The Cytokinins BAP and 2-iP Modulate Different Molecular Mechanisms on Shoot Proliferation and Root Development in Lemongrass (Cymbopogon citratus)

The known activities of cytokinins (CKs) are promoting shoot multiplication, root growth inhibition, and delaying senescence. 6-Benzylaminopurine (BAP) has been the most effective CK to induce shoot proliferation in cereal and grasses. Previously, we reported that in lemongrass (Cymbopogon citratus) micropropagation, BAP 10 µM induces high shoot proliferation, while the natural CK 6-(γ,γ-Dimethylallylamino)purine (2-iP) 10 µM shows less pronounced effects and developed rooting. To understand the molecular mechanisms involved, we perform a protein–protein interaction (PPI) network based on the genes of Brachypodium distachyon involved in shoot proliferation/repression, cell cycle, stem cell maintenance, auxin response factors, and CK signaling to analyze the molecular mechanisms in BAP versus 2-iP plants. A different pattern of gene expression was observed between BAP- versus 2-iP-treated plants. In shoots derived from BAP, we found upregulated genes that have already been demonstrated to be involved in de novo shoot proliferation development in several plant species; CK receptors (AHK3, ARR1), stem cell maintenance (STM, REV and CLV3), cell cycle regulation (CDKA-CYCD3 complex), as well as the auxin response factor (ARF5) and CK metabolism (CKX1). In contrast, in the 2-iP culture medium, there was an upregulation of genes involved in shoot repression (BRC1, MAX3), ARR4, a type A-response regulator (RR), and auxin metabolism (SHY2).

We have previously reported that, when using an MS medium supplemented with BAP 10 µM, 5% sucrose, and 5 g/L gelrite TM , a high shoot proliferation was induced (23.3 shoots/ explant) [4].In contrast, when using the natural CK 6-(γ,γ-Dimethylallylamino)purine (2-iP) 10 µM, the expected activation of shoots was substantially decreased (8.8 shoots/explant) and rooting developed in one step [4].In this scenario, two different molecular mechanisms were activated under the BAP and 2-iP culture media in the lemongrass tissue culture.To understand the differences between the BAP and 2-iP effects in lemongrass, we reconstructed a network using the STRING database v11.5 [18] with the genes involved in shoot proliferation/repression, stem cell maintenance, CK signaling, cell cycle, and auxin signaling, based on the Brachypodium distachyon homologous genes present in the Arabidopsis thaliana genome (Figure 1).We analyzed 13 genes that are involved in CK signaling (AHK3, ARR1, and ARR4), stem cell maintenance (STM, REV, and CLV3), cell cycle regulation (CDKA-CYCD3 complex), auxin signaling (ARF5), CK metabolism (CKX1), shoot repression (BRC1 and MAX3), and auxin signaling repressor (SHY2) for RT-qPCR analysis.Gene expression analysis revealed that in the BAP culture medium (CM), the genes involved in shoot proliferation, such as CK signaling (AHK3, ARR1), stem cell maintenance (STM, REV, and CLV3), cell cycle regulation (CDKA-CYCD3 complex), auxin signaling (ARF5), and CK metabolism (CKX1), were higher upregulated.In contrast, in 2-iP CM, there was an upregulation of genes involved in shoot repression (BRC1 and MAX3), auxin (SHY2), and CK signaling activating the A-type response regulator (RR) ARR4.

Shoot Proliferation
Lemongrass plants grown in a medium supplemented with BAP 10 µM, 5% sucrose, and 5 g/L gelrite TM are consistent with the canonical effects of CKs developed 23.3 shoots The goal of this study was to elucidate the molecular mechanisms that BAP uses to activate shoot proliferation and 2-iP root development in lemongrass (C.citratus).

