A subtilisin-like serine protease specifically expressed ... - Springer Link

0 downloads 0 Views 254KB Size Report
rice subtilisin-like serine protease (RSP1). Sequence comparisons revealed that RSP1 has several characteris- tics in common with preproproteins. RT-PCR ...
Sex Plant Reprod (2001) 13:193–199

© Springer-Verlag 2001

O R I G I N A L A RT I C L E

Kaoru T. Yoshida · Tsutomu Kuboyama

A subtilisin-like serine protease specifically expressed in reproductive organs in rice

Received: 27 February 2000 / Revision accepted: 24 October 2000

Abstract To identify genes specifically expressed in flowering pistils and that are related to reproductive phenomena, simplified differential display was performed with cDNA obtained from pistils and ovaries at several stages. One clone preferentially expressed in pistils at flowering and 1 day before flowering was identified as a rice subtilisin-like serine protease (RSP1). Sequence comparisons revealed that RSP1 has several characteristics in common with preproproteins. RT-PCR, northern blot, and in situ hybridization analysis revealed that RSP1 mRNA accumulates in pistils and in the filaments of stamens, whereas mRNA was undetectable in tissue from leaves, roots, panicles, and embryos. The mRNA levels in pistils increased slightly at flowering and decreased afterwards. Possible roles of the subtilisin-like serine protease in plant reproduction are discussed. Keywords Filament · Pistil · Preproprotein · Rice · Subtilisin-like serine protease

Introduction Plant reproduction is a very complex and highly controlled process that is regulated by a number of distinct groups of genes. Identification of particular genes that play important roles in pollen recognition, pollen-tube guidance, and fertilization is highly desirable for understanding the molecular mechanisms associated with plant reproduction. A direct way to study gene expression during pollen-pistil interaction events is to isolate cDNA clones by differential screening of mRNA populations in pistils. To isolate rarely expressed genes, simplified differential display (Yoshida et al. 1994; Nakazono and Yoshida 1997) was applied in this study. cDNA from K. T. Yoshida (✉) · T. Kuboyama Graduate School of Agricultural Life Sciences, The University of Tokyo, Yayoi 1-1-1, Bunkyo-ku, Tokyo 113-8657, Japan e-mail: [email protected] Tel.: +81-3-5841-5064, Fax: +81-3-5841-5063

pistils at flowering and 1 day before flowering was compared with that from ovaries 1 day after flowering. Using simplified differential display, we isolated a number of pistil-abundant cDNA clones that are related to flowering events (Yoshida and Kuboyama 1997). These clones included those predicted to encode phospholipase D and methylmalonate semialdehyde dehydrogenase. In this report, we present data on a subtilisin-like serine protease that was cloned by further screening of the simplified differential display. Subtilisin-like serine proteases are a class of endoproteases secreted into the extracellular space. In many animal systems, signal molecules, including peptide hormones and neuropeptides, are synthesized as larger precursors and are activated by endoproteolytic cleavage of the precursor. Subtilisin-like serine proteases are the major enzymes responsible for producing these peptide signals (Steiner 1998). Recently, subtilisinlike serine proteases were revealed to exist widely in the plant kingdom; they have been found in melon (Yamagata et al. 1994), lily (Kobayashi et al. 1994), Alnus glutinosa (Ribeiro et al. 1995), Arabidopsis (Ribeiro et al. 1995), and tomato (Tornero et al. 1996). Although we do not know whether plant subtilisin-like serine proteases are proprotein convertases like animal subtilases, potential roles for subtilisin-like serine protease in rice reproduction are discussed.

