Stabilization of vaccines and antibiotics in silk
and eliminating the cold chain
Jeney Zhanga,b,1, Eleanor Pritcharda,1, Xiao Hua, Thomas Valentina, Bruce Panilaitisa, Fiorenzo G. Omenettoa, and
David L. Kaplana,2
aTufts University, Department of Biomedical Engineering, Medford, MA 02155; and
Medford, MA 02155
bTufts University, Department of Chemical & Biological Engineering,
Edited by Arnold L. Demain, Drew University, Madison, NJ, and approved June 12, 2012 (received for review April 12, 2012)
Sensitive biological compounds, such as vaccines and antibiotics,
traditionally require a time-dependent “cold chain” to maximize
therapeutic activity. This flawed process results in billions of dollars
worth of viable drug loss during shipping and storage, and se-
verely limits distribution to developing nations with limited infra-
structure. To address these major limitations, we demonstrate self-
standing silk protein biomaterial matrices capable of stabilizing
labile vaccines and antibiotics, even at temperatures up to 60°C
over more than 6 months. Initial insight into the mechanistic basis
for thesefindings is provided.Importantly,thesefindingssuggesta
transformative approach to the cold chain to revolutionize the way
many labile therapeutic drugs are stored and utilized throughout
measles ∣ mumps ∣ rubella ∣ penicillin
and antibiotics are important components of an effective infec-
tious disease containment strategy; antibiotics represent a rescue
measure while vaccination can be a primary mode of disease pre-
vention. Unfortunately, the use of vaccines and antibiotics is
severely limited in the poorest countries where infectious diseases
account for more than half of all deaths (2). Due to temperature
sensitivity, vaccine and antibiotic formulations must be main-
tained within a specific refrigeration temperature range. Because
ambient temperatures in the developing world deviate signifi-
cantly from refrigeration temperatures, the successful delivery
of active vaccines and antibiotics depends on the “cold chain”
system, a distribution network to maintain optimal cold tempera-
tures during transport, storage, and handling. Cold chain require-
ments represent a major economic and logistical burden, particu-
larly in lower resource settings, where refrigeration and electricity
can be limited (3, 4). The cold chain alone can account for 80% of
the financial cost of vaccination (1) and is estimated to cost vac-
cine programs $200–300 million per year (5). Deficiencies in the
process frequently occur even in industrialized countries (6, 7).
For temperature sensitive compounds like vaccines and antibio-
tics, maintaining the cold chain is critical for adequate bio-
activity (8, 9). Failures in the cold chain result in costly waste
and the loss of nearly half of all global vaccines (10). Such failures
can also result in the delivery of ineffective, subtherapeutic doses.
For antibiotics, this problem can be associated with the develop-
ment ofantibiotic-resistant strains, a major public health concern.
Silk fibroin is a biologically-derived protein polymer purified
from domesticated silkworm (Bombyx mori) cocoons that has
demonstrated excellent properties for biomedical applications,
including biocompatibility (11–14), robust mechanical strength
(15), and slow, controlled degradation to nontoxic products in
vivo (16). Silk can be prepared in a range of material formats,
including films, hydrogels and microspheres (17, 18). Silk can
be processedentirely in aqueous systemsusing mild, ambient con-
ditions of temperature and pressure, allowing the incorporation
of labile compounds without loss of bioactivity (18, 19). Silk is
composed of a block copolymer structure with large hydrophobic
ore than 17 million people die every year from infectious
diseases, particularly in the developing world (1). Vaccines
domains interspersed with small hydrophilic regions that form
organized crystalline domains (β-sheets) that stabilize viaphysical
crosslinks. This assembly forms nanoscale pockets that can immo-
bilize bioactive molecules and improve their stability by minimiz-
ing water content and reducing protein chain mobility, thus pre-
venting a transition from the native to denatured state (20, 21).
Extensive physical crosslinking, the hydrophobic nature of the
protein and high glass transition temperature (around 178 °C)
render silk highly thermodynamically stable to changes in tem-
perature and moisture, and mechanically robust due to the heav-
ily networked β-sheet structures (21, 22). Due to its unique struc-
ture, encapsulation of therapeutic compounds in silk matrices
could stabilize labile antibiotics and vaccines, akin to enzyme sta-
bilization in silk (22).
Our model vaccine to investigate stabilization effects of silk
matrices was the live measles, mumps and rubella (MMR) vac-
cine. The commercially available MMR vaccine contains a variety
of protein and salt stabilizers; however, even with the stabilizers
the vaccine rapidly loses potency at temperatures above the re-
commended 2–8°C (23–25). Our model antibiotics were penicil-
lin and tetracycline. Dating back to Fleming’s original 1929 paper
on penicillin (26), penicillin is unstable in solution, breaking down
within weeks at 25°C and within 24 h at 37°C (27). Tetracycline
possesses broad specificity and low cost, but undergoes rapid
photolysis and hydrolysis in solution (28, 29).
