Cloning and Differential Expression Analysis of FaGF14-B and FaGF14-C Genes in Tall Fescue

Xiaoli WANG, Jiahai WU*, Xiaodong LI, Jianhong SHU, Xiaoxia LIU, Shuying WANG, Yiming CAI

1. Guizhou Institute of Prataculture, Guiyang 550006, China;

2. Guizhou Academy of Agricultural Sciences, Guiyang 550006, China

As an important small molecular modulator of protein, 12-3-3 proteins are ubiquitous in a variety of eukaryotes by binding to the target proteins in the form of monomer for dimer. They are referred to as GF14 proteins in plants. The sequences of 14-3-3 proteins are highly conserved[1]. Currently, it has been found that in plants, 14-3-3 proteins participate in the responses to biotic and abiotic stresses[2]. Under cold stress,the expression of 14-3-3 protein was up-regulated in callus and seedlings of rice[3]. In tomato, a lot of 14-3-3 protein was expressed after three consecutive hours of iron deficiency[4].Xu et al.found that in tomato roots, only the expression of TFT17 gene, belonging to the 14-3-3 gene family, was up-regulated after 48-h treatment of iron deficiency, and its maximum expression level was increased by four times[5].This suggests that the 14-3-3 gene is specifically in response to iron deficiency stress in tomato roots. In Arabidopsis, the transcript levels of 14-3-3ψ and 14-3-3μ declined under phosphorus deficiency;while under potassium deficiency, the expression level of 14-3-3χ protein was reduced, but of 14-3-3ω protein was increased[6].Under high salt stress,the binding of plasma membrane H+-ATPase and 14-3-3 protein induced the ion homeostasis inside and outside the membrane[2,7]. Under osmotic stress, the accumulation of 14-3-3protein significantly increased the plasma membrane H+-ATPase activity in maize root cells[8]. The overexpression of ZmGF14-6 gene from maize could enhance the drought resistance in rice[9].The overexpression of GF14-λ from Arabidopsis could maintain the evergreen state and enhance the drought resistance of cotton[10].

Tall fescue is a major cool-season pasture and turf grass, and it plays an important role in soil and water conservation, environmental protection and urban green space and playing fields planting.Tall fescue has a strong resistance.It is widely used in China’s turf and landscaping,and has become one of the major domestic turfgrass applications[11-12]. Qiancao 1, as a tall fescue cultivar, was bred by the Guizhou Institute of Prataculture. It has a strong resistance,and is tolerant to low nitrogen. In the production,Qiancao 1 has a low demand for nitrogen fertilizer. Currently, the physiological response mechanism and molecular regulation foundation of Qiancao 1 under nitrogen stress have been studied from the perspectives of morphology, physiology, transcriptomics and proteomics.It is found that the expression levels of 14-3-3 protein and gene are relatively high under nitrogen stress. Therefore, the genetic sequence fragment of 14-3-3 of tall fescue, obtained by transcriptome sequencing, was used as a template,and the GF14-B and GF14-C genes were amplified using the 3’RACE and 5’RACE techniques. Moreover, their differential expression under nitrogen stress was studied by real-time fluorescence quantitative PCR. This study will lay a theoretical basis for screening of low nitrogen-tolerant genes and breeding of low nitrogen-tolerant forage germplasms.

Materials and Methods

Materials

Plant material Qiancao 1 is new national forage cultivar bred by the Guizhou Institute of Prataculture in 2005.Its registration number is 299.

Main reagents The E. coli DH5α competent cells were purchased from the Tiangen Biotech (Beijing) Co., Ltd.The cloning vector pMD 19-T, RNA extraction kit (TaKaRa RNAiso Reagent), RNA reverse transcription kit (RevertAid First cDNA Synthesis Kit) were purchased from the Thermo Scientific Company. The 5’RACE kit(5’RACE System for Rapid Amplification of cDNA Ends, Version 2.0) was purchased from the Shanghai Invitrogen Biotechnology Co., Ltd. The3’RACE kit (SMARTerTMRACE cDNA Amplification Kit) was purchased from the Clontech Laboratories Inc. The DNA Marker DL2000, 100 bp DNA Ladder Marker, LA Taq enzyme,EASY Dilution, DNaseI and In Vitro Transcription T7 kit were purchased from the TaKaRa Biotechnology(Dalian) Co., Ltd. The Taq enzyme(Advantage 2 PCR kit)was purchased from the Clontech Laboratories Inc.The agarose DNA extraction kit was purchased from the Omega BioTek Inc.The yeast extract,peptone and agarose were purchased from the Oxoid Inc.The X-gal,IPTG and Ampicillin were purchased from the Beijing Soledad Technology Co.,Ltd.

