Polymer–drug conjugation, recent achievements and general strategies

ARTICLE IN PRESS

0079-6700/$ - se

doi:10.1016/j.pr

Abbreviations

CT-2103, PGA

and Drug Adm

factor; HPMA,

IV, intravenous

NCS, neocarcin

PAcM, poly(ac

hydroxyethyl)-L

(Gly–Phe–Leu–

polyvinylalcoho

SS–PEG, PEG

vinylsulfone; w�CorrespondE-mail addr

Prog. Polym. Sci. 32 (2007) 933–961

www.elsevier.com/locate/ppolysci

Polymer–drug conjugation, recent achievementsand general strategies

G. Pasut, F.M. Veronese�

Department of Pharmaceutical Sciences, University of Padua, Via F. Marzolo 5, 35131-Padova, Italy

Received 17 April 2007; received in revised form 23 May 2007; accepted 24 May 2007

Available online 5 June 2007

Abstract

The field of drug delivery is fast expanding and its potentials have already been proved by the many products in the

market. Among all approaches, polymer conjugation is a well known and widely exploited technique useful to improve

therapeutic properties of peptides, proteins, small molecules or oligonucleotides. Polymer-conjugated drugs generally

exhibit prolonged half-life, higher stability, water solubility, lower immunogenicity and antigenicity and often also specific

targeting to tissues or cells. This technology, exploited for the first time in the fifties and sixties, received a great

development both for the introduction and study of new and different polymers and for the progresses in the chemical

strategies of coupling. Polymer–drug conjugates are already in the market for the treatment of different diseases,

demonstrating the potentials of the technology. Furthermore, new polymers, in addition to the most known N-(2-

hydroxypropyl)methacrylamide copolymer (HPMA), polyglutamic acid (PGA) and poly(ethylene glycol) (PEG), are

continuously investigated and proposed. The review will discuss the most recent achievements in polymer conjugation with

special emphasis on PEG application strategies and approved products.

r 2007 Elsevier Ltd. All rights reserved.

Keywords: Polymer conjugation; Drug-delivery; PEG; PEGylation; Polymer therapeutics

e front matter r 2007 Elsevier Ltd. All rights reserved.

ogpolymsci.2007.05.008

: A, Amstrong; AL–PEG, PEG aldehyde; BTC–PEG, PEG benzotriazolyl carbonate; CDI–PEG, PEG carbonylimidazole;

–paclitaxel conjugate; Da, Dalton; EGF, epidermal growth factor; EPR, enhanced permeability and retention; FDA, Food

inistration; G-CSF, granulocyte colony stimulating factor; HCV, extrahepatic hepatitis C virus; hGH, human growth

N-(2-hydroxypropyl)methacrylamide copolymer; IA–PEG, PEG–iodoacetamide; IFN, interferons; IL2, interleukine-2;

; kDa, kilo Dalton; MAL–PEG, PEG–maleimide; mPEG, monomethoxy-poly(ethylene glycol); MW, molecular weight;

ostatin; NSCLC, non-small cell lung cancer; NHS, N-hydroxysuccinimide; OPSS–PEG, PEG–orthopyridyl-disulfide;

roloylmorpholine); PEG, poly(ethylene glycol); PEI, poly(ethyleneimine); PGA, polyglutamic acid; PHEG, poly((N-

-glutamine); PK1, HPMA–(Gly–Phe–Leu–Gly–doxorubicin)n conjugate; PK2, (N-acylated galactosamine)m–HPMA–

Gly–doxorubicin)n conjugate; pKa, �log(dissociation constant); pNPC–PEG, PEG p-nitrophenyl carbonate; PVA,

l; PVP, poly(vinylpyrrolidone); SC–PEG, PEG succinimidyl carbonate; SMA, poly(styrene-co-maleic acid/anhydride);

succinimidyl succinate; TCP–PEG, PEG trichlorophenyl carbonate; TGase, transglutaminase; VS–PEG, PEG–

t%, weight percent

ing author. Tel.: +39049 8275694; fax: +39 049 8275633.

esses: [email protected] (G. Pasut), [email protected] (F.M. Veronese).

www.elsevier.com/locate/ppolysci

dx.doi.org/10.1016/j.progpolymsci.2007.05.008

mailto:[email protected]

mailto:[email protected]

ARTICLE IN PRESSG. Pasut, F.M. Veronese / Prog. Polym. Sci. 32 (2007) 933–961934

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 934

1.1. Proteins–polymer conjugates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 935

1.2. Low molecular weight drug–polymer conjugates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 936

2. Polymers for bioconjugation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 937

2.1. Vinyl polymers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 938

2.2. Poly(amino acids) and analogues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 938

2.3. Polysaccharides. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 939

2.4. Poly(styrene-co-maleic acid/anhydride) (SMA). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 939

3. Poly(ethylene glycol) (PEG) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 939

3.1. PEG–small drug conjugates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 942

3.2. Protein PEGylation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 945

4. Innovative PEGylation procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 950

5. Selected examples of PEGylated proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 952

6. Aptamer PEGylation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 954

7. Cells pegylation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 954

8. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 955

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 955

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 955

1. Introduction

Biotech-derived drugs are increasingly used in theclinical practice and represent today an importantshare of the research and development budget ofbiopharmaceutical companies. Nowadays, thanksto the development of pharmaceutical biotechnolo-gies, proteins and peptides are becoming potent andspecific therapeutic agents, useful as replacementtherapy or as inhibitors or regulators of the immunesystem for the treatment of important multifactorialdiseases. However, these products still possess manyintrinsic limitations to large-scale applications, suchas low stability in vivo, short half-life and immuno-genicity. The innovation in the design and produc-tion of biotech drugs, by recombinant DNAtechniques, has been paralleled by importantdiscoveries and improvements in the field of drugdelivery, necessary to extend the half-life of thesefragile products and to avoid the rapid clearanceobserved after their systemic administration. Dif-ferent drug delivery systems have been developed inthe last few years to improve the pharmacokineticand pharmacodynamic profile of such compounds[1]. These approaches are based on the preparationof more favorable genetic variants or on tailor-madeformulations of the drug, as liposomal preparations,controlled release systems, covalent modifications ofthe drug by low molecular weight reagent or bypolymer conjugation. The last one is a fast growingtechnique that already produced several molecules

available in the market [2], as shown in Table 1. Therationale for polymer conjugation is the possibilityto prolong the plasma half-life of therapeuticallyactive agents by increasing their hydrodynamicvolume and hence decreasing their excretion rate.Furthermore, polymer chains can prevent theapproach of antibodies, proteolytic enzymes or cellson conjugated molecules, an effect obtained by thesteric hindrance of polymer strands.

Immunogenicity is likely to be one of the mostserious problems, especially when dealing withheterologous proteins that commonly cause adverseresponse when recognized as non-self by the bodyimmune system. This problem was addressed also inthe first two papers of protein polymer conjugationhighlighting, as an important result, the immuno-genicity reduction of bovine albumin and catalaseafter PEG coupling [3,4]. The prevention ofimmunogenicity can be attributed to the shieldingeffect of polymeric chains surrounding the protein.This steric hindrance prevents interaction of anti-bodies or degrading enzymes with the protein(Fig. 1). In general, the conjugation of hydrophilicpolymers deeply changes the behavior of the parent(free) compound both in vitro and in vivo. Thischange happens with both proteins and lowmolecular weight agents. Some advantages are (i)increased water solubility (important for very lowsoluble molecules as taxol, camptothecin or plati-nate derivatives); (ii) enhanced bioavailability andprolonged plasma half-life due to the increased

ARTICLE IN PRESS

Table 1

Protein (A), oligonucleotide (A) and non-peptide (B) drugs polymer conjugates present on the market or in clinical trials

Conjugates Indication Year to

market or

status

Company

(A) High molecular weight drugs

SMANCS (Zinostatin, Stimalamer) [60] Hepatocellular carcinoma 1993 Yamanouchi

Pharmaceutical

PEG–adenosine deaminase (Adagens) [97] SCID sindrome 1990 Enzon

PEG–asparaginase (Oncaspars) [98] Acute lymphoblastic leukaemia 1994 Enzon

Linear PEG–interferon a2b (PEG–Introns)

[99,168]

Hepatitis C, clinical evaluation on cancer,

multiple sclerosis and HIV/AIDS

2000 Schering Plough/Enzon

Branched PEG–interferon a2a (Pegasyss)[100] Hepatitis C 2002 Roche/Nektar

PEG–growth hormone receptor antagonist

(Pegvisomant, Somaverts) [102]

Acromegaly 2002 Pfizer (Pharmacia)

PEG–G-CSF (Pegfilgrastim, Neulastas) [103] Prevention of neutropenia associated with

cancer chemotherapy

2002 Amgen

Branched PEG–anti-VEGF aptamer

(Pegaptanib, MacugenTM) [157]

Age-related macular degeneration 2004 EyeTech Pharmaceuticals

(now OSI

Pharmaceutical)/Pfizer

PEG–anti-TNF Fab (CDP870; Certolizumab

pegol, Cimzia) [169,170]

Rheumatoid arthritis and Crohn’s disease 2008 UCB (formerly Celltech)

A PEGylated diFab antibody. Targets

VEGFR-2 (CDP791)

Solid tumors Phase II UCB-ImClone System

(B) Low molecular weight drugs

HPMA copolymer–doxorubicin (PK1;

FCE28068) [44,45]

Cancer, in particular lung, breast cancers Phase II Pfizer (CRC/Pharmacia)

HPMA copolymer–doxorubicin–galactosamine

(PK2; FCE28069) [46]

hepatocellular carcinoma Phase I/II Pfizer (CRC/Pharmacia)

HPMA copolymer–camptothecin (MAG–CPT;

PNU166148) [14]

Clinical evaluation on several solid cancers Phase I Pfizer (Pharmacia)

HPMA copolymer–paclitaxel (PNU166945)[15] Clinical evaluation on several solid cancers Phase I Pfizer (Pharmacia)

HPMA copolymer–platinate (AP5280)[47] Clinical evaluation on several solid cancers Phase II Access Pharmaceutical

HPMA copolymer–platinate (AP5346) Clinical evaluation on several solid cancers Phase I/II Access Pharmaceutical

Polyglutamate–paclitaxel (XYOTAX; CT-

2103) [10–13,170]

Cancer, in particular lung, ovarian and

esophageal cancers

Phase II/III Cell Therapeutics

Polyglutamate–camptothecin (CT-2106)

[10–13,171]

Clinical evaluation on colorectal, lung, ovarian

cancer

Phase I/II Cell Therapeutics

PEG–camptothecin (PROTHECAN) [172] Clinical evaluation on several solid cancers Phase II Enzon

PEG–paclitaxel [173] Clinical evaluation on several solid cancers Phase I Enzon

G. Pasut, F.M. Veronese / Prog. Polym. Sci. 32 (2007) 933–961 935

hydrodynamic volume that reduces the kidneyclearance; (iii) protection towards degrading en-zymes; (iv) prevention or reduction of aggregation,immunogenicity and antigenicity; and (v) specificaccumulation in organs, tissues or cells, byactively targeted polymers or exploiting theknown ‘‘enhanced permeability and retention(EPR) effect’’ [5].

1.1. Proteins– polymer conjugates

In the field of protein–polymer conjugation, thefirst pioneering studies were carried out by Torchilinet al. [6] with dextran. There since, a number of

alternative polymers have been explored for poly-mer conjugation and poly(ethylene glycol) (PEG)emerged as the best candidate for protein modifica-tion [3,4]. Indeed, its leading position reflects thefact that the majority of conjugates on the marketare PEGylated products and that many others arealready in advanced clinical investigation (Table 1).It is evident from Table 1 that polymer conjugatesare applicable for the treatment of several severediseases, demonstrating that the potentials of thistechnique are not restricted to few therapeutic areas.The great economic success of these products willalso be a driving force for the development of newpolymeric drugs (Table 2).

ARTICLE IN PRESS

Antibodies and

degrading enzymes

Polymer

chain

Conjugated protein

Fig. 1. Protein surface shielding effect offered by conjugated polymer chains.

Polymeric

backbone

Spacer Drug

Targeting moiety

Solubilizingresidue

Fig. 2. Small drug–polymer conjugate model according to

Ringsdorf [8].

Table 2

Worldwide sales of most important PEGylated proteins [174]

Product Worldwide

sales, 2005 in

US $M

Worldwide

sales, 2004 in

US $M

Worldwide

sales, 2003 in

US $M

PEG–

Introns1369 1851 1851

Pegasyss 1374 1382 762

Neulastas 2288 1700 1300

G. Pasut, F.M. Veronese / Prog. Polym. Sci. 32 (2007) 933–961936

1.2. Low molecular weight drug– polymer conjugates

Polymer–low molecular weight drugs conjugateshave been already deeply reviewed before in thisJournal [7]. These conjugates are well describedby the model that Ringsdorf [8] proposed in 1975(Fig. 2). Since then, many steps have been made inthe understanding of their mechanism of actionboth at cellular level and in vivo that allowed animproved design of each conjugate component (i.e.,polymer backbone, spacers, targeting agents andsolubilising agents). Although, a large number ofstudies have been carried out for the preparation ofsmall-drug conjugates [2,9], unfortunately none hasyet reached the market. At present, odds are thatXYOTAXTM (paclitaxel conjugated to polygluta-mic acid (PGA)) will be the first of its class to reachthe market: it is now under phase III clinical trials

for the treatment of non-small cell lung cancer(NSCLC) in combination with carboplatin [10–13].

The difficulties encountered in the development ofsuccessful conjugates of low molecular weight drugscan be attributed to the vast number of chemical

ARTICLE IN PRESSG. Pasut, F.M. Veronese / Prog. Polym. Sci. 32 (2007) 933–961 937

and biological factors that has to be taken intoconsideration, namely:

�
conjugate features, e.g., size, polydispersivity,solubility, hydrophilic/lipophylic balance, stabi-lity, biodegradability, drug loading, free drugamount as impurity, mechanism of drug release; � in vivo behavior, e.g., biodistribution, pharmaco-
kinetic, interaction with blood components andcells, intracellular trafficking, specific targets,metabolism.

Furthermore, it is essential to know how theconjugated drug acts and in which form (i.e.,whether drug release is essential or the conjugatedcompound is also active). Usually, a covalent andstrategically positioned linkage with the polymerprevents the activity of small drugs. To ensure drug-release several methods have been developed pri-marily based on either hydrolytically unstable bondor enzymatically labile spacers between the drugand the polymer. These spacers or bonds (furtherdiscussed below) may control the release rate whichis also important, because ideally the drug has to bereleased only at the site of action to reduce the drugtoxicity. Indeed, a too fast release can abolish theadvantages of polymer conjugation and yieldconjugates with the same toxicity of the free drugs[14,15], whereas a too slow release can impair drugactivity.

Since most of the polymer–drug conjugates thathave been explored were designed for anticancertherapy, it seems worthwhile to shortly discuss thebiological basis that is making conjugation soappealing.

