PSK-α promotes root growth in Arabidopsis


Author for correspondence:
Margret Sauter
Tel: +49 431 880 4210
Fax:+49 431 880 4222


  • • Phytosulfokine-α (PSK-α) is a disulfated pentapeptide described to act as a growth factor in suspension cells. In this study, the involvement of PSK signaling through the PSK receptor gene AtPSKR1 in Arabidopsis root growth was assessed.
  • • Expression studies of PSK precursor genes and of AtPSKR1 were performed in roots with RT-PCR and P:GUS analyses. Root elongation, lateral root formation, cell production and root cell elongation were analyzed in wild-type (wt) and in the receptor knockout mutant Atpskr1-T treated with or without synthetic PSK-α.
  • • Phytosulfokine and AtPSKR1 genes are differentially expressed in roots. PSK-α induced root growth in a dose-dependent manner without affecting lateral root density. Kinematic analysis established that enhancement of root growth by PSK-α was mainly caused by an increase in cell size. In Atpskr1-T, the primary roots were shorter as a result of reduced mature cell size and a smaller root apical meristem composed of fewer cells than in wt.
  • • The results indicate that PSK-α signaling through AtPSKR1 affects root elongation primarily via control of mature cell size. Root organogenesis, on the other hand, is not controlled by PSK-α.


The rate at which plant roots grow is an adaptive trait. This is obvious in natural ecotypes which are adapted to different environmental conditions. Up to fourfold differences in primary root growth rates were observed when 18 ecotypes of Arabidopsis thaliana were compared (Beemster et al., 2002). Direction and rate of root growth are also influenced by external factors, such as gravity, water and nutrients. The rate at which roots grow is determined by two parameters: cell production rate and mature cell length. The cell production rate in the meristem in turn depends on the number of meristematic cells and the average duration of a cell cycle. An increase in cell division rate can thus directly promote root growth. Accelerated root elongation rates during Arabidopsis development were shown to be almost exclusively the result of increased cell production in the meristem, with little change in cell expansion rates (Beemster & Baskin, 1998). In roots, cell division occurs in the apical meristem but also in pericycle cells that initiate formation of lateral roots, and in lateral root primordia. Cell growth, on the other hand, also occurs in meristems and in the elongation zone proximal to the meristem (Beemster & Baskin, 1998).

Phytosulfokine-α (PSK-α) is an autocrine growth factor found in cell cultures from dicot and monocot plants where it promotes proliferation of cells kept at low density (Yang et al., 2000). PSK-α is perceived by leucine-rich repeat (LRR) receptor kinases conserved in carrot (Matsubayashi et al., 2002) and Arabidopsis (Matsubayashi et al., 2006). In Arabidopsis, two PSK-α receptors were identified based on the homology towards the receptor from carrot. They were named AtPSKR1 (encoded by At2g02220) and AtPSKR2 (encoded by At5g53890). [3H]PSK-α binding activity was enhanced in microsomal fractions of Arabidopsis cells overexpressing AtPSKR1 (Matsubayashi et al., 2006) and a genetic approach confirmed the redundant role of AtPSKR2 in PSK-α signaling (Amano et al., 2007). Genes encoding PSK precursors appear to be ubiquitously expressed in higher plant organs (Yang et al., 2001; Lorbiecke & Sauter, 2002; Matsubayashi et al., 2006). However, detailed analysis in reproductive tissues of Zea mays L. revealed cell type-specific and gene-specific expression patterns for several members of the maize PSK precursor gene family (Lorbiecke et al., 2005). Some of the sites of PSK precursor gene expression identified in planta coincided with functions of PSK-α predicted from cell culture experiments. For instance, germination of pollen from Nicotiana tabacum L. kept at low density was promoted by PSK-αin vitro in a concentration-dependent manner (Chen et al., 2000). It was coincidentally found in maize (Lorbiecke et al., 2005) and Arabidopsis (A. Kutschmar & M. Sauter, unpublished) that pollen contains high amounts of PSK-α precursor mRNA. Since PSK-α was shown to act as a growth factor in tissue-cultured cells, the question arose as to whether it may play a growth regulatory role in planta as well. Matsubayashi et al. (2006) provided evidence that root elongation was enhanced in AtPSK4 overexpressing Arabidopsis seedlings. Insertional knockout of the receptor gene AtPSKR1, on the other hand, led to reduced root elongation. In this study, the putative sites of PSK-α activity in roots were analyzed in detail by the use of promoter:Gus lines of the five PSK genes present in the Arabidopsis genome and of the AtPSKR1 receptor gene. To specify the effect of PSK-α on root growth, a detailed analysis of main and lateral root elongation and of lateral root initiation was performed. Finally, cell elongation was compared in wt and Atpskr1 mutant treated with or without PSK-α. Our findings indicate that PSK-α signaling through AtPSKR1 regulates root growth mainly by controlling cell size.

Materials and Methods

Plant growth

Experiments were performed with Arabidopsis thaliana (L.) Heynh. ecotype Columbia (Col-0). Unless stated otherwise, plants were grown in a 1 : 1 sand : humus mixture that was frozen at –80°C for 2 d to avoid insect contamination and watered regularly with tap water. Before germination, seeds were stratified at 4°C in the dark for 2 d and then transferred to a growth chamber under long-day conditions.