Methods

Treatment of plant material The tall fescue plants with similar genetic background were selected as tested material. Their full seeds were selected and soaked in water at 50 ℃overnight. Subsequently, the seeds were soaked and stirred in 75%ethanol for 30 s. And then, they were rinsed three times with sterile water.After soaked in 0.1% HgCl for 4 min,the seeds were rinsed 5-6 times with sterile water. The filter paper was laid in petri dishes,and then sterile distilled water was added into the dishes with a pipette.The seeds were placed on the filter paper, which was maintained moist with sterile distilled water. The seeds were inoculated in a lighting incubator(lighting time of 16h;dark time of 8 h; growth temperature of 25 ℃).After the germination, the seeds were transferred to self-made hydroponic culture cups containing Hoagland solution. After a 30-d culture, the nitrogen-free treatment (nitrogen stress treatment,and the NO3-in the solution was substituted by Cl-) and normal nitrogen treatment (control treatment,and the NO3-concentration in the Hoagland solution was as usual) were performed for the seeds.The nutrient solution was replaced every three days.At the same time, in order to reduce the effect of microenvironment in the greenhouse,the planting pots were rotated regularly. The leaves of tall fescue in both of the two treatment groups were sampled every day.They were frozen immediately with liquid nitrogen, and then preserved at -80 ℃for use.

Total RNA extraction The total RNA in leaves of Qiancao 1 was extracted using extraction kit (TaKaRa RNAiso Reagent).The absorbances of the extracted RNA at 260 and 280 nm were determined with a UV spectrophotometer. The A260/A280ratio was calculated, and the concentration(μg/ml) of total RNA was calculated(A260×40×dilution time).The integrity of extract RNA was examined by 1.0%agarose gel electrophoresis.And then,the RNA was preserved at -80 ℃for use.

Primer design The genetic sequence fragments of FaGF14-B and FaGF14-C in tall fescue, obtained by transcriptome sequencing were used as templates. Using the Primer Premier 5.00, three specific downstream primers were designed for each of them. Combining with the primers in the 5’RACE kit, the 5’ ends of FaGF14-B and FaGF14-C were cloned. The primers were named as FaGF14-B-rev1, FaGF14-B-rev2, Fa-GF14-B-rev3, FaGF14-C-rev1, FaGF-14-C-rev2, FaGF14-C-rev3, respectively. At the same time, two specific upstream primers were designed, and combining with the primers in 3’RACE kit, the 3’ ends of FaGF14-B and FaGF14-C were cloned. The primers were named as FaGF14-Bfwd1, FaGF14-B-fwd2, FaGF14-Cfwd1, FaGF14-C-fwd2, respectively.After the fragments were assembled correspondingly, the accuracy of the assembly was verified using primers designed based on the coding regions of FaGF14-B and FaGF14-C. The primers were named as FaGF14-Bfwd4, FaGF14-B-rev4, FaGF14-Cfwd4,FaGF14-C-rev4,respectively.In addition, two pairs of primers were designed for differential expression analysis of FaGF14-B and FaGF14-C under nitrogen stress by real-time fluorescence quantitative PCR. The primers were named as FaGF14-B-fwd5, FaGF14-B-rev5, FaGF14-Cfwd5 and FaGF14-C-rev5, respectively. In the PCR analysis, the Ubiquitinconjugating enzyme (EG974074) in tall fescue was used as a reference gene. The designed primers for the reference gene were named as UBIfwd1 and UBI-rev1.

Full-length cDNA cloning of FaGF14-B and FaGF14-C Combing the designed primers and provided primers by the 3’RACE and 5’RACE kits, the 5’ and 3’ ends of the FaGF14-B and FaGF14-C genes were amplified as described by instructions.The recovered PCR products were inserted in the T vector and cloned. The amplified fragments were sequenced and then spliced with known sequences by DNAMAN. Thus the fulllength cDNA sequences of FaGF14-B and FaGF14-C were obtained.

