evaluation of biocompatibility of polypyrrole in vitro and in vivo

12
Evaluation of biocompatibility of polypyrrole in vitro and in vivo Xioadong Wang, 1 Xioasong Gu, 1 Chunwai Yuan, 2 Shujian Chen, 2 Peiyun Zhang, 1 Tianyi Zhang, 1 Jian Yao, 1 Fen Chen, 1 Gang Chen 1 1 Jiangsu Key Laboratory of Neuron Regeneration, Nantong Medical College, Nantong 226001, People’s Republic of China 2 Nanional Laboratory of Molecular and Biomolecular Electronics, Southeast University, Nanjing, People’s Republic of China Received 4 February 2003; revised 16 July 2003; accepted 8 August 2003 Abstract: In this study, the biocompatibility of the electri- cally conductive polymer polypyrrole (PPy) with nerve tis- sue was evaluated in vitro and in vivo. The extraction solu- tion of PPy powder, which was synthesized chemically, was tested for acute toxicity, subacute toxicity, pyretogen, quan- titative measure of cell viability, hemolysis, allergen, and micronuclei. The PPy membrane was synthesized electro- chemically on the indium tin oxide conductive borosilicate glass. The dorsal root ganglia from 1–3-day-old Sprague- Dawley rats were cultured above PPy membrane and ob- served by light or scanning electron microscopy. The PPy- silicone tube (PPy membrane on the inner surface of the silicone tube) also synthesized electrochemically was used to bridge across 10-mm sciatic nerve gap in rats. Twenty-four weeks after the operation to rats, the regenerated tissues were observed by electrophysiological and histological tech- niques. PPy extraction solution showed no evidence of acute and subacute toxicity, pyretogen, hemolysis, allergen, and mutagenesis, and the Schwann cells from the PPy extraction solution group showed better survival rate and proliferation rate as compared with the saline solution control group. The migration of the Schwann cells and the neurite extension from dorsal root ganglia on the surface of PPy membrane- coated glass was better than those of bare glass. There was only lightly inflammation during 6 months of the postop- eration, when the PPy-silicone tube bridged across the gap of the transected sciatic nerve. The regeneration of nerve tissue in the PPy-silicone tube was slightly better than that in the plain silicone tube by means of electrophysiological and histological examination. The results of this study indicate that PPy has a good biocompatibility with rat peripheral nerve tissue and that PPy might be a candidate material for bridging the peripheral nerve gap. © 2003 Wiley Periodicals, Inc. J Biomed Mater Res 68A: 411– 422, 2004 Key words: conducting polymer; polypyrrole; biomaterial; biocompatibility; nerve regeneration INTRODUCTION Damage to a peripheral nerve often results in a nerve gap, followed by poor functional recovery. The conventional treatment of such injury is by use of an autologous nerve graft. However, there are several disadvantages of this procedure, including loss of end-organ innervation by harvesting the autograft at a second operation site, mismatch of the injured nerve, and shortage of autologous nerve as graft material for extensive nerve injures. 1,2 An alternative treatment of such damage could be achieved by use of a synthetic guidance channel to bridge the nerve gap. Several new materials have been investigated for preparing a guid- ance channel to bridge the transected nerve gap. How- ever, it has been found that physical and chemical factors have effects on the bio-regenerative process. For example, applied electrical fields influenced the extension and direction of neurite outgrowth from neurons cultured in vitro, 3,4 and pulsed electromag- netic fields stimulated sciatic nerve regeneration in vivo. 5,6 In addition, the galvanotropic currents pro- duced by silicone channels fitted with electrode cuffs were shown to enhance the regeneration of the periph- eral nerve system in vivo. 7 Because electromagnetic fields have a positive influ- ence on nerve regeneration, it is important that the repairing material has electrical conductivity. The electrically conductive polymers are a class of materi- als that could be fabricated to generate either transient or static electrical charges because of their physico- Correspondence to: X. Wang; e-mail: [email protected] Contract grant sponsor: National Natural Science Foun- dation of China; contract grant number: 39425006 © 2003 Wiley Periodicals, Inc.