The objective of the present study was to evaluate silk proteins
as a matrix for vaccine and antibiotic storage and stabilization.
We demonstrate remarkable stabilization of these labile com-
pounds and also provide mechanistic insight. Because silk can
be formed into any variety of material formats, including films,
hydrogels, microspheres and microneedles (17, 30), the proposed
system has potential use ex vivo or in vivo as drug delivery vehi-
cles. Further, the results provide a feasible path forward to revo-
lutionize the cold chain and provide more efficient and wide-
spread distribution of labile therapeutics throughout the world
Antibiotic Stabilization in Silk Films. Tetracycline stability was com-
pared stored in solution (Fig. 1A–B) or entrapped in silk films
(Fig. 1C) over 4 wks at 4°C, 25 °C, 37 °C, and 60 °C with light
protection, and 25 °C with ambient light exposure. In solution,
loss of tetracycline activity was observed at all storage tempera-
tures, even refrigeration at 4°C. Entrapped in silk, tetracycline
Author contributions: J.Z., E.P., F.G.O., and D.L.K. designed research; J.Z., E.P., X.H., T.V., and
B.P. performed research; J.Z. and E.P. contributed new reagents/analytic tools; J.Z., E.P.,
X.H., T.V., B.P., F.G.O., and D.L.K. analyzed data; and J.Z., E.P., X.H., B.P., F.G.O., and
D.L.K. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
1J.Z. and E.P. contributed equally to this work.
2To whom correspondence should be addressed. E-mail: firstname.lastname@example.org.
This article contains supporting information online at www.pnas.org/lookup/suppl/
July 24, 2012
activity was lost only for films stored at 60°C, and even at this
temperature, activity loss was relatively low; approx. 10% and
20% loss after 2 and 4 wks, respectively, compared to 100% loss
after 2 wks for tetracycline stored in solution. Storage at body
temperature (37°C) resulted in no activity loss for storage in silk
films, compared to 80% lost within 4 wks for storage in solution.
Tetracycline storage stability was also investigated in dry powder
over 9 mo at various temperatures with light protection: 4 °C
(Fig. 1D), 25 °C (Fig. 1E), 37 °C (Fig. 1F) and 60°C (Fig. 1G).
With the exception of samples stored for 2 mo at 37 °C, retention
of tetracycline bioactivity in silk films was equal to or higher than
that of tetracycline stored in dry powder format. At the 6 and
9 mo sampling times, residual tetracycline bioactivity was signifi-
cantly improved for storage in silk films compared with storage in
dry powder format for all storage temperatures tested (one tailed
t-test, df ¼ 4, p < 0.05).
Stability studies were also carried out comparing the residual
activity of penicillin stored in silk films versus other material for-
mats (including storage in solution, storage as dry powder and
entrapped in collagen films) at 4 °C , 25°C , 37°C and 60 °C
(Fig. S2). With a few exceptions the stability of penicillin incor-
porated into silk films was higher than that of penicillin stored in
any other form tested, including dry powder at 4°C. While sta-
bility rapidly declined for penicillin solutions stored at 25 °C
and 37°C (consistent with previous reports as seen in ref. 27),
approximately 50% of the initial penicillin activity was retained
after 183 d of storage in silk films for all temperatures tested. For
the first 40 d of storage, incubation enhanced penicillin activity
above 100% of the initial value, a phenomenon also observed for
previously reported enzyme stabilization in silk films and likely
related to initial aggregation and subsequent reactivation of the
bioactive component (22). Total activity loss was observed within
24 h for penicillin stored in solution at 60 °C. Penicillin stored in
collagen films or in dry powder exhibited more than 20% activity
loss over 30 d. In contrast, no loss of activity was observed for
penicillin stored in silk films at 60°C for 30 d.
Vaccine Stabilization in Silk Films. Encapsulation in silk films
enhanced stability (expressed as the residual potency remaining
post-storage compared with initial potency) of the measles,
mumps, and rubella vaccines at elevated temperatures (Table S1).