For growth experiments on sterile plates, Arabidopsis seeds were surface-sterilized for 10 min in 1 ml 0.1% (w/v) sodium hypochlorite, washed five times with autoclaved water and laid out under sterile conditions on square plates containing half-concentrated modified Murashige and Skoog-medium (Murashige & Skoog, 1962) supplemented with 1.5% sucrose and PSK-α, dsPSK-α, or a mixture of amino acids present in dsPSK at the concentrations indicated (1 µm of amino acid mixture contained 2 µm Tyr, 1 µm Ile, 1 µm Thr, and 1 µm Gln). The media were solidified with 0.8% agarose. Plants were grown under long-day conditions.

Atpskr1-T mutant analysis

The mutant line SALK_008585 of A. thaliana ecotype Columbia-0 was obtained from the NASC (Nottingham Arabidopsis Stock Centre, University of Nottingham, Nottingham, UK). It carries a T-DNA insertion in the AtPSKR1 gene At2g02220 and was therefore termed Atpskr1-T to distinguish it from the Atpskr1-1 knockout mutant previously described in the ecotype Landsberg erecta (Matsubayashi et al., 2006). Insertion of the T-DNA in At2g02220 was confirmed by PCR. Effective knockout of gene expression from At2g02220 in Atpskr1-T was tested through reverse transcription of mRNA using oligo-dT primers and subsequent PCR amplification using the AtPSKR1-specific forward primer 5′-GAGCGTTGCAATACAATCAG-3′ and reverse primer 5′-CAGTACTTACATGCGTCTCGT-3′.

Growth measurements

Lengths of main and lateral roots were measured using a binocular (Olympus, Hamburg, Germany) and analyzed with Image J (National Institute of Health, Bethesda, MD, USA). For PCycB1;1:GUS plants, the number of lateral roots, Gus-expressing lateral root primordia and division active pericycle cells was determined and lateral root density was calculated per millimeter main root. Shoot and root fresh weights were measured with a precision balance (Sartorius, Göttingen, Germany).

Cell size measurements

Using the method described by Beemster et al. (2002), we established cell length profiles in the root tip of 5-d-old wt and Atpskr1-T seedlings. However, instead of using differential interference contrast microscopy, we used fluorescence microscopy to visualize cell outlines stained with FM4-64 (Invitrogen, Karlsruhe, Germany), a dye that specifically inserts into the plasma membrane. In our hands, FM4-64 proved to be more reliable than propidium iodide, which failed to stain cell walls in the apical root meristem in a reproducible manner. Pictures were taken with a TCS SP confocal laser scanning microscope (Leica, Bernsheim, Germany). Using the software LCS Lite (Leica), cell size was measured in cortical cell files that could be continuously followed from the QC up to a distance of at least 600 µm from the QC. In order to be able to average cell size as a function of its distance from the QC, each cell length profile was smoothed and interpolated by means of a kernel smoothing algorithm essentially as described by Beemster & Baskin (1998) but using the locpoly function in the statistical package R (

Cloning of promoter:GUS vectors, transformation, and histochemical Gus detection

For expression analysis of PSK genes and of the PSK receptor gene AtPSKR1 by promoter:Gus analysis, genomic fragments were amplified upstream of the start codon by PCR as follows. For AtPSK1 (At1g13590), the forward primer 5′-CACCCTATACTCCACCTCCCAAATTGA-3′ and the reverse primer 5′-CAACACTTCACTTTTCGTCTTCA-3′ amplified a fragment of 1121 bp. For AtPSK2 (At2g22860), a 1229 bp genomic fragment was amplified with the forward primer 5′-CACCGACTGAGTTACGATTGAATCAC-3′ and the reverse primer 5′-GACGGGTTTGGCTTCTTTGT-3′. For AtPSK3 (At3g44735), a 1102 bp genomic fragment was amplified with the forward primer 5′-CACCTATAGCATTAAGGGACAGAAGTGG-3′ and the reverse primer 5′-GTGAGAAGTTTGTGATAGACAGATAC-3′. For AtPSK4 (At3g49780), a 1116 bp genomic fragment was amplified with the forward primer 5′-CACCGTGTAAGAAGACTTTACTAACCT-3′ and the reverse primer 5′-AGCAGATGAAGCCAGTTAGG-3′. For AtPSK5 (At5g65870), a 1198 bp genomic fragment was amplified with the forward primer 5′-CACCTATGATATCACAGGTCTTTC-3′ and the reverse primer 5′-GAAGAGGAAGTTATGAAATAGAGG-3′, and for AtPSKR1 (At2g02220) a 1363 bp genomic fragment was amplified with the forward primer 5′-CACCTCGTCTACAACTCACATACGTTACACT-3′ and the reverse primer 5′-CGATGAACACGCATTTCAAGAACAGAG-3′. The fragments were cloned into the vector pBGWFS7 (VIB, Gent, Belgium) in front of the β-glucuronidase coding sequence using the Gateway system (Invitrogen). Arabidopsis plants were transformed with Agrobacterium tumefaciens. Selection of transformed plants was performed by spraying with 200 µm BASTA (AgroEvo, Berlin, Germany). For each construct, several Gus-expressing lines were analyzed. Gus assays were performed as described (Vielle-Calzada et al., 2000). Stained plant material was cleared in chloral hydrate (Yadegari et al. 1994) and observed under bright-field illumination using Leica DM LS (Leica) or Olympus BX41 (Olympus, Hamburg, Germany) microscopes. To obtain tissue sections, 14-d-old cleared plants were washed and dehydrated using an ethanol series at 30, 50, 70, 90 and 100% for 60 min each. Plants were embedded in Technovit 7100 according to the manufacturer's instructions (Heraeus Kulzer, Wehrheim, Germany). Sections 10 µm thick were cut with a Leica RM 2255 microtome and analyzed using a Leica DM LS microscope (Leica, Bensheim, Germany).

