A robust protocol for efficient generation, and genomic characterization of insertional mutants of Chlamydomonas reinhardtii
© The Author(s) 2017
Received: 19 October 2016
Accepted: 22 March 2017
Published: 3 April 2017
Random insertional mutagenesis of Chlamydomonas reinhardtii using drug resistance cassettes has contributed to the generation of tens of thousands of transformants in dozens of labs around the world. In many instances these insertional mutants have helped elucidate the genetic basis of various physiological processes in this model organism. Unfortunately, the insertion sites of many interesting mutants are never defined due to experimental difficulties in establishing the location of the inserted cassette in the Chlamydomonas genome. It is fairly common that several months, or even years of work are conducted with no result. Here we describe a robust method to identify the location of the inserted DNA cassette in the Chlamydomonas genome.
Insertional mutants were generated using a DNA cassette that confers paromomycin resistance. This protocol identified the cassette insertion site for greater than 80% of the transformants. In the majority of cases the insertion event was found to be simple, without large deletions of flanking genomic DNA. Multiple insertions were observed in less than 10% of recovered transformants.
The method is quick, relatively inexpensive and does not require any special equipment beyond an electroporator. The protocol was tailored to ensure that the sequence of the Chlamydomonas genomic DNA flanking the random insertion is consistently obtained in a high proportion of transformants. A detailed protocol is presented to aid in the experimental design and implementation of mutant screens in Chlamydomonas.
Over the past decade, Chlamydomonas reinhardtii (hereafter referred to as Chlamydomonas) has been successfully used as a model system to help answer biological questions related to a wide variety of cellular processes. With a sequenced genome and a growing experimental toolbox to facilitate large-scale forward and reverse genetic studies, this unicellular microalga now provides an even stronger functional genomics template for the further dissection of biological processes, and both metabolic and regulatory pathways. Experimental work using Chlamydomonas will not only contribute to our increased understanding of its own physiology and biochemistry, but will continue to reveal the genetic basis of similar processes in other organisms such as bacteria, fungi, vascular plants, animals and even humans.
A valuable resource for studying biological processes is the availability of stable mutations that disrupt key genes that encode components of those processes. In Chlamydomonas, insertional mutagenesis has been routinely used for this purpose, and with great success. A large number of transformants can be generated using this technique with the goal of tagging a single functional gene within the nucleus of each transformant. When transforming nuclear DNA, a short DNA cassette consisting of an antibiotic resistance gene marker flanked by endogenous promoter and terminator sequences is used to ensure optimum marker gene expression and prevent transcriptional read through. In Chlamydomonas, the absence of homologous recombination in the nuclear genome means that insertion of the antibiotic cassette occurs at random genomic sites. Identification of the location of the insert in the genome is therefore pivotal to the success of this method. In forward genetic studies, the position of the inserted DNA in mutants with phenotypes of interest would help identify genes disrupted in the transformants. However, this approach could also be valuable for reverse genetic screens since there are many Chlamydomonas genes, both with known and unknown functions, in which knockouts might not result in readily discernible phenotypes.
With the current availability of a sequenced and largely annotated nuclear genome [1, 2], a relatively short sequence flanking the genomic insert is often provides enough information to use the genomic database to identify the disrupted genomic locus. Several techniques were successfully used in the past to identify insertion sites within the genome. These involved modified protocols for plasmid rescue , Thermal asymmetric interlaced PCR or TAIL PCR , Restriction enzyme site-directed amplification PCR or RESDA-PCR , 3′-Rapid Amplification of cDNA ends or 3′RACE  and Site Finding PCR . Recently, a high throughput Mme1-based Insertion site Sequencing strategy for Chlamydomonas insertional mutants, called ChlaMmeSeq [8, 9], was used for the simultaneous screening of large numbers of mutagenic insertion sites. Keeping in mind the available technology used with varying degrees of success in Chlamydomonas, this study proposes the use of a different protocol aimed at the successful recovery of genomic regions flanking an insert within transformants generated in large scale insertional mutagenesis efforts. This protocol is based on an adaptor-linked PCR method that has been modified and refined for success with Chlamydomonas. It is robust, time efficient, and identifies the location of inserts in the majority of transformants. This method also detects the number of insertions and their direction, and any deletions/rearrangements at the insertion site.
