Patent Application
20190016782
The ASCII file, entitled 74493SequenceListing.txt, created on Jul. 4, 2018, comprising 5,550 bytes, submitted concurrently with the filing of this application is incorporated herein by reference.
The present invention, in at least some embodiments, relates to a stabilized form of a SCP (Signal Converting Protein) fusion protein, and in particular, to a homohexamer form of such a fusion protein.
Signal-Converting-Proteins (SCP) which are currently known in the art are bi-functional fusion proteins that link an extracellular portion of a type I membrane protein (extracellular amino-terminus), to an extracellular portion of a type II membrane protein (extracellular carboxyl-terminus), forming a fusion protein with two active sides (see for example U.S. Pat. No. 7,569,663). CTLA4-FasL is a SCP in which the N-terminal side is composed of the extracellular domain of CTLA-4, a Type-I membrane protein that binds with high affinity to B7 receptors, and the C-terminal side is composed of extracellular domain of Fas-ligand (FasL), a Type-II membrane ligand that induces cell apoptosis.
CTLA-4 (Cytotoxic T-Lymphocyte Antigen 4), also known as CD152, is a protein receptor that naturally down-regulates the immune system via inhibition of T-cell activation. T-cell activation requires co-stimulatory binding between the CD28 receptor to the CD80 and CD86 receptors, also called B7-1 and B7-2 respectively, on antigen-presenting cells. CTLA-4 is similar in function to CD28 in that they both bind to B7, however, whereas CD28 transmits a positive T-cell activation stimulatory signal, CTLA-4 binds B7 receptors at higher affinity but does not activate T-cells, thus, competing with CD28 and down-regulating the immune-system. The membrane-bound CTLA-4 is known to function as a homodimer that is interconnected by a disulfide bond.
CTLA4's strong binding affinity to B7 led to the design of protein-based therapeutics, linking the CTLA4 extracellular domain to an antibody Fc domain (CTLA4-Fc), that is already approved for use in autoimmune diseases and transplantation (Herrero-Beaumont G, Martinez Calatrava M J, Castafieda S. Abatacept mechanism of action: concordance with its clinical profile. Rheumatol Clin. 2012 March-April 2012; 8(2):78-83). In these chimeric constructs, both the CTLA4 and the Fc domains form a natural homo-dimer (Arora S, Tangirala B, Osadchuk L, Sureshkumar K K. Belatacept: a new biological agent for maintenance of immunosuppression in kidney transplantation. Expert Opinion on Biological Therapy. 2012; 12(7):965-979).
FasL is a Type-II membrane protein that naturally binds and activates Fas-receptors (FasR), which induce cellular apoptosis, or programmed cell death. FasL and FasR belongs to the tumor necrosis factor (TNF) family and FasL/FasR interactions play an important role in the regulation of the immune system and the progression of cancer. FasL, like other TNF super-family members, functions as a non-covalently bound homo-trimeric protein that signals through trimerization of FasR, which usually leads to apoptosis of the “target” cell. Upon FasL binding and trimerization of FasR, a death-inducing signaling complex (DISC) is formed within the target cell, and subsequently apoptosis is induced. Studies indicate that two adjacent trimeric FasL are required for efficient FasR signaling and the formation of DISC (Holler et-al, Molecular and Cellular Biology, February 2003, p. 1428-1440. Eisele et-al, Neuro-Oncology 13(2):155-164, 2011).
Non-Hodgkin lymphomas (NHLs), as a disease set, is among the ten most prevalent malignant tumors, accounting for approximately 4% of all malignancies in both men and women. NHLs are of B or T-lymphocytes lineage with most (80-90%) of them being of B-cell origin. Though prognosis and treatment depend on specific type and stage, irradiation and chemotherapy have been proven effective in many NHL patients. New protein-based therapeutics, such as anti-CD20, have been recently added to the treatment toolbox. The overall 5-year survival rate has increased to approximately 50%, but there is still need for new effective treatment for the more aggressive and relapsing forms of the disease.
Activated B-cells are known to express high levels of B7 receptors, also known as CD80 (B7.1) and CD86 (B7.2), which are required for T-cell activation as part of a co-stimulatory signal between the T-cell CD28 receptor and the B7 receptors on antigen-presenting cells including B lymphocytes. Similarly to activated B-cells, B-cell lymphoma cells also express high levels of B7 molecules.
Signal-Converting-Proteins (SCP) are a novel type of bi-functional fusion proteins that are formed by directly linking an extracellular domain of a type I membrane protein (extracellular amino-terminus), to the extracellular domain of a type II membrane protein (extracellular carboxyl-terminus), creating a fusion protein with two active sides. CTLA4-FasL is one such SCP, in which the N-terminal side is the extracellular domain of CTLA-4 and the C-terminal side is composed of the extracellular domain of Fas-ligand (FasL) (J.H. H, M.L. T. CTLA-4-Fas ligand functions as a trans signal converter protein in bridging antigen-presenting cells and T cells. International Immunology. 2001; 13 (4):529-539). Since CTLA4-FasL has the ability to bind to B7 molecules and to FasR, and in doing so, concurrently, to inhibit co-stimulation and induce apoptosis. CTLA4-FasL has been shown to efficiently induce apoptosis of activated T-cells (Orbach A, Rachmilewitz J, Parnas M, Huang J H, Tykocinski M L, Dranitzki-Elhalel M. CTLA-4. FasL induces early apoptosis of activated T cells by interfering with anti-apoptotic signals. J Immunol. December 2007; 179(11):7287-7294) and to function as a strong immunomodulator in multiple autoimmune and transplantation animal models (Zhang W, Wang F, Wang B, Zhang J, Yu J Y. Intraarticular gene delivery of CTLA4-FasL suppresses experimental arthritis. Int Immunol. June 2012; 24(6):379-388; Jin Y, Qu A, Wang G M, Hao J, Gao X, Xie S. Simultaneous stimulation of Fas-mediated apoptosis and blockade of costimulation prevent autoimmune diabetes in mice induced by multiple low-dose streptozotocin. Gene Ther. June 2004; 11(12):982-991; Shi W, Chen M, Xie L. Prolongation of corneal allograft survival by CTLA4-FasL in a murine model. Graefes Arch Clin Exp Ophthalmol. November 2007; 245(11):1691-1697; Feng Y G, Jin Y Z, Zhang Q Y, Hao J, Wang G M, Xie S S. CTLA4-Fas ligand gene transfer mediated by adenovirus induce long-time survival of murine cardiac allografts. Transplant Proc. June 2005; 37(5):2379-2381). Recently, the present inventors have shown that CTLA4-FasL can induce robust apoptosis of B cell lymphoma cell lines by activating pro-apoptotic signals in parallel to abrogating anti-apoptotic ones (Orbach A, Rachmilewitz J, Shani N, et al. CD40-FasL and CTLA-4•FasL fusion proteins induce apoptosis in malignant cell lines by dual signaling. Am J Pathol. December 2010; 177(6):3159-3168).
