Genome-wide analyses revealed that the distribution of cyclophilin genes on different chromosomes in plants is uneven (Table 3). The cyclophilin genes in allopolyploids such as B. napus and wheat occur in pairs, with each member originating from one progenitor chromosomal set. These pairs are highly identical and share localization patterns (Hanhart et al., 2017; Singh et al., 2019). Structural analysis of cyclophilin genes in plants has been carried out for soybean, cotton, wheat and Medicago truncatula (Mainali et al., 2014; Chen et al., 2019; Singh et al., 2019; Ge L. et al., 2020). These studies revealed considerable variability in the distribution and size of introns in the open reading frames (ORFs) and untranslated regions (UTRs) as compared to other organisms (Table 4). The cyclophilin genes with the highest number of introns include cotton (20 in GbCYP142; Chen et al., 2019), wheat (13 each in TaCYP64-1-7A, TaCYP64-2-7B, and TaCYP64-3-7D; Singh et al., 2019) and soybean (13 each in GmCYP56 and GmCYP59; Mainali et al., 2014). The largest intron (28618 bp) was observed in TaCYP26-5-2B, while the smallest (39 bp) was noticed in GmCYP5 (Table 4). Information about variations in the structure of cyclophilin genes in rice, Arabidopsis and Brassica, which is lacking, may provide further insights into the evolution of these families in plants. Loss or gain of introns, an important aspect of structural variation, is vital for gene evolution (Roy and Gilbert, 2006). The intron size may be correlated with the genome size and longer introns have been proposed to confer a selective advantage by improving the recombination, and also by counterbalancing the mutational bias towards deletions (Carvalho and Clark, 1999; McLysaght et al., 2000). Thus, the variability in introns in plant cyclophilins may have important implications in their functionalization which needs to be explored further. Since 5′ and 3′ UTRs are structurally important and regulate the expression of eukaryotic genes (Wilkie et al., 2003), differences in these regions may likely enable differential regulation of plant cyclophilins, leading to divergence in their physiological roles.

The cyclophilins in plants and other organisms, though predominantly cytosolic, are also predicted to be localized in the chloroplast, nucleus, mitochondria, extracellular/secretory and plasma membrane (Table 1). The presence of cyclophilins in different organelles of plants signifies their specific and distinct roles in the cell (Tables 1, 2). Based on domain organization, the cyclophilins are classified as single- (SD) or multi-domain (MD) forms (Table 1). The SD cyclophilins possess the characteristic cyclophilin-like domain (CLD), while the MD cyclophilins also contain additional specific functional domains (Table 5). Analysis of CLD in the typical human cyclophilin, hCYPA (hCYP18-A/CYPA), demonstrated that the residues Arg55, Phe60, Met61, Glu63, Ala101, Phe113, Trp121, Leu122 and His126 are essential for PPIase activity (Zydowsky et al., 1992b; Ke et al., 1994; Zhao et al., 1997; Howard et al., 2003; Davis et al., 2010). Arg55, in particular, plays a critical role in PPIase functions, whereas Trp121, though not involved in cis-trans isomerization, is essential for CsA binding (Liu et al., 1991; Zydowsky et al., 1992b; Howard et al., 2003). Interestingly, in the plant MD cyclophilins, the TPR and WD40 repeats are observed more commonly compared to other domains (Table 5). The domains such as TPR, WD40, F-box, coiled-coil, etc., have been reported to facilitate protein-protein interactions in the cell (Lamb et al., 1995; Craig and Tyers, 1999; Van Nocker and Ludwig, 2003; Liu et al., 2006). Hence, the cyclophilins consisting of these motifs may be acting as platforms for assembling protein complexes or mediate transient interactions among other proteins, further indicating their functional versatility (Bandziulis et al., 1989; Van Nocker and Ludwig, 2003; Stirnimann et al., 2010; Earley and Poethig, 2011). Compared with yeast and human cyclophilins, the presence of various additional domains such as PsbQ-like, F-box, Helical bundle, ATPase and PAN_4 domain in the plant MD cyclophilins (Figure 1 and Table 5) signifies divergence of their roles that are yet to be explored completely (Dornan et al., 2003; Romano et al., 2004b; Mainali et al., 2014; Kumari et al., 2015; Hanhart et al., 2017; Chen et al., 2019; Singh et al., 2019).

