Raw material considerations.

Raw materials (RM) testing and control strategy form a major part of the foundation of a well-characterized protein (WCP) or biopharmaceutical. Raw materials may be present in the final vial as excipients or may have product contact earlier in processing. Manufacturers of WCPs should use a scientific approach to set acceptance criteria and test methods for bulk raw materials which are not present in dosage forms in substantial amounts. These methods and standards should enable the procurement of specified RMs which enable the reliable preparation of a product which consistently meets its quality attributes. Manufacturers of WCPs must use available pharmacopoeial standards for excipients and bulk process RMs which cannot be substantially removed during purification or other processing steps. Manufacturers should support efforts to harmonize pharmacopoeial standards for excipients.

Viral validation strategy for recombinant products derived from established animal cell lines.

For products derived from continuous cell lines, regulatory agencies worldwide require that the purification process be validated for its ability to remove or inactivate potential contaminants such as viruses and virus-like particles. New guidance suggests a requirement for statistical evaluation of these studies but the industry has yet to develop such standards. The task of estimating excess capacity is also complicated by variable assays, accumulation of variability in clearance estimates over unit operations, dependence of clearance capacity on operating parameters, and expense of experiments. We propose an experimental strategy to determine the excess clearance capacity of a biopharmaceutical process and to provide statistical estimation of excess capacity in an efficient way. Clearance estimates and their variances are calculated for each orthogonal unit operation and estimates are combined to form an interval estimate of overall process clearance capacity. Poisson regression is suggested as an efficient technique for data analysis of clearance studies. We believe that this approach should meet regulatory guidelines in a cost effective way, while clarifying the roles of qualitative and quantitative components in setting requirements.

Human macrophage colony-stimulating factor heterogeneity results from alternative mRNA splicing, differential glycosylation, and proteolytic processing.

The single gene for human macrophage colony-stimulating factor (M-CSF, or CSF-1) generates multiple mRNA species that diverge within the coding region. We have characterized translation products of these mRNA species from native and recombinant sources. Immunoblots of reduced native M-CSF indicate that multiple glycosylated species ranging from 25 kd to 200 kd are secreted by human monocytes and cell lines. In contrast, CV-1 cells expressing a short M-CSF clone secrete only 24 kd recombinant M-CSF. Synthetic peptide antibodies were developed to distinguish between secreted recombinant M-CSF from long and short mRNA splicing variants. Immunoblot analysis indicates that alternative mRNA splicing generates some M-CSF protein heterogeneity. Most secreted MIA PaCa-2 M-CSF reacts with long-clone-specific antibody. Lectin affinity chromatography shows that variable glycosylation contributes significantly to MIA PaCa-2 M-CSF size heterogeneity. In addition, cell lysates also contain larger M-CSF species that apparently undergo proteolytic processing before secretion. The data indicate that M-CSF protein heterogeneity results from both pre- and post-translational processing.

Detection of endogenous macrophage colony-stimulating factor (M-CSF) in human blood.

We have detected endogenous human macrophage colony-stimulating factor (M-CSF) in blood of normal individuals, using a novel RIA that accurately measures M-CSF concentrations as low as 60 U/ml (1.2 ng/ml) in the presence of serum proteins. The RIA uses an antibody to highly purified recombinant human M-CSF and is calibrated to a mouse bone marrow colony-forming assay. Ten samples of normal human blood plasma contained an average 118 +/- 9 U/ml of M-CSF, and similar concentrations were detected in serum prepared from the same individuals. RIA-positive samples contained biologically active M-CSF, as determined in a colony assay performed on mouse bone marrow cells. The M-CSF biological activity was removed by specific immune precipitation and inhibited by addition of M-CSF antibody. Physical characterization of plasma M-CSF was done by immunoblotting after partial purification on controlled pore glass and immunoaffinity chromatography. The major reduced protein species of plasma M-CSF had an apparent molecular weight of about 24 kd, and minor species of 30, 45, and 60-70 kd were also present. The RIA results on ten normal individuals suggest that endogenous circulating M-CSF is present at a low but detectable concentration. This RIA can be used to measure M-CSF in clinical samples that contain serum proteins and other growth factors that may interfere with accurate bioassay determinations.

