Patent Grant
7981417
This invention relates generally to antibodies recognizing endothelial cell surface receptors and methods of identifying such antibodies.
Treatment modalities for brain and neurological diseases are extremely limited due to the impermeability of the brain's blood vessels to most substances carried in the blood stream. The blood vessels of the brain, referred to collectively as the blood-brain barrier (BBB), are unique when compared to the blood vessels found in the periphery of the body. Tight apposition of BBB endothelial cells (EC) to neural cells like astrocytes, pericytes and neurons induces phenotypic features that contribute to the observed impermeability. Tight junctions between ECs comprising the BBB limit paracellular transport, while the lack of pinocytotic vesicles and fenestrae limit non-specific transcellular transport. These factors combine to restrict molecular flux from the blood to the brain to those molecules that are less than 500 daltons and also lipophilic. Thus, using the large mass transfer surface area (over 21 m2 from 400 miles of capillaries in human brain) of the bloodstream as a delivery vehicle is largely infeasible except in those circumstances where a drug with the desired pharmacological properties fortuitously possesses the size and lipophilicity attributes allowing it to pass freely through the blood vessel. Because of such restrictions, it has been estimated that greater than 98% of all small molecule pharmaceuticals and nearly 100% of the emerging class of protein and gene therapeutics do not cross the BBB.
In addition to the physical barrier presented by the BBB, efflux transporters such as p-glycoprotein (MDR1) and members of the multi-drug resistance-associated protein family (MRP) serve to further limit brain uptake of even those small molecules that are small and lipophilic.
Although neurological diseases such as brain cancer, Alzheimer's disease, Parkinson's disease, and stroke continue to afflict people worldwide, there has been a paucity of new therapies to treat such diseases. Lack of treatment modalities can, in part, be attributed to the lack of effective brain delivery strategies. Due to this lack of treatment methods the National Institutes of Health tumor and stroke progress review groups have even identified the search for innovative strategies for drug or gene targeting through the blood-brain barrier as a top research priority. The National Institutes of Health (NIH) guidelines instruct that such breakthroughs in basic neuroscience can be delivered to the clinic and “require an agent delivery strategy and/or the ability to target specific areas of the brain”. Thus, in the absence of appropriate vehicles for targeting and trans-BBB transport, the pipeline of new CNS medicines will likely continue to be inadequate for the people suffering from neurological disorders. As examples, protein therapeutics known as neurotrophins have been investigated recently for their protective capacity in stroke, reversal of Parkinson's disease symptoms after direct infusion into the brains of human subjects, and for their ability to direct specialized differentiation of neural stem cells for potential treatment of Parkinson's disease and other neurodegenerative diseases such as Alzheimer's disease, Huntington's disease and multiple Sclerosis. Although extremely promising therapeutics, these trophic factors do not readily cross the BBB and will require a noninvasive delivery system for widespread, effective administration.
Present brain delivery strategies are particularly invasive and require circumvention of the BBB. Strategies requiring neurosurgery are used for implantation of polymer particles infused with drug, but the treatment volume is limited because the extracellular fluid in brain tissue is quiescent, and simple molecular diffusion only allows for a modest penetration distance of 2-3 mm. A similar method is direct injection of a drug into the brain ventricles, but the penetration into brain tissue is also limited for such intraventricular injections because the cerebral spinal fluid is rapidly cleared and is turned over 4-5 times daily. In addition, protein therapeutics, such as, for example, Glial-cell Line-Derived, Neurotrophic Factor (GDNF) for Parkinson's disease have been delivered through neurosurgically implanted catheters with continual drug infusion by a peripheral pump. Disruption of the BBB has been investigated using hyperosmolar solutions and vasoactive agents like serotonin and bradykinin peptides to allow free passage of molecules from the blood to the brain. This method is primarily used clinically only in terminal patients because alterations in the BBB can lead to toxic effects due to the free access of solutes and immune factors that are normally excluded from brain.
The aforementioned invasive strategies can have success for those diseases with limited treatment volumes, such as, for localized, non-metastatic brain tumors. However, for chronic conditions requiring a repetitive treatment regimen, or for those diseases present in large portions of the brain such as Alzheimer's disease, a noninvasive drug delivery strategy would be substantially more preferable and practical.
A variety of noninvasive brain drug delivery methods have been investigated that make use of the brain blood vessel network to gain widespread drug distribution. The brain capillary network has an average spacing of just 40 microns between capillaries and is sufficiently dense that each brain cell essentially has its own vessel for nutrient supply (
Non-specific uptake mechanisms, while allowing some transport across the BBB, lack sensitivity and specificity. Cationic protein transduction domains fall into the realm of non-specific carriers, and although the HIV TAT peptide was shown to gain access to the brain interstitium after intraperitoneal injection, subsequent pharmacokinetic analysis indicated that the rapid clearance and broad organ uptake would necessitate very high doses to gain a pharmacologic effect. Another non-specific uptake mechanism is the surfactant coating of nanoparticles with polysorbate 80. Although the mechanism of brain uptake is still unresolved, the labile nature of the particles in vivo leads to short-lived pharmacologic effects and possible BBB permeabilization. Both of these non-specific methods suffer from a lack of selective targeting and result in widespread distribution of the active compound throughout the body with concomitant systemic effects.
Carrier-mediated drug transport relies on the presence of endogenous transmembrane proteins that are selective and stereospecific for small molecule solutes. For instance, L-dopa, administered to treat Parkinson's disease gains entry to the brain by utilizing the large amino acid transporter (LAT-1). The successful transport through the blood-brain barrier is a result of L-dopa mimicking the structure of phenylalanine with only the substitution of two hydroxyl groups on the aromatic ring of phenylalanine. Utilization of the saturable biotin transport system for delivery of biotinylated drugs has also been attempted. In addition, it is likely that the efflux of the AIDS drug AZT progresses in a carrier-mediated fashion. However, due to the stereospecificity and steric constraints imposed by these selective membrane pores, applications are potentially limited.
Receptor-mediated transport involves the binding of an exofacial epitope of a cell surface receptor and triggering of an energy intensive transcellular transport process known as transcytosis (
Antibodies are particularly well suited for targeting BBB receptor-mediated transcytosis systems given their high affinity and specificity for their ligands. As examples, appropriately-targeted antibodies that recognize extracellular epitopes of the insulin and transferrin receptors can act as artificial transporter substrates that are effectively transported across the BBB and deposited into the brain interstitium via the transendothelial route. Additionally, when conjugated to drugs or drug carriers of various size and composition, the BBB targeting antibodies mediate brain uptake of the therapeutic cargo. Noninvasive transport of small molecules such as methotrexate has been achieved using anti-transferrin receptor antibodies. Proteins such as nerve growth factor, brain derived neurotrophic factor, and basic fibroblast growth factor were delivered to the brain after intravenous administration by using an anti-transferrin receptor antibody. The latter two cases promoted reduction in stroke volume in rat middle cerebral artery occlusion models. Liposomes and liposomes containing genes have also been delivered to the brain in vivo using anti-transferrin receptor antibodies. In particular, gene-containing antibody-targeted liposomes have been targeted to rat brain for restoration of tyrosine hydroxylase activity in an experimental Parkinson's disease model and to primate brain using a humanized anti-insulin receptor antibody. In addition, brain delivery of the new class of RNA interference drugs via pegylated immunoliposomes has been demonstrated to increase survival of mice implanted with an experimental human brain tumor model. Further, anti-transferrin receptor conjugated nanoparticles have been produced. Finally, even if an antibody binds to the brain microvsculature or internalizes without full transcytosis, it can have drug delivery benefits. When conjugated to a liposome or nanoparticle loaded with lipophilic small molecule drugs, it might be possible to raise the local BBB concentration of drug and help circumvent brain efflux systems, thereby facilitating brain uptake. Finally, in the event binding occurs without transcytosis or even internalization, the identification of BBB-specific antibody receptors or ligands will further help to characterize and identify components of the transporter system and help to further optimize antibodies that do internalize and trancytose. Taken together, these results indicate the potential utility of antibody-targeted transcytosis systems for noninvasive trafficking of drugs into the brain.
Regardless of the promise shown for antibody mediated transport in prior studies, the current antibody targeting reagents lack specificity and transport efficiency. Although early results derived from the receptor-mediated transcytosis process are promising due to its robustness in delivery of drugs or drug carriers in many formats, it also has some serious drawbacks that need to be addressed for general clinical success. The present methods rely on receptors that are ubiquitously expressed like the transferrin and insulin receptors. This leads to mis-targeting of potentially expensive drugs that also may have unwanted side effects in tissues other than the brain. In addition, the present methodologies generally result in a low fraction (1-4%) of the injected dose actually reaching the brain target as a consequence of poor targeting and nonideal BBB permeability. This loss of between 96-99% of the administered therapeutic could hamper the development of these delivery approaches given the cost of drug manufacture, especially for protein and gene-based medicines that currently comprise nearly 700 drugs in various stages of clinical trials. Finally, the antibodies used in the aforementioned proof-of-concept experiments are either of murine origin or partially humanized, and this could lead to unwanted immunogenic reactions in human patients. Therefore, the identification of fully human antibodies that specifically recognize brain endothelial receptors would vastly improve the targeting and efficiency of drug delivery while minimizing side-effects.
This invention provides antibodies that bind endothelial cell receptors resulting in endocytosis of the receptor and bound ligands. In one exemplary embodiment, the invention comprises an isolated antibody fragment having the amino acid sequence set forth in any one of SEQ ID NOs:1-34. In another embodiment, the invention provides an isolated nucleic acid having a sequence coding for an amino acid as set forth in any one of SEQ ID NOs:1-34. In some preferred versions, the isolated antibody fragment is a single chain fragment variable (scFv) fragment. In other preferred versions, the isolated antibody is a Fab, an IgG or any other ligand specific to the endothelial cell receptor.
In another preferred embodiment, the invention is a pharmaceutical composition comprising an antibody linked to a pharmaceutically active compound that is useful in transferring the pharmaceutically active compound across the blood brain barrier (BBB).
In another exemplary embodiment, the invention is an isolated expression vector that includes a polynucleotide encoding the amino acid sequence set forth in SEQ ID NOs:1-34. In other embodiments, the invention includes a purified and isolated host cell comprising the expression vector containing the isolated nucleic acid encoding the amino acid sequence set forth in any one of SEQ ID NOs:1-34. It should be appreciated that the host cell can be any cell capable of expressing antibodies, such as, for example fungi, mammalian cells, including the Chinese hamster ovary cells; insect cells, using, for example, a baculovirus expression system; plant cells, such as, for example, corn, rice, Arabidopsis, and the like.