Upload: xioadong-wang

Post on 11-Jun-2016

222 views

Category:

Documents


4 download

TRANSCRIPT

Page 1: Evaluation of biocompatibility of polypyrrole in vitro and in vivo

Evaluation of biocompatibility of polypyrrole in vitro andin vivo

Xioadong Wang,1 Xioasong Gu,1 Chunwai Yuan,2 Shujian Chen,2 Peiyun Zhang,1 Tianyi Zhang,1 Jian Yao,1

Fen Chen,1 Gang Chen1

1Jiangsu Key Laboratory of Neuron Regeneration, Nantong Medical College, Nantong 226001,People’s Republic of China2Nanional Laboratory of Molecular and Biomolecular Electronics, Southeast University, Nanjing,People’s Republic of China

Received 4 February 2003; revised 16 July 2003; accepted 8 August 2003

Abstract: In this study, the biocompatibility of the electri-cally conductive polymer polypyrrole (PPy) with nerve tis-sue was evaluated in vitro and in vivo. The extraction solu-tion of PPy powder, which was synthesized chemically, wastested for acute toxicity, subacute toxicity, pyretogen, quan-titative measure of cell viability, hemolysis, allergen, andmicronuclei. The PPy membrane was synthesized electro-chemically on the indium tin oxide conductive borosilicateglass. The dorsal root ganglia from 1–3-day-old Sprague-Dawley rats were cultured above PPy membrane and ob-served by light or scanning electron microscopy. The PPy-silicone tube (PPy membrane on the inner surface of thesilicone tube) also synthesized electrochemically was used tobridge across 10-mm sciatic nerve gap in rats. Twenty-fourweeks after the operation to rats, the regenerated tissueswere observed by electrophysiological and histological tech-niques. PPy extraction solution showed no evidence of acuteand subacute toxicity, pyretogen, hemolysis, allergen, andmutagenesis, and the Schwann cells from the PPy extraction

solution group showed better survival rate and proliferationrate as compared with the saline solution control group. Themigration of the Schwann cells and the neurite extensionfrom dorsal root ganglia on the surface of PPy membrane-coated glass was better than those of bare glass. There wasonly lightly inflammation during 6 months of the postop-eration, when the PPy-silicone tube bridged across the gapof the transected sciatic nerve. The regeneration of nervetissue in the PPy-silicone tube was slightly better than that inthe plain silicone tube by means of electrophysiological andhistological examination. The results of this study indicatethat PPy has a good biocompatibility with rat peripheralnerve tissue and that PPy might be a candidate material forbridging the peripheral nerve gap. © 2003 Wiley Periodicals,Inc. J Biomed Mater Res 68A: 411–422, 2004

Key words: conducting polymer; polypyrrole; biomaterial;biocompatibility; nerve regeneration

INTRODUCTION

Damage to a peripheral nerve often results in anerve gap, followed by poor functional recovery. Theconventional treatment of such injury is by use of anautologous nerve graft. However, there are severaldisadvantages of this procedure, including loss ofend-organ innervation by harvesting the autograft at asecond operation site, mismatch of the injured nerve,and shortage of autologous nerve as graft material forextensive nerve injures.1,2 An alternative treatment ofsuch damage could be achieved by use of a syntheticguidance channel to bridge the nerve gap. Several new

materials have been investigated for preparing a guid-ance channel to bridge the transected nerve gap. How-ever, it has been found that physical and chemicalfactors have effects on the bio-regenerative process.For example, applied electrical fields influenced theextension and direction of neurite outgrowth fromneurons cultured in vitro,3,4 and pulsed electromag-netic fields stimulated sciatic nerve regeneration invivo.5,6 In addition, the galvanotropic currents pro-duced by silicone channels fitted with electrode cuffswere shown to enhance the regeneration of the periph-eral nerve system in vivo.7

Because electromagnetic fields have a positive influ-ence on nerve regeneration, it is important that therepairing material has electrical conductivity. Theelectrically conductive polymers are a class of materi-als that could be fabricated to generate either transientor static electrical charges because of their physico-

Correspondence to: X. Wang; e-mail: [email protected] grant sponsor: National Natural Science Foun-

dation of China; contract grant number: 39425006

© 2003 Wiley Periodicals, Inc.

Page 2: Evaluation of biocompatibility of polypyrrole in vitro and in vivo

chemical properties. Polypyrrole (PPy), an electricallyconductive polymer, has been chosen for a candidatematerial of nerve regeneration because of its inherentelectrical conductive properties, ease of preparation,and flexibility of surface characteristics. Based on this,the biocompatibility of PPy becomes an importantissue of interest and has been extensively studied inseveral in vitro and in vivo systems (including nerve).