The high hydrodynamic volume of the macro-molecular carriers offers ‘‘per se’’ a passive targetingto solid tumors [16] due to the anatomical andphysiological modifications of such tissues. Inparticular tumors have (i) an increased vasculardensity (the result of highly active angiogenesis); (ii)vessels with both wide fenestrations and lack ofsmooth muscle layer; and (iii) a decreased lymphaticdrainage. The extensive production of vascularmediators in solid tumors contributes to theincreased vascular permeability. The EPR effectcan therefore lead to concentrations of a macro-molecular carrier in tumor tissue 5–10-fold higherthan those in blood plasma, a result difficult toachieve with unmodified low molecular weightdrugs. In addition, the polymeric prodrug can entercells only through endocytosis [17,18], a process that

is increased in tumor cells further enhancing drugspecificity [19]. Beyond passive targeting, the addi-tion of a targeting agent (such as antibodies) isreceiving great attention for the potential benefitthat may result from increased selectivity towardscells, tissues or organs [20].

To maximize the outcomes and better tailor thepolymer conjugation, a number of different poly-mers and chemical approaches were also developed,yielding a selection of new structures like dendri-mers [21–23], dendronized polymers [24], graftpolymers [25,26], block copolymers [27] branchedpolymers [28], multivalent polymers [29], stars [30]and hybrid glycol [31] and peptide derivatives [32].

Nevertheless, the difficulty to afford an efficientoral administration of polymeric derivatives willalways be a strong limitation of this technology,since parenteral administration usually reduce thepatient compliance.

From what is discussed above, it is clear that amultidisciplinary approach is essential for a success-ful drug design in this field and an intensecollaboration between polymer and pharmaceuticalchemists, biologists and immunologists is manda-tory. Hopefully now, in the ‘‘nanomedicine era’’, thegreat interest towards all the aspects of polymersand polymer conjugates, stimulated by the increaseof dedicated funding, will help scientists to reachnew goals [9,33] and will also enlarge the classes oftherapeutic agents considered for polymer conjuga-tion beyond the most studied antitumor agents [9].

Many reviews (Refs. [2,7,34,35]), with differentperspectives, are already available on this wideresearch area of small drug–polymer conjugates.

2. Polymers for bioconjugation

Many polymers have been investigated as candi-dates for the delivery [36] of natural or syntheticdrugs. In general, an ideal polymer for drug deliveryshould be characterized by (i) biodegradability oradequate molecular weight that allows eliminationfrom the body to avoid progressive accumulation in

vivo; (ii) low polydispersity, to ensure an acceptablehomogeneity of the final conjugates; (iii) longerbody residence time either to prolong the conjugateaction or to allow distribution and accumulation inthe desired body compartments; and (iv) for proteinconjugation, only one reactive group to avoid cross-linking, whereas for small drug conjugation, manyreactive groups to achieve a satisfactory drugloading. Despite the long list of available polymers


described in literature, lamentably no one combinesall the above reported requisites; we will brieflydiscuss some of the most popular ones, while PEG,due to its relevant presence in the market, willreceive a separate paragraph.

�
Synthetic polymers: PEG, N-(2-hydroxypropyl)-methacrylamide copolymers (HPMA), poly(ethy-leneimine) (PEI), poly(acroloylmorpholine)(PAcM), poly(vinylpyrrolidone) (PVP), polyami-doamines, divinylethermaleic anhydride/acid co-polymer (DIVEMA), poly(styrene-co-maleicacid/anhydride) (SMA), polyvinylalcohol (PVA); � Natural polymers: dextran, pullulan, mannan,
dextrin, chitosans, hyaluronic acid, proteins;
� Pseudosynthetic polymers: PGA, poly(L-lysine),
poly(malic acid), poly(aspartamides), poly((N-hydroxyethyl)-L-glutamine) (PHEG).

2.1. Vinyl polymers

These polymers are prepared by radical polymer-ization of the respective vinyl monomer or bycopolymerization of two or more different mono-mers, to modulate the properties of the finalproduct. As polymeric carrier, they can reach highlevels of drug loading thanks to the number ofpotentially reactive pendant groups. Vinyl polymersare not biodegradable and therefore their molecularweight must fall below the renal threshold filtrationfor these molecules (i.e., 40–50 kDa). HPMA is themost widely studied [37–39]: its derivative with theantitumor agent doxorubicin (PK1) was the firstexample of drug–conjugate designed and optimizedfollowing the Ringsdorf’s model that entered theclinical trials [40]. The drug was linked to thepolymeric backbone through a peptidyl spacer(Gly–Phe–Leu–Gly) specifically designed to bestable during plasma circulation, but promptlycleaved by lysosomal cathepsin B after cellularendocytosis [41]. The polymer molecular weight inPK1 is 30 kDa, chosen to ensure body clearanceand, at the same time, enough long circulation timeto allow tumor targeting [42]. The amount of bounddrug (�8.5wt%) was required to reach therapeuticconcentrations in tumor cells. Following PK1another conjugate was developed. This was namedPK2 and contained doxorubicin and an N-acylatedgalactosamine residue as targeting group [43]. Thismolecule confers liver targeting properties thanks to

its interaction with hepatocyte asialoglycoproteinreceptors. PK2 has a molecular weight of 25 kDa, adoxorubicin content of �7.5wt%, and a galactosa-mine content of �1.5–2wt%. PK1 reached phase IIclinical trials for the treatment of breast, colon, andsmall-cell lung cancer [44,45], while PK2 was thefirst targeted polymer to enter clinical trials, and itwas used for the treatment of primary andsecondary liver cancer [46]. PK1 and PK2 showeda two- to five-fold reduction in anthracyclinetoxicity, and despite the high cumulative doses ofdoxorubicin administered, no cardiotoxicity wasobserved.

HPMA copolymer was studied also in camp-tothecin [14], paclitaxel [15] and Pt–malonateconjugation [47,48], drugs that suffer from lowsolubility in water, a problem well overcome bypolymer conjugation. Camptothecin was linked toHPMA copolymer through a Gly–NH–(CH2)6–NH–Gly spacer to form an ester linkage with theC-20 hydroxyl group of the drug; paclitaxel wasconjugated at the C-2 hydroxyl group, again bymeans of an ester bond, using the spacer Gly–Phe–Leu–Gly. Both HPMA camptothecin and paclitaxelconjugates entered phase I clinical trials, but theresults indicated the need for optimization of thederivatives design, in particular regarding thestability of the linkage between polymer and drugs(esters in both cases) and drug loading, because theproducts showed the toxicity of the respective freedrugs due to the rapid hydrolysis of the ester bondin vivo.

2.2. Poly(amino acids) and analogues

PGA, poly(L-lysine), poly(aspartamides), poly(N-hydroxyethyl-L-glutamine) (PHEG), are easilysynthesized and, having a peptidyl structure, theyare biodegradable. Similarly to HPMA, the drugloading is high because any monomer possesses aside reactive group (e.g., amine, carboxyl, orhydroxyl) for coupling. Of this class of polymers,the conjugate PGA–paclitaxel (CT-2103, XYO-TAXTM, [10–12,49] from Cell Therapeutics reachedthe most advanced clinical stage (phase III). In thiscase, a PGA with �17,000Da molecular weight wasconjugated to paclitaxel through an ester bondreaching the impressive high drug loading of�37wt% [50], the final conjugate molecular weightwas �49,000Da. Initially, the future looked verypromising for XYOTAXTM since in phase I/IIstudies a significant number of partial responses or


stable disease were shown particularly in patientswith mesothelioma, renal cell carcinoma, NSCLCand in paclitaxel-resistant ovarian cancer [49].However, in phase III clinical trials, XYOTAXTM

was compared to gemcitabine or vinorelbine as afirst-line treatment for poor performance status(PS2) NSCLC patients and the outcome dataexpressed as survival days, indicated that thepolymer failed to show a significant improvedsurvival with respect to the two free small drugs[13,51].

Interestingly, the phase III studies highlightedthat patients with pre-menopausal estrogen levels(serum estrogen 430pg/mL), regardless of age, hada significant improvement in median survival whentreated with poliglumex–paclitaxel/carboplatin,compared to patients treated with paclitaxel/carbo-platin while it failed to improve survival in femaleswith post-menopausal estrogen levels (serum estro-gen o30 pg/mL) [52,53]. This difference has beencorrelated with the increase expression of cathepsinB by estrogen [54], the lysosomal enzyme respon-sible for drug release. A gender-specific individua-lized medicine is attractive, even if in this case morethan 50% of the NSCLC patients are aged 65 orolder and therefore patient population would belimited. Now XYOTAX is under in a phase III trial,known as the PIONEER trial, in women withadvanced lung cancer to validate the earlier clinicalresults.

2.3. Polysaccharides

Polysaccharides have been widely studied indrug delivery; their pharmacokinetic being largelyinfluenced by molecular weight, electric charge,chemical modifications, and degree of polydispersityand/or branching. Their applications range fromdelivery of small drugs to preparation of proteinconjugates [55].

Dextran (initially approved as plasma expander)is the most widely used polymer of this class [36]. Itsconjugate with doxorubicin (AD-70, molecularweight of dextran was �70 kDa) entered phase Iclinical trials, but displayed a toxicity attributed touptake of dextran by the liver reticuloendothelialcells [56].

From the protein side, dextran was conjugated tostreptokinase (streptodekases) [6] and to super-oxide dismutase [57]. Streptodekases was marketedin Russia and it was the first protein conjugatecommercialized. To perform the coupling, the

polymer was oxidized by periodate yielding alde-hyde groups that in turn were reacted with proteinamino groups. The method however was abandonedsince the multiplicity of binding groups in thepolymer may yield to undesired cross-linking in thereaction with the proteins, yielding to high hetero-geneity of the products.

Renewed interest in these polymers is nowcoming from the coupling with sialic acid terminat-ing polysaccharides to mimic the effect of naturalglycosilation in proteins, since sialic acid conjuga-tion was seen to increase the protein residence inblood [58]. The method, however, is still on benchscale feasibility and need to be investigated forlarge-scale production.

2.4. Poly(styrene-co-maleic acid/anhydride)

(SMA)

SMA is a hydrophobic copolymer obtained frommaleic anhydride and styrene. Its most knownderivative is a conjugate with the protein neocarci-nostatin (NCS), which exhibits cytotoxicity againstmammalian cells and Gram-positive bacteria. Moreprecisely, two small polymer chains of 1.6 kDa werelinked to the e-amino group of Lys-20 and to thea-amino group of Ala-1, obtaining a product knownas SMANCS (SMA–NCS). The conjugationallowed for a half-life increase of 10–20 times withrespect to the native protein and, by the EPR effect,the accumulation in tumor tissue was 30-fold that inmuscle [59]. Thanks to its increased hydrophobicity,SMANCS must be solubilized in lipid media(Lipiodol) and administered intra-arterially fortreatment of tumors. The derivative was marketedin Japan in 1993 (Zinostatin Stimalamer) for thetreatment of hepatocellular carcinoma [60].

3. Poly(ethylene glycol) (PEG)

PEG is synthesized by ring opening polymeriza-tion of ethylene oxide using methanol or water asinitiator to yield methoxy–PEG or diol PEG,respectively. It has been used for a long time asexcipient in many pharmaceutical or cosmeticformulations [61] while studies in the field of drugconjugation and delivery started only 30 years ago.PEG presents unique properties such as (i) lack ofimmunogenicity, antigenicity and toxicity; (ii) highsolubility in water and in many organic solvents;(iii) high hydration and flexibility of the chain,which is at the basis of the protein rejection

ARTICLE IN PRESS

Table 3

PEG derivatives that maintain the charge of the native protein in the final conjugate

Structure Alkylating PEGs Topics

PEG H

O PEG–aldehyde (also in the

form of more stable

acetale)

A two-step reaction, the first product (a Shiff base) is reduced by

NaBH3CN. When the coupling reaction is carried out at low pH

(4.5–5) it labels only the a-amino group

PEG O SO2 CH2CF3PEG–tresyl or tosyl (not

here reported)

Not much used since the chemistry leads to a mixture of products

N

N

N

Cl

Cl

PEG O

N

N

N

O

OPEG

PEG

Cl

PEG–dichlorotriazine or

chlorotriazine

Now they are abandoned for therapeutic application because of

their toxicity

PEG O CH2

O PEG–epoxide Slowly reactive, rarely used


properties; and (iv) approval by FDA for humanuse.

Another property, important for pharmaceuticalapplications, is the low polydispersity, with a Mw/Mn spanning from 1.01 for PEGo5000Da and upto 1.1 for PEG as high as 50 kDa. The lone hydroxylgroup in the case of the methoxy form, or the two inthe case of PEG diol, can be modified to be reactivetowards different chemical groups by severalactivation strategies. In addition, a number ofactivated PEGs are commercially available (seeTables 2–4). The reported PEG derivatives inTables 2–4 represents just a brief list of the mostimportant activated PEGs, the basic chemicaldifferences of these activated forms are alsodescribed. As shown, the derivatives are mainlyfocused on amino coupling, because these groupsare easily modified and commonly present inproteins, but reactive polymers towards othergroups have also been prepared, such as towardsthe thiol group of cysteine (Table 4). These reportedderivatives are all monofunctional PEGs, as neededfor protein modification, either in their linear orbranched structure, because these don’t give riseto cross-linking; while PEG diols, PEG dendrons[62–64], or PEG dendrimers [65], are generally used

for small drug delivery thanks to the possibility toreach higher drug to polymer ratios.

Nowadays researchers look to PEGylationmainly to prolong the pharmacokinetic propertiesof drugs [66], but also the possibility to reduce theimmunogenicity of heterologous proteins is appeal-ing. Only in one case PEGylation was exploited toenhance the pharmacodynamic response of aprotein, obtaining the product Somaverts, a casediscussed in more detail in paragraph 5. The PEGcontribution to the final hydrodynamic volume ofconjugates is of particular relevance, as demon-strated in the case of a PEGylated antibody scFvwith a branched 40 kDa PEG [67]. The conjugatethat had the calculated mass of about 65 kDa(40 kDa for the mass of PEG and 25 kDa for hemass of scFv) showed to have an apparentmolecular weight of �670 kDa. This behavior isdue to PEG abilities to coordinates a lot watermolecules per each ethylene unit and to its highlychain flexibility [68,69], thus giving to PEG anapparent molecular weight 5–10 times higher thanthat of a globular protein of a comparable mass, asverified by gel permeation chromatography [70,71].

PEG is considered a non-biodegradable poly-mer; in fact only slow degradation by alcohol

ARTICLE IN PRESS

Table 4

PEG derivatives that, after amino coupling, lead to a loss of positive charge in the final conjugate with respect to starting protein

Structure PEG–carboxylates Topics

PEG O X

O

OSun

Several PEG derivatives where X

is a linear or substituted alkyl

group (e.g. CH2, CH2CH2,

CH2CH2(CH3), etc.)

