Appl Microbiol Biotechnol DOI 10.

1007/s00253-012-4071-7

BIOTECHNOLOGICALLY RELEVANT ENZYMES AND PROTEINS

Heterologous expression of the C-terminal antigenic domain of the malaria vaccine candidate Pfs48/45 in the green algae Chlamydomonas reinhardtii
Carla S. Jones & Tiffany Luong & Michael Hannon & Miller Tran & James A. Gregory & Zhouxin Shen & Steven P. Briggs & Stephen P. Mayfield

Received: 20 February 2012 / Revised: 28 March 2012 / Accepted: 29 March 2012 # Springer-Verlag 2012

Abstract Malaria is a widespread and infectious disease that is a leading cause of death in many parts of the world. Eradication of malaria has been a major world health goal for decades, but one that still remains elusive. Other diseases have been eradicated using vaccination, but traditional vaccination methods have thus far been unsuccessful for malaria. Infection by Plasmodium species, the causative agent of malaria, is currently treated with drug-based therapies, but an increase in drug resistance has led to the need for new methods of treatment. A promising strategy for malaria treatment is to combine transmission blocking vaccines (TBVs) that prevent spread of disease with drug-based therapies to treat infected individuals. TBVs can be developed against surface protein antigens that are expressed during parasite reproduction in the mosquito. When the mosquito ingests blood from a vaccinated individual harboring the Plasmodium parasite, the antibodies generated by vaccination prevent completion of the parasites life-cycle. Animal studies have shown that immunization with Pfs48/ 45 results in the production of malaria transmission blocking antibodies; however, the development of this vaccine candidate has been hindered by poor expression in both prokaryotic and eukaryotic hosts. Recently, the chloroplast of Chlamydomonas reinhardtii has been used to express complex recombinant proteins. In this study, we show that the C-terminal antigenic region of the Pfs48/45 antigen can be expressed in the chloroplast of the green algae C.
C. S. Jones : T. Luong : M. Hannon : M. Tran : J. A. Gregory : Z. Shen : S. P. Briggs : S. P. Mayfield (*) The San Diego Center for Algae Biotechnology and the Division of Biological Sciences, University of California, San Diego, 9500 Gilman Drive, MC0368, La Jolla, CA 92093, USA e-mail: smayfield@ucsd.edu

reinhardtii and that this recombinant protein has a conformation recognized by known transmission blocking antibodies. Production of this protein in algae has the potential to scale to the very large volumes required to meet the needs of millions at risk for contracting malaria. Keywords Chlamydomonas reinhardtii . Recombinant protein . Pfs48/45 . Malaria vaccine . Transmission blocking vaccine

Introduction Malaria is an infectious disease affecting nearly 300 million people every year and resulting in the annual death of about 780,000 people (Schwartz et al. 2012; Targett and Greenwood 2008). Malaria infection results in severe flu-like symptoms that appear when the parasite invades human red blood cells following infection from a bite by the parasites biological vector, the Anopheles mosquito (Greenwood et al. 2008; Todryk and Hill 2007). Although malaria is considered a tropical disease due to its prevalence in these climates, Anopheles mosquitoes thought capable of harboring the Plasmodium parasite are present in more temperate regions around the world including the United States and Canada (Greenwood 2008; Center for Disease Control and Prevention). Increased international travel to regions endemic with the disease combined with a changing global climate could result in the reintroduction of this deadly parasite into unsuspecting and unprepared communities (Berrang-Ford et al. 2009). The enormous prevalence of the disease combined with its toll on humans, both physical and economic, makes malaria a significant global threat, and one with no existing vaccine (Greenwood 2008).

Appl Microbiol Biotechnol

Previous attempts to reduce malaria prevalence, such as vector control through the use of pesticides and parasite treatment through the use of anti-malaria drugs, have proven less effective at eradication than hoped due to resistance developed in both the mosquito and the parasite. This resistance has led to the exploration of other methods for the control of malaria including the development of malaria vaccines (Greenwood and Targett 2011). The presence of immune resistance to malaria in communities with high exposure to the parasite supports the idea that vaccines can act as a potential malaria treatment (Todryk 2007). Three types of malaria vaccine candidates are being studied and developed with a primary focus on those related to infection by the deadliest of the Plasmodium genera, Plasmodium falciparum. These include pre-erythrocytic antigens such as RTS,S (currently in phase III clinical trials), erythrocytic antigens such as the merozoite surface proteins (MSP-1 and MSP-2), and transmission blocking antigens including sexual stage antigenic surface proteins such as Pfs25, Pfs28, Pfs230, and Pfs48/45 (Todryk 2007). Transmission blocking vaccines are those that prevent the fertilization and development of P. falciparum gametes and ookinetes, thus these vaccine candidates result in the prevention of transmission of P. falciparum and protect a human community from further infection. The development of transmission-blocking vaccines that can be used in conjunction with the other two types of vaccine candidates is likely to play a critical role in the future control of malaria (Targett 2008). Pfs48/45 is considered one of the best transmissionblocking vaccine candidates due to its ability to induce antibodies capable of preventing the male gamete from participating in the fertilization process (Van Dijk et al. 2001, 2010), and unlike antigens present in later stages of fertilization, Pfs48/45 is present on gametocytes within the human host (Saul 2007). Previous research in malaria endemic regions has shown that the presence of Pfs48/45 in the human host supports naturally acquired immunity (Oudraogo et al. 2011; Bousema et al. 2010; Sutherland 2009), and repeated exposure to the Pfs48/45 antigen would likely lead to immune boosting and enhanced immune protection against transmission of the parasite, a key component to a successful transmission blocking vaccine (Outchkourov et al. 2008). Unfortunately, the continued development of Pfs48/45 as a vaccine candidate has been slowed by limited success in currently accepted expression platforms, which either lack the sophistication to fold the complex protein, or the costeffectiveness required to vaccinate millions of individuals (Chowdhury et al. 2009; Outchkourov 2008; Outchkourov et al. 2007; Milek et al. 2000, 1998; Kocken et al. 1993). The difficulty in expressing this protein is likely due to the complex structure of the Pfs48/45 antigen, which requires the formation of multiple disulfide bonds to present the proper conformation-dependent epitopes for transmission blocking activity (Sutherland 2009).

