Controlling Deamidation Rates in a Model Peptide:
`Effects of Temperature, Peptide Concentration, and Additives
`
`LEWIS P. STRATTON,1,3 R. MICHAEL KELLY,1 JARED ROWE,1 JESSE E. SHIVELY,1 D. DAVID SMITH,2
`JOHN F. CARPENTER,1 MARK C. MANNING1
`
`1Center for Pharmaceutical Biotechnology and Department of Pharmaceutical Sciences, School of Pharmacy,
`Campus Box C238, University of Colorado Health Sciences Center, Denver, Colorado 80262
`
`2Department of Biomedical Sciences, School of Medicine, Creighton University, Omaha, Nebraska
`
`3Department of Biology, Furman University, 3300 Poinsett Highway, Greenville South Carolina 29613-0418
`
`Received 27 November 2000; revised 2 July 2001; accepted 10 July 2001
`
`ABSTRACT: The rate of deamidation of the Asn residue in Val-Tyr-Pro-Asn-Gly-Ala
`(VYPNGA), a model peptide, was determined at pH 9 (400 mM Tris buffer) as a function
`of temperature and peptide concentration. Over the temperature range 5–658C,
`deamidation followed Arrhenius behavior, with an apparent activation energy of
`13.3 kcal/mol. Furthermore, increasing the peptide concentration slows the rate of
`deamidation. Self-stabilization with respect to deamidation has not been reported
`previously. The rate of deamidation was also determined in the presence of sucrose and
`poloxamer 407 (Pluronic F127). In both cases, the rate of deamidation was retarded by
`up to 40% at 358C. In aqueous solutions containing poloxamer 407, the degree of
`stabilization is independent of formation of a reversible thermosetting gel. With sucrose,
`maximum reduction in the deamidation rate was attained with as little as 5% (w/v).
`Addition of sucrose results in a greater conformational preference for a type II b-turn
`structure, which presumably is less prone to intramolecular cyclization and subsequent
`deamidation. (cid:223) 2001 Wiley-Liss, Inc. and the American Pharmaceutical Association J Pharm Sci
`90:2141–2148, 2001
`Keywords: deamidation; poloxamer; gel; peptide stability; Arrhenius plot
`
`INTRODUCTION
`
`Deamidation of asparagine (Asn) residues is
`probably the most common pathway for chemical
`inactivation of protein pharmaceuticals1,2 The
`reaction rate for deamidation in aqueous solution
`is dependent on a number of extrinsic factors,
`such as pH,3–6 solvent dielectric,7 buffer concen-
`tration,5 and temperature,5 as well as intrinsic
`factors, such as the primary sequence8–10 and the
`presence of secondary2,11,12 and tertiary struc-
`ture.13 Despite numerous studies, little has been
`reported on the ability of additives to affect the
`
`Correspondence to: M.C. Manning (Telephone: 303-315-
`6162; Fax: 303-315-6281; E-mail: mark.manjing@uchsc.edu)
`
`Journal of Pharmaceutical Sciences, Vol. 90, 2141–2148 (2001)
`(cid:223) 2001 Wiley-Liss, Inc. and the American Pharmaceutical Association
`
`rate of deamidation in peptides and proteins. In
`the case of deamidation of human epidermal
`growth factor (hEGF), a wide variety of excipients
`were reported to have an effect, but with no
`explanation as to why some were effective and
`others were not.14
`It is now well established that the conformation
`of the peptide backbone (i.e., the secondary struc-
`ture), as well as the side chain dihedral angles,
`can affect the rate of deamidation.2,3 Therefore,
`one approach for slowing deamidation might be to
`control peptide conformation. One way this con-
`trol might be accomplished is through the use of
`preferentially excluded solutes. These solutes are
`additives that are selectively excluded from the
`hydration sphere of the protein. In other words,
`there will be a relatively higher concentration of
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 12, DECEMBER 2001
`
`2141
`
`Novo Nordisk Ex. 2044, P. 1
`Mylan Institutional v. Novo Nordisk
`IPR2020-00324
`
`

`

`2142
`
`STRATTON ET AL.
