots of many species, like tomato, tea, etc. (Yuan et al., 2018, 2019; Khan et al., 2019; Liu et al., 2020; Zhou et al., 2020). In roots, however, auxin represses Chl biosynthesis by inhibiting the activities of HY5 and GLK1/GLK2. In this case, the auxin‐responsive transcription factors ARF7 and ARF19 repress GLK2, HY5, and GNC/CGA1 expression, which results in the inhibition of Chl synthesis (Kobayashi et al., 2012; Richter et al., 2013).

(3) The third group of hormones, including jasmonate (JA), salicylic acid (SA), and abscisic acid (ABA), mainly come into play under stress conditions. The nuclear‐targeted SIGMA FACTOR‐BINDING PROTEIN 1 (SIB1) functions as a defense‐related transcriptional coregulator by interacting with GLK1/2 in response to an increase in the SA content of leaves (Lv et al., 2019). Under such stress conditions, lesion‐simulating disease 1 (LSD1) and SIB1 antagonistically regulate GLK activity to fine tune the expression of TBS genes (Li et al., 2022a). The role of ABA in modulating plant growth and stress responses is directly linked to its function in regulating chloroplast biogenesis, including the number of chloroplasts, and the expression of photosynthesis genes (Rock et al., 1992; Galpaz et al., 2007; Fujita et al., 2011; Yamburenko et al., 2015). ABI4 and HY5 antagonistically regulate the expression of COP1, leading to higher levels of Pchlide accumulation and lower levels of PORA transcripts in the dark‐grown abi4 mutant (Xu et al., 2016). Long‐term ABA treatment activates COP1 and promotes the ubiquitination of GLK1 for degradation, which then leads to reduced expression of HEMA1 and PORA (Lee et al., 2021). Although the direct regulatory role of JA in Chl synthesis is unknown, both JA and SA play important roles in ROS stress‐induced cell death (Sánchez‐Corrionero et al., 2017; Geng et al., 2024; Shi et al., 2024).

POST‐TRANSLATIONAL REGULATION OF THE TBS PATHWAY

Post‐translational modifications (PTMs) of enzymes play a prominent role in the regulation of protein stability and activity, and many studies have identified diverse PTMs on enzymes involved in Chl biosynthesis (Herbst et al., 2019; Wang et al., 2022). Recent studies have confirmed that thiol‐based redox regulation, including redox‐active thioredoxins (TRX) and NADPH‐dependent thioredoxin reductase (NTRC)‐mediated redox reactions, methylation of lysine and arginine residues and nitric oxide‐mediated S‐nitrosylation, play important roles in chloroplast TBS (Herbst et al., 2019). This section of our review focuses on post‐translational regulation by chaperones and auxiliary factors (Figure 3).

Figure 3.

The post‐translational regulatory network that controls the functions of tetrapyrrole biosynthesis (TBS) enzymes

Post‐translational regulation of TBS enzymes encompasses several modifiers of specific amino‐acid residues, auxiliary factors and chaperones, all of which are involved in the control of protein stability and enzyme activity. The precursors of TBS enzymes are transported into chloroplasts by the TOC–TIC complex, with the aid of molecular chaperones that include the cytoplasmic chaperones Hsp90 and Hsp70 (Heat Shock Proteins 90 and 70) and the chloroplast chaperone cpHsp70 (chloroplast stromal Hsp70). Interactions of the chaperones ClpC1, ClpC2, and ClpD with GluTR (Glutamyl‐tRNA Reductase) or CAO (Chlorophyll a oxygenase) lead to their degradation by Clp protease. CPP1/CDF1 (Chaperone of POR protein 1/Cell growth deficient factor 1) enhances the stability of POR. cpSRP43 (chloroplast Signal Recognition Particle 43) interacts with GluTR, GUN4, and GUN5 and prevents them from aggregating during heat stress. Heat‐induced dissociation of cpSRP43 from cpSRP54 enables cpSRP43 to protect these TBS enzymes under heat stress. MORF2 and 9 (Multiple organellar RNA‐editing factors) function as chaperones to inhibit the aggregation of PORB, and promote MgCh activity through interaction with GUN4. FLU (Fluorescent in blue light) and GBP (GluTR‐binding protein) act as an inhibitor and a stabilizer of GluTR, respectively, through their interactions with GluTR. LCAA interacts with CHL27 and promotes its enzyme activity. BCM1 and 2 (Balance of Chlorophyll Metabolism) optimize chlorophyll (Chl) biosynthesis by stimulating MgCh activity via their interactions with GUN4, and attenuating Chl degradation by inducing the degradation of SGR1 (Stay‐green 1). FC2 (Ferrochelatase 2) interacts with POR and FLU to stabilize the FLU‐mediated GluTR‐inactivation complex and reduce the synthesis of heme for plastid‐localized heme‐dependent proteins. PCD8 (Programmed Cell Death 8) interacts with HEMC, CHLD, and PORC to stimulate their proteolysis. In addition, heme and Pchlide contribute to the inhibition of GluTR activity, and Chl b causes the feedback‐induced breakdown of CAO (Yamasato et al., 2005; Herbst et al., 2019). The modifications of amino acids of TBS enzymes by thiol‐based redox switches, S‐nitrosylation, and phosphorylation have been summarized in an earlier review (Herbst et al., 2019) and are not shown in this figure.

