Purification and characterization of recombinant human macrophage colony-stimulating factor and generation of a neutralizing antibody useful for Western analysis

Recombinant human macrophage colony-stimulating factor 1 (rCSF-1, also known as M-CSF) has been purified in milligram quantities from culture supernatants of SV40-infected CV-1 monkey cells that were transformed with a plasmid (pcCSF17) containing a human CSF-1 cDNA (Kawasaki et al. (1985) Science 230, 291–296). The rCSF-1 was purified using a 4-step procedure which resulted in a 285-fold purification and a yield of 40%. This rCSF-1 was shown to be a dimeric, disulfide-linked glycoprotein with an apparent native molecular weight of 65 kDa. The specific biological activity and amino-terminal sequence of this rCSF-1 were shown to be identical to that reported for native CSF-1 from MIA PaCa-2 cells. Although the pcCSF17 CSF-1 cDNA sequence coded for a mature polypeptide of 224 amino acids in length, C-terminal analysis of purified rCSF-1 indicated that C-terminal proteolytic processing had occurred at or near residue 158.A high-titer, polyclonal antibody to rCSF-1 was produced in rabbits and shown to specifically neutralize the biological activity of both CV-1 rCSF-1 and native CSF-1 from MIA PaCa-2 cells. In addition, the anti-CSF-1 antibody has been used to detect native and recombinant CSF-1 on Western blots.     

Osteopontin, a transformation-associated cell adhesion phosphoprotein, is induced by 12-O-tetradecanoylphorbol 13-acetate in mouse epidermis

A murine mRNA (provisionally called 2ar) is described whose abundance is greatly increased by the tumor promoter 12-O-tetradecanoylphorbol 13-acetate both in JB6 epidermal cells in vitro and in epidermis in vivo. We have previously shown induction of 2ar in epidermal or fibroblast cell lines by tumor promoters, growth factors, and transformation with H-ras. The 2ar mRNA appears to be derived from a single copy gene. It encodes the mouse homolog of rat osteopontin, a 41.5-kDa glycosylated bone phosphoprotein that binds to fibroblasts and osteosarcoma cells and to hydroxylapatite (bone matrix). The rat and mouse sequences are 84% identical at the amino acid level and 87% identical at the nucleotide level. Many of the primary structural features are conserved, including a run of 9-10 aspartic residues and a Gly-Arg-Gly-Asp-Ser cell adhesion sequence. Antiserum raised against portions of the predicted polypeptide immunoprecipitated proteins of apparent Mr 55,000-70,000 both from reticulocyte lysates containing the translation products of hybrid-selected mRNA and from cell culture medium containing metabolically labeled proteins secreted by JB6 cells. The results presented here demonstrate that osteopontin is identical to a transformation-associated phosphoprotein whose level of expression by cultured cells and abundance in human sera has been correlated with tumorigenicity. These results suggest a role for osteopontin in carcinogenesis. The murine version of osteopontin has been given the formal name "secreted phosphoprotein 1" and the designation spp.     

Macrophage growth factor CSF-1 stimulates human monocyte production of interferon, tumor necrosis factor, and colony stimulating activity

CSF-1, a macrophage colony stimulating factor that causes proliferation and differentiation of progenitor cells, may also have effects on mature cells. Human peripheral blood monocytes were used to examine this possibility. Monocytes, separated from normal blood by density centrifugation and adherence, were incubated for 3 days with or without CSF-1 (1,000 U/ml, purified from the MIA PaCa pancreatic carcinoma line). The two groups of cells were then washed and tested for the ability, when induced, to produce several factors. When induced for 2 days with LPS and PMA, the monocytes produced a factor that was cytotoxic to L929 cells, and this factor was completely neutralized by polyclonal antibody to tumor necrosis factor. The cells preincubated with CSF-1 consistently produced an average of 12 times more of this factor than cells not exposed to CSF-1. Monocytes induced with LPS and PMA also produced a colony stimulating activity, as measured by colony formation when using mouse bone marrow. Cells preincubated with CSF-1, washed, and induced with LPS and PMA produced more than three times as much activity compared with control monocytes. When monocytes were induced with poly-I.C, 22-fold higher levels of interferon were produced by the cells exposed to CSF-1. These results show that CSF-1 has direct stimulating effects on mature human monocytes, and suggest that the macrophage growth factor may have clinical application in the treatment of infectious diseases and cancer.     