In yet another exemplary embodiment, the invention comprises a process for expressing an antibody fragment capable of binding to a brain endothelial cell receptor comprising: (a) displaying an antibody fragment on a yeast cell; (b) panning the antibody displaying yeast cell against a brain endothelial cell culture; (c) identifying displayed antibody fragments that specifically bind brain endothelial cell receptors; (d) inserting an isolated nucleic acid coding for the antibody fragment identified in step (c) in an expression vector; and (e) transforming a host cell with the expression vector. In some versions of this embodiment, the host cell is selected from the group consisting of: yeast, bacteria, and combinations thereof. In some preferred embodiments the host cell is Saccharomyces cerevisiae or E. coli.
In another exemplary embodiment, the invention is a method of identifying a brain endothelial cell specific antibody. This embodiment includes displaying antibody fragments on a yeast cell surface, panning the displayed antibody against a brain endothelial cell culture, isolating specific binders to the membrane receptors on the brain endothelial cell, and identifying the specific binders, thereby identifying an endothelial cell specific antibody. In some preferred embodiments the endothelial cell culture is provided in a cell monolayer.
In still another exemplary embodiment, the invention comprises a method of identifying endothelial cell receptors functioning in or related to endocytosis and transcytosis comprising, providing a cultured endothelial cell monolayer and panning a yeast-displayed antibody library against the endothelial cell monolayer, isolating antibody bound endothelial cells; and identifying the cognate receptor.
Novel human antibodies that target BBB transport systems have immense utility. Unlike the antibody delivery strategies that utilized well studied systems such as the transferrin and insulin systems, identifying novel receptor-mediated transport systems is quite difficult given that these ligands are membrane proteins that have yet to be identified. Therefore, the inventors have developed a novel selection methodology to identify fully human antibodies having the required functionality of BBB binding. The power of this technique is the concomitant selection of potential drug delivery vector and its cognate cell surface receptor (receptor-mediated carrier system) with no prior knowledge as to the origin of either component.
Another aspect of the present invention is in regard to antibody targets at the blood brain barrier, particularly, targets that represent the molecular machinery for internalization (endocytosis) and transcytosis. Accordingly, the present invention is directed to a purified scFvA antigen having a molecular weight of approximately 124 kDa when immunoprecipitated by human scFvA under non-reducing conditions. The inventors identified and demonstrated this antigen is expressed in intact brain capillaries.
Accordingly, the invention further encompasses an antibody-blood brain barrier transporter system. Such a system includes: (a) an antigen having a molecular weight of approximately 124 kDa when immunoprecipitated by human scFvA under non-reducing conditions, the antigen expressed in intact brain capillaries; and (b) a purified antibody bound to the scFvA antigen. The purified antibody has the amino acid sequence set forth in any one of SEQ ID NOs:1-5, more preferably the antibody is scFvA having the amino acid sequence set forth in SEQ ID NO:1. In certain embodiments, the purified antibody is provided in combination with a pharmaceutically active compound.
As demonstration of a similar exemplary embodiment, the inventors have identified the scFvJ antibody's (SEQ ID NO:14) target antigen as neural cell adhesion molecule (NCAM). Thus, the invention further encompasses an antibody-blood brain barrier transporter system including (a) an antigen comprising NCAM; and (b) a purified antibody bound to the antigen. In certain embodiments, the purified antibody has the amino acid sequence set forth in SEQ ID NO:14. Preferably, the purified antibody is provided in combination with a pharmaceutically active compound.
Based upon the disclosed antibody-blood brain barrier transporter system, the invention also contemplates a method of delivering a pharmaceutically active compound across the blood brain barrier to a subject's brain. Such a method includes administering a pharmaceutically active compound in combination with a purified antibody having an amino acid sequence of any one of SEQ ID NOs:1-5, SEQ ID NOs:11-12, SEQ ID NO:15, and SEQ ID NO:21 to a subject such that the antibody directs delivery of the pharmaceutically active compound across the blood brain barrier to the subject's brain.
These and other features and advantages of various exemplary embodiments of the articles and methods according to this invention are described in, or are apparent from, the following detailed description of various exemplary embodiments of the methods according to this invention.
Various exemplary embodiments of the methods of this invention will be described in detail, with reference to the following figures, wherein:
The current invention is based on the observation that cellular mechanisms may be the most efficient and effective way to transport drugs and/or therapeutics across the blood brain barrier (BBB). While a variety of cellular transporter and receptor mechanisms have been identified, they are limited by the specificity of the ligand to the endothelial receptor and by both the affinity and avidity of the ligand for the receptor. Further, identification of, as yet, unidentified antibody ligands and receptors will provide better tools by which to transport compounds across the BBB.
The inventors have mined a human single-chain antibody fragment (scFv) library for scFvs that bind to membranes of brain endothelial cells. A subset of the scFv identified as binding to brain endothelial cells also endocytose into these cells, indicating that they may be substrates for BBB transcytosis systems. The scFv also specifically labels vessels of all sizes in brain tissue sections in a pattern that represents a transport system of endothelial origin. No labeling is detected in other brain cells. This class of scFv and their associated BBB transport systems represent a novel mechanism for noninvasive drug delivery to the brain.
Before the present invention is described, it is understood that this invention is not limited to the particular methodology, protocols, and reagents described, as these may vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to limit the scope of the present invention which will be limited only by the appended claims.
It must be noted that as used herein and in the appended claims, the singular forms “a”, “an”, and “the” include plural reference unless the context clearly dictates otherwise. Thus, for example, reference to “a cell” includes a plurality of such cells and equivalents thereof known to those skilled in the art, and so forth. As well, the terms “a” (or “an”), “one or more” and “at least one” can be used interchangeably herein. It is also to be noted that the terms “comprising”, “including”, and “having” can be used interchangeably.
Unless defined otherwise, all technical and scientific terms used herein have the same meanings as commonly understood by one of ordinary skill in the art to which this invention belongs. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, the preferred methods and materials are now described. All publications mentioned herein are incorporated herein by reference for the purpose of describing and disclosing the chemicals, cell lines, vectors, animals, instruments, statistical analysis and methodologies which are reported in the publications which might be used in connection with the invention. Nothing herein is to be construed as an admission that the invention is not entitled to antedate such disclosure by virtue of prior invention.
As used herein “subject” means mammals and non-mammals. “Mammals” means any member of the class Mammalia including, but not limited to, humans, non-human primates such as chimpanzees and other apes and monkey species; farm animals such as cattle, horses, sheep, goats, and swine; domestic animals such as rabbits, dogs, and cats; laboratory animals including rodents, such as rats, mice, and guinea pigs; and the like. Examples of non-mammals include, but are not limited to, birds, and the like. The term “subject” does not denote a particular age or sex.
As used herein, “administering” or “administration” includes any means for introducing compositions of the invention into the body, preferably into the systemic circulation. Examples include but are not limited to oral; buccal, sublingual, pulmonary, transdermal, transmucosal, as well as subcutaneous, intraperitoneal, intravenous, and intramuscular injection.
As used herein, “pharmaceutical composition” means therapeutically effective amounts of the BBB transcytosis compound (also termed “pharmaceutically active compound”) together with suitable diluents, preservatives, solubilizers, emulsifiers, and adjuvants, collectively “pharmaceutically-acceptable carriers.” As used herein, the terms “effective amount” and “therapeutically effective amount” refer to the quantity of active therapeutic agent or agents sufficient to yield a desired therapeutic response without undue adverse side effects such as toxicity, irritation, or allergic response. The specific “effective amount” will, obviously, vary with such factors as the particular condition being treated, the physical condition of the subject, the type of animal being treated, the duration of the treatment, the nature of concurrent therapy (if any), and the specific formulations employed and the structure of the compounds or its derivatives. In this case, an amount would be deemed therapeutically effective if it resulted in one or more of the following: (a) the prevention of a neurological or brain disease (e.g., Alzheimers, Parkinson's and/or cancer); and (b) the reversal or stabilization of a neurological or brain disease (e.g., Alzheimers, Parkinson's and/or cancer). The optimum effective amounts can be readily determined by one of ordinary skill in the art using routine experimentation.
Pharmaceutical compositions are liquids or lyophilized or otherwise dried formulations and include diluents of various buffer content (e.g., Tris-HCl, acetate, phosphate), pH and ionic strength, additives such as albumin or gelatin to prevent absorption to surfaces, detergents (e.g., Tween 20, Tween 80, Pluronic F68, bile acid salts), solubilizing agents (e.g., glycerol, polyethylene glycerol), anti-oxidants (e.g., ascorbic acid, sodium metabisulfite), preservatives (e.g., Thimerosal, benzyl alcohol, parabens), bulking substances or tonicity modifiers (e.g., lactose, mannitol), covalent attachment of polymers such as polyethylene glycol to the protein, complexation with metal ions, or incorporation of the material into or onto particulate preparations of polymeric compounds such as polylactic acid, polyglycolic acid, hydrogels, etc, or onto liposomes, microemulsions, micelles, milamellar or multilamellar vesicles, erythrocyte ghosts, or spheroplasts. Such compositions will influence the physical state, solubility, stability, rate of in vivo release, and rate of in vivo clearance. Controlled or sustained release compositions include formulation in lipophilic depots (e.g., fatty acids, waxes, oils).
Further, as used herein “pharmaceutically acceptable carriers” are well known to those skilled in the art and include, but are not limited to, 0.01-0.1M and preferably 0.05M phosphate buffer or 0.9% saline. Additionally, such pharmaceutically acceptable carriers may be aqueous or non-aqueous solutions, suspensions, and emulsions. Examples of non-aqueous solvents are propylene glycol, polyethylene glycol, vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate. Aqueous carriers include water, alcoholic/aqueous solutions, emulsions or suspensions, including saline and buffered media.
For purposes of the present invention, “treating” or “treatment” describes the management and care of a subject for the purpose of combating the disease, condition, or disorder. The terms embrace both preventive, i.e., prophylactic, and palliative treatment as well as therapeutic treatment. Treating includes the administration of a compound of present invention to prevent the onset of the symptoms or complications, alleviating the symptoms or complications, or eliminating the disease, condition, or disorder.
In one exemplary embodiment, the invention comprises an isolated antibody fragment having the amino acid sequence set forth in any one of SEQ ID NOs:1-34. In some preferred versions, the isolated antibody fragment is a single chain fragment variable (scFv) fragment.
