state, water may affect deamidation in glassy formulations
by increasing the polarity of the matrix. Water may also
facilitate deamidation by serving as a medium for proton
transfer during the cyclization process.5,16
Ala) in lyophilized formulations containing PVP and glyc-
erol was investigated. Increases in moisture and glycerol
contents increased the rate of peptide deamidation. This
increase was strongly correlated with Tg at constant water
content and activity, suggesting that increased matrix
mobility facilitates deamidation. In rubbery systems (T >
Tg), deamidation rates appeared to be independent of water
content and activity in formulations with similar Tgs.
However, in glassy formulations with similar Tgs, deami-
dation increased with water content, suggesting a solvent/
medium effect of water on reactivity in this regime. An
increase in water content also affected the degradation
product distribution; less of the cyclic imide intermediate
(Asu) and more of the hydrolytic products, isoAsp and Asp,
were observed as water content increased. Under low
moisture conditions, the water-catalyzed hydrolysis of the
cyclic imide intermediate to produce the isoAsp and Asp is
suppressed. Thus, residual water appears to facilitate
deamidation in these solid PVP formulations both by
enhancing molecular mobility and by solvent/medium
effects, and also participates as a chemical reactant in the
subsequent breakdown of the cyclic imide.
In addition to having an effect on reactant mobility,
matrix mobility may also affect the ability of a solvent to
adequately solvate and stabilize charged reaction centers.
In the glassy state, the solid polymer solvent may not have
sufficient mobility to rearrange itself to “solvate” the
charged transition state during Asu formation. Although
the polymer may be dynamically constrained, Oksanen and
Zografi have shown that water in solid PVP maintains a
high degree of mobility relative to the polymer.27 The more
mobile water may be able to facilitate deamidation by
“solvating” the charged transition state during Asu forma-
tion. In the rubbery state, PVP is more mobile than in the
glassy state. With greater mobility, PVP may be more able
to solvate and stabilize the development of a charged
transition state during deamidation. Thus, matrix mobility
may have an impact on chemical reactions beyond influ-
encing reactant mobility.
Because water appears to have a solvent effect on
deamidation in these solid systems, glycerol may have had
a solvent effect on the reaction rates in addition to a
plasticizing effect. We cannot rule out the possibility that
the increase in deamidation rates with decreasing Tg
(increasing glycerol content) is in part due to a solvent
effect (Figure 5).
Effect of Wa ter on Dea m id a tion P r od u ct Distr ibu -
tion sThree major degradation products were observed in
this study: the cyclic imide hexapeptide (Asu), the isoAsp-
hexapeptide (isoAsp), and the Asp-hexapeptide (Asp). In
these lyophilized PVP-glycerol systems, the Asn-hexa-
peptide appears to deamidate through intramolecular
cyclization to form Asu, which may degrade to produce
isoAsp and Asp. The deamidation product distribution
would then depend on Asu formation (deamidation of Asn)
and breakdown (hydrolysis to form isoAsp and Asp). In
formulations with minimal moisture content (<0.004 g
water/g wet solid), Asu is the dominant degradation
product with little isoAsp or Asp observed at the sampling
time evaluated. As water content increases, the product
distribution shifts toward less Asu and greater fractions
of isoAsp and Asp. These observations are consistent with
the hydrolytic formation of isoAsp and Asp from Asu, as
observed in solution (Scheme 1). In solution at neutral pH,
Asu undergoes spontaneous hydrolysis to form isoAsp and
Asp, where the attack of water or hydroxide ion on the
cyclic imide is the rate-limiting step.16,17,28
Acknowledgments
M.C.L. thanks J ulie Bauer, J ohn Biermacher, Bob Dalga, and
Kate Weaver of Pharmacia & Upjohn for their help in this project,
and thanks Pharmacia & Upjohn, Inc. for the opportunity to carry
out an industrial externship. This project was supported by the
Takeru Higuchi Predoctoral Fellowship (M.C.L.), an NIGMS
Biotechnology Training Grant, Pharmacia & Upjohn, Inc., and by
NIH grant #GM-54195.
