RELEASE OF DNA FROM CRYOGEL PVA-DNA
MEMBRANES
A. J. M. Valente1*, S. M. A. Cruz1, C. Morán1, D. B. Murtinho1, M. G. Miguel1, B. Lindman1,2
1*
2
Departamento de Química, Universidade de Coimbra, 3004-535 Coimbra, Portugal– [email protected]
Physical Chemistry 1, Centre for Chemical and Chemical Engineering, University of Lund, S-221 00 Lund, Sweden
Poly(vinyl alcohol) (PVA) hydrogels have been used for numerous biomedical and pharmaceutical
applications, as a consequence of their non-toxic, non-carcinogenic and bioadhesive properties. In this
communication the effect of different factors, such as ionic strength, symmetric and asymmetric electrolytes,
surfactants, pH (at physiological values: 2.0, 5.0 and 7.0) and temperature (ranging from 25 ºC to 50 ºC) on
the distribution coefficients (α) and desorption kinetics (kR) of DNA from PVA-DNA blend gel matrices (of
a sheet shape), will be presented and discussed. The release kinetic constant and the partition coefficient of
DNA are sensitive to the surrounding matrix media (e.g., kR ranges from 1.3×10-6 to 4.7×10-5 s-1, and α from
0.12 to 0.87, for 10 mM NaI and 50 mM NaBr, respectively). The analysis of the temperature dependence on
kR shows that the activation energy for the DNA desorption to an aqueous solution is equal to 22.9 kJ/mol.
These results constitute a step forward towards the design of controlled DNA release PVA-based devices.
Keywords: DNA, PVA, hydrogel, controlled release, kinetics.
Introduction
Hydrogels are cross-linked polymeric networks that can be swollen in water, buffered or physiological
solutions. Hydrogels show a high water content, soft and rubbery consistency and low interfacial
tension with water or biological fluids [1]. The ability of molecules of different size to diffuse into
(drug loading), and out (release drug) of hydrogels, allows the use of hydrogels as delivery systems
[2,3].
In recent years, considerable work has been done in the development of cross-linked polymeric
networks which are sensitive to their surrounding physiological environment and therefore will be
desirable systems for site-specific drug-delivery [4-9].
One of the most well known hydrogels is poly(vinyl alcohol) (PVA). PVA and its copolymers have
been widely employed in controlled drug release systems [10]. PVA is hydrophilic and easily swells
upon hydration [11]. Furthermore, PVA is non-toxic, non-carcinogenic, shows bioadhesive
characteristics and is easily processed [12]. These properties make it ideal for biomedical uses,
especially drug delivery.
Recently we have reported the encapsulation of DNA into PVA hydrogels, obtained by a technique of
repeated freezing and thawing [13]. The obtained cryogels were chemically and physically
charaterized, and show a good mechanical resistance and a white and opaque appearance due to a
heterogeneous porous structure. Furthermore, the encapsulated DNA molecules can be compacted or
extended in the PVA matrix by tailoring the crystallinity degree of the PVA network.
In this communication the effect of different factors, such as ionic strength, symmetric and asymmetric
electrolytes, cationic and di-cationic surfactants and temperature (ranging from 25 ºC to 50 ºC) on the
distribution coefficients (α) and desorption kinetics (kR) of DNA from PVA-DNA blend gel matrices
(of a sheet shape), will be presented and discussed. The kinetics of release will be evaluated by a
reversible first-order kinetic law equation, developed by Muniz et al. [14] and based on the assumption
that the release of a solute from a hydrogel is treated as a partition phenomenon. These results
constitute a step forward towards the design of controlled DNA release PVA-based devices.
Experimental
Reagents and materials
Poly(vinyl alcohol) (PVA) (MW 72000, degree of polymerization ~1600, degree of hydrolysis 97.599.5 mol%) was supplied from Fluka. Double stranded (ds) DNA (sodium salt from salmon testes)
was purchased by Sigma. Sodium bromide (Merck), sodium iodide (Merck), sodium nitrate (Riedel-de
Haën), sodium chloride (Riedel-de Haën), calcium nitrate (Riedel-de Haën) and aluminium nitrate
(Fluka), all of pro analysis grade, were used as received.
