Pigment epithelium-derived factor: characterization, genomic organization and sequence of the PEDF gene
Nucleic acids encoding the neurotrophic protein known as pigment epithelium-derived factor (PEDF), a truncated version of PEDF referred to as rPEDF, and equivalent proteins, vectors comprising such nucleic acids, host cells into which such vectors have been introduced, recombinant methods for producing PEDF, rPEDF, and equivalent proteins, the rPEDF protein and equivalent proteins of rPEDF and PEDF-BP, -BX and BA, and the PEDF protein produced by recombinant methods. Effects and uses of these variants on 1) neuronal differentiation (neurotrophic effect) 2) neuron survival (neuronotrophic effect) and 3) glial inhibition (gliastatic effect) are described.
Latest The United States of America, As Represented by the Dept. of Health and Human Services Patents:
This application is a continuation-in-part of application Ser. No. 08/257,963 filed on Jun. 7, 1994, which is a continuation-in-part of application Ser. No. 07/952,796 filed on Sep. 24, 1992.
TECHNICAL FIELD OF THE INVENTIONThis invention relates to a neurotrophic, neuronotrophic and gliastatic protein. More specifically, this invention relates to the biological properties of a protein known as pigment epithelium-derived factor (PEDF) and recombinant forms of the protein. This invention also relates to a truncated version of PEDF that is referred to as rPEDF. In addition to PEDF and rPEDF and functionally equivalent proteins, this invention relates to nucleic acids that encode rPEDF, and fragments thereof, to vectors comprising such nucleic acids, to host cells into which such vectors have been introduced, and to the use of these host cells to produce such proteins.
BACKGROUND OF THE INVENTIONPigment epithelium-derived factor, otherwise known as pigment epithelium differentiation-factor, was identified in the conditioned medium of cultured fetal human retinal pigment epithelial cells as an extracellular neurotrophic agent capable of inducing neurite outgrowth in cultured human retinoblastoma cells (Tombran-Tink et al. (1989) Invest. Ophthalmol. Vis. Sci., 30 (8), 1700-1707). The source of PEDF, namely the retinal pigment epithelium (RPE), may be crucial to the normal development and function of the neural retina. A variety of molecules, including growth factors, are synthesized and secreted by RPE cells. Given that the RPE develops prior to and lies adjacent to the neural retina, and that it functions as part of the blood-retina barrier (Fine et al. (1979) The Retina, Ocular Histology: A Text and Atlas, New York, Harper & Row, 61-70), the RPE has been implicated in vascular, inflammatory, degenerative, and dystrophic diseases of the eye (Elner et al. (1990) Am. J. Pathol., 136, 745-750). In addition to growth factors, nutrients and metabolites are also exchanged between the RPE and the retina. For example, the RPE supplies to the retina the well-known growth factors PDGF, FGF, TGF-α, and TGF-β (Campochiaro et al. (1988) Invest. Ophthalmol. Vis. Sci., 29, 305-311; Plouet (1988) Invest. Ophthalmol. Vis. Sci., 29, 106-114; Fassio et al. (1988) Invest. Ophthalmol. Vis. Sci., 29, 242-250; Connor et al. (1988) Invest. Ophthalmol. Vis. Sci., 29, 307-313). It is very likely that these and other unknown factors supplied by the RPE influence the organization, differentiation, and normal functioning of the retina.
In order to study and determine the effects of putative differentiation factors secreted by the RPE, cultured cells have been subjected to retinal extracts and conditioned medium obtained from cultures of human fetal RPE cells. For example, U.S. Pat. No. 4,996,159 (Glaser) discloses a neovascularization inhibitor recovered from RPE cells that is of a molecular weight of about 57,000+/−3,000. Similarly, U.S. Pat. No. 1,700,691 (Stuart), U.S. Pat. No. 4,477,435 (Courtois et al.), and U.S. Pat. No. 4,670,257 (Guedon born Saglier et al.) disclose retinal extracts and the use of these extracts for cellular regeneration and treatment of ocular disease. Furthermore, U.S. Pat. No. 4,770,877 (Jacobson) and U.S. Pat. No. 4,534,967 (Jacobson et al.) describe cell proliferation inhibitors purified from the posterior portion of bovine vitreous humor.
PEDF only recently has been isolated from human RPE as a 50-kDa protein (Tombran-Tink et al. (1989) Invest. Ophthalmol. Vis. Sci., 29, 414; Tombran-Tink et al. (1989) Invest. Ophthalmol. Vis. Sci., 30, 1700-1707; Tombran-Tink et al. (1991) Exp. Eye Res., 53, 411-414). Specifically, PEDF has been demonstrated to induce the differentiation of human Y79 retinoblastoma cells, which are a neoplastic counterpart of normal retinoblasts (Chader (1987) Cell Different., 20, 209-216). The differentiative changes induced by PEDF include the extension of a complex meshwork of neurites, and expression of neuronal markers such as neuron-specific enolase and neurofilament proteins. This is why the synthesis and secretion of PEDF protein by the RPE is believed to influence the development and differentiation of the neural retina. Furthermore, PEDF is only highly expressed in undifferentiated human retinal cells, like Y79 retinoblastoma cells, but is either absent or down-regulated in their differentiated counterparts. Recently, it was reported that PEDF mRNA is expressed in abundance in quiescent human fetal W1 fibroblast cells and not expressed in their senescent counterparts (Pignolo et al., 1993).
Further study of PEDF and examination of its potential therapeutic use in the treatment of inflammatory, vascular, degenerative, and dystrophic diseases of the retina and central nervous system (CNS) necessitates the obtention of large quantities of PEDF. Unfortunately, the low abundance of PEDF in fetal human eye and furthermore, the rare availability of its source tissue, especially in light of restrictions on the use of fetal tissue in research and therapeutic applications, make further study of PEDF difficult at best. Therefore, there remains a need for large quantities of PEDF and equivalent proteins. Accordingly, the obtention of nucleic acids that encode PEDF and equivalent proteins, and the capacity to produce PEDF and equivalent proteins in large quantities would significantly impact upon the further study of PEDF, its structure, biochemical activity and cellular function, as well as the discovery and design of therapeutic uses for PEDF.
SUMMARY OF THE INVENTIONIt is an object of the present invention to provide nucleic acids encoding for PEDF and functional fragments thereof, vectors comprising such nucleic acids, host cells into which such vectors have been introduced, and a recombinant method of producing PEDF and equivalent proteins. It is another object of the present invention to obtain the genomic DNA sequences encoding for PEDF, identify the intron-exon junctions, the chromosome location in the human genome, and to provide the regulatory regions of the gene which flank the genomic sequence. The present invention relates to such genomic PEDF DNA.
It is a further object of the present invention to provide structural characteristics of PEDF and its similarities to the serpin family of serine protease inhibitors, both structural and functional.
It is yet another object of the present invention to provide PEDF and equivalent proteins produced in accordance with such a recombinant method, wherein the PEDF and equivalent proteins so produced are free from the risks associated with the isolation of PEDF from naturally-occurring source organisms.
Another object of the present invention is to provide nucleic acids for a truncated version of PEDF, referred to as rPEDF, and equivalent proteins, vectors comprising such nucleic acids, host cells into which such vectors have been introduced, and a recombinant method of producing rPEDF and equivalent proteins. It is also an object of the present invention to provide rPEDF and equivalent proteins produced in accordance with such a recombinant method.
It is a further object of the invention to provide a PEDF protein having neuronotrophic and gliastatic activity. The neuronotrophic activity is seen in the prolonged survival of neuronal cells. The gliastatic activity is observed in the inhibition of growth of glial cells in the presence of PEDF or active fragment thereof. It is another object of the invention to provide methods for treating neuronal cells so as to promote/enhance neuron survival and prevent growth of glial cells, comprising treating such cell populations with an effective amount of PEDF or an active fragment thereof.
It is yet another object of the present invention to provide antibodies which specifically recognize PEDF, either monoclonal or polyclonal antibodies, raised against native protein, the recombinant protein or an immunoreactive fragment thereof. It is an object of the invention to provide methods for detecting PEDF by immunoassay using such antibody preparation in determining aging and/or other degenerative diseases. Another object of the invention relates to a method of using PEDF antibodies to specifically inhibit PEDF activity.
These and other objects and advantages of the present invention, as well as additional inventive features, will be apparent from the description of the invention provided herein.
DESCRIPTIONS OF THE FIGURES
FIGS. 6A & 6B: Relationship between cell density plated and optical density measured by MTS assay. Different concentrations of postnatal-day 8 cerebellar granule cells were added to 96 well plate and cultured in serum-containing medium (6A), or chemically defined medium (6B). Optical density was measured on days in vitro (DIV) 1, 4, or 7. Square, DIV 1; Solid circle, DIV 4; Open circle, DIV7. The data are plotted as function of cell density (n=6).
The present invention relates to a protein having novel, important and unobvious properties. Pigment epithelium-derived factor (PEDF) is a protein having neurotrophic, neuronotrophic and gliastatic characteristics. The present invention further relates to the DNA sequences coding for the PEDF gene, the genomic DNA containing the PEDF gene and fragments of the PEDF gene encoding for protein fragments of PEDF having biological activity.
“Neurotrophic” activity is defined herein as the ability to induce differentiation of a neuronal cell population. For example, PEDF's ability to induce differentiation in cultured retinoblastoma cells is considered neurotrophic activity.
“Neuronotrophic” activity is defined herein as the ability to enhance survival of neuronal cell populations. For example, PEDF's ability to act as a neuron survival factor on neuronal cells is neuronotrophic activity.
“Gliastatic” activity is defined herein as the ability to inhibit glial cell growth and proliferation. For example, PEDF's ability to prevent growth and/or proliferation of glial cells is gliastatic activity.
Based upon the protein amino acid sequence elucidated in the present invention, PEDF has been found to have extensive sequence homology with the serpin gene family, members of which are serine protease inhibitors. Many members of this family have a strictly conserved domain at the carboxyl terminus which serves as the reactive site of the protein. These proteins are thus thought to be derived from a common ancestral gene. However the developmental regulation differs greatly among members of the serpin gene family and many have deviated from the classical protease inhibitory activity (Bock (1990) Plenum Press, New York Bock, S. C., Protein Eng. 4, 107-108; Stein et al. (1989) Biochem. J. 262, 103-107). Although PEDF shares sequence homology with serpins, analysis of the cDNA sequence indicates that it lacks the conserved domain and thus may not function as a classical protease inhibitor.
