Identification and molecular characterization of a novel Chlamydomonas reinhardtii mutant defective in chlorophyll biosynthesis

The green micro-alga Chlamydomonas reinhardtii is an elegant model organism to study all aspects of oxygenic photosynthesis. Chlorophyll (Chl) and heme are major tetrapyrroles that play an essential role in energy metabolism in photosynthetic organisms and are synthesized via a common branched tetrapyrrole biosynthetic pathway. One of the enzymes in the pathway is Mg chelatase (MgChel) which inserts Mg 2+ into protoporphyrin IX (PPIX, proto) to form magnesium-protoporphyrin IX (MgPPIX, Mgproto), the first biosynthetic intermediate in the Chl branch. MgChel is a multimeric enzyme that consists of three subunits designated CHLD, CHLI and CHLH. Plants have two isozymes of CHLI (CHLI1 and CHLI2) which are 70%-81% identical in protein sequences. Although the functional role of CHLI1 is well characterized, that of CHLI2 is not. We have isolated a non-photosynthetic light sensitive mutant 5A7 by random DNA insertional mutagenesis that is devoid of any detectable Chl. PCR based analyses show that 5A7 is missing the CHLI1 gene and at least eight additional functionally uncharacterized genes. 5A7 has an intact CHLI2 gene. Complementation with a functional copy of the CHLI1 gene restored Chl biosynthesis, photo-autotrophic growth and light tolerance in 5A7. We have identified the first chli1 (chli1-1) mutant of Chlamydomonas reinhardtii and in green algae. Our results show that in the wild type Chlamydomonas CHLI2 protein amount is lower than that of CHLI1 and the chli1-1 mutant has a drastic reduction in CHLI2 protein levels although it possesses the CHLI2 gene. Our chli1-1 mutant opens up new avenues to explore the functional roles of CHLI1 and CHLI2 in Chl biosynthesis in Chlamydomonas, which has never been studied before.


Introduction
The green micro-alga Chlamydomonas reinhardtii possesses a photosynthetic apparatus very similar to that of higher plants, can grow photo-autotrophically and heterotrophically (it can metabolize exogenous acetate as a carbon source) and possesses a completely sequenced genome 1 . These attributes make it an elegant model organism to study oxygenic photosynthesis and chloroplast biogenesis 2,3 . In photosynthetic organisms, tetrapyrroles like Chl and heme are essential for energy metabolism (i.e. photosynthesis and respiration). Biosynthesis of Chl and heme occur via a common branched pathway that involves both nuclear-and chloroplast-encoded enzymes in most photosynthetic organisms 4 . In photosynthetic eukaryotes, 5-aminolevulinic acid (ALA) is synthesized from glutamine through glutamyl-tRNA 5 . Conversion of ALA through several steps yields protoporphyrin IX (PPIX), the last common precursor for both heme and Chl biosynthesis 5 . Ferrochelatase inserts iron in the center of PPIX thus committing it to the heme branch of the pathway. Insertion of Mg 2+ in PPIX by MgChel leads to Mgproto, the first biosynthetic intermediate in the Chl branch 6 . Magnesium chelatase has three subunits, which are CHLD, CHLH and CHLI 7 . The ATP-dependent catalytic mechanism of the heterotrimeric MgChel complex includes at least two steps 7-9 : an activation step, followed by the Mg 2+ insertion 8 . Activation of MgChel with ATP involves CHLD and CHLI while CHLH is required for the chelation step 10 . CHLI belongs to the AAA+ family of ATPases. Plants have two isozymes of CHLI1 (CHLI1 and CHLI2) which are 70%-81% identical in protein sequences 10 . Although the functional role of CHLI1 is well characterized, that of CHLI2 is not. Most of the data on CHLI comes from studies on Arabidopsis thaliana chli mutants and the functional significance of CHLI1 and CHLI2 has not been studied in green algae 10-16 . In Arabidopsis CHLI2 plays a limited role in Chl biosynthesis because of its lower expression level compared to that of CHLI1 12-15 . In Arabidopsis the CHLI2 protein amount is lower than that of CHLI1. When overexpressed, CHLI2 can fully rescue an Arabidopsis chli1chli2 double mutant 12 .
We have isolated the first (chli1-1) mutant of Chlamydomonas reinhardtii (5A7) which possesses an intact CHLI2 gene. Transformation of 5A7 with a functional copy of the CHLI1 gene restored Chl biosynthesis. Western analyses show that the CHLI2 protein level is lower than that of CHLI1 in the wild type strain and CHLI2 protein is barely detectable in the mutant strain. In this study, we present our molecular data on the identification of the mutation locus in 5A7 and its complementation.