In another exemplary embodiment, the invention provides antibodies that bind endothelial cell receptors resulting in endocytosis of the receptor and bound ligands. In some embodiments, the invention is a pharmaceutical composition comprising an antibody linked to a pharmaceutically active compound that is useful in transferring the pharmaceutically active compound across the blood brain barrier (BBB).
In another exemplary embodiment, the invention is an expression vector that includes a polynucleotide encoding the amino acid sequence set forth in any one of SEQ ID NOs:1-34. In other embodiments, the invention includes a purified and isolated host cell comprising the expression vector containing the isolated nucleic acid encoding the amino acid sequence set forth in any one of SEQ ID NOs:1-34. It should be appreciated that the host cell can be any cell capable of expressing antibodies, such as, for example fungi, mammalian cells, including the Chinese hamster ovary cells; insect cells, using, for example, a baculovirus expression system; plant cells, such as, for example, corn, rice, Arabidopsis, and the like. See, generally, Verma, R. et al., J Immunol Methods. 1998 Jul. 1; 216(1-2):165-81.
In yet another exemplary embodiment, the invention comprises a process for expressing an antibody fragment capable of binding to a brain endothelial cell receptor comprising: (a) displaying an antibody fragment on a yeast cell; (b) panning the antibody-displaying yeast cell against a brain endothelial cell culture; (c) identifying displayed antibody fragments that specifically bind brain endothelial cell receptors; (d) inserting an isolated nucleic acid coding for the antibody fragment identified in (c) in an expression vector; and (e) transforming a host cell with the expression vector. In some versions of this embodiment, the host cell is selected from the group consisting of: yeast, bacteria, and combinations thereof. In some preferred embodiments the host cell is Saccharomyces cerevisiae or E. coli.
In another exemplary embodiment, the invention is a method of identifying a brain endothelial cell specific antibody. This embodiment includes displaying antibody fragments on a yeast cell surface, panning the displayed antibody against a brain endothelial cell culture, isolating specific binders to the membrane receptors on the brain endothelial cell, and identifying the specific binders, thereby identifying an endothelial cell specific antibody. In some preferred embodiments the endothelial cell culture is provided in a cell monolayer.
In still another exemplary embodiment, the invention comprises a method of identifying endothelial cell receptors functioning in or related to endocytosis and transcytosis comprising, providing a cultured endothelial cell monolayer and panning a yeast-displayed antibody library against the endothelial cell monolayer, isolating antibody bound endothelial cells; and identifying the cognate receptor.
Novel human antibodies that target BBB transport systems have immense utility. Unlike the antibody delivery strategies that utilized well studied systems such as the transferrin and insulin systems, identifying novel receptor-mediated transport systems is quite difficult given that these ligands are membrane proteins that have yet to be identified. Therefore, the inventors have developed a novel selection methodology to identify fully human antibodies having the required functionality of BBB binding. The power of this technique is the concomitant selection of potential drug delivery vector and its cognate cell surface receptor (receptor-mediated carrier system) with no prior knowledge as to the identity of either component.
If antibodies are used to target receptor-mediated transport systems at the BBB, then drug molecules and drug carriers can be effectively transcytosed across the BBB endothelium into brain tissue. Such noninvasive delivery from blood to brain is a result of the antibody acting as a surrogate ligand for the endogenous transport systems. Current known antibody-targeted brain delivery systems include the transferrin and insulin receptor systems. These receptors are expressed ubiquitously throughout the body and lead to mistargeting of expensive pharmaceuticals.
Depending on the neuronal disorder targeted, a variety of brain drug cargoes, e.g. pharmacologic compounds or, equivalently, pharmaceutically active compounds, can be delivered successfully in vivo using antibody-based targeting according to the invention. As used herein, the terms “pharmaceutically active compound” and “pharmacologic compound” shall refer to any compound useful in treating or ameliorating the effects of a disease or disorder. For example, diseases including neurodegenerative diseases such as Alzheimer's disease, Parkinson's disease, Huntington's disease, amyotrohpic lateral sclerosis (ALS, Lou Gehrig's disease) and multiple sclerosis can be targeted by use of such drugs as neurotrophic factors, including, but not limited to, nerve growth factor (NGF), brain derived neurotrophic factor (BDNF), ciliary neruotrophic factor (CNTF), glial cell-line neurotrohphic factor (GDNF) and insulin-like growth factor (IGF). In addition, other compounds that have been shown to have therapeutic potential and may be delivered by antibodies of the invention are neuropeptides, including, but not limited to, Substance P, neuropeptide Y, vasoactive intestinal peptide (VIP), gamma-amino-butyric acid (GABA), dopamine, cholecystokinin (CCK), endorphins, enkephalins and thyrotropin releasing hormone (TRH). Further, therapeutics may include cytokines, anxiolytic agents, anticonvulsants, polynucleotides and transgenes, including, for example, small-interfering RNAs which may be used for such neuronal disorders, including, but not limited to, psychiatric illnesses, such as, for example anxiety, depression, schizophrenia, and sleep disorders, as well as epilepsies, seizure disorders, stroke and cerebrovascular disorders, encephalitis and meningitis, memory and cognition disorders, pain therapeutics and physical trauma.
Antibodies are particularly well suited for targeting BBB receptor-mediated transcytosis systems given their high affinity and specificity for their ligands. As examples, appropriately-targeted antibodies that recognize extracellular epitopes of the insulin and transferrin receptors can act as artificial transporter substrates that are effectively transported across the BBB and deposited into the brain interstitium via the transendothelial route. Additionally, when conjugated to drugs or drug carriers of various size and composition, the BBB targeting antibodies mediate brain uptake of the therapeutic cargo. Noninvasive transport of small molecules such as methotrexate has been achieved using anti-transferrin receptor antibodies. Proteins such as nerve growth factor, brain derived neurotrophic factor, and basic fibroblast growth factor were delivered to the brain after intravenous administration by using an anti-transferrin receptor antibody. The latter two cases promoted reduction in stroke volume in rat middle cerebral artery occlusion models. Liposomes and liposomes containing genes have also been delivered to the brain in vivo using anti-transferrin receptor antibodies. In particular, gene-containing antibody-targeted liposomes have been targeted to rat brain for restoration of tyrosine hydroxylase activity in an experimental Parkinson's disease model and to primate brain using a humanized anti-insulin receptor antibody. In addition, brain delivery of the new class of RNA interference drugs via pegylated immunoliposomes has been demonstrated to increase survival of mice implanted with an experimental human brain tumor model. In addition, anti-transferrin receptor conjugated nanoparticles have been produced. Finally, even if an antibody binds to the brain microvasculature or internalizes without full transcytosis, it can have drug delivery benefits. When conjugated to a liposome or nanoparticle loaded with lipophilic small molecule drugs, it might be possible to raise the local BBB concentration of drug and help circumvent brain efflux systems, therefore facilitating brain uptake. Further, the ability to bind without internalization or with internalization but without full trancytosis will provide for the identification of BBB endothelial receptors or ligands that, once known, will allow for characterization and identification of the transporter system and for optimization of antibodies that are internalized and/or trancytosed. Taken together, these results indicate the utility of antibody-targeted transcytosis systems for noninvasive trafficking of drugs into the brain.
As can be appreciated, one aspect of the present invention is in regard to antibody targets at the blood brain barrier, particularly, targets that represent the molecular machinery for internalization (endocytosis) and transcytosis. Accordingly, the present invention is directed to a purified antigen having a molecular weight of approximately 124 kDa when immunoprecipitated by human scFvA under non-reducing conditions. The inventors identified and demonstrated this antigen is expressed in intact brain capillaries. The inventors have further demonstrated that scFvB, scFvC, scFvG, and scFvK each recognize and are capable of immunoprecipitating the approximately 124 kDa antigen immunoprecipitated by scFvA. Accordingly, scFvA, scFvB, scFvC, scFvG, and scFvK (SEQ ID NOs:1-5, respectively) are each useful as inventive blood brain targeting antibodies.
Accordingly, the invention further encompasses an antibody-blood brain barrier transporter system. Such a system includes: (a) an antigen having a molecular weight of approximately 124 kDa when immunoprecipitated by human scFvA under non-reducing conditions, the antigen expressed in intact brain capillaries; and (b) a purified antibody bound to the antigen wherein said antibody has the amino acid sequence set forth in any one of SEQ ID NOs:1-5. The purified antibody is preferably an scFvA antibody (SEQ ID NO:1), and, even more preferably, the scFvA antibody is provided in combination with a pharmaceutically active compound.
Furthermore, the inventors have identified neural cell adhesion molecule (NCAM) as an antigen that is immunoprecipitated by scFvJ antibody (SEQ ID NO:14). Accordingly, the invention further encompasses an antibody-blood brain barrier transporter system including: (a) an antigen comprising neural cell adhesion molecule (NCAM); and (b) a purified antibody bound to the antigen. In such a transporter system, the purified antibody binds to the NCAM antigen in the cell surface and is transported through the blood-brain barrier. In certain embodiments, the antibody included in the transporter system has the amino acid sequence set forth in SEQ ID NO:14. Preferably, the purified antibody is provided in combination with a pharmaceutically active compound, and the transporter system acts to facilitate the transfer of the pharmaceutically active compound across the blood-brain barrier.
Based upon the disclosed antibody-blood brain barrier transporter system, the invention also contemplates a method of delivering a pharmaceutically active compound across the blood brain barrier to a subject's brain. Such a method includes administering a pharmaceutically active compound in combination with a purified antibody having an amino acid sequence set forth in any one of SEQ ID NOs:1-5, SEQ ID NOs:11-12, SEQ ID NO:15, and SEQ ID NO:21 to a subject such that the antibody directs delivery of the pharmaceutically active compound across the blood brain barrier to the subject's brain. The purified antibody is preferably an scFv antibody having the amino acid sequence set forth in any one of SEQ ID NOs:1-5, SEQ ID NO:11, SEQ ID NO:15, and SEQ ID NO:21, and the purified antibody is more preferably an scFvA antibody having the amino acid sequence set forth in SEQ ID NO:1.
