WO2011056877A1 - Malaria transmission-blocking vaccine - Google Patents

Abstract

The present invention relates to malaria vaccines. Particularly, the present invention relates to antigens that are effective to prevent the transmission of malaria. The antigen comprises a fragment of AnAPN1.

Description

MALARIA TRANSMISSION-BLOCKING VACCINE

This application claims the priority of U.S. Provisional Patent Application Serial No. 61/257,573, filed November 3, 2009, which is incorporated herein by reference.

FIELD OF THE INVENTION

The present invention relates to malaria vaccines. Particularly, the present invention relates to antigens that are effective to prevent the transmission of malaria. BACKGROUND OF THE INVENTION

Malaria continues to be a tremendous public health burden worldwide, with more than 2 to 4 million people dying from the disease each year. One of the most prevalent forms of the disease is caused by the protozoan parasite Plasmodium vivax, which is found in tropical and sub-tropical regions. Interestingly the parasite can complete its mosquito cycle at temperatures as low as 15 degrees Celsius, which has allowed the disease to spread in temperate climates. The acute form of malaria is caused by the protozoan parasite, Plasmodium falciparum which is responsible for most of the mortality attributable to malaria.

The life cycle of Plasmodium is complex, requiring two hosts, man and mosquito for completion. The infection of man is initiated by the inoculation of sporozoites into the blood stream through the bite of an infected mosquito. The sporozoites migrate to the liver and there infect hepatocytes where they differentiate, via the exoerythrocytic intracellular stage, into the merozoite stage which infects red blood cells (RBC) to initiate cyclical replication in the asexual blood stage. The cycle is completed by the differentiation of a number of merozoites in the RBC into sexual stage gametocytes, which are ingested by the mosquito, where they develop through a series of stages in the midgut to produce sporozoites which migrate to the salivary gland.

Vaccines have been at the forefront of global research efforts to curb the disease, yet despite several vaccine candidates, this goal has not been completely realized. To date, only a single preventative vaccine against malaria has been developed (RTS,S). The RTS, S vaccine has been shown to confer between -38-54% protection in vaccinated individuals in Sub-Saharan Africa.

Other vaccines include those disclosed in WO 2006/088597, WO 2006/088597, and WO 2008/009650, which are incorporated herein by reference.

Although disease severity can be lessened with the afore-mentioned malaria vaccine, the fact remains that individuals who receive the vaccine are still carrying infectious parasites that can be transmitted by mosquitoes. Such deficiencies will unlikely lead to malaria elimination or eradication. In order to reach such a goal, a transmission-blocking component to the vaccine is necessary. The global research effort has since focused on developing a "cocktail" vaccine, wherein different life stages of the parasite can be targeted specifically by separate vaccine components. The key component to such an eradication vaccine would be one that blocks the cascade of secondary infections from immunized but parasitized human hosts. The current paradigm for developing a malaria transmission-blocking vaccine (TBV) involves developing antibodies against gametocyte or ookinete surface molecules. This approach is indeed effective but since there are two main human parasite species (P. falciparum and P. vivax), two separate vaccines must be produced. One cannot claim malaria eradication a success if one parasite species remains prevalent in human populations.

Therefore, there remains a need to effectively combat the spread of malaria.

SUMMARY OF THE INVENTION

The present invention relates to a composition to combat the spread of malaria by using transmission-blocking vaccines (TBVs), which prevent the development of malarial parasites within its mosquito vector; thereby abrogating the cascade of secondary infections in humans. The TBV contains a fragment of the Anopheline midgut specific membrane-bound alanyl aminopeptidase (AnAPNl). The present inventors have discovered that AnAPNl is an excellent TBV due to its highly conserved molecule present across diverse Anopheles vector species and is a putative ligand for P. falciparum and P. vivax ookinetes. Antibodies raised in rabbits against the N-terminal domain of AnAPNl (NT135APN1) (SEQ ID NO.: 1) are capable of completely (100%) blocking P. falciparum invasion of An. gambiae in Cameroon. The same antibodies exhibit virtually complete blocking (95-98%) of P. vivax invasion of An. dims in Thailand. Anti- AnAPNl antibodies exhibit no cross-reactivity with normal human lung, kidney, liver, and small intestine tissue sections. These results strongly support the use of AnAPNl fragments as TBVs. Preferably, the AnAPNl fragment is NT135APN1 (SEQ ID NO.: 1); more preferably, the fragment is amino acids 32-64 of SEQ ID NO.: 1. The fragment of AnAPNl can be used as a vaccine for preventing the transmission of the parasites that cause malaria. Once an immune response against AnAPNl is stimulated in a subject, preferably a person, the antibodies produced against the AnAPNl is effective to disrupt the development of the parasites. Thus, without being held to a particular theory, when a mosquito ingests the blood of the subject, it is believed that the antibodies disrupt the parasites in the mosquito gut, thereby preventing the transmission of the parasites to other subjects. In an embodiment the TBVs of the present invention can be used with currently available vaccine for malaria, including the RTS,S vaccine, and those disclosed in WO 2006/088597, WO 2006/088597, and WO 2008/009650. The TBV can be administered concurrently with the malaria vaccine. Alternatively, the TBV can be admixed with the malaria vaccine to produce a combination vaccine and administered as a single dose.

The TBV can also include an adjuvant. The term "adjuvant" generally refers to any material that increases the humoral or cellular immune response to an antigen.

Adjuvants are used to accomplish two objectives: 1) slow the release of antigens from the injection site; and 2) stimulate the immune system. Synthetic and subunit vaccines are expensive to produce. The addition of an adjuvant may permit the use of a smaller dose of antigen to stimulate a similar immune response, thereby reducing the production cost of the vaccine. Thus, the effectiveness of some injectable medicinal agents may be significantly increased when the agent is combined with an adjuvant.

In another embodiment, the present invention also relates to methods for using the TBV, by itself or in a combination vaccine, to stimulate an immune response against AnAPNl. The method involves administering the AnAPNl fragment to a subject. The amount of AnAPNl fragment administered should be sufficient to stimulate an immune response against AnAPNl. The TBV may be administered by any known method, such as by intramuscular injection, intradermal injection, gene gun or electroporation. A skilled person in this art would readily be able to determine the appropriate dosage and schedule for administering the vaccine.

