Vascular expression of the grp1.8 promoter is controlled by three specific regulatory elements and one unspecific activating sequence

The bean grp1.8 full-length promoter is specifically active in vascular tissue during normal development of tobacco. Deletion of a negative regulatory element resulted in ectopic activity of the promoter in cortical cells of hypocotyls, roots and stems. A 169 bp fragment (–205 to –36) of the grp1.8 promoter conferred vascular-specific expression to CaMV 35S minimal promoters whereas a 141 bp fragment (–205 to –64) strongly activated these minimal promoters both in vascular and cortical cells. These experiments defined a new regulatory element (VSE) that is essential for vascular-specific expression and is located between –64 and –36. The 141 bp grp1.8 promoter sequence had enhancer-like properties as it was active in both orientations. A 24 bp sequence (bp –119 to –96, corresponding to the SE1 regulatory element) enhanced expression from several minimal promoters strongly but unspecifically, whereas a 26 bp sequence (–98 to –73, corresponding to the RSE regulatory element) induced vascular-specific expression. Thus, the grp1.8 promoter is regulated by a combinatorial mechanism that can integrate the action of different, non-additively acting regulatory elements into vascular-specific expression.     

Motifs specific for the ADR1 NBS–LRR protein family in <Emphasis Type="Italic">Arabidopsis</Emphasis> are conserved among NBS–LRR sequences from both dicotyledonous and monocotyledonous plants

The activated disease resistance (ADR) 1 gene encodes a protein that possesses an N-terminal coiled-coil (CC) motif, nucleotide-binding site (NBS) and C-terminal leucine-rich repeat (LRR) domains. ADR1 belongs to a small, atypical Arabidopsis thaliana sub-class containing four CC–NBS–LRR genes. The NBS region of most NBS–LRR proteins possesses numerous conserved motifs. In contrast, the LRR domain, which is subject to positive selection, is highly variable. Surprisingly, sequence analysis revealed that the LRR domain of the ADR1 sub-class was more conserved than the NBS region. Sequence analysis identified two novel conserved motifs, termed TVS and PKAE, specific for this CC–NBS–LRR sub-class. The TVS motif is adjacent to the P-loop, whereas the PKAE motif corresponded to the inter-domain region termed the NBS–LRR linker, which was conserved within the different CC–NBS–LRR classes but varied among classes. These ADR1-specific motifs were employed to identify putative ADR1 homologs in phylogenetically distant and agronomically important plant species. Putative ADR1 homologs were identified in 11 species including rice and in 3 further Poaceae species. The ADR1 sub-class of CC–NBS–LRR proteins is therefore conserved in both monocotyledonous and dicotyledonous plant species.     

Physical and Genetic Maps of thedeafwaddlerRegion on Distal Mouse Chr 6

Thedeafwaddler(dfw) mutation, displaying motor ataxia and profound deafness, arose spontaneously in a C3H/HeJ colony and was mapped previously to distal mouse Chr 6. In this study, a high-resolution genetic map was generated by positioning 10 microsatellite markers and 5 known genes on a 968-meioses intersubspecific backcross segregating fordfw[(CAST/Ei–+/+ × C3HeB/FeJ–dfw/dfw) × C3HeB/FeJ–dfw/dfw], giving the following marker order and sex-averaged distances:D6Mit64–(0.10 + 0.10 cM)–Pang–(1.24 + 0.36 cM)–Itpr1–(0.62 + 0.25 cM)–D6Mit108–(0.52 + 0.23 cM)–D6Mit54–(0.21 + 0.15 cM)–D6Mit23, D6Mit107, D6Mit328–(0.72 + 0.27 cM)–D6Mit11–(0.21 + 0.15 cM)–dfw–(0.93 + 0.31 cM)–Gat4, D6Mit55–(0.10 + 0.10 cM)–D6Mit63–(0.31 + 0.18 cM)–Syn2–(0.62 + 0.25 cM)–D6Mit44(Rho). Female and male genetic maps are similar immediately surrounding thedfwlocus, but show marked differences in other areas. A yeast artificial chromosome-based physical map suggests that the closest markers flanking thedfwlocus,D6Mit11(proximal) andGat4, D6Mit55(distal), are contained within 650–950 kb. The human homologues of the flanking lociItpr1(proximal) andSyn2(distal) map to chromosome 3p25–p26, suggesting that the human homologue of thedfwgene is located within this same region.     