Reverse transcription-polymerase chain reaction

Total RNA was isolated from shoots and roots of 8-d-old seedlings and 14-d-old plants with Tri-Reagent following manufacturer's instructions (Sigma-Aldrich, Steinheim, Germany). A 5 µg quantity of DNAse-treated RNA (MBI Fermentas, Leon-Roth, Germany) was reverse-transcribed with oligo-dT primers (Invitrogen) and used for PCR amplification. AtActinII was amplified with gene-specific primers and used as a control. For PCR amplification of AtPSK1, AtPSK2, AtPSK3, AtPSK4 and AtPSK5 genes, cDNA from 0.5 µg RNA was used; for amplification of AtPSKR1 cDNA from 1 µg RNA was used; and for amplification of AtActinII, cDNA from 0.05 µg RNA was used. To be able to distinguish between fragments derived from cDNA or genomic DNA, gene-specific oligonucleotides were positioned to include an intron, except for the intronless AtPSKR1. Genomic DNA was isolated and PSK precursor genes and AtPSKR1 were amplified as an additional control. For AtPSK1 (At1g13590), the forward primer 5′-ATG GCT TCA AGT GTT ATT TTA AGA G-3′ and the reverse primer 5′-ATC TTG TGT ATA GAT GTA ATC GGT G-3′ amplified a cDNA fragment of 203 bp and a genomic DNA fragment of 574 bp. For AtPSK2 (At2g22860), the forward primer 5′-CAC CAT GGC AAA CGT CTC CGC-3′ and the reverse primer 5′-TCA AGG ATG CTT CTT CTT CTG GGT-3′ amplified a cDNA fragment of 262 bp and a genomic DNA fragment of 409 bp. For AtPSK3 (At3g44735), the forward primer 5′-ATG AAG CAA ACC TTG TGC C-3′ and the reverse primer 5′-TCA ATG CTT ATG GTG CTG TG-3′ amplified a cDNA fragment of 245 bp and a genomic DNA fragment of 728 bp. For AtPSK4 (At3g49780), the forward primer 5′-CAC CAT GGC TCT TCT TTG CTC T-3′ and the reverse primer 5′-TTA GGG CTT GTG ATT CTG AGT-3′ amplified a cDNA fragment of 238 bp and a genomic DNA fragment of 502 bp. For ATPSK5 (At5g65870), the forward primer 5′-ATG GTT AAG TTC ACA ACT TTC CTC-3′ and the reverse primer 5′-TTA GGG ATT GTG GTT TTG AGT GTA-3′ amplified a cDNA fragment of 231 bp and a genomic DNA fragment of 462 bp. For the intronless AtPSKR1 (At2g02220), the forward primer 5′-GAG CGT TGC AAT ACA ATC AG-3′ and the reverse primer 5′-CAG TAC TTA CAT GCG TCT CGT-3′ amplified a fragment of 1142 bp. For AtActinII, the forward primer 5′-CAA AGA CCA GCT CTT CCA TCG-3′ and the reverse primer 5′-CTG TGA ACG ATT CCT GGA CCT-3′ amplified a cDNA fragment of 427 bp and a genomic DNA fragment of 513 bp. PCR was performed for 30 cycles. PCR fragments were separated on a 1% (w/v) agarose gel and stained with ethidium bromide.


PSK gene expression in roots

In order to identify putative sites of PSK activity and to link PSK-α synthesis to its growth-promoting effect on roots, we performed expression studies of the five PSK genes that are expressed in Arabidopsis. All PSK precursor genes encode the same conserved pentapeptide backbone YIYTQ characteristic of PSK-α (Lorbiecke & Sauter, 2002). Semiquantitative RT-PCR analysis showed expression of all five genes in roots and in shoots (Fig. 1). However, AtPSK1 and, to a minor degree, AtPSK5 were expressed at higher levels in roots as compared with shoots. Transcript abundance of AtPSK3 increased with age in shoots. To obtain a more detailed picture of PSK gene expression, promoter:GUS studies were performed (Fig. 2).

Figure 1.

Phytosulfokine (PSK) and PSK receptor genes are expressed in shoots and roots of Arabidopsis. Semiquantitative RT-PCR analyses were performed to study expression of AtPSK1, AtPSK2, AtPSK3, AtPSK4, AtPSK5 and of AtPSKR1 in shoots and roots of 8- and 14-d-old Arabidopsis plants. RT-PCR analysis of AtActinII was included as a control for RNA input, and PCR amplification of genomic DNA was performed as a control for genomic contamination.

Figure 2.

Promoter:GUS analysis of phytosulfokine (PSK) genes and of the PSK receptor gene AtPSKR1 in Arabidopsis thaliana. β-Glucuronidase activity in promoter:Gus lines of AtPSK1, AtPSK2, AtPSK3, AtPSK4, AtPSK5 and AtPSKR1 was analyzed in roots at the shoot-to-root junction (A, F, K, P, U, a), in lateral root primordia (B, G, L, Q, V, b), lateral roots (C, H, M, R, W, c), root tips (D, I, N, S, X, d), and in root cross-sections (E, J, O, T, Y, e) of 14-d-old seedlings.