Calcium binding protein from porcine intestine binds to phosphatidylserine vesicles in the presence of calcium.

Protein II, a 32K cytoskeleton-associated protein isolated from porcine intestinal epithelium, binds to vesicles composed of phosphatidylserine in the presence, but not the absence, of 10 microM Ca2+. Binding was saturable and was specifically inhibited by chelation of free Ca2+ with EGTA. Binding was also inhibited by trifluophenothiazine. Vesicles composed of dimyristoylphosphatidylcholine did not bind protein II, suggesting that interaction with phosphatidylserine was selective. These properties are consistent with a possible role for protein II in Ca-regulated cytoskeleton-cell membrane events.

Three Ca2+-binding proteins from porcine liver and intestine differ immunologically and physicochemically and are distinct in Ca2+ affinities.

Intestinal brush-border-derived membrane vesicles contain, after demembranation in the presence of Ca2+, a subset of polypeptides that are specifically solubilized by the addition of Ca2+ chelators. As described previously, this fractionation scheme leads to the enrichment of two major proteins (I and II), one of which has been shown to be identical to the cellular p36K target of Rous sarcoma virus-encoded tyrosine-specific protein kinase (Gerke, V., and Weber, K., (1984) EMBO J. 3, 227-233). We have applied a similar protocol to membrane vesicles from porcine liver and purified a third Ca2+-binding protein (III). All three proteins had wide tissue distributions, and were absent from brain, red blood cells, and cardiac and skeletal muscle. Relative amounts varied between tissues, with protein I low in liver and protein III very low in intestine. Despite their similar extractability the three proteins (I, II, and III) are clearly distinct as far as immunological, biochemical, and physicochemical properties are concerned. They also show characteristic differences in their affinities for Ca2+ ions. The association constants of Ca2+ binding for proteins I and III have been estimated by means of indirect methods to be 10(4) M-1 (protein I) and 10(6) M-1 (protein III), while the direct Hummel-Dreyer method reveals Ca2+ binding to protein II, characterized by an association constant of 0.4 X 10(5) M-1 in the absence and 0.2 X 10(5) M-1 in the presence of 2 mM MgCl2. Conformational changes upon binding Ca2+ are described for protein II using circular dichroism, fluorescence emission, and UV difference spectra. These alterations could be attributed to an increased exposure of tyrosine and tryptophan residues to a more aqueous environment, and led to increased hydrophobicity of protein II that would explain the observed Ca2+-dependent interaction with hydrophobic matrices like phenyl-Sepharose.

Platelet-collagen adhesion: inhibition by a monoclonal antibody that binds glycoprotein IIb.

To identify platelet surface structures involved in adhesion to collagen, the effect of 16 murine antiplatelet membrane hybridoma antibodies were tested in a defined, in vitro assay. Four of these antibodies inhibited platelet-collagen adhesion and reacted with a polypeptide with Mr approximately 125,000, as determined by immunoblots after gel electrophoresis under reducing conditions. Through detailed studies with one of these antibodies, the monoclonal antibody PMI-1, the relevant antigen was identified as platelet glycoprotein IIb alpha, based upon (a) co-migration with this glycoprotein in two-dimensional gel electrophoresis and (b) co-purification by immunoaffinity chromatography with a protein with apparent Mr identical to that of glycoprotein III, under conditions in which glycoproteins IIb and III form a complex. Univalent antibody fragments prepared from monoclonal antibody PMI-1 inhibited greater than 80% of platelet-collagen adhesion, and inhibition was completely blocked by the immunopurified antigen. These results indicate that glycoprotein IIb participates in some aspect of platelet-collagen adhesion. In contrast, the purified antigen only partially neutralized a polyclonal antiserum that blocked platelet-collagen adhesion, to a maximum of approximately 25%, at saturating antigen concentrations. Thus, by these immunological criteria, glycoprotein IIb is not the only molecule involved in this process.