Requirement of hydrophilic amino-terminal residues for granulocyte-macrophage colony-stimulating factor bioactivity and receptor binding.

Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a glycoprotein required for the proliferation and differentiation of granulocyte and macrophage precursors. Previous investigations have identified regions in human and murine GM-CSF that are required for bioactivity. In the present study, alanine substitution mutagenesis was undertaken to define more precisely specific amino-terminal residues in murine GM-CSF that are involved in bioactivity and receptor binding. Five double alanine mutants were identified that showed at least 10-fold reductions in bioactivity (K14AK20A, K14AE21A, H15AK20A, H15AE21A, K20AE21A). Each of these mutants maintained a normal N-linked glycosylation pattern when expressed in COS-1 cells, suggesting that native polypeptide backbone conformation was preserved. The purified prokaryotic expression products of two mutants (K14AE21A and H15AE21A) had a 100-fold decrease in bioactivity and a decrease in receptor binding, indicating that the side chains of K14, H15, and E21 are required for optimal receptor binding and maximal bioactivity.     

Glucose metabolic alterations in isolated and perfused rat hepatocytes induced by pancreatic cancer conditioned medium: a low molecular weight factor possibly involved

A serious insulin resistance characterizes pancreatic cancer-associated diabetes mellitus. Elsewhere, we demonstrated that MIA PaCa2 cultured cells secrete a soluble factor responsible for reduced glucose tolerance induced in SCID mice. The intracellular mechanism of insulin resistance was investigated in isolated and perfused rat hepatocytes incubated with MIA PaCa2 conditioned medium. Lactate production was reduced compared to hepatocytes incubated with control medium while 1,2-DAG was increased and PKC was activated in the hepatocytes incubated with MIA PaCa2 conditioned medium. This behavior was not reproduced treating the hepatocytes with the growth factors EGF, interleukin Ibeta, interleukin-6, and TGF-beta1. In an attempt to make a biochemical identification of the hypothesized tumor associated-diabetogenic factors we observed a low molecular weight protein in the conditioned medium, absent in the nonconditioned one, that may be responsible for the described behaviors.     

Sciatin Is a Transferrin-Like Polypeptide

Abstract: Sciatin, an acidic glycoprotein from chicken sciatic nerve, has myotrophic effects on avian skeletal muscle cells in culture. As sciatin was found to have certain structural similarities to transferrin, we further investigated the physicochemical characteristics of sciatin in order to determine the relationship between these two proteins. Sciatin was found to be strikingly similar to ovotransferrin in amino acid composition. In addition, amino acid sequence analysis revealed that sciatin and ovotransferrin had identical amino-terminal sequences for at least the first 20 amino acid residues. Chicken ovotransferrin, but not human serum transferrin, cross-reacted with rabbit antisciatin antibodies upon rocket immunoelectrophoresis and double immunodiffusion in agar. In addition, in the presence of bicarbonate, sciatin bound approximately 2 mol ferrous iron/mol protein. Using the purification procedure developed for sciatin, we purified a protein from chicken serum that cross-reacted with antisciatin serum, migrated at a position identical to that of sciatin or ovotransferrin on two-dimensional gel electrophoresis, had an amino composition very similar to ovotransferrin and sciatin, and had myotrophic effects on cultured muscle cells. From these data, we conclude that sciatin is a growth-promoting polypeptide closely related in structure to transferrin.     

3D modeling and characterization of the human CD115 monoclonal antibody H27K15 epitope and design of a chimeric CD115 target

The humanized monoclonal antibody H27K15 specifically targets human CD115, a type III tyrosine kinase receptor involved in multiple cancers and inflammatory diseases. Binding of H27K15 to hCD115 expressing cells inhibits the functional effect of colony-stimulating factor-1 (CSF-1), in a non-competitive manner. Both homology modeling and docking programs were used here to model the human CD115 extracellular domains, the H27K15 variable region and their interaction. The resulting predicted H27K15 epitope includes mainly the D1 domain in the N-terminal extracellular region of CD115 and some residues of the D2 domain. Sequence alignment with the non-binding murine CD115, enzyme-linked immunosorbent assay, nuclear magnetic resonance spectroscopy and affinity measurements by quartz crystal microbalance revealed critical residues of this epitope that are essential for H27K15 binding. A combination of computational simulations and biochemical experiments led to the design of a chimeric CD115 carrying the human epitope of H27K15 in a murine CD115 backbone that is able to bind both H27K15 as well as the murine ligands CSF-1 and IL-34. These results provide new possibilities to minutely study the functional effects of H27K15 in a transgenic mouse that would express this chimeric molecule.     

