TY - JOUR
AU - Hansson, Mats
AB - Abstract The barley (Hordeum vulgare L.) mutants fch2 and clo-f2 comprise an allelic group of 14 Chl b-deficient lines. The genetic map position of fch2 corresponds to the physical map position of the gene encoding chlorophyllide a oxygenase. This enzyme converts chlorophyllide a to chlorophyllide b and it is essential for Chl b biosynthesis. The fch2 and clo-f2 barley lines were shown to be mutated in the gene for chlorophyllide a oxygenase. A five-base insertion was found in fch2 and base deletions in clo-f2.101, clo-f2.105, clo-f2.2800 and clo-f2.3613. In clo-f2.105 and clo-f2.108, nonsense base exchanges were discovered. All of these mutations led to a premature stop of translation and none of the mutants formed Chl b. The mutant clo-f2.2807 was transcript deficient and formed no Chl b. Missense mutations in clo-f2.102 (leading to the amino acid exchange D495N) and clo-f2.103 (G280D) resulted in a total lack of Chl b, whereas in the missense mutants clo-f2.107 (P419L), clo-f2.109 (A94T), clo-f2.122 (C320Y), clo-f2.123 (A94T), clo-f2.133 (A376V) and clo-f2.181 (L373F) intermediate contents of Chl b were determined. The missense mutations affect conserved residues, and their effect on chlorophyllide a oxygenase is discussed. The mutations in clo-f2.102, clo-f2.103, clo-f2.133 and clo-f2.181 may influence electron transfer as illustrated in the active site of a structural model protein. The changes in clo-f2.107, clo-f2.109, clo-f2.122 and clo-f2.123 may lead to Chlb deficiency by interfering with the regulation of chlorophyllide a oxygenase. The correlation of mutations and phenotypes strongly supports that the barley locus fch2 encodes chlorophyllide a oxygenase. The nucleotide sequences reported in this paper have been submitted to GenBank under accession numbers JQ627619 (chlorophyllide a oxygenase in Sejets Tron), JQ627620 (chlorophyllide a oxygenase in Svalöfs Bonus) and JQ627621 (chlorophyllide a oxygenase in Ackermanns Donaria). Introduction Leaves with aberrant Chl pigmentation are easily spotted in large groups of plants even at an early stage of growth. Thus, Chl mutation frequencies have been used to evaluate the success rates of various mutagenic treatments (Lundqvist 1992). Many of those Chl mutants have been deposited in mutant stock collections of various species (e.g. Arabidopsis Biological Resource Center; National Small Grains Collection; Maize Genetics Cooperation Stock Center; Nordic Genetic Resource Center–NordGen). In barley (Hordeum vulgare L.), the Chl-related mutants kept at NordGen and the Carlsberg Laboratory have been assigned to 105 different loci (Simpson and von Wettstein 1992). The genetic identity of the loci has been clarified at the DNA level in five cases only: Xantha-f, -g and -h encode the three subunits of magnesium chelatase (Hansson et al. 1999, Olsson et al. 2004, Axelsson et al.2006); Xantha-l encodes a component of the magnesium protoporphyrin IX monomethyl ester cyclase (Rzeznicka et al. 2005); and Tigrina-d encodes a tetrapyrrole biosynthesis regulatory component (Lee et al. 2003). The barley mutant 28-3398/fch2.c (Franckowiak 1997) has a light-green chlorina phenotype. It is Chl b deficient and survives as a homozygous mutant (Highkin 1950, Highkin and Frenkel 1952). The mutant occurred spontaneously in a cross of the cultivars Moister, CI 2799 (Lion × Manchuria), and California Coast, CI 6115 (Tsuchiya 1972). The locus of the monofactorial, recessive trait was designated chlorina seedling 2 and initially given the symbol f2 (Tsuchiya and Robertson 1971). Later, the symbol was changed to fch2 and the locus was mapped to the long arm of barley chromosome 3H (3HL), about 7.7 cM distal to the locus curly 2 (cur2; Franckowiak, 1997). Additional alleles of fch2 are light green 5 (lg5), which was designated fch2.y (Takahashi et al. 1975, Franckowiak 1997), and the three mutants 2800, 2807 and 3613 that were induced by X-ray treatment of seeds of the cultivar Ackermanns Donaria (Machold et al. 1977). Ten more allelic mutants were induced by sodium azide treatment and referred to as clo-f2 (Simpson et al. 1985). The mutants clo-f2.101, clo-f2.102, clo-f2.103, clo-f2.105, clo-f2.107, clo-f2.108 and clo-f2.109 were induced in the cultivar Sejets Tron, whereas clo-f2.122, clo-f2.123 and clo-f2.133 were induced in the cultivar Svalöfs Bonus. The three mutants derived from Donaria were included in this nomenclature: clo-f2.2800, clo-f2.2807 and clo-f2.3613 (Simpson et al. 1985). Another mutant induced in Sejets Tron was clo-106, which was reported to have a reduced Chl b content, yet it was not an allele of fch2 (Simpson et al. 1985). The fch2 and clo-f2 lines are characterized by a reduced amount or total lack of Chl b (Highkin 1950, Highkin and Frenkel 1952, Apel 1967, Sagromsky 1974, Simpson et al. 1985). Absence of Chl b results in a different composition of the light-harvesting apparatus, which is the reason why many studies have been conducted with fch2 (see, for example, Goodchild et al. 1966, Thornber and Highkin 1974, Henriques and Park 1975, Hiller and Raison 1980, Mullet et al. 1980, Bhalla and Bennett 1987, Harrison et al. 1993, Preiss and Thornber 1995, Bossmann et al. 1997, Havaux and Tardy 1997, Rudoi and Shcherbakov 1998, Saito et al. 2010, Tyutereva and Voitsekhovskaja 2011a, Tyutereva and Voitsekhovskaja 2011b). Despite the detailed characterization of the fch2 and clo-f2 mutants, the identity of the fch2 locus has not yet been described. Chl b is formed by the enzyme chlorophyllide a oxygenase, which converts the 7-methyl group of chlorophyllide a to the 7-formyl group of chlorophyllide b (Tanaka et al. 1998, Espineda et al. 1999, Oster et al. 2000). Alignment of the chlorophyllide a oxygenase sequences of Arabidopsis thaliana, Oryza sativa and the prochlorophyte Prochlorothirx hollandica led to the assignment of three domains in the polypeptide. Domain A is regulatory, B is a linker, and C is the catalytic domain (Nagata et al. 2004, Sakuraba et al. 2007, Yamasato et al. 2008). Domain A contains a degradation signal sequence (degron) which has been suggested to be involved in the regulation of chlorophyllide a oxygenase activity by interaction with a protease (Sakuraba et al. 2009). The C domain contains three conserved sequence motifs: a Rieske FeS cluster-binding site, a mononuclear non-heme iron-binding site and a conserved sequence unique to chlorophyllide a oxygenase polypeptides (Espineda et al. 1999). The oxygen atom incorporated in the 7-formyl group of chlorophyllide b originates from O2 (Schneegurt and Beale 1992) and, to activate O2 for the oxygenation reaction, electron transfer from a reduced donor through the FeS cluster and the mononuclear iron has been discussed (Tanaka et al. 1998). This electron transfer sequence suggests an essential functional and structural involvement of the conserved metal-binding sites of chlorophyllide a oxygenase in the synthesis of Chl b. The Chl b-deficient phenotype of the fch2 and clo-f2 mutants and genetic mapping data suggested that the locus fch2 and the barley gene encoding chlorophyllide a oxygenase (HvCAO) are identical. To test this and to identify the barley fch2 locus, HvCAO cDNA in fch2 and the 13 allelic clo-f2 mutants was amplified and sequenced. This revealed severe mutations in HvCAO in the fch2 and clo-f2 barley genetic stocks. We suggest structural and functional consequences of the missense mutations on chlorophyllide a oxygenase by comparison with a model structure. The findings provide strong evidence that fch2 does encode barley chlorophyllide a oxygenase. Results The fch2 and clo-f2 phenotype Barley mutated in the locus fch2 has a light green chlorina phenotype due to the reduction in Chl b. The color varies between alleles (Fig. 1). Homozygous plants are viable, but they are delayed in their development (Highkin 1950, Franckowiak 1997). The pigment contents of the fch2 and clo-f2 plants are given in Table 1. Fig. 1 View largeDownload slide Wild-type Tron, fch2 and clo-f2 mutants. The two leaves to the left are Tron, the two to the right are (A) fch2, (B) clo-f2.108, (C) clo-f2.109 and (D) clo-f2.2807. Fig. 1 View largeDownload slide Wild-type Tron, fch2 and clo-f2 mutants. The two leaves to the left are Tron, the two to the right are (A) fch2, (B) clo-f2.108, (C) clo-f2.109 and (D) clo-f2.2807. Table 1 Pigment content in wild-type, fch2 and clo-f2 barley Line  Chl a (nmol g−1 fresh leaf)  Chl b (nmol g−1 fresh leaf)  Chl a/b ratio  Chl a as pecentage of Ch a in Tron  Mother cultivar  Tron  613 (±11)  161 (±6)  3.8  100  –  Tron*  567  152  3.7  92  –  Bonus  640 (±16)  171 (±6)  3.7  104  –  Donaria  602  162  3.7  98  –  fch2  386  ND  –  63  (Lion × Manchuria) × California Coast  clo-f2.101  529  ND  –  86  Tron  clo-f2.102  343  ND  –  56  Tron  clo-f2.103  233  ND  –  38  Tron  clo-f2.105  385  ND  –  63  Tron  clo-f2.107  816  142  5.7  133  Tron  clo-f2.108  616  ND  –  100  Tron  clo-f2.109  563  59  9.5  92  Tron  clo-f2.122*  342  16  21  60  Bonus  clo-f2.123  185  20  9.3  30  Bonus  clo-f2.133  571  67  8.5  93  Bonus  clo-f2.181  398  26  15  65  Tron  clo-f2.2800*  459  ND  –  81  Donaria  clo-f2.2807  464  ND  –  76  Donaria  clo-f2.3613  420  ND  –  69  Donaria  clo-106  452  131  3.4  74  Tron  Line  Chl a (nmol g−1 fresh leaf)  