In yet another embodiment, the present invention also provides methods for making the TBVs. The TBV can be isolated from natural sources or, preferably, made recombinantly. Recombinant methods are well known in the art; however, Pichia and E. coli are the preferred systems for making the AnAPNl fragment, particularly NT135APN1.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 shows the details of AnAPNl. (a) Schematic of the AnAPNl glycoprotein, a highly conserved molecule found on the apical face of midgets of anopheline vectors as determined by Western blot. The N-terminal 135 amino acid fragment target antigen (NT135APN1) is indicated. Predicted motifs for glycosaminoglycans (GAG) and O- linked glycans and GPI-anchors are shown. The GPI anchor was confirmed by releasing AnAPNl into the supernatant with PI-PLC. (b) Homology model of AnAPNl based off the A4 hydrolase crystal structure indicating the NT135 epitope (orange) and GAGs (aqua). Image portrays view from the top looking down on AnAPNl. The predicted Mr of AnAPNl is indicated to the left by the apparent Mr is 125 kDa.

Fig. 2 shows that anti- AnAPNl IgG blocks P. falciparum development in An. Gambiae mosquitoes in Cameroon, (a) Antibodies were mixed with gametocymetic blood from volunteers (case codes indicated below each bar panel) and fed to mosquitoes. Asterisks indicates statistical significance (a=0.05). Dilution series of IgG are indicated between panels (a) and (b). A 1:2 dilution is equivalent to 10 μg/mLof AnAPNl -specific IgG. (b) An alternative presentation of the inhibition data depicted in (a) is reduction in mosquito infection prevalence. Data is from a second cohort of volunteers during the following year's transmission season. The data clearly show that the antibodies reduce mosquito infection prevalence to zero, i.e., no oocysts form in the mosquito midgets. This suggests that the antibodies are in fact more potent against field isolates of P.

falciparum than against laboratory stains and represent a true transmission-blocking vaccine approach. Fig. 3 shoes that anti-AnAPNl IgG blocks P. vivax development in An. Dims A mosquitoes in Thailand. Experiments conducted as was described for Fig. 2 above.

Asterisks indicates statistical significance (a=0.05). Dilution series of IgG are indicated on top of the graph. A 1:2 dilution is equivalent to 10 μg/mLof AnAPNl- specific IgG.

Fig. 4 shows mouse immunizations with recombinant AnAPNl elicit a potent transmission-blocking response against P. berghei. (a) Antigen specific antibody titers elicited in BALB/c (5 mich/cohort) from tree bleeds (pooled) at two week intervals following the first (Bleed 1, ·), second (Bleed 2,■) and third (Bleed 3, Δ) booster immunizations as determined by ELISA. This data is representative of replicated experiments. Optical density (O.D. 630 nm) and serum dilutions are plotted. Error bars indicate the standard error of the mean from quadruplicate wells. End point titers were defined as the highest serum dilution giving an O.D. reading greater than that of pre- immune serum +2 standard deviations. (bO mice (M1-M3) immunized with AnAPNl elected P. berghei transmission-blocking antibody titers in two biological experiments indicated by the stippled (cohort 1) and solid vertical bars (cohort 2). Asterisks denote statistical significance (*) at P<0.05 (Mann-Whitmey U Test). Error bars indicate 1 standard deviation.

Fig. 5 shows data generated by Epitope Identification Suite (Merck Research labs) suggest that the NT135APN1 sequence is "clean" of potential cross-reactive human epitopes. A search for potential T-cell epitopes with significant sequence identity with human 9-mer peptides suggests (1) that the NT135APN1 has an immunogenic hot spot between amino acids 32-64 and (2) there is little to no homology with human peptides currently annotated in the human proteome. Three MHC II alleles common in Tanzanian populations are indicated in the gray box on the right. The eight MHC II alleles used in the query (shown o n the left) are common across Caucasian populations in the US. The complete NT135APN1 sequence is shown in the red box on the bottom with the peptide sequence corresponding to a potential cross -reactive epitope in humans underlined.

Fig. 6 shows immunofluorescence staining of normal human kidney, lung, liver and small intestine sections with anti-AnAPNl IgG. Sections were stained with polyclonal rabit anti-AnAPNl IgG and detected with ant-rabbit Texas red-conjugated antibody. Secondary antibody alone was not used as a control in follow up assays since there was no detectable signal to primary antibody. Note that the red signal in the overlay image for the liver is completely due to leakage of autofluorescence into the red channel. Images were acquired at 200X. Bright field images are provides for orientation. DAPI was used to stain nuclei and appear blue.

Fig. 7 shows H&E staining and immunofluorescence microscopy of a normal lung section with anti-AnAPNl IgG. Lund sections were stained with polyclonal rabbit anti- ANAPN1 IgG and anti-actin (human) monoclonal antibodies and detected with anti- rabbit Texas Red-conjugated and anti-mouse Alexa 488-conjugated secondary antibodies, respectively. Two different compartments of the lung were examined, the alveolus and smooth muscle/connective tissues, which are also shown by H&E staining below. Actin appears green and AnAPNl cross reactivity appears red. DAPI was used to stain nuclei and appear blue. Note that the actin signal and AnAPNl signal overlap (orange) but that the fluorescence signal is predominantly contributed by bleed-through of the green signal into the red channel from detection of actin, not unlike what was observed in Fig. 6.

H&E staining of compartments of human lung from a serial section are shown in panels (a) and (b). In panel (a), the arrow indicates staining of resident alveolar lumen macrophages. AS indicates alveolar sac and AD indicates alveolar duct. In panel (b), the arrow indicates junction between sub-endothelial connective and smooth muscle tissues within the lung. Bright field images are provided for orientation. All images were acquired at 400X.

Fig. 8 shows expression of recombinant AnAPNl in different yeast constructs. 1. AnAPNl/pPinka-HC/PichiaPink#4. 2. AnAPNl-M/pPICZaA/P.pastoris X-33. 3.

AnAPNl-D/pPICZaA/P.pastoris X-33. Expression gels suggest that using strains that are deficient of proteases is a better strategy than changing the makeup of the protein.