Materials and methods Rice (Oryza sativa L. var. japonica cv. Kamenoo) plants were grown in a greenhouse under natural light conditions (for panicles, flowers, and seeds), or in a growth chamber at 28°C with 14 h of illumination per day at an intensity of approximately 200 µEm–2s–1 (to obtain roots and leaves of 7-day-old seedlings). For emasculation, rice flowers just before flowering were treated with hot water at 43°C for 7 min. After the treatment, rice flowers opened immediately and pollen grains were shed on the stigma, but the pollen sterilized with hot water did not germinate. For simplified differential display, mRNA from pistils 1 day before flowering, pistils at flowering, and ovaries 1 day after flowering was prepared using Oligotex-dT30 (Takara Shuzo, Kyoto, Japan) from total RNA isolated using an ISOGEN kit (Nippon

194 Gene, Tokyo, Japan). Conditions for simplified differential display were described previously (Nakazono and Yoshida 1997). Primer sets of 12 bases each were used (Common Primers, Bex, Tokyo, Japan). The PCR products were fractionated in 1.6% agarose gels. Differential bands identified by ethidium bromide staining were recovered from the gels and cloned into pCRII vector (Invitrogen, San Diego, Calif.). For RT-PCR analysis, mRNA was prepared from panicles, anthers, pistils, 14-day-old embryos, roots, and leaves of seedlings 7 days after germination. An actin gene was used as a control in the RT-PCR analysis. Internal primer sets of eighteen bases each were designed for amplification of RSP1 and actin gene fragments. Conditions for RT-PCR analysis were denaturation at 94°C for 1 min, annealing for 1 min, and extension at 72°C for 2 min. Annealing temperatures for RSP1 and actin were 58°C and 56°C, respectively. The number of PCR cycles for amplification of RSP1 and actin were 28 and 26, respectively. To construct a cDNA library, total RNA was isolated from pistils 0 to 4 h after flowering using an isogen system (Nippon Gene). Poly(A)-RNA was prepared using Oligotex-dT30 (Takara Shuzo). The poly(A)-RNA was used to construct a cDNA library with a ZAP-cDNA Synthesis Kit (Stratagene, La Jolla, Calif.). Both strands of the cDNA insert were sequenced using a dye-primer cycle-sequencing kit (Prism, PE Applied Biosystems) and an automatic DNA sequencer (model 377A, PE Applied Biosystems). Database searches were performed using the BLAST algorithm (Altschul et al. 1990). Genomic Southern hybridization was performed as described by Yoshida et al. (1999). Restriction enzymes that do not digest the RSP1 cDNA fragment were used to digest the genomic DNA. The blot was probed with the fragment labeled with [α32-P]dCTP using a random primer DNA labeling kit, ver. 2 (Takara Shuzo). For northern analysis, total RNA was fractionated on 1.2% agarose gels containing formaldehyde and transferred to HybondN+ membranes (Amersham, UK). The fragment of RSP1 cDNA was labeled with [α32-P]dCTP using a random primer DNA labeling kit, ver. 2 (Takara Shuzo). Hybridization was carried out at 42°C according to the manufacturer’s instructions (Amersham, UK). The final wash was in 1×SSPE/0.1% SDS at 42°C. In situ hybridization was carried out as described by Kouchi and Hata (1993) with some modifications. Stamens and pistils harvested 1 day before flowering and 2 h after flowering were fixed in FAA solution and embedded in paraplast plus (Oxford, St. Louis, Mo.), and 8-µm sections were mounted on slides coated with vectabond (Vector Laboratories, Burlingame, Calif.). The sections were hybridized with DIG-labeled RNA probes synthesized from the fragment of pRSP1 cDNA using a labeling kit (Boehringer Mannheim).

Results Isolation of a reproduction-associated cDNA from pistils In preliminary studies, the time course of pollen tube elongation in the rice pistil was examined. A rice flower is only open for about 1 h from the time it begins to open. During this time, pollen tubes elongate and reach the micropyle opening at the bottom of the ovary. Within 90 min of the start of flowering, a pollen tube enters the micropyle of the ovule. To isolate genes associated with sexual reproduction, we compared mRNA populations from pistils at flowering and 1 day before flowering with mRNA from ovaries 1 day after flowering. Simplified differential display was used to isolate the specific genes (Yoshida et al. 1994). Nineteen cDNA fragments that were candidates for genes preferentially expressed in pistils at flowering or 1 day before flowering were recovered from agarose gels

