JXB Advance Access originally published online on September 24, 2004
Journal of Experimental Botany 2004 55(408):2607-2616; doi:10.1093/jxb/erh267
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RESEARCH PAPER |
Genetic analysis of the function of major leaf proteases in barley (Hordeum vulgare L.) nitrogen remobilization
1Department of Plant Sciences and Plant Pathology, Montana State University, Bozeman, MT 59717-3150, USA
2Pioneer Hi-Bred Intl., Inc., Johnston, IA 50131-0552, USA
3Department of Plant Pathology, Kansas State University, Manhattan, KS 66506, USA
* To whom correspondence should be addressed. Fax: +1 406 994 7600. E-mail: fischer{at}montana.edu
Received 23 April 2004; Accepted 26 July 2004
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Most of the nitrogen harvested with the seeds of annual crops is remobilized and retranslocated within the plant between anthesis and plant death. While chloroplasts contain most of the reduced nitrogen present in photosynthetically active leaf cells, the (major) pathway(s) involved in the degradation of their proteins prior to the retranslocation of the resulting amino acids are unknown. In this study, a population of 146 recombinant inbred barley lines (RIL), derived from the cross between two varieties with a highly inheritable difference in grain protein concentration, was used to map quantitative trait loci (QTL) for leaf amino-, carboxy- and endopeptidase activities relative to previously determined QTL for grain protein, leaf N storage, and remobilization. The results strongly suggested that major endopeptidases, assayed at both acidic and slightly alkaline pH values (favouring vacuolar and extravacuolar enzymes, respectively) are not instrumental in leaf N remobilization or the control of grain protein accumulation. Similarly, QTL determined for aminopeptidases (relative to QTL for N remobilization) indicated no functional role for the enzyme(s) assayed in plant N recycling. By contrast, careful evaluation of QTL data suggested that one or several carboxypeptidase isoenzymes may be involved in this physiologically and economically important process. As these proteases (in contrast to aminopeptidases) have previously been localized in vacuoles, this result appears intriguing. These data, while shedding new light on an old problem, also indicate that the described approach may prove useful in evaluating the functional roles of additional (not assayed in this study) proteolytic systems in whole-plant nitrogen recycling.
Key words: Aminopeptidase, barley, carboxypeptidase, endopeptidase, Hordeum vulgare L., nitrogen remobilization, peptide hydrolase, QTL, senescence
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Nitrogen remobilization from senescing vegetative tissues, such as leaves and stems, is an exceedingly important process for the N economy of both annual and perennial plants. It has been estimated that 70% or more of the nitrogen harvested in seeds of annual crop species is derived from this process (Peoples and Dalling, 1988
); consequently, only a minor fraction of the harvested nitrogen is taken up from the soil and/or reduced during the period between anthesis and plant maturity.
Up to 75% of the reduced nitrogen present in photosynthetically active cells of cereal leaves is located in the chloroplasts (Peoples and Dalling, 1988; Hörtensteiner and Feller, 2002). Stromal (especially Rubisco, but also other Calvin cycle enzymes) and thylakoidal proteins represent the major fractions of chloroplast nitrogen. Therefore, plastidial proteins are the predominant source of nitrogen recycled during leaf senescence. It has been found that leaf photosynthetic capacity and the levels of major chloroplast proteins decline during the early phases of the senescence process (before dark respiration; Feller and Fischer, 1994), indicating a rapid reallocation of these resources within the plant.
It has been demonstrated that chloroplasts contain a number of proteases, such as the ClpC/P system, FtsH, and the DegP and Lon proteases (Ostersetzer and Adam, 1997; Nakabayashi et al., 1999; Adam et al., 2001; Haussühl et al., 2001; Adam and Clarke, 2002). In addition, some evidence indicates that a metallo-endopeptidase may be involved in the first steps of plastidial protein degradation (Bushnell et al., 1993; Roulin and Feller, 1998), and a large fraction of the aminopeptidases present in plant cells are located in plastidial compartments (Thayer et al., 1988). Isolated chloroplasts, and extracts prepared from isolated chloroplasts, have the capacity to initiate the degradation of major stromal proteins (e.g. Rubisco and plastidial glutamine synthetase; Mitsuhashi and Feller, 1992; Mitsuhashi et al., 1992; Kokubun et al., 2002). Therefore, it appears likely that degradation of chloroplast proteins is initiated inside the intact organelles in vivo at the onset of senescence (Hörtensteiner and Feller, 2002). On the other hand, proteolytic activities are also present in most other cellular compartments and might be active in the degradation of peptides released from chloroplasts (Distefano et al., 1999; Brouquisse et al., 2001). Cytosolic aminopeptidases and oligopeptidases may cleave peptides released from plastids to free amino acids; in addition, the proteasome and other endoproteolytic systems are also located in the cytosol (Hörtensteiner and Feller, 2002). Plant biologists have been intrigued for a number of years by the high activity of cysteine endopeptidases and carboxypeptidases present in lytic vacuoles, especially during senescence. Using molecular methods, several authors have also described increased transcript levels of cysteine protease genes in senescing tissues of different species (Buchanan-Wollaston, 1997; Ueda et al., 2000; Guo et al., 2004). Some cellular evidence points to a possible role of vacuolar proteases in the degradation of plastidial proteins after vacuolar autophagy of chloroplasts or chloroplast fragments, but evidence for such a process is, at present, not clear-cut (Minamikawa et al., 2001; Hörtensteiner and Feller, 2002). It has therefore been suggested that a role for these enzymes in the final stages of the senescence process, after loss of membrane integrity, is more likely (Hörtensteiner and Feller, 2002).