Fig. 9 shows SDS PAGE of product purification efforts from Pichia. (a) In process samples from Butyl HP column (SDS PAGE/CBB). See Blue MW marker (M), Starting material (SM), Flow-through (FT), peak fraction A4 (elution with buffer containing no salt), and peak fraction A5 (elution with buffer containing no salt) were run on a 14% Tris-Glycine gel (Invitrogen) and stained with Coomassie Brilliant Blue (CBB). Starting material was diluted and very faint on the gel. No AnAPNl was seen in the collected fractions, (b) In process samples from QXL column (SDS PAGE/CBB). See Blue MW marker (M), Starting material (SM), peak fraction A8 (elution with buffer containing 200 mM NaCl), peak fraction Al l (elution with buffer containing 400 mM NaCl) and peak fraction B2 (elution with buffer containing 600 mM NaCl) were run on a 14% Tris-Glycine gel (Invitrogen) and stained with CBB. No AnAPNl was seen in the collected fractions, (c) In process samples from QXL column (SDS PAGE/SS). See Blue MW marker (M), Starting material (SM), peak fraction A8 (elution with buffer containing 200 mM NaCl), peak fraction Al l (elution with buffer containing 400 mM NaCl) and peak fraction B2 (elution with buffer containing 600 mM NaCl) were run on a 14% Tris-Glycine gel (Invitrogen) and silver stained. There is very little AnAPNl seen in the collected fractions and is relatively unpure.

Fig. 10 shows E. coli in process samples during the homogenization step (SDS PAGE /CBB) of the recovery of NT135APN1 from E. coli. To determine if the

NT135APN1 was solubilized, 10 ul of each in-process sample were run on a reduced 14% Tris-Glycine gel (Invitrogen). For each cycle with the homogenizer samples were centrifuged at 14,000 RPM for 5 minutes at room temperature to separate the soluble portion in the supernatant and the insoluble pellet. A whole cell sample (total cell suspension after Cycle 1) was also included to show the total protein contained in both supernatant and pellet.

Fig. 11 shows E. coli in process samples from Ni Column (SDS PAGE / CBB). Starting material (SM), flow-through (FT), wash of unbound sample (W), wash with corresponding buffer containing 20 mM Imidazole (Labeled as 20 mM), and elution (Labeled as 500 mM).

Fig. 12 shows E. coli purification yield based on densitometry (SDS PAGE / CBB). For NT135APN1, 2.5 ul of sample was determined to be within the linear range of the machine, and for BSA 4 ug was determined to be within the linear range of the machine (l.See Blue; 2. NT135APN1 - 2.5μ1; 3. BSA, 4μ§).

Fig. 13 shows an SDS-PAGE of pre-dialyzed material 2 μg (non-reduced: Lane 1; reduced: Lane 2) and 2 μg Lot 032410JLP (non-reduced: Lane 3; Reduced: Lane 4) followed by Coomassie Staining (A), Silver Staining (B), Western blot analysis with anti- APN1 (C) and anti-His (D). Fig. 14 shows binding of APN-1 to Alhydrogel® in 10% glucose/10 mM imidazole/10 mM SB 3-12 or 10% glucose/10 mM imidazole/1 mM SB 3-12. Vaccine supernatants were analyzed by SDS-PAGE alongside NT135APN1 standards

(unformulated). The standards show the amounts of AnAPNl that would be present in the vaccine supernatants if 50%, 25%, 10% and 1% of the formulated NT135APN1 was unbound and free in solution. No protein at the expected size of approximately 15 kDa migrates in the formulated vaccine supernatant lanes, indicating that NT135APN1 was 100% adsorbed to Alhydrogel® in these formulations. MW markers (in kDa) are shown in the far left lane.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The present inventors have discovered that fragments of AnAPNl, preferably NT135APN1 (SEQ ID NO.: 1), is an attractive malaria TBV because it is highly immunogenic, can block P. falciparum and P. vivax transmission through mosquitoes in the field, and can be easily microencapsulated in slow-release polymers for delivery. The features that allow AnAPNl fragments to be an effective TBV includes:

• Mosquito-based antigen as opposed to a parasite antigen which is difficult to express correctly;

• Mosquito-based antigen is a highly conserved (i.e., present in all species of

anopheline mosquitoes) mosquito midgut specific membrane bound alanyl aminopeptidase (APN)

• The selected protein fragment to be used as antigen contains immunological hotspots corresponding to common HLA epitopes that would be recognized across different human populations (targeted for vaccination);

· Rabbit IgG completely blocks (100% inhibition) field isolates of P. falciparum;

• Rabbit IgG is a potent inhibitor (95-98% inhibition) of field isolates of P. vivax;

• Rabbit IgG works against parasites in different mosquito species from different parts of the world;

• At present, polymorphisms in APN in these mosquitoes do not appear to effect potency;

• IgG shows no cross-reactivity with human APN in several different tissues;

• At present, mouse-immunization experiments do not result in death; • Mouse IgG exhibits potent inhibitory activity against Plasmodium berghei (murine model of malaria);

• Antigen can be expressed as a completely soluble product ~ 100 mg/L in Pichia pastoris or E. coli;

• E. coli codon-harmonized gene can be used to express protein in bacteria;

• Antigen can be microencapsulated in biodegradable micro- and/or nanoparticles for delivery. Additional proof of principle data shows that a single dose of the microparticle:antigen:alum formulation in mice results in comparable inhibition profiles to that of controls which were immunized following the standard prime and 3 boost schedule;

• Antigen can be co-administered with other vaccine candidates targeting parasite proteins from the sporozoite, gametocyte, gamete and ookinete stages;

• Antigen can be easily complexed with other adjuvants, e.g., CpG

oligonucleotides, cholera toxin, synthetic bacterial lipopolysaccharides, TLR agonists as well as other adjuvants currently available (e.g., from Infectious Disease Research Institute); and

• Antigen:adjuvant delivery systems can take the form of needle-free, microparticle sheet delivery, liposomes, as naked DNA, via adenoviral vectors, as well as other novel vehicles and standard emulsions.