Genomic sequencing and analysis of PEDF has provided sequences of introns and exons as well as approximately 4 kb of 5′-upstream sequence. The present invention demonstrates the localization of the gene for PEDF to 17p13.1 using both in situ hybridization and analyses of somatic cell hybrid panels (Tombran-Tink, et al., (1994) Genomics, 19:266-272). This is very close to the p53 tumor suppressor gene as well as to the chromosomal localization of a number of hereditary cancers unrelated to mutations in the p53 gene product. PEDF thus becomes a prime candidate gene for these cancers.
The full length genomic PEDF sequence is represented by SEQ ID NO:43. The PEDF gene encompasses approximately 16 Kb and contains 8 exons all of which have conventional consensus splice-sites. The 5′ flanking region of the PEDF gene contains two Alu repetitive elements which cover approximately two thirds of the first 1050 bp of the putative promoter sequence. There are also several sequence motifs which may be recognized by members of several families of transcription factors. The presence of two possible binding sites for the ubiquitous octamer family of transcription factors, may explain the presence of PEDF in most tissues tested. The presence of other more specific elements, however, suggests that PEDF is under precise control and supports previous work including its effects on such diverse processes as neuronal differentiation and fibroblast senescence.
The genomic PEDF sequence or fragments thereof are useful as a probe for detecting the gene in a cell. In addition, such a probe is useful in a kit for identification of a cell type carrying the gene. Mutations, deletions or other alternations in the gene organization can be detected through the use of a DNA probe derived from the PEDF genomic sequence.
Tissue DistributionAlthough PEDF is particularly highly expressed by RPE cells, it is detectable in most tissues, cell types, tumors, etc. by Northern and Western blot analyses. It is readily detected, for example in vitreous and aqueous humors. The important question of subcellular localization of PEDF has also been addressed. Although the bulk of the PEDF appears to be secreted, we have used a PEDF antibody to probe cultured monkey RPE-cells and found that PEDF is associated with the nucleus as well as with very specific cytoskeletal structures in the cytoplasm. Importantly, this varies as to the age of the cells and the specific cell-cycle state examined. For example, the protein appears to concentrate at the tips of the pseudopods of primate RPE cells that interact with the substratum during the initial stages of attachment. Later though, this staining disappears and there is appearance of the protein in association with specific cytoskeletal structures and the nucleus. Thus it appears that PEDF plays an important intracellular role in both nucleus and cytoplasm.
Involvement in Cell CycleThe present invention indicates that there is expression in dividing, undifferentiated Y-79 cells and little or no expression in their quiescent, differentiated counterparts (Tombran-Tink, et al. (1994) Genomics, 19:266-272). Pignolo et al. (1993) J. Biol. Chem., 268:2949-295) have demonstrated that the synthesis of PEDF in WI-38 fibroblast cells is restricted to the G0 stage of the cell cycle in young cells. Moreover, in old senescent cells, PEDF messenger RNA is absent.
Production of Recombinant PEDFSegmentation of the PEDF polypeptide is basic to studies on structure-function. For this purpose, expression vectors containing fragments of PEDF coding sequences provide an excellent source for synthesizing and isolating different regions of the PEDF polypeptide. Expression of human fetal PEDF sequences was achieved with E. coli expression vectors and the human fetal PEDF cDNA. We have shown that the recombinant PEDF product (rPEDF) is a biologically-active neurotrophic factor and is obtained in yields on the order of 1.3 mg/g of wet E. coli cells. Truncated peptides can also be made from appropriate molecular biological constructs and expressed in E. coli. Using these products, we have evidence that two distinct regions on the PEDF primary structure can be distinguished: 1) an “active site” conferring neurotrophic activity on the molecule that is located within amino acid residues 44-121 near the N-terminal of the protein and 2) a region near the C-terminal with homology to a serpin exposed loop i.e., the “classical” serpin active site. These results suggest 1) that the overall native conformation of PEDF is not required for neurite outgrowth and 2) that inhibition of serine proteases can not account for the biological activity of PEDF. We now have a series of truncated rPEDF constructs that span the protein sequence and can pinpoint the specific neurotrophic “active site” near the N-terminal.
Characterization with a Highly Specific Polyclonal AntibodyPurified recombinant human PEDF was used to develop a polyclonal antibody (“Anti-rPEDF”) that specifically blocks the PEDF-mediate neurotrophic activity. Furthermore, the anti-rPEDF completely blocks the IPM-induced neurotrophic activity.
Neuronotrophic Properties of PEDF In addition to demonstrating that native PEDF and rPEDF are neurotrophic in the Y-79 and Weri tumor cell systems, the present invention determined whether PEDF had an effect on normal neurons in primary culture. For this purpose, studies were conducted using cultures of normal cerebellar granule cells (CGCs) prepared from the 8-day postnatal rat. Cells treated with rPEDF did not respond to treatment by exhibiting a more neuronal morphological appearance. However, PEDF had a large effect on granule cell survival. Since these cells are not tumorous or transformed cells, they have a finite life in culture, dying in about 21 days depending on the culture medium. PEDF-treated culture, however, contained up to 10-fold more cells after 10 days of culture in serum-free medium compared to non-treated culture (
In General Tissue Culture Research:
Two problems that generally plague any tissue culture experiment using neurons and glia is that the neurons tend to die quickly and that glia tend to overrun the culture dish. PEDF or its peptides can help in both regards. Thus, one commercial use of PEDF might be as a general culture medium additive when CNS cells are to be cultured.
In CNS Transplantation Studies:
It is thought that transplantation of neurons may cure certain pathologies. For example, in Parkinson's disease, transplantation of specific fetal brain cells into patients could alleviate or cure the problems associated with the disease. One of the major problems to contend with, though, would be to prolong the life of the transplanted cells and to keep them differentiated, e.g. secreting the proper substances, etc. Pretreatment of the cells with PEDF could aid in both of these areas. Similarly, transfection of either neurons or astroglia with the PEDF gene before implantation can be a long-term source of PEDF at the transplantation site.
There is much activity in attempts at transplantation of neural retina and photoreceptor cells to help cure blindness. Attempts to date have not been fruitful both due to non-differentiation and death of the grafts. Again, PEDF may help in both regards. Specifically, photoreceptor neurons to be transplanted can be pretreated with PEDF or the gene transfected into the cells before surgery. Alternatively, PEDF can be transfected at high levels into adjacent retinal pigment epithelial (RPE) cells where they can serve as a supranormal source of the protein. Several investigators have now shown that cultured RPE cells survive very well after transplantation into the interphotoreceptor space of test animals. Transfection of human RPE cells in vitro with the PEDF gene then use of them in retinal transplantation thus is feasible.
In Neurodegenerative Diseases:
Many neurodegenerative diseases and other insults to the CNS (brain and retina) are typified by death of neurons and overpopulation by glia (gliosis). PEDF can be used effectively in these conditions to prolong the life and functioning of the primary neurons and to stave off the glial advance. PEDF can be effective, for example, in blocking microglial activation in response to CNS injury as well as prolonging/sparing the lives of neurons.
In the retina, it is predictable that PEDF inhibits the Muller glial cells. Since Muller cells are similar to astroglia, PEDF would be similarly effective in blocking gliosis in conditions such as retinal detachment, diabetes, Retinitis Pigmentosa, etc. as well as sparing the lives of the retinal neurons.
In Glial Cancers:
Most of the major forms of cancer that strike the CNS involve glial elements, PEDF is a gliastatic factor that can be used in combination with other forms of therapy. For example, along with surgery, PEDF can effectively inhibit the spread or reoccurrence of the disease.
Genetic AnalysisThe present invention relates to the determination of the organization of the human PEDF gene and its promoter and analysis of its evolutionary relatedness and expression in a variety of human fetal and adult tissues.
The present invention provides, among other things, a nucleic acid which encodes PEDF. In particular, a cDNA sequence is provided as set forth in SEQ ID NO:1. This cDNA sequence codes for PEDF, which has the amino acid sequence set forth in SEQ ID NO:2. Further genomic sequences are mapped in
The term “nucleic acid” refers to a polymer of deoxyribonucleic acid (DNA) or ribonucleic acid (RNA), which can be derived from any source, can be single- or double-stranded, and can optionally contain synthetic, non-natural, or altered nucleotide which are capable of being incorporated into DNA or RNA polymers. The nucleic acid of the present invention is preferably a segment of DNA.
The present invention further provides truncated versions of PEDF. The largest of these is referred to as rPEDF, and comprises the amino acid sequence Met-Asn-Arg-Ile fused to Asp44 . . . Pro418 of PEDF, the amino terminus of which has been deleted. The rPEDF protein comprises the amino acid sequence of SEQ ID NO:3. The present invention also provides a nucleic acid which encodes a protein comprising the amino acid sequence of rPEDF, i.e., the amino acid sequence of SEQ ID NO:3.
One who is skilled in the art will appreciate that more than one nucleic acid may encode any given protein in view of the degeneracy of the genetic code and the allowance of exceptions to classical base pairing in the third position of the codon, as given by the so-called “Wobble rules”. Accordingly, it is intended that the present invention encompass all nucleic acids that encode the amino acid sequences of SEQ ID NO:2 and SEQ ID NO:3, as well as equivalent proteins. The phrase “equivalent nucleic acids” is intended to encompass all of these nucleic acids.
It also will be appreciated by one skilled in the art that amino acid sequences may be altered without adversely affecting the function of a particular protein. In fact, some alterations in amino acid sequence may result in a protein with improved characteristics. The determination of which amino acids may be altered without adversely affecting the function of a protein is well within the ordinary skill in the art. Moreover, proteins that include more or less amino acids can result in proteins that are functionally equivalent. Accordingly, it is intended that the present invention encompass all amino acid sequences that result in PEDF protein or functional protein fragments thereof.
Some examples of possible equivalent nucleic acids and equivalent proteins include nucleic acids with substitutions, additions, or deletions which direct the synthesis of the rPEDF protein and equivalent protein fragments thereof; nucleic acids with different regulatory sequences that direct the production of rPEDF proteins; variants of rPEDF which possess different amino acids and/or a number of amino acids other than four fused to the amino terminal end of the protein; and PEDF and rPEDF and functional protein fragments thereof with amino acid substitutions, additions, deletions, modifications, and/or posttranslational modifications, such as glycosylations, that do not adversely affect activity. Since the neurotrophic activity has been correlated to a particular portion of the PEDF protein fragments containing these residues are clearly within the scope of the present invention.