So far, only five different plant cyclophilins viz., TaCYPA-1 (Sekhon et al., 2013), CsCYP (Campos et al., 2013), Catharanthus roseus Cat r 1 (Ghosh et al., 2014), BnCYP19-1 (Hanhart et al., 2019) and AtCYP38 or CYP38 (Vasudevan et al., 2012) have been characterized for their crystal structures. While the former four are single-domain proteins and show PPIase activity, the AtCYP38 is a MD cyclophilin that lacks cis-trans isomerization capability (Vasudevan et al., 2012). The crystal structures of TaCYPA-1, CsCYP, BnCYP19-1 and CLD of AtCYP38 are similar to “archetypal” human cyclophilin hCYPA, and consist of eight-stranded antiparallel β-barrel capped at either end by two α-helices (Vasudevan et al., 2012; Campos et al., 2013; Sekhon et al., 2013; Hanhart et al., 2019). However, Cat r 1 (PDB: 2MC9) shows variability in its structure since the β-barrel in this protein consists of seven antiparallel β-strands instead of eight (Ghosh et al., 2014). The CsA-binding site in hCYPA and other such cyclophilins is composed of seven aromatic and other hydrophobic residues that constitute the hydrophobic core within the barrel (Kallen et al., 1991). The topology of this β-barrel structure is unique in the sense that it remains occupied with a set of closely packed aromatic groups making no room for binding of either CsA or the Pro containing peptides (Ke, 1992). Therefore, the CsA and other substrates bind to an active site that is formed by amino acid residues located on the outer face of the β sheet. The active sites consist of 13 residues which are identical in CsCYP, TaCYPA-1, BnCYP19-1 and hCYPA (Ke et al., 1994; Campos et al., 2013; Sekhon et al., 2013; Hanhart et al., 2019). However, the electrostatic surface map studies indicated that despite conservation of all the 13 active site residues, the active site pocket in Cat r 1 appears to be slightly broader and is more acidic in nature, which might be imparting precision for binding of peptides with a specific amino acid composition (Ghosh et al., 2014). While the conservation of CLD structure in cyclophilins underlines its fundamental role in the cell, the remarkable diversity in their domain architecture could have subtle or profound effects on the structure of these proteins which may, in turn, affect their biochemical activities differently, enabling them to perform a wide variety of roles in different cellular processes. Elucidation of crystal structures of different cyclophilins and identification of their interacting proteins is, thus, imperative to gain further insights into their specific functions.

Recent studies have provided evidence for the involvement of cyclophilins in several hormone-mediated responses in plants. Brassinosteroids and gibberellic acid (GA) are key regulators of plant stem elongation, and defects in the biosynthetic or signaling pathways of these hormones result in dwarf phenotype (Wang and Li, 2008). Genes contributing to dwarfness are of agronomic importance due to their potential for developing crops that are resistant to lodging under water-logging and strong wind conditions. DELLA proteins (named after conserved N-terminal D-E-L-L-A amino acid sequence) are inhibitors of stem growth and have been implicated in dwarf phenotype in Arabidopsis, B. napus and peach (Lawit et al., 2010; Zhao et al., 2017; Cheng et al., 2019). GA degrades DELLA proteins via the ubiquitin-proteasome pathway to promote stem growth (Sun, 2008, 2010). Mutations in the DELLA domain that abrogate interaction with F-box containing proteins SLY1, GID1 and GID2 prevent their GA-dependent degradation (Dill et al., 2004; Ueguchi-Tanaka et al., 2005; Nakajima et al., 2006; Lou et al., 2016). Functional impairment of DELLA proteins was reported to result in the dominant GA-insensitive dwarf phenotype (gaid) in wheat and B. rapa (Brrga1-d) (Ho et al., 1981; Muangprom et al., 2005). The gaid phenotype in wheat was also associated with higher levels of a 20 kDa cyclophilin, TaCYP20-2, overexpression of which in the wild-type wheat lead to gaid-like phenotype (Li et al., 2010), implying that this protein plays an essential role in maintaining GA homoeostasis by regulating the DELLA proteins. However, elucidation of the precise mechanism of action requires further intense experimentations.

