Current Proteomics

3

Current Proteomics

Send Orders for Reprints to [email protected] Current Proteomics, 2018, 15, 3-12

REVIEW ARTICLE ISSN: 1570-1646 eISSN: 1875-6247

Conjugative Post-Translational Modifications for Pharmacological Improvement of Therapeutic Proteins

The journal for current and in-depth reviews on proteomics

Impact Factor: 0.59

BENTHAM SCIENCE

Diego de Carvalho Carneiro1,* and Suzana Telles da Cunha Lima2 1

Department of Biointeraction, Institute of Health Sciences, Federal University of Bahia, Salvador, Brazil; 2Department of General Biology, Institute of Biology, Federal University of Bahia, Salvador, Brazil

ARTICLE HISTORY Received: February 26, 2017 Revised: May 18, 2017 Accepted: June 14, 2017 DOI: 10.2174/1570164614666170623103718

Abstract: Therapeutic proteins can be pharmacologically improved by conjugative post-translational modifications through the rational design of their structure and production process. N- and O- linked glycosylation can confer advantages to proteins and can be controlled by the process conditions, producing cell lines and enzymatic expression or activity. The resulting glycan profile influences their pharmacological features such as in the biopharmaceutical velaglucerase alfa. Conjugation of polyethylene glycol onto therapeutic proteins has been used to overcome pharmacological limitations. They can be site-specifically or randomly linked through the reaction between a terminal group of the polyethylene glycol molecule and an amino acid residue group of the protein. Both components (protein and polyethylene glycol) participate in pharmacological mechanisms, enabling a longer half-life for peginterferon beta-1a, for example. Fatty acylation of proteins occurs in cells via some known reactions that may involve different linkages, fatty acids, modified residues, and enzymes. Insulin degludec is a fatty acylated biopharmaceutical formulated to solve the problem of variability in insulin exposure associated with other products. Conjugative post-translational modifications can be used to pharmacologically improve even therapeutic proteins produced by prokaryotic cell lines while exploring them may result in the development of other potential uses as well as new generations of biopharmaceuticals.

Keywords: Biopharmaceuticals, fatty acylation, glycosylation, PEGylation, pharmacology, PTMs. 1. INTRODUCTION Therapeutic proteins regularly undergo post-translational modifications (PTMs) in order to consistently form their functional structure, especially when produced by eukaryotic cell lines. Some PTMs are processed by the cellular metabolism during industrial production of protein biopharmaceuticals while others are artificially engineered into the protein structure to improve their pharmacological features [1]. PTMs are additional chemical modifications taken after mRNA templates are readout by ribosomes. They are usually executed by enzymes through peptide bond cleavages or modification of amino acid residues. The functionality of many proteins requires the conjugation of molecules (e.g. glycans) that take part in the protein’s chemical activity, intracellular localization, or tertiary structure [2]. PTMs can change the chemical and physical properties of proteins such as shape, charge, molecular weight, and interactions with *Address correspondence to this author at the Department of Biointeraction, Institute of Health Sciences, Federal University of Bahia, Salvador, Brazil; Tel: +55-71-99405-8012; E-mail: [email protected]

1875-6247/18 $58.00+.00

other molecules, enabling the exploitation of those changes through some techniques to separate, identify, and characterize proteins and their PTMs (e.g. two-dimensional electrophoresis, liquid chromatography, western blotting, immunoaffinity-based approaches, etc.) [3]. Tandem mass spectrometry (MS/MS) has been used to delve the composition and structure of proteins. In this method, a previous mass spectrometry (MS) analysis of the precursor ions is performed before MS analysis of the dissociated ions produced by gas-phase activation. Since the amide bonds in proteins require collision energies higher than e.g. glycosidic or phosphate bonds present in some PTMs, ion activation by electron transfer dissociation (ETD) or electron capture dissociation (ECD) can be used to prevent the loss of PTMs during MS/MS analysis, giving means to elucidate their composition and structure [4]. Eukaryotic cells (mainly mammalian cell lines) are preferentially selected over prokaryotic ones in processes involving biopharmaceutical manufacturing since they are the best expression systems for production of proteins with proper post-translational modifications. Chinese hamster ovary © 2018 Bentham Science Publishers

4 Current Proteomics, 2018, Vol. 15, No. 1

Carneiro and Lima

(CHO) cells are the most regularly utilized mammalian cells for industrial manufacturing of therapeutic proteins. CHO cells are more advantageous than other cell lines because they can vigorously grow in suspension culture with chemically-defined or serum-free media, they are safe for pathogenic viruses, and they are able to express proteins with PTMs similar to those found in humans. Moreover, generating engineered mammalian cells expressing a specific gene for production of proteins with quality and high-yield is easily achieved by CHO cell cultures [5].

industries to improve the pharmacological features of therapeutic proteins for humans. Table 1 summarizes the therapeutic proteins approved for treatment in humans by the FDA from 2002 to 2016 that were pharmacologically improved by conjugative PTMs [9, 10]. 2. GLYCOSYLATION Glycosylation is the PTM most commonly found in proteins, and above 60% of therapeutic proteins are glycosylated. This PTM consists of the attachment of glycans to amino acid residues and takes place via complex pathways in the endoplasmic reticulum and Golgi apparatus, generally by an N-linked glycosylation or an O-linked glycosylation. Nlinked glycosylation occurs through the nitrogen atom of asparagine side chain, and O-linked glycosylation occurs through the oxygen atom of either serine or threonine side chain. Figs. (1a and 1b) show a simplified chemical reaction representation of O-linked and N-linked glycoprotein formation [11]. Less frequently found, C-mannosylation is a Clinked glycosylation that occurs through the C5 carbon atom of a tryptophan residue and the C1 carbon atom of a mannose molecule [12]. The biological advantages of attaching glycans to proteins have been uncovered. For instance, hydrophilic clusters of oligosaccharides are able to alter the protein’s polarity and solubility, and oligosaccharides added after protein synthesis work as markers for its destination in the cell (along the secretory pathway, for example) and for the degradation of misfolded proteins. Glycans can also contribute to tertiary structure formation of proteins, confer protection against proteolytic enzymes, and provide specific biologic activity to glycoproteins [13].