Some studies were in vitro performed with endothe-lial cells, calvarial osteoblast, chick nerve tissue, chro-maffinoma cell line (PC-12 cell), or a neuro 2a neuro-blastoma cell,8–12 the last two of which are mainly notnormal mammalian nerve tissue but tumor cells. Theimplants of PPy to rat muscle, hypodermis tissues, orguinea pig brain, which have been reported in litera-ture,11–14 were in vivo examined over a shorter periodof time (2–4 weeks). In addition, to our knowledge,there seem to be few studies involved in the evalua-tion of the biocompatibility between PPy and periph-eral nerve, especially evaluation of practical applica-tion of PPy to peripheral nerve repair. Hence, itsbehaviors in vivo have not been fully studied.14

This study concentrates on coculture of neurons andSchwann cells of rat peripheral nerve with PPy in vitro,and on PPy bridging the mammalian peripheral nervegap in vivo within a long-term (6 months) period.Furthermore, this study was first designed to examinea more thorough series of toxicity using many differ-ent animal models because we think that a few indi-vidual toxicity tests12,15 are not enough for PPy bio-compatibility studies. The results of this studyindicate that PPy has a good biocompatibility with ratperipheral nerve tissue and that PPy membrane mightbe a candidate material for bridging peripheral nervegaps.

MATERIALS AND METHODS

Preparation of PPy

Pyrrole (yellow-brown as received; Aldrich ChemicalCompany, Milwaukee, WI) was vacuum-distilled at 129–131°C until becoming a colorless liquid, and then stored at4°C in dark until use.

One molar pyrrole and 1M ferric chloride solutions wereprepared, respectively. After 4.5 mL of 1M ferric chloridesolution was added to 2 mL of 1M pyrrole solution drop bydrop with agitation, the PPy was chemically polymerizedand grew in its powder size.

In this study, both the PPy membrane and the PPy-sili-cone tube were prepared by electropolymerization, the ex-perimental setup of which was composed of three parts: anelectrochemistry station, a constant flow pump (HL-2 type),and a flow cell. Before use, an indium tin oxide (ITO)-conductive borosilicate glass slide, used as a support for thePPy membrane, was cleaned by supersonic bath in hexane,

methanol, and methylene chloride for 5 min each; a siliconetube, the inner surface of which was used for the PPy-silicone tube preparation, was cleaned by supersonic bath inalcohol for 20 min.

To prepare the PPy membrane, the flow cell consisted ofan ITO-conductive borosilicate glass slide as a working elec-trode, platinum gauze as a counter electrode, and a satu-rated calomel electrode (SCE) as a reference electrode. Theelectrolyte solution containing 0.1M pyrrole (pyrrole mono-mer) and 0.1M potassium chloride was pumped through theflow cell by the constant flow pump. In this way, the PPymembrane was deposited on the surface of ITO glass at aconstant potential of 1 V (vs SCE).

To prepare the PPy-silicone tube, a silicone tube is used asa backstop of the template synthesis. The flow cell consistedof a silicone tube oriented vertically, the inner surface ofwhich had two stainless steel rings as a working electrodeand a counter electrode, respectively, and a SCE as a refer-ence electrode. The electrolyte solution containing 0.2M pyr-role and 0.1M potassium chloride was pumped through theflow cell by the constant flow pump. In this way, the PPyfilm was deposited on the inner surface of the silicone tubeat a constant potential of 10 V (vs SCE), forming a PPy-silicone tube.

All PPy preparations were rinsed in distilled water for 30min and sterilized by autoclave at 1.3 Mpa for 30 min, thendried at 37°C and stored in dry and aseptic conditions.

Surface morphology analysis

The surfaces of the PPy membrane and PPy-silicone tubewere observed under scanning electron microscopy (JES-T300 SEM, JEOL Inc., Japan), respectively, for identificationof their overall surface features.

Systemic toxicity tests

The experimental procedure of a series of toxicity testsfollowed the methods and prescribes on the evaluation ofbiomaterials by the International Organization for Standard-ization and the American Society for Testing and Materialsin the United States.16,17 All experiments were performed induplicate.

Preparation of extracts

The PPy powder (20 g) was placed in an extraction con-tainer with the addition of 200 mL of sterile saline solution,and stored in an incubator at 37°C for 72 h with agitation.The extraction solution was then filtered through a 0.22-�mfilter membrane into a dry sterile container.

Acute toxicity test

Ten male and 10 female ICR mice (weight 19–21 g) wererandomly divided into two groups: a treatment group and a

412 WANG ET AL.

Page 3: Evaluation of biocompatibility of polypyrrole in vitro and in vivo

control group, five male and five female mice for eachgroup. The mice of the treatment group were injected with50 mL/kg PPy extraction solution into the abdominal cavityonce, and the mice of the control group were injected with 50mL/kg saline solution into the abdominal cavity once. Allmice were fed with standard food, and monitored for theirfatality and adverse response for 72 h.

