Peptide PEGylation: The Next Generation

Pharmaceutical Technology, Pharmaceutical Technology-05-01-2011, Volume 2011 Supplement, Issue 3

Linking peptides to polyethylene glycol, or PEGylation, has helped improve pharmaceutical therapeutics in several ways. A wave of new techniques is now ushering in further advances.

A number of peptides have been used to effectively treat diseases, such as Adagen (Sigma-Tau Pharmaceuticals), which treats severe combined immunodeficiency. However, the use of peptides continues to be hampered by their extremely short half-life. Peptides are small and are often cleared by the kidneys or mononuclear phagocyte system within minutes of administration (1). They are also susceptible to degradation by proteolytic enzymes.

The problems associated with peptides can be overcome by PEGylation, that is, linking peptides to polyethylene glycol (PEG). Once linked to a peptide, each PEG subunit becomes tightly associated with two or three water molecules, which have the dual function of rendering the peptide more soluble in water and making its molecular structure larger. As the kidneys filter substances according to size, the addition of PEG's molecular weight prevents the premature renal clearance undergone by small peptides. PEG's globular structure also acts as a shield to protect the peptide from proteolytic degradation, and reduces the immunogenicity of foreign peptides by limiting their uptake through the dendritic cells (1, 2). PEG itself is not immunogenic or toxic, and allows for lower doses and less-frequent administrations. In some instances, PEG can increase the circulating half-life of a peptide drug by more than 100 times (1, 3). This saves money and resources, promotes patient compliance, and reduces the development of toxicity, tolerance and allergic reactions compared with their non-PEGylated counterparts (4). In addition to improving the pharmacokinetic and pharmacodynamic properties of peptide drugs once inside the body, PEGylation can also aid drug delivery because PEGylated petides act as permeation enhancers for nasal drug delivery. Synthetic PEGylated glycoproteins have been used in lieu of viruses for targeted gene delivery (5,6).

Despite the benefits associated with PEGylation, there are many factors that must be considered, including the peptide's primary and secondary structures, molecular weight, the number and placement of the linked PEG moieties, PEG's size and shape, and conjugation chemistry. All of these can impact the final product's physical, chemical, and biological properties.

First-generation PEGylation techniques

Originally, monomethoxy PEG (mPEG), which has relatively clean chemistry because of the simplicity of its monofunctionality (CH3O-(CH2CH2O)n-CH2CH2-OH), was used for polypeptide modification. This "first-generation" PEG chemistry was used to create Adagen and Oncaspar (Enzon), where mPEG was linked to the N-terminal amino group or the alpha or epsilon amino groups of a lysine residue.

Lysines are among the more common amino acids comprising proteins—sometimes comprising 10% of a protein's overall amino acid sequence. Before PEG can be conjugated to an amino group, its terminal hydroxyl group must be activated by adding a functional group that is reactive with the peptide. This can be achieved by either acylating the PEG or alkylating the PEG. Most first-generation PEG chemistries used acylation, which generates a neutral amide. Alkylation, on the other hand, maintains a positively charged amine.

Most acylated PEGs are hydroxysuccinimidyl esters (-OSu) of carboxylated PEGs. The distance between the active ester (-COOSu) and the terminal ether in PEG can vary by up to four methylene units, and can greatly affect the reaction rate of PEG with both water and amino groups. Other acylating chemistries include activating the terminal PEG hydroxyl group with chloroformates or carbonylimidazole. These reagents tend to have a slower reaction rate than –Osu activated PEGs, and enable the reaction to be more easily stopped at the desired degree of modification, after attachment to the desired number of lysines.

If a positive charge is important for biological activity, then alkylation is the method of choice. One of the more common alkylated PEGs is PEG-aldehyde, which provides a permanent linkage after Shiff base formation and cyanoborohydride reduction. However, the reaction rate for the Shiff base formation is very slow—sometimes taking more than 24 h. When generating alkylating PEGs, the pH of the reaction is also critical for selectivity.

The problem. After these first-generation PEG-protein derivatives came to market, it quickly became evident that the technology's potential would only be fully realized when further methods of conjugation were developed. One major hindrance was the polydispersivity inherent in the PEG molecule; the presence of some PEGdiol—PEG dimmers, or PEG molecules that interact with each other to form complexes—even in monomethoxylated PEG, can yield unwanted cross-linked conjugates. Moreover, these firstgeneration chemistries are limited to low-molecular weight PEG, can generate unstable linkages, and are difficult to selectively modify. A population of modified peptides can contain a mixture of molecules with PEG attached to different lysines, as well as molecules with different numbers of linked PEGs. This variability in modification diminishes the purity of the finished product, impedes reproducibility, and can have pharmacological ramifications.

Second-generation PEGylation techniques

Since the first wave of PEGylated peptide drugs entered the market, many more methods for PEGylating peptides have been developed to confer PEG's advantages on molecules with varying properties and to overcome some of the problems associated with first-generation PEGylation.

Site-specific conjugation. One new technique is to modify the peptide, rather than the PEG (7). This may be necessary if the peptide lacks lysines, or if the lysine is located in an active site. A cysteine can be added, where desired, to generate site-specific PEGylation at places chosen to minimize interference with the peptide's biological function, while maximally decreasing the peptide's immunogenicity. PEG-maleimide, PEG-vinylsulfone, PEG-iodoacetamide, and PEG orthopyridyl disulfide are thiol reactive PEGs that have been created to PEGylate free cysteine residues. This approach has been used in a number of ways including: to make monoPEGylated human growth hormone analog, which could potentially be used to treat growth hormone deficiency and wasting in AIDS patients with an approximately 8-fold longer half-life compared with unmodified growth hormone (8); a long-acting highly potent interferon a-2 conjugate for certain cancers and viral conditions (not PEGIntron) with a 20-to 40-fold longer half-life than unmodified interferonalpha2 (IFN-a2) (9); and PEGylated recombinant human granulocyte macrophagecolony stimulating factor (GM-CSF) for neutropenia and other myeloid disorders.