Surfactant synthesis
The alkyldibromide, 1,10-dibromodecane or 1,12-dibromododecane, (10 mmol) was dissolved in 2025 mL dry ethanol and 200 mmol of amine (triethylamine or an ethanolic solution of trimethylamine,
31-35%) were added. The reaction mixture was refluxed until the alkylbromide was consumed, usually
48 h, as monitored by TLC. The solvent and the excess amine were evaporated under reduced pressure
and the residue was crystallized several times from the appropriate solvent or solvent mixtures.
Decane-1,10-bis(trimethylammonium bromide), C10Me6: Crystallized in ethanol, 90% yield.
Dodecane-1,12-bis(trimethylammonium bromide), C12Me6: Crystallized in ethanol, 93% yield.
Decane-1,10-bis(triethylammonium bromide), C10Et6: Crystallized in CH2Cl2/acetone, 84% yield.
Dodecane-1,12-bis(triethylammonium bromide), C12Et6: Crystallized in CH2Cl2/ethyl acetate, 85%
yield.
All aqueous solutions were prepared by using Millipore-Q water.
Preparation of PVA-DNA gel matrices
A PVA solution of 14 wt% concentration was prepared by dissolving the appropriate amount of PVA
into distilled water at 80 ºC under continuous stirring for three hours. An accurately weighed amount
Anais do 10o Congresso Brasileiro de Polímeros – Foz do Iguaçu, PR – Outubro/2009
of ds DNA (ca. 1 % w/w), using a Scaltec SBC22 balance with a resolution of ± 0.01 mg, is added to 1
g of PVA solution and mixed, under continuously stirring, during 4 hours. After that, the solution was
casted in 2 mm diameter cylinder flasks and submitted to freezing for 12 hours at -20 ºC and, after
that, thawed for 12 hours at +25 ºC. The cycles of freezing and thawing were repeated three times. The
incorporation of DNA in the PVA gel matrix was checked by fluorescence microscopy, FM, (Figure
1), using an Olympus BX51M Microscope and acridine orange as a fluorescent probe.
Figure 1 - Fluorescence microscopy images of (a) PVA and (b) PVA-DNA blend matrix. Acridine orange has been used as DNA fluorescent probe.
Desorption kinetics of DNA
DNA desorption kinetics were performed by immersing a PVA-DNA gel membrane sample (as it was
synthesised) in 100 mL of liquid (water, salt or surfactant solution). Experiments have been carried out
at 20 ºC. The effect of temperature on the release kinetics of DNA has been studied at 20 ºC to 40 ºC
temperature range. In all experiments, temperature was kept constant by using a thermostatic bath
Multistirrer 6 from Velp Scientifica. During DNA release experiments gel-containing solutions were
stirred at ca. 220 rpm. At defined intervals, aliquots of the supernatant were collected. The secondary
structure of DNA desorbed from PVA-DNA gel was monitoried by FM and no evidences of ss DNA
have been found (Figure 2). Consequently, the amount of substance of DNA released from polymeric
matrices, nr,t, to the supernatant solution was determined by UV-vis spectrophotometry, by measuring
the absorbance at 260 nm with a Jasco V-530 spectrophotometer, and using the extinction coefficient
of 6600 M-1 cm-1 [15,16].
Figure 2 - Fluorescence microscopy photograph of an aqueous supernatant solution of water in contact with PVA-DNA gel membrane.
Anais do 10o Congresso Brasileiro de Polímeros – Foz do Iguaçu, PR – Outubro/2009
Mathematical Model
The quantification of the release kinetics of DNA has been done taking into account the initial and
border conditions of a non-steady state diffusion transport occurring in a stirred solution of limited
volume [17]. Recently, Muniz et al. [14] developed a set of equations to model the release of dyes
from poly(N-isopropylacrylamide)–polyacrylamide taking into account the previous mentioned
conditions, and treating the release of a solute from the gel as a partition phenomena, where the
partitioning of such solute occurs between a solvent (or solution) phase and the hydrogel. This
behaviour can be quantified through the distribution coefficient, α, that characterizes the physical
chemical affinity of the solute for both phases
α = nR,max / n0
(1)
where nR,max is the maximum amount of substance release from the gel and n0 is the initial number of
moles of the loaded solute into gel. When α>0, the diffusion of the solute between the hydrogel and
the solution phase occurs, and the process of release and absorption of solute occurs simultaneously.