`
`the excluded solute in the bulk relative to near the
`surface of the protein. The mechanism by which
`these solutes stabilize proteins has been described
`in detail by Timasheff and co-workers.15–17
`It has also been demonstrated that preferen-
`tially excluded solutes can lead to compaction of
`the native structure of a globular protein.15,16 The
`same additives may be able to alter the conforma-
`tional distribution of a flexible peptide in aqueous
`solution.15 In short, one might be able to use
`any number of excluded solutes (e.g., polymers,
`sugars, salts, or amino acids) to affect the solution
`conformation of the peptide, thereby having an
`effect on the deamidation rate. If this effect can be
`demonstrated, it would provide a new strategy for
`stabilizing hydrolytically sensitive peptides in
`aqueous formulations. It is important to note that
`preferentially excluded solutes are maximally
`effective at relatively high concentrations of
`solute (as much as 1 M).
`Herein we describe the deamidation of a
`previously studied model peptide, Val-Tyr-Pro-
`Asn-Gly-Ala (VYPNGA),4–7,10 as a function of
`temperature, peptide concentration, and the addi-
`tion of solutes. We chose to examine two different
`excipients, sucrose and poloxamer 407, both of
`which are known to act as preferentially excluded
`solutes. Sucrose is well known to alter the confor-
`mation and flexibility of proteins through a prefe-
`rential exclusion mechanism.17 As has been
`shown for interleukin-1 receptor antagonist, addi-
`tion of sucrose produces a more compact protein
`structure,15,17 which is less prone to aggregation
`as well as deamidation.18 Similar effects have
`been observed for sucrose interacting with inter-
`feron-g.16 However, the effect of excluded solutes
`on the solution conformation of a peptide has not
`been assessed.
`The possible effects of poloxamer 407 (Pluronic
`F127), a triblock copolymer of poly(oxyethylene)
`and poly(oxypropylene), on deamidation of a
`peptide is less clear. Within a certain concentra-
`tion range, aqueous solutions of poloxamer 407
`will undergo a cooperative phase transition from a
`sol state to a gel with an increase in temperature.
`This transitition property has made it an attrac-
`tive vehicle for controlled drug delivery.19,20 It
`was hypothesized that poloxamer 407 might
`retard deamidation as well because it has been
`shown to salt out native proteins presumably due
`to preferential exclusion,19 but that the stabiliza-
`tion may also be related to the sol–gel transition
`rather than solely due to poloxamer 407 effects on
`peptide conformation.
`
`MATERIALS AND METHODS
`
`Materials
`
`Poloxamer 407 (Pluronic F127) was obtained from
`BASF and used as received. Sucrose was pur-
`chased from Pfannstiehl. All solvents were ob-
`tained from Aldrich.
`
`Peptide Synthesis
`
`Amino acid derivatives and resin were purchased
`from Applied Biosystems Incorporated (Foster
`City, CA). All other solvents and reagents were
`purchased from Fisher Scientific (Pittsburgh,
`PA). The peptide was assembled on Boc-Ala-
`PAM resin (0.5 meq) using an ABI 430A auto-
`mated peptide synthesizer. Boc groups were
`removed with 33% trifluoroacetic acid (TFA) in
`dichloromethane. Subsequent Boc amino acid
`derivatives were coupled to the resin in fourfold
`excess using diisopropylcarbodiimide and hydro-
`xybenzotriazole. Coupling reactions were moni-
`tored by the quantitative ninhydrin test.1 All
`yields were > 99% after a single coupling. Once
`the desired sequence was assembled, the peptide
`was simultaneously deprotected and cleaved from
`the resin using a mixture of TFA, trifluorometha-
`nesulphonic acid, ethanedithiol and thioanisole
`(80/8/4/8, v/v/v/v). The crude peptide product was
`loaded onto a Vydac 218TP101550 preparative
`C18 column (250 (cid:2) 50 mm, 300, 10–15 m) from
`The Separations Group (Hesperia, CA) that was
`previously equilibrated with triethylammonium
`phosphate buffer (100 mM, pH 2.5). The concen-
`tration of acetonitrile in the eluent was raised to
`11% over a period of 50 min. The eluent was
`continuously monitored at 230 nm and collected
`in 50 mL fractions. Fractions containing only the
`peptide were pooled, diluted twofold with water
`and desalted on the same column now equili-
`brated with 0.1% TFA. The concentration of
`acetonitrile was increased to 30% over a period
`of 30 min. Fractions containing the product were
`pooled and lyophilized to yield 150 mg of
`VYPNGA as a fluffy white powder (48% overall
`yield based on the initial resin substitution). FAB-
`MS [M (cid:135) H](cid:135) found 620.4, theoretical 620.3.
