Treatment with supplementary UV-B resulted in decreases in transcripts of the photosynthetic genes Lhcb and psbA and concomitant increase in transcripts of two pathogen-related genes, PR-1 and PDF1·2, in Arabidopsis thaliana. UV-B exposure caused increases in jasmonic acid (JA) levels and ethylene production. UV-B treatment of jar1 and etr1-1 mutants, which are insensitive to JA and ethylene, respectively, showed that the increase in PR-1 transcripts was dependent on ethylene and PDF1·2 transcripts on both JA and ethylene. In contrast, the down-regulation of photosynthetic transcripts was independent of both compounds. Previous studies have indicated a role for reactive oxygen species (ROS) in the UV-B-induced down-regulation of the photosynthetic genes and up-regulation of PR-1 genes. Here we have shown that ROS are also required for the UV-B-induced up-regulation of PDF1·2 genes. The results indicate that the effects of UV-B on the three sets of genes are mediated through three distinct signal transduction pathways which are similar, but not identical, to pathways initiated in response to pathogen infection. In addition, the increased sensitivity of both jar1 and etr1-1 mutants to UV-B radiation, as compared with wild-type plants, indicated that intact JA and ethylene signal pathways are required for defence against UV-B damage.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
In this paper, we have investigated the role of ROS, JA and ethylene in signal pathways leading to changes in gene expression in Arabidopsis in response to UV-B exposure, by making use of mutant plants which are insensitive to JA (jar1, Staswick, Su & Howell 1992) or ethylene (etr1-1;Bleecker et al. 1988 ). We have demonstrated that UV-B leads to increases in two defence-associated genes, PDF1·2 and PR-1, and decreases in photosynthetic genes through three separate pathways with ROS as a component common to all three. In addition, we have shown that, in Arabidopsis, some protection against UV-B radiation is conferred by defence mechanisms which require both JA and ethylene-dependent signalling pathways.
MATERIALS AND METHODS
Plant material, experimental conditions and feeding experiments
Arabidopsis thaliana (Columbia (MGH)), jar-1 ( Staswick et al. 1992 ) or etr1-1 ( Bleecker et al. 1988 ) were used in experiments as indicated in the text. Seeds were treated for 3 d at 4 °C after sowing on a compost : sand : vermiculite mixture and then transferred to a Sanyo (Leicester, UK) fitotron with 12 h light (22 °C), 12 h dark (16 °C) cycles at 70% humidity. At 14 d the seedlings were transplanted into individual pots and grown for a further 14 d. Incident irradiation was provided by Philips (Croydon, UK) warm white fluorescent tubes giving a fluence rate of 150 μmol m−2 s−1 photosynthetically active radiation (400–700 nm). Half the plants were then given supplementary UV-B radiation from four UV lamps (Philips TL 12, 40 W) during the photoperiod. The fluence rate between 280 and 320 was 3·2 μmol m−2 s−1. The UV lamps were covered with cellulose acetate sheets, which were changed daily, to exclude UV radiation below 290 nm. The plants in the other cabinet (controls) were subjected to the same treatment but the UV lamps were covered with Mylar film to exclude radiation below 320 nm.
For feeding experiments, 3-amino-1,2,4-triazole (3-AT: 8 mol m−3), ascorbic acid (AsA: 10 mol m−3) and jasmonic acid (JA: 1 mmol m−3) were applied, at the concentrations indicated, as a spray until runoff. The compounds were applied 1 h before the beginning of the light period. For ethylene treatment the plants were enclosed in an airtight jar and treated with 100 μL L−1 of gaseous ethylene for the times indicated in the text. Ethylene concentrations were confirmed by gas chromatography.
Each experiment was repeated at least three times with similar results. Samples were taken at random from different parts of the cabinets, 6 h into each photoperiod (unless otherwise stated) and at least three whole plants were taken per sample for analysis. For determination of the sensitivity of plants to UV-B irradiation ( Fig. 6) the plants were exposed to supplementary UV-B for up to 12 h. At hourly intervals three to five plants were removed and left to recover under constant white light. The plants were monitored daily for symptoms of stress (glazing or bronzing of the upper surface of the leaves). The experiment was repeated twice with similar results. The exposure times shown are those which best illustrated differences between wild-type and mutants and the plants shown are representative of the appearance of the majority of the plants within an exposure time.
