Proline affects the size of the root meristematic zone in Arabidopsis

We reported previously that root elongation in Arabidopsis is promoted by exogenous proline, raising the possibility that this amino acid may modulate root growth.

Biancucci et al BMC Plant Biology (2015) 15:263 DOI 10.1186/s12870-015-0637-8 RESEARCH ARTICLE Open Access Proline affects the size of the root meristematic zone in Arabidopsis Marco Biancucci, Roberto Mattioli, Laila Moubayidin, Sabrina Sabatini, Paolo Costantino and Maurizio Trovato* Abstract Background: We reported previously that root elongation in Arabidopsis is promoted by exogenous proline, raising the possibility that this amino acid may modulate root growth Results: To evaluate this hypothesis we used a combination of genetic, pharmacological and molecular analyses, and showed that proline specifically affects root growth by modulating the size of the root meristem The effects of proline on meristem size are parallel to, and independent from, hormonal pathways, and not involve the expression of genes controlling cell differentiation at the transition zone On the contrary, proline appears to control cell division in early stages of postembryonic root development, as shown by the expression of the G2/M-specific CYCLINB1;1 (CYCB1;1) gene Conclusions: The overall data suggest that proline can modulate the size of root meristematic zone in Arabidopsis likely controlling cell division and, in turn, the ratio between cell division and cell differentiation Keywords: Proline, Root meristem, Arabidopsis, SHY2, CYB1;1, Plant hormones, Cell cycle genes Background Thanks to its unique cyclic structure and physicalchemical properties, proline is of paramount importance in plants, both as building block for protein synthesis and as a compatible osmolyte accumulating during, and protecting from, environmental stress It is synthesized in the cytosol from glutamate in a two-step pathway catalyzed by δ-pyrroline-5-carboxylate synthetase (P5CS), and δ-pyrroline-5-carboxylate reductase (P5CR) The first enzyme of this pathway, catalyzing the rate-limiting step of proline synthesis in higher plants, is encoded in Arabidopsis by two paralog genes P5CS1 and P5CS2, while a single gene, P5CR, encodes the second committed enzyme of proline synthesis in plants [1] In the last years it has been increasingly evident that the amino acid proline, in addition to its role in protein synthesis and stress response, plays a key role in plant development, particularly in developmental processes related to reproduction [2], such as flowering [3–6], pollen development [7, 8] and embryogenesis [2, 9] Accordingly, Arabidopsis mutants carrying a knock out T-DNA insertion in P5CS2 (FLAG_139H07, GABI_ 452G01) are embryo lethal in homozygosis, and can be * Correspondence: maurizio.trovato@uniroma1.it Dipartimento di Biologia e Biotecnologie, Sapienza, Università di Roma, P.le Aldo Moro 5, 00185 Rome, Italy propagated only as heterozygotes, unless complemented by exogenous proline [2, 9] Furthermore, Arabidopsis mutants homozygous for p5cs1 and heterozygous for p5cs2 (p5cs1 p5cs2/P5CS2) are late flowering [2] and male sterile [7, 8] A more general role as signal molecule involved in plant development, however, might also be assigned to proline on the basis of the claim that micromolar concentrations of exogenous proline promote root growth [6] Intriguingly, the first indications of a role of proline in plant development, beyond protein synthesis and stress adaptation, came from the study of the adventitious roots induced by the soil bacterium Agrobacterium rhizogenes [10] Virulent strains of this bacterium harbor a plasmid capable to transfer to, and integrate in the plant genome a portion of its own DNA, called T-DNA The expression of some of the genes borne on this transferred DNA, notably rolA, rolB, rolC and rolD, are responsible of hairy root insurgence and elongation This latter, by insertional mutagenesis [11], has been specifically attributed to rolD, later on recognized as a proline-producing ornithine cyclodeaminase gene [4], providing a direct correlation