Synthesis of bioadhesive poly nano- and microparticles using an inverse emulsion polymerization method for the entrapment of hydrophilic drug candidates

(1)

Synthesis of bioadhesive poly(acrylic acid) nano- and microparticles using an inverse emulsion polymerization method for the

entrapment of hydrophilic drug candidates

Burkhard Kriwet, Elke Walter, Thomas Kissel *

Department of Pharmaceutics and Biopharmacy Philipps-University Marburg, Ketzerbach 63, D-35032 Marburg, Germany Received 15 November 1996; accepted 8 May 1998

Abstract

Bioadhesive latices of water-swollen poly(acrylic acid) nano-and microparticles were synthesized using an inverse (W/ O)

TM TM

emulsion polymerization method. They are stabilized by a co-emulsifier system consisting of Span 80 and Tween 80 dispersed in aliphatic hydrocarbons. The initial polymerization medium contains emulsion droplets and inverse micelles which solubilize a part of the monomer solution. The polymerization is initiated by free radicals, and particle dispersions with a narrow size distribution are obtained. The particle size is dependent on the type of radical initiator used. With water-soluble initiators, for example ammonium persulfate, microparticles were obtained in the size range of 1 to 10mm indicating that these microparticles originate from the emulsion droplets since the droplet sizes of the W/ O emulsion show similar distribution. When lipophilic radical initiators, such as azobis-isobutyronitrile, are used, almost exclusively nanoparticles are generated with diameters in the range of 80 to 150 nm, due to the limited solubility of oligomeric poly(acrylic acid) chains in the lipophilic continuous phase. These poly(acrylic acid) micro- and nanoparticles yielded excellent bioadhesive properties in an in-vitro assay and may, therefore, be suitable for the encapsulation of peptides and other hydrophilic drugs. 1998 Elsevier Science B.V. All rights reserved.

Keywords: Poly(acrylic acid); Nanoparticles; Inverse emulsion polymerization; Inverse micelles; Bioadhesive drug delivery

1. Introduction testinal tract an enhancement of drug penetration might be achieved. An important factor for bioadhe- Bioadhesive drug delivery systems have received sion is the particle size of the drug delivery system.

considerable interest in recent years because they Bioadhesive properties were optimized by a reduc- may provide an attractive approach to overcome tion of the size of the microparticulates and these some of the bioavailability problems associated with improvements were attributed to several factors, such peroral peptide delivery [1]. By increasing the resi- as an increase of the adhesive force [2], or a dence time at the site of absorption in the gastroin- prolongation of the gastrointestinal transit time [3], leading to a higher bioavailability of drugs [4]. For these reasons our aim was to produce a bioadhesive

*Corresponding author. Tel.:149 6421 285881; fax:149 6421 drug delivery system with a very small particle size,

287016; e-mail: [email protected] preferably in the lower micrometer or nanometer

P I I : S 0 1 6 8 - 3 6 5 9 ( 9 8 ) 0 0 0 7 8 - 9

(2)

range, by using poly(acrylic acid) (PAA) as matrix Alternatively, co-surfactant systems containing

TM TM

material [5]. Span 80 with a low HLB (4.3 [11]) and Tween

Two polymerization methods that would allow the 80 (polyethyleneglycol-sorbitan monooleate) with a incorporation of hydrophilic drug candidates seem to higher HLB (15 [12]) can be used to optimize the be suitable for obtaining poly(acrylic acid) mi- emulsion stability. The mechanism of the inverse croparticles (PAA-MP) in this size range. For one, emulsion polymerization in aliphatic hydrocarbons is the precipitation polymerization in an organic solvent still unknown and it is unclear whether emulsion or the inverse emulsion polymerization, which was droplets or inverse micelles are the sites of poly- proposed by Vanderhoff et al. [6] in the early 60’s. merization. In a similar inverse emulsion polymeri- An aqueous solution of the corresponding hydro- zation system, containing liquid paraffin and acryl- philic monomer-such as acrylic acid in this case-is amide no micelles could be detected [13]. Neverthe- dispersed in a continuous lipophilic medium using less, inverse micelles were postulated to be the site surfactants, which promote the formation of a water- of polymerization, suggesting that formation of in-oil (W/ O) emulsion. The polymerization is inverse micelles would be a prerequisite for initiated by free radicals. Using a lipophilic initiator, nanoparticle formation.

