Apple Pollen Soluble Inorganic Pyrophosphatase MdPPa Involves In Self-incompatibility

Self-incompatibility (SI) provides an intraspecific reproductive barrier that is widely adopted by many flowering plants to prevent inbreeding. Apple (Malus×domestica Borkh.), the main fruit crop of temperate regions of the world, shares typical S-RNase-mediated SI with many others plants in the Rosaceae, Solanaceae and Plantaginaceae families. A recent model proposed that when taken up non-specifically through the action of an ABCF protein, S-RNase is recognized by collaborative non-self SLFs/SFBBs proteins and degraded via the ubiquitin-26S-proteasome pathway in a non-self-pollen tube. In contrast, when recognition does not occur the S-RNase has a cytotoxic effect inside the self-pollen tube, resulting in an inhibition of pollen tube growth. The cytotoxicity of S-RNase has generally been attributed to its degradation of pollen rRNA, resulting in the arrest of protein synthesis. However, to date there has been no reported evidence that such ribonuclease activity is required for the apple SI response. Here we used the mature peptides of S2-RNase (S2Mat) as a bait to screen a yeast cDNA library prepared from M. domestica cv.’Ralls Janet’ pollen and obtained a sPPase. We report that apple S-RNase causes a decrease in sPPase activity in self-pollen tubes, resulting in continuous PPi accumulation. We also provide evidence that this PPi accumulation inhibits tRNA aminoacylation, leading to the reduction of pollen growth. The detailed results are showed as follows:1. To identify candidate proteins potentially involved in SI through an interaction with apple S-RNase, we used the mature peptides of S2-RNase (S2Mat) as a bait to screen a yeast cDNA library prepared from M. domestica cv.’Ralls Janet’ pollen. Two identical 484 bp fragments, A14/24, were identified in the screen and the sequence was located on chromosome 1 of the apple genome. We cloned the corresponding cDNA sequences from pollen, and obtained 654 bp full-length sequence, which we temporarily named A14. Y2H, pull-down and BiFC assay showed that A14 interacted with S1, S2, S3, S9-RNase, suggesting that A14 interacts with different S-haplotype S-RNases.2. A14 is predicted to encode a 24.4-kDa protein, comprising 217 amino acids. A search of NCBI, database revealed that A14 is homologous to Family I sPPase proteins. A14 contains the canonical sPPase signature sequence that is essential for catalytic activity, Recombinant A14 was observed to have Mg2+- dependent sPPase specific activity was inhibited by Ca2+, and so we renamed the corresponding gene, MdPPa. Quantitative reverse transcription PCR (qRT-PCR) analysis showed that MdPPa had significantly higher expression in pollen grains and pollen tubes than in other tissues. We fractionated pollen protein extracts to determine MdPPa protein localization, and it was detected only in the soluble fraction.To further confirm its subcellular localization, the MdPPa protein was transiently expressed in growing ’Ralls Janet’ pollen tubes as a fusion with GFP. Fluorescence was observed in the cytosol, which was also consistent with the localization of MdPPa when expressed in maize protoplasts.3. Immunofluorescence assays showed that S-RNase and MdPPa co-localized in the cytosol of in vitro growing apple pollen tubes. In vitro immunoprecipitation assays demonstrated that S-RNase combines with MdPPa in self-pollen tubes. We found that SPI treated pollen tube growth was strongly inhibited and showed the greatest difference from the CPI treated pollen tubes at 120 min. The sPPase specific activities in pollen extracts were notably lower after the SPI treatment at 120 min than those associated with the CPI treatment. As expected, the concentration of PPi in ’Ralls Janet’ SPI treated pollen was greater than those in CPI treated and normally growing pollen tubes. We compared the sPPase activity of recombinant His-tagged MdPPa in the presence or absence of various concentrations of His-tagged S-RNases. We found that the specific sPPase activity of MdPPa showed an obvious reduction after being incubated with equal amounts of recombinant S1-, S2-, S3-, and S9-RNases, respectively. To further elucidate the mechanism of this inhibition, we initially divided the protein into an N-variable region (N-Vb,1-58 aa), a conserved function domain (CFD,59-181 aa) and a C-variable region (C-Vb,182-217 aa), and assessed their interaction with S2Mat/S2-RNase using a Y2H and BiFC assay, respectively. The result showed that both the N-Vb and C-Vb region of MdPPa exhibited obviously interactions with S2-Mat, whereas the CFD showed a no interaction with S2Mat. To determine the mechanism that S-RNase decreases the MdPPa sPPase activity, further analysis of the enzyme kinetics of the reaction was performed. The result showed that the S-RNase inhibition of sPPase activity of MdPPa exhibited a pure noncompetitive inhibition mechanism when measured against the variable substrate, PPi.4. To determine whether MdPPa is similarly required in apple, we performed an antisense oligonucleotide assay to down-regulate its expression in pollen tubes. We observed a significant reduction in pollen tube length and sPPase activity in the presence of antisense oligonucleotides (as-ODN) targeted to an MdPPa specific sequence, and an increased concentration of PPi was also observed in as-ODN treated pollen tubes, suggesting that MdPPa plays an important role in pollen tube growth. To further elucidate the link between PPi accumulation and the inhibition of pollen tube growth, we used an RNA gel blot assay to evaluate the levels of charged and uncharged tRNA, with a 5’-digoxigenin-labeled probe complementary to nucleotides 20 to 47 of two major glycine and alanine isoacceptor tRNAs (tRNAGGG and tRNAGCU). In all the control and CPI treated pollen tubes, the detected tRNAGGGs and tRNAGCUs were mostly present in the charged form. However, in SPI treated pollen tubes, the detected tRNAGGGs and tRNAGCUs in ’Rolls Janet’,’Fuji’ and ’Golden Delicious’ were mostly uncharged. Additionally, in MdPPa-silenced (as-ODN) pollen tubes, the detected tRNAGGGs and tRNAGCUs were also mostly present in an uncharged form, suggesting that tRNA aminoacylation was inhibited in SPI treated and as-ODN pollen tubes. This inhibitory effect may trigger suppression of protein synthesis, which is required to maintain pollen tube elongation, ultimately leading to the inhibition of pollen tube growth. In addition, we characterized a naturally occurring self-compatible apple cultivar,’CAU-1’(S1S9), and found the self-compatibility of’CAU-1’was due to an S-locus unlinked pollen-part mutation (PPM). qRT-PCR revealed that there was no significiant difference in MdPPa expression between ’CAU-1’ and ’Fuji’, suggesting that there might be other factors causing the self-compatibility of ’CAU-1’.