Report

A Plant Peptide Encoded by CLV3 Identified by in Situ MALDI-TOF MS Analysis

See allHide authors and affiliations

Science  11 Aug 2006:
Vol. 313, Issue 5788, pp. 845-848
DOI: 10.1126/science.1128439

Abstract

The Arabidopsis CLAVATA3 (CLV3) gene encodes a stem cell–specific protein presumed to be a precursor of a secreted peptide hormone. Matrix-assisted laser desorption/ionization–time-of-flight mass spectrometry (MALDI-TOF MS) applied to in situ Arabidopsis tissues determined the structure of a modified 12–amino acid peptide (MCLV3), which was derived from a conserved motif in the CLV3 sequence. Synthetic MCLV3 induced shoot and root meristem consumption as cells differentiated into other organs, displaying the typical phenotype of transgenic plants overexpressing CLV3. These results suggest that the functional peptide of CLV3 is MCLV3.

Several lines of evidence indicate that peptides are important for plant growth and development. Three plant peptide hormones, systemins, phytosulfokine, and SCR/SP11, have been biochemically identified. These bioactive peptides are originated and posttranslationally modified from corresponding precursor proteins that possess signal sequences in each N-terminal region (1). More than 200 leucine-rich repeat (LRR) type receptor-like kinases (RLKs) have been identified in the Arabidopsis genome, and these are presumed to act as receptors for peptides or small molecules. The mutants of these receptors show interesting phenotypes in plant growth and development (2). From the presence of these putative receptors, we hypothesize that putative peptides exist that interact with these receptors, indicating that many unidentified hormones may play important roles in plant growth and development. However, the mature forms of predicted peptide hormones are unknown, and we cannot identify the mature forms via “in silico” (computer analysis) approaches.

Stem cells in higher plants proliferate and supply the new cells destined to become various organs while also maintaining undifferentiated cells in the apical meristem. The CLAVATA genes (CLV1, 2, and 3) control the size of the shoot apical meristem (SAM) in Arabidopsis (3, 4). Loss-of-function mutants of CLAVATA genes accumulate stem cells in the SAM, resulting in an enlarged, dome-shaped meristem. The CLV1 gene encodes an LRR-RLK, whereas CLV2 encodes a similar protein lacking the kinase domain (5, 6). CLV1 is expressed only in the central L3 layer of the SAM, and the resulting protein is hypothesized to form a heteromeric receptor complex with CLV2. CLV3 is expressed at the outer L1 and middle L2 layer, adjacent to the zone expressing CLV1, and is hypothesized to encode a ligand of the CLV1/CLV2 receptor complex. The CLV3 gene encodes a 96–amino acid protein with an 18–amino acid N-terminal signal peptide (7). CLV3 belongs to the CLE gene family (CLAVATA3/ESR-related), which shares a conserved, 14–amino acid CLE motif at the C-terminal region (8).

We generated transgenic plants constitutively overexpressing CLV3 under the control of the cauliflower mosaic virus 35S promoter (CaMV35S). CaMV35S::CLV3 transgenic plants (CLV3OX) ceased initiating organs from the SAM after the emergence of the first leaves, which were occasionally misshapen (9). Four lines of plants exhibiting premature cessation of leaf development were selected, and dissected leaf tissues were used to induce calli. During dedifferentiation and callus formation, we observed no differences between wild-type and CLV3OX plants (fig. S1A). We maintained dedifferentiated calli, and CLV3 mRNA overexpression in the CLV3OX calli was confirmed by Northern blotting (fig. S1B).

Matrix-assisted laser desorption/ionization–time-of-flight mass spectrometry (MALDI-TOF MS) was used to identify the neuropeptides involving neural signal transduction in crustaceans (10). We characterized the structure of the mature peptide encoded by the CLV3 gene (MCLV3) in CLV3OX calli using MALDI-TOF MS. We performed in situ MALDI-TOF MS experiments with CLV3OX calli slices to detect typical ions (fig. S2). We observed a CLV3OX-specific ion at a mass-to-charge ratio (m/z) of 1345.6 (Fig. 1A) when calli were cultured on hormone-free agar medium for 6 days. The observed ion was due to a peptide containing 12 amino acid residues from Arg70 to His81 in CLV3, in which two of three proline residues were modified to hydroxyproline (Ph or Hyp). This was the only peptide specifically detected in CLV3OX calli and, therefore, we hypothesize that this peptide is the only CLV3-derived peptide present in the tissue. Thus, MCLV3 is the same peptide as the identified peptide.