between proline availability and root growth Moreover, proline was found, at low water potential, to accumulate preferentially in the root meristem growth zone of the maize primary root [12, 13] © 2015 Biancucci et al Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated Biancucci et al BMC Plant Biology (2015) 15:263 In recent years our understanding of the genetic and molecular mechanisms underlying root growth and development has tremendously improved thanks to the exploiting of the model species Arabidopsis thaliana The simplicity of its cellular organization, the possibility to be grown on agar plates under well-defined conditions, and the wealth of genetic and molecular resources available for Arabidopsis, have greatly contributed to build a solid picture of the molecular mechanisms behind growth and development of the Arabidopsis root [14] The dimension of the root meristematic zone, which relies on the ratio between cell division in the meristem region, and cell differentiation in the transition zone (TZ), is pivotal for postembryonic root growth, and is regulated by the plant hormones auxin, cytokinin and gibberellin which, in turn, control a short regulatory circuit converging on the gene SHY2 [15] According to the current model [15], SHY2 is induced by the cytokinin-responsive transcription factors ARR1 and ARR12, and regulates the size of the root meristem by downregulating the PIN-FORMED (PIN) genes that encode auxin efflux facilitators In addition to plant hormones, however, novel effectors have been recently proposed to affect root meristem size in Arabidopsis [16–18] and others are likely to be found, as plants, being sessile organisms, must be able to respond to a multiplicity of different stimuli To test the hypothesis that proline may be one of such effectors, we used a combination of genetic, molecular and pharmacological analyses to study the growth of the primary root in proline-deficient mutant p5cs1 p5cs2/P5CS2, compared to wild type Here we show that proline can modulate the size of root meristematic zone in Arabidopsis by controlling cell division and, in turn, by modulating the ratio between cell division and cell differentiation Page of 14 Results Proline stimulates growth of the root meristematic zone We reported previously that root elongation in Arabidopsis is promoted by micromolar concentrations of exogenous proline [2] In order to verify whether a proline-deficient mutant is hampered in root growth, we analyzed the length, relative to wild type, of roots from the proline-deficient partial double mutant p5cs1 p5cs2/ P5CS2 [2] from to 12 days after germination (dag) The proline content of this partial double mutant was measured at and 14 dag in roots, confirming that this mutant contains, on average, one fourth as much proline as a wild type (0.050 ± 0.03 compared to 0.23 ± 0.02 μmoles/g (fresh weight), for proline-deficient mutants and wild types, respectively; p < 0.01) As shown in Fig 1a to c, from dag on, the roots of these mutants are shorter than wild type supporting the notion that proline stimulates root elongation To further verify the correlation between proline and root growth, we analyzed the proline content and the length of roots from heterozygous p5cs2/P5CS2 and homozygous p5cs1 parental lines, compared to partial double mutant p5cs1 p5cs2/P5CS2 and wild type lines The proline content of the parental lines turned out to be intermediate between p5cs1 p5cs2/P5CS2 and wild type lines, with measured values of 0.15 ± 0.05 μmoles/g of proline for homozygous p5cs1 roots, and 0.11 ± 0.02 μmoles/g of proline for heterozygous p5cs1/P5CS2 roots In spite of the reduction in proline content, roots from homozygous p5cs1 mutants appeared indistinguishable from wild type roots, while roots from heterozygous p5cs1/P5CS2 looked slightly shorter (not shown) suggesting that the levels of endogenous proline present in these mutants are reduced, relative to wild type, but still sufficient (p5cs1) or nearly sufficient (p5cs2/P5CS2) to sustain normal root growth Fig Proline specifically modulates root growth a-b Roots from wild