The approaches used so far to investigate plastidial protein degradation have not yet been successful in clearly delineating the major pathway(s) responsible for this important process. Fortunately, in the past few years, with the development of molecular marker techniques, it has become easier to obtain detailed genetic maps for model or major crop species such as cereals, and to use these maps to locate chromosomal regions containing genes for physiologically and/or economically important traits (Obara et al., 2001; Yamaya et al., 2002; Gallais and Hirel, 2004). A population of 146 recombinant inbred lines (RIL) was developed at Montana State University from the cross between a high- and a low-grain protein barley variety and used to map quantitative trait loci (QTL) for grain protein (See et al., 2002), leaf N storage, and N remobilization (Mickelson et al., 2003). A major grain protein QTL was detected near marker HVM74 on barley chromosome 6; gene(s) present at this locus and influencing this trait may be identical to those present at a recently identified locus on wheat chromosome 6B (Olmos et al., 2003; Distelfeld et al., 2004). In addition, utilizing the barley RIL, a number of QTL for leaf nitrogen storage and remobilization were detected, especially on chromosomes 3, 6, and 7 (Mickelson et al., 2003). To initiate a novel approach to the in vivo functional characterization of major leaf proteases (for which simple, high-throughput assays are available), the goal of this work was to map such enzymes relative to previously detected QTL for grain protein, yield, and N remobilization. The assays used were aimed at detecting major (vacuolar) endo- and carboxypeptidases as well as enzymes potentially active in the cytosol and/or chloroplast (aminopeptidases and endopeptidases active at neutral and slightly alkaline pH values).
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Plant material
Two barley (Hordeum vulgare L.) varieties, ‘Lewis’ (CI15856) and ‘Karl’ (CI15487), were chosen as parents for this study, based on marked differences in grain protein concentration at maturity. Karl is a six-rowed variety which produces grain of consistently lower protein concentration than most other varieties. Lewis is a commonly grown two-rowed barley. A 146-member recombinant inbred population was developed by single seed descent from a cross and used to map the genes responsible for barley grain protein concentration (See et al., 2002
), leaf nitrogen storage, and remobilization (Mickelson et al., 2003). For the present study, F8 and F9 plants were grown in two independent replicates in a randomized block design near Bozeman, MT in summer 2000 and 2001 using standard farming practice. Flag leaves (10 leaves per sample) from all 146 lines as well as the parental lines, Lewis and Karl, were collected at mid-grain fill and plant maturity, immediately frozen in liquid nitrogen, transported to the laboratory in liquid nitrogen, and stored at –80 °C. Leaves were weighed, ground to a fine powder in liquid nitrogen using a mortar and pestle, and stored again at –80 °C until analysis. The same plant samples as those used for a previous publication (Mickelson et al., 2003) were used to facilitate comparison between the two studies.
Protease activity assays
Enzymes were extracted from liquid nitrogen powder (1:4, w/v) in 25 mM TRIS-HCl pH 7.5 containing 1% (w/v) insoluble polyvinylpolypyrrolidone and 0.1% (v/v) ß-mercaptoethanol using a mortar and pestle. Samples were centrifuged (10 min, 20 000 g), and the supernatants were directly used for protease activity assays.