Extractive fermentation for enhanced gellan-hydrolysing enzyme production by Bacillus thuringiensis H14

A new extractive fermentation process using PEG and potassium phosphate aqueous two-phase system (ATPS) was developed for enhanced production of gellan-hydrolysing enzyme by Bacillus thuringiensis H14. Five different Bacillus sp. were tested for their ability to synthesize gellan-hydrolysing enzyme. Bacillus thuringiensis H14 was found to be the best organism for gellan-hydrolysing enzyme production. The enzyme showed maximum activity at pH 7.5 and 40 °C. The partition studies of gellan-hydrolysing enzyme in the system using PEG X (X = 9000, 6000, 4000) and potassium phosphate–water and PEG–sodium citrate–water system indicated at PEG (4000)– potassium phosphate–water is the best system for partitioning of gellan-hydrolysing enzyme into the PEG phase (K = 4.99). Gellan-hydrolysing enzyme production by Bacillus thuringiensis H14 was studied in ATPSs composed of PEG X (X = 9000, 6000, 4000) and potassium phosphate. The top phase is continuous and rich in PEG while the bottom phase is dispersed and is rich in phosphate, microbial cells being mainly retained in the bottom phase. The gellan-hydrolysing enzyme produced during fermentation partitioned into the upper PEG phase and total gellan-hydrolysing enzyme produced was 2.12, 2.29 and 2.40 times higher than that of homogeneous fermentation when the fermentations were carried out using PEG 9000–potassium phosphate–water, PEG 6000–potassium phosphate–water, PEG 4000–potassium phosphate–water systems respectively.     

Linkage of Genes Controlling Morphological Traits with Isozyme Markers of Rye Chromosomes

Data on linkage of 12 rye genes controlling morphological traits (el, Vs, ln, w, np, ct2, Hs, Ddw, cb, mn, vi1, mp) with one or several isozyme markers of individual rye chromosomes (2R–7R) are presented. Linkage of the following gene pairs was established: chromosome 2R: Est3/5–el, el–-Glu, Sod2–el, Sod2–Vs; chromosome 3R: ln–Got4; chromosome 4R: w–Got1, np–Got1; chromosome 5R: Est4–ct2, Est6/9–ct2, ct2–Est2, ct2–Aco2, Est2–Hs, Aco2–Hs, Est2–Ddw, Aco2–Ddw; chromosome 6R:Lap2–cb, cb–Aco1, Est10–mn; chromosome 7R: Acph2/3–vi1, Got2–vi1, mp–Acph2/3. The reasons for mapping a very small number of genes in rye in spite of high intraspecific variability of this species are discussed. An approach is suggested to improve this situation by simultaneous identification and mapping of all diverse spontaneous mutations maintained in heterozygous state in various rye cultivars.     

Comparison of the average structures,from molecular dynamics,of complexes of GTPase activating protein (GAP) with oncogenic and wild-type ras-p21: identification of potential effector domains

GTPase activating protein (GAP) is a known regulator of ras-p21 activity and is a likely target of ras-induced mitogenic signaling. The domains of GAP that may be involved in this signaling are unknown. In order to infer which domains of GAP may be involved, we have performed molecular dynamics calculations of GAP complexed to wild-type and oncogenic (Val 12–containing) ras-p21, both bound to GTP. We have computed and superimposed the average structures for both complexes and find that there are four domains of GAP that undergo major changes in conformation: residues 821–851, 917–924, 943–953, and 1003–1020. With the exception of the 943–953 domain, none of these domains is involved in making contacts with ras-p21, and all of them occur on the surface of the protein, making them good candidates for effector domains. In addition, three ras-p21 domains undergo major structural changes in the oncogenic p21-GAP complex: 71–76 from the switch 2 domain; 100–108, which interacts with SOS, jun and jun kinase (JNK); and residues 122–138. The change in conformation of the 71–76 domain appears to be induced by changes in conformation in the switch 1 domain (residues 32–40) and in the adjacent domain involving residues 21–31. In an accompanying paper, we present results from microinjection of peptides corresponding to each of these domains into oocytes induced to undergo maturation by oncogenic ras-p21 and by insulin-activated wild-type cellular p21 to determine whether these domain peptides may be involved in ras signaling through GAP.     