type (a) and p5cs1 p5cs2/P5CS2 (b) grown on agar plates for dag c Primary root lengths of p5cs1 p5cs2/P5CS2 (orange line) and wild type (blue line) plotted over time from to 12 dag The data are means ± SE of at least 90 samples from independent experiments Biancucci et al BMC Plant Biology (2015) 15:263 Overall these data confirm the positive correlation between proline content and root growth Clearly mutations on P5CS2 have a stronger effect on proline accumulation and root growth, compared to mutations on P5CS1 However both genes seem to contribute to the overall proline content in roots, as indicated by the lower proline level and by the shorter roots exhibited by the p5cs1 p5cs2/ P5CS2 partial double mutant, compared to parental lines In Arabidopsis, the maintenance of the root meristematic zone and, consequently, of the root growth is ensured by the balance between the rate of cell division in the root meristematic zone and the rate of cell differentiation in the TZ [15, 19] To establish whether the reduction in root length of the proline-deficient mutant may derive from a reduction in meristem size, we measured, in p5cs1 p5cs2/ Page of 14 P5CS2 mutant and in wild type, the size of the root meristematic zone expressed as number of cortex cells spanning from the quiescent center (QC) to the first elongated cell in the TZ As shown in Fig 2, from 2a to 2g, and in Fig 2j, the shorter roots of p5cs1 p5cs2/P5CS2 are accounted for by smaller meristems that stop growing at dag with an average number of cells of 16.4 ± 0.47 (Fig 2j) The wild-type meristem, by contrast, reaches the balance between dividing and differentiating cells between and dag (Fig 2j) with an average number of cells of 28.3 ± 0.33 (p < 0.001; wild type v/s p5cs1 p5cs2/P5CS2) To confirm the effect of proline on the size of the root meristematic zone, we scored the number of meristem cells in wild-type roots, grown either in presence or in absence of proline (Fig 2j), at the optimal concentration of 10 μM as inferred Fig Proline-deficient mutants have meristems smaller than wild types a-f Root meristems from wild type (a, c, e) and p5cs1 p5cs2/P5CS2 (b, d, f), at (a, b), (c, d) and dag (e, f) Bottom black arrowheads indicate the QC, top black arrowheads indicate the cortex TZ g Root meristem from p5cs1 p5cs2/P5CS2 treated, at dag, with 10 μM exogenous proline h-i Wild-type root, at 10 dag, treated with 10 μM exogenous proline (i) compared with an untreated control (h) Bars = 20 μm (a-i) j Root meristem cell number of plants described in (a) to (i) plotted over time from to 10 dag The data are the means ± SE of at least independent experiments A minimum of 50 roots per line was analyzed at each time point Biancucci et al BMC Plant Biology (2015) 15:263 by the dose-response curves shown in Additional file 1: Figure S1 Proline treatment (Fig 2j) significantly (p < 0.001) increased meristem size from to 10 dag, with an average number of 37.3 ± 0.54 cells, as compared to 28.5 ± 0.42 of the untreated controls (Fig 2h and i) More importantly, the addition of exogenous proline was able to fully complement the reduction in meristem size of p5cs1 p5cs2/P5CS2 roots (Fig 2g and j) Moreover, in the meristem cells of homozygous p5cs1 and heterozygous p5cs2/P5CS2 roots, we scored an average number of 28.1 ± 0.41 and 25.5 ± 0.21 cells, respectively, in good correlation with their measured proline levels of 0.15 ± 0.05 and 0.094 ± 0.08 μmoles/g To further investigate the specificity of proline to modulate meristem size and to rule out a generic stimulatory effect of amino acids as a source of supplemental nitrogen, we analyzed, in 7-days-old wildtype roots, the effects of different amino acids on the size of the root meristem Wild-type seedlings were grown on different Petri dishes, each one supplemented with one of the amino acids shown in Fig 3e, at the concentration of 10 μM As shown in Fig (a to e), most of the tested amino acids (tyrosine, arginine, tryptophan, glycine, histidine, threonine and leucine) had no significant effect on the size of the root meristem size Two amino acids (methionine, asparagine), however, had