the polymerization will start in the continuous or- The study consists of two parts. First we analyzed ganic phase analogus to the conventional emulsion the structural properties of the emulsifier mixtures in polymerization, where a lipophilic monomer is dis- hexane and those of the emulsions, formed from persed in water with surfactants, and the polymeri- partially neutralized acrylic acid in hexane. In the zation is started by water-soluble initiators. At the second part we used electron microscopy and laser end of the polymerization, a dispersion is produced light scattering to characterize the particle disper- which consists of water-swollen particles in an sions obtained after polymerization. The bioadhesive organic solution. The removal of the reaction heat, properties of the polymerized particles were also the low viscosity of the dispersion and the high examined.

molecular weight of the polymer is advantageous [7].

The main point for the use as a drug delivery system is the direct production of microparticulate disper-

2. Materials and methods sions. Additionally, the properties of the polymer can

be easily varied by changing monomer composition

2.1. Materials and concentration. Vanderhoff [6] listed several hy-

drophilic monomers, but the main interest was

Acrylic acid, the cross-linker polyethyleneglycol focused on acrylamide in the recent years [8,9].

(400)-dimethacrylate and the lipophilic initiator azo- Attempts to produce microparticulate systems by

bis-isobutyronitrile (AIBN) were gifts of Roehm an inverse emulsion polymerization of acrylic acid

GmbH (D- Darmstadt). Acrylic acid contained 200

have failed [10]. It was not possible to obtain _TM

ppm hydrochinonmethylether as stabilizer. Span

particles with sizes ,40 mm, despite vigorous stir- _TM

80 (sorbitan monooleate) and Tween 80 (poly- ring of the emulsion during polymerization. Since

ethyleneglycol-sorbitan monooleate) were purchased the droplet sizes of the emulsion determine the

from Atlas-ICI (D-Essen). All other materials of particle size of the PAA-MP dispersion, this phenom-

analytical quality were obtained from E. Merck (D- enon can be attributed to a lack of suitable emul-

Darmstadt).

sifiers in the polymerization mixture. Trijasson et al.

[10] used SpanTM 80 (sorbitan monooleate) as emul-

sifier and noted that the emulsion stability is a 2.2. Polymerization critical factor, when changes of the electrolyte

concentrations occur as a function of the degree of The continuous lipophilic phase consisted of 68%

neutralization. Therefore, an adaptation of the hydro- (w / w) liquid paraffin and 2% emulsifiers. Usually a

TM TM

philic–lipophilic balance (HLB) of the emulsifier 75: 25 ratio of Span 80: Tween 80 was used.

system is necessary to obtain stable emulsions. For the aqueous phased acrylic acid (10%), sodium

(3)

hydroxide (5%) and water (15%) were mixed, 2.4. Solubilization of acrylic acid solution yielding a W/ O emulsion with 70% lipophilic and

30% hydrophilic phase. The aqueous phase con- A 1 ml volume of sodium acrylate solution (38%

tained both acrylic acid and sodium acrylate, the w / w) was added to 3 ml of the emulsifier solutions degree of neutralization was 90%. The polymeri- in hexane and sonicated twice for 30 s. After two h zation was carried out in a water-jacketed glass emulsion droplets and micelles were separated and vessel by gradually adding the aqueous phase to the 1.0 ml of the clear supernatant was removed. The oil phase, containing the dissolved emulsifiers, under concentration of water in the micellar system was magnetic stirring. The emulsion was heated to 508C examined by the Karl–Fischer- method.

and purged with argon to remove residual oxygen.