In general, methods of conjugating, linking and coupling antibodies to pharmacologically active compounds are well known in the field. For example, see, Wu A M, Senter P D, Arming antibodies: prospects and challenges for immunoconjugates, Nat. Biotechnol. 2005 September; 23(9):1137-46 and Trail P A, King H D, Dubowchik G M, Monoclonal antibody drug immunoconjugates for targeted treatment of cancer, Cancer Immunol Immunother. 2003 May; 52(5):328-37; Saito G, Swanson J A, Lee K D. Drug delivery strategy utilizing conjugation via reversible disulfide linkages: role and site of cellular reducing activities, Adv Drug Deliv Rev. 2003 Feb. 10; 55(2):199-215. As well, the present antibodies may be provided in combination with liposome, nanoparticles or other analogous carriers loaded with a pharmaceutically active compound. Methods of preparing such compositions are known in the field (see, for example, Sugano et al., Antibody Targeting of Doxorubicin-loaded Liposomes Suppresses the Growth and Metastatic Spread of Established Human Lung Tumor Xenografts in Severe Combined Immunodeficient Mice Cancer Research 60, 6942-6949, Dec. 15, 2000 and Martin et al., Nanomaterials in Analytical Chemistry, Analytical Chemistry News & Features, May 1, 1998; pp. 322 A-327 A). As used herein, the phrase “antibody in combination with a pharmaceutically active compound” shall not be limited by the method of manufacture and such compositions may be produced by, but not limited to, techniques of conjugating, linking, coupling and decorating known in the art.
The examples described below disclose antibody delivery vectors that are a critical component for disease-based applications. Identification of such vectors combines an innovative combinatorial antibody screening technology with expertise in the BBB and brain drug delivery fields to attack a complex delivery problem.
Various exemplary embodiments of compounds obtained as generally described above and methods according to this invention will be understood more readily by reference to the following examples, which are provided by way of illustration and are not intended to be limiting of the invention in any fashion.
In order to simultaneously identify both brain endothelial cell surface proteins (potential BBB-resident transport systems) and antibody targeting reagents (drug delivery vector), the inventors have performed validation of a novel cell surface profiling or “panning” technique. This technique uses a combinatorial library of naïve single-chain antibodies (scFv,
The effectiveness of the panning method was first investigated using a model scFv-surface ligand system (Wang, X. X. and Shusta, E. V. The use of scFv-displaying yeast in mammalian cell surface selections. J Immunol Methods 2005, 304, 30-42) (incorporated herein in its entirety for all purposes). Briefly, the hapten fluorescein was used as the surface ligand and an anti-fluorescein scFv (4-4-20) was used as the yeast-displayed antibody. This system allowed detailed investigation of the factors governing panning success. Yeast displaying 4-4-20 exhibited specific interactions with fluoresceinated rat brain endothelial cells (RBE4 cell line Roux, F., et al., Regulation of gamma-glutamyl transpeptidase and alkaline phosphatase activities in immortalized rat brain microvessel endothelial cells. J Cell Physiol 1994, 159, 101-13.) (
The effects of scFv affinity on conjugate formation were examined since the naïve yeast library contains antibodies with affinities in the nanomolar range. The extent of enrichment was analyzed for dilute mixtures containing both 4-42-0 (Kd=1.3 nM) and a high affinity mutant 4M5.3 (Kd=270 fM) [55] at varying cell surface FITC intensities. Briefly, 4-4-20:4 M5.3:nonbinder mixtures of 1:1:105 were applied to NAFITC-labeled monolayers having varying densities of fluorescein ligand. After a two hour incubation at 4° C., the wells were washed to remove non-specific binders. The simultaneous enrichment of 4-4-20 and 4M5.3 was analyzed over three consecutive enrichment rounds (
Recognizing the limitations of phage-display, the strategy illustrated in
The inventors possess a human nonimmune scFv library in the identical yeast display format as that used for the studies described in Example 1. Methods of making such a library are known in the art and have been described by, e.g., Feldhaus et al. at Nat. Biotechnol. 21(2):163-70 (2003), incorporated by reference herein. The library is comprised of ˜1×109 scFvs that represent much of the in vivo human antibody heavy and light chain gene diversity. Since the library is naïve, it has not been subjected to any prescreening against antigen and thus can provide antibodies that bind a widely diverse range of antigens. The library has been successfully used to isolate panels of scFvs that bind with high affinity (Kd=1-1000 nM) to hapten antigens (fluorescein), peptide antigens having different phosphorylation states (p53), protein antigens (lysozyme), and the extracellular portion of cell surface receptors (epidermal growth factor receptor). The yeast-EC targeting method described in Example I was used for the profiling of RBE4 rat brain endothelial cells using the scFv library. Briefly, the yeast library was applied to 100 cm2 of RBE4 cells, and the binding yeast clones recovered and amplified (
The inventors developed the biopanning antibody library screening method based on the model system described in EXAMPLE 2. This model was then applied to the YSD scFv library to identify BBB-binding and internalizing antibodies. Preparation of the library is described by Wang et al. at Nat. Methods, 4(2):143-5 (2007) which, along with the references cited therein, is incorporated by reference into the present disclosure.
While the capability of the inventors' panning system was illustrated by the results shown in
The germline family usages of these clones are summarized in
A total of approximately 2000 clones from the fourth round of panning have been analyzed using this method. Of the 34 unique scFvs identified thus far, only a few clones were recovered multiple times, and based on statistical considerations, many unique clones remain in the enriched round 4 pool. The germline origins of the heavy and light variable regions for the recovered clones are indicated in
While the test for efficacy of the yeast library began with an identification of the binding clones, their use in transcytosis further requires that the scFvs could be internalized. Therefore, the scFv's were also examined to see if they could be internalized by RBE4 cells. Because yeast are too big to be internalized the scFv was produced as a soluble protein in yeast cell culture. To do this, scFv gene in the YSD plasmid was subcloned into a secretion plasmid (described below in EXAMPLE 10) and the scFv was then secreted into the yeast cell culture medium. Briefly, the scFv open reading frames for several of the scFvs in
Several of the class 1, 2, 3, 4, 5, 7, 10, 12, 15, 16, and 17 scFv were evaluated for their ability to mediate cellular internalization. The scFv were predimerized with the anti-cmyc epitope tag antibody to provide the bivalency often required for receptor clustering and endocytosis. The dimerized scFv were then applied to living RBE4 cells at 4° C. to yield cell surface labeling. Subsequently, the RBE4 cells were shifted to 37° C. for 30 minutes to enable cellular trafficking (
More specifically, RBE4 cells were labeled with pre-dimerized scFv A, scFv D, scFv 4-4-20 (dimerized to promote receptor clustering often required for endocytosis) or OX26 monoclonal antibody at 4° C. for cell surface labeling and then shifted to 37° C. to promote cellular trafficking. The cells were then labeled with AlexaFluor 555 conjugated anti-mouse IgG at 4° C. followed by AlexaFluor 488 conjugated anti-mouse IgG with or without cell permeabilization by saponin (SAP) treatment. Merged images of the AlexaFluor-labeled images are shown in
The inventors determined that class 1 scFv exhibited a clear binding signal with either supernatant or purified material at 3-4 μg/mL, but class 2 (scFvD) and 3 scFv (scFvF) required approximately 10-fold higher purified concentrations of 20-80 μg/mL to yield cell surface immunolabeling. In terms of binding affinity to live RBE4 cells, class 1 scFvA possessed an affinity of Kd=82±15 nM, whereas the affinity of class 2 scFvD as a monomeric protein could not be determined using this method. Instead, the inventors evaluated the binding properties of scFvD after predimerization with an epitope tag antibody (avidity=2.0±0.1 nM). As a comparison, the affinity of an anti-transferrin receptor scFv isolated using phagemid panning was measured to be 135 nM, of similar affinity to scFvA isolated with our yeast panning system.
Thus, the results disclosed herein clearly indicate that yeast culture supernatants can be used directly for facile biochemical testing similar to approaches that use hybridoma-conditioned medium as a source for MAb. Further, the scFv can be purified for labeling and transport studies using the c-terminal six histidine epitope as described in EXAMPLE 11.
To assess the nature of the antigens recognized by the scFv, the inventors developed a novel yeast immunoprecipitation procedure. They used yeast displaying scFv to directly immunoprecipitate the cognate plasma membrane antigens from detergent-solubilized, biotinylated RBE4 lysates. Conveniently, the use of yeast as the immunoprecipitation particle allowed the sizing of antigens without any additional subcloning, production or immobilization of scFv proteins required by traditional immunoprecipitation methods. The inventors assessed the immunoprecipitated products by anti-biotin Western blotting (
The high-throughput method described so far allowed the identification of scFv clones that bind specifically to RBE4 cells and may be endocytosed in culture. However, the inventors wanted to confirm that clones identified in culture also bound brain endothelial cells in vivo. Therefore, it was necessary to first confirm that the antigen targeted by scFvA in culture was also present on the endothelial cells in vivo.
The right panel of
The following provides a detailed description of the materials and methods utilized in the foregoing example sections.
Growth and Induction of scFv Library.
The nonimmune human scFv library in EBY100 yeast8 (a, GAL1-AGA1::URA3 ura3-52 trp1 leu2Δ1 his3Δ200 pep4::HIS2 prb1Δ1.6R can1 GAL) was grown at 30° C. in SD-CAA (20.0 g/L dextrose, 6.7 g/L yeast nitrogen base, 5.0 g/L casamino acids, 10.19 g/L Na2HPO4·7H2O, 8.56 g/L NaH2PO4·H2O) plus 50 μg/mL kanamycin for 24 hours (OD600˜10). Yeast at 10-fold excess of the library diversity (5×109) were subsequently induced in 500 mL SG-CAA medium (same as SD-CAA except dextrose replaced by galactose) at 20° C. for 22 hours prior to panning against RBE4 monolayers.
Panning of scFv Library Against RBE4 Cell Monolayers.