The chloroplast of Chlamydomonas reinhardtii has been shown to be capable of expressing complex human proteins, including those that require the formation of disulfide bonds (Rasala et al. 2010; Tran et al. 2009; Mayfield et al. 2003). This suggests that C. reinhardtii chloroplasts may also be capable of folding complex P. falciparum surface proteins requiring antigenic dependent disulfide-based conformations like Pfs48/45. These chloroplasts also lack the ability to glycosylate proteins, an important feature of any malaria protein expression platform, as the Plasmodium parasite is one of the few eukaryotes that lacks the machinery for N-linked glycosylation (Outchkourov 2008; Mayfield 2003). In this study, we show that C. reinhardtii is capable of expressing and accumulating the C-terminal region of the Pfs48/45 protein that has previously been found to contain the epitopes necessary for transmissionblocking activity (Outchkourov 2007, 2008). We also show that this recombinant protein accumulates in a conformation recognized by an established transmission blocking antibody.

Materials and methods Plasmid construction The amino acid sequence for Pfs48/45 was obtained from the National Center for Biotechnology Information (NCBI). Nucleotides representing amino acids 27448 of this gene were sequenced de novo using codon bias for the C. reinhardtii chloroplast with the addition of a C-terminal Flag sequence (5-GATTATAAAGATGATGATGACAAA-3) and restriction sites NdeI, AgeI, and XbaI (GeneArt). Plasmids D1 and D2 were constructed using vectors previously optimized for expression in the chloroplast of C. reinhardtii (Manuell et al. 2007; Barnes et al. 2005). D1 homologously recombines in the region of the psbA gene using kanamycin selection and contains the promoter and 5UTR for psbA as well as the 3UTR for psbA. D2 recombines in the region of psbH under photosynthetic selection and contains the promoter and 5UTR for psbD and the 3UTR for psbA. Nucleotide fragments prepared by Hi-Fidelity polymerase chain reaction (PCR) and restriction enzyme digestion were used to clone nucleotide sequences containing variations in two expression tags (FLAG and serum amyloid A protein (SAA)) and four c.r.pfs48/45 nucleotide regions representing various amino acid lengths of the c.r.pfs48/45 gene (including 27448, 27426, 178427, and 178448). These plasmids were transformed into Escherichia coli and verified by restriction digestion and nucleotide sequencing (Retrogen and Eton Biosciences). Algal strains, transformation, and growth conditions Three algal strains were used for transformations, including C. reinhardtii wild-type (w/t) strain 137c (Mt+) (CC-125,

Appl Microbiol Biotechnol

the Chlamydomonas CoreCollection Center, Duke University), C. reinhardtii psbA mutant strain (W1.1) (Manuell 2007), and C. reinhardtii non-photosynthetic psbH mutant strain (psbH-) (Rasala et al. 2011). Prior to transformation, all strains were grown to mid-logarithmic phase (approximately 8105 to 2106 cells/mL) in TAP (Trisacetatephosphate) medium at 29 C on a rotary shaker. Cells were harvested using centrifugation and resuspended using TAP medium at a concentration of 3107 cells/mL. Both w/t and W1.1 strains were plated on TAP solid medium containing 150 mg/mL kanamycin. The psbH- strain was plated on high salt solid medium. Approximately 1.5107 cells were plated for each transformation. About 10 g of plasmid DNA was prepared for each construct. This DNA was bound to gold particles (Seashell Technology), and 10 L of the DNA/gold mixture was added to a carrier membrane and used for particle bombardment transformation. D1-based constructs were transformed onto individual plates containing W1.1 and w/t strains separately while D2based constructs were only transformed on to plates containing the psbH- strain. Following transformation, plates containing W1.1 and w/t strains were placed under low constant illumination while plates containing the psbH- strain were placed under high constant illumination to allow for the formation of transformed colonies. These plates were allowed to grow under these light conditions for 2 to 4 weeks before colonies were assessed. PCR screening Successful transformants were initially identified based on primary selection measures. For w/t and W1.1 transformed strains, colonies with kanamycin resistance grew following the loss of non-resistant background, and for the psbH- strain, colonies grew that were capable of photosynthesis under the high-light conditions. Colonies were patched onto their respective solid medium plates (W1.1 and w/t were plated on TAP medium containing 100 g/mL kanamycin) and allowed to grow for about 12 weeks. These colonies were then screened using primer-specific PCR to identify the insertion of the pfs48/ 45 gene (5-GTGCTAGGTAACTAACGTTTGATTTTT-3 and 5-AATATTACTTGGTTCTAATTCTTC-3). Colonies found to contain the pfs48/45 gene were plated onto a fresh solid medium plate containing their respective antibiotics. These Pfs48/45 strains were repeatedly screened by PCR using primer sets designed to indicate the loss of the endogenous C. reinhardtii chloroplast gene compared with a control region. These strains were continuously screened and replated about every 12 weeks until homoplasmic lines were identified. RNA isolation and cDNA analysis For RNA isolation, algal cultures were grown to a concentration of about 5105 cells/mL under low light conditions and then induced under high light conditions for 24 h.