Reconstituted vaccine loses potency rapidly in solution (Fig. S3),
a result consistent with manufacturer specifications that any un-
used MMR vaccine be discarded 8 h post reconstitution. To mini-
mize loss of vaccine potency during the solution stage of silk
encapsulation, lyophilized MMR-silk films were prepared. Com-
pared to the initial potency recovered from the air-dried silk
25°C with ambient light exposure at 1 and 4 wks. (B) Residual bioactivity over 4 wks storage of tetracycline stored in solution or (C) in 6% (w∕v) silk films at 4 °C,
25°C, 37°C, 60 °C, and 25°C with ambient light exposure. Bioactivity measured using a zone of inhibition assay in S. aureus lawns. N ¼ 4, error bars represent
standard deviations (D) Comparison of tetracycline stored in silk films and as dry powder at 4°C (refrigeration), (E) 25°C (room temperature), (F) 37°C (body
temperature)and (G) 60°C over 9 months. Activitymeasured using a zone ofinhibition assay inS. aureus lawns.N ¼ 3, error bars represent standard deviations.
Data analyzed by one-tailed t-test, df ¼ 4; significance levels of individual tests are indicated: *P < 0.05, **P < 0.01, ***P < 0.005.
Tetracycline Storage Stability. (A) Aliquots of tetracycline in solution and tetracycline loaded silk films stored at −20°C, 4°C, 25 °C, 37°C and 60°C, and
www.pnas.org/cgi/doi/10.1073/pnas.1206210109Zhang et al.
films, the lyophilized films improved the recovery of measles,
mumps, and rubella to 94.7%, 87.0%, and 98.4%, respectively.
Encapsulation in silk film (either air-dried or lyophilized)
enhanced measles, mumps, and rubella viral particle stability over
six months, particularly at elevated storage temperatures (Fig. 2).
Both silk films showed improved residual potency of the measles
component compared to lyophilized MMR vaccine powder stored
at 25°C, 37°C and 45°C. After 6 months stored at 25°C, measles
encapsulated in silk films retained 83.9% potency compared to
74.5% for the lyophilized vaccine powder. After 6 mo stored at
37°C, silk films showed a dramatic improvement in stability of
to 9.9% from the powder. At 45°C, the measles vaccine lost all
potency after 20 wks in storage while the silk films retained
53.4% activity after 24 wks. The mumps (Fig. 2B) and rubella
Lyophilization of the silk films not only improved vaccine
activity recover postencapsulation, but also provided even greater
thermostabilization of the vaccines. Stabilization of all vaccine
components in lyophilized silk films was found to be independent
of storage temperature: After 6 mo of storage at 37 °C and 45°C
in lyophilized silk films, ≥85% initial potency was retained for all
components of the vaccine (measles, mumps, and rubella). Be-
cause activity loss during storage in solution or as lyophilized vac-
cine powder increased with increasing temperature, the improve-
ment in stabilization resulting from silk film encapsulation is most
dramatic at elevated storage temperatures. In plots of predicted
half-lives versus storage temperature, slopes for vaccine stored
as powder were consistently higher than the slopes for vaccine
stored in either silk film systems, indicating that storage in pow-
der results in greater degradation rate increases with increased
storage temperature (Fig. 3B). As seen in the comparison of
estimated degradation rates and corresponding half-lives of the
three vaccine systems (Table S2), stability improvements observed
for silk systems become more dramatic as the storage tempera-
ture increases. With the exception of storage at 4 °C, the silk films
and lyophilized silk films exhibited a greater predicted half-life of
all three viral components of the vaccine. The difference was
especially pronounced in the predicted half-life of the virus at
37°C and 45 °C. For the measles component, entrapment in silk
film increased the viral half-life at 37°C from 9.4 wks for dry
powder to 22.0 wks for silk films and 93.8 wks for lyophilized silk
films. Storage at 45°C provided similarly impressive stabilization
results: virus half-lives for powder, silk films and lyophilized silk
films were 5.0, 19.8, and 107.6 wks, respectively. The Arrhenius
plot of vaccine stored as powder has the steepest slope, indicating
the greatest increase in rate ofdegradation from storage in 4 °C to
45°C and thus a greater temperature-dependence of degradation
compared to the silk film systems, which have more shallow
slopes (Fig. 3A).
Mechanisms of Stabilization. The shelf life of lyophilized vaccines is
dependent on both adherence to the cold chain and maintenance
of low residual moisture content (25). The residual moisture of
MMR powder, MMR-silk films and lyophilized MMR-silk films
was 2.47% ? 0.25, 4.60% ? 0.83, and 1.85% ? 0.30, respectively
(Table S3). Due to processing conditions, MMR-silk films had
higher residual moisture, but the net increase in moisture at 6 mo
of storage at 45°C was 28.3%, compared to a 59.5% increase for
the MMR powder system. Residual moisture content of the vac-
cine storage systems was found to correspond to residual potency.
At elevated temperatures, MMR powder exhibited greater losses
Residual potency of the (A) measles, (B) mumps, and (C) rubella components of the MMR vaccine. (⧫) lyophilized MMR-silk films, (○) MMR-silk films, (▪) MMR
powder. N ¼ 3, error bars represent standard deviations.