Pigeon flight in a wind tunnel

Summary Core temperatureT _c, breast temperatureT _s–br and leg temperatureT _s–1 were measured simultaneously in pigeons during rest and flight in a wind tunnel, using thermistors.MeanT _c at rest is 39.8±0.7°C and is independent of ambient temperatureT _a (10–30°C). In the first minutes of flight,T _c increases to 1.5–3.0°C above resting level and remains at this higher level. This hyperthermia increases withT _a (v=const.). It is±constant in the lowT _a range (10.6–13.9°C) at flight speeds v ranging from 10–18 m s^–1 and normal body mass, but increases with v and elevated body mass in the highT _a range (23.7–28.8°C). T _s–1 is adapted toT _a at rest and increases in flight up to 3–4°C belowT _c. This increase inT _s–1 is linear toT _a. T _s–br is always lower thanT _c, in extreme cases reaching restingT _c in flight.Supported by the Deutsche Forschungsgemeinschaft     

Production and characterization of long-acting recombinant human albumin–EPO fusion protein expressed in CHO cell

A long-lasting recombinant human albumin-linker-erythropoietin (EPO) is a human albumin gene fused to the N-terminal of EPO with a (GGSGG)_n-repeated linker inserted between albumin and EPO. Albumin–EPO fusion genes were co-transfected with the dhfr gene. Albumin–EPO fusion protein has three kinds of sub-types (IALE, AD2LE, AD1LE). Albumin–EPO fusion protein was quantified with human EPO ELISA. The in vitro efficacy of albumin–EPO fusion protein was estimated using F-36E cell, and in vivo efficacy of albumin–EPO fusion protein was estimated using normocythemic mice (B6D2F1). We also determined the in vivo half-life in a Sprague–Dawley rat. A PLA program analysis result demonstrated that the albumin–EPO fusion protein IALE is about 7.8-fold more potent than rHuEPO in increasing the hematocrit of normal mice.     

Minimal enhancer elements of the leghemoglobinlba andlbc3 gene promoters fromGlycine max L. have different properties

The characteristics of the soybean leghemoglobinlba gene promoter were analyzed and important promoter elements from thelba andlbc3 promoters were compared using transgenicLotus corniculatus plants. A 5 deletion analysis of thelba promoter delimited twocis-acting elements controlling expression: a distal positive element (–1254, –884) required for expression and a proximal element (–285, –60) essential for full-level activity. In contrast to the corresponding region of thelbc3 promoter, thelba proximal element is unable to control expression from the heterologous CaMV 35S enhancer. The upstream positive element of thelba gene contains a position- and orientation-independent enhancer between positions (–1091, –788). The sequence of this enhancer region is conserved in thelbc3 gene upstream (–1333, –1132) of the previously assigned strong positive element (SPE; –1090, –947). The present analysis revealed some of the properties of this extendedlbc3 SPE element. The extended element (–1364, –947) functions in both orientations from 5 locations whereas the SPE2 subcomponent (–1364, –1154) containing the conserved sequence is only active in the correct orientation. Removal of the SPE2 by internal deletion demonstrates that the SPE2 subcomponent is indispensable for the activity of thelbc3 upstream positive element. These results indicate that the upstream positive elements of thelba andlbc3 genes possess different properties although their conserved minimal enhancer sequence has similar function. This may reflect the differential expression of the twolb genes ofGlycine max L.     

Direct analysis of nitrocatechol-type glucuronides in urine by capillary electrophoresis–electrospray ionisation mass spectrometry and tandem mass spectrometry

Direct, quantitative capillary electrophoresis–electrospray ionisation mass spectrometric (CE–ESI-MS) and tandem mass spectrometric (CE–ESI-MS–MS) methods are described for the quantitation of 3-O-glucuronides of E- and Z-entacapone isomers (EEG and EZG) and tolcapone (TG) in urine. 3-O-Glucuronide of nitecapone was used as internal standard. Good separation of glucuronides was achieved with 20 mM ammonium acetate as separation solution at pH 6.84. Stacking was used to increase the sensitivity of the method by introducing samples in 5 mM ammonium acetate. CE–ESI-MS and CE–ESI-MS–MS methods are linear with correlation coefficients better than 0.9983 and 0.9982, and repeatable with relative standard deviations below 9 and 14%, respectively. The limit of detection (LOD) in CE–ESI-MS at signal-to-noise ratio 3 is 100 ng/ml for EEG and EZG and 250 ng/ml for TG. The CE–ESI-MS–MS method was the more sensitive; LOD was 7 ng/ml for all compounds, without any concentration of the sample.     

