Advance Access published February 27, 2004 Journal of Experimental Botany, Page 1 of 11 DOI: 10.1093/jxb/erh095
RESEARCH PAPER
Molecular cloning and characterization of a cDNA encoding asparaginyl endopeptidase from sweet potato (Ipomoea batatas (L.) Lam) senescent leaves Hsien-Jung Chen1, Wen-Chi Hou2, Jih-Shiou Liu3, Chih-Yuan Yang4, Dong-Jiann Huang3 and Yaw-Huei Lin3,* 1 2 3
Institute of Botany, Academia Sinica, Nankang, 115 Taipei, Taiwan Life Science, Liberal Arts Center, HsiWu College, 224 Taipei, Taiwan
Received 19 November 2003; Accepted 24 November 2003
Abstract Asparaginyl endopeptidase is a cysteine endopeptidase that has strict substrate speci®city toward the carboxyl side of asparagine residues, and is possibly involved in the post-translational processing of proproteins. In this report one full-length cDNA, SPAE, was isolated from senescent leaves of sweet potato (Ipomoea batatas (L.) Lam). SPAE contained 1479 nucleotides (492 amino acids) in the open reading frame, and exhibited high amino acid sequence homologies (c. 61±68%) with asparaginyl endopeptidases of Vicia sativa, Phaseolus vulgaris, Canavalia ensiformis, and Vigna mungo. SPAE probably encoded a putative precursor protein. Via cleavage of the N- and C-termini, it produced a mature protein containing 325 amino acids (from the 51st to the 375th amino acid residues), the conserved catalytic residues (the 173rd His and 215th Cys amino acid residues), and the putative N-glycosylation site (the 332nd Asn amino acid residue). Semi-quantitative RTPCR and western blot hybridization showed that SPAE gene expression was enhanced signi®cantly in natural senescent leaves and in dark- and ethephoninduced senescent leaves, but was much less in mature green leaves, stems, and roots. Phylogenic analysis showed that SPAE displayed close association with vacuolar processing enzymes (legumains/ asparaginyl endopeptidases), which function via cleavage for proprotein maturation in the protein bodies during seed maturation and germination. In
conclusion, sweet potato SPAE is probably a functional, senescence-associated gene and its mRNA and protein levels were signi®cantly enhanced in natural and induced senescent leaves. The possible role and function of SPAE associated with bulk protein degradation and mobilization during leaf senescence were also discussed. Key words: Asparaginyl endopeptidase, dark, ethephon, leaf senescence, sweet potato.
Introduction During leaf senescence, breakdown of leaf proteins by proteinases provides a large pool of cellular nitrogen for recycling (Makino and Osmond, 1991). In plants, three major degradation pathways have been described: (a) the ubiquitin-dependent pathway, (b) the chloroplast degradation pathway, and (c) the vacuolar degradation pathways (Vierstra, 1996). Among these pathways, vacuolar degradation is assumed to be involved in bulk protein degradation by virtue of the resident proteinases in the vacuole. Two types of vacuoles have been described in plants: (a) the storage vacuole and (b) the lytic central vacuole (Marty, 1999). Protein storage vacuoles are often found in seed tissues and accumulate proteins that are remobilized and used as the main nutrient resource for germination (Senyuk et al., 1998; Schlereth et al., 2001). Most cells in vegetative tissues have a large central
* To whom correspondence should be addressed. Fax: +886 2 27827954. E-mail:
[email protected] Journal of Experimental Botany, ã Society for Experimental Biology 2004; all rights reserved
Downloaded from http://jxb.oxfordjournals.org/ by guest on December 7, 2014
4
Department of Horticulture, Chinese Culture University, 111 Taipei, Taiwan Graduate Institute of Pharmacognosy Science, Taipei Medical University, 110 Taipei, Taiwan
2 of 11 Chen et al.
responsible for the degradation of seed proteins accumulated in protein bodies (protein storage vacuoles). SH-EP belongs to the papain proteinase family and has an N-terminal and a C-terminal prosegment (Okamoto and Minamikawa, 1999; Okamoto et al., 1999). Okamoto and Minamikawa (1995) isolated a processing enzyme, designated VmPE-1. VmPE-1 is a member of the asparaginyl endopeptidases and is involved in the post-translational processing of SH-EP. In addition, the cleavage sites of the in vitro processed intermediates and the mature form of SH-EP were identical to those of SH-EP puri®ed from germinated cotyledons of V. mungo. Therefore, it is proposed that the asparaginyl endopeptidase (VmPE-1)mediated processing mainly functions in the activation of proSH-EP during seed germination (Okamoto et al., 1999). The activated SH-EP plays a major role in the degradation of seed storage proteins accumulated in cotyledonary vacuoles of Vigna mungo seedlings (Mitsuhashi et al., 1986). These reports demonstrate a role of asparaginyl endopeptidase associated with cleavage and maturation of seed proteins during seed development and germination. It is reported here that a full-length cDNA, SPAE, was isolated from senescent leaves of sweet potato and exhibited high amino acid sequence homologies to seed vacuolar legumains/asparaginyl endopeptidases of kidney bean (Phaseolus vulgaris), spring vetch (Vicia sativa), and jack bean (Canavalia ensiformis). Gene expression of SPAE was signi®cantly enhanced in both natural and induced senescent leaves. A possible role of SPAE in senescent leaves associated with the cleavage and maturation of proproteins, which are involved in protein degradation and recycling similar to the mechanism utilized in seeds for bulk storage protein mobilization, is also addressed. Materials and methods Plant materials The storage roots of sweet potato (Ipomoea batatas (L.) Lam) were purchased from a local supermarket and grown in the greenhouse. Plantlets were grown from the storage roots. Mature green leaves near the top of stems and leaves with different levels of senescence were used for experiments. PCR-based subtractive hybridization and RACE PCR Total RNAs were isolated separately from the mature green leaves and senescing leaves of sweet potato, basically according to the method of Sambrook et al. (1989). The mRNAs were puri®ed with a puri®cation kit (Promega) and used for the differentially-expressed ®rst strand cDNA synthesis with a PCR-based subtractive hybridization kit (Clontech) following the protocols supplied by the manufacturer. The double-strand cDNAs of senescing leaves were subtracted by that of mature green leaves, then ligated to pGEM-T vector for E. coli DH5a competent cell transformation. Recombinant plasmids were isolated for DNA sequencing using an ABI PRIZM 337 DNA Sequencer. Nucleotide sequence data were analysed using the Genetics Computer Group (GCG) programs. The RACE PCR method with the Marathon cDNA ampli®cation kit