Measles, Mumps, and Rubella Vaccine Stability. Vaccines prepared in 9% (w∕v) silk and lyophilized films over 6 mo at 4°C, 25°C, 37°C, and 45°C.
Zhang et al.PNAS
July 24, 2012
in potency and greater increases in residual moisture than the silk
MMR-silk films corresponded to greater viral potency retention.
The enhanced thermostability provided by silk in contact
with the vaccines was assessed with differential scanning calorime-
try (DSC) and light scattering. DSC demonstrated that the silk en-
capsulation increased glass transition (Tg) of the vaccine (Fig. 4B).
DSC thermogram of the MMR powder showed a Tg at 68.9°C,
while the lyophilized MMR-silk films showed a Tg at 89.2°C as
well as two peaks at 116.6°C and 164.8°C indicative of a Tm and
a Td (melting and degradation temperatures, respectively) of the
complex vaccine system. It was unclear whether the Tg shift was
due to structural stabilization by the silk or by the interaction be-
tween silk and the various excipients present in the vaccine sample.
Also, the DSC curve of MMR powder includes a complicated
mixed melting and degradation process. Without the protection
at a much lower temperature (starting around approximately 110–
120°C) and this fully or partially covered the melting process of
MMR powders. Therefore, the vaccine was purified to remove the
excipients. As the viral sample is a liquid solution, nano-DSC was
run on these samples. The nano-DSC thermogram of the purified
viral particles in water showed a transition point (Tp) at 16.8°C,
indicating the viral proteins were undergoing a conformational
change (Fig. 4C). The solution of purified viral particles in silk
showed an elevated transition point at 68.3°C. While the range
of the nano-DSC does not extend far enough to show the Tg of
the silk, a thermogram of silk solution is still shown to illustrate
that neither of the transition point values can be attributed to a
change in the silk structure alone, indicating a positive impact
of silk proteins on the vaccine as reflected in the elevated Tg.
The elevated transition point of the vaccine-encapsulated silk
solution is due to structural stabilization provided by the silk to
prevent viral protein denaturation and aggregation (Fig. 4A-ii).
The transition point value at 16.8°C is likely due to the unfolding
of the viral surface glycoproteins (Fand N of measles and mumps
and E1 and E2 of rubella) due to the elevated heat applied to the
viral particles. This denaturation most likely leads to the aggre-
gation of the viral particles (Fig. 4A-iv), preventing them from
binding and fusing with the animal cells used in testing, thus lead-
ing to loss of infectivity of the viral particles and vaccine. The
broadness of the transition peaks is likely a result of the combined
contribution of the unfolding of several proteins present in the
sample. Furthermore, protein aggregation may also contribute
to the broad peaks by overlapping with the endotherm of dena-
turation (31). The infectivity of MMR is dependent on the con-
formational stability of the viral proteins. The presence of silk
molecules increased transition point values, indicating that the
silk provided structural stability to the viral proteins, shielding
them from thermal denaturation. The interaction between the
viral particles and the silk hydrophobic regions, as well as limited
chain mobility, might prevent viral protein aggregation, thereby
preserving viral and vaccine infectivity.
To further investigate whether transition points corresponded
to viral particle denaturation due to aggregation, viral particle
size was examined by dynamic light scattering (DLS see Fig. 4D).
The average mean effective diameter of a naked viral particle was
around 250 nm, consistent with reported diameter ranges of
measles and mumps (32, 33). Purified virus solution exhibited an
increase in mean effective diameter around 16 °C, indicating the
presence of viral protein aggregation due to the increased ther-
mal input. In contrast, viral particles in silk solution did not show
signs of aggregation until 70 °C, indicating silk provided structural
stability that prevented thermal-induced aggregation of the viral
proteins. These DLS results correspond well with the TP values
determined by the nano-DSC, as the aggregation detected by
light scattering occurred within the temperature range of the pro-
tein unfolding measured by DSC.
Activity loss at body temperature represents a serious problem for
long-term, implantable drug delivery systems, as well as for the
cold chain requirements. Formulations which are unstable at tem-
peratures ≥25°C are difficult to transport and store, particularly in
developing countries where refrigeration is limited. Immobiliza-
tionofbioactivemoleculesleads toanincreaseinstability bymain-
taining constant environmental conditions to protect the bioactive
molecules against potential degradative variables such as pH, tem-
perature or ionic strength changes and reducing molecular mobi-
lity (34). The chemistry, structure, and assembly of silk generates a
unique nanoscale environment (35) and makes this protein poly-
mer an attractive candidate for the stabilization of bioactive mo-
lecules over extended periods of time. Without chemical proces-
sing, aqueous silk solution can be used to entrap bioactive mole-
cules in amphiphilic, self-assembly domains.