Because malarial parasites are transmitted among human hosts by the bite of its obligate insect vector, the Anopheles mosquito, targeting mosquito midgut antigens that serve as ligands for different Plasmodium parasite species and strains is one viable approach to breaking the transmission cycle. The present inventors have recently identified in AnAPNl as a predominant ligand for different malaria parasite species and strains. Rabbit antibodies directed against a 135 amino acid N-terminal fragment (NT135APN1) prevented both the murine malaria parasite Plasmodium berghei and the human malaria parasite, P. falciparum from developing in the mosquito vector. The antibodies are quite potent, blocking field isolates of P. falciparum (100% inhibition) and P. vivax (95-98% inhibition) in completely divergent anopheline vectors (An. gambiae s.s. and An. dims A, respectively). Moreover, immunization of mice following a standard regimen, which are then subsequently infected with P. berghei, showed that the mice elicit a high titer of IgG that completely block parasite transmission to mosquitoes. It is important to note that the critical goal for TBVs should be the complete prevention of mosquito infection and not simply the reduction in oocyst intensity in the mosquito. After all, a single oocyst is all that is needed to continue the transmission cycle of Plasmodium.

APN1 antigen appears to be an abundant midgut glycoprotein that is conserved across diverse anopheline vectors an is referred to herein as AnAPNl (Fig. 1). Anti- AnAPNl IgG recognition of orthologous aminopeptidase antigens in the midguts of several anopheline vectors and its efficacy in blocking both human malaria parasite species implies a significant utility of AnAPNl or its fragments as an effective transmission-blocking target that is effective for all malaria causing parasites, especially P. falciparum and P. vivax.

The AnAPNl fragments can be obtained from natural sources, or preferably through recombinant techniques known in the art. Appropriate host cells for the expression of the fragments include, but are not limited to, yeast cells, insect cells, mammalian cells, or bacteria. In exemplary embodiments of the invention, the fragments are produced in yeast cells such as Saccharomyces cerevisiae, Hansenula polymorphs, Pichia pastoris, Kluyvermyces fragilis, Kluveromyces lactis, and Schizosaccharomyces pombe. Alternatively, the preferred bacterial system for producing the fragments is E. coli. The preferred expression systems offer the advantages of being cost-effective and easily adapted to large-scale growth in fermenters.

Transmission blocking efficiency

Anti-AnAPNl polyclonal IgG efficiently and effectively (10 μg) confers complete transmission-blocking immunity against field isolates of P. falciparum (Fig. 2) and between 95-98% inhibition of P. vivax (Fig. 3). In contrast, the matching concentrations of IgG in pre-immune sera from rabbits, mice or malaria-naive human AB serum exhibit no inhibition of oocyst development in mosquito. It remains unclear why P. vivax is not blocked to the same extent at the 10 μg concentration of IgG.

Immunization of mice with AnAPNl recombinant protein in incomplete Freunds

Adjuvant or alum elicits a potent transmission-blocking antibody titer against P. berghei (Fig. 4). These results suggest that eliciting transmission -blocking antibody titer in human is highly likely.

In silico analysis

Work leading to the selection of NT135APN1 as an antigen focuses on identifying highly hydrophilic, relatively low structural complexity (e.g. the absence cysteins), and potentially immunogentic domains using a variety of web-based tools. Further in silico analyses using simple BLAST analysis as well as proprietary software from Merck Research Laboratories (Epitope Identification Suite) has confirmed the initial findings and verified the suitability of NT135APN1 as a highly antigenic fragment.

Briefly, Epitope Identification Suite is a program used by Merck Research labs to identify potential microbial pathogen vaccine targets (e.g. HIV vaccines). We used this software to 1) identify areas of the protein fragment which show high sequence identity with human proteins through a search against the entire human proteome; and 2) identify peptides along NT135APN1 that are predicted to be high-binders to eight of the most frequently found MHC II HLA-DR alleles (DR-1, 3, 4, 7, 8, 11, 13, and 15) in the world population (Fig. 5).

A search for potential T-cell epitopes with significant homology to human proteins (80% or higher amino acid identity to any human peptide, i.e., eight or more identical amino acids in a span of 9), suggests that the NT135APN1 is "clean"

(WO2006/124406). We found only two possible peptides (MSSAKVSSL and

SSAKVSSLP) that match, albeit poorly. An independent BLASTP analysis using the NT135APN1 as 'query' resulted in poor matches to a puromycin- sensitive human aminopeptidase (with a known cytosolic localization) with e-values with low statistical significance (e^"°⁹). Alignments with other human APNs show lower e-values (le^"°⁵ to 2^" °⁵). Furthermore, sequence alignments confirm the absence of a string of amino acids in one linear location that is above 7 aa/10 aa sequence. NT135APN1 shows the highest sequence identity with mosquito APNs only (both. Anopheles and Aedes).

Loosening the stringency (2 mismatches in a 9 amino acid linear location), the total number of possible peptide matches increased to four. It should be noted that highly safe and acceptable vaccines that are already available, e.g., against tetanus (TeTx) exhibits more sequence identity with human sequences than NT135APN1. Moreover, recent immunizations of mice, with NT135APN1 in alum, did not result in any obvious adverse side-effects in the mice 6 months post-immunization (i.e., no unexpected mortality due to antigen vs. controls). Taken together, the data suggests that NT135APN1 is relatively "clean" and potentially safe. This data suggests NT135APN1 would be a safe vaccine no little human cross-reactivity.

It is important to note that Epitope Identification Suite also found a potential immunogenic "hot spot" in sub-fragment amino acids 39-62 of the NT135APN1 sequence (Fig. 5). Not only is this sub-fragment "clean" (as described above) but contains multiple predicted binders for all eight MHC II alleles. In addition, two other immunogenic sub-fragments are clearly identifiable. At present it is not clear which are the protective epitopes or if polyclonal targeting of these epitopes is the major contributing factor to the potency of the antibodies.