For the analysis of endo- and exoproteolytic activities, extracts were desalted by centrifugation through Sephadex G-25 columns equilibrated with 25 mM TRIS-HCl pH 7.5, containing 0.1% (v/v) ß-mercaptoethanol (Feller et al., 1977). Aminopeptidase activities were assayed using 2 mM L-leu-p-nitroanilide in 100 mM Na-phosphate buffer pH 7.0 with 1% (v/v) DMSO as substrate. The assays were performed kinetically, at room temperature, for 10 min at 405 nm, using a SPECTRAmax PLUS384 spectrophotometer (Molecular Devices, Sunnyvale, CA.). Carboxypeptidase activities were performed using 2 mM N-carbobenzoxy-L-phe-L-ala in 100 mM Na-acetate pH 5.0, containing 2% (v/v) DMSO as substrate. Substrate was omitted for blanks. Enzyme assays (in wells of microtitration plates) were incubated for 1 h at 37 °C, and liberated 
-amino groups were assayed with a TNBS reagent as described by Fischer et al. (1998). The method was calibrated with 0–50 nmol Gly. Endopeptidases were quantified using 1% (w/v) azocasein as substrate, either in 0.2 M Na-acetate buffer pH 5.4 containing 0.1% (v/v) ß-mercaptoethanol, or in 0.2 M TRIS-HCl buffer, pH 7.5. After 3 h of incubation at 37 °C, undigested azocasein and large fragments were precipitated with cold TCA (final concentration: 5%, w/v), and small fragments resulting from proteolytic digestion were assayed in microplates as described by Fischer et al. (1998).
Statistical and QTL analysis
Data were collected for both independent replications for all nitrogen uptake, storage, and remobilization traits from both years (2000 and 2001) and analysed as previously described (Mickelson et al., 2003). Data analysis was conducted using the General Linear Model procedure of SAS (SAS Institute Inc., 1990). Phenotypic correlations were calculated among traits using least square mean values for each genotype, and narrow sense heritability on an entry-mean basis was determined.
A 110-point linkage map developed by See et al. (2002) for the barley RIL population was used for genetic analyses. The PlabQTL Version 1.1 mapping program (Utz and Melchinger, 1996), composite interval mapping function was selected for QTL analysis employing the covariate SELECT option which uses step-wise multiple regression to select cofactors. For QTL model building and detection of QTLxenvironment interactions, a LoD threshold of 2.5 was used which allowed comparison of the resultant QTL models with those previously reported for N remobilization in this barley population (Mickelson et al., 2003). The additive effect of a marker was calculated by PlabQTL as [(mean of the homozygous Karl class–mean of the homozygous Lewis class)/2]. QTL support intervals are calculated as the point along the significance peak at which the LoD score is 1.0 unit less than the peak LoD score. The phenotypic variance 
explained by a single QTL was estimated by the square of the partial correlation coefficient (R2). The phenotypic variance explained by the QTL model was estimated by the adjusted correlation coefficient 
which accounts for the number of predictors in the QTL model.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Genotypic variation and heritability
Considerable variation was observed in the RIL for all proteolytic activities assayed in this study (Fig. 1). Differences in the parental lines, which were originally selected for their variance in grain protein concentration, were smaller for most traits measured (not shown), demonstrating transgressive segregation in both directions. Narrow-sense heritability was 60.7/35.6% for aminopeptidases, and 19.3/25.3% for carboxypeptidases at mid-grain fill and maturity, respectively. For endopeptidases, narrow-sense heritability was 20.5/11.2% (assayed at pH 5.4), and 8.5/25.4% (assayed at pH 7.5) at the two developmental stages.
|
An analysis of variance was performed on all parameters measured. This analysis indicated significant (P <0.01) genotypexyear interactions for aminopeptidases present in mature leaves, and close to this level (P=0.012) for endopeptidases (pH 7.5), also measured from fully senesced leaves. Interactions were predominantly associated with differences in magnitude of effects between years; therefore, for the analyses presented here (Tables 1–3
; Figs 1, 2), data from both years (2000 and 2001) were combined for these, as well as all the other traits. In addition, correlation and QTL analyses were performed on separate years for the two traits with significant genotypexyear interactions (data not shown) in order to investigate situations for which data interpretation may be influenced by year-to-year variation.
|
|
|
|
Aminopeptidase activities were considerably higher around mid-grain fill (Fig. 1A; mean of 0.86 µmol min–1 g–1 FW) than in fully senesced leaves from mature plants (Fig. 1B; mean of 0.12 µmol min–1 g–1 FW). This finding is in agreement with earlier studies indicating that aminopeptidase activities are higher in young than in senescing or senesced tissues (Feller and Fischer, 1994
). Mean values for carboxypeptidases were 44.8 (Fig. 1C) and 120 µmol h–1 g–1 FW (Fig. 1D) at the two sample dates. Endopeptidases measured at pH 5.4 were higher than those measured at pH 7.5, with means of 2.8 and 4.2 mg h–1 g–1 FW at the acidic, and 0.89 and 2.8 mg h–1 g–1 FW at the slightly alkaline pH. By contrast with aminopeptidases, it has been shown by several authors that carboxy- and endopeptidases increase in senescing tissues, with some endopeptidases reaching maximal activities in completely senesced plant material (Feller and Fischer, 1994). Values obtained for all measured traits approached a normal distribution (Fig. 1); however, as none of the data sets were statistically normal, square root or logarithmic transformations were used prior to QTL analysis as appropriate.