Algal media and cultures
Chlamydomonas strains 4A+ (a gift from Dr. Krishna Niyogi (UC, Berkeley), 5A7/chli1-1 (generated by our laboratory) and chli1-1 rescued transformants (generated by our laboratory) were grown either in Tris-Acetate Phosphate (TAP) heterotrophic media or in Sueoka's High Salt (HS) photo-autotrophic media. TAP and HS liquid media and agar plates were prepared in the lab using reagents from Fisher Scientific (Pittsburg, PA) according to the protocol given in Gorman and Levine (1965) 17 and Sueoka (1960) 18 , respectively. The 4A+ strain and chli1-1 rescued transformants were maintained on TAP agar plates and TAP+zeocin (Sigma, St. Louis, MO) plates, respectively under dim light intensities (10-15 µmol photons m -2 s -1 ) at 25°C. The final zeocin concentration was 15 µg/ml). The chli1-1 mutant (5A7) was maintained in the dark on TAP 1.5% agar plates containing 10 µg/ml of paromomycin (Sigma, St. Louis, MO). Liquid algal cultures used for RNA and genomic DNA extractions and protein analyses were grown in 100 ml flasks on the New Brunswick Scientific Excella E5 platform shaker (Enfield, CT) at 150 rpm in the dark or in the dim light.
Generation of the 5A7 mutant The purified pBC1 plasmid from the DH5α Escherichia coli clone harboring the pBC1 plasmid (obtained from Dr. Krishna Niyogi's laboratory at UC, Berkeley) was used for random DNA insertional mutagenesis. This plasmid contains two antibiotic resistance genes: APHVIII and Amp R (Figure 1). APHVIII confers resistance against the antibiotic paromomycin (Sigma, St. Louis MO) and was used as a selection marker for screening of Chlamydomonas transformants. Amp R was used as a selection marker for screening of E. coli clones harboring the pBC1 plasmid. E. coli was grown in 1 l of Luria Bertani (LB) broth containing 1% tryptone, 0.5% of yeast extract, 1% NaCl and ampicillin [final concentration of ampicillin:100 µg/ml]. LB reagent was prepared in the laboratory using reagents purchased from Fisher (Pittsburgh, PA). Ampicillin was purchased from Fisher (Pittsburgh, PA). The culture was incubated at 37°C overnight. Plasmid purification from E. coli cells was facilitated by Qiagen plasmid mega kit according to the protocol given in the technical manual (Qiagen, Valencia, CA). Once purified from E. coli, the circular pBC1 vector was linearized with the restriction enzyme Kpn1 (NEB, Beverly, MA) according to the protocol given in the technical manual. The linearized DNA was purified using a QIAEX II gel extraction kit (Qiagen, Valencia, CA) according to the protocol given in the technical manual. All agarose DNA gel electrophoresis was visualized by BioRad Molecular Imager Gel Doc XR+ (BioRad, Hercules, CA). Transformation of parental strain 4A+ by the linearized pBC1 vector was performed utilizing the glass bead transformation technique described by Kindle et al. (1989) 19 and Dent et al. (2005) 2 . Transformants were plated onto fresh TAP agar plates containing 10 µg/ml paromomycin (TAP+P) in the dark. Single colonies of mutants were picked and transferred onto fresh TAP+P plates using a numbered grid layout. Screening of photosynthetic and pigment deficient mutants was done by visual inspection and monitoring of growth under different light intensities in heterotrophic, mixotrophic and photo-autotrophic conditions 2 .
Genomic DNA and RNA extraction 4A+, chli1-1 rescued transformants complements and 5A7/chli1-1 were grown in TAP liquid media in the dark to a cell density of