Downloaded from http://jxb.oxfordjournals.org/ by guest on December 7, 2014
vacuole, containing a wide range of proteinases in an acidic environment. Substrate proteins must be transported and sequestered into this vacuole for degradation. The molecular mechanisms for vacuolar protein degradation and the nutrient recycling pathway in senescent leaves are generally not clear. Recently, a novel group of cysteine endopeptidases was found in seeds. They have strict cleavage speci®city for the peptide bonds of seed storage proteins with asparagines at the P1 position, and are called asparaginyl endopeptidases (Ishii, 1994). The substrate speci®city has been observed with puri®ed asparaginyl endopeptidases from developing seeds of castor bean (Hara-Nishimura et al., 1991) and soybean (Scott et al., 1992; Hara-Nishimura et al., 1995), from mature seeds of jack bean (Abe et al., 1993), and from germinating seeds of vetch (Becker et al., 1995). Many seeds accumulate sizeable reserves of proteins in storage vacuoles during seed development, and a number of these proteins undergo proteolytic cleavage as they accumulate in the vacuole including the 7S and 11S seed storage globulins and 2S seed storage albumins (MuÈntz and Shutov, 2002). The 11S seed storage proteins are synthesized as precursors that are cleaved post-translationally in storage vacuoles by an asparaginyl endopeptidase (Ishii, 1994). In castor bean and soybean seeds, vacuolar processing enzymes were detected in the protein bodies and probably associated with the conversion of proproteins to their corresponding mature forms in vacuoles (Hara-Nishimura et al., 1991; Shimada et al., 1994). Asparaginyl endopeptidases also play a role with bulk degradation and mobilization of storage proteins during seed germination and seedling growth. An asparaginylspeci®c cysteine endopeptidase which was named `legumain-like proteinase' (LLP) was isolated from cotyledons of kidney bean (Phaseolus vulgaris) seedlings. It was the ®rst proteinase ever known which in vitro extensively degrades native phaseolin, the major storage globulin of this grain legume (Senyuk et al., 1998). In vetch (Vicia sativa) seeds, the legumain-like VsPB2 and proteinase B, together with additional papain-like cysteine proteinases, were responsible for the bulk breakdown and mobilization of storage globulins during seed germination (Schlereth et al., 2000). In Arabidopsis, the seed protein pro®les were compared between the wild type and a seed-type vacuolar processing enzyme bVPE mutant using a two-dimensional gel/mass spectrometric analysis. A signi®cant increase in the accumulation of several legumin-type globulin propolypeptides was found in bVPE mutant seeds (Gruis et al., 2002). The mechanism of asparaginyl endopeptidases (SH-EP and VmPE-1) associated with bulk storage protein degradation in seed has been studied in Vigna mungo. A vacuolar cysteine proteinase, designated SH-EP, is synthesized in cotyledons of germinated Vigna mungo seeds and is
Sweet potato asparaginyl endopeptidase 3 of 11 the NJ plot and redrawn by the graphic software of a Macintosh computer.
Southern blot hybridization Young (not expanded) leaves of sweet potato were harvested and ground in liquid N2. The powder was transferred to a centrifuge tube, mixed gently and thoroughly with cetyltrimethylammonium bromide (CTAB) buffer (2% CTAB, 1.4 mol l±1 NaCl, 20 mmol l±1 ethylenediaminetetraacetate (EDTA), 0.2% b-mercaptoethanol, and 100 mmol l±1 TRIS-HCl pH 8.0) in a 20:1 (v:w) ratio, and kept at 60 °C in a water bath for genomic DNA extraction according to the method of Chen et al. (2000). The total nucleic acid after precipitation with an equal volume of isopropanol was redissolved in sterile water, digested with restriction enzymes of HindIII or XbaI and separated on a 0.8% agarose gel. For semi-quantitative detection of PCR products from different tissues, dark-treated or ethephontreated mature green leaves, the ampli®ed cDNAs were separated on 1.5% agarose gels. After electrophoresis, the DNA was transferred onto a Hybond-N+ nylon membrane (Amersham) following the protocol of Molecular Cloning (Sambrook et al., 1989) for Southern blot hybridization.
Construction, overexpression, and puri®cation of SPAE fusion protein from E. coli The full-length SPAE cDNA was used as the template to amplify the PCR products encoding the mature SPAE protein with primers (Y192-5¢M: GGTATTGAGGGTCG CTCCGTCGGCACTCGTTGGGCC and Y192-3¢S: AGAGGAGAGTTAGAGCCC TGGTTCACAGCCCTTCTT). The ampli®ed PCR products were puri®ed ®rst and then cloned directly into PET32Xa/LIC vector (Novagen) according to the protocols provided by the supplier. After induction with 1 mM IPTG, the expressed fusion proteins were extracted from cells with 8 M urea, and puri®ed with a His-tag af®nity column according to the protocols from Novagen. The puri®ed fusion protein was digested with a protease Xa factor to release the expressed SPAE mature protein for N-terminal amino acid sequencing and as an antigen for polyclonal antibody production in rabbits.