The regulation of TBS by chaperones

More than 90% of plastid‐localized proteins are encoded in the nucleus and synthesized in cytoplasm, before being translocated into plastids. These proteins generally have a transit peptide (TP) at the N‐terminal end and are referred to as precursor proteins. They are imported into plastids by successive interactions with the translocon on the outer chloroplast membrane (TOC) and the inner envelope membrane (TIC) complex (Demarsy et al., 2014; Chu and Li, 2018; Jin et al., 2022; Liu et al., 2023). The TPs of the precursor proteins are cleaved off by the stromal processing peptidase (SPP) and the mature protein is sorted within the chloroplasts (Lee et al., 2018; Xu et al., 2021b). Many chaperones are involved in the import, suborganellar localization, and regulation of the stability of chloroplast‐located proteins (Li et al., 2020a; Sun and Jarvis, 2023). Outside of chloroplasts, two cytosolic chaperones,heat shock protein 90 (Hsp90) and heat shock protein 70 (Hsp70), guide and deliver precursors to the chloroplasts (May and Soll, 2000; Fellerer et al., 2011). Inside chloroplasts, chaperones like chloroplast Hsp70 (cpHsp70), chloroplast heat shock protein 90 (Hsp90C), and Hsp93 (ClpC) assist in the import of precursors (Su and Li, 2010; Inoue et al., 2013; Liu et al., 2014). Genomes uncoupled 1 (GUN1) interacts with cpHSC70 to regulate protein uptake, including the TBS proteins GluTR, GluTR‐binding protein (GBP), and CHL27. Moreover, GUN1 regulates the maintenance of chloroplast proteostasis following treatment with lincomycin or norflurazon (Wu et al., 2019). The Clp protease is located in the plastidal stroma, and is composed of a proteolytic core complex and a chaperone complex, including ClpC1, ClpC2, and ClpD (Nishimura and van Wijk, 2015; Rodriguez‐Concepcion et al., 2018). GluTR is degraded by the Clp protease upon interaction with the substrate recognition subunits ClpS1 and ClpF, and the chaperones ClpC1 and ClpC2 (Nishimura et al., 2013; Nishimura and van Wijk, 2015; Apitz et al., 2016). The Clp protease also affects the stability of CAO, which catalyzes Chl b synthesis (Nakagawara et al., 2007; Sakuraba et al., 2009). The Chaperone of POR 1/cell growth deficient factor 1 (CPP1/CDF1) serves as a chaperone that stabilizes POR in cyanobacteria, Arabidopsis, and Nicotiana tabacum (Lee et al., 2013; Liu et al., 2016).

Recent research has characterized the role of a new chaperone in Chl biosynthesis, chloroplast Signal Recognition Particle 43 (cpSRP43), which binds directly to GluTR, thereby preventing its aggregation and enhancing its stability (Wang et al., 2018). cpSRP43 also protects other proteins of Chl synthesis (including CHLH, GUN4, and POR) from heat‐induced aggregation, while cpSRP54 itself competitively inhibits the chaperone activity of cpSRP43 (Ji et al., 2021, 2023). PPO1 interacts with multiple organellar RNA‐editing factors (MORF2, MORF8, and MORF9). These proteins have been shown previously to control plastidal RNA editing (Zhang et al., 2014). More recent studies have found that MORF2 and MORF9 directly interact with multiple enzymes and regulators of TBS, including PORB and GUN4, display a holdase chaperone activity that alleviates aggregation of PORB in vitro, and contribute to the stimulated MgCh activity (Yuan et al., 2022).