The carboxylic group is activated as N-hydroxy

succinimidyl ester, imidazole or benzotriazole. The

kinetic rate of OSu hydrolysis, and therefore

reactivity, depends on the group linked to the carboxyl

group

PEG O

O

CH2CH2

O

OSu

PEG–succinimidylsuccinate The ester bond between succinic acid and PEG is

easily hydrolyzed

PEG X

O

OSu

PEG–amino acid–succinimidyl

ester

Nle or bAla as amino acid moiety allows an easy

quantification of the number of linked PEG chains by

amino acid analysis

PEG X

O

OSun

PEG–peptide–succinimidyl ester The Met–Nle or Met–bAla allows the removal of PEG

by CNBr treatment for an easy localization of

PEGylation site. Lysosomal cleavable sequences, as

Gly–Phe–Lue–Gly, allow the release of the bound

drug inside the cell

PEG O CH2

O

SN

S PEG–thiazolidine-2-thione Long half-life compared to those of traditional

succinimidyl linkers. Reacts with proteins under mild

conditions

N

PEG

O

S

PEG–ortho-pyridylthioester This derivatives can provide a specific N-terminus

PEGylation; the requirement for the coupling is that

the N-terminal amino acid be a cysteine with an

unhindered thiol group. When not naturally present

the N-terminal cysteine can be added to a peptide by

genetic engineering

PEG– carbonates

PEG O

O

O NO2

PEG–p-nitrophenyl carbonate Slowly reactive, yield a carbonate linkage with amine

N

NN

PEG O

O

O

PEG–benzotriazol carbonate

PEG O

O

O

Cl

Cl

Cl

PEG–2,3,5 trichlorophenyl

carbonate

PEG O

O

OSu

PEG–succinimidyl carbonate


dehydrogenase [72], aldehyde dehydrogenase [73]and cytochrome P-450 [74] has been documented,especially for PEG oligomer. Therefore, its bodyclearance depends upon its molecular weight: below

20 kDa it is easily secreted into the urine, whilehigher molecular weight PEGs are eliminated moreslowly and the clearance through liver becomespredominant. The threshold for kidney filtration is


about 40–60 kDa (a hydrodynamic radius of ap-proximately 45 A [75]), which represents the albuminexcretion limit. Over this limit the polymer remainsin circulation even longer and accumulates in liver.The plasma kinetic of PEGs has been studied and itwas found to be dependent on both molecular weightand site of injection [76,77]. It was demonstrated thatusing PEGs of increasing molecular weights e.g., 6,20, 50, and 170 kDa the AUC and the half-life valuesincreased (Table 5); in particular the half-life had asigmoidal relationship with polymer whose molecu-lar weight agrees with the theoretical models of renalexcretion of macromolecules based on the pore sizesof the glomerular capillary wall [77]. Polymer–protein conjugates are excreted as such (unmodified)by the kidneys or after hydrolysis at the level ofprotein moiety. For example, a study on PEGylatedasialofetuin [78] demonstrated that the conjugate wasinternalized, by receptor-mediated transport, anddegraded as reported for the free protein, eventhought at lower rate due to PEG steric hindrance onboth enzyme degradation and complex formation forcellular uptake. Small peptide fragments linked toPEG were in fact isolated in urine after interferonadministration.

3.1. PEG– small drug conjugates

Several studies have been carried out in this fieldbut unfortunately no product has yet reached the

Table 5

PEGs reactive towards a thiol group

Structure Thioreactive PEGs

N

PEG S SPEG–pyridyldisulfide

NPEG

O

O PEG–maleimide

PEG S

O

O

CH CH2

PEG–vinylsulfone

PEGNH

O

I

PEG–iodo acetamide

market. The main limitation of PEG as drug carrieris the presence of only two reactive groups whichleads to an intrinsically low drug payload. Toovercome this limitation, the construction of den-dron structure at the PEG’s end chain has beenafforded, leading to enhanced drug loading [62–64].Even if this is a very interesting approach, one canexpect problems at the stage of large-scale produc-tion, because it requires several chemical steps thatwill much increase the overall production cost. Asalready mentioned in the introduction, polymer–small drug conjugates need to be properly designedto allow drug release only after the target site isreached. Since this factor is key, here we arereporting a brief description of some linkages thathave been exploited for drug release in addition toester bonds:

�

T

T

a

G

w

L

Linkers that respond to pH changes [79] aremainly used for small drugs conjugation. Follow-ing cell internalization by endocytosis the con-jugates are exposed to the acidic pH ofendosomes and lysosomes. Furthermore, thepH of tumor tissue is slightly more acidic thanthat of healthy tissues [80]. This parameter can beexploited using acid-labile spacers, such as N-cis-aconityl acid, which was the first approachemployed to reversibly link daunorubicin toaminoethyl polyacrylamide or poly(D-lysine)[81]. In this case, a prompt drug release was

opics

he most specific towards thiol but yields a cleavable linkage by

reducing agent also in vivo

ive stable linkage by double bond addition but may also react

ith amines at high pH values

ess reactive, not much used


evident at pH 4 or lower, while in blood or at pH6 the linker was stable. A hydrazone linkage canalso take advantage of the lower pH needed forfreeing the drug; this linker was exploited inseveral conjugates to release adriamycin [82] orstreptomycin [83] from the carrier.
� Linkers releasable by lysosomal enzymes are
mainly and exclusively used for small drugs.These spacers (usually an oligopeptide) arespecifically designed to be stable in blood butpromptly cleaved by the lysosomal enzymes,allowing a lysosomotropic drug delivery. Themacromolecular carriers are exposed to lysoso-mal enzymes when the newly formed endosome,produced after polymer–drug internalization intothe cell by endocytosis, fuses with the lysosome.Furthermore, lysosomes are overexpressed intumor cells, where cathepsins B or D and othermetalloproteinases play a role in tumor growth,and this being a potential advantage for this dugrelease strategy. Examples of such oligopeptidelinkers include those studied by Duncan andKopecek for optimization of PK1; among theseGFLG and GLFG appear to be the mosteffective [41,84]
� Drug release by anchimeric-assisted hydrolysis.
This sophisticated strategy uses linkers that aredesigned to form a double prodrug system. Thedrug-linker is first released from the polymer byhydrolysis (first prodrug), which triggers thelinker (second prodrug) that finally releases thefree and active drug. Examples of these doubledrug delivery systems include the 1,6-eliminationreaction or trimethyl lock lactonization (Fig. 3)[85,86].

A novelty may be considered the use of hetero-bifunctional PEGs [87,88], derivatives that havedifferent reactive groups at the two polymer ends;they are already commercially available and allowlinkage of different molecules to the same PEGchain, such as a drug and a targeting molecule ortwo different drugs [89]. Below some examples ofsmall drug conjugates based on PEG are described.

A PEG– camptothecin conjugate (PROTHECANor Pegamotecan by Enzon Inc.) was obtained bylinking two drug molecules, via an ester linkageinvolving the C-20 drug hydroxyl group, at thetermini of a 40 kDa polymer chain. The drugloading corresponds to 1.7wt%, which is ratherlow if compared with other multivalent polymers(i.e., PGA, HPMA copolymers). The product

passed phase I trials [90] and the maximumtolerated dose is 200mgm�2 (camptothecin equiva-lent). In phase II studies, the product, administeredevery 3 weeks at �mgm�2, showed to be promisingin treatment of adenocarcinoma of the stomach andgastroesophageal (GE) junction; and appeared to bewell tolerated, with a low incidence of toxicities [91];

PEG conjugates of a camptothecin derivative, SN-

392 (10-amino-7-ethyl camptothecin), were ob-tained linking the drug to the polymer through tri-or tetrapeptide linkers and showed to be less toxicthan the free drug and the camptothecin analogCPT-11, but they are still highly effective againstMeth A fibrosarcoma cell lines. The peptidyl linkerallowed exploiting the drug release inside the cellonly, because they are stable in plasma but rapidlyhydrolyzed in the presence of lysosomal enzymecathepsin B [92].

PEG conjugates with a technetium chelating agent,N-(N-(3-diphenylphosfino-propionyl)glycyl)-S-trytil-cysteine (PN2S–OH), were designed as new imagingagent for tumor diagnosis [93]. It was found thatPEGs of 5 and 20 kDa coupled to the carboxylicgroup of PN2S–OH enhanced the water solubility ofthe chelating agent and modified its excretion routefrom liver to kidney. The aim was to get conjugatespossessing passive targeting to solid tumor by EPReffect, and also to develop a platform for preparingactive targeted derivatives, employing heterobifunc-tional PEGs bearing specific binding moieties.Surprisingly it was found that the PEG–PN2Sconjugates were able to reduce directly 99mTcO4

�

(the form of technetium as eluted from generator) to[99mTcO]3+ (the form suitable for chelation), with-out the need of an external reducing agent, as usedin the standard procedure of complex formation.This new and simplified procedure for technetiumlabeling, that was proved to rely on an internaloxidation–reduction reaction between the phos-phorous of the ligand and 99mTcO4

�, which doesnot take place in the non-PEGylated chelatingagent. The study demonstrated that the conju-gates were able to form stable micelles in solution[93], and these favoured the reaction by micellarcatalysis.

PEG– doxorubicin conjugates were synthesized.During drug design, particular attention was givento parameters such as polymer molecular weight(5–20 kDa), polymer structure (linear or branched),and the choice of peptide linker between drug andcarrier (i.e., GFLG, GLFG, GLG, GGRR, orRGLG) [94]. Surprisingly, the conjugate with the

ARTICLE IN PRESS

PEG Spacer

O

O

CH3

R2

R1

O

NH

Drug

OH

CH3

R2

R1

O

NH

Drug

O

CH3

R2

R1

O

Drug NH2

PEG

Controllable rate

In vivo

Ester cleavageby enzyme

FastAmide cleavage by lactonization

+

+

R1 = R2 = H or CH3

Trimethyl Lock Lactonization System

O

OPEG

O

O

NH

Protein

OH

O

O

NH

Protein

PEG COOH

O CH2 CO

2NH

2ProteinOH

OH

In vivocleavage +

+ +OH-

1,6-Elimination

hydrolysis anddecarboxylation

Fig. 3. Two systems of drug-controlled release from PEG chain (A) 1,6-elimination system and (B) trimethyl lock lactonization system.


lowest molecular weight PEG displayed the highestantitumor activity in rats (mPEG5000–GFLF–doxorubicin). This seems in contrast with the needto of molecular weights above 20–30 kDa to allowaccumulation in tumors by the EPR effect. How-ever, further characterization by light scatteringshowed that this low-molecular-weight PEG–dox-orubicin conjugate forms micelles in solution withan apparent molecular weight of 120 kDa. Thissupramolecular assembly ensures effective EPReffect and explains the higher antitumor activity ofthe conjugate.

Heterobifunctional PEG, HO–PEG–COOH, hasbeen used to prepare new epirubicin conjugates. Theanthracycline epirubicin (EPI) is an effective antic-ancer drug although its use is limited by severe

cardiotoxicity. Nitric oxide (NO) proved to increasethe antitumoral activity of several chemotherapeuticagents [95], while it provides protection againstanthracycline cardiotoxicity [96]. The aim of thestudy was to exploit the beneficial effects of NO ona PEGylated epirubicin. For this purpose NO andEPI were conjugated on the same PEG chain andfurthermore, to increase the NO potential, theconstruction of a dendron structure at the level ofone PEG chain end was carried out. In particularthe conjugate bearing a 16 NO releasing moleculesand one EPI was more efficient than EPI ininducing apoptosis in Caco-2 cell. Interestingly,the same conjugate spared HUVEC, H9c2 cellsand adult cardiomyocytes from EPI-induced toxi-city. This study indicated that addition of an


NO-releasing moiety to PEGylated-EPI increased theantineoplastic activity of the drug, while it reduced itscytotoxicity against non-neoplastic cells [89].

3.2. Protein PEGylation

The first protein–PEG conjugates were basedon low molecular weight linear polymers (o10–12 kDa) that contained a significant percentage ofPEG diol (HO–PEG–OH), as an impurity inmonomethoxy–PEG batches that, once activated,could act as a potential cross-linking agent. Thechemistry employed in mPEG–OH activation oftenyielded side reaction products or weak and rever-sible linkages between polymer and protein. In anycase the conjugation methods were not site-specificand random PEGylation always took place. Evenwith these limitations some interesting productshave been obtained. PEG–adenosine deaminase(Adagens [97]) and PEG–asparaginase (Oncaspars

[98]) are typical examples of random multi-PEGy-lated proteins that got FDA approval.

Improvements in polymer homogeneity, polydis-persivity, and chemistry of coupling allowed thedesign and marketing of new and better character-ized products, such as: linear PEG–interferon a2b(PEG–Introns [99]), branched PEG–interferona2a (Pegasyss [100,101]), PEG–growth hormonereceptor antagonist (Pegvisomant, Somaverts

[102]) and PEG–granulocyte colony-stimulatingfactor (pegfilgrastim, Neulastas [103]) that will belater described.

Since the high number of PEGylation studies thathave been already conducted, it is now possible tomake several considerations that may be used as aguide for the preparation of new PEG–proteinconjugates.

Net charge of conjugate: protein amino groups arethe most exploited sites for polymer modification.At physiological conditions these groups are proto-nated allowing protein solubility and may beinvolved in recognition process with receptors. Thepreservation of the charge following conjugationdepends on the type of chemistry employed. Inparticular, alkylation maintains the charges, whilethey are lost using acylating methods of PEGyla-tion. Total charge retention is very important forpeptides, for instance, in the case of somatostatinanalogous ocreotide its activity was maintained onlythrough alkylation [104].

Total mass of coupled PEG: This (i.e., the sum ofthe molecular weights of each PEG strand attached

on a protein) determines the conjugate residencetime in vivo. The desired PEG mass may be reachedby linking either a long strand or several small PEGoligomers. To reach a prolonged body residence it isnecessary that the molecular weight of the conjugateovercomes a minimum threshold. In addition, a lowmass of bound PEG cannot ensure a great protec-tion from degrading enzymes.

How to reach the needed mass of coupled PEG:For signaling proteins achievement of the requiredPEG mass by linking one high molecular weightchain rather than several small ones can prove morebeneficial. In fact, a multipoint attachment of PEGis expected to reduce or prevent protein’s bioactivityby masking the receptor binding sites. On thecontrary a single big PEG strand, attached to theappropriate site, can better leave free the recogni-tion surface. The effect of both number and mass oflinked PEG chains on recognition and pharmaco-kinetic parameters are well documented in theliterature [76,105–107]. The studies on G-CSFmuteins or growth hormone-releasing hormoneanalogous (GRF) are of great interest as theydemonstrate the correlation between PEG mass andin vitro activity of PEG conjugates. It was foundthat the higher the PEG mass the lower the potency[105,106], while the contrary occurred for the in vivo

activity [106,107]. Several in vivo data indicated thatthe reduced receptor binding affinity, as conse-quence of high mass polymer conjugation, wascounterbalanced by the in vivo prolonged residencetime of conjugates. This is a very importantobservation because it also highlights the risk ofdrawing conclusions on polymer–protein conjugatesactivity based on in vitro assays only. The situationis different in case of enzymes working on smallsubstrates where the use of low molecular weightPEGs may be preferred because a multi-PEGattachment allows covering completely the proteinsurface, better contrasting immunogenicity andproteolysis.

The site of PEGylation: This is critical if thecoupling site resides in, or in proximity to theprotein recognition area or the active site in case ofenzymes [67,108,109]. In enzymes, this can beprevented by performing the modification in thepresence of substrates or inhibitors that protect thecatalytic region. For signaling proteins the protec-tion is more difficult, even though PEGylation inthe presence of protein’s receptor has been reported.

Polymer shape: It was found that not only themass and the chemistry of binding are important for

ARTICLE IN PRESS

Enzyme or protein

modification by:

Branched or Linear PEGs

enzyme’s active site= Buried group or

enzyme= Antibody, proteolytic

Fig. 4. Relevance of PEG shape on protein surface coverage or enzyme active site access. The higher steric hindrance of branched PEGs

ensures more complete protein shielding and makes difficult the access to recognition or active site cleft.


the biological behavior of the coupled protein butalso the shape. In fact, the ‘Y’ shaped branchedPEG [110] improved the PEG shielding effect on aprotein surface, thus being more effective inprotecting the conjugated protein from proteolyticenzymes and antibodies (Fig. 4). Moreover, proteinsmodified with branched PEG retain activity morethan those coupled to linear PEGs because thehigher steric hindrance prevents the PEG fromreaching the enzyme active site clefts or other lessaccessible sites involved in the biological activity(Fig. 4).