Unexpectedly, the present inventors found that CTLA4-FasL fusion proteins are more stable in solution as homo-hexamers, and that in fact the purification and production process for this fusion protein may optionally be adjusted so that the homo-hexamer form is the majority form of the fusion protein. Furthermore, the increased stability of such fusion proteins may be extended to other SCP (Signal Converting Protein) fusion proteins having properties as described below.
By “CTLA4-FasL fusion protein” it is meant a bi-component protein featuring a CTLA4 domain and a FasL domain as described herein which are linked covalently. This fusion protein is also referred to herein as “CTLA4-FasL”. Optionally and preferably, the bi-component protein comprises the extracellular domain of CTLA-4 and the extracellular domain of Fas-ligand (FasL). Optionally and more preferably, the bi-component protein has an N-terminal side which is the extracellular domain of CTLA-4 and a C-terminal side which is composed of the extracellular domain of Fas-ligand (FasL).
Although this form is referred to as a homo-hexamer, it may optionally also be described as a multimer of approximately 250 kD. Without wishing to be limited by a single hypothesis, it is believed that two different types of homo-hexamer structures may optionally form in terms of the interactions, which may be covalent or non-covalent. In a first type, the homo-hexamer forms via interactions of and hence dimerization of two FasL trimers. In a second type, the homo-hexamer forms via interactions of and hence trimerization of three CTLA4 dimers. Of course these different types are presented as non-limiting examples only.
These findings (that CTLA4-FasL fusion proteins are more stable in solution as homo-hexamers) are particularly unexpected because nothing in the characteristics of either component of the fusion protein would have previously lead one of ordinary skill in the art to predict such an outcome. Fusion of a dimeric protein such as CTLA-4 to a trimeric TNF super-family member such as FasL opens up the possibility that oligomers may form. For example, covalently linked homo-dimers can form via the formation of the natural, disulfide-linked, dimer of CTLA-4, while homo-trimers can form via the natural requisite of FasL to trimerize. One article that studied this question by using chemical cross-linking and gel filtration analyses showed that CTLA4-FasL complexes do contain intermolecular disulfide-bridges but indicated a trimeric molecular stoichiometry (Huang and Tykocinski, Int Immuno 2001, vol 13, no 4, pp 529-539). The authors indicated that since the main structures they identified were homo-trimers and the formation of a homo-hexamer in solution is not likely, homo-hexamers might form on the surface of the target cell, inducing the extremely effective apoptosis they have recorded.
According to at least some embodiments of the present invention, there is provided a stable CTLA4-FasL fusion protein, wherein in solution, a majority of the fusion proteins are in the homo-hexamer form.
According to at least some embodiments of the present invention, there is provided a method for producing a stable CTLA4-FasL fusion protein, wherein the protein is characterized in that in solution, a majority of the fusion proteins are in the homo-hexamer form. Optionally and preferably, the homo-hexamer form is present as the majority form after initial harvesting, but before any purification methods have been applied to the protein.
By “initial harvesting” it is meant for example after cell media has been harvested, such that cells are removed from the media (optionally through centrifugation or filtration for example), as the protein is secreted into the cell media from the cells. The initial harvesting process is preferably performed such that the homo-hexamer form is present as the majority form in the cell media. Furthermore, the purification process is optionally performed so as to maintain the homo-hexamer form as the majority form during purification. By “majority form” it is meant that the homo-hexamer form is at least 51% of the fusion proteins. Unless otherwise indicated, percentages of the fusion protein are given with regard to the total fusion protein amount, and not necessarily with regard to the total protein content.
According to at least some embodiments, the homo-hexamer form is optionally at least 51%, preferably at least 60%, more preferably at least 70%, most preferably at least 80%, optionally and most preferably at least 90%, also optionally and most preferably at least 95%, 96% and so forth up to 100% of the total fusion protein; optionally any percentages between these amounts may also be contemplated within these embodiments.
According to at least some embodiments, the homo-hexamer form is optionally the majority form before purification and is preferably the majority form after purification. If the homo-hexamer form is not a majority form, then it is at least a significant minority (preferably at least 20%, more preferably at least 30%, most preferably at least 40% and optionally and most preferably 50%; optionally any percentages between these amounts may also be contemplated within these embodiments).
According to at least some embodiments, a dodecamer form is present, optionally before and/or during and/or after purification. Optionally the dodecamer form is the majority form during purification; optionally and preferably, regardless of whether the dodecamer form is the majority form during purification, the homo-hexamer form is the majority form at least after purification and optionally before purification. Although this form is referred to as a dodecamer, it may optionally also be described as a multimer of approximately 500 kD.
According to at least some embodiments, the dodecamer form is optionally no more than 5% of the total fusion protein. Preferably, the dodecamer form is no more than 4% of the total fusion protein. More preferably the dodecamer form is no more than 3% of the total fusion protein. Most preferably, the dodecamer form is no more than 2% of the total fusion protein. Optionally and most preferably, the dodecamer form is no more than 1% of the total fusion protein.
According to at least some embodiments, the above embodiments, regarding the amounts of the dodecamer and the amounts of the homo-hexamer, are optionally combined.
According to at least some embodiments, there is provided a stable homo-hexamer SCP (Signal Converting Protein) fusion protein in majority form. As used herein, the term “fusion protein” generally refers to a protein in which the N-terminal end forms a homodimer and has a specific biological function, and the C-terminal side forms a homotrimer and has a different biological function. The N-terminal end is referred to as the “component 1 protein” while the C-terminal end is referred to as the “component 2 protein.
According to at least some embodiments of the present invention, the fusion protein is a fusion of a Type-I protein and a Type-II protein, in which the Type-I is known to form a homodimer and the type-II is known to form a homotrimer. Based on the data presented herein for CTLA4-FasL fusion protein, these fusion proteins should form homo-hexamers and offer improved activity.
The above description regarding purification of the stable homo-hexamer and the definition of majority form also apply to these general fusion proteins. Examples are given below for component 1 and component 2 proteins. It is understood that optionally a functional portion of the protein may be used to form the fusion protein, such as (where applicable) an extracellular portion of the protein.
For the first protein (component 1), non-limiting examples include many receptors or ligands that naturally form homodimers, optionally including disulfide-linked dimers, e.g., BTN3A1, CD27, CD80, CD86, ENG, NLGN4X, CD84, TIGIT, CD40, IL-8, IL-10, CD164, LY6G6F, CD28, CTLA4, TYROBP, ICOS, VEGFA, CSF1, VEGFB, BMP2, BMP3, GDNF, PDGFC, PDGFD, TGFB1, LY96, CD96 and GFER. Of these component 1 proteins, at least CD28, CTLA4, TYROBP, ICOS, VEGFA, CSF1, VEGFB, BMP2, BMP3, GDNF, PDGFC, PDGFD, TGFB1, LY96, CD96 and GFER are expected to form disulfide links to form the homodimers as part of the formation of the homo-hexamer, without wishing to be limited by a single hypothesis.
For the second protein (component 2), some non-limiting examples include all TNF-superfamily ligands, e.g., FasL, TRAIL, TNF-alpha, TNF-beta, OX40L, CD40L, CD27L, CD30L, 4-1BBL, RANKL, TWEAK, APRIL, BAFF, LIGHT, VEGI, GITRL, EDA1/2, Lymphotoxin alpha and Lymphotoxin beta (reference: Tansey and Szymkowski, Drug Discovery Today, 2009 December; 14(23-24)).