Purification of total RNA, northern blotting and hybridization
Arabidopsis rosettes (0·5 g) were frozen in liquid nitrogen and stored at –70 °C. Total RNA was extracted as described in Jordan et al. (1991) and stored at –70 °C. The RNA was separated on a 1·5% agarose gel containing 6% (v/v) formaldehyde in 2 × MOPS ( Maniatis, Fritsch & Sambrook 1982). Each gel was checked under UV light to ensure both RNA integrity and that equal amounts of RNA were loaded in each lane (data not shown). The RNA blotting, pre-hybridization, hybridization and washes were as described in A.-H. -Mackerness, Thomas & Jordan (1997). Autoradiography of the filters was at –70 °C using Fuji (Felsted, UK) RX film with two intensifying screens. Blots presented are representative of three or more independent experiments. Relative amounts of radioactivity bound to specific bands were quantified using a Phosphorimager SI (Molecular Dynamics Ltd, Bucks., UK). Loading control analysis of each blot was carried out by hybridization using an 18S rRNA probe from pea.
The pooled aerial portions (approximately 1 g) of four Arabidopsis plants were extracted by first homogenizing in 25 cm3 of cold (4 °C) 80% methanol containing 20 mg L−1 of butylated hydroxytoluene and then stirring the homogenate overnight at 4 °C. The purification procedures were based on those of Finch-Savage, Blake & Clay (1996) with the following modifications. The extracts were filtered, the residues washed twice with methanol and the aqueous extracts partitioned three times with equal volumes of hexane. An equal volume of 0·4 M sodium acetate buffer pH 4·0 was then added and the extracts partitioned three times with equal volumes of dichloromethane. The organic layers were combined and dried over anhydrous magnesium sulphate, filtered and passed through two ‘Sep-Pak’ (Waters Associates, Watford, UK) silica cartridges (pre-washed with diethyl ether : methanol (3 : 2 v : v) and dichloromethane). The cartridges were washed with dichloromethane and the JA recovered in diethyl ether : methanol (3 : 2 v : v). The methyl esters were prepared by adding excess ethereal diazomethane; after 30 min the diazomethane was removed under a stream of dry, oxygen-free nitrogen gas. The extracts were reduced to dryness using a ‘Speed-Vac’ centrifugal vacuum concentrator (Savant Instruments, Bierbeek, Belgium), redissolved in dichloromethane and line-loaded onto pre-washed (methanol) 20 cm × 20 cm, 0·2 mm silica gel, aluminium-backed thin layer chromatography (TLC) plates (Merck 5553, Lutterworth, UK). Standards of authentic methyl jasmonate (Firmenich, Geneva, Switzerland) and methyl dihydrojasmonate were applied and the plates allowed to develop for 25 cm in the vertical direction using dichloromethane : acetone (95 : 5 v : v) as solvent. A 2 cm band of silica was removed from the lanes containing the samples that corresponded to the relative front (Rf) of the standards. The silica gel was packed into glass-wool-plugged Pasteur pipettes and eluted with acetone, reduced to dryness under dry oxygen-free nitrogen gas and redissolved in hexane : ethyl acetate (9 : 1). The methyl esters of JA was analysed using gas chromatography-mass spectrometry, selected ion monitoring system operating under the conditions described by Finch-Savage et al. (1996) . All samples were spiked with 100 ng of (±) 9,10 dihydrojasmonic acid (DHJA) during the first extraction stage to calculate the efficiency of recovery. DHJA was obtained by the alkaline hydrolysis of (±) methyl dihydrojasmonate (Firmenich) as described by Fan et al. (1997) .
Gas analysis was carried out on three plants in 2 L vessels sealed with a rubber serum cap. Plants were placed in the vessels for 24 h after treatment with or without UV-B for the times indicated in the text. Ethylene was measured from 1 mL of gas from the headspace of the vessel. The concentration was determined on a gas chromatograph (GC-8 A, Shimadzu, Dyson Instruments, Tyne & Wear, UK) equipped with a flame ionization detector and a column packed with Poropack Q. The temperature was maintained at 150 and 120 °C for injector/detector and columns, respectively. Identification and quantification was carried out by comparison with the retention times and peak areas of standard samples. Each sample was measured at least three times and results presented are from three independent experiments.
