Divalent metal-dependent catalysis and cleavage specificity of CSP41, a chloroplast endoribonuclease belonging to the short chain dehydrogenase/reductase superfamily

Abstract

CSP41 is a ubiquitous chloroplast endoribonuclease belonging to the short chain dehydrogenase/reductase (SDR) superfamily. To help elucidate the role of CSP41 in chloroplast gene regulation, the mechanisms that determine its substrate recognition and catalytic activity were investigated. A divalent metal is required for catalysis, most probably to provide a nucleophile for cleavage 5' to the phosphodiester bond, and may also participate in cleavage site selection. This requirement distinguishes CSP41 from other Rossman fold-containing proteins from the SDR superfamily, including several RNA-binding proteins and endonucleases. CSP41 is active only in the presence of MgCl2 and CaCl2. Although Mg2+- and Ca2+-activated CSP41 cleave at identical sites in the single-stranded regions of a stem-loop-containing substrate, Mg2+-activated CSP41 was also able to cleave within the double-stranded region of the stem-loop. Mixed metal experiments with Mg2+ and Ca2+ suggest that CSP41 contains a single divalent metal-binding site which is non-selective, since Mn2+, Co2+ and Zn2+ compete with Mg2+ for binding, although there is no activity in their presence. Using site-directed mutagenesis, we identified three residues, Asn71, Asp89 and Asp103, which may form the divalent metal-binding pocket. The activation constant for Mg2+ (K(A,Mg) = 2.1 +/- 0.4 mM) is of the same order of magnitude as the stromal Mg2+ concentrations, which fluctuate between 0.5 and 10 mM as a function of light and of leaf development. These changes in stromal Mg2+ concentration may regulate CSP41 activity, and thus cpRNA stability, during plant development.

Figure 1

(A and B) Cleavage of petDΔ18 (A) or petDΔ3 RNA (B) by CSP41. A 0.5 µg aliquot of recombinant CSP41 was incubated with 20 fmol petD RNA in buffer containing 20 mM MgCl₂. Samples were removed at the indicated times and analyzed in a 5% denaturing polyacrylamide gel. (C) The spinach petD 3′ UTR substrates used in this study. Numbering begins at the 5′ end of the in vitro transcript. The petD stop codon, major cleavage sites of CSP41 and the mature petD 3′ end are indicated. Secondary structures are based on folding predictions (RNA Structure 3.7) and structural mapping experiments.

Figure 2

CSP41 binds to petD RNA in the absence of divalent metals. Inset: gel mobility shift assay of petDΔ3 binding by CSP41. Uniformly labeled petDΔ3 RNA at the indicated concentrations was incubated with 5 µg of CSP41. The samples were then treated with 40 U of RNase T1 and analyzed in a 5% native polyacrylamide gel. Reactions were performed in duplicate. The intensity of the shifted band from the gel mobility shift assay was plotted versus the concentration of petDΔ3 RNA. The error in intensity at each data point was typically <5%. The curve represents the best fit to equation 2 (see Materials and Methods).

Figure 3

CSP41 cleaves petDΔ3 RNA 5′ to the phosphodiester bond. petDΔ3 RNA, either uniformly labeled or unlabeled, was incubated in the presence of 0.5 µg of CSP41 and buffer containing 20 mM MgCl₂ for 20 min. The unlabeled reactions were then treated with or without 10 U alkaline phosphatase as indicated. All reactions were subsequently incubated with 10 U of polynucleotide kinase and [γ-³²P]ATP. The reactions were then precipitated and analyzed in a 5% denaturing polyacrylamide gel.

Figure 4

The divalent metal specificity of CSP41. (A) A 0.5 µg aliquot of recombinant CSP41 was incubated with 20 fmol petDΔ18 RNA in buffer containing 20 mM concentrations of the indicated metals. Samples were removed after 10 min and analyzed in a 5% denaturing polyacrylamide gel. Sizes of CSP41 cleavage products are indicated on the left. (B) A 0.5 µg aliquot of recombinant CSP41 was incubated with 20 fmol petDΔ3 RNA in buffers containing varying concentrations of MgCl₂ as indicated. Each data point represents the average of duplicate measurements. The variation in the rate at each of the indicated Mg²⁺ concentrations was typically <10%. The curve represents the best fit to equation 1 (see Materials and Methods). (C) A 0.5 µg aliquot of recombinant CSP41 was incubated with 20 fmol petDΔ18 RNA in buffer containing 2 mM MgCl₂ and 20 mM of the indicated divalent metals. The control reaction (C) contained no CSP41. Reactions were stopped after 10 min and analyzed in a 5% denaturing polyacrylamide gel. Sites of CSP41 cleavage are indicated on the left.

Figure 5

Mg²⁺ and Ca²⁺ bind to the same site on CSP41. (A) CSP41 was incubated for 10 min in the presence of 20 fmol petDΔ3 RNA and divalent metals as indicated. The reaction products were analyzed in a 5% denaturing polyacrylamide gel. Sizes of CSP41 cleavage products are indicated on the left. (B) Relative cleavage rates at position 136 under different mixed metal conditions. Reactions were run in triplicate, the 136 nt band was quantified on a Storm Scanner, and the values were averaged. Cleavage in the presence of 10 mM MgCl₂ was adjusted to 100% as a reference.

Figure 6

Site-directed mutagenesis of conserved acidic residues of CSP41. (A) Sequence alignment of spinach (AAC49424), Arabidopsis CSP41a (NM_116179), tomato (T52071), tobacco (AY324804), Arabidopsis CSP41b (NM_100804), Synechocystis (NP_440784) and Nostoc (NP_488871) CSP41. The sequence is numbered according to the spinach protein, with arrowheads indicating the positions of residues mutated in this study. N indicates the N-terminus of the mature spinach protein, and ΔP and ΔS indicate the C-termini of the CSP41ΔP and CSP41ΔS deletion proteins. (B) Relative rates of degradation of petDΔ18 RNA by CSP41ΔS and site-directed mutants as indicated under each bar. Samples were assayed in duplicate, degradation of the 179 nt fragment was quantified on a Storm Scanner, and the values were averaged. Cleavage by CSP41ΔS was adjusted to 100% as a reference.

Figure 7

Cleavage of petD RNA by CSP41, CSP41ΔS and mutants. The top row represents cleavage of petDΔ18 (left) and petDΔ3 (right) in the presence of Mg²⁺, and the bottom row in the presence of Ca²⁺. CSP41 was incubated with either 20 fmol petDΔ18 or 20 fmol petDΔ3 and either 20 mM MgCl₂ or 20 mM CaCl₂ as indicated. The enzyme used is indicated at the top of each lane. The sizes of substrates and cleavage products are indicated to the left of each gel. (–) indicates the non-enzymatic control. The band marked with an asterisk is discussed in the text.