Another means of achieving site-specific conjugation is to use an N-terminal serine or threonine if it exists. These residues can be converted to glyoxylyl derivatives by periodate oxidation. The N-terminal-introduced reactive carbonyl group specifically reacts, under mild acidic conditions, with an aminooxy-functionalized poly(ethylene glycol) to form a stable oxime bond. Unlike most previous methods, this approach places a single PEG chain at a defined site on the protein and should, therefore, be more likely to conserve biological activity. PEGylated interleukin8 (IL-8), G-CSF, and interleukin1 receptor agonist (IL-1ra) were made this way (10).

HIPEG. HiPEG (PolyTherics, UK) was developed to specifically attach PEG to histidine sequences expressed on the N or C terminal of proteins. After PEGylation, the histidine tag remains available for affinity purification of the protein or peptide. This technology has been shown to be scalable, robust and reproducible (9), and has been successfully used to PEGylate an anti-TNFa domain antibody, which can be used to treat autoimmune disorders, such as rheumatoid arthritis, inflammatory bowel disease, psoriasis, and refractory asthma (9).

Branched and forked PEG. Branched PEGs (PEG2) can be used in lieu of linear PEG molecules, and enable a larger and purer PEG to be linked with only one reactive group. This is particularly useful in cases where the incorporation of many PEG molecules could block a protein–protein interaction. The bulkier nature of branched PEGs helps to repel approaching macromolecules from a peptide's active site, thereby preserving biological activity. They are also more effective at protecting peptides from proteolysis, and reducing immunogenicity and antigenicity. Branched PEG was used to PEGylate lysostaphin, which is used to fight multidrugresistant strains of Staphylococcus aureus.

Forked PEG is the opposite of branched PEG. Rather than one functional group attached to two PEG chains, it has two reactive groups at the end of one PEG chain. Forked PEG can be used to bring two moieties into proximity.

Releasable PEGs. Releasable PEGs (rPEGs) may be desired if the PEG interferes with biological activity. With rPEGs, the PEG conjugation achieves its goal by protecting the peptide in circulation, but it is then discarded to release an active protein. PEG-Intron (originally made by Ezon, but now marketed by Merck), which is currently used to treat Hepatitis C, was made this way: Enzon conjugated PEG-SC to His34 of interferon a-2b using a carbamate linkage. Over time, this linkage degrades to yield native interferon a-2b. Enzymatic degradation, hydrolytic cleavage, or reduction can also be used to release PEG for rPEGs. To date, many rPEGs generated use a linker with an ester bond to control the rate of protein release.

Heterobifunctional PEG. Heterobifunctional PEGs have two different terminal groups. Because they are hydrophilic, flexible, and biocompatible, they can be used as a spacer to link two groups, or to target drugs, or even in peptide synthesis. Preferred end groups include NHS esters, maleimide, vinyl sulfone, pyridyl disulfide, amines, and carboxylic acids.

Effective PEGylation

The solutions presented herein provide many options to help one decide how to PEGylate a given peptide most effectively. However, there are three factors that must be taken into account because they can affect the efficiency of PEGylation: the number of chains attached to a peptide; the molecular weight and structure of the PEG chains; and the attachment site of PEG.

In recent work (11), a synthetic peptide inhibiting the binding of human neonatal Fc receptor to immunoglobulin was PEGylated in a number of different ways. In vitro and in vivo activity was then compared to determine which method was the most effective. Three different amino acids were either replaced with lysine, or a C-terminal amine was generated, or a linker amine was generated. The PEGylated peptides were more active in mice compared with the nonPEGylated version, but only when linked through a specific linker. The linear version was more active compared with the branched version. Because technology now exists to make different versions, a few should be tested for each specific peptide.

As PEGylation becomes more well used, such comparative studies may help optimize future PEGylation strategies. Additionally, as more peptides become drug candidates, testing different ways of linking them to PEG may become a routine part of fine-tuning them for their final clinical manifestations.

Baosheng Liu is vice-president of research and manufacturing at the American Peptide Company, 777 E. Evelyn Avenue, Sunnyvale, CA 94086.


1. G. Pasut and F.M. Veronese, Drugs of Today 45 (9), 687–695 (2009).

2. H.F. Gaertner and R.E. Offord, Bioconjug. Chem. 7 (1), 38–44 (1996).

3. M.J. Roberts, M.D. Bentley, and J.M. Harris, Advanced Drug Delivery Reviews 54 (4), 459–476 (2002).

4. V.M. Veronese, Biomaterials 22 (5), 405–417 (2001).

5. A.D. Sezer and A.F. Yagci, Acta. Pharmaceutica. Sciencia 52(4), 376–388 (2010).

6. C.P. Chen et al., Bioconjug. Chem. 18 (2), 371–378 (2007).

7. F.F. Schumacher, Bioconjug. Chem. 22 (2), 132–136 (2011).

8. G.M. Cox et al., Endocrinology 148 (4), 1590–1597 (2007).

9. M.S. Rosendahl, Bioconjug. Chem. 16 (1), 200–207 (2005).

10. D.H. Doherty, Bioconjug. Chem. 16 (5), 1291–1298 (2005).

11. A.R. Mezo et al., Bioorg. Med. Chem. 16 (12), 6394–405 (2008).