Response to UV-B radiation in wild-type Arabidopsis
In order to determine the effects of UV-B on the photosynthetic, PR-1 and PDF1·2 transcript levels, Arabidopsis plants were treated with supplementary UV-B for up to 30 h and RNA levels analysed by Northern blot analysis. Supplementary UV-B treatment led to a rapid reduction in the level of both the photosynthetic transcripts with effects of UV-B being more severe for the nuclear-encoded Lhcb transcripts than the chloroplast-encoded psbA transcripts throughout the sampling period ( Fig. 1). The circadian rhythm of Lhcb was apparent, hence the drop at 12 h, which was greater in the UV-B-treated samples than in control samples ( Fig. 1).
In contrast to the photosynthetic genes, there was an increase in the level of PR-1 and defencin transcripts, PDF1·2 ( Fig. 1). Transcripts of PR-1 increased within 6 h to approximately 266% of control values whereas the increase in PDF1·2 transcripts was not detectable until 9 h, by which time levels were at approximately 150% of control values. These smaller rises in transcript levels were followed by larger increases at the subsequent sampling times ( Fig. 1).
Responses of wild-type Arabidopsis plants treated with chemical inducers
In order to assess the ability of ROS, JA and ethylene to mimic effects of UV-B on the four transcripts, wild-type plants were treated with 3-AT, JA, ethylene and a combination of JA and ethylene. Effects on transcript levels were determined by northern blot analysis ( Fig. 2). 3-AT, an oxidant, was found to mimic effects of UV-B on all four genes, leading to a decrease in both photosynthetic transcripts and an increase in PDF1·2 and PR-1 transcripts. Both JA and ethylene separately resulted in a greater reduction of Lhcb transcripts but did not lead to a decrease in psbA transcript levels or increases in the level of PR-1 transcripts. Both of these compounds, provided separately, also resulted in a small increase in PDF1·2 transcript levels, but treatment with ethylene and JA simultaneously led to a synergistic increase in PDF1·2 transcripts ( Fig. 2).
Previous studies have illustrated that spraying plants with antioxidants prior to UV-B treatment can block the increase in PR transcripts and the decrease in photosynthetic transcripts (A.-H. -Mackerness et al. 1998 ; Surplus et al. 1998 ). This was taken as an indication that ROS were involved in the pathway leading to changes in the level of these transcripts in response to UV-B irradiation. In order to assess the role of ROS in induction of PDF1·2 transcript levels, we determined the effect of ascorbic acid (AsA) pre-treatment on the effects of UV-B on this transcript ( Fig. 3). Consistent with the involvement of ROS in regulation of PDF1·2 gene expression in response to UV-B, the increase in transcripts was blocked by pre-treatment with AsA ( Fig. 3).
Changes in jar1 and etr1-1 mutants
Determination of effects of various chemicals on transcript levels provides an indication of the ability of that chemical to affect gene expression. However, it does not establish whether that compound is an essential component in the signalling pathway in response to a particular stimulus. Therefore, in order to clarify the role of JA and ethylene in regulation of photosynthetic, PR-1 and PDF1·2 genes in response to UV-B radiation, we investigated expression of these genes in the ethylene insensitive mutant, etr1-1, and the JA-insensitive mutant, jar1. The effects of supplementary UV-B on Lhcb and psbA transcripts in both the etr1-1 and jar1 mutants ( Fig. 4) were similar to those observed for the wild-type plants ( Fig. 1) with a comparable reduction in the level of both transcripts in the mutant and wild-type plants ( Table 1). However, in both the etr1-1 and jar1 plants, the UV-B-induced up-regulation of PDF1·2 ( Fig. 4) was considerably reduced compared with wild-type plants ( Fig. 1) indicating a role for both JA and ethylene in up-regulation of these genes in response to UV-B exposure ( Table 1). Interestingly, the up-regulation of PR-1 transcript was also compromised in the etr1-1 plants but not in the jar1 plants ( Fig. 4) compared with wild-type plants ( Fig. 1) indicating a role for ethylene in regulation of PR-1 transcripts in response to UV-B exposure ( Table 1).