By injecting 0.125 % (w / w) of the oil- soluble

2.5. Particle size characterization of poly(acrylic lipophilic initiator AIBN or the water-soluble

acid) nano- and microparticles initiator ammonium persulfate (NH ) S O_{4 2} ₂ ₈ (APS),

the polymerization reaction was started. The tem-

Dispersion of PAA nanoparticles were diluted with perature was maintained at 508C for four h. After

a solution (2%) of the emulsifiers in hexane. The cooling to room temperature, the particles were

size distribution of the nanoparticles was investigated isolated by centrifugation (5000 rpm for 30 min,

by photon correlation spectroscopy (Zetasizer 4) Sorvall RC-5B, Du Pont de Nemours, D-Bad Hom-

equipped with a 5 mW helium neon laser operating burg), washed several times with hexane, and then

633 nm. As refractive index of the microparticles freeze- dried to remove residual solvents and water.

1.48 was chosen, the index of hexane was deter- The PAA-MP were stored in a desiccator at room

mined with a refractometer, the viscosity with a temperature.

capillary viscometer. The time dependent auto correlation function of the scattered intensity was derived 2.3. Characterization of the W/O micelles in

using a 64 channel correlator. The multimodal hexane by dynamic laser light scattering

exponential sampling method of the computer pro-

TM TM

gram Zetasizer Size Mode 1.1 was used to calculate Mixtures of Span 80 and Tween 80 with

the mean of the volume distribution.

different ratios were dissolved in hexane at a con-

21 The size of PAA-MP was determined by laser light

centration of 2.0% (m V ). The sizes of the inverse

diffraction (Mastersizer X, Malvern, D-Herrsching).

micelles were determined by photon correlation

The presentation 2JFE of the computer program spectroscopy (procedure see below). Since the scat-

Mastersizer X 1.1 was used to calculate the mean of tering intensity was low, a Autosizer LoC (Malvern,

the volume distribution.

D-Herrsching) was used, equipped with a 35 mW helium neon laser operating at 633 nm (Spectra

Physics, D-Darmstadt). For a ratio SpanTM 80: 2.6. Transmission electron microscopy TweenTM 80 of 75:25, the emulsifier concentration

was varied from 0.1 to 3.0% and the micelle size The polymer dispersions were attached between

determined. two gold supports. They were rapidly frozen in the

Water-in-oil emulsions, containing 1 ml of sodium liquid phase of melting nitrogen. Freeze fracture was acrylate solution (38%) and 5 ml of the emulsifier carried out at 21008C in a Balzers BAF 400 solutions in hexane were sonicated twice for 30 s in apparatus (Balzers, D-Wiesbaden). Replicas were an ultrasounic bath (Branson 2220, Branson, USA- produced by evaporation from a platinum–carbon Shelton CT). The stability of the emulsion was source, the specimens were shadowed at 458C and examined optically for two h and the particle size in the replicas stabilized with a further carbon layer.

the supernatant was measured with the Zetasizer 4 After cleaning in methanol–chloroform, replicas (Malvern, D-Herrsching). The size of the inverse were washed with distilled water, mounted on copper micelles was also investigated as a function of the grids and observed with a Zeiss transmission electron sodium acrylate concentration. microscope EM 109 (Zeiss, D-Oberkochen).

(4)

2.7. Scanning electron microscopy allow a better retention in the gastrointestinal tract and might hence provide higher local drug con- The microparticles were dried in a vacuum centrations. To achieve these goals the polymeri- chamber to remove residual water, sputter-coated zation medium was optimized with respect to surfac- with a gold layer (Sputter coater S150, Edwards tant composition and concentration. Fig. 1 shows the GmbH, D-Marburg), and viewed in a scanning hydrodynamic radii of inverse micelles dispersed in electron microscope (Hitachi, S-510, Hitachi Denshi, hexane containing different ratios of SpanTM 80 /

D-Rodgau). TweenTM 80. The overall emulsifier concentration

was kept constant at 2%. SpanTM80 was not soluble 2.8. In vitro bioadhesion in hexane, only when the ratio of TweenTM 80 exceeded 20% of the emulsifier mixture, the solution The bioadhesive properties of the microparticles became clear and inverse micelles could be detected.

were determined by measuring the force of detach- The size of the micelles increases with the proportion