3D modeling and characterization of the human CD115 monoclonal antibody H27K15 epitope and design of a chimeric CD115 target

The humanized monoclonal antibody H27K15 specifically targets human CD115, a type III tyrosine kinase receptor involved in multiple cancers and inflammatory diseases. Binding of H27K15 to hCD115 expressing cells inhibits the functional effect of colony-stimulating factor-1 (CSF-1), in a non-competitive manner. Both homology modeling and docking programs were used here to model the human CD115 extracellular domains, the H27K15 variable region and their interaction. The resulting predicted H27K15 epitope includes mainly the D1 domain in the N-terminal extracellular region of CD115 and some residues of the D2 domain. Sequence alignment with the non-binding murine CD115, enzyme-linked immunosorbent assay, nuclear magnetic resonance spectroscopy and affinity measurements by quartz crystal microbalance revealed critical residues of this epitope that are essential for H27K15 binding. A combination of computational simulations and biochemical experiments led to the design of a chimeric CD115 carrying the human epitope of H27K15 in a murine CD115 backbone that is able to bind both H27K15 as well as the murine ligands CSF-1 and IL-34. These results provide new possibilities to minutely study the functional effects of H27K15 in a transgenic mouse that would express this chimeric molecule.     

Human liver alkaline phosphatase, purification and partial sequencing: homology with the placental isozyme

Human liver alkaline phosphatase (AP) has been purified to homogeneity. The enzyme has a molecular weight of 150,000 in its native state and consists of two identical subunits of Mr 75,000. After treatment with endoglycosidase F the molecular weight is reduced to 50,000 indicating a high degree of glycosylation. The amino-terminal sequence up to 22 residues was found to be Leu-Val-Pro-Glu-Lys-Glu-Lys-Asp-Pro-Lys-Tyr-(Ala)-Arg-Asp-Gln-Ala-Gln-?- Thr-Leu-Lys-Tyr. The amino-terminal portions of human and bovine liver AP are identical. The amino termini of the human liver and human placental AP isozymes have appreciable homology. Conformationally the amino termini are very similar.     

Sequence analysis of the large and small subunits of human ribonucleotide reductase.

We have isolated and sequenced overlapping cDNA clones from a breast carcinoma cDNA library containing the entire coding region of both the R1 and R2 subunits of the human ribonucleotide reductase gene. The coding region of the human R1 subunit comprises 2376 nucleotides and predicts a polypeptide of 792 amino acids (calculated molecular mass 90,081). The sequence of this subunit is almost identical to the equivalent mouse ribonucleotide reductase subunit with 97.7% homology between the mouse and human R1 subunit amino acid sequences. The coding region of the human R2 subunit of ribonucleotide reductase comprises 1170 nucleotides and predicts a polypeptide of 389 amino acids (calculated molecular mass 44,883), which is one amino acid shorter than the equivalent mouse subunit. The human and mouse R2 subunits display considerable homology in their carboxy-terminal amino acid sequences, with 96.3% homology downstream of amino acid 68 of the human and mouse R2 proteins. However, the amino-terminal portions of these two proteins are more divergent in sequence, with only 69.2% homology in the first 68 amino acids.     

A receptor binding domain of mouse interleukin-4 defined by a solid-phase binding assay and in vitro mutagenesis.

Interleukin 4 (IL-4) is a potent, pleiotropic lymphokine that affects a variety of cells, especially those of hematopoietic origin. Although murine and human IL-4 are homologous proteins, they display a species specificity in which murine IL-4 acts only upon mouse cells, and human IL-4 only upon human cells. We have used a mutagenesis strategy to define both the structural determinants of this specificity and a receptor binding domain of murine IL-4. To do this, we developed convenient solid-phase binding assays for mouse and for human IL-4, each utilizing receptor-immunoglobulin fusion proteins and alkaline phosphatase-tagged ligands. These were employed to assess the receptor binding activities of wild type and mutant forms of IL-4. In a separate biological assay, we measured the ability of each version of IL-4 to induce proliferation of a cultured mouse T-cell line. By replacing regions of mouse IL-4 with homologous segments of human IL-4, we found that the amino-terminal 16 residues and the carboxyl-terminal 20 residues of murine IL-4 are required for species-specific receptor binding as well as for T-cell proliferation. A major portion of the amino acid sequence between these regions can be substituted between mouse and human without loss of receptor binding or biological activity. Further, alanine-scanning mutagenesis revealed specific residues in the amino- and carboxyl-terminal regions (Glu-12, Ile-14, Leu-104, Asp-106, Phe-107, and Leu-111) that bear side chains critical for function. An analysis of the carboxyl-terminal region of murine IL-4 and its comparison with carboxyl-terminal regions of other related cytokines suggest an evolutionary conservation of structural and functional features.     