The present invention also provides a vector which comprises a nucleic acid of SEQ ID NO:1, a nucleic acid which encodes a protein comprising the amino acid sequence of SEQ ID NO:2 or an equivalent protein, a nucleic acid which encodes a protein comprising the amino acid sequence of SEQ ID NO:3 or conservatively modified variant proteins, and conservatively modified variant nucleic acids thereof.
In particular, the present invention provides the vector πFS17, which comprises the nucleic acid of SEQ ID NO:1, and the vector pEV-BH, which comprises a nucleic acid which encodes a protein comprising the amino acid sequence of SEQ ID NO:3. It will be appreciated by those skilled in the art that the cDNA inserts described can be present in alternative vectors. For example, inserts can be in vectors of different nature, such as phages, viral capsids, plasmids, cosmids, phagemids, YACs, or even attached to the outside of a phage or viral capsid. The vectors can differ in host range, stability, replication, and maintenance. Moreover, the vectors can differ in the types of control exerted over cloned inserts. For example, vectors can place cloned inserts under the control of a different promoter, enhancer, or ribosome binding site, or even organize it as part of a transposon or mobile genetic element.
The present invention also provides a host cell into which a vector, which comprises a nucleic acid of SEQ ID NO:1, a nucleic acid which encodes a protein comprising the amino acid sequence of SEQ ID NO:2 or an equivalent protein, a nucleic acid which encodes a protein comprising the amino acid of SEQ ID NO:3 or an equivalent protein, or an equivalent nucleic acid thereof, has been introduced. In particular, the host cell may have the vector πFS17, which comprises the nucleic acid of SEQ ID NO:1, or the vector pEV-BH, which comprises a nucleic acid which encodes a protein comprising the amino acid sequence of SEQ ID NO:3.
The vectors of the present invention can be introduced into any suitable host cell, whether eukaryotic or prokaryotic. These host cells may differ in their preferred conditions for growth, their nutritive requirements, and their sensitivity to environmental agents. Any appropriate means of introducing the vectors into the host cells may be employed. In the case of prokaryotic cells, vector introduction may be accomplished, for example, by electroporation, transformation, transduction, conjugation, or mobilization. For eukaryotic cells, vectors may be introduced through the use of, for example, electroporation, transfection, infection, DNA coated microprojectiles, or protoplast fusion.
The form of the introduced nucleic acid may vary with the method used to introduce the vector into a host cell. For example, the nucleic acid may be closed circular, nicked, or linearized, depending upon whether the vector is to be maintained as an autonomously replicating element, integrated as provirus or prophage, transiently transfected, transiently infected as with a replication-disabled virus or phage, or stably introduced through single or double crossover recombination events.
The present invention also provides a method of producing PEDF, rPEDF, and equivalent proteins, which method comprises expressing the protein in a host cell. For example, a host cell into which has been introduced a vector which comprises a nucleic acid of SEQ ID NO:1, a nucleic acid which encodes a protein comprising the amino acid sequence of SEQ ID NO:2 or an equivalent protein, a nucleic acid which encodes a protein comprising the amino acid of SEQ ID NO:3 or an equivalent protein, or an equivalent nucleic acid thereof, may be cultured under suitable conditions to produce the desired protein. In particular, a host cell into which has been introduced the vector πFS17, which comprises the nucleic acid of SEQ ID NO:1, or the vector pEV-BH, which comprises a nucleic acid which encodes a protein comprising the amino acid sequence of SEQ ID NO:3, may be cultured under suitable conditions to produce the proteins comprising the amino acid sequences of SEQ ID NO:2 and SEQ ID NO:3, respectively.
The present invention also provides recombinantly produced PEDF, and functional protein fragments thereof which have been produced in accordance with the aforementioned present inventive method of culturing an appropriate host cell to produce the desired protein. The production of a protein such as PEDF by recombinant means enables the obtention of large quantities of the protein in a highly purified state, free from any disease-causing agents which may accompany the protein isolated or purified from a naturally occurring source organism, and obviates the need to use, for example, fetal tissue as a source for such a protein.
Recombinant PEDF and functional protein fragments thereof may be supplied as active agents to cells by a variety of means, including, for example, the introduction of nucleic acids, such as DNA or RNA, which encode the protein and may be accordingly transcribed and/or translated within the host cell, the addition of exogenous protein, and other suitable means of administration as are known to those skilled in the art. In whatever form in which supplied, the active agent can be used either alone or in combination with other active agents, using pharmaceutical compositions and formulations of the active agent which are appropriate to the method of administration. Pharmaceutically acceptable excipients, i.e., vehicles, adjuvants, carriers or diluents, are well-known to those who are skilled in the art, and are readily available. The choice of excipient will be determined in part by the particular compound, as well as by the particular method used to administer the compound. Accordingly, there is a wide variety of suitable formulations which can be prepared in the context of the present invention. However, pharmaceutically acceptable excipients not altering the neurotrophic, neuronotrophic and gliastatic activities of the recombinant protein are preferred.
The following examples serve to illustrate further the present invention and are not to be construed as limiting its scope in any way.
EXAMPLE 1This example describes the trypsin digestion of PEDF and the amino acid sequencing of the resulting fragments.
PEDF was purified from the medium of a primary culture of human fetal RPE cells by high performance liquid chromatography (HPLC). The HPLC-purified PEDF was then reduced and alkylated. Afterwards, it was dried and redissolved in 50 μl of CRA buffer (8 M urea, 0.4 M ammonium carbonate, pH 8.0), and 5 μl of 45 mM dithiothreitol (DTT) (Calbiochem, San Diego, Calif.) were added. After heating at 50° C. for 15 minutes, the solution was cooled, and 5 μl of 100 mM iodoacetic acid (Sigma Chem. Co., St. Louis, Mo.) were added. After 15 minutes, the solution was diluted to a concentration of 2 M urea and subjected to trypsin digestion (Boehringer-Mannheim, Indianapolis, Ind.) for 22 hours at 37° C. using an enzyme:substrate ratio of 1:25 (wt/wt). Tryptic peptides were separated by narrowbore, reverse-phase HPLC on a Hewlett-Packard 1090 HPLC, equipped with a 1040 diode array detector, using a Vydac 2.1 mm×150 mm C18 column. A gradient of 5% B at 0 minutes, 33% B at 63 minutes, 60% B at 95 minutes, and 80% B at 105 minutes, with a flow rate of 150 μl/minute, was used. In this gradient, buffer A was 0.06% trifluoroacetic acid/H2O, and buffer B was 0.055% trifluoroacetic acid/acetonitrile. Chromatographic data at 210 and 277 nm, and UV spectra from 209 to 321 nm, of each peak were obtained. Samples for amino-terminal sequence analysis were applied to a polybrene precycled glass fiber filter and subjected to automated Edman degradation (Harvard. Microchemical Facility, Boston, Mass.) on an ABI model 477A gas-phase protein sequencer (program NORMAL 1). The resulting phenylthiohydantoin amino acid fractions were manually identified using an on-line ABI Model 120A HPLC and Shimadzu CR4A integrator.
Trypsin digestion of purified PEDF and amino acid analysis of the resulting fragments yielded nonoverlapping peptide sequences, including the sequences JT-3 (SEQ ID NO:6):
This example describes the construction of oligonucleotides, based on the peptide sequences of Example 1, the use of the oligonucleotides in the isolation of PEDF cDNA, and the sequencing of PEDF cDNA.
Based on the JT-3 and JT-8 peptide sequences of Example 1 and codon usage data, the oligonucleotides oFS5665 (SEQ ID NO:4): 5′-AGYAAYTTYTAYGAYCTSTA-3′ and oFS5667 (SEQ ID NO:5): 5′-CTYTCYTCRTCSAGRTARAA-3′ were constructed on an ABI 392 DNA/RNA Synthesizer and used as primers in a polymerase chain reaction (PCR).
A human fetal eye Charon BS cDNA library (obtained from Dr. A. Swaroop of the Kellog Eye Institute) was amplified once (Sambrook et al., Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Press, Cold Spring Harbor, N.Y. (1989)) and screened by PCR (Friedman et al., Screening of λgt11 Libraries, In: PCR Protocols: A Guide to Methods and Applications, Innis et al., eds., Academic Press, NY (1990), pp. 253-260) using a Techne thermal cycler and standard reagents (GeneAMP, Perkin-Elmer Cetus), except that MgSO4 was used at 3 mM. A PCR amplification fragment of about 350 bp was isolated on a 3% NuSieve 3:1 gel (FMC Biochemicals, Rockland, Me.) using NA-45 DEAE-cellulose paper (Schleicher and Scheull) (Sambrook et al., supra). The fragment was labeled with α32P-dCTP (Amersham Corp., Arlington Heights, Ill.) by random priming (Random Priming kit, Boehringer-Mannheim, Indianapolis, Ind.), and used to screen 200,000 plaque-forming units (PFUs) of the human fetal eye library.
Eight positive clones were isolated (Sambrook et al., supra), and DNA of the positive clones was purified according to Qiagen Maxi preparation protocols (Qiagen, Inc., Chatsworth, Calif.). The inserts of the positive clones were cut out with Not I (BRL, Gaithersburg, Md.), circularized with T4 DNA ligase (New England Biolabs, Beverly, Mass.), transformed into Escherichia coli Epicurian Sure competent cells (Stratagene, Inc., La Jolla, Calif.), and plated onto Luria broth (LB) plates containing ampicillin and 5-bromo-4-chloro-3-indolyl-β-D-galactoside (X-gal).
White colonies were selected on the basis that such colonies should possess an insert, and plasmid DNA from single colony cultures were isolated by the Qiagen plasmid miniprep protocol. Purified plasmids were digested with EcoR I and Hind III (BRL). These restriction sites were added during library construction through the ligation of linkers to the 5′ and 3′ ends of the insert, thus EcoR I-Hind III digestion excises the insert present in isolated plasmids. These fragments were electrophoresed on a 0.7% agarose gel to determine insert size. The plasmid possessing the largest insert, namely πFS17, was selected for mapping and subsequent sequencing using the Sequenase 2.0 sequencing kit (United States Biochemical Corp., Cleveland, Ohio) to confirm the identity of the clone. Sequence analysis was performed using the MacVector software package (International Biotechnologies, Inc.) and the GenBank® Sequence Data Bank (Intelligenetics, Mountain View, Calif.).