TM TM

ment for microparticle-coated disks from rat intestine of Tween 80 from 7 to 13 nm. At a ratio Span in Tyrode’s solution at 378C. A modified tensile 80 / TweenTM 80 of 50:50 the resulting surfactant tester (Lauda TE1, Lauda Dr. R. Wobser GmbH and system became unstable. Therefore the range, where Co. KG, D-Lauda) was used [14]. The microparticles the two surfactants create inverse micelles, ranges were dispersed in propan-2-ol at a concentration of from 75 to 50% SpanTM80. For a given ratio of the 1% (m / v). A 20 ml sample of the dispersion were two emulsifiers the micelle size was independent of added on the surface of a metal disk (diameter 5 mm) the emulsifier concentration. At a concentration and dried in a heater at 808C for 15 min. Rats were range between 0.1 and 3.0% the co-emulsifier mix-

TM TM

sacrificed and the small intestine was removed. 3 cm ture Span 80: Tween 80 (75:25) did not in- parts of gut were cut open, washed and mounted on a fluence the micelle sizes which were in the order of 6 platform in Tyrode’s solution (pH 7.4) at 378C. The nm and did not change significantly with the surfac- test disk, suspended from a force recorder, was tant concentration, as shown in Fig. 2.

applied to the mucosal surface. After 1 min of The addition of sodium acrylate solution yields contact time the platform was lowered at a velocity 5 water-in-oil emulsions. At high relative ratios of

21 TM

mm min while recording the required force of Span 80 (100–85% of the emulsifier mixture), the detachment. Polycarbophil, a high molecular weight emulsion was unstable and separated without stir- poly(acrylic acid), served as reference. The poly- ring, at ratios below 65% the emulsion droplets carbophil microparticles were prepared by a spray- coalesced. The mixture 75:25 of SpanTM 80 and drying technique [15] (Buechi 190 Mini Spray Drier,

Buechi Laboratoriumstechnik, Flawil, Switzerland).

3. Results

As outlined above the formation of PAA nanoparticles will depend critically on the composition of the polymerization medium. Since hydrophilic drug candidates such as peptides or proteins are the most attractive proposition for perorally administered bioadhesive delivery systems, we were primarily interested in a W/ O emulsion polymerization method, which would allow the incorporation

of these molecules in high yields during the poly- Fig. 1. Hydrodynamic radii of inverse W/ O micelles as a function

21 TM

merization process. Nanoparticles were preferred of the co-emulsifier composition (m V ) of Span 80 and

over pellets, because smaller particle size might TweenTM80 in hexane; Mean1S.D. (n55).

(5)

Fig. 2. Micelle size in dependency on the emulsifier concentration

TM TM Fig. 4. Micelle size in dependency on the concentration of sodium

of Span 80 / Tween 80 mixture (75:25) in hexane; Mean1 21 TM TM

acrylate solution, solubilized by 2% (m V ) Span 80 / Tween S.D. (n54–9).

80 (75: 25) in hexane; Mean1S.D. (n55).

TweenTM 80 formed a stable emulsion. In the almost independent of the monomer concentration.

continuous phase inverse micelles were present. The The concentrated sodium acrylate solution had a high size of the emulsion droplets is larger than that of the ionic strength, possibly changing the interfacial micelles in the pure emulsifier solution. This implies properties of emulsifier system.

that sodium acrylate solution had been solubilized by In the absence of TweenTM 80, the capacity of

TM TM TM

the micelles (Fig. 3). For a Span 80 / Tween 80 Span 80 to solubilize aqueous sodium acrylate in ratio of 75:25 the particle size increased from 23 to hexane was relatively small. TweenTM 80 increased 30 nm, when the concentration of the inner phase the micellar solubilization capacity of the surfactant was increased from 0.25 to 2% (Fig. 4). An increase system considerably. As shown in Fig. 5 approxi- of the micelle volume reflects a swelling of the mately 5% monomer solution is solubilized in hex-

TM TM

micelles by the monomer solution as it seems to be ane at a 50:50 ratio of Span 80 / Tween 80. The not proportional to the concentration of the inner

phase. A preferred size of 28 nm was found, which is

Fig. 5. Solubilization of sodium acrylate solution by emulsifier

TM TM 21

Fig. 3. Micelle size with emulsified sodium acrylate solution in mixtures of Span 80 / Tween 80, 2% (m V ) in hexane;

hexane; Mean1S.D. (n53). Mean1S.D. (n55).