`
`HPLC Protocol
`
`The degradation of VYPNGA was monitored
`using a LKB Bromma high-performance liquid
`chromatography (HPLC) system with a Applied
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 12, DECEMBER 2001
`
`Novo Nordisk Ex. 2044, P. 2
`Mylan Institutional v. Novo Nordisk
`IPR2020-00324
`
`

`

`FACTORS AFFECTING DEAMIDATION OF A MODEL PEPTIDE
`
`2143
`
`Biosystems variable wavelength detector set at
`214 nm. Separation was achieved using a Vydac
`reverse phase C-18 column (4.6 (cid:2) 25 mm length, 5
`mm bead diameter and 80 A˚ pore size). The mobile
`phase was a 24/76 mixture of water and 20%
`acetonitrile in 0.1% aqueous TFA. The flow rate
`was 1.5 mL/min. This protocol is similar to those
`reported previously for deamidation studies of
`VYPNGA.5,10
`
`Deamidation Studies
`
`Samples were prepared by dissolving a weighed
`amount of VYPNGA in 1 mL of 400 mM aqueous
`Tris buffer (pH 9), with or without poloxamer 407
`present. The sample was placed in a water bath to
`control the temperature. Periodically, samples
`were removed to assay for the extent of deamida-
`tion using the HPLC protocol already described.
`
`Circular Dichroism Studies
`
`Samples were prepared by dissolving 1 mg of
`VYPNGA in 1 mL of 400 mM Tris buffer contain-
`ing various amounts of sucrose or poloxamer 407.
`
`Samples were then placed in a 0.1 mm quartz
`cuvette. Far ultraviolet (UV) spectra (l & 250–
`180 nm) were collected using an AVIV 62-DS
`circular dichroism (CD) spectrometer with a ther-
`moelectric temperature control unit regulated to
`(cid:6)0.18C. Data were taken every 0.25 nm, using an
`averaging time of 4 s and a bandwidth of 1.5 nm.
`
`RESULTS AND DISCUSSION
`
`Deamidation of a polypeptide proceeds through a
`cyclic imide intermediate formed by intramolecu-
`lar attack of the succeeding peptide nitrogen on
`the carbonyl of the Asn side chain above pH 63,4
`(see Figure 1). This intermediate can hydrolyze,
`via nucleophilic attack of water, at either of the
`carbonyl groups; one attack will form a product
`containing normal aspartic acid (Asp) in place
`of the Asn residue, and one leads to a rear-
`ranged form of aspartic acid referred to as iso-
`aspartic acid (isoAsp). As a model of a peptide
`prone to deamidation, we chose the model hex-
`apeptide, VYPNGA, which has been used widely
`in the past.4–7,10 This peptide represents the
`
`Figure 1. General mechanism of deamidation of Asn residues under basic conditions.
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 12, DECEMBER 2001
`
`Novo Nordisk Ex. 2044, P. 3
`Mylan Institutional v. Novo Nordisk
`IPR2020-00324
`
`

`

`2144
`
`STRATTON ET AL.
`
`deamidation site in the peptide hormone, ACTH.
`The chromatographic properties of this peptide
`and its degradation products are well estab-
`lished,4–7,10 making determination of the reaction
`kinetics straightforward. To have the reaction
`proceed quickly, we used conditions known to
`accelerate deamidation; namely, high pH and
`high buffer concentration. For this study, 400 mM
`Tris buffer at pH 9 was employed throughout.
`Under these conditions, a 1 mg/mL sample of
`VYPNGA will deamidate almost completely at 45
`8C within 6 h.