Following this 24-h period, 10 mL of algal cells were centrifuged and resuspended in 0.5 mL of plant RNA reagent. Following separation and chloroform extraction, RNA was treated with isopropyl alcohol and washed with ethanol. The RNA pellet was then resuspended in RNase-free water. RNA was treated to remove genomic DNA using the Ambion Turbo DNA-free kit and resuspended in 2050 L of RNase-free water. RNA was quantified, and 500 ng of the RNA was used for cDNA synthesis using the Biorad iScript cDNA synthesis kit. cDNA was then used in a standard PCR reaction using pfs48/45 primers (5-CATGGTTGTAATTTCTCATC-3 and 5-GATTCTGGTTGATATACTTG-3) and rbcL control primers (5 -AGCAGGTGCTGGATTCAAAG-3 and 5 CAGCTACAGCAGCACCACAT-3). PCR products were analyzed on a 1 % agarose gel containing ethidium bromide. Protein expression, purification, and Western blot analysis Algal cultures were grown to a concentration of about 8 105 cells/mL under low light conditions and then induced under high light conditions for 24 h. Following this 24h period, algal cells were centrifuged and resuspended in lysis buffer (50 mM TrisHCl, pH 8.0; 500 mM NaCl; 0.5 % Tween-20). Cells were sonicated and centrifuged at 13,000 rpm for 15 min. The soluble protein fraction was incubated overnight with FLAG resin at 4 C on an endover-end shaker. Resin was rinsed twice with lysis buffer and once with 1 Tris-buffered saline (TBS) before elution with six column volumes of elution buffer into a microcentrifuge tube containing 1:20 1 M TrisHCl, pH 8.0. The total soluble and eluted protein fractions were then analyzed using Western blot. Proteins were separated on RunBlue 12 % PAGE gels (Expedeon) and transferred to nitrocellulose membranes for Western analysis. Membranes were blocked with 5 % dry milk resuspended in 1 TBS for 2 h. An anti-FLAG antibody preconjugated with alkaline phosphatase (monoclonal anti-FLAG M2-alkaline phosphatase M2, Sigma Aldrich) was added to the solution at a concentration of 1:5,000 and allowed to incubate at 4 C overnight. The blots were rinsed with 1 TBS and then developed using nitro-blue tetrazolium (NBT) and 5-bromo-4-chloro-3-indoylphosphate (BCIP) in 1 AP buffer. Conformation Western blot and Elisa assays Equivalent amounts of eluted Pfs48/45 protein were separated on RunBlue 12 % PAGE gels. Reduced samples were prepared by adding 160 mM urea and 490 mM dithiothreitol (DTT) and heating at 60 C for 15 min while non-reduced samples contained only the native protein loading dye (62.5 mM TrisHCl, pH 6.8; 40 % glycerol; 0.01 % bromophenol blue, Biorad). The gel was run using a non-reducing running buffer (800 mM Tricine, 1.2 M Tris, 69.4 mM

Appl Microbiol Biotechnol

SDS). Proteins were subsequently transferred to nitrocellulose membranes. These membranes were then cut into quarters to allow for analysis with both anti-FLAG and antiPfs48/45 and to ensure even antibody incubation within each sector of the Western blot. After blocking with 5 % milk, anti-FLAG antibody (Monoclonal Anti-FLAG M2, Sigma Aldrich) was added at a concentration of 1:5,000, and the Pfs48/45 conformation-specific antibody anti-Pfs48/45 IIC510 MRA-26 (Malaria Research and Reference Reagent Resource Center) was added at a concentration of 1:1,000 to the respective blot sections. Following overnight incubation at 4 C, all blots were rinsed in 1 TBS and then incubated for 2 h at room temperature with the anti-mouse secondary antibody preconjugated with alkaline phosphatase (anti-mouse IgG AP conjugate; Sigma Aldrich). These blots were then rinsed and developed using NBT and BCIP in 1 AP buffer. Enzyme-linked immunosorbent assays (ELISA) were completed following protein quantification using a Lowry assay. The 8 g of eluted protein was separated into two aliquots prior to plating on Nunc Blot 96-well plates. The first aliquot was suspended in 1 PBS and treated with 175 mM urea and 630 mM DTT as well as heated to 60 C for 15 min while the second aliquot was suspended in an equivalent total volume of 1 PBS. Protein solutions were allowed to incubate on the plate overnight at 4 C before being removed and the wells rinsed with 1 PBS. Primary antiFLAG antibody at a concentration of 1:6,000 and MRA-26 at a concentration of 1:1,000 were resuspended in 1 PBS and added to each respective well. These were incubated for 4 h before being removed and the wells rinsed with 1 PBS. Secondary anti-mouse-HRP antibody (stabilized goat antimouse IgG peroxidase conjugated; Thermo Scientific) was added at a concentration of 1:20,000 and incubated for 2 h at room temperature. These wells were then rinsed four times with 1 PBS before being developed with (TMB Substrate Kit; Thermo Scientific). The plates were then analyzed on a plate reader at 450 nm. Mass spectrometry Elution aliquots of protein obtained from strain 391 were used for mass spectrometry identification of the Pfs 48/45 amino acid sequence. Pfs48/45 protein solution (0.3 mg/ml in 50 mM Hepes buffer, pH 7.2) was reduced and alkylated using 1 mM Tris(2-carboxyethyl) phosphine (Fisher, AC36383) at 94 C for 5 min and 2.5 mM iodoacetamide (Fisher, AC12227) at 37 C in dark for 30 min, respectively. Proteins were digested on beads with 1 ug trypsin (Roche, 03 708 969 001) at 37 C overnight. Nanoflow liquid chromatography-tandem mass spectrometry (MS/MS) analysis was performed using LTQ tandem mass spectrometer (Thermo Electron Corporation, San Jose, CA) employing automated data-dependent acquisition.

