Characterization of biosynthetic IgG oligomers resulting from light chain variable domain duplication.

We have characterized a human IgG1 monoclonal antibody composed of altered light chains. Each light chain consists of two identical variable domains and a kappa constant domain, in association with a normal gamma chain. This antibody assembled biosynthetically into a mixture of stable oligomers and monomers. Employing gel filtration, PAGE, and electron microscopy, we examined the antibody and the nature of the associations involved in oligomer formation. By engineering a protease factor Xa site between the duplicated light chain variable domains and examining the fragments produced following factor Xa cleavage, we demonstrated the association of the IgG monomers occurred through their duplicated VL domains. Electron microscopy showed the oligomeric antibody to be predominantly dimers and trimers in which the monomeric units were associated through the tips of the Fab portion of the antibody, presumably through the protruding N-terminal VL domains. Similar examination of monomers demonstrated several molecular forms, including individual molecules with self-crosslinked Fab arms and others displaying the open Y and T shapes typically observed for IgG antibodies. The monomers also displayed distally protruding domain-like structures. The oligomers produced by this cell line therefore occurred through the noncovalent interaction between the extra light chain variable domains.

Stability of thiotepa (lyophilized) in 0.9% sodium chloride injection.

The stability of thiotepa in a new formulation of the drug was studied. Vials of Thioplex (Immunex), a relatively new lyophilized formulation of thiotepa, were reconstituted with sterile water and diluted with 0.9% sodium chloride injection in polyvinyl chloride infusion bags to thiotepa concentrations of 0.5, 1, and 3 mg/mL. The solutions were stored at 8 and 25 degrees C in ambient light and analyzed at 0, 8, 24, and in most cases 48 hours for thiotepa concentration and chloro-adduct formation by stability-indicating high-performance liquid chromatography. Thiotepa 1 and 3 mg/mL was stable for 48 hours at 8 degrees C and for 24 hours at 25 degrees C. Thiotepa 0.5 mg/mL was not stable at either temperature. Storage at 8 degrees C slowed but did not prevent chloro-adduct formation and loss of potency. The pH tended to increase with time; turbidity remained low. Thiotepa (lyophilized) 1 and 3 mg/mL in 0.9% sodium chloride injection was stable for 48 hours at 8 degrees C and for 24 hours at 25 degrees C; the drug was unstable when diluted to 0.5 mg/mL and stored under the same conditions.

Homogeneous enzyme immunoassay (EMIT) protocol for monitoring tricyclic antidepressants on the COBAS-BIO centrifugal analyzer.

EMIT tests are available for quantitative determination of the tricyclic antidepressants amitriptyline (AMI), nortriptyline (NORT), imipramine (IMI), and desipramine (DMI). An extraction step before analysis eliminates cross-reacting polar metabolites. Excellent correlation between EMIT and high-performance liquid chromatography (HPLC) has been previously established. Most published protocols for using EMIT reagents on COBAS-BIO centrifugal analyzers were designed for analytes present at 1-50 mg/L in serum. AMI, NORT, IMI, and DMI are usually present at far less than 1 mg/L. We describe a COBAS-BIO EMIT protocol for assaying these analytes. Patient sample correlation between COBAS-BIO EMIT and EMIT-AutoLab were greater than 0.99. Between-run precision for single-point determinations was comparable to SYVA AutoLab performance [less than or equal to 11% at 40 micrograms/L AMI, NORT, IMI (80 micrograms/L DMI), and less than or equal to 4% at 200 micrograms/L AMI, NORT, IMI (400 micrograms/L DMI)]. With stored-curve updating, working reagents were usable for at least 14 days (AMI) or 23 days (DMI).

Quantitative homogeneous enzyme immunoassays for amitriptyline, nortriptyline, imipramine, and desipramine.

We describe specific EMIT homogeneous enzyme immunoassays for amitriptyline, nortriptyline, imipramine, and desipramine in patients' serum samples. Before analysis, an easily performed extraction step involving the use of 500 microL of sample and a 1-mL disposable column eliminates cross-reacting polar metabolites. The range of the standard curve for the first three drugs is 25 to 250 micrograms/L, and for desipramine is 50 to 500 micrograms/L. Within-run and between-run CVs are less than 10% throughout the range of the assays. Results for patients' samples obtained by this method and by "high-performance" liquid chromatography compare well, showing a slope range of 0.94-1.04 and correlation coefficients ranging from 0.93 to 0.96, depending on the assay.