Impact of defoliation by the biocontrol agentZygogramma bicolorataon the weed Partheniumhysterophorus in Australia

The leaf-feeding beetle Zygogrammabicolorata Pallister was introduced from Mexico intoAustralia in 1980 as a biocontrol agent for the weedParthenium hysterophorus L. (Asteraceae). Z. bicolorata became abundant in 1990, and since 1992there has been regular outbreaks resulting in thedefoliation of the weed in central Queensland. In thisstudy we evaluated the impact of defoliation by Z. bicolorata on P. hysterophorus from 1996 to1998. Z. bicolorata caused 91–100% defoliationresulting in reductions in weed density by 32–93%,plant height by 18–65%, plant biomass by 55–89%,flower production by 75–100%, soil seed-bank by13–86% and seedling emergence in the following seasonby 73–90%. At sites with continued outbreaks ofZ. bicolorata, it is expected that the existing soilseed-bank will be minimised, resulting in reduceddensity of parthenium in 6 to 7 years.     

Response of chloride efflux from skeletal muscle ofRana pipiens to changes of temperature and membrane potential and diethylpyrocarbonate treatment

Summary Efflux of³⁶Cl^– from frog sartorius muscles equilibrated in two depolarizing solutions was measured. Cl^– efflux consists of a component present at low pH and a pH-dependent component which increases as external pH increases.For temperatures between 0 and 20°C, the measured activation energy is 7.5 kcal/mol for Cl^– efflux at pH 5 and 12.6 kcal/mol for the pH-dependent Cl^– efflux. The pH-dependent Cl^– efflux can be described by the relationu=1/(1+10n(pK _a -pH)), whereu is the Cl^– efflux increment obtained on stepping from pH 5 to the test pH, normalized with respect to the increment obtained on stepping from pH 5 to 8.5 or 9.0. For muscles equilibrated in solutions containing 150mm KCl plus 120mm NaCl (internal potential about –15 mV), the apparent pK _a is 6.5 at both 0 and 20°C, andn=2.5 for 0°C and 1.5 for 20°C. For muscles equilibrated in solutions containing 7.5mm KCl plus 120mm NaCl (internal potential about –65 mV), the apparent pK _a at 0°C is 6.9 andn is 1.5. The voltage dependence of the apparent pK _a suggests that the critical pH-sensitive moiety producing the pH-dependent Cl^– efflux is sensitive to the membrane electric field, while the insensitivity to temperature suggests that the apparent heat of ionization of this moiety is zero. The fact thatn is greater than 1 suggests that cooperativity between pH-sensitive moieties is involved in determining the Cl^– efflux increment on raising external pH.The histidine-modifying reagent diethylpyrocarbonate (DEPC) applied at pH 6 reduces the pH-dependent Cl^– efflux according to the relation, efflux=exp(–k·[DEPC]·t), wheret is the exposure time (min) to DEPC at a prepared initial concentration of [DEPC] (mm). At 17°C,k ^–1=188mm·min. For temperatures between 10 and 23°C,k has an apparent Q₁₀ of 2.5. The Cl^– efflux inhibitor SCN^– at a concentration of 20mm substantially retards the reduction of the pH-dependent Cl^– efflux by DEPC. The findings that the apparent pK _a is 6.5 in depolarized muscles, that DEPC eliminates the pH-dependent Cl^– efflux, and that this action is retarded by SCN^– supports the notion that protonation of histidine groups associated with Cl^– channels is the controlling reaction for the pH-dependent Cl^– efflux.     

Familial Correlations of the Ages of Manifestation of Lung Cancer and Large-Intestine Cancer

The probability to develop lung cancer before 80 years of age is 1.67 and 0.18% for the male and female populations of Kharkov, respectively; the probability to develop large-intestine cancer is 0.92 and 0.49%, respectively. The correlation coefficient (r) of the age of manifestation (AM) of cancer in parent–offspring pairs is 0.47. These correlation coefficients for the father–son, mother–daughter, mother–son, and father–daughter pairs are 0.64, 0.49, 0.44, and 0.37, respectively. If the parent has lung cancer, the correlation is stronger (r = 0.71). On average, cancer is manifested in offspring earlier than in parents (57 and 63 years, respectively); the differences in the father–daughter and mother–son pairs are 8.2 and 2.8 years, respectively. The best prognostic parameter is the AM of cancer in the father with respect to the AM in the son (b _yx = 0.45).     