Changes from Version 1
We have incorporated all the suggestions of the reviewers in our revised version. We have changed the name of our chli1 mutant to chli1-1 and have changed the word "complements" to "rescued transformants". We have corrected the figure numbers for Figure 2 and Figure 6 in the manuscript text. We have also changed our discussion section with respect to the tetrapyrrole mediated nuclear gene expression changes in Chlamydomonas, according to Dr. Masuda's suggestion. Six new references have been added based on the changes made in the discussion section 32-37 .

See referee reports
about 5 × 10 6 cells/ml of the culture. Genomic DNA was purified using a phenol-chloroform extraction method 19 . RNA extraction was facilitated by TRIzol reagent from Invitrogen (Carlsbad, CA) following the protocol in the technical manual. DNA and RNA concentrations were measured using a Nanodrop 1000 spectrophotometer from Thermo Fisher Scientific (Wilmington, DE). DNase treatment was performed using Ambion's TURBO DNA-free kit from Invitrogen (Carlsbad, CA) following the protocol in the technical manual to remove genomic DNA from the RNA preparation. Generation of cDNA was performed using Life Technologies Superscript III First-Strand Synthesis System from Invitrogen (Carlsbad, CA) following the protocol in the technical manual.
Thermal asymmetric interLaced PCR TAIL (Thermal Asymmetric InterLaced) PCR was implemented, following the protocol of Dent et al. (2005) 2 . This protocol was implemented with one modification of utilizing a non-degenerate primer (AD2) derived from the original degenerate primer (AD) for TAIL PCR as this non-degenerate primer was giving us optimum yield without generating excess nonspecific products.HotStar Taq Plus DNA polymerase kit reagents (Qiagen, Valencia, CA) were used for PCR. The PCR reaction mixture consisted of 1× PCR buffer, 200 µM of each dNTP, 1× Q-solution, 2.5 units of HotStar Taq Plus DNA polymerase, 60 pmoles of the non-degenerate primer AD2 and 5 pmol of the APHVIII specific primer. Primers were ordered from IDT (Skokie, IL; Table 1). The non-degenerate primer AD2 has a T m of 46°C while the three APHVIII specific primers used had T m ranging from 58°C to 64°C. PCR cycling programs were created using the program given in Dent et al. (2005) 2 . TAIL1 PCR product was diluted 10 and 25-fold and 2 µl of the diluted TAIL1 PCR product was used for TAIL2 PCR reactions. The TAIL2 PCR product was gel purified using QIAEX II gel extraction kit (Qiagen, Valencia, CA) according to the protocol given in the technical manual. The purified TAIL2 PCR product was sequenced at the UC, Berkeley DNA Sequencing Facility (Berkeley, CA).
Genomic and reverse transcription PCR Primers were designed based on genomic DNA sequences available in the Chlamydomonas genome database in Phytozome. Amplifications of genomic DNA and cDNA were executed using MJ Research PTC-200 Peltier Thermal Cycler (Watertown, MA). HotStar Taq Plus DNA polymerase kit (Qiagen, Valencia, CA) was used for PCR following the cycling conditions given in the Qiagen protocol booklet. Annealing temperature was between 55 and 60°C depending on the T m of the primers. Extension time was varied according to the size of the PCR product amplified. Final extension was set at 72°C for ten minutes. All genomic and reverse transcription PCR products were amplified for a total of thirty-five cycles. 50-150 ng of genomic DNA or cDNA templates were used for PCR reactions. For semiquantitative RT-PCR using CHLI1 and CHLI2 gene specific primers, 3 µg of total RNA was converted into cDNA and then 150 ng of cDNA templates were used for RT-PCR. Sequences of primers used for genomic and RT-PCR are shown in Table 2, Table 3 and Table 4.
Cloning of the CHLI1 cDNA in the pDBle vector The pDBle vector (obtained from Dr. Saul Purton, University College London, UK) was double-digested with restriction enzymes EcoR1 and Nde1 (NEB, Beverly, MA) according to the protocol given in the technical manual. The CHLI1 cDNA template was amplified using primers given in Table 4. Ligation of the double digested (NdeI and EcoR1 digested) CHLI1 cDNA and the NdeI/EcoRI doubledigested pDBle vector was done using the T4 ligase and 1 mM ATP (NEB, Beverly, MA). Chemically competent (CaCl 2 treated) E. coli cells were used for transformation. After transformation, E. coli cells were plated on LB+ampicillin (100 µg/ml) plates and incubated at 37°C overnight. Single colonies were picked the next Figure 1. Linearized pBC1 plasmid used for random insertional mutagenesis. The cleavage site of the Kpn1 restriction enzyme, used for linearization of the vector is shown. APHVIII is under the control of combo promoters consisting of the promoter of the gene encoding the small subunit of Rubisco (RbcS2) and the promoter of the gene encoding the heat shock protein 70A (Hsp70A). pBC1 is a phagemid and its F1 origin (F1 ori) and pUC origin (pUC ori) are shown. The size of the plasmid is 4763 bp. Table 1. List of primers used for Thermal Asymmetric InterLaced (TAIL) PCR, verification of TAIL-PCR product and DNA sequencing. These primers were used to generate the data in Figure 3 and Figure 4.