The RBE4 rat brain endothelial cell line was used as the brain endothelial cell source as RBE4 cells have previously been demonstrated to display many attributes characteristic of the BBB in vivo. RBE4 cells exhibit a nontransformed phenotype, express typical endothelial markers, respond to astrocyte cues, and exhibit BBB-specific properties such as the expression and correct localization of the tight junction protein occludin. In addition, plasma membrane-localized transporters characteristic to brain endothelial cells including those that transport glucose (GLUT1), large neutral amino acids (LAT1), and iron (transferrin receptor), and those that function in active efflux at the BBB (p-glycoprotein, MDR1), are expressed by RBE4 cells. RBE4 cells were a kind gift from Dr. Françoise Roux and were maintained as described previously. RBE4 cells were seeded on collagen type I-coated (Sigma) 6-well plates at 25% confluency two days prior to panning. Induced yeast cells at 10-fold excess of the library size (5×109 yeast) were washed twice with 0.01 M PBS, pH 7.4, supplemented with 1 mM CaCl2, 0.5 mM Mg2SO4 and 0.1% bovine serum albumin (BSA) (Wash buffer) and the yeast mixture was added dropwise onto 100 cm2 of RBE4 cell monolayer to ensure even distribution across the monolayer. The density of yeast (5×107 yeast/cm2) is at the upper limit for panning in that the yeast completely coat the RBE4 monolayer. Panning at high density allows ˜30% recovery of binding yeast while also providing appropriate oversampling of the library diversity. The monolayers were then incubated at 4° C. for 2 hours to allow yeast-RBE4 cell contacting. The washing strategy was optimized to recover a model scFv that binds to RBE4 cells with nanomolar affinity. The resulting method involved washing the RBE4 layers with ice cold wash buffer by gently rocking the plate twenty-five times, rotating the plate five times (repeated twice), and rotating the plate ten times. The washing supernatant was removed after each step and replaced with fresh wash buffer. After the washing steps, 1 mL of wash buffer was added into each well and all cells were scraped off the plate and pooled together. The yeast/RBE4 cell mixture was resuspended in 5 mL kanamycin-supplemented SD-CAA and grown at 30° C. overnight, followed by SG-CAA induction for 20 hours at 20° C. In parallel, a small fraction of the recovered cells were plated on SD-CAA agar plate to quantify the total number of recovered yeast cells after each round. Since the pool diversity was greatly reduced after round 1, the yeast panning density was lowered to 5×106 yeast/cm2, and the RBE4 area was reduced to 20 cm2 for round 2 and 10 cm2 for rounds 3-5. After round 2 of panning, the recovered yeast clones numbered 8.2×104, and a parallel experiment with control yeast displaying an anti-fluorescein scFv (4-4-20) showed very little background using the same washing regimen indicating that the panning strategy was yielding primarily RBE4-binding yeast clones. To confirm that the yeast-RBE4 interactions were scFv-based, the scFv-encoding plasmids for several RBE4-binding yeast clones (7 from round 3, 12 from round 4) were recovered using the Zymoprep yeast miniprep kit (Zymo Research). The scFv-encoding plasmid was then retransformed into yeast surface display parent strain, EBY100, using the lithium acetate method and Trp+ transformants were selected. After RBE4 binding with the retransformed clones was confirmed, the plasmids were sequenced with the Gal1-10 (5′-CAACAAAAAATTGTTAATATACCT-3′; SEQ ID NO:35) and alpha terminator primers (5′-GTTACATCTACACTGTTGTTAT-3′; SEQ ID NO: 36) (UW-Madison Biotechnology Center).
High-throughput Analysis of Recovered Yeast Clones.
As described above, yeast are typically grown first in SD-CAA followed by SG-CAA to promote scFv expression. However, this technique yielded comparatively low levels of scFv surface expression level and lowered percentages of yeast displaying scFv when a 96-well format was used. Therefore, the scFv display methodology was optimized for 96-well plates, and it was found that simultaneous growth and induction in SG-CAA allowed for scFv display having similar efficiency to that observed using the traditional yeast display methods. Thus, for high throughput screening, yeast clones were inoculated into 200 μL of SG-CAA (induced sample) and SD-CAA (control uninduced sample) in a 96-well plate and incubated at 30° C. for 24 hours. After removing 160 μl, of SD culture to ensure similar total yeast numbers as the parallel SG culture, the 96-well plate of yeast was centrifuged, and the supernatant was carefully removed. The yeast were then washed once with 150 μL wash buffer and resuspended in 150 μL wash buffer. In parallel, RBE4 cells cultured to confluency in a 96-well plate were washed once with ice-cold wash buffer. The yeast clones were then transferred into corresponding wells containing RBE4 monolayers and incubated at 4° C. for 2 hours. After washing, light microscopy was used to assess the binding capacity of the scFv yeast clones. After visual inspection, a yeast clone was defined as RBE4-binding if induced yeast remained bound while uninduced yeast originating from the same clone were washed away.
The scFv genes harbored by binding yeast clones were directly amplified by whole yeast cell PCR. Briefly, a small amount of a fresh, uninduced yeast colony was transferred into 30 μL 0.2% SDS, vortexed, frozen at −80° C. for 2 minutes and incubated at 95° C. for 2 minutes (temperature shift repeated once). One microliter of the cell lysis solution was then used as a PCR reaction template with primers, PNL6 Forward (5′-GTACGAGCTAAAAGTACAGTG-3′; SEQ ID NO: 37) and PNL6 Reverse (5′-TAGATACCCATACGACGTTC-3′; SEQ ID NO: 38). Subsequently, 20 μL of PCR product was subjected to BstNI (New England Biolabs) restriction digest at 60° C. for 14 hours. The digested products were resolved on a 3% agarose gel for unique scFv clone identification. The PCR product of each clone displaying a unique BstNI digestion pattern was sequenced with Rev Seq P2 (5′-CCGCCGAGCTATTACAAGTC-3′; SEQ ID NO: 39) and For Seq P2 (5′-TCTGCAGGCTAGTGGTGGTG-3′; SEQ ID NO: 40) primers. The sequence was then analyzed by the IgBLAST program to identify the human germline origin (IgBLAST available at NCBI website: www.ncbi.nlm.nih.gov).
Yeast Colony Northern Blotting.
Yeast colony Northern blotting was used to detect and presubtract class 1 and class 2 scFv from the yeast binding pool. Reagents and instruments were prepared as in standard Northern blotting experiments to eliminate RNase contamination. Yeast clones were cultured on SD-CAA agar plates and the resulting colonies were transferred onto ethanol-sterilized nitrocellulose membranes. The colony-loaded membrane was then layered on top of SG-CAA agar plates, cell side facing up, and incubated at 30° C. for 2 days to induce transcription of the scFv gene. To prepare the induced yeast colonies for Northern blotting, the nitrocellulose membranes were layered onto Whatman filter paper soaked with 10% SDS and incubated at 65° C. for 30 minutes. The filters were then fixed by transferring to formaldehyde-soaked filter paper at 65° C. for 30 minutes (3×SSC, 10% formaldehyde in ddH2O). Air-dried membranes were subsequently baked for 2 hours at 80° C. under vacuum. Oligonucleotide probes corresponding to Class 1 VHCDR2 and Class 2 VHCDR2 were radiolabeled with a 10 residue 32P-dATP tail using the STARFIRE kit according to manufacturer's instructions (IDT), and their specific radioactivity was determined by scintillation counting. Being part of the germline V-region, the VHCDR2 regions exhibited 100% homology within class 1 and class 2, and were therefore amenable to hybridization-based subtraction. The membranes were blocked in prehybridization buffer (50% formamide, 5×Denhardt's solution, 5×SSPE, 1% SDS, 0.1% salmon sperm DNA) at 43° C. for 2 hours, and then hybridized (prehybridization buffer with 8×105 cpm/ml of each probe) at 43° C. overnight. After hybridization, the nitrocellulose membranes were washed as follows: 2×SSC, 0.1% SDS at room temperature for 8 minutes, 0.5×SSC, 0.1% SDS at room temperature for 8 minutes, 0.1×SSC, 0.1% SDS at room temperature for 8 minutes, 0.1×SSC, 1% SDS at 50° C. for 30 minutes. The membranes were then exposed to ECL Hyperfilm (Amersham) at −80° C. for 24 or 72 hours. Although VHCDR2 was used as the probe in these subtractive screens, the diversity of the recovered scFv clones can be readily expanded as desired via subtraction using any combination of CDR probes.
scFv Secretion and Purification.
Open reading frames for scFv were isolated from the PCR products used for BstNI typing by NheI-HindIII restriction digest and were shuttled to an scFv yeast secretion vector (pRS316-GAL4-4-20) that has been used extensively for scFv secretion. The resultant pRS316-GALscFv plasmids were then transformed into YVH10, a yeast strain overexpressing protein disulfide isomerase. Yeast harboring scFv secretion vector were grown in minimal SD medium (2% dextrose, 0.67% yeast nitrogen base) supplemented with 2×SCAA amino acid (190 mg/L Arg, 108 mg/L Met, 52 mg/L Tyr, 290 mg/L Ile, 440 mg/L Lys, 200 mg/L Phe, 1260 mg/L Glu, 400 mg/L Asp, 480 mg/L Val, 220 mg/L Thr, 130 mg/L Gly, 20 mg/L tryptophan lacking leucine and uracil) at 30° C. for 72 hours. Subsequently, scFv secretion was induced at 20° C. for 72 hours in SG-SCAA (dextrose substituted by galactose) with 1 mg/ml BSA as a nonspecific carrier. For experiments requiring purified scFv, Ni-NTA columns (Qiagen) were used to purify the six histidine-tagged scFv from 50 mL or 1 L batches as described previously in the literature.
The size, purity, and secretion yields of scFv were analyzed by SDS-polyacrylamide gel electrophoresis (PAGE) with a 4% stacking and 12.5% separating gel followed by Coomassie blue staining. Protein concentrations were estimated by comparison to a series of carbonic anhydrase standards (31 kDa) and by BCA protein assay (Pierce). In parallel, the SDS-PAGE resolved proteins were also blotted onto a nitrocellulose membrane (BioRad) for Western blotting. The nitrocellulose membrane was blocked at 4° C. overnight in TBST solution (8 g/L NaCl, and 0.1% Tween-20, buffered to pH 7.6 with 20 mM Tris) supplemented with 5% nonfat milk and probed with 1 μg/mL 9E10 anti c-myc antibody (Covance) followed by an anti-mouse IgG horse radish peroxidase conjugate (Sigma). Detection was performed using enhanced chemiluminescence and multiple time point exposures to ECL hyperfilm (Amersham) were evaluated by NIH ImageJ software for quantification.
Affinity Determination.