Optionally and preferably, the TNF-superfamily ligands for the component 2 protein form a non-covalent homo-trimer as part of the formation of the homo-hexamer. Although it is well documented that TNF-superfamily ligands naturally form homotrimers, their most efficient activation receptor-activation structure seems to be as two adjacent trimers or as a homo-hexamer (Eisele et al, APO010, a synthetic hexameric CD95 ligand, induces human glioma cell death in vitro and in vivo; Neuro-Oncology vol 13(2): pp 155-164, 2011; Wyzgol et al, Trimer Stabilization, Oligomerization, and Antibody-Mediated Cell Surface Immobilization Improve the Activity of Soluble Trimers of CD27L, CD40L, 41BBL, and Glucocorticoid-Induced TNF Receptor Ligand, J. Immunol. 2009; vol 183; pp 1851-1861; and Holler et al, Two Adjacent Trimeric Fas Ligands Are Required for Fas Signaling and Formation of a Death-Inducing Signaling Complex, Mol. Cell. Biol. 2003, vol 23(4): pp 1428-1440).
According to at least some embodiments of the present invention, the component 1 protein is a type-I protein, while the component 2 protein is a type-II protein. Optionally and preferably for such embodiments, the bi-component protein comprises the extracellular domain of the type-I protein and the extracellular domain of the type-II protein. Optionally and more preferably, the bi-component protein has an N-terminal side which is the extracellular domain of the type-I protein and a C-terminal side which is composed of the extracellular domain of the type-II protein.
Exceptions to the above embodiment involving extra-cellular proteins may optionally include proteins which do not feature transmembrane domains, yet which may still function as component 1 or component 2 proteins. Non-limiting examples of such proteins for component 1 proteins include VEGFB, BMP2, BMP3, GDNF, PDGFC, PDGFD, TGFB1, GFER and LY96. Non-limiting examples of such proteins for component 2 proteins include April and TNF-beta. Of course, other exceptions to the above embodiment may also occur and are encompassed within various embodiments of the present invention.
Non-limiting examples of disease indications and potential fusion protein combinations are given below, although of course other/different disease indications and/or fusion protein combinations are possible and are encompassed within various embodiments of the present invention. By “combination” as described herein with regard to fusion protein components, it is meant bi-component fusion proteins comprising a component 1 protein (or functional portion thereof) and a component 2 protein (or functional portion thereof).
1. For immune induction (e.g., for treatment of cancer, although optionally also for treatment of other diseases for which the subject would benefit from immune induction), combine component 1 homodimer candidates that may have immune-targeting or immune-induction activities (e.g., CD28, CSF1, LY96, CD96, BTN3A1, TIGIT, IL8, IL10, CD80, CD86, LY96), with component 2 homotrimer candidates that may induce the immune system and induce growth (e.g., TNF, OX40L, CD40L, CD27L, CD30L, 4-1BBL, TWEAK, APRIL, BAFF, LIGHT, GITRL).
2. For immune suppression and/or cancer cell apoptosis, combine component 1 homodimer candidates that may have immune-targeting or immune-inhibitory activities (e.g., CTLA4, CD40, TYROBP, CD27, ENG), with component 2 homotrimer candidates that may activate apoptosis or suppress growth (e.g., FasL, TRAIL, VEGI).
3. For induction of angiogenesis for treatment of any disease for which the subject would benefit from angiogenesis induction, combine component 1 homodimer candidates that may have angiogenic activities (e.g., VEGFA), with component 2 homotrimer candidates that may activate angiogenesis or activate growth (e.g., TWEAK, APRIL).
4. For inhibition of angiogenesis for treatment of any disease for which the subject would benefit from angiogenesis inhibition, combine component 1 homodimer candidates that may have anti-angiogenic activities (e.g., ENG), with component 2 homotrimer candidates that may suppress angiogenesis or activate apoptosis (e.g., FasL, TRAIL, VEGI).
5. For induction of bone formation, combine component 1 homodimer candidates that may have bone formation activities (e.g., BMP2), with component 2 homotrimer candidates that may activate cell growth and bone formation (e.g., TWEAK, APRIL).
6. For inhibition of bone formation, combine component 1 homodimer candidates that may have inhibitory bone formation activities (e.g., BMP3), with component 2 homotrimer candidates that may activate bone resorption, apoptosis or suppress growth (e.g., RNAKL, FasL, TRAIL, VEGI).
7. For liver regeneration for treatment of any disease for which the subject would benefit from liver regeneration, combine component 1 homodimer candidates that may have liver regeneration activities (e.g., GFER), with component 2 homotrimer candidates that may activate cell growth and (e.g., TWEAK, APRIL).
Table 11 shows some purification attempts and results (given in the Figures).
According to at least some embodiments, the present invention relates to a stable homo-hexamer SCP fusion protein, comprising a Component 1 protein and a Component 2 protein. The Component 1 protein is optionally selected from the group consisting of many receptors or ligands that naturally form homodimers, especially disulfide-linked dimers, e.g., BTN3A1, CD27, CD80, CD86, ENG, NLGN4X, CD84, TIGIT, CD40, IL-8, IL-10, CD164, LY6G6F, CD28, CTLA4, TYROBP, ICOS, VEGFA, CSF1, VEGFB, BMP2, BMP3, GDNF, PDGFC, PDGFD, TGFB1, LY96, CD96 and GFER. Of these component 1 proteins, at least CD28, CTLA4, TYROBP, ICOS, VEGFA, CSF1, VEGFB, BMP2, BMP3, GDNF, PDGFC, PDGFD, TGFB1, LY96, CD96 and GFER form disulfide links to form the homodimers as part of the homo-hexamer formation process.
The Component 2 protein optionally comprises any TNF-superfamily ligand and is preferably selected from the group consisting of FasL, TRAIL, TNF-alpha, TNF-beta, OX40L, CD40L, CD27L, CD30L, 4-1BBL, RANKL, TWEAK, APRIL, BAFF, LIGHT, VEGI, GITRL, EDA1/2, Lymphotoxin alpha and Lymphotoxin beta (reference: Tansey and Szymkowski, Drug Discovery Today, 2009 December; 14(23-24)).
The homo-hexamer may optionally be formed through the formation of covalent bonds, non-covalent bonds or a combination thereof. Optionally, at least the following component 1 proteins form disulfide bonds as part of the homo-hexamer formation process: CD28, CTLA4, TYROBP, ICOS, VEGFA, CSF1, VEGFB, BMP2, BMP3, GDNF, PDGFC, PDGFD, TGFB1, LY96, CD96 and GFER.
Additional information with regard to component 1 proteins is provided in Table 1 below. This information includes details with regard to binding and interaction partners, the biological mechanisms associated with binding to and/or interacting with such binding and interaction partners, and tissue specificity. A similar table is provided for component 2 proteins as Table 2.