`The first studies were conducted to determine
`whether the mechanism of deamidation was
`altered by the use of high Tris concentrations at
`pH 9, conditions that have not been reported
`previously for this peptide. Therefore, the tem-
`perature dependence of the reaction was followed
`over a range spanning 5 to 658C. It should be
`noted that there is the possibility that Tris will
`decompose at elevated temperatures,21 producing
`reactive species, such as formaldehyde. However,
`there was no chromatographic evidence for the
`formation of species other than the deamidated
`forms of VYPNGA.
`Within experimental error, the reaction follows
`Arrhenius behavior (Figure 2). The activation
`energy was 13.3 kcal/mol, with an A-value of
`1014.9 (r2 (cid:136) 0.98), which is significantly less than
`the (cid:24)20 kcal/mol reported for this reaction.4,5
`However,
`those activation energies were for
`reactions extrapolated to zero buffer concentra-
`tion. At high buffer concentrations, buffer cata-
`lysis becomes significant, and the activation
`energy could be affected. The ratio of isoAsp to
`Asp was similar to that reported in the literature
`for VYPNGA,4,5,10 averaging (cid:24)3.0–3.5 to 1.0 at all
`temperatures tested, as determined by HPLC.
`
`Concentration Effects
`
`One aspect of peptide deamidation that has not
`been reported is whether peptide concentration
`has any effect. Presumably, increased concentra-
`tion could lead to aggregates that might be more
`stable or increased concentration may favor a
`solution conformation that is more or less reac-
`tive, depending on the relative spatial arrange-
`ment of the peptide backbone and Asn side chain.8
`On increasing the peptide concentration from 0.1
`to 100 mg/mL, an approximate threefold decrease
`in reaction rate was observed (Figure 3). A
`number of possible explanation exist for such an
`observation. First, it is possible that the peptide is
`interacting with itself, assembling to form a
`structure that is less reactive than the free
`monomer in solution. Second, small changes in
`pH (< 1 pH unit) could cause a modest decrease in
`rate, and high concentrations of the peptide may
`be able to shift the pH to this degree.22 However,
`no significant change in overall pH was measured
`for the concentrated solutions. Third, a peptide
`concentration of 100 mg/mL will cause increases
`in the viscosity of the solution. It has been
`reported that viscosity increase of <5 cP can slow
`the reaction to this extent.23 Whereas this most
`plausible explanation may be self-assembly, these
`other mechanisms cannot be excluded.
`
`Additive Effects-Sucrose
`
`Sucrose is well known to alter both the conforma-
`tion and flexibility of proteins through a prefer-
`ential exclusion mechanism.15–17 Briefly, an
`exclusion of solute, such as sucrose, means that
`the solute would rather exist in the bulk solution
`rather than at the surface of the protein. Such
`
`Figure 2. Arrhenius
`deamidation
`for
`plot
`VYPNGA (1 mg/mL) at pH 9 (400 mM Tris buffer).
`
`of
`
`Figure 3. Concentration effects on the rate of dea-
`midation of VYPNGA (pH 9, 400 mM Tris buffer, 358C).
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 12, DECEMBER 2001
`
`Novo Nordisk Ex. 2044, P. 4
`Mylan Institutional v. Novo Nordisk
`IPR2020-00324
`
`

`

`FACTORS AFFECTING DEAMIDATION OF A MODEL PEPTIDE
`
`2145
`
`‘negative’ binding is thermodynamically destabi-
`lizing. The same exclusion exists for the unfolded
`state, as well as the native state. Because the
`unfolded state is typically larger in surface area
`than the folded state, the effect is larger in
`magnitude, meaning that
`the denatured or
`unfolded state is destabilized more than the
`native state. Together, these effects result in
`an increase in the free energy required to go
`from the folded to unfolded state; that is, there is a
`net stabilization. Moreover, for most proteins,
`excluded solutes also induce the protein to adopt a
`more compact native state.15,16 For interleukin-1
`receptor antagonist, the more compact protein
`structure is less prone to aggregation15 as well
`as deamidation.18 However, the effects of sucrose
`on peptide stability have not been well charac-
`terized. On addition of increasing amounts of
`sucrose,
`the reaction rate for deamidation
`decreased, even with as little as 2% sucrose
`present. The maximum effect could be achieved
`with as little as 5% sucrose (Figure 4). With 5%
`sucrose present, the relative deamidation rate
`was 77%. With 10% sucrose present, the relative
`rate was 78% of comparable solutions with no
`sucrose present.