Measurement of pigments For quantitative analysis of pigment contents, the mature green leaves (S0) and senescing leaves (S1 and S2) of sweet potato were collected separately and extracted with 80% acetone (pH 7.8) buffered with 2.5 mM sodium phosphate according to the method of Chen et al. (2000). The absorbance of extracts was measured at wavelengths of 663.8 nm, 646.8 nm, and 470 nm, respectively. Quantitative values of pigments for aqueous 80% acetone extracts were calculated from the absorbance data according to the report of Lichtenthaler (1987). For dark treatment, mature green leaves were detached, and placed on a wet paper towel containing 3 mM 2-(Nmorpholino)ethanesulphonic acid (MES) buffer pH 7.0, then kept at room temperature in the dark for 0, 1, 3, 6, 9, and 12 d. For ethephon treatment, the detached mature green leaves were placed on a wet paper towel containing 3 mM MES (pH 7.0) and 1 mM ethephon, then kept in the dark for 0, 1, 2, and 3 d. Leaves were individually collected for quantitative analysis of pigment contents as described previously, and also for semi-quantitative RT-PCR as described below. Semi-quantitative RT-PCR Total RNA was isolated from (a) different tissues including roots, stems, mature green leaves, and senescing leaves, and (b) darktreated, or 1 mM ethephon-treated mature green leaves of sweet potato. The primers (Y192A: TGTGCTGACGATGATTCGCT CCGTC and Y192B: ACTGCAAATTATGCACTGAATCCTC) were used to amplify the full lengths of double-strand cDNAs for semi-quantitative analysis according to the method of Jonson et al. (2000). The full-length cDNA of SPAE in the recombinant plasmid was labelled with digoxigenin-11-dUTP nucleotides as the probe to detect semi-quantitative RT-PCR products using Southern blot hybridization and CSPD substrate (Boehringer Mannheim) as described previously. Phylogenic analysis Amino acid sequence alignment of SPAE with published plant cysteine endopeptidases after GCG/fasta comparison was used for phylogenic tree construction. The distances among entries were calculated with the Neighbor±Joining method (Thompson et al., 1994), and the BLOSUM series matrix (80, 62, 45, 30) of software Clustal W (version 1.7) was used as parameters in the alignment with default setting for gap penalty. The phylogenic tree was drawn using
Determination of N-terminal amino acid sequence of the puri®ed SPAE fusion protein After digestion with the Xa factor, the puri®ed proteins were mixed with the SDS sample extraction buffer and boiled in a 100 °C water bath for 5 min. The samples were subjected to protein SDS PAGE in 12.5% gels, then, transferred onto Millipore PVDF membranes. The band with molecular weight corresponding to the expressed SPAE mature protein was cut from the PVDF membrane and used for N-terminal amino acid determination. Western blot hybridization and activity staining After digestion with the Xa factor, the band with a molecular weight corresponding to the expressed SPAE mature protein was cut from the 12.5% polyacylamide gel, and was then mixed with the appropriate amount of pH 7.5 phosphate buffer saline (PBS) containing 0.1% SDS. The eluted proteins in PBS containing 0.1% SDS were precipitated with acetone containing 10% trichloroacetic acid (TCA) at ±20 °C for 2 h. After centrifugation at 13 000 g for 20 min, the pellet was washed with acetone twice before being dried at room temperature. The acetone powder was redissolved in a small amount of PBS buffer containing 0.1% SDS and used as the antigens for subcutaneous injections (Taiwan Bio-Pharm Inc.). The polyclonal antibody obtained from rabbit antiserum was used for western blot hybridization to study the gene expression of SPAE in different tissues, leaves with various levels of senescence, and with dark or ethephon treatments as described previously. For activity staining, basically it follows the method reported by Lee and Lin (1995). The crude protein extract from S1 leaves was analysed for protease activity in the presence or absence of inhibitor E-64 in gelatincontaining polyacrylamide gel.
Results DNA and amino acid sequences of SPAE
In order to isolate senescence-associated genes, sweet potato leaves were divided into three stages based on their cellular pigment contents. S0 was the stage of fully expanded green leaves, the chlorophyll content of which was assigned as 100% for a comparison standard. S1 was the stage of senescing leaves with c. 25% chlorophylls, and the completely yellow leaves were classi®ed as S2 with less than 5% chlorophylls (Chen et al., 2003). With PCRbased subtractive hybridization and RACE PCR tech-
Downloaded from http://jxb.oxfordjournals.org/ by guest on December 7, 2014
(Clontech) was used to isolate the 5¢ and 3¢ ends of the necessary cDNAs according to the protocols provided by the manufacturer.
4 of 11 Chen et al.
niques, a full-length cDNA, Y192 (GenBank accession number AF260827), was cloned from senescent leaves and renamed as SPAE. There were 1479 nucleotides (492 amino acids) in its open reading frame (Fig. 1). GCG/fasta comparison showed that SPAE exhibited high amino acid sequence homologies (61±68%) with asparaginyl endo-
peptidases of Vicia sativa, Phaseolus vulgaris, Canavalia ensiformis, and Vigna mungo. SPAE probably encoded an asparaginyl endopeptidase precursor. After comparison of SPAE with Phaseolus vulgaris legumain-like proteinase (LlP) (Senyuk et al., 1998) and Vigna mungo vacuolar processing enzyme 1
Downloaded from http://jxb.oxfordjournals.org/ by guest on December 7, 2014
Fig. 1. DNA and amino acid sequences of SPAE isolated from sweet potato senescent leaves. ATG (underlined) and TAA (underlined) represent the start codon and the stop codon, respectively. The His (H) and Cys (C) residues within the grey boxes represent the conserved catalytic amino acid residues. The amino acid ASN (N) within a grey box is the putative N-glycosylation site. The arrow () represents the possible cleavage sites of N- and C-termini of the SPAE-encoded protein precursor.
Sweet potato asparaginyl endopeptidase 5 of 11
(VmPE1) (Okamoto and Minamikawa, 1999), the ®rst 50 amino acids of the N-terminus and the last 117 amino acids of the C-terminus of SPAE-encoded protein precursor were cleaved off. Thus, the mature SPAE protein probably contained 325 amino acid residues starting from the 51st to the 375th amino acid residues. The 332nd amino acid residue (N) was the putative N-glycosylation site shaded in grey (Fig. 1).