Regulation of TBS by other auxiliary factors

The first step of TBS is the ATP hydrolysis‐dependent loading of glutamate onto tRNAGlu to make glutamyl‐tRNA(Glu), followed by the two‐step formation of ALA (Grimm, 1998; Brzezowski et al., 2015; Wang and Grimm, 2015). In transplastomic tobacco (Nicotiana tabacum) plants, overexpression of tRNAGlu does not affect TBS. However, the accumulated mutant tRNAGlu_C56U leads to reduced ALA synthesis rates, but does not affect chloroplast protein biosynthesis (Agrawal et al., 2020). It is an interesting question how the flux of glutamyl‐tRNA(Glu) is regulated between the demands of protein and Chl biosynthesis. 5‐Aminolevulinic acid is the common precursor of both heme and Chl synthesis. The balanced metabolic flow between the magnesium and iron branches plays an important role in ensuring that the accumulation of light‐sensitive metabolic intermediates, like PPIX, MgPPIX, and Pchlide, is avoided. When heme binds to GBP,which was identified in a search for GluTR‐interacting protein partners, its interaction with GluTR is attenuated. As a result, GluTR is released and becomes accessible to proteolysis by the Clp protease (Zhao et al., 2014; Richter et al., 2019). A recent report has emphasized the heme binding of the DUF2470 domain, which seems to be a conservative heme‐binding motif that is found in unrelated proteins (Grosjean et al., 2024). This domain is also present in the C‐terminal segment of GBP.

In Chlamydomonas reinhardtii, the heme breakdown product phycocyanobilin (PCB) interacts with CrGUN4 to promote the activity of MgCh and maintain the stability of CrCHLH1 to regulate Chl synthesis (Zhang et al., 2021b). A structural comparison revealed that the bilin‐binding site is conserved from cyanobacteria up to higher plants (Hu et al., 2021). GUN1 interacts with many TBS enzymes (such as PBGD, UROD2, CHLD, and FC1), and inhibits both heme and Pchlide synthesis in the dark (Tadini et al., 2016). Heme and other porphyrins bind directly to GUN1 to inhibit Pchlide synthesis (Shimizu et al., 2019). In addition, the accumulation of Chl b can lead to the degradation of CAO by the Clp protease, providing a negative feedback control (Yamasato et al., 2005). Apart from Pchlide, heme, and bilins, other mechanisms of tetrapyrrole metabolite‐mediated control of TBS are unclear and await further studies.

Auxiliary factors other than chaperones have also been shown to regulate TBS enzyme activity by means of protein–protein interactions. The tetratricopeptide‐repeat protein Fluorescent in blue light (FLU) was initially found to be a key negative regulator of Chl biosynthesis in Arabidopsis (Meskauskiene et al., 2001). FLU interacts with GluTR in the dark and forms an ALA synthesis inhibition complex that contains PORB, PORC, CHL27, CHLP, and Pchlide (Kauss et al., 2012; Fang et al., 2016). Upon light exposure, FLU modulates GluTR activity by altering its distribution between the thylakoid membrane and the stroma (Schmied et al., 2018; Hou et al., 2019).

Biochemical studies have uncovered a tripartite protein complex comprised of GluTR, GSAAT, and GBP that indicates that the interaction between GluTR and GSAAT is facilitated by GBP (Sinha et al., 2022). Low chlorophyll accumulation A (YCF54/LCAA) is an essential component of the O₂‐dependent MgPME cyclases, and has been proposed to act as a scaffold protein for the formation of the unique fifth ring of Chl molecules in oxygenic phototrophs, including cyanobacteria, green algae, and plants (Hollingshead et al., 2016; Chen and Hunter, 2020). YCF54 functions as an auxiliary factor to regulate the stability and activity of CHL27 (Albus et al., 2012; Hollingshead et al., 2012; Bollivar et al., 2014; Herbst et al., 2018).