Intrasexual Competition and Mating Behavior in <Emphasis Type="Italic">Ptomascopus morio</Emphasis> (Coleoptera: Silphidae Nicrophorinae)

In nicrophorine beetles, genus Nicrophorus care their larva using small vertebrate carrion, whereas genus Ptomaucopusreproduce with small vertebrate carrion but show no parental care. Aggression and sexual behavior were examined in Ptomascopus morio and Nicrophorus quadripunctatus. Nicrophorus quadripunctatus had intense female–female as well as male–male contests. In Ptomascopus morio, by contrast, female–female aggression was rarely observed. Male–male aggression (pushing, biting, male–male mounting) in Ptomascopus morio was observed when a resource for breeding was present, whether or not a female was present. The lack of female–female aggression, and male–male aggression when resources but not females are present, suggest that the mating system of Ptomascopus morio is resource defense polygyny. Large males of Ptomascopus morio were also found to exhibit mate choice, preferring large females over small females.     

Chaetosphaeria tortuosa, the newly discovered teleomorph of Menispora tortuosa, with a key to known Menispora species

Chaetosphaeria tortuosa is described as the newly discovered teleomorph of Menispora tortuosa, based on specimens from Canada and the Czech Republic, and single spore isolations from both morphs. The fungus produces superficial, more or less globose, papillate, dark brown to black smooth perithecia (200–)220–250 × (220–)230–260 μm. The asci are unitunicate, 8-spored, cylindrical-fusiform, (110–)120–133(–145) × 12–14 with a distinct apical, nonamyloid annulus 1–1.5 μm high, 3.5–4 μm wide. The ascospores are fusiform, 19–24 × 5–6 μm, hyaline, 3-septate, smooth, and 2-seriate in the ascus. The morphology of the teleomorph and anamorph are similar to that of C. ovoidea (anamorph: M. glauca), differing in dimensions of asci and ascospores, and in the disposition and morphology of the phialides of the anamorphs. The generic concept and phylogeny of Menispora is briefly discussed, and a key to the 11 species currently accepted in the genus is provided.     

Characterization of a weak allele of the GL1 gene of Arabidopsis thaliana

A genomic clone containing the gl1–2 allele has been isolated and sequenced. The predicted amino acid sequence of the gl1–2 protein is identical to that of the GL1-Col allele up to position 201. At this point in the coding region of gl1–2 there is a deletion relative to the wild-type sequence that results in an in-frame stop codon at position 202. This deletion removes 27 amino acid residues, including a highly negatively charged region, from the predicted gl1–2 polypeptide. The loss of this negatively charged carboxy-terminal region from the gl1–2 product is most likely the cause of the partial loss of gene activity which results in a reduction in leaf trichome initiation.     

Current-voltage relationships and voltage sensitivity of the Cl− pump inHalicystis parvula

Summary The current-voltage (I–V) relationships for internally perfused and nonperfused cells ofHalicystis parvula were determined. In both types of cells theI–V curve shows a conspicuous region of negative slope, beginning at vacuole potentials around –30 mV and continuing to values of +20 to +40mV. The negative slope in perfused cells is abolished by the metabolic inhibitors, darkness and low temperature. In order to determine the origin of this negative slope, we measured the voltage sensitivity of the unidirectional fluxes of Cl^–, Na⁺ and K⁺ in perfused cells. The results show that the Cl^– influx, which is mediated primarily by a Cl^– pump, increases as the vacuole potential is clamped at increasingly morenegative values up to –50 mV, while the other fluxes measured changed in the directions predicted by the change in electrical driving force. The voltage sensitivity of the Cl^– pump quantitatively accounts for the negative slope of theI–V curve. Also, we observed a large transient outward current of 10–20-sec duration following an abrupt depolarization by voltage clamping. This transient current was reduced or abolished by low temperature, which suggests that it may be due to the voltage-sensitive Cl^– pump. Finally, we found an inverse relationship between the transprotoplasm resistance (R _m ) and thePD under standard conditions, which suggests that the activity of the electrogenic Cl^– pump lowerR _m , i.e., it is a conductive pump.     