stimulatory effects on root meristem size (Fig 3b, c and e), and one amino acid Page of 14 (glutamic acid) caused a reduction of root meristem size (Fig 3d and e) Overall, these experiments indicate that amino acids not have per se a generic stimulatory effect on meristem size, and that proline and few others amino acids may have a special role as metabolic or signaling molecule To additionally validate the specific effects of proline on root growth, we analyzed, at dag, the total protein profile of root tips (Additional file 2: Lane and 4) of either wild type or p5cs1 p5cs2/P5CS2 plants As shown in supplemental Fig we found no significant difference in the accumulation of total proteins between wild type and p5cs1 p5cs2/P5CS2 mutants indicating that the difference in root length and root meristem size between p5cs1 p5cs2/P5CS2 and wild type are not caused by gross variations in protein accumulation The effect of proline on root meristem size is independent from hormone action Since root growth is regulated by the combined action of auxin, cytokinin and gibberellin, we searched for possible interactions between proline and these hormones As a first approach, we analyzed the size of the root meristematic zone of p5cs1 p5cs2/P5CS2, compared to wild-type plants, upon exogenous treatment with either gibberellin (GA), indol-3-acetic acid (IAA) or cytokinin As described previously [15, 19], supplementation of either GA or IAA to wild-type roots results in larger Fig Effects of different amino acids on root meristem size a-d Wild type roots at dag treated with (b) 10 μM asparagine (Asp), (c) 10 μM methionine (Met), (d) 10 μM glutamate (Glu) compared to wild type (a) Bottom black arrowheads indicate the QC, top black arrowheads indicate the cortex TZ Bars = 20 μm e Bar plot showing the effects of 10 amino acids on the size of a wild-type root meristem The amino acids were supplied in vitro at the concentration of 10 μM and the root meristem cells were scored at dag Apart from Asp, Met and Glu treatment, which led to meristems significantly larger (Asp, Met), or smaller (Glu) than untreated wild type meristems, all the other amino acids produced no effect on meristem size when supplied exogenously Error bars indicate Standard Error (SE) The data are the means ± SE of at least independent experiments Significance levels for each amino acid treatment were calculated, relative to untreated controls, with a paired Student’s t-Test p*** < 0.001; p** < 0.01 Biancucci et al BMC Plant Biology (2015) 15:263 Page of 14 Fig Proline effects on meristem size are independent from GA, IAA and cytokinin a Root meristem sizes, measured at dag as number or cortex cells spanning from the QC to the TZ, of wild types (dark grey bars) and p5cs1 p5cs2/P5CS2 (light grey bars), upon pharmacological treatment with either 20 μM GA3, 0.1 nM IAA, or 10 μM kinetin Error bars indicate Standard Error (SE) All pairwise comparisons (Student’s t test) showed that proline mutants treated with either GA3- or IAA, had root meristems highly significantly larger than untreated mutants (p < 0.001 +++), and highly significantly smaller than treated wild types (p < 0.001***) Kinetin-treated proline mutants, however, had root meristems highly significantly smaller than either treated mutants (p < 0.001***) , but only significantly smaller than untreated wild types (p < 0.01 ++) b Root meristem sizes, at dag, of genetic combinations (light grey bars) mimicking either GA3 treatment (p5cs1 p5cs2/P5CS2, gai-t6, rga-24), or kinetin treatment (either p5cs1 p5cs2/P5CS2, arr1 and p5cs1 p5cs2/P5CS2, arr12), compared to parental lines (dark grey bars) c qRT-PCR analysis shows no statistically significant differences in the expression levels of ARR1, ARR12, GAI and RGA between p5cs1 p5cs2/P5CS2 (light grey) and wild-type controls (dark grey) ACTIN was used as a reference gene to normalize the qRT-PCRs meristems, while cytokinin produces smaller meristems (Fig 4a) After treatment with either GA or IAA the root meristematic zone of p5cs1 p5cs2/P5CS2 appears significantly larger than untreated controls (p