The recently discovered regulatory factors BCM1 and BCM2 (balance of Chl metabolism 1/2) act as functionally conserved scaffold proteins in Chl homeostasis. In developing leaves, the thylakoid‐bound BCM1 promotes Chl biosynthesis by stimulating MgCh activity through its interaction with GUN4, and simultaneously promotes Chl accumulation through interaction‐induced degradation of Stay‐green 1 (SGR1). The second isoform BCM2 mainly functions at the beginning of leaf senescence to destabilize SGR1 (Wang et al., 2020c). Interestingly, BCM orthologs are absent from both cyanobacteria and algae. Ferrochelatase 2 (FC2), a key enzyme for heme biosynthesis in chloroplasts, is also involved in the regulation of ALA synthesis in the dark. By interacting with PORB, FC2 enhances the inhibitory effect of the FLU–POR–CHL27–GluTR complex (Fan et al., 2023). Programmed Cell Death 8 (PCD8), a newly identified thylakoid‐localized protein in Arabidopsis, interacts with a number of enzymes, such as HEMC, CHLD, and PORC, to provide further fine‐tuned regulation of Chl homeostasis (Geng et al., 2023). A novel auxiliary factor from the tetratricopeptide‐repeat protein family has recently been identified as the membrane‐bound factor; TBS‐regulating tetratricopeptide‐repeat protein 1 (TTP1) interacts with GluTR, POR, CHL27, GBP, and FLU to maintain their stability (Herbst et al., 2019, 2024).

Other modifications of TBS enzymes

Proteomic data have shown that several TBS enzymes (GSA, PBGD, UROD, PPO, CHLI, CHLH, POR, and DVR) have potential phosphorylation sites (Brzezowski et al., 2015). In A. thaliana, the S264 residue of GUN4 is such a site and the phosphorylated protein inhibits the activity of MgCh so that over‐accumulation of Pchlide is prevented in the dark (Richter et al., 2016). In C. reinhardtii, the CHLI and CHLD subunits of MgCh can be phosphorylated. CHLI2 has been shown to phosphorylate CHLD at a conserved histidine residue, which boosts MgCh activity (Adams and Reid, 2013; Sawicki et al., 2017). A recent study also reported the light‐dependent acetylation of PORA in the regulation of greening of Arabidopsis cotyledons. The key lysine residues are conserved in plants and algae (Liang et al., 2023). The identification of enzymes involved in acetylation of TBS enzymes awaits further investigation.

HEME SIGNALING AND TETRAPYRROLE TRANSPORT

Plastid‐derived retrograde signaling and the role of tetrapyrroles

As a semi‐autonomous organelle, the chloroplast can transmit so‐called retrograde signals to the nucleus that then regulate nuclear gene expression in response to environmental and developmental cues. The metabolic intermediates of the Chl branch were previously proposed to serve as putative retrograde signaling molecules (Chi et al., 2015; Zhang et al., 2015). A signaling role for MgPPIX in A. thaliana was first hypothesized more than two decades ago (Strand et al., 2003), but numerous subsequent studies have shown that MgPPIX is highly unlikely to act as a chloroplast‐derived signaling molecule in the regulation of nuclear gene expression (Larkin, 2016; Shimizu and Masuda, 2021). Chlorophyll intermediates, such as MgPPIX and Pchlide, exhibit phototoxicity and tend to generate ROS in the light (Jung et al., 2008). Singlet oxygen (¹O₂) has been suggested to function as a chloroplast retrograde signaling molecule that modulates the expression of nuclear genes via the EX1/2 pathway (Wagner et al., 2004; Li and Kim, 2022) (Figure 4).

Figure 4.

Tetrapyrrole signaling and transport

The maintenance of chloroplast function in photosynthetic eukaryotes requires coordinated regulation between the nuclear genome and the plastid genome. Tetrapyrroles, such as heme and bilin, play crucial roles in plastid‐to‐nucleus retrograde signaling in plants and algae. Heme regulates the expression of nuclear genes, including PhANGs, HSP70A, and others. In Chlamydomonas, bilin serves as a chloroplast retrograde signal that regulates the expression of nuclear genes involved in ROS detoxification and genes coding for O₂‐dependent enzymes. In addition, singlet oxygen (¹O₂) produced by tetrapyrrole intermediates mediates retrograde signaling through the EX1/2 pathways, and regulates the expression of ¹O₂‐responsive genes. Specific membrane transporter proteins for tetrapyrroles have not yet been identified. Heme can be transported from the chloroplast to the cytosol by TSPO, GSTUs, or other unidentified transport proteins. Heme is also transported to other organelles and interacts with heme‐binding proteins in the nucleus, endoplasmic reticulum, Golgi apparatus, mitochondria, and peroxisomes. Heme released from mitochondrial hemoproteins may also be transported to the cytosol. During chlorophyll (Chl) degradation, Chl breakdown products, such as pFCC and Hydroxy‐pFCC, can be transported into vacuoles by the vacuolar ABC transport proteins AtMRP1–3.