Sequence analysis of πFS17 revealed a base sequence comprising SEQ ID NO:1, with a long, open reading frame (ORF) encoding the 418 amino acids of SEQ ID NO:2, a typical ATG start codon, and a polyadenylation signal (not shown in SEQ ID NO:1). The coding sequence of the clone aligns exactly with all previously determined PEDF peptide sequences. The deduced amino acid sequence also contains a stretch of hydrophobic amino acids that could serve as a signal peptide. A comparison of the coding sequence and peptide sequence with the GenBank® Data Bank indicates that PEDF is a unique protein having significant homology to the serpin (serine protease inhibitor) gene family, which includes human [α]-1-antitrypsin. Although some of the members of this gene family exhibit neurotrophic activity (Monard et al. (1983) Prog. Brain Res., 58, 359-364; Monard (1988) TINS, 11, 541-544), PEDF lacks homology to the proposed consensus sequence for the serpin reactive domain.
EXAMPLE 3This example describes the construction of an expression vector for the production of recombinant PEDF.
An expression vector was constructed using the plasmid πFS17, which contains the full-length cDNA for human PEDF as described in Example 2. The PEDF coding sequence was placed under the control of a bacteriophage lambda PL promoter present in the plasmid pEV-vrf2 (Crowl et al., Gene, 38, 31-38 (1985)) to obtain the vector pEV-BH. This was accomplished by obtaining a BamH I-Hind III fragment of πFS17 comprising a portion of the PEDF coding region (namely, nucleotide 245 to 1490 of SEQ ID NO:1), digesting plasmid pEV-vrf2 with EcoR I-Hind III, rendering both fragments blunt by means of a fill-in reaction at the BamH I and EcoR I ends with DNA polymerase I (Klenow fragment), and ligating the resultant blunt-ended/compatible-ended fragments to each other. The resultant vector pEV-BH places a distance of 8 nucleotide between the Shine-Dalgarno (SD) sequence and the PEDF coding region. The construct specifies Met-Asn-Arg-Lle-Asp44 - - - Pro418 such that a protein of 379 amino acids, known as rPEDF, is encoded as indicated in SEQ ID NO:3. The amino acids at the amino terminus of the rPEDF protein do not occur in native PEDF and result from the fusion of nucleic acids during the construction of pEV-BH.
To verify production of the recombinant PEDF protein by pEV-BH, the plasmid was propagated in E. coli strain RRI (Maniatis et al. (1982) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), bearing the low copy-number compatible plasmid pRK248cIts that contains a gene for encoding a temperature-sensitive λcIAt2 repressor (Bernard et al. (1979) Methods in Enzymology, 68, 482-492). Protein induction was performed as described in Becerra et al. (1991) Biochem., 30, 11707-11719, with the following modifications. Bacterial cells containing pEV-BH were grown in LB medium containing 50 μg/ml ampicillin at 32° C. to early logarithmic phase, such that OD600nm=0.2. The temperature of the culture was rapidly increased to 42° C. by incubating the flask in a 65° C. water bath, and the bacteria were subsequently grown at 42° C. for 2-3 hours in an air-flow incubator at 340 rpm. Aliquots were taken for absorbance readings at 600 nm.
Nascent proteins, synthesized following protein induction, were radiolabeled. After the temperature of the culture had reached 42° C., 150 μCi of L-[35S]methionine (1040 Ci/mmol, Amersham Corp., Arlington Heights, Ill.) were added per ml of culture, and incubation was continued at 42° C. for 10 minutes and 30 minutes. Cells were harvested by centrifugation and washed with TEN buffer (10 mM Tris-HCl, pH 7.5, 1 mM EDTA, and 100 mM NaCl). 35S-labeled peptides from total bacterial extracts were resolved and analyzed on SDS-12% PAGE followed by fluorography. A band corresponding to a 42,820 Mr polypeptide was detected 10 and 30 minutes post-induction. The size obtained for the recombinant protein expressed by pEV-BH matched the expected size for the coding sequence subcloned in pEV-BH. In a similar manner, smaller fragments (BP=28,000 Mt; BX=24,000 Mt; BA=9,000 Mt) can be synthesized and purified. BP peptide includes PEDF amino acids 44 through 269, BX peptide includes PEF amino acids 44 through 227, and BA peptide includes PEDF amino acids 44 through 121.
EXAMPLE 4This example describes the construction of expression vectors containing the full-length PEDF cDNA.
In a manner similar to that described in Example 3 for the construction of pEV-BH, the PEDF ORF of plasmid πFS17 was placed under the control of the bacteriophage lambda PL promoter present in the plasmids pRC23 and pEV-vrf1 (Crowl et al. Gene, 38, 31-38 (1985)). This was accomplished by obtaining the SfaN I-Hind III fragment of πFS17 comprising a portion of the PEDF cDNA (namely, nucleotide 107 to 1490 of SEQ ID NO:1), digesting the plasmids with EcoR I-Hind III, rendering the fragments blunt by means of a fill-in reaction at the SfaN I and EcoR I ends with DNA polymerase I (Klenow fragment), and ligating the resultant blunt-ended/compatible-ended fragments to each other. The resulting vectors pRC-SH and pEV-SH place a distance of 14 and 8 nucleotide, respectively, between the SD sequence and the PEDF coding region. The construct pRC-SH encompasses the full-length PEDF ORF, and specifies a PEDF protein of 418 amino acids, with its naturally occurring amino terminus, as set forth in SEQ ID NO: 2. The construct pEV-SH encompasses the full-length PEDF ORF, and specifies a PEDF amino-terminal fusion protein of 425 amino acids, with Met-Asn-Glu-Leu-Gly-Pro-Arg (SEQ ID NO:8) preceding the PEDF sequence of SEQ ID NO:2. These additional amino acids at the amino terminus do not occur in native PEDF, and the codons in pEV-SH specifying these additional amino acids result from the fusion of nucleic acids during the construction of pEV-SH.
To verify production of the recombinant proteins specified by the two vectors, the vectors were introduced into E. coli strain RRI [pRK248cIts], and protein induction was performed and monitored by metabolic labeling with 35S-methionine during induction in a manner similar to that set forth in Example 3. The induced expression of the proteins specified by pRC-SH and pEV-SH had a negative effect on bacterial cell growth. In comparison with bacterial cultures containing the parental plasmids, cultures containing pRC-SH and pEV-SH grew and divided more slowly. This negative effect on bacterial growth correlated with the distance between the initiation codon and the SD, which may suggest that a shorter such distance results in more efficient translation of the recombinant protein. A 46,000 candidate polypeptide for PEDF was not detected in the media or cell lysates of bacterial cultures containing pRC-SH and pEV-SH. However, a 35,000 Mr protein was observed in extracts of cultures containing pRC-SH and pEV-SH, but not in extracts of cultures containing parental plasmids. This may indicate that the amino-terminal end of PEDF is protease-sensitive and that recombinant full-length PEDF is metabolized in this particular host. Alternatively, failure to observe the anticipated-sized recombinant PEDF proteins may reflect an experimental artifact which could be overcome through the use of alternative expression vectors, hosts, inducible promoters, subcloning sites, methods of recombinant protein isolation or detection, or means of protein induction.
EXAMPLE 5This example describes a method for producing large quantities of recombinantly produced PEDF.
A total of 1 g of E. coli cells containing rPEDF was resuspended in 50 ml 20 mM-Tris-HCl, pH 7.5, 20% sucrose, and 1 mM EDTA. The cells were maintained on ice for 10 minutes, sedimented by centrifugation at 4000×g, and were resuspended in 50 ml of ice-cold water for 10 minutes. Lysed outer cell walls were separated from spheroplasts by centrifugation at 8000×g.
The pelleted spheroplasts were resuspended in 10 ml of phosphate buffered saline (PBS) containing 5 mM EDTA, 1 μg/ml pepstatin and 20 μg/ml aprotinin. The suspension was probe-sonicated with a sonicator (Ultrasonics, Inc., model W-225) to lyse the cell membranes. Three bursts at 30 second pulses with a 30 second pause were performed while the sample was immersed in an ice-water bath. RNase TI (1300 units, BRL) and DNase I (500 μg, BRL) were added to the sonicated cell suspension, and the suspension was incubated at room temperature for 10 minutes. This suspension was diluted by the addition of 40 ml of phosphate buffered saline (PBS) containing 5 mM EDTA, 1 μg/ml pepstatin and 20 μg/ml aprotinin, and the crude inclusion bodies were sedimented by centrifugation at 13,000×g for 30 minutes. The particulate material consisting of inclusion bodies was resuspended in 40 ml of PBS containing 25% sucrose, 5 mM EDTA, and 1% Triton X-100, incubated on ice for 10 minutes, and centrifuged at 24,000×g for 10 minutes. The washing step was repeated three times. Finally, the inclusion bodies were resuspended in 10 ml of denaturation buffer containing 50 mM Tris-Cl, pH 8.0, 5 M guanidine-Cl, and 5 mM EDTA. The suspension was probe-sonicated briefly for 5 seconds in an ice-water bath. The resulting suspension was incubated on ice for an additional hour. After centrifugation at 12,000×g for 30 minutes, the supernatant was added to 100 ml of renaturation buffer containing 50 mM Tris-Cl, pH 8.0, 20% glycerol, 1 mM DTT, 1 μg/ml pepstatin, and 20 μg/ml aprotinin, and stirred gently at 4° C. overnight to renature the protein. The soluble and insoluble fractions were separated by centrifugation at 13,500×g for 30 minutes.