Shoot Proliferation
Lemongrass plants grown in a medium supplemented with BAP 10 µM, 5% sucrose, and 5 g/L gelrite TM are consistent with the canonical effects of CKs developed 23.3 shoots per initial explant after 2 months in culture (Figure 2A).
We performed histological analysis of longitudinal sections of roots from both CKs in which a higher number of axillary meristems (AMs) were observed in plants grown on BAP 10 µM compared to the 2-iP 10 µM CM (Figure 2C,D).
Several publications indicate that high levels of CK promote shoot growth and suppress root formation.However, our results suggest a different role in 2-iP inducing roots (Figure 2B).Plants grown on BAP 10 µM did not produce roots (Figure 2A).Instead, in plants grown on 2-iP 10 µM, the root phenotypes were thicker and pigmented, possibly due to the biosynthesis of anthocyanins (Figure 2B).

Discussion
CKs have been mainly used in plant micropropagation, including Poaceae members like cereals and grasses, due to the positive regulation of shoot and the negative regulation of root development [19][20][21].
Based on that, we aimed to understand by using a PPI network STRINGv11.5 database containing the molecular mechanisms involved in lemongrass grown in the high shoot proliferation medium (BAP 10 µM) versus the low shoot proliferation and root development medium (2-iP 10 µM) (Figures 2 and 5).
The interpretation of the molecular mechanism found in this work is as follows:

CK Signaling Genes
The expression of AHK3 and the B-type RR ARR1 were upregulated in BAP CM (2.8 and 3.5 folds, respectively) compared to 2-iP CM (0.4 and 0.3 folds, respectively), while upregulation of the A-type RR, ARR4, was found in 2-iP-treated plants (5 fold) compared to BAP (0.2 fold) (Figure 4).

Auxin Signaling Gene
The ARF5 expression was revealed to be the most highly upregulated gene in BAP CM (5.4 fold) compared to the treatment with 2-iP (0.2 fold) (Figure 4).

Auxin Signaling Repressor Gene
The analysis reveals that SHY2 was the most upregulated gene in 2-iP CM (3.5 fold) compared to the treatment with BAP (0.3 fold) (Figure 4).

CK Metabolism Gene
The CKX1 expression remains upregulated in BAP CM (3.2 fold) compared to the CM with 2-iP (0.3 fold) (Figure 4).

Shoot Proliferation Repressor Genes
The analysis revealed that the levels of BRC1 and MAX3 remained elevated in the 2-iP CM (1.7 and 1.8 folds, respectively) when compared to the BAP CM (0.6 and 0.6 folds, respectively) (Figure 4).

Discussion
CKs have been mainly used in plant micropropagation, including Poaceae members like cereals and grasses, due to the positive regulation of shoot and the negative regulation of root development [19][20][21].
Based on that, we aimed to understand by using a PPI network STRINGv11.5 database containing the molecular mechanisms involved in lemongrass grown in the high shoot proliferation medium (BAP 10 µM) versus the low shoot proliferation and root development medium (2-iP 10 µM) (Figures 2 and 5).
The interpretation of the molecular mechanism found in this work is as follows:

Cytokinin Signaling
Interestingly, we found that different molecular mechanisms between BAP and 2-iP relied on CK signaling, specifically with the type of RRs.Type-A RRs evolved with land plants, indicating that their signaling pathway and negative feedback regulation were established simultaneously to activate vascularization and root development, whereas type-B RRs were already present, advocating that they had CK-independent functions [36].
In our analysis, AHK3 and ARR1 were higher expressed in BAP than the 2-iP-containing medium and in agreement with our results, where BAP has been demonstrated to be involved in the expression/mutant restitution of type-B RRs that function as repressors of root development [27,[37][38][39][40][41][42][43] (Figures 4 and 5).
Type-B PeRR12 is a negative regulator of root development in poplar, repressing the WUSCHEL-related homeobox PeWOX5, which is involved in the maintenance of the stem cells in the root apical meristem (RAM), and PeWOX11, which interacts with the auxin signaling pathway regulating the founder cells during de novo root regeneration [44].PeRR12 also represses PePIN1 and PePIN3 transcripts, two auxin efflux carriers involved in root and shoot development [45].