Heme, a cofactor of hemoproteins that are found in all subcellular compartments, is essential for many activities, such as redox control, electron transfer, catalysis, gas binding, and transport (Reedy and Gibney, 2004; Poulos, 2014). Heme has also been suggested to function as a versatile signaling molecule that dynamically regulates numerous proteins and pathways. Hundreds of heme‐binding proteins in A. thaliana and Cyanidioschyzon merolae have been identified by proteomic approaches. Several candidates, such as the transcription factors bHLH110 and brother of LUX ARRHYTHMO (BOA), DEAD‐box ATP‐dependent RNA helicase, NAD⁺‐dependent protein deacetylase SIRT1, and so forth, are predicted to be localized in the nucleus. This is compatible with the idea that heme might serve as a signaling molecule in the control of nuclear gene expression, RNA metabolism, and other biological processes (Shimizu et al., 2020).

In A. thaliana, FC1‐specific production of heme is more likely to be a trigger of plastid‐derived retrograde signal transduction that coordinates the expression of photosynthesis‐associated nuclear gene (PhANG) expression with chloroplast development (Woodson et al., 2011). Using a genetically encoded fluorescent heme sensor, the pool of free heme in the cytoplasm of tobacco plants overexpressing AtFC1 was found to be significantly enhanced (Wen and Grimm, 2024). In C. reinhardtii, exogenous heme treatment in the dark transiently affected the abundance of hundreds of transcripts, although genes of endosymbiotic origin were surprisingly not overrepresented (Voß et al., 2011). Detailed analysis identified a possible plastid response element (PRE) in the promoter region of HSP70A and several other nuclear genes, and its transcriptional inducibility was confirmed by hemin feeding (von Gromoff et al., 2008).

Indirect heme signaling involves heme degradation products, such as carbon monoxide (CO) and biliverdin, which affect multiple downstream targets, and influence cell signaling and metabolism (Donegan et al., 2019). In this context, heme can be regarded as a composite form of CO and biliverdin, whose signaling function could be activated by heme oxygenase. In plants and algae, the molecular components of heme signaling await further clarification.

Intracellular transport of tetrapyrrole molecules

Despite extensive research on heme biosynthesis and its regulatory mechanisms, as well as advances in heme transport in other species, little information is known about heme transport between and within plant cells. Recent findings have shown that both PPO isoforms are localized in the chloroplasts of A. thaliana, albeit with different subplastidal localizations. In Arabidopsis, PPO1 is exclusively found in the thylakoid membrane, while PPO2 is attached to the envelope membrane. Localization of the two FC isoforms resembles that of the two PPO variants, with FC2 in the thylakoid membrane and FC1 mainly in the stroma and partially at the envelope membrane. This distribution of the isoforms of PPO and FC enables the synthesis of different pools of Proto and heme for intraplastidal and extraplastidal allocations of these tetrapyrroles (Hedtke et al., 2023).

Heme synthesized in plastids must cross the envelope membranes and subsequently be incorporated into hemoproteins in the cytoplasm and various organelles, including the nucleus, mitochondria, endoplasmic reticulum (ER) and the Golgi network, peroxisomes, and the plasma membrane. The details of heme export from the plastid into the cytoplasm and its subsequent transport into other organelles remain open and, to date, plant‐specific heme transporters have not been identified. Transport of heme across the lipid bilayer from plastids is energy dependent. The porphyrin transporter TSPO was first discovered in Rhodobacter sphaeroides, and was proposed to be essential for the homeostasis and transport of uroporphyrinogen III (Yeliseev and Kaplan, 1999). Arabidopsis TSPO is a stress‐induced ER/Golgi‐localized membrane protein that can bind to tetrapyrrole molecules, including heme and PPIX. AtTSPO may contribute to porphyrin/heme binding and scavenging under stressful conditions (Vanhee et al., 2011). Notably, TSPO is present in almost every living species, although its biological functions may differ significantly among animals, plants, and microorganisms (Veenman et al., 2016).

Intracellular transport mechanisms for downstream heme metabolites have also been reported. Heme is degraded and oxidized by heme oxygenase (HO) to form biliverdin IXα (BV IXα). In photosynthetic organisms, different types of ferredoxin‐dependent biliverdin reductases (FDBRs) can convert BV IXα into various bilins (Rockwell et al., 2014). Studies on the HO mutant hmox1 in C. reinhardtii have revealed that bilins can act as chloroplast retrograde signaling molecules (Duanmu et al., 2013, 2017; Wittkopp et al., 2017). These findings suggest that bilins synthesized in the algal chloroplasts and destined for retrograde signaling must be transported into the cytoplasm. However, at present, no specific protein(s) involved in bilin transport can be pinpointed. A recent study showed that the Arabidopsis protein crumpled leaf (CRL) retains the bilin‐binding pocket, although mutations of the key bilin‐binding residues could still rescue multiple lesions of the crl mutant (Wang et al., 2020a).