Correlation analysis
The main purpose of this research was to contribute to the functional analysis of major leaf proteases with respect to cereal nitrogen recycling, utilizing a genetic approach. To achieve this goal, in addition to QTL analyses (see below), simple statistical correlations were calculated (i) between the different proteolytic activities measured (Table 1), and (ii) between protease activities and previously assayed and mapped parameters associated with leaf N storage and remobilization (Mickelson et al., 2003; Table 2).
Aminopeptidases at mid-grain fill were positively correlated with remaining aminopeptidase activity in completely senesced leaves, as well as with endopeptidases (measured at pH 5.4) at mid-grain fill and endopeptidase activities (measured at pH 7.5) in completely senesced leaves (Table 1). Aminopeptidases remaining at maturity were positively correlated with endopeptidases (pH 5.4) present in fully senesced leaves. Carboxypeptidases detectable at mid-grain fill with the substrate used showed no significant correlations with the other peptide hydrolases assayed; on the other hand, carboxypeptidase activity in leaves of mature plants were positively correlated with endopeptidases (pH 5.4) present at the same developmental stage. Besides the correlations mentioned with aminopeptidases, endopeptidases measured at an acidic pH value (5.4, at mid-grain fill) were strongly positively correlated with endopeptidases measured at pH 7.5 at the same developmental stage, but only weakly (not significant at P < 0.05) with these enzyme(s) in fully senesced leaves. Similarly, there was a strong positive correlation between endoproteolytic activities measured at both pH values in leaves from mature plants.
While correlations between protease activities indicate which proteolytic enzymes are up- or down-regulated under similar physiological conditions, correlations between such enzymes and traits indicating N storage and remobilization may represent a first step towards the identification of peptide hydrolases functionally involved in plastidial protein degradation and nitrogen recycling within the plant (Table 2). In this context, the most conspicuous result for leaf aminopeptidases present at both mid-grain fill and in fully senesced leaves is their negative correlation with 
N (a measure of leaf N remobilization), with yield and with protein yield. In addition, no correlation was found between these enzymes and grain protein concentration. On the other hand, aminopeptidases are mostly positively correlated with parameters indicating increased N content in vegetative tissues, such as total leaf nitrogen, nitrate, and amino acid contents.
By contrast with aminopeptidases, carboxypeptidases were not significantly correlated with any leaf N storage or remobilization parameters used in this study.
Endopeptidases assayed at an acidic pH value were generally positively correlated with total leaf N content, but negatively correlated with N. In addition, at mid-grain fill, these enzymes were negatively correlated with leaf nitrate levels measured at an earlier developmental stage (anthesis), and positively correlated with free amino acids. The same enzymes, when measured in fully senescent leaves, were negatively correlated with heading date and grain protein. Endopeptidases measured at a slightly alkaline pH value (at mid-grain fill) were negatively correlated with leaf nitrates, assayed at both anthesis and mid-grain fill, but showed no significant correlations with any of the other traits associated with leaf N storage or remobilization. These enzymes, when assayed at maturity, were again positively correlated with leaf N content, but negatively correlated with N, heading date, and grain protein. A slightly negative correlation was also found with protein yield, but was not statistically significant at P <0.05. Clearly, from simple correlation analysis in this population of RIL, no possible role of the proteases measured in leaf N remobilization can be postulated.
QTL analysis
Since simple correlation analysis of quantitative traits (such as proteolytic activities) is unable to discern the number and relative contribution of loci which may be influencing each given trait, these loci were identified using a previously published map (See et al., 2002) for the population of RIL utilized in this study (Table 3; Fig. 2). To gain additional information on the potential functional involvement of the assayed leaf proteases in N recycling (missed with simple correlation analysis), protease QTL are shown in the same map as was previously utilized for leaf N storage and recycling QTL (Mickelson et al., 2003; Fig. 2).