Identification of a novel serum protein secreted by lung carcinoma cells

The murine anti-human lung tumor monoclonal antibody L3 recognizes antigens found both in the medium of cultured carcinoma cells and in normal human serum. Sequential immunoprecipitation experiments indicate that the L3 antigen is also recognized by a previously described monoclonal antibody directed against a melanoma-associated antigen [Natali, P. G., Wilson, B. S., Imai, K., Bigotti, A., & Ferrone, S. (1982) Cancer Res. 42, 583-589]. This antibody precipitated a Mr 76000 glycoprotein from metabolically labeled extracts of the lung carcinoma cell line Calu-1 and a Mr 94 000 glycoprotein from labeled culture medium. Pulse-chase experiments suggested a precursor-product relationship between these molecules. Analysis of glycosidase sensitivities of the two forms indicated that maturation of carbohydrate side chains correlated with the apparent increase in molecular weights. L3 antigenic activity, measured in a competitive radiometric cell binding assay, was purified more than 90-fold from serum-free medium of Calu-1 cells and more than 3000-fold from normal human serum. The major immunoreactive components purified from culture medium and serum were identical with respect to apparent molecular weight, electrophoretic mobility, pI, glycosidase sensitivity, and V8 protease fingerprints. In addition, the sequence of the amino-terminal 16 N-terminal amino acid residues of the major immunoreactive species from both sources was identical. The properties of the L3 antigen did not correspond to those of any known protein, suggesting that this serum protein has not been previously characterized.     

Stimulation of macrophage colony-stimulating factor synthesis by interleukin-1.

Interleukin-1 (IL-1), which plays an important role in the inflammatory response, was found to induce colony-stimulating factor-1 (CSF-1) expression in the MIA PaCa-2 cells. IL-1-induced CSF-1 production was markedly suppressed (70%) by pertussis toxin. This inhibition by pertussis toxin was reversed by benzamide, an inhibitor of ADP-ribosylation reactions. Similarly, IL-1-induced CSF-1 production was inhibited by cholera toxin and this inhibition was reversed by an arginine analog, p-methoxy-benzylaminodecamethylene guanidine sulfate. Dibutyryl-cAMP as well as other cAMP elevating agents such as theophylline and forskolin also suppressed IL-1-induced CSF-1 production, suggesting that cAMP concentrations inversely regulate the biosynthesis of CSF-1. Measurement of cAMP concentration indicated that IL-1 treatment of MIA PaCa-2 cells did not change the cAMP level. IL-1-induced CSF-1 production was not suppressed by the protein kinase C (PKC) inhibitor, H7, under conditions in which 12-O-tetradecanoylphorbol-13-acetate-induced CSF-1 production was completely abolished. These data suggest that IL-1-induced CSF-1 production is not mediated via the activation of PKC. Analysis of oncogene c-fos and c-jun expression has shown the enhancement of expression of both protooncogenes prior to CSF-1, suggesting that the expression of these two oncogenes may be the mechanism which triggers CSF-1 gene expression.     

The c-fms proto-oncogene product is related to the receptor for the mononuclear phagocyte growth factor, CSF-1

The feline c-fms proto-oncogene product is a 170 kd glycoprotein with associated tyrosine kinase activity. This glycoprotein was expressed on mature cat macrophages from peritoneal inflammatory exudates and spleen. Similarly, the receptor for the murine colony-stimulating factor, CSF-1, is restricted to cells of the mononuclear phagocytic lineage and is a 165 kd glycoprotein with an associated tyrosine kinase. Rabbit antisera to a recombinant v-fms-coded polypeptide precipitated the feline c-fms product and specifically cross-reacted with a 165 kd glycoprotein from mouse macrophages. This putative product of the murine c-fms gene exhibited an associated tyrosine kinase activity in immune complexes, specifically bound murine CSF-1, and, in the presence of the growth factor, was phosphorylated on tyrosine in membrane preparations. The murine c-fms proto-oncogene product and the CSF-1 receptor are therefore related, and possibly identical, molecules.     