The soluble fraction was further purified by concentrating it to 1 ml using a Centricon 30 microconcentrator (Amicon Div., W.R. Grace & Co., Beverly, Mass.), and dialyzing it against Buffer A (50 mM sodium phosphate, 1 mM DTT, 20% glycerol, 1 mM EDTA, 1 μg/ml pepstatin, and 1 mM benzamidine) at 4° C. for 3 hours. The dialyzed extract was centrifuged at 14,000 rpm in an Eppendorf Centrifuge (Model 5415C) for ten minutes. The supernatant fraction was layered on a S-Sepharose fast-flow (Pharmacia, New Market, N.J.) column (1 ml bed volume) pre-equilibrated with buffer A. The column was washed with two column-volumes of buffer A. Finally, recombinant rPEDF was eluted with a step gradient of 50, 100, 150, 200, 300, 400, 500, and 1000 mM NaCl in buffer A. Fractions of 1 ml were collected by gravity flow, and were dialyzed against buffer A. Fraction 300, containing recombinant rPEDF, was stored at −20° C. The recovery in fraction 300 was 50 μg per gram of packed cells, which represents 25% of the total protein.
Most of the rPEDF was recovered from the insoluble fraction by dissolving the fraction in 10 ml of 6M guanidinium-Cl in buffer B (50 mM Tris-Cl, pH 8.0, 1 mM DTT, 2 mM EDTA). The solution was centrifuged at 10,000×g for 5 minutes. The supernatant was layered onto a Superose-12 (Pharmacia, New Market, N.J.) column attached in tandem to a second Superose-12 column (each column 2.6 cm×95 cm) pre-equilibrated with buffer containing 4 M guanidinium-Cl in buffer B. The flow rate was 3 ml/minute. Recombinant rPEDF containing fractions from the Superose-12 column were pooled and dialyzed against buffer C (4 M urea, 50 mM sodium phosphate, pH 6.5, 1 mM benzamidine, 1 μg/ml pepstatin, 4 mM EDTA). The dialyzed fraction was passed through a 0.22 μm filter (Miller-GV, Millipore Corp., Bedford, Mass.). The filtered solution was layered onto a mono-S (Pharmacia, New Market, N.J.) column (1 cm×10 cm, d×h) pre-equilibrated with buffer C. The column was washed with buffer C, and recombinant rPEDF was eluted with a gradient of 0 mM-500 mM NaCl in buffer C at 0.5 ml/min. Two-ml fractions were collected, and the peak fractions of recombinant rPEDF were pooled. The recovery in the pooled fractions was 0.5 mg of recombinant PEDF per gram of packed cells.
EXAMPLE 6This example describes the use of purified recombinant PEDF as a differentiation agent.
Y79 cells (ATCC, HTB18) were grown in Eagle's Minimal Essential Medium with Earl's salts (MEM) supplemented with 15% fetal bovine serum and antibiotics (10,000 u/ml penicillin and 10 mg/ml streptomycin) at 37° C. in a humidified incubator under 5% CO2. Cells were propagated for two passages after receipt from the ATCC, and then frozen in the same medium containing 10% DMSO. A few of the frozen aliquots were used for each differentiation experiment. All experiments were performed in duplicate.
After thawing, the cells were kept, without further passaging, in the serum-containing medium until the appropriate number of cells were available. Cells were collected by centrifugation and washed twofold in PBS, resuspended in PBS, and counted. At that point, 2.5×105 cells were plated into each well of a 6-well plate (Nunc, Inc., Roskilde, Denmark) with 2 ml of serum-free medium (MEM, supplemented with 1 mM sodium pyruvate, 10 mM HEPES, 1× non-essential amino acids, 1 mM L-glutamine, 0.1% ITS mix (5 μg/ml insulin, 5 μg/ml transferrin, 5 ng/ml selenium, Collaborative Research, Bedford, Mass.), and antibiotics as described above.
Differentiation effectors and control buffers were added 12-16 hours after plating, and the cultures were incubated and left undisturbed for 7 days. On the eighth day, cells were transferred to poly-D-lysine-coated six-well plates (Collaborative Research, Bedford, Mass.), and the old medium was replaced with 2 ml of fresh serum-free medium, upon attachment of the cells to the substrate. The cultures were maintained under these conditions for up to 11 days. Post-attachment cultures were examined daily for morphological evidence of differentiation as well as quantification of neurite outgrowth using an Olympus CK2 phase-contrast microscope.
In comparison with untreated cells, only Y79 cultures that were exposed to recombinant rPEDF showed any significant evidence of neuronal differentiation. Some neurite outgrowth (below 5%) was detectable in control cultures treated with the same buffer used to solubilize rPEDF, and no evidence of differentiation was found in cultures processed in the same manner without the addition of rPEDF or buffer (
The PEDF and rPEDF cDNA clones not only provide means to produce large quantities of the PEDF and rPEDF proteins but also serve as sources for probes that can be used to study the expression and regulation of the PEDF gene. In addition, these sequences can be used in the antisense technique of translation arrest to inhibit the translation of endogenous PEDF.
The recombinantly produced PEDF and rPEDF proteins and equivalent proteins can be used as potent neurotrophic agents in vitro and in vivo. Additional biochemical activities of these proteins as neurotrophic agents can be determined through standard in vitro tests, which will enable the development of other therapeutic uses for these proteins in the treatment of inflammatory, vascular, degenerative and dystrophic diseases of the retina. Given that these proteins are such potent neurotrophic agents, it can be envisioned that these proteins could be modified for therapeutic utility in the treatment of tissues other than the retina, which also respond to neurotrophic factors. These proteins may even find more generic utility as “differentiation” factors for non-neural tissues and certain types of cancer.
EXAMPLE 7Along with the 3,000 mol. wt. recombinant PEDF, smaller recombinant constructs have been synthesized to determine if they have neurotrophic activity. Smaller peptides could offer a variety of advantages over the full-length construct such as greater solubility, better membrane penetration, less antigenicity, greater ease in preparation, etc.
Table 1: Exons are in upper case and introns sequences in lower case. The 5′ donor GT and 3′ acceptor AG are underlined. Exon and intron sizes are given in bp and kb respectively.
EXAMPLE 8 Cloning and Sequencing of the Human PEDF GeneMaterials—Restriction enzymes, SuperScripts RT and Kanamycin were purchased from GIBCO-BRL (Gaithersburg, Md.). Dynabeads® Oligo dT(25) were purchased from Dynal Inc. (Lake Success, N.Y.). Retrotherm™ RT was obtained from Epicentre Technologies (Madison, Wis.). RNAsin® was purchased from Promega (Madison, Wis.). Taq polymerase was purchased from Perkin-Elmer (Norwalk, Conn.), or Stratagene (La Jolla, Calif.). The plasmid vector pBlueScript® used for subcloning was purchased from Stratagene (La Jolla, Calif.). Total RNA from neural retina and retinal pigment epithelium was purified from human tissue obtained from the National Disease Research Interchange (NDRI, Philadelphia, Pa.) as previously described (Chomczynki and Sacchi, 1987). [32P]α-dATP and [32P]γ-ATP (3000 Ci/mmol) used for labeling and sequencing (respectively) were purchased from Amersham) Arlington Hts, Ill.). Superbroth (Bacto-Tryptone 12 g/L, yeast extract 24 g/L, K2 HPO4 12.5 g/L, HK2PO4 3.8 g/L and glycerol 5 mL/L), denaturing solution (0.2 N NaOH, 1.5 M NaCl), neutralizing solution (1 M Tris-Cl pH 7.0, 1.5 M NaCl), 20×. SSC (3.0 M NaCl, 0.3 mM sodium citrate), 10×TBE (1 M Tris-borate, 2 mM EDTA, pH 8.3), and 50×TAE (2 M Tris-acetate 50 mM EDTA, pH 8.0) were purchased from Quality Biologicals (Gaithersburg, Md.). 20×SSPE (3M NaCl, 0.2 M NaH2PO4, 20 mM EDTA pH 7.4) was purchased from Digene Diagnostics, Inc. (Silver Spring, Md.). Ampicillin was purchased from Sigma Chemical Co. (St. Louis, Mo.) dissolved in water and filter-sterilized.
Polymerase chain reaction (PCR). A 2×PCR mix was prepared containing 1.6 μmoles/mL of GeneAmp® dNTPs (400 μM each), 2×GeneAmp® PCR buffer and 50 U/mL Taq polymerase. These reagents were purchased from Perkin-Elmer (Norwalk, Conn.). In general, the template and oligonucleotides (100 ng of each oligo) were mixed in 25 μL volume and 25 μL of the 2× mix were then added followed by 50 μL of mineral oil. The template was initially denatured for 2 min at 95° C., 30 sec annealing (temperature between 55 and 65° C. depending on the primers) and an extension at 72° C. for 1-5 min depending on the length of the product amplified.
cDNA synthesis on Dynabeads® oligo (dT)25. The cDNA was synthesized on Dynabeads as previously described (Rodriguez and Chader 1992). The Dynabeads (0.5 mg) were washed with 100 μL of 10 mM Tris-Cl pH 7.0, 1 mM EDTA, 1 M KCl. The total RNA 30 μL, (30 μg, ˜1 μL), in water was mixed with 30 μL of the above buffer and the equilibrated Dynabeads (0.5 mg) then heated to 55° C. for 2 minutes. The poly+ A RNA was allowed to anneal to the beads for 15 min at room temperature and the excess RNA removed by binding the beads for 15 min at room temperature and the excess RNA removed by binding the beads to the MPC-E magnetic separator (Dynal Inc.). The beads with the annealed poly+ A mRNA were then suspended in 2.5 μL buffer A (200 mM Tris-Cl pH 8.3, 1.0 M KCl), 2.5 μL buffer B (30 mM MgCl2, 15 mM MnCl), 20 μL 10 mM dNTP's (2.5 mM each), 1 μL RNAsin, 2 μL SuperScript RT, 5 μL of Retrotherm RT (1 Unit/μI) and 16 μL of H2O to make a final volume of 50 μL. The reaction mixture was incubated at 40° C. for 10 min, than at 65° C. for 1 hr. The beads were again bound to the MPC-E magnetic separator and the excess RT reaction mix removed. The beads were then washed once with 100 μL 0.2N NaOH, once with 10×SSPE, and twice in 1×TE. The cDNA-containing beads were suspended in a final volume of 100 μL 1×TE.
5′ Rapid Amplification of cDNA Ends (RACE). The 5′-RACE was performed using a modified method based on the 5′-AmpliFINDER RACE kit purchased from Clontech (Rodriguez et al. 1994). First, cDNA was synthesized on Dynabeads® Oligo dT(25) as described above (Rodriguez and Chader, 1992). The AmpliFINDER anchor primer (Clontech) was ligated to the 3′ ends tips of the Dynabead-immobilized retinal pigment epithelium cDNA using the same conditions as for soluble cDNA described in the 5′-AmpliFINDER RACE kit. The AmpliFINDER anchor primer was used in combination with an PEDF-specific primer #2744 to PCR amplify the 5′ prime end. The amplification was done as described above with 2 μL of anchor-ligated human retinal pigment epithelium-Dynabeads cDNA used as template. The amplification was performed for 30 cycles.