The inhibition of hypocotyl growth and the expansion of cotyledons by light after the emergence of shoot from the soil in Arabidopsis is regulated by the photoreceptors phytochromes (PHYA to PHYE) and cryptochromes (CRY1 and CRY2) (Cashmore et al., 1999; Quail, 2005). Screening of the transgenic Arabidopsis 35S-cDNA lines for defective de-etiolation under a combination of blue and far-red light resulted in the isolation of a mutant (roc1-1D) that depicted enhanced expression of a cytoplasmic cyclophilin, AtCYP18-3 (ROC1, Rotamase Cyclophilin 1). The roc1-1D plants exhibited long hypocotyls and poorly unfolded cotyledons under blue and far-red light, and lower anthocyanin under far-red or blue light (Trupkin et al., 2012). Further analysis revealed that the mutant plants were hypersensitive to brassinosteroids in light but not in the dark. Inhibition of brassinosteroid synthesis and mutations in the genes responsible for brassinosteroid signaling abolished the mutant phenotype, implying that AtCYP18-3 links cryptochrome and phytochrome to brassinosteroid sensitivity (Trupkin et al., 2012).

Subsequent studies also provided evidence that functionality of AtCYP18-3 is highly sensitive to single amino acid substitution, since plants which over-expressed its variant containing phenylalanine instead of serine at position 58 exhibited reduced height, increase in shoot branching and higher sensitivity to photoperiod and temperature (Ma et al., 2013). The wild type AtCYP18-3 though does not appear to control stem elongation, likely conformation changes due to amino acid substitution might have resulted in the identification of new targets, thereby, affecting the stem growth. Therefore, structural analysis and identification of interacting proteins are imperative to understand the molecular mechanisms by which the mutated AtCYP18-3 controls growth and development in plants. Further, whether the mutated AtCYP18-3 can facilitate cross-talk between brassinosteroid signaling and photoreceptors is also a subject of future studies.

Besides brassinosteroid and GA signaling, cyclophilins have also been demonstrated to mediate auxin response. At low levels of auxin, the expression of auxin-responsive genes is kept in check by the unstable transcriptional repressors Aux/IAA proteins that bind to and inhibit the activity of auxin response factors (ARFs), a family of transcriptional activators (Figure 2; Theologis et al., 1985; Ainley et al., 1988; Conner et al., 1990; Yamamoto et al., 1992; Guilfoyle et al., 1993; Abel et al., 1995). The Aux/IAA genes are also induced by IAA and control the auxin response through a negative feedback loop (Reed, 2001). The Aux/IAA proteins consist of four highly conserved domains I-IV and bind to the ARFs either directly or through recruitment of transcriptional corepressor such as TOPLESS (TPL), the interactions being mediated by domain I that contains Leu-rich motif (Tiwari et al., 2004; Szemenyei et al., 2008). At high levels, the auxin binds to its receptor TRANSPORT INHIBITOR RESPONSE1/AUXIN SIGNALING F-BOX PROTEINS (TIR1/AFBs), an F-box containing protein, and the auxin-responsive genes are activated through auxin-dependent proteasomal degradation of Aux/IAA proteins that require ubiquitination (Wang and Estelle, 2014). The ubiquitination of proteins is catalyzed by a cascade of three enzymes viz., the Ub-activating enzyme (E1), the Ub-conjugating enzyme (E2) and the Ub-protein ligase (E3). The SCF (Skp1-Cul1-F box) E3, one of the four different types of E3s described in plants, is a complex of four different polypeptides viz., SKP1 (a member of an ASK family in plants), CDC53 or Cullin (Cul1), an F Box protein and RBX. The Cul1 acts as a central scaffold protein, while the SKP1 interacts with the F-box protein that further binds to the substrate proteins (Smalle and Vierstra, 2004). Transfer of Ub from Ub-E2 to the substrate protein is catalyzed by the fourth subunit (RBX1, ROC, or Hrt1) of the SCF complex (Petroski and Deshaies, 2005). The TIR1 interacts with SKP1 to form the SCF^TIR1 complex (Ruegger et al., 1998; Gray et al., 1999). Auxin acts as a molecular glue and after binding to TIR1, it enhances the interaction of the latter with the highly conserved ‘degron’ motif GWPPV in domain II of Aux/IAAs, leading to ubiquitination and proteolytic degradation of the latter (Figure 2; Gray et al., 2001; Reed, 2001; Tan et al., 2007). The Aux/IAA proteins bind to SCF^TIR1-Auxin complex only when the ‘degron’ motif GWPPV is in the cis W-P isomer (Tan et al., 2007; Acevedo et al., 2019). Recent studies have provided insights into the implications of cyclophilin-associated PPIase activity in mediating the interaction of Aux/IAA with the SCF^TIR1-Auxin complex. The LATERAL ROOTLESS 2 (LRT2) in rice encodes a cyclophilin PPIase OsCYP2, and disruption of this gene leads to an auxin-resistant phenotype and defective development of lateral roots (Kang et al., 2013; Zheng et al., 2013). The OsCYP2 was demonstrated to physically interact with the rice OsAux/IAA and TIR proteins, and catalyze the cis-trans isomerization of the OsIAA11 degron motif (Jing et al., 2015). These findings, thus, imply that the equilibrium of cis to trans populations of Aux/IAA proteins acts as a molecular timer to regulate auxin signal transduction (Acevedo et al., 2019). Since transcription of genes responsive to jasmonic acid, GA and strigolactone is also dependent on proteasome-mediated degradation of their specific repressors, the involvement of PPIases in controlling regulatory circuits of other hormones cannot be ruled out and should be the subject of future studies.