(6)

particle size distribution of the emulsion droplets covered the 1–10mm range with an average volume size of about 5 mm, determined by laser light diffraction.

3.1. Polymerization products

The polymerization reached complete conversion after two h yielding a dispersion of swollen poly- (acrylic acid) particles in paraffin. Using the lipophilic initiator AIBN, the particles obtained were in the nanometer range. A typical particle size distribution determined by photon correlation spectros-

copy is given in Fig. 6 with an average diameter Fig. 7. SEM micrograph of nanoparticles (oil-soluble initiator).

between 80 and 150 nm. The SEM micrograph (Fig.

7) demonstrates that the nanoparticle size distribu-

tion is exceptionally narrow. The water-soluble two particle classes. Using AIBN as initiator, the initiator APS produced particles of larger sizes, transmission electron micrograph of the dispersion usually the average diameter was about 5mm. A size (Fig. 10) displays small particles in the range of distribution determined by laser light diffraction is 60–150 nm. Initiation of the polymerization by APS shown in Fig. 8, demonstrating an almost bimodal generates in the final latex the same nanoparticles, distribution with centers at 1 and 5mm. The surface yet in reduced number. The main fraction of the structure of these particles was analyzed by scanning volume distribution of the polymerization product electron microscopy (Fig. 9), showing the spherical consists of microparticles with a size of 1–5 mm structure without any pores. The freeze fracture of (Fig. 11).

latices of PAA particles confirmed the existence of

3.2. Bioadhesion

The bioadhesive force of the polymerized mi-

Fig. 6. Size distribution of poly(acrylic acid) nanoparticles (oil-

soluble initiator). Fig. 8. Size distribution of microparticles (water-soluble initiator).

(7)

croparticles was determined with rat intestine. The force of detachment of microparticles in Tyrode’s buffer solution is given in Table 1. Spray-dried microparticles of polycarbophil, which served as reference, have been reported to possess excellent bioadhesive properties [16]. The data show that the polymerized microparticles were as bioadhesive as polycarbophil. Neither the addition of a cross-linker nor the model substance lead to a significant change of the adhesive properties of the particles.

4. Discussion

Fig. 9. SEM micrograph of microparticles (water-soluble initiator).

Pharmaceutical applications of the inverse emulsion polymerization method for generating nanoparticles were not extensively reported. Ekman and Sjoeholm [17] incorporated proteins, such as car- boanhydrase, in nanoparticles of cross-linked polyacrylamide, reaching a particle size of 300 nm [18].

The enzyme retained its activity and was protected against heat and proteolytic enzymes. Similar results were obtained with polyacryldextrane [19]. The residual solvents chloroform and toluene, used in the polymerization process, raised some concern. Similar problems were encountered with the method of Birrenbach and Speiser [20], where high concentrations of surfactants and deactivation of drug candidates during polymerization of the nanoparti-

Fig. 10. TEM micrograph of nanoparticles latex after inverse

cles have limited the application to pharmaceutical

emulsion polymerization. Bar represents 170 nm.

problems.

The selection of suitable solvents for the inverse emulsion polymerization procedure is of critical importance, since residual aromatic hydrocarbons may cause toxicological problems and since aliphatic hydrocarbon solvents influence the reaction kinetics, the molecular weight and the particle size of the nanoparticle dispersion. This was illustrated in the case of acrylamide, where the size increased from 0.23 mm to 1.12 mm when the polymerization medium was changed from toluene to liquid paraffin [13].

One drawback for the polymerization of acrylic acid in a water-in-oil emulsion was the instability of the emulsion itself. Due to the solubility of undis- sociated acrylic acid in organic solvents, which

Fig. 11. TEM micrograph of latex with micro- and nanoparticles

increases with the formation of dimers and oligomers

after inverse emulsion polymerization (water-soluble initiator). Bar

represents 720 nm. [21,22], acrylic acid has to be partially neutralized.

