Are cuproproteins part of the multi-protein framework for making the Casparian strip?

ABSTRACT

Casparian strip (CS) is a lignified structure localized on the cell wall between adjacent root endodermal cells and functions as an apoplastic diffusion barrier in the root. The polarly localized, lignin-based CS is an excellent system for studying peptide signaling and position recognition. In this short review, we summarize advances in the past decade on the molecular mechanism governing CS development. In addition to the multi-protein framework underlying the CS membrane domain, we discuss recently observed participation of cell wall located cuproproteins in CS formation. These new discoveries shed light on a potential CS wall domain that coordinates with the membrane domain to provide bidirectional positional information for guiding precise CS development.

KEYWORDS:

• Casparian strip
• cuproproteins
• cell wall
• position recognition

Introduction

Plants must respond and adapt to various soil conditions for survival because they are sessile and rely on the roots to acquire water and mineral nutrients.¹ The protective sheath called Casparian strip (CS) between adjacent endodermal cells allows the root to block the apoplastic pathway and take in minerals selectively.² This apoplastic diffusion barrier is formed by impregnating the cell wall with lignin, a hydrophobic phenylpropanoid polymer that seals the entire intercellular space between two neighboring cytoplasmic membranes.³ CS occupies the central region of transversal and anticlinal cell walls of endodermal cells⁴ (Figure 1a). Collectively, connected CSs are woven into a fishnet-like structure that stretches from the differentiation zone to the root-shoot junction.⁴ The mechanism underlying the establishment of a polarly localized, lignin-based CS has been an intriguing question since 1865, when it was first described by the German botanist Robert Caspary.⁵

Figure 1. Schematic representation of the multi-protein framework for CS development. (a) The morphology of CS observed from different views. (b) Model for how the wall domain coordinates with the membrane domain to spatially define CS. Before the diffusion barrier is established, as shown on the left portion, SGN3 and SGN1 recognize the CIF peptides emanating from the stele to act as a surveillance system that activates localized ROS and downstream gene expression. As shown on the right portion, LAC3 participates in forming the cell wall domain, which together with EXO70A1 help focalize CASPs to demark the CS membrane domain and recruit PER64 for further lignification and proper sealing of the extracellular space. ESB1 and UCC1 are involved in this scenario with unknown functions. Once the CS is sealed, the CIF peptides are trapped in the pericycle side and thus cease signaling.

The upstream signaling pathway for CS formation

The first proteins found to be directly involved in CS development are Casparian strip membrane domain proteins (CASPs). Five members of this family in Arabidopsis, CASP1-5, mark a membrane domain that is attached to CS and excludes most other plasma membrane proteins.^6,7 The interaction between CASP1 and CASP3 is thought to form a scaffold to recruit other proteins into the CS region. These characteristics of CASPs are conserved in other plant species such as rice and tomato.^8–11 To search for genes that regulate CS development, Niko Geldner and colleagues developed an ingenious assay to screen for mutants with leaky CS, employing the stele active promoter proSHR to drive β-glucuronidase as a reporter.¹² Four genes were identified from this screen and named SCHENGEN (SGN). SGN4 encodes an NADPH oxidase (also known as RBOHF) and its mutants display severe delay of lignin deposition on CS.¹³ Pharmacological inhibition of other components of the ROS pathway as well as generation of a quintuple peroxidase (PER) mutant abolishes lignin deposition on CS,^13,14 firmly establishing peroxidases as the main enzymes for lignin crosslinking in CS. SGN3 is a leucine-rich-repeat receptor-like kinase flanking the central CS membrane domain,¹⁵ whereas SGN1 localizing to the outer plasma membrane belongs to the receptor-like cytoplasmic kinase family.¹² The ligands for SGN3, named CASPARIAN STRIP INTEGRITY FACTOR 1 & 2 (CIF1 & 2), are small peptides expressed from the stele.^16,17 These peptides are sulfated by SGN2, the only tyrosylprotein sulfotransferase in Arabidopsis.¹⁸ Function of the CIF-SGN3-SGN1 signaling module has been elucidated. On one hand, SGN3 bound by CIFs phosphorylates SGN1, which furtherly phosphorylates and activates RBOHF and RBOHD to induce localized ROS production for lignin deposition. On the other hand, CIF/SGN3 induces MAP kinase phosphorylation and activates downstream gene expression.¹⁹ Once the CS is sealed, the CIF peptides physically separate from the receptor. Thus, the CIF-SGN3-SGN1 module acts as a surveillance system to detect porous CS and trigger compensatory lignification that starts at the middle lamellae of cell corners^18,20 (Figure 1b). It is subsequently shown that this signaling pathway acts in parallel with the SHORTROOT-SCARECROW (SHR-SCR) module to control endodermis identity altogether.^21,22

Other CS-associated proteins

Mutants resulting in discontinuous CS alter the elemental profile in the shoot, which is often coupled with enhanced suberin deposition at the CS, presumably to compensate for the compromised integrity of the lignin ring. This phenomenon in retrospect has inspired reexamination of mutants with altered elemental profiles for isolating determinants of CASP localization and CS formation. Enhanced suberin1 (ESB1) was characterized this way.²³ ESB1 is a dirigent domain-containing protein with unknown function and specifically localizes to the CS. MYB36 is a transcription factor induced by the SHR-SCR module and activates the expression of many CS related genes such as CASPs, PER64 and ESB1.^24,25 Among another set of mutants (termed lord of the rings) that compromises CS integrity, LOTR1 is a member of an uncharacterized family and LOTR2 is a subunit of the exocyst complex.^26,27 LOTR2/EXO70A1 transiently accumulates on CS membrane domain prior to CASP1. In its absence, CASP1 secretion is not affected but displays dramatic de-localization into random microdomains, indicating that EXO70A1 generates a transient positional information that guides CASP localization²⁷ (Figure 1b). Despite these findings, how the various factors integrate together to form a polarly localized CASP domain remains to be elucidated.