Subacute toxicity test

Ten male and 10 female ICR mice (weight 19–21 g) wererandomly divided into two groups: a treatment group and acontrol group, five male and five female mice for eachgroup. The mice of the treatment group were injected with10 mL/kg PPy extraction solution into the abdominal cavityonce every other day during 30 days, and the mice of thecontrol group were injected with 10 mL/kg saline solutioninto the abdominal cavity once every other day during 30days. All mice were fed with standard food, and monitoredfor their fatality and adverse response. At the end of 30 days,the mice were sacrificed, and their heart, lung, liver, spleen,and kidney were fixed in 10% formalin, embedded in par-affin, sectioned, and stained with hematoxylin and eosin.The normality of histological structure of these organs waschecked under light microscopy.

Pyretogenesis test

The body temperature of three healthy New Zealand rab-bits (female, weight approximately 2.2–2.8 kg) was mea-sured once every hour within 4 h. The range of the bodytemperature was approximately 38.5–39.2°C; the tempera-ture change of each rabbit was �0.4°C. Before testing, therabbits were fasted for 2 h, and their temperature was mea-sured once every 0.5 h within 2 h. The temperature changefor each rabbit was �0.2°C. The average value of the last twotemperatures measured for each rabbit was considered itsnormal temperature. The PPy extract at 38°C was injectedintravenously (10 mL/kg) into the rabbit, and its tempera-ture measured once every hour for three times. The value ofthe temperature change equals the difference of its temper-ature measured after injection and its normal temperaturefor each rabbit.

MTT test

The Schwann cells obtained from the sciatic nerve of 20newborn Sprague-Dawley rats were cultured in L15 (Sigma)medium supplemented with 15% fetal bovine serum (FBS)in a humidified incubator with 5% CO2 at 37°C. In all ex-periments, a 96-well microtiter plate with 0.1 mL of mediumper well (1 � 107 cells/mL) was incubated in 5% CO2 at 37°Cfor 48 h. Then the medium was removed from all wells, andwas replaced by 100 �L of PPy extraction solution and L15medium supplemented with 15% FBS (1:1), or 100 �L ofsaline solution and L15 medium supplemented with 15%FBS (1:1), or 100 �L of L15 medium supplemented with 15%FBS alone at 37°C for 24 h, followed by addition of 10 �L of

MTT solution (5 mg of MTT per milliliter of medium) to eachwell for the assay, and incubation of the plate at 37°C for 4 h.One hundred microliters of acid-isopropanol (containing0.04N HCl in isopropanol) was added to each well andmixed completely until the dark blue crystals dissolved. Toensure that all crystals had dissolved, the solution was keptat room temperature for an additional several minutes. Theplate was read by an Elx-800 Microelisa reader (Bio-TEKInc., Burlington, VT) at 540 nm.

Hemolysis test

A 12-mL blood sample was freshly collected from threefemale normal rabbits into an anticoagulin tube and gentlymixed. The pooled blood was diluted to 15 mL with salinesolution and then mixed with (1:8) PPy extracts, or salinesolution, or distilled water. The contents were gently mixedand incubated at 37°C for 4 h. After incubation, the suspen-sion was centrifuged at 750g for 5 min. The absorbance of thesuspension was measured with a UV 500 UV-visible spec-trometer (Thermo Inc.) at 540 nm, and the hemolysis ratiowas calculated according to the following formula: hemoly-sis ratio � (absorbance of treatment group � absorbance ofnegative control group) � (absorbance of positive controlgroup � absorbance of negative control group) � 100%.

Allergic test

This experiment was designed to assess any allergic reac-tion after animals were injected with PPy extraction solu-tion. Eight female guinea pigs (body weight 280–320 g) wererandomly divided into two groups. Each animal was in-jected with 1 mL of PPy extraction solution intraperitoneallyonce every other day for three times. After 14 days of thefirst injection, the animals from one group were injectedwith 1 mL of PPy extraction solution intravenously again;and after 21 days of the first injection, the animals from theother group were injected with 1 mL of PPy extractionsolution intravenously again.

Micronuclei test

Fifteen male and 15 female ICR mice (weight 19–21 g)were randomly divided into three groups: a treatmentgroup, a positive control group, and a blank control group,five male and five female mice for each group. The mice ofthe treatment group were injected with the PPy extractionsolution (50 mL/kg) into the abdominal cavity once, and themice of the blank control group or positive control groupwere injected with saline solution (50 mL/kg) or cyclophos-phamide (80 mg/kg) into the abdominal cavity once, respec-tively. Six hours and 30 h before being killed, each animalwas injected with the same solution as the above intraperi-toneally. The bone marrow of the animal was smeared andstained with Wright’s stain. The micronuclei of polychro-matic erythrocyte on smear of bone marrow were observedand counted, and the appearing rate of micronuclei per 1000polychromatic erythrocytes was calculated.

