Poly(Ethylene Glycol) Chemistry

At first glance, the polymer known as poly(ethylene glycol) or PEG appears to be a simple molecule. It is a linear or branched, neutral polyether, available in a variety of MWs, and soluble in water and most organic solvents. Despite its apparent simplicity, this molecule is the focus of much interest in the biotechnical and biomedical communities. Primarily this is because PEG is unusually effective at excluding other polymers from its presence when in an aqueous environment. This property translates into protein rejection, formation of two-phase systems with other polymers, nonimmunogenicity, and nonantigenicity. In addition, the polymer is nontoxic and does not harm active proteins or cells although it interacts with cell membranes. It is subject to ready chemical modification and attachment to other molecules and surfaces, and when attached to other molecules it has little effect on their chemistry but controls their solubility and increases their size. These properties, which are described in more detail below, have led to a variety of important biotechnical and biomedical applications, a summary of which is also presented below.

As this volume attests, poly(ethylene glycol) or PEG is a material of growing importance in the biomedical world. It has been used in free solution as an agent for cell fusion¹ and protein precipitation.² It has also been conjugated to proteins and drugs to reduce immunological responses and control pharmacodynamics.³ Finally, it has been used in biocompatible materials, either as a coating or incorporated into a hydrogel.⁴ These surfaces are expected to be highly biocompatible because protein adsorption to them is low.^4,5 Both the amount of protein adsorption and the magnitude of other biochemical events, such as platelet adhesion, rapidly decline as the PEG molecular weight rises.^6,7 This decline is most marked at molecular weights up to 1000, after which the biointeractions tend to level out gradually.

Poly(ethylene glycol) (PEG) is a water-soluble polymer that exhibits properties such as protein resistance, low toxicity, and nonimmunogenicity.^1–5 These properties have been attributed to its segmental flexibility and its polar, but uncharged, chemical composition. This segmental flexibility produces a high degree of steric exclusion and entropy at PEG-water interfaces which in turn leads to protein resistance. Its exclusion property also enables the precipitation of proteins without denaturation.^6–9

In the last few years the use of poly(ethylene glycol) (or PEG) as an agent with which to modify the properties of macromolecules and surfaces has greatly increased, as witnessed by the contributions in this volume. In most instances, PEG is used because it exhibits the interesting property of being highly compatible with water (i.e., highly water soluble) while exhibiting strong incompatibility with a wide variety of other water-soluble substances. Incompatibility means that an unfavorable free-energy change occurs when a second species interacts with a solvated PEG molecule, resulting in a statistical tendency for the second species to be excluded from the region within or near the PEG chain. Such excluded volume effects are manifested in a variety of ways, including phase separation in mixtures with a second water-soluble polymer of salt, enhanced exclusion of PEG from chromatographic gel beads relative to other polymers of similar molecular weight, protein precipitation and reduced binding of external proteins to surfaces or molecules derivatized with PEG.^1,2 Many of these kinds of interactions are described in this book.

Water and other liquids can be divided into two fluid compartments or phases (in direct contact with each other) by using them as solvents for poly(ethylene glycol) (PEG) together with another polymeric substance.¹ The incompatibility of polymers in solution that gives rise to this phase separation also has the consequence that the two polymers are accumulated in opposite phases. The difference in polymer structures and polymer concentrations in the phases may cause significant divergences in the solvating properties for high molecular weight substances added in low concentration. When water is used as the solvent, added salts partition more or less evenly between the phases. Proteins, on the other hand, partition more unequally. The actual partition of a substance is described by the partition coefficient, K, which is defined as the ratio of the concentration of partitioned substance between the upper and lower phase. The most popular aqueous two-phase systems for partitioning of biological substances have been the ones containing PEG and dextran.^2–4 The top phases of these systems contain, besides water, mainly PEG (5–15%) while the bottom phases contain dextran (10–25%) and some PEG (0.2–2%). Three PEG-dextran systems which have identical phase compositions are shown in Figure 1.

Aqueous polymer two-phase systems are increasingly being used in biochemistry and cell biology for the separation of macromolecules, membranes, cell organelles, and cells.^1,2 The great interest in aqueous phase partitioning is due to the unique separation properties of the systems and the mild conditions during the separation process. The unique properties of the systems make them also very interesting for large-scale industrial applications. In the biotechnical industry this technique is starting to be used for large-scale enzyme extractions.^3,4 Many applications of aqueous polymer two-phase systems in biotechnology are currently being explored, both for separations of biomolecules, cell organelles, and cells, and for bioconversions.