Fig. 1A, B Comparison of patterns obtained by differential display and RT-PCR analysis. A Electrophoretic pattern of a simplified differential display stained with ethidium bromide. Fragments indicated by the arrow were isolated and used for further analysis. Positions and sizes (in kb) of markers are indicated. B RT-PCR assay of a specific pattern detected by simplified differential display. cDNA templates prepared from pistils 1 day before flowering (1), pistils at flowering (2), and ovaries 1 day after flowering (3). An actin gene was used as a control

and cloned into vectors for further analysis. The nucleotide sequences at both ends of the cDNA inserts were determined. To investigate the possible identities of these clones, the sequences were used to search the DNA and protein databases using the program BLAST. One out of 19 clones was identified as a rice subtilisin-like serine protease, pRSP1. RT-PCR analysis was performed to confirm that the simplified differential display pattern was consistent with the actual expression pattern of this gene. An internal primer set was designed to amplify the RSP1 gene fragment. The expression pattern of rsp1 was consistent with the pattern observed in the simplified differential display (Fig. 1). The rsp1 gene is expressed in pistils both 1 day before flowering and during flowering, while its level of expression in ovaries 1 day after flowering is very low. Structural features of RSP1 To isolate full-length cDNA clones, we screened a cDNA library generated from pistils harvested 0 to 4 h after flowering. We cloned nine independent clones. They had multiple polyadenylation sites within the 3’-untranslated region (UTR). The longest cDNA clone for pRSP1 is shown in Fig. 2 (accession number AB037371). It contained an open reading frame for a protein of 789 amino acids with a 243-bp 5’-UTR, and a

195 Fig. 2 Nucleotide and deduced amino acid sequence of RSP1 cDNA. The putative signal peptide is double underlined and the putative pro-region is shaded. The amino acids that form the catalytic triad and the substrate-binding site are boxed. Potential consensus sequences for N-glycosylation sites are shown in bold with asterisks. Arrowheads mark the positions of polyadenylation sites. Arrows indicate internal primer sets used for RT-PCR analysis. The DNA fragment used as a probe for northern, Southern, and in situ hybridization analyses is underlined

398-bp 3’-UTR. Of the nine clones examined, we could detect seven alternative polyadenylation sites in the single 3’-UTR (Fig. 2). A preferential polyadenylation site was not observed. The deduced amino acid sequence was used for homology searches of databases, and sequence comparison showed that rsp1 encodes a polypeptide with about 40% homology to the subtilisin-like serine proteases in plants. The highest homology was found with the p69 subfamily of serine protease from tomato (Meichtry et al. 1999). The deduced RSP1 protein possesses all the characteristics of serine proteases belonging to the subtilisin-like serine protease family. The amino acids that form the catalytic triad (Ser-555, Asp-164, and His-230) and the substrate-bind-

ing site (Asn-334) in the subtilisin family were also conserved in RSP1 (Fig. 2). Homology is particularly high (about 80%) around the four residues forming the active site of the enzyme (Fig. 3A). A long insertion between Asn-334 and Ser-555 observed in the subtilisin-like serine protease from plants was also observed in RSP1. A hydropathy profile of RSP1 predicts that the N-terminal region (residues 1–28) is hydrophobic; this is thought to serve as a signal peptide for secretion (Fig. 3B). In addition, a computer analysis using PSORT, a computer program for predicting protein localization sites in cells (Nakai and Horton 1999), suggests that RSP1 has a cleavable N-terminal signal sequence and exists in the extracellular space. Subtilisin-like serine

196

Fig. 4A, B RT-PCR analysis of RSP1 and actin. A cDNA templates prepared from roots (1) and leaves (2) 7 days after germination, young panicles (3), stamens (4), and pistils (5) 1 day before flowering, and embryos 14 days after flowering (6). B cDNA templates prepared from pistils 1 day before flowering (1), pistils from emasclulated open flowers (2), pistils at flowering (3), ovaries 1 day after flowering (4), and embryos 14 days after flowering (5)

