DNA Encoding Macrophage Migration Inhibition Factor From Ocular Lens

The lymphokine, Macrophage Migration Inhibition Factor (MIF), has been identified as a mediator of the function of macrophages in host defence and its expression correlates with delayed hypersensitivity and cellular immunity. A 12,000 da protein with MIF activity was identified by Weiser et al, Proc. Natl. Acad. Sci. U.S.A., 86, 7522-7526 (1989). MIF was first characterized by expression cloning from activated human T-cells, however, the abundance of the product is low in these cells. No MIF protein is commercially available, although the human cDNA is marketed by R&D Systems Inc., Minneapolis, Minn.

The eye lens contains high concentrations of soluble proteins, Harding et al, The Eye, ed. Davson, H., Academic Press, New York, Vol. 1B, pp. 207-492; Wistow et al, Ann. Rev. Biochem., 57, 479-504 (1988); and Wistow et al, Nature, 326, 622-624 (1987). The most abundant proteins, the crystallins, are structural, comprising the refractive material of the tissue. Some crystallins are specialized for the lens, others are identical to enzymes expressed in lower amounts in other tissues. Individual crystallins may account for a quarter or more of total lens protein, Wistow et al, Nature, 326, 622-624 (1987) and Wistow et al, PNAS, 87, 6277-6280 (1990). However, other proteins are also present at moderate abundance, typically in the range 0.1-1% of total protein. Some of these are also enzymes, such as .alpha.-enolase or aldehyde dehydrogenase, found as crystallins in some species, Wistow et al, J. Mol. Evol., 32, 262-269 (1991).

It has been discovered that a moderately abundant protein in the eye lens, "10K protein", which accounts for as much as 1% of total protein in young or embryonic lenses is similar to MIF. An equivalent protein is present in all lenses examined, including bovine lenses from slaughtered animals. Accordingly, eye lenses of various animals, especially birds and mammals, can be used as a source of MIF.

MIF is extremely abundant in lens compared with other known sources. Proteins accumulate to high levels in lens, which has low proteolytic activity. Lenses may be removed from eyes quickly and simply with one incision. Moreover, no other abundant lens proteins are close to lens MIF in size, thus facilitating its separation. Lenses can similarly be used as abundant sources of active enzymes including lactate dehydrogenase B and argininosuccinate lyase.

The lens MIF can be obtained by homogenizing ocular lens to form a homogenate, separating a soluble extract and an insoluble membrane fraction from said homogenate and recovering purified MIF from said soluble extract.

The present invention is also directed to purified lens MIF. In preparing the purified natural lens MIF of the present invention, a stimulant such as Con-A is not added to the preparation and therefore this possible source of contamination is avoided.

MIF plays an important role in the inflammatory response. Lenses could become a useful source of MIF protein for research and therapeutic purposes. In lens, MIF expression is associated with cell differentiation and with expression of the proto-oncogene N-myc. Lens MIF may be a growth factor in addition to its role as a lymphokine. Like other lymphokines, such as IL-2, MIF could have specific therapeutic value in stimulation of immune system and other cells. In particular, lens MIF may play a role in some inflammatory conditions in the eye. MIF isolated from lens could be modified to derive antagonists for the inflammatory process.

The MIF of the present invention can be produced by recombinant DNA techniques. The invention therefore is also directed to recombinant DNA which encodes MIF, replicable expression vectors which contain the DNA and which can express MIF and transformed cells and/or microorganisms which contain the DNA and which can express large amounts of MIF.

The MIF can be separated from the lens by a variety of different procedures.

As a first step, the lens should be homogenized in an aqueous solution, preferably an aqueous buffered solution having a pH of about 7 to 7.6, preferably 7 to 7.4 which does not adversely affect the MIF, in order to allow the soluble materials to dissolve in the buffer. The buffered solution will usually not contain any other solvents. Homogenization is preferably achieved by physically breaking up the lenses by use of a glass rod, blender or other suitable devices or procedures. The volume ratio of lens to the solution is usually 1:1 to 5, preferably 1:1.5 to 3 (v/v).

After the lens is homogenized, the insoluble membranes are separated from the aqueous solution containing the soluble extract. This can be accomplished in any known manner but centrifugation appears to be especially useful.

The MIF is then recovered from the soluble extract. In the experiments reported herein, this is accomplished by subjecting the soluble extract to sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE). However, if it is desired to separate the MIF from the soluble extract on a larger scale, various procedures such as column chromatography (sizing columns) and/or isoelectric focusing can be utilized.