Chemical modification of proteins by attachment of poly(ethylene glycol) (or PEG) is of much interest in medical applications,^1,2 enzymatic organic synthesis,³ and affinity separations.⁴ Some of the desired properties obtained by PEG modification are reduced antigenicity, control of partitioning of affinity ligands, and solubility in organic solvents.

Enzymes are naturally designed catalysts with high efficiency and specificity. To know how to design such enzymes is a major goal for enzymologists, and they have tried to elucidate the structural basis of enzyme catalysis. Information on structure-function relationships has made it possible to design or redesign enzymes by protein engineering and also to design artificial enzymes.^1–3

The therapeutic value of enzymes as drugs would be considerably increased if drawbacks such as immunogenicity and antigenicity, rapid clearance from circulation, difficulty in targeting, instability, and inadequate supply were overcome. Genetic engineering seems to be promising in obtaining large amounts of useful enzymes, although doubts exist concerning the correct folding of expressed proteins.¹ Nevertheless, the disadvantages of limited tissue distribution and rapid clearance from circulation of enzymes remain a major problem. Moreover, genetic methods cannot be used for producing enzymes carrying post-transcriptional modifications, but with this aim the so-called transgenic animal technology appears to be quite promising. For these reasons, alternative strategies are being actively investigated in several laboratories. These strategies include surface modification of the enzymes by chemical modification or compartmentalization of the enzyme onto complex structures which isolate it from body cells, tissues, and proteolytic enzymes.²³

The possibility of selectively downregulating the host’s immune response to a given antigen represents one of the most formidable challenges of modern immunology in relation to the development of new therapies for IgE-mediated allergies, autoimmune diseases, and the prevention of immune rejection of organ transplants. Similar considerations apply to an increasing number of promising therapeutic modalities for a broad spectrum of diseases, which would involve the use of foreign biologically active agents potentially capable of modulating the immune response, provided they were not also immunogenic. Among these agents, one may cite (1) xenogeneic monoclonal or polyclonal antibodies (collectively referred to here as xIg) against different epitopes of the patients’ CD4⁺ cells,^1,2 administered alone or in combination with immunosuppressive drugs for treatment of rheumatoid arthritis and other autoimmune diseases, or for the suppression of graft versus host reactions and of the immune rejection of organ transplants,^1,2 and (2) “magic bullets” for the destruction of tumor cells,^3–5 which consist of anti-tumor xIg to which are coupled toxins (Tx), or radionuclides, or chemotherapeutic drugs. However, in most cases the patients produce antibodies to the injected xIg and to the even more immunogenic immunotoxins (xIg-Tx); consequently, the therapeutic effectiveness of these immunological strategies is undermined by the patients’ antibodies which prevent these “bullets” from reaching their target cells. In addition, the repeated administration of these agents may result in serious complications, namely, serum sickness, anaphylactic symptoms (i.e., bronchospasm, dyspnea, and hypotension), and/or the deposition in the liver of toxic immune complexes leading frequently to hepatotoxicity.^6,7 Similar limitations apply to the use of hormones or other regulatory factors, such as lymphokines and growth factors, synthesized by recombinant DNA technology, which are often immunogenic⁸ probably because of small differences in their conformational characteristics or in their glycosidic constituents in relation to their natural, human counterparts.

Bilirubin, the end product of heme catabolism, is generally regarded as toxic and highly fatal in newborn infants and fulminant hepatitis. Bilirubin encephalopathy (kernicterus) is usually considered to be caused by the entry of circulating, free (albumin-unbound), unconjugated bilirubin into the cerebral tissue.^1,2 Bilirubin conjugation with glucuronic acid takes place in the liver and the process is impaired in liver diseases. Our tactic for the treatment of jaundice is to decompose toxic bilirubin by the enzyme bilirubin oxidase, and for that purpose the enzyme is made into a polymer conjugate to improve its pharmacological properties.³

A need is strongly recognized for a safe red cell substitute that can carry oxygen to the hypoxic tissues of an anemic body. Banked blood is in short supply because of mounting fears of donors and limited storage life. A safe, stable, oxygen carrier would eliminate many problems such as cross-matching of blood types, danger of virus infections, short shelf life, and availability. Not only could this be used as an oxygen-carrying plasma expander for trauma victims and patients in surgery, such as cardiac bypass and angioplasty, this product could also be used as a perfusate for the preservation of isolated organs for transplantation. Cold storage is the popular choice for preservation at present, but there is an increasing demand for a safe organ perfusate that would promote the use of perfusion preservation to obtain a longer period of preservation.