Coupling chemistry: The appropriate PEGylationstrategy must be chosen and optimized for anyprotein to improve conjugation yield, conjugatepurity and biological activity. Amino groups are themost widely exploited residues for protein modifica-tion. However, since they are widely represented inproteins, PEGylation can result in a mixture ofisomers. The difference between isomers mightdepend on the number of PEG chains linked tothe protein as well as on the amino acids involved inthe conjugation. Initial coupling methods werecharacterized by a random PEGylation, and in thiscase acceptable degree of product homogeneityrelayed only on the strength of purification proce-dures. Several activated PEGs were synthesized for

amino coupling [35] (some reported in Tables 3 and4) and the difference between these PEGs relies inthe kinetic rate of conjugation and in the linkbetween polymer and drug. Information of thereactivity of activated ester or carbonate PEGderivatives can be obtained by analyzing the rateof active group hydrolysis in buffer solutions,expressed as hydrolysis half-life time. Usually adifference in the order of magnitude can be foundamong different active PEG esters. For exampletaking into consideration two N-hydroxysuccinimi-dyl PEG esters, the half-life of active grouphydrolysis is less than 1min for the carboxymethy-lated PEG (SCM–PEG) while rises to 45min for thea-methylbutanoate–PEG (SMB–PEG). This differ-ence is due to the upstream groups of active moiety.Similarly these groups can lead to different rate in aPEG–protein conjugation also. Therefore, it ispossible to take advantage of the derivatives witha lower reactivity to obtain a certain degree ofselective conjugation within the amino groupspresent in a protein, according to their nucleophi-licity or accessibility [111].

Relevance of amino groups’ pKa in a protein

modification: The three-dimensional protein confor-mation is responsible for the degree of solventexposition of the amino acid residues. Furthermore,


the nucleophilicity of the amines can be influencedby the neighboring groups through the formation ofhydrogen bonds. This means that different aminogroups in a protein do not present the samenucleophilicity. In addition, it is instructive toremember that in some cases certain protein lysinespossess a pKa as lower as two-three units of theusual value of 9.3–9.5 [112]. Certain reactionconditions or particular induced conformations(e.g., using structure promoting solvents [105]) canenhance the degree of PEGylations at the level ofspecific residues only. A general method to reach aselective amino modification exploits the differencein reactivity between the e-amino and the a-aminogroup of proteins, due to their pKa: 9.3–9.5 for thee-amino residue of lysine and 7.6–8 for the a-aminogroup (N-terminal amino group). This was the caseof Neulastas in which the use of PEG–aldehyde atslightly acidic pH ensured the formation of a singleisomer (a case discussed in paragraph 5) thatfacilitated also the purification procedures. Alter-natively, specific PEGylation may be performed byblocking some of the reactive groups with areversible protecting group as reported for insulin[113]. In this case, the coupling reaction wasperformed following protection of two of the threeamino groups with (tert-butyl carbamate) (N-BOC).The removal of BOC protection at the end ofPEGylation leads to mono-PEGylated insulin at thedesired site.

Selective PEGylation exploiting genetic engineer-

ing: several protein mutants were designed to reachhigh selective PEGylation. Two different ap-proaches are mainly used, in one case lysines inthe protein recognition site can be replaced withnon-reactive amino acids thus preventing the PEGlinking [114], and in a second a cysteine can beintroduced in a desired position allowing the link ofPEG using specific thiol reactive polymer deriva-tives (Table 5) [67,108].

Controlled release: Linkers that allow the releaseof the conjugated protein were recently proposed.As already mentioned for non-peptide drugs astable bond between polymer and protein oftendecreases, and sometimes prevents, biological activ-ity. Releasable PEGs, able to link the proteinthrough hydrolysable bonds, have been studied toobtain slow release of the fully active free form ofthe conjugated proteins. The PEGs, which arereported below, may be seen as the develop-ment of those previously described for small drugrelease:

�
Linkers based on N-modified bis-2-hydroxyethyl-glycinamide (bicin). The system started from theobservation that when the carboxylic group ofbicin is linked to an amino group, the amidebond, which is usually very stable, undergoesanchimeric-assisted hydrolysis by the hydroxylgroups of bicin itself. Exploiting this mechanism,two classes of PEG bicin derivatives wereprepared: (i) one composed by a branchedPEG–bicin, where PEG chains were linked byester bond to both bicin hydroxyl groups and (ii)the other obtained by coupling one bicin hydro-xyl group with a PEG chain and the second witha molecule that controls the kinetic rates ofhydrolysis (e.g., an acetyl residue). The releasemechanism (shown in Fig. 5) firstly involves theliberation of at least one bicin hydroxyl groupsby water. Subsequently the release of conjugateddrugs in its fully active form is promoted [115].Through this bicin spacer PEGylated interferonb-1b demonstrated an increase of the area underthe curve (approximately 20-fold) compared tothe native protein. � Anchimeric assisted hydrolysis was also exploited
in the method proposed by Schechter based onthe bifunctional reagent 2-sulfo-9-fluorenyl-methoxycarbonyl spacer [116], In this case, therate of protein release is determined by thehydrolysis of the spacer.
� A new releasable system is based on a spacer
containing a molecule of b-alanine, where thecarboxyl group is conjugated to an amino PEG,while the amino group is activated asN-[(succinimidoxy)carbonyl]. The reactivity of thisPEG–NH–CO–bAla–NH–CO–NHS was signifi-cantly lower of that of PEG–O–CH2–CO–NHS(SCM–PEG) making the former more selective,reacting preferably with the most nucleophilic andexposed amino groups of the protein. This canhopefully reduce the number isomers obtainedduring PEGylation. Furthermore, as verified withhuman growth hormone it showed a slow releaseof the native protein [117,118].
� The link between a carboxylic activated PEG and
histidine side chain residue is unstable leading torelease of the native protein in vivo. Acyl-imidazoles are in fact well known to be markedlyunstable and PEG–acyl imidazoles were there-fore already used as PEGylating agents. Themost important example of this type of PEGlinkage is the PEG–Introns [99] that, marketedsince 2000, was obtained using PEG succinimidyl

ARTICLE IN PRESS

mPEG NH

O

O

O

O

mPEG NH

O

O

O

O

N NH

O

Drug

mPEG NH

O

O

O

O

N NH

O

Drug

O

H

mPEG NH

O

O

O

O

N

O

O

DrugNH2

OH

N NH

O

Drug

O

H

DrugNH2

OH

N

O

O

+

+

Fig. 5. Bicin system of drug controlled release from PEG chains.


carbonate (SC–PEG; 12 kDa) at pH 6.5 takingadvantage of the lower pKa of His (6.7–7.1) withrespect of that of Lys (9.3–9.5). Analysis of thepurified monoPEGylated product showed thepresence of small amounts of several positionalisomers while the histidine-34 conjugate repre-sented approximately 47% of the PEGylatedspecies [119,120]. The activity of this interferonpreparation was related to its ability to releasefree and fully active interferon by slow hydrolysisof the His–PEG bond.

Conjugates purification: theoretically, in a con-jugation reaction conducted with an excess of PEG,one could expect that all of the reactive groups ofthe protein are modified thus yielding mainly asingle product. However, to prevent loss of biolo-gical activity, a lower amount of PEG is generallyemployed which usually yields to a mixture ofpositional isomers [121] as well as products withdifferent extent of PEGylation (i.e., mono-, di-, tri-,tetra- and more PEGylated protein). These lattercan be quite easily separated by gel filtration or ionexchange chromatography while reverse phasechromatography was found less efficient [122,123].On the other hand, a mixture of positional isomerscan be fractionated by highly selective ionicexchange chromatography in HPLC [124] thanks

to the subtle differences in isoelectric point of eachisomer.

Evaluation of the extent of PEGylation: colori-metric methods measuring the residual uncoupledamino groups may give a gross estimation of theextent of PEGylation [125]. Information can also beobtained by gel-filtration chromatography or bySDS-page electrophoresis. However, the last twomeasures can be of difficult estimation due to thedifferent molecular weights/hydrodynamic volumesrelationships for PEGs and proteins (as abovereported in paragraph 3). Matrix-assisted laserdesorption ionization–time-of-flight (MALDI–TOF) mass spectroscopy and capillary electrophor-esis are definitely the best procedure for conjugatecharacterization. Usually the electrospray massspectroscopy cannot be used otherwise employingmonodisperse PEGs that are unfortunately avail-able so far at the very low molecular weights only.

PEG conjugates characterization: the usual ap-proach to identify the PEGylation site into theprimary sequence involves enzymatic digestion ofthe conjugate, purification of the peptides and theiridentification by mass spectroscopy or amino acidanalysis. The site of PEGylation in this case is foundby the identification of the missed peptide inthe proteolysis mixture as compared to that ofthe unconjugated form. The characterization of


PEGylated interferon a-2a represents a goodexample [126]. In this study, the comparison of thepeptide fingerprint of the conjugated protein withthat of the native protein allowed identification ofthe region of PEGylation based on the disappear-ance of peptide signal. Besides the lengthy proce-dure, the conjugated polymer may interfere by sterichindrance with the proteolytic enzymes, resultingin an incomplete cleavage that complicates theinterpretation of the peptide finger printing. Analternative procedure exploits the use of PEG–Met–Nle–COOH or PEG–Met–bAla–COOH in theconjugation step. These PEG derivatives possess achemically labile bond in the peptide spacer at thelevel of methionine, which can be cleaved bytreatment with CNBr, this leaving a nor-leucine oran b-alanine tags on the protein primary sequence.The amino acid-tagged peptides can be easilyidentified by standard sequence methods or by massspectrometry analysis in the enzymatically digestedmixture [127]. A method to evaluate the effect ofPEGylation on a signaling protein is Biacore, whichoffers also the opportunity to quantify the affinityof a modified protein towards its receptor [71,128].However, the biological data obtained by testingthe conjugate on a specific cell culture (takinginto consideration the possible problems dis-cussed above) or in vivo remain the most valuableones.

Primary amino acid sequence degradation: pro-teins are known to have labile bonds, such as theAsp–Pro bond or the amide group of Asn sidechains. The former may undergo cleavage, breakingthe amino acid sequence, while the second can beeasily deaminated with likely formation of asparticor iso-aspartic; either process is pH dependent.Since in certain conjugation procedures and inpurification steps proteins are exposed to non-physiological pH values, these primary amino acidsequence modification might occur. These modifica-tions can be detected using appropriate methodssuch as analysis of the protein proteolytic digestion.Further indications of the occurrence of theseunwanted reactions may also come from other,more common, analysis like ion exchange andreverse-phase chromatography in HPLC.

Three-dimensional structure evaluation: any che-mical modification of proteins can lead to changesin their tertiary structure that, although slightly,may influence the biological properties. Therefore,suitable methods to assay the conformation mayhelp to better design the conjugation technique.

Circular dichroism (CD) investigation is the firstchoice for determination of the secondary structureof a protein. In addition, fluorescence, fluorescencequenching or thermal unfolding evaluated by CD(CD melting profile) are useful techniques for thethree-dimensional conformation [71,129].

Solubility and aggregation studies: for its highhydrophilicity PEG is expected to increase thesolubility of linked molecules, either small drugsor proteins. Two thorough investigations on stabi-lity and aggregation were carried out for PEGylatedproteins: one involving PEG-b interferon conjugates[71] and the other the PEG20 kDa–G-CSF (Neu-lastas) [130]. Both native proteins have hightendency to aggregate. In the first study it wasshowed a direct relationship between PEG–INFbstability and polymer molecular weight [71]. In thecase of G-CSF it was demonstrated that both thenative protein and the PEGylated one formaggregates, but those obtained from native proteinsolutions were insoluble while PEG–G-CSF leads tosoluble aggregates. Interestingly, the pathways ofaggregation was the same for both proteins,involving a secondary structural transitions andthen a disulfide bridge cross-linking, but for thePEGylated G-CSF the rate of aggregation wasvastly reduced. This may be accounted either to thesteric hindrance of PEG, thus impeding protein–protein association, or to the shielding of hydro-phobic areas on the protein surface. Furtherinteresting data from this paper [130] demonstratedthat also the G-CSF conjugation with a 5 kDa PEGhas beneficial effect of aggregation, and these weremainly comparable to those obtained by the 20 kDaPEG. Therefore, this opens a debate on the pointmainly driving the choosing of PEG length duringdesign of PEGylated proteins; in fact so farresearchers based their decision looking to pharma-cokinetics profiles while also the aggregation aspectshould be taken into consideration. PEGylation,especially with low molecular weight polymers, maybe useful also for those molecules that do not need aprolonged residence time but have stability pro-blems. It is noteworthy that also the chemistry ofPEGylation influences the aggregation rate, in factPEG conjugated to G-CSF by alkylation signifi-cantly decreased aggregation with the respect to thatobtained by acylation [131]. This result highlightsthe importance of maintaining the protein nativecharge to avoid aggregation. Finally, the authorsclaim the formation of micelles containing a hydro-phobic protein core and a hydrophilic PEG exterior


as explanation for the solubility of aggregates in thecase of PEG–G-CSF. PEG-conjugates aggregationleading to micelles was also observed in PEGslinked to low molecular weight hydrophobic mole-cules, as for the PEG conjugates of doxorubicin[94] and PN2S-chelating agent [93]. In the firstcase, micelles contributed a lot to enhance thein vivo distribution and antitumor activity, whilein the second they accounted for an unexpectedmicellar catalysis that favored an oxido-reductionreaction.

Immunogenicity evaluation: immunogenicity, eval-uated in animals or sometimes in humans, isgenerally decreased following conjugation andexamples of these investigations are reported[3,4,132,133]. Of interest, recent studies on PEGurate oxidase showed that the conjugate is clearedfaster from the blood circulation of certain patientscompared to the average rate. This effect was due tothe formation of IgM and IgG antibodies, un-expectedly direct towards the PEG moiety [134].However, this phenomenon is not new since the onpurpose elicitation of antibodies against PEG wasobtained by administering PEGylated proteins inFreund’s complete adjuvant to rabbits [135]; but onthe other hand it was demonstrated that theadministration of PEG alone or PEGylated proteinswithout the adjuvant did not lead to formation ofantibodies. Commercially available PEG antibodiescan be useful for detection analysis or in vivo

removal and neutralization of PEGylated products[136,137].

S

S

S-

SHR

O

CH

PEG

Disulfide

reduction

11°° AdditionAddition

S R

O

PEG

S-

22°° AdditionAddition

Fig. 6. Selective PEGylation at the level of disulfide br

4. Innovative PEGylation procedures

Since the high number of papers that alreadyreviewed this matter [35,69,111,138,139], a descrip-tion of only few and most recent methods of proteinPEGylation is reported here.

