Journal of Experimental Botany, Vol. 54, No. 391, pp. 2251-2263,
October 1, 2003
© 2003 Oxford University Press
Differential gene expression for suicide-substrate serine proteinase inhibitors (serpins) in vegetative and grain tissues of barley
Received 21 February 2003; Accepted 14 June 2003
1 Biochemistry and Nutrition, BioCentrum, Building 224, Technical University of Denmark, DK-2800 Lyngby, Denmark
2 Department of Crop Science, Swedish University of Agricultural Sciences, PO Box 44, SE-23053 Alnarp, Sweden
3 Plant Research Department, Building 301, Risø National Laboratory, Frederiksborgvej 399, PO Box 49, DK-4000 Roskilde, Denmark
* Present address and to whom correspondence should be sent: Department of Biological Sciences, Building E8A, Macquarie University NSW 2109, Australia. Fax: +61 2 98508245. E-mail: troberts{at}els.mq.edu.au
Abbreviations: BSZ, barley serpin (protein Z family); dpa, days post-anthesis; rBSZ, recombinant barley serpin (protein Z family); RT-PCR, reverse-transcription polymerase chain reaction.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Proteins of the serpin superfamily (
43 kDa) from mature cereal grains are in vitro suicide-substrate inhibitors of specific mammalian serine proteinases of the chymotrypsin family. However, unlike the ‘standard-mechanism’ serine proteinase inhibitors (<25 kDa), the biological functions of plant serpins are unknown. Expression studies of genes encoding members of three subfamilies of serpins (BSZx, BSZ4 and BSZ7) in developing grain and vegetative tissues of barley (Hordeum vulgare L.) showed that transcripts encoding BSZx, which inhibits distinct proteinases at overlapping reactive centres in vitro, were ubiquitous at low levels, but the protein could not be detected. EST analysis showed that expression of genes for serpins with BSZx-type reactive centres in vegetative tissues is widespread in the plant kingdom, suggesting a common regulatory function. For BSZ4 and BSZ7, expression at the protein level was highest in the maturing grain (
15 d post-anthesis), where these serpins were localized by immunomicroscopy to the central and peripheral starchy endosperm, subaleurone, and (at lower levels) to the aleurone. Serpins were also localized to the meristem and vascular tissues of roots, and to the phloem of coleoptiles and leaves. The identification of BSZ4 in vegetative tissues by western blotting was confirmed for the roots by purification and amino acid sequencing, and for the leaves by in vitro reactive-centre loop cleavage studies. Plant serpins are likely to use their irreversible inhibitory mechanism in the inhibition of exogenous proteinases capable of breaking down seed storage proteins, and in the defence of specific cell types in vegetative tissues. Key words: Barley, Hordeum vulgare, gene expression, plant defence, protease inhibitor, serine proteinase inhibitor, serpin.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Research on the major antigens of beer (Hejgaard and Sørensen, 1975; Hejgaard, 1977; Hejgaard and Kærsgaard, 1983) led to the identification of a group of
43 kDa proteins from mature grains of barley (Hordeum vulgare L.) as members of the serpin superfamily (Hejgaard et al., 1985). Most serpins, including the well-studied blood-plasma proteins antithrombin and 
1-antitrypsin, are regulatory proteins that control the proteolytic activity of specific endogenous serine proteinases of the chymotrypsin family, for example, thrombin and elastase (see Silverman et al., 2001; Gettins, 2002, for reviews). Several serpins inhibit cysteine proteinases (Komiyama et al., 1996; Schick et al., 1998), and a few, such as ovalbumin and angiotensinogen, are non-inhibitory. Inhibitory serpins constitute one of at least 16 families of proteinaceous serine proteinase inhibitors recognized on the basis of sequence similarity, topological similarity, and binding mechanism (Laskowski and Kato, 1980; Koiwa et al., 1997; Bode and Huber, 2000), but the serpins are unique in several important aspects. They are metastable proteins (Carrell and Owen, 1985) with a ‘suicide-substrate’, irreversible mechanism of action (Huber and Carrell, 1989; Huntington et al., 2000) very different from that of the smaller, ‘standard-mechanism’ serine proteinase inhibitors such as those of the Bowman-Birk and Kunitz families. Serpins from the grains of barley (Hejgaard, 1982; Lundgard and Svensson, 1989; Dahl et al., 1996a), wheat (Triticum aestivum L.; Rosenkrands et al., 1994; Dahl et al., 1996a; Østergaard et al., 2000), rye (Secale cereale L.; Hejgaard, 2001), and oat (Avena sativa L.; Hejgaard and Hauge, 2002) are potent, irreversible inhibitors of serine proteinases of the chymotrypsin family in vitro, as is a serpin from the phloem exudate of pumpkin (Cucurbita maxima Duch.; Yoo et al., 2000). In their native, inhibitory form, serpins possess a flexible, solvent-exposed ‘reactive-centre loop’, which acts as a bait that mimics a proteinase substrate (Elliott et al., 1996). The inhibitory specificity of serpins is determined primarily, but by no means fully (Djie et al., 1996), by the identity of the P1 residue of the reactive-centre loop (Olson et al., 1995). The amino acid residues in the reactive-centre loop between which a serpin is cleaved by a cognate proteinase are defined as the P1 and P1' residues, with those further towards the NH2-terminus defined as P2, P3, P4, etc, and those towards the COOH-terminus as P2', P3', P4', etc (Schechter and Berger, 1967). Intriguingly, the reactive-centre sequences (P2 to P1') of most serpins from wheat and rye grain resemble the glutamine-rich repetitive sequences of prolamin storage proteins of the endosperm (Østergaard et al., 2000; Hejgaard, 2001). This resemblance, along with the rarity of glutamine residues in the reactive centres of other serpins, suggests that serpins from wheat and rye grain function in the inhibition of proteinases specifically adapted to the breakdown of grain prolamins. Østergaard et al. (2000) found that wheat serpins were unable to inhibit microbial and endogenous grain proteinases tested, suggesting that the wheat (and rye) serpins may function to protect the prolamins from the digestive proteinases of insects (or untested pathogens).
In barley, resemblance between storage protein repeat sequences and reactive centres of the grain serpins is not apparent, and no clues to the physiological function of barley serpins have been found other than their ability to inhibit mammalian serine proteinases in vitro. There are three recognized barley serpin subfamilies: BSZ4, BSZ7 and BSZx. Gene structures have been described for Paz1, which encodes a member of the BSZ4 subfamily (Brandt et al., 1990), and Pazx, which encodes BSZx (Rasmussen, 1993). A cDNA corresponding to Paz7, which encodes BSZ7, has also been studied (Rasmussen et al., 1996). The basic molecular properties of one member of the BSZ7 subfamily have been investigated (Lundgard and Svensson, 1989), and the inhibitory properties of recombinant BSZ4 (rBSZ4) and rBSZx have been studied in detail (Dahl et al., 1996a, b). The reactive-centre residues of BSZ4 are P1-P1' Met–Ser, while both BSZ7 and BSZx have P1-P1' Arg–Ser. Each of these serpins has a distinct inhibitory specificity, with rBSZx displaying overlapping reactive centres at P1 Arg and P2 Leu.