Figure 2 showed the alignment of the SPAE putative catalytic domain (from the 163rd to the 237th amino acid residues) with the other cysteine proteinases. A conservation of the catalytic residues within the domains was observed among these cysteine proteinases. The 173rd His (H) and the 215th Cys (C) amino acids of SPAE labelled with asterisks and printed in white on grey were identi®ed as conserved catalytic residues (Figs 1, 2). There is a block
Downloaded from http://jxb.oxfordjournals.org/ by guest on December 7, 2014
Fig. 2. Alignment of the putative SPAE catalytic domain (from the 163rd to the 237th amino acid residues) with the other cysteine proteinase families. The 173rd His (H) and 215th Cys (C) amino acids printed in white on grey represent the conserved catalytic residues. There is a block of four predominantly hydrophobic residues printed in white on black that are two residues N-terminal to each of the two catalytic residues. (A) Legumains/asparaginyl endopeptidases. AF260827 (Ipomoea batatas SPAE); AL000990 (mouse); Q99538 (human); AB025310 (Oryza sativa); AB081464 (Oryza sativa); AF169973 (Sesamum indicum); AF238384 (Vigna radiata); AJ131718 (Zea mays); D31787 (Canavalia ensiformis); D89971 (Vigna mungo); D89972 (Vigna mungo); JX0344 (Canavalia ensiformis); P49046 (Canavalia ensiformis); S49175 (Vicia sativa); T12043 (Phaseolus vulgaris); T12044 (Phaseolus vulgaris); Z99956 (Phaseolus vulgaris); Z99957 (Phaseolus vulgaris). (B) Vacuolar processing enzymes. AB075947 (Nicotiana tabacum); AB075948 (Nicotiana tabacum); AB075949 (Nicotiana tabacum); AB075950 (Nicotiana tabacum); AF521661 (Arabidopsis thaliana); AJ243876 (Lycopersicon esculentum); AJ309173 (Beta vulgaris); D17401 (Ricinus communis); D61393 (Arabidopsis thaliana); D61394 (Arabidopsis thaliana); D61395 (Arabidopsis thaliana); JQ2387 (Ricinus communis); S60050 (Arabidopsis thaliana); T02629 (Arabidopsis thaliana); T05302 (Arabidopsis thaliana). (C) Cysteine proteinases. AJ007743 (Vicia sativa); D28876 (Glycine max); S51117 (Citrus sinensis); T07132 (Glycine max); T10944 (Vicia sativa); T19231 (Caenorhabditis elegans).
6 of 11 Chen et al.
of four predominantly hydrophobic residues printed in white on black that are two residues N-terminal to each of the two catalytic residues. These two blocks were also conserved among these cysteine proteinases and were contained in two central b-strands that were reported to support the catalytic residues, His and Cys (Chen et al., 1998). Southern blot hybridization showed that there was one band (c. 1.7 kb) for HindIII digestion and two bands (c. 18 and 6.5 kb) for XbaI digestion detected with the SAPE probe (Fig. 3). These data suggest that SPAE encodes a putative asparaginyl endopeptidase and may comprise a low copy number in the sweet potato genome. Gene expression of SPAE is enhanced in senescent leaves
Fig. 3. Southern blot hybridization of sweet potato SPAE. The genomic DNA of sweet potato was digested with restriction enzymes (HindIII, or XbaI) and detected with the digoxigenin-labelled SPAE probe.
Downloaded from http://jxb.oxfordjournals.org/ by guest on December 7, 2014
The temporal and spatial expression patterns of SPAE were studied with semi-quantitative RT-PCR according to Jonson et al. (2000). The amount of PCR-ampli®ed product was detected by ethidium bromide (EtBr) staining and dig-labelled probe hybridization. SPAE gene expres-
sion was signi®cantly enhanced during natural leaf senescence. The ampli®ed PCR product was remarkably increased from the S0 to the S1 and S2 stages. However, the amount in mature green leaves (S0 stage), root, and stem was much less than that of senescent leaves (S1 and S2 stages). A metallothionein-like protein gene, G14, which was isolated previously from sweet potato leaves, exhibited a constitutive expression pattern in all tissues assayed (Chen et al., 2003) and was used as a control. No signi®cant variation in G14 gene expression level was found among the tissues and stages analysed (Fig. 4A). These data suggest that SPAE is likely to be a senescenceassociated gene that exhibits an enhanced expression pattern during natural leaf senescence. Western blot hybridization with polyclonal antibody raised against SPAE showed that a band with a molecular
Fig. 4. Temporal and spatial gene expression of sweet potato SPAE detected with both RT-PCR and Western blot. (A) RT-PCR products. The ampli®ed full-lengths of SPAE RT-PCR products were detected with both ethidium bromide staining (EtBr staining) and nonradioactive digoxigenin-labelled SPAE probe. (B) Western blot. Total proteins extracted from leaves were separated by SDS-PAGE on 12.5% gels and detected with a polyclonal antibody raised against SPAE. S0, S1, and S2 represented different senescent stages (100%, 25%, and 5% chlorophyll content, respectively) of leaves. G14, which encoded a metallothionein-like protein and exhibited a constitutive expression pattern, was used as a control. The experiments were performed three times and a representative one was shown. (C) Activity staining. The crude protein extract from S1 leaves was analysed for protease activity in the presence or absence of inhibitor E-64 in gelatin-containing polyacrylamide gel.