`At these concentrations, lower rates due solely
`to increased viscosity is unlikely, especially
`because the rate-limiting step is intramolecular
`cyclization.6 At a sucrose concentration of 2%,
`there should be little change in the viscosity of an
`aqueous solution (< 5 cP). Although the spectro-
`scopic evidence given later suggests that rate
`retardation is due to a conformational change, the
`effects of the increased viscosity, although mod-
`est, cannot be excluded as having some modulat-
`
`ing effect on deamidation rates, especially in light
`of the work of Li et al.23
`As just hypothesized, retardation of deamida-
`tion rates by an excluded solute suggests that the
`peptide has changed conformation, making intra-
`molecular cyclization to the succinimide inter-
`mediate more difficult. The effect of secondary
`structure on deamidation has been attributed to
`this type of effect, where the peptide and the
`asparagine side chain are in less favorable
`positions for intramolecular attack when in an
`ordered secondary structure.2,3,8 In this case, the
`deamidation site is not within a fixed secondary
`structure, but the distribution within the native
`state ensemble is being shifted.
`CD spectra were obtained for VYPNGA both in
`the absence and presence of sucrose at pH 9. The
`spectra were taken at 258C and within 10 min of
`sample preparation to minimize the extent of
`deamidation. A difference spectrum for a sample
`in 5% sucrose versus no sucrose reveals that a
`certain secondary structure is induced by the
`addition of sucrose (Figure 5). The induced
`structure is easier to detect using difference
`spectroscopy because the original CD spectra are
`dominated by contributions from the Tyr side
`chain. The difference CD spectrum shown a
`negative band near 225 nm and a positive band
`indicative of a type II b-turn
`near 200 nm,
`structure,24,25 presumably with Pro at position
`i (cid:135) 1. This structure would place the Asn residue
`at the corner of a type II turn. It is known that
`there is a subset of conformations that favors
`deamidation by positioning the side chain for
`attack by the succeeding nitrogen atom in the
`polypeptide chain.3 Such a structure would be
`
`Figure 4. Effect of added sucrose on the deamidation
`rate of VYPNGA (1 mg/mL) at pH 9, 400 mM Tris
`buffer) at 358C.
`
`Figure 5. Difference CD spectrum of VYPNGA in the
`(A) presence and (B) absence of 5% (w/v) sucrose (pH 9,
`400 mM Tris buffer, 258C).
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 12, DECEMBER 2001
`
`Novo Nordisk Ex. 2044, P. 5
`Mylan Institutional v. Novo Nordisk
`IPR2020-00324
`
`

`

`2146
`
`STRATTON ET AL.
`
`predicted to slow deamidation for two reasons.
`First, the orientation of the carbonyl group of Asn
`relative to the succeeding amide nitrogen of the
`Gly residue is not favorable for nucleophilic
`attack. Second, the amide NH is presumably
`involved in stabilizing the b-turn, making its
`deprotonation more difficult. The deprotonation
`must occur to generate a species sufficiently
`nucleophilic to form the cyclic imide intermediate.
`Other researchers have observed that residues
`located in b-turns do deamidate more slowly than
`residues in unordered or random coil struc-
`ture.2,26,27 Analysis of the sequence preferences
`for certain amino acids in various positions of
`b-turns indicates that the combination of Pro-Asn
`is highly favored for occupying positions i (cid:135) 1 and
`i (cid:135) 2 in a type II b-turn.28,29
`Although these data do not constitute definitive
`proof for rate retardation arising from an altered
`solution conformation of a peptide, the magnitude
`of the rate changes and the substantial increase in
`the b-turn population present a compelling obser-
`vation. The ability of additives to modulate
`peptide conformation and thereby affect chemical
`degradation rates appears to be one possible
`formulation strategy. Recently, a similar finding
`was observed for active site oxidation retardation
`by the addition of sucrose.30
`
`Additive Effects- Poloxamer 407
`
`Few data have been reported on the ability of
`excipients to slow chemical degradation in pro-
`teins, especially for deamidation. Son and Kwon
`reported that numerous polymers were effective
`at retarding deamidation in hEGF.14 However, no
`mechanistic explanation was given and the
`additives were very different in chemical and
`physical properties (various polymers, surfac-
`tants, etc.), making generalization difficult.