Functional analysis of proteins encoded by FaGF14-B and FaGF14-C The amino acid sequences were spliced using DNAMAN. Based on amino acid sequences, the phylogenic tree of FaGF14-B , FaGF14-C and GF14s in other plants was constructed using Mega5.03. The open reading frames of FaGF14-B and FaGF14-C genes searched in NCBI, and the functional domains of proteins encoded by FaGF14-B and FaGF14-C were analyzed(http://www.ncbi.nlm.nih.gov/gorf/gorf.html; http://www.ncbi.nlm.nih.gov/structure/c dd/wrpsbcgi).The relative molecular weight, isoelectric point and hydrophobicity of FaPHYA were analyzed using ExPASy (http://cn.expasy.org/tools/protparam. Html;http://www.expasy.org/cgi-bin/protscale.pl). The secondary structures of FaGF14-B and FaGF14-C were analyzed using SOPMA (http://npsa-pbil.ibcp.fr/cgi-bin/npsa_automat. pl). The subcellular localization analysis was performed at http://www.csbio.sjtu.edu.cn/bioinf/Cell-PLoc. The new accession numbers were applied in Gen-Bank.

Real-time FQ-PCR The amplification based on SYBR Green I fluorescent dye was performed in a PCR instrument(Eppendorf realplex2realtime PCR detection system, Eppendorf).Thus the relative quantitative results of samples were obtained.

Table 1 Sequences of primers for amplifying cDNA sequences of FaGF14-B and FaGF14-C

Results and Analysis

Cloning and amino acid sequence analysis FaGF14-B and FaGF14-Cin tall fescue

Using the sequence fragments of FaGF14-B and FaGF14-C as the templates, the 5’ ends of FaGF14-B and FaGF14-C were cloned using realtime fluorescence quantitative PCR by combining the 6 designed downstream primers(Table 1)and provided primers by the 5’RACE kit (Fig.1,Fig.2). There were overlapping region between obtained PCR product and the known sequence,so the obtained PCR product was successfully spliced with theknown sequence. Combing the 4 designed upstream primers and provided primers by the 3’RACE kit, the 3’ ends of FaGF14-B and FaGF14-C were amplified by real-time PCR(Fig.3, Fig.4). There were overlapping region between obtained PCR product and the known sequence, so the obtained PCR product was successfully spliced with the known sequence.Thus, the FaGF14-B and FaGF14-C genes in tall fescue were cloned.Since the sequence fragments of FaGF14-B and FaGF14-C genes in tall fescue,obtained by transcriptome sequencing,were used as templates, there might be deletions in cloned sequences.Therefore, specific primers were designed for the two ends of coding regions of FaGF14-B and FaGF14-C to verify the full lengths of cloned sequences (Fig.5, Fig.6). The results showed that the sequences of amplified PCR products were consistent with the known sequences, suggesting that the assembly of cloned sequences of FaGF14-B and FaGF14-C was correct. The full length of FaGF14-B gene is 1 548 bp, and of FaGF14-C gene is 1 250 bp. The similarity between sequences of FaGF14-B and FaGF14-C genes was 50.74%. In GenBank, the accession numbers of FaGF14-B and FaGF14-C were KR095168 and KR095169,respectively.

Analysis of proteins encoded by FaGF14-B and FaGF14-C in tall fescue

The analysis of full lengths of FaGF14-B and FaGF14-C cDNA sequences showed that there is one complete open reading frame in FaGF14-B, and its initiation codon starts at the 449thnucleotide and termination codon ends at the 1 228thnucleotide. The encoded protein by FaGF14-B gene presumably contains 260 amino acids (Fig.7, Fig.8). There is one complete open reading frame in FaGF14-C, and its initiation codon starts at the 66thnucleotide and termination codon ends at the 848thnucleotide. The encoded protein by FaGF14-C presumably contains 261 amino acids (Fig.9, Fig.10). The relative molecular weight and isoelectric point of FaGF14-B were speculated as 29.36 kDa and 4.62 respectively, and of FaGF14-C were speculated as29.54 kDa and 4.74 respectively. The subcellular localization analysis showed that FaGF14-B and FaGF14-C are all located in nucleus. The hydrophobicity of FaGF14-B and FaGF14-C were analyzed using Ex-PASy’s ProtScale. The results showed that the maximum hydrophobic value of FaGF14-B is 1.722, and minimum hydrophobic value is-3.078;above the score of 1.5, there is one hydrophobic peak, and below the score of -2, there are four hydrophilic peaks (Fig.11). The minimum hydrophobic value of FaGF14-C is 1.722,and minimum value is -3.708; above the score of 1.5, there are two hydrophobic peaks, and below the score of -2, there are five hydrophilic peaks(Fig.12). The secondary structures of FaGF14-B and FaGF14-C were predicted using SOPMA. FaGF14-B protein contains about 57.14% of the α-helix, 10.81% of extended strand,5.79% of β-corner and 26.25% of irregular curling (Fig. 13); and FaGF14-C protein contains about 64.23% of the α-helix, 10.00% of extended strand, 5.38% of β-corner and 20.38%of irregular curling(Fig.14).