Prolonged exposure to temperatures above recommended sto-
rage conditions can damage vaccines in various ways, most nota-
bly by altering the tertiary structure of the viral proteins, reducing
viral infectivity and thereby decreasing vaccine potency (36). The
main cause of viral inactivation is disruption of viral surface pro-
teins and stresses such as elevated temperature can induce con-
formational changes in viral proteins (37). These conformational
changes may affect the stability of the virus by inducing viral par-
ticle aggregation that prevents viral fusion and uptake, thus lead-
ing to virus inactivation (Fig. 4A and ref. 38). The silk films may
have improved recovery of active vaccine by minimizing exposure
to degradative enzymes (39). Generally, storage in either silk film
format decreased observed degradation rates for all three virus
components. Lyophilized silk systems exhibited the lowest degra-
dation rates and longest half-lives, while lyophilized vaccine pow-
der degraded fastest at all storage temperatures.
plots of the degradation rates of the measles, mumps, and rubella compo-
nents of the vaccine as a function of the inverse of the absolute temperature.
(B) Predicted half-lives of the measles, mumps and rubella viral components
as a function of temperature and the corresponding upper and lower limits
of the half-life. The predicted half-lives represent the estimated time re-
quired for the viral component to degrade to 50% of the initial value.
(⧫) lyophilized MMR-silk films, (○) MMR-silk films, (▪) MMR powder.
Kinetics of Degradation of MMR-Silk Film Systems. (A) Arrhenius
www.pnas.org/cgi/doi/10.1073/pnas.1206210109 Zhang et al.
Humidity can also have a significant effect on vaccine products
as the excess water introduced to the system can lead to increase in
mobility and corresponding reactivity of the viral proteins (40).
The residual moisture analysis of the films revealed increased re-
sidual moisture overthe course of6 moin storage,especially in the
high temperature ranges. While MMR powder and lyophilized
MMR-silk films were stored in low humidity conditions provided
by lyophilization vials and sealing in a nitrogen-rich environment,
MMR-silk films stored in Eppendorf tubesallowed greater absorp-
tion of humidity. The increase in residual moisture in silk films
at elevated temperatures could also be explained by water deso-
rption from the silk, because they still contain traces of intermo-
lecular tubes bound water (41). Though humidity absorption is
higher in nonlyophilized silk films, silk films stored in Eppendorf
demonstrate the feasibility of silk as a simple, straightforward sto-
to existing more complicated vaccine formulations. Lyophilized
silk films combine the advantages of silk storage and freeze drying,
further reducing protein mobility and improving stabilization com-
pared to air-dried films. Though the absolute residual moisture of
the silk films is higher than that of the MMR powder, the percent
increase of the residual moisture in the powder over the tested
temperature range was greater than that observed in the silk films.
The increase in temperature appears to have had a greater impact
on moisture in the powder than the silk films. This suggests silk
inhibited molecular mobility during storage to prevent protein un-
folding and subsequent aggregation.
The mechanisms involved in antibiotic stabilization in silk may
be attributable to a combination of low water content and silk
surface chemistry to reduce aggregation or degradation (42).
We have shown silk as an effective carrier material for enhanced
thermostability of both antibiotics and vaccines. Both silkfilm sys-
tems were able to increase the half-lives of the vaccine compared
to the manufacturer supplied vaccine formulation at 25 °C, 37 °C,
and 45 °C. Silk reduced the temperature-induced protein unfold-
ing and subsequent aggregation by reducing residual moisture
during storage at elevated temperatures, and also provided struc-
tural stability to the vaccine to elevate the temperature at which
the viral proteins denature.
Pronounced stabilization by silk at the elevated temperatures
that result in vaccine spoilage or antibiotic activity loss in con-
ventional formulation when the cold chain is broken (37°C and
45°C) suggest silk films would provide sufficient stability over a
wide range of storage temperatures. When the stabilization data
presented here are combined with the remarkable mechanical fea-
tures and tunable release kinetics characteristic of silk carriers (30,
43, 44), a robust stabilization and delivery platform for antibiotics
can be envisioned, extending even to microneedle formats (30). In
chain,providing new venues towardsimprovedprocessing,distribu-
tion and use of labile therapeutics such as antibiotics and vaccines.
Materials and Methods
Silk Fibroin Purification. Silk fibroin aqueous solutions were prepared as pre-
viously described (16).