Degradation products of Chl metabolism have been reported to be transported into vacuoles by the vacuolar ABC transporters AtMRP1‐3. This provides some evidence for the potential involvement of ABC transporters in the redistribution of other tetrapyrrole molecules (Lu et al., 1998; Tommasini et al., 1998). In C. reinhardtii, in total, 75 ABC transporter proteins were identified and categorized into eight subfamilies. This has created a basis for the functional investigation of ABC proteins in microalgae, for example in the transport of bilins and Chl intermediates (Li et al., 2022b).

PERSPECTIVES

Chl biosynthesis and its regulation remain a vibrant research area with significant implications for plant biology, agriculture, and biotechnology. Using advanced technologies and interdisciplinary approaches, future research can uncover new regulatory mechanisms, improve our understanding of plant physiology, and contribute to sustainable agricultural practices. Comprehensive multi‐omics approaches, including genomics, transcriptomics, proteomics and metabolomics, can provide a holistic understanding of Chl biosynthesis and its regulation. The integration of these datasets can help to identify novel regulatory components and interactions. Further investigation of PTMs such as phosphorylation, ubiquitination, and acetylation could reveal new players of regulation.

Understanding how environmental factors influence Chl biosynthesis is crucial for the improvement of crop resilience. Research on stress‐responsive regulatory networks and their impact on Chl metabolism could lead to the development of stress‐tolerant plant varieties. Comparative studies across different plant species, including non‐vascular plants and algae, could provide insight into the evolution of Chl biosynthesis pathways. Understanding the evolutionary adaptations could inform strategies for optimizing photosynthesis in diverse plant systems.

The use of synthetic biology approaches to engineer Chl biosynthesis pathways in heterologous systems could provide valuable insight and uncover potential applications. These approaches range in complexity from increasing pigment concentrations per leaf, extending the lifespan of Chl and broadening the light absorption spectrum, to constructing artificial photosynthetic systems. It is possible to exploit exotic Chls by integrating them with endogenous intermediates in plants or engineering cell factories with light‐powered metabolism. In fact, it has been demonstrated that Chl b and Chl c can be heterologously produced, and Chl d can be assembled into plant antenna complexes in vitro (Xu et al., 2001; Elias et al., 2021, 2024a; Jinkerson et al., 2024). With regard to Chl f, recent substitutions of two amino acids in the D1 protein have led to the production of Chl f in cyanobacteria (Trinugroho et al., 2020). This heterologous Chl may be adaptable to photosynthetic eukaryotes. Researchers have also been evaluating the complete set of enzymes required for Chl biosynthesis in heterotrophic organisms such as Escherichia coli; this is a first toward building the apparatus of the light reactions de novo (Pearlstein, 1996; Chen et al., 2018). Further investigations into the biosynthesis and regulation of the relative levels of Chl derivatives could open up new avenues for biotechnological applications. It has also been verified experimentally that ROS generated by photoactivated porphyrins can disrupt the structure of Tick‐borne encephalitis virus (TBEV) viral particles, thus inhibiting virus activity (Holoubek et al., 2024). Chlorophylls also have photosensitizing properties that could be used for early and accurate cancer diagnosis and treatment (Sun et al., 2024). Finally, in addition to analyzing the biological significance and functions of Chl, techniques based on Chl fluorescence (especially that of Chl a) are now being used to determine photosynthetic performance rapidly, non‐invasively, and with high sensitivity. Furthermore, the development of high‐throughput and stable Chl fluorescence‐based methods will facilitate the quantitative assessment of photosynthesis for agricultural crops.

CONFLICTS OF INTEREST

The authors declare no competing financial interests.

AUTHOR CONTRIBUTIONS

Y.L, T.C., Y.G., B.G., X.L., D.D., and R.L. wrote and revised the paper. All authors read and approved its content.

Biographies

Li, Y. , Cao, T. , Guo, Y. , Grimm, B. , Li, X. , Duanmu, D. , and Lin, R. (2025). Regulatory and retrograde signaling networks in the chlorophyll biosynthetic pathway. J. Integr. Plant Biol. 67: 887–911.