Analysis of PAtPSK1:GUS plants indicated expression in lateral root primordia (Fig. 2B) and in growing lateral roots, including the growing region at the apex where Gus staining appeared in a somewhat patchy pattern (Fig. 2A,D). Cross-sections through lateral roots indicated that PAtPSK1:Gus was expressed in all cell layers with stronger expression in the epidermis (Fig. 2E). Promoters of AtPSK2, AtPSK3, AtPSK4, and AtPSK5 were active in the central cylinder of main and lateral roots (Fig. 2). Cross-sections revealed that AtPSK2 and AtPSK3 expression was confined to the central cylinder excluding the endodermis (Fig. 2J,O). Gus activity driven by the promoters of AtPSK4 and AtPSK5 was detected in the central cylinder and in the endodermal cell layer (Fig. 2T,Y). Gus staining resulting from the activities of the AtPSK2, AtPSK3 and AtPSK4 promoters was detected at sites where lateral roots were connected to the central cylinder of the parent root, whereas lateral root primordia or young lateral roots were only weakly stained or not stained (Fig. 2G,L,R). With respect to the root tip, two types of expression patterns were obvious. AtPSK1 and AtPSK3 showed promoter activity in the root tip (Fig. 2C,D,M,N) whereas the promoters of AtPSK2, AtPSK4, and AtPSK5 were not active in the very root tips (Fig. 2H,I,R,S,W,X).

Taken together, all PSK genes are expressed in roots with differential and partly overlapping expression patterns. Promoters of AtPSK2, AtPSK4 and AtPSK5 displayed strongly overlapping activities in the central cylinder of the differentiated part of main and lateral roots and no activities in growing root tips. AtPSK1 and AtPSK3 were both expressed in root tips but displayed otherwise different cell-specific expression patterns. Promoters of AtPSK2, AtPSK3 and AtPSK4 were particularly active in the central cylinder of parent roots at sites where new lateral roots are formed.

Expression of the PSK-α receptor gene AtPSKR1 in roots

The PSK-α receptor AtPSKR1, encoded by At2g02220, was identified in the Arabidopsis genome by sequence homology to the PSK receptor from carrot (Matsubayashi et al., 2002). The function of AtPSKR1 as a PSK-α receptor was confirmed by Matsubayashi et al. (2006). In order to compare sites of PSK-α synthesis with sites of PSK-α perception, and in order to specify the cells and tissues that can perceive, and thus respond to, PSK-α, we analyzed expression of the PSK-α receptor gene AtPSKR1 in Arabidopsis seedlings. Semiquantitative RT-PCR studies indicated that AtPSKR1 was expressed in shoots and roots at seemingly low levels (Fig. 1). AtPSKR1 promoter-driven GUS activity was observed in the primary and lateral roots, including root primordia, but not in the hypocotyl (Fig. 2a,b). In the primary root and in lateral roots, activity was particularly high in the very root tip, possibly representing the root cap, but was present at lower levels throughout the growing and mature regions (Fig. 2c,d). Cross-sections revealed that the receptor gene was expressed in all cell layers of the root. Based on publicly available microarray data (Genevestigator at; Zimmermann et al., 2004), the second PSK-α receptor AtPSKR2 is expressed in the root tip. Together with the promoter:GUS data shown here, this indicates that PSK-α can potentially be perceived by all cells in the root tip.

PSK-α promotes root but not shoot growth of Arabidopsis plants

An AtPSKR1 T-DNA insertion mutant was identified in the Arabidopsis thaliana ecotype Columbia and obtained from NASC. It was named Atpskr1-T to distinguish it from the Ds insertion mutant pskr1-1 described previously in the ecotype Landsberg erecta (Matsubayashi et al., 2006). The T-DNA was inserted 2337 bp downstream of the start codon of the intronless AtPSKR1 gene (Fig. 3a) and led to loss of detectable AtPSKR1 mRNA (Fig. 3b).

Figure 3.

The phytosulfokine (PSK) receptor T-DNA insertion mutant Atpkr1-T does not express AtPSKR1. (a) The mutant Atpskr1-T of Arabidopsis thaliana ecotype Col-0 has a T-DNA insertion 2337 bp downstream of the start codon of the AtPSKR1 gene At2g02220 which disrupts the intronless open reading frame. (b) RT-PCR analysis of AtPSKR1 expression in Arabidopsis wt and Atpskr1-T mutant showed that AtPSKR1 gene expression is knocked out in Atpskr1-T. mRNA of AtActinII was reverse-transcribed and amplified as a control. wt, wild-type.

Wild-type Arabidopsis plants were grown for 14 d in the presence of PSK-α at concentrations between 0.1 nm and 10 µm, or without PSK-α. Shoot fresh weight was not affected by PSK-α treatment at any concentration in either wt or Atpskr1-T (Fig. 4a). In roots, a significant increase in fresh weight was observed at 10 nm and higher concentrations of PSK-α in the media (Fig. 4b). Atpskr1-T plants grown on MS medium for 14 d were smaller than wt, indicating that disrupted AtPSKR1 signaling was limiting plant growth (Fig. 5a). Shoot fresh weight was reduced by c. 20% in Atpskr1-T plants compared with wt (Fig. 5b), while root mass was c. 40% lower (Fig. 5c). It is conceivable that shoot growth was delayed as a result of reduced root mass. When grown on PSK-α concentration between 100 nm and 10 µm, plants displayed an approx. 20% elevated root fresh weight over control plants. Interestingly, root fresh weight was increased in the presence of PSK-α not only in wt but also in the Atpskr1-T mutant, which is compatible with the existence of a second receptor for PSK-α. In Atpskr1-T, some root growth was induced by the unsulfated PSK pentapeptide. In wt, treatment with the unsulfated control peptide also appeared to induce minor root growth, albeit the effect was not statistically significant (Fig. 5c). Treatment with a mixture of the amino acids composing PSK did not promote root growth. In summary, PSK-α induced root growth, and roots of Atpskr1-T produced c. 40% less root mass compared with wt, supporting the idea that PSK signaling influences root growth.