Chl b (nmol g−1 fresh leaf)  Chl a/b ratio  Chl a as pecentage of Ch a in Tron  Mother cultivar  Tron  613 (±11)  161 (±6)  3.8  100  –  Tron*  567  152  3.7  92  –  Bonus  640 (±16)  171 (±6)  3.7  104  –  Donaria  602  162  3.7  98  –  fch2  386  ND  –  63  (Lion × Manchuria) × California Coast  clo-f2.101  529  ND  –  86  Tron  clo-f2.102  343  ND  –  56  Tron  clo-f2.103  233  ND  –  38  Tron  clo-f2.105  385  ND  –  63  Tron  clo-f2.107  816  142  5.7  133  Tron  clo-f2.108  616  ND  –  100  Tron  clo-f2.109  563  59  9.5  92  Tron  clo-f2.122*  342  16  21  60  Bonus  clo-f2.123  185  20  9.3  30  Bonus  clo-f2.133  571  67  8.5  93  Bonus  clo-f2.181  398  26  15  65  Tron  clo-f2.2800*  459  ND  –  81  Donaria  clo-f2.2807  464  ND  –  76  Donaria  clo-f2.3613  420  ND  –  69  Donaria  clo-106  452  131  3.4  74  Tron  The measurements of Tron and Bonus were done in triplicate; the standard deviations are given. The other values are based on single or duplicate measurements. An asterisk denotes leaves grown in ambient light in a greenhouse. ND, not detected. View Large The analysis of the pigment content of the individual fch2 and clo-f2 lines (Table 1) detected no Chl b in the mutants fch2, clo-f2.101, clo-f2.102, clo-f2.103, clo-f2.105, clo-f2.108, clo-f2.2800, clo-f2.2807 and clo-f2.3613. The wild type had a Chl a to Chl b (Chl a/b) ratio of 3.7–3.8. The weakest Chl b-deficient phenotype was that of mutant clo-f2.107, which had a Chl a/b ratio of 5.7. In the lines clo-f2.109, clo-f2.122, clo-f2.123, clo-f2.133 and clo-f2.181 the ratio was 9.5, 21, 9.3, 8.5 and 15, respectively. The Chl a content of the Chl b-deficient lines compared with wild-type barley varied between 30 and 133% (clo-f2.123 and clo-f2.107, respectively), while the majority of lines had between 56 and 93% Chl a. The pigment analyses showed that the mutations in the fch2/clo-f2-lines specifically affect the biosynthesis of Chl b, while the ability per se to synthesize Chl was not affected in any of the lines. As chlorophyllide a oxygenase is the enzyme that catalyzes the formation of Chl b, it was hypothesized that locus fch2 encodes barley chlorophyllide a oxygenase. Mapping identifies HvCAO as a candidate gene for fch2 In the GrainGenes Hordeum-Bins-3H map, fch2 is listed at position 166.8 cM on chromosome arm 3HL. The distance to curly2 (cur2 at 155.5 cM) is somewhat greater than the 7.7 cM reported by Franckowiak (1997). Recently, Druka et al. (2011) reported on a collection of 881 near-isogenic barley lines representing mutant alleles in various loci, which were genotyped using 3,072 single nucleotide polymorphisms (SNPs). Line BW358 contains the original fch2 mutation (22-3398/fch2.c) and resulted from six backcrosses to the barley cultivar Bowman. In BW358, four SNP markers co-segregated with the fch2 phenotype. Two markers are located at position 167.8 cM (GrainGenes map Hordeum-OPA-2009-3H, Pilot OPA markers 1_1410 and 1_0681), and one is located at 168.4 cM (1_0694). The fourth marker, 1_1411, was not mapped. Interestingly, the marker 1_0629, which is also at 167.8 cM in this map, is the HarvEST barley unigene HAR35_2495 that has been annotated as chlorophyllide a oxygenase (HvCAO). This finding suggested that fch2 and HvCAO might be identical. To corroborate the location of HvCAO at the position of the barley genome to which fch2 was mapped, the synteny of barley 3HL to the chromosomes of rice and Brachypodium distachyon was analyzed (Fig. 2). In rice, the genes for two isoforms of chlorophyllide a oxygenase are located on chromosome 10 (LOC_Os10g41760.1 and LOC_Os10g41780.1), while the orthologs to the barley markers flanking unigene HAR35_2495 and the position of fch2 are found on rice chromosome 1 (Fig. 2). On B. distachyon chromosome 2, however, the order of open reading frames including that annotated chlorophyllide a oxygenase is the same as the orthologous markers in barley flanking unigene HAR35_2495 and the position of fch2. This conserved synteny between the genetic map of barley chromosome 3HL and the physical map of B. distachyon chromosome 2 supported the strong connection between fch2 and HvCAO. Fig. 2 View largeDownload slide Synteny of B. distachyon, barley and rice in the region containing barley chlorophyllide a oxygenase (HAR35_2495). Chlorophyllide a oxygenase and flanking markers are syntenic on barley chromosome 3HL and B. distachyon chromosome 2. In contrast, the arrangement of loci between rice and barley is not syntenic. The loci annotated or shown to be chlorophyllide a oxygenase are marked in gray. Fig. 2 View largeDownload slide Synteny of B. distachyon, barley and rice in the region containing barley chlorophyllide a oxygenase (HAR35_2495). Chlorophyllide a oxygenase and flanking markers are syntenic on barley chromosome 3HL and B. distachyon chromosome 2. In contrast, the arrangement of loci between rice and barley is not syntenic. The loci annotated or shown to be chlorophyllide a oxygenase are marked in gray. Cloning and characterization of HvCAO In Arabidopsis, AtCAO is a gene of 3,273 bp that encodes a protein of 536 amino acid residues (GenBank AF177200 encoding AAD54323.1; Espineda et al. 1999). The B. distachyon gene Bradi2g61500.1 is annotated to be orthologous to AtCAO. Rice (Oryza sativa) has two isoforms of chlorophyllide a oxygenase. OsCAO1 (LOC_Os10g41780) is induced by light, whereas OsCAO2 (LOC_Os10g41760) is induced under dark conditions and is expressed mainly in non-photosynthetic tissues (Lee et al. 2005). A BLASTN search of the HarvEST database ‘Barley 2943 Mapped SNPs’ with the coding sequence of rice OsCAO1 yielded unigene HAR35_2495 with 83% identity. In order to investigate the possible presence of two CAO genes in barley, a TBLASTN search of the NCBI ‘est other’ database, limited by ‘Hordeum’, was done using OsCAO1 and OsCAO2 as probes. Both queries resulted in the same expressed sequence tags (ESTs) (scores down to 200 were compared), hinting at one isoform of HvCAO. A consensus sequence for HvCAO was constructed from the ESTs, and gene-specific primers were designed according to it (Table 2). HvCAO cDNA was sequenced in the three cultivars Sejets Tron, Svalöfs Bonus and Ackermanns Donaria. The sequences were entered into GenBank with the accessions JQ627619 (Tron), JQ627620 (Bonus) and JQ627621 (Donaria). The cDNA sequences of Tron and Bonus were identical. Donaria differed from them in single nucleotides (Table 3). However, the deduced chlorophyllide a oxygenase polypeptides of the three cultivars were identical. The Tron sequences will be referred to as the ‘wild-type barley’ sequences in the following. Table 2 PCR primers used in the study Gene  Name  Sequence  Position  HvCAO  up1  5′-CCC TCT TTT CCC GCT GCT CGA CG-3′  −51 to −29  up2  5′-CGG CAG CCA GCG TAG CGT CC-3′  −20 to −1  up3  5′-CCA CAG TCG CCT CGC TCT CCC-3′  5 to 25  up4  5′-GCA CTT CCT CCA CGG TCC GAG TCG-3′  584 to 607  lo5  5′-CGA CTC GGA CCG TGG AGG AAG TGC-3′  584 to 607  lo7  5′-CCA TGC AGG GTG GTC GGA ACT CC-3′  1242 to 1264  lo8  5′-GCT GAA CGG TAA CCG ATC GAC GCC-3′  1606 to 1629  lo10  5′-CGG GTC ATC TGG CTG CCA TTG CC-3′  1707 to 1729  chlH  UpU20  5′-TAC TTG TCC CAG ACC AAG TT-3′  2761 to 2780  LoU21  5′-CAT CTC GAG GGA CTC GGA AA-3′  4088 to 4107  dvr  up6  5′-GCC AAG GTG TTC CCG GGG CTG G-3′  640 to 661  lo9  5′-GCG TCA CTT ACA AGC CGG AAG GCC-3′  855 to 878  por  up5  5′-CGT CGG CGT CAA CCA CCT CGG-3′  564 to 584  lo7  5′-CGA CGA GCT TCT CGC TGA GCT CC-3′  1155 to 1177  Gene  Name  Sequence  Position  HvCAO  up1  5′-CCC TCT TTT CCC GCT GCT CGA CG-3′  −51 to −29  up2  5′-CGG CAG CCA GCG TAG CGT CC-3′  −20 to −1  up3  5′-CCA CAG TCG CCT CGC TCT CCC-3′  5 to 25  up4  5′-GCA CTT CCT CCA CGG TCC GAG TCG-3′  584 to 607  lo5  5′-CGA CTC GGA CCG TGG AGG AAG TGC-3′  584 to 607  lo7  5′-CCA TGC AGG GTG GTC GGA ACT CC-3′  1242 to 1264  lo8  5′-GCT GAA CGG TAA CCG ATC GAC GCC-3′  1606 to 1629  lo10  5′-CGG GTC ATC TGG CTG CCA TTG CC-3′  1707 to 1729  chlH  UpU20  5′-TAC TTG TCC CAG ACC AAG TT-3′  2761 to 2780  LoU21  5′-CAT CTC GAG GGA CTC GGA AA-3′  4088 to 4107  dvr  up6  5′-GCC AAG GTG TTC CCG GGG CTG G-3′  640 to 661  lo9  5′-GCG TCA CTT ACA AGC CGG AAG GCC-3′  855 to 878  por  up5  5′-CGT CGG CGT CAA CCA CCT CGG-3′  564 to 584  lo7  5′-CGA CGA GCT TCT CGC TGA GCT CC-3′  1155 to 1177  HvCAO encodes barley chlorophyllide a oxygenase, chlH encodes the H-subunit of magnesium chelatase, dvr encodes 3,8-divinyl protochlorophyllide a 8-vinyl reductase, and por encodes NADPH:protochlorophyllide a oxidoreductase. The ‘lo’ primers are reversely complementary to the coding sequence. Positional information is for the coding sequence with the translation start at +1. View Large Table 3 Variations in HvCAO between the cultivars Sejets Tron, Svalöfs Bonus, Ackermanns Donaria and Haruna Nijo Position  Tron  Bonus  Donaria  Haruna Nijo  117  C  C  A  A  177  C  C  G  G  297  T  T  C  C  462  T  T  C  C  732  G  G  A  A  1037  C  C  C  T  1368  G  G  T  T  1392  C  C  T  T  Position  Tron  Bonus  Donaria  Haruna Nijo  117  C  C  A  A  177  C  C  G  G  297  T  T  C  C  462  T  T  C  C  732  G  G  A  A  1037  C  C  C  T  1368  G  G  T  T  1392  C  C  T  T  The first three cultivars were determined in this study, and the latter is AK359686.1 (Matsumoto et al. 2011). View Large Recently, the sequences