Cell Cycle
The CK BAP induces the CYCD3 expression at the G1-S cell cycle phase transition, showing that CYCD3 induction may be a direct response to BAP or an indirect response of cells reaching a particular position in the cell cycle under the influence of BAP [70], where the same occurred in our results with BAP at the concentration of 10 µM (Figures 4 and 5).CYCD3-1 forms a complex with CDKA, which is required for the maintenance of undifferentiated cells in the SAM [71,72], determines the cell number and expansion in developing lateral organs (shoots), and mediates CK effects in apical growth and development [73] (Figure 5).

BRC1 Shoot Repressor
In our results, we found that BAP activates poorly BRC1 compared to 2-iP (Figures 4  and 5), as described by Xu et al. [74], in which the formation of mutant phenotypes in Panicum virgatum L. is rescued via the BAP application.In another example, the work of Lewis et al. [75] showed that the overexpression of BRC1 results in a reduced number of shoots.BRC1 encodes a shoot growth inhibitory TF [76], which in 2-iP (10 µM), its relative expression was elevated compared to BAP (10 µM) and the associated plant phenotypes show fewer AMs, but with developed roots (Figure 2).

SHY2 (IAA3/Short Hypocotyl 2)
CKs cooperate with other plant growth regulators to regulate root meristem devel- Type-B PtRR13 is a negative regulator of root development in Populus.A transcriptome analysis showed that it altered the expression levels of RING1, a negative regulator of vascularization, PDR9, an auxin efflux transporter, and also two apetala/ethylene responsive factors [46].
In our analysis AHK3 was expressed lower and ARR4 higher in the 2-iP-containing medium.2-iP has demonstrated to be involved in the expression of type-A RRs that function as activators of root development (Figures 4 and 5).Roots developed in 2-iP were pigmented and four times thicker than normal lemongrass roots.The histological analysis from RT revealed that they contain a prominent root cap (RC) composed by several layers of cells containing statocytes (statenchyma), CX, PD, QC, DM, P, MZ, and EZ.
When cultured in a 2,4-D-containing medium, the explants of roots from 2-iP were capable of developing the somatic embryogenesis process while in normal roots, it was not possible.
However, the BAP-containing medium activates B-type RRs, TFs involved in shoot proliferation, and root inhibition.
The overexpression of type-A RRs in Arabidopsis substantially increased lateral root development and inhibits shoot development in ARR3,-5,-6,-16 and -17 when supplemented with 2-iP, where mutated type-A RRs led to reduced root architecture [47].The RcRR1, a type-A RRs of Rosa canina, has been found to be involved in CK-modulated rhizoid organogenesis [48].
Successful rooting from shoots using 2-iP has been observed in date palm [49], amaryllis [50], sweet cherry [51], and apple [52,53].Moreover, 2-iP is involved in nitrogen signaling and regulates root architecture and development by modulating polar auxin transport and vascular patterning in the root meristem [54].
CK signaling loss correlates with a reduced meristem size, whereas enhanced cytokinin action stimulates meristem activity.The shoot architecture is normally promoted in the stem cells located in the central zone of the shoot apical meristem (SAM) [55][56][57].
CLV3 is involved in maintaining the stem cell population in the SAM and is central to continuing shoot growth [68].The TF REV gene regulates the STM expression [67] and is an activator of shoot proliferation, where REV mutants develop fewer shoots [69].

Cell Cycle
The CK BAP induces the CYCD3 expression at the G 1 -S cell cycle phase transition, showing that CYCD3 induction may be a direct response to BAP or an indirect response of cells reaching a particular position in the cell cycle under the influence of BAP [70], where the same occurred in our results with BAP at the concentration of 10 µM (Figures 4 and 5).CYCD3-1 forms a complex with CDKA, which is required for the maintenance of undifferentiated cells in the SAM [71,72], determines the cell number and expansion in developing lateral organs (shoots), and mediates CK effects in apical growth and development [73] (Figure 5).