Sweet potato asparaginyl endopeptidase 7 of 11
mass between 36 kDa and 50 kDa was detected and its amount increased remarkably from the S0 to the S1 and S2 stages (Fig. 4B). The data are consistent with the semiquantitative RT-PCR results (Fig. 4A) and provide further evidence to support SPAE as a functional gene. Activity staining with crude protein extract from S1 leaves in gelatin-containing polyacrylamide gel showed that an activity band with a molecular mass between 36 kDa and 50 kDa was also detected that was insensitive to inhibitor E-64 inhibition. However, the other activity band with a molecular mass between 30 kDa and 36 kDa exhibited sensitivity to inhibitor E-64 inhibition. These data agree with immunoblot results and provide indirect evidence to support the existence of an asparaginyl endopeptidase with a molecular mass between 36 kDa and 50 kDa in senescent leaves (Fig. 4C).
In order to characterize the gene expression patterns of SPAE, mature green leaves were treated with different inducers that have been reported to cause senescence. A decrease of the pigment contents was used as an indication of leaf senescence. For dark treatment, contents of chlorophylls (a+b) decreased gradually and the amount at day 12 was about one-third that of day 0 (Fig. 5A). Gene expression of SPAE detected by ethidium bromide staining
Fig. 5. Effects of dark treatment on SPAE gene expression in mature green leaves within a 12 d period. (A) Changes of SPAE RT-PCR products in dark-treated mature green leaves. The RT-PCR products were detected with both ethidium bromide (EtBr) staining and nonradioactive digoxigenin-labelled SPAE probe. G14, which encoded a metallothionein-like protein and exhibited a constitutive expression pattern, was used as a control. D denotes dark treatment. (B) Western blot. Total proteins extracted from leaves were separated by SDSPAGE on 12.5% gels and detected with a polyclonal antibody raised against SPAE. The experiments were performed three times and a representative one was shown.
Fig. 6. Effect of ethephon, an ethylene-releasing compound, on SPAE gene expression in mature green leaves within a 3 d period. (A) Changes of SPAE RT-PCR products in 1 mM ethephon-treated and dark-treated mature green leaves. The RT-PCR products were detected with both ethidium bromide (EtBr) staining and non-radioactive digoxigenin-labelled SPAE probe. G14, which encoded a metallothionein-like protein and exhibited a constitutive expression pattern, was used as a control. E and D denote ethephon and dark treatments, respectively. (B) Western blot. Total proteins extracted from leaves were separated by SDS-PAGE on 12.5% gels and detected with a polyclonal antibody raised against SPAE. The experiments were performed three times and a representative one was shown.
Downloaded from http://jxb.oxfordjournals.org/ by guest on December 7, 2014
Gene expression of SPAE is enhanced by dark and ethephon treatments
or Southern blot hybridization was not signi®cantly increased from day 0 to day 6; however, it was remarkably enhanced from day 9 to day 12 after dark treatment (Fig. 5A). For G14, no signi®cant variation of RT-PCR products was found during the time intervals after dark treatment (Fig. 5A). Western blot hybridization with a polyclonal antibody raised against SPAE showed that a band with molecular weight between 36 kDa and 50 kDa was detected from day 9 to day 12 (Fig. 5B). For 1 mM ethephon treatment, loss of contents of chlorophylls (a+b) from treated mature green leaves were much faster than that of the untreated dark control and the amount at day 3 was about one-third that of day 0 (Fig. 6A). The semi-quantitative RT-PCR products of SPAE detected by ethidium bromide staining or Southern blot hybridization increased from day 1 to day 3 in ethephon-treated samples. Whereas, no signi®cant change among SPAE RTPCR products from untreated dark controls was observed. The increased amounts in ethephon-treated samples were much higher and faster than that of the untreated dark controls from day 1 to day 3, and were correlated with the rate of leaf senescence using changes of pigment contents as a senescent indicator. For G14, no signi®cant variation of RT-PCR products was found during the time intervals after ethephon treatment (Fig. 6A). Western blot hybridi-
8 of 11 Chen et al.
zation with polyclonal antibody against SPAE also detected a band with molecular weight between 36 kDa and 50 kDa from day 2 till day 3 after ethephon treatment (Fig. 6B), which is consistent with the semi-quantitative RT-PCR results (Fig. 6A). These data of dark and ethephon treatments, therefore, further support SPAE as a functional, senescence-associated gene during leaf senescence.
Discussion In this report, it was found that sweet potato SPAE exhibited high amino acid sequence homologies (61±68%) with plant asparaginyl endopeptidases of Vicia sativa, Phaseolus vulgaris, Canavalia ensiformis, and Vigna mungo (Fig. 1). The conserved catalytic residues (His and Cys) and central b-strands that supported the catalytic residues of human and mouse legumains (Chen et al., 1998) were also found in SPAE, plant legumain/asparaginyl endopeptidase, vacuolar processing enzymes, and the other cysteine proteinases (Fig. 2). From RT-PCR products and western blot hybridization, SPAE-encoded products could be detected in mRNA and protein levels
Downloaded from http://jxb.oxfordjournals.org/ by guest on December 7, 2014
Phylogenetic analysis of SPAE The phylogenic relationship of SPAE with the other cysteine proteinases was shown in Fig. 7. Sweet potato SPAE (AF260827) was grouped together with the vacuolar processing enzymes of sugar beet (Beta vulgaris Vp1 gene (AJ309173)) and tobacco (Nicotiana tabacum AB075947, AB075948, AB075949, AB075950). The sugar beet Vp1 gene was speci®cally expressed in taproots, but not in leaves, stems, and in¯orescences (Kloos et al., 2002). SPAE also exhibited a close relationship with another group of vacuolar processing enzymes (legumains/asparaginyl endopeptidases) from germinating seeds of Vigna radiata (AF238384), Phaseolus vulgaris (Z99956) (Becker et al., 1995), Vigna mungo (D89972) (Okamoto and Minamikawa, 1999), Vicia sativa (Z34899, proteinase B) (Schlereth et al., 2001), and Vigna mungo (D89971, VmPE-1) (Okamoto and Minamikawa, 1999). SPAE, however, displayed a more distantly related association with other vacuolar processing enzymes (legumains/ asparaginyl endopeptidases) from the seeds of Arabidopsis thaliana (AF521661) (Gruis et al., 2002), Phaseolus vulgaris (Z99957) (Senyuk et al., 1998), Canavalia ensiformis (D31787) (Takeda et al., 1994), Glycine max (D28876) (Shimada et al., 1994), Ricinus communis (D17401) (Hara-Nishimura et al., 1991), Vicia sativa (AJ007743, VsPB2) (Schlereth et al., 2000, 2001); and Oryza sativa (AB025310 and AB081464). These data suggest that SPAE may encode a putative vacuolar asparaginyl endopeptidase with a possible function similar to the vacuolar processing enzymes (legumains/asparaginyl endopeptidases) of seeds during maturation and/or germination.