Modification of therapeutic proteins by PTMs can have an impact on the protein’s yield, stability, pharmacokinetics, pharmacodynamics, and immunogenicity. Understanding the association between structural modification of proteins and their function can give means to enhance pharmacological features and processing controls to secure product quality through the rational design of therapeutic proteins. Preselected modifications and improved properties may result in the development of new generations of biopharmaceuticals [6]. Although fusion proteins are successful approaches to improve biological and pharmacological properties of biopharmaceuticals such as plasma half-life, the majority of fusion proteins results from genetic fusion technologies (tandem fusion and domain insertion). The conjugation of proteins is also an auspicious technology to make biobetters. This approach may circumvent the limitation of combining domains’ diversity seen in the genetic fusion technologies by cross-linking tagged proteins using chemical reagents or enzymes. However, the post-translational protein conjugation is not yet an established technology for production of therapeutic proteins [7, 8].

Nutrient constitution, pH, temperature, oxygenation, and ammonia content are some of the culture conditions that may contribute to the distribution of carbohydrate moieties of

This article intends to review the conjugative posttranslational modifications currently used by pharmaceutical Table 1.

Therapeutic proteins pharmacologically improved by PTM conjugates and approved by the FDA for human treatment from 2002 to 2016. Protein

Analogue Product

Company

Therapeutic Indication

PTM

Insulin

TRESIBA ®

Novo Nordisk

Diabetes type 1 and 2

Fatty acylation

Biogen

Relapsing forms of multiple sclerosis

PEGylation

Savient Pharmaceuticals

Chronic refractory gout

PEGylation

Shire Pharmaceutical

Gaucher disease

Glycosylation

Schering-Plough

Chronic hepatitis C

PEGylation

Interferon beta-1a Urate oxidase β-glucocerebrosidase

PLEGRIDY

®

KRYSTEXXA VPRIV

®

® ®

Interferon alfa-2b

PEGINTRON

Tumor necrosis factor monoclonal antibody Fab' fragment

CIMZIA ®

UCB

Crohn's disease

PEGylation

Erythropoietin

MIRCERA ®

Hoffmann-La Roche

Anemia associated with chronic renal failure

PEGylation

Insulin

LEVEMIR ®

Novo Nordisk

Diabetes type 1 and 2

Fatty acylation

Pharmacia Corp.

Acromegaly

PEGylation

Hoffmann-La Roche

Chronic hepatitis C

PEGylation

Amgen

Febrile neutropenia

PEGylation

Human growth hormone

SOMAVERT

®

®

Interferon alfa-2a

PEGASYS

Granulocyte-colony stimulating factor

NEULASTA ®


Current Proteomics, 2018, Vol. 15, No. 1

5

(a) N-linked Glycosylation NH H2O

CH2OH O OH

HN

GlcNAc OH

+

C O

O

C O H2 C CH C

C

C

GlcNAc OH

C O H2 C CH Asn NH

O

Asn NH

NH

O

CH2OH H ON

NH

H

C

O

NH

O

C

O

O

CH3

CH3 (b) O-linked Glycosylation O

O OH

CH2

NH

CH2 O

GalNAc OH

+ OH

Ser C O H2 O C CH H

NH

C

NH

O

C O H2 C CH Ser NH

O

NH

H2O

C

OH GalNAc OH

NH

O

C

O

O

C

O

CH3

CH3 (c) Inhibition of Mannosidase I in an N-linked Glycan Kifunensine HOH2C Man

HO HO

OH O

HOH2C Man

HO HO

Mannosidase I O

O

HOH2C

O

NH H N

O

C O HO GlcNAc NHCOCH 3

C O H2 C CH Asn NH C

O

Fig. (1). (a, b) N- and O-linked glycosylation of therapeutic proteins. (c) Inhibition of mannosidase I α-1,2-mannose cleavage by kifunensine in an N-glycan. GlcNAc and GalNAc are monosaccharides at the reducing end of a glycan. The dashed lines indicate the sites where the oligosaccharide or polypeptide chains continue. Asparagine (Asn) and Serine (Ser) are amino acid residues from the primary structure of a protein. Ser has been used to illustrate the O-linked glycosylation reaction although Thr can also be involved.

therapeutic proteins. For instance, specific glutamine and glucose levels in the culture medium may produce less sialylated or more mannosylated glycans. The presence of ammonia can decrease the formation of terminal sialic acid content in O-linked glycans. The medium’s pH can interfere in the pH of the Golgi apparatus, causing reduced activity of enzymes participating in glycosylation, and a glycan profile change may occur at high dissolved oxygen levels [14]. Glycosylation profile and level can also vary with the producing cell line and the culture production mode (e.g. CHO cell’s increased sialylated glycoforms in perfusion mode rather than fed-batch mode). Overexpressing enzymes such as β 1,4-galactosyltransferase and α-2,3-sialyltransferase can render higher terminal sialylation of glycans [15]. In addition, glycoengineering P. pastoris is an auspicious approach to express more mammalian homologous counterparts of re-

combinant therapeutic proteins by cleaving unwanted yeasttype glycans, especially O-linked glycans with linear chains of α-linked mannose residues or β-linked mannose content. It can be achieved by genetic elimination of beta- and phosphomannose followed by reduction of O-linked glycans through the addition of mannosidases such as jack bean mannosidase or, preferably, lysosomal mannosidases [16]. The pharmacology of therapeutic glycoproteins relies on their oligosaccharide constituents as determinants for their pharmacological properties. Glycosylation is involved in modulation of interactions between proteins, and a therapeutic protein’s immunogenicity may be related to a glycosylation pattern different from that generated in the human body or the lack of a specific glycosylation pattern. For example, B-cell activating factor receptors with a Fc region have de-