Controlled release of TGF-beta 1 from a biodegradable matrix for bone regeneration.

Although bone has a remarkable capacity for regenerative growth, there are many clinical situations in which the bony repair process is impaired. TGF-beta 1 is a 25 kD homodimeric protein which modulates the growth and differentiation of many cell types. The ability of TGF-beta 1 to promote bone formation suggests that it may have potential as a therapeutic agent in disease of bone loss. However, there still exists a need for an effective method of delivering TGF-beta 1 to the site of an osseous defect for the promotion of bone healing. This paper describes a novel biodegradable controlled release system for TGF-beta 1 comprised of poly (DL-lactic-co-glycolic acid) (PLPG) and demineralized bone matrix (DBM). The amount and activity of TGF-beta 1 released was determined using several methods including 125I-labeled TGF-beta 1 as a tracer, an enzyme linked immunosorbent assay (ELISA) and a growth inhibitory assay (GIA). Protein was released from the devices for time periods of more than 600 h. The amount of TGF-beta 1 released was directly proportional to both the TGF-beta 1 loading and the weight percent of DBM in the device. The release kinetics could be further controlled by applying polymeric coatings of varying porosity to the devices. The GIA indicated that between 80 and 90% of the TGF-beta 1 released from the delivery system retained its bioactivity. The PLPG and DBM existed in phase separated domains within the device as determined by differential scanning calorimetry. Scanning electron microscopy suggested that the devices were sufficiently porous to allow bone ingrowth.

Stimulation of bone healing by transforming growth factor-beta 1 released from polymeric or ceramic implants.

The ability to transforming growth factor-beta 1 (TGF-beta 1), to stimulate bone healing was evaluated in a rat critical calvarial defect model. Both a low dose and a high dose of TGF-beta 1 were incorporated into two different types of implants: one made from a composite of poly(lactic-co-glycolic acid) (PLPG) (50:50) and demineralized bone matrix (DBM), and the other from calcium sulfate (CaSO 4). Scanning electron microscopy showed that the CaSO 4 implants were more porous than the PLPG/DBM samples. Both types of implants released biologically active TGF-beta 1 for over 300 h in vitro. The samples were implanted in a 9-mm diameter rat calvarial defect for 6 weeks along with contralateral control implants containing no TGF-beta 1. Microradiography and histological analysis were used to assess the bone healing in the defects. Microradiography revealed that the greatest amount of calcified bone (67.5%) was present in in the CaSO 4 implants containing a high dose of TGF-beta 1 while minimal new bone formation occurred in the PLPG/DBM implants. Histologically, the PLPG/DBM implants exhibited an inflammatory response with little mineralization or bone formation. The defects containing the PLPG/DBM implants consisted of a connective tissue stroma with large void spaces. Giant cells and numerous polymorphonuclear leukocytes were present throughout the implants. In contrast, the CaSO 4 implants had only a few inflammatory cells and the presence of mineralization and true bone was a more consistent feature. These preliminary studies show that TGF-beta 1 is capable of inducing new bone formation. Furthermore, the materials used to deliver the growth factor can play a significant role in the bone healing process.

The enhancement in wound healing by transforming growth factor-beta 1 (TGF-beta 1) depends on the topical delivery system.

Transforming growth factor-beta 1 (TGF-beta 1) has beneficial effects on wound healing. However, the ideal method for its administration to the wound site remains unknown. Our aim was to analyze the release of TGF-beta 1 from different formulations and to study whether the changes in wound healing by TGF-beta 1 depend on its topical delivery system. For the studies the TGF-beta 1 was incorporated into phosphate-buffered saline, into a polyoxamer gel, into DuoDERM hydroactive paste, and into a poly(ethylene oxide) hydrogel. The release of 125I-labeled TGF-beta 1 from carriers was measured in full-thickness wounds in rats and the healing of the wounds was analyzed by histology and wound area measurements. The TGF-beta 1 was released from all formulations at a different rate and in an active form as determined by growth inhibition assay. Wound size measurements and the analysis on the amount of cellular influx, fibroplasia, and granulation tissue showed that a single dose (1 microgram/wound) of locally administered TGF-beta 1 significantly (P < 0.01) enhanced the wound healing. This effect was most prominent with polyoxamer gel formulation, which provided the most sustained release of TGF-beta 1. Our finding that the enhancement in wound healing by TGF-beta 1 was significantly dependent on the carrier used for its topical delivery to the wound site is novel and shows the importance of using adequate delivery systems when growth factors are used to enhance wound repair.