Immunohistology

Alanyl aminopeptidases, in mammalian general, are membrane -bound

metalloproteases that are expressed on the surface of normal human cells in the lung, liver, small intestine and kidney. We conducted immunohistochemical experiments to assess the potential cross -reactivity of anti-AnAPNl antibodies against commercially available human tissue sections (corresponding to those listed above). Sections were stained with polyclonal rabbit anti-AnAPNl IgG followed by detection with Texas Red- conjugated anti-rabbit secondary Abs. Images were acquired in the green/red channel to acquire the background autofluorescence signal and then in the red channel alone to assess anti-AnAPNl cross-reactivity. Replicate immunofluorescence, staining experiments suggest that any detection of signal in the red channel is largely due to a strong background autofluorescence signal extending into the red channel as determined by overlays (Fig. 6). In spite of this initial data, closer examination (400X) of

compartments of the human lung suggest that AnAPNl IgG staining of alveolar macrophages appeared distinct from background fluorescence. In follow-up experiments, lung sections were stained with both anti-AnAPNl antibodies (rabbit polyclonal IgG) and anti-Human Actin (mouse MAb) and detected with Texas Red (red channel) conjugated anti-rabbit and Alexa 488 (~FITC/green channel) conjugated anti-mouse secondary antibodies, respectively. These experiments show that the AnAPNl cross -reactivity signal colocalizes with cytosolic actin staining suggesting that the detection may be either artefactual spill-over from the anti-actin signal in the green channel (as was seen with the autofluorescence data) or that macrophages in the alveolar lumen can be stained with anti-AnAPNl IgG (Fig. 7).

Expression of NT135APN1 in yeast

The native DNA sequence (SEQ ID NO: 2) encoding NT125APN1 has been cloned into different yeast expression vectors and transformed into different strains of yeast. Preliminary expression feasibility is assessed at the shaker flask level and optimized prior to conducting larger 10L scale fermentations. The expression results are summarized in the Table 1. Table 1. Status of NT125APN1 expression in yeast. Summary of a variety of AnAPNl constructs that were engineered and evaluated. Stability and yield indicators (+'s) are to compare between strains. +++ for yield is approximately lOOmg/L of culture broth.

Stability

DNA His-taa Vector Host Solubility Yield DH5.0 DH3.Q

AnAPNl + pPICZaA P.pastoristX-33 soluble ++ + ++

AnAPNl - pPICZaA PpastorisiX-SZ soluble +++ + ++

AnAPNl + pPICZaA P.pasfors/SMD1 168 soluble + ++

AnAPNl - pPICZaA P.pasiQ«s/S D1168 soluble ++ ++ +++

AnAPNl +/- pGAPZaA PpastorisfX-33 soluble ++ + ++

AnAPNl + pGAPZaA P.pasfc¾7s/SMD1 168 soluble + + ++

AnAPNl - pPinka-HC PichiaPink Strain#1 soluble ++ + ++

As shown in Table 1, NT125APN1 with or without His-tag at C-terminus was expressed in most of the yeast strains with high yield. The best yield (> 100 mg/liter) was observed when NT125APN1 without a His-tag was expressed in Pichia pastoris X- 33 under induction of 0.5% methanol. However, the expressed recombinant AnAPNl was not stable in yeast culture, with product derived (confirmed by western blot and amino terminal sequencing) break-down fragments readily forming. Amino acid sequencing results of different fragments showed that the cleavage happened at multiple sites on the C-terminus of the recombinant protein, possibly resulting from the hydrolytic activities of Pichia derived protease(s).

In an attempt to evaluate how to inhibit or prevent the proteolytic breakdown of recombinant NT135APN1, experiments were initiated by transforming NT135APN1 into protease deficient Pichia strains like P. pastoris SMD1168 or PichiaPink Strain #4 and by lowering pH of culture to 3-5. The best condition for expressing NT135APN1 with high yield and high stability has been optimized as being induced in minimal methanol Adenine media (MMA) with pH 4 for clone of AnAPNl /pPinka-HC transformed PichiaPink strain#4 (shown as pink in the table 1, Fig. 8). The construct without the His- tag was evaluated in PichiaPink, strain#4.

In addition and to reduce the degradation of recombinant full-length NT135APN1 caused by yeast derived protease(s) during induction, the actual and predicted cleavage sites at C-terminus of NT135APN1 based on the amino acid sequencing results were mutated to create a mutated clone (AnAPNl -M), or the C-terminus containing all cleavage sites was deleted to create a deletion clone (AnAPNl-Ό). These two clones were transformed into P. pastoris X-33 to create high yield transformants with improved stability, but still minor degradation remains (Fig. 8).

Finally and to increase stability of recombinant NT135APN1, the dimer form of NT135APN1 (Di- NT135APN1) was expressed in P. pastoris X-33 by linking two AnAPNl coding sequences with a EcoRI site (GAATTC encoding for EF). However, the expressed recombinant dimer form of NT135APN1 remained problematic in terms of stability with the dimer becoming separated and the proteolytic cascade continuing.

Yeast fermentation

Seed Culture. For a 10 L-scale fermentation process, a 2.5 L Tunair shake flask (Shelton Scientific, CT), containing 1 L of buffered minimal glycerol was inoculated with 2 mL of the P. pastoris PichiaPink™ Strain 4 seed stock. The flask was rotated on a shaker at 180 RPMs and incubated at 30 + 2 °C for 24-28 hours to a final OD₆₀₀ of 5-15. Approximately 0.5 L of this culture was used to inoculate 5 L of heat- sterilized basal salt media (BSM) containing 3.5 mL/L (17.5 mL total) of a filter-sterilized trace element (PTM4) solution.

Fermentation Parameters. Fermentation is conducted in a BioFlo 3000 fermentor (New Brunswick Scientific Co. Inc.). The pH of the 5 L of BSM is adjusted and maintained with a 14% ammonium hydroxide feed. The following fermentation parameters are set and verified.

Table 2. Initial fermentation parameters

Batch Phase. After inoculation, a dissolved oxygen spike to a value of 50% for 5 minutes marks the depletion of glycerol (17-20 hours after inoculation) and the end of the batch phase.