(Figs 4, 5, 6). These data provide evidence to support the isolated sweet potato SPAE as a functional gene. SPAE encoded a putative mature protein, which contained 325 amino acid residues and an N-glycosylation site at the C-terminus (Fig. 1). The deduced molecular weight of mature SPAE protein was, therefore, probably more than 36 kDa and agreed with the protein band detected by western blot hybridization that had a molecular weight between 36 kDa and 50 kDa (Figs 4B, 5B, 6B). Asparaginyl endopeptidase is an atypical cysteine endopeptidase with a reported insensitivity to inhibitor L-3carboxy-2,3-trans-epoxypropionyl-leucyl-amino (4-guanidino)butane (E-64) (Okamoto and Minamikawa, 1999). Cysteine proteinase activities were assayed with total crude protein extract from senescent leaves (S1 stage) in gelatin-containing polyacryamide gels. A cysteine proteinase activity band with a molecular mass between 36 kDa and 50 kDa similar to the western blot results (Figs 4, 5, 6) was also detected and exhibited insensitivity to the E-64 inhibitor. These data provide indirect evidence to support the existence of asparaginyl endopeptidase in senescent leaves. In sweet potato, the SPAE gene expression level is higher in dark- or ethephon-induced senescent leaves similar to that in natural senescent leaves (S1 and S2 stages) (Figs 4, 5, 6); but is much lower under non-induced state (S0 stage). The effect of ethephon as a senescence accelerator is indirect. It is ®rst decomposed into HCl, phosphate, and ethylene in a 1:1:1 ratio before the ethylene action. In order to prevent the side-effects caused by HCl and phosphate, ®nal concentrations of 1 mM HCl and 1 mM phosphate were added to the control when 1 mM ethephon was used for treatment. No signi®cant effects of HCl and phosphate on SPAE gene induction were observed (Fig. 6A). Hormones such as jasmonic acid (2 mM) also caused the decrease of chlorophyll contents in treated leaves; whereas it did not signi®cantly alter the SPAE gene expression level compared with that of the untreated dark control in mature green leaves within a 3 d period (data not shown). These data suggest that SPAE is a senescenceassociated gene and its expression in natural or induced senescent leaves is probably controlled by ethylene, but not by JA. The role and function of SPAE in sweet potato senescent leaves are not clear. However, vacuolar processing enzymes (legumains/asparaginyl endopeptidases) have been implicated in the deposition and mobilization of storage proteins, globulins, during seed maturation and germination/seedling growth in Phaseolus vulgaris (Senyuk et al., 1998), Vigna mungo (Okamoto et al., 1999), Vicia sativa (Schlereth et al., 2000, 2001), and Arabidopsis thaliana (Gruis et al., 2002). Phylogenic analysis placed SPAE in close association with plant vacuolar processing enzymes (legumains/asparaginyl endopeptidases) from seeds during maturation (Hara-
Sweet potato asparaginyl endopeptidase 9 of 11
Nishimura et al., 1991; Shimada et al., 1994; Takeda et al., 1994) and/or germination (Becker et al., 1995; Senyuk et al., 1998; Okamoto et al., 1999; Gruis et al., 2002) (Fig. 7). In castor bean and soybean seeds, vacuolar processing enzymes were detected in the protein bodies and are probably associated with the conversion of
proproteins to their corresponding mature forms in vacuoles (Hara-Nishimura et al., 1991; Shimada et al., 1994). In Vigna mungo, VmPE-1 has been demonstrated to increase in the cotyledons of germinating seeds and was involved in the post-translational processing of a vacuolar cysteine endopeptidase, designated SH-EP, which de-
Downloaded from http://jxb.oxfordjournals.org/ by guest on December 7, 2014
Fig. 7. Phylogenic analysis of plant cysteine proteinases. AF238384 (Vigna radiata asparaginyl endopeptidase); Z99956 (Phaseolus vulgaris asparaginyl endopeptidase); D89972 (Vigna mungo asparaginyl endopeptidase); Z34899 (Vicia sativa cysteine proteinase); D89971 (Vigna mungo asparaginyl endopeptidase VmPE-1), Z47793 (Citrus sinensis cysteine proteinase); AB075948 (Nicotiana tabacum vacuolar processing enzyme1b); AB075947 (Nicotiana tabacum vacuolar processing enzyme-1a); AB075950 (Nicotiana tabacum vacuolar processing enzyme-3); AB075949 (Nicotiana tabacum vacuolar processing enzyme-2); AJ309173 (Beta vulgaris vacuolar processing enzyme); AF260827 (Ipomoea batatas SPAE); D61395 (Arabidopsis thaliana gamma vacuolar processing enzyme); D61393 (Arabidopsis thaliana alpha vacuolar processing enzyme); AF082347 (Zea mays C13 endopeptidase NP1); AJ131719 (Zea mays legumain-like protease); AJ131718 (Zea mays legumain-like protease); AB025310 (Oryza sativa asparaginyl endopeptidase); AJ243876 (Lycopersicon esculentum vacuolar processing enzyme); AF521661 (Arabidopsis thaliana delta vacuolar processing enzyme); Z99957 (Phaseolus vulgaris legumain-like proteinase); D31787 (Canavalia ensiformis asparaginyl endopeptidase); D28876 (Glycine max cysteine proteinase); Z99174 (Vicia narbonensis cysteine proteinase); AJ007743 (Vicia sativa cysteine proteinase); D17401 (Ricinus communis vacuolar processing enzyme); AF169973 (Sesamum indicum asparaginyl endopeptidase); D61394 (Arabidopsis thaliana vacuolar processing enzyme); and AB081464 (Oryza sativa asparaginyl endopeptidase REP-2).