An Agilent 1100 high-performance liquid chromatography (HPLC) system (Agilent Technologies, Wilmington, DE) was used to deliver a flow rate of 500 nL min1 to the mass spectrometer through a splitter. Using an in-houseconstructed pressure cell, 5 um Zorbax SB-C18 (Agilent) packing material was packed into a fused silica capillary tubing (200 m ID, 360 um OD, 20 cm long). One end of the fused silica tubing was pulled to a sharp tip using a laser puller (Sutter P-2000) as the electro-spray tip. The peptide mixture was loaded onto the HPLC column using the same in-house pressure cell. Peptides were eluted from the HPLC column using a 0 % to 80 % acetonitrile gradient for 120 min. Mass spectrometer was programmed to perform data-dependent MS/MS scans on the seven most intense ions from the full MS scan (4502,000 Da). Raw data were extracted and searched using Spectrum Mill (Agilent, version A.03.02.060b). MS/MS spectra with a sequence tag length of 1 or less were considered as poor spectra and discarded. The filter of the MS/MS spectra were searched against a database containing the Pfs48/45 protein sequence and common contaminants including trypsin and keratin. The enzyme parameter was limited to full tryptic peptides with a maximum mis-cleavage of 1. All other search parameters were set to SpectrumMills default settings (carbamidomethylation of cysteines, +/2.5 Da for precursor ions, +/0.7 Da for fragment ions, and a minimum matched peak intensity of 50 %). Oxidized-methionine and pyroglutamate was defined as variable modifications. A maximum of two modifications per peptide was used.

Results Nine plasmids containing variations in the c.r.pfs48/45 coding sequence, protein expression tags, promoters, and 5 UTRS were successfully integrated into the chloroplast genome of C. reinhardtii (Fig. 1). Specifically, these nine strains differed by variations in two expression tags, FLAG and SAA, usage of either the psbA or psbD promoter and 5 UTR, and they contained variable lengths of coding sequence for the c.r.pfs48/45 gene. The coding sequence variations included the presence or absence of the C-terminal anchor domain (AA 427448) and the presence or absence of the domain containing the first six cysteine amino acids within the original Pfs48/ 45 sequence (AA 27427 versus AA 178427). Of these nine strains, only two (strains 391 and 389) containing the shorter amino acid sequence encoding the domain containing ten cysteines and the C-terminal anchor domain were shown to result in protein accumulation. Both strains 391 and 389 contain the codon optimized gene sequence for expression of amino acids 178448 of the native Pfs48/45 peptide with a FLAG expression tag located at the N terminus. These two strains only differed in the site

Appl Microbiol Biotechnol

a
C. reinhardtii Strain Designation 389 391 399 401 472 484 547 549 551 Plasmid Design D2:Flag-c.r.-Pfs 48/45 AA 178-448:D1 D1:Flag-c.r.-Pfs 48/45 AA 178-448:D1 D2:SAA-c.r.-Pfs 48/45 AA 178-427:D1 D2:SAA-c.r.-Pfs 48/45 AA 178-448:D1 D1:SAA-c.r.-Pfs 48/45 AA 27-448:D1 D2:SAA-c.r.-Pfs 48/45 AA 27-448:D1 D1:SAA-c.r.-Pfs 48/45 AA 27-426:D1 D2:Flag-c.r.-Pfs 48/45 AA 27-426:D1 D1:Flag-c.r.-Pfs 48/45 AA 27-426:D1 5 psbA Homology Region

1X Flag

1XFlag

5 psbH Homology Region

NdeI c.r.pfs 48/45

XbaI

3 Homology Region

C.

NdeI c.r.pfs48/45

XbaI
aphA VI

3 Homology Region

psbD promoter and 5 UTR BamHI Silent Site

psbA 3 UTR

psbA promoter and 5 UTR

atpA psbA 3 UTR promoter and 5 UTR psbA exons and introns

rbcL 3 UTR

psbH psbN

5 psbA Homology Region

psbA promoter and 5 UTR

psbA 3 UTR

3 Homology Region

Fig. 1 Nine plasmids containing variations in the c.r.pfs48/45 gene were created and integrated into the C. reinhardtii chloroplast genome using homologous recombination. a List of C. reinhardtii strains created showing variations in the promoter and 5 UTR regions, the expression tag (FLAG or SAA) and in the length of the gene sequence integrated. b Diagram of the integrated region of plasmids containing

the psbD promoter and 5 UTR (D2) and the site of homologous recombination within the C. reinhardtii chloroplast genome resulting in the recovery of photosynthesis. c Diagram of the integrated region of plasmids containing the psbA promoter and 5 UTR (D1) and the site of homologous recombination within the C. reinhardtii chloroplast genome resulting in kanamycin resistance

of integration within the chloroplast genome and the use of the promoter and 5 UTR from either psbA (D1, 391) or psbD (D2, 389) (Fig. 1bc). Due to strain 389 having much lower expression levels based on Western blotting, strain 391 was used for larger scale production and further analysis of the protein expressed in the chloroplast of C. reinhardtii. The presence of the c.r.pfs48/45 gene sequence in strain 391 was confirmed by primer-specific PCR, and the recombinant gene was shown to occupy the psbA insertion site in all copies of the chloroplast genome through colony PCR screening (Fig. 2ab). In order to confirm that the gene sequence was being transcribed within the chloroplast of C. reinhardtii, total RNA was isolated and used for a qualitative RNA analysis. The analysis confirmed the c.r.pfs48/45 gene was being transcribed within the chloroplast of the C. reinhardtii strain 391. (Fig. 2c). Following growth to middle logarithmic phase, gene expression in algal cultures was induced by high light exposure for 24 h (Manuell 2007). The 391 algal cells were pelleted and used for c.r.Pfs48/45 protein isolation and purification. Western blot analysis of total protein extracts using a monoclonal anti-FLAG-M2 alkaline phosphatase conjugated antibody recognized proteins of approximately the predicted size for strain 391 at 34 kD. This flag-tagged