RBE4 cells were labeled at various scFvA concentrations at 4° C., and the bound scFv detected by anti-cmyc (9E10) antibody labeling followed by anti-mouse IgG AlexaFluor 555. Fluorescence intensity was monitored by FACSCalibur flow cytometer and used to quantitate fractional bound ligand. The scFvA binding data was fit to an equilibrium binding model to determine the monovalent affinity dissociation constant (Kd). The cell-labeling assay was not sensitive enough to produce a binding curve using monomeric scFvD, so scFvD was predimerized with anti-epitope tag antibody, 9E10, to provide the requisite avidity for the ligand binding measurements. ScFv D dimer-labeled cells were then probed with anti-mouse IgG AlexaFluor 555 conjugate and assessed by flow cytometry. The resulting data were fit to an equilibrium binding model to derive an apparent affinity (avidity). For antigen density experiments, living RBE4 cells were labeled at 4° C. with 62.5 nM of pre-dimerized scFvA, scFvD, 4-4-20 scFv, OX26 monoclonal antibody, or IgG2a isotype control. A uniform secondary antibody (anti-mouse IgG AlexaFluor 555 conjugate) was used for each sample to facilitate quantitative comparisons of labeling intensity. These antibody labeling concentrations were adequate for saturation binding of the cell surface antigens, and the cell surface labeling was quantitatively assessed by flow cytometry.
ScFv-RBE4 Immunocytochemistry.
Predimerization of scFv via the c-myc epitope tag by the 9E10 antibody was used as a method to provide bivalency which is often an important component for promoting cellular internalization of scFv. To this end, the RBE4-binding scFv was first incubated with 9E10 to form artificial dimers. Equal volumes of purified scFv (diluted to 8 μg/mL for scFvA or 32 μg/mL for scFvD and 4-4-20 using 40% goat serum in PBS supplemented with 1 mM CaCl2, 0.5 mM Mg2SO4) and 10 μg/mL 9E10 were mixed and incubated at room temperature for 1 hour to form artificial dimer RBE4 cells at about 90% confluency were washed 3× with wash buffer. RBE4 cells were then incubated with scFv artificial dimer or OX26 monoclonal antibody (10 μg/mL) (Serotec) at 4° C. for 30 minutes and then switched to 37° for another 30 minutes. An anti-mouse IgG secondary antibody conjugated with AlexaFlour555 (Molecular Probes) was applied for 30 minutes at 4° C. to label cell surface-bound scFv. The cells were then permeabilized with 0.5% saponin (SigmaAldrich) diluted in wash buffer at 4° C. for 5 minutes, and subsequently labeled with an anti-mouse IgG antibody conjugated with AlexaFluor488 (Molecular Probes) for 30 minutes at 4° C. to detect internalized scFv. Labeled cells were then fixed with 4% paraformaldehyde and examined using a fluorescence microscope (Olympus IX70).
Yeast Display Immunoprecipitation.
ScFv-displaying yeast cells selected from the human scFv library were directly used to immunoprecipitate the cognate plasma membrane antigens. Yeast cells displaying anti-hen egg lysozyme (D1.3) scFv were used as a negative control. As a positive control, an anti-transferrin receptor OX26 scFv yeast display plasmid was created by excising OX26 scFv open reading frame from pRS316-GALOX26 as an NheI-XhoI fragment and ligating into pCT-LWHI. Yeast clones were grown and induced in 50 mL cultures as described above. Induced yeast were collected by centrifugation, washed and fixed with 3% vol/vol formalin in PBS. RBE4 plasma membrane proteins were biotinylated using 0.5 mg/mL Sulfo-NHS-LC-Biotin (Pierce). To prepare RBE4 cell lysate, approximately 5×106 biotinylated RBE4 cells were lysed using a 1% (w/v) n-octyl-beta-D-glucopyranoside (scFvA, B, C and J Sigma) or 0.1% (w/v) Triton X-100 (scFvD, I, J, and OX26, Sigma) detergent solution in PBS, supplemented with a protease inhibitor cocktail (Calbiochem). For immunoprecipitation, 400 μg of cell lysate protein was mixed with approximately 108 yeast cells and incubated overnight at 4° C. Elution of immunoprecipitated product was performed by resuspending yeast cells in 30 μL of 0.5% SDS in 0.4 M Tris (pH 6.8) for 15 minutes. The eluates were separated with SDS-PAGE (8% separating gel) with or without reducing agent (DTT) present, and blotted onto a nitrocellulose membrane (BioRad). Western blotting was subsequently performed with an anti-biotin monoclonal antibody (0.5 μg/mL, clone BTN.4, Labvision), OX26 monoclonal antibody (5 μg/mL, Serotec), or anti-insulin receptor □-subunit monoclonal antibody (1 μg/mL, clone CT-3, Labvision) as described above. Neither scFvA nor scFvD were active in Western blotting format with immunoprecipitated products or with cell lysates, likely a result of selections being performed under native conditions with living cells.
Immunohistochemical Labeling of Rat Brain Sections by scFv A.
Brain tissue sections were prepared from the brain of an adult male Sprague Dawley rat. The brain was snap-frozen with tissue freezing medium (Triangle Biomedical Sciences) using a liquid nitrogen bath, and 7 μm coronal sections were cut from the frozen brain. The brain sections were blocked with 40% goat serum and 0.2% TritonX-100 in PBSCM at room temperature for 30 minutes. Purified scFv A or 4-4-20 was diluted with 40% goat serum and incubated with an equal volume of 10 μg/mL 9E10 for 1 hour at room temperature to form artificial dimer. Brain sections were then incubated with scFv A artificial dimer at 4° C. for 1 hour. A secondary labeling solution consisting of phycoerythrin-conjugated anti-mouse IgG and FITC conjugated Griffonia simplicifolia lectin (GSA-FITC 10 μg/ml, Sigma) was applied for 30 minutes at 4° C. After washing, the brain sections were immediately fixed with 4% paraformaldehyde for 10 minutes on ice and examined by fluorescence microscopy.
In the above Examples, yeast library screening methods were used to identify antibodies against cell surface antigens. Specifically, the inventors have demonstrated that yeast cells displaying scFvs can be used for immunoprecipitation and characterization by Western blotting of target cell (RBE04) surface antigens (see Examples 10 and 12). In this Example, the inventors show the full capability of the yeast display immunoprecipitation technique (YDIP) for recovery and analysis of both soluble and plasma membrane antigens. Specifically, the inventors report the use of YDIP with tandem mass spectrometry to identify the RBE4 plasma membrane antigen immunoprecipitated by scFvJ as the neural cell adhesion molecule (NCAM).
Materials and Methods:
Cells, media and plasmids. Saccharomyces cerevisiae strain EBY100 was used for surface display of scFvs. Surface display plasmids pCT201-D1.3 and pCT302 were used for the display of anti-hen egg lysozyme D1.3 scFv and anti-fluorescein 4420 scFv, respectively. All human scFvs (scFvA, scFvD, scFvJ, and scFvK) in the surface display format were selected from the previous examples. Yeast cells were grown in SD-CAA medium (20.0 g/L dextrose, 6.7 g/L yeast nitrogen base, 5.0 g/L casamino acids, 10.19 g/L Na2HPO4.7H2O, 8.56 g/L NaH2PO4.H2O) at 30° C. to reach an OD600 nm of approximately 1.0 and induced in the same volume of SG-CAA medium (dextrose replaced by galactose in SD-CAA) for 16-18 h at 20° C. for scFv display. The rat brain endothelial cell line (RBE4) was a kind gift from Dr. Françoise Roux. RBE4 cells were grown at 37° C. in 5% CO2, in 45% Alpha Minimum Essential Medium, 45% Ham's F10 medium, and 10% heat inactivated fetal bovine serum (FBS, Invitrogen, Carlsbad, Calif.) supplemented with 100 μg/mL streptomycin, 100 units/mL penicillin G (Invitrogen, Carlsbad, Calif.), 0.3 mg/mL geneticin (Fisher Scientific, Pittsburgh, Pa.) and 1 μg/L basic FibroblastGrowth Factor (bFGF) (Roche Diagnostics, Indianapolis, Ind.).
Preparation of biotinylated RBE4 cell lysates. Plasma membrane proteins of RBE4 cells were biotinylated by incubating live RBE4 cells with 0.5 mg/mL sulfo-NHS-LCBiotin (Pierce, Rockford, Ill.) in 10 mM phosphate buffered saline (PBS, pH 7.4) supplemented with 1 mMCaCl2, 0.5 mM Mg2SO4 (PBSCM) for 30 min with rocking at room temperature. The charged sulfoxide group of the biotinylating reagent prevents the biotinylation of cytosolic proteins by hindering the diffusion through cell membranes. After the biotinylation, approximately 5×106 biotinylated RBE4 cells were lysed at 4° C. by scraping the cells into 1 mL of PBS supplemented with a protease inhibitor cocktail (Calbiochem, Gibbstown, N.J.), 2 mM EDTA and containing one of the following detergents: 0.1% (w/v) Triton X-100 (TX), 1% (w/v) n-octyl-β-D-glucopyranoside (OG) (Anatrace, Maumee, Ohio), 0.5% (w/v) CHAPS (Fisher Scientific), or radioimmunoprecipitation assay (RIPA) buffer (0.1% (w/v) SDS, 0.5% (w/v) sodiumdeoxycholate (DOC), and 1% (w/v) Triton X-100 in PBS). The initial cell lysates were then centrifuged at 18,000×g for 15 min at 4° C. to remove insoluble debris. The total protein concentration remaining in the supernatant was determined using the BCA assay per manufacturer's instructions (Pierce, Rockford, Ill.).
Detection of scFv activity in detergent solutions. To assess the effects of detergents on scFv affinity, an anti-hen egg lysozyme antibody D1.3 scFv and hen-egg lysozyme (HEL) from chicken egg white (Sigma, St. Louis, Mo.) were used. Yeast cells were fixed with 3% v/v formaldehyde in PBS for 1 h at room temperature when indicated. The HEL was biotinylated using sulfo-NHS-LC-biotin (Pierce, Rockford, Ill.) and the degree of biotinylation was determined to be 2.1 biotins per HEL using the 4-Hydroxyazobenzene-2-carboxylic acid (HABA) assay (Sigma, St. Louis, Mo.). Serial dilutions of HEL were prepared in the various lysate buffers described above minus protease inhibitors and incubated with yeast cells displaying anti-HEL D1.3 scFv for 2 h at 20° C. The incubated yeast cells were washed once with the corresponding detergent solution, once with PBS containing 0.1% (w/v) bovine serum albumin (PBS-BSA) and subsequently incubated with a mouse anti-c-myc antibody (9E10, 30 μg/mL, Covance, Berkeley, Calif.) to selectively analyze the antibody-displaying yeast population. Next, goat antimouse IgG-Alexa488 conjugate (αM488, 1/1000 dilution, Invitrogen, Carlsbad, Calif.) and streptavidin-phycoerythrin conjugate (SAPE, 1/80 dilution, Sigma) were added to quantify ligand binding. All of the washing and labeling steps were performed at 4° C. The fluorescence intensities were quantified using the FACSCalibur flow cytometer (Becton Dickinson, Franklin Lakes, N.J.) and fitted to a bimolecular equilibrium binding model to determine the dissociation constants (Kd). To test the binding activity of scFvA, scFvK, scFvD, and scFvJ, yeast displaying scFv were incubated with biotinylated RBE4 cell lysate (1% OG, 200 μg total protein) for 2 h at 4° C. Yeast were then washed twice with PBS containing 1% OG. Labeling with 9E10, SAPE, and αM488 was identical to that described above, except for scFvJ, for which mouse anti-HA antibody (12CA5, Roche Diagnostics, Indianapolis, Ind.) was used instead of 9E10, since scFvJ is actually a single domain Fv consisting of only the variable heavy chain and lacking the C-terminal c-myc epitope. To measure the relative affinity of scFvA and scFvK, all procedures were identical as in measuring affinity of D1.3 scFv, except that serially diluted RBE4 cell lysate (biotinylated, 1% OG) was used as a ligand source.