Two QTL were found for carboxypeptidases at mid-grain fill. The first locus, on chromosome 3, overlaps with a QTL for leaf nitrate at anthesis; low nitrate contents are associated with high carboxypeptidase activity. The second locus, on chromosome 7, colocalizes with a QTL for N; at this locus, allele(s) associated with high carboxypeptidase activity are also associated with high values for N remobilization. Six QTL were found for carboxypeptidase activity in mature leaves, on all chromosomes except 3 and 5 (Table 3; Fig. 2). Loci on chromosomes 1, 6, and 7 (near marker acgt63) do not overlap with any other QTL present in the map. On chromosome 2, a carboxypeptidase QTL overlaps with a QTL for yield and is close to a QTL for leaf amino acids; at this locus, high carboxypeptidase activity is associated with high yield. On chromosome 4, at markers acag273 and acac272, a carboxypeptidase QTL colocates with QTL for leaf amino acids (at mid-grain fill) and total nitrogen in fully senesced leaves. At this locus, low carboxypeptidase activity is associated with high values for nitrogen remaining in fully senesced leaves. Similarly, a second carboxypeptidase QTL on chromosome 7 partially overlaps with a QTL for non-mobilized leaf N; again, at this locus, low carboxypeptidase activities appear to be associated with poor N retranslocation from flag leaves. Results obtained for carboxypeptidases convincingly demonstrate the power of QTL mapping for the functional analysis of enzymes; none of the relationships between N remobilization and carboxypeptidase activities were detected with simple correlation analyses (Table 2).
QTL for aminopeptidase activity measured at mid-grain fill were present on chromosomes 3 and 7. The QTL on chromosome 3 broadly overlapped with a QTL for yield, and partially overlapped with a QTL for leaf amino acids; at this locus, allele(s) associated with high aminopeptidase activity and leaf amino acid levels were correlated with low yield (Fig. 2). Two loci are present on chromosome 7 for this trait. For the first locus, allele(s) associated with high aminopeptidase activity are also associated with high total leaf protein at mid-grain fill; the second QTL broadly overlapped with QTL for leaf N remobilization (N) and residual N in fully senesced leaves. In this case, low aminopeptidase activity was associated with high N and low residual leaf N, suggesting that aminopeptidases regulated or encoded by genes at this locus are not functionally involved in leaf N remobilization. QTL for aminopeptidase activities remaining in mature leaves were found on chromosomes 1 and 3. The locus on chromosome 1 is close to a QTL for heading date; on chromosome 3, one locus overlapped with QTL for leaf nitrogen, with high aminopeptidase activity correlating with high leaf N content, including high N in fully senesced leaves. The second aminopeptidase QTL overlaps with a QTL for amino acid levels at mid-grain fill; again, high aminopeptidase activities are positively correlated with high leaf amino acid levels.
Two QTL were detected for endopeptidases (pH 5.4) at mid-grain fill, on chromosomes 3 and 7. The major QTL, with a LoD of 6.7, near marker HVM33 on chromosome 3, did not overlap with any other QTL located on the map; however, the second QTL, near markers pinb1 and pinb2 on chromosome 7, overlaps with QTL for carboxypeptidase activity (at mid-grain fill) and N. At this locus, in contrast to carboxypeptidases, low endopeptidase activity is associated with high values for N. One QTL was detected for endopeptidases (pH 5.4) present in fully senesced leaves. This locus, at marker hvbkasi on chromosome 2, overlaps with QTL for endopeptidases (pH 7.5) at maturity, heading date, and grain protein concentration. At this locus, alleles associated with high endoproteolytic activity (at both pH values) are associated with low grain N and early flowering. Two QTL for endopeptidase at mid-grain fill (pH 7.5) were identified in this study. For the first locus, at marker acaa210 on chromosome 2, alleles associated with high protease activity are also associated with late heading and low yield. At the second locus, near marker acgc140 on chromosome 7, low endoproteolytic activity was found to be associated with low aminopeptidase activity and efficient N remobilization from senescing leaves.
Because of the significant genotypexyear interactions detected for both aminopeptidases and endopeptidases (pH 7.5) present in fully senesced leaves, QTL analyses for these two traits were also performed separately for both years included in this study (data not shown). For aminopeptidases, the QTL on chromosome 1 (at marker acat128) was not found in the separate data sets. While no QTL (with LoD >2.5) were present in the separately analysed 2001 data, two loci were found on chromosome 3 in the 2000 data set. The position of the first QTL (at marker acgc155) remained unchanged with respect to the combined analysis of both years (Table 3; Fig. 2), while the position of the second locus was shifted to marker acgc469.
For endopeptidases (pH 7.5) measured at maturity, the QTL present in the combined analysis from both years (at marker hvbkasi, on chromosome 2) was also found in the 2001 (but not the 2000) data set. An additional QTL for this trait was found on chromosome 1, at marker actt239 (not overlapping with any other QTL), in the 2000 data analysis.