Antibiotic Stabilization Studies. The bacteria strains used were E. coli ATCC
25922 and S. aureus ATCC 25923 (American Type Culture Collection, Manassas,
For long-term stability studies, tetracycline and penicillin loaded silk films
were prepared as previously described (preparation, loading and treatment
details in SI Materials). Collagen films were prepared by dissolving Avitene®
Microfibrillar Collagen Flour (Bard Davol, Warwick, RI) in sterile ultrapure
water in a weight∕volume concentration equal to the silk solution used for
film preparation, then mixing with antibiotic solution, casting films and dry-
ing overnight at ambient conditions. Storage temperatures tested were 4 °C,
25°C, 37°C and 60 °C. For tetracycline studies, samples were wrapped in foil
to protect from light exposure, except a group ofsamples stored at 25 °C with
exposure toambient light. Residual bioactivity ofantibiotic in various storage
systems was evaluated using a direct zone of inhibition assay based on the
principle of the Kirby-Bauer Susceptibility Test (45, 46) (SI Materials).
Vaccine Stabilization Studies. Trivalent vaccine. A commercial source of triva-
lent measles, mumps, rubella vaccine was used for potency estimation:
in nucleocapsids within a lipid bilayer. The viral envelop consist of the matrix (M), haemaglutinin (H) and fusion (F) protein. (ii) By a combination of hydrophobic
interaction and limited chain mobility, silk-entrapped viral particles maintain structural activity at elevated temperatures. (iii) Structurally intact F and
H proteins bind and fuse to the receptors CD46 and CD150 to gain entry into the cell to initiate viral replication. (iv) Denaturation of the surface proteins
can result in aggregation of the viral particles and binding and fusion cannot occur. (B) Solid-state DSC. Tg¼ glass transition point. (C) Nano-DSC.
TP¼ protein transition point. The presence of silk increases the TPof the viral particles, indicating the effect of silk on the encapsulated viral proteins.
The drop following the TPis most likely indicative of an aggregation event as a result of the protein unfolding. (D) Comparison of dynamic light scattering
of purified viral particles in water and purified viral particles in silk solution.
Thermal Analysis of Stabilization of MMR-Silk Films. (A) (i) Measles and mumps structures consist of single-stranded, negative-sense RNA enclosed
Zhang et al.PNAS
July 24, 2012
MMR® II (Merck & Co., Inc., USA), a sterile lyophilized live virus vaccine con-
taining the Enders’ attenuated Edmonston measles, the Jeryl Lynn mumps
and Wistar RA 27/3 rubella. Virus was purified by reconstituting lyophilized
vaccine powder in sterile water, loading into 0.5 kDa dialysis tubing (Sigma
Aldrich) and dialyzing against a 0.15 M NaCl solution to remove excipients.
The recovered vaccine solution was then run through a PD-10 desalting col-
umn (GE Healthcare) per manufacturer specifications to remove excess salt.
Recovered purified viral particle solution was collected and stored in an Ep-
pendorf tube at −80°C until use. Viral infectivity was evaluated using a quan-
titative real-time RT-PCR viral infectivity assay (SI Materials). Degradation
behavior was characterized in terms of predicted viral half-life, degradation
rate and slope of the Arrhenius plot (SI Materials).
Vaccine entrapment in silk films. Standard MMR-silk films were prepared by
casting a 1∶1 weight ratio mixture of MMR: silk on Teflon-coated molds and
drying films at room temperate for 12 h in a sterile hood protected from
light. Individual films were placed in Eppendorf tubes, under ambient con-
ditions, and stored at 4°C, 25°C, 37 °C, and 45 °C for stability studies. Lyophi-
lized MMR-silk films were prepared by aliquoting a 1∶1 weight ratio MMR:
silk solution into 96-well plates and freeze dried using a VirTis 25L Genesis SQ
Super XL-70 Freeze Dryer. The samples were frozen at −45°C for 480 min. The
primary drying occurred at −20°C for 2,400 min and secondary drying at 35°C
for 620 min. The samples were held at −45°C until they were removed from
the lyophilizer. The films were then removed from the well plates and trans-
ferred to 5 cc glass serum vials. Five mm lyophilization stoppers were applied
to the vials under nitrogen and vacuum conditions in a MBRAUN LABmaster
glovebox (Garching, Germany) and a 5-mm crimper was used to tighten the
5-mm alum seal on the vials. The vials were stored at 4 °C, 25 °C, 37 °C, and 45
°C for stability studies. The vials, stoppers, seals and crimper were supplied by
VWR (Bridgeport, NJ).