of >24,000 full-length cDNAs from barley were published (Matsumoto et al. 2011). GenBank accession AK359686.1 and AK248476.1 are two HvCAO sequences from the cultivar Haruna Nijo. Alignment of the sequences to the HvCAO sequences determined in this study showed Haruna Nijo AK359686.1 to have the same SNP alleles as Donaria (Table 3), except for a unique C to T exchange at position 1037. Interestingly, this exchange is not found in AK248476.1 or any barley EST in the database. This base exchange also leads to a change in the primary sequence of the resulting polypeptide: Pro346 in Tron is replaced by leucine in Haruna Nijo (BAJ90895.1; Fig. 3). Fig. 3 View largeDownload slide Multiple ClustalW 2.1 alignment of chlorophyllide a oxygenase from barley wild-type Tron (this study), B. distachyon (Bradi2g61500.1), O. sativa CAO1 (LOC_Os10g41780) and A. thaliana (AAD54323.1). Triangle 1 indicates the transit peptide cleavage site in Tron, Brachypodium and rice. The cleavage site for the Arabidopsis sequence is marked by triangle 2. Domain A is underlaid light gray and the degron sequence (Sakuraba et al. 2009) within it is highlighted in yellow. Linker domain B is underlaid dark gray (Nagata et al. 2004). The Rieske FeS site is in blue, the mononuclear Fe site is white on red, and the chlorophyllide a oxygenase unique conserved site is in green (Espineda et al. 1999). The residues affected by missense mutations in the clo-f2 mutants are white on black in the barley sequence: Ala94Thr in clo-f2.109 and clo-f2.123; Gly280Asp in clo-f2.103; Cys320Tyr in clo-f2.122; Leu373Phe in clo-f2.181; Ala376Val in clo-f2.133; Asp495Asn in clo-f2.102. The Tron residue Pro346, which is exchanged to leucine in Haruna Nijo, is boxed (GenBank BAJ90895.1; Matsumoto et al. 2011). Fig. 3 View largeDownload slide Multiple ClustalW 2.1 alignment of chlorophyllide a oxygenase from barley wild-type Tron (this study), B. distachyon (Bradi2g61500.1), O. sativa CAO1 (LOC_Os10g41780) and A. thaliana (AAD54323.1). Triangle 1 indicates the transit peptide cleavage site in Tron, Brachypodium and rice. The cleavage site for the Arabidopsis sequence is marked by triangle 2. Domain A is underlaid light gray and the degron sequence (Sakuraba et al. 2009) within it is highlighted in yellow. Linker domain B is underlaid dark gray (Nagata et al. 2004). The Rieske FeS site is in blue, the mononuclear Fe site is white on red, and the chlorophyllide a oxygenase unique conserved site is in green (Espineda et al. 1999). The residues affected by missense mutations in the clo-f2 mutants are white on black in the barley sequence: Ala94Thr in clo-f2.109 and clo-f2.123; Gly280Asp in clo-f2.103; Cys320Tyr in clo-f2.122; Leu373Phe in clo-f2.181; Ala376Val in clo-f2.133; Asp495Asn in clo-f2.102. The Tron residue Pro346, which is exchanged to leucine in Haruna Nijo, is boxed (GenBank BAJ90895.1; Matsumoto et al. 2011). Barley chlorophyllide a oxygenase consists of 550 amino acid residues. Alignment of the barley chlorophyllide a oxygenase polypeptide with those encoded by AtCAO, OsCAO1, OsCAO2 and BdCAO showed 67, 87, 78 and 94% identity, respectively. In the alignment in Fig. 3, the domains of chlorophyllide a oxygenase (Nagata et al. 2004) as well as conserved and functional sites of the Arabidopsis sequence (Espineda et al. 1999, Sakuraba et al. 2009) are indicated. Sequencing of HvCAO in the fch2 and clo-f2 mutants Analysis of genetic mapping data had indicated that fch2 might be HvCAO. To investigate this further, the coding region of HvCAO was sequenced in the fch2 and clo-f2 mutants to see if chlorophyllide a oxygenase is defective in these plants. Mutations were found in all available lines, with the exception of clo-f2.2807. The location of the mutations in the gene and the changes to the resulting polypeptide are given in Table 4 and Fig. 4. Fig. 4 View largeDownload slide The locations of the mutations identified in the HvCAO gene in relation to functional domains in the polypeptide. Missense mutations are shown with gray vertical bars. Frameshift deletions, insertions and nonsense mutations that result in truncated polypeptides are shown by red vertical bars. The domains (TP, transit peptide; A, B and C, domains) and conserved sites (FeS, Rieske FeS-binding site; Fe, mononuclear iron site; ucs, unique conserved site) of chlorophyllide a oxygenase are shown. Fig. 4 View largeDownload slide The locations of the mutations identified in the HvCAO gene in relation to functional domains in the polypeptide. Missense mutations are shown with gray vertical bars. Frameshift deletions, insertions and nonsense mutations that result in truncated polypeptides are shown by red vertical bars. The domains (TP, transit peptide; A, B and C, domains) and conserved sites (FeS, Rieske FeS-binding site; Fe, mononuclear iron site; ucs, unique conserved site) of chlorophyllide a oxygenase are shown. Table 4 Allelic mutations in the fch2 and clo-f2 mutants Line  Mother cultivar  Mutation       DNA  Polypeptide  fch2  (Lion × Manchuria) × California Coast  Insertion G 791-CCTGG-A 792  264 native residues, followed by LDVSRIHVLT EPALLILVQ(283)  clo-f2.101  Tron  Deletion C 427–G 446  142 native residues, followed by SVVGP(147)  clo-f2.102  Tron  G 1483 A  Asp495Asn  clo-f2.103  Tron  G 839 A  Gly280Asp  clo-f2.105  Tron  Deletion A 1061–T 1064  354 native residues  clo-f2.107  Tron  C 1256 T  Pro419Leu  clo-f2.108  Tron  G 990 A  329 native residues  clo-f2.109  Tron  G 280 A  Ala94Thr  clo-f2.122  Bonus  G 959 A  Cys320Tyr  clo-f2.123  Bonus  G 280 A  Ala94Thr  clo-f2.133  Bonus  C 1127 T  Ala376Val  clo-f2.181  Tron  C 1071 T  Leu373Phe  clo-f2.2800  Donaria  Deletion A 1240–T 1241  413 native residues, followed by GVPTTLHGLVNHWHLQAWKA RREEHTAMCNASPPAPCMLAFIYE(457)  clo-f2.2807  Donaria  Transcript deficiency    clo-f2.3613  Donaria  Deletion T 1362–T 1363  454 native residues, followed by IYE(457)  Line  Mother cultivar  Mutation       DNA  Polypeptide  fch2  (Lion × Manchuria) × California Coast  Insertion G 791-CCTGG-A 792  264 native residues, followed by LDVSRIHVLT EPALLILVQ(283)  clo-f2.101  Tron  Deletion C 427–G 446  142 native residues, followed by SVVGP(147)  clo-f2.102  Tron  G 1483 A  Asp495Asn  clo-f2.103  Tron  G 839 A  Gly280Asp  clo-f2.105  Tron  Deletion A 1061–T 1064  354 native residues  clo-f2.107  Tron  C 1256 T  Pro419Leu  clo-f2.108  Tron  G 990 A  329 native residues  clo-f2.109  Tron  G 280 A  Ala94Thr  clo-f2.122  Bonus  G 959 A  Cys320Tyr  clo-f2.123  Bonus  G 280 A  Ala94Thr  clo-f2.133  Bonus  C 1127 T  Ala376Val  clo-f2.181  Tron  C 1071 T  Leu373Phe  clo-f2.2800  Donaria  Deletion A 1240–T 1241  413 native residues, followed by GVPTTLHGLVNHWHLQAWKA RREEHTAMCNASPPAPCMLAFIYE(457)  clo-f2.2807  Donaria  Transcript deficiency    clo-f2.3613  Donaria  Deletion T 1362–T 1363  454 native residues, followed by IYE(457)  The one-letter amino acid code is used to describe longer changes to the sequence of the polypeptide resulting from the mutation. View Large In six of the mutants, frameshift insertion and deletion or nonsense mutations were found, which all led to truncated proteins: the deletion in clo-f2.101 and the insertion in the original fch2 mutant eliminate all catalytic domains in the resulting polypeptide; in the nonsense mutant clo-f2.108 and the deletion mutant clo-f2.105, the truncation is after the Rieske FeS-binding site. In the deletion mutants clo-f2.2800 and clo-f2.3613, the conserved sequence unique to chlorophyllide a oxygenases is missing or incomplete, respectively. In eight clo-f2 mutants, missense mutations were discovered in HvCAO. These led to changes of amino acid residues that are conserved between the chlorophyllide a oxygenases of barley, rice, B. dystachyon and A. thaliana (Fig. 3). While domain A is the one differing most between the chlorophyllide a oxygenases, it contains many conserved residues, particularly in and around the degron sequence. Alanine at position 94 is one of these bordering residues and it is changed to a threonine in the mutants clo-f2.109 and clo-f2.123 (Figs. 3, 4, Table 4). The six remaining missense mutations led to changes of conserved residues in the catalytic C domain (Figs. 3, 4, Table 4): in clo-f2.103 a conserved glycine of the Rieske FeS site is changed to an aspartate, in clo-f2.122 the cysteine at position 320 between the Rieske and the mononuclear iron site is changed to a tyrosine. The mutations in clo-f2.181 and clo-f2.133 change amino acids in the mononuclear iron-binding site itself, a leucine to a phenylalanine and an alanine to a valine, respectively. The change of proline at 419 to leucine in clo-f2.107 is between the mononuclear iron site and the unique conserved site, the aspartate to asparagine (at position 495) in clo-f2.102 is C-terminal of the latter site in a conserved region of previously unnoticed function. The Tron-derived line clo-106 had previously been reported to be Chl b deficient, yet non-allelic with fch2 (Simpson et al. 1985), and was included in the present study as a control. Surprisingly, the sequencing chromatograms showed equally high signals for C and T at position 1071 (Tron has C at position 1071), suggesting a mutation in HvCAO in part of the seeds in the clo-106 stock. In order to better understand the situation in this mutant, the cDNA from six individual