Adaptor linked PCR has been used with prokaryotes  and eukaryotes for genome walking, as well as to identify the location of T-DNA inserts in the Arabidopsis genome . This procedure usually involves restriction and blunting of genomic DNA, followed by ligation of an asymmetric adaptor DNA oligonucleotide that has been modified to prevent self-ligation. The adaptor linked to the ends of genomic DNA provides a template for designing primers based on known sequences of the adaptor and of the inserted cassette for a series of nested PCRs with varying stringency. These PCRs produce DNA fragments that are sequenced and aligned to the nuclear genome sequence. In this study, the adaptor linked PCR method was modified for efficient use in determining insert locations in Chlamydomonas transformants derived from a large-scale mutagenesis effort. Several combinations of restriction enzymes and PCR conditions were tested to provide a robust and relatively fail-safe method of determining insert location on the Chlamydomonas nuclear genome.
A detailed protocol is provided to aid researchers interested in high throughput determination of insert locations with a population of transformants. Several recommendations in areas ranging from primer design to the use of combinations of restriction enzymes that might further increase the probability of insert recovery are presented. Many shortcuts have also been suggested such as one step digestions and ligations, rapid methods for agarose gel electrophoresis, and the use of shorter PCR cycles to reduce both time and cost. The execution of the technique proposed in this study will contribute to both the generation of genome wide mutant libraries and the characterization of the insertion sites.
Generation of the insertional DNA fragment
The primers used for this work
Used to amplify cassette
TGT GTG GAA TTG TGA GCG G
Used to amplify cassette
CTT TCC ATC GGC CCA GCA
3′ insert primer
CGG TAT CGG AGG AAA AGC TG
3′ insert primer
GCT GTT GGA CGA GTT CTT CTG
5′ insert primer
TTC CAA GCG ATC ACC AGC AC
5′ insert primer
GCT GGC ACG AGT ACG GGT TG
5′ insert primer
AGC TTT TGT TCC CTT TAG TG
GTA ATA CGA CTC ACT ATA GAG T
ACT ATA GAG TAC GCG TGG T
5′ insert primer pSL18
GCC CTC ATA GCC CGC CAA ATC AG
5′ insert primer pSL18
AAG CCG ATA AAC ACC AGC CC
Composition of the 50 µL PCR reaction mixture used to amplify the 1812 bp insertional DNA fragment
10× polymerase buffer #3
2.5 mM dNTP mix
RIM-f2 (20 µM)
RIM-r1 (20 µM)
pSL72 plasmid template
Polymerase mix (5 U µL−1)
Preparation of the DNA insert
After electrophoresis in an 0.8% (W/V) agarose gel, the DNA fragments were purified from the gel by excising the 1812 bp fragment (PCR generated) or the 1813 bp fragment (restriction generated) with a razor blade (visualized with ethidium bromide and a low energy UV light). DNA from the gel slices were purified using a gel extraction kit (Qiagen). The purified DNA was filter sterilized and quantified by comparing the fluorescence of the DNA to that of a known DNA mass ladder (HindIII digested lamda DNA (New England Biolabs); or a 1 Kb ladder (New England Biolabs).
Preparing chlamydomonas cells for transformation
500 mL of TAP medium  was inoculated with a 4 mm spherical scraping (a small green pea volume equivalent) of D66 cells from a TAP agar plate. When cells reached a density of approximately 3 × 106 cells mL−1 they were harvested by centrifugation at 2500×g and resuspended in a final volume of 3.5–4.0 mL TAP containing 60 mM sorbitol to achieve a final cell concentration of 2–4 × 108 cells per mL. Resuspended cells were transferred to a 15 mL Corning centrifuge tube (Corning #430790) and placed on ice. Electroporation cuvettes (Bio-Rad #165-2091) with a gap-width of 0.4 cm were used for electroporation. 250 µL of cells was transferred to a sterile cuvette with no added DNA, labeled “No DNA control”, and placed on ice. Approximately 2.5 µg of the prepared insert DNA fragment was added to the remaining 3.25–3.75 mL of suspended cells, gently inverted 5–6 times to make the DNA/cell mixture homogeneous and then placed on ice. 250 µL aliquots of the DNA/cell suspension were added to the sterile electroporation cuvettes, yielding approximately 14 transformation reactions. All cuvettes were placed on ice for 10–20 min before electroporation. This step is critical in obtaining a pulse rate that does not lyse the cells (see below). Care must be taken to ensure the cells do not warm up before the pulse is generated.