Glycerol Fed-Batch Phase. After the dissolved oxygen spike, 50% (v/v) glycerol is introduced into the cell culture media at a set flow rate of 15mL/L of BSM/hr for 8 hours. The pH of the cell culture media is decreased linearly from 5.0 to 4.5 and the temperature decreased linearly from 30 °C to 26 °C over the last two hours of this 8 hour period. Excessive foaming is controlled with 10% (v/v) KF0673 antifoam (Emerald Performance Materials) in DI water. Methanol Induction Ramp-Up Phase. The methanol induction phase, inducing the expression and secretion of AnAPN-1, is initiated when the wet cell weight (WCW) reached approximately 200 g/L. 100% Methanol is added at an initial flow rate of 1.5 mL/L of BSM/h increasing to 8.0 mL/L of BSM/h (ramp) over an 8-hour period.

10X YPYNB Feed. During the methanol induction phase, a feed of 10X Yeast

Extract, Phytone Peptone, Yeast Nitrogen Base (10X YPYNB) is initiated to try to prevent proteolytic cleavage of AnAPN-1. The 10X YPYNB is added at a flow rate of 3 ml/L/h over a period of 65 hours to the end of the fermentation.

Methanol Soaking Phase. Methanol induction is continued for another 57 hours by adding 100% Methanol at a flow rate of 8 ml/L of BSM/Hour. The dissolved oxygen concentration is maintained at >20% using a 30% set point and monitored to provide information on methanol accumulation.

Harvest. The fermentation was harvested with a final wet cell weight of approximately 258 g/L.

Recovery and purification from yeast fermentation

Approximately 7.5 L of fermentation cell culture is centrifuged for 30 minutes at 7000 RPM and 4 °C. The -6.5 L of supernatant is filtered with a 0.22 urn Millipak-60 filter (Millipore) and stored at -80°C, until further purification.

Purification is performed on AKTA Explorer (GE Healthcare) Air 100 purification system. The chromatography columns evaluated for the first purification step were a hydrophobic interaction Butyl Sepharose HP column, a cation exchange SP Sepharoase XL column, and an anion exchange Q Sepharose XL column. Filtered fermentation supernatant was the starting material for the following chromatographic experiments

Butyl Sepharose HP. Butyl Sepharose HP is a hydrophobic affinity column that binds proteins based upon their hydrophobicity; material is then eluted from the column with a chaotropic agent. All steps of the Butyl Sepharose HP chromatography are performed at room temperature at a flow rate of -155 cm/h (1 mL/min) while monitoring the absorbance of the column eluate at 280 nm.

A pre-packed HiTrap Butyl HP 1 mL column (GE Healthcare) was equilibrated with 50 mM Succinic acid, pH 6.0. A solution of 50mM Succinic acid/3 M ammonium sulfate was added to the fermentation supernatant to achieve a concentration of 1.5 M ammonium sulfate, and the pH was adjusted to 6.0 with 5N NaOH. The adjusted supernatant was filtered with a 0.22 um filter (Millipore).

The AnAPNl starting material was loaded onto the HiTrap Butyl HP 1 mL column, and unbound sample was washed with buffer (50 mM Succinic acid, 1.5 M ammonium sulfate, pH 6.0). The column was then washed with 50 mM Succinic acid, pH 6.0 applying step wise gradiant of six different concentrations of ammonium sulfate (1.35M, 1.2M, 1.05M, 0.9M, 0.75M, and 0M). Two fractions were collected during a wash with buffer containing no ammonium sulfate while monitoring the absorbance at 280 nm. They were also analyzed by SDS-PAGE (Fig. 9a).

SP Sepharose XL. SP Sepharose XL binds proteins in a low ionic strength buffer; material is then eluted from the column by increasing the ionic strength of the buffer. All steps of the SP Sepharose XL chromatography are performed at room temperature at a flow rate of -155 cm/h (1 mL/min) while monitoring the absorbance of the column eluate at 280 nm.

A pre-packed HiTrap SPXL 1 mL column (GE Healthcare) was equilibrated with 50 mM Succinic acid, 1M NaCl, pH 4.0. A solution of 50mM Succinic acid, pH 4.0 was added to the fermentation supernatant to reduce the sample conductivity to <5 mS/cm. The pH was approximately 4.0 and did not need adjustment. The diluted supernatant was filtered with a 0.22 um filter (Millipore).

After filtration of the sample, the material was loaded onto the HiTrap SPXL 1 mL column, and unbound sample was washed with buffer (50 mM Succinic acid, pH 4.0). The column was then washed with 50 mM Succinic acid, pH 4.0 containing six different concentrations of NaCl (20 mM, 50 mM, 150 mM, 300 mM, 400 mM, and 800 mM). No peak fractions were seen while monitoring the absorbance at 280 nm.

Q Sepharose XL. Q Sepharose XL binds proteins in a low ionic strength buffer; material is then eluted from the column by increasing the ionic strength of the buffer. All steps of the Q Sepharose XL chromatography are performed at room temperature at a flow rate of -155 cm/h (1 mL/min) while monitoring the absorbance of the column eluate at 280 nm.

A pre-packed HiTrap QXL 1 mL column (GE Healthcare) was equilibrated with 20 mM Tris-HCl, 1M NaCl, pH 8.5. A solution of 20mM Tris-HCL, pH 8.5 was added to the fermentation supernatant to reduce the sample conductivity to <5 mS/cm and the pH was adjusted to 8.5 with 5N NaOH. The adjusted supernatant was filtered with a 0.22 um filter (Millipore). After filtration, the material was loaded onto the HiTrap QXL 1 mL column, and unbound sample was washed with 20 mM Tris-HCL, pH 8.5. The column was then washed with 20 mM Tris-HCl, pH 8.5 containing seven different concentrations of NaCl (25 mM, 50 mM, 100 mM, 200 mM, 400 mM, 600 mM, and 1M). Peak fractions were collected during the washes containing 200 mM, 400 mM, and 600 mM of NaCL while monitoring the absorbance at 280 nm. They were also analyzed by SDS-PAGE (Fig. 9b and 9c).