Research on the expression of barley serpin genes has been limited to the grain (Giese and Hejgaard, 1984; Giese and Hopp, 1984; Mundy et al., 1986; Sørensen et al., 1989; Jakobsen et al., 1991; Rasmussen et al., 1991), and BSZx protein has never been positively identified in barley grain (or other tissue), despite the use of highly reactive, specific antibodies raised against rBSZx. Early work on BSZ4 and BSZ7 (then known collectively as ‘protein Z’) established that these proteins are among the few to be synthesized beyond 15 d post-anthesis (dpa), and that they are among the dominant salt-soluble endosperm proteins at 20 dpa (Hejgaard and Boisen, 1980; Giese and Hopp, 1984). Sørensen et al. (1989) found that mRNA levels for serpins (and hordeins) increased 3–4-fold from 8 dpa to 25 dpa and then decreased; by contrast, mRNA levels for the glycolytic enzyme GAPDH and histone H3 were constant from 8 dpa to 15 dpa and then declined. Interest in barley serpins was supported by an intention to understand the expression of genes encoding lysine-rich grain proteins: the high-lysine alleles lys3a and lys1 on chromosome 7 were found to repress and enhance, respectively, the expression of Paz1 on chromosome 4 (Balasaraswathi et al., 1984). The level of serpin gene expression in barley grain is strongly stimulated by increased levels of nitrogen nutrition (Giese and Hopp, 1984). The abundance of serpins in maturing barley grain was quantified in an ELISA-based study of serpin content (Evans and Hejgaard, 1999), which showed that the mean levels of BSZ4 and BSZ7 (in combined PBS and thiol extractions of malt) among 93 varieties were 944 and 206 µg g–1 grain, respectively. Interestingly, cv. Pirkka, which lacks the gene for BSZ4 (Brandt et al., 1990), was found to have the highest content of BSZ7 (771 µg g–1).
Taken together, the timing of serpin gene expression in grain, the influence of nitrogen nutrition, and the effects of mutations are consistent with a role for barley grain serpins as storage proteins. However, once cleaved in their reactive centre, serpins are extremely resistant to further degradation. This is reflected in the abundance of barley serpins in beer (Hejgaard and Sørensen, 1975), the production of which involves exposure of proteins to proteolytic enzymes produced during germination, as well as to boiling. Clearly, proteins resistant to degradation upon germination are unsuitable as storage proteins. Furthermore, all plant serpins tested are inhibitory in vitro, suggesting that their functions involve the utilization of their inhibitory capability, as in animals. The starchy endosperm of mature grains is dead tissue, with no means to produce inhibitors de novo, suggesting that serpins may be laid down in the latter stages of endosperm development in large amounts as insurance against proteolysis of the grain storage proteins by digestive proteases of pests (Østergaard et al., 2000).
As a step towards elucidating the physiological functions of plant serpins, a detailed expression study of serpin genes in the vegetative and grain tissues of barley has been conducted here.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Plant material
Grains of barley (H. vulgare L.) cv. Alexis and cv. Pirkka were obtained from the Risø National Laboratory, Roskilde, Denmark. Barley (cv. Alexis) tissues used for the isolation of poly(A)+ mRNA were obtained as follows. Plants were grown either in sand layered over soil in trays (for plants harvested <10 d after sowing) or in soil in pots in a glasshouse with a daily regime of 16/8 h light/dark at 18/15 °C). Samples of roots, shoots, coleoptiles, leaves, whole seeds, endosperm, and embryos were taken at the times indicated in the Results. Roots were washed extensively in water. All samples were immediately wrapped in aluminium foil, snap-frozen in liquid N2 and stored frozen.
Identification of Hordeum serpin ESTs
Translated BLAST searches were performed to identify serpin ESTs among all Hordeum EST libraries in GenBank. Full-length amino acid sequences for BSZ4 (accession no. S13822), BSZ7 (CAA64599
[GenBank]
), BSZx (444778, plus a 401-residue altered BSZx sequence based on fusion of EST sequences), and wheat serpin WSZ1a (S65782
[GenBank]
) were used as query sequences. Default searching parameters (BLOSUM62 matrix, expect=10, default gap costs) were used except that the low complexity filter was turned off to allow for sequences similar to the Ala-rich hinge region of the serpin to be identified. Partial sequences specific for BSZ4, BSZ7 and BSZx taken from variable regions in the serpin sequence were also used as query sequences.
The following H. vulgare libraries included serpin ESTs: cv. Morex HVcDNA0009 spike 5–45 dpa, HVcDNA0013 testa/pericarp, HVcDNA0010 spike 20 dpa, HVcDNA0012 spike at 0, 1, 2, 3, 4, 5, 6, and 8 d after Fusarium graminearum inoculation, HVcDNA0015 rachis, HVcDNA0008 pre-anthesis spike (white to yellow anther), HVcDNA0007 seedling root; cv. Optic EBma08 maternal tissue from grains 28 dpa; cv. Barke (no library numbers given) developing caryopsis 3–15 dpa, roots 2 d; cv. Akashinriki (no library number given) leaves; cultivar not given etiolated leaf/culm.
Isolation of poly(A)+ RNA
Frozen tissue samples (typically 1 g) were ground under liquid N2 using a mortar and pestle. Poly(A)+ RNA was isolated immediately from 0.1 g ground tissue using the Dynabeads mRNA DIRECT kit (Dynal, Oslo) and a Pellet Pestle Mixer (Kontes Biotechnology, Vineland, NJ, USA) in a 1.5 ml Eppendorf tube. The isolation was performed according to the Dynal instructions except that homogenization was performed with 0.5 ml lysis/binding buffer prior to the addition of a further 0.5 ml buffer. The same poly(A)+ RNA samples were used for the northern dot-blot and RT-PCR experiments.
Northern dot-blotting
Northern dot-blotting was performed according to Brown and Mackey (1997). In brief, aliquots (10 µl, 0.3 µg) of poly(A)+ RNA were heated with 30 µl denaturing solution (1.5 M NaCl, 0.5 M NaOH; used for both RNA and DNA) for 15 min at 65 °C, plunged into ice and applied to a 12x8 cm Hybond N uncharged nylon membrane (Amersham Biosciences, Uppsala). Plasmid DNA samples corresponding to each of the three serpin genes, Paz1, Paz7 and Pazx, were used on the same blot as controls to check for hybridization specificity. The concentration of the undiluted plasmid samples was checked using a GeneQuant DNA/RNA quantifier (Amersham Biosciences) and a Hoeffer TKO fluorometer (Hoeffer, San Fransisco) using Hoechst 33258 dye. Values (averages obtained with the two methods) were 122, 381 and 133 µg ml–1 DNA for plasmids encoding BSZ4, BSZ7 and BSZx, respectively. The undiluted plasmid DNA samples plus four 10-fold serial dilutions (10 µl) were heated at 100 °C with 30 µl denaturing solution for 10 min, plunged into ice and applied to the membrane. The blot was exposed to UV light for 5 min to cross-link the poly(A)+ RNA samples to the membrane. The radioactive probes used for Paz1, Paz7 and Pazx hybridization were prepared by generating PCR products from the plasmid DNA described above using the PCR primers employed for RT-PCR experiments. These PCR products were purified using a GeneClean Spin Kit (BIO-101) and labelled using a Ready.To.Go kit (Amersham Biosciences). Unincorporated radioactivity was separated using a syringe packed with Sephadex G-25 (Amersham Biosciences). Washing at moderate stringency (twice in prewarmed 0.2x SSC/0.1% SDS for 15 min each at 42 °C with rotation) or high-stringency (the same except at 68 °C) was performed. The integrity of the poly(A)+ RNA samples on the blot was tested by hybridization with a radiolabelled probe specific for the gene encoding the glycolytic enzyme triose phosphate isomerase (TPI, accession no. U83414
[GenBank]
). Amersham Hyperfilm MP was used for the autoradiograms. The blot was stripped at 100 °C following each hybridization and the absence of signal confirmed by autoradiography.