Table 1. . Quantitative changes in Lhcb, psbA, PR-1 and PDF1·2 transcript levels in wild-type, jar1 and etr1-1 plants exposed to supplementary UV-B radiation. The amount of radioactivity was quantified using a phosphorimager; the data shown have been corrected for loading differences by using the counts obtained with 18S rRNA. Values are means (SE) of three to four independent experiments
Quantitative analysis of jasmonic acid and ethylene levels following UV-B treatment
In order to determine whether UV-B exposure resulted in increases in JA and emissions of ethylene in Arabidopsis, the levels of JA and ethylene in wild-type plants were measured after exposure to supplementary UV-B. In comparison with control wild-type plants, UV-B exposure resulted in a significant rise in both JA and ethylene levels ( Table 2). The levels of ethylene began to rise after 1 h while JA levels had increased after 3 h of UV-B exposure. The level of both hormones continued to rise throughout subsequent sampling periods. The levels of both JA and ethylene remained relatively constant in the control plants ( Table 2).
Table 2. . Changes in jasmonic acid and ethylene levels in response to supplementary UV-B treatment. 4 week-old Arabidopsis wild-type plants were grown in the presence or absence of supplementary UV-B. JA and ethylene levels were determined from the rosettes. Results presented are from at least two independent experiments ± S.E.
Response of photosynthetic, PDF1·2 and PR-1 transcripts in Arabidopsis jar1 and etr1-1 plants treated with chemical inducers
In order to assess the relationship between ROS, JA and ethylene in regulation of PDF1·2 and PR-1 in our system, we treated the jar1 and etr1-1 mutants with ROS, JA and ethylene and determined their effect on transcript levels. Treating jar1 and etr1-1 with JA and ethylene, respectively, did not lead to a significant effect on any of the genes studied, indicating that these mutants are sufficiently blocked in these pathways for the purpose of these experiments ( Fig. 5).
There was no detectable increase in PDF1·2 transcripts in the jar1 plants treated with ethylene or etr1-1 plants treated with JA ( Fig. 5), further confirming our previous results that both compounds are required for full induction of PDF1·2 ( Fig. 2). In addition, 3-AT did not lead to an increase in PDF1·2 in either mutant indicating that ROS must lie up-stream of both JA and ethylene in regulation of PDF1·2 transcript levels.
As in wild-type plants ( Fig. 2), no induction of PR-1 transcripts was observed in either mutant sprayed with ethylene or JA ( Fig. 5). However, unlike wild-type plants, 3-AT was unable to increase PR-1 transcripts in the etr1-1 plants ( Fig. 5), which is similar to the response seen on UV-B treatment ( Fig. 4). Thus neither UV-B nor ROS can up-regulate PR-1 transcripts in the absence of ethylene perception ( Figs 4 & 5). However, ethylene alone is not sufficient to induce PR-1 expression ( Fig. 2).
Sensitivity of wild-type, jar1 and etr1-1 mutants to UV-B exposure
In order to assess the importance of JA and ethylene signalling pathways in defence against UV-B stress, we determined the exposure time at which UV-B pre-treatment resulted in visible tissue damage. The plants were treated with the same fluence of UV-B but for increasing lengths of exposure and then left to recover under continuous light. We assessed the sensitivity of the two mutants, as compared with wild-type plants, by determining the minimum exposure required for the plant to show symptoms of stress. Figure 6 illustrates the effect of varying UV-B exposure time (0, 4, 6, 8 and 12 h) on wild-type, jar1 and etr1-1 plants. No signs of stress were evident while plants were treated with supplementary UV-B (data not shown), but after 2 d of recovery there were differences in visible symptoms between the wild-type and mutant plants. Figure 6 illustrates plants after 3 d recovery at which time the differences in symptoms were most apparent. Both etr1-1 and jar1 plants that had been treated with 4 h of UV-B showed signs of UV-B stress (glazing and bronzing of the upper surface of the leaves), with the effects clearly visible in plants treated with 6 h of UV-B. In contrast, the wild-type plants that had been treated with 4 h of UV-B showed no sign of damage and minimal symptoms were evident on plants treated with 6 h of UV-B ( Fig. 6). Progressively longer exposure times further illustrated the greater tolerance of wild-type as compared to mutants with effects being less pronounced in wild-type treated with 8 and 12 h of UV-B as compared with the mutants at these exposure times.