Platelet-collagen adhesion: evidence for participation of antigenically distinct entities.

Univalent antibody fragments prepared from a rabbit antiserum raised against whole human platelets completely inhibited adhesion of platelets to immobilized trimeric collagen in a defined, Mg2+-dependent, adhesion assay. An octylglucoside extract of whole platelets completely neutralized this antibody, and all neutralizing activity bound to immobilized wheat germ agglutinin. Further fractionation on concanavalin A gave rise to subfractions that each neutralized only partially at saturation, when tested against antibody concentrations that inhibit 50% of platelet-collagen adhesion. When tested against higher antibody concentrations that completely inhibited adhesion, each subfraction had no detectable neutralizing effect, although the combined subfractions neutralized completely. This and other evidence suggests that more than one platelet entity participates in platelet-collagen adhesion. Although distinct, they appear to play interdependent roles in a single adhesion process.

Adhesion of human platelets to immobilized trimeric collagen.

Human platelets adhere to trimeric Type 1 chick collagen that was covalently linked to plastic slides, providing the basis for a well-defined quantitative assay. The number of platelets that adhere is a function both of platelet concentration and of collagen density on the slides. In contrast with other in vitro assays using collagen that is not covalently linked to the substratum, we found no platelet-platelet aggregation. Adhesion was absolutely dependent on Mg2+, whereas Ca2+ was ineffective. Native trimeric collagen conformation was required for adhesion, since platelets did not bind to slides containing heat-denatured collagen, or isolated alpha 1(1) or or alpha 2(1) chains. Modifications of collagen oligosaccharides had no effect on adhesion. Adhesion was inhibited by cytochalasin D but was not affected by prostaglandin E1, apyrase, acetylsalicylic acid, or theophylline. Because this assay measures platelet-collagen adhesion in the absence of platelet-platelet aggregation, it should facilitate identification of the platelet surface components that directly mediate this adhesion.

Heparin-inhibitable lectins: marked similarities in chicken and rat.

Extracts of young rat lung contain a heparin-inhibitable lectin that closely resembles one recently purified from chicken liver. Both lectins interact with heparin and N-acetyl-D-galactosamine, and were purified by gel filtration on Sepharose CL-2B followed by affinity chromatography on heparin-Sepharose. They both behave as high molecular weight aggregates that can be dissociated into two peptides with apparent molecular weights of 13,000 and 16,000 by gel electrophoresis in SDS. Samples of purified lectin contained up to 20% DNA by weight, and the degree of lectin aggregation and hemagglutination activity was greatly reduced by treatment with micrococcal nuclease without inhibiting heparin-binding activity. Association of lectin with DNA is an artifact of homogenization in high salt, since only 2% of the lectin is found associated with a purified nuclear fraction.

Extracellular lectin and its glycosaminoglycan inhibitor in chick muscle cultures.

Embryonic chick muscle contains two developmentally regulated lectins, which may be involved in cell interactions. These endogenous lectins are assayed as agglutinins of appropriate test erythrocytes. One of these, called lectin-2, interacts with specific glycosaminoglycans, especially heparin and dermatan sulfate. Lectin-2 is present at constant levels in both chick fibroblast and chick muscle cells throughout 14 days of culture but is released into the medium of cultured embryonic muscle after 7-8 days of culture, soon after myoblast fusion. Lectin-2 interacts strongly with a component of substrate-attached material in embryonic muscle cultures which is extractable from the cutlure dishes with alkali after the cells have been removed with ethylediaminetetraacetic acid. The active component in the substrate-attached material appears to be a glycosaminoglycan that is a more potent inhibitor of lectin-2 agglutination activity than any of the known glycosaminoglycans that we have tested. The active material is degraded by chondroitinase ABC but not by chondrotinase AC, hyaluronidase, or proteolytic enzymes and thus appears to be similar to dermatan sulfate. The results of these studies raise the possibility that lectin-2 functions by interacting with glycosaminoglycans, either associated with the cell surface or with the extracellular matrix.