Figure 4.

Root and shoot fresh weights of phytosulfokine (PSK)-treated Arabidopsis plants. Arabidopsis plants were grown for 14 d on plates containing PSK-α (circles) or desulfated PSK (dsPSK; squares) at the concentrations indicated. Shoot (a) and root (b) fresh weights were determined and averages (± SE) were calculated as a percentage of control. Fresh weights were determined in three independent experiments with a total of 10–15 plants per treatment. Values that are significantly different from the control (P < 0.05) are marked with an asterisk.

Figure 5.

Root growth is reduced in Atpskr1-T. (a) Arabidopsis wild-type (wt) and Atpskr1-T plants were grown for 14 d on plates. Average shoot (b) and root (c) fresh weights per plant were determined in wt and Atpskr1-T mutant plants grown on 1 µm PSK-α, 1 µm desulfated PSK (dsPSK), 1 µm of each amino acid (aa) present in dsPSK or without effector (control), and shoot and root fresh weights were determined for each plant. Numbers are averages (± SE) from three independent experiments with a total of 12–20 plants per treatment. Different letters indicate statistically significant differences between treatments in one genotype (P < 0.05). Root fresh weights of wt and Atpskr1-T shown in (c) are statistically different with P < 0.001.

PSK-α affects root length

Increased root fresh weight in the presence of PSK-α may result from increased lateral root formation or from longer roots, or from both. In order to distinguish between these possibilities, we analyzed both the number of lateral roots and root length in 7-d-old wt and Atpskr1-T mutant seedlings.

Growth of wt seedlings for 7 d on 1 µm PSK-α resulted in an approx. 20% longer primary root (Fig. 6a,b). Primary roots of Atpskr1-T seedlings, on the other hand, were only approximately two-thirds the length of roots from wt seedlings (Fig. 6a,b). In the presence of 1 µm PSK-α, the main root of receptor mutant seedlings elongated more than in untreated seedlings but did not reach the length of the main root of wt. These observations indicated that main root length was controlled by PSK-α acting, at least in part, through AtPSKR1. In Arabidopsis, two genes encoding PSK receptors were described (Amano et al., 2007). Analysis of a T-DNA insertion mutant of the second receptor gene, AtPSKR2 (Supporting information, Fig. S1a,b), showed that knockout of this receptor also reduced main root length (Fig. S1c). However, root growth was not as impaired in Atpskr2 as it was in Atpskr1-T. As with Atpskr1-T, Atpskr2 was still capable of responding to PSK-α with increased primary root growth. Atpskr1-T/Atpskr2 plants in which both receptors were knocked out displayed the shortest roots. Furthermore, Atpskr1-T/Atpskr2 plants were no longer capable of responding to PSK-α (Fig. S1c). The results support the conclusion that PSK signaling of root growth occurs mainly through AtPSKR1 and, to a lower extent, through AtPSKR2.

Figure 6.

Phytosulfokine-α (PSK-α) promotes root growth. (a) Arabidopsis wild-type (wt) and Atpskr1-T (pskr1-T) seedlings were grown for 7 d on plates without effector. (b) Arabidopsis wt and Atpskr1-T (pskr1-T) seedlings were grown for 7 d on plates containing 1 µm PSK-α, 1 µm desulfated PSK (dsPSK), 1 µm of each amino acid (aa) present in dsPSK or without addition of an effector (control), and the length of the primary root and the total length of lateral roots (c) were determined for each seedling. Average numbers (± SE) per plant were obtained from three independent experiments with a total of 12–20 seedlings analyzed per treatment. Different letters within one genotype indicate statistically significant differences between treatments. Lengths of main roots in wt and Atpskr1-T, as shown in (b), are significantly (P < 0.001) different, while the lengths of lateral roots between genotypes, as shown in (c), do not differ.

We further analyzed regulation of lateral root length by PSK (Fig. 6c). In wt seedlings, 1 µm PSK-α caused an approx. 45% increase in lateral root length. In Atpskr1-T, unlike main root length, lateral root length was not reduced compared with wt, indicating that AtPSKR1 plays different roles in the regulation of main and lateral roots. Treatment of Atpskr1-T seedlings with 1 µm PSK-α appeared to increase average lateral root length, but this difference was not statistically significant (Fig. 6c). Overall, the number of lateral roots appeared to be slightly lower in Atpskr1-T receptor knockout mutant seedlings than in wt (Fig. 7a). When seedlings were treated with 1 µm PSK-α for 7 d, the number of lateral roots per seedling increased by c. 27% in both wt and Atpskr1-T mutant as compared with controls, indicating once again that the Atpskr1-T mutant was capable of responding to PSK-α (Fig. 7a). Lateral root numbers were not significantly different when wt or Atpskr1-T plants were treated with 1 µm desulfated PSK or 1 µm amino acids as compared with either untreated plants or plants treated with PSK-α.