Sequence of oligonucleotides. Oligonucleotide primers were synthesized in an Applied Biosystems Inc. (Foster City, Calif.) DNA synthesizer model 392. The oligonucleotides were deprotected and used without further purification.
Screening of genomic libraries. The human genomic cosmid library (Clontech) was plated on LB plates containing 150 mg/mL ampicillin, 20 mg/mL Kanamycin at a density of 10,000 colonies per plate. Nitrocellulose filters were used to lift the colonies and the filters were treated and hybridized as described in Sambrook et al., (1989). The library was probed with [32P]-labeled PCR product obtained from amplifying a PEDF cDNA clone (Steele et al. 1993) using T7/T3 primers. This resulted in the isolation of the p10A cosmid. A λDASH™II library (Stratagene) was screened by Lark Sequencing Technologies Inc. (Houston, Tex.) using the insert from the PEDF cDNA clone mentioned above. This resulted in the isolation of the 7 Kb NotI-Not fragment (JT6A). A P-1 clone, p147, containing the entire PEDF gene and flanking regions was isolated using oligos 1590/1591 by Genome Systems (St. Louis, Mo.).
Cloning of PCR products: Four sets of primers, 603:604; 605:606; 2238:354 and 2213:2744 designed from the internal coding regions of the PEDF cDNA sequenced were synthesized as decribed above for use as primers in a polymerase chain reaction (PCR) experiments. The primer sequences are as follows: 603: 5′-ACA AGC TGG CAG CGG CTG TC-3′ (SEQ ID NO: 13), 604: 5′-CAG AGG TGC CAC AAA GCT GG-3′ (SEQ ID NO: 14); 605: 5′-CCA GCT TTG TGG CAC CTC TG-31 (SEQ ID NO: 15), 606: 5′-CAT CAT GGG GAC CCT CAC GG-3′ (SEQ ID NO: 16), 2213: 5′-AGG ATG CAG GCC CTG GTG CT-3′ (SEQ ID NO: 17), 2744: 5′CCT CCT CCA CCA GCG CCC CT-3′ (SEQ ID NO: 18); 2238: 5′-ATG ATG TCG GAC CCT AAG GCT GTT-31 (SEQ ID NO: 19), 354: 5′-TGG GGA CAG TGA GGA CCG CC-3′ (SEQ ID NO: 20). The amplifications, subcloning and sequencing of the PCR products generated with primers 603:604 and 605:606 was performed by Lark Sequencing Technologies Inc. using human genomic DNA as template. The product generated from 603:604 is −2 kb (jt8A) and expands from exon 3 to exon 5. The product generated using 605:606 is −3.3 kb (jt 9) and expands from exon 5 to exon 6. The primers set 2213-2744 was used to amplify a ˜2.5 Kb product (jt15; also referred to as JT115) from the P1 clone p147. This product was then sent to Lark Sequencing Technologies Inc. for subcloning and sequencing. The 2238:354 primers were used to amplify from exon 6 to exon 7 across intron E. This product was not subcloned but was sequenced directly and entirety by us.
DNA sequencing. The P-1 clone (p147), subclones of this clone and PCR products from this clone were sequenced. Most of the sequencing was performed by Lark Sequencing Technologies Inc. using standard sequencing techniques. All important areas (e.g. intron-exon boundaries), and junctions between clones were sequenced in our laboratory. DNA from the PCR products was prepared for sequencing using Wizard™ PCR Preps DNA purification kit purchased from Promega Corp. (Madison, Wis.). The P-1 clone, and plasmid subclones were purified using Qiagen Inc. (Chatsworth, Calif.) Midi plasmid purification kit. The purified PCR products and plasmids were sequenced using the PRISM™ DyeDeoxy Terminator Cycle Sequencing Kit (Applied Biosystems a Division of Perkin-Elmer Corp., Foster City, Calif.), following the manufacturer's protocol. Typically, 0.5 pmoles of template and 3 pmoles of primer were used per sequencing reaction. The sequencing reaction products were purified using Select-D G-50 columns (5 Prime-3 Prime; Boulder, Colo.) and dried. Each sample was then dissolved in 5 μL formamide, 1 μL 50 mM EDTA, heated and located in a Model 370A Automated Fluorescent Sequencer (ABI, Foster City, Calif.). All splice-sites junctions, intron F and junctions across clones were sequenced.
Southern blot. An EcoRI digested genomic (8 μg) blot of DNA from a variety of species was purchased from BIOS Laboratories, New Haven, Conn. The blot was probed with the PEDF cDNA using standard techniques (Sambrook et al., 1989).
5′ RACE of PEDF. The 5′ RACE was performed as described above by ligating the anchor oligo to human retinal pigment epithelium cDNA previously synthesized on Dynabeads. The 51 end was amplified using the anchor primer (AmpliFinder's kit) and the PEDF-specific primer 2744. The amplification was performed for 30 cycles. One main band was observed at ˜230 bp. The PCR products were cloned in pGEM-T (Promega Corp., Madison, Wis.) and sequenced. The longest of these clones was found to extend the 5′ end of PEDF by 20 bp.
Isolation of the PEDF gene. The PEDF gene was isolated in a P-1 clone (p147) by Genome Systems (St. Louis, Mo.) using primers 1590 and 1591 (1590: 5′-GGA CGC TGG ATT AGA AGG CAG CAA A-3′ (SEQ ID NO: 23); and 1591: 5′-CCA CAC CCA GCC TAG TCC C-3′ (SEQ ID NO: 24)). In order to determine if this clone contained the entire PEDF gene, both p147 and human genomic DNA were digested with BamHI, EcoHI, HindIII and PstI then separated by agarose gel electrophoresis in a pulse field apparatus. The agarose gel was blotted and probed with the PEDF cDNA clone (Steele et al. (1993) Proc. Natl. Acad. Sci. USA 90:1526-1530). Comparison of the band pattern between the P-1 clone and genomic DNA indicates that the entire PEDF gene is contained in this clone. Furthermore, this result is also indicative that there is only one gene for PEDF.
Sequence of the PEDF gene. A scale map of the gene is shown in
jt1: A 7.1 kb cosmid clone isolated from a human genomic cosmid library (Clontech) containing exon 7, exon 8 and the 3′ flanking region of the PEDF gene. The 5′ end of this clone, an area of approximately 2.1 Kb, is not part of PEDF. This was apparently caused by a rearrengement of the cosmid. This clone was sequenced entirely by Lark Sequencing Technologies Inc.
jt6A: This is a 7.2 kb Not I fragment isolated by Lark Sequencing Technologies Inc. from a λDASHII human genomic library (Statagene). This clone contained >6 Kb of the 5′ flanking region, exon1 and 424 bp of intron A of the PEDF gene. This clone was sequenced entirely by Lark Sequencing Technologies Inc.
jt8A: This cloned PCR product JT8A generated from genomic DNA using primers 603:604. This clones expands from exon 3 to exon 5 including exon 4 and introns C and D. It was amplified, cloned and sequenced entirely by Lark Sequencing Technologies Inc.
jt9: This cloned PCR product JT8A was generated from genomic DNA using primers 605:606. It contains the entire intron E and portions of exon 5 and exon 6. It was amplified, cloned and sequenced entirely by Lark Sequencing Technologies Inc.
jt15: This clone was obtained from a PCR product-amplified using the primer pair 2213:2744 from p147. The clone expands from exon 2 to exon 3 across intron B. The PCR product was submitted to Lark Sequencing Technologies Inc. for subcloning and sequencing.
P1 clone p147: This clone was isolated by Genome Systems Inc. using oligonucleotides 1590:1591. This clone was used to obtain the sequence of intron F (2238:354), and the subclone jt14. It was also used to confirm the intron-exon boundaries initially obtained from the above mentioned clones. All the exons and intron boundaries were amplified (using p147 as template) using intron-specific oligos and the products sequenced. All splice junctions sequences were confirmed as well as the sizes of introns and exons.
jt14: This is a subclone of p147 containing most of intron A, exon 2 and a portion of intron B. This clone was isolated by us and sent to Lark Sequencing Technologies Inc. for sequencing.
Thus from the sequence analysis of all the above mentioned clones and PCR products the structure and size of exons and introns of the human PEDF gene were determined. The 5′ splice donor and 3′ splice acceptor sites in all junctions conform to the GT/AG consensus.
EXAMPLE 9 Analysis of the PEDF Promoter In order to obtain some understanding as to the possible transcriptional elements that may regulating PEDF and guidance for future experiments on PEDF expression, we performed a theoretical analysis of the PEDF 5′ flanking region (
Gene Expression Analysis
Multiple human tissue mRNA Northern blots (Clonetech) with 2 ug Poly-(A) RNA per lane were hybridize with a radioactively-labelled 667 bp PCR amplified PEDF product (Tombran-Tink et al., 1994 Genomics, 19:266-272). Blots were prehybridized for 15 min at 68° C. in QuickHyb rapid hybridization solution (Stratagene, La Jolla, Calif.) and hybridized for 1 hr at 68° C. in the same solution containing 5×106 cpm DNA/ml. Hybridized blots were washed twice with 100 ml of 2×SSC, 0.1% SDS for 15 min at room temperature and once with 200 ml of 0.1×SSC, 0.1% SDS for 30 min at 68° C. The blots were autoradiographed at −70° C. for 2 hr using Kodax XAR-5 film and DuPont intensifying screens.
Gene Expression:
In order to determine whether expression of the PEDF messenger RNA occurs in human tissues other than in cultured human fetal RPE cells, we analyzed multiple tissue human adult and fetal RNA blots containing equal amounts of poly-(A) RNA for each tissue examined. The results are shown in
In contrast to the single 1.5 kb transcript observed in the adult tissues, an additional minor transcript of less than 500 bp is labelled variably and with lower intensity in fetal heart, lung and kidney. This may be due to partial degradation of the message or an alternative splicing phenomenon. PEDF is also only expressed in early passaged monkey RPE cells (1st-5th passage) and not in late passaged cells (10th passage). These data demonstrate the relevance of PEDF to senescence.