Currently much interest has been shown in attaching poly(ethylene glycol) (PEG) to proteins as a means of increasing solubility and serum lifetime.^1–3 Modification of proteins by PEG has been used to reduce the immunogenicity of various enzymes.^4–6 PEG is well known to be nontoxic, nonantigenic, biocompatible, and soluble in water and organic solvents.^7–9

This brief resumé is a personal view, offered as a historical narrative to complement the other contributions to this volume dealing with the role of PEO in blood-surface interactions.

Poly(ethylene oxide), or as it is frequently denoted in the literature, poly-(ethylene glycol) (PEG), is a nonionic, water-soluble polymer widely used for stabilizing colloids in food and paints and in formulating pharmaceuticals and cosmetics. The reason for the extensive use of this polymer is that it acts as a dispersant and yet is inert, e.g., it does not interfere adversely with other functional ingredients in the dispersion.

Poly(ethylene oxide) (or PEO) surfaces represent an important class of biomaterials because of their low capacity for protein adsorption. A potentially wide range of applications exists for PEO surfaces including blood contacting devices, drug delivery systems, contact lenses, intraocular lenses, vascular grafts, catheters, immunoassays, biosensors, and media for protein and cell separations, to name a few.

Poly(ethylene glycols) (or PEGs) have been utilized in various forms in pharmaceutical preparations for many years. They are used as additives to creams, as solubilizing agents, and as components of injectable formulations. In the form of surface active agents they are widely used in the formulation of a variety of products in both the foodstuffs and in the pharmaceutical industries. Their toxicity is considered to be adequate for many uses on or in the body. They thus provide a very attractive group of materials for the synthesis of novel “tailor-made” polymers for drug delivery or other therapeutic use. In the form of copolymers these materials have been licensed for use as both drug delivery and wound healing. This chapter describes work done to synthesize new and useful polymers based on this class of starting materials.

Thrombus formation is a serious problem in surgical therapy and clinical application of artificial organs. Therefore, the need for the creation of highly antithrombogenic biomaterials has been increasing. Although a substantial amount of work in the improvement of the blood compatibility of polymeric materials has been carried out, the results are still inconclusive. This is caused partly by the fact that the relationship between surface properties and surface-induced thrombosis has not been thoroughly evaluated.

Grafting of poly(ethylene glycol) (or PEG) to solid surfaces has been recognized as a technique for obtaining low protein adsorption and low cell adhesion characteristics.^1,2 For instance, PEG coating is reported to give a marked suppression of plasma protein adsorption and platelet adhesion leading to reduced risk of thrombus formation, as demonstrated both in vitro and in vivo.^3–5

Poly(ethylene glycol) (or PEG) is very compatible with peptides and proteins. It is soluble in water and almost all organic solvents, with the exception of aliphatic hydrocarbons and ether. This polymer was shown to be a valuable support for peptide and nucleotide synthesis in homogeneous solution (liquid-phase method^1–5) as an alternative to the solid-phase method of Merrifield.¹³ In general, PEG of molecular masses 3000–20,000 daltons are used in liquid-phase peptide synthesis. Even insoluble free peptides often are solubilized, if covalently linked to PEG. On the other hand, the conformation of the peptide bound to PEG is the same as the conformation of the free peptide in the same solvents. A synthetic cycle using the liquid-phase procedure is shown in Scheme 1. The couplings are carried out in homogeneous solution.

The unique properties of poly(ethylene glycol) (or PEG) and its general compatibility with polypeptide materials facilitated development of a variety of different applications of this polymer.^1–11 A marked proportion of these applications involve the use of covalently linked polypeptide-PEG adducts (reviewed elsewhere^5–11). For example, a number of PEG-enzymes were shown to be useful as catalysts, soluble and active in organic solvents.⁷ Due to the affinity to the upper phase of PEG/Dextran and PEG/salt two-phase systems, PEG-modified proteins were proven useful both as diagnostic tools⁸ and in preparative separations of biological cells.⁹

The chapters of this book describe the synthesis and use of a variety of active PEG derivatives designed to couple PEG to other materials. Despite the availability of these derivatives, there remains a need for new derivatives with presently unavailable properties and work continues in this area. Desirable properties include selectivity, stability, and ease of preparation. For example, it would be desirable to have derivatives that react with nucleophilic groups on proteins, but which do not react with water. Derivatives of a wide range of reactivities are always in demand. Similarly, there would be advantages to having derivatives that react with groups other than the commonly used amino groups. And, of course, the need for derivatives that can be prepared cheaply and easily in large quantity is critical for commercialization of the many biomedical and biotechnical applications of PEG chemistry.