Identification of tyrosine 706 in the kinase insert as the major colony-stimulating factor 1 (CSF-1)-stimulated autophosphorylation site in the CSF-1 receptor in a murine macrophage cell line.

The receptor for colony-stimulating factor 1 (CSF-1) is a ligand-activated protein-tyrosine kinase. It has been shown previously that the CSF-1 receptor is phosphorylated on serine in vivo and that phosphorylation on tyrosine can be induced by stimulation with CSF-1. We studied the phosphorylation of the CSF-1 receptor by using the BAC1.2F5 murine macrophage cell line, which naturally expresses CSF-1 receptors. Two-dimensional tryptic phosphopeptide mapping showed that the CSF-1 receptor is phosphorylated on several different serine residues in vivo. Stimulation with CSF-1 at 37 degrees C resulted in rapid phosphorylation on tyrosine at one major site and one or two minor sites. We identified the major site as Tyr-706. The identity of Tyr-706 was confirmed by mutagenesis. This residue is located within the kinase insert domain. There was no evidence that Tyr-973 (equivalent to Tyr-969 in the human CSF-1 receptor) was phosphorylated following CSF-1 stimulation. When cells were stimulated with CSF-1 at 4 degrees C, additional phosphotyrosine-containing phosphopeptides were detected and the level of phosphorylation of the individual phosphotyrosine-containing phosphopeptides was substantially increased. In addition, we show that CSF-1 receptors are capable of autophosphorylation at six to eight major sites in vitro.     

Transduction of human colony-stimulating factor-1 (CSF-1) receptor into interleukin-3-dependent mouse myeloid cells induces both CSF-1-dependent and factor-independent growth.

A retroviral vector encoding the receptor for human colony-stimulating factor-1 (CSF-1) was introduced into murine myeloid FDC-P1 cells which require interleukin-3 (IL-3) for their proliferation and survival in culture. Cells expressing the CSF-1 receptor (CSF-1R), selected by fluorescence-activated cell sorting in the continued presence of murine IL-3, formed colonies in semisolid medium and were able to proliferate continuously in liquid cultures containing human recombinant CSF-1. Thus, although they do not synthesize endogenous murine CSF-1R, FDC-P1 cells express the downstream components of the CSF-1 mitogenic pathway necessary for its signal-response coupling. After receptor transduction, slowly proliferating factor-independent variants that produced neither CSF-1 nor growth factors able to support the proliferation of parental FDC-P1 cells also arose. When the human CSF-1R was expressed in FDC-P1 cells under the control of an inducible metallothionein promoter, the frequencies of both CSF-1-responsive and factor-independent variants increased after heavy-metal treatment. In addition, a monoclonal antibody to human CSF-1R arrested colony formation by both the CSF-1-dependent and factor-independent cells but did not affect their growth in response to IL-3. Therefore, the induction of both the CSF-1-dependent and factor-independent phenotypes depended on expression of the transduced human CSF-1R.     

Identification of Two Tyrosine Residues Required for the Intramolecular Mechanism Implicated in GIT1 Activation

GIT1 is an ArfGAP and scaffolding protein regulating cell adhesion and migration. The multidomain structure of GIT1 allows the interaction with several partners. Binding of GIT1 to some of its partners requires activation of the GIT1 polypeptide. Our previous studies indicated that binding of paxillin to GIT1 is enhanced by release of an intramolecular interaction between the amino-terminal and carboxy-terminal portions that keeps the protein in a binding-incompetent state. Here we have addressed the mechanism mediating this intramolecular inhibitory mechanism by testing the effects of the mutation of several formerly identified GIT1 phosphorylation sites on the binding to paxillin. We have identified two tyrosines at positions 246 and 293 of the human GIT1 polypeptide that are needed to keep the protein in the inactive conformation. Interestingly, mutation of these residues to phenylalanine did not affect binding to paxillin, while mutation to either alanine or glutamic acid enhanced binding to paxillin, without affecting the constitutive binding to the Rac/Cdc42 exchange factor βPIX. The involvement of the two tyrosine residues in the intramolecular interaction was supported by reconstitution experiments showing that these residues are important for the binding between the amino-terminal fragment and carboxy-terminal portions of GIT1. Either GIT1 or GIT1-N tyrosine phosphorylation by Src and pervanadate treatment to inhibit protein tyrosine phosphatases did not affect the intramolecular binding between the amino- and carboxy-terminal fragments, nor the binding of GIT1 to paxillin. Mutations increasing the binding of GIT1 to paxillin positively affected cell motility, measured both by transwell migration and wound healing assays. Altogether these results show that tyrosines 246 and 293 of GIT1 are required for the intramolecular inhibitory mechanism that prevents the binding of GIT1 to paxillin. The data also suggest that tyrosine phosphorylation may not be sufficient to release the intramolecular interaction that keeps GIT1 in the inactive conformation.     