To detect any possible influence of allele(s) associated with the major grain protein QTL on chromosome 6 (which explains 
46% of the variance in this trait; See et al., 2002; Mickelson et al., 2003) on leaf proteolytic activity, the region around marker HVM74 was studied in more detail. No protease QTL overlapping with this grain protein QTL were detected using mean values from both years of analysis (2000 and 2001; Fig. 2); however, detailed separate analysis of each year (data not shown) allowed some interesting observations. First, a QTL with a LoD of 3.66 (and a support interval from 250–270 cM) was detected at HVM74 for endopeptidases (pH 5.4) at mid-grain fill in the 2000 data set; in this case, the ‘Karl’ allele(s) (low grain protein content) were also associated with low protease activity. Second, a QTL for endopeptidases (pH 5.4) in fully senesced leaves with a LoD of 2.93 (and a support interval of 228–246 cM, i.e. not overlapping with the grain protein QTL) was also detected in the 2000 data set. In this case, however, the ‘Karl’ allele(s) were associated with an increase in leaf proteolytic activity. These findings should be associated with previously reported data (Mickelson et al., 2003), indicating that allele(s) present at this locus may also influence leaf nitrate and amino acid levels at mid-grain fill. While these QTL are not stable under changing environmental conditions, their colocation with the major grain protein QTL appears intriguing.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Over the last few years, QTL mapping has been successfully used to improve our understanding of basic physiological problems (Obara et al., 2001
; Yamaya et al., 2002; Gallais and Hirel, 2004). In this context, the main purpose of this work was to analyse the function of major leaf proteases in nitrogen remobilization from senescing leaves to developing seeds of annual plants. A population of RIL, derived from a cross between two barley varieties with a highly heritable difference in grain protein concentration, appeared to be the most promising starting material for this research, especially as a genetic map with good coverage was available (See et al., 2002), and QTL relating to leaf N storage and remobilization had previously been mapped in this population (Mickelson et al., 2003).
Role of major leaf proteases in nitrogen remobilization
The most conspicuous result of this study, underlining the power of QTL mapping for functional studies, is the strong indication that one or several carboxypeptidase isoenzymes are involved in nitrogen retranslocation from senescing barley leaves. This result appears especially intriguing with regard to the ongoing discussion on the compartmentation of plastidial proteolysis, as carboxypeptidases assayed at acidic pH values have previously been localized in vacuolar compartments (Brouquisse et al., 2001). Among the two carboxypeptidase QTL found at mid-grain fill, the locus on chromosome 7 is of special interest, as allele(s) associated with high carboxypeptidase activity are also associated with high values for N, and low values for N remaining in fully senesced leaves. Since allele(s) present at this locus influence several parameters potentially associated with N recycling, this locus appears to be promising for closer analysis. In this context, synteny between the model species rice and other grasses has been successfully used for positional cloning efforts in cereals (Goff et al., 2002; Distelfeld et al., 2004), and may also prove useful for this project. Among the six carboxypeptidase QTL identified when these enzyme(s) were assayed in mature plants, three demonstrated overlaps with other QTL indicating a role for these protease(s) in N remobilization. On chromosome 2, allele(s) associated with high carboxypeptidase activity were also associated with high yield; on chromosome 4 and 7, negative correlations were observed between carboxypeptidase activity and total N in fully senesced leaves. Together, these data indicate that one or several barley carboxypeptidase genes are involved in leaf N remobilization, favouring yield and protein yield.
In marked contrast to carboxypeptidases, these data are clearly not in favour of a functional role of the assayed amino- and endopeptidases (measured at both pH 5.4 and 7.5) in nitrogen retranslocation from senescing barley leaves. Simple correlation analysis (Table 2) did not indicate any significant, positive correlations between these enzymes and parameters associated with efficient N remobilization; by contrast, negative correlations were found between some of the activities assayed on one side and N, grain protein concentration, or protein yield on the other side. In addition, while overlaps of some amino- and endopeptidase QTL with QTL associated with N retranslocation were found on chromosomes 2, 3, and 7 (Fig. 2), the direction of observed allelic effects contradicts a role of these enzymes in N retranslocation at all the loci concerned. The fact that some protease QTL (most notably the major QTL for endopeptidases measured at pH 5.4, at marker HVM33 on chromosome 3) do not co-localize with any loci for N retranslocation suggests that the principal functional role of these enzymes is independent of N metabolism. Therefore, understanding the reason(s) causing senescing plant tissues to accumulate some of these proteases to high levels remains, at present, a fascinating problem.