Electroporation of chlamydomonas
As stated above, 250 µL containing between 0.5 × 108 and 1 × 108 cells with 180 ng of the paromomycin cassette were electroporated using the Bio-Rad Gene Pulser II system, as modified from the method reported by Shimogawara et al. . One significant modification was that no carrier DNA was added. A voltage setting of 0.8 kV, a capacitor setting of 25 µF, and no shunt resistor, was used for electroporation. The measured pulse time, an excellent predictor of the success of the electroporation transformation, generally ranged from 10 to 13 ms. If the pulse time was shorter, the transformation efficiency decreased approximately 100-fold as cells were lysed during the pulse. The cuvettes were placed at room temperature for 5 min following electroporation and prior to transfer to overnight recovery medium.
Overnight recovery of electroporated cells
Within 30 min of electroporation, the mixture from each cuvette was transferred using a 200 µL wide orifice pipet tip (E&K #3502-R96S) (to prevent shearing of the cells) to 10 mL of TAP medium containing 60 mM sorbitol in a 15 mL Corning centrifuge tube (Corning #430790). The tubes were placed in low light (10–20 µmol photons m−2 s−1), gently rocked to keep the cells suspended, and allowed to recover overnight (12–16 h). Used electroporation cuvettes were rinsed with water and stored in 100% ethanol until required for another round of transformations. As long as the cuvettes did not crack they were reused several times. Cuvettes were dried in a sterile hood immediately before use.
Plating recovered electroporated cells on selective medium
Genomic DNA preparation
Total DNA was isolated from 50 mL of mutant cells grown in TAP medium under continuous low light (50–80 µmol photons m−2 s−1) according to Newman et al.  with several modifications. Briefly, cells were pelleted by centrifugation at 2500×g in 50 mL sterile centrifuge tubes (Corning #430828) and resuspended in 400 µL of dH20 in two 1.5 mL Eppendorf tubes and 800 µL disruption buffer containing SDS was added (2% SDS, 400 mM NaCl, 40 mM EDTA, 100 mM Tris–HCl, pH 8.0). The nucleic acids were extracted three times using a phenol/chloroform/isoamyl alcohol mixture until the interface of the inorganic and organic layers contained no residual protein residue. The aqueous phase was then extracted a final time with chloroform. Nucleic acids were then precipitated with two volumes of ethanol and washed twice with 70% ethanol. The pellet was air-dried for 10 min and dissolved in 100 µL of TE (10 mM TRIS, pH7.5, 1 mM EDTA). See Additional file 1 for a more detailed description of the DNA preparation protocol.
Preparation of blunt-ended restriction digest fragments of genomic DNA
200 ng of genomic DNA was digested with a mixture of AleI (10 units), NaeI (10 units), PmlI (10 units) and PvuII (1 unit) restriction endonucleases (New England Biolabs) in a volume of 100 µL in NEB buffer 2 supplemented with 100 µg mL−1 BSA. The reactions were incubated at 37 °C for 16–18 h. These four restriction endonucleases were chosen because they do not recognize sequences in the insertion DNA fragment and because they recognize sequences that occur, on average, every 200 bps in the C. reinhardtii genome, creating blunt-ended fragments. The last hour of the reaction was supplemented with 1 µL RNAse to degrade RNA. After the incubation, 5 µL of digested genomic DNA was separated by agarose gel electrophoresis to verify that the digestions were complete, which was observable as a smear of DNA from approximately 1 to 6 kb on a 1% agarose gel. Digested DNA was extracted once with an equal volume (95 µL) of phenol:chloroform:isoamyl alcohol (25:24:1 v/v), and once with chloroform:isoamylalcohol (24:1 v/v), precipitated with three volumes (285 mL) of ethanol, the pellet washed once with ice-cold 80% (V/V) ethanol, air-dried for 10 min and then resuspended in 20 µL of TE.