Reverse-transcription PCR
Differences in the C-terminal regions of the three barley serpins allowed the design (using GeneWorks software, Oxford Molecular) of specific primer pairs for the Paz1, Paz7 and Pazx genes, with the downstream primer of each pair derived from the 3'-non-coding region. The primer pairs (T-A-G-Copenhagen, Copenhagen) used (upstream and downstream, respectively) were 5'-CCTCTTCCGACAACTTGAAGG-3' and 5'-GGACAGCAGATCATAGCAAGC-3' for Paz1, 5'-CTTCCATACCAACATGGTGG-3' and 5'-GGTCTACAATCACATCGCCG-3' for Paz7, and 5'-CGGAGCTTGTATGTCTCATCC-3' and 5'-CACATGTGACAGTTGAGGCC-3' for Pazx. The predicted lengths of the amplified sequences were 560, 376, and 282 bp for primers specific for Paz1, Paz7 and Pazx, respectively. The predicted melting temperatures for the six primers were within the range 59–61 °C and their %G+C was 50–55%. The optimal PCR annealing temperature (55 °C) for the Pazx primers was determined using an Eppendorf MasterCycler Gradient thermocycler, and this temperature was used for the other primer pairs. The MgCl2 concentration for all PCR reactions was 2.0 mM, and the primer concentration was 10 pmol per 100 µl reaction (0.1 µM). Control PCR experiments were performed to demonstrate the specificity of the primer pairs using Paz1, Paz7 and Pazx plasmid DNA and the same final concentrations of components as used for the PCR step of the RT-PCR experiments. These controls showed that the primers pairs were specific in amplifying only their corresponding DNA sequences providing the template concentrations were similar to the concentrations of the poly(A)+ RNA used in RT-PCR.
The GeneAmp RNA PCR Kit (Perkin Elmer) was used for RT-PCR according to the manufacturer’s recommended protocol, with the reverse transcriptase and DNA polymerase used consecutively in a single-tube reaction, and a combined annealing/extension step employed for the PCR. Reverse transcription was performed using oligo d(T)16 primers in 20 µl for 15 min at 42 °C, and stopped by heating at 5 min at 99 °C. As template, 1 µl (0.03 µg) poly(A)+ RNA was used in each reaction, and each tissue sample was tested at least twice in independent RT-PCR reactions. An additional set of primers for Pazx, which in this case spanned the (single) Pazx intron, was designed to test for the presence of genomic DNA in the poly(A)+ RNA preparations. These primers were 5'-AACTCAAGCCTTCCTTCAAGG-3' and 5'-GGATGAGACATACAAGCTCCG-3', with the predicted lengths of the amplified sequences being 1625 bp including the intron and 656 bp excluding the intron.
All RT-PCR runs were conducted using a Techne Genius thermal cycler (Techne, Cambridge) with the following PCR program: 1 cycle of 1 min 45 s at 95 °C, 35 cycles of 15 s at 95 °C and 30 s at 55 °C, and 1 cycle of 7 min at 72 °C. Products were run on 1.4% agarose gels in TAE buffer containing ethidium bromide and visualized using UV light. PCR products were sequenced using an Applied Biosystems 373A DNA sequencer.
Polyclonal and monoclonal antibodies
Antibody specificities refer to Western blots. Polyclonal (rabbit) antibody R360, raised against rBSZx, reacts very strongly with rBSZx, moderately with BSZ4 and more weakly with BSZ7 (Østergaard et al., 2000). Monoclonal (mouse) antibodies: 5C11, raised against purified BSZ7, reacts with BSZ4 and BSZ7; 8E8, raised against purified BSZ4, reacts strongly with rBSZx and more weakly with BSZ4 (Dahl et al., 1996b); 11C7, raised against purified BSZ7b, reacts only with this molecular form of BSZ7 (Evans and Hejgaard, 1999).
Purification of serpins from barley vegetative tissues
Roots (20 g) were harvested from seedlings (7 d) of Alexis barley and homogenized at 4 °C with a mortar and pestle in 40 ml 25 mM TRIS-HCl, pH 8.0, 1 mM EDTA (Buffer A) containing one-half of a Complete Inhibitor Tablet (Roche). The mixture was centrifuged at 4 °C to remove debris and the supernatant applied to a DEAE Sephadex A-50 column (2.2x15 cm) previously equilibrated with Buffer A at room temperature. The column was washed with Buffer A until the A280 returned to the baseline level and then a linear gradient from 0–1.0 M KCl in a total of 60 ml was applied. A flow rate of 1 ml min–1 was used. Fractions containing serpins were identified by dotting 5 µl aliquots directly onto nitrocellulose membranes and testing with R360, 5C11, 8E8, and nonsense antibodies. The serpin-containing fractions were pooled, the salt removed by gel filtration through a PD-10 column (Amersham Biosciences), and the sample applied to a Mono-Q HR 5/5 column (Amersham Biosciences) previously equilibrated with Buffer A. The column was washed with 5 ml Buffer A and the serpins eluted with a gradient of 0–0.5 M KCl in a total volume of 40 ml Buffer A. The serpin-containing fractions were identified using dot-blotting as above. Aliquots of the selected fractions were run on 8% native PAGE (Østergaard et al., 2000), transferred onto PVDF membranes, and western blotting and Coomassie Blue staining (prior to band excision for sequencing) performed.
SDS-PAGE, immunoelectrophoresis, western blotting and amino acid sequencing
A Novex Xcell II Mini Cell electrophoresis apparatus and Novex precast 10%-acrylamide TRIS-glycine gels were used. SDS-PAGE was performed as described previously (Østergaard et al., 2000), as was silver staining (Dahl et al., 1996a) and immunoelectrophoresis (Hejgaard and Sørensen, 1975). Western blotting to nitrocellulose or PVDF membrane was performed with a Semi-Dry Blotter II (KemEnTec) using a three-buffer system (Kyhse-Andersen, 1984). Amino acid sequencing was performed with a Procise 494 sequencer from Applied Biosystems.
Immunohistochemistry
All antibodies were purified using protein A affinity chromatography (Amersham Biosciences). Paraffin sections were prepared from developing barley grains (30 dpa) of cv. Alexis and cv. Pirkka as in Marttila et al. (1996). Briefly, grains were cut in halves either vertically or horizontally, and fixed in 4% paraformaldehyde in PBS (10 mM Na-phosphate, pH 7.4, 150 mM NaCl) with gentle shaking overnight at 4 °C. Fixative was changed after 3–4 h. After fixation, seeds were washed with PBS, dehydrated and infiltrated with paraffin within 3 d. Sections (10 µm) were collected on Superfrost Plus slides (Menzel-Gläser, Braunschweig, Germany).
For immunolocalization, sections were first deparaffinized and blocked with 5% (w/v) goat normal serum and 1% BSA (w/v) in PBS for 30 min. The sections were incubated with the primary antibody either for 2 h at 37 °C (polyclonal R360-IgG diluted 1:10 000 with 1% BSA in PBS) or overnight at 4 °C (monoclonals 5C11, 8E8 and 11C7 diluted 1:50, 1:1000 and 1:1000, respectively). After washing, the slides were treated with the goat anti-rabbit or anti-mouse secondary IgG coupled with gold (diameter 1 nm; British Biocell International, Cardiff, UK) for 60 min at 37 °C, washed with PBS and H2O, and enhanced with silver (British Biocell). The slides were mounted with water-based 20% Mowiol 4-88 (Calbiochem-Novabiochem Corporation, La Jolla, CA). Positive labelling was seen as a dark precipitate in bright-field microscopy. Sections were studied under a Leica DM LB microscope (Leica Microsystems, Wetzlar, Germany), digital imaging was done with a Leica DC200 camera, and image handling was performed in Adobe Photoshop 6.0.