leaves was sequenced: three showed equally intense C and T signals at position 1071, two leaves had a T and one a C. This supported the assumption that the seeds in the present clo-106 stock were a mixture of wild-type, homozygous and heterozygous genotypes concerning the mutation in HvCAO. However, all plants originating from the old clo-106 seed batch showed a light-green phenotype indistinguishable by eye. To obtain a true clo-106 line with the wild-type allele of HvCAO, 28 plants of the present clo-106 seeds were grown, harvested and genotyped. Of these, 10 were clo-106 (C 1071), nine were heterozygous for the mutation in HvCAO, and nine constituted a new line named clo-f2.181 (T at position 1071). This new fch2 allele was described together with the other mutants above (Figs. 3, 4, Table 4). Analysis of the pigments (Table 1) showed that clo-f2.181 leaves contained Chl a and b in a ratio of 15 (wild-type Tron had 3.8). In clo-106, the amount of Chl was reduced compared with the wild type (74% of the Chl a of the wild type), while the Chl a/b ratio was higher in this line (3.4). In mutant clo-f2.2807, HvCAO could not be amplified from cDNA initially, while in wild-type barley and the other fch2 and clo-f2 lines HvCAO amplicons were as abundant as those of genes encoding other enzymes of the Chl biosynthetic pathway [H-subunit of magnesium chelatase (chlH), 3,8-divinyl protochlorophyllide a 8-vinyl reductase (dvr) and NADPH: protochlorophyllide a oxidoreductase (por); Fig. 5B]. The result of two different amplicons of HvCAO of clo-f2.2807 is shown in Fig. 5B (highlighted with asterisks). If a 5-fold higher amount of clo-f2.2807 RNA was used in the production of first-strand cDNA, faint bands of amplicons of HvCAO were detected. This observation suggests that the level of transcript of HvCAO in clo-f2.2807 is lower than in the wild type. Fig. 5 View largeDownload slide Mutant clo-f2.2807 is reduced in HvCAO mRNA. (A) Total RNA preparations from Donaria (lane 1) and clo-f2.2807 (lanes 2 and 3). In lane 2, the mass of nucleotides loaded is five times that in lane 3. (B) PCR-amplified cDNA of the Chl biosynthesis loci chlH (lanes 1), dvr (lanes 2) and por (lanes 3) as controls. Amplicons with HvCAO primers up1 and lo7 are in lanes 4 (highlighted with *) and those with primers up2 and lo10 are in lanes 5 (**). In the reactions labeled ‘clo-f2.2807 (5×)’, the amount of RNA used in cDNA first-strand synthesis was five times that in the cDNA syntheses for the reactions ‘Donaria’ and ‘clo-f2.2807’. Fig. 5 View largeDownload slide Mutant clo-f2.2807 is reduced in HvCAO mRNA. (A) Total RNA preparations from Donaria (lane 1) and clo-f2.2807 (lanes 2 and 3). In lane 2, the mass of nucleotides loaded is five times that in lane 3. (B) PCR-amplified cDNA of the Chl biosynthesis loci chlH (lanes 1), dvr (lanes 2) and por (lanes 3) as controls. Amplicons with HvCAO primers up1 and lo7 are in lanes 4 (highlighted with *) and those with primers up2 and lo10 are in lanes 5 (**). In the reactions labeled ‘clo-f2.2807 (5×)’, the amount of RNA used in cDNA first-strand synthesis was five times that in the cDNA syntheses for the reactions ‘Donaria’ and ‘clo-f2.2807’. Discussion Identification of multiple independent alleles is the most effective approach to validate a candidate gene. In the original fch2 line and in 13 allelic clo-f2 genetic stocks, mutations in the coding sequence of HvCAO were found. The lines with nonsense and frameshift mutations are devoid of Chl b. The changes to the chlorophyllide a oxygenase polypeptide resulting from missense mutations are reflected in the observed phenotypes of Chl b deficiency in the mutants, as will be discussed below, and allow insights into the mechanism and regulation of chlorophyllide a oxygenase in barley. We are therefore confident that the Chl b deficiency in the fch2 and clo-f2 mutants is due to mutations in HvCAO and that fch2 and HvCAO are identical. Missense mutants in the regulatory A domain A mutation in the regulatory A domain of chlorophyllide a oxygenase is found in both clo-f2.109 and clo-f2.123. Finding the same mutation in these two lines is interesting as the mutants were isolated after mutagenic treatment of two different mother cultivars, Tron in the case of clo-f2.109 and Bonus in the case of clo-f2.123 (Simpson et al. 1985). In the mutants, the Chl a/b ratio is increased to 9.5 and 9.3, respectively; the proportion of Chl b of total Chl is about half of that found in the wild type. The mutation results in the exchange of an alanine residue by a threonine. The residue in position 94 is conserved and it is close to the degron (Q97DLLTIMILH106 in Arabidopsis; Fig. 3; Sakuraba et al. 2009). This protease degradation signal was suggested to become accessible upon accumulation of Chl b in the chloroplast and in this way to be involved in regulation of chlorophyllide a oxygenase activity by proteolysis (Sakuraba et al. 2009). In Arabidopsis, the removal of the degron caused increased chlorophyllide a oxygenase stability and Chl b formation. Likewise, deletion of the entire A domain led to decreases in Chl a/b ratios (Yamasato et al. 2008). In addition, random missense mutations in the A domain were shown to enhance the stability of the chlorophyllide a oxygenase polypeptide (Sakuraba et al. 2009). The missense mutation found in clo-f2.109 and clo-f2.123, in contrast, decreases the amount of Chl b and thus probably does not stabilize chlorophyllide a oxygenase. Barley residue Ala94 corresponds to Ala95 in Arabidopsis, which was not exchanged in the random mutagenesis of the A domain (Sakuraba et al. 2009). We propose that in clo-f2.109 and clo-f2.123, the exchange of the conserved alanine adjacent to the degron disrupts the structure so that the chlorophyllide a oxygenase degron is exposed to protease attack. The resulting degradation of the polypeptide results in Chl b deficiency. Structural model for the catalytic C domain Four of the residues affected by missense mutations in HvCAO are predicted to be in the Rieske FeS- or in the mononuclear iron-binding site. In this connection it is instructive to relate the exchanges to known protein structures containing these sites. A PSI-BLAST (one repeat) of chlorophyllide a oxygenase against the Protein Data Bank (PDB) found 3,6-dichloro-2-methoxybenzoic acid monooxygenase [Dicamba-MO; PDB accession 3GKE; substrate-bound form 3GL2; Dumitru et al. 2009; Supplementary Fig. S1] as one model structure (Expect = 10−17). Other significant BLAST hits (Expect < 3 × 10−6) were the Rieske FeS and mononuclear iron site containing carbazole 1,9a-dioxygenase (PDB accession 3GKQ) and 3-ketosteroid-9-α-hydroxylase (PDB accession 2ZYL). Comparing the structures of these three polypeptides illustrates structural similarity and conservation of binding residues in the Rieske FeS and mononuclear iron sites (Supplementary Fig. S2). Dicamba-MO catalyzes a somewhat similar reaction to chlorophyllide a oxygenase: Where chlorophyllide a oxygenase oxidizes a chlorophyllide a chlorin ring methyl substituent, Dicamba-MO is involved in the oxidation of the O-methyl substituent of the aromatic ring in the herbicide 3,6-dichloro-2-methoxybenzoic acid. The O-hydroymethyl group is subsequently removed (Herman et al. 2005, Dumitru et al. 2009). The alignment of Dicamba-MO with the chlorophyllide a oxygenases from barley, the evolutionarily distantly related prochlorophyte P. hollandica and the moss Physcomitrella patens is shown in Fig. 6. Fig. 6 View largeDownload slide Multiple Clustal W 2.1 alignment of 3,6-dichloro-2-methoxybenzoic acid monooxygenase (PDB accession 3GKE) and the chlorophyllide a oxgenases from barley, P. patens (accession EDQ57502) and P. hollandica (accession BAD02269.1). Residues involved in the binding of the Rieske FeS cluster are boxed in blue; those involved in binding the mononuclear iron are white on red. The ‘gatekeeper’ aspartate is boxed in yellow. The amino acids changed in six of the clo-f2 lines compared with wild-type Tron are white on black. Fig. 6 View largeDownload slide Multiple Clustal W 2.1 alignment of 3,6-dichloro-2-methoxybenzoic acid monooxygenase (PDB accession 3GKE) and the chlorophyllide a oxgenases from barley, P. patens (accession EDQ57502) and P. hollandica (accession BAD02269.1). Residues involved in the binding of the Rieske FeS cluster are boxed in blue; those involved in binding the mononuclear iron are white on red. The ‘gatekeeper’ aspartate is boxed in yellow. The amino acids changed in six of the clo-f2 lines compared with wild-type Tron are white on black. The residues binding the FeS cluster [C-x-H-x(16)-C-x(2)-H] and the mononuclear iron [D-x(2)-H-x(4)-H-x(n)-D] in the Dicamba-MO structure are identical in the four proteins (Figs. 6, 7). The conservation of the metal-binding and surrounding residues suggests that the Dicamba-MO structure can be used as a model for barley chlorophyllide a oxygenase in the following discussion. Fig. 7 View largeDownload slide Structure of the active center of 3,6-dichloro-2-methoxybenzoic acid monooxygenase (Dicamba-MO; PDB accession 3GL2). The Rieske FeS site of one monomer (blue) and the mononuclear iron site of a second monomer (red) form the active site. The proposed electron path involving ‘gatekeeper’ Asp157 is shown as a dashed line (D’Ordine et