Fig. 1.

In situ MALDI-TOF MS analysis of CLV3OX callus. (A) A typical ion (m/z = 1345.6) observed from a CLV3OX callus. (B) MS/MS spectrum of fragment ions from MCLV3. One major (m/z = 842.4) and four minor (m/z = 503.3, 825.4, 1189.6, 1207.6) ions were detected. (C) The predicted structure of MCLV3; “b-series” or “y-series” fragment ions originating from N- and C-terminal fragments from peptide bond cleavage between -CO- and -NH- are identified. Subscript numbers indicate the length of each fragment. (D) MS/MS spectrum of synthetic MCLV3. Five signals assigned to b8, b8-NH3, b11, b11 + H2O, and y4 fragments correspond with those in spectrum B. Several ions are assigned to fragment ions (b5, b6, y2, y9, y10) originating from synthetic MCLV3.

We subjected the pseudomolecular ion (m/z = 1345.6) to MS/MS analysis to identify which of the three prolines were hydroxylated. We identified one strong (at m/z = 842.4) and four weak ion peaks (at m/z = 503.3, 825.4, 1189.6, and 1207.6) (Fig. 1B). We determined that the m/z = 842.4 ion was derived from the N-terminal RTVPhSGPhD fragment and that the m/z = 503.3 ion was derived from the C-terminal PLHH fragment (Fig. 1C). We hypothesize that the C-terminal end contains a His residue, because the observed fragment ion (m/z = 1189.6) originated from the loss of a C-terminal His residue accompanied by a hydrated ion (m/z = 1207.6).

To confirm the structure of MCLV3, we synthesized three dodecapeptides containing two Hyp residues covering all possible combinations. The MS and MS/MS spectra of one synthetic peptide agreed with that of the natural peptide, confirming its structure (Fig. 1D). The MS/MS analysis of the two peptides that contained Hyp residues at incorrect positions resulted in different-fragment ions from those of MCLV3 (fig. S3), leading us to conclude that RTVPhSGPhDPLHH is the structure of the MCLV3 peptide.

We examined the biological activities of MCLV3 by observing the effect of synthesized CLV3 peptides of various lengths (Fig. 2A) on roots on agar plates. Wild-type seeds were exposed to 1 μM MCLV3 and control buffer, resulting in main root lengths at 14 days after germination of 18.6 ± 2.0 mm (mean ± SEM) and 85.0 ± 6.3 mm, respectively (Fig. 2, B and C); these results show that MCLV3 limits root growth. Treatment with the 14–amino acid CLV3, CLE19, and CLE40 peptides also result in a shortened root (11). The Arg70 (R70) position in CLV3 is hypothesized to be excluded from the functional CLV3 peptide (12). After exposure to CLV3L and CLV3S (1 μM), the lengths of the main root were 18.4 ± 2.4 and 78.2 ± 10.5 mm, respectively (Fig. 2B). These results indicate that the functional CLV3 peptide contains the R70 residue at the N terminus, which is the second R of the CLE motif, consistent with the results of the in situ MALDI-TOF MS analysis.

Fig. 2.

MCLV3 and related peptides function in the root apical meristem. (A) The amino acid sequences of CLV3 (residues 69 to 96) and MCLV3-related peptides. Blue, hydroxylated proline residues; red, amino acids of CLE motif. (B) The effect of MCLV3 and related peptides on root meristem. Wild-type plants were grown for 14 days on agar plates containing buffer (Mock) or 1 μM each of MCLV3, CLV3L, or CLV3S, and primary root morphology was observed. Scale bar: 100 μm. Arrowheads indicate border between root meristematic and elongation zones. (C) Effect of various peptide concentrations on wild-type root development. Main root length was measured after 14 days of growth on media containing peptide (n = 10 for each peptide). Abbreviations for the amino acid residues are as follows: A, Ala; D, Asp; E, Glu; F, Phe; G, Gly; H, His; L, Leu; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; and V, Val.