At present, little is known about the mechanisms by which plants perceive UV-B radiation and the signal transduction mechanisms by which UV-B regulates gene expression. We have demonstrated previously that UV-B exposure leads to an increase in ROS and SA levels in Arabidopsis and that the regulation of the photosynthetic genes, Lhcb and psbA, and the pathogenesis-related gene, PR-1, by UV-B radiation are mediated via two separate pathways. Both pathways are ROS-dependent but regulation of PR-1 also requires SA accumulation ( Surplus et al. 1998 ). In this paper we have extended our previous studies and investigated the role of JA, ethylene and ROS in the regulation of the photosynthetic genes, PR-1 and a defencin gene, PDF1·2. Our results are summarized in Fig. 7.
Our data show that as with PR-1 ( Fig. 1 and Surplus et al. 1998 ), there was a similar increase in PDF1·2 transcript levels in response to UV-B ( Fig. 1) as previously reported in response to pathogens ( Penninckx et al. 1996 , 1998). In contrast, confirming earlier studies ( Surplus et al. 1998 ; Jordan, James & A.-H.-Mackerness 1998), UV-B exposure led to a decrease in transcripts encoding two photosynthetic proteins, the chloroplast encoded psbA and the nuclear encoded Lhcb ( Fig. 1).
The UV-B-induced up-regulation of PDF1·2 transcript was considerably reduced in both the jar1 and etr1-1 plants ( Fig. 4). In addition, feeding with JA and ethylene simultaneously resulted in a higher induction of this gene than the two compounds provided separately ( Fig. 2). Thus, together, these results strongly indicate that increases in both JA and ethylene, concomitantly, are essential for the maximal induction of the PDF1·2 gene by UV-B radiation ( Fig. 7). Similar synergistic effect of JA and ethylene on the induction of PDF1·2 in Arabidopsis ( Penninckx et al. 1998 ) and an osmotin gene in tobacco ( Xu et al. 1994 ) following pathogen infection has been recently reported.
Previous studies have indicated that, in response to pathogens, ethylene is not required for chemically or biologically induced acquired resistance, and thus increases in PR transcripts ( Lawton et al. 1994 , 1995), but serves as a modulator of tissue responsiveness to low levels of SA ( Lawton et al. 1994 ). However, interestingly UV-B exposure did not result in a significant rise in PR-1 transcripts in the etr1-1 plants ( Fig. 4), indicating an important role for ethylene in the up-regulation of PR-1 transcripts in response to UV-B exposure. Therefore, although ethylene alone cannot induce PR-1 transcripts ( Fig. 2), unlike in pathogen responses, it must play a critical role in increases in PR-1 transcripts following UV-B treatment ( Fig. 4). In addition, as SA can lead to increases in PR-1 transcripts in the etr1-1 plants ( Lawton et al. 1994 ; our data not shown), ethylene is likely to lie up-stream of SA in the signal pathway involved in the regulation of PR-1 transcripts ( Fig. 7). The absence of the absolute requirement for ethylene in PR-1 induction in pathogens thus indicates that the up-regulation of PR-1 by UV-B radiation and pathogen infection is not through identical signal pathways.
Lhcb transcripts, but not psbA transcript levels, were reduced in response to JA feeding ( Fig. 2) as has been previously reported in barley (reviewed in Reinbothe et al. 1994 ). We are not aware of any studies on the effects of ethylene on photosynthetic transcripts but similar to JA, ethylene only resulted in a decrease in Lhcb transcripts ( Fig. 2). However, the effect of UV-B on photosynthetic genes was similar in the wild-type and jar 1 plants ( Fig. 4), indicating that, unlike responses to wounding (reviewed in Reinbothe et al. 1994 ), JA is not required for the regulation of Lhcb transcript levels by UV-B radiation. Similarly, the effect of UV-B on these transcripts was not affected in the etr1-1 mutants ( Fig. 4) indicating that ethylene is also not involved in this pathway. Little information is available on the signal components important in regulation of photosynthetic genes in response to UV-B radiation.