EXAMPLE 11 Comparative Analysis of PEDF in a Variety of Phylogenetically Related SpeciesEvolutionary Conservation Analysis
8 ug of genomic DNA from lymphocytes of a variety of species including a number of mammalian and primate species (BIOS laboratories, New Haven Conn.) was digested with Eco-R1 and separated in 1% agarose gels. The gels were transblotted and membranes containing the digested DNA hybridized using the same procedure and conditions as that for Northern analysis.
Evolutionary Conservation:
The evolutionary conservation of PEDF among a number of phylogenetically related species was examined. The results are presented in
Cell Culture
Cerebellar granule cells (CGC) were prepared from 5 or 8-day-old Sprague-Dawley rat pups as described by Novelli et al. (1988, Brain Res., 451:205-212). In brief, tissue free of meninges was minced in a buffer containing 124 mM-NaCl, 1 mM NaH2PO4, 1.2 mM MgSO4, 3 mg/ml bovine serum albumin (BSA), 27 μM phenol red, and 25 mM HEPES (pH 7.4), and centrifuged at 550×g for 3 min. The tissue pellet from 10-20 animals was resuspended and trypsinized (15 min, 37° C.) in 30 ml of the same buffer containing 250 μg/ml trypsin; a further 15 ml of buffer was added containing 26 μg/ml DNase I, 166 ug/ml soybean trypsin inhibitor, and 0.5 mM additional MgSO4 and the tissue was centrifuged again as described above. The pellet was resuspended in 1 ml of buffer supplemented with 80 μg/ml DNase, 0.52 mg/ml of trypsin inhibitor, and 1.6 mM additional MgSO4, and triturated 60 times with a Pasteur pipette. The suspension was diluted with 2 ml of buffer containing 0.1 mM CaCl2 and 1.3 mM additional MgSO4, and undissociated material allowed to settle for 5 min. The supernatant was transferred to another tube, cells were recovered by brief centrifugation and resuspended in serum-containing medium (Eagle's basal medium with 25 mM KCl, 2 mM glutamine, 100 μg/ml gentamycin, and 10% heat inactivated fetal calf serum) or chemically defined medium (DMEM:F 12 (1:1) with 5 μg/ml insulin, 30 nM selenium, 100 μg/ml transferrin, 1000 nM putrescine, 20 nM progesterone, 50 U/ml penicillin, 50 μg/ml streptomycin, and 2 mM glutamine) (Bottenstein, 1985 Cell Culture in the Neurosciences, J. E. Bottenstein and G. Sato, eds. New York Plenum Publishing Corp. p. 3-43). Cells were plated in poly-L-lysine-coated 96 well plates (for MTS assay and neurofilament ELISA assay) or 8-well chamber slides (for immunocytochemistry and BrdU labelling) at 2.5×105 cells/cm2 and grown at 37° C. in an atmosphere consisting of 5% CO2 in air. After 1 day in culture, cytosine arabinoside (Ara-C) was added only to cells in serum-supplemented medium (final concentration 50 μM).
MTS AssayCerebellar granule cells in 96 well plates were incubated in a CO2 incubator for 4 hours with MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt) and PMS (phenazine methosulfate) final concentration; 333 μg/ml MTS and 25 μM PMS) (Promega Corp.). In the presence of PMS, MTS is converted to a water-soluble formazan by a dehydrogenase enzyme found in metabolically active cells (Cory et al. (1991) Cancer Comm, 3:207-212). The quantity of formazan product was determined by spectrophotometry at 490 nm.
ImmunocytochemistryAfter 7 days in vitro (DIV), the cells were washed three times in calcium- and magnesium-free phosphate-buffered saline (PBS) and fixed with 2% paraformaldehyde for 10 min, followed by 10 min at −20° C. in 95% ethanol/5% acetic acid. Incubation with primary antibodies against NSE (neuron specific enolase), GABA, calbindin, or glial fibrillary acidic protein (GFAP) was carried out for 60 min at RT. Antibodies were applied at 1:1000-1:5000 in the presence of 2% normal goat serum and 0.2% BSA. The antibodies were visualized using the ABC system (Vector Laboratories) and diaminobenzidine. At least 20 fields were counted from 2-3 wells for each experiment. The average number of cells per field was then calculated to determine the ratio for the number of cells stained by the other antibodies relative to NSE-positive cells in control cultures.
Bromodeoxyridine (BrdU) LabelingBrdU labeling was performed by the method of Gao et al. (1991 Neuron, 6: 705-715) with the following modification. The cells were plated in 8-well chamber slides and rPEDF added immediately. After 24 hours, BrdU (1:100; Amersham cell proliferation kit) was added to the culture medium for 24 hours, after which the cells were fixed in 2% paraformaldehyde (10 min), treated with 95% ethanol/5 acetic acid (10 min), and incubated with an anti-BrdU monoclonal antibody (1:20 for 2 hrs). The cultures were then incubated with a horseradish peroxidase-conjugated goat anti-mouse secondary antibody for 60 min. After diaminobenzidine-peroxidase, the cells were mounted in Gel Mount. The mitotic index was determined by counting the percentage of labeled cells with a microscopy. For each value, a random sample of 3000 cells was counted.
Neurofilament ELISA AssayThe neurofilament ELISA was performed according to the method of Doherty et al. (1984 J. Neurochem., 42:1116-1122) with slight modification. Cultures grown in 96-well microtiter plates were fixed with 4% paraformaldehyde in PBS at 4° C. for 2 hr. The fixed cells were permeabilized by treatment for 15 min with 0.1% Triton X-100 in PBS, followed by incubation for 60 min with PBS containing 10% goat serum to block nonspecific binding. The cultures were then incubated with a monoclonal anti-neurofilament antibody overnight at 4° C. (RMO-42 at 1:100; which stains only neurites in the cultures of cerebellar granule cells). After washing twice with PBS containing 10% goat serum, cells were incubated with secondary antibody (horseradish peroxidase-conjugated goat anti-mouse at 1:1000) for 1 hr. Following sequential washing with PBS and water, the cultures were incubated with 0.2% O-phenylenediamine and 0.02% H2O2 in 50 mM citrate buffer (pH 5.0) for 30 min. The reaction was stopped by adding an equal volume of 4.5 M H2SO4. Product formation was quantitated by reading the optical density (O.D.) of an aliquot of the reaction product at 490 nm using a microplate reader.
In order to validate the MTS assay as a measure of live cells, and to determine the range of cell number over which the results would be linear, the experiments shown in.
In order to determine whether the increase in O.D. (MTS assay) in response to PEDF reflected an increase in surviving cells or an increase in proliferation, a BrdU labeling study was performed using cultures from postnatal day 5 (P5) animals (a time when cerebellar granule cells are still dividing in the animal).
Immunocytochemistry was used to identify the cells present in cultures before and after treatment with PEDF. P8 cultures grown for 7 days with and without PEDF (500 ng/ml) were stained with four different antibodies: a polyclonal rabbit antibody to neuron-specific enolase (NSE), which recognizes all cerebellar neurons (Schmechel et al. (1978) Science, 199:313-315); a polyclonal antibody to GABA, which is synthesized in all cerebellar neurons except cerebellar granule cells (Gruol and Crimi (1988) Dev. Brain Res., 41:135-146); an antibody to calbindin, which is a neuron-specific protein and GFAP, an intermediate filament protein present only in astrocytes. The results are summarized in Table 2. PEDF significantly increased the number of NSE-positive cells in both SCM (30% increase) and in CDM (60% increase). There was a small, not statistically significant, increase in the number of GABA-positive neurons and Purkinje cells (calbindin-positive). Thus, PEDF is neurotrophic only for granule neurons. In addition, PEDF significantly decreased the number of GFAP-positive astrocytes present in the cultures (30% decrease in SCM and 40% decrease in CDM). This “gliastatic” property of PEDF is further discussed in Example 14.
Postnatal-day 8 cerebellar granule cells were cultured in 8-well chamber slides. PEDF (500 ng/ml) was added at DIV 0, the cells were fixed on DIV 7, and the immunocytochemistry was carried out using antibodies against NSE, GABA, Calbindin and GFAP. At least 20 fields were counted from 2-3 wells for each experiment. Data are expressed as percent of control of NSE-positive cells. Each experiment value represents mean cell number ± SEM.
*P < 0.005 compared with each other control by using non-paired test.
In order to investigate the effects of PEDF on neurite outgrowth, a neurofilament ELISA assay was used. Immunocytochemistry had shown that the monoclonal antibody RMO-42, stained only the neurites of cerebellar granule cells in culture, so this antibody was used as a direct measure of neurofilament present only in processes and not the cell body (
Described in the previous sections on the “neuronotrophic” activity of PEDF is the fact that we can produce relatively large amounts of a recombinant PEDF (rPEDF) that exhibits potent neurotrophic activity. Using appropriate recombinant molecular biological technology, we can also produce smaller fragments of the PEDF molecule that can be tested for either neurotrophic or neuronotrophic activity.
Along with neurons in the primary cultures of rat cerebellar granule cells are a small number of different types of glia. Glia are the “support” elements in the CNS for neurons, forming the architectural framework and the metabolic support system on which neurons depend. Glia are also of clinical importance since tumors of the brain are mostly formed by glia and gliosis is a problem in several neurodegenerative diseases. In our system, we first noticed an effect of PEDF on glia when we immunocytochemically stained the cultured mixed population of cells with antibodies specific for neurons and other antibodies specific for different types of glia. For this purpose, we used the standard markers Neuron-Specific Enolase (NSE) and others to demonstrate the presence of neurons, Glial Fibrillary Acidic Protein (GFAP) to demonstrate the presence of astroglia and OX-42 to stain microglia. In this experiment (Table 2), we found the expected increase in NSE staining with PEDF treatment since we then knew that the neurons were living longer but we found an unexpected decrease in GFAP staining. This indicated the possibility of fewer astrocytes in the PEDF-treated cultures.