Environmental influence and heritability
Since significant genotypexyear interactions were detected for two of the assayed traits (aminopeptidases and endopeptidases [pH 7.5]) in fully mature leaves, QTL analyses of the combined data set (from 2000 and 2001) were complemented by separate analyses for each year (see Results). While this analysis was judged to be essential, it does not change the interpretation presented in the last paragraph. In the same context, it has to be borne in mind that narrow-sense heritability of the protease activities measured varied considerably (see Results); while aminopeptidase activities (especially at mid-grain fill) were highly heritable, lower values were found for some of the other proteases, suggesting an important environmental influence on these parameters.
Of special interest was the detailed, separate (by year) analysis of the area around marker HVM74 (chromosome 6) for all traits. This analysis indicated that allele(s) present at this locus, associated with low grain protein, were also associated with low endopeptidase activity (pH 5.4) at mid-grain fill in the 2000 (but not 2001) data set. This result is the only indication from this study potentially involving major leaf endopeptidase(s) in N translocation to the grains; however, again, this effect appears to be influenced by environmental conditions. Similarly, Mickelson et al. (2003) reported QTL for leaf nitrate and amino acids in the 2001 (but not the 2000) dataset at this locus; in this case, the Karl allele(s) associated with low grain protein were associated with higher levels of leaf nitrogen. It appears intriguing that the major grain protein QTL, explaining 46% of the variability in this trait, is not stably associated with QTL for parameters such as N and/or residual N in fully senesced leaves; this indicates that there is no direct, functional correlation between grain protein concentration and N remobilization from vegetative plant parts. Based on both this and the previous study in this series (Mickelson et al., 2003), it appears that efficient N remobilization is associated with high yield and protein yield, but not with grain protein concentration.
In summary, with the exception of carboxypeptidase(s) (for which more detailed biochemical and molecular studies appear justified), these results are not in favour of a quantitatively important role for the studied proteases in N remobilization from leaves to the developing cereal kernels, supporting the hypothesis that the decisive steps may be catalysed by less conspicuous (i.e. harder to detect) plastidial or extraplastidial proteases. This study also indicates that the approach presented here could be used to evaluate the function of such proteases on a case-by-case basis if sufficiently simple, high-throughput quantitative assays are first developed.
|
Acknowledgements |
|---|
This work has been supported by award number 0101019 from the USDA-NRI Competitive Grants program to AMF.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Adam Z, Adamska A, Nakabayashi K, et al. 2001. Chloroplast and mitochondrial proteases in Arabidopsis. A proposed nomenclature. Plant Physiology 125, 1912–1918.
Adam Z, Clarke AK. 2002. Cutting edge of chloroplast proteolysis. Trends in Plant Science 7, 451–456.[CrossRef][ISI][Medline]
Brouquisse R, Masclaux C, Feller U, Raymond P. 2001. Protein hydrolysis and nitrogen remobilization in plant life and senescence. In: Lea PJ, Morot-Gaudry JF, eds. Plant nitrogen. Berlin: Springer, 275–293.
Buchanan-Wollaston V. 1997. The molecular biology of leaf senescence. Journal of Experimental Botany 48, 181–199.[ISI]
Bushnell TP, Bushnell D, Jagendorf AT. 1993. A purified zinc protease of pea chloroplasts, EP1, degrades the large subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase. Plant Physiology 103, 585–591.[Abstract]
Distefano S, Palma JM, McCarthy I, del Río LA. 1999. Proteolytic cleavage of plant proteins by peroxisomal endoproteases from senescent pea leaves. Planta 209, 308–313.[CrossRef][ISI][Medline]
Distelfeld A, Uauy C, Olmos S, Schlatter AR, Dubcovsky J, Fahima T. 2004. Microcolinearity between a 2 cM region encompassing the grain protein content locus Gpc-6B1 on wheat chromosome 6B and a 350 kb region on rice chromosome 2. Functional and Integrative Genomics 4, 59–66.
Feller UK, Soong TS-T, Hageman RH. 1977. Leaf proteolytic activities and senescence during grain development of field-grown corn (Zea mays L.). Plant Physiology 59, 290–294.
Feller U, Fischer A. 1994. Nitrogen metabolism in senescing leaves. Critical Reviews in Plant Sciences 13, 241–273.
Fischer A, Brouquisse R, Raymond P. 1998. Influence of senescence and carbohydrate levels on the pattern of leaf proteases in purple nutsedge (Cyperus rotundus). Physiologia Plantarum 102, 385–395.
Gallais A, Hirel B. 2004. An approach to the genetics of nitrogen use efficiency in maize. Journal of Experimental Botany 55, 295–306.
Goff SA, Ricke D, Lan T-H, et al. 2002. A draft sequence of the rice genome (Oryza sativa L. ssp japonica). Science 296, 92–100.
