Modification of biopharmaceutical molecules by covalent conjugation of polyethylene glycol (PEG) molecules is known to enhance pharmacologic and pharmaceutical properties of proteins and other large molecules and has been used successfully in 12 approved drugs. Both linear and branched-chain PEG reagents with molecular sizes of up to 40 kDa have been used with a variety of different PEG derivatives with different linker chemistries. This review describes the properties of PEG itself, the history and evolution of PEGylation chemistry, and provides examples of PEGylated drugs with an established medical history. A trend toward the use of complex PEG architectures and larger PEG polymers, but with very pure and well-characterized PEG reagents is described. Nonclinical toxicology findings related to PEG in approved PEGylated biopharmaceuticals are summarized. The effect attributed to the PEG part of the molecules as observed in 5 of the 12 marketed products was cellular vacuolation seen microscopically mainly in phagocytic cells which is likely related to their biological function to absorb and remove particles and macromolecules from blood and tissues. Experience with marketed PEGylated products indicates that adverse effects in toxicology studies are usually related to the active part of the drug but not to the PEG moiety.

SummaryPatients with thrombotic thrombocytopenic purpura (TTP) have a deficiency of von Willebrand factor (vWF)-cleaving protease, whereas patients with hemolytic-uremic syndrome (HUS) show normal activity of this protease. Present methods for assaying vWF-cleaving protease by immunoblotting are time-intensive and cumbersome. We therefore developed a new functional assay based on the preferential binding of high-molecular-weight forms of vWF to collagen. In this assay, the diluted plasma sample to be tested is added to normal human plasma in which protease activity had been abolished. The vWF present in the protease-depleted plasma is digested by the vWF-cleaving protease in the test plasma. The proteolytic degradation leads to low-molecular-weight forms of vWF, which show impaired binding to microtiter plates coated with human collagen type III. The collagen-bound vWF is quantified using a peroxidase-conjugated rabbit antibody against human vWF. The values of vWF-cleaving protease activity in tested plasma samples are read from a calibration curve achieved by incubating the vWF-substrate with dilutions of a normal human plasma pool (NHP). Testing of plasma from patients with TTP and HUS showed that the assay can be used to distinguish between these two syndromes. The presence of an inhibitor can be detected by carrying out the test after incubation of NHP with the patient plasma sample, thus enabling differentiation of patients with familial TTP from those with non-familial TTP.

Memory B cells are responsible for the rapidly emerging antibody response after antigen reexposure. The signals required for the restimulation of memory B cells have not been fully explained. We used a murine model of anti-factor VIII (FVIII) antibody responses in hemophilia A to study the requirements for the restimula-tion of FVIII-specific memory B cells and their differentiation into anti-FVIII anti-body-producing cells. We were particularly interested in the significance of activated T cells and costimulatory interactions. Our results indicate that the re-stimulation of FVIII-specific memory B cells is strictly dependent on interactions with activated T cells. These activated T cells can be specific for either FVIII or third-party antigens. Restimulation by T cells specific for third-party antigens requires the presence of FVIII, indicating that signals induced by B-cell receptor (BCR) triggering and by interactions with activated T cells are important. The blockade of B7-1 or B7-2 as well as the blockade of CD40L inhibits the restimulation and differentiation of FVIII-specific memory B cells in vitro and in vivo. The interference with inducible costimulator-inducible costimulator ligand (ICOS-ICOSL) interactions, however, does not cause any modulation. As expected, the production of anti-FVIII antibodies by plasma cells is not dependent on any of the costimulatory interactions tested.

Human alpha1-antitrypsin (A1PI) is a well-known glycoprotein in human plasma important for the protection of tissues from proteolytic enzymes. The three N-glycosylation sites of A1PI contain diantennary N-glycans but also triantennary and even traces of tetraantennary structures leading to the typical IEF pattern observed for A1PI. Here we present an approach to characterize A1PI isoforms from human plasma and its PTMs by LC-ESI-MS and LC-ESI-MS/MS of peptides obtained by proteolytic digestion. The single cysteine residue of A1PI formed a disulfide bridge with free cysteine. The variability of the number of antennae and hence sialic acids on glycosylation site N107, which even contained minute amounts of tetraantennary structures, emerged as a major cause for the IEF pattern of A1PI. Only negligible amounts of triantennary structures were identified attached to N70, and exclusively diantennary structures were present on site N271 in each of the isoforms analyzed. Exoglycosidase digests revealed alpha2,6-linked neuraminic acids on diantennary N-glycans, and triantennary contained additionally one single alpha2,3-neuraminic acid per N-glycan, which, together with a fucose, formed a sialyl Lewis X determinant on the beta1,4-linked N-acetylglucosamine, as shown by 2-D-HPLC of pyridylaminated asialoglycans. Fucosylation of diantennary structures was marginal and of the core alpha1,6 type.