Controls used for immunolocalization were substitution of the primary antibody with: (i) the IgG fraction of rabbit normal serum (polyclonals); (ii) highly diluted antibody (monoclonals); (iii) dilution buffer; (iv) the primary antibody blocked with its antigen.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Serpin transcripts in developing grain
Searching databases at NCBI (http://www.ncbi.nlm.nih. gov) among H. vulgare ESTs revealed that those encoding BSZ4 and BSZ7 were numerous in libraries from developing and maturing spikes and testa/pericarp, but notably absent in libraries from pre-anthesis spikes. By contrast, (only) two BSZx ESTs were identified in each of the collections from pre-anthesis, developing and maturing spikes (grains). In northern dot-blot experiments, radioactive probes complementary to RNA/DNA encoding serpins BSZ4, BSZ7, BSZx, and the ubiquitous glycolytic enzyme triose phosphate isomerase (TPI, positive control) were used to detect serpin gene transcripts in tissues of cv. Alexis (Fig. 1). The TPI probe bound to the poly(A)+ RNA samples from all tissues tested, confirming their integrity, and (weakly) to only the highest concentrations of the serpin plasmid DNA controls on the blot (Fig. 1A); the signal variation obtained between samples with the TPI probe was expected. The specificity of each of the serpin probes was demonstrated by hybridization to plasmid DNA controls (Fig. 1B–D). The BSZ4-probe hybridized to poly(A)+ RNA from endosperm isolated 15, 20 and 30 dpa and from 45 dpa whole seeds; no substantial hybridization to any other samples was revealed by an overnight exposure (Fig. 1B). The BSZ7 probe hybridized weakly to transcript from 15 dpa endosperm, strongly to the 20 dpa and 30 dpa endosperm and 45 dpa whole seed poly(A)+ RNA, and not substantially to the other samples (Fig. 1C). The probe for BSZx did not hybridize to any of the poly(A)+ RNA samples as detected with an overnight exposure (Fig. 1D); a replicate hybridization using a freshly prepared BSZx probe produced the same result. When this blot was left to expose film for 11 d, weak hybridization could be detected to most of the samples (not shown), but since the sensitivity of the probe for its corresponding plasmid DNA was not more than 100-fold higher than for the other controls (Fig. 1D), the 11 d signals may have been non-specific.
|
RT-PCR allowed a more sensitive analysis of serpin transcripts. Identical amplification conditions were used with each of the primer pairs and all of the poly(A)+ RNA samples; however, direct comparisons cannot be made between intensities obtained with different sets of primers. Comparisons were made of the RT-PCR product-band intensity (trace, low, moderate, and high) obtained for different tissues for each set of primers (Table 1). RT-PCR products with primers specific for genes encoding BSZ4, BSZ7 and BSZx were of the predicted size (560, 376 and 282 bp, respectively). The absence of amplicons for the negative controls lacking reverse transcriptase confirmed that the cDNA amplified was not contaminated by genomic DNA. For BSZ4 RT-PCR, moderate- to high-intensity product bands of the predicted size were found for the embryo samples and whole seeds. High-intensity bands were found for the endosperm samples, in accordance with the northern dot-blot results. For BSZ7 RT-PCR, high-intensity bands were found for 20 dpa and 30 dpa endosperm and 45 dpa whole seeds, again corresponding to the northern dot-blot. Bands of moderate intensity were found for 5 d whole seeds, 15 dpa endosperm, and 20 dpa and 30 dpa embryos. Interestingly, no products were obtained for both 15 d embryos, indicating different timing of expression for Paz7 compared to Pazx and Paz1, and 10 d whole seeds (RT-PCR performed in triplicate), indicating biphasic expression Paz7 in the grain. In Fig. 1A, the TPI dot intensity for 10 d whole seeds is lower than that for 5 d whole seeds and 15 d endosperm, but this difference is not sufficient to explain the complete lack of RT-PCR product obtained using 10 d poly(A)+ RNA as template (see also whole-seed results for BSZx and BSZ7).
|
BSZx RT-PCR gave product bands of the predicted size (282 bp) with moderate to high intensity for all grain tissues tested (Table 1). However, since the northern dot-blots showed that the level of Pazx expression was much lower than the highest level of expression of Paz1 and Paz7, differences in amplification kinetics between the different primer sets must account for the relatively high intensity of the RT-PCR bands obtained with the Pazx primers. The identity of the BSZx RT-PCR products was confirmed by sequencing. Confirmation of Pazx gene expression at the mRNA level was important with respect to the apparent absence of BSZx protein in barley tissues, as discussed below. A PCR experiment using Pazx plasmid DNA (containing the intron) and a set of intron-spanning Pazx primers resulted in the expected 1625 bp (intron-containing) product alone. Using these primers, RT-PCR with poly(A)+ RNA from 45 dpa seeds resulted exclusively in the expected 656 bp product, indicating no apparent contamination of the poly(A)+ RNA with genomic DNA. This PCR product was sequenced and positively identified as a BSZx cDNA.
Serpin abundance in the starchy endosperm and subaleurone
The tissue distribution of serpins within maturing grains (including embryonic tissues) was studied by immunomicroscopy using purified poly- and monoclonal antibodies combined with silver-enhanced gold labelling. Antibody specificities determined from Western blotting are given in the Materials and methods. An abundance of BSZ4-related serpins in the starchy endosperm (both peripheral and central) and the subaleurone layer of cv. Alexis is seen as a dark brown precipitate (Fig. 2A, C, E) absent in control sections for which the primary monoclonal antibody 8E8 was omitted (Fig. 2B, D, F). A similar labelling pattern to that obtained with 8E8 was given by monoclonal 5C11 and polyclonal R360 antibodies (not shown), which suggested similar tissue distributions for BSZ4 and BSZ7. The presence of BSZx in these tissues cannot be excluded, but many tests for the presence of BSZx involving purification and sequencing of barley serpins were negative. The abundance of serpins in the protein matrix of the cells of the starchy endosperm is contrasted with the lack of label in the cell walls in this tissue (Fig. 2C, D). In repeated experiments the aleurone layer (two to three cells thick) was always weakly (but clearly) labelled by the serpin antibodies. The aleurone label appeared to be outside the nucleus, and the cell walls were not labelled. While the subaleurone cells were intensively labelled, no label was seen in the outer layers of the grain, i.e. in the testa, pericarp and husk (Fig. 2E, F). Weak but clear labelling was observed in the scutellum and epithelial cells of the maturing grain (Fig. 2H, I). Phenolic compounds in the testa cause a characteristic dark brown colour (Duffus and Cochrane, 1993) as seen in the controls.
|
To examine expression of BSZ7 more closely, immunomicroscopy was performed on grain sections of Pirkka (which lacks BSZ4) using 11C7, a monoclonal antibody that reacts with only one specific form of BSZ7 not found in Alexis. This antibody gave intense labelling in a single subaleurone layer immediately adjacent to the aleurone, while only weak label in the aleurone and starchy endosperm was detected (Fig. 2G). Labelling of the subaleurone was also apparent with 8E8 in the Alexis sections, but with less contrast due to a greater intensity of labelling obtained. These results show that distribution of different serpin isoforms may vary in grain tissues.