Figure 7.

Effect of phytosulfokine-α (PSK-α) on lateral root formation. (a) Treatment with PSK-α results in more lateral roots per plant. Arabidopsis wild-type (wt) and Atpskr1-T (pskr1-T) seedlings were grown for 7 d on plates containing 1 µm PSK-α, 1 µm desulfated PSK (dsPSK), 1 µm of each amino acid (aa) present in dsPSK or without addition of an effector (control), and the number of lateral roots was determined for each seedling. Numbers are averages (± SE) from three independent experiments with a total of 12–20 plants per treatment. PCycB1;1:Gus plants show staining of apical root meristems (b), lateral root primordia (c), and pericycle cells (d) that are primed to develop into lateral roots. (e) Treatment with PSK-α does not increase lateral root density. Average numbers of lateral roots (LR), lateral root primordia (LRP) and proliferation-primed pericycle regions (pericycle) were determined in 7-d-old wt Arabidopsis seedlings treated with 1 µm PSK-α, 1 µm desulfated PSK (dsPSK), or without effector (control). Numbers are averages (± SE) from three independent experiments with a total of 12–15 seedlings analyzed per treatment. No statistically significant differences were detected between treatments.

To test if the increase in root fresh weight in PSK-α-treated plants was a result of increased lateral root density, we used a PCycB1;1:Gus line which displays promoter activity of the mitotic CycB1;1 gene in meristematic cells such as in root primordia or in the root apical meristem. This reporter line also offered the possibility to study early events of lateral root initiation since the CycB1;1 promoter is active in pericycle cells from which lateral roots develop (Beeckman et al. 2001). We distinguished between lateral roots that had emerged from the parent root (Fig. 7b), lateral root initials present within the cortex (Fig. 7c) and cell division-primed, CycB1;1-expressing pericycle cells (Fig. 7d). PCycB1;1:Gus seedlings were grown for 7 d with 1 µm PSK-α, 1 µm desulfated PSK or without effector. The number of lateral roots, lateral root primordia or CycB1;1 expressing pericycle cells per millimeter main root were not significantly altered by any treatment (Fig. 7e). Thus, the observed increase in root fresh weight in PSK-treated plants was not a result of increased lateral root density.

We conclude from our measurements that the PSK-α-dependent increase in root mass was the result of a longer main root with, overall, more lateral roots, which also grew longer in the presence of PSK-α. We further conclude that control of main, but not lateral, root length by PSK-α is mostly exerted through AtPSKR1 and only to a minor degree through an AtPSKR1-independent signaling pathway likely involving AtPSKR2.

PSK-α signaling promotes root growth by increasing meristem and cell size

We next asked whether cell division activity in the root apical meristem or cell elongation in the elongation zone of the primary root were regulated through PSK-α. To that end, cell length profiles were obtained from the tips of primary roots and final cell lengths were measured in the cortex of wt and Atpskr1-T seedlings grown in the absence or presence of 1 µm PSK-α for 5 d. It was previously shown that the apical limit of the meristem is reached when cells grow above 40 µm (Beemster & Baskin, 1998). Based on averaged smoothed data calculated according to Beemster et al. (2002), we determined from root cell length profiles (Fig. 8) the number of cells in the apical meristem and the meristem size. By dividing root elongation rates by the average mature cell length, we also determined cell production rates (Table 1).

Figure 8.

Differences in cell length profiles associated with phytosulfokine (PSK) signaling. Cell length was measured in cortical files of primary roots of 5-d-old Arabidopsis seedlings up to a distance of 700 µm from the quiescent center (QC) in wild-type (wt, open circles), wt + PSK-α (closed circles), Atpskr1-T (open triangles) and Atpskr1-T + PSK-α (closed triangles). Results are averages (± SE) of interpolated values calculated from smoothed profiles of at least five roots, with two cortical cell files measured per root. The basal border of the meristem as shown on the x-axis was defined as the distance from the QC at which cells reach an average length of 40 µm.

Table 1.  Root elongation rate, mature cell size, cell production rate, meristem size, and number of cells per meristem were determined in primary roots of Arabidopsis thaliana wild-type (wt) and Atpskr1-T grown for 5 d in the presence or absence of 1 µm phytosulfokine-α (PSK)
Genotype + treatmentRoot elongation rate (µm h−1± SE)Mature cell length (µm ± SE)Cell production rate (cells h−1)Meristem size (µm ± SE)Cells per meristem (± SE)
  • Root elongation rates were calculated by dividing final root lengths by growth duration. The mature cortical cell length was measured at 5 mm from the root tip.