`Poloxamer 407, when dissolved in water at
`high concentrations, has the ability to form either
`a fluid solution (at low temperatures) or a stiff
`gel (at higher temperatures). It was of interest
`to see if formation of an aqueous gel would
`retard deamidation in a peptide, particularly as
`a function of physical state. Relatively little is
`known about the detailed interaction of polox-
`amers and proteins. Recently, we demonstrated
`that poloxamer 407 is a preferentially excluded
`solute that is able to salt-out proteins when
`dissolved in aqueous buffer at high concentra-
`tions.19 Therefore, as an excluded solute, it should
`be possible for poloxamer 407 to affect the solution
`
`conformation of VYPNGA in the same fashion as
`sucrose, and thereby alter the rate of cyclic imide-
`mediated deamidation.
`At concentrations > 17% (w/w), solutions of
`poloxamer 407 can form stiff gels as the tempera-
`ture is increased. The process is highly coopera-
`tive, occurring within a narrow temperature
`range. This process can be monitored by observing
`the physical state of the solution or by spectro-
`scopic methods, such as Fourier transform infra-
`red spectroscopy.19 Using a 22% gel solution,
`which gels at (cid:24)15 8C,
`the deamidation of
`VYPNGA was investigated under conditions
`analogous to those in solution (pH 9, 400 mM
`Tris). Whether the poloxamer solution was in the
`gel or sol state, there was a decrease in the rate of
`deamidation by > 40% versus the peptide in
`aqueous solution. Above the gel transition tem-
`perature (358C),
`the deamidation rate was
`58.6 (cid:6) 2.9% (n (cid:136) 3) relative to solutions without
`poloxamer present. Below the gel
`transition
`temperature (58C), the relative deamidation rate
`was 45 (cid:6) 3% (n (cid:136) 2).
`In the presence of 22% poloxamer, the tem-
`perature dependence of the reaction was investi-
`gated for temperatures above the transition
`temperature of the gel, meaning all measure-
`ments were made on poloxamer solutions existing
`in the gel state. Under those conditions, the
`system displays Arrhenius behavior, and the acti-
`vation energy was determined to be 12.1 kcal/mol,
`with an A-value of 1014.2 (Figure 6, r2 (cid:136) 0.92),
`similar to that determined for VYPNGA in buffer
`alone. Considering the variance in the data, the
`
`Figure 6. Arrhenius
`of
`deamidation
`for
`plot
`VYPNGA (1 mg/mL) at pH 9 (400 mM Tris buffer) in
`a 22 % (w/w) aqueous solution of poloxamer 407.
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 12, DECEMBER 2001
`
`Novo Nordisk Ex. 2044, P. 6
`Mylan Institutional v. Novo Nordisk
`IPR2020-00324
`
`

`

`FACTORS AFFECTING DEAMIDATION OF A MODEL PEPTIDE
`
`2147
`
`activation energies are probably not significantly
`different.
`Given that sucrose appears to slow deamida-
`tion by altering the conformational preference of
`the peptide, similar CD studies were conducted
`looking for evidence of a structural change in the
`presence of poloxamer 407. Difference spectra
`showed a similar far UV CD pattern as with the
`addition of sucrose, but much weaker. Given that
`the solute concentration was much greater (22
`versus 5%), it appears that although poloxamer
`can act as an excluded solute, it is much less
`effective than sucrose on a weight-to-weight basis.