BRC1 Shoot Repressor
In our results, we found that BAP activates poorly BRC1 compared to 2-iP (Figures 4 and 5), as described by Xu et al. [74], in which the formation of mutant phenotypes in Panicum virgatum L. is rescued via the BAP application.In another example, the work of Lewis et al. [75] showed that the overexpression of BRC1 results in a reduced number of shoots.BRC1 encodes a shoot growth inhibitory TF [76], which in 2-iP (10 µM), its relative expression was elevated compared to BAP (10 µM) and the associated plant phenotypes show fewer AMs, but with developed roots (Figure 2).

SHY2 (IAA3/Short Hypocotyl 2)
CKs cooperate with other plant growth regulators to regulate root meristem development; for example, the SHY2 gene is an essential axis of the interaction between CKs, auxin, and brassinosteroids (BR) [77,78]; it is therefore upregulated after the CM with 2-iP (Figure 5).
The expression of SHY2 upon 2-iP CM negatively regulates polar auxin transport carried out via PIN, redistributes the auxin concentration, and induces cell proliferation.Alternatively, auxin and CK are regulated via antagonism in the meristematic and differentiation zones of the roots to control the meristem size [79].

Strigolactone (SL) Biosynthesis
MAX3 and BRC1, which are involved in SL biosynthesis and repress shoot proliferation, were upregulated in 2-iP CM (Figures 4 and 5).The exogenous supply of SL can upregulate the BRC1 expression independently of any new protein synthesis, suggesting that BRC1 is the most direct target of SL signaling [80,81].MAX3 transcript profiling revealed that the upregulation of SL biosynthetic genes was indirectly triggered by CKs and led to the activation of the BRC1 expression [82].

CK Oxidase/Dehydrogenase
Exposure to 10 µM BAP caused the induction of the CK oxidase/dehydrogenase gene CKX1, which was also reported in the work of Vylíčilová et al. [83], suggesting that BAP has an independent mode of action or may act as a negative regulator of CK perception by upregulating specific genes, such as CKX1 to balance the endogenous CK concentration (Figures 4 and 5).

Auxin Response Factor
ARF5 is an auxin response factor that regulates auxin-responsive elements (Figure 5).It plays a crucial role in plant development, especially in embryonic development, cell differentiation, and organogenesis [84].Mutations result in a defective embryo and SAM development [84].

Histology of C. citratus Shoots and Roots
Shoots and roots of C. citratus were randomly selected (10 samples of BAP CM and 10 samples of 2-iP CM for shoot histology; 10 samples of 2-iP CM and 10 samples of control CM for root histology).These were fixed in FAE (5% formaldehyde, 10% acetic acid, and 50% ethanol), followed by dehydration in a series of ethanol dilutions (20%, 40%, 60%, 80%, and 100% ethanol) for 2 h each.Shoot samples were embedded in Technovit 7100 (Heraeus Kulzer, Hesse-Hanau, Deu) and root samples in a mix of acetonitrile/Spurr EMS (Hatfield, UK) following the manufacturer's instructions.Shoot sections (14 µm) were obtained on a rotary microtome (Reichert-Jung 2040, Leica (Hernalser Hauptstrasse, Vienna); and the root semithin sections (500 nm) with a Leica EM UC7 ultramicrotome Leica (Hernalser Hauptstrasse, Vienna).Tissue sections were stained with a 0.02% toluidine blue solution (HYCEL, Zapopan, Mexico) for 3 min, washed with distilled water for 1 min, and then air dried [86].Shoot images were taken with a DM6000B microscope (Leica) and root photos were taken with a DMi8 inverted microscope (Leica).The open source image processing software FIJI (Fiji Is Just ImageJ), based on ImageJ2 [87], was used.