Given the diversity of PPIases in plants, it is likely that parallel regulatory mechanisms may be operating for several other processes in plants that, nonetheless, are yet to be identified. The presence of different functional domains, several of which facilitate protein-protein interactions, may enable the cyclophilins to identify a multitude of proteins as targets, thereby controlling complex regulatory circuits that enable the plants to respond to various developmental and environmental cues. It is apparent that, as proposed earlier for several biological processes such as cell division, gene expression, immune response and neural functions in animals (Lu et al., 2002, 2007), the PPIase catalyzed cis-trans conversion may act as a molecular switch in plants as well.

Transcript turnover and translational control are important post-transcriptional mechanisms of regulation of gene expression. Several cyclophilins have been reported to contain RNA Recognition Motif (RRM), a 90 amino acid long conserved RNA binding motif that is a characteristic feature of RNA-interacting proteins known to actively participate in pre-mRNA processing events (Kenan et al., 1991; Birney et al., 1993). This group of proteins is popularly known as cyclophilin-RNA interacting proteins (CRIPs). The first gene belonging to this group, KIN241, was identified in Paramecium and demonstrated to play an essential role in cell morphogenesis, cortical organization and nuclear reorganization (Krzywicka et al., 2001). The Arabidopsis cyclophilin AtCYP59, which besides PPIase domain also contains an RRM motif, a Zn-knuckle and a charged C-terminal domain consisting of RS/RD (arginine/serine and arginine/aspartate) repeats, was proposed to regulate transcription through its interaction with the immature mRNA (Gullerova et al., 2006; Bannikova et al., 2012). However, contrary to human RRM-containing cyclophilin hCYP33 (CYPE), that showed enhanced PPIase activity after binding to RNA (Wang et al., 2008), the catalytic activity of AtCYP59 was repressed by RNA, indicating a possible negative feedback loop. The physiological significance of this observation in plants is, however, still to be established. Though the presence of RRM along with other domains is also observed in other cyclophilins viz., BnCYP52, BnCYP55 and BnCYP112 in B. napus, and TaCYP37-1-3D, TaCYP38-1-3B, TaCYP45-1-3A, TaCYP53-1-4B, TaCYP54-1-4A, TaCYP55-1-4D, TaCYP64-1-7A, TaCYP64-2-7B and TaCYP64-3-7D in wheat (Hanhart et al., 2017; Singh et al., 2019), the precise role of these proteins in RNA processing or transcriptional regulation is only speculative. A multi-domain cyclophilin, BnCYP146, the largest cyclophilin in B. napus, exhibits the presence of a putative Fip1 domain that has not been identified earlier in any of the cyclophilins. As Fip1 is a transmembrane motif involved in polyadenylation of mRNAs via interaction with the poly(A) polymerase (Hanhart et al., 2017), BnCYP146 might have a role in the stabilization of target RNA molecules and, hence, in the regulation of translation. This, however, requires further validation.