References and Notes
1. Carpenter, J . F.; Pikal, M. J .; Chang, B. S.; Randolph, T. W.
Rational design of stable lyophilized protein formulations:
Some practical advice. Pharm. Res. 1997, 14, 969-975.
2. Hageman, M. J . Water sorption and solid-state stability of
proteins, In Stability of Protein Pharmaceuticals, Part A:
Chemical and Physical Pathways of Protein Degradation;
Ahern, T. J .; Manning, M. C. Eds.; Plenum: New York, 1992;
pp 273-309.
3. Pikal, M. J .; Dellerman, K. M.; Roy, M. L.; Riggin, R. M. The
effect of formulation variables on the stability of freeze-dried
human growth hormone. Pharm. Res. 1991, 8, 427- 436.
4. Constantino, H. R.; Langer, R.; Klibanov, A. M. Moisture-
induced aggregation of lyophilized insulin. Pharm. Res. 1994,
11, 21-29.
5. Strickley, R. G.; Anderson, B. D. Solid-state stability of
human insulin. I. mechanism and the effect of water on the
kinetics of degradation in lyophiles from pH 2-5 solutions.
Pharm. Res. 1996, 13, 1142-1153.
Because the amount of Asu observed depends on the
rates of Asn-hexapeptide deamidation and Asu hydrolysis,
the decrease in the percentage of Asu among the degrada-
tion products with increasing water content suggests that
Asu hydrolysis becomes more rapid than Asu formation
under these conditions. The shift in product distribution
with increasing glycerol content (Figures 6, 7a, and 7b)
suggests that mobility may also affect Asu hydrolysis
through its role as a plasticizer to increase formulation
mobility, although direct evidence for a plasticizing role
in this reaction was not obtained in these experiments.
These differences in product distribution may be due to
differences in the reaction time course. For example,
reactions in formulations with higher glycerol contents
occur at a faster rate. Therefore, these reactions would be
more complete than slower reactions (lower glycerol con-
tent) at the time the reactions were sampled.
6. Oliyai, C.; Patel, J .; Carr, L.; Borchardt, R. T. Solid-state
stability of lyophilized formulations of an asparaginyl residue
in a model hexapeptide. J . Parenteral Sci. Technol. 1994, 48,
167-173.
7. Shalaev, E. Y.; Zografi, G. How does residual water affect
the solid state degradation of drugs in the amorphous state?
J . Pharm. Sci. 1996, 85, 1137-1141.
8. Levine, H.; Slade, L. The glassy state phenomenon in food
molecules. In The Glassy State in Foods; Blanshard, J . M.
V.; Lillford, P. J ., Eds.; Nottingham: Nottingham, U.K. 1993;
pp 35-101.
9. Ferry, J . D. Viscoelastic Properties of Polymers, 3rd ed.; Wiley
& Sons: New York, 1980; pp 486-544.
10. Angell, C. A. Formation of glasses from liquids and biopoly-
mers. Science 1995, 267, 1924-1935.
11. Williams, M. L.; Landel, R. F.; Ferry, J . D. The temperature
dependence of relaxation mechanisms in amorphous poly-
mers and other glass-forming liquids. J . Am. Chem. Soc.
1955, 77, 3701-3707.
12. Hancock, B. C.; Zografi, G. Characteristics and significance
of the amorphous state in pharmaceutical systems. J . Pharm.
Sci. 1997, 86, 1-12.
13. Bell, L. N.; Hageman, M. J . Differentiating between the
effects of water activity and glass transition dependent
Conclusion
The mechanistic role of water in the deamidation of an
Asn-containing model hexapeptide (Val-Tyr-Pro-Asn-Gly-
mobility on
a solid-state chemical reaction: Aspartame
degradation. J . Agric. Food Chem. 1994, 42, 2398-2401.
1088 / Journal of Pharmaceutical Sciences
Vol. 88, No. 10, October 1999