Assuming a first order kinetic process, changes of solute concentration in solution at a given time t,
can be expressed as
dC R ,t
dt
= k R (C 0 −C R ,t ) − k A (C R , 0 − C A,t )
(2)
where CR,0 and CR,t are the concentration of the release solute at t=0 and at time t, C0 is the initial
concentration of DNA in the gel and CA,t is the concentration of absorbed solute at specific time t. kR
and kA are the constants of the released and absorption rates, respectively.
From the kinetic law equation (eq. 2) and taking into account considerations reported elsewhere [14],
the release kinetics of DNA can be treated by using the following equation
FR = FR ,max (1 − e
− ( k R / FR , max ) t
)
(3)
where (FR = CR,t/C0) and (FR,max = CR,max/C0), and CR,max is the maximum concentration of solute in
solution released from the gel. By fitting eq. (3) to experimental data (e.g., Figure 3) it is possible to
calculate the following parameters: FR,max and kR , and so to characterise the release not only in terms
of kinetics as well as the partition coefficient. It is worthwhile to note that the release of DNA, in all
systems mentioned in this communication, follows a first-order kinetic law. The experimental data
were analyzed using a nonlinear least-squares fitting procedure (Origin 8.0), using a 95 % confidence
level. The uncertainty in the fit of Eq. (3) to the data is in general smaller than 5%.
Anais do 10o Congresso Brasileiro de Polímeros – Foz do Iguaçu, PR – Outubro/2009
0.2
0.1
0.0
0
100
200
300
400
500
FR x 10
2
0.2
0.1
0.0
0
50
100
150
200
250
300
0.05
0.00
0
20
40
60
80
100
120
140
160
180
time / hours
Figure 3 - Desorption kinetics of DNA from PVA-DNA membranes to cationic surfactant solution: (□) C12TAB, (o) C12Me6, (∆)C12Et6. Solid lines
represent the fraction of release DNA predicted by Eq. (3).
Results and Discussion
Effect of symmetrical and unsymmetrical electrolytes
Figure 1 shows the effect of sodium salts on the distribution coefficient (α) and released rate constants
(kR) of DNA release from PVA-DNA gel membranes. It is possible to observe that comparing the
release of DNA to sodium salts, at a given concentration, there is a decrease of both α and kR in the
order: NaI > NaNO3 > NaCl > NaBr. A simple explanation for this behaviour comes from the capacity
of different salts to solubilise DNA, according to the Hofmeister serie [18,19]. Among these salts,
iodide solution has the strongest effect as water structure breaking, being the highest effective on the
solubilization of DNA. Using the same arguments, NaBr acts as the most kosmostropic salt on
solubilisation of DNA. Such behaviour shows a variation in the Hofmeister sequence [20]. A possible
hypothesis to justify that behaviour comes from the specific effect of bromide ions on the crystallinity
of PVA. Crystallinity of PVA cryogels decreases in the presence of bromide ions [13]. Consequently,
it can be suggested that in more amorphous gels, due to the higher mobility of the PVA chains, DNA
molecules show stronger self-interaction, generating globular structures and, thus, increasing the
retention of DNA inside the gel membrane [13].
It is also possible to conclude that DNA distribution coefficient and release rate constant decrease by
increasing the ionic strength. This can be due to the response of PVA gels to the presence of ionic
solutions; in the presence of unsymmetrical salts or increasing salt concentration PVA collapses [21]
and, consequently, the polymer volume fraction increases as well as obstacles for DNA release. This
Anais do 10o Congresso Brasileiro de Polímeros – Foz do Iguaçu, PR – Outubro/2009
may justify the decrease of distribution coefficients in the order: Al(NO3)3 > Ca(NO3)2 > NaNO3. A
further, and possibly concomitant, justification for this behaviour results from the affinity of these
cations to DNA bases, inducing the DNA compaction [22,23].