All lenses examined by SDS polyacrylamide gel electrophoresis have a prominent minor band with subunit size around 10-12 kDa, "10K protein". MIF is the major component in the 10-12,000 da subunit size range, as visualized by SDS polyacrylamide gel electrophoresis. In aged and cataractous lenses, fragments of .alpha.-crystallin have been found in this size range, Harding et al, The Eye, ed. Davson, H. (Academic Press, New York), Vol. 1B, pp. 207-492 (1984). However, even embryonic lenses, in which proteolysis is unlikely to have occurred to a great extent, have a distinct 10K subunit band. This band was isolated from embryonic chick lens and sequenced. Surprisingly, the sequence obtained showed close identity to a recently described lymphokine, human MIF, Weiser et al, Proc. Natl. Acad. Sci. U.S.A., 86, 7522-7526 (1989). The polymerase chain reaction (PCR) was used to clone the mRNA for chick lens 10K protein. This provided a probe to clone cDNA for chick and three week old mouse lens 10K protein. PCR was also used to clone 10K protein from fetal human lens.

The presence of MIF at high levels in lens suggests it may have a wide role as a polypeptide growth factor rather than a restricted function as a lymphokine. Preliminary experiments using PCR suggest that MIF in embryonic chick lens is expressed in equatorial and fiber cells but not in central epithelium, consistent with a role in the differentiation of lens cells. Northern blot analysis with the cDNA for mouse lens 10K/MIF shows that the message is present in various tissues, including lens, brain and kidney.

The present invention is also directed to a vector comprising a replicable vector and a DNA sequence encoding the MIF inserted into the vector. The vector may be an expression vector and is conveniently a plasmid.

The MIF preferably comprises one of the sequences described in the SEQUENCE LISTING or a homologous variant of said MIF having 5 or less conservative amino acid changes, preferably 3 or less conservative amino acid changes. In this context, "conservative amino acid changes" are substitutions of one amino acid by another amino acid wherein the charge and polarity of the two amino acids are not fundamentally different. Amino acids can be divided into the following four groups: (1) acidic amino acids, (2) neutral polar amino acids, (3) neutral non-polar amino acids and (4) basic amino acids. Conservative amino acid changes can be made by substituting one amino acid within a group by another amino acid within the same group. Representative amino acids within these groups include, but are not limited to, (1) acidic amino acids such as aspartic acid and glutamic acid, (2) neutral polar amino acids such as valine, isoleucine and leucine, (3) neutral non-polar amino acids such as asparganine and glutamine and (4) basic amino acids such as lysine, arginine and histidine.

In addition to the above mentioned substitutions, the MIF of the present invention may comprise the specific amino acid sequences shown in the SEQUENCE LISTING and additional sequences at the N-terminal end, C-terminal end or in the middle thereof. The "gene" or nucleotide sequence may have similar substitutions which allow it to code for the corresponding MIF.

In processes for the synthesis of the MIF, DNA which encodes the MIF is ligated into a replicable (reproducible) vector, the vector is used to transform host cells, and the MIF is recovered from the culture. The host cells for the above-described vectors include prokaryotic microorganisms including gram-negative bacteria such as E. coli, gram-positive bacteria, and eukaryotic cells such as yeast and mammalian cells. Suitable replicable vectors will be selected depending upon the particular host cell chosen. Alternatively, the DNA can be incorporated into the chromosomes of the eukaryotic cells for expression by known techniques. Thus, the present invention is also directed to recombinant DNA, recombinant expression vectors and transformed cells which are capable of expressing MIF.

For pharmaceutical uses, the MIF is purified, preferably to homogeneity, and then mixed with a compatible pharmaceutically acceptable carrier or diluent. The pharmaceutically acceptable carrier can be a solid or liquid carrier depending upon the desired mode of administration to a patient. If the MIF is used to stimulate growth or differentiation of cells, specifically mammalian or bird cells, the MIF is contacted with the cells under conditions which allow the MIF to stimulate growth or differentiation of the cells. MIF could be administered to stimulate macrophages, which might be useful under some circumstances. For suppression of inflammation, macrophages would need to be unstimulated, this might be achievable using modified MIF as an antagonist.

Lenses:

Chick lenses were excised from 11 day post fertilization embryos. Mouse lenses were from 3 week old BALB/C mice. Human fetal lenses were from a 13.5 week fetus obtained in therapeutic abortion in 1986 and saved at -80.degree. C. Bovine lenses were obtained from approximately 1 year-old animals from slaughter.

Lens Protein:

For protein analysis, lenses were homogenized with a Teflon tipped rod in Eppendorf tubes (1.5 ml) in TE buffer (10 mM Tris-HCl, pH 7.3; 1 mM EDTA) in an amount of about 1:2 (v/v). Membranes were spun down by microcentrifugation and the soluble extract retained. Proteins were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis, using 15% acrylamide, 1% SDS, Laemmli, Nature, 227, 680-685 (1970). Loading buffer contained 1 mM mercaptoethanol. For sequencing, gels were electroblotted onto nitrocellulose. The 10K band was excised. The Harvard Microchemistry facility performed microsequencing as a service, as described before, Wistow et al, J. Cell Biol., 107, 2729-2736 (1988). The protein was digested off the nitrocellulose by trypsin. Peptides were separated by HPLC and the major peaks sequenced using an Applied Biosystems automated sequencer. An initial N-terminal sequence was obtained by direct microsequencing of a fragment eluted from a coomassie blue stained gel slice.