It should be noted that TYROBP is listed below (and for the above embodiments) with multiple types of functions, because it is a signaling adapter with dual functionality. Although in association with several receptors it can activate effector cell functions that play a critical role in mediating immune responses against bacterial and viral infections and tumors. TYROBP can also negatively regulate immune responses, including NK cells and macrophages and can play an important role in limiting cytokine production mediated by TLR and FcERIg pathways.
Table 3 provides a description for determining which component 1 and component 2 proteins can partner. Specifically, Table 3 lists various disease indications, followed by component 1 activities and then component 1 proteins, followed by component 2 activities and component 2 protein partners for the component 1 proteins.
According to at least some embodiments of the present invention, for each disease indication, any listed component 1 protein in that row can partner with any listed component 2 protein in that row. However, this list is meant for the purposes of illustration only and is not intended to be limiting in any way. Furthermore, the combinations described below may optionally be used to treat other (alternative or additional) diseases, while the diseases below may optionally be treated with other (alternative or additional) combinations.
As described herein, many different component 1 and component 2 proteins, as well as functional portions thereof, may optionally form fusion proteins that have stable homo-hexamer forms. CTLA4-FasL fusion protein is a non-limiting example of such a fusion protein, for which detailed experimental methods and results are described below. However, the present inventors believe that the other component 1 and component 2 proteins (and functional portions thereof) would form SCP fusion proteins showing at least similar behavior.
Materials and Methods
Construction and Isolation of a CHO-S Production Clone
The CTLA4-FasL amino acid sequence of
CHO-S cells were transfected with the UCOE/CTLA4-FasL expression vector, shown in
Clones with the highest expression were then expanded into 10 ml Mix6 in 125 ml shaker flasks at 50 rpm and maintained on puromycin selection. After 4 days the shaker speed was increased to 100 rpm and cells were counted for viability at regular intervals and expanded as necessary. Expression analysis was performed on clones under standard non-supplemented media conditions at small scale in 125 ml shaker flasks. Cultures were maintained until cell viability approached 80% and expression levels were then estimated by FasL ELISA. The protein product was analyzed by SDS-PAGE and Western blots showing that all clones produce a protein product of similar size and one clone, showing the highest levels of expression, was selected for limiting dilution. Limiting dilutions were carried out in two different occasions with 7 days in between; in the first seeding occasion, cells were seeded in 96 well plates in 50% conditioned media at cells/well ratios of 800, 400, 200, 100, 50, 25, and 10 cells per well. Based on FasL ELISA analysis of the first seeding ratios, cells were diluted to lower seeding densities of; 50, 25, 12.5, 6.25, 3.13, 1.56, and 0.78 cells per well. The 96-well plates were incubated in a HeraCell incubator for 2-3 weeks at which point all plates were assessed visually for growth. Clones with the lowest seeding cell density were selected from the 96-well plates and transferred to 24-well plates and diluted 1:10 in cell culture media. After 7 days of culture the cell suspension was harvested and analyzed by ELISA and the ten highest producing clones were transferred from 24-well plates to 6-well plates by a dilution of 1:4 in cell culture media. When sufficient cells had been received in 6-well plates, the ten clones were transferred from 6-well plates to 125 mL shake flasks and their growth profile and CTLA4-FasL expression investigated. One clone was selected as the final clone based on that analysis.
SDS-PAGE and Western Blotting
Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was performed by using 4-12% Bis-Tris gel (1 mm, 12 wells, NP0322BOX, Life Technologies) and “See Blue Plus 2” MW markers (LC5925, Life Technologies). Proteins were transferred to a PVDF-membrane and the membrane blocked using dried milk. The primary antibodies were either goat anti-human CTLA4 antibody (AF-386-PB, RandD Systems, 1:300 dilution) or goat anti-human Fas Ligand (AB126, RandD Systems, 1:100 dilution). The secondary antibody was a donkey anti-Sheep/Goat Immunoglobulins (HRP, AP360, The Binding Site, 1:10,000 dilution), detected by HRP substrate 3,3′, 5,5′-Tetramethylbenzidine (TMB, Liquid Substrate System for Membranes, T0565, Sigma).
For Western blot analysis of intracellular proteins, whole cell lysates were separated on 10% SDS-PAGE and blotted according to standard procedures.
Membranes were incubated with the following primary antibodies: anti Caspase-3, anti Caspase-8, anti Caspase-9, PARP, Bcl-2, IAP-1,2, pNFkB, pJNK, pERK1/2 (1:1000), (Cell Signaling Technology, Danvers, Mass., USA); XIAP (1:100) (Santa Cruz Biotechnology, Santa Cruz, Calif., USA); FLIP (1:500) (Enzo, Calif., USA); BID (1:4000), IkB (1:20,000) (RandD, Minneapolis, Minn., USA); Bcl-x (1:1000) (BD Biosciences, NJ, USA); GAPDH (1:500; Millipore, Billerica, Mass., USA). Secondary detection was performed with HRP-conjugated antibodies (1:10,000; Biorad, Hercules, Calif., USA).
Native-PAGE analysis was performed with NativePAGE™ Novex® 4-16% Bis-Tris Gel (Invitrogen), according to the manufacturer protocol. Samples were prepared with and without G-250 sample additive. 10 uL of the CTLA4-FasL sample and of the NativeMark were loaded to each gel lane. Coomassie G-250 was added to the cathode buffer and to the samples, resulting in staining of the proteins during gel electrophoresis.
Con-A/SEC Purification Process
Thawed production harvest was centrifuged at 5000×g, followed by 0.2 μm filtration (10 kDa cut-off cellulose centrifugal filters; Sartorius-Stedim, Goettingen, Germany) and loaded onto the Concavalin-A (Con A) (GE Healthcare, Little Chalfont, UK) at 7 mg/mL resin. The Con-A eluate was loaded onto a Size-Exclusion-Chromatography (SEC) Sephacryl S-200 column (GE Healthcare). The SEC eluate was 0.2 μm filtered (Minisart syringe filter) (Sartorius-Stedim) and frozen at −70° C.
Table 4 lists some non-limiting materials and Table 5 lists some non-limiting equipment. Tables 6-8 provide more chromatographic details.
Analytical SE-HPLC
Analytical size-exclusion (SE) was performed using a Dionex HPLC instrument (Pump P580, Auto sampler ASI-100/ASI-100T Injector, UVNIS Detector UVD340U, Chromeleon 6.80 Software) with Tosoh Bioscience TSK-Gel G3000SWXL 7.8×300 mm column. Phosphate Buffered Saline (PBS) was used as the mobile phase and samples of <50 μg or 100 μg injected. Reference standards and 25% GFS (gastric fluid simulant) were run before and after the samples. The column was equilibrated by running with mobile phase (PBS) at flow rate 0.1 ml/min. The separation was performed using an isocratic separation method with a runtime of 20 min and a flow rate of 1 ml/min. The column oven was set at 25° C. and the sample holder at 8° C.
Iso Electric Focusing (IEF): CTLA4-FasL was separated on IEF gels (Novex, Life technologies, NY, USA), pH3-7 and pH3-10 according to the manufacturer's instructions.