Fig. 3A–D Primary structure of RSP1. A Sequence conservation around the amino acids forming the active site of subtilisin-like serine proteases. B Hydropathy analysis of RSP1 according to Kyte and Doolittle (1982). Values above zero are hydrophobic. C Amino acid sequences of plant subtilisin-like serine proteases at the junction of the proprotein and mature protein. D Diagram of the three-domain structure of RSP1 from NH2 to the COOH terminus. A possible signal peptide (28 amino acid residues), the NH2terminal prosequence (99 residues), and the mature autolyzed serine protease (662 residues) are shown. The amino acids forming the catalytic triad in the active site (Asp, His, Ser) and the conserved Asn residue are shown

proteases are produced as preproenzymes, remaining inactive until secreted and processed (Siezen et al. 1991; Zhou et al. 1995). The N-terminus protein sequences of the mature proteins of several plant subtilisin-like proteases have been determined (Yamagata et al. 1994; Tornero et al. 1996; Taylor et al. 1997). The starting point of the mature protein of RSP1 was deduced from sequence comparisons with other subtilisin-like serine proteases from plants (Fig. 3C). A pair of threonine residues, which is highly conserved in plant subtilisin-like proteases, is also observed in RSP1. After processing 28 amino acid residues of the presequence and 99 amino acid residues of the prosequence, the predicted mature protein of RSP1 consisted of 662 amino acid residues (Fig. 3D). The deduced mature RSP1 protein included four asparagine-linked glycosylation sites (Fig. 2).

Fig. 5A, B Northern and Southern blot analyses of RSP1. A Total RNA (10 µg per lane) isolated from leaves (1) and roots (2) 7 days after imbibition, and from pistils (3) 1 day before flowering was hybridized with the RSP1 probe. The arrow indicates the hybridized band position. Positions and sizes (in kb) of rRNA are indicated. B Rice genomic DNA (5 µg per lane) digested with EcoRI (1), PstI (2) or SacI (3) was hybridized with the RSP1 probe. Positions and sizes (in kb) of markers are indicated

RT-PCR analysis of RSP1 mRNA expression To determine the tissue-specific expression pattern of the gene corresponding to RSP1, RT-PCR analysis was performed. RSP1 transcripts were detected only in floral organs, pistils and stamens; they were not detected in roots, leaves, panicles, or embryos (Fig. 4A). The

197

Fig. 6A–G In situ localization of RSP1 transcripts in a rice flower 1 day before flowering (A–E) and 2 h after flowering (F and G). A, B, F and G are longitudinal sections. C, D, and E are cross sections. A stigma; B anthers and filaments; C upper part of the ovary; D middle part of the ovary; E basal part of the ovary; F and G style and ovary. Sections were hybridized with the DIG-labeled antisense probe (A–F). Positive hybridization signals are shown in blue. Mature pistil section was hybridized with the DIG-labeled sense probe (G). an anther; fi filament; es embryo sac; nu nucellus; po pollen; sy style. The arrowhead shows the outer integument. Arrows show vascular bundles. Bars represent 200 µm

RSP1 transcripts were most abundant in mature stamens. In pistils, the expression level of RSP1 increased slightly at flowering and decreased afterwards (Fig. 4B). We could not detect RSP1 mRNA in developing embryos 14 days after flowering. It is noteworthy that the expression level of the transcript was lower in pistils at flowering after emasculation with hot water than in untreated pistils at flowering. The expression level observed in the emasculated pistils was almost the same as in the pistils 1 day before flowering.

198

In situ localization of RSP1 mRNA in the flower To determine whether the 541-bp fragment (Fig. 2) could be used for in situ hybridization analysis to define the spatial expression pattern during rice development, northern blot analysis and genomic Southern blot analysis were performed. The 541-bp fragment probe detected a single band of the predicted molecular weight in the northern analysis (Fig. 5A). The probe also detected a single band using several restriction enzymes in the Southern analysis (Fig. 5B). Then, in situ RNA hybridization experiments were performed using RSP1 fragment-specific digoxigenin-labeled antisense RNA probes. In pistils, RSP1 mRNA was expressed in almost the entire organ: stigma, style, ovary, integument, and nucellus (Fig. 6). The signal corresponding to the RSP1 transcript was strong in the ovary wall around the ovule, especially in the outer integument and the vascular bundle, while the signal was weak in the basal region of the ovary. The signal was also observed in the stamens. Transcripts were detected in the filament of stamens; on the other hand, no signal was detected in mature anthers or mature pollen (Fig. 6).