Expression of NT135APNl in E. Coli

In order to produce soluble NT135APN1 in E. coli, the coding sequence was harmonized based on the E. coli codon preference and protein structure analysis. The harmonized DNA sequence (SEQ ID NO.: 3) was synthesized and cloned into E. coli expression vector pET41a with the deletion of the vector GST tag. The recombinant NT135APN1 with 6His-tag at C-terminus was expressed in E. coli BL21(DE3) under induction of 1 mM IPTG. However, the majority of recombinant NT135APN1 that was expressed in E. coli formed insoluble inclusion bodies. Initial experiments have been performed to refold the denatured recombinant NT135APN1 by rapid dilution into various refolding buffers. The best recovery of about 40% soluble protein was achieved by diluting guanidine denatured inclusion bodies into 1.1 M guanidine, 0.44 Arginine and 5mMDTT, however, the recovered soluble NT135APN1 precipitated out after being further dialyzed against PBS, pH7.4.

To increase the solubility of recombinant NT135APN1 expressed in E. coli, the induction condition was optimized by decreasing the concentration of IPTG, and lowering induction temperature. The induced E. coli cells were lysed in the presence of ionic detergent sarkosyl. The results showed that more that 10% of recombinant

NT135APN1 expressed in E. coli became soluble when NT135APN1/E. coli was induced with 0.5 mM IPTG at 25 °C for 6 hours and recovered in the presence of sarkosyl. The total yield of soluble NT135APN1 was about 80 mg/L under this induction condition.

However, the efficiency and specificity of His-tag NT135APN1 binding on nickel column was low during initial experiments (possibly due to aggregation and subsequent lack of His-tag accessibility).

Due to the initial promising results, the strain AnAPN/pET41aAGST/BL21 was identified to undergo evaluation for process development and scale -up. A Research seed stock was made by growing An APN/pET41 aAGS T/BL21 in LB till the OD600 reached 2.1, adding glycerol to 20% and storing at -80°C. Process development and scale-up of soluble NT135APN1 were accelerated and evaluated by performing fermentation and purification at the 10 liter scale.

E. coli fermentation

Fermentation is conducted in a BioFlo 3000 fermentor (New Brunswick

Scientific Co. Inc.). The pH of the 9 L of heat-sterilized LB media is adjusted and maintained with a 20% phosphoric acid feed. Prior to fermentor inoculation the LB media in the fermentor is supplemented with 30μg/ml of Kanamycin by adding 27ml of lOmg/ml Kanamycin stock solution. Approximately 130 mL of seed culture was used to inoculate 9 L of LB media. Target OD in fermentor after inoculation is 0.05-0.1 (Table 3). Table 3. 10-L Production Fermentor Operating Parameters

Cell Growth Phase. After fermentor inoculation the level of dissolved oxygen

(DO) is maintained above 30%. Cells culture was growing at 37 + 0.5 °C, at 400 RPM for

1.3 hours to target OD₆oo of 0.4. The temperature of the cell culture media is decreased linearly from 37 to 25 °C in one hour period. Excessive foaming is controlled with 10% (v/v) KF0673 antifoam in DI water.

IPTG Induction Phase. The IPTG induction phase, inducing the expression of NT135APN1, is initiated when the OD₆oo of cell culture reached approximately 0.6. Concentration of 0.5 mM IPTG in fermentor was achieved by adding 9 mL of 5M IPTG stock solution. After 6 hours of induction at 25 °C final OD₆oo of cell culture in fermentor was 2.8.

Recovery and purification from E. coli fermentation

Centrifugation. The cell culture is subjected to a centrifugation step in order to separate cells from fermentation media. After centrifugation of ~9 L cell culture at 7000 RPM, 30 minutes at 4°C, approximately 20 g of cell pellet was collected. Cell pellet was resuspended in 0.8L of ice-cold buffer 20 mM Tris-HCl, pH 8.5 (40 mL buffer/gram of cell pellet) and centrifuged at 7000 RPM, 30 minutes at 4°C.

Homogenization. Approximately 5g of the cell pellet (1/4 of the full scale) was resuspended in 250 mL buffer 20 mM Tris-HCl, 2% Sarkosyl, pH 8.5. Cells were opened and homogenized by passing the resuspended cell culture three times through Emulsiflex C-55B machine. Homogenization was performed at 20,000 psi, maintaining the temperature of cell suspension below 13 °C. Disrupted cell pellet was centrifuged at 7000 RPM, for 30 minutes at 4°C in order to separate supernatant from inclusion bodies. The supernatant, containing soluble portion of AnAPNl is filtrated (0.8 - 0.2 um) and it serves as the starting material for downstream purification. In process samples were taken and analyzed (Fig. 10). Purification. The current purification process for NT135APN1 consists of one column chromatography step. Purification was performed on AKTA Explorer (GE Healthcare) Air 100 purification system.

Sample was adjusted with the Imidazole concentration to 5 mM, by adding 62mL of 5X 5mM Imidazole, 20mM Tris-HCl, 500mM NaCl, pH 8.5 solution, 300 mL of supernatant (1/4 of the full scale) was loaded onto the Ni column at flow rate of 10 mL/min. Unbound sample was washed from column with 50 mL (10 CV) of 20 mM Tris-HCL, 5 mM Imidazole and 500 mM NaCl, pH 8.5 to remove all unbound material. Weak bounded impurities are washed from the column with 50 mL (10 CV) of 20 mM Tris-HCL, 20 mM Imidazole and 500 mM NaCl, pH 8.5.

The NT135APN1 is then eluted from the column with 20 mM Tris-HCl, 500 mM Imidazole, 500 mM NaCl, pH 8.5. The eluate (19 mL) is collected as a single peak. The eluted protein was stored at 2-8°C for the subsequent purification or formulation step.

This purification steps are performed at room temperature at flow rate of 300cm/h (10 mL/min). The column eluate is monitored for protein content by absorbance at 280 nm.

To assess the purity of the protein eluted with 20 mM Tris-HCl, 500 mM

Imidazole, 500 mM NaCl, pH 8.5 (eluate), 3.5 ul of sample from each step of the purification were loaded on a non-reduced 14% Tris-glycine gel (In vitro gen) and stained with Coomassie Brilliant Blue (CBB) (Fig. 11). The majority of impurities seen in the starting material (SM) did not bind to the Ni column and were found in the flow-through (FT) and wash of unbound sample (W), leaving the eluate with an approximate relative purity more than 90%. The wash with 20 mM Tris-HCL, 20 mM Imidazole and 500 mM NaCl, pH 8.5 did not show significant removal of protein from the column on this gel.