`
`SUMMARY
`
`These results demonstrate that the addition of
`excluded solutes can alter the conformational
`preference of a flexible peptide while also altering
`its propensity to deamidate. The effect is analo-
`gous to the role of secondary structure on
`modulating deamidation by constraining back-
`bone dihedral angles. In the case of VYPNGA, the
`addition of sucrose drives the conformational
`ensemble toward a greater preference for a type-
`II b-turn structure, most likely around the Pro-
`Asn residues.28,29
`Furthermore, it is becoming clear that polox-
`amers, much like PEGs and other polymers, act as
`excluded solutes at high concentrations (as
`demonstrated herein and previously19). Possibly,
`the mechanism of stabilization is the same as with
`sucrose, although viscosity effects and small
`changes in pH cannot be ruled out. Certainly,
`one must consider the possibility for retardation
`of deamidation in such concentrated solutions is
`that it is simply a viscosity effect, even for
`poloxamer in the sol state.23 However, the results
`with sucrose suggest that the effect is conforma-
`tional, not a function of microviscosity, especially
`because the rate-limiting step is an intramolecu-
`lar event. Furthermore, the fact that the extent of
`stabilization of poloxamer solutions was similar in
`both the sol and gel states, where the viscosity is
`dramatically different, also argues against visc-
`osity playing a major role in slowing deamidation
`in flexible peptides. Together, these findings
`suggest that control of the conformational state
`of a peptide may not only retard physical pro-
`cesses, such as aggregation, but can also slow
`degradation due to hydrolytic reactions, such as
`deamidation.
`
`REFERENCES
`
`1. Manning MC, Patel K, Borchardt RT. 1998.
`Stability of protein pharmaceuticals. Pharm Res
`6:903–918.
`2. Xie M, Schowen RL. 1999. Secondary structure and
`protein deamidation. J Pharm Sci 88:8–13.
`3. Clarke S. 1987. Propensity for spontaneous succi-
`nimide formation from aspartyl and asparaginyl
`residues in cellular proteins. Int J Peptide Protein
`Res 30:808–829.
`4. Geiger T, Clarke S. 1987. Deamidation, isomeriza-
`tion, and racemization at asparaginyl and aspartyl
`residues in peptides. Succinimide-linked reactions
`that contribute to protein degradation, J Biol Chem
`262:785–794.
`5. Patel K, Borchardt RT. 1990. Chemical pathways of
`peptide degradation. II. Kinetics of deamidation of
`an asparaginyl residue in a model hexapeptide.
`Pharm Res 7:703–711.
`6. Xie M, Morton M, Vander Velde D, Borchardt RT,
`Schowen RL. 1996. pH-induced change in rate-
`determining step for the hydrolysis of the Asn/Asp
`derived cyclic imide intermediate in protein degra-
`dation. J Am Chem Soc 118:8955–8956.
`7. Brennan TV, Clarke S. 1993. Spontaneous degra-
`dation of polypeptides at aspartyl and asparaginyl
`residues: Effects of solvent dielectric., Protein Sci
`2:331–338.
`8. Wright TE. 1991. Sequence and structure determi-
`nants of the nonenzymatic deamidation of aspar-
`agine and glutamine residues in proteins. Protein
`Eng 4:283–294.
`9. Bischoff R, Lepage P, Jaquinod M, Cauet G, Acker-
`Klein M, Clesse D, Laporte M, Bayol A, Van
`Dorsselaer A, Roitsch C. 1993. Sequence specific
`deamidation: isolation and biochemical character-
`ization of succinimide intermediates of recombi-
`nant hirudin. Biochemistry 32:725–734.
`10. Patel K, Borchardt RT. 1990. Chemical pathways of
`peptide degradation.
`III. Effect
`of primary
`sequence on the pathways of deamidation of
`asparaginyl residues in hexapeptides. Pharm Res
`7:787–793.
`11. Lura R, Schirch V. 1988. Role of peptide conforma-
`tion in the rate and mechanism of deamidation of
`asparaginyl residues. Biochemistry 27:7671–7677.
`12. Stevenson CL, Friedman AR, Kubiak TM, Donlan
`ME, Borchardt RT. 1993. Effect of secondary
`structure on the rate of deamidation of several
`growth hormone releasing factor analogs. Int J
`Pept Protein Res 42:497–503.
`13. Kossiakoff AA. 1988. Tertiary structure is a
`principal determinant
`to protein deamidation.
`Science 240:191–194.
`14. Son K, Kwon C. 1995. Stabilization of human
`epidermal growth factor (hEGF) in aqeuous for-
`mulation. Pharm Res 12:451–454.
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 12, DECEMBER 2001
`
`Novo Nordisk Ex. 2044, P. 7
`Mylan Institutional v. Novo Nordisk
`IPR2020-00324
`
`

`

`2148
`
`STRATTON ET AL.