10 of 11 Chen et al.
ing of downstream proteases for protein degradation and recycling in senescing tissues (Rojo et al., 2003). Whether a similar mechanism is adopted in sweet potato senescent leaves similar to that utilized in Arabidopsis senescing leaves awaits further investigation. It can be concluded that sweet potato SPAE is a senescence-associated gene that encodes a functional asparaginyl endopeptidase probably associated with the clevage and maturation of proproteinases that are involved in bulk protein degradation and mobilization during leaf senescence. Whether seeds and senescent leaves utilize similar mechanisms for macromolecule such as protein and lipid degradation and mobilization will be an important issue to be addressed. Acknowledgement The authors thank the National Science Council, Taiwan, ROC, for ®nancial support (NSC92-2313-B-034-002). References Abe Y, Shirane K, Yokosawa H, Matsushita H, Mitta M, Kato I, Ishii S. 1993. Asparaginyl endopeptidases of jack bean seeds. Journal of Biological Chemistry 268, 3523±3529. Becker C, Shutov AD, Nong VH, Senyuk VI, Jung R, Horstmann C, Fischer J, Nielsen NC, Muntz K. 1995. Puri®cation, cDNA cloning and characterization of proteinase B, an asparaginyl endopeptidase from germinating vetch (Vicia satava L.) seeds. European Journal of Biochemistry 228, 456± 462. Chen HJ, Hou WC, Jane WN, Lin YH. 2000. Isolation and characterization of an isocitrate lyase gene from senescent leaves of sweet potato (Ipomoea batatas cv. Tainong 57). Journal of Plant Physiology 157, 669±676. Chen HJ, Hou WC, Yang CY, Huang DJ, Liu JS, Lin YH. 2003. Molecular cloning of two metallothionein-like protein genes with differential expression patterns from sweet potato (Ipomoea batatas (L.) Lam.) leaves. Journal of Plant Physiology 160, 547± 555. Chen JM, Rawlings ND, Stevens RAE, Barrett AJ. 1998. Identi®cation of the active site of legumain links it to caspases, clostripain and gingipains in a new clad of cysteine endopeptidases. FEBS Letters 441, 361±365. Chrispeels MJ, Herman EM. 2000. Endoplasmic reticulumderived compartments function in storage and as mediators of vacuolar remodeling via a new type of organelle, precursor protease vesicles. Plant Physiology 123, 1227±1233. Gruis DF, Selinger DA, Curran JM, Jung R. 2002. Redundant proteolytic mechanisms process seed storage proteins in the absence of seed-type members of the vacuolar processing enzyme family of cysteine proteases. The Plant Cell 14, 2863±2882. Hara-Nishimura I, Inoue K, Nishimura M. 1991. A unique vacuolar processing enzyme responsible for conversion of several proprotein precursors into the mature forms. FEBS Letters 294, 89±93. Hara-Nishimura I, Shimada T, Hiraiwa N, Nishimura M. 1995. Vacuolar processing enzyme responsible for maturation of seed proteins. Journal of Plant Physiology 145, 632±640. Hayashi Y, Yamada K, Shimada T, Matsushima R, Nishizawa NK, Nishimura M, Hara-Nishimura I. 2001. A proteinase-
Downloaded from http://jxb.oxfordjournals.org/ by guest on December 7, 2014
graded seed storage proteins (Okamoto and Minamikawa, 1999). During vetch seed germination and seedling growth, the legumain-like proteins, VsPB2 and proteinase B, were found in protein bodies not only in the cotyledons, but also in the radicle, axis, and shoot (Schlereth et al., 2000). VsPB2 has been formed during seed maturation, and proteinase B is formed de novo during seed germination. Both legumain-like VsPB2 and proteinase B together with additional papain-like cysteine proteinases such as CPR1 (accession number X75749), CPR2 (accession number Z30338), and CPR4 (accession number Z99172) are responsible for the bulk degradation and mobilization of globulins, one of storage proteins. A papain-type cysteine proteinase, Y166 (accession number AF242373) was also isolated in the authors' laboratory and exhibited high amino acid sequence homology to Vicia sativa CPR2 (accession number Z30338). These data provide further evidence to support a similar mechanism for bulk protein degradation and mobilization that occurs in both germinating seeds and senescent leaves. Based on these reports, a speculated role is suggested. SPAE may function for the clevage and maturation of other proproteinases (possibly cysteine proteinase propolypeptides). Together with these processed, mature cysteine proteinases it is probably involved in bulk protein degradation and mobilization during leaf senescence similar to Vicia sativa VsPB2 and proteinase B (Schlereth et al., 2000), and Vigna mungo VmPE-1 (Okamoto and Minamikawa, 1999) during seed maturation and germination. Many vacuolar enzymes are synthesized as pro-proteins, which are proteolytically processed to become active. In seed storage tissues, speci®c endoplasmic reticulum (ER)derived compartments containing precursors of cysteine proteinases have been described (Chrispeels and Herman, 2000; Toyooka et al., 2000; Hayashi et al., 2001; Schmid et al., 2001). Germination of the seeds induces the expression and processing of those proteases into the mature active forms, which, in turn, participate in the degradation of cellular materials in storage tissues and provide nutrients to the growing embryo. Recently, similar compartments have also been described in vegetative tissues of Arabidopsis (Hayashi et al., 2001). These precursor protease vesicles derived from ER are plantspeci®c compartments containing precursor protease vesicle-localized vacuolar processing enzyme (gVPE), which is critical for the maturation of the vacuolar protease AtCPY. The vacuolar protease AtCPY in turn participates in the degradation of cellular components including vacuolar invertase AtFruct4 and various proteins in organs undergoing senescence in Arabidopsis (Rojo et al., 2003). A mechanism of senescence-induced activation of precursor protease vesicle-localized vacuolar processing enzyme, possibly by releasing the inactive precursor form from the precursor protease vesicle into the acidic lumen of the vacuole, is suggested. This activation triggers the process-