Primer name Sequence of primer Gene
5´-TTAGAAGCACTTCCGGTGCACGAT-3´Actin Table 4. List of primers for amplifying CHLI transcripts and complement testing. These primers were used in the experiments that generated the data in Figure 7 and Figure 10 and also used for CHLI1 cDNA amplification for cloning.

Western analysis
Electrophoretic transfer of the SDS-PAGE resolved proteins onto an Immobilon P-PVDF membrane (Millipore, Billerica, MA) was carried out for 2 hours at a constant current of 400 mA in the transfer buffer (25 mM Tris, 192 mM glycine and 20% methanol). The CHLI1 polyclonal antibody was raised in rabbit against the full length Arabidopsis thaliana CHLI1 mature protein that lacks the predicted transit peptide 24 . This antibody is a gift from Dr. Robert Larkin (Michigan State University). CHLI1 primary antibodies were diluted to a ratio of 1:2,000 before being used as a primary probe. The secondary antibodies used for Western blotting were conjugated to horseradish peroxidase (Pierce protein research product, Thermo Fisher Scientific, Rockford, IL) and diluted to a ratio of 1:20,000 with the antibody buffer. Western blots were developed by using the Supersignal West Pico chemiluminescent substrate kit (Pierce protein research product, Thermo Fisher Scientific, Rockford, IL).

Cell counts and chlorophyll extraction
Cell density (number of cells per ml of the culture) was calculated by counting the cells using a Neubauer ultraplane hemacytometer (Hausser Scientific, Horsham, PA). Pigments from intact cells were extracted in 80% acetone and cell debris was removed by centrifugation at 10,000 g for 5 minutes.

Results
Generation and identification of the mutant 5A7 Mutant 5A7 was generated by random insertional mutagenesis of the Chlamydomonas reinhardtii wild type strain 4A+ (137c genetic background). 5A7 lacks detectable chlorophyll, appears yellowishbrown in color and grows only under heterotrophic conditions in the dark or in the dim light in the presence of acetate in the growth media ( Figure 2). It is incapable of photosynthesis and is sensitive to light intensities higher than 20 µmol photons m -2 s -1 (Figure 2A).