BIOCOMPATIBILITY OF POLYPYRROLE 413

Page 4: Evaluation of biocompatibility of polypyrrole in vitro and in vivo

Test for cell culture of peripheral nerve tissue

The dorsal root ganglia (DRG) were obtained from 15newborn Sprague-Dawley rats and stripped with connectivetissue at its surface. The DRG were seeded on the PPymembrane-coated surface of ITO glass, or on bare ITO glass,or on collagen membrane formed by collagen solution ofrat’s tail in culture plate. L15 medium supplemented with15% FBS was added to the culture plate in a humidifiedincubator with 5% CO2 at 37°C for 3 weeks, respectively.The growth of DRG was observed and measured with amicrometer under inverted light microscopy at differentculture times. The specimen of PPy membrane was observedby scanning electron microscopy at the end of 3 weeks.

Test for PPy-silicone tube inserted into theperipheral nerve

Ten male Sprague-Dawley rats (body weight approxi-mately 250 g) were anesthetized with sodium pentobarbitalin the abdominal cavity at approximately 3–5 mg/100 gbody weight, and the sciatic nerve of rats was exposed in theleft leg. The nerve was transected at 5 mm away frominferior border of the piriformis muscle, and transected forthe second time at 6 mm away from the initial transection,with the loose segment removed. A 14-mm-long PPy-sili-cone tube was inserted into the nerve stump for randomlyselected five rats (as the treat group). A 14-mm-long plainsilicone tube was inserted into the nerve stump for other fiverats (as the control group). In both cases, the epineuriums oftwo nerve stumps were sutured at either ends of the PPy-silicone tube or plain silicone tube with a 9-0 stitch so as tomaintain a 10-mm gap. All rats were kept routinely for 6months after the operation and monitored for changes intheir appearance, appetite, body weight, wound, and oper-ated leg. At the end of 6 months postoperation, the nerveand the channel tube were dissected and separated from thesurrounding tissue. The multiple action potentials were re-corded with routine electrophysiology. At the end of theexperiments, the nerve, the PPy-silicone tube, and the plainsilicone tube were also harvested for histological evaluation.The specimens were qualitatively and quantitatively evalu-ated by light microscopy and transmission electron micros-copy with routine fixation, embedment, section, and stainingwith trichrome stain18 (hematoxylin, fast green FCF, chro-motrope 2R, and phosphotungstic acid).

Statistical analysis

A one-way analysis of variance was used to compare themeans of different groups, and statistical significance wasaccepted at the 0.05 confidence level.

RESULTS

Surface morphology

The scanning electron microphotographs of PPymembrane and PPy-silicone tube are shown in Figure 1.

The PPy membrane on the ITO glass was flat, smooth,and dense with few microporosities. The conductance ofPPy synthesized electrochemically was good. The innersurface of the silicone tube was completely and evenlycovered by PPy film. The thickness of the PPy film in thetube was 650 nm. There were two major features of thesurface of PPy tube: one was flat, smooth, and densewith few microporosities, which was similar to the PPymembrane on the ITO glass slide, another was catkin-like or granule-like with many microporosities. The con-ductance of the PPy tube was also good.

Tests for systemic toxicity

Acute toxicity test

The injection of PPy did not cause any deaths, con-vulsions, or paralysis in either the treatment or controlgroup during 72 h after injection. All animals had anormal appetite and movement, and their normalbody weight increased. These results are summarizedin Table I. No significant difference was observedbetween the treatment group and control group.

Subacute toxicity test

All animals showed a normal appetite and move-ment, and a normal body weight increased in both thetreatment and control groups during 30 days of thesubacute toxicity test. Under light microscopy, thehistological structures of the heart, lung, liver, spleen,and kidney were normal, and no degenerated or ne-crotic cells or tissues were found; no lymphocytes orinflammatory cells were found.

Pyretogenesis test

The normal rabbit’s temperature was 38.64° � 0.15°C.The maximum changes in the rabbits’ temperature were0.15°C, 0.33°C, and 0.25°C, respectively, after injection ofthe PPy extracts to veins, which were under 0.6°C crite-rion. The total rising temperature of three rabbits was0.73°C, which was under 1.4°C criterion.

MTT test

The results of the survival and proliferation ofSchwann cells in the different groups are presented inTable II. The survival and proliferation rates of Schwanncells in L15 medium containing 50% PPy extraction so-lution were significantly higher than those in the controlgroup of L15 medium containing 50% saline solution,

414 WANG ET AL.

Page 5: Evaluation of biocompatibility of polypyrrole in vitro and in vivo

but those were significantly lower than that in the con-trol group of plain L15 medium. The data demonstratethat the PPy extraction solution was not cytotoxic.

Hemolysis test

As revealed in Table III, the blood sample mixedwith the PPy extraction solution showed no evidence

of hemolysis. The sample from the treatment grouphad nearly 0% of hemolysis ratio, which was farsmaller than 5% international standard.