The stabilization of a human IgM monoclonal antibody with poly(vinylpyrrolidone).

An IgM anti-group B Streptococcus monoclonal antibody (4B9) was found to undergo irreversible heat-induced aggregation at 50 degrees C. A variety of excipients was tested for their ability to inhibit antibody aggregation. The amount of 4B9 aggregation, which was determined by analysis on a size-exclusion HPLC, was significantly reduced in the presence of low concentrations [between 0.1 and 1.0% (w/v)] of poly(vinylpyrrolidone) (PVP) molecules ranging in molecular weight from 10 to 40 kDa. When the PVP concentration was greater than 1.0%, antibody aggregation was enhanced, and with the highest molecular weight PVP, antibody precipitation occurred. HPLC was used to show that more PVP was associated with the 4B9 at 50 degrees C than at 25 degrees C. Differential scanning calorimetry revealed that PVP concentrations greater than 2.0% decreased the antibody thermal transition temperature. Enzyme-linked immunosorbent assays were used to assess the effects of PVP on the antigen binding capacity of 4B9 and on 4B9 quantitation. At 4 degrees C, PVP solutions of up to 5.0% had no effect on either 4B9 quantitation or antigen binding. At 50 degrees C, however, less 4B9 was detected in the 5.0% PVP solution. The heat stabilization of the 4B9 antibody by low concentrations of PVP can be explained by a weak binding of PVP to the native protein. The PVP may sterically interfere with protein-protein interactions, thus reducing aggregation. Higher concentrations of PVP lead to protein aggregation and precipitation, probably by a volume-exclusion mechanism.(ABSTRACT TRUNCATED AT 250 WORDS)

Characterization of poly(glycolide-co-D,L-lactide)/poly(D,L-lactide) microspheres for controlled release of GM-CSF.

PURPOSE: This study describes the preparation and characterization of a controlled release formulation of granulocyte-macrophage colony-stimulating factor (GM-CSF) encapsulated in poly(glycolide-co-D,L-lactide) (PLGA) and poly(D,L-lactide) (PLA) microspheres. METHODS: GM-CSF was encapsulated in PLGA/PLA microspheres by a novel silicone oil based phase separation process. Several different blends of PLGA and low molecular weight PLA were used to prepare the microspheres. The microspheres and the encapsulated GM-CSF were extensively characterized both in vitro and in vivo. RESULTS: Steady release of GM-CSF was achieved over a period of about one week without significant "burst" of protein from the microspheres. Analysis of microsphere degradation kinetics by gel permeation chromatography (GPC) indicated that low molecular weight PLA enhanced the degradation of the PLGA and thereby affected release kinetics. GM-CSF released from the microspheres was found to be biologically active and physically intact by bioassay and chromatographic analysis. Analysis of serum from mice receiving huGM-CSF indicated that the GM-CSF was biologically active and that a concentration of greater than 10 ng/mL was maintained for a period lasting at least nine days. MuGM-CSF was not detected following in vivo administration of muGM-CSF microspheres. The tissues of mice receiving muGM-CSF microspheres were characterized by infiltration of neutrophils, and macrophages which were in significant excess of those found in mice administered with placebo controls (i.e. microspheres without GM-CSF). CONCLUSIONS: This study demonstrates the influence of formulation parameters on the encapsulation of GM-CSF in PLGA/PLA microspheres and its controlled release in biologically active form. The intense local tissue reaction in mice to muGM-CSF microspheres demonstrates the importance of the mode of delivery on the pharmacologic activity of GM-CSF.