Ligation of the adaptor to the digested DNA
The adaptor was ligated to the digested genomic DNA overnight (16–20 h) at 16 °C. The ligation reaction consisted of 2.0 µL of adaptor, 4 µL of digested genomic DNA, 2 µL of 10× New England Biolabs ligase buffer, 1 µL of T4 DNA ligase (NEB #M0202S). The reaction was stopped by incubating the mixture at 80 °C for 20 min. Seventy µL of TE was added to the reaction before use as template during the PCR amplification.
Alternative one-step restriction digest and ligation reaction
Alternatively, the genomic DNA was restricted and ligated to the adaptor in a one-step reaction. The one step reaction was performed in the restriction enzyme buffer (NEB#2) with the addition of 10 µM ATP. The reaction was performed overnight at room temperature.
PCR amplification from the insertion to the flanking adaptor
Results and discussion
This report details a method that enables researchers to generate insertional mutants and to successfully identify the location of the paromomycin resistance cassette in the Chlamydomonas genome over 75% of the time. The method is rapid, fairly inexpensive and does not require exotic equipment, except for the electroporator. This method differs significantly from other methods which primarily use TAIL PCR or a modification of that method [4, 5]. First, the transformation method described in this report was electroporation versus glass beads, to reduce genomic deletions. Secondly, a DNA cassette was used containing only the paromomycin resistance gene instead of a plasmid [4, 5]. The use of a cassette makes it easier to identify the ends of the inserted DNA and makes the adaptor method a viable choice for researchers. In addition, a combination of restriction enzymes is described that enhances the likelihood of cutting the flanking genomic DNA within 1000 bp of the insert, increasing the success of obtaining a PCR fragment long enough to identify the site of insertion. Finally, adaptor PCR was used instead of TAIL PCR, which offers an additional tool to use if a researcher is having trouble with one method. Recently, a library of Chlamydomonas insertional mutants became available to researchers . This major advance will allow researchers to order Chlamydomonas strains containing an insert in specific genes. However, for gene discovery, scientists will still need to generate and screen new insertional mutants in their own laboratories. After conducting these screens and selecting strains with the desired phenotype, researchers need to be able to identify the disrupted gene. Furthermore, the new insertion library is not complete (less than 40% of the genes covered by more than one allele, and only a small number of the mutants have been validated). Finally, the method for generating insertional mutants described here will result in a high percentage of the colonies having single, simple insertions with few large genomic deletion. This transformation procedure also yields a relatively low number of strains with multiple DNA insertions. We found that multiple inserts were present in 11 out of the 156 mutants (~6%), although some insertions will not be detected using this method. Zhang et al.  observed that about 15% of the paromomycin resistant transformants had more than one insert when using a similar transformation protocol. We also observed very few deletions of genomic DNA flanking the inserts. One disadvantage of the glass bead transformation method is that it sometimes results in large DNA deletions resulting in the loss of more than one gene, making the results much harder to analyze [17, 18]. The electroporation method appears to be less likely to cause these large deletions although deletions have been reported using this method also [19, 20]. While we did not find large deletions in the transformants that we characterized, we were unable to recover the other side of the insert in 25% of the colonies, which could indicate the occurrence of large deletions or rearrangements.
Cosegregation of the DNA insert with the phenotype being studied
For any study using insertional mutagenesis, cosegregation of the insert with the desired phenotype must be genetically demonstrated. Genetic linkage was investigated in 15 strains generated using this method and paromomycin resistance cosegregated with the SLC phenotype about 40% of the time (6 of 15 transformants). Cells with an SLC phenotype grow normally on elevated CO2 concentrations but more slowly than wild-type cells at low CO2 concentrations. This rate of cosegregation is similar rates reported by others . Since random insertion involves double strand DNA breaks the generation of insertions and point mutations is likely. Clearly it remains essential that a genetic analysis of any interesting insertional mutant be done before proceeding with a complete physiological characterization of the mutant. Tetrad analysis was used in this study but random spore analysis could also be used. It is also critical to demonstrate rescue of the mutant phenotype by introduction of a wild type copy of the disrupted locus.