An innovative method has been devised byBrocchini and coworkers in which PEG is selectivelylinked at the level of protein disulfide bonds, whichare almost ubiquitous in proteins [140]. Thechemistry requires a prior reduction of existingdisulfide to obtain the two free thiols, one of whichreacts with a special PEG derivative (a,b-unsaturedb0-monosulfone PEG of 5 kDa). Consequently, thisfirst linking of the PEG reagent rearranges andreacts with the second thiol forming, through thisbis-alkylation, a three carbon bridge between thetwo cysteines. The new link is a three-carbon bridgeto which PEG is covalently bound (Fig. 6). Thereaction is carried out by a careful control ofstoichiometry to modify only the disulfides moreexposed on the protein surface. Molecular modelinganalysis displayed that this three-carbon bridge,inserted in the protein disulfide bonds, does notapparently modify the protein secondary andtertiary structure.

A recent study also demonstrated that it ispossible to transiently denature a protein to favorPEG linkage at the level of buried and otherwise notaccessible groups. In particular, different thiolreactive PEGs were conjugated at the lone free butburied cysteine of G-CSF by exposing the protein to

SO2R'

2

S SO2R'

ORPEG

SH

S

S

R

O

PEG

EliminationElimination

BridgedBridged PEGylation PEGylation proteinprotein

idge as proposed by Brocchini and Duncan [36].

ARTICLE IN PRESS

Table 6

AUC and half-life in blood of PEGs with different molecular

weight after iv administration in mice

Parameter PEG

6kDa

PEG

20 kDa

PEG

50 kDa

PEG

170 kDa

AUC (%dose h/mL) 6.17 110 600 1110

t1/2 (min) 17.6 169 987 1390

Table 7

Pharmacokinetic properties of Interferon a-2a and its PEGylated

form in rats [101]

Protein Half-life (h) Plasma

residence

time (h)

Interferon a-2a 2.1 1.0

PEG2 (40 kDa)–interferon a-2a 15.0 20.0

Table 8

Comparison of IL-2 conjugates activities between random

PEGylation and site direct PEGylation by TGase [142]. The

rTG1 abbreviation represents the N-terminal peptide fused to

IL-2 as specific TGase substrate

Proteins %

Activitya

rhIL-2 100

PEG10–rhIL-2 (random PEGylation) 74

(PEG10)2–rhIL-2 (random PEGylation) 36

rTG1–IL-2 (chimeric protein for enzymatic

PEGylation)

72

PEG10–rTG1–IL-2 (enzymatic PEGylation) 69

(PEG10)2–rTG1–IL-2 (enzymatic PEGylation) 72

aThe amount of % activity was expressed as the percent

residual bioactivity as compared to the rhIL-2.


partially denaturant conditions (e.g., solutions of3M NH4OH or 4M guanidine hydrochloride) butwithout reducing the disulfide bridges. After mod-ification with PEG, the native conformation ofG-CSF was recovered by removal of the denaturant[141], to regain the three-dimensional structure andtherefore the protein native activity.

Traditionally, polymer conjugation has beenperformed by chemical procedures but recently theuse of specific enzymes has been investigated. Infact, enzymes can offer a degree of specificity andselectivity that it is not easily reached by chemicalmethods, and furthermore they can catalyze reac-tion that otherwise cannot be conducted. Their useis seen as an alternative solution to obtain homo-geneous PEG–protein conjugates under mild reac-tion conditions or to perform PEGylation at thelevel of unusual amino acids, which might preservethe protein activity better than with the usualconjugation techniques. An example of feasibilityof this approach is the PEGylation of interleukin-2(IL-2) at the level of glutamines using the enzymetransglutaminase (TGase) [142]. This also representsan example of PEGylation involving an amino acidthat was never considered before for conjugation ofa protein. In particular in this study two differentTGase were investigated and PEGylation wasconducted either on native IL-2 or on a chimericIL-2, the last prepared by fusing IL-2 with a specificpeptide sequence that is a substrate for one of theused TGase. TGase catalyzed the transfer of theamino group of a PEG–NH2 to a glutamine residuepresent in a protein, forming an amide bondbetween PEG and the glutamine side chain group.The aim of PEGylation of the chimeric protein wasto demonstrate that this approach can still beapplied even when native glutamines are notsuitable substrate for TGase, circumventing thisproblem by fusing the proper peptide. Therefore,this allows also to choose the preferred location ofPEGylation site on the protein surface.

The derivatives prepared using chimeric IL-2maintained almost the same activity of the nativeprotein, whereas the classical chemical conjugationwith mPEG–NHS yielded products with decreasedactivity (Tables 6–8).

A two-step enzymatic PEGylation, called Glyco-PEGylation, was developed by Neose Technologies.In this case, Escherichia coli-expressed proteins wereglycosylated at the level of specific serine andthreonine with N-acetylgalactosamine (GalNAc)by in vitro treatment with the recombinant

O-GalNAc-transferase. The obtained glycosylatedproteins were subsequently PEGylated using the O-GalNAc residue as the acceptor site of the cytidinemonophosphate derivative of a sialic acid–PEG(20 kDa), a reaction selectively catalyzed by asialyltransferase [143]. This method was tested onG-CSF, INF-a2b and GM–CSF, and for the firsttwo proteins the formation of homogeneous mono-PEGylated derivatives at threonine 133 and threo-nine 106 respectively was demonstrated. In the caseof GM–CSF a different approach was necessary toincorporate the PEG derivative. Since two GalNAcmolecules were linked to the protein the authorswant to obtain a di-PEGylated species but theGalNAc proximity, Ser7 and Ser9, prevented theformation of the di-PEGylated form; therefore to


circumvent this problem, it was firstly introduced asugar residue spacer, at the level of GalNAcmolecules, providing a more favorable substratefor the addition of two sialic acid–PEG chains.Using this approach, approximately 75% of di-PEGylated GM–CSF formation was achieved. Thegreat advantage of this technology is the possibilityof PEGylating proteins produced in E. coli in orderto mimic the mammalian ones, since the PEG chainsreplace the native sugar moiety at the precise site ofglycosylation, forming conjugates that retain thecorrect structure for receptor recognition andextended plasma half-life.

5. Selected examples of PEGylated proteins

Interferon a-2 (IFNa-2) has been the standardtreatment for chronic hepatitis C virus (HCV). Itsactivity is mediated by IFN-inducible gene activa-tion [144] that occurs when the IFN binds to amultimeric cell surface receptor. The interferon lowmolecular weight (�20 kDa) is the cause of therelatively short serum half-life (�4–16 h [145]) ofthis protein, which limits its efficacy against HCV.PEGylation was investigated to improve the phar-macokinetic profile of INFa-2 and two differentconjugation strategies were developed by randomPEGylation: (i) PEG12 000–INFa-2b based on SC–PEG 12 kDa, to obtain a mixture of monoPEGy-lated positional isomers in which the most repre-sented is the conjugate at His34 (�47%) [99] and (ii)PEG240000–INFa-2a prepared using branchedPEG2–NHS 40 kDa, leading also to a mixture ofmonoPEGylated positional isomer, in particular thepolymer is attached for the 94% at one of thefollowing lysines: Lys31, Lys121, Lys131 or Lys134[100,101]. The former method yielded a productmarketed with the name PEG–Introns, and thesecond has been marketed with the name Pegayss.The different polymer molecular weights led to adifferent half-life of the conjugates, 27–37 and 65 h,respectively. Furthermore, the MW of PEG chainalso affected the residual antiviral activities. Thesewere 28% and 7% for the mixtures of PEG12 000–INFa-2b and PEG2 40000–INFa-2a, respectively.The higher in vitro activity of PEG–Introns

compared to Pegayss is ensured not only by thelower molecular weight of PEG but also by therelease of native INF from the one of the conjugateisomers, namely the one having the PEG linked toHis34 [119,120]. On the other hand, Pegayss

overcomes its reduced in vitro activity exploiting

the extended pharmacokinetics that finally leads toa more prolonged activity in vivo.

G-CSF was studied by site-direct amino-PEGyla-tion performed taking advantage of the lower pKavalue of the a-amino group at the N-terminus withrespect to the pKa of e-amine of lysines (see above).Indeed, when the reaction is carried out underneutral or mildly acidic conditions, PEGylation atthe level of lysine can be prevented whilst leavingthe N-terminal amino group reactive [146]. Themost successful example of this strategy is thealkylation of r-metHuG–CSF with PEG–aldehyde,proposed by Kinstler et al. [103]. The reaction wascarried out at pH 5.5 in the presence of sodiumcyanoborohydride to reduce the Shiff’s base initiallyformed [103,131]. The PEG–G-CSF conjugate,obtained using a linear PEG of 20 kDa, showed amarked increase in protein residence time and theconjugate has been on the market since 2002 asPegfilgastrim, Neulastas.

The growth hormone-releasing hormone

(hGRF1– 29) was conjugated to PEG performingthe reaction in structure promoting solvents, whichreduced the number of PEGylation site by limitingthe accessibility of the amino groups. Nuclearmagnetic resonance (NMR) and CD analysisindicated that the percentage of a-helix inhGRF1–29, which is only 20% in water, rises to90% in structure-promoting solvents such asmethanol/water or 2,2,2-trifluoroethanol (TFE),thus facilitating a region-selective modification.Indeed, when PEGylation was performed in TFEthe yield of the conjugate at Lys-12 reached 80%.This approach represented a marked improvementif one considers that the same reaction carried out inDMSO leads to an almost equimolar mixture ofLys-12 and Lys-21 conjugate isomers [148].

Recombinant human growth hormone mutein

(B2036) has an antagonist effect compared to thegrowth hormone and can be useful in the treatmentof acromegaly. This protein was the object of anunusual approach of protein PEGylation. GHreceptor is constituted by two proteins; one moreresponsible for GH recognition and the other forthe signal transduction. GH binds both receptorsthanks to two different binding sites. The muteinB2036, with mutations at the level of both bindingsites, retains the binding capacity to the firstreceptor protein but blocks the interaction withthe second, thus avoiding receptor signaling [149].Furthermore, the mutation was designed to obtainpreferential PEGylation at the level of the second


recognition site. This prevented more efficiently theinteraction with second receptor enhancing theantagonist effect, while increasing residence time in

vivo. In details, a lysine residue in the first site wasexchanged with a non-reactive amino acid, whilst alysine was added to the second site. SPA–PEG(5 kDa) was employed for PEGylation, forming amixture of derivatives with 4–6 PEG chains for eachB2036 protein. This example of PEGylation inwhich the polymer contributes in enhancing theprotein response can prompt studies with otherprotein antagonists.

Interferon b-1b (INFb-1b) PEGylation is a veryinformative case of protein modification and it isreported by Enzon’s researchers [71]. The aim wasto overcome the intrinsic problems of the nativeprotein. PEGylation increased solubility, bloodresidency, reduced the tendency to aggregation,expanded in vivo exposure of bioactive protein, andreduced IgG response against the bioconjugate. Inthis study the authors exploited almost all aspects ofPEGylation approaches discussed so far, in factpolymer coupling was performed either in a randomor site-specific manner. A random PEGylation wasdirected to all amines, while a site-specific PEGyla-tion was performed at the level of N-terminal (astrategy above reported for Neulastas) or exploit-ing INFb-1b muteins prepared by genetic engineer-ing. The latter was based on two alternativeapproaches for PEG–INFb-1b engineering: i) site-direct PEGylation at the level of cysteine inserted ina position far from the protein binding site (e.g.,N-terminus, C-terminus, or position 79 that is theN-glycan site in the human protein), an approachfor protein PEGylation that already proved itsefficacy in other studies [67,108], ii) preparation ofmuteins in which few lysines present in binding sitewere substituted with non-reactive amino acids toavoid PEG linkage at this level, as devised byanother study [114]. The effect of polymer mole-cular weights (e.g., 12, 20, 30 and 40 kDa),structures (linear or branched), extent of PEGyla-tion (e.g., mono-, di- or tri- PEGylated specie),activation group (e.g., succinimidyl carbonate-,thiazolidine-2-thione-,aldehyde- and various NHSactivated PEGs) were also investigated. Twenty oneconjugates were prepared and evaluated for struc-tural, physical–chemical and biological behavior.This study can be considered the most completecomparison on the influence of the differentparameters on protein PEGylation. In this studyproblems were encountered with the engineered

proteins, due to the low activity of muteinsfollowing lysine replacement and, in the case ofcysteine muteins to the impossibility to perform asatisfactory thiol conjugation since the disulfidebridges were cleaved under the reducing conditionsused. Unexpectedly similar physical and biologicalproperties were displayed by different bioconjugatessuch as the 40 kDa mono-PEGylated NHS–PEG2the aldehyde PEG2 derivative or the di-PEGylatedsuccimidyl propionate (2� 20 kDa) conjugates.

Recombinant mammalian urate oxidase PEGyla-tion (PEG–uricase) represents a typical example ofenzyme modification in which multiple attachmentsof small PEG chains is preferred to the linkage ofone large PEG strand. In this case, about ninechains of pNPC–PEG were coupled to each subunitof the tetrameric enzyme [150]. At present, theconjugate is undergoing phase II clinical evaluation.Data from phase I trial on 13 markedly hyperur-icemic patients showed the ability of the product tonormalize plasma uric acid [134]. In eight subjectsenzyme activity was still detectable after 21 days, onthe contrary in the other five the plasma uricaseactivity could not be measured beyond ten daysafter injection. The authors correlated this with theappearance of IgM and IgG antibodies againstPEG–uricase. Unexpectedly, the trial showed thatthe antibodies were not directed against the proteinbut against PEG. This research contrasts with theother studies conducted so far, in which the polymeralways reduced the immuno-response against alinked protein, but it recalls the observation byIshida et al. [151] of formation of antibodies againstPEG when PEGylated liposomes were administeredto rats. Further investigations are still needed toelucidate the reasons of this immuno-responseagainst PEG in the uricase case.

A recombinant arginine deiminase (ADI) fromMycoplasma arthritidis is an enzyme that metabo-lizes arginine, an important metabolite for thefunction of several biological systems and argininedeprivation inhibits tumor growth, angiogenesisand nitric oxide synthesis [152]. This enzyme (adimer protein) has been the object of a PEGylationstudy by coupling lot chains of polymer per proteinmolecule [153], as already above mentioned forother heterologous enzymes [150] and following thesuccessful method exploited in the development ofADAGENs and ONCOSPARs. In particular theinvestigation was carried out using two PEGmolecular weights, 12 and 20 kDa, both activatedas succinimidyl carbonate; the conjugation reached


in both cases about 33% of protein amino groupsmodification that means 2272 PEG strands perprotein dimer, about 264 and 452 kDa of linkedPEG mass, respectively. These huge amount of PEGmass considerably prolonged the plasma meanresidence time of the enzyme over 30-fold in mice.Since immunogenicity is one of the most importantconcern for protein drugs, the ADI–PEG conju-gates were tested in presence of neutralizing anti-serum, developed using native ADI, andinterestingly the antibodies were able to reduce theactivity of both conjugates although the enzyme isalmost completely covered by PEG strands; there-fore also in these cases, PEGylated proteins are notexpected to be completely preserved from anyprotein–protein interaction in vivo. The pharmaco-dynamics of conjugates were quite similar andtherefore the authors considered the derivative ofPEG 12 kDa most interesting because it simplifiedthe purification procedure (i.e., the separation ofunreacted PEG by ultrafiltration/diafiltration wasmore fast and efficient using the lower molecularweight PEG), gave less viscose solution and reducedthe final cost production.

Further examples of PEGylated proteins orenzymes are well reviewed elsewhere [35,154,155].