Allergic test

After being injected with PPy extraction solution,Guinea pigs in two groups did not show any symp-

Figure 1. The scanning electron microscopy photomicrographs of the PPy membrane and PPy-coated tube. (A) The PPymembrane on the ITO glass slide was flat, smooth, and dense with few microporosities. (B) The inner surface in the siliconetube was completely and evenly covered by the PPy-coated tube. There were two major characterizations of the surface ofPPy-coated tube: (C) one feature was flat, smooth, and dense with few microporosities, and (D) another feature wascatkin-like or granule-like with many microporosities.

BIOCOMPATIBILITY OF POLYPYRROLE 415

Page 6: Evaluation of biocompatibility of polypyrrole in vitro and in vivo

toms of allergic reaction, such as nose scratching, pi-loerection, dyspnea, and spasm.

Micronuclei test

The results of the appearing rate of micronuclei per1000 polychromatic erythrocytes are indicated in Ta-ble IV. The appearing rate of micronuclei showed nosignificant difference between the PPy extraction so-lution group (the treatment group) and the saline so-lution group (the blank control group), but bothgroups were significantly lower in appearing rates ofmicronuclei compared with the cyclophosphamidegroup (the positive control group).

Test for coculture of PPy with peripheral nervetissue

After 1 day of culture, the cells migrated from 32%of the DRG on the PPy membrane, and the distance ofmigration was approximately 50 �m away from theDRG. After 2 days of culture, Schwann cells and fi-broblasts migrated from approximately 70% of theDRG on the PPy membrane, whereas the neurites bythe cells traction, about 96 �m long, appeared on partof the DRG. After 3 days of culture, the cells grew outof 85% of the DRG on the PPy membrane, and

Schwann cells emigrated radially from the DRG, andneurites extended approximately 253 �m (as shown inFig. 2). When the culture took up to 1 week, more andmore Schwann cells migrated further from the DRGby the neurites on the PPy membrane. The cells mi-grated by the neurites aggregated and formed a clus-ter of cells at the end of the neurites. The number ofneurites increased continuously and adjacent neuritesunited together to form thick processes. After 2 weeksof culture, whereas the cells and processes constantlyemigrated from the DRG on the PPy membrane, thecluster of cells at the end of the neurites also extendedto form new neurites. The growing phenomena ofDRG on the PPy membrane at the second week wereproceeding continuously in 3 weeks. The cells migrat-ing and neurites extending on collagen membranewere similar to that on the PPy membrane. Althoughthe cells migrated and the neurites extended on thebare glass slide, the distance of migrated cells and thelength neurites extended were shorter than that on thePPy membrane or on collagen membrane.

The relations between total length of neurites ofDRG and culture time on three different surfaces arepresented in Figure 3. These curves fitted logistic pass-ing through zero point (0,0). The difference betweenneurites of DRG on the PPy membrane and on thebare glass slide was significant (p � 0.01, t test ofmodeling parameter) and the difference between neu-

TABLE IIResults of the MTT Test for PPy Extraction Solution

(x� � SD)

Group OD Value

L15 medium plus 50% PPy extractionsolution 1.204 � 0.070*,**

L15 medium plus 50% saline solution 1.147 � 0.085***L15 medium 1.281 � 0.061

Compared with L15 medium group, *p � 0.05; ***p � 0.01.Compared with saline solution group, **p � 0.05.

TABLE IResults of the Acute Toxicity Test for PPy Extraction Solution

PPy Extraction Solution Saline Solution

AnimalNo.

Weight (g) Average WeightIncrease (%)

(x� � SD)Animal

No.

Weight (g) Average WeightIncrease (%)

(x� � SD)Day 0 Day 3 Day 0 Day 3

1 20 24

0.205 � 0.053

1 19 24

0.206 � 0.040

2 19 23 2 21 253 21 25 3 20 244 20 25 4 19 225 19 24 5 21 256 20 22 6 19 247 20 25 7 20 238 21 24 8 20 259 20 25 9 20 24

10 21 25 10 21 25

TABLE IIIResults of the Hemolysis Test for PPy Extraction

Solution

Group

Animal No. Total ofHemolysisRatio (%)1 2 3 4

PPy extraction solution 0.003 0.03 0.005 0.014Saline solution 0.002 0.02 0.001 0.005 0.84Distilled water 0.77 0.75 0.68 0.88Hemolysis ratio (%) 0.39 1.37 0.59 1.03

416 WANG ET AL.

Page 7: Evaluation of biocompatibility of polypyrrole in vitro and in vivo

rites of DRG on the PPy membrane and on collagenmembrane was not significant. The logistic curvesthrough zero point (0,0) exhibited strict monotonouselevation tendency. The longest neurites of DRG onthe PPy membrane were longer than that on the bareglass slide. The speed increase on logistic curves in thePPy membrane group was faster than that in the bareglass slide group, and the inflection point on logisticcurves for the PPy membrane group was 0.45 day laterthan that for the bare glass slide group. The relationsbetween neurites number of DRG and time of cellculture on three different surfaces are showed in Fig-

ure 4. The Poisson’s regression analysis verified thatthe neurites number of DRG on PPy membrane or oncollagen membrane was significantly different fromthat on bare glass slide (p � 0.01), but the difference ofthe number of neurites on the PPy membrane and oncollagen membrane was not significant.