Molecular characterization of the mutation in 5A7
The linearized plasmid pBC1 was used to generate 5A7 (Figure 1). To find the insertion of the APHVIII end of the plasmid in 5A7, a modified TAIL (Thermal Asymmetric InterLaced) PCR method was used. Figure 3A shows the position of the vector specific TAIL PCR primers and also shows the arbitrary position of the random nondegenerate primer. A 850 bp DNA product from TAIL2 PCR was purified from the agarose gel ( Figure 3B, Table 1). This purified DNA product was used for PCR using internal primers specific to the 3´UTR (UnTranslated Region) of the APHVIII gene. The PCR results confirmed that the 850 bp DNA product contains the 3´UTR of the APHVIII gene ( Figure 3C). Sequencing of the 850 bp TAIL2 PCR product revealed that the APHVIII end of the plasmid has been inserted in the fourth exon of a hypothetical gene which we have named as UP6 (Figure 4). UP6 (Cre06.g306500) is located on chromosome 6. Figure 5 shows a schematic map of the UP6 locus with its eight neighboring genes UP4 (Cre06.g306100), UP3 (Cre06.g306150), UP1 (Cre06.g306250), UP2 (Cre06.g306200) CHLI1 (Cre06. g306300), FDX3 (Cre06.g306350), AMT (g7098) and UP5 (Cre06. g306450). It is to be noted that we have named all of these genes arbitrarily for our study except for the CHLI1 and FDX3 genes, which were annotated in the Chlamydomonas genome database.
Readers are requested to identify these unknown genes by the gene locus number (Cre or g number) in the Phytozome database. PCR analyses with the genomic DNA of 4A+ and 5A7 were performed using primers specific to four neighboring genes upstream of the CHLI1 (including UP6) and four neighboring genes downstream of the CHLI1 locus (Table 2 and Table 3; Figure 6A and 6B). PCR analyses revealed that all eight genes neighboring the CHLI1 locus were deleted or displaced from their native location ( Figure 6A and 6B). UP5 primers gave nonspecific multiple products in 5A7 ( Figure 6B). The first two exons of UP6 are present in the 5A7 genome as the UP6 primers spanning the first and the second exon, gave similar genomic DNA PCR product of the expected size as in the 4A+ lane ( Figure 6B). Reverse transcription (RT-PCR) analyses using the same UP6 primers on 5A7 and 4A+ cDNA did not yield a PCR product in 5A7 unlike that in 4A+ ( Figure 6C; Table 3). This shows that the insertion of the plasmid in the fourth exon of UP6 in 5A7 has hampered the transcription of the UP6 gene.   Taken together the data shows that at least a 35,715 bp genomic region has been deleted and or/displaced when the plasmid got inserted in the 5A7 genome. Except for the CHLI1 gene, the functions of the remaining eight genes (including UP6) are not known. We do not yet know the exact location of the pUC origin (pUC ori) end of the plasmid (Figure 1) in the 5A7 genome.
Checking for the absence/presence of the CHLI1 transcript and the CHLI2 gene and transcript As CHLI plays a role in Chl biosynthesis, we checked for the presence/absence of the CHLI1 and CHLI2 in 5A7. RT-PCR results show that CHLI1 transcript is absent and CHLI2 transcript is present in 5A7 ( Figure 7A, Table 4). Figure 7B shows the presence of the CHLI2 gene in 5A7.