6. Aptamer PEGylation

The highly specific interaction between oligonu-cleotides and proteins is commonly present innature, for example transcription factors specificallyrecognize the DNA sequence of promoters andrestriction enzymes selectively cleave their restric-tion site DNA sequence. Exploiting this highspecificity of binding was developed a new class oftherapeutic agents: aptamers, which are RNA orDNA oligonucleotides that have been selected fortheir capacity to bind a specific protein. Theselection of new aptamers is performed by systema-tic evolution of ligands by exponential enrichment(SELEX) [156].

Aptamers great potentials can be put on risk by acommonly too short in vivo half-life, mainly due tonuclease and fast clearance by kidney. To overcomethe first problem, several types of nucleotidemodification have been proposed, leading to satis-factory results, for the second conjugation with highmolecular weight polymer can be a solution. As anexample, the development of Pegaptanib (Macugen;OSI/Pfizer) is remarkable, which is the first aptamerapproved for clinical use [157].

Macugen is a 27mer RNA oligonucleotide thatbinds to extracellular vascular endothelial growthfactor (VEGF) and preventing its receptor bindingand signal transduction it blocks the angiogenesispromoted by this factor. This aptamer is used in thetreatment of neovascular macular degeneration, aneye pathology characterized by increased vascular-ization that finally leads to vision loss. Some specialnucleotides were used to increase nuclease resis-tance, for example it was included some 20-O-methylated purines and 20-fluorine modified pyr-imidines, and to prolong the half-life a PEGbranched of 40 kDa was linked to the 50 side.Despite a fourfold reduction in binding affinity forVEGF after PEG conjugation the product showed amarked increase to 83% of VEGF inhibition [158].

This product paved the way for the developmentof further PEGylated aptamer showing the greatpotentials coming from the combination of thesetwo technologies, SELEX and PEGylation.

7. Cells pegylation

Recently PEGylation extended its outlook, be-yond the field of small drugs and proteins, takinginto consideration also quite complex biologicalsystem, such as cells and tissue. Great challengescan be expected in for PEGylation of these systems,and standardization will be a basic troublesome forthe ones dealing with these projects.

PEGylation of red blood cells (RBCs) directedtowards membrane proteins, carbohydrates, or lipidhead groups, was the first example of this newresearch direction. It was devised mainly fortransfusion purposes, but also for the preparationof stealth RBCs useful as drug delivery systems[159,160]. PEG linking leads to immunocamouflageof cells by covering antigenic sites and membranesurface charges, and by reducing receptor–ligandand cell–cell interactions. Interesting results wereobtained in the PEGylation of white blood cells andlater also with the more challenging pancreatic islets(containing the b-cells that produce insulin) [161].The latter represents a more challenging systemthan RBC for PEGylation because the pancreaticcells have to retain not only the capacity to produceinsulin but also the ability to fine-tune the release ofthe hormone following glucose response. Theauthors demonstrated the functionality of theLangerhans islets after PEG modification by trans-planting the PEGylated cells into rats via portal veininjection for engraftment within liver vasculature.


8. Conclusion

There are already hundreds of papers, patentsand reviews on PEGylation of proteins, peptidesand small drugs. The wealth of this body ofliterature shows that this technique has becameextremely important although it is relatively young(no more than 30 years old).

A critical examination of the large production ofstudies of the last years shows however only fewreally original advances or new chemical ap-proaches. Among these we may list for instancethe enzymatic methods of PEGylation [142,143],which allow addressing amino acids residues other-wise difficult to modify by chemical approaches, orthe procedure to introduce PEG at the level ofdisulfides [140].

At present the research on PEGylation is mainlyfocusing on the application to different proteins ofthe conjugation methods already developed. Thesestudies are in fact directed to establish suitableapplications of the numerous and therapeuticallyinteresting proteins that otherwise exhibit unfavor-able properties for clinical use. Currently, this is aneed due to the continuously increasing number oftherapeutic proteins coming from the proteomicdiscoveries.

Important interdisciplinary studies are thereforeemerging from large industrial research groups, verymuch stimulated by the market opportunities ofthese new drugs. These studies take into considera-tion all the physical–chemical, conformational,immunological, pharmacokinetic and formulationaspects because they aim to reach a humanapplication.

Even thought many studies have been carried outon the behavior of PEG and PEG conjugates,several aspects need deeper investigations to becompletely clarified, especially those related to PEGinteractions with proteins or membranes [162]. Thisknowledge will have interesting implications inprojecting and preparing biocompatible conjugateswith well defined and predictable in vivo properties.On the other hand, deeper evaluation studies haveto be carried out on the formulation of PEGylatedproteins and on their stability. In this regard, it isworth mentioning the recent data demonstratingthat PEGylation has positive effects on lyophilizedformulation; in fact lyophilized PEGylated proteinswere found to possess shorter dissolution time,lower moisture content and increased amorphicity[163]. Furthermore, PEG may also be useful to

prevent protein aggregation and precipitation,problems of great concern in therapeutic proteinhandling and use [164]. Higher knowledge in thisfield may also improve protein stability during thestorage and avoid for example protein aggregationthat is one of the main causes of a immunogenicresponse after recombinant proteins administration[165].

A recent important novelty in the field ofPEGylation is represented by the extension ofconjugation to oligonucleotides, a recently emergingarea of therapeutics. This new application is verychallenging for the difficulties to formulate anddeliver nucleotides, and the developments in thisfield may therefore represent a really new avenue totreat pathologies.

In conclusions, one should not forget that, inaddition to all advantages obtainable by PEGyla-tion, there is also the important clinical- and cost-effectiveness of PEGylated proteins as demon-strated by clinic and economic analysis [166,167].

Acknowledgments

We thank Dr. F Greco and Dr. A Guiotto for thehelp in revising the manuscript.

References

[1] Reddy KR. Controlled-release, pegylation, liposomal for-

mulations: new mechanisms in the delivery of injectable

drugs. Ann Pharmacother 2000;34:915–23.

[2] Duncan R. Dawning era of polymer therapeutics. Nat Rev

Drug Discov 2003;2:347–60.

[3] Abuchowski A, Van Es T, Palczuk NC, Davis FF.

Alteration of immunological properties of bovine serum

albumin by covalent attachment of polyethylene glycol.

J Biol Chem 1977;252:3578–81.

[4] Abuchowski A, Mccoy R, Palczuk NC, Van Es T, Davis

FF. Effect of covalent attachment of polyethylene glycol on

immunogenicity and circulating life of bovine liver catalase.

J Biol Chem 1977;252:3582–6.

[5] Maeda H, Wu J, Sawa T, Matsumura Y, Hori K. Tumor

vascular permeability and the EPR effect in macromole-

cular therapeutics: a review. J Control Release 2000;65:

271–84.

[6] Torchilin VP, Voronkov JI, Mazoev AV. The use of

immobilized streptokinase. (Streptodekaza) for the therapy

of thromboses. Ter Ark (Ther Arch Russ) 1982;54:21–5.

[7] Khandare J, Minko T. Polymer–drug conjugates: progress

in polymeric prodrugs. Progr Polym Sci 2006;31:359–97.

[8] Ringsdorf H. Structure and properties of pharmacologi-

cally active polymers. J Polym Ci Symp 1975;51:135–53.

[9] Duncan R. Polymer conjugates as anticancer nanomedi-

cines. Nat Rev Cancer 2006;6:688–701.


[10] Langer CJ. CT-2103: a novel macromolecular taxane with

potential advantages compared with conventional taxanes.

Clin Lung Cancer 2004;6:S85–8.

[11] Sabbatini P, Brown J, Aghajanian C, Hensley ML, Pezzulli

S, O’Flaherty C, et al. A phase I/II study of PG–paclitaxel

(CT-2103) in patients (pts) with recurrent ovarian, fallopian

tube, or peritoneal cancer. Proc Am Soc Clin Oncol

2002;871.

[12] Kudelka AP, Verschraegen CF, Loyer E, Wallace S,

Gershenson DM, Han J, et al. Preliminary report of a

phase I study of escalating dose PG–paclitaxel (CT-2103)

and fixed dose cisplatin in patients with solid tumors. Proc

Am Soc Clin Oncol 2002;2146.

[13] Langer CJ, Obyrne K, Ross H, et al. Xyotax/carboplatin

vs. paclitaxel/carboplatin for the treatment of PS2 patients

with chemotherapy naıve advanced non small cell lung

cancer (NSCLC): the STELLAR 2 phase III study. Lung

Cancer 2005;49(Suppl 2):S36 [Abstract O-101].

[14] Schoemaker NE, van Kesteren C, Rosing H, Jansen S,

Swart M, Lieverst J, et al. A phase I and pharmacokinetic

study of MAG–CPT, a water soluble polymer conjugate of

camptothecin. Br J Cancer 2002;87:608–14.

[15] Meerum Terwogt JM, et al. Phase I clinical and pharma-

cokinetic study of PNU166945, a novel water soluble

polymer-conjugated prodrug of paclitaxel. Anticancer

Drugs 2001;12:315–23.

[16] Matsumura Y, Maeda H. A new concept for macromole-

cular therapeutics in cancer chemotherapy: mechanism of

tumouritropic accumulation of proteins and antitumour

agent SMANCS. Cancer Res 1986;25:865–9.

[17] Russell-Jones GJ. The potential use of receptor-mediated

endocytosis for oral drug delivery. Adv Drug Deliv Rev

1996;20:83–97.

[18] Takakura Y, Mahoto RI, Hashida M. Extravasa-

tion of macromolecules. Adv Drug Deliv Rev 1998;34:

93–108.

[19] Okamoto CT. Endocytosis and transcytosis. Adv Drug

Deliv Rev 1998;29:215–28.

[20] Allen TM. Ligand-targeted therapeutics in anticancer

therapy. Nat Rev Cancer 2002;2:750–63.

[21] Tomalia DA, Baker H, Dewald J, Hall M, Kallos G,

Martin S, et al. A new class of polymers—starburstden-

dritic macromolecules. Polym J 1985;17:117–32.

[22] Frechet JMJ. Dendrimers and hyperbranched polymers:

two families of three-dimensional macromolecules with

similar but clearly distinct properties. J Macromol Sci Pure

Appl Chem 1996;33:1399–425.

[23] Frechet JMJ, Tomalia DA. Dendrimers and other dendritic

polymers. Chichester: Wiley; 2001.

[24] Malefant PRL, Frechet JMJ. In: Frechet JMJ, Tomalia

DA, editors. Dendrimers and other dendritic polymers.

Chichester: Wiley; 2001. p. 171–96.

[25] Ferruti P, Knobloch S, Ranucci E, Duncan R, Gianasi E. A

novel modification of poly(L-lysine) leading to a soluble

cationic polymer with reduced toxicity and with potential

as a transfection agent. Macromol Chem Phys 1998;199:

2565–75.

[26] Dautzenberg H, Zintchenko A, Konak C, Reschel T, Subr

V, Ulbrich K. Polycationic graft copolymer as carriers for

oligonucleotides delivery. Complexes of oligonucleotides

with polycationic graft copolymers. Langmuir 2001;17:

3096–102.

[27] Pechar M, Ulbrich K, Subr V. Poly(ethyleneglycol) multi-

block copolymers as carrier of anticancer drug doxorubi-

cin. Bioconj Chem 2000;11:131–9.

[28] Stiriba SE, Krautz H, Frey H. Hyperbranched molecular

nanocapsules: comparison of the hyperbranched architec-

ture with the perfect linear analogue. J Am Chem Soc 2002;

124:9698–9.

[29] Mammen M, Choi SK, Whitesides GM. Polyvalent

interaction in biological system: implications for design

and use of multivalent ligands and inhibitors. Angew Chem

Int Edn Engl 1998;37:2754–94.

[30] Mirhra MK, Kobayashi S. Star and hyperbranched

polymers. Basel: Marcel Dekker; 1999.

[31] Roy R. Recent developments in the rational design of

multivalent glycoconjugates. Top Curr Chem 1997;187:

241–74.

[32] Chaves F, Calvo JC, Carvajal C, Rivera Z, Ramirez L,

Pinto M, et al. Synthesis, isolation and characterization of

Plasmodium falciparum antigenic tetrabranched polypep-

tide dendrimers obtained by thiazolidine linkages. J Pept

Res 2001;58:307–16.

[33] Ferrari M. Cancer nanotechnology: opportunities and

challenges. Nat Rev Cancer 2005;5:161–71.

[34] Greenwald RB. PEG drugs: an overview. J Control Release

2001;74:159–71.

[35] Pasut G, Guiotto A, Veronese FM. Protein, peptide and

non-peptide drug PEGylation for therapeutic application: a

review. Exp Opin Ther Patents 2004;14:859–94.

[36] Brocchini S, Duncan R. In: Mathiowitz E, editor.

Encyclopaedia of controlled drug delivery. New York:

Wiley; 1999. p. 786–816.

[37] Kopecek J, Bazilova H. Poly(N-(2-hydroxypropyl)metha-

crylamide)-1. Radical polymerisation and copolymerisa-

tion. Eur Polym J 1973;9:7–14.

[38] Lloyd JB, Duncan R, Pratten MK. Soluble synthetic

polymers as targetable agents for intracellular drug release.

Br Polym J 1983;15:158–9.

[39] Duncan R, Kopecek J, Lloyd JB. Development of N-(2-

hydroxypropyl)methacrylamide copolymers as carriers of

therapeutic agents. In: Chiellini E, Giusti P, editors.

Polymers in medicine: biomedical and pharmacological

applications. New York: Plenum Press; 1983. p. 97–113.

[40] Duncan R. Polymer therapeutics. In: Cooper E, editor.

Business briefing pharmatech. World Markets Research

Centre; 2001. p. 178–84.

[41] Duncan R, Cable R, Lloyd HC, Rejmanova P, Kopecek J.

Polymers containing enzymatically degradable bonds. 7:

design of oligopeptide sidechains in poly(N-(2-hydroxypro-

pyl)methacrylamide) copolymers to promote efficient de-

gradation by lysosomal enzymes. Makromol Chem

1983;184:1997–2008.

[42] Seymour LW, Duncan R, Strohalm J, Kopecek J. Effect of

molecular weight of N-(2-hydroxypropyl)methacrylamide

copolymers on body distribution and arte of excretion after

subcutaneous, intraperitoneal and intravenous administra-

tion to rats. J Biomed Mater Res 1987;21:1341–58.

[43] Duncan R, Kopecek J, Rejmanova P, Lloyd JB. Targeting

of N-(2-hydroxypropyl)methacrylamide copolymers to liver

by incorporation of galactose residues. Biochem Biophys

Acta 1983;755:518–21.

[44] Duncan R, Coatsworth JK, Burtles S. Preclinical toxi-

cology of novel polymeric antitumour agent: HPMA


copolymer–doxorubicin (PK1). Hum Exp Toxicol 1998;17:

93–104.

[45] Vasey P, Kaye SB, Morrison R, Twelves C, Wilson P,

Duncan R, et al. Phase I clinical and pharmacokinetic

study of PK1 (N-(2-hydroxypropyl)methacrylamide copo-

lymers doxorubicin): first member of a new class of

chemotherapeutic agents–drug–polymer conjugates. Clin

Cancer Res 1999;5:83–94.

[46] Seymour LW, Ferry DR, Anderson D, Heslewood S,

Julyan PJ, Poyner R, et al. Hepatic drug targeting: phase I

evaluation of polymer-bound doxorubicin. J Clin Oncol

2002;20:1668–76.

[47] Gianasi E, Buckley RG, Latigo J, Wasil M, Duncan R.