Residual moisture determination. Residual moisture of the lyophilized vaccine
powder, MMR-silk films, and lyophilized MMR-silk films was measured by the
thermo-gravimetric method, modified from Worrall et al. (47)
Differential scanning calorimetry (DSC). Five mg of solid samples were encap-
sulated in Al pans and heated in a TA Instrument Q100 DSC (New Castle, DE)
with a purged dry nitrogen gas flow of 50 mL∕min. Tg was recorded as
the onset temperature of the discontinuity curve of the heat flow versus tem-
perature. All measurements were made at 10°C∕min. The samples were
initially equilibrated at −20°C for 5 min and then heated to 200 °C, held
at 200°C for 5 min, followed by cooling to 20°C. Nano-DSC measurements
were taken on a CSC Model 6100 Nano II Differential Scanning Calorimeter
Dynamic light scattering (DLS). The size of the measles, mumps and rubella
viral particles as a function of temperature was monitored by DLS. A
400 μL aliquot of 2 mg∕mL sample solution was filtered through a
0.45 μm syringe filter (GE, Fairfield, CT). DLS was conducted using the Dyna-
Pro DLS system (Wyatt Technology, Santa Barbara, CA) with parameters set at
60 s acquisition time, 10 number of acquisition and laser power of 75 mW. A
100 μL aliquot of the sample was transferred into an RNAse-free, DNAse-free,
protein-free UVette Eppendorf cuvette to be inserted into the DLS.
Guziewicz, Michael Lovett, Nikola Kojic and Reid McCabe is gratefully
acknowledged. We thank the NIH (P41 Tissue Engineering Resource Center
EB002520) and EY020856, DE017207-01, EB003210 and AFOSR 9550-10-1-
0172zsAJMn for support of this work.
The technicalassistance oradvice ofNicholas
1. Chen X, et al. (2011) Improving the reach of vaccines to low-resource regions, with a
needle-free vaccine delivery device and long-term thermostabilization. J Controlled
2. Yager P, et al. (2006) Microfluidic diagnostic technologies for global public health.
3. Cheyne J (1989) Vaccine delivery management. Rev Infect Dis 11:S617–S622.
4. Wolfson LJ, et al. (2008) Estimating the costs of achieving the WHO-UNICEF Global
Immunization Vision and Strategy, 2006–2015. Bull W H O 86:27–39.
5. Das P (2004) Revolutionary vaccine technology breaks the cold chain. Lancet Infect Dis
6. Bishai DM (1992) Vaccine storage practices in pediatric offices. Pediatrics 89:193–196.
7. Cheriyan E (1993) Monitoring the vaccine cold chain. Arch Dis Child 69:600–601.
8. Adu FD, Adedeji AA, Esan JS, Odusanya OG (1996) Live viral vaccine potency: an index
for assessing the cold chain system. Public Health 110:325–330.
9. Arya SC (2001) Stabilization of vaccines: To be or not to be. Vaccine 19:595–597.
10. UNAIDA, UNICEF, UNESCO, UNFPA, WB, WHODavey S, ed. (2000) A key to prosperity:
Success stories in developing countries. Health (World Health Organization, Commu-
nicable Diseases, Geneva).
11. Leal-Egaña A, Scheibel T (2010) Silk-based materials for biomedical applications.
Biotechnol Appl Biochem 55:155–167.
12. Meinel L, et al. (2005) The inflammatory responses of silk films in vitro and in vivo.
13. Seo Y, et al. (2009) Increase in cell migration and angiogenesis in a composite silk
scaffold for tissue-engineered ligaments. J Orthop Res 27:495–503.
14. Panilaitis B, et al. (2003) Macrophage responses to silk. Biomaterials 24:3079–3085.
15. Altman GH, et al. (2003) Silk-based biomaterials. Biomaterials 24:401–416.
16. Wang X, et al. (2008) Controlled release from multilayer silk biomaterial coatings to
modulate vascular cell responses. Biomaterials 29:894–903.
17. Omenetto F, Kaplan DL (2010) From silk cocoon to medical miracle.SciAm303(5):76–77.
18. Vepari C, Kaplan DL (2007) Silk as a biomaterial. Prog Polym Sci 32(8–9):991–1007.
19. Lawrence BD, Cronin-Golomb M, Georgakoudi I, Kaplan DL, Omenetto FG (2008) Bioac-
20. Guziewicz N, Best A, Perez-Ramirez B, Kaplan DL (2011) Lyophilized silk fibroin hydro-
gels for the sustained local delivery of therapeutic monoclonal antibodies. Biomater-
21. Pritchard EM, Kaplan DL (2011) Silk fibroin biomaterials for controlled release drug
delivery. Expert Opin Drug Delivery 8:797–811.
22. Lu Q, et al. (2010) Stabilizationand release of enzymes from silk films. Macromol Biosci
23. Colinet G, Rossignol J, Peetermans J (1982) A study of the stability of a bivalent
measles-mumps vaccine. J Biol Stand 10:341–346.