al. 2009). Residue numbering is according to Dicamba-MO. Two of the residues affected by missense mutations in barley shown and named here are Gly280 to aspartate in clo-f2.103 (dark blue) and Asp495 to asparagine in clo-f2.102. Fig. 7 View largeDownload slide Structure of the active center of 3,6-dichloro-2-methoxybenzoic acid monooxygenase (Dicamba-MO; PDB accession 3GL2). The Rieske FeS site of one monomer (blue) and the mononuclear iron site of a second monomer (red) form the active site. The proposed electron path involving ‘gatekeeper’ Asp157 is shown as a dashed line (D’Ordine et al. 2009). Residue numbering is according to Dicamba-MO. Two of the residues affected by missense mutations in barley shown and named here are Gly280 to aspartate in clo-f2.103 (dark blue) and Asp495 to asparagine in clo-f2.102. Mononuclear iron site mutant clo-f2.102 The alignment in Fig. 6 shows the conservation of a C-terminal aspartate residue (position 294 in Dicamba-MO) in the evolutionarily disparate sequences. Although separated by >100 residues from the nearest other mononuclear iron ligand (histidine at position 165), the structure (Fig. 7) shows its involvement in the mononuclear iron site in Dicamba-MO. The mutation in the clo-f2.102 mutant affects this C-terminal, conserved iron-liganding aspartate and exchanges it to asparagine. In barley chlorophyllide a oxygenase, aspartate at position 495 is separated by 114 residues from the nearest other mononuclear iron ligand. The result of this mutation is total Chl b deficiency (Table 1). Electron transfer inhibition in clo-f2.103, clo-f2.181 and clo-f2.133 In the Dicamba-MO monomer, the Rieske FeS cluster and the mononuclear iron site are far apart from each other (Supplementary Fig. S1A). The formation of a trimer results in three intersubunit Rieske FeS/mononuclear iron sites (Supplementary Fig. S1B; Dumitru et al. 2009). The crystal structure (Fig. 7) illustrates how electron transport occurs between the Rieske FeS cluster of one subunit of the protein and the mononuclear iron of an adjacent subunit where it reductively activates O2, enabling substrate oxidation (D’Ordine et al. 2009). Functional oligomers of chlorophyllide a oxygenase have not been described to our knowledge, but the similarity to Dicamba-MO indicates oligomerization. In the following we show how the electron transport between the Rieske FeS and mononuclear iron sites (dashed line in Fig. 7) in Dicamba-MO is likely to be inhibited by three missense mutations corresponding to those identified in HvCAO in the clo-f2 lines. This explains how the ability to synthesize Chl b is affected in the barley mutants. The mutation in the conserved Rieske FeS site in the line clo-f2.103 results in total Chl b deficiency. The amino acid changed is glycine (position 280) which in Dicamba-MO (Gly59) is at the end of a hairpin of β-sheet connecting the two C-x(-x)-H motifs which bind the two irons of the Rieske FeS center (Gly59 is dark blue in Fig. 7). This glycine is conserved in the three evolutionarily divergent sequences of barley, P. patens and P. hollandica chlorophyllide a oxygenases and also in Dicamba-MO (Fig. 6). The mutation in clo-f2.103 introduces a charged aspartate residue at the FeS center in the interior of the protein which will certainly disturb charge balance in the active site of the protein and inevitably destroy activity as observed. The Chl b free phenotype of clo-f2.103 is the consequence. The exchange in clo-f2.181 (Leu373 is replaced by phenylalanine) is adjacent to the conserved Asp372 which corresponds to the ‘gatekeeper’ Asp157 of Dicamba-MO (Figs. 6, 7). The gatekeeper passes electrons from the Rieske FeS center of one subunit over a distance of 12.5 Å to the mononuclear iron of another subunit (D’Ordine et al. 2009). The presence of a bulky, hydrophobic and electron-rich phenylalanine residue instead of leucine neighboring the gatekeeper results in an increase in the Chl a/b ratio to 15 in clo-f2.181 (the Chl b proportion of total Chl is a quarter of that in the wild type). In rice there are two isoforms of chlorophyllide a oxygenase (Lee et al. 2005). Interestingly, only the product of light-induced OsCAO1 has a leucine at position 364 (corresponding to barley position 373), while the polypeptide encoded by dark-induced OsCAO2 has an isoleucine residue there. The mutation in clo-f2.133 causes changes in the conserved mononuclear iron-binding site. Ala376 is exchanged to valine, which reduces the Chl b proportion of total Chl to about half that in the wild type (the Chl a/b ratio is 8.5). The alanine residue is identical in Dicamba-MO and the three chlorophyllide a oxygenases (Fig. 6). In Dicamba-MO, this alanine (position 161) is the neighbor of the mononuclear iron-binding ligand His160 which is suggested to be an electron transport component (Fig. 7; D’Ordine et al. 2009). The presence of a valine residue which has a 50% greater molecular volume than alanine (Wang et al. 1999) next to the mononuclear iron site might negatively affect chlorophyllide a oxygenase activity and lead to the observed deficiency in Chl b. Missense mutant clo-f2.122 The residue changed by the missense mutation in clo-f2.122 (Cys320 is replaced by tyrosine) is between the Rieske FeS- and mononuclear iron-binding sites. The alignment of barley, rice, B. dystachyon and A. thaliana chlorophyllide a oxygenases (Fig. 3) shows that Cys320 is conserved, and also in P. patens chlorophyllide a oxygenase (Fig. 6). In P. hollandica, the orthologous residue is a valine (Fig. 6). Interestingly, when P. hollandica chlorophyllide a oxygenase was overexpressed in a Chl b-less Arabidopsis mutant, the Chl a/b ratio observed was extremely low and it decreased from 1.1 to 0.9 in high light conditions. In contrast, the Chl a/b ratio in wild-type Arabidopsis increased in high light (Hirashima et al. 2006). In agreement with this, the effect of light intensity on the prochlorophyte has been reported to be opposite to the effect on plants: in P. hollandica, the Chl a/b ratio increases in low light and decreases in high light (Burger-Wiersma and Post 1989). A redox- and/or light intensity-dependent regulatory system has been demonstrated for many chloroplast enzymes (Montrichard et al. 2009). If the conserved cysteine in position 320 in barley is involved in such regulation, the presence of a valine in this position in P. hollandica might explain the different behavior toward light intensities. The involvement of Cys320 in redox regulation of chlorophyllide a oxygenase might be through the formation of a disulfide bridge. However, it has been suggested that chlorophyllide a oxygenase is not regulated via the thioredoxin system that was shown to regulate pheophorbide a oxygenase (Bartsch et al. 2008). It should be noted that this result considers only a certain group of thioredoxin targets (C-x-x-C) where an intramolecular disulfide bridge is reduced by thioredoxin. Immunoprecipitates of Chlamydomonas chlorophyllide a oxygenase suggest the co-precipitation of a thioredoxin (Eggink et al. 2004). It will be very interesting to see if chlorophyllide a oxygenase is regulated by thioredoxin and the formation of intermonomeric disulfide bridges in addition to the regulation by protein degradation (Sakuraba et al. 2009) and gene expression (Tanaka and Tanaka 2005). Missense mutant clo-f2.107 In missense mutant clo-f2.107, the proline residue in position 419 is changed to a leucine. This affects one of a pair of proline residues in a strongly conserved stretch between the mononuclear iron site and the chlorophyllide a oxygenase unique site. This mutant has the highest Chl b content of all fch2 and clo-f2 mutants (Chl a/b is 5.7). Interestingly, the analysis of pigments showed more Chl a in clo-f2.107 than in the wild type (Table 1). The alignment of evolutionarily diverse sequences (Fig. 6) shows that the Pro–Pro pair is conserved in barley, P. hollandica and P. patens. Of the Pro–Pro sequences in the PISCES-culled PDB database (Wang and Dunbrack 2003), 56% were found to adopt a polyproline helix structure (Rai et al. 2006). Polyproline stretches are considered to be important for protein–protein interactions (Mansiaux et al. 2011). The exchange of the proline residue at position 419 might therefore impair the interaction of chlorophyllide a oxygenase with a possible other component of the reaction such as ferredoxin (Oster et al. 2000). Nonsense and frameshift mutations In the present study, two nonsense mutations, three mutants with deletions and one with an insertion were identified. All of these lead to premature stop codons and eventually polypeptides that are truncated, without conserved sites or entire domains (Fig. 4; Espineda et al. 1999, Nagata et al. 2004) of chlorophyllide a oxygenase, and thus without catalytic function. The deletions in clo-f2.2800 and clo-f2.3613 result in polypeptides lacking the unique conserved site. The translation products in the mutants clo-f2.108 and clo-f2.105 are truncated before the mononuclear iron-binding site, whereas the insertion in mutant fch2 leads to an omission of the C-terminus including the Rieske FeS-binding site. The mutation in clo-f2.101 is even more severe, allowing translation of the N-terminal first quarter of chlorophyllide a oxygenase only. Given the severe mutations in