Yield was determined by analyzing protein bands from SDS-PAGE with a GS- 800 self-calibrating densitometer. The NT135APN1 eluate from the Ni column and BSA, a known standard, were run on a reduced 14% Tris-Glycine gel (Invitrogen) and stained with Coomassie Brilliant Blue (Fig. 12). Based on calculations, the concentration of NT135APN1 in the Ni column eluate was 1.65 mg/ml. The volume of the eluate was 19 mL, therefore the yield was 31.3 mg.

10 mg of protein was dialyzed in 15% Sucrose, 10 mM Tris, 0.2% Tween-80 pH 7.8 buffer (GWU Lot# 032410JLP). The protein was characterized by SDS-PAGE (non- reducing, reducing) followed by coomassie blue staining (with densitometry purity analysis), silver staining, western blotting with anti-His and anti-APNl (Supplied by Rhoel) antibodies, and LAL endotoxin testing. The lot had a relative purity (by densitometry after coomassie blue staining) of 95.2% and an endotoxin content of < 25 EU/mL. The characterization is presented in the following Fig. 13 and Table 4.

Table 4. Research COA for Lot 032410JLP of AnAPNl in 15% Sucrose, 10 mM Tris, 0.2% Tween-80, pH 7.8

Formulation

Initial formulation studies based on SE-HPLC and visual color and appearance analysis showed that NT135APN1 was highly insoluble in lx PBS, but that a buffer composed of 10% glucose/10 mM imidazole/10 mM sulfobetaine 3-12 (SB 3-12, a non- ionic detergent) was beneficial in terms of reducing precipitation and aggregation of APN-1.

Once this initial buffer was identified, a preliminary Alhydrogel® (aluminum hydroxide gel adjuvant) binding study was performed. Based upon the theoretical pi of APN-1 (approx. 6.4), it was predicted that NT135APN1 would readily adsorb onto Alhydrogel®. This prediction was based on the fact that at physiological pH,

NT135APN1 is relatively negatively charged and Alhydrogel® is relatively positively charged. Electrostatic interactions are the most important attractive forces between protein and aluminum hydroxide when the protein and adjuvant have opposite charges (as is the case here).

With this in mind, 0.1 mg/mL NT135APN1 was formulated with 0.8 mg/mL

Alhydrogel® in two different buffers: 10% glucose/10 mM imidazole/10 mM SB 3-12, pH 7.4; and 10% glucose/10 mM imidazole/1 mM SB 3-12, pH 7.4. Binding of

NT135APN1 was allowed to proceed with gentle rotation at room temperature for 30 minutes. Samples of each formulation were then centrifuged at 16,100 x g for 2 minutes to pellet the Alhydrogel® (along with adsorbed AnAPNl). The supernatants were aspirated and run in duplicate on a 4-20% Tris-glycine SDS-PAGE gel in the presence of NT135APN1 standards (unformulated) representing the amount of NT135APN1 that would be present in the vaccine supernatants if 50%, 25%, 10% and 1% were unbound (i.e., free in solution) (Fig. 14). NT135APN1 was 100% adsorbed to Alhydrogel® in both formulation buffers since no protein is visible at approximately 15 kDa on the gel. The level of detection of this assay is approximately 600 pg, which allows for detection of 0.05% unbound NT135APN1. This study clearly shows that NT135APN1 readily binds to Alhydrogel® at physiological pH and in the presence of the stabilizing excipient glucose and the non-ionic detergent SB 3-12.

Although certain presently preferred embodiments of the invention have been specifically described herein, it will be apparent to those skilled in the art to which the invention pertains that variations and modifications of the various embodiments shown and described herein may be made without departing from the spirit and scope of the invention. Accordingly, it is intended that the invention be limited only to the extent required by the appended claims and the applicable rules of law.

Claims

What is claimed is

1. An immunogenic composition for use in prevention of malaria comprising a fragment of AnAPNl.

2. The method of claim 1, further comprising a malaria antigen.

3. The immunogenic composition of claim 2, wherein the malaria antigen is an antigen from Plasmodium falciparum or Plasmiodium vivax.

4. The immunogenic composition of claim 2, wherein the malaria antigen is an immunogenic particle RTS,S.

5. The immunogenic composition of claim 1, wherein the malaria antigen is an antigen from the sporozoite, gametocyte, gamete, or ookinete stages of Plasmodium falciparum or Plasmiodium vivax.

6. The immunogenic composition of claim 1, wherein the fragment of AnAPNl NT135APN1.

7. The immunogenic composition of claim 1, further comprising an adjuvant.

8. The immunogenic composition of claim 1, wherein the fragment of AnAPNl comprises amino acids 32-64 of SEQ ID NO.: 1.

9. The immunogenic composition of claim 1, wherein the fragment of AnAPNl is a recombinant polypeptide.

10. A method for preventing malaria comprising the step of administering an

effective amount of the composition of claim 1 to an individual.

11. A method for preventing the transmission of malaria comprising the step of stimulating an immune response against AnAPNl .

12. The method of claim 11, wherein the stimulating step comprises administering to an individual a fragment of AnAPNl.

13. The method of claim 12, wherein the fragment of AnAPNl is NT135APN1.

14. The method of claim 12, wherein the fragment of AnAPNl comprises amino acids 32-64 of SEQ ID NO.: 1.

15. The method of claim 12, wherein the fragment of AnAPNl is a recombinant polypeptide.

16. A method for making an immunogenic composition for use in prevention or treatment of malaria comprising the step of admixing a fragment of AnAPNl and a malaria antigen.

17. The method of claim 16, further comprising the step of adding an adjuvant to the admixture.

18. The method of claim 16, wherein the malaria antigen is an antigen from

Plasmodium falciparum or Plasmiodium vivax.

19. The method of claim 16, wherein the malaria antigen is an immunogenic particle RTS,S.

20. The method of claim 16, wherein the malaria antigen is an antigen from the sporozoite, gametocyte, gamete, or ookinete stages of Plasmodium falciparum or Plasmiodium vivax.