Many stresses including pathogen infection, ozone treatment, wounding and UV-B radiation have been shown to stimulate the generation of ROS ( Murphy & Huerta 1990; Arnotts & Murphy 1991; Rao et al. 1996 ; Dai et al. 1997 ). Recent studies have demonstrated the importance of ROS as signalling components involved in the regulation of PR-1 and photosynthetic genes in response to UV-B exposure ( Green & Fluhr 1995; A.-H. -Mackerness et al. 1998 ; Surplus et al. 1998 ). Using a similar approach, we have shown that ROS must play a role in up-regulation of PDF1·2 gene in response to supplementary UV-B ( Figs 2 & 3) as previously reported in response to pathogen infection ( Penninckx et al. 1996 , 1998). In addition, as ROS (3-AT) were unable to increase PDF1·2 transcript levels in either jar1 or etr1-1 mutants ( Fig. 5) then ROS must lie upstream of both JA and ethylene in this signal pathway. Using a similar approach, Penninckx et al. (1996 , 1998) identified the same pathway in experiments directed at determining the role of ROS, JA and ethylene in plant disease resistance. Therefore, the up-regulation of PDF1·2 by UV-B radiation is through a similar, if not identical, signal pathway that is employed in response to pathogen infection ( Fig. 7).
Previous studies have indicated a role for ROS in up-regulation of PR-1 in response to UV-B exposure and pathogen infection ( Green & Fluhr 1995; Surplus et al. 1998 ). In contrast to wild-type plants ( Fig. 2), spraying etr1-1 plants with 3-AT did not lead to an increase in PR-1 transcripts ( Fig. 5) indicating that ethylene must lie down-stream of ROS in this signal pathway ( Fig. 7). To our knowledge no such spraying experiment has been carried out using etr1-1, but the results presented in this paper ( Figs 2, 4 & 5) and in Surplus et al. (1998) indicate that ROS generation, and subsequent SA accumulation, requires ethylene perception in order to lead to PR-1 induction in response to UV-B exposure ( Fig. 7). This pathway, due to the absolute requirement for ethylene, is thus similar, but not identical, to that involved in PR-1 up-regulation in response to pathogen infection ( Lawton et al. 1994 , 1995).
Our results have shown that both JA and ethylene are components of signal pathways leading to changes in gene expression in response to UV-B radiation ( Fig. 7). These results, however, do not indicate the importance of these compounds in defence against UV-B radiation. Thus we determined the UV-B exposure time which resulted in the appearance of visible symptoms in the two mutants, as compared with the wild-type plants, as an indicator of sensitivity to UV-B radiation. Both jar1 and etr1-1 plants showed symptoms of UV-B damage at earlier exposure times in comparison with wild-type plants ( Fig. 6) indicating that, visibly, these plants are more susceptible to UV-B-induced tissue damage. Both mutants have been previously shown to be susceptible to opportunistic pathogens indicating that both are affected in signalling events leading to disease resistance ( Knoester et al. 1998 ; Pieterse et al. 1998 ) but no such studies have been carried out with respect to UV-B sensitivity. Our results indicate that as in pathogen defence, intact JA and ethylene signal pathways are required for maximum defence against UV-B-induced tissue damage. However, our data do not prove that the effect is determined by the particular set of proteins that we have studied as many yet unknown effector molecules may also be involved.
It is clear from Fig. 7 that although many components of the pathways leading to changes in gene expression in response to UV-B and pathogens are similar or identical (up-regulation of PDF1·2), there are also pathways which are distinct (up-regulation of PR-1). The existence of distinct pathways in response to UV-B exposure strongly indicate that effects of UV-B on gene expression are unlikely to be due to non-specific damage but involve, as yet unidentified, UV-B photoreceptor(s). Isolation and characterization of these photoreceptor(s) will greatly increase our understanding of how plants perceive and thus respond to UV-B radiation and stress.
The research was supported by BBSRC competitive strategic grant at HRI. We are grateful to the Nottingham Arabidopsis Stock Centre, University Park, Nottingham for providing the etr1-1 and jar1 mutants, to Novartis Corporation, 3054 Cornwallis Road, Research Triangle Park, North Carolina, USA for supplying the PR-1 cDNA and Firmenich, Geneva, Switzerland, for the generous gifts of methyl jasmonate and methyl dihydrojasmonate. We would also like to thank Steve Robertson for running and maintenance of the Sanyo cabinets, Lynette Shakespeare for help with the ethylene analysis, Mike Smith for prepara-tion of the figures and finally to Drs Richard Napier and Mahmut Tör for critical reading of this manuscript and many helpful comments.