Discussion In this study, we isolated a rice subtilisin-like serine protease, RSP1, which is specifically expressed in reproductive organs. Recently, cDNA clones with high homology to subtilisin-like serine proteases have been reported from several plant species, including melon (Yamagata et al. 1994), lily (Kobayashi et al. 1994), Alnus glutinosa (Ribeiro et al. 1995), Arabidopsis (Ribeiro et al. 1995), and tomato (Tornero et al. 1996; 1997 Jorda et al. 1999; Meichtry et al. 1999). A gene family of subtilisin-like serine proteases in tomato has been extensively investigated (Jorda et al. 1999; Meichtry et al. 1999) and includes at least 15 members. Although we detected a single band in rice by genomic Southern blot analysis using a fragment of the RSP1 gene as a probe, it is possible that a gene family of subtilisin-like serine proteases also exists in rice. Based on sequence homology, the 15 tomato serine proteases can be grouped into five subfamilies (Meichtry et al. 1999). Each serine protease produced a distinctive expression pattern in tomato organs (Jorda et al. 1999; Meichtry et al. 1999). Proprotein convertases are also expressed in a highly tissue-specific manner in animals (Seidah et al. 1994). The diverse expression pattern with cell type, developmental stage, and environmental state, may contribute to the substrate specificity of these enzymes. RSP1 shares the highest similarity with the tomato p69 subfamily, which contains pathogen-induced serine proteases p69B and p69C (Jorda et al. 1999). It is well known that many pathogenesis-related proteins (PR proteins) are expressed in mature pistils (Atkinson et al. 1993). Pistils have to play antagonistic roles simultaneously, since they accept the pollen of their own species and refuse any other organ-

ism, including pollen of other species. It is possible that RSP1 as well as the PR proteins play defensive roles against these external invaders at flowering time. A stretch of 28 hydrophobic amino acid residues is found at the N-terminus of RSP1. This is typically observed in signal peptides responsible for targeting the protein to the secretory system. The lack of possible transmembrane segments and the computer analysis using the PSORT program (Nakai and Horton 1999) also support the idea that RSP1 is a protein secreted into the intercellular space of pistils, where pollen tubes elongate. A computer-based comparison of amino acid sequences indicated the existence of a 99-amino acid prosequence following a presequence in RSP1. The propeptide in subtilisin-like serine protease is likely responsible for repressing the activity of the enzyme. After secretion outside the cell, the prosequence of RSP1 may be removed to activate RSP1, as it occurs with other subtilisin-like proteases (Steiner et al. 1992). We detected a multiple poly(A) site in rsp1. In animals, numerous examples of mRNA with multiple poly(A) sites within a 3’-UTR have been described in the past few years (Edwards-Gilbert et al. 1997). Alternative polyadenylation is a consequence of a tandem array of poly(A) signals within a single 3’-UTR. In some cases, a different poly(A) site is selected in different tissues or at different developmental stages. Although seven poly(A) sites spread over 310 nucleotides were detected in this experiment, no preferential poly(A) site was observed. Whether the different lengths of mRNA have different stability or translatability has to be examined in the future. In animals, subtilases are generally involved in the maturation of peptide hormones, neuropeptides, growth factors, and receptor proteins (Steiner et al. 1992; Seidah et al. 1994; Nakayama 1997). These signal molecules are always derived from larger precursors and are proteolytically processed by members of a family of site-specific subtilisin-related proteases in animal signaling systems (Steiner et al. 1992). As stated, tomato subtilisin-like serine proteases p69B and p69C are induced in response to pathogen attack (Tornero et al. 1996; Jorda et al. 1999). In Alnus glutinosa, an actinorhizal nodule-specific gene, ag12, is expressed in infected cells before the onset of nitrogen fixation (Ribeiro et al. 1995). In pistils, the female reproductive system recognizes pollen grains and pollen tubes. All these phenomena (pathogenesis, nodulation, and reproduction) include cell-cell interaction events. It is noteworthy that subtilisin-like serine proteases, which are secreted into the extracellular space and are responsible for processing peptide signals, are commonly expressed in these spaces. Several peptide hormones are also reported in plant systems: a wound signal, systemin (McGurl et al. 1992), an early nodulin ENOD40 (Kouchi and Hata 1993; Yang et al. 1993), and growth factors PSKα and PSKβ (Matsubayashi and Sakagami 1996). A 50-kDa systemin-binding protein, which closely resembles a subtilase, was identified in tomato leaf (Schaller and Ryan 1994). There is, however,