Earlier versions of the transformation procedure described here were used to generate and characterize insertional mutants in two large scale experiments. In one study, over 30,000 insertional mutants were screened for aberrant responses to sulfur limitation (SAC) . In a separate investigation, again over 30,000 insertional mutants were generated and screened for a ‘sick in low carbon dioxide’ phenotype (SLC). In some cases, mutants showing the desired phenotype were subjected to adaptor PCR to determine the genomic location of the paromomycin resistance cassette in their genomes. In the study in which the cells were screened for aberrant responses to sulfur limitation, a number of novels genes were discovered, including proteins involved in the responses of Chlamydomonas to sulfur deprivation [19, 21]. In the screen for mutants unable to grow photoautotrophically on low CO2, insertions in CIA6 , bestrophin, MITC11 and LCI9  and CIA8 (Machingura, Bajsa-Hirschel and Moroney, unpublished) were identified using the forward genetics approach described by González-Ballester et al. . No large deletions were observed in these studies.
In this communication we have detailed a method to generate insertional mutants that will have mostly single simple DNA inserts. We have described an adaptor-PCR based method that reliably can identify the location of the cassette in Chlamydomonas. Using this method, we were able to identify the genomic DNA flanking the insertion over 75% of the time. The combination of employing the electroporation method to generate insertional mutants in conjunction with the adaptor method should provide researchers using Chlamydomonas an excellent chance to quickly generate and characterize useful insertional mutant strains at a relatively low cost.
SVP, ARG and JVM designed the method, all authors conducted the experiments and helped write the manuscript. All authors read and approved the final manuscript.
The authors thank Susan Laborde for excellent technical assistance.
The authors declare that they have no competing interests.
Consent for publication
All authors have consented for publication.
Supported by NSF Award IOS1146597 and subcontract from the University of Illinois to JVM and NSF Award MCB 0951094 to ARG.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Merchant SS, Prochnik SE, Vallon O, Harris EH, et al. The Chlamydomonas genome reveals the evolution of key animal and plant functions. Science. 2007;318:245–50.View ArticlePubMedPubMed CentralGoogle Scholar
- Blaby IK, Blaby-Haas CE, Tourasse N, Hom EFY, Lopez D, Aksoy M, Grossman A, Umen J, Dutcher S, Porter M, King S, Witman GB, Stanke M, Harris EH, Goodstein D, Grimwood J, Schmutz J, Vallon O, Merchant SS, Prochnik S. The Chlamydomonas genome project: a decade on. Trends Plant Sci. 2014;19:672–80.View ArticlePubMedPubMed CentralGoogle Scholar
- Tam LW, Lefebvre PA. Cloning of flagellar genes in Chlamydomonas reinhardtii by DNA insertional mutagenesis. Genetics. 1993;135:375–84.PubMedPubMed CentralGoogle Scholar
- Dent RM, Haglund CM, Chin BL, Kobayashi MC, Niyogi KK. Functional genomics of eukaryotic photosynthesis using insertional mutagenesis of Chlamydomonas reinhardtii. Plant Physiol. 2005;137:545–56.View ArticlePubMedPubMed CentralGoogle Scholar
- González-Ballester D, Montaigu A, Galván A, Fernández E. Restriction enzyme site-directed amplification PCR: a tool to identify regions flanking a marker DNA. Anal Biochem. 2005;340:330–5.View ArticlePubMedGoogle Scholar
- Meslet-Cladière L, Vallon O. A new method to identify flanking sequence tags in Chlamydomonas using 3′-RACE. Plant Methods. 2012;8:1.View ArticleGoogle Scholar
- Li X, Moelleringa ER, Liu B, Johnny C, Fedewac M, Sears BB, Kuoc MH, Benning C. A galactoglycerolipid lipase is required for triacylglycerol accumulation and survival following nitrogen deprivation in Chlamydomonas reinhardtii. Plant Cell. 2012;24:4670–86.View ArticlePubMedPubMed CentralGoogle Scholar