1.0
50
0.8
30
0.4
20
0.2
kR / (10 −8 s -1)
α
40
0.6
10
10
[ sa
NaI
NaNO3
NaCl
NaBr
20
30
lt]
/
40
mM
50
Ca(NO3)2
Al(NO3)3
10
20
[sa
lt]
/
30
mM
40
50
NaI
NaNO3
NaCl
NaBr
Ca(NO3)2
Al(NO3)3
Figure 4 - Distribution coeffcients, α, and rate constants (kR) for the release of DNA, from PVA-DNA membranes, to aqueous solutions of different
electrolytes at 20 ºC.
Effect of cationic surfactants
Cationic surfactants interact strongly with DNA. Those interactions depend on the surfactant
hydrophobic chain length, charge and headgroup [24,25]. Figure 5 shows the effect of those three
factors on the release kinetics and distribution coefficient of DNA from DNA-containing PVA gel
matrices. Several interesting comments can arise from the analysis of Figure 5. Comparing the effect
of cationic surfactants on the release of DNA, we can conclude that only a very small fraction of DNA
is released from gel (α=0.05) when this is immersed in C12TAB, whilst when immersed in C10TAB the
α value is 2.07. This behaviour is in agreement with that reported elsewhere [13] and can be explained
by the fact that strong interactions between cationic surfactants and DNA are dependent on both
electrostatic and hydrophobic interactions [26]. Distribution coefficients of DNA in dodecyl-based
surfactant systems increase by increasing the charge, but do not depend on the bolaform surfactant
headgroup [25]. Although in decyl-based surfactant systems α values are always higher than in C12
chain length surfactants, which can be explained by the decrease of hydrophobic role on the
interactions with DNA, there is a decrease of α not only with the charge density but also with
increasing hydrophobicity of bolaform surfactant headgroup. We can hypothesise that with a so short
alkyl chain the intramolecular charge repulsion can not be neglected [27] and, consequently, the
bolaform can not be seen as a strong electrolyte. The main conclusion is that the release of DNA to a
surfactant solution is quite sensitive to the balance between hydrophobic/hydrophilic and electrostatic
Anais do 10o Congresso Brasileiro de Polímeros – Foz do Iguaçu, PR – Outubro/2009
interactions. However, it should be noted that the rate constants increase in the same order kR(CnTAB)
< kR(CnMe6) < kR(CnEt6) and are higher for n=10.
10
6
-8
-1
kR / (10 s )
8
4
2
0
α
2
1
0
C10TAB
C10Me6
C10Et6
C12TAB
C12Me6
C12Et6
Figure 5 - Distribution coeffcients, α, and rate constants (kR) for the release of DNA, from PVA-DNA membranes, to aqueous solutions of different
electrolytes at 20 ºC.
Effect of Temperature
The effect of temperature on the release of DNA from PVA-DNA gels to unbuffered water was
evaluated at 20, 30, 37 and 40 ºC. Figure 6a, shows the temperature dependency of drug release. It can
be seen that the release is temperature-dependent, and α and kR increase by increasing temperature.
2.0
a)
b)
-15.2
24
1.6
-8
ln kR
α
kR / (10 s )
20
1.2
-15.4
-1
-15.6
16
0.8
-15.8
0.4
12
20
25
30
T / °C
35
40
0.0032
0.0033
0.0034
(1 / T) / K-1
Figure 6 – (a) Effect of temperature on the rate constants and distribution coefficients of DNA released from PVA-DNA
membranes. (b) Arrhenius treatment of the temperature dependence of the rate constant, kR.
Anais do 10o Congresso Brasileiro de Polímeros – Foz do Iguaçu, PR – Outubro/2009
The temperature dependence of the rate constant, kR, can be described by the Arrhenius relationship:
k R = k 0 e − ( Ea / RT ) )
(4)
where ko is the preexponential factor, Ea is the activation energy, R is the universal gas constant, and T
is the absolute temperature. One can determine the value of Ea from the slope of the linear relationship
between ln(k) and the reciprocal of absolute temperature as demonstrated in Figure 2b. It was found
that Ea = 21.2 (±0.8) kJ/mol.