Conserved domain analysis of FaGF14-B and FaGF14-C in tall fescue

The conserved domain functional analysis showed that FaGF14-B and FaGF14-C proteins all have a typical 14-3-3 protein domain (Fig.15,Fig.16).The amino acid sequences and secondary structures of FaGF14-B and FaGF14-C in tall fescue were aligned with those of GF14s in Brachypodium distachyon, Arabidopsis thaliana, Oryza sativa and Zea mays. The results showed that the FaGF14-B and FaGF14-C in tall fescue have the typical secondary structure of GF14s. Their secondary structures all contain nine conserved α-helixes and non-conserved N-and C-terminals (Fig.17). The function of their last α-helixes is mainly to bind with functional proteins.Therefore,the differences in secondary structure have decisive significances for functions of GF14s.

Phylogenic analysis

The phylogenic tree of GF14s was constructed with neighbor-jointing method using Mega 5.03(Fig.18).The GF14s were classified into four major large groups. Cicer arietinum is clustered into one group. The FaGF14-B and FaGF14-C from tall fescue and GF14s from gramineous plants have genetic relationships, and they areclustered into one group.This might be because that tall fescue is also a gramineous plant, and the GF14s of gramineous plants have a common conserved region.As shown in Fig.18,there are differentiations in GF14 genes among different families and genera. In conclusion, in the evolution of plants, the GF14 genes are changed and differentiated constantly.

Differential expression analysis of FaGF14-B and FaGF14-C in tall fescue under low nitrogen stress

The expression levels of FaGF14-B and FaGF14-C genes in tall fescue under nitrogen stress were analyzed by real-time fluorescence quantitative PCR. The results showed that there were no significant differences in transcript level of FaGF14-B gene between treatment and control groups on day 1 and 3; the transcript level of FaGF14-B was increased sharply on day 4;the transcript level of FaGF14-B was decreased sharply on day 5; from day 5 to day 12, the transcript level of FaGF14-B gene was not changed significantly (Fig.19).There were no significant differences in transcript level of FaGF14-C gene between treatment and control groups from day 1 to day 8; the transcript level of FaGF14-C gene was increased gradually with the increased intensity of nitrogen stress from day 9 to day 12 (Fig.20). It suggests that both the FaGF14-B and FaGF14-C genes are in response to nitrogen stress.

Conclusions and Discussion

FaGF14-B and FaGF14-C proteins having conserved secondary structures in tall fescue

The dimer of 14-3-3 protein is composed of two connected monomers.Its spatial structure has amphipro-tic grooves.Each monomer consists of nine α-helixes (α1-α9), which form an anti-parallel mode. There is a short inter-ring connection between every two adjacent α-helixes. Among the nine αhelixes, the dimer interface is composed of α1, α2, α3 and α4. The α1 and α4 are two necessary components for dimer. The amino acid sequences of α3 and α4 are the longest(30-34 amino acids). The α3, α5, α7 and α9 form the asymmetry grooves on both sides.On one side,α7 and α9 form the hydrophobic interface containing four Leu side chains; while on the other side, α3 formed three basic side chains, and α5 formed a charged group. The polar residues in α3 and α5 form the polar interface,and the hydrophobic residues in α7 and α9 form the non-polar interface[13-14].