Serpin transcripts in vegetative tissues
H. vulgare ESTs encoding BSZ4 and BSZ7 were notably absent in libraries from shoots (early stage of development), leaves and roots. By contrast, 11 BSZx ESTs were found in leaf and root libraries. Northern dot-blots showed that serpin transcripts (generally) were at undetectably low levels in vegetative tissues relative to those for BSZ4 and BSZ7 in the endosperm of Alexis (Fig. 1). RT-PCR showed that poly(A)+ RNA encoding BSZ4 was present in all tissues, with moderate- to high-intensity bands (560 bp) obtained for roots and leaves during early and mid-development (Table 1). In contrast, no poly(A)+ RNA encoding BSZ7 was detected by RT-PCR for any of the leaf samples, and only minute traces of transcript (376 bp) were detected in coleoptiles, 7 d shoots, and embryo-derived callus. Low-intensity bands were found for each of the root samples (Table 1). BSZx RT-PCR gave product bands (282 bp) with moderate to high intensity for all vegetative tissues tested (Table 1). In conjunction with the northern dot-blots, the RT-PCR results indicate ubiquitous, but low, levels of BSZx transcript were present in the vegetative tissues.
The major serpin in vegetative tissues identified as BSZ4
Preliminary rocket immunoelectrophoresis experiments (not shown) performed with two polyclonal serpin antibodies (including R360) resulted in no formation of rockets with leaf and root crude extracts. Positive control experiments gave clear rockets with grain extracts and with vegetative tissue extracts to which rBSZx had been added. No evidence was found for the presence of BSZx in the grain or for any serpin in vegetative tissues. However, more-sensitive western blot experiments with polyclonal antibody R360 identified serpins in salt extracts of vegetative tissues (Fig. 3). Two molecular forms were identified in extracts of young leaves and coleoptiles (43 kDa and 39 kDa, respectively), whereas only the 39 kDa form was detected in root extracts (Fig. 3A). Very similar results were obtained with extracts of leaves and roots from older plants (not shown). The observed difference in size indicated that the upper band represents intact serpin and the lower band serpin cleaved in the reactive-centre loop (Østergaard et al., 2000).
|
In contrast to BSZ7, the major serpin of Alexis grain, BSZ4, is sensitive to cleavage with pancreas elastase in the reactive-centre loop (J Hejgaard, unpublished data). Identical incubations of grain and young leaf extracts with elastase resulted in conversion of the larger molecular form to the smaller form (Fig. 3B). Incubation of the leaf extract without proteinase for 60 min at 30 °C did not change the pattern of a minor 43 kDa band and a major 39 kDa band (Fig. 3C, lane 1; compare with Fig. 3A, lane 1). By contrast, the addition of root extract (3 µl) to the leaf extract (15 µl) resulted in the disappearance of the 43 kDa band after incubation at 30 °C, but not at 0 °C (Fig. 3C, lanes 2 and 3). The root extract appeared to contain additional proteinase activity capable of cleaving the intact leaf serpin when added to the extract. Thus, in spite of the addition of a Complete proteinase inhibitor tablet during extraction, a major part of the leaf and coleoptile serpin and all the root serpin was in a form cleaved in the reactive-centre loop after extraction (Fig. 3a).
The observations above were supported by column chromatography-based purification to identify the major serpins in roots. Crude extraction followed by DEAE-Sephadex- and MonoQ chromatography, using western blotting to detect serpins in column fractions, resulted in a substantial partial purification (not shown). This root preparation was run on Tricine SDS-PAGE and blotted to PVDF membrane for band excision and sequencing. Both a 39 kDa and a 4 kDa band on the blot yielded N-terminal sequence, the latter band corresponding to the peptide released from the purified serpin after boiling with SDS. The 39 kDa band gave the sequence ATDVRLSIAH, identical to residues 5–14 from the N-terminus of BSZ4 (Brandt et al., 1990). The 4 kDa band gave the sequence KVDLVD, corresponding to BSZ4 residues P5' to P10', and confirming that the band resulted from cleavage of the serpin reactive-centre loop at P4' Leu by an endogenous proteinase(s) present in the root extract. Thus the root serpin purified belongs to the BSZ4 subfamily. Serpin antigen corresponding to BSZ7 or BSZx could not be detected with poly- or monoclonal antibodies during the purification. Together, these results strongly suggest that BSZ4 is the major serpin in barley leaf and root. Attempts to purify leaf serpins using column chromatography were unsuccessful.
Many of the well-studied serpins in animals are exported from the cell and circulate in the blood. In plants, many defence-related proteins are extracellular and their genes induced in vegetative tissues by stress. In the supernatant of barley cell cultures, exported proteins such as chitinases and pathogenesis-related proteins can be detected on western blots of proteins separated by SDS-PAGE (Kragh et al., 1991). Poly- and monoclonal serpin antibodies failed to recognize serpins on blots in identical experiments (data not shown). Similarly, vacuum infiltration of unstressed leaves, as well as leaves stressed with nickel according to Hejgaard et al. (1992), followed by SDS-PAGE and western blotting, failed to detect serpins in the extracellular fluid.
Serpin localization to tissues of young roots and shoots
The tissue distribution of serpins within roots of barley was studied by immunomicroscopy using monoclonal antibody 5C11 and polyclonal antibody R360 (Fig. 4). Monoclonal antibody 8E8 gave similar results, but with less-intense labelling. Longitudinal sections from the root tip and an early differentiated zone of a 10 d root were labelled with 5C11 in the apical meristem, cortex and vascular cylinder (Fig. 4A, B). Transverse sections labelled with R360 showed that serpins were abundant in some phloem cells of the root, while the parenchymatous tissues, i.e. the cortex outside the vascular cylinder and pith in the middle of the cylinder, were less strongly labelled (Fig. 4C, D). The undifferentiated cells in the maturing vascular cylinder near the root tip were not labelled, but serpins were clearly present in the apical meristem and in the newly-formed root-cap cells (Fig. 4A, B). Longitudinal sections of the embryonic root (Fig. 4E, F) showed that the structures that protect the rootlet, i.e. the root-cap and coleorhiza, were strongly labelled. Labelling was also obtained for the apical meristem of the embryonic root (not shown), which reflected the localization of serpins to the apical meristem of the young root (Fig. 4A).
|
Serpin was clearly present in the embryonic leaf and surrounding coleoptile, whereas the undifferentiated vascular tissue in the embryonic leaf was not labelled (not shown). Vascular bundles of 10 d coleoptile and leaf gave very specific labelling with monoclonals 8E8 and 5C11, respectively, in only a few cells in the phloem (Fig. 5A–D). Weak label seen in the thick walls of some xylem cells was probably unspecific. Bundle sheath, mesophyll and epidermal cells were not labelled. Monocotyledonous phloem tissue contains sieve-tube elements and companion cells; however, the identity of the labelled phloem cells remains unclear.