protein was present in the total protein extract and was significantly concentrated when proteins were eluted from an anti-FLAG affinity resin (Fig. 3a). Soluble protein concentrate from strain 391 c.r.Pfs48/45 shows the presence of two proteins of distinct sizes in the Western blot. The top band of approximately 40 kD is likely the reduced protein containing the entire 391 c.r.Pfs48/45 protein while the smaller band of approximately 21 kD is likely a cleavage product of the protein. Above these two bands are two faint bands that likely represent the unreduced folded 391 c.r.Pfs48/45 protein at approximately 60 kD and a potential protein multi-mer at 150 kD. To confirm that the FLAG antibody was recognizing the correct 391 c.r.Pfs48/45 protein, an eluted protein extract was analyzed by mass spectrometry. Three peptides matching those expected in the 391 c.r.Pfs48/45 protein were identified (Fig. 3b). To confirm that 391 c.r.Pfs48/45 folded in a manner conducive to the formation of the conformational-dependent epitopes necessary for antigenicity, both a Western blot and ELISA assay were completed using the conformational specific Pfs48/45 antibody IIC5-10 (MRA-26). Figure 4a shows the Western blot using the MRA-26 antibody (lanes 4 and 5) compared with a FLAG-specific antibody (lanes 1 and 2).

Appl Microbiol Biotechnol

a
61 600bp

b
61 500bp 61 61 61 Pos Neg

c
+ 200bp A

Pfs48/45 Primers + B 391-61 wildtype + -

rbcL Primers
+ A + B wildtype + -

391-61

Fig. 2 Confirmation of complete integration and transcriptional expression for strain 391 (line 61). a Gene-specific PCR confirmation of the presence of the c.r.pfs48/45 gene. b PCR screen showing the loss of the endogenous psbA gene when compared with a control PCR product indicating 100 % replacement of the psbA gene within all copies of the C. reinhardtii chloroplast genome. c Qualitative cDNA analysis showing the transcription of the c.r.pfs48/45 gene compared with the rbcL control. A and B represent duplicate samples of the c.r.pfs48/45 cDNA product

When the 391 c.r.Pfs48/45 protein was under reduced conditions, the FLAG-specific antibody clearly recognized the peptides of the appropriate size while the unreduced sample revealed bands at higher molecular weights. However, when the 391 c.r.Pfs48/45 protein was reduced prior to exposure to the MRA-26 antibody, no peptides were recognized, as predicted because the epitope was denatured. Exposure of the MRA-26 antibody to the unreduced 391 c.r.Pfs48/45 protein revealed the presence of a band on the Western blot at approximately 130 kD. These results show that the conformationspecific Pfs48/45 antibody, MRA-26, only recognizes the 391 c.r.Pfs48/45 protein when it is in its unreduced conformation and suggest that this algal chloroplast accumulated protein folded in the correct conformation. The Western results were further confirmed with an ELISA assay using both reduced and non-reduced protein samples incubated with the MRA-26 antibody (Fig. 4b). The exposure of both reduced and unreduced 391 c.r.Pfs48/45 and a control bovine serum albumin (BSA) standard to the MRA-26 antibody clearly shows that the recognition of this antibody is specific to the unreduced 391 c.r.Pfs48/45 protein.

Discussion Despite its promise as a transmission blocking malaria vaccine candidate, the P. falciparum surface protein Pfs48/45

Fig. 3 Heterologous expression of c.r.Pfs 48/45 protein (C10) fragment and peptide identification by mass spectrometry. a Western blot of protein purification extracts including total protein (T), soluble protein (S), wash (W), and five elutions (E). FLAGspecific primary antibody conjugated with amino peroxidase recognizes protein of approximately the correct size in the total protein extract and clearly in the second and third elutions. b Strain 391 c.r.Pfs48/45 amino acid sequence showing the location of three peptides (bold and underlined) identified by mass spectrometry. c Strain 391 c.r.Pfs48/45 peptide identification from by mass spectrometry

kDa 100 70 55 35 27 15 10

T S W

E1 E2 E3 E4 E5

MDYKDDDDKSGGGGSRSGENLYFQGS GGGGSRVLKYPHNILFTNLTNDLFTYLP KTYNESNFVSNVLEVELNDGELFVLACE LINKKCFQEGKEKALYKSNKIIYHKNLTIF KAPFYVTSKDVNTECTCKFKNNNYKIVL KPKYEKKVIHGCNFSSNVSSKHTFTDSL DISLVDDSAHISCNVHLSEPKYNHLVGL NCPGDIIPDCFFQVYQPESEELEPSNIV YLDSQINIGDIEYYEDAEGDDKIKLFGIV GSIPKTTSFTCICKKDKKSAYMTVTIDSA YYGFLAKTFIFLIVAILLYI*
Spectrum

Peptide

Score

(K)YPHNILFTNLTNDLFTYLPK(T)

17.82

(K)APFYVTSK(D)

14.53

(K)LFGIVGSIPK(T)

18.41

Appl Microbiol Biotechnol

b
Absorbance (450nm)

0.045 0.04 0.035 0.03 0.025 0.02 0.015 0.01 0.005 0

BSA unreduced BSA reduced 391 c.r.Pfs48/45 unreduced 391 c.r.Pfs48/45 reduced

Samples

Fig. 4 Recognition of strain 391 c.r.Pfs48/45 recombinant protein conformation through Western blot and Elisa assays. a Western blot using MRA-26, a conformation-specific anti-Pfs48/45 antibody showing that reduction of the 391 c.r.Pfs48/45 recombinant protein prevents recognition by antibody when compared with Western analysis using a