CS localized cuproproteins

Lignin is a polymer formed by the dehydrogenative polymerization of monolignols.²⁸ Many essential phenylpropanoid biosynthesis genes are expressed in the root endodermis, suggesting that this cell layer is able to synthesize monolignols autonomously.²⁹ A monolignol transporter AtABCG29 is found localizing on the plasma membrane of endodermis,³⁰ though monolignols can also be exported via passive membrane crossing.³¹ Taken together, these observations indicate that precise positioning of CS lignification is not dependent on polar transport of lignin monomers, but localized monolignol oxidation. Laccases are secreted multi-copper oxidases that have been implicated in oxidative coupling of monolignols within the cell wall.²⁸ A particular member of the laccase family in Arabidopsis, LAC3, is specifically expressed in the endodermis and its transcript abundance is further upregulated by CIF/SGN3 signaling,¹⁹ suggesting that LAC3 participates in CS development. Recently, it was reported that LAC3 localizes on the primary cell wall in the CS domain, initially preceding CASP1 localization to the membrane, then accumulating between adjacent CASP1 domains, and eventually localizing on the borders of CS.³² Moreover, copper chelation and inhibition of N-glycosylation cause CASP1 dispersal and abnormal lignification.³² These findings indicate that cuproproteins are required for CASP1 focalization and proper CS lignification. However, a nonuple mutant of LAC genes in Arabidopsis exhibits no apparent defects in CS formation,¹⁴ implying that additional cuproproteins are involved.

Uclacyanins (UCCs) are a glycosylphosphatidylinositol-anchored sub-family of plant-specific blue copper proteins possessing a characteristic mononuclear copper site and a putative arabinogalactan glycomodule resembling that of cell wall structural proteins.³³ Thus, UCCs are likely anchored to the cell surface and associate with the cell wall. Transcriptomic data in Arabidopsis indicate that UCC1 is differentially expressed in myb36 mutants.^24,34 UCC1 is compartmentalized and forms a central nanodomain within the CS in contrast to LAC3 and other CS-located proteins that mainly accumulate at the periphery of the CS^32,34 (Figure 1b). Loss-of-function of UCC1 & 2 reduces lignification specifically in this central CS nanodomain,³⁴ similar to CS defects caused by copper chelation.³² These observations suggest that strong toxicity of copper chelation on CS development may reflect a comprehensive effect of interfering both LAC3 and UCC1 & 2 or other yet-unidentified cuproproteins. Interestingly, the transcripts of LAC3 as well as its close homologs LAC12 & 13 and UCC1 are targeted by a conserved microRNA, miR408.³⁵ This miRNA belongs to the so-called copper-miRNAs because they mostly target transcripts encoding cuproproteins and their abundance is regulated by the copper-responsive transcription factor SQUAMOSA PROMOTER BINDING PROTEIN-LIKE7.^36,37 These findings indicate that regulation and function of LAC3 and UCC1 & 2 are orchestrated.

Different from tight junctions in animal cells that are directly sealed by large protein complexes, two adjacent root endodermal cells share one CS position with the plasma membranes physically separated by the cell wall. Thus, it was proposed that cuproproteins such as LAC3 marks a possible cell wall domain that coordinates with the membrane domain to define the positional information for accurate CS formation³² (Figure 1b). In this proposal, anchoring LAC3 to the cell wall is one of the earliest events in CS development. LAC3 and UCC1 & 2 compartmentalized respectively to the periphery and the central nanodomain of the CS are hypothesized to define a cell wall domain that provides an interface for restricting or priming the peroxidase pathway for CS lignification (Figure 1b). Likewise, a study using scr mutants supports the notion that endodermal cell-cell contact is required for the positioning of CS.³⁸ It can be extrapolated that this cell wall domain coordinates with the CASP domain and contributes to CS integrity that is under surveillance by the CIF-SGN3 peptide signaling pathway.³² Thus, precise CS lignification across a cell wall in between neighboring endodermal cells may require separate wall domain and membrane domain that coordinate to provide bidirectional positional information.

Conclusions

During the last decade, studies of the molecular mechanism underlying CS development have made impressive progress. So far, most of the experimentally identified CS factors show specific localization on the plasma membrane of endodermal cells. Despite the long debate on whether laccases or peroxidases oxidize monolignols to polymerize lignin,^14,28 recent characterization of cuproproteins exemplified by LAC3 and UCC1 supports a model for CS positioning involving a cell wall domain (Figure 1b). This notion brings copper homeostasis, which consists of sophisticated gene regulation programs and elaborate transport systems,^39,40 into the realm of CS development. Further elucidation of the function of CS cuproproteins and related copper homeostatic events should provide more insights into the cellular mechanisms governing position recognition and focalized lignification in plants.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Additional information

Funding

This work was supported by the National Key Research and Development Program of China [2017YFA0503800] and the National Natural Science Foundation of China [31621001].