Complementation of 5A7
We will be referring to strain 5A7 as chli1-1 from here onward. As our chli1-1 lacks Chl and CHLI1 is involved in Chl biosynthesis, we cloned the CHLI1 cDNA in the pDBle vector to transform chli1-1 (Figure 8, Table 4). CHLI1 expression is driven by the constitutive PsaD promoter in the CHLI1-pDBle construct (Figure 8). pDBle has two Ble genes that confer resistance to the antibiotic zeocin. Figure 9 shows growth phenotypes of two chli1-1 rescued transformants (chli1-7 and chli1-8); chli1-1 and 4A+. chli1-1 rescued transformants are able to synthesize Chl, are not light sensitive and are capable of photosynthesis ( Figure 9). As the chli1-1 rescued transformants harbor the Ble gene (from the pDBle vector) and APHVIII gene (derived from the parental strain chli1-1), they can grow both on zeocin and paromomycin media plates unlike chli1-1 and 4A+ (Figure 9).
Chl analyses show that both chli1-1 rescued transformants are about 33-46% Chl deficient. chli1-1 rescued transformants have a similar Chl a/b ratio as that of the wild type (Table 5, Data File below). Figure 10A and 10B show a schematic figure of the native Chlamydomonas CHLI1 gene and the trans CHLI1 gene used for complementation, respectively. PCR analyses using the genomic DNA show that the chli1-1 rescued transformants have the trans CHLI1 gene ( Figure 10C and 10D). In Figure 10D the genomic DNA PCR product sizes in the two chli1-1 rescued transformant lanes are smaller than that in the 4A+ lane as we have cloned the CHLI1 cDNA for complementation. The Chlamydomonas CHLI1 protein has about 71% sequence identity to the Arabidopsis CHLI1 protein. Figure 11A shows a stained protein gel. The two chli1-1 rescued transformants and the 4A+ were loaded on an equal Chl basis in each lane in the protein gel ( Figure 11A). As chli1-1 lacks Chl, the maximum amount of protein (40 µg) that can be loaded in a mini protein gel, was used ( Figure 11A). Light harvesting complex proteins (LHCs) can barely be detected in the chli1-1 mutant ( Figure 11A).  (Table 1), specific to the 3´UTR of the APHVIII gene was used for sequencing the 850 bp Thermal Asymmetric InterLaced 2 (TAIL2) PCR product. 3´UTR sequence of APHVIII is in bold black, extra nucleotide additions are in bold blue. The flanking Chlamydomonas UP6 genomic sequence is denoted in red. The APHVIII end of the plasmid has been inserted after the eighth nucleotide in the fourth exon of UP6 gene. Western analyses of the two chli1-1 rescued transformants with a CHLI1 antibody show that the CHLI1 protein is absent in the chli1-1 mutant but present in the chli1-1 rescued transformants ( Figure 11B). Western analyses also show that the Arabidopsis CHLI1 antibody detects both the CHLI1 (40 kDa) and CHLI2 (42 kDa) protein in Chlamydomonas as the Chlamydomonas CHLI2 has about 62% sequence identity to the Arabidopsis CHLI1 ( Figure 11B). In the wild type the CHLI2 protein amount is much lower than that of CHLI1. As the chli1-1 rescued transformants are Chl deficient Odd numbered lanes denote genomic DNA PCR products; even numbered lanes denote cDNA products. All primers used spanned an intron. M denotes 50 bp DNA ladder. All DNA samples were run on a 1.8% agarose gel. Gene names are given at the bottom of the gel. Primer sequences are shown in Table 2 and Table 3.  Table 4.  Table 4. CHLI1 expression is driven by the constitutive PsaD promoter. NdeI and EcoRI restriction sites are labeled. pDBle contains two copies of Ble R genes driven by the Rubisco (RbcS2) promoter. The size of the CHLI1-pDBle construct is 7957 bp. Black arrow and white arrow denotes CHLI1 cDNA and Ble R gene, respectively. Grey boxes denote UnTranslated regions (UTR). lower expression of the CHLI1 protein in these chli1-1 rescued transformants ( Figure 11). Semi-quantitative RT-PCR shows that the CHLI2 transcript level in chli1-1 is much lower than that in the wild type strain (Figure 7). Western analyses show that the CHLI2 protein level is severely reduced in the chli1-1 mutant (Figure 11). Real Time PCR analyses can be used to confirm whether the reduction in the CHLI2 protein level is due to a low abundance of the CHLI2 transcript. Additionally, the roles of any of the other missing eight genes in Chl biosynthesis cannot be ruled out as currently the functions of these genes are unknown.
In Arabidopsis, it has been shown that CHLI2 does play a limited role in Chl biosynthesis in the absence of CHLI1 12-15 . chli1-1 possesses an intact CHLI2 gene but the CHLI2 protein is barely detectable in the mutant. This raises two questions: 1) Is the low abundance of the CHLI2 protein a general effect or is it a specific effect of the CHLI1 mutation?
2) Is the total absence of Chl in strain chli1-1 due to the specific absence of CHLI1 or due to the absence of both CHLI1 and the near absence of CHLI2 protein?  Table 4. compared to the wild type, the two rescued transformant lanes show higher amount of protein loadings ( Figure 11A). Although more protein was loaded in the chli1-1 lane in the protein gel compared to that in the 4A+ and the chli1-1 rescued transformant lanes, the CHLI2 protein was barely detectable in 5A7 ( Figure 11B).