|
Widespread occurrence of serpins with a BSZx-type reactive centre
All of the 11 serpin ESTs found in libraries from barley vegetative tissues (five from roots, three from leaves, and three from shoots including two in the apex) coded for BSZx. To determine whether genes for serpins with BSZx-type reactive centres are expressed in vegetative tissues of other plants, a search was made among viridiplantae cDNA libraries originating solely from non-seed tissues such as leaf, root, rhizome, flower, and fibre (cotton) for EST sequences containing the highly conserved serpin motifs P19 to P8 and P10' to P33'. These motifs represent both sides of the variable reactive-centre loop, with numbering based on the length of this loop in BSZx (Rasmussen, 1993). Expressed serpins were identified in monocots (Oryza sativa, Sorghum propinquum and Zea mays), eudicots (Arabidopsis thaliana, Glycine max, Gossypium arboreum, Lycopersicon esculentum, Mesembryanthemum crystallinum, and Medicago truncatula), a gymnosperm (Pinus taeda), and a moss (Physcomitrella patens). All these species contained one or more cDNAs for a serpin with the BSZx reactive centre P2 to P1' Leu–Arg–Ser, or very similar variants thereof (Leu–Lys–Ser or Leu–Arg–Gly). For example, several ESTs from P. patens were found to encode a serpin with a P3–P3' sequence identical to that of BSZx. In only three of the species (sorghum, soybean and tomato) were serpin ESTs for proteins with non-BSZx-like reactive centres (suggesting the presence of a serpin with a different inhibitory specificity) identified in vegetative tissue libraries. Thus a serpin with a reactive centre similar to that of BSZx appears to be widespread in the plant kingdom.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Serpins in animals regulate proteolysis in blood clotting, fibrinolysis, protein folding, complement activation, cell differentiation, and turn-over of extracellular matrix. They also function in hormone transport, tumour suppression and chromatin folding (Silverman et al., 2001). Even serpins that cluster at the same chromosome position and have the same gene structures can have distinct functions (Rollini et al., 1997). Expressed serpin genes are found in photosynthetic eukaryotes as various as green algae, mosses and cereals (EMBL/GenBank), but whether plant serpins are functionally diverse or share few common functions remains unknown. No target proteinases have been identified for plant serpins, and no concrete predictions of physiological function have resulted from experimental characterization of reactive centres and inhibitory specificity. A putative role in defence for the abundant inhibitory serpins in the mature grains of wheat and rye constitutes the most convincing prediction thus far (Østergaard et al., 2000; Hejgaard, 2001).
Here it has been shown that transcripts for barley serpins BSZ4 and BSZ7 in the maturing endosperm (from 15 dpa) dominate serpin gene expression at the mRNA level (Fig. 1; Table 1), in general agreement with results of Sørensen et al. (1989). The RT-PCR experiments showed that levels of mRNA for BSZ7 (in contrast to those for BSZ4 and BSZx) were much higher at 5 dpa and 15 dpa than at 10 dpa, indicating biphasic (and differential) expression. A similar biphasic expression has been reported for genes encoding thaumatin-like proteins (defence-related proteins (PR-5) that inhibit fungal growth) in barley and oat grain (Skadsen et al., 2000). At the protein level in maturing grains, immunomicroscopy revealed serpins throughout the starchy endosperm, at relatively high levels in the subaleurone, and also in the aleurone itself (Fig. 2). This is consistent with the in vitro synthesis of BSZ4 from mRNA isolated from either endosperm or developing aleurone cells (20 dpa and 30 dpa) detected by Mundy et al. (1986). Through RT-PCR it has been demonstrated that serpin mRNA is present in embryos, and through immunomicroscopy that serpins are present in the embryonic scutellum, shoot and root, together providing further evidence that serpin gene expression is not limited to those grain tissues where storage proteins are found.
Serpins in grain tissues may function in defence, but those in aleurone (living cells in mature grain) might also be involved in protecting cells from damage from endogenous proteinases, as has been suggested for some intracellular mammalian serpins (Scott et al., 1999).
Plant serpins are unlikely to act as in vivo inhibitors of digestive proteinases in animals with acidic stomachs (such as mammals) because serpins are inactivated by low pH. However, seed serpins may protect the storage proteins of whole seeds that pass through to the lower gut where chymotrypsin and trypsin are found, thus promoting seed dispersal; this concept might also apply to seed-eating birds. Many insects are known to have digestive serine proteinases of the chymotrypsin family and to have neutral or alkaline gut pH, thus providing an environment in which exogenous inhibitory serpins could act. In the context of digestion, it may be important that serpins form kinetically stable, covalent complexes with cognate proteinases, while reversible standard-mechanism inhibitors form only non-covalent complexes in competition with storage protein substrate molecules present at high concentrations. The long lifetime (hours or days) of serpin–proteinase inhibitory complexes (Huntington et al., 2000) would continue to keep exogenous proteinases inactive under conditions in which the presence of competing substrate proteins would lower the efficacy of standard-mechanism inhibitors (with the same inhibitory specificity as the serpins).
Despite the very low level of serpin gene expression in the vegetative tissues relative to the grain, serpin transcript can be detected using RT-PCR and serpin protein by western blotting (Fig. 3) and immunomicroscopy (Figs 3, 5). In leaves, the presence of transcripts for BSZ4 and BSZx and the absence of transcript for BSZ7 (Table 1) suggest the possibility of organ-specific functions for specific plant serpins. Elastase-specific cleavage of a leaf serpin identified it as BSZ4 (Fig. 3). In roots, transcripts for all three serpins can be detected by RT-PCR, and western blotting using a serpin-specific polyclonal antibody identified serpins (Fig. 3) which were also found in leaves and coleoptiles. Purification and partial sequencing of a root serpin identified BSZ4; thus, the Paz1 gene appears to be widely expressed at the protein level.
Many animal serpins are extracellular proteins (a characteristic of many plant defence proteins) exported from the cell through a mechanism involving cleavable N-terminal signal peptides. No cleavable signal peptides have been identified in plant serpin sequences, and no extracellular serpins have been found in the present study or others focusing on identifying extracellular proteins from leaves or in cell culture supernatants of barley and other plants. However, here (reminiscent of blood serpins) serpins of barley coleoptiles and leaves have been localized to the phloem (Fig. 5), consistent not only with the expression of standard-mechanism inhibitors in the stem phloem of other plants (Habu et al., 1996), but with a study by Yoo et al. (2000) in which a serpin from pumpkin was identified in the phloem exudate and characterized. Meristem and vascular tissues may require the presence of specific defence proteins, reflecting the importance of these cell types in plant growth and development. De Leo et al. (2001) found that expression of the trypsin inhibitor-2 gene (mti-2) from mustard (Sinapis alba) in Arabidopsis thaliana (following treatment with jasmonic acid or wounding) was limited to meristem and vascular tissues. Yoo et al. (2000) found that in vivo feeding assays with Myzus persicae, a piercing-sucking aphid, established a close correlation between the developmentally regulated increase of the serpin cloned from pumpkin phloem sap and the reduced ability of the aphids to survive and reproduce on the pumpkin plants. However, a direct effect of the serpin on aphid survival in feeding experiments was not found. A defence hypothesis for vegetative tissue serpins is supported by work on insect herbivory showing the effect of over-expressing a serpin gene (cloned from an insect, Manduca sexta) in diverse crop species including tobacco (Thomas et al., 1995).
Addressing the expression of the gene for BSZx was critical, not only because of the potent inhibitory activity of BSZx (Dahl et al., 1996b), but because this serpin could not be detected in grain or vegetative extracts on western blots, or during serpin purification. Database searching in grain/endosperm libraries revealed BSZx ESTs, although only very few and only in those libraries containing material from early stages of grain development. Interestingly, ESTs encoding serpins with BSZx-type reactive centres were widespread in the plant kingdom in vegetative tissues, and it is speculated that this serpin has a conserved regulatory or ‘housekeeping’ function. It seems unlikely that mRNA for BSZx is never translated in vivo; thus, BSZx might be converted into a functional form that is not detected by antibodies raised against the recombinant full-length protein. Of relevance to this possibility is the finding that the mammalian serpin leukocyte elastase inhibitor is converted to the apoptosis-related L-DNAase II through proteolytic cleavage (Torriglia et al., 1998).
Serpin-mediated regulation of endogenous proteolytic events remains a possibility. To demonstrate that this type of interaction occurs in plant cells demands that one or more cellular target proteinases be identified. This has proved problematic, not the least because there is no confirmed evidence for the existence of serine proteinases of the chymotrypsin family or their genes in the plant kingdom. Searches for these genes included BLASTING the fully-sequenced Arabidopsis thaliana genome, in which around eight full-length serpin genes and several other serpin-related sequences are found. Experiments whereby active rBSZx was bound to a nickel-nitrilotriacetic (Ni-NTA) matrix and extracts from barley tissues eluted through the column have failed to identify cognate enzymes.