(8)

Table 1

Detachment forces obtained poly(acrylic acid) microparticles

Material Mean force6S.D. [mN cm22]

Polymerized PAA particles 81.67624.18 (n56)

Polymerized PAA particles with crosslinker 65.50626.73 (n58)

Polymerized PAA particles with sodium chloride 81.10611.89 (n55)

Spray-dried Polycarbophil particles 64.28612.64 (n56)

Otherwise the polymerization could not be carried the polyethyleneglycol chains, which are attached to the sorbitan ring requiring more space than the out in a controlled and reproducible manner. The

hydrophilic domain of SpanTM 80 molecules. The monomer solution used in our case consists of 90%

TM TM TM

(w / w) sodium acrylate and 10% acrylic acid. Span 75:25 mixture of Span 80 and Tween 80 80 is often used as emulsifier for the inverse stabilized the W/ O emulsion ideally, even when the emulsion polymerization of acrylamide, but when concentrated aqueous sodium acrylate solution is acrylic acid is chosen as monomer, the high ionic dispersed in the continuous hydrophobic paraffin strength of the aqueous phase destabilizes the emul- phase. This size of the inverse W/ O micelles in this sion, causing a phase separation. Hence, we attempt- system were found to be in the range of 25–30 nm.

ed to stabilize the W/ O emulsion by a suitable These two conditions, emulsion stability and forma- co-emulsifier system in pharmaceutically acceptable tion of inverse W/ O micelles, are necessary to aliphatic hydrocarbons. generate poly(acrylic acid) particles in the nanometer

One remarkable property of surfactants in nonpo- range.

lar solvents is their small aggregation number com- The inverse emulsion polymerization of sodium pared to aqueous solutions. Solvents that serve as acrylate yields latices of neutralized poly(acrylic hydrogen bond donors or acceptors, for example acid) particles dispersed in liquid paraffin. Two size dioxane or ethylacetate, may completely suppress the fractions can be differentiated, those in the lower association [23]. The determination of the critical micrometer range and secondly nanoparticles in the micelle concentration is notoriously difficult in hy- 100–200 nm range. The main factor, influencing the drophobic solvents and several techniques are only particle size under the conditions described above, applicable in aqueous systems [24]. The micelle seems to be the solubility of the radical initiator in formation is driven on the one hand by the dipole– the continuous phase. Using a lipophilic initiator, dipole interactions between the polar groups of the such as AIBN, nanoparticles are formed with a surfactant molecules [25] and on the other hand on narrow particle size distribution. As shown both by the repulsion between the nonpolar solvent and the scanning and transmission electron microscopy, hydrophilic regions of the emulsifiers. In the case of larger particles were rarely found. This effect may be nonionic surfactants containing polyethyleneglycol attributed to the possible polymerization mechanism:

chains, the repulsive force is reduced in comparison The lipophilic initiator decomposes into radicals to ionic substances, so that in alcohols no micelles within the continuous lipophilic paraffin phase of the are formed [26]. Kitahara reported that these W/ O emulsion. The radicals generated react with nonionic surfactants do not produce micelles in monomer molecules of acrylic acid present in the solvents like benzene and cyclohexane unless traces paraffin phase according to their equilibrium dis- of water are added [27]. tribution. Subsequently the inverse micelles, which Even in the absence of water our co-emulsifier are more numerous and have a greater contact

TM TM

system consisting of Span 80 and Tween 80 surface than the emulsion droplets, become the promotes the formation of W/ O micelles in hexane, reaction sites. The mechanism may be driven by the as shown by photon correlation spectroscopy. The concentration gradient, monomer being supplied micelle size is approximately 6 nm and increases in from emulsion droplets and other micelles either by proportion to the TweenTM 80 ratio. Possibly due to diffusion or by collisions with the growing micelles.

(9)

Fig. 12. Mechanism of the inverse emulsion polymerization (oil-soluble initiator).