The subunits of 14-3-3 protein all have a conserved core region, but the N- and C-terminals were not conservative. The analysis of primary structures of 14-3-3 family members in Arabidopsis showed that the N-terminal has a low homology (14%), while the homology of middle region reaches 51%[15]. If the region composed of 12-30 amino acids at the N-terminal is mutated, the 14-3-3 protein will lose the ability to form dimer and to bind with some target proteins.In the 7th-9thα-helixes at the C-terminal, there are binding sites that participate directly in interaction between proteins and proteins[16]. It indicates that the N- and C-terminal domains are essential for the formation of 14-3-3 protein dimer.In addition, the conservatism of the N- and C-terminal domains determines the functional diversity of 14-3-3 protein family.

In this study, the FaGF14-B and FaGF14-C proteins in tall fescue have the typical secondary structure of GF14s. Their secondary structures all contain nine conserved α-helixes and non-conserved N- and C-terminals,which all meet the structure characteristics of 14-3-3 proteins.

FaGF14-B and FaGF14-C genes making responses to nitrogen stress in tall fescue

14-3-3 plays an important role in the regulation of plant metabolic balance under different nutritional conditions.In Arabidopsis,14-3-3χ,14-3-3κ and 14-3-3ψ are the target proteins of SnRK2.8 kinase under nutrient deficiency. Under nutrient deficiency,SnRK2.8 kinase reduces the expression of 14-3-3ψ and regulates nutrient metabolism of plants through phosphorylation of 14 -3 -3ψ and other proteins. Under K and N deficiencies,the overexpression of 14-3-3ψ elongates the newly-grown roots in Arabidopsis[17]. In the metabolism of N, P and K in plants, 14-3-3 proteins interact with a variety of proteins to change the activity of target proteins[18].The affinity chromatography shows that the nitrate reductase (NR), alanine-glyoxylate aminotransferase,glutamine synthetase(GS)and S-adenosylmethionine synthetase all can interact with 14-3-3χ, 14- 3-3κ and 14-3-3ψ proteins[19-20]. These types of enzymes are all related to the metabolic pathways of nitrogen in plants. 14-3-3 proteins regulate nitrogen metabolism by regulating the activity of GS[21]. Finnemann studied the GS in senescent leaves of Brassica, and he found that the phosphorylation level of GS is affected by ATP/AMP, and the binding of GS and 14-3-3 proteins can significantly increase enzyme activity in the senescence process of leaves[22].Only binding with 14-3-3 proteins can the GS activity in tobacco chloroplasts reach the maximum level[23]. As a key enzyme in nitrogen metabolism, NR is widely distributed in plant cells, and the binding of 14-3-3 proteins and phosphorylated NR inactivates NR[24].Previous studies have shown that the 14-3-3 proteins balance the nitrogen metabolism in plants by regulating enzyme activities in nitrogen metabolismr elated pathways. In this study, both the FaGF14-B and FaGF14-C genes in tall fescue are in response to nitrogen stress,suggesting that they are all involved in the nitrogen metabolism in plants, which is consistent with previous study results. This study will lay a theoretical basis for the functional analysis of FaGF14-B and FaGF14-C genes.

[1]HIRSCH S,AITKEN A,BERTSCH U,et al. A plant homologue to mammalian brain 14-3-3 protein and protein kinase C inhibitor[J]. FEBS Letters, 1992, 296:222-224.

[2]SEHNKE PC, HENRY R, CLINE K, et al.Interaction of a plant 14-3-3 protein with the signal peptide of a thylakoidtargeted chloroplast precursor protein and the presence of 14-3-3 isoforms in the chloroplast stoma[J]. Plant Physiology,2000,122:235-242.

[3]BIHN EA, PUL AL, WANG SW, et al.Localization of 14-3-3 protein in the nuclei of Arabidopsis and maize [J].Plant J,1997,12:1439-1445.

[4]WANG YH,GARVIN DF,KOCHIAN LV.Rapid induction of regulatory and transporter genes in response to phosphorus, potassium, and iron deficiencies in tomato roots: evidence for cross-talk and root/rhizosphere-mediated signals[J]. Plant Physiology, 2002, 130: 1361-1370.

[5]XU WF,SHI WM.Expression profiling of the 14-3-3 gene family in response to salt stress and potassium and iron deficiencies in young tomato (Solanum lycopersicum)roots:analysis by real-time RT-PCR [J]. Ann Bot, 2006, 98: 965-974.