199

no evidence to date that plant subtilases are responsible for producing these peptide hormones. To determine in which signaling system the plant subtilisin-like serine proteases operate, the enzyme substrates must be identified. Transcripts corresponding to RSP1 were abundant only in pistils and filaments of stamens. The transcripts were scarce or undetectable in leaves, roots, panicles, anthers, pollen, and embryos. The level of rsp1 gene expression in pistils increased slightly at flowering and decreased afterwards. These findings indicate that rsp1 expression is developmentally regulated. We do not know whether RSP1 plays similar roles inpistils and filaments. Rapid cell-elongation is observed in the filament of stamens at flowering time (Hoshikawa 1993). After flowering, the filaments abort rapidly. In pistils, rapid cell elongation and cell abortion are also observed (Hoshikawa 1993). To ensure rapid seed development, rapid cell elongation must occur in the ovary walls and integument. After pollination, tissues of the stigma, style, and ovary wall abort immediately. Cell elongation and cell abortion are, therefore, common phenomena observed in both pistils and filaments of stamens. Perhaps RSP1 is expressed in relation to these phenomena, if RSP1 plays similar roles in pistils and filaments. The question of downstream target proteins of RSP1 remains. Experiments to identify molecules interacting with RSP1 may allow elucidation of RSP1-dependent signal transduction.

References Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. J Mol Biol 215: 403–410 Atkinson AH, Heath RL, Simpson RJ, Clarke AE, Anderson MA (1993) Proteinase inhibitors in Nicotiana alata stigmas are derived from a precursor protein which is processed into five homologous inhibitors. Plant Cell 5: 203–213 Edwards-Gilbert G, Veraldi KL, Milcarek C (1997) Alternative poly(A) site selection in complex transcription units: means to an end? Nucleic Acids Res 25: 2547–2561 Hoshikawa K (1993) Anthesis, fertilization and development of caryopsis. In: Matsuo T, Hoshikawa K (eds) Science of the rice plant (Morphology, vol.1.) Food and Agriculture Policy Research Center, Tokyo Jorda L, Coego A, Conejero V, Vera P (1999) A genomic cluster containing four differentially regulated subtilisin-like processing protease genes is in tomato plants. J Biol Chem 274: 2360–2365 Kobayashi T, Kobayashi E, Sato S, Hotta Y, Miyajima N, Tanaka A, Tabata S (1994) Characterization of cDNAs induced in meiotic prophase in lily microsporocytes. DNA Res 1: 15–26 Kouchi H, Hata S (1993) Isolation and characterization of novel nodulin cDNA representing genes expressed at early stages of soybean nodule development. Mol Gen Genet 238: 106–119 Kyte J, Doolittle RF (1982) A simple method for displaying the hydropathic character of a protein. J Mol Biol 157: 105–132 Matsubayashi Y, Sakagami Y (1996) Phytosulfokine, sulfated peptides that induce the proliferation of single mesophyll cells of Asparagus officialis L. Proc Natl Acad Sci USA 93: 7623– 7627