Yeast Display Immunoprecipitation.
Immunoprecipitation from complex mixtures. Approximately 5×107 of yeast cells displaying D1.3 scFvs were used for YDIP. Yeast cells were fixed with 3% v/v formaldehyde in PBS for 1 h at room temperature when indicated. Yeast were incubated 2 h at 4° C. with 34 nM HEL diluted into RBE4 cell lysate prepared with 1% (w/v) TX and washed twice with and eluted with 50 μL of 0.2 M glycine-HCl solution (pH 2.0). Yeast displaying the anti-fluorescein 4-4-20 scFv were used as a negative control.
Elution methods for antigen recovery. To monitor the amount of antigen obtained by various elution methods, D1.3 scFv and biotinylated HEL were used. A batch of 3×107 yeast cells displaying D1.3 scFv were incubated with 10 nM HEL in PBS-BSA for 2 h at 4° C. The yeast cells were washed twice with PBS at 4° C. prior to elution. Elution of immunoprecipitated product was performed by resuspending yeast cells in 30 μL of 0.2M glycine-HCl solution (pH 2.0), 9 M urea, 0.2 M NaOH, 3 M NaCl, or 0.2% (w/v) SDS for 10 min at 4° C. Where indicated, yeast cells were fixed with 3% v/v formaldehyde in PBS for 1 h at room temperature before incubation with the antigen, to reduce the co-elution of yeast proteins. The eluates were separated with SDS-PAGE (15% separating gel) and either probed with streptavidin-HRP conjugate (1/2000 dilution, GE healthcare, Piscataway, N.J.) or silver-stained following standard protocols. The HEL bands from Western blot films or silver stained gels were quantified using densitometry with ImageJ (US National Institutes of Health, Bethesda, Md.) to compare the relative amount of biotinylated HEL eluted under each condition.
Immunoprecipitaion of cell surface antigens and Western Blotting. For Western blotting analysis of immunoprecipitated antigens, 3-10×107 yeast cells were used. Induced yeast cells were collected by centrifugation and washed with PBS-BSA. Biotinylated RBE4 cell lysate prepared using the various detergents containing approximately 400 μg of total protein (corresponding to approximately 2×106 RBE4 cells) was mixed with yeast cells and incubated overnight at 4° C. in a total volume of 200 μL. After the incubation, yeast cells were washed three times by resuspension in 1 mL of the corresponding lysis buffer (without protease inhibitor cocktail) and incubated for 15 min at 4° C. Elution of immunoprecipitated product was performed by resuspending yeast cells in 30 μL of 0.2 M glycine-HCl solution (pH 2.0) for 10 min. The mixtures were briefly centrifuged and the eluted supernatants were separated with non-reducing SDSPAGE (8% separating gel), and blotted onto a nitrocellulose membrane (BioRad, Hercules, Calif.). Yeast displaying D1.3 scFv were used as a negative control for these experiments. Western blotting was subsequently performed with an anti-biotinmonoclonal antibody (0.5 μg/mL, clone BTN.4, Labvision, Fremont, Calif.). To test the serial elution process, scFvJ was used. The IP procedure is identical to that above except the eluted supernatant was serially applied for the indicated number of times to elute additional batches of yeast that have undergone the immunoprecipitation process. To confirm the scFvJ antigen neural cell adhesion molecule (NCAM), a monoclonal anti-NCAM antibody (clone 5B8, Developmental studies Hybridoma Bank, Iowa Cities, Iowa) was used for Western blotting.
Identification of immunoprecipitated proteins using tandem mass spectrometry. To sequence hen-egg lysozyme using tandem mass spectrometry (MS/MS), approximately 3×108 yeast displaying the D1.3 scFv were incubated for 2 h at 4° C. with 30 nM HEL in PBSTX, washed twice with PBSTX and eluted with 50 μL of 0.2 M glycine-HCl solution (pH 2.0). The elution product was precipitated for 2 h by adding 250 μL of ice-cold acetone at −20° C. The protein pellet was dried and reconstituted in 5 μL of 8 M urea and then diluted with 40 μL of 50 mM ammonium bicarbonate buffer with 10 mM DTT. Approximately 100 ng of sequencing grade modified trypsin (Promega, Madison, Wis.) was added and digestion was allowed to proceed for 16 h at 37° C. Peptide fractions were individually analyzed by nanoLC-MS/MS using 1100 series LC/MSD Trap SL spectrometer (Agilent, Palo Alto, Calif.). Chromatography of peptides prior to mass spectral analysis was accomplished using C18 reverse phase HPLC trap column (Zorbax 300SB-C18, 5 μM, 5×0.3 mm, Agilent) and separation column (Zorbax 300SB-C18, 3.5 μm, 0.075×150 mm, Agilent) onto which 40 μL of each extracted peptide fraction was automatically loaded. An Agilent 1100 series HPLC delivered solvents A: 0.1% (v/v) formic acid in water, and B: 95% (v/v) acetonitrile, 0.1% (v/v) formic acid at either 10 μL/min, to load sample, or 0.28 μL/min, to elute peptides directly into the nano-electrospray source over a 60 min 20% (v/v) B to 80% (v/v) B gradient. As peptides eluted from the HPLC-column/electrospray source, MS/MS spectra were collected over 5 channels from 300 to 2200 m/z; redundancy was limited by dynamic exclusion (120 s). MS/MS data were converted to mgf file format using Data Analysis Software (Agilent). Resulting mgf files were used to search the nonredundant NCBI database with an in-house licensed Mascot search engine (Matrix Science, London, UK) with methionine oxidation, cysteine carbamidomethylation and glutamine and asparagine deamidation as variable modifications.
For MS/MS sequencing of the RBE4 cell-surface antigen recognized by scFvJ, 3×109 formaldehyde fixed yeast cells displaying scFvJ were divided into eight microfuge tubes and RBE4 cell lysate containing 3 mg of total protein (both biotinylated and non-biotinylated ID was successful) in 1 mL of 1% OG with protease inhibitors was added to each tube. The yeast cells were incubated overnight at 4° C. in the cell lysate, and then 4 product pools created each by serial elution of two tubes with 70 μL of 0.2 M glycine-HCl solution (pH 2.0). To concentrate the antigen, the four elution products were combined and precipitated at 4° C. for 2 h by the addition of trichloroacetic acid (Fisher Scientific) to a final concentration of 13% (w/v). Protein was pelleted by centrifuging at 12,000×g for 15 min at 4° C. The pelleted protein was then dissolved in 2× Laemmli sample buffer for separation by SDS-PAGE (8%). Gels were stained using colloidal coomassie to visualize the eluted proteins. In-gel digestion was performed using standard protocols. In short, gel pieces were excised, dehydrated, and reduced in 50 μL 25 mMDTT in 25 mM ammonium bicarbonate for 20 min at 56° C., and alkylated with 50 μL of 55 mM iodoacetamide (Sigma, St. Louis, Mo.) in 100 mM ammoniumbicarbonate for 20 min at room temperature. Then the gel pieces were washed, dried and digested for 18 h at 37° C. with 400 ng of sequencing grade modified trypsin in 25 mM ammonium bicarbonate with 5% (v/v) acetonitrile. Digested peptides were eluted off from a C18 micro ZipTip (Millipore, Billerica, Mass.) with acetonitrile/H2O/TFA (60%:40%:0.2%) directly onto the Opti-TOF™ 384 Well plate (Applied Biosystems, Foster City, Calif.) and re-crystallized with 0.75 μL of matrix Y. K. Cho et al./Journal of Immunological Methods 341 (2009) 117-126 119 (10 mg/ml α-Cyano-4-hydroxycinnamic acid in acetonitrile/H2O/TFA (70%:30%:0.2%)). Peptide identification via result dependent MS/MS analysis was performed on a 4800 Matrix-Assisted Laser Desorption/Ionization-Time of Flight-Time of Flight (MALDI TOF-TOF) mass spectrometer (Applied Biosystems, Foster City, Calif.). In short, a peptide fingerprint was generated by scanning the 700-4,000 Da mass range using 1500 shots acquired from 20 randomized regions of the sample spot using an OptiBeam™on-axis Nd:YAG laser (4200 intensity, 200 Hz firing rate, 3-7 ns pulses). The fifteen most abundant precursors, excluding trypsin autolysis peptides and sodium/potassium adducts, were selected for subsequent tandem MS analysis where 2000 total shots were taken with 4700 laser intensity and 2 kV collision induced activation (CID) using air. Post-source decay (PSD) fragments from the precursors of interest were isolated by timed-ion selection and reaccelerated into the reflectron to generate the MS/MS spectrum. Raw data was deconvoluted using GPS Explorer™ software and submitted for peptide mapping and MS/MS ion search analysis against non-redundant NCBI database with an in-house licensed Mascot search engine (Matrix Science, London, UK) with fixed modification of carboxyamidomethylation of cysteines and parent/fragment ion mass tolerance of 0.3 Da.