The human double-stranded DNA-activated protein kinase phosphorylates the 90-kDa heat-shock protein, hsp90 alpha at two NH2-terminal threonine residues

The 90-kDa heat-shock protein, hsp90, is an abundant cytoplasmic protein that can be phosphorylated in vitro by a double-stranded (ds) DNA-activated protein kinase found in cells from several species. Here we show that the dsDNA-activated protein kinase from human HeLa cells phosphorylates 2 threonine residues in the sequence PEETQTQDQPME at the amino terminus of human hsp90 alpha. Hsp90 beta, which is 97% identical to hsp90 alpha but lacks both amino-terminal threonines, is not phosphorylated by the dsDNA-activated protein kinase. Mouse hsp86 and rabbit hsp90 alpha are homologous to human hsp90 alpha; both heterologous proteins are phosphorylated at the same amino-terminal threonines by the human dsDNA-activated protein kinase.     

Amino acid sequence and molecular characterization of murine lamin B as deduced from cDNA clones

The nuclear lamina is the karyoskeletal structure, intimately associated with the nuclear envelope, that is widespread among the diverse types of eukaryotic cells. A family of proteins, termed lamins, has been shown to be a prominent component of this lamina, and various members of this family are differentially expressed in different cell types. In mammals, three major lamins (A, B, C) have been identified, and in all cells so far examined lamin B is constitutively expressed while lamins A and C are not, suggesting that lamin B is sufficient to form a functional lamina. Because of this key importance of lamin B, cDNA clones encoding mammalian lamin B were isolated by screening murine cDNA libraries, representing F9 teratocarcinoma cells and fetal liver, with the corresponding cDNA probe of lamin LI of Xenopus laevis. The nucleotide sequence of the murine lamin B mRNA (approximately 2.9 kb) was determined. The deduced amino acid sequence of the encoded polypeptide (587 amino acids; mol. wt. 66760) is highly homologous to X. laevis lamin LI (72.9% identical residues) but displays lower similarity to A-type lamins (53.8% identical amino acid residues with human lamin A). Lamin B also conforms to the general molecular organization principle of the members of the intermediate filament (IF) protein family, i.e., an extended alpha-helical rod domain that is interrupted by two non alpha-helical linkers and flanked by non-alpha-helical head (amino-terminal) and tail (carboxy-terminal) domains. The tail domain, which does not reveal a hydrophobic region of considerable length, contains a typical karyophilic signal sequence and an uninterrupted stretch of eight negatively charged amino acids.(ABSTRACT TRUNCATED AT 250 WORDS)     

Structural and immunological comparison of indigenous human O6-methylguanine-DNA methyltransferase with that encoded by a cloned cDNA

O6-Methylguanine-DNA methyltransferase, a ubiquitous and unusual DNA repair protein, eliminates mutagenic and cytotoxic O6-alkylguanine from DNA by transferring the alkyl group to one of its cysteine residues in a second-order suicide reaction. This 22-kDa protein was immunoaffinity-purified to homogeneity from cultured human lymphoblasts (CEM-CCRF line) and compared with the O6-methylguanine-DNA methyltransferase purified to homogeneity from Escherichia coli expressing a cloned human cDNA. The cellular and recombinant proteins were identical in size, as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis of intact molecules and their peptides. Immunoprobing of Western blots with three monoclonal antibodies specific for human cellular O6-methylguanine-DNA methyltransferase further indicated identity of the two proteins. The amino acid sequence of the cellular protein was experimentally determined for 87 out of a total of 207 residues and was found to be identical to that deduced from the cDNA sequence. A unique cysteine residue at position 145 was identified as the methyl acceptor site by autoradiographic analysis of peptides and sequence analysis of 3H-methylated O6-methylguanine-DNA methyltransferase. These observations establish that the cloned O6-methylguanine-DNA methyltransferase cDNA encodes the full-length O6-methylguanine-DNA methyltransferase polypeptide that is normally present in human cells. Moreover, the cellular protein does not appear to be significantly modified by posttranslational processes.