  • a–d

    Averages significantly different from each other (with P = 0.0016 for the difference between a and d, and P < 0.0001 for all other pairs). Cell production rates were calculated by dividing the root elongation rate by the average cell size. Meristem size and number of cells per meristem were calculated from cell length profiles shown in Fig. 9, taking 40 µm as a threshold size for meristematic cells (Beemster et al., 2002).

wt115 ± 3175 ± 31a0.6543489 ± 1749 ± 2
wt + PSK120 ± 7204 ± 43b0.5894521 ± 1246 ± 2
Atpskr1-T78 ± 3133 ± 24c0.5893385 ± 639 ± 1
Atpskr1-T+ PSK87 ± 5157 ± 34d0.5506460 ± 1945 ± 1

In wt, treatment with PSK-α induced cells to elongate more (Fig. 8). Final cell length was on average 175 µm in wt, and increased significantly, by c. 15%, to 204 µm in the presence of 1 µm PSK-α (Table 1). The estimated size of the meristem also increased when wt seedlings were treated with PSK-α. This increase most likely resulted from increased cell size, because the number of cells in the meristem remained constant and the cell production rate decreased only slightly (Table 1). An increase in the size of the apical meristem in response to PSK-α was confirmed using an independent approach. The PCycB1;1:GUS reporter line was used to determine the region in the root tip that showed meristematic activity as indicated by the G2/M phase-specific CycB1;1 promoter (Fig. 9a). For comparison, the region of the root tip that expressed AtPSKR1 is also shown. It clearly overlapped with, but was not confined to, the meristem. PCycB1;1:GUS activity was observed over an approx. 10% longer distance in the root tip when seedlings were grown with PSK-α, a difference that was statistically significant (Fig. 9b).

Figure 9.

Phytosulfokine-α (PSK-α) increases root meristem size. (a) PAtPSKR1:GUS expression (left) and PCycB1;1:GUS expression (right) in primary roots of 5-d-old Arabidopsis seedlings. PCycB1;1:Gus seedlings were grown without (control) or with 1 µm PSK-α (PSK-α). (b) The length of the primary root meristems was estimated in plants of the PCycB1;1:GUS line treated without (control) or with 1 µm PSK-α (PSK-α). Averages (± SE) were calculated from 46 to 49 roots and were significantly different based on a two-way Student's t-test (P < 0.05).

In the Atpskr1-T mutant, the elongation rate was significantly lower than in wt (Student's t-test, P < 0.0003) as was the final average cell length (Table 1). It was 133 µm in Atpskr1-T as compared with 175 µm in wt, which corresponds to c. 24% shorter cells in the receptor mutant (Table 1; P < 0.0001). The average final cell length increased significantly in Atpskr1-T with PSK-α treatment, to 157 µm, but overall remained significantly smaller than in wt (P = 0.0016), indicating that redundant PSK-α signaling pathways did not fully compensate for the lack of AtPSKR1 activity. The root apical meristem was on average smaller in Atpskr1-T than in wt (P = 0.0006), essentially because it contained fewer cells of similar size than in the wt (Table 1). In the presence of PSK-α, the meristem in Atpskr1-T became larger, as it did in wt; however, unlike in wt, the increase in meristem size was essentially the result of an increased number of cells (P = 0.0039).

Taken together, the results indicate that the promoting effect of PSK-α on root elongation in wt is essentially the consequence of a positive effect on cell elongation. In agreement with this conclusion, the Atpskr1-T receptor mutant has a reduced final cell length. Nonetheless, a redundant receptor, likely AtPSKR2, provides responsiveness to PSK-α in Atpskr1-T such that final cell lengths are partially restored when seedlings are treated with PSK-α. PSK-α signaling in Atpskr1-T also promotes meristem size such that it is partially restored to wt size (Table 1). In summary, the results indicate that PSK-α signaling through AtPSKR1 regulates cell elongation.


PSK-α can be synthesized and perceived in the root apex

Phytosulfokine-α is considered an autocrine factor (Yang et al., 2000). Yet it is conceivable that this small, water-soluble molecule moves or is transported over longer distances to act distal from its site of synthesis. Therefore, it was of interest to see to what extent gene expression of ligand and receptor overlap. Promoter:Gus analyses of AtPSK2, AtPSK3, AtPSK4, AtPSK5 and of the PSK receptor gene AtPSKR1 were shown previously for whole seedlings but cell- or tissue-specific expression patterns were not given (Matsubayashi et al., 2006). Our detailed analysis indicated that each PSK gene had a specific expression pattern. Yet taking all patterns into account, it can be concluded that each root cell produces a PSK precursor mRNA.

In order to promote root growth rate, PSK-α must be active in the root apex. Activity in the meristem is required to accelerate the rate of cell division for a continuous supply of new cells. Acceleration of cell elongation contributes to enhanced growth but is not sufficient to promote long-term growth without production of new cells from the meristem. On the other hand, the final length that cells reach can determine root length. A change in final cell length is sufficient to alter root size. AtPSK1 and AtPSK3 were expressed throughout the root, including the apex of primary and lateral roots, and could thus directly contribute to synthesis of PSK-α not only in the cell division zone but also in the elongation zone of roots. Importantly, AtPSK1 was expressed not only in the central cylinder but also in cortex and epidermal cells as revealed in cross-sections. It should also be noted that, at the base of the meristem, cells move rather quickly. The time it takes to cross the entire elongation zone is only c. 8 h (Beemster & Baskin, 1998). Hence, it is conceivable that the expression of both PSK precursors and receptors in the meristem will also result in active signaling in the elongation zone.

Thus, AtPSK1 is expressed in all dividing and elongating cells and could provide active PSK-α in an autocrine fashion. The promoter of the PSK receptor AtPSKR1 was also active throughout the root, including the root tip, supporting the idea that perception of locally synthesized PSK-α can occur in the growth zone of roots.

AtPSK2, AtPSK4 and AtPSK5 were specifically expressed in the central cylinder of roots where PSK-signaling may contribute to differentiation of vascular cells. Such a role was shown for in vitro cultured mesophyll cells from Zinnia. These were induced to differentiate into tracheal elements upon treatment with PSK-α in a dose-dependent manner (Matsubayashi et al., 1999). The observed elevated expression of AtPSK2, AtPSK3, and AtPSK4 at sites where a lateral root emerged from the main root may elevate PSK-α activity to participate in regulation of vascular development or proper connection of vasculatures between parent and lateral roots.