Discussion
5A7 is the first chli1-1 mutant to be identified in C. reinhardtii and in green algae. CHLI1 deletion has affected Chl biosynthesis and photosynthetic growth in the chli1-1 mutant (Figure 2). Overaccumulation of photo-excitable PPIX leads to photo-oxidative damage to the cells in presence of light and oxygen 4-6 . The light sensitivity of the chli1-1 is most probably due to an over-accumulation of PPIX which occurs due to the inactivity of MgChel enzyme which converts PPIX to MgPPIX. Future HPLC (High Performance Liquid Chromatography) analyses of steady state tetrapyrrole intermediates will confirm this hypothesis.
Based on the current molecular analyses, our chli1-1 mutant has a deletion of at least nine genes (including the CHLI1 gene). Currently we are investigating the exact insertion point of the pUC ori end of the plasmid in the chli1-1 genome (Figure 1). This will provide us with a precise estimate of the number of gene deletions in chli1-1. Although complementation of chli1-1 with the CHLI1 gene restored Chl biosynthesis, tolerance to high light levels and photo-autotrophic growth, chli1-1 rescued transformants are still Chl deficient to some extent (Table 5). This is probably due to a In Chlamydomonas, hemin and MgPPIX has been shown to induce global changes in nuclear gene expression in Chlamydomonas, unlike that in Arabidopsis 27-28,32-37 . In Chlamydomonas, the above mentioned tetrapyrroles altered expressions of genes encoding TCA cycle enzymes, heme binding proteins and stress response proteins as well as proteins involved in protein folding and degradation (eg. heat shock proteins) 34-37 . Hence the roles of tetrapyrroles in retrograde signaling appears to be distinct in green algae and in higher plants. In summary, in the future our chli1-1 mutant can be used to clarify the functional role of CHLI1 and CHLI2 in Chl biosynthesis in C. reinhardtii.
Author contributions MM and PG conceived the study, designed the experiments and took the lead role in preparing the manuscript. DW generated the mutant chli1-1 and isolated genomic DNA from the strains. PG did the PCR analyses, isolated genomic DNA, performed growth and Chl analyses of different strains, prepared cDNA, cloned the CHLI1

Open Peer Review
Reviewer Report 28 August 2013 https://doi.org/10.5256/f1000research.1956.r1587 © 2013 Masuda T. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Tatsuru Masuda
Environmental Sciences, University of Tokyo, Tokyo, Japan I have reviewed the manuscript by Grovenstein et al. In this manuscript, the authors identified a Chl deficient mutant of Chlamydomonas reinhardtii by insertional mutagenesis.
As described in the manuscript, Arabidopsis possesses two CHLI isoforms, CHLI1 and CHLI2, like Chlamydomonas. However, it is not clear whether these isoforms are paralogous or not among these two photosynthetic organisms. The authors should perform phylogenetic analysis to show whether these isoforms are truly phylogenetically related or not.
Considering the chli1-1 mutant is Chl deficient, it is reasonable to assume that CHLI2 is not functional in Chlamydomonas. In my understanding, the expression of CHLI2 is not affected by the mutation of CHLI1 gene in Arabidopsis. The upper band of Fig.11B could be a non-specific band of CHLI1 protein that hides the original band of CHLI2 in Western blot analysis. In this sense, the expression analysis of CHLI2, as well as functional complementation with CHLI2 in the chli1-1 mutant, are apparently necessary for further understanding.
Specific comments: In Fig. 7, the authors performed semi-quantitative RT-PCR analysis of CHLI1 and CHLI2. As the authors discussed quantities of these transcripts in the Discussion section (p.10), it is necessary to show the levels of housekeeping genes such as ACTIN8 or Ubiquitin10 as loading control.