these lines, it is understandable that no Chl b could be detected in either of them (Table 1). It is possible that truncated Chl a oxygenase polypeptides are not present in these mutants in vivo. Transcript instability by the mechanism of nonsense-mediated decay has been observed in barley xantha mutants (Gadjieva et al. 2004). In addition, premature stop codons or other modifications leading to a reduction in structural stability of the polypeptide have been suggested to increase degradation of polypeptides in other barley mutants (Olsson et al. 2004). The transcript-deficient mutant clo-f2.2807 No mutation was detected in the coding region of HvCAO of clo-f2.2807. The decreased amount of transcript suggests a mutation in a regulatory element of HvCAO. In the two other mutants induced by X-ray irradiation of seeds, clo-f2.2800 and clo-f2.3613 (Machold et al. 1977), base deletions were detected (Table 4). The same type of mutation might be expected in clo-f2.2807. The transcript reduction must critically affect chlorophyllide a oxygenase levels since Chl b cannot be detected in clo-f2.2807 (Table 1). The mutations that were identified in this study are distributed all over the HvCAO gene (Fig. 4). This will be utilized in future work focusing on the chlorophyllide a oxygenase protein and the function of the various sites. An expression system to produce chlorophyllide a oxygenase recombinantly will allow in vitro analyses, including in vitro activity measurements of the mutated versions of chlorophyllide a oxygenase identified in the present study. The availability of the mutations of HvCAO in viable and well characterized fch2 and clo-f2 mutant lines (Highkin 1950, Sagromsky 1974, Simpson et al. 1985) will allow additional in vivo studies that provide an important complement to the in vitro studies. This is of particular interest as transformation of barley—and thus directed modification of polypeptides in vivo—is only possible with certain genotypes (Harwood 2012) at present. By correlating detailed knowledge of the seven different missense mutations and the ability of the mutants to make Chl b, it will be possible to test the ideas about the catalytic mechanism of chlorophyllide a oxygenase detailed above. In the leaky mutants with detectable amounts of Chl b, the presence of Chl b implies that chlorophyllide a oxygenase is only partly active, but present. In the mutants with no detectable Chl b, in contrast, the phenotype could also be explained by a lack of the protein as a consequence of the mutation. Future studies of the chlorophyllide a oxygenase protein will also analyze for the actual presence of the truncated or missense mutated versions of chlorophyllide a oxygenase in the fch2 and clo-f2 plants. Mutants clo-106 and clo-f2.181 We showed that the stock clo-106 was a mixture of seeds with wild-type, heterozygous and homozygous mutant genotypes for the replacement of C by T at position 1071 in HvCAO. However, all plants, irrespective of their genotype, showed a light-green phenotype, indicating the presence of another mutation in that batch. By analyzing the SNP at position 1071, a line with the wild-type allele of HvCAO (homozygous for C at position 1071) was isolated which keeps the name clo-106. The pigment determination showed no Chl b deficiency in this non-fch2 mutant; the Chl a/b ratio was 3.4. The chlorina phenotype of clo-106 is due to an unknown mutation conferring a reduction in total Chl: it contains 74% of the amount of Chl a in Tron. In the same manner, a line homozygous for T at position 1071 was isolated which we designate chlorina-f2.181 (clo-f2.181; since chlorina 180 is the highest numbered chlorina mutant mentioned in Simpson and von Wettstein 1992). It should be noted that clo-f2.181 carries the unknown clo-106 mutation in addition to the mutation in HvCAO. The new line clo-f2.181 has a low Chl b content; the Chl a/b ratio is 15 (Table 1). Pigment determinations Comparing the Chl a contents of the fch2 and clo-f2 mutants, most are in a range between 60 and 90% of that of the wild type. Exceptions to this are clo-f2.103 and clo-f2.123 with 38 and 30%, respectively. The reductions in Chl a content in the fch2 and clo-f2 lines were ascribed to the different composition of the photosynthetic apparatus resulting from the lack of Chl b (Preiss and Thornber 1995). The presence of a second mutation in these lines that decreases the Chl content as such is not probable as the lines have been repeatedly backcrossed. In addition, this hypothetical second mutation would have to be present in all of these lines. The Chl b-deficient lines clo-f2.108 and clo-f2.107 contain wild-type levels of Chl a or more Chl a than Tron, respectively. How the variations in Chl content can be explained will be the focus of further studies. The pigment content of the fch2 and clo-f2 mutants has been determined before. Highkin (1950) reported Chl a as the only green pigment in fch2. We did not detect Chl b in fch2 either. In a later study, Sagromsky (1974) determined the Chl a/b ratio in Donaria as 2.8. The difference from values determined here may be explained by the absorption coefficients used (Hager and Bertenrath 1962, Hager and Meyer-Bertenrath 1966) and the growth conditions of the plants. Other than that, the results agreed with the results found here in that no Chl b was found in clo-f2.2800, clo-f2.2807 and clo-f2.3613. Simpson et al. (1985) determined the pigmentation of the clo-f2 mutants by HPLC and reported amounts of Chl b (of total Chl) of up to 2.0% in clo-f2.101, clo-f2.102, clo-f2.103 and clo-f2.105, while no Chl b was detected in these mutants in the present study. We observed a signal in the elution from HPLC that had a retention time comparable with that of Chl b but no Chl b-typic absorption spectrum. In the study of Simpson et al., only the absorption of the eluate at 430 nm was monitored and this signal was possibly taken to be Chl b. We therefore assume a detection threshold in Simpson et al. (1985) of 2% and, with that in mind, the only disagreement between the HPLC-based pigment analyses was the pigmentation of clo-f2.122, where Simpson et al. found 14.3% Chl b, while we measured 4.6%. This disagreement cannot be explained. Materials and Methods Plant material The barley (Hordeum vulgare L.) mutants used in the present study can be obtained from NordGen (www.nordgen.org) or the Carlsberg Laboratory (www.carlsberglab.dk/professors/Hansson). Seeds of mutant lg5 and the historic crossing (Lion × Manchuria) × California Coast, in which the original fch2 mutant occurred, could not be obtained. Identification of mutations For RNA isolation, mutant and wild-type barley was grown in moist vermiculite in ambient light at room temperature. The seedling leaves were harvested after 10 d and frozen in liquid nitrogen. TRIzol Reagent (Invitrogen) was used according to the manufacturer’s instructions to extract RNA from ground leaves. The RNA was treated with DNase I and reverse transcribed to DNA using dT18 primers and RevertAid M-MuLV reverse transcriptase (all from Fermentas/Thermo Fisher Scientific). The first-strand cDNA was amplified with gene-specific primers (Eurofins MWG Operon) and High Fidelity PCR Enzyme Mix (Fermentas/Thermo Fisher Scientific) according to the manufacturer’s instructions. The cycling conditions were: 94°C for 3 min, followed by 39 cycles of 94°C for 1 min, annealing temperature (5°C below the lowest primer-specific annealing temperature as given by the manufacturer) for 30 s, and 72°C for 1 min per 1,000 bp. After a final incubation at 72°C for 10 min, the products were stored at 4°C. The gene-specific primers (Table 2) were designed according to an EST consensus sequence as described in the Results. Amplicons produced with primer combinations up1/lo5, up1/lo7, up2/lo8, up2/lo10, up3/lo8, up3/lo10, up4/lo8 and up4/lo10 usually resulted in satisfactory coverage of the sequence. Sequencing was done by StarSEQ (Mainz, Germany). FinchTV 1.4.0 (Geospiza Research Team) and Jalview 2.6.1 (Waterhouse et al. 2009) were used to analyze DNA sequence data. As controls, primer pairs specific for the genes encoding the large magnesium chelatase subunit (chlH; Olsson et al. 2004), 3,8-divinyl protochlorophyllide a 8-vinyl reductase (dvr) and NADPH:protochlorophyllide a oxidoreductase (por) were used (Table 2). The latter two pairs were designed according to EST consensus sequences determined by the procedure laid out for HvCAO in the Results. For analytical agarose gel electrophoresis, 1% gels stained with ethidium bromide (0.5 µg ml−1) were used. O’Gene Ruler 1 kb DNA Ladder Plus (0.3 µg; Fermentas/Thermo Fischer Scientific) was used as a marker. Pigment determinations For the pigment determinations, barley was grown in moist vermiculite in 16/8 h light/dark cycles with 80 µmol photons m−2 s−1 at 21°C for 10 d. The plants were harvested and frozen in liquid nitrogen. This could not be done for the lines clo-f2.122 and clo-f2.2807 due to a shortage of seeds. Instead, leaves grown in a greenhouse in winter in moist vermiculite (ambient light, no temperature control) for an earlier experiment were used to determine the pigments in these lines. The samples were otherwise treated in the same way. The wild-type samples used for reference amounts of pigments were grown under the same