To delimit the minimal C-terminal region required for activity, and to examine the effect of the C-terminal region outside of the CLE motif, we examined CLV3L, -A, -B, and -C (Fig. 2). The root meristem was inhibited by 1 μM applications of CLV3L, CLV3-A, and CLV3-B. In contrast, applications of the CLV3-C peptide did not affect the root meristem (Fig. 2C), indicating that the 13th His in the CLE motif is essential for activity. Therefore, the C-terminal region outside the CLE motif is superfluous for activity and MCLV3, the 12–amino acid CLV3 peptide, spans the 2nd “R” to the 13th “H” in the CLE motif. As identified by the in situ MALDI-TOF MS, this molecule was the most active and shortest functional CLV3 peptide. Hydroxylation of the two proline residues did not affect root growth inhibition when exposed to MCLV3 compared to MCLV3′, and CLV3-C compared to CLV3-C′ (Fig. 2C).

To examine the effect of synthetic CLV3 peptides on the SAM, we cultured seedlings in liquid medium and exposed them to 1 μM applications of different synthetic peptides. Applications of CLV3S and CLV3-C peptides had no effect on the SAM (Fig. 3, A to C). In contrast, CLV3L, CLV3-A, CLV3-B, and MCLV3 reduced the size of the SAM (Fig. 3, D to G). Therefore, the effects of various CLV3 peptides on the SAM resembled those on root growth.

Fig. 3.

MCLV3 and related peptide functions in SAM. Longitudinal sections (A to H) and scanning electron micrographs (I to L) of wild-type plants (Col-0) incubated with 1-μM peptides in liquid Murashige-Skoog medium for 21 days after germination. Scale bar: 100 μm.

Scanning electron microscopy revealed that MCLV3 (Fig. 3K) and CLV3L (Fig. 3J), as well as CLV3-A, CLV3-B, and MCLV3′ (fig. S4), induced random phyllotaxis and the occasional severe reduction of the SAM region compared to the control (Fig. 3I). These phenotypes are similar to those of CLV3OX or wuschel seedlings (9, 13). On the contrary, phyllotaxis was normal in seedlings exposed to CLV3S and CLV3-C peptides (fig. S4). In about 5% of plants treated with these functional CLV3 peptides, a leaflike structure was produced at the top of the SAM (Fig. 3, H and L), but these leaflike structures maintained phenotypically normal abaxial-adaxial polarity.

Thus, MCLV3 is the minimal unit that mimics CLV3 signaling at the SAM. MCLV3 was composed of 12 amino acid residues containing two Hyp residues, the size and the position of which were identical to those of the CLE peptide, TDIF (tracheary element differentiation inhibitory factor), which was isolated from Zinnia as an inhibitory factor of xylem cell differentiation (14). We hypothesize that the chemical structure of MCLV3 should be common to many CLE peptides. The fact that the activity of the longer peptide, CLV3L, was weaker than that of MCLV3 indicates that processing into 12–amino acid peptides is crucial to the construction of an active and mature CLV3 peptide.

The 4-hydroxylation of proline residues is a posttranslational modification found in animal tissues (15). In higher plants, this modification occurs in many glycoproteins, including arabinogalactan proteins and elastins (16), as well as in the bioactive peptide systemins in tobacco and tomato (17). Hydroxylation had no effect on MCLV3 activity. Therefore, the Hyp residues may not be responsible for the affinity with which MCLV3 binds to its receptor. Hydroxylation may serve other functions such as trafficking, storage, and peptide stability.

The 216 LRR-RLKs represent the largest group of RLKs in Arabidopsis (18). It has long been known that CLV3-CLV1 interactions are critical for the proper development of the SAM but, because the active form of CLV3 was unknown, the direct ligand-receptor interaction in the CLAVATA pathway had not been demonstrated. Our identification of the chemical structure of the CLV3 peptide will allow the interaction between CLV3 and its putative receptors, CLV1 and CLV2, to be clarified.

Supporting Online Material

www.sciencemag.org/cgi/content/full/313/5788/845/DC1

Materials and Methods

Figs. S1 to S4

References

References and Notes

View Abstract

Stay Connected to Science

Navigate This Article