Under scanning electron microscopy, the long oli-vary Schwann cells emigrated radially from DRG onthe PPy membrane with 2–3 large processes and smallbranch, and the cells were connected to each other bylarge processes (as shown in Fig. 5). The otherSchwann cells emigrated from DRG along the neuritesto form a cluster of cells at the neurite terminals. Manyneurites extended from DRG, and adjacent neuritesunited together to form thick processes (as shown inFig. 6).

Figure 2. After 3 days of culture, the cells were foundproductively to grow out of DRG on the PPy membrane, andSchwann cells emigrated radially from DRG, and neuronsextended neurites. Original magnification �360. [Color fig-ure can be viewed in the online issue, which is available atwww.interscience.wiley.com.]

TABLE IVResults of the Micronuclei Test for PPy Extraction

Solution (x� � SD)

Group Rate of Micronuclei (‰)

PPy extraction solution 2.90 � 0.738*Saline solution 2.80 � 1.135*Cyclophosphamide 28.5 � 2.593

Compared with cyclophosphamide group, *p � 0.01.

Figure 3. The relation between total length of neurites ofDRG and time of culture on three different surfaces. Thesecurves were logistic curves through zero point (0,0) by curvefitting. The difference between neurites of DRG on the PPymembrane and on the bare glass slide was significant (p �0.01, t test of modeling parameter), but the difference be-tween neurites of DRG on the PPy membrane and on colla-gen membrane was not significant.

Figure 4. The relation between the number of neurites ofDRG and time of cell culture on three different surfaces. ThePoisson’s regression analysis verified that the number ofneurites of DRG on PPy membrane or on collagen mem-brane was significantly different from that on the bare glassslide (p � 0.01), but the difference of the number of neuriteson the PPy membrane and on collagen membrane was notsignificant.

BIOCOMPATIBILITY OF POLYPYRROLE 417

Page 8: Evaluation of biocompatibility of polypyrrole in vitro and in vivo

Test for the PPy-silicone tube inserted into theperipheral nerve

During 6 months of postoperation, all rats exhibitednormal appetite, normal body weight increase, as wellas normal luster of hair. However, the skin on theankle joint of the operated legs of all rats was red andswollen, and muscle atrophy occurred to differentextent in 2 weeks. Different extent of skin ulcer ap-peared on the operated leg of two rats in each group in2 months. Their toe could not spread in 6 months. Thesymptom of extensive rejecting reaction and inflam-matory reaction was not found in any rats during 6months of postoperation in the two groups.

The nerve and the silicone tube were exposed againafter 6 months postoperation. The silicone tube andthe nerve terminals were connected well in the two

groups, and there was connective tissue with muchsmaller blood vessel adhesion and coating on the sil-icone tube, which had no edema, hematoma, abscess,or inflammatory reaction at the operation site. Thetransmission speed of regenerated nerve tissue was9.33 � 4.28 m/s and the action potential was 9.07 �4.83 mV in the PPy-silicone tube. The transmissionspeed of regenerated nerve tissue was 6.28 � 5.43 m/sand the action potential was 5.92 � 3.61 mV in theplain silicone tube. Four rats of the treatment groupproduced action potential in regenerated nerves, andonly two rats of the control group did.

Under light microscopy, there were abundant bloodvessels and 0–10 white blood cells per high-powerfield in the section of regenerated nerve tissue in two

Figure 5. Scanning electron microscopy photomicrographof the Schwann cells emigrated radially from DRG on thePPy membrane. The long olivary Schwann cells with 2–3large processes and small branch, and the cells were con-nected to each other by large processes. Original magnifica-tion �800.

Figure 6. Scanning electron microscopy photomicrographof the Schwann cells emigrated radially and neurites ex-tended from DRG on the PPy membrane. Part of theSchwann cells emigrated from DRG along the neurites toform a cluster of cells at the neurite terminals. Many neuritesextended from DRG, and adjacent neurites united togetherto form thick processes. Original magnification �400.