At the endpoint of the reaction nanoparticles are stable W/ O emulsion, but rather a quasi- emulsion.

formed, which are stabilized by the surfactants. Their Therefore, relatively broad particle size distributions diameter varies from 80 to 150 nm. Fig. 12 illustrates are observed and high energy input using the proposed mechanism for the formation of homogenizers is necessary to obtain PAA-MP in the nanoparticles by an inverse W/ O emulsion poly- 40mm range. These shortcomings are avoided, when merization method. Similar observations have been the polymerization medium is optimized.

made by Baade et al. [13] studying polymerization of In summary, we succeeded in preparing polyacrylamide in a hydrophobic medium. (acrylic acid) nanoparticles by an inverse W/ O When a water-soluble initiator is used, the number emulsion polymerization technique, using a lipo- of microparticles per volume is much higher, but philic initiator. An inverse W/ O emulsion polymeri- also some nanoparticles can be detected. The poly- zation has been reported for acrylamide and merization is started mainly in the emulsion droplets, methacrylic acid [12,13], using similar conditions.

which contain much more of the aqueous phase than PAA nanoparticles of 100–200 nm size have not the micelles, by the radicals formed from the been described before and may have potential as initiator. Since the droplets are the reaction sites in bioadhesive carrier systems for the peroral delivery this system, the particle diameter almost equals that of peptides and proteins. The extend of nanoparticle of the emulsion droplets. After polymerization the uptake in the gastrointestinal tract was shown to be particles are obtained as a dispersion of water-swol- clearly size-dependent and therefore, investigation of len PAA-MP in paraffin. After removing the continu- rate and extend of PAA-MP uptake would be of ous phase by centrifugation and washing several general interest. The method for the preparation of times with hexane, the water entrapped by the these nanoparticles would lend itself to the encapsu- microparticles is removed by freeze-drying. The lation of hydrophilic drug candidates, since the micrographs obtained with scanning electron micro- external phase of the emulsion systems consists of scopy display no pores or changes of the spherical paraffin. The bioadhesive properties of the polymer- shape afterwards. In accordance with the transmis- ized PAA particles are very promising because they sion electron micrographs it can be postulated that exhibit adhesive forces comparable to those of the surface of the particles is covered by an emul- polycarbophil.

sifier layer. Trijasson et al. [10] have obtained PAA- MP with sizes of at least 40 mm, when they

polymerized PAA in an inverse W/ O emulsion Acknowledgements system using SpanTM 80 as surfactant. Their poly-

merization medium does not allow the formation of a Support of this investigation by Roehm GmbH

(10)

[13] W. Baade, K.H. Reichert, Polymerization of acrylamide in

(D-Darmstadt, Germany) is gratefully acknowl-

disperision with paraffinic and aromatic liquids as oil phase,

edged.

Macromol. Chem. Rapid. Commun. 7 (1986) 235–241.

[14] J.D. Smart, An in vitro assessment of some mucosa-adhesive dosage forms, Int. J. Pharm. 73 (1991) 69–74.

References [15] P. Vidgren, M. Vidgren, J. Arppe, et al., In vitro evaluation of spray-dried mucoadhesive microspheres for nasal administra- tion, Drug Dev. Ind. Pharm. 18 (1992) 581–597.

[1] M.R. Jimenez-Castellanos, H. Zia, C.T. Rhodes, Mucoadhe-

[16] P. Vidgren, M. Vidgren, P. Vainio, et al., Double-labeling sive drug delivery systems, Drug. Dev. Ind. Pharm. 19 (1983)

technique in the evaluation of nasal mucoadhesion of di- 143–194.

sodium-cromoglycate microspheres, Int. J. Pharm. 73 (1991) [2] D. Chickering, J. Jacob, G. Panol, et al., A tensile technique

131–136.

to evaluate the interaction of bioadhesive microspheres with

[17] B. Ekman, I. Sjoeholm, Improved stability of protein im- intestinal mucosa, Proc. Intn. Symp. Contr. Rel. Bioact.

mobilized in microparticles prepared by a modified emulsion Mater. 19 (1992) 88–89.

polymerization technique, J. Pharm. Sci 67 (1978) 693–696.