Sweet potato asparaginyl endopeptidase 11 of 11 laboratory manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Schlereth A, Becker C, Horstmann C, Tiedemann J, Muntz K. 2000. Comparison of globulin mobilization and cysteine proteinases in embryonic axes and cotyledons during germination and seedling growth of vetch (Vicia sativa L.). Journal of Experimental Botany 51, 1423±1433. Schlereth A, Standhardt D, Mock HP, Muntz K. 2001. Stored cysteine proteinases start globulin mobilization in protein bodies of embryonic axes and cotyledons during vetch (Vicia sativa L.) seed germination. Planta 212, 718±727. Schmid M, Simpson DJ, Sarioglu H, Lottspeich F, Gietl C. 2001. The ricinosomes of senescing plant tissue bud from the endoplasmic reticulum. Proceedings of the National Academy of Sciences, USA 98, 5353±5358. Scott MP, Jung R, Muntz K, Nielsen NC. 1992. A protease responsible for post- translational cleavage of a conserved AsnGly linkage in glycinin, the major seed storage protein of soybean. Proceedings of the National Academy of Sciences, USA 89, 658±662. Senyuk V, Rotari V, Becker C, Zakharov A, Horstmann C, Muntz K, Vaintraub I. 1998. Does an asparaginyl-speci®c cysteine endopeptidase trigger phaseolin degradation in cotyledons of kidney bean seedlings? European Journal of Biochemistry 258, 546±558. Shimada T, Hiraiwa N, Nishimura M, Hara-Nishimura I. 1994. Vacuolar processing enzyme of soybean that converts proproteins to the corresponding mature forms. Plant Cell Physiology 35, 713±718. Takeda O, Miura Y, Mitta M, Matsushita H, Kato I, Abe Y, Yokosawa H. 1994. Isolation and analysis of cDNA encoding a precursor of Canavalia ensiformis asparaginyl endopeptidase (legumain). Journal of Biochemistry 116, 541±546. Thompson JD, Higgins DG, Gibson TJ. 1994. Clustal W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position- speci®c gap penalties and weight matrix choice. Nucleic Acids Research 22, 4673±4680. Toyooka K, Okamoto T, Minamikawa T. 2000. Mass transport of proform of a FDEL-tailed cysteine proteinase (SH-EP) to protein storage vacuoles by endoplasmic reticulum-derived vesicle is involved in protein mobilization in germinating seeds. Journal of Cell Biology 148, 453±463. Vierstra RD. 1996. Proteolysis in plants: mechanisms and functions. Plant Molecular Biology 32, 275±302.
Downloaded from http://jxb.oxfordjournals.org/ by guest on December 7, 2014
sorting body that prepares for cell death or stress in the epidermal cells of Arabidopsis. Plant Cell Physiology 42, 894±899. Ishii S. 1994. Legumain: asparaginyl endopeptidase. Methods of Enzymology 244, 604±615. Jonson T, Mahlamaki EH, Karhu R, Gorunova L, Johansson B, Huglund M. 2000. Characterization of genomically ampli®ed segments using PCR: optimizing relative-PCR for reliable and simple gene expression and gene copy analyses. Genes, Chromosomes and Cancer 29, 192±199. Kloos DU, Oltmanns H, Dock C, Stahl D, Hehl R. 2002. Isolation and molecular analysis of six taproot expressed genes from sugar beet. Journal of Experimental Botany 53, 1533±1534. Lee TM, Lin YH. 1995. Trypsin inhibitor and trypsin-like protease activity in air- or submergence-grown rice (Oryza sativa L.) coleoptiles. Plant Science 106, 43±54. Lichtenthaler HK. 1987. Chlorophylls and carotenoids: pigments of photosynthetic biomembranes. Methods of Enzymology 148, 350±382. Makino A, Osmond B. 1991. Effects of nitrogen nutrition on nitrogen partitioning between chloroplasts and mitochondria in pea and wheat. Plant Physiology 96, 355±362. Marty F. 1999. Plant vacuoles. The Plant Cell 11, 587±599. Mitsuhashi W, Koshiba T, Minamikawa T. 1986. Separation and characterization of two endopeptidases from cotyledons of germinating Vigna mungo seeds. Plant Physiology 80, 628±634. MuÈntz K, Shutov AD. 2002. Legumains and their functions in plants. Trends in Plant Science 7, 340±344. Okamoto T, Minamikawa T. 1999. Molecular cloning and characterization of Vigna mungo processing enzyme 1 (VmPE1), an asparaginyl endopeptidase possibly involved in posttranslational processing of a vacuolar cysteine endopeptidase (SH- EP). Plant Molecular Biology 39, 63±73. Okamoto T, Minamikawa T. 1995. Puri®cation of a processing enzyme (VmPE-1) that is involved in post-translational processing of a plant cysteine endopeptidase (SH-EP). European Journal of Biochemistry 231, 300±305. Okamoto T, Yuki A, Mitsuhashi N, Mimamikawa T. 1999. Asparaginyl endopeptidase (VmPE-1) and autocatalytic processing synergistically activate the vacuolar cysteine proteinase (SH-EP). European Journal of Biochemistry 264, 223±232. Rojo E, Zouhar J, Carter C, Kovaleva V, Raikhel NV. 2003. A unique mechanism for protein processing and degradation in Arabidopsis thaliana. Proceedings of the National Academy of Sciences, USA 100, 7389±7394. Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning: a