PSK-α promotes root growth in Arabidopsis

Treatment of Arabidopsis plants with PSK-α increased root biomass in a dose-dependent manner. The dose–response curve indicated a half-maximal root growth response at c. 10 nm PSK-α. This value agreed well with the affinity of the Arabidopsis PSK receptor AtPSKR1 to PSK-α determined in in vitro binding assays using microsomal membrane fractions in which the dissociation constant of PSK-α was estimated to be 7.7 nm (Matsubayashi et al., 2006). In Atpskr1-T plants, root fresh weight was increased significantly not only when treated with PSK-α but also when treated with desulfated PSK (dsPSK) pentapeptide. On the other hand, treatment with equimolar amounts of the amino acids present in dsPSK did not show any growth-promoting effect. dsPSK was shown to weakly compete for PSK-α-binding to rice cells and to plasma membrane fractions from rice cells and could thus have weak PSK activity (Matsubayashi et al., 1997; Matsubayashi & Sakagami, 1999). It is conceivable that application of a relatively high concentration of 1 µm dsPSK was sufficient to trigger a weak PSK response.

Shoot growth was not stimulated in plants supplied with PSK-α. Leaf expansion proceeds two-dimensionally. It is controlled by mechanisms which balance cell size and cell number to maintain a given leaf size. In case of reduced cell production from the meristem, cells become larger to compensate for reduced cell number. If cells become larger, for example as a consequence of altered ploidy, the cell number is adapted. Unlike leaves, roots do not reciprocally compensate cell size through cell number (Ferjani et al., 2007). Rather it seems that regulation of the cell production rate in the root apical meristem is independent of cell size control and that root growth rate is directed by the cell production rate in the root apical meristem (Beemster & Baskin, 1998; Chen et al., 2003). It is possibly because of a lack of compensation in roots that a PSK-α effect on growth is only seen in roots. Longer roots were observed in Arabidopsis plants constitutively overexpressing the PSK precursor gene AtPSK4 (Matsubayashi et al., 2006), supporting the finding that availability of PSK-α was limiting root growth in wt plants. Unlike supply of PSK-α to growing roots, overexpression of the PSK receptor AtPSKR1 did not promote root growth (Matsubayashi et al., 2006), indicating that root growth was limited by the abundance of the ligand and not by the abundance of receptor protein. It should be interesting to learn under which circumstances plants increase or reduce synthesis of PSK-α to boost or limit root elongation. Auxin and ethylene also interact to alter root growth through regulation of cell expansion (Swarup et al., 2007). It is conceivable that either hormone interacts with PSK-α production or PSK signaling.

PSK-α signaling through AtPSKR1 promotes cell elongation

Pharmacological studies and analysis of the PSK-α receptor mutant Atpskr1-T revealed a role for PSK signaling in root growth regulation. The cellular basis of PSK-α action was studied by measuring root cell length profiles. Mature cortex cells were shorter in Atpskr1-T than in wt, pointing to cell expansion as a major target of PSK-α signaling through AtPSKR1. Furthermore, meristematic cells became larger in wt plants treated with PSK-α, resulting in an overall larger meristem. The size of the meristem in wt seedlings was estimated to be c. 490 µm based on the previously published assumption that cells that leave the meristem have a size of 40 µm (Beemster et al., 2002). Based on the activity of the mitotic CycB1;1 promoter, the length of the meristem was estimated at c. 200 µm. CycB1;1 expression is a marker for entry into mitosis, whereas kinematic data depend on new cell wall being visible microscopically (i.e. it is a much later event). Also, CycB1;1 expression is a stochastic event which always underestimates the meristem. Despite the differing values of estimated total meristem size, the relative changes in meristem size that were measured in response to treatment with PSK-α were highly comparable when determined by either of the two methods. PSK-α increased meristem size. Based on cell size measurements, this increase was the result of increased average cell lengths. Thus it is concluded that PSK-α is a positive regulator of cell elongation growth that acts on meristematic as well as elongating cells, ultimately resulting in longer mature cells. The increase in main root length in the presence of PSK-α was c. 24%. Mature cell length was increased by c. 17% in the presence of PSK-α, indicating that cell elongation can account for the largest part of the growth-promoting effect of PSK-α observed in wt roots.

Knockout of AtPSKR2 slightly reduced root lengths, and knockout of both PSK receptor genes in the Atpskr1-T/Atpskr2 double mutant resulted in somewhat shorter roots than in the Atpskr1 single mutant, indicating that AtPSKR2 participates in root growth regulation. This conclusion is supported by the observation that Atpskr1-T knockout mutants were capable of responding to PSK-α with accelerated root growth even though wt root length was not completely restored. When Atpskr1-T plants were treated with PSK-α, the number of cells in the root apical meristem increased significantly, by c. 15%, resulting in an enlarged meristem. These results can be interpreted to mean that PSK signaling through AtPSKR2 affects cell proliferation rather than cell elongation. PSK-α signaling through AtPSKR1 and AtPSKR2 might thus initiate divergent signaling pathways, leading to the activation of different cellular processes, both of which contribute to enhanced root growth.


Support of this work by the Deutsche Forschungsgemeinschaft is gratefully acknowledged.