Conclusion
The kinetics and equilibrium properties of the release of DNA from PVA-DNA cryogel membranes to
different solutions and temperatures have been studied. The first order kinetic law equation was found
to be an excellent model to describe the experimental release data. The kinetics of release and the
maximum amount of DNA desorbed are quite sensitive to the media where the PVA-DNA gel matrix
is immersed. The rate of DNA release from PVA-DNA gels increased with increasing temperature
with activation energy (Ea) of 21.2 kJ/mol.
Acknowledgements
We thank FCT (Project PTDC/QUI/67962/2006) for financial support.
References
1. N. A. Peppas, Hydrogels in Medicine and Pharmacy, Vols. I-III, CRC Press, Boca Raton, 1986.
2. B. Amsden Macromolecules 1998, 31, 8382.
3. C. S. Satish; K. P. Satish; H. G. Shivakumar Ind. J. Pharm. Sci. 2006, 68, 133.
4. C. L. Bayer; N. A. Peppas J. Control. Release 2008, 132, 216.
5. J. Leach; V. Schmidt Biomaterials 2005, 26, 125.
6. E. M. M. del Valle; M. A. Galan; R. G. Carbonell Ind. Eng. Chem. Research 2009, 48, 2475.
7. C. L. He; S. W. Kim; D. S. Lee J. Control. Release 2008, 127, 189.
8. J. K. Oh; R. Drumright; D. J. Siegwart; K. Matyjaszewski Prog. Polym. Sci. 2008, 33, 448.
9. L. Klouda; A. G. Mikos Eur. J. Pharm. Biopharm. 2008, 68, 34.
10. K. Juntanon; S. Niamlang; R. Rujiravanit; A. Sirivat Int. J. Pharm. 2008, 356, 1.
11. R. Morita; R. Honda; Y. Takahashi J. Control. Release 2000, 63, 297.
12. C. C. DeMerlis; D. R. Schoneker Food Chem. Toxicology 2003, 41, 319.
13. A. Papancea; A. J. M. Valente; M. G. Miguel; B. Lindman Langmuir 2008, 24, 273.
14. A. V. Reis, M. R. Guilherme, A. F. Rubira, E. C. Muniz J. Colloid Interf. Sci. 2007, 310, 128.
Anais do 10o Congresso Brasileiro de Polímeros – Foz do Iguaçu, PR – Outubro/2009
15. L. Beach, C. Schweitzer, J. C. Scaiano Org. Biomol. Chem. 2003, 1, 450.
16. M. Rosa, R. Dias, M.G. Miguel, B. Lindman Biomacromolecules 2005, 6, 2164.
17. J. Crank, The Mathematics of Diffusion, 2nd Ed., Oxford Sci. Publ., Oxford, 1975.
18. B. Lindman, S. Forsen, E. Forslind, J. Phys. Chem. 1968, 72, 2805.
19. Y. Marcus Chem. Rev. 2009, 109, 1346.
20. M. Fortini, D. Berti, P. Baglioni, B.W. Ninham Curr. Opin. Colloid Interface Sci. 2004, 9, 168.
21. S. Patachia, A.J.M. Valente, C. Baciu Eur. Polym. J. 2007, 43, 460.
22. E.V. Hackl, S.V. Kornilova, Y.P. Blagoi Int. J. Biol. Macromol. 2005, 35, 175.
23. D. Costa, M.G. Miguel, B. Lindman J. Phys. Chem. B, 2007, 111, 8444.
24. R.S. Dias, B. Lindman, M.G. Miguel J. Phys. Chem. B 2002, 106, 12608.
25. R.S. Dias, L.M. Magno, A.J.M. Valente, D. Das, P.K. Das, S. Maiti, M.G. Miguel, B. Lindman J.
Phys. Chem B 2008, 112, 14446.
26. M.C. Moran, M.G. Miguel, B. Lindman Biomacromolecules 2007, 8, 3886.
27. M. Nilsson, A.J.M. Valente, G. Olofsson, O. Soderman, M. Bonini J. Phys. Chem. B 2008, 112,
11310.
Anais do 10o Congresso Brasileiro de Polímeros – Foz do Iguaçu, PR – Outubro/2009