HPMA copolymers platinates containing dicarboxylato

ligands. Preparation, characterisation and in vitro and in

vivo evaluation. J Drug Target 2002;10:549–56.

[48] Rademaker-Lakhai JM, Terret C, Howell SB, Baud CM,

De Boer RF, Pluim D, et al. A Phase I and pharmacolo-

gical study of the platinum polymer AP5280 given as an

intravenous infusion once every 3 weeks in patients with

solid tumors. Clin Cancer Res 2004;10:3386–95.

[49] Todd R, Sludden J, Boddy AV, Griffin MJ, Robson L,

Cassidy J, et al. Phase I and pharmacological study of CT-

2103, a poly(L-glutamic acid)-paclitaxel conjugate. Proceed-

ings of ASCO annual meeting 2001;439.

[50] Singer JW, Baker B, de Vries P, Kumar A, Shaffer S,

Vawter E, et al. Poly-(L)-glutamic acid–paclitaxel (CT-

2103) [XYOTAXTM], a biodegradable polymeric drug

conjugate. Adv Exp Med Biol 2003;519:81–99.

[51] Cell Therapeutics. Improving Outcomes in PS2 Patients:

results of the XYOTAXTM Phase III STELLAR trials. In:

11th world congress on lung cancer, Barcelona, Spain July

2005.

[52] Socinski M. XYOTAX in NSCLC and other solid tumors.

Emerging evidence on biological sex differences: is gender-

specific therapy warranted? In: Chemotherapy Foundation

XXIII symposium innovative cancer therapy for tomorrow,

Mount Sinai November 2005.

[53] Ross H, Bonomi P, Langer P, O’Brien M, O’Byrne K, Paz-

Ares L, et al. Effect of gender on outcome in two

randomized phase III trials of paclitaxel poliglumex

(PPX) in chemonaıve pts with advanced NSCLC and

poor performance status (PS2). ASCO annual meeting

2006;7039.

[54] Kremer M, Judd J, Rifkin B, Auszmann J, Oursler MJ.

Estrogen modulation of osteoclast lysosomalenzyme secre-

tion. J Cellular Biochem 1995;57:271–9.

[55] Mehvar R. Recent trends in the use of polysaccharides for

improved delivery of therapeutic agents: pharmacokinetic

and pharmacodynamic perspectives. Curr Pharma Biotech-

nol 2003;4:283–302.

[56] Danhauser-Riedl S, Hausmann E, Schick HD, Bender R,

Dietzfelbinger H, Rastetter J, et al. Phase I clinical and

pharmacokinetic trial of dextran conjugated doxorubicin

(AD-70, DOX-OXD). Invest New Drugs 1993;11:187–95.

[57] Fujita T, Nishikawa M, Tamaki C, Takakura Y, Hashida

M, Sezaki H. Targeted delivery of human recombinant

superoxide dismutase by chemical modification with mono-

and polysaccharide derivatives. J Pharmacol Exp Ther

1992;263:971–8.

[58] Elliott S, Lorenzini T, Asher S, Aoki K, Brankow D, Buck

L, et al. Enhancement of therapeutic protein in vivo

activities through glycoengineering. Nat Biotechnol

2003;21:414–21.

[59] Maeda H. SMANCS and polymer-conjugated macromole-

cular drugs: advantages in cancer chemotherapy. Adv Drug

Del Rev 1991;6:181–202.

[60] Maeda H, Konno T. In: Maeda H, Edo K, Ishida N,

editors. Neocarzinostatin: the past, present, and future of

an anticancer drug. Berlin: Springer; 1997. p. 227–67.

[61] Working PK, Newman SS, Johnson J, Cornacoff JB.

Safety of poly(ethylene glycol) derivatives. In: Harris JM,

Zalipsky S, editors. Poly(ethylene glycol) chemistry and

biological applications. Washington, DC: ACS Books;

1997. p. 45–54.

[62] Choe YH, Conover CD, Wu D, Royzen M, Gervacio Y,

Borowski V, et al. Anticancer drug delivery systems: multi-

loaded N4-acyl poly(ethylene glycol) prodrugs of ara-C: II.

Efficacy in ascites and solid tumors. J Control Release

2002;79:55–70.

[63] Schiavon O, Pasut G, Moro S, Orsolini P, Guiotto A,

Veronese FM. PEG–Ara-C conjugates for controlled

release. Eur J Med Chem 2004;39:123–33.

[64] Pasut G, Scaramuzza S, Schiavon O, Mendichi R, Veronese

FM. PEG–epirubicin conjugates with high loading. J Bioac

Comp Pol 2005;20:213–30.

[65] Berna M, Dalzoppo D, Pasut G, Manunta M, Izzo L,

Jones AT, et al. Novel monodisperse PEG–dendrons as

new tools for targeted drug delivery: synthesis, character-

ization and cellular uptake. Biomacromolecules 2006;7:

146–53.

[66] Hamidi M, Azadi A, Rafiei P. Pharmacokinetic conse-

quences of pegylation. Drug Deliv 2006;13:399–409.

[67] Yang K, Basu A, Wang M, Chintala R, Hsieh MC, Liu S,

et al. Tailoring structure–function and pharmacokinetic

properties of single-chain Fv proteins by site-specific

PEGylation. Prot Eng 2003;16:761–70.

[68] Israelachvili J. The different faces of poly(ethylene glycol).

Proc Natl Acad Sci USA 1997;94:8378–9.

[69] Harris JM, Chess RB. Effect of PEGylation on pharma-

ceuticals. Nat Rev Drug Discov 2003;2:214–21.

[70] Manjula BN, Tsai A, Updhya R, Perumalsamy K, Smith

PK, Malavalli A, et al. Site-specific PEGylation of

hemoglobin at Cys-93(b): correlation between the colliga-

tive properties of the PEGylated protein and the length

of the conjugated PEG Chain. Bioconjug Chem 2003;14:

464–72.

[71] Basu A, Yang K, Wang M, Liu S, Chintala M, Palm T,

et al. Structure–function engineering of interferon-beta-1b

for improving stability, solubility, potency, immunogeni-

city, and pharmacokinetic properties by site-selective

mono-PEGylation. Bioconjug Chem 2006;17:618–30.

[72] Kawai F. Microbial degradations of polyethers. Appl

Microbiol Biotechnol 2002;58:30–8.

[73] Mehvar R. Modulation of the pharmacokinetics and

pharmacodynamic sod proteins by polyethylene glycol

conjugation. J Pharm Pharm Sci 2000;3:125–36.

[74] Beranova M, Wasserbauer R, Vancurova D, Stifter M,

Ocenaskova J, Mara M. Effect of cytochrome P-450

inhibition and stimulation on intensity of polyethylene

degradation in microsomial fraction of mouse and rat

livers. Biomaterials 1990;11:521–4.

[75] Petrak K, Goddard P. Transport of macromolecules across

the capillary walls. Adv Drug Deliv Rev 1989;3:191–214.


[76] Yamaoka T, Tabata Y, Ikada Y. Distribution and tissue

uptake of poly(ethylene glycol) with different molecular

weights after intravenous administration to mice. J Pharm

Sci 1994;83:601–6.

[77] Yamaoka T, Tabata Y, Ikada Y. Fate of water-soluble

administered via different routes. J Pharm Sci 1995;84:

349–54.

[78] Roseng L, Tolleshaug H, Berg T. Uptake, intracellular

transport, and degradation of polyethylene glicol-modified

asialofetuin in hepatocytes. J Biol Chem 1992;267:

22987–93.

[79] Kratz F, Beyer U, Schutte MT. Drug–polymer conjugates

containing acid-cleavable bonds carrier system. Crit Rev

Ther Drug Carrier Syst 1999;16:245–88.

[80] Thistlethwaite AJ, Lepper DB, Moylan DJ, Nerlinger RE.

pH distribution in human tumors. Int J Radiat Oncol

1985;11:1647–52.

[81] Shen WC, Ryser HJP. Cis-acconityl spacer between

daunomicin and macromolecula carriers: a model of pH-

sensitive linkage releasing drug from a lysosomotropic

conjugate. Biochem Byophysic Res Commun 1981;102:

1048–54.

[82] Kaneko T, Willner D, Monokovic I, Knipe JO, Braslawsky

GR, Greenfield RS, et al. New hydrazone derivatives of

adriamycin and their immunoconjugates. A correlation

between acid stability and cytotoxicity. Biocon Chem

1991;2:133–41.

[83] Coessen V, Schacht E, Domurando D. Synthesis of

polyglutamine and dextran conjugates of streptomycin

with an acid-sensitive drug–carrier linkage. J Control

Release 1996;38:141–50.

[84] Rejmanova P, Pohl J, Baudys M, Kostka V, Kopecek J.

Polymers containing enzimatically degradable bonds, 8:

degradation of oligopeptide sequences in N-(2-hydroxy-

propyl)methacrylamide copolymers by bovine spleen cathe-

psin B. Makromol Chem 1984;184:2009–20.

[85] Greenwald RB, Yang K, Zhao H, Conover CD, Lee S,

Filpula D. Controlled release of proteins from their

poly(ethylene glycol) conjugates: drug delivery system

employing 1,6-elimination. Bioconjug Chem 2003;14:

395–403.

[86] Greenwald RB, Choe YH, Conover CD, Shum K, Wu D,

Royzen M. Drug delivery systems based on trimethyl lock

lactonization: poly(ethylene glycol) prodrugs of amino-

containing compounds. J Med Chem 2000;43:475–87.

[87] Akiyama Y, Otsuka H, Nagasaki Y, Kato M, Kataoka K.

Selective synthesis of heterobifunctional poly(ethylene

glycol) derivatives containing both mercapto and acetal

terminals. Bioconjug Chem 2000;11:947–50.

[88] Zhang S, Du J, Sun R, Li X, Yang D, Zhang S, et al.

Synthesis of heterobifunctional poly(ethylene glycol) with a

primary amino group at one end and a carboxylate group

at the other end. React Funct Polym 2003;56:17–25.

[89] Santucci L, Mencarelli A, Renga B, Pasut G, Veronese FM,

Zacheo A, et al. Nitric oxide modulates proapoptotic and

antiapoptotic properties of chemotherapy agents: the case

of NOpegylated epirubicin. FASEB 2006;20:765–7.

[90] Rowinsky EK, Rizzo J, Ochoa L, Takimoto CH, For-

ouzesh B, Schwartz G, et al. A phase I and pharmacoki-

netic study of pegylated camptothecin as a 1-hour infusion

every 3 weeks in patients with advanced solid malignancies.

J Clin Oncol 2003;21:148–57.

[91] Scott LC, Evans T, Yao JC, Benson AI, Mulcahy M,

Thomas AM, et al. Pegamotecan (EZ-246), a novel

PEGylated camptothecin conjugate, for treatment of

adenocarcinomas of the stomach and gastroesophageal

(GE) junction: preliminary results of a single-agent phase 2

study. J Clin Oncology—ASCO annual meeting proceed-

ings (Post-Meeting ed.) 2004;22(14S July 15 Suppl):4030.

[92] Guiotto A, Canevari M, Orsolini P, Lavanchy O, Deuschel

C, Kaneda N, et al. Synthesis, characterization, and

preliminary in vivo tests of new poly(ethylene glycol)

conjugates of the antitumor agent 10-amino-7-ethylcamp-

tothecin. J Med Chem 2004;47:1280–9.

[93] Visentin R, Pasut G, Veronese FM, Mazzi U. Highly

efficient technetium-99m labeling procedure based on the

conjugation of N-[N(3-diphenylphosphinopropionyl)gly-

cyl]cysteine ligation with poly(ethylene glycol). Biocon

Chem 2004;15:1046–54.

[94] Veronese FM, Schiavon O, Pasut G, Mendichi R,

Andersson L, Tsirk A, et al. PEG–doxorubicin conjugates:

influence of polymer structure on drug release, in vitro

cytotoxicity, biodistribution, and antitumor activity. Bio-

con Chem 2005;16:775–84.

[95] Evig CB, Kelley EE, Weydert CJ, Chu Y, Buettner GR,

Burns CP. Endogenous production and exogenous expo-

sure to nitric oxide augment doxorubicin cytotoxicity for

breast cancer cells but not cardiac myoblasts. Nitric Oxide

2004;10:119–29.

[96] Maejima Y, Adachi S, Morikawa K, Ito H, Isobe M. Nitric

oxide inhibits myocardial apoptosis by preventing caspase-

3 activity via S-nitrosylation. J Mol Cell Cardiol

2005;38:163–74.

[97] Levy Y, Hershfield MS, Fernandez-Mejia C, Polmar SH,

Scrudiery D, Berger M, et al. Adenosine deaminase

deficiency with late onset or recurrent infections: response

to treatment with polyethylene glycol modified adenosine

deaminase. J Pediatr 1988;113:312–7.

[98] Graham LM. PEGASPARAGINASE: a review of clinical

studies. Adv Drug Deliv Rev 2003;10:1293–302.

[99] Wang YS, Youngster S, Grace M, Bausch J, Bordens R,

Wyss DF. Structural and biological characterisa-

tion of pegylated recombinant interferon a-2b and its

therapeutic implications. Adv Drug Deliv Rev 2002;54:

547–70.

[100] Bailon P, Palleroni A, Schaffer CA, Spence CL, Fung WJ,

Porter JE, et al. Rational design of a potent, long lasting

form of interferon: a 40 kDa branched poly-ethylene glycol-

conjugated interferon alpha-2a for the treatment of

hepatitis C. Bioconjug Chem 2001;12:195–202.

[101] Reddy KR, Modi MW, Pedder S. Use of peginterferon a2a(40KD) (Pegasyss) for the treatment of hepatitis C. Adv

Drug Deliv Rev 2002;54:571–86.

[102] Trainer PJ, Drake WM, Katznelson L, Freda PU, Herman-

Bonert V, van der Lely AJ, et al. Treatment of acromegaly

with the growth hormone-receptor antagonist pegvisomant.

N Engl J Med 2000;342:1171–7.

[103] Kinstler O, Moulinex G, Treheit M, Ladd D, Gegg C.

Mono-N-terminal poly(ethylene glycol)–protein conju-

gates. Adv Drug Deliv Rev 2002;54:477–85.

[104] Na DH, Lee KC, DeLuca PP. PEGylation of octreotide:

II. Effect of N-terminal mono-PEGylation on biolo-

gical activity and pharmacokinetics. Pharm Res 2005;22:

743–9.


[105] Esposito P, Barbero L, Caccia P, et al. PEGylation of

growth hormone-releasing hormone (GRF) analogues. Adv

Drug Deliv Rev 2003;55:1279–91.

[106] Bowen S, Bowen S, Tare N, et al. Relationship between

molecular mass and duration of activity of polyethylene

glycol conjugated granulocyte colony-stimulating factor

mutein. Exp Hematol 1999;27:425–32.

[107] Eliason JF, Greway A, Tare N, et al. Extended activity in

cynomolgus monkeys of a granulocyte colony-stimulating

factor mutein conjugated with high molecular weight

polyethylene glycol. Stem Cells 2000;18:40–5.

[108] Chapman AP. PEGylated antibodies and antibody frag-

ments for improved therapy: a review. Adv Drug Deliv Rev

2002;54:531–45.

[109] Caliceti P, Schiavon O, Sartore L, Monfardini C, Veronese

FM. Active site protection of proteolytic enzymes by

poly(ethylene glycol) surface modification. J Bioact Bio-

comp Polym 1993;8:41–50.