`
`15. Kendrick BS, Chang BS, Arakawa T, Randolph T,
`Manning MC, Carpenter JF. 1997. Preferential
`exclusion of sucrose from recombinant human
`interleukin-1 receptor antagonist: Role of re-
`stricted conformational mobility and compaction
`of native state. Proc Natl Acad Sci USA 94:11917–
`11922.
`16. Kendrick BS, Carpenter JF, Cleland JL, Randolph
`TW. 1998. A transient expansion of the native
`state precedes aggregation of recombinant human
`interferon-g. Proc Natl Acad Sci, USA 95:14142–
`14146.
`17. Lee JC, Timasheff SN. 1981. Stabilization of
`proteins by sucrose. J Biol Chem 256:7193–7201.
`18. Chang BS, Reeder G, Carpenter JF. 1996. Devel-
`opment of a stable freeze-dried formulation of
`recombinant
`interleukin-1 receptor antagonist.
`Pharm Res 13:243–249.
`19. Stratton LP, Dong A, Manning MC, Carpenter JF.
`1997. Drug delivery matrix containing native
`protein precipitates suspended in a poloxamer gel.
`J Pharm Sci 86:1006–1010.
`20. Suh H, Jun HW. 1996. Physicochemical and release
`studies of naproxen in poloxamer gels. Int J Pharm
`129:13–20.
`21. Song Y, Schowen RL, Borchardt RT, Topp EM.
`2001. Effect of ‘pH’ on the rate of asparagine
`deamidation in polymeric formulations: ‘pH’-rate
`profile. J Pharm Sci 90:141–156.
`22. Brennan TV, Clarke S. 1995. Deamidation and
`isoaspartate formation in model synthetic peptides:
`The effects of sequence and solution environment.
`In: Aswad D, editor. Deamidation and isoaspartate
`formation in peptides and proteins. Boca Raton, FL:
`CRC Press.
`
`23. Li R D’Souza AJ, Laird BB, Schowen RL, Borchardt
`RT, Topp EM. 2000. Effects of solution polarity and
`viscosity on peptide deamidation. J Peptide Res
`56:326–334.
`24. Perczel A, Hollo´si M, Foxman BM, Fasman GD.
`1991. Conformational analysis of pseudo cyclic
`hexapeptides based on quantitative circular dichro-
`ism (CD), NOE and X-ray data. The pure CD
`spectra of type I and type II beta-turns. J Am Chem
`Soc 112:9772–9784.
`25. Perczel A, Fasman GD. 1992. Quantitative analysis
`of cyclic b-turn models. Protein Sci 1:378–
`395.
`26. Sharma S, Hammen PK, Leung A, Georges G,
`Hestenberg W, Klevit RE, Waygood EB. 1993.
`Deamidation of HPr, a phosphocarrier protein of
`the phosphoenolpyruvate: Sugar phosphotransfer-
`ase system, involves asparagine 38 (HPr-1) and
`asparagine 12 (HPr-2) in isoaspartyl acid forma-
`tion. J Biol Chem 268:17695–17704.
`27. Brennan TV, Anderson JW, Jia Z, Waygood EB,
`Clarke S. 1993. Repair of spontaneously deami-
`dated HPr phosphocarrier protein catalyzed by the
`L-isoaspartate (D-aspartate) O-methyltransferase.
`J Biol Chem 269:25486–25595.
`28. Sibanda BL, Thornton JM. 1993. Accommodating
`sequence changes in beta-hairpins in proteins. J
`Mol Biol 229:428–447.
`29. Wilmot CM, Thornton JM. 1988. Analysis and
`prediction of the different types of beta-turn in
`proteins. J Mol Biol 203:221–232.
`30. De Paz RA, Barnett CC, Dale DA, Carpenter JF,
`Gaertner AL, Randolph TW. 2000. The excluding
`effects of sucrose on a protein chemical degradation
`pathway: Methionine oxidation in subtilisin. Arch
`Biochem Biophys 384:123–132.
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 90, NO. 12, DECEMBER 2001
`
`Novo Nordisk Ex. 2044, P. 8
`Mylan Institutional v. Novo Nordisk
`IPR2020-00324
`
`