418 WANG ET AL.

Page 9: Evaluation of biocompatibility of polypyrrole in vitro and in vivo

groups, which indicated that the PPy-silicone tubecaused only minimal reaction.19 The transected area ofregenerated nerve tissue was close to the proximal endof the sciatic nerve in two rats of the treatment group,which were coated by the connected tissue, and ap-proximately 2–3 nerve tracts were found in their re-generated nerve tissue. As shown in Figure 7, stainedby trichrome stain, the green dots are the regeneratedneurites (the other shape green structure is connectivetissue), and a light red substance coating the outer partof the neurites is the myelin sheath (the dark redsubstance in the blood vessels is the red blood cell) onthe section of regenerated nerve tissue in the PPy-silicone tube. Similar structures could be found on thetransected distal end of regenerated nerve tissue, but

the number of both neurites and myelin sheath on thedistal end were less than those on the proximal end (asshown in Fig. 8). In the PPy-silicone tube of the otherthree rats of the treatment group and in the plainsilicone tube of three rats of the control group, thesparsely regenerated neurites and myelin sheath werefound and the section area of regenerated nerve tissuewas small. The regenerated tissue in the plain siliconetube of the other two rats was connective tissue, andthe diameter of the regenerated tissue was only 0.2mm.

Under electron microscopy, the regenerated neu-rites and thin but high electron density myelin sheathwere found in the PPy-silicone tube. Many regener-ated neurites were coated by a Schwann cell to formunmyelinated nerve fiber (as shown in Fig. 9). Theregenerated neurites and myelinated nerve fiber werefew in the plain silicone tube.

A comparison of the section area of regenerated

Figure 8. The transected distal end of regenerated nervetissue in the PPy-silicone tube was stained by trichromestain. Similar structures could be found on the transectedregenerated nerve tissue, but the number of neurites (greendots) and myelin sheath (light red substance) on the distalend of the regenerated nerve tissue was less than that on theproximal end of the regenerated nerve tissue. Original mag-nification �180. [Color figure can be viewed in the onlineissue, which is available at www.interscience.wiley.com.]

Figure 7. The transected proximal end of regeneratednerve tissue in the PPy-silicone tube was stained bytrichrome stain. The regenerated nerve was coated by theconnected tissue. The green dots are the regenerated neu-rites (the other shape green structure is connective tissue),and a light red substance coating the outer part of theneurites is the myelin sheath (the dark red substance in theblood vessels is the red blood cell) on the section of regen-erated nerve tissue in the PPy-silicone tube. Original mag-nification �180. The middle part of the photomicrographwas amplified and is located at the bottom right of thisfigure as an inset. Original magnification �270. [Color figurecan be viewed in the online issue, which is available atwww.interscience.wiley.com.]

BIOCOMPATIBILITY OF POLYPYRROLE 419

Page 10: Evaluation of biocompatibility of polypyrrole in vitro and in vivo

nerve and number of regenerated axons between thetwo groups are shown in Table V.

DISCUSSION

Of the various conducting materials for biomedicalapplications, PPy is probably one of the most widely

studied polymers in virtue of its inherent advantage,namely, the surface properties of PPy (e.g., surfacecharge, wettability, and conformational and dimen-sional changes) can be altered reversibly by chemicalor electrochemical oxidation or reduction for cellgrowth.20,21 It has been accepted that evaluation of abiocompatibility of any biomaterial must be related tothe principle of biosafety. That is why this study as-sumes a more thorough series of toxicity tests for theuse of PPy to different animal models. The resultssuggested that the extraction solution of PPy powderchemically prepared does not have acute toxicity andsubacute toxicity leading to lethality to tested animals,and indicated that it does not cause change of bodytemperature, hemolysis of red blood cells, allergic re-sponse of the immune system, or mutagenesis of cells.The results also showed that the biocompatibility be-tween Schwann cells and PPy was good and PPy wassuitable for growth of mammalian nerve tissue. Allthese conclusions are in accordance with internationalstandard (ISO 10993 and ASTM F1748-82).16,17 In aquantitative measure of cell viability (MTT) of thisstudy, which is the chiefly chosen method of the cy-totoxic test in vitro on evaluation of biocompatibility ofbiomaterials at the present time, the results on thecultured Schwann cells, a kind of normal neuroglialcell in the peripheral nerve system, explain whetherPPy is cytotoxic to peripheral nerve tissue or not, sothat it affords a promising application of PPy to pe-ripheral nerve reconstruction in the future.

To examine the biocompatibility between PPy andnerve tissue, it is important to conduct in vitro and invivo studies in an environment that is as close to whatwill be applied to as possible. We tried to culture thenormal mammal nerve tissue on the PPy membrane,and to use the PPy membrane deposited on the