conditions as the respective fch2/clo-f2 mutant. To extract pigments from frozen plant material, leaves were ground in liquid nitrogen and 50 mg were extracted with 200 µl of 80% acetone (acetone/50 mM Tris–HCl, pH 8.0, 80 : 20, v/v; Müller et al. 2011). The extraction was repeated with 200 µl and 100 µl of 80% acetone (acetone/50 mM Tris–HCl, pH 8.0, 80 : 20, v/v) and the green phases were pooled. The preparation of samples for HPLC analysis used the method described by Martinson and Plumley (1995), with modifications: 50 µl of the pigment extract in 80% acetone (acetone/50 mM Tris–HCl, pH 8.0, 80 : 20, v/v) was mixed with 20 µl of 5 M NaCl and extracted twice using 50 µl of 2-butanol. The 2-butanol phases were combined and made up to 500 µl with 80% acetone (acetone/50 mM Tris–HCl, pH 8.0, 80 : 20, v/v). To analyze the pigments in this solution by HPLC, the protocol of Hobe et al. (2000) was modified (personal communication Dr. Hobe). A 20 µl sample was analyzed using a Chromolith SpeedROD RP-18 50–4.6 mm column (Merck) and a linear gradient from 70 to 100% acetone over 3.5 min at a flow rate of 1.5 ml min−1. Total run time was 6.5 min. The diluent water for the acetone was buffered with 0.2 mM Tris–HCl, pH 7.0. The eluting pigments were identified by their absorption properties and quantified based on the absorption at 440 nm. Average retention times (± SD; 12 samples) were: neoxanthin 1.48 min (±0.08 min), violaxanthin 1.93 min (±0.09 min), lutein 2.82 min (±0.07 min), Chl b 3.80 min (±0.07 min), Chl a 3.96 min (±0.07 min) and β-carotene 4.53 min (±0.10 min). Bioinformatics The GrainGenes database (wheat.pw.usda.gov), HarvEST (harvest.ucr.edu) and the Triticeae Mapped EST DataBase (trimedb.psc.riken.jp) were used for barley marker and EST sequences. Rice data was from rice.plantbiology.msu.edu; B. distachyon sequences were from db.brachypodium.org. Accession numbers are for GenBank at NCBI. BLAST searches (Altschul et al. 1997) were performed at NCBI (blast.ncbi.nlm.nih.gov) and HarvEST (138.23.178.42/blast). Chloroplast transit peptides were determined by ChloroP 1.1 (www.cbs.dtu.dk/services/ChloroP/; Emanuelsson et al. 1999). Sequence alignments used ClustalW 2.1 (www.ebi.ac.uk). Structural analyses used UCSF Chimera (www.cgl.ucsf.edu/chimera/; Pettersen et al. 2004). Funding This work was supported by The Danish Council for Independent Research Natural Sciences [to M.H.]; the Deutsche Forschungsgemeinschaft [DO14821/1-1 to C.D.]. Acknowledgments The authors are grateful to Lisbeth Faldborg for technical assistance. We are also grateful to Dr. Stephan Hobe for support in determining pigment content. Abbreviations Abbreviations Dicamba-MO 3,6-dichloro-2-methoxybenzoic acid monooxygenase EST expressed sequence tag SNP single nucleotide polymorphism. References Altschul SF,  Madden TL,  Schäffer AA,  Zhang J,  Zhang Z,  Miller W, et al.  Gapped BLAST and PSI-BLAST: a new generation of protein database search programs,  Nucleic Acids Res. ,  1997, vol.  25 (pg.  3389- 3402) Google Scholar CrossRef Search ADS PubMed  Apel P.  Photosynthesemessungen an Chlorophyllmutanten von Gerste (Lichtkurven, ‘Lichtatmung’, Starklichempfindlichkeit),  Stud. Biophys. ,  1967, vol.  5 (pg.  105- 110) Axelsson E,  Lundqvist J,  Sawicki A,  Nilsson S,  Schröder I,  Al-Karadaghi S, et al.  Recessiveness and dominance in barley mutants deficient in Mg-chelatase subunit D, an AAA protein involved in chlorophyll biosynthesis,  Plant Cell ,  2006, vol.  18 (pg.  3606- 3616) Google Scholar CrossRef Search ADS PubMed  Bartsch S,  Monnet J,  Selbach K,  Quigley F,  Gray J,  von Wettstein D, et al.  Three thioredoxin targets in the inner envelope membrane of chloroplasts function in protein import and chlorophyll metabolism,  Proc. Natl Acad. Sci. USA ,  2008, vol.  105 (pg.  4933- 4938) Google Scholar CrossRef Search ADS   Bhalla P,  Bennett J.  Chloroplast phosphoproteins: phosphorylation of a 12-kDa stromal protein by the redox-controlled kinase of thylakoid membranes,  Arch. Biochem. Biophys. ,  1987, vol.  252 (pg.  97- 104) Google Scholar CrossRef Search ADS PubMed  Bossmann B,  Knoetzel J,  Jansson S.  Screening of chlorina mutants of barley (Hordeum vulgare L.) with antibodies against light-harvesting proteins of PS I and PS II: absence of specific antenna proteins,  Photosynth. Res. ,  1997, vol.  52 (pg.  127- 136) Google Scholar CrossRef Search ADS   Burger-Wiersma T,  Post AF.  Functional analysis of the photosynthetic apparatus of Prochlorothrix hollandica (Prochlorales), a chlorophyll b containing procaryote,  Plant Physiol. ,  1989, vol.  91 (pg.  770- 774) Google Scholar CrossRef Search ADS PubMed  D’Ordine RL,  Rydel TJ,  Storek MJ,  Sturman EJ,  Moshiri F,  Bartlett RK, et al.  Dicamba monooxygenase: structural insights into a dynamic Rieske oxygenase that catalyzes an exocyclic monooxygenation,  J. Mol. Biol. ,  2009, vol.  392 (pg.  481- 497) Google Scholar CrossRef Search ADS PubMed  Druka A,  Franckowiak J,  Lundqvist U,  Bonar N,  Alexander J,  Houston K, et al.  Genetic dissection of barley morphology and development,  Plant Physiol. ,  2010, vol.  155 (pg.  617- 627) Google Scholar CrossRef Search ADS PubMed  Dumitru R,  Jiang WZ,  Weeks DP,  Wilsom MA.  Crystal structure of dicamba monooxygenase: a Rieske nonheme oxygenase that catalyzes oxidative demethylation,  J. Mol. Biol. ,  2009, vol.  392 (pg.  498- 510) Google Scholar CrossRef Search ADS PubMed  Eggink LL,  LoBrutto R,  Brune DC,  Brusslan J,  Yamasato A,  Tanaka A, et al.  Synthesis of chlorophyll b: localization of chlorophyllide a oxygenase and discovery of a stable radical in the catalytic subunit,  BMC Plant Biol. ,  2004, vol.  4 pg.  5  Google Scholar CrossRef Search ADS PubMed  Emanuelsson O,  Nielsen H,  von Heijne G.  ChloroP, a neural network-based method for predicting chloroplast transit peptides and their cleavage sites,  Protein Sci. ,  1999, vol.  8 (pg.  978- 984) Google Scholar CrossRef Search ADS PubMed  Espineda CE,  Linford AS,  Devine D,  Brusslan JA.  The AtCAO gene, encoding chlorophyll a oxygenase, is required for chlorophyll b synthesis in Arabidopsis thaliana,  Proc. Natl Acad. Sci. USA ,  1999, vol.  96 pg.  10507 10511  Google Scholar CrossRef Search ADS   Franckowiak JD.  Stock number: BGS 117,  Barley Genet. Newslett. ,  1997, vol.  26 (pg.  151- 152) Gadjieva R,  Axelsson E,  Olsson U,  Vallon-Christersson J,  Hansson M.  Nonsense-mediated mRNA decay in barley mutants allows the cloning of mutated genes by a microarray approach,  Plant Physiol. Biochem. ,  2004, vol.  42 (pg.  681- 685) Google Scholar CrossRef Search ADS PubMed  Goodchild DJ,  Highkin HR,  Boardman NK.  The fine structure of chloroplasts in a barley mutant lacking chlorophyll b,  Exp. Cell Res. ,  1966, vol.  43 (pg.  684- 688) Google Scholar CrossRef Search ADS PubMed  Hager A,  Bertenrath T.  Verteilungschromatographische Trennung von Chlorophyllen und Carotinoiden grüner Pflanzen an Dünnschichten,  Planta ,  1962, vol.  58 (pg.  564- 568) Google Scholar CrossRef Search ADS   Hager A,  Meyer-Bertenrath T.  Die Isolierung und quantitative Bestimmung der Carotinoide und Chlorophylle von Blättern, Algen und isolierten Chloroplasten mit Hilfe dünnschichtchromatographischer Methoden,  Planta ,  1966, vol.  69 (pg.  198- 217) Google Scholar CrossRef Search ADS PubMed  Hansson A,  Kannangara CG,  von Wettstein D,  Hansson M.  Molecular basis for semidominance of missense mutations in the XANTHA-H (42-kDa) subunit of magnesium chelatase,  Proc. Natl Acad. Sci. USA ,  1999, vol.  96 (pg.  1744- 1749) Google Scholar CrossRef Search ADS   Harrison MA,  Nemson JA,  Melis A.  Assembly and composition of the chlorophyll a-b light harvesting complex of barley (Hordeum vulgare L.): immunochemical analysis of chlorophyll b-less and chlorophyll b-deficient mutants,  Photosynth. Res. ,  1993, vol.  38 (pg.  141- 151) Google Scholar CrossRef Search ADS PubMed  Harwood WA.  Advances and remaining challenges in the transformation of barley and wheat,  J. Exp. Bot. ,  2012, vol.  63 (pg.  1791- 1798) Google Scholar CrossRef Search ADS PubMed  Havaux M,  Tardy F.  Thermostability and photostability of photosystem II in leaves of the chlorina-f2 barley mutant deficient in light-harvesting chlorophyll a/b protein complexes,  Plant Physiol. ,  1997, vol.  113 (pg.  913- 923) Google Scholar CrossRef Search ADS PubMed  Henriques F,  Park RB.  Further chemical and morphological characterization of chloroplast membranes from a chlorophyll b-less mutant of Hordeum vulgare,  Plant Physiol. ,  1975, vol.  55 (pg.  763- 767) Google Scholar CrossRef Search ADS PubMed  Herman PL,  Behrens M,  Chakraborty S,  Chrastil BM,  Barycki J,  Weeks DP.  A three-component dicamba O-demethylase from Pseudomonas maltophila, strain DI-6,  J. Biol. Chem. ,  2005, vol.  280 (pg.  24759- 24767) Google Scholar CrossRef Search ADS PubMed  Hirashima M,  Satoh S,  Tanaka R,  Tanaka A.  Pigment shuffling in antenna systems achieved by expressing prokaryotic chlorophyllide a oxygenase in Arabidopsis,  J. Biol. Chem. ,  2006, vol.  281 (pg.  15385- 15393) Google Scholar CrossRef Search ADS PubMed  Hobe S,  Niemeier H,  Bender A,  Paulsen H.  Carotenoid binding sites in LHCIIb,  Eur. J. Biochem. ,  2000, vol.  267 (pg.  616- 624) Google Scholar CrossRef Search ADS PubMed  Highkin HR.  