creased clearance rate as its content of sialic acid increases. The amount of sialylated glycans can also induce production of neutralizing antibodies by T- and B- cells against therapeutic proteins [17]. Therapeutic IgG Fc antibodies constituted of glycans with low fucose content can enhance antibody-dependent cell-mediated cytotoxicity and antibodydependent cell-mediated phagocytosis. As a result, an effort has been made to engineer low fucose or afucosylated IgG Fc antibodies through genetic modification or enzymatic inhibition of biosynthetic pathways in host cell lines and enzymatic modification into a desired glycoprofile [18]. The affinity of erythropoietin (EPO) to its receptor is modulated by the EPO’s glycoprofile. Hematopoietic activity increases with increasing amount of sialic acid in EPO’s glycans and prolonged half-life. Also, sialylated oligosaccharides at Asn 38 and 83 are required to cover EPO’s hydrophobic surface in order to keep its solubility and prevent protein-protein aggregation, which prolongs the half-life and improves glomerular filtration of EPO as well. When the same sialylated glycans are attached to nonnative positions, EPO does not display hematopoietic activity, indicating that its bioactivity depends on glycosylation position and pattern [19]. In this section, a discussion about an example of therapeutic protein and its pharmacological improvements using a glycosylation modification is presented below. Gaucher disease (GD) is caused by a hereditary deficiency of β -glucocerebrosidase, which is responsible for cleaving the glucose and ceramide molecules that form glucocerebroside, resulting in its accumulation in the lysosomes of macrophages. VPRIV® or velaglucerase alfa has the normal human sequence of β-glucocerebrosidase and provides a more advantageous enzyme replacement therapy [20]. A long-term study of phase III clinical trial data has shown that VPRIV® has satisfactory clinical response and good safety and tolerability standards [21]. A complete characterization has demonstrated that velaglucerase alfa has highmannosylated glycans whose exposed mannose residues increase its cellular uptake by macrophages while targeting their endocytotic system mediated by mannose receptors. This glycoprofile is achieved by adding a mannosidase I inhibitor called kifunensine in the culture medium during the production of velaglucerase alfa (Fig. 1c). On the other hand, imiglucerase, which is another biopharmaceutical used to treat GD, has its mannose residues exposed by using exoglycosidases. However, it is contentious if differences in the glycosylation of those therapeutic proteins in fact impact on their uptake into macrophages [22, 23]. 3. PEGYLATION PEGylation is a PTM strategy that has been developed to overcome pharmacokinetic (PK) and pharmacodynamic (PD) limitations of therapeutic proteins. It consists of the attachment of polyethylene glycol (PEG) chains to the protein structure. PEG is a polymer of ethylene oxide monomers that has been considered safe for pharmacological use in humans. Besides proteins, PEG can also be used for pharmacological improvement of low molecular weight drugs and drug delivery systems such as hydrogels, liposomes, microparticles, and nanocarriers [24]. Even though this method has become

Carneiro and Lima

the most widely used to improve the pharmacological features of therapeutic proteins, it has presented concerns like signs of vacuolation caused by PEGylated proteins in animal models, lack of degradation in systemic circulation for conjugated PEGs with molecular weight higher than renal clearance, and detection of anti-PEG antibodies in animal models and patients. For this reason, an effort to search for PEG alternatives that solve those limitations or bring additional benefits has been growing. Non-degradable and degradable alternative polymers have been investigated such as poly(Nvinylpyrrolidone), polyglycerol, poly[oligo(ethyleneglycol) methyl methacrylate], poly(zwitterions), poly(amino acid)based hybrid materials, and others [25]. The first generation of PEGylated therapeutic proteins is characterized by linear PEGs with low molecular weight. To conjugate PEG to a protein, it is necessary the activation of the PEG through a functional group from one of its terminal sides that is suitable for reaction with an available reactive group on the protein (usually the amino groups on the side chain of lysine and N-terminus). Other end groups can also be part of the PEG’s structure when it is prepared with a suitable initiator or termination reagent. Monomethoxy PEG (mPEG) is more suitable to conjugate proteins since it can produce reactive PEGs that prevent formation of crosslinked polypeptides. Fig. (2) lists the scope of first generation PEG derivatives of PEGylated proteins with a PEG and a protein’s amino acid residue linked through either an alpha or epsilon amino group. It is suggested that the PEG’s ability to precipitate proteins, minimize immunogenicity and antigenicity, keep away proteins and cells from surfaces, and avoid degradation by enzymes or mammalian cells is related to its amphiphilic property, capacity to bind to water molecules, and backbone chain’s flexibility [26]. The second generation has been developed to solve the problems of the first generation such as impurities, limitation to low molecular weight, weak linkages, side reactions, and deficiency of site-specificity. Site-specific PEGylation can form a homogeneous product with PEG chains attached to specific amino acid residues that do not impact on the biological activity of the therapeutic protein while maintaining the PEG’s pharmacological benefits. Encoding single amino acid residues or sequences is a method used for site-specific modification available to create specific PEG conjugates. Likewise, incorporating a functionality that does not exist naturally in the cell (i.e. bio-orthogonal group) can be used to form a covalent bond with a complementary functional group at the terminal side of PEG. [27]. For example, a free cysteine can be encoded into a protein so that a terminal maleimide group attached to PEG can react with the cysteine, forming a covalent bond (Fig. 3a) [28]. In the same fashion, when a polyhistidine tag (His-tag) is encoded into a protein, it can be PEGylated using the reagent Ni– nitrilotriacetic acid (Ni-NTA) covalently bound to a PEG, which forms a noncovalent complex between its nickel ion and two of the histidine residues (Fig. 3b) [29]. Noncovalent PEGylation is a prospective approach to overcome the affinity reduction of PEGylated therapeutic proteins. While linking PEG in a nonpermanent way, the PEG chains can be released from the protein at a particular moment in order to allow the protein to perform its pharmacological activity