His6-Tagged Protein:
In-vitro experiments were performed with a His6 tagged version of CTLA4-FasL24. The activity of the tagged His6CTLA4-FasL was compared to that of the purified non-tagged CTLA4-FasL and found to be identical (not shown).
Cell Lines.
Liver adenocarcinoma SK-HEP-1 cell line, A498 Renal Carcinoma Cell line and Raji B cell lymphoma cell line were purchased from ATCC (Manassas, Va., USA). Other lymphatic cell lines were a kind gift from the Gene Therapy institute and Hepatology Unit, Hadassah Hebrew University Medical Center in Jerusalem, Israel. Attached cells were grown in DMEM (Gibco) supplemented with 10% FBS, 2 mM glutamine, 100 IU/mL penicillin and 100 μg/mL streptomycin, and were detached using Trypsin-EDTA solution. Suspended cells were grown in RPMI (Gibco) with the same additives. All cell lines were cultured at 37° C., 6% CO2.
Immuno-Histo-Chemistry
B cell lymphoma tissue microarray (TMA) (US Biomax; # LM801a, lymphoma tissue array with adjacent normal lymph node and spleen tissue as control, 80 cases) paraffin sections were deparaffinized in xylene (3×3′) and rehydrated in graded alcohol (3×1′ 100% ethanol; 3×1′ 96% ethanol). Following 5′ incubation in 3% H202 for endogenous peroxidase inactivation, slides were incubated in Citrate buffer (pH6; #005000; Invitrogen) and boiled in electric pressure cooker (DC2000; BioCare Medical) for antigen retrieval. Samples were blocked for 20′ in CAS-BLOCK (#00-8120; Invitrogen) prior to overnight incubation with the primary antibodies at 4c in humidified box (see below table).
Following washing (3×2′ in Super Sensitive wash buffer, #HK583-5K; BioGenex), samples were incubated for 30′ in RT with the relevant secondary antibody (see below table). Diaminobenzidine (DAB; UltraVision Detection System, TA-125-HDX, Thermo scientific) was used as the chromogen according to manufacturer instructions, and 20″ incubation in hematoxylin (MHS16, SIGMA-Aldrich) was used as the nuclear counter-stain. Following dehydration steps (2′ 80% ethanol, 2′ 96% ethanol, 2′ 100% ethanol, 2′ xylene) and mounting (Histomount mounting solution, #0080-30; Invitrogen), Staining intensity was quantified using the Ariol SL50 automated robotic image analysis system, according to manufacturer instructions.
FACS Analysis
Approximately 1×106 cells were washed with PBS in a FACS tube and re-suspended in 95 ul of staining buffer (1% BSA, 0.1% azide in PBS) and 5 ul of human Fc blocker (#422302; e-Bioscience), and incubate for 5′ on ice. The appropriate antibody/isotype (see below Tables 10A and 10B) were added and incubated in the dark on ice for 30′. Cells were washed with PBS, re-suspended in 350 ul staining buffer, and filtered into a clean FACS tube via a 40 uM filter and kept in the dark on ice until analyzed by a BD™ LSR II Flow Cytometer, according to the manufacturer instructions; 20,000 events per sample were counted. Data were analyzed using CellQuest software (Becton Dickinson).
CTLA4-FasL Quantification by Gyrolab
To efficiently quantify CTLA4-FasL, a Gyrolab platform immunoassay (Gyrolab Workstation, Gyros) was developed. This immunoassay utilizes capture and detection by two different antibodies specific to different regions of CTLA4-FasL. The assayed sample was transferred by centrifugal force through Streptavidin beads coated with anti-CTLA4 antibodies, that were packed in minute columns (15 nL) and quantification was performed by the detection of an anti-FasL antibody that was bound to the product in the column, by laser induced fluorescence.
An anti-Human CTLA-4 polyclonal goat antibody (AF-386PB, RandD systems) was selected as capture antibody. The polyclonal antibody was generated using recombinant human CTLA4 Ala37-Phe162 (Accession # Q6GR94) expressed in S. frugiperda insect ovarian cell line Sf 21. The polyclonal antibody was biotinylated using EZ-link Sulfo-NHS-LC-Biotin as described in the instructions supplied with the kit. The biotinylated material was purified by a size-exclusion spin column and the material kept in PBS.
An anti-Human Fas Ligand monoclonal mouse IgG2B antibody (MAB-126, RandD systems) was selected as detection antibody. The monoclonal antibody was generated using recombinant human Fas Ligand/TNFSF6, Pro134-Leu281 (Accession #P48023) expressed in Chinese hamster ovary cell line CHO. The monoclonal antibody was Alexa labeled using the Alexa Fluor™ 647 Monoclonal Antibody Labelling Kit as described in the instructions supplied with the kit. The labeled material was purified by using a size-exclusion spin column included in the kit. The Alexa labeling was measured to be 6.9 Alexa/Ab and the material diluted to 1 μM with 1% BSA in PBS and kept in the freezer.
CTLA4-FasL sample was transferred by centrifugal force through Streptavidin beads coated with anti-CTLA4 antibodies, that were packed in minute columns (15 nL) and quantification was performed by the detection of an anti-FasL antibodies that were bound to the CTLA4-FasL in the column, by laser induced fluorescence.
In-Vitro Activity Bioassay
For in vitro examination of the CTLA4-FasL cytotoxic effect on different human cell lines, 32,000 cells per well (suspended cultures) or 8000 cells per well (attached cells) in 50 ul of complete RPMI (suspended cultures) or DMEM (attached cells) medium without Phenol Red, were seeded in triplicates, in a flat 96-wells plate (Nunc or similar), and 50 ul of CTLA-4•FasL (or his6CTLA-4•FasL) dilutions (in growth media; 3000 ng/ml-0.1 ng/ml, triplicates), or dilution media as negative control were added. Calibration curve wells contained serial dilution from 64,000 to 2000 cells per well for suspended cultures or 16,000 to-2000 cells for attached cells in triplicates. Plates were incubated for 24 hours at 37c in 5% CO2 humidified incubator. Cell viability was quantified by a MTS kit (Promega, CellTiter 96® Aqueous Non-Radioactive Cell Proliferation Assay) according to manufacturer instructions.
Mouse Disease Models
Xenograft Lymphoma Model:
Athymic-NUDE female mice (Harlan, Israel), 4-6 weeks of age, were maintained under defined flora conditions at the Hebrew University Pathogen-Free Animal Facility. All experiments were approved by the Animal Care Committee of the Hebrew University. The JY cells used in this study were harvested from subcutaneous JY xenograft tumor, and expanded in culture. Mice were irradiated (300R), and two days later JY cells in exponential growth were harvested, washed with PBS, and injected subcutaneously (7-10×106/mouse) into the right flanks of mice. When tumors were palpable, at day 5, treatment was started. Mice were treated for 4 days with two 100 micro-liter subcutaneous injections per day of CTLA4-FasL or the vehicle buffer (PBS). Tumor size was measured by a micro caliber and volume was calculated by the equation: (w 2*length/2). Mice bearing tumor of >1000 mm3 or necrotic tumors were sacrificed. In some experiments, and to further assess CTLA4-FasL effect on JY-derived tumors, mice were sacrificed one hour post the 1st injection, at the 4th injection day (20 micro-gram CTLA4-FasL per day). SC tumors were removed and fixated for in 4% formaldehyde, routinely processed, and embedded in paraffin. Transverse sections (5 μm) were stained with hematoxylin and eosin (HandE).