FLAG antibody. b ELISA assay confirming the Western blot results showing that unreduced 391 c.r.Pfs48/45 recombinant protein is recognized at a higher level than either the reduced 391 c.r.Pfs48/45 protein sample or the BSA control standards

has proven to be a difficult target for multiple heterologous expression platforms (Chowdhury 2009; Outchkourov 2007, 2008; Milek 1998, 2000; Kocken 1993). These expression complications are likely due to structural discrepancies preventing antigen recognition including the complexity of the disulfide bridges formed within the repeated six-cysteine-containing domain structure that is common to this family of P. falciparum surface proteins (van Dijk 2010; Sutherland 2009; Outchkourov 2008). Additionally, in expression systems that can accumulate complex proteins, post-translational modifications of the recombinant Pfs48/45 protein may prevent the formation of the correct conformation due to the lack of N-linked glycosylation machinery for endogenous Plasmodium proteins (Outchkourov 2008). However, the model green alga C. reinhardtii whose chloroplast has been shown to provide an environment conducive to the formation of proteins containing complex disulfide bridges that lack post-translational modifications such as glycosylation is an ideal candidate for the production of a malarial transmission blocking vaccine (Tran 2009; Mayfield 2003). In this study, a gene sequence codon optimized for expression in the chloroplast of C. reinhardtii was created and used to construct a series of nine plasmids to test for expression of the Pfs48/45 protein in the chloroplast. Of the nine C. reinhardtii strains, only two (strains 389 and 391) resulted in detectable levels of protein expression. Expression in these two strains may be a result of increased protein stability due to the expression of a smaller protein domain, decreased disulfide bridge formation complexity, or the presence of the C-terminal anchor domain. Strains 389 and 391 contain only ten of the 16 possible cysteines from the full-length c.r.Pfs48/45 protein. These ten cysteines should result in the formation of only five disulfide bridges

compared with eight in other strains containing the fulllength peptide and may result in enough stability to allow for protein accumulation. Previous studies have shown that a similar C-terminal ten-cysteine containing protein region optimized for expression in E. coli had greater stability, solubility, and epitope recognition resulting in better transmission blocking activity (Outchkourov 2007, 2008). Another factor potentially impacting the expression of the c.r.Pfs48/45 protein in four of the C. reinhardtii strains failing to accumulate protein (strains 399, 547, 549, and 551) may be the absence of the Pfs48/45 C-terminal anchor domain. Strains 389 and 391 both contain this C-terminal anchor domain sequence, and the accumulation of protein from these strains may be due to protein interaction with the membrane allowing for added stability in chloroplasts. However, strains 401, 472, and 484 contain this Cterminal anchor domain but show no protein accumulation. This lack of accumulation may be due to the presence of the large and very soluble SAA protein expression tag on the N terminus of the peptide in these strains compared with the smaller FLAG expression tag in strains 389 and 391. This SAA expression tag may keep the recombinant protein in the soluble protein fraction for a longer period of time allowing for protease degradation resulting in undetectable recombinant protein accumulation levels. Ongoing studies focused on understanding the impact of variations in expression tags and coding sequence lengths as well as the impact of the presence or absence of the C-terminal anchor domain in c.r.Pfs48/45 strains, and its interaction with the chloroplast membrane will help guide further research on Pfs48/45 protein expression in the chloroplast of C. reinhardtii. Lower levels of protein expression in strain 389 led to the use of strain 391 for all further characterization. To begin,

Appl Microbiol Biotechnol

PCR analysis showed the homoplasmic incorporation of the c.r.pfs48/45 gene within the chloroplast genome of strain 391. The complete homoplasmic replacement of the psbA gene is critical for high levels of expression, as the psbA gene product (the D1 protein) has been shown to play a role in autoattenuation of translation from any construct containing the psbA promoter and UTR (Manuell 2007). In order to rule out expression problems at the transcriptional level, qualitative cDNA analysis was also used to show that 391 c.r.Pfs48/45 RNA was present in the light induced C. reinhardtii cultures. To show protein expression, the accumulation of the 391 c.r.Pfs48/45 protein was analyzed through Western blot analyses. The presence of the 391 c.r.Pfs48/45 protein in the total protein extract, but limited presence within the soluble protein extract, through multiple attempts, indicates that this protein may be found at higher concentrations associated with the membrane portion of the total protein extract. As mentioned earlier, the presence of the Pfs48/ 45 C-terminal anchor on the 391 c.r.Pfs48/45 protein may be the reason for the lower levels of protein accumulation shown in the soluble fraction. Ongoing research in the laboratory is focused on optimizing protein expression conditions to increase recombinant protein yields in both the total and soluble portions of the protein extract. Accumulation results also show a cleavage product of about 20 kD. This cleavage product is not surprising due to previous studies showing parts of the Pfs48/45 protein to be more resistant to trypsin digestion (Outchkourov 2007). These trypsin resistant regions may also be more resistant to other types of proteases both during the in vivo expression of the protein in the chloroplast and during protein isolation and purification. This cleavage product is further supported by the initial isolation results for the native fulllength Pfs48/45 protein from malaria parasites where there was a similar pattern of cleavage. This isolation also presented a very large protein likely due to unreduced tertiary and quarternary protein structures (Rener et al. 1983). To further support the accumulation data and confirm the identity of the recombinant 391 c.r.Pfs48/45 protein, this protein was analyzed by mass spectrometry. This analysis found three peptides matching the Pfs 48/45 amino acid sequence. These three peptides were also located in three different regions of the protein indicating that the entire 391 c.r.Pfs48/45 domain is being expressed. Finally, the biological activity of the Pfs48/45 protein is dependent on the presence of conformation-dependent epitopes (Sutherland 2009; Outchkourov 2007; Vermeulen et al. 1985). Of these five epitopes present on the native protein, three (epitopes I, IIb, and III) have been shown to result in naturally occurring antibody responses and have the greatest potential impact on transmission blocking activity (Graves et al. 1992). A conformation-specific antibody, IIC5-10, (MRA-26) recognizes epitope III on Pfs48/45 when the