O

N

R-NH2

N

nO

N

O

*

N

nO N

N

HN

Cl O

*

* O

O2 S

nO

H2 C

*

O

O

CF3

O O H H O 2 2 O C C C C O N n

R-NH2

R-NH2

O

*

nN H

R

R

O H H O 2 2 H O C C C C N n

* O

R

O

O *

7

Cl

Cl *


O nO

C

O

N O N N

O nO

C

O

R-N H

2

N R-NH 2

O *

O

nO

C

O

NO2 Cl

O

nO

C

O

O *

O

nO

C

O *

O

nO

C

H N

R

H2 R-N

O *

R-NH2

Cl

N

H2 R-N

Cl

N

Fig. (2). First generation PEG derivatives of PEGylated proteins. R-NH2 represents an amino group from either the side chain or the N-terminus of a therapeutic protein’s amino acid residue.

without steric hindrance or folding alteration [30]. Furthermore, an enzyme that recognizes a specific amino acid sequence, amino acid residue, glycan’s terminal monosaccharide, or bio-orthogonal group can be used to catalyze the linkage of PEG chains. For instance, transglutaminase (TGase) can catalyze the link between a PEG molecule with a terminal primary amine and the carboxamide group of a glutamine that must be positioned in a flexible loop in order for the acyl transfer reaction occurs (Fig. 3c) [31]. The oral absorption rate of polyethylene glycol decreases as its molecular weight (MW) increases until 3350 kDa (when absorption becomes almost absent) while, after parenteral administration, the metabolic clearance decreases and the elimination in urine increases with higher MW. Also, compared to non-PEGylated proteins, the PEGylated ones usually have slower absorption rate after parenteral administration and a range of 1 to 3 days of time to maximum concentration. Initially, the PK of PEGylated proteins is governed by both the protein and PEG components. Then, their

PK and biodistribution are driven by mechanisms associated to the PEG only. Cellular mechanisms of denaturation and proteolysis are involved in removing the protein part from circulation whereas target-mediated uptake and pinocytosis participate in PEG-related processing mechanisms of clearance. Since renal filtration of PEG from PEGylated proteins is the last elimination pathway of high MW PEGs (≥ 30 kDa), it is proposed that rod-shaped and rigid molecules are able to pass through renal glomeruli, rendering their observed clearance. PEG and/or PEGylated proteins may be cleared directly by the kidney or through a fluid-phase pinocytosis, so the tissues and cells where the PEG components of the proteins are not processed by lysosomal enzymes may form lysosomal distension and vacuolation [32]. Some biopharmaceuticals have been associated with anti-PEG antibodies and PEG-induced complement activation although many therapeutic proteins have been considered safe for a long time. A response against their protein component is usually explained as the reason for the immunogenic reactions observed, and a validated standard anti-PEG assay


Carneiro and Lima

0 N

0 n*

0

0 X

0 0

0 n

N

NH2 0

HHHHHH

0 0

as

e

n*

0

0 N 0 Ni+2

0 n*

0

0

NH2

TG

*

S

(a)

X= X=

0

SH

X=

(b)

0

(c)

* n

0 0

0

H N

0

HN

N 0 Ni+2 0 0 N N H N

HHHH

0

0 n*

0 0 0 NH

0H

0 0

Fig. (3). Site-specific PEGylation of therapeutic proteins. X generically represents a specific site of a therapeutic protein used for PEGylation. (a) Linkage of PEG through the reaction of the maleimide group on the terminal side of PEG and the sulfide group of a free cysteine previously encoded into the protein. (b) PEGylation of a therapeutic protein using the reagent Ni-NTA-PEG, which is able to form a complex between two histidine residues of a His-tag and the nickel ion. (c) Enzymatic linkage of PEG via TGase catalysis of a primary amine on the terminal side of PEG and the carboxamide group of a glutamine.

needs to be developed to screen for anti-PEG antibodies [33]. A discussion about an example of therapeutic protein and its pharmacological improvements using PEGylation is described below. Interferons type 1 (mainly IFN-α and IFN-β) belong to a family of related proteins that play a variety of biological functions such as antiretroviral, antiproliferative, and immunomodulatory activities, which provide the basics for the pharmacological benefits of IFN therapies. As many therapeutic proteins with low molecular weight, INFs have short

serum half-life, so PEGylation has been used as a strategy to improve its PK and PD properties. Plegridy® or peginterferon beta-1a is a recombinant IFN-β-1a therapeutic protein with an N-terminus PEGylation, which is specific to a site that does not participate in receptor binding, and it is used for treatment of relapsing-remitting multiple sclerosis (RRMS). RRMS is a complex pathogenesis that encompasses inflammation, demyelination, and axonal damage. Conjugating PEG to IFN-β-1a resulted in growth of systemic exposure during different routes of administration. Compar-