A Mouse/Mouse Model for Lymphatic Cancer:
Balb/c mice were injected intravenously with 1×106 Bcl-1 cells in 200 ul PBS (murine B cell leukemia splenocytes; Ref: Slavin S, Nature, v 272, p 624, 1978). From the next day, mice were treated for 3.5 days with two 100 ul subcutaneous injections per day of CTLA4-FasL or the vehicle buffer (PBS). Disease parameters were monitored by measuring mouse weight, spleen weight and blood counts.
Pharmacokinetics:
For analysis of pharmacokinetics, CTLA4-FasL at different doses was subcutaneously injected to mice at a total volume of 150 microliter per mouse.
Mice were sacrificed at various time points post injection. Blood was collected in heparin, kept on ice, centrifuged at 1000 g (˜3000 rpm) for 10′, plasma was kept at −70c. CTLA4-FasL was quantified by LEGEND MAX™ Human Soluble CTLA-4 ELISA kit (Biolegend #437407), according to the manufacturer instructions.
Results and Discussion—Purification
As shown in the previous results, CTLA4-FasL was found to be most stable in a form which is described herein as either a homo-hexamer or a multimer of approximately 250 kD. Without wishing to be limited by a single hypothesis, it is believed that this highly stable form of CTLA4-FasL is in fact a homo-hexamer in solution, based upon the data; however, it should be noted that the data definitively supports the concept that the protein is a multimer of approximately 250 kD molecular weight.
Such results are surprising, because previous authors did not realize that CTLA4-FasL would actually preferentially form such a stable homo-hexamer upon secretion to cell media, and that this preferential form would maintain its configuration during purification in solution. Furthermore, previous authors did not realize that this form would be a highly stable form of CTLA4-FasL in solution, and that purification peaks which preferentially contain the homo-hexamer form would contain the majority of in vitro functional activity for the CTL4-FasL fusion protein.
As noted previously, in order to produce the fusion protein CTLA4-FasL, a gene coding for the human sequence of CTLA4-FasL, linked to the human Urokinase signal peptide, was cloned into an expression vector (
Western blot analysis of the production media, using monoclonal antibodies to both human CTLA-4 and FasL (not shown), showed a band that specifically reacted with both anti-CTLA4 and anti-FasL. Although the predicted molecular weight of CTLA4-FasL is approximately 31 kD, the CTL4-FasL fusion protein migrated in a reduced SDS-PAGE as a protein of approximately 43 kD. The difference between the calculated and observed molecular weight of CTLA4-FasL was not investigated in the past and by treating production media samples with the “Peptide N-Glycosidase F” enzyme, that removes N-glycan chains from the protein followed by Western blot analysis, it was found that treatment with the enzyme caused a shift in molecular weight from ˜45 kDa to ˜33 kDa, which appears to suggest that the apparent difference in MW is due to protein glycosylation (
Initial attempts to purify this protein were not successful. As shown in Table 11 (included within the figures), attempts to purify the protein from host-cell proteins, using art-known chromatographic methods, such as cation-exchange, anion-exchange and hydrophobic-interactions chromatography for example, which had proved reliable and successful for other proteins in the past, failed to be successful for purifying CTLA4-FasL at its expected molecular weight and PI.
Utilizing the glycosylation of CTLA4-FasL, a preliminary purification process was developed, in which Concanavalin-A (Con-A) chromatography was used as the main capture step, followed by size-exclusion chromatography (SEC), yielding CTLA4-FasL at over 90% purity as measured by SDS-PAGE (
The theoretical iso-electric point (pI) of CTLA4-FasL is 6.59. To measure the actual protein pI, the purified CTLA4-FasL was analyzed by Iso-Electric-Focusing. Surprisingly, the actual pI of the protein is approximately 4.5, significantly different from the theoretical one.
The in-vitro killing activity of the purified CTLA4-FasL was measured on both malignant and non-malignant human cell-lines and, as can be seen in
To further study the actual structure of CTLA4-FasL, purified CTLA4-FasL was initially analyzed by gel-filtration chromatography (specifically by running purified CTLA4-FasL (after ConA/SEC chromatography) on a Seperose-12 column (GE Healthcare, 100×1.6 cm˜200 ml). As can be seen in
Since this observed product size of about 250 kD was significantly larger than the predicted homo-trimer (e.g., ˜130 kD) suggested by others, analytical Size-Exclusion High-performance Liquid Chromatography (SE-HPLC) and native-PAGE were used to study the actual product size at higher resolution, as shown in
To test if the CTLA4-FasL homo-hexamer structure is formed only at high-concentrations of pure preparations of the protein, a similar SE-HPLC analysis was performed on harvested production media, before any purification was carried out, and the amount of CTLA4-FasL in the SE-HPLC fractions was quantified by CTLA4-FasL Gyrolab analysis. As can be seen in
The natural stochiometry of CTLA4-FasL is of great functional significance to its activity since the optimal functionality of FasL-related apoptosis is predicted to be linked to the formation of two FasL trimers, namely a homo-hexamer, which activates two FasR trimers at the target cell membrane (Holler et-al, MOLECULAR AND CELLULAR BIOLOGY, February 2003, p. 1428-1440. Eisele et-al, Neuro-Oncology 13(2):155-164, 2011). Thus, a homo-hexamer of CTLA4-FasL is predicted to be more competent than other oligomeric forms. To test that theory, a preparative SEC of CTLA4-FasL was performed and the bioactivity of fractions representing the homo-hexamer were compared to fractions representing lower and larger CTLA4-FasL oligomer forms. Concentrated clarified harvest was partially purified by Con-A chromatography followed by SEC fractionation on a Superdex 200 column. The SEC fractions were then analyzed by SEC-HPLC and the relative percentage of different product types [250 kD product (Main Peak), Low Molecular Weight (LMW) and High Molecular Weight (HMW, presumably the dodecamer form)] are indicated in Table 12.