The expression of cyclophilins in plants and other organisms is regulated by several different stress conditions (Table 7), supporting their role in the adaptation process (Marivet et al., 1994; Godoy et al., 2000; Sharma et al., 2003; Sekhar et al., 2010; Kumari et al., 2013, 2015). Our studies on sorghum were the first in plants to demonstrate that stress-induced PPIase activity is associated with drought tolerance (Sharma and Singh, 2003a,b; Sharma et al., 2003). Since then, conclusive evidence for the role of cyclophilins in the adaptation of plants to abiotic stress has been provided by several transgenic studies (Table 7). Heterologous expression of pigeonpea (CcCYP) and Golgi-localized rice (OsCYP21-4) cyclophilins imparted tolerance against salt and oxidative stress in Arabidopsis and rice (Oryza sativa), respectively (Sekhar et al., 2010; Lee et al., 2015a). Ectopic expression of a cold-induced cyclophilin PPIase, OsCYP19-4, in transgenic rice resulted in a significant increase in the number of tillers, spikes, grain weight, and was associated with cold resistance (Yoon et al., 2016). Due to high similarity (70 %) to AtCYP19-4 (Ahn et al., 2010), that interacts with guanine nucleotide exchange factor (GNOM protein) which is involved in polar localization of the auxin efflux carrier PIN1, the enhanced performance of OsCYP19-4 overexpressing plants was ascribed to alteration in auxin homeostasis (Yoon et al., 2016). Determination of the auxin levels is required to support the proposed mechanism.

The ability to confer tolerance against a broad range of abiotic stress conditions was also observed for the rice cyclophilin OsCYP2 (Table 7), which is localized to cytosol and nucleus, and shares 62.79 % and 32.08 % identity with OsCYP19-4 and OsCYP21-4, respectively (Kumari et al., 2013, 2015). The OsCYP2-induced tolerance to stress in transgenic tobacco plants was attributed to the regulation of ion homeostasis due to an enhanced K⁺/Na⁺ ratio (Kumari et al., 2015). The drought tolerance in the OsCYP18-2 over-expressing transgenic Arabidopsis, on the contrary, was ascribed to reduced transpiration rate due to a decrease in stomatal aperture (Lee et al., 2015b). Though OsCYP18-2 was also shown to interact with the Ski interacting protein (OsSKIP) in rice (Lee et al., 2015b), the role of this interaction in stress tolerance is not understood. The abrogation of this interaction by engineering OsCYP18-2 and OsSKIP will provide further insights into its functional significance.

The plastidic cyclophilins have also been demonstrated to impart protection against stress. Ectopic expression of the thylakoid localized cyclophilins, OsCYP20-2 and AtCYP38, resulted in enhanced tolerance to various abiotic stresses in the transgenic Arabidopsis and tobacco plants (Kim et al., 2012; Wang et al., 2015; Ge Q. et al., 2020). While the OsCYP20-2-induced-tolerance was ascribed to higher chloroplast PPIase activity and maintenance of NADH dehydrogenase-like complex that protects the stroma against over-reduction under stress conditions, the AtCYP38-stimulated protection against high light intensity was due to inhibition of PsbO₂ activity which is an important component of photosystem II (Wang et al., 2015). Recent studies have demonstrated the presence of two different variants of OsCYP20-2 in rice, and the two isoforms contribute to chilling stress tolerance through different mechanisms (Ge Q. et al., 2020). While the chloroplast-localized OsCYP2 contributes to scavenging of ROS by enhancing the activity of a superoxide dismutase, OsFSD2, the nuclear-localized isoform, generated following truncation of the chloroplast signal peptide, interacts with a DELLA protein, SLENDER RICE1, and stimulates its degradation to promote growth. These studies, hence, highlight the crucial role of OsCYP20-2 in integrating plant growth and abiotic stress response. As observed in transgenic tobacco plants that constitutively expressed GjCYP-1, a cyclophilin gene from red alga Griffithsia japonica, the PPIase-induced stress tolerance might also be associated with adverse effects on growth and yield (Cho and Kim, 2008), thereby, necessitating the use of stress-inducible promoters.

Though molecular processes underlying the cyclophilin-induced stress tolerance are not fully understood for the majority of the cyclophilins, prevention of protein aggregation, as reported for GjCYP-1, may be one of the protective mechanisms (Cho et al., 2005). The heat stress tolerance in E. coli that overexpressed a redox-regulated wheat cytosolic cyclophilin, TaCYPA-1, was however attributed to its PPIase activity (Kaur et al., 2016, 2017). Since the redox status of plants undergoes reversible changes under stress conditions (Jubany-Mari et al., 2010), application of a redox-sensing GFP (c-roGFP1) for real-time monitoring of cytosol redox status (Brossa et al., 2013) is needed to explore the role of TaCYPA-1 in the maintenance of redox homeostasis in the cell under stress conditions. Further, our studies also demonstrated that TaCYPA-1 and AtCYP19-3, that are 74 % identical, interact with calmodulin (CaM) in a Ca²⁺-dependent fashion (Popescu et al., 2007; Kaur et al., 2015). As Ca²⁺ is a transducer of stress signals (Snedden and Fromm, 2001; Virdi et al., 2015), cyclophilins may likely constitute an important component of Ca²⁺-CaM signaling pathway. Whether interaction with CaM is a property shared by all cyclophilins is still a matter of speculation and requires further investigations for elucidating the role of these proteins in CaM-mediated responses

The expression of cyclophilin genes is also regulated by CO₂ and nitrogen. Transcript levels of a tobacco cyclophilin gene were reported to increase under low nitrogen conditions (Yang et al., 2013), but the physiological implication of this observation is yet to be ascertained. Due to the competitive nature of ribulose-1, 5-bisphosphate carboxylase oxygenase (Rubisco) catalyzed carboxylation and oxygenation reactions, the photosynthetic activity is low in plants and algae. Hence, under low CO_2, the CO₂-concentrating mechanism (CCM) is induced in several algae such as Chlamydomonas reinhardtii (Moroney and Ynalvez, 2007). CCM leads to a high ratio of CO₂ to O₂ at the site of Rubisco and stimulates the carboxylation reaction under depleted CO₂ conditions (Badger et al., 1980; Moroney and Mason, 1991). The establishment of CCM under low CO₂ conditions in C. reinhardtii was reported to coincide with a transient increase in expression of a cyclophilin gene, indicating its likely role in this mechanism (Somanchi and Moroney, 1999). It was conjectured that this cyclophilin may be required for protecting the proteins against photodamage since CO₂ is an electron receptor and a decrease in CO₂ concentration at the same light imposes photooxidative stress. Similar roles cannot be ruled out for other cyclophilins, particularly the chloroplast-localized ones, and warrants further experimentation.

The cyclophilins from extremophiles such as Piriformospora indica and Thellungiella halophila also offer an attractive alternative to improve stress tolerance in crop plants (Table 7) (Chen et al., 2007; Trivedi et al., 2013b,c). PiCYPA cloned from the xerophytic fungus P. indica, despite lacking the canonical RRM, demonstrated interaction with RNA. It is likely that protection against stress in the PiCYPA-overexpressing transgenic E. coli and tobacco plants might be due to its role in the stabilization of RNA transcripts (Trivedi et al., 2013b). Induction of a 17.5 kDa cyclophilin PmCYP in dinoflagellate algae Prorocentrum minimum in response to different pollutants viz., copper and polychlorinated biphenyl (Ponmani et al., 2015) further suggests that the role of cyclophilins as stress proteins is conserved. The role of cyclophilins as universal stress proteins is also substantiated by studies on Brucella, an intracellular bacterial pathogen in humans and cows which causes the disease brucellosis (Young, 1995). Comparative proteomic analysis in B. abortus resulted in the identification of two cyclophilins (CYPA and CYPB) which were differentially expressed and implicated in bacterial intracellular adaptation (Roset et al., 2013). Studies employing ΔcypAB mutants revealed that these genes were essential for virulence and tolerance to various abiotic stresses such as oxidative, acidic pH and detergents (Roset et al., 2013).

It is apparent that despite being distinct, protection against stress-induced damage is a property common to several cyclophilins (Table 7), suggesting an overlap of their roles. However, the precise mechanisms by which these proteins protect the cellular machinery against stress-induced damage are still elusive for the majority of these proteins. Although except for AtCYP38, all the cyclophilins implicated in stress tolerance are SD proteins, similar roles for the MD cyclophilins cannot be ruled out and should be the subject of future studies. Further investigations are therefore necessary to unravel the physiological implications of cyclophilins in plants that will enable their applications in crop improvement through biotechnological interventions or conventional breeding.