ing to the non-PEGylated form, PEG-IFN-β-1a has longer terminal half-life and elevated PD markers for receptor activation by IFNs type 1. Subcutaneous administration of 125 µg of PEG-IFN-β-1a every two weeks during 48 weeks can significantly reduce relapses and annualized relapse rate, risk of disability progression, T2 lesions counts, and tertiary magnetic resonance imaging measures [34, 35]. Dosing every 2 weeks was demonstrated to be more efficient than every 4 weeks. Maximum serum concentration was observed after 1-1.5 days of administration, and mono-phasic decrease and median half-life were found between 2 and 3 days. The sustained neopterin growth during 10-14 days indicated cumulative duration of peginterferon beta-1a’s activity. The increase of induced neopterin and reduction of reversible lymphocyte count are consistent with the pharmacology of the IFN type 1 family [36, 37]. 4. FATTY ACYLATION The major cellular fatty acylation reactions of proteins are N-myristoylation, S-palmitoylation, and membranebound O-acyl transferase (MBOAT) fatty acylation of secreted proteins. Fatty acylation can improve membrane binding and form interactions between proteins through a hydrophobic pocket insertion. Table 2 summarizes the major cellular fatty acylation reactions of proteins currently known [38]. N-myristoylated proteins generally have the sequence MetGly-X-X-X-Ser/Thr on N-terminus. Once, Gly is exposed as the N-terminus via the removal of Met by methionine aminopeptidases, N-myristoyl transferase (NMT) catalyzes the transfer of a myristate molecule (through myristoyl-CoA) to the N-terminal Gly, forming an amide covalent bond. NMT has a hydrophobic biding pocket that accommodates 14 carbons, which confers specificity for myristate [39]. S-palmitoylation is catalyzed by palmitoyl acyltransferases, which are members of the DHHC protein family (named after its conserved sequence motif Asp-His-His-Cys). In S. Cerevisiae, it initiates with the reaction between palmitoylCoA and the cysteine of the DHHC sequence, forming an acyl-DHHC intermediate. Then, this intermediate transfers the palmitate molecule to a cysteine positioned near the Nor C- terminus or within the acceptor protein substrate [40]. MBOAT enzymes such as Hedgehog acyltransferase (Hhat), Porcupine (Porcn), and ghrelin O-acyl transferase (GOAT) catalyzes fatty acylation of secreted proteins, also by intermediation of CoA. Hhat couples a palmitate molecule to an N-terminal Cys of hedgehog family proteins via amide linkage Table 2.


9

(N-palmitoylation) [41]. Porcn catalyzes an oxyester linkage between cis-Δ9-palmitoleate (palmitoleic) and a conserved serine residue (Ser209) in Wnt proteins (Opalmitoleoylation) [42]. GOAT links an octanoate molecule to the serine on the third position of the N-terminal sequence GSSFL of the protein ghrelin (O-octanoylation) [43]. Chemical reporters have been developed to assay the biological functions of protein fatty acylation using bioorthogonal linkage approaches. The reporter alk-12 can label N-myristoylated proteins and other types of fatty acylated proteins such as S-palmitoylated ones. Hence, hydroxylamine is used to hydrolyze the ester and thioester linkages found in the other forms of fatty acylation, selecting the Nmyristoylated proteins for detection and profiling. In addition, a combination of alk-12 and in-gel fluorescence has been developed to study N-myristoylation in cancer cells and NMT as an antimalarial drug target. N-palmitoylation has been shown to have a significant role in Sonic Hedhehog signaling in human carcinoma cells using the alk-15 reporter. S-palmitoylation can be detected and profiled using the alk16 reporter in several applications such as revealing Spalmitoylated proteins and their roles. Alk-16 can also label the O-palmitoleoylation of Ser209 by porcupine in Wnt proteins [44]. Since S-acylation is a reversible protein modification (i.e. carried out by acyltransferases and reversed by acylthioesterases), it is used as a dynamic control of many ligand- and voltage-gated ion channels. It may regulate channel trafficking and control the kinetics and modulation of ion channels by other PTMs. S-acylation may modulate ion channel function through the modification of preforming and regulatory subunits and control of signaling, scaffolding, and adapter proteins [45]. Reversible lysine fatty acylation (e.g. Lys-myristoylation) has been reported as a regulator of sTNF-α secretion by helping to target TNF-α for degradation in lysosomes and reducing its placement in recycling endosomes [46]. Numerous proteins have been identified to be palmitoylated such as TLRs 2, 5, and 10 (toll-like receptors), which may represent a new mechanism for modulatory regulation of inflammatory activity [47]. The fatty acylation of insulin discussed below is an example of the potential for pharmacological improvement of therapeutic proteins by fatty acid conjugation. Insulin degludec (IDeg) or Tresiba® is a long-acting human insulin analog for basal glycemia control of diabetes

Major cellular fatty acylation reactions of proteins. Reaction

Molecule

Linkage

Modified Residue

Enzyme

N-myristoylation

Myristate (14:0)

Amide

Gly

N-myristoyl transferase (NMT)

S-palmitoylation

Palmitate (16:0)

Thioester

Cys

Palmitoyl acyltransferase (DHHC)

N-palmitoylation

Palmitate (16:0)

Amide

Cys

Hedgehog acyltransferase (Hhat)

O-octanoylation

Octanoate (8:0)

Oxyester

Ser

Ghrelin O-acyl transferase (GOAT)

O-palmitoleoylation

Palmitoleic (16:1 Δ9)

Oxyester

Ser

Porcupine (Porcn)


Carneiro and Lima

type 1 and 2 produced through recombinant DNA technology for subcutaneous injection and engineered to solve the problem of variable insulin exposure associated with other basal insulins. The name “degludec” addresses three of its features: absence of ThrB30 (“de”), addition of a glutamic acid on the side chain of LysB29 through a non-standard peptide bond (“glu”), and attachment of a dicarboxylic acid (thapsic or hexadecanedioic acid) to the α -amino group of the added Glu (“dec”). Fig. (4) illustrates the structure of insulin degludec. Since IDeg loses the positive charge of LysB29 and gains one negative charge from the hexadecanedioic acid, it can be formulated into a solution with phenol at pH 7.4, which is more suitable for the subcutaneous neutral environment, forming stable dihexamers. After subcutaneous injection, these dihexamers cluster into long multihexamers that are absorbed slowly because of their large molecular weight, creating a hypodermic depot. The insulin degludec’s monomers gradually dissociate from the multihexamers into the circulatory system, where the conjugated fatty acyl group also mediates biding to serum albumin, resulting in a slow and even delivery with consistent day-to-day exposure and no hyperglycemic peaks. IDeg has a half-life of 25.4 hours and can be found in the bloodstream for more than 120 hours. Its exposure reaches a stable and steady state after 2 or 3 doses, and the dose does not need to be adjusted every day [48-50].