24. McAleer WJ, Markus HZ, McLean AA, Buynak EB, Hilleman MR (1980) Stability on
storage at various temperatures of live measles, mumps and rubella virus vaccines
in new stabilizer. J Biol Stand 8:281–287.
25. GalazkaA,MilstienJ, ZaffranM (1998) ThermostabilityofVaccines.GlobalProgramme
for Vaccines and Immunization (World Health Organization, Geneva).
26. Fleming A (1929) On the antibacterial action of cultures of a penicillium with special
reference to their use in the isolation of B. influenza. Br J Exp Pathol 3:226–236.
27. Benedict RG, Schmidt WH, Coghill RD, Oleson PA (1945) The stability of penicillin in
aqueous solution. J Bacteriol 49:85–95.
28. Xuan R, Arisi L, Wang Q, Yates SR, Biswas KC (2010) Hydrolysis and photolysis of
oxytetracycline in aqueous solution. J Environ Sci Health part B 45:73–78.
29. Sun Y, Scruggs DW, Peng Y, Johnson JR, Shukla AJ (2004) Issues and challenges in
developing long-acting veterinary antibiotic formulations. Adv Drug Delivery Rev
30. Tsioris K, et al. (2012) Fabrication of silk microneedles for controlled-release drug
delivery. Adv Funct Mater 22:330–335.
31. Packer L, Colowick SP, Kaplan NO (2002) Methods in enzymology, ed L Packer
(Academic Press, California, USA), Volume 349.
32. Hall WW, Martin SJ (1973) Purification and Characterization of Measles Virus. J Gen
33. Merz DC, Wolinsky JS (1983) Conversion of nonfusing mumps virus infections to fusing
infections by selective proteolysis of the HN glycoprotein. Virology 131:328–340.
34. Gianfreda L, Modafferi M, Greco G, Jr (1984) Enzyme stabilization towards thermal,
chemical and proteolytic deactivation. Enzyme Microb Technol 7:78–82.
35. Jin HJ, Kaplan DL (2003) Mechanism of silk processing in insects and spiders. Nature
36. Chen D, Kristensen D (2009) Opportunities and challenges of developing thermostable
vaccines. Expert Rev Vaccines 8:547–558.
37. Ausar SF, Hudson TR, Hudson MH, Vedvick TS, Middaugh CR (2006) Conformational
stability and disassembly of Norwalk virus like particles: effect of pH and temperature.
J Biol Chem 281:19478–19488.
38. Brandau DT, Jones LS, Wiethoff CM, Rexroad J, Middaugh CR (2003) Minireview: Ther-
mal stability of vaccines. J Pharm Sci 92(2):218–231.
39. Li S, Schoneich C, Borchardt RT (1994) Chemical instability of protein pharmaceuticals:
mechanisms of oxidation and strategies of stabilization. Biotechnol Bioeng 48:490–500.
40. Waterman K, Adami RC (2005) Accelerated aging: Prediction of chemical stability of
pharmaceuticals. Int J Pharm 293:101–125.
41. Hu X, Kaplan DL, Cebe P (2007) Effect of water on the thermal properties of silk
fibroin. Thermochim Acta 461:137–144.
42. Pritchard EM, Dennis PB, Omenetto F, Naik RR, Kaplan DL (2012) Physical and chemical
aspects of stabilization of compounds in silk. Biopolymers 97:479–498.
43. Wang Y, Kim H-J, Vunjak-Novakovic G, Kaplan DL (2006) Stem cell-based tissue
engineering with silk biomaterials. Biomaterials 2:6064–6082.
44. Pritchard EM, Szybala C, Boison D, Kaplan DL (2010) Silk fibroin encapsulated powder
reservoirs for sustained release of adenosine. J Controlled Release 144:159–167.
45. Boyle VJ, Fancher ME, Ross RW (1973) Rapid, modified Kirby-Bauer susceptibility test
with single, high-concentration antimicrobial disks. Antimicrob Agents Chemother
46. Hu ZQ, Zhao W-H, Hara Y, Shimamura T (2001) Epigallocatechin gallate synergy with
ampicillin/sulbactam against 28 clinical isolates of methicillin-resistant Staphylococcus
aureus. J Antimicrob Chemother 28:361–364.
47. Worrall EE, Litamoi JK, Seck BM, Ayelet G (2001) Xerovac: an ultra rapid method for
the dehydration and preservation of live attenuated Rinderpest and Peste des Petits
ruminants vaccines. Vaccine 19:834–839.
www.pnas.org/cgi/doi/10.1073/pnas.1206210109Zhang et al.