Quantitation of scFv and NCAM expression levels using flow cytometry. To quantitate the expression level of scFvs on the yeast cell surface, Quantum simply cellular anti-mouse IgG microbeads (Banglabs, Fishers, Ind.), which contain four populations of beads with known mouse IgG binding sites were used. Yeast cells displaying either D1.3 scFv or scFvA were incubated with a mouse anti-c-myc antibody (9E10, 30 μg/mL, Covance, Berkeley, Calif.) in PBS-BSA for 1 h. For scFvJ quantitation, a mouse anti-HA antibody (12CA5, Roche Diagnostics, Indianapolis, Ind., 1/100 dilution) was used instead. The microbeads were also incubated under the same conditions. Next, the yeast cells and the microbeads were washed, incubated with a goat anti-mouse IgG-Alexa555 conjugate (1/500 dilution, Invitrogen, Carlsbad, Calif.) for 30 min. and analyzed using the FACSCalibur flow cytometer. To quantitate the expression level of NCAM in RBE4 cells, RBE4 cells were permeabilized with 0.5% (w/v) saponin in PBSCM with 40% (v/v) goat serum (PBSCMG, Sigma, St. Louis, Mo.) and incubated with a mouse anti-NCAM antibody (clone 5B8, diluted to 50 μg/mL) in PBSCMG for 1 h. The microbeads were also incubated under the same conditions. The secondary reagents used were identical to the scFv quantitation.
Results:
Effect of Detergents on yeast-displayed scFv binding activity. Yeast display immunoprecipitation (YDIP) takes advantage of scFv expressed as fusions to the yeast cell wall for capture of antigens from complex mixtures such as cell lysates (
The lone exception was SDS, which resulted in a complete inhibition of the antigen-antibody interaction. Interestingly however, when 0.1% SDS was mixed with 1% TX and 0.5% DOC as in the RIPA buffer, the scFv remained active (
ScFvs (scFvA, scFvK, scFvD, and scFvJ) that were previously selected as capable of binding to rat brain endothelial cell line (RBE4) plasma membrane proteins (see previous Examples) were used to assess the general compatibility of yeast displayed scFv for immunoprecipitation both from the standpoint of detergent tolerance and applicability to membrane protein targets. Yeast cells displaying these scFvs were used to immunoprecipitate antigens from RBE4 cell lysates generated by TX detergent-solubilization of biotinylated plasma membranes (
Elution, antigen purity and quantity by YDIP. Next, we further assessed the purity and quantity of antigens isolated using YDIP under a variety of elution conditions. The anti-HEL D1.3 scFv was again tested as a model system representing an antibody-soluble antigen combination. We first sought to identify the most efficient method for dissociation of bound antigens using YDIP of biotinylated HEL from a buffer solution (
In addition to coextracted yeast proteins, another confounding factor for antigen analysis is the non-specific binding of mammalian cell lysate proteins to yeast. Lysate-derived impurities can be more detrimental than the coextracted yeast proteins in applications such as antigen identification, since yeast proteins can be distinguished from mammalian proteins by sequence in many cases, while contaminants from the same species cannot. To evaluate lysate-derived impurities, YDIP was performed to recover HEL that had been doped into RBE4 cell lysates. First, using unfixed yeast, several bands other than HEL were detected on a silver-stained gel, especially in the lower molecular weight range (
Another important aspect to consider is the amount of antigen isolated, because 50-500 ng are desired for characterization by peptide sequencing using MS/MS. We first estimated the quantity of antigen that could be reasonably isolated using YDIP. Using the HEL-D1.3 pair that has an affinity of Kd=0.488 nM in TX lysate (Table 1) with 31,000 scFv per yeast cell (see Materials and methods for scFv quantification details) and 10 nM or higher antigen concentration, approximately 150 ng of antigen can be isolated with 3×108 D1.3-displaying yeast cells (15 mL overnight culture with density of 2 OD600 nm). Using 5×107 yeast cells (˜3 mL overnight culture), as was the case in
YDIP of mammalian cell-surface antigens. Since YDIP exhibited good recovery and purity for the soluble HEL antigen, the optimized YDIP protocol was applied to analyze plasma membrane antigens. We have shown that yeast cells displaying scFvs selected from a non-immune human scFv library can be used to immunoprecipitate mammalian cell surface antigens from cell lysates for Western blot analysis using various detergents and elution conditions (
Identification of immunoprecipitated proteins using tandem mass spectrometry. To assess the compatibility of YDIP with mass spectrometry for peptide sequencing and antigen identification, we first tested the D1.3-HEL system. To this end, HEL was immunoprecipitated from a TX containing buffer using 3×108 formaldehyde-fixed yeast cells. HEL was subsequently eluted with low pH buffer, acetone precipitated and the entire elution product was trypsinized in solution. The trypsinized elution product was directly subjected to LC-MS/MS to attempt detection of HEL and to monitor the effects of yeast protein contaminants in the MS/MS analysis. As expected by the fact that HEL was isolated in high purity (
aProteins that have a summed peptide ion score greater than 67 carry a significance of p < 0.05.
Next, YDIP was applied for de novo identification of the RBE4 plasma membrane antigen immunoprecipitated by scFvJ. Although the molecular weight of the antigen was previously determined by anti-biotin Western blotting (
Discussion:
This study demonstrates that scFv-displaying yeast cells are effective affinity reagents allowing the immunoprecipitation and identification of antigens from detergent solubilized cell lysates. We have shown that yeast displaying scFvs retain their activities in detergent solutions by quantitative assessment of antigen-antibody interactions and two-fold differences in antibody affinity could be accurately determined using cell lysate as an antigen source. Furthermore, the immunoprecipitated proteins could be readily characterized by well-established downstream procedures such as Western blotting, silver/coomassie staining, and mass spectrometry. Optimized YDIP conditions that combine fixed yeast, a low pH serial elution strategy and concentration by TCA precipitation allowed isolation of enough cell-surface receptor for MS/MS antigen identification.
Although it is known that various non-ionic detergents such as Triton X-100, Tween 20, and Brij do not affect the activity of antibodies, this generalization may not apply for antibody fragments such as scFvs that could be less stable than the whole immunoglobulin molecule. Here we have shown that scFvs expressed on yeast surface are active in non-ionic detergents (TX, OG), zwitterionic detergent (CHAPS), and a mixture of ionic and non-ionic detergents (RIPA buffer). It was important to test a wide variety of detergents, since antigens have varying solubility depending on the detergent. Interestingly, while scFvs were inactive in the highly denaturing ionic detergent, SDS, as previously seen for antibodies, when SDS was in a mixture with TX and DOC(RIPA buffer), the scFvs remained active. This might be explained by the fact that proteins denatured in SDS can be renatured by the addition of TX.
For the elution of antigens, we have applied conditions classically used in immunoprecipitations (
Another important consideration is the quantity of antigen isolated with YDIP. When the antigen concentration is in the nanomolar range, N200 ng of antigens could be easily isolated. However, as an example, this antigen concentration corresponds to a protein expressed at approximately 100,000 copies per mammalian cell for the case of a 14 kDa protein, with 5×107 mammalian cells in 1 mL of lysis buffer. This concentration of antigen may not be achievable in every situation, since normal protein expression levels may vary between 100 and 100,000 copies per cell, averaging about 50,000 copies per cell, with plasma membrane proteins being on the low end of this spectrum. In the case of scFvJ-NCAM pair that was evaluated in this study, quantification of NCAM expression level in RBE4 cells showed that there are 72,000 copies of NCAM per RBE4 cell corresponding to a 2 nM antigen concentration in the cell lysate. This result suggests that for antigens with average expression level, YDIP can be applied for de novo identification. On the yeast side of the equation, when the antibodies are isolated from a non immune or synthetic library, Kd values can be in the low micromolar range limiting recovery from dilute cell lysates. Theoretical calculations based on bimolecular equilibrium show that even with antibodies with micromolar affinities, it is estimated that N100 ng of antigens can be isolated using 109 yeast cells. Indeed this was the case for the scFvJ-NCAM system where 3×109 yeast cells from 150 mL of overnight culture were sufficient. In terms of scFv expression levels which can also affect YDIP yield, quantification of displayed D1.3, scFvA and scFvJ showed expression levels ranging from 31,000 to 68,000 scFv per yeast cell, differences that can be accommodated for simply by scaling the number of yeast used in the YDIP procedure.
The major immunoprecipitation product that appears in anti-biotin Western blotting of the scFvJ elution product was identified to be the neural cell adhesion molecule (NCAM) (
In conclusion, the results show that YDIP provides an efficient and facile method for antigen identification compared to conventional immunoprecipitation approaches since it eliminates additional cloning and production steps needed to obtain affinity reagents. Therefore, we anticipate that YDIP will enable a wide range of applications in combination with yeast antibody library screening technology.
As described in the preceding EXAMPLES, the invention provides a method of panning a yeast display antibody library against cultured endothelial cells and simultaneously isolating multiple binders of different affinities against one or more antigens. The binding clones were isolated and their ability to trigger endocytosis and transcytosis was then confirmed using endothelial cells. The invention also provides an improved immunoprecipitation method capable of identifying specific antigen targets. Using the improved immunoprecipitation method, the antigen binding target for scFvJ was identified to be the neural cell adhesion molecule (NCAM).
While this invention has been described in conjunction with the various exemplary embodiments outlined above, various alternatives, modifications, variations, improvements, and/or substantial equivalents, whether known or that are or may be presently unforeseen, may become apparent to those having at least ordinary skill in the art. Accordingly, the exemplary embodiments according to this invention, as set forth above, are intended to be illustrative, not limiting. Various changes may be made without departing from the spirit and scope of the invention. Therefore, the invention is intended to embrace all known or later-developed alternatives, modifications, variations, improvements, and/or substantial equivalents of these exemplary embodiments.
Sequence Listing. Applicants are submitting as part of this Application a computer readable sequence listing txt file, which is incorporated by reference herein.
This application is a continuation-in-part of U.S. application Ser. No. 11/759,723, filed Jun. 7, 2007, which claims the benefit of U.S. Provisional application 60/811,618, filed Jun. 7, 2006, both of which are incorporated herein by reference in their entirety for all purposes.
This invention was made with United States government support awarded by the following agency: National Institutes of Health, Grant No. NS052649 & EY018506, and National Science Foundation, Grant No. 0238864. The United States government has certain rights in this invention.
| Number | Name | Date | Kind |
|---|---|---|---|
| 7138501 | Ruben et al. | Nov 2006 | B2 |
| 20070009972 | Chao et al. | Jan 2007 | A1 |
| Number | Date | Country |
|---|---|---|
| 2005049802 | Jun 2005 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20100209425 A1 | Aug 2010 | US |
| Number | Date | Country | |
|---|---|---|---|
| 60811618 | Jun 2006 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 11759723 | Jun 2007 | US |
| Child | 12540897 | US |