CONCLUSION AND FUTURE PERSPECTIVES Conjugative PTMs can be engineered onto the structure of therapeutic proteins by biotechnology industries at the upstream or downstream processing. Even though eukaryotic cells are able to produce therapeutic proteins with complex folding, multiple subunits, and consistent profile of PTMs (which is essential for the safety and efficacy of biopharmaceuticals), prokaryotic cell lines grow faster in relatively inexpensive culture medium, reaching high cell densities, and can have their behavior and product yield better controlled and predicted during the bioprocess, making the production more efficient and economic. Therefore, conjugative PTMs are advantageous for industrial production since therapeutic proteins can also be pharmacologically improved by them even when produced by prokaryotic cell lines, resulting in an efficient and high yield manufacture of biopharmaceuticals. Other potential uses for conjugative PTMs could be explored such as helping complex folding, preventing formation of inclusion bodies in production systems with prokaryotic cells, and assisting protein isolation and purification. Also, conjugative PTMs could be used to decrease or reverse tissue cross-reactivity (TCR) and exaggerated on-target effects of therapeutic proteins by promoting steric hindrance between the specific sites of the therapeutic protein and the off-target receptor responsible for the TCR, and by minimizing the therapeutic protein’s affinity for the receptor responsible for the exaggerated on-target effect.

Insulin degludec has been confirmed to be safe and effective in patients with diabetes mellitus type 1 and 2. It is able to significantly reduce insulin dose and frequency of hypoglycemia in diabetes type 1, but not in diabetes type 2. However, comparing with other long-acting insulin analogs, IDeg presents better glycemic control for diabetes type 2. Also, IDeg can minimize hyperglycemic peaks along the day, including dawn and dusk, which lowers the risk of nocturnal episodes of hypoglycemia [51, 52]. The combination of insulin degludec and insulin aspart (IAsp) can integrate the longacting and the rapid-acting features of IDeg and IAsp, respectively, and it may be a promising alternative, especially for treatment of diabetes type 2. This combination has the additional benefit of targeting postprandial glucose, which is proposed to be related to cardiovascular disease [53].

Whereas numerous PTMs still need to have their functions and mechanisms of action uncovered and developed to new perspectives, the diversity of PTMs can be expanded by new conjugative molecules. They may be chosen from a pool of molecules that do not naturally participate in cellular mechanisms but have the potential to alter the protein’s structure, interaction, and pharmacological activity. Those PTMs need to be ideally chosen for efficient and economic production as well as for suitability to isolation and purification strategies. Moreover, it is essential that they do not form harmful side products in the organism and produce side

Des(B30) LysB29(!-Glu N"-hexadecandioyl) human insulin S A1

G

I

V

E

Q

C

C

T

S S

I

C

S

L

Y

Q

L

E

N

Y

S F

V

N

Q

H

L

C

A21

N

LysB29 HO O

S

S B1

C

S G

S

H

L

V

E

A

L

Y

L

V

C

G

E O

HO O

Hexadecandioyl

R HO

G

F

F

Y

T

P

DesB30

Thr

O H N

N H

L-!-Glu O

Fig. (4). Structural composition of insulin degludec. A1 and A21 indicate the 1st and 21st amino acids of insulin chain A. B1 and B29 (Lys) indicate the 1st and 29th amino acids of insulin chain B. ThrB30 (30th amino acid of chain B) has been omitted in insulin degludec. L-γ-Glu indicates the location of the Glu residue linked to the ε-amino group of LysB29. Hexadecandioyl is the fatty acid side chain attached to the αamino group of Glu.


effects, toxicity, immunogenicity, or other detriments such as the cellular vacuolation that seems to be caused by PEGylated proteins. As can be seen, exploring the potentials of conjugative PTMs can bring many benefits, which may result in the development of new generations of biopharmaceuticals.

[18]

CONSENT FOR PUBLICATION

[21]

[19]

[20]

Not applicable. CONFLICT OF INTEREST The authors declare no conflict of interest, financial or otherwise.

[22]

ACKNOWLEDGEMENTS

[23]

Declared none. REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9]

[10] [11] [12] [13] [14] [15] [16]

[17]