[110] Monfardini C, Schiavon O, Caliceti P, Morpurgo M,

Harris JM, Veronese FM. A branched monomethoxypo-

ly(ethylene glycol) for protein modification. Bioconjug

Chem 1995;6:62–9.

[111] Veronese FM, Morpurgo M. Bioconjugation in pharma-

ceutical chemistry. Farmaco 1999;54:497–516.

[112] Mokouyama EB, Oquchi M, Kodera Y, Maeda T, Suzuki

H. Low pKa lysine residues at the active site of sarcosine

oxidase from Corynebacterium sp. U-96. Biochem Biophys

Res Commun 2004;320:846–51.

[113] Hinds KD, Kim SW. Effects of PEG conjugation on insulin

properties. Adv Drug Deliv Rev 2002;54:505–30.

[114] Yamamoto Y, Tsutsumi Y, Yoshioka Y, Nishibata T,

Kobayashi K, Okamoto T, et al. Site-specific PEGylation

of a lysine-deficient TNF-a with full bioactivity. Nat

Biotechnol 2003;21:546–52.

[115] Zhao H, Yang K, Martinez A, Basu A, Chintala R, Liu

HC, et al. Linear and branched bicin linkers for releasable

PEGylation of macromolecules: controlled release in vivo

and in vitro from mono- and multi-PEGylated proteins.

Bioconjug Chem 2006;17:341–51.

[116] Peleg-Shulman T, Tsubery H, Mironchik M, Fridkin M,

Schreiber G, Shechter Y. Reversible PEGylation: a novel

technology to release native interferon alpha2 over a

prolonged time period. J Med Chem 2004;7:4897–904.

[117] Veronese FM, Pasut G, Tonon G, Schrepfer R. New

activated poly(ethylene glycols) and related polymers and

their applications. IT MI 2006A000612.

[118] Pasut G, Caboi F, Schrepfer R, Tonon G, Schiavon O,

Veronese FM. New active poly(ethylene glycol) deri-

vative for amino coupling. React Func Polym 2007;67:

529–39.

[119] Wylie DC, Voloch M, Lee S, Liu YH, Cannon-Carlson S,

Cutler C, et al. Carboxyalkylated histidine is a pH-

dependent product of pegylation with SC–PEG. Pharm

Res 2001;18:1354–60.

[120] Wang YS, Youngster S, Bausch J, Zhang R, Mcnemar C,

Wyss DF. Identification of the major positional isomer of

pegylated interferon alpha-2b. Biochemistry 2000;39:

10634–40.

[121] Hooftman G, Herman S, Schacht E. Review: poly(ethylene

glycol)s with reactive endgroups. II. Pratical consideration

for the preparation of protein–PEG conjugates. J Bioact

Comp Polym 1996;11:135–59.

[122] Clark R, Olson K, Fuh G, Marian M, Mortesen D,

Teshima G, et al. Long-acting growth hormone produced

by conjugation with polyethylene glicol. J Biol Chem

1996;271:21969–77.

[123] Snider J, Neville C, Yuan L-C, Bullock J. Characterization

of the heterogeneity of polyethylene glycol-modified super-

oxide dismutase by chromatographic and electrophoretic

techniques. J Chromatogr 1992;599:141–55.

[124] Delgado C, Francis GE, Malik F, Fisher D, Parkes V.

Polymer-derivatized proteins: analytical and preparative

problems. Pharm Sci 1997;3:59–66.

[125] Snyder SL, Sobocinski PZ. An improved 2,4,6-trinitroben-

zenesulfonic acid method for the determination of amines.

Anal Biochem 1975;64:284–8.

[126] Monkarsh SP, Ma Y, Aglione A, Bailon P, Ciolek D,

DeBarbieri B, et al. Positional isomers of mono-pegylated

interferon a-2a: Isolation, characterization, and biological

activity. Anal Biochem 1997;247:434–40.

[127] Veronese FM, Sacca B, Polverido de Laureto P, Sergi M,

Caliceti P, Schiavon O, et al. New PEGs for peptide and

protein modification, suitable for identification of the

PEGylation site. Biocon Chem 2001;1:62–70.

[128] Kubetzko S, Sarkar CA, Pluckthun A. Protein PEGylation

decreases observed target association rates via a dual

blocking mechanism. Mol Pharmacol 2005;68:1439–54.

[129] Federico R, Cona A, Caliceti P Veronese FM. Histaminase

PEGylation: preparation and characterization of a new

bioconjugate for therapeutic application. J Control Rel

2006;115:168–74.

[130] Rajan RS, Li T, Aras M, Sloey C, Sutherland W, Arai H,

et al. Modulation of protein aggregation bu polyethylene

glycol conjugation: GCSF as a case study. Prot Sci

2006;15:1063–75.

[131] Kinstler OB, Brems DN, Lauren SL, Paige AG, Hambur-

ger JB, Treuheit MJ. Characterization and stability of

N-terminally pegylated rhG–CSF. Pharm Res 1996;13:

996–1002.

[132] He XH, Shaw PC, Tam SC. Reducing the immunogenicity

and improving the in vivo activity of trichosanthin by site-

directed pegylation. Life Sci 1999;65:355–68.

[133] Trakas N, Tzartos N. Conjugation of acetylcholine

receptor-protecting Fab fragments with polyethylene

glycol results in a prolonged half-life in the circulation

and reduced immunogenicity. J Neuroimmunol 2001;120:

42–9.

[134] Ganson NJ, Kelly SJ, Scarlett E, Sundy JS, Hershfield MS.

Control of hyperuricemia in subjects with refractory gout,

and induction of antibody agaist poly(ethylene glycl)

(PEG), in a phase I trial of subcutaneous PEGylated urate

oxidase. Arthritis Res Ther 2006;8:R12.

[135] Richter AW, Akerblom E. Antibodies against polyethylene

glycol produced in animals by immunization with mono-

methoxy polyethylene glycol modified proteins. Int Arch

Allergy Appl Immunol 1983;70:124–31.

[136] Cheng TL, Cheng CM, Chen BM, Tsao DA, Chuang KH,

Hsiao SW, et al. Monoclonal antibody-based quantitation

of poly(ethylene glycol)-derivatized proteins, liposomes,

and nanoparticles. Bioconjug Chem 2005;16:1225–31.

[137] Cheng TL, Chen BM, Chem JW, Wu MF, Roffler SR.

Efficient clearance of poly(ethylene glycol)-modified im-

munoenzyme with anti-PEG monoclonal antibody for

prodrug cancer therapy. Bioconjug Chem 2000;11:258–66.


[138] Pasut G, Veronese FM. PEGylation of proteins as tailored

chemistry for optimized bioconjugates. In: Satchi-Fainaro

R, Duncan R, editors. Advance in polymer science, vol.

192. Berlin, Heidelberg: Springer; 2005. p. 95–134.

[139] Veronese FM, Harris JM. Introduction and overview of

peptide and protein pegylation. Adv Drug Del Rev 2002;

54:453–6.

[140] Balan S, Choi J-W, Godwin A, Teo I, Laborde CM,

Heidelberger S, et al. Site-specific pegylation of protein

disulfide bonds using a three-carbon bridge. Bioconjug

Chem 2007;18:61–76.

[141] Veronese FM, Berna M. Novel G-CSF conjugates. Patent

no. EP1737496.

[142] Sato H. Enzymatic procedure for site-specific pegylation of

proteins. Adv Drug Deliv Rev 2002;54:487–504.

[143] DeFrees S, Wang Z, Xing R, Scott AE, Wang J, Zopf D,

et al. GlycoPEGylation of recombinant therapeutic pro-

teins produced in Escherichia coli. Glycobiology 2006;16:

833–43.

[144] Uze G, Lutfalla G, Mongensen. Alpha and beta interferons

and their friends and relations. J Interferon Cytokine Res

1995;5:3–26.

[145] Wills RJ. Clinical pharmacokinetics of interferons. Clin

Pharmacokinet 1990;19:390–9.

[146] Wong SS. Reactive groups of proteins and their modifying

agents. In: Chemistry of protein conjugation and cross-

linking. Boston: CRC Press; 1991. p. 13.

[148] Piquet G, Gatti M, Barbero L, Traversa S, Caccia P,

Esposito P. Set-up of a large laboratory scale chromato-

graphic separation of poly(ethylene glycol) derivatives

of the growth hormone-releasing factor 1–29 analogue.

J Chromatogr 2002;A944:141–8.

[149] Goffin V, Bernichtein S, Carriere O, Bennett WF,

Kopchick JJ, Kelly PA. The human growth hormone

antagonist B2036 does not interact with the prolactine

receptor. Endocrinology 1999;140:3853–6.

[150] Kelly SJ, Delnomdedieu M, Oliviero MI, Williams LD,

Saifer MGP, Sherman MR, et al. Diabetes insipidus in

uricase-deficient mice: a model for evaluating therapy with

poly(ethylene glycol)-modified uricase. J Am Soc Nephrol

2001;12:1001–9.

[151] Ishida T, Ichihara M, Wang X, Yamamoto K, Kimura J,

Majima E, et al. Injection of PEGylated liposomes in rats

elicits PEG-specific IgM, which is responsible for rapid

elimination of a second dose of PEGylated liposomes.

J Control Release 2006;112:15–25.

[152] Cheng PN, Leung YC, Lo WH, Tsui SM, Lam KC.

Remission of hepatocellular carcinoma with arginine

depletion induced by systemic release of endogenous

hepatic arginase due to transhepatic arterial embolisation,

augmented by high-dose insulin: arginase as a potential

drug candidate for hepatocellular carcinoma. Cancer Lett

2005;224:67–80.

[153] Wang M, Basu A, Palm T, Hua J, Youngster S, Hwang L,

et al. Engineering an arginine catabolizing bioconjugate:

biochemical and pharmacological characterization of pe-

gylated derivatives of arginine deiminase from Mycoplasma

arthritidis. Bioconjug Chem 2006;17:1447–59.

[154] Morpurgo M, Pasut G, Veronese FM. PEGylated proteins

as cancer therapeutics. In: Torchilin VP, editor. Delivery of

protein and peptide drugs in cancer. London: Imperial

College Press; 2006. p. 85–110.

[155] Veronese FM, Pasut G. Drug–polymers conjugates.

In: Triggle DJ, Taylor JB, editors. Comprehensive

medicinal chemistry II, vol. 5. Oxford: Elsevier; 2007.

p. 1043–68.

[156] Tuerk C, Gold L. Systematic evolution of ligands by

exponential enrichment: RNA ligands to bacteriophage T4

DNA polymerase. Science 1990;249:505–10.

[157] Ng EWM, Shima DT, Calias P, Cunningham Jr ET, Guyer

DR, Adamis AP. Pegaptanib, a targeted anti-VEGF

aptamer for ocular vascular disease. Nat Rev Drug Discov

2006;5:123–32.

[158] Ruckman J, Green LS, Beeson J, Waugh S, Gillette WL,

Henninger DD, et al. 20-Fluoropyrimidine RNA-based

aptamers to the 165-amino acid form of vascular endothe-

lial growth factor (VEGF 165). Inhibition of receptor

binding and VEGF-induced vascular permeability through

interactions requiring the exon 7-encoded domain. J Biol

Chem 1998;273:20556–67.

[159] Murad KL, Mahany KL, Kuypers FA, Brugnara C, Eaton

JW, Scott MD. Structural and functional consequences of

antigenic modulation of red cells with methoxypoly(ethy-

lene glycol). Blood 1999;93:2121–7.

[160] Tan Y, Qiu Y, Xu H, Ji S, Li S, Gong F, et al. Decreased

immunorejection in unmatched blood transfusions by

attachment of methoxypolyethylene glycol on human red

blood cells and the effect on D antigen. Transfusion 2006;

46:2122–7.

[161] Scott M D, Chen A M. Beyond the red cell: pegylation of

other blood cells and tissues. Transfus Clin Biol 2004;11:

40–6.

[162] Sheth SR, Leckband D. Measurements of attractive

forces between proteins and end-grafted poly(ethylene

glycol) chains. Proc Natl Acad Sci USA 1997;94:

8399–404.

[163] Mosharraf M, Malmberg M, Fransson J. Formulation,

lyophilization and solid-state properties of a pegylated

protein. Int J Pharmaceut 2007; doi:10.1016/j.ijpharm.

2006.11.064.

[164] Hermeling S, Schellekens H, Maas C, Gebbink MFBG,

Crommelin DJA, Jiskoot W. Antibody response to

aggregated human interferon alpha2b in wildtype and

transgenic immune tolerant mice depends on type and level

of aggregation. J Pharm Sci 2006;95:1084–96.

[165] Schellekens H. Immunogenicity of therapeutic proteins:

clinical implications and future prospects. Clin Ther 2002;

24:1720–40.

[166] Kurre HA, Ettinger AG, Veenstra DL, Gaynon PS,

Franklin J, Sencer SF, et al. A pharmacoeconomic analysis

of pegaspargase versus native Escherichia coli L-asparagi-

nase for the treatment of children with standard-risk, acute

lymphoblastic leukemia: the Children’s Cancer Group

study (CCG-1962). J Pediatr Hematol Oncol 2002;24:

175–81.

[167] Sheperd J, Brodin HF, Cave CB, Waugh NR, Price A,

Gabbay J. Clinical- and cost-effectiveness of pegylated

interferon alfa in the treatment of chronic hepatitis C: a

systematic review and economic evaluation. Int J Technol

Assess Health Care 2005;21:47–54.

[168] Bukowski R, Ernstoff MS, Gore ME, Nemunaitis JJ,

Amato R, Gupta SK, et al. PEGylated interferon a2btreatment for patient with solid tumors: a phase I/II study.

J Clin Oncol 2002;20:3841–9.

dx.doi.org/10.1016/j.ijpharm.2006.11.064

dx.doi.org/10.1016/j.ijpharm.2006.11.064


[169] Chapman AP, Antoniw P, Spitali M, West S, Stephens S,

King DJ. Therapeutic antibody fragments with prolonged

in vivo half-lives. Nat Biotechnol 1999;17:780–3.

[170] Schreiber S, Rutgeerts P, Fedorak RN, Khaliq-Kareemi M,

Kamm MA, Boivin M, et al. CDP870 Crohn’s Disease

Study Group. Gastroenterology 2005;129:807–18.

[171] O’Brien M, Sandler A, Popovish A, et al. Xyotax vs.

gemcitabine or vinorelbine for the treatment of perfor-

mance status 2 patients with chemotherapy naive non small

cell lung cancer (NSCLC): the STELLAR 4 phase III

study. Lung Cancer 2005;49(Suppl 2):S37 [Abstract O-104].

[172] De Vries P, Bhatt R, Stone I, Klein P, Singer J.

Optimisation of CT2106: a water soluble poly-L-glutamic

acid (PG)–camptothecin conjugate with enhanced in vivo

antitumor efficacy. In: Proceedings of the AACR-NCI-

EORTC symposium, 2002, Miami Beach, FL.

[173] Greenwald RB, Gilbert CW, Pendri A, Conover CD,

Xia J, Martinez A. Drug delivery systems: water

soluble taxol 20-poly(ethylene glycol) ester prodrugs-

design and in vivo effectiveness. J Med Chem 1996;39:

424–31.

[174] /http://www.i-s-b.org/business/rec_sales.htmS.

http://www.i-s-b.org/business/rec_sales.htm

Documents

Polymer–drug conjugation, recent achievements and general strategies