Specific fractions were then pooled to represent four different product types (250 kD, HMW (dodecamer), LMW1, LMW2) and an overlay of the pools can be seen in
To assess the natural stability of the CTLA4-FasL homo-hexamer, purified CTLA4-FasL was subjected to different chemical and physical conditions, and their effects were studied by SE-HPLC analysis to determine whether the homo-hexamer form was maintained:
Freeze/Thaw Cycles:
Purified CTLA4-FasL went through repeated freeze/thaw (FT) cycles. 1×FT, 3×FT and 5×FT were performed when in each cycle samples were moved from −80° C. to RT until the samples were completely thawed. As can be seen in
Reduction and Oxidation:
Since disulfide-bonds between CTLA-4 domains are thought to play a major role in the formation of the CTLA4-FasL homo-hexamer, the effects of reduction and oxidation on hexamer stability were assessed via incubation (2-8° C. for ˜24 h) in increasing amounts of glutathione (GSH; 2.5, 5, 7.5 and 10 mM) or copper sulfate (CuSO4; 5, 50, 100 and 200 mM). Surprisingly, both reduction and oxidation leads to a dramatic shift from CTLA4-FasL homo-hexamer to dodocamers with no observed formation of lower MW oligomers (
An additional detailed purification process was performed to determine yield but also the ratio of CTLA4-FasL homo-hexamer to dodocamers. This process was performed as follows. The medium harvest to be purified (including CTLA4-FasL with other proteins) was applied to a 300 kDa UF/DF retentate membrane system. UF stands for ultrafiltration while DF stands for diafiltration. In this case, the membrane pore size had a MW (molecular weight) cut-off of 300 kDa. Proteins and other molecules that have a lower MW than 300 kDa pass through the membrane and are collected as filtrate. Molecules with a MW greater than 300 kDa are retained by the membrane and collected as retentate. The process enables concentration and buffer exchange. Even though CTLA4-FasL homo-hexamer has 250 kDa MW, it is retained by the membrane and collected as retentate.
Next, the material was applied to at least a phenyl sepharose column (additional columns could also be used at this point). Viral inactivation may also optionally be performed after application of the material to one or more columns and was in fact performed in this example. Optionally, a 30 kDa UF/DF retentate membrane system could be used and was in fact used in this example. After that a Superdex column was used for chromatography, which included the Superdex material from GE Healthcare Life Sciences (made from dextran covalently attached to cross-linked agarose). Again one or more other columns could also optionally be applied.
The resultant fractions after application to the Superdex column were analyzed as shown in
As previously described, the in-vitro killing activity of the purified CTLA4-FasL was measured on both malignant and non-malignant human cell-lines and, as can be seen in
The exceptional cancer killing activity of CTLA4-FasL, observed on lymphatic cancer cells (e.g., EC50 of 0.02 nM) as compared to other cell types, might be related to high expression of B7 receptors on these cells. To investigate possible correlation between drug activity and receptor expression, FACS was used to quantify the expression of the three target receptors of CTLA4-FasL, namely CD80 (B7.1), CD86 (B7.2) and CD95 (FasR), on the different human cancer cell lines. Results shown in
CTLA4-FasL Killing Activity is Apoptosis-Based and is Superior to its Two Subunits or their Combination
It was previously shown that his6-CTLA4-FasL induces efficient apoptosis of lymphatic cancer cells by utilizing a dual signaling pathway that includes Fas-mediated apoptosis of CD95 expressing cells, coupled to the abrogation of cFLIP expression in B7 expressing cells (Orbach A, Rachmilewitz J, Shani N, Isenberg Y, Parnas M, Huang J H, Tykocinski M L, Dranitzki-Elhalel M., Am J Pathol. 2010 December; 177(6):3159-68). It was also previously shown that CTLA4-FasL inhibitory effect on T lymphocyte activation is mediated by apoptosis induction, through the cascade of caspases (Orbach A, Rachmilewitz J, Parnas M, Huang J H, Tykocinski M L, Dranitzki-Elhalel M., J Immunol. 2007 Dec. 1; 179(11):7287-94). To further investigate CTLA4-FasL mode-of-action in cancer cell line, an experiment was performed to determine whether CTLA4-FasL killing effect can be abrogated by the pan-caspase inhibitor, Z-VAD, caspase 8 inhibitor (Z-IETD-FMK) and caspase 9 inhibitor (Z-LEHD-FMK) on malignant cell lines positive for FasR only. As can be seen in
SCP chimeras have been shown to confer superior activity over their parts, separately or in combination. However, this was tested previously only in target cells that express binding molecules to both SCP sides. As the hepatocellular carcinoma (HCC) cell line SK-HEP1 does not express B7 molecules (
Raji (B7+) and A498 (B7-) cells were incubated with 50 or 500 ng/ml CTLA-4•FasL, sFasL, CTLA-4-Fc or the combination of the last for 2 h. Cell lysates were subjected to immunoblotting with the following antibodies: caspase 3, caspase 8, caspase 9, FLIP and GAPDH. As could be seen from the activation of caspase 3, 8 and 9, in the A498 cell line, CTLA-4•FasL induces effective apoptosis only at high concentrations of 500 ng/ml (
Immuno-histo-chemistry (IHC) was then used to analyze the expression of these receptors on tissue arrays of lymphoma and normal human tissues. It was found that the three receptors can be quantified by IHC on human tissues and that these receptors show high expression levels in a significant fraction of patients (
CTLA4-FasL Reduces Tumor Size in a Dose-Dependent Manner and Increases Survival in a B-Cell Lymphoma Xenograft Model
Prior to initiation of studies in a mouse disease model, the basic pharmacokinetic (PK) parameters of CTLA4-FasL in mice were measured. The protein levels in the peripheral blood of Balb/c mice were quantified by a CTLA4 commercial ELISA at specific time points following subcutaneous injections. CTLA4-FasL levels were shown to reach the highest values approximately 2 hours post injection with TI/2 of approximately 4-5 hours post injection (
The in-vivo activity of the purified CTLA4-FasL protein was assessed in two mouse cancer disease models; the first, human lymphoma in mouse model, measures the effect of CTLA4-FasL treatment on growth of a xenograft human lymphatic tumor within immune-compromised mice. In this model, mice were treated with twice-daily subcutaneous injections of CTLA4-FasL for several days. As illustrated in
In agreement with tumor volume and the survival indexes, the high efficacy of CTLA4-FasL treatment of JY xenograft tumors was further illustrated by the histological analysis of JY tumors removed from the mice, showing clear atrophy of tumor tissue in mice treated with CTLA4-FasL injections. Immunostaining with anti-cleaved casapase 3 demonstrated that tumor cells in CTLA4-FasL treated mice undergo apoptosis (
Results and Discussion—Dodecamer
As previously described, one of the forms of the CTLA4-FasL fusion protein is a dodecamer (12 monomers); both before and following purification, it is typically a very minor component of the total fusion protein (approximately 5%) but during purification it can become a major component of the fusion protein.
Without wishing to be limited by a single hypothesis, the hexamer may optionally be formed in two different ways, as a dimer of trimers or as a trimer of dimers. The “Dimer of Trimers” model assumes that FasL trimerization is stronger or forms quicker than CTLA4 dimerization. In this model the hexamer forms via dimerization of two FasL trimers. This type of hexamer is very stable since all CTLA4 domains are dimerized and all FasL domains are trimerized.
In the “Trimer of Dimers” model, it is assumed that CTLA4 dimerization is stronger or forms quicker than FasL trimerization. In this model the hexamer will form by trimerization of three CTLA4 dimers. This hexamer is only partially stable since, although all CTLA4 are dimerized, three of the FasL are not trimerized. This latter type of hexamer could break apart, forming free CTLA4 dimer subunits, which could then join a “Trimer of Dimers” hexamer to form a stable dodecamer. This dodecamer is predicted to be very stable since, as with the “Dimer of Trimers” hexamer model, here also all CTLA4 domains are dimerized and all FasL domains are trimerized.