Walsh, G. Post-translational modifications of protein biopharmaceuticals. Drug Discov. Today, 2010, 15(17-18), 773-780. Knorre, D.G.; Kudryashova, N.V.; Godovikova, T.S. Chemical and functional aspects of posttranslational modification of proteins. Acta Naturae, 2009, 1(3), 29-51. Černý, M.; Skalák, J.; Cerna, H.; Brzobohatý, B. Advances in purification and separation of posttranslationally modified proteins. J. Proteomics, 2013, 92, 2-27. Leurs, U.; Mistarz, U.H.; Rand, K.D. Getting to the core of protein pharmaceuticals - Comprehensive structure analysis by mass spectrometry. Eur. J. Pharm. Biopharm., 2015, 93, 95-109. Fischer, S.; Handrick, R.; Otte, K. The art of CHO cell engineering: A comprehensive retrospect and future perspectives. Biotechnol. Adv., 2015, 33(8), 1878-1896. Zhong, X.; Wright, J.F. Biological insights into therapeutic protein modifications throughout trafficking and their biopharmaceutical applications. Int. J. Cell Biol., 2013, 2013, 273086. Strohl, W.R. Fusion proteins for half-life extension of biologics as a strategy to make biobetters. BioDrugs, 2015, 29, 215-239. Yu, K.; Liu, C.; Kim, B.; Lee, D. Synthetic fusion protein design and applications. Biotechnol. Adv., 2015, 33, 155-164. Biotechnology Information Institute. BIOPHARMA: Biopharmaceutical Products in the U.S. and European Markets. http://www.biopharma.com/approvals.html (Accessed January, 2017). U.S. Department of Health & Human Services. Food and Drug Administration. http://www.accessdata.fda.gov/scripts/cder/drugsatfda/ index.cfm (Accessed January, 2017). Planinc, A.; Bones, J.; Dejaegher, B.; Van Antwerpen, P.; Delporte, C. Glycan characterization of biopharmaceuticals: Updates and perspectives. Anal. Chim. Acta, 2016, 921, 13-27. Furmanek, A.; Hofsteenge, J. Protein C-mannosylation: Facts and questions. Acta. Biochim. Pol., 2000, 47(3), 781-789. Lehninger, A.L.; Nelson, D.L.; Cox, M.M. Lehninger Principles of Biochemistry, 6th ed.; Worth Publishers: New York, 2014. Butler, M. Optimisation of the cellular metabolism of glycosylation for recombinant proteins produced by mammalian cell systems. Cytotechnology, 2006, 50(1-3), 57-76. Hossler, P.; Khattak, S.F.; Li, Z.J. Optimal and consistent protein glycosylation in mammalian cell culture. Glycobiology, 2009, 19(9), 936-949. Hopkins, D.; Gomathinayagam, S.; Hamilton, S.R. A practical approach for O-linked mannose removal: The use of recombinant lysosomal mannosidase. Appl. Microbiol. Biotechnol., 2015, 99(9), 3913-3927. Li, H.; ad’Anjou, M. Pharmacological significance of glycosylation in therapeutic proteins. Curr. Opin. Biotechnol., 2009, 20(6), 678684.

[24] [25] [26] [27] [28]

[29]

[30] [31] [32]

[33]

[34]

[35]

[36]

[37]


11

Yu, X.; Marshall, M.J.E.; Cragg, M.S.; Crispin, M. Improving antibody-based cancer therapeutics through glycan engineering. BioDrugs, 2017, 31(3), 1-16. Murakami, M.; Kiuchi, T.; Nishihara, M.; Tezuka, K.; Okamoto, R.; Izumi, M.; Kajihara, Y. Chemical synthesis of erythropoietin glycoforms for insights into the relationship between glycosylation pattern and bioactivity. Sci. Adv., 2016, 2(1), e1500678. Elstein, D. Recent advances in treatment approaches to Gaucher disease. Curr. Pharm. Biotechnol., 2011, 12(6), 854-860. Hughes, D.A.; Gonzalez, D.E.; Lukina, E.A.; Mehta, A.; Kabra, M.; Elstein, D.; Kisinovsky, I.; Giraldo, P.; Bavdekar, A.; Hangartner, T.N.; Wang, N.; Crombez, E.; Zimran, A. Velaglucerase alfa (VPRIV) enzyme replacement therapy in patients with Gaucher disease: Long-term data from phase III clinical trials. Am. J. Hematol., 2015, 90(7), 584-591. Ye, H.; Hill, J.; Gucinski, A.C.; Boyne, M.T.; Buhse, L.F. Direct site-specific glycoform identification and quantitative comparison of glycoprotein therapeutics: Imiglucerase and velaglucerase alfa. AAPS J., 2015, 17(2), 405-415. Tekoah, Y.; Tzaban, S.; Kizhner, T.; Hainrichson, M.; Gantman, A.; Golembo, M.; Aviezer, D.; Shaaltiel, Y. Glycosylation and functionality of recombinant β-glucocerebrosidase from various production systems. Biosci. Rep., 2013, 33(5), 771-781. Kolate, A.; Baradia, D.; Patil, S.; Vhora, I.; Kore, G.; Misra, A. PEG - A versatile conjugating ligand for drugs and drug delivery systems. J. Control Release, 2014, 192, 67-81. Qi, Y.; Chilkoti, A. Protein-polymer conjugation-moving beyond PEGylation. Curr. Opin. Chem. Biol., 2015, 28, 181-193. Roberts, M.J.; Bentley, M.D.; Harris, J.M. Chemistry for peptide and protein PEGylation. Adv. Drug Deliv. Rev., 2002, 54(4), 459476. Dozier, J.K.; Distefano, M.D. Site-specific PEGylation of therapeutic proteins. Int. J. Mol. Sci., 2015, 16(10), 25831-25864. Gao, M.; Tong, Y.; Gao, X.; Yao, W. Development of a C-terminal site-specific PEGylated analog of GLP-1 with improved antidiabetic effects in diabetic mice. Drug Dev. Res., 2013, 74, 186193. Cong, Y.; Pawlisz, E.; Bryant, P.; Balan, S.; Laurine, E.; Tommasi, R.; Singh, R.; Dubey, S.; Peciak, K.; Bird, M.; Sivasankar, A.; Swierkosz, J.; Muroni, M.; Heidelberger, S.; Farys, M.; Khayrzad, F.; Edwards, J.; Badescu, G.; Hodgson, I.; Heise, C.; Somavarapu, S.; Liddell, J.; Powell, K.; Zloh, M.; Choi, J.W.; Godwin, A.; Brocchini, S. Site-specific PEGylation at histidine tags. Bioconjug. Chem., 2012, 23(2), 248-263. Reichert, C.; Borchard, G. Noncovalent PEGylation, an innovative subchapter in the field of protein modification. J. Pharm. Sci., 2016, 105(2), 386-390. Da Silva, F.D.; Mero, A.; Pasut, G. Chemical and enzymatic site specific PEGylation of hGh. Bioconjug. Chem., 2013, 24(3), 456463. Baumann, A.; Tuerck, D.; Prabhu, S.; Dickmann, L.; Sims, J. Pharmacokinetics, metabolism and distribution of PEGs and PEGylated proteins: Quo vadis? Drug Discov. Today, 2014, 19(10), 1623-1631. Verhoef, J.J.; Carpenter, J.F.; Anchordoquy, T.J.; Schellekens, H. Potential induction of anti-PEG antibodies and complement activation toward PEGylated therapeutics. Drug Discov. Today, 2014, 19(12), 1945-1952. Pepinsky, R.B.; LePage, D.J.; Gill, A.; Chakraborty, A.; Vaidyanathan, S.; Green, M.; Baker, D.P.; Whalley, E.; Hochman, P.S.; Martin, P. Improved pharmacokinetic properties of a polyethylene glycol-modified form of interferon-beta-1a with preserved in vitro bioactivity. J. Pharmacol. Exp. Ther., 2001, 297(3), 1059-1066. Calabresi, P.A.; Kieseier, B.C.; Arnold, D.L.; Balcer, L.J.; Boyko, A.; Pelletier, J.; Liu, S.; Zhu, Y.; Seddighzadeh, A.; Hung, S.; Deykin, A. Pegylated interferon beta-1a for relapsing-remitting multiple sclerosis (ADVANCE): A randomised, phase 3, doubleblind study. Lancet Neurol., 2014, 13(7), 657-665. Hu, X.; Cui, Y.; White, J.; Zhu, Y.; Deykin, A.; Nestorov, I.; Hung, S. Pharmacokinetics and pharmacodynamics of peginterferon beta1a in patients with relapsing-remitting multiple sclerosis in the randomized ADVANCE study. Br. J. Clin. Pharmacol., 2015, 79(3), 514-522. Hu, X.; Olivier, K.; Polack, E.; Crossman, M.; Zokowski, K.; Gronke, R.S.; Parker, S.; Li, Z.; Nestorov, I.; Baker, D.P.; Clarke,