Guo Y, Cai Z, Gan S. 2004. Transcriptome of Arabidopsis leaf senescence. Plant, Cell and Environment 27, 521–549.[CrossRef]
Haussühl K, Andersson B, Adamska I. 2001. A chloroplast DegP2 protease performs the primary cleavage of the photodamaged D1 protein in plant photosystem II. EMBO Journal 20, 713–722.[CrossRef][ISI][Medline]
Hörtensteiner S, Feller U. 2002. Nitrogen metabolism and remobilization during senescence. Journal of Experimental Botany 53, 927–937.
Kokubun N, Ishida H, Makino A, Mae T. 2002. The degradation of the large subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase into the 44 kDa fragment in the lysates of chloroplasts incubated in darkness. Plant and Cell Physiology 43, 1390–1395.
Mickelson S, See D, Meyer FD, Garner JP, Foster CR, Blake TK, Fischer AM. 2003. Mapping of QTL associated with nitrogen storage and remobilization in barley (Hordeum vulgare L.) leaves. Journal of Experimental Botany 54, 801–812.
Minamikawa T, Toyooka K, Okamoto T, Hara-Nishimura I, Nishimura M. 2001. Degradation of ribulose-bisphosphate carboxylase by vacuolar enzymes of senescing French bean leaves: immunocytochemical and ultrastructural observations. Protoplasma 218, 144–153.[CrossRef][ISI][Medline]
Mitsuhashi W, Crafts-Brandner SJ, Feller U. 1992. Ribulose-1, 5-bisphosphate carboxylase/oxygenase degradation in isolated pea chloroplasts incubated in the light or in the dark. Journal of Plant Physiology 139, 653–658.
Mitsuhashi W, Feller U. 1992. Effects of light and external solutes on the catabolism of nuclear-encoded stromal proteins in intact chloroplasts isolated from pea leaves. Plant Physiology 100, 2100–2105.
Nakabayashi K, Ito M, Kiosue T, Shinozaki K, Watanabe A. 1999. Identification of clp genes expressed in senescing Arabidopsis leaves. Plant and Cell Physiology 40, 504–514.
Obara M, Kajiura M, Fukuta Y, Yano M, Hayashi M, Yamaya T, Sato T. 2001. Mapping of QTLs associated with cytosolic glutamine synthetase and NADH-glutamate synthase in rice (Oryza sativa L.). Journal of Experimental Botany 52, 1209–1217.
Olmos S, Distelfeld A, Chicaiza O, Schlatter AR, Fahima T, Echenique V, Dubcovsky J. 2003. Precise mapping of a locus affecting grain protein content in durum wheat. Theoretical and Applied Genetics 107, 1243–1251.[CrossRef][ISI][Medline]
Ostersetzer O, Adam Z. 1997. Light-stimulated degradation of an unassembled Rieske FeS protein by a thylakoid-bound protease: the possible role of the FtsH protease. The Plant Cell 9, 957–965.
Peoples MB, Dalling MJ. 1988. The interplay between proteolysis and amino acid metabolism during senescence and nitrogen reallocation. In: Noodén LD, Leopold AC, eds. Senescence and aging in plants. San Diego, CA: Academic Press, 181–217.
Roulin S, Feller U. 1998. Light-independent degradation of stromal proteins in intact chloroplasts isolated from Pisum sativum L. leaves: requirement for divalent cations. Planta 205, 297–304.[CrossRef]
SAS Institute Inc. 1990. SAS/Stat user's guide: Version 6, 4th edn. Vol. 2. Cary, NC: SAS Institute.
See D, Kanazin V, Kephart K, Blake T. 2002. Mapping genes controlling variation in barley grain protein concentration. Crop Science 42, 680–685.
Thayer SS, Choe HT, Rausser S, Huffaker RC. 1988. Characterization and subcellular localization of aminopeptidases in senescing barley leaves. Plant Physiology 87, 894–897.
Ueda T, Seo S, Ohashi Y, Hashimoto J. 2000. Circadian and senescence-enhanced expression of a tobacco cysteine protease gene. Plant Molecular Biology 44, 649–657.[CrossRef][ISI][Medline]
Utz HF, Melchinger AE. 1996. PlabQTL: a program for composite interval mapping of QTL. Journal of Quantitative Trait Loci 2, 1.
Yamaya T, Obara M, Nakajima H, Sasaki S, Hayakawa T, Sato T. 2002. Genetic manipulation and quantitative-trait loci mapping for nitrogen recycling in rice. Journal of Experimental Botany 53, 917–925.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Uauy, J. C. Brevis, and J. Dubcovsky The high grain protein content gene Gpc-B1 accelerates senescence and has pleiotropic effects on protein content in wheat J. Exp. Bot., August 1, 2006; 57(11): 2785 - 2794. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||