Inhibition of Oncogenic and Activated Wild-Type ras-p21 Protein-Induced Oocyte Maturation by Peptides from the ras-Binding Domain of the raf-p74 Protein, Identified from Molecular Dynamics Calculations

In the preceding paper we found from molecular dynamics calculations that the structure of the ras-binding domain (RBD) of raf changes predominantly in three regions depending upon whether it binds to ras-p21 protein or to its inhibitor protein, rap-1A. These three regions of the RBD involve residues from the protein–protein interaction interface, e.g., between residues 60 and 72, residues 97–110, and 111–121. Since the rap-1A–RBD complex is inactive, these three regions are implicated in ras-p21-induced activation of raf. We have therefore co-microinjected peptides corresponding to these three regions, 62–76, 97–110, and 111–121, into oocytes with oncogenic p21 and microinjected them into oocytes incubated in in insulin, which activates normal p2l. All three peptides, but not a control peptide, strongly inhibit both oncogenic p21- and insulin-induced oocyte maturation. These findings corroborate our conclusions from the theoretical results that these three regions constitute raf effector domains. Since the 97–110 peptide is the strongest inhibitor of oncogenic p21, while the 111–121 peptide is the strongest inhibitor of insulin-induced oocyte maturation, the possibility exists that oncogenic and activated normal p21 proteins interact differently with the RBD of raf.     

Inhibition of oncogenic and activated wild-type ras-p21 protein-induced oocyte maturation by peptides from the guanine-nucleotide exchange protein, SOS, identified from molecular dynamics calculations. Selective inhibition of oncogenic ras-p21

In the preceding paper we performed molecular dynamics calculations of the average structures of the SOS protein bound to wild-type and oncogenic ras–p21. Based on these calculations, we have identified four major domains of the SOS protein, consisting of residues 631–641, 676–691, 718–729, and 994–1004, which differ in structure between the two complexes. We have now microinjected synthetic peptides corresponding to each of these domains into Xenopus laevis oocytes either together with oncogenic (Val 12)-p21 or into oocytes subsequently incubated with insulin. We find that the first three peptides inhibit both oncogenic and wild-type p21-induced oocyte maturation, while the last peptide much more strongly inhibits oncogenic p21 protein-induced oocyte maturation. These results suggest that each identified SOS region is involved in ras–stimulated signal transduction and that the 994–1004 domain is involved uniquely with oncogenic ras–p21 signaling.     

Functions of lumican and fibromodulin: lessons from knockout mice

Lumican and fibromodulin are collagen-binding leucine-rich proteoglycans widely distributed in interstitial connective tissues. The phenotypes of lumican-null (Lum ^–/–), Fibromodulin-null (Fmod ^–/–) and compound double-null (Lum ^–/– Fmod ^–/–) mice identify a broad range of tissues where these two proteoglycans have overlapping and unique roles in modulating the extracellular matrix and cellular behavior. The lumican-deficient mice have reduced corneal transparency and skin fragility. The Lum ^–/– Fmod ^–/– mice are smaller than their wildtype littermates, display gait abnormality, joint laxity and age-dependent osteoarthritis. Misaligned knee patella, severe knee dysmorphogenesis and extreme tendon weakness are the likely cause for joint-laxity. Fibromodulin deficiency alone leads to significant reduction in tendon stiffness in the Lum ^+/+ Fmod ^–/– mice, with further loss in stiffness in a lumican gene dose-dependent way. At the level of ultrastructure, the Lum ^–/– cornea, skin and tendon show irregular collagen fibril contours and increased fibril diameter. The Fmod ^–/– tendon contains irregular contoured collagen fibrils, with increased frequency of small diameter fibrils. The tendons of Lum ^–/– Fmod ^–/– have an abnormally high frequency of small and large diameter fibrils indicating a de-regulation of collagen fibril formation and maturation. In tissues like the tendon, where both proteoglycans are present, fibromodulin may be required early in collagen fibrillogenesis to stabilize small-diameter fibril-intermediates and lumican may be needed at a later stage, primarily to limit lateral growth of fibrils Published in 2003.