The induction of genes encoding standard-mechanism proteinase inhibitors in the vegetative tissues of eudicot plants by insect attack and other wounding stimuli such as the application of jasmonic acid has been well characterized (Ryan, 1990). Systemic induction of proteinase inhibitors has also been shown for cereal tissues such as germinating maize embryos in response to fungal infection (Cordero et al., 1994). Proteinase inhibitor accumulation in barley leaves in response to aphid attack has been studied (Casaretto and Corcuera, 1998), but the (weakly) induced chymotrypsin inhibitors did not include serpins. As far as is known, none of the large number of studies on induced gene expression in plants includes reports of the accumulation of serpin transcripts or protein, but serpins may provide an innate barrier defence, especially in dead tissues such as mature starchy endosperm.
In conclusion, it has been shown that barley serpin genes are expressed not only in specific tissues within the developing grain, but also in the meristem and phloem of vegetative tissues. Results suggest that expression of Paz1 and Paz7 is driven by distinct promoters. The low-level but constitutive expression of Pazx at the mRNA level is clearly distinct from that of Paz1 and Paz7, possibly reflecting a regulatory function for BSZx. Further attempts will be made to localize this serpin to specific tissues. While plant serpins appear to be intracellular and their levels unaffected by stress, their localization within cells has yet to be established and more work on possible induction mechanisms is required. It is likely that the unique ability of serpins to inactivate proteinases irreversibly has ensured their evolutionary success in the plant kingdom alongside the standard-mechanism protease inhibitors, for which a role in defence has substantial experimental support (Ryan, 1990). The results presented remain consistent with a defence hypothesis for plant serpins, and further work is being conducted using a variety of experimental approaches to elucidate the physiological functions of these fascinating proteins.
|
Acknowledgements |
|---|
This study was supported by Grant 9601066 from the Danish Agricultural and Veterinary Research Council. We thank Tran Duc Tuan Tung, Susanne Blume, Bente Isbye, and Mette Madsen for excellent technical assistance.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Balasaraswathi R, Køie B, Doll H. 1984. The concentration and yield of hordein and some lysine-rich proteins as influenced by the lys gene of Hiproly barley. Hereditas 100, 225–231.
Bode W, Huber R. 2000. Structural basis of the endoproteinase-protein inhibitor interaction. Biochimica et Biophysica Acta 1477, 241–252.[CrossRef][Medline]
Brandt A, Svendsen I, Hejgaard J. 1990. A plant serpin gene. Structure, organization and expression of the gene encoding barley protein Z4. European Journal of Biochemistry 194, 499–505.[Web of Science][Medline]
Brown T, Mackey K. 1997. Analysis of RNA by northern and slot blot hybridization. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K, eds, Current protocols in molecular biology, Vol. I. John Wiley & Sons, 4.9.1–4.9.16.
Carrell RW, Owen MC. 1985. Plakalbumin, alpha 1-antitrypsin, antithrombin and the mechanism of inflammatory thrombosis. Nature 317, 730–732.[CrossRef][Medline]
Casaretto JA, Corcuera LJ. 1998. Proteinase inhibitor accumulation in aphid-infested barley leaves. Phytochemistry 49, 2279–2286.[CrossRef]
Cordero MJ, Raventos D, San Segundo B. 1994. Expression of a maize proteinase inhibitor gene is induced in response to wounding and fungal infection: systemic wound-response of a monocot gene. The Plant Journal 6, 141–150.[CrossRef][Medline]
Dahl SW, Rasmussen SK, Hejgaard J. 1996a. Heterologous expression of three plant serpins with distinct inhibitory specificities. Journal of Biological Chemistry 271, 25083–25088.
Dahl SW, Rasmussen SK, Petersen LC, Hejgaard J. 1996b. Inhibition of coagulation factors by recombinant barley serpin BSZx. FEBS Letters 394, 165–168.[CrossRef][Web of Science][Medline]
De Leo F, Ceci LR, Jouanin L, Gallerani R. 2001. Analysis of mustard trypsin inhibitor-2 gene expression in response to developmental or environmental induction. Planta 212, 710–717.[CrossRef][Web of Science][Medline]
Djie MZ, Le Bonniec BF, Hopkins PC, Hipler K, Stone SR. 1996. Role of the P2 residue in determining the specificity of serpins. Biochemistry 35, 11461–11469.[CrossRef][Medline]
Duffus CM, Cochrane MP. 1993. Formation of the barley grain—morphology, physiology and biochemistry. In: Bhatty RS, MacGregor W, eds. Barley: chemistry and technology. St Paul, Minnesota: The American Association of Cereal Chemists, 31–72.
Elliott PR, Lomas DA, Carrell RW, Abrahams JP. 1996. Inhibitory conformation of the reactive loop of alpha 1- antitrypsin. Nature Structural Biology 3, 676–681.[CrossRef][Web of Science][Medline]
Evans DE, Hejgaard J. 1999. The impact of malt derived proteins on beer foam quality. Part 1. The effect of germination and kilning on the level of protein Z4, protein Z7 and LTP1. Journal of the Institute of Brewing 105, 159–169.
Gettins PG. 2002. Serpin structure, mechanism, and function. Chemical Reviews 102, 4751–4804.[CrossRef][Web of Science][Medline]
Giese H, Hejgaard J. 1984. Synthesis of salt-soluble proteins in barley. Pulse-labelling study of grain filling in liquid-cultured detached spikes. Planta 161, 172–177.[CrossRef]
Giese H, Hopp HE. 1984. Influence of nitrogen nutrition on the amount of hordein, protein Z and ß-amylase messenger RNA in developing endosperms of barley. Carlsberg Research Communications 49, 365–383.
Habu Y, Fukushima H, Sakata Y, Abe H, Funada R. 1996. A gene encoding a major Kunitz proteinase inhibitor of storage organs of winged bean is also expressed in the phloem of stems. Plant Molecular Biology 32, 1209–1213.[CrossRef][Web of Science][Medline]
Hejgaard J. 1977. Origin of a dominant beer protein: immunochemical identity with a ß-amylase-associated protein from barley. Journal of the Institute of Brewing 83, 94–96.
Hejgaard J. 1982. Purification and properties of protein Z—a major albumin of barley endosperm. Physiologia Plantarum 54, 174–182.
Hejgaard J. 2001. Inhibitory serpins from rye grain with glutamine as P1 and P2 residues in the reactive center. FEBS Letters 488, 149–153.[CrossRef][Web of Science][Medline]
Hejgaard J, Boisen S. 1980. High-lysine proteins in Hiproly barley breeding: identification, nutritional significance and new screening methods. Hereditas 93, 311–320.
Hejgaard J, Hauge S. 2002. Serpins of oat (Avena sativa) grain with distinct reactive centers and inhibitory specificity. Physiologia Plantarum 116, 155–163.[CrossRef][Medline]
Hejgaard J, Jakobsen S, Bjørn SE, Kragh KM. 1992. Antifungal activity of chitin-binding PR-4 type proteins from barley grain and stressed leaf. FEBS Letters 307, 389–392.[CrossRef][Web of Science][Medline]
Hejgaard J, Kærsgaard P. 1983. Purification and properties of the major antigenic beer protein of barley origin. Journal of the Institute of Brewing 89, 402–410.
Hejgaard J, Rasmussen SK, Brandt A, Svendsen I. 1985. Sequence homology between barley endosperm protein Z and protease inhibitors of the 1-antitrypsin family. FEBS Letters 180, 89–94.