Chlorophyll studies on barley mutants,  Plant Physiol. ,  1950, vol.  25 (pg.  294- 306) Google Scholar CrossRef Search ADS PubMed  Highkin HR,  Frenkel AW.  Studies of growth and metabolism of a barley mutant lacking chlorophyll b,  Plant Physiol. ,  1952, vol.  37 (pg.  814- 820) Google Scholar CrossRef Search ADS   Hiller RG,  Raison JK.  The fluidity of chloroplast thylakoid membranes and their constituent lipids—a comparative study by ESR,  Biochim. Biophys. Acta ,  1980, vol.  599 (pg.  63- 72) Google Scholar CrossRef Search ADS PubMed  Lee KP,  Kim C,  Lee DW,  Apel K.  TIGRINA d, required for regulating the biosynthesis of tetrapyrroles in barley, is an ortholog of the FLU gene of Arabidopsis thaliana,  FEBS Lett. ,  2003, vol.  553 (pg.  119- 124) Google Scholar CrossRef Search ADS PubMed  Lee S,  Kim J-H,  Yoo ES,  Lee C-H,  Hirochika H,  An G.  Differential regulation of chlorophyll a oxygenase in rice,  Plant Mol. Biol. ,  2005, vol.  57 (pg.  805- 818) Google Scholar CrossRef Search ADS PubMed  Lundqvist U.  Mutation research in barley,  1992  PhD thesis. The Swedish University of Agricultural Sciences, Svalöv Machold O,  Meister A,  Sagromsky H,  Høyer-Hansen G,  von Wettstein D.  Composition of photosynthetic membranes of wild-type barley and chlorophyll b-less mutants,  Photosynthetica ,  1977, vol.  11 (pg.  200- 206) Mansiaux Y,  Joseph AP,  Gelly JC,  de Brevern AG.  Assignment of polyproline II conformation and analysis of sequence–structure relationship,  PLoS ONE ,  2011, vol.  6 pg.  e18401  Google Scholar CrossRef Search ADS PubMed  Martinson TA,  Plumley FG.  One-step extraction and concentration of pigments and acyl lipids by sec-butanol from in vitro and in vivo samples,  Anal. Biochem. ,  1995, vol.  228 (pg.  123- 130) Google Scholar CrossRef Search ADS PubMed  Matsumoto T,  Tanaka T,  Sakai H,  Amano N,  Kanamori H,  Kurita K, et al.  Comprehensive sequence analysis of 24,783 barley full-length cDNAs from 12 clone libraries,  Plant Physiol. ,  2011, vol.  156 (pg.  20- 28) Google Scholar CrossRef Search ADS PubMed  Montrichard F,  Alkhalfioui F,  Yano H,  Vensel WH,  Hurkman WJ,  Buchanan BB.  Thioredoxin targets in plants: the first 30 years,  J. Proteomics ,  2009, vol.  72 (pg.  452- 474) Google Scholar CrossRef Search ADS PubMed  Müller AH,  Gough SP,  Bollivar DW,  Meldahl M,  Willows RD,  Hansson M.  Methods for the preparation of chlorophyllide a: an intermediate of the chlorophyll biosynthetic pathway,  Anal. Biochem. ,  2011, vol.  419 (pg.  271- 276) Google Scholar CrossRef Search ADS PubMed  Mullet JE,  Burke JJ,  Arntzen CJ.  A developmental study of photosystem I peripheral chlorophyll proteins,  Plant Physiol. ,  1980, vol.  65 (pg.  823- 827) Google Scholar CrossRef Search ADS PubMed  Nagata N,  Satoh S,  Tanaka R,  Tanaka A.  Domain structures of chlorophyllide a oxygenase of green plants and Prochlorothrix hollandica in relation to catalytic functions,  Planta ,  2004, vol.  218 (pg.  1019- 1025) Google Scholar CrossRef Search ADS PubMed  Olsson U,  Sirijovski N,  Hansson M.  Characterization of eight barley xantha-f mutants deficient in magnesium chelatase,  Plant Physiol. Biochem. ,  2004, vol.  42 (pg.  557- 564) Google Scholar CrossRef Search ADS PubMed  Oster U,  Tanaka R,  Tanaka A,  Rüdiger W.  Cloning and functional expression of the gene encoding the key enzyme for chlorophyll b biosynthesis (CAO) from Arabidopsis thaliana,  Plant J. ,  2000, vol.  21 (pg.  305- 310) Google Scholar CrossRef Search ADS PubMed  Pettersen EF,  Goddard TD,  Huang CC,  Couch GS,  Greenblatt DM,  Meng EC, et al.  UCSF Chimera—a visualization system for exploratory research and analysis,  J. Comput. Chem. ,  2004, vol.  25 (pg.  1605- 1612) Google Scholar CrossRef Search ADS PubMed  Preiss S,  Thornber JP.  Stability of the apoproteins of light harvesting complex I and II during biogenesis of thylakoids in the chlorophyll b-less barley mutant chlorina f2,  Plant Physiol. ,  1995, vol.  107 (pg.  709- 717) Google Scholar CrossRef Search ADS PubMed  Rai R,  Aravinda S,  Kanagarajadurai K,  Raghothama S,  Shamala N,  Balaram P.  Diproline templates as folding nuclei in designed peptides. Conformational analysis of synthetic peptide helices containing amino terminal Pro–Pro segments,  J. Amer, Chem. Soc. ,  2006, vol.  28 (pg.  7916- 7928) Google Scholar CrossRef Search ADS   Rudoi AB,  Shcherbakov RA.  Analysis of the chlorophyll biosynthetic system in a chlorophyll b-less barley mutant,  Photosynth. Res. ,  1998, vol.  58 (pg.  71- 80) Google Scholar CrossRef Search ADS   Rzeznicka K,  Walker CJ,  Westergren T,  Kannangara CG,  von Wettstein D,  Merchant S, et al.  Xantha-l encodes a membrane subunit of the aerobic Mg-protoporphyrin IX monomethylester cyclase involved in chlorophyll biosynthesis,  Proc. Natl Acad. Sci. USA ,  2005, vol.  102 (pg.  5886- 5891) Google Scholar CrossRef Search ADS   Sagromsky H.  Zur physiologischen Bedeutung von Chlorophyll b,  Biochem. Physiol. Pflanz. ,  1974, vol.  166 (pg.  95- 104) Google Scholar CrossRef Search ADS   Saito A,  Iino T,  Sonoike K,  Miwa E,  Higuchi K.  Remodeling of the major light-harvesting antennae protein of PSII protects the young leaves of barley (Hordeum vulgare L.) from photoinhibition under prolonged iron deficiency,  Plant Cell Physiol. ,  2010, vol.  51 (pg.  2013- 2030) Google Scholar CrossRef Search ADS PubMed  Sakuraba Y,  Tanaka R,  Yamasato A,  Tanaka A.  Determination of a chloroplast degron in the regulatory domain of chlorophyllide a oxygenase,  J. Biol. Chem. ,  2009, vol.  284 (pg.  36689- 36699) Google Scholar CrossRef Search ADS PubMed  Sakuraba Y,  Yamasato A,  Tanaka R,  Tanaka A.  Functional analysis of N-terminal domains of Arabidopsis chlorophyllide a oxygenase,  Plant Physiol. Biochem. ,  2007, vol.  45 (pg.  740- 749) Google Scholar CrossRef Search ADS PubMed  Schneegurt MA,  Beale SI.  Origin of the chlorophyll b formyl oxygen in Chlorella vulgaris,  Biochemistry ,  1992, vol.  31 (pg.  11677- 11683) Google Scholar CrossRef Search ADS PubMed  Simpson DJ,  Machold O,  Høyer-Hansen G,  von Wettstein D.  Chlorina mutants of barley (Hordeum vulgare L.),  Carlsberg Res. Commun. ,  1985, vol.  50 (pg.  223- 238) Google Scholar CrossRef Search ADS   Simpson DJ,  von Wettstein D.  Coordinator’s report: nuclear genes affecting the chloroplast. Stock list of mutants kept at the Carlsberg Laboratory,  Barley Genet. Newslett. ,  1992, vol.  21 (pg.  102- 108) Takahashi R,  Hayashi J,  Konishi T,  Moriya I.  Linkage analysis of barley mutants,  Barley Genet Newslett. ,  1975, vol.  5 (pg.  56- 60) Tanaka A,  Ito H,  Tanaka R,  Tanaka NK,  Yoshida K,  Okada K.  Chlorophyll a oxygenase (CAO) is involved in chlorophyll b formation from chlorophyll a,  Proc. Natl Acad. Sci. USA ,  1998, vol.  95 (pg.  12719- 12723) Google Scholar CrossRef Search ADS   Tanaka R,  Tanaka A.  Effects of chlorophyllide a oxygenase overexpression on light acclimation in Arabidopsis thaliana,  Photosynth. Res. ,  2005, vol.  85 (pg.  327- 340) Google Scholar CrossRef Search ADS PubMed  Thornber JP,  Highkin HR.  Composition of the photosynthetic apparatus of normal barley leaves and a mutant lacking chlorophyll b,  Eur. J. Biochem. ,  1974, vol.  41 (pg.  109- 116) Google Scholar CrossRef Search ADS PubMed  Tsuchiya T.  BGS 0117 Chlorina f2,  Barley Genet. Newslett. ,  1972, vol.  2 pg.  183  Tsuchiya T,  Robertson DW.  Association of a chlorina mutant with chromosome 3,  Barley Genet. Newslett. ,  1971, vol.  1 (pg.  64- 65) Tyutereva EV,  Voitsekhovskaja OV.  Responses of chlorophyll b-free chlorina 3613 barley mutant to a prolonged decrease in illuminance: 1. Dynamics of chlorophyll content, growth, and productivity,  Russ. J. Plant Physiol. ,  2011, vol.  58 (pg.  1- 8) Google Scholar CrossRef Search ADS   Tyutereva EV,  Voitsekhovskaja OV.  Responses of chlorophyll b-less chlorina 3613 mutant of barley to a prolonged decrease in illuminance: 2. Dynamics of carotenoids in leaf chloroplasts,  Russ. J. Plant Physiol. ,  2011, vol.  58 (pg.  218- 225) Google Scholar CrossRef Search ADS   Wang G,  Dunbrack RL.  PISCES: a protein culling server,  Bioinformatics ,  2003, vol.  19 (pg.  1589- 1591) Google Scholar CrossRef Search ADS PubMed  Wang J,  Yan Z,  Zhuo K,  Lu J.  Partial molar volumes of some alpha-amino acids in aqueous sodium acetate solutions at 308.15 K,  Biophys. Chem. ,  1999, vol.  80 (pg.  179- 188) Google Scholar CrossRef Search ADS PubMed  Waterhouse AM,  Procter JB,  Martin DMA,  Clamp M,  Barton GJ.  Jalview version 2: a multiple sequence alignment and analysis workbench,  Bioinformatics ,  2009, vol.  25 (pg.  1189- 1191) Google Scholar CrossRef Search ADS PubMed  Yamasato A,  Tanaka R,  Tanaka A.  Loss of the N-terminal domain of chlorophyllide a oxygenase induces photodamage during greening of Arabidopsis seedlings,  BMC Plant Biol. ,  2008, vol.  8 pg.  64  Google Scholar CrossRef Search ADS PubMed  © The Author 2012. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oup.com
TI - Characterization of Mutations in Barley fch2 Encoding Chlorophyllide a Oxygenase
JO - Plant and Cell Physiology
DO - 10.1093/pcp/pcs062
DA - 2012-04-25
UR - https://www.deepdyve.com/lp/oxford-university-press/characterization-of-mutations-in-barley-fch2-encoding-chlorophyllide-a-bBn4BY0REH
SP - 1232
EP - 1246
VL - 53
IS - 7
DP - DeepDyve
ER -