protein is folded, and the correct pattern of disulfide bonds is present (Carter et al. 1990; Targett et al. 1990; Vermeulen 1985; Rener 1983). To determine if the 391 c.r.Pfs48/45 protein domain contained this folded epitope, the MRA-26 antibody was used in both Western blot and ELISA analysis of reduced and unreduced samples. These resulted in the specific recognition of only unreduced 391 c.r.Pfs48/45, confirming the presence of the folded epitope. The MRA-26 antibody has previously been shown to specifically recognize Pfs48/45 antigens associated with that antigens transmission blocking activity (Targett 1990; Rener 1983). The recognition of only unreduced 391 c.r.Pfs48/45 by MRA-26 likely demonstrates that the C. reinhardtii expressed protein has the conformation necessary to induce transmission blocking activity. Further analysis of antibody elicitation in mice and transmission blocking bioactivity assays are ongoing to confirm these results. This study shows that the chloroplast of C. reinhardtii is a viable platform for the production of the transmissionblocking vaccine candidate Pfs48/45 and that this host accumulates a protein that contains a conformation conducive to the induction of transmission blocking antibodies. The expression of this P. falciparum antigen in the chloroplast of C. reinhardtii is significant because it demonstrates a novel method to produce this protein at large scales and also because algae have the potential to be developed for oral delivery of a malaria vaccine, similar to those shown from the chloroplast of higher plants (Davoodi-Semiromi et al. 2010; Webster et al. 2009). The expression of a transmission-blocking vaccine antigen in a scalable and edible expression system may enable the availability of such a vaccine to people around the world. The accumulation of the C-terminal region of the Pfs48/45 antigen in the chloroplast of C. reinhardtii is an important step, and further studies of this protein to gauge its effectiveness in malaria are underway.
Acknowledgments We acknowledge SAIC subcontract P010020844 and Gabe Gutierrez for initial support for this project. This work was also supported by the San Diego IRACDA Postdoctoral Fellowship from the NIH grant GM06852 and the San Diego Foundation Blasker Science and Technology Grant C-2011-00208. We would also like to acknowledge Joseph Vinetz and laboratory at the University of California, San Diego for the IIC5-10 antibody.

References
Barnes D, Franklin S, Schultz J, Henry R, Brown E, Coragliotti A, Mayfield SP (2005) Contribution of the 5 and 3-untranslated regions of the plastid mRNAs to expression of Chlamydomonas reinhardtii chloroplast genes. Mol Gen Genom 274:625636. doi:10.1007/s00438-005-0055-y Berrang-Ford I, MacLean JD, Gyorkos TW, Ford JD, Ogden NH (2009) Climate change and malaria in Canada: a systems approach. Interdiscip Perspect Infect Dis 113. doi:10.1155/2009/ 385487