Because of the distinctive morphology of astroglia and microglia in the culture dishes and their selective staining for GFAP or OX-42, it is possible to individually count their numbers under the microscope under different experimental conditions. This has now been done as outlined in
Since the specific antibody indicated the presence of PEDF in the adult IPM, we used bovine IPM washes as a source for purification of native PEDF. Although RPE and retinal cells express PEDF mRNA, anti-BH could not detect PEDF bands on Western transfers in these cell extracts, suggesting a rapid PEDF release into the IPM. We now estimate that PEDF is present in bovine IPM at less than 1% of the total soluble protein (i.e. about 2-5 ng/bovine eye). At physiological temperatures, the PEDF protein in the IPM remains stable for extended periods of time and does not form non-reduced complexes resistant to SDS. Thus, its potential usefulness in culture experiments and transplantation in vivo. is greatly enhanced due to its stable nature.
Purification to apparent homogeneity is achieved by a simple two-step procedure (
Starting with a wash of soluble IPM components, the first step involves removal of the most abundant protein, IRBP, by size exclusion chromatography. PEDF elutes as a monomeric polypeptide around 50 kDa in size. Since we have determined that PEDF's isoelectric point is 7.2-7.8, we have used S-sepharose column chromatography at pH 6.0 in the second step of our procedure to simultaneously purify and concentrate the protein. Purified protein is recovered at about 2 ug protein per adult bovine eye with a recovery of about 40%. Native PEDF behaves like a monomeric glycoprotein with an apparent molecular weight of 49,500±1,000 on SDS-PAGE.
The purified protein is sensitive to glycosidase F, revealing N-linked oligosaccharides that account for up to 3,000-Mr of the native protein (
The present invention provides the tools for determining the effect of authentic PEDF on the expression of neuronal and glial markers in the CGC cultures and Y-79 tumor cells including NSE, GFAP, neurofilament (NF-200) protein.
EXAMPLE 16 Pigment Epithelium-Derived Factor: Characterization Using a Highly Specific Polyclonal Antibody We have used purified recombinant human PEDF produced in E. coli to develop polyclonal antibodies against PEDF. Anti-rPEDF specifically recognized one polypeptide on Western transfer of IPM wash from adult bovine eyes (
Anti-BH specifically recognizes human PEDF on Western transfers at dilutions as low as 1:50,000, importantly, it does not recognize serum α-antitrypsin. The antibody recognizes one major band on Western transfers of conditioned medium from juvenile monkey RPE cells in culture as well as of IPM from adult bovine eyes. Anti-rPEDF blocked the IPM-promoting neurotrophic activity (
The anti-rPEDF also blocked the neurite-outgrowth activity promoted by the purified PEDP. Our data indicate that PEDF is the only neurotrophic factor in the IPM. These results also suggest that the anti-rPEDF will be useful in probing the PEDF neurotrophic active site as well as the physiological role of PEDF in the IPM and other tissues (e.g. brain) as well. Further, these results indicate that PEDF is a bona fide component of the IPM and is probably the sole neurotrophic component in the extracellular matrix. Moreover, the protein is present in a wide range of tissues and extracellular spaces. The blocking antibody is useful in studies probing the physiological functions of PEDF.
EXAMPLE 17 Pigment Epithelium-Derived Factor: A Serpin with Neurotrophic ActivityThe amino acid sequence derived from a fetal human PEDF cDNA shares identity of its primary structure (˜30%) with the serine protease inhibitor (serpin) family, preserving 90% of the residues essential for the structural integrity of serpins. However, recombinant PEDF does not inhibit the serine proteases trypsin, chymotrypsin, elastase or cathepsin G. A natural target for PEDF has not yet been identified. We have analyzed proteins from the interphotoreceptor matrix (IPM), the space between the retinal pigment epithelium and the retina by immunodetection on western blots with antibodies raised against PEDF and by zymography in gels containing casein as a proteolytic substrate. Our results show that bovine IPM contains a stable, glycosylated PEDF polypeptide (50,000 Mr) at about 2-5 μg per eye. Limited proteolysis of bovine PEDF produced a polypeptide of 46,000 Mr with trypsin, subtilisin, chymotrypsin and elastase, suggesting a globular structure with a hinge region susceptible to proteolytic cleavage. On the other hand, casein SDS-PAGE zymography revealed low protease activity in the IPM which migrated as a double of about 80,000±5,000 Mr. The caseinolytic activities were inhibited 100% with 1 μg/ml aprotinin and 10 mM PMSF added to the gel mixture, but were not affected by E64 or EDTA. Importantly, IPM protein did not react with antibody against plasminogen, a serine protease of about 80,000 Mr. When rPEDF protein was added at 1 μg/ml, the signal for these caseinolytic activities, as well as another serine protease activity of unknown origin, diminished by about 50%. Our results suggest the IPM as a natural extracellular site for a novel serine protease and the serpin PEDF, both present at ≦1% of the total protein.
All of the references cited herein are hereby incorporated in their entireties by reference.
The present invention discloses the general structural features of PEDF and beginnings of understanding of how these relate to function of the protein. PEDF possesses the structural features and general tertiary characteristics previously attributed to serpins but not its anti-protease activity. PEDF is a neurotrophic protein and appears to be the sole component of the IPM that promotes neurite-outgrowth on retinoblastoma cells. However, the reactive center for serine protease inhibition found near the carboxy terminal of classical serpins is not necessary for PEDF's neurotrophic biological activity. Specifically, a polypeptide chain containing a domain from the amino-terminal portion of the molecule (BA) is sufficient for neurotrophic and neuron-survival activity. The present invention further allows for determination of whether the CGC neurons normally die by apoptosis and whether PEDF is an apoptosis inhibitor. In other words, the present invention allows one to determine by what mechanism PEDF “saves” neurons and “inhibits” glia growth or proliferation.
The present invention is useful in determining the specific neurotrophic “active site”. Further, the use of rPEDF truncated peptides allows us to define the elements necessary for neuronotrophic and perhaps gliastatic activity of PEDF. The present invention further provides necessary tools to study the interactions of PEDF that trigger the signal for differentiation of retinoblastoma. Recent experiments demonstrate that 125I-BH binds to retinoblastoma cells in competitive fashion only when added in medium that had been previously “conditioned” by retinoblastoma cells. This suggests that one or more co-factors produced by the cells could be required for binding. The present invention further provides the tools necessary to identify and characterize a putative cell-surface receptor for PEDF or for a PEDF complex from our CGC and retinoblastoma test systems.
Recombinant mutated proteins, proteolytic products and synthetic peptides have become instrumental in domain mapping of functional sites of proteins. Further, the recombinant proteins of the present invention allow the mapping of neurotrophic and neuronotrophic “active sites” on the PEDF molecule and the determination of the cellular transduction mechanism through which this interesting protein exerts its dramatic biological effects.
While this invention has been described with an emphasis upon preferred embodiments, it will be obvious to those of ordinary skill in the art that variations in the preferred nucleic acids coding for, and the amino acid sequences of, PEDF, rPEDF, and equivalent proteins, (BP, BX, BA) the vectors utilizing any such nucleic acids, the recombinant methods of producing such proteins, and the methods of using such proteins, may be realized and that it is intended that the invention may be practiced otherwise than as specifically described herein. Accordingly, this invention includes all modifications encompassed within the spirit and scope of the invention as defined by the following claims.
Claims
1. An isolated and purified nucleic acid selected from the group consisting of nucleic acids comprising SEQ ID NO:1, SEQ ID NO:43 and conservatively modified variants thereof.
2. A protein selected from the group consisting of proteins comprising SEQ ID NO:3 and conservatively modified variants thereof.
3. An isolated and purified nucleic acid which encodes a protein selected from the group consisting of proteins comprising the sequence of SEQ ID NO:3 and conservatively modified variants thereof.
4. A vector which comprises a nucleic acid molecule of claim 1.
5. A vector according to claim 4, wherein the nucleic acid comprises SEQ ID NO:1.
6. A vector according to claim 4, wherein the nucleic acid comprises SEQ ID NO:43.
7. A vector which comprises the nucleic acid of claim 2.
8. A vector which comprises the nucleic acid of claim 3.
9. A host cell containing the vector of claim 5.
10. A host cell containing the vector of claim 6.
11. A host cell containing the vector of claim 7.
12. A host cell containing the vector of claim 8.
13. A method of producing the protein encoded by a nucleic acid of SEQ ID NO:1 and conservatively modified variants thereof, said method comprising:
- expressing said protein in a host cell of claim 9; and
- recovering said protein from the host cell.
14. A method of producing a protein comprising SEQ ID NO:2 and conservatively modified variants thereof, said method comprises:
- expressing said protein in a host cell of claim 9; and
- recovering said protein from the host cell.
15. A method of producing a protein comprising SEQ ID NO:3 and conservatively modified variants thereof, said method comprises:
- expressing said protein in a host cell of claim 11; and
- recovering said protein from the host cell.
16. A recombinantly produced PEDF protein, which has been produced in accordance with the method of claim 14 and which is substantially free from proteins normally associated with a naturally occurring source organism.
17. An isolated and purified human genomic DNA molecule which encodes a protein having a sequence of SEQ ID NO:2.
18. The method of claim 14, wherein said method further comprises lysing said host cell and then isolating said protein.
19. The method of claim 18, wherein said method further comprises purifying said protein.
20. The method of claim 15, wherein said method further comprises lysing said host cell and then isolating said protein.
21. The method of claim 20, wherein said method further comprises purifying said protein.
22. The nucleic acid of claim 1 which comprises the sequence of SEQ ID NO:1.
23. The vector of claim 4 comprising a nucleic acid encoding a protein of SEQ ID NO:3.
24. A recombinantly produced PEDF protein, which has been produced in accordance with the method of claim 15 and which is substantially free from proteins normally associated with a naturally occurring source organism.
25. An isolated and purified PEDF protein comprising SEQ ID NO:2 and conservatively modified variants thereof.
26. A DNA encoding PEDF comprising SEQ ID NO:43.
Type: Application
Filed: Aug 10, 2005
Publication Date: Jan 12, 2006
Applicant: The United States of America, As Represented by the Dept. of Health and Human Services (Rockville, MD)
Inventors: Gerald Chader (Bethesda, MD), Ignacio Rodriguez (Rockville, MD), Joyce Tombran-Tink (Derwood, MD), Sofia Becerra (Bethesda, MD), Fintan Steele (Washington, DC), Lincoln Johnson (Pasadena, CA)
Application Number: 11/201,443
International Classification: C12N 5/06 (20060101); C07H 21/04 (20060101); C07K 14/705 (20060101);