[38] [39] [40]

[41] [42] [43]

[44] [45] [46]

J.; Subramanyam, M. In vivo pharmacology and toxicology evaluation of polyethylene glycol-conjugated interferon {beta}-1a. J. Pharmacol. Exp. Ther., 2011, 338(3), 984-996. Resh, M.D. Fatty acylation of proteins: The long and the short of it. Prog. Lipid Res., 2016, 63, 120-131. Resh, M.D. Covalent lipid modifications of proteins. Curr. Biol., 2013, 23(10), R431-R435. Mitchell, D.A.; Mitchell, G.; Ling, Y.; Budde, C.; Deschenes, R.J. Mutational analysis of saccharomyces cerevisiae Erf2 reveals a two-step reaction mechanism for protein palmitoylation by DHHC enzymes. J. Biol. Chem., 2010, 285(49), 38104-38114. Buglino, J.A.; Resh, M.D. Hhat is a palmitoyl acyltransferase with specificity for N-palmitoylation of Sonic Hedgehog. J. Biol. Chem., 2008, 283(32), 22076-22088. Nile, A.H.; Hannoush, R.N. Fatty acylation of Wnt proteins. Nat. Chem. Biol., 2016, 12(2), 60-69. Darling, J.E.; Zhao, F.; Loftus, R.J.; Patton, L.M.; Gibbs, R.A.; Hougland, J.L. Structure-activity analysis of human ghrelin Oacyltransferase reveals chemical determinants of ghrelin selectivity and acyl group recognition. Biochemistry, 2015, 54(4), 1100-1110. Thinon, E.; Hang, H.C. Chemical reporters for exploring protein acylation. Biochem. Soc. Trans., 2015, 43(2), 253-261. Shipston, M.J. Ion channel regulation by protein S-acylation. J. Gen. Physiol., 2014, 143(6), 659-678. Jiang, H.; Zhang, X.; Lin, H. Lysine Fatty acylation promotes lysosomal targeting of TNF-α. Sci. Rep., 2016, 6, 24371.

Carneiro and Lima [47]

[48] [49] [50] [51]

[52]

[53]

Chesarino, N.M.; Hach, J.C.; Chen, J.L.; Zaro, B.W.; Rajaram, M.V.; Turner, J.; Schlesinger, L.S.; Pratt, M.R.; Hang, H.C.; Yount, J.S. Chemoproteomics reveals toll-like receptor fatty acylation. BMC Biol., 2014, 12, 91. Berenson, D.F.; Weiss, A.R.; Wan, Z.L.; Weiss, M.A. Insulin analogs for the treatment of diabetes mellitus: Therapeutic applications of protein engineering. Ann. N. Y. Acad. Sci., 2011, 1243, E40-E54. Berard, L.; MacNeill, G. Insulin degludec, a long-acting once-daily basal analogue for type 1 and type 2 diabetes mellitus. Can. J. Diabetes, 2015, 39(1), 4-9. Atkin, S.; Javed, Z.; Fulcher, G. Insulin degludec and insulin aspart: Novel insulins for the management of diabetes mellitus. Ther. Adv. Chronic Dis., 2015, 6(6), 375-388. Kobuke, K.; Yoneda, M.; Nakanishi, S.; Ohno, H.; Maeda, S.; Egusa, G. Efficacy and safety of insulin degludec in Japanese patients with type 1 and type 2 diabetes: 24-week results from the observational study in routine clinical practice. J. Diabetes Investig., 2016, 7(1), 94-99. Kaku, K.; Eid, M.A. Safety, efficacy, and early clinical experience of insulin degludec in Japanese people with diabetes mellitus: A first-year report from Japan. J. Diabetes Investig., 2015, 6(6), 610619. Dardano, A.; Bianchi, C.; Del Prato, S.; Miccoli, R. Insulin degludec/insulin aspart combination for the treatment of type 1 and type 2 diabetes. Vasc. Health Risk Manag., 2014, 10, 465-475.

DISCLAIMER: The above article has been published in Epub (ahead of print) on the basis of the materials provided by the author. The Editorial Department reserves the right to make minor modifications for further improvement of the manuscript.