Hejgaard J, Sørensen SB. 1975. Characterization of a protein-rich beer fraction by two-dimensional immunoelectrophoretic techniques. Comptes Rendus des Travaux du Laboratoire Carlsberg 40, 187–203.
Huntington JA, Read RJ, Carrell RW. 2000. Structure of a serpin-protease complex shows inhibition by deformation. Nature 407, 923–926.[CrossRef][Medline]
Huber R, Carrell RW. 1989. Implications of the three-dimensional structure of alpha 1-antitrypsin for structure and function of serpins. Biochemistry 28, 8951–8966.[CrossRef][Medline]
Jakobsen KS, Kalvenes C, Olsen O-A. 1991. mRNA levels in the developing aleurone and starchy endosperm in wild type and a high lysine (lys 3a) mutant of barley. Physiologia Plantarum 83, 201–208.
Koiwa H, Bressan RA, Hasegawa PM. 1997. Regulation of protease inhibitors and plant defense. Trends in Plant Science 2, 379–384.[CrossRef][Web of Science]
Komiyama T, Quan LT, Salvesen GS. 1996. Inhibition of cysteine and serine proteinases by the cowpox virus serpin CRMA. Advances in Experimental Medicine and Biology 389, 173–176.[Medline]
Kragh KM, Jacobsen S, Mikkelsen JD, Nielsen KA. 1991. Purification and characterization of three chitinases and one beta-1,3-glucanase accumulating in the medium of cell-suspension cultures of barley (Hordeum vulgare L.). Plant Science 76, 65–77.
Kyhse-Andersen J. 1984. Electroblotting of multiple gels: a simple apparatus without buffer tank for rapid transfer of proteins from polyacrylamide to nitrocellulose. Journal of Biochemical and Biophysical Methods 10, 203–209.[CrossRef][Web of Science][Medline]
Laskowski M, Kato I. 1980. Protein inhibitors of proteinases. Annual Review of Biochemistry 49, 593–626.[CrossRef][Web of Science][Medline]
Lundgard R, Svensson B. 1989. A 39 kDa barley seed protein of the serpin superfamily inhibits alpha- chymotrypsin. Carlsberg Research Communications 54, 173–180.[Web of Science][Medline]
Marttila S, Tenhola T, Mikkonen A. 1996. A barley (Hordeum vulgare L.) LEA3 protein, HVA1, is abundant in protein storage vacuoles. Planta 199, 602–611.
Mundy J, Hejgaard J, Hansen A, Hallgren L, Jorgensen KG, Munck L. 1986. Differential synthesis in vitro of barley aleurone and starchy endosperm proteins. Plant Physiology 81, 630–636.
Olson ST, Stephens AW, Hirs CH, Bock PE, Bjork I. 1995. Kinetic characterization of the proteinase binding defect in a reactive site variant of the serpin, antithrombin. Role of the P1' residue in transition-state stabilization of antithrombin- proteinase complex formation. Journal of Biological Chemistry 270, 9717–9724.
Østergaard H, Rasmussen SK, Roberts TH, Hejgaard J. 2000. Inhibitory serpins from wheat grain with reactive centers resembling glutamine-rich repeats of prolamin storage proteins. Cloning and characterization of five molecular forms. Journal of Biological Chemistry 275, 33272–33279.
Rasmussen SK. 1993. A gene coding for a new plant serpin. Biochimica et Biophysica Acta 1172, 151–154.[Medline]
Rasmussen SK, Klausen J, Hejgaard J, Svensson B, Svendsen I. 1996. Primary structure of the plant serpin BSZ7 having the capacity of chymotrypsin inhibition. Biochimica et Biophysica Acta 1297, 127–130.[CrossRef][Medline]
Rasmussen SK, Welinder KG, Hejgaard J. 1991. cDNA cloning, characterization andexpression of an endosperm-specific barley peroxidase. Plant Molecular Biology 16,317–327.[CrossRef][Web of Science][Medline]
Rollini P, Fournier RE. 1997. A 370-kb cosmid contig of the serpin gene cluster on human chromosome 14q32.1: molecular linkage of the genes encoding alpha1-antichymotrypsin, protein C inhibitor, kallistatin, alpha 1-antitrypsin, and corticosteroid-binding globulin. Genomics 46, 409–415.[CrossRef][Web of Science][Medline]
Rosenkrands I, Hejgaard J, Rasmussen SK, Bjorn SE. 1994. Serpins from wheat grain. FEBS Letters 343, 75–80.[CrossRef][Web of Science][Medline]
Ryan CA. 1990. Proteinase inhibitors in plants: genes for improving resistance against insects and pathogens. Annual Review of Phytopathology 28, 425–449.[Web of Science]
Schechter I, Berger A. 1967. On the size of the active site in proteases. I. Papain. Biochemical and Biophysical Research Communications 27, 157–162.[CrossRef][Web of Science][Medline]
Schick C, Pemberton PA, Shi GP, Kamachi Y, Cataltepe S, Bartuski AJ, Gornstein ER, Bromme D, Chapman HA, Silverman GA. 1998. Cross-class inhibition of the cysteine proteinases cathepsins K, L, and S by the serpin squamous cell carcinoma antigen 1: a kinetic analysis. Biochemistry 37, 5258–5266.[CrossRef][Medline]
Scott FL, Hirst CE, Sun J, Bird CH, Bottomley SP. 1999. The intracellular serpin proteinase inhibitor 6 is expressed in monocytes and granulocytes and is a potent inhibitor of the azurophilic granule protease, cathepsin G. Blood 93, 2089–2097.
Silverman GA, Bird PI, Carrell RW, et al. 2001. The serpins are an expanding superfamily of structurally similar but functionally diverse proteins. Evolution, mechanism of inhibition, novel functions, and a revised nomenclature. Journal of Biological Chemistry 276, 33293–33296.
Skadsen RW, Sathish P, Kaeppler HF. 2000. Expression of thaumatin-like permatin PR5 genes switches from the ovary wall to the aleurone in developing barley and oat seeds. Plant Science 156, 11–22.[Medline]
Sørensen MB, Cameron-Mills V, Brandt A. 1989. Transcriptional and post-transcriptional regulation of gene expression in developing barley endosperm. Molecular and General Genetics 217, 195–201.[CrossRef]
Thomas JC, Adams DG, Keppenne VD, Wasmann CC, Brown JK, Kanost MR, Bohnert HJ. 1995. Manduca sexta encoded protease inhibitors expressed in Nicotiana tabacum provide protection against insects. Plant Physiology and Biochemistry 33, 611–614.
Torriglia A, Perani P, Brossas JY, Chaudun E, Treton J, Courtois Y, Counis MF. 1998. L-DNase II, a molecule that links proteases and endonucleases in apoptosis, derives from the ubiquitous serpin leukocyte elastase inhibitor. Molecular and Cellular Biology 18, 3612–3619. (Erratum in Molecular and Cellular Biology 18, 4947.)
Yoo BC, Aoki K, Xiang Y, Campbell LR, Hull RJ, Xoconostle-Cazares B, Monzer J, Lee JY, Ullman DE, Lucas WJ. 2000. Characterization of Cucurbita maxima phloem serpin-1 (CmPS-1). A developmentally regulated elastase inhibitor. Journal of Biological Chemistry 275, 35122–35128.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. l. C. Petersen, J. Hejgaard, G. A. Thompson, and A. Schulz Cucurbit phloem serpins are graft-transmissible and appear to be resistant to turnover in the sieve element-companion cell complex J. Exp. Bot., December 1, 2005; 56(422): 3111 - 3120. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||