- Zhang RR, Patena W, Armbruster U, Gang SS, Blum SR, Jonikas MC. High-throughput genotyping of green algal mutants reveals random distribution of mutagenic insertion sites and endonucleolytic cleavage of transforming DNA. Plant Cell. 2014;26:1398–409.View ArticlePubMedPubMed CentralGoogle Scholar
- Li X, Zhang R, Patena W, Gang SS, Blum SR, Ivanova N, Yue R, Robertson JM, Lefebvre PA, Fitz-Gibbon ST, Grossman AR, Jonikas MC. An indexed, mapped mutant library enables reverse genetics studies of biological processes in Chlamydomonas reinhardtii. Plant Cell. 2016;28:367–87.View ArticlePubMedPubMed CentralGoogle Scholar
- Wang S, He J, Cui Z, Li S. Self-formed adaptor PCR: a simple and efficient method for chromosome walking. Appl Environ Microbiol. 2007;73:5048–51.View ArticlePubMedPubMed CentralGoogle Scholar
- O’Malley RC, Alonso JM, Kim CJ, Leisse TJ, Ecker JR. An adapter ligation-mediated PCR method for high-throughput mapping of T-DNA inserts in the Arabidopsis genome. Nat Protoc. 2007;2:2910–7.View ArticlePubMedGoogle Scholar
- Pollock SV, Colombo SL, Prout DL Jr, Godfrey AC, Moroney JV. Rubisco activase is required for optimal photosynthesis in the green alga Chlamydomonas reinhardtii in a low CO2 atmosphere. Plant Physiol. 2003;133:1854–61.View ArticlePubMedPubMed CentralGoogle Scholar
- Sizova I, Fuhrmann M, Hegemann P. A Streptomyces rimosus aphVIII gene coding for a new type phosphotransferase provides stable antibiotic resistance to Chlamydomonas reinhardtii. Gene. 2001;277:221–9.View ArticlePubMedGoogle Scholar
- Sueoka N. Mitotic replication of deoxyribonucleic acid in Chlamydomonas reinhardtii. Proc Natl Acad Sci USA. 1960;46:83–91.View ArticlePubMedPubMed CentralGoogle Scholar
- Shimogawara K, Fujiwara S, Grossman AR, Usuda H. High efficiency transformation of Chlamydomonas reinhardtii by electroporation. Genetics. 1998;148:1821–8.PubMedPubMed CentralGoogle Scholar
- Newman SM, Boynton JE, Gillham NW, Randolph-Anderson BL, Johnson AM, Harris EH. Transformation of chloroplast ribosomal RNA in Chlamydomonas: molecular and genetic characterization of integration events. Genetics. 1990;126:875–88.PubMedPubMed CentralGoogle Scholar
- Dent RM, Sharifi MN, Malnoë A, Haglund C, Calderon RH, Wakao S, Niyogi KK. Large-scale insertional mutagenesis of Chlamydomonas supports phylogenomic functional prediction of photosynthetic genes and analysis of classical acetate-requiring mutants. Plant J. 2015;82:337–51.View ArticlePubMedGoogle Scholar
- Wang Y, Spalding MH. An inorganic carbon transport system responsible for acclimation specific to air levels of CO2 in Chlamydomonas reinhardtii. Proc Natl Acad Sci USA. 2006;103:10110–5.View ArticlePubMedPubMed CentralGoogle Scholar
- Aksoy M, Pootakham W, Pollock SV, Moseley JL, González-Ballester D, Grossman AR. Tiered regulation of sulfur deprivation responses in Chlamydomonas reinhardtii and identification of an associated regulatory factor. Plant Physiol. 2013;162:195–211.View ArticlePubMedPubMed CentralGoogle Scholar
- Aksoy M, Pootakham W, Grossman AR. Critical function of a Chlamydomonas reinhardtii putative polyphosphate polymerase subunit during nutrient deprivation. Plant Cell. 2014;26:4214–29.View ArticlePubMedPubMed CentralGoogle Scholar
- Gonzalez-Ballester D, Pollock SV, Pootakham W, Grossman AR. The central role of a SNRK2 kinase in sulfur deprivation responses. Plant Physiol. 2008;147:216–27.View ArticlePubMedPubMed CentralGoogle Scholar
- Ma Y, Pollock SV, Xiao Y, Cunnusamy K, Moroney JV. Identification of a novel gene required for normal pyrenoid formation in Chlamydomonas reinhardtii. Plant Physiol. 2011;156:884–96.View ArticlePubMedPubMed CentralGoogle Scholar
- Jungnick N, Ma Y, Mukherjee B, Cronan JC, Speed DJ, Laborde SM, Longstreth DJ, Moroney JV. The carbon concentrating mechanism in Chlamydomonas reinhardtii—finding the missing pieces. Photosynth Res. 2014;121:159–73.View ArticlePubMedGoogle Scholar