Computer analysis:

Sequences were compared with the translated GenBank database, v65 using the SEQFT program of the IDEAS package, Kanehisa, IDEAS User's Manual (Frederick Cancer Research Facility, Frederick, Md.) (1986), run on the CRAY XMP at the Advanced Scientific Computing Laboratory, Frederick, Md.

RNA Preparation and Analysis:

Chick, mouse and human lenses and other tissues were homogenized in RNAzol (Cinna/Biotecx, Friendswood, Tex.) and subjected to RNA extraction, Chomczynski et al, Anal. Biochem., 162, 156-159 (1987). RNA was quantitated by UV absorption. For Northern blots, equal amounts of RNA were run on formaldehyde gels, Davis et al, Basic Methods in Molecular Biology, Elsevier Science Publishing Co., New York, N.Y. (1986) and electroblotted onto nitrocellulose or nylon membranes, Towbin et al, Proc. Natl. Acad. Sci. U.S.A., 76, 4350-4354 (1979).

PCR:

Oligonucleotides were designed from the sequence of chick lens 10K protein peptides and from the sequence of human MIF. Bam HI and Sal I sites were incorporated as shown.

oligo sequences: ##STR1##

Chick and human lens RNA were amplified using one step of the reverse transcriptase reaction primed with either 3' or oligo dT primers, Innis et al, PCR Protocols, A Guide to Methods and Applications, Academic Press, Inc., New York, N.Y., 1st Ed. (1990). First strand cDNA was then amplified by 30 cycles of PCR using an annealing temperature of 55.degree. C. Product was visualized using 1% agarose gels and ethidium bromide staining.

cDNA cloning and Sequencing:

The 300 bp chick lens cDNA PCR product was subcloned in Bluescript II (Stratagene, La Jolla, Calif.) following digestion with Bam HI and Sal I. A Bam HI site in the chick sequence resulted in two fragments which were cloned separately. Multiple clones were sequenced using Sequenase reagents (USB, Cleveland, Ohio) and .sup.35 S-dATP label (Amersham, Arlington Hts., Ill.). The human lens PCR product was subcloned as a single Bam HI-Sal I fragment and sequenced. The chick PCR product was also used as a probe by labelling with .sup.32 P-dCTP and random priming using a kit from Bethesda Research Laboratory, Gaithersburg, Md. This was used to screen an embryonic chick lens cDNA library in .lambda.gt11 (Clontech, Palo Alto, Calif.) and a newborn mouse lens library in .lambda.zap (vector from Stratagene, library a gift from Joan McDermott, NEI). Clones were screened, purified and sequenced by standard methods, Davis et al, Basic Methods in Molecular Biology, Elsevier Science Publishing Co., New York, N.Y. (1986).

The partial cDNA sequences obtained are as follows:

Protein sequence:

Microsequence for 5 tryptic peptides of chick lens MIF and an N-terminal sequence were obtained and are shown in Table 1.

The four sequences compared are:

Human T-cell MIF is from Weiser et al, Proc. Natl. Acad. Sci. U.S.A., 86, 7522-7526 (1989). Lens sequences are from cDNA library and PCR derived clones. Parts of the human lens 10K sequence were derived from the PCR oligos and are not shown. Peptides of chicken 10K/MIF are indicated by underline. The asterisk (*) shows the only difference between human lens and T-cell sequences.

The N-terminus of the lens 10K protein is at least partly unblocked. All sequences gave a partial match with the sequence of human MIF cloned from activated T-cells, Weiser et al, Proc. Natl. Acad. Sci. U.S.A., 86, 7522-7526 (1989). Sequences deduced from PCR and cDNA library clones confirmed this relationship. PCR clones for the coding region of human lens 10K protein were identical in sequence to the published sequence of human T-cell MIF except for one base identical in different PCR clones. Different PCR clones confirmed the difference. This single base change alters a predicted Serine residue to Asparagine, the identical amino acid found at the same position in mouse and chick cDNA clones and in chick protein sequence. It is possible that this conservative difference with the T-cell sequence results from conservative polymorphism or cloning or sequencing artifact. Such a change may or may not significantly change the properties of the protein.

Distribution of 10K/MIF:

PCR of RNA from dissected central epithelium, equatorial epithelium and fiber cells from 6, 12 and 14-day chick embryos showed that RNA for 10K/MIF is present in eqatorial and fiber cells at all stages but is absent from the central epithelium. Protein gels also confirm that 10K protein is detectable from 6 days and throughout chick lens development. A similar band is seen in all species examined, including bovine lenses. Northern blot analysis of mouse tissues using a mouse cDNA probe, show that 10K/MIF RNA is present in several tissues in addition to lens, particularly in brain and kidney.

The mouse cDNA is subcloned into a eukaryotic expression vector, pMAMNeo. PCR with added linker sequences is utilized to accomplish this so that a complete mouse MIF will be produced from its own initiator ATG in mammalian cells such as COS or NIH 3T3 cells.

The same clone of Example 2 is inserted into prokaryotic expression vector pKK233-2 to produce mouse MIF in E. coli.