[6]BUNNEY TD,WALRAVEN HSV,BOER ADH. 14-3-3 protein is a regulator of the mitochondrial and chloroplast ATP synthase [J]. Proceedings of the National Academy of Sciences, 2001, 98:4249-4254.

[7]KIDOU S, UMEDA M, KATO A, et al.Isolation and characterization of a rice cDNA similar to the bovine brain-specific 14-3-3 protein gene [J].Plant Mol Biol,1993,21:191-194.

[8]COMPAROT S, LINGIAH Q, MARTIN T. Function and specificity of 14-3-3 proteins in the regulation of carbohydrate and nitrogen metabolism[J].J Exp Bot,2003,54:595-604.

[9]ZHU JK. Regulation of ion homeostasis under salt stress[J].Curr Opin Plant Biol,2003,6:441-445.

[10]AKSAMIT A, KOROBCAZK A, SKALAJ J, et al. The 14-3-3 gene expression specificity in response to stress is promoter-dependent[J]. Plant Cell Physiology,2005,46(10):1635-1645.

[11]HAN XG (韩晓光), XUE ZY (薛哲勇),ZHI DY(支大英),et al.High-efficiency inducement and salt-tolerant mutant selection of embryonic calli of Festuca arundinacea (高羊茅胚性愈伤组织的高效诱导及其耐盐突变体筛选)[J].Acta Prataculturae Sinica (草业学报),2005,14(6):112-118.

[12]CHEN Q (陈群),YUAN XJ (袁晓君),HE YL (何亚丽).Screening molecular markers for heat tolerance and its relation to summer tolerance in tall fescue single plants (高羊茅单株耐热性相关分子标记的筛选及其与越夏性的关系研究)[J].Acta Prataculturae Sinica(草业学报),2013,22(5):84-95.

[13]XIAO B, SMERDON SJ, JONES DH,et al. Structure of a 14-3-3 protein and implications for coordination of multiple signaling pathways [J]. Nature, 1995,376:188-191.

[14]LIU D,BIENKOWSKAJ,PETOSAC,et al.Crystal structure of the zeta isoform of the 14-3-3 protein [J].Nature,1995,376:191-194.

[15]YANG X, LEE WH, SOBOTT F, et al.Structural basis for protein-protein interactions in the 14-3-3 protein family[J].PNAS,2006,103:17237-17242.

[16]WU K, ROONEY MF, FERL RJ. The Arabidopsis 14-3-3 multigene family[J].Plant Physiology, 1997, 114: 1421 -1431.

[17]CHANG IF, CURRARI A, WOOISEY R, et al. Proteomic profiling of tandem affinity purified 14-3-3 protein complexes in Arabidopsis thaliana [J]. Proteomics,2009,9(11):2967-2985.

[18]SHIN R, BERG RH, SCHACHTMAN DP.Reactive oxygen species and root hairs in Arabidopsis root response to nitrogen, phosphorus and potassium deficiency [J]. Plant Cell Physiology,2005,46(8):1350-1357.

[19]SHIN R, JEZ JM, BASRA A, et al. 14-3-3 Proteins fine-tune plant nutrient metabolism [J]. FEBS Letters, 2011,585(1):143-147.

[20]COTELLE V, MEEK SEM, PROVAN F,et al.14-3-3s regulate global cleavage of their diverse binding partners in sugar-starved Arabidopsis cells [J].EMBO J,2000,19(12):2869-2876.

[21]COMPAROT S, LINGIAH G, MARTIN T. Function arid specificity of 14-3-3 proteins in the regulation of carbohydrate and nitrogen metabolism [J].Journal of Experimental Botany, 2003,54(382):595-604.

[22]FINNEMANN J, SCHJOERRING JK.Post-translational regulation of cytosolic glutamine synthetase by reversible phosphorylation and 14-3-3 protein interaction [J]. Plant Journal,2000,24:171-181.

[23]RIEDEL J, TISCHNER R, MAECK G.The chloroplastic glutamine synthetase(GS-2)of tobacco is phosphorylated and associated with 14-3-3 proteins inside the chloroplast [J].Planta,2001,213:396-401.

[24]ATHWAL GS, HUBER SC. Divalent cations and polyamines bind to loop 8 of 14-3-3 proteins, modulating their interaction with phosphorylated nitrate reductase[J]. The Plant Journal, 2002,29(2):119-129.