McGurl B, Pearce G, Orozco-Cardenas M, Ryan CA (1992) Structure, expression, and antisense inhibition of the systemin precursor gene. Science 255: 1570–1573 Meichtry J, Amrhein N, Schaller A (1999) Characterization of the subtilase gene family in tomato (Lycopersicon esculentum Mill.). Plant Mol Biol 39: 749–760 Nakai K, Horton P (1999) PSORT: a program for detecting sorting signals in proteins and predicting their subcellular localization. Trend Biochem Sci 24: 34–35 Nakayama K (1997) Furin: a mammalian subtilisin/Kex2p-like endoprotease involved in processing of a wide variety of precursor proteins. Biochem J 327: 625–635 Nakazono M, Yoshida KT (1997) A rapid and efficient method for the isolation of differentially expressed genes: simplified differential display. Plant Biotechnol 14: 187–190 Ribeiro A, Akkermans ADL, Kammen A van, Bisseling T, Pawlowski K (1995) A nodule-specific gene encoding a subtilisin-like protease is expressed in early stages of actinorhizal nodule development. Plant Cell 7: 785–794 Schaller A, Ryan CA (1994) Identification of a 50-kDa systeminbinding protein in tomato plasma membranes having Kex2plike properties. Proc Natl Acad Sci USA 91: 11802–11806 Seidah NG, Chretien M, Day R (1994) The family of subtilisin/ kexin like pro-protein and pro-hormone convertases: divergent or shared functions. Biochimie 76: 197–209 Siezen RJ, Vos WM de, Leunissen JAM, Dijkstra BW (1991) Homology modeling and protein engineering strategy of subtilases, the family of subtilisin-like serine proteinases. Protein Eng 4: 719–737 Steiner DF (1998) The proprotein convertases. Curr Opin Chem Biol 2: 31–39 Steiner DF, Smeekens SP, Ohagi S, Chan SJ (1992) The new enzymology of precursor processing endoproteases. J Biol Chem 267: 23435–23438 Taylor A, Horsch A, Rzepczyk A, Hasenkampf CA, Riggs CD (1997) Maturation and secretion of a serine proteinase is associated with events of late microsporogenesis. Plant J 12: 1261–1271 Tornero P, Conejero V, Vera P (1996) Primary structure and expression of a pathogen-induced protease (PR-P69) in tomato plants: similarity of functional domains to subtilisin-like endoproteases. Proc Natl Acad Sci USA 93: 6332–6337 Tornero P, Conejero V, Vera P (1997) Identification of a new pathogen-induced member of the subtilisin-like processing protease family from plants. J Biol Chem 272: 14412–14419 Yamagata H, Masuzawa T, Nagaoka Y, Ohnishi T, Iwasaki T (1994) Cucumisin, a serine protease from melon fruits, shares structural homology with subtilisin and is generated from a large precursor. J Biol Chem 269: 32725–32731 Yang WC, Katinakis P, Hendriks P, Smolders A, Vries F de, Spee J, Kammen A van, Bisseling T, Franssen H (1993) Characterization of GmENOD40, a gene showing novel patterns of cellspecific expression during soybean nodule development. Plant J 3: 573–585 Yoshida KT, Kuboyama T (1997) Isolation of sexual reproductionrelated genes in rice by simplified differential display. Rice Genet Newslet 14: 126–129 Yoshida KT, Naito S, Takeda G (1994) cDNA cloning of regeneration-specific genes in rice by differential screening of randomly amplified cDNAs using RAPD primers. Plant Cell Physiol 35: 1003–1009 Yoshida KT, Wada T, Koyama H, Mizobuchi-Fukuoka R, Naito S (1999) Temporal and spatial patterns of accumulation of the transcript of myo-inositol-1-phosphate synthase and phytincontaining particles during seed development in rice. Plant Physiol 119: 65–72 Zhou A, Paquet L, Mains RE (1995) Structural elements that direct specific processing of different mammalian subtilisin-like prohormone convertases. J Biol Chem 270: 21509–21516