The next panel shows a dimer, which are two bonded fusion proteins. Next the trimer panel shows three bonded fusion proteins. The two different homo-hexamer structures are shown, followed by the dodecamer structure on the far right.
Unfortunately, as shown by the data below, the resultant dodecamer is quite toxic to mice. Injecting mice with very high dosages of the purified CTLA4-FasL preparation (10-20 times the therapeutic dose) was found to cause toxic effects, and investigation of these effects shown liver toxicity alongside a sharp increase in cytokine levels (data not shown). Since both SEC-HPLC and Native-PAGE analysis have shown that over 90% of the purified protein preparation is a homo-hexamer structure (250 kD), while the remaining 5-10% is a higher molecular weight (HMW) form, consisting with a homo-dodecamer structure (500 kD), experiments were performed to determine whether the efficacy and/or toxicity associated with this preparation is a result of the homo-hexamer or the dodecamer.
To study toxicity, experiments were performed to determine the levels of ALT/AST enzymes, which are released to the serum as a result of liver dysfunction (
Mice were injected with 10 ug or 50 ug of the CTLA4-FASL preparation, 10 ug of the purified HMW (dodecamer) fraction and PBS as control. Blood was collected 5 hours post injection. Quantitative determinations of ALT/AST from serum were performed using Reflotron test strip (Roche). Detection of Cytokines in serum was measured by flow cytometry (FACS) using FlowCytomix Kit (eBioscience). The cytokines detected were as follows: IL-5 (
As can be seen in
In the present study the unique properties of the signal converter protein CTLA4-FasL as a potent apoptosis inducer of malignant cells were investigated. Without wishing to be limited by a closed list, at least some of the findings include the following: 1. CTLA4-FasL naturally forms a stable homo-hexamer; 2. CTLA4-FasL induces robust apoptosis of malignant cell lines while relatively sparing non-malignant ones; 3. The CTLA4-FasL killing effect is more efficient when both relevant receptors (e.g. B7 and FasR) are expressed on target cells; 4. Even in non-B7 expressing cells, CTLA4-FasL exhibited significantly higher apoptotic activity than its parts, alone or in combination; 5. CTLA4-FasL efficiently inhibited the growth of human B cell lineage tumors in a xenograft model.
Bi-specific and multi-specific biological drugs are believed to develop into the “next generation” of protein-based drugs. Mostly combining functional units of two known biological targets, this drug-development field is currently lead by bi-specific antibodies, while other bi-specific technologies, such as Signal Convertor Proteins, are being assessed as well. Without being limited by a closed list, among the many advantages of bi-specific biological drugs over existing biological drugs, that comprise only one target, is a significant synergistic effect which cannot be obtained by simply administering the functional activity units alone or in combination. These synergistic effects have been mainly suggested to stem from the ability of bi-functional molecules to influence two or more biological pathways concomitantly. Notably, the efficient apoptotic activity induced by CTLA4-FasL can be seen in human B cell lymphoma cells that express both a functional Fas receptor and B7 receptors, supporting the notion that more than one biological signaling pathway is involved. Indeed, in B7 expressing cells, CTLA4-FasL provoked activation of the cascade of caspases and abrogated anti-apoptotic signals at very low concentrations, a phenomenon that could not be mimicked by CTLA4-Fc, sFasL or their combination. Importantly, this also suggests that measuring the expression of FasR, CD80 and CD86 in patient tumor samples may be used as a biomarker for patient treatment selection.
Intriguingly, CTLA4-FasL potency was higher than that of FasL, CTLA4-Fc or the combination of the latter two when incubated with non-B7 expressing cells as well, making other explanations for its robust potency plausible. The above presented data suggests higher-order structures may play a significant role in the activity and potency of these novel bi-specific drugs, as for example the homo-hexamer structures described herein.
As reported for other TNF-super family members, activation of the Fas apoptosis pathway requires trimerization of Fas receptors upon binding of FasL trimers. Moreover, it was previously shown that efficient Fas activation requires two adjacent such trimerization events. Therefore, the finding that the natural stochiometry of soluble CTLA4-FasL is a homo-hexamer is of great significance for understanding its unique, robust apoptotic capabilities. Being a hexamer, CTLA4-FasL is capable of presenting two functional trimers of FasL to their relevant receptors, resulting in optimal initiation of the apoptosis signaling pathway to the malignant cells.
The formation of a membrane bound CTLA4-FasL homo-hexamer was suggested previously. Since only homo-trimers were identified at that earlier study, the authors suggested that two CTLA4-FasL trimers may form a homo-hexamer on target cell's surface when anchored to B7 molecules, thereby inducing an extremely efficient apoptotic effect that would explain the high efficacy of CTLA4-FasL observed in that report. Here data is presented suggesting that CTLA4-FasL naturally form a soluble and stable homo-hexamer as early as it is produced and that this structure maintains its stability through a purification process that includes harsh conditions and multiple freeze/thaw cycles (
As described above, one possibility of dodecamer formation may optionally occur through a less stable “trimer of dimers” hexamer. Such a dodecamer has been shown to cause liver toxicity in mice. Therefore according to at least some embodiments, preferably the CTLA4-FasL fusion protein has less than 10% dodecamer, less than 7.5% dodecamer, less than 5% dodecamer, less than 2.5% dodecamer or less than 1% dodecamer.
Using a xenograft human-mouse disease model it was shown that CTLA4-FasL has the ability to inhibit the growth of tumors originating from B lymphocytes lineage, and to provide a significant beneficial effect on mice survival, in a dose dependent manner and at very low dosages. It was shown that this in-vivo effect is mediated by activation of the caspases cascade, as can be seen by the increased cleaved caspase 3 in immunohistochemistry of the tumors.
In summary, data is presented that the fusion protein, CTLA4-FasL induces effective apoptosis of B lymphoblastoid cells, in-vitro and in-vivo, in a highly efficient way. Also, in the case of B7 expressing cells, its potency stems from the combination of its synergistic effect of activating the cascade of caspases while abrogating the anti-apoptotic signaling, with its unique natural hexameric structure. Without wishing to be limited by a single hypothesis, it appears that that this combination of properties makes CTLA4-FasL an extremely potent apoptosis inducer of at least B7 expressing tumors, such as B cell lymphomas.
While the invention has been described with respect to a limited number of embodiments, it will be appreciated that many variations, modifications and other applications of the invention may be made, and that various combinations and subcombinations of embodiments are also possible and encompassed within the scope of this application.
This application is a continuation of U.S. patent application Ser. No. 14/655,752 filed on Jun. 26, 2015, which is a National Phase of PCT Patent Application No. PCT/IL2013/051098 having International Filing Date of Dec. 31, 2013, which claims the benefit of priority under 35 USC § 119(e) of U.S. Provisional Patent Application No. 61/748,079 filed on Jan. 1, 2013. The contents of the above applications are all incorporated by reference as if fully set forth herein in their entirety.
| Number | Date | Country | |
|---|---|---|---|
| 61748079 | Jan 2013 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 14655752 | Jun 2015 | US |
| Child | 16027382 | US |