[3] D. Harris, J.T. Fell, H. Sharma, et al., Studies on potential

[18] I. Sjoeholm, P. Edman, Acrylic microspheres in vivo: I.

bioadhesive systems for oral drug delivery, S.T.P. Pharma 5

Distribution and elimination of polyacrylamide microparti- (1989) 852–856.

cles after intravenous and intraperitoneal injection in mouse [4] C.M. Lehr, J.A. Bouwstra, W. Kok, et al., Effects of the

and rat, J. Pharmacol. Exp. Ther. 211 (1979) 656–662.

mucoadhesive polymer polycarbophil on the intestinal ab-

[19] P. Edman, B. Ekman, I. Sjoeholm, Immobilization of pro- sorption of a peptide drug in the rat, J. Pharm. Pharmacol. 44

teins in microspheres of biodegradable polyacryldaxtran, J.

(1992) 402–407.

Pharm. Sci 69 (1980) 838–843.

[5] J.D. Smart, I.W. Kellaway, H.E.C. Worthington, An in vitro

[20] G. Birrenbach, P.P. Speiser, Polymerized micelles and their investigation of mucosa-adhesive materials for use in con-

use as adjuvants in immunology, J. Pharm. Sci. 65 (1976) trolled drug delivery, J. Pharm. Pharmacol. 36 (1984) 295–

1763–1766.

299. [21] A. Simon, M. Muecklich, D. Kunath, et al., Ueber raman-und

[6] J.W. Vanderhoff, E.B. Branford, H.L. Tarkowski, et al.,

ultrarotspektroskopische Untersuchungen an Hoch- Inverse emulsion polymerization, Adv. Chem. Ser. 34 (1962)

polymeren, J. Polymer Sci. 30 (1958) 201–226.

32–51.

[22] A. Chapiro, J. Dulieu, Influence of solvents on the molecular [7] F.L. Buchholz, Polyacrylamides and Poly(acrylic acid). in:

associations and on the radiation initiated polymerization of Ullmann’s Encyclopedia of Industrial Chemistry, Verlag

acrylic acid, Europ. Polym. J. 13 (1977) 563–577.

Chemie, Weinheim, 5th Edition, 1992, A21, pp. 143–156.

[23] H.F. Eicke, H. Christen, On the stability of micelles in apolar [8] M.V. Dimonie, C.M. Boghina, N.N. Marinescu, et al., Inverse

media, J. Colloid Interf. Sci. 46 (1974) 417–427.

suspension polymerization of acrylamide, Eur. Polym. J. 18

[24] D. Myers, Surfactant Science and Technology - 2nd Edition, (1982) 639–645.

VCH Publishers, New York, NY, 1992.

[9] F. Candau, Y.S. Leong, G. Pouyet, et al., Inverse microemul-

[25] K.L. Mittal, Micellisation, Solubilization, and Microemul- sion polymersization of acrylamide: characterization of the

sions, Plenum Press, New York, NY, 1977.

water-in-oil microemulsions and the final microlatices, J.

[26] A. Ray, Solvophobic interactions and micelle formation in Colloid Interf. Sci. 101 (1984) 167–183.

structure forming nonaqueous solvents, Nature 231 (1971) [10] P. Trijasson, T. Pith, M. Lambla, Hydrophilic polyelectrolyte

313–315.

gels by inverse suspension polymerization, Macromol.

[27] A. Kitahara, Solubility behaviour of polyoxyethylene Chem. Macromol. Symp. 35–36 (1990) 141–169.

nonylphenyl ethers in cyclohexane and the effect of water by [11] H.P. Fiedler, Lexikon der Hilfsstoffe fuer Pharmazie, Kos-

a light scattering method, J. Phys. Chem. 69 (1965) 2788–

metik und angrenzende Gebiete. Editio Cantor Verlag, Au-

2791.

lendorf, 1981.

[12] C. Pichot, C. Graillat, V. Glukhikh, Some aspects of inverse emulsion copolymerization of acrylamide with methacrylic acid, Macromol. Chem. Suppl. 10–11 (1985) 199–214.