Appl Microbiol Biotechnol Bousema T, Roeffen W, Meijerink H, Mwerinde H, Mwakalinga S, van Gemert GJ, van de Vegte-Bolmer M, Mosha F, Targett G, Riley EM, Sauerwein R, Drakeley C (2010) The dynamics of naturally acquired immune responses to Plasmodium falciparum sexual stage antigens Pfs230 & Pfs48/45 in a low endemic area in Tanzania. PLoS One 5:e14114. doi:10.1371/journal.pone.0014114 Carter R, Graves PM, Keister DB, Quakyi IA (1990) Properties of epitopes of Pfs 48/45, a target of transmission blocking monoclonal antibodies, on gametes of different isolates of Plasmodium falciparum. Parasite Immunol 12:587603. doi:10.1111/j.13653024.1990.tb00990.x Chowdhury DR, Angov E, Kariuki T, Kumar N (2009) A potent malaria transmission blocking vaccine based on codon harmonized full length Pfs48/45 expressed in Escherichia coli. PLoS One 4:e6352. doi:10.1371/journal.pone.0006352 Davoodi-Semiromi A, Schreiber M, Nalapalli S, Verma D, Singh ND, Chakrabarti D, Daniell H (2010) Chloroplast-derived vaccine antigens confer dual immunity against cholera and malaria by oral or injectable delivery. Plant Biotech J 8:223 242. doi:10.1111/j.1467-7652.2009.00479.x Graves PM, Doubrovsky A, Sattabongkot J, Battistutta D (1992) Human antibody responses to epitopes on the Plasmodium falciparum gametocyte antigen Pfs 48/45 and their relationship to infectivity of gametocyte carriers. AmJTrop Med Hyg 46:711719 Greenwood BM, Targett GAT (2011) Malaria vaccines and the new malaria agenda. Clin Microbiol Infect 17:1600 1607. doi:10.1111/j.1469-0691.2011.03612.x Greenwood BM, Fidock DA, Kyle DE, Kappe SHI, Alonso PL, Collins FH, Duffy PE (2008) Malaria: progress, perils and prospects for eradication. J Clin Invest 4:12661276. doi:10.1172/JCI33996 Kocken CHM, Jansen J, Kaan AM, Beckers PJA, Ponnudurai T, Kaslow DC, Konings RNH, Schoenmakers JGG (1993) Cloning and expression of the gene coding for the transmission blocking target antigen Pfs48/45 of Plasmodium falciparum. Mol Biochem Parasitol 61:5968. doi:10.1016/0166-6851(93)90158-T Manuell AL, Beligni MV, Elder JH, Siefker DT, Tran M, Weber A, McDonald TL, Mayfield SP (2007) Robust expression of a bioactive mammalian protein in Chlamydomonas chloroplast. Plant Biotech J 5:402412. doi:10.1111/j.1467-7652.2007.00249.x Mayfield SP, Franklin SE, Lerner RA (2003) Expression and assembly of a fully active antibody in algae. Proc Nat Acad Sci 100:438 442. doi:10.1073/pnas.0237108100 Milek RLB, DeVries AAF, Roeffen WFG, Stunnenberg H, Rottier PJM, Konings NH (1998) Plasmodium falciparum: heterologous synthesis of the transmission-blocking vaccine candidate Pfs48/ 45 in recombinant Vaccinia virus-infected cells. Exper Parasitol 90:165174. doi:10.1006/expr.1998.4315 Milek RLB, Stunnenberg HG, Konings RNH (2000) Assembly and expression of a synthetic gene encoding the antigen Pfs48/45 of the human malaria parasite Plasmodium falciparum in yeast. Vaccine 18:14021411. doi:10.1016/S0264-410X(99) 00392-8 Oudraogo AL, Roeffen W, Luty AJF, de Vlas SJ, Nebie I, Edith IS, Cuzin-Ouattara N, Teleen K, Tiono K, Tiono AB, Sirima SB, Verhave JP, Bousema T, Sauerwein R (2011) Naturally acquired immune responses to Plasmodium falciparum sexual stage antigens Pfs48/45 and Pfs230 in an area of seasonal transmission. Infect Immun 79:49574964. doi:10.1128/IAI.05288-11 Outchkourov N, Vermunt A, Jansen J, Kaan A, Roeffen W, Teelen K, Lasonder E, Braks A, van de Vegte-Bolmer M, Qiu LY, Sauerwein R, Stunnenberg HG (2007) Epitope analysis of the malaria surface antigen Pfs48/45 identifies a subdomain that elicits transmission blocking antibodies. J Biol Chem 282:17148 17156. doi:10.1074/jbc.M700948200 Outchkourov NS, Roeffen W, Kaan A, Jansen J, Luty A, Schuiffel D, van Gemert GJ, van de Vegte-Bolmer M, Sauerwein RW, Stunnenberg HG (2008) Correctly folded Pfs48/45 protein of Plasmodium falciparum elicits malaria transmission-blocking immunity in mice. Proc Nat Acad Sci 105:43014305. doi:10.1073/pnas.0800459105 Rasala BA, Muto M, Lee PA, Jager M, Cardoso RMF, Behnke CA, Kirk P, Hokanson CA, Crea R, Mendez M, Mayfield SP (2010) Production of therapeutic proteins in algae, analysis of expression of seven human proteins in the chloroplast of Chlamydomonas reinhardtii. Plant Biotech J 8:719733. doi:10.1111/j.1467-7652.2010.00503.x Rasala BA, Muto M, Sullivan J, Mayfield SP (2011) Improved heterologous expression in the chloroplast of Chlamydomonas reinhardtii through promoter and 5 untranslated region optimization. Plant Biotech J 9:674683. doi:10.1111/j.1467-7652.2011.00620.x Rener J, Graves PM, Carter R, Williams JL, Burkot TR (1983) Target antigens of transmission-blocking immunity on gametes of Plasmodium falciparum. J Exp Med 158:976981. doi:10.1084/ jem.158.3.976 Saul A (2007) Mosquito stage, transmission blocking vaccines for malaria. Curr Opin Infect Dis 20:476 481. doi:10.1097/ QCO.0b013e3282a95e12 Schwartz L, Brown GV, Genton B, Moorthy VS (2012) A review of malaria vaccine clinical projects based on the WHO rainbow table. Malaria J 11:11. doi:10.1186/1475-2875-11-11 Sutherland CJ (2009) Surface antigens of Plasmodium falciparum gametocytesa new class of transmission-blocking vaccine targets? Mol Biochem Parasitol 166:93 98. doi:10.1016/ j.molbiopara.2009.03.007 Targett GA, Greenwood B (2008) Malaria vaccines and their potential role in the elimination of malaria. Malaria J 7:S10. doi:10.1186/ 1475-2875-7-S1-S10 Targett GAT, Harte PG, Eida S, Rogers NC, Ong CSL (1990) Plasmodium falciparum sexual stage antigens: immunogenicity and cellmediated responses. Immunol Let 25:7782. doi:10.1016/01652478(90)90095-8 Todryk SM, Hill AVS (2007) Malaria vaccines: the stage we are at. Nat Rev Microbiol 5:487489. doi:10.1038/nrmicro1712 Tran M, Zhou B, Pettersson PL, Gonzalez MJ, Mayfield SP (2009) Synthesis and assembly of a full-length human monoclonal antibody in algal chloroplasts. Biotech Bioeng 104:663 673. doi:10.1002/bit.22446 Van Dijk MR, Janse CJ, Thompson J, Waters AP, Braks JAM, Dodemont HJ, Stunnenberg HG, van Gemert GJ, Sauerwein RW, Eling W (2001) A central role for Pfs48/45 in malaria parasite male gamete fertility. Cell 104:153164. doi:10.1016/S0092-8674(01)00199-4 Van Dijk MR, van Schaijk BCL, Khan SM, van Dooren MW, Ramesar J, Kaczanowski S, van Gemert GJ, Kroeze H, Stunnenberg HG, Eling WM, Sauerwein RW, Waters AP, Janse CJ (2010) Three members of the 6-Cys protein family of Plasmodium play a role in gamete fertility. PLOS Pathogens 6:e1000853. doi:10.1371/journal.ppat.1000853 Vermeulen AN, Ponnudurai T, Beckers PJA, Verhave JP, Smits MA, Meuwissen JHE (1985) Sequential expression of antigens on sexual stages of Plasmodium falciparum accessible to transmissionblocking antibodies in the mosquito. J Exp Med 162:14601476. doi:10.1084/jem.162.5.1460 Webster DE, Wang L, Mulcair M, Ma C, Santi L, Mason HS, Wesselingh SL, Coppel RL (2009) Production and characterization of an orally immunogenic Plasmodium antigen in plants using a virus-based expression system. Plant Biotech J 7:846855. doi:10.1111/j.14677652.2009.00447.x