Isolation of RNA and Real Time Gene Expression Analysis
Total RNA was isolated from the shoots derived from cultures containing BAP and/or 2-iP using Trizol (Invitrogen, Carlsbad, CA, USA).The RNA concentration was measured using its absorbance at 260 nm, where a ratio of 260 nm/280 nm was assessed and its integrity was confirmed via electrophoresis in agarose 1% (w/v) gels.The samples of cDNA were amplified via RT-qPCR using Maxima SYBR Green/ROX qPCR (Thermo Scientific, Waltham, MA, USA) in a Real-Time PCR System (CFX96 BioRad).Elongation factor 1α (EF1α), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and actin (ACT) were used as reference genes for qPCR normalization [88] (Table 2).* The sequences of the reference genes used were obtained from Meena et al. [88].
Retfinder, NormFinder, Bestkeeper, and Delta-Ct analyses were used.Three replicates were created for each of these reference genes, where the relative expression and weighted Ct. were calculated.Next, a Delta-Ct of each gene and the relative amount of target gene expression were analyzed using the 2 −∆∆Ct method [89].RT-qPCR analysis was based on at least three biological replicates with three technical repetitions for each sample and the control treatment.
The gene identifier was made according to UniProt (http://www.uniprot.org)and NCBI (http://www.ncbi.nlm.nih.gov)databases.Sequences of B. distachyon genes were analyzed using blastN and blastP.Oligonucleotides were designed for qPCR (2 −∆∆Ct method analysis) gene expression or transcriptional analysis.

Conclusions and Perspectives
In lemongrass (C.citratus), shoots developed in the BAP-containing medium drive the mechanisms to activate gene expression for AMs; stem cell maintenance mediated by STM, REV, and CLV3; CK metabolism CKX1; CK signaling mediated by AHK3, ARR1; cell cycle progression by CDKA-CYCD3; and auxin signaling by ARF5.
Plants developed in 2-iP showed higher expression levels in the genes involved in root development; MAX3; BRC1, a major shoot repressor; SHY2 and ARR4.
The molecular mechanisms of BAP activating ARR1 (B-type RRs) to produce shoots and repress root development are in agreement with the high shoot proliferation in C. citratus.
The molecular mechanisms of 2-iP activating ARR4 (A-type RRs) to produce roots and repress shoot proliferation are in agreement with the 2-iP phenotype showed in C. citratus.
Further experiments on transcriptomic/proteomic/metabolomics analysis are required to satisfactorily corroborate our findings.

Figure 1 .
Figure 1.Protein-Protein interaction (PPI) network of genes involved in cytokinin (CK) signaling, stem cell maintenance, cell cycle regulation, auxin signaling, auxin repression, CK metabolism, and shoot proliferation/repressor in C. citratus.This PPI network was reconstructed from a gene network with an average confidence of 0.4 in STRING (https://string-db.org/,v11.5).

Figure 5 .
Figure 5. Molecular mechanisms describing proliferation and inhibition of shoots and root development in lemongrass (C.citratus) under BAP-and 2-iP-containing medium.Red indicators represent inhibition effects, and the green arrows represent activations.Green elements are the components of protein interactions, yellow elements represent biological functions, blue elements represent plant hormones, red elements are branching repressors, white elements represent auxin influence, orange elements represent the activation of type-A and -B response regulators (RRs) via CK signaling, and light blue element represents strigolactone (SL).

Figure 5 .
Figure 5. Molecular mechanisms describing proliferation and inhibition of shoots and root development in lemongrass (C.citratus) under BAP-and 2-iP-containing medium.Red indicators represent inhibition effects, and the green arrows represent activations.Green elements are the components of protein interactions, yellow elements represent biological functions, blue elements represent plant hormones, red elements are branching repressors, white elements represent auxin influence, orange elements represent the activation of type-A and -B response regulators (RRs) via CK signaling, and light blue element represents strigolactone (SL).

Table 1 .
Nuclei and nucleolus size from root-tip sections of plants derived from 2-iP and controls without hormones.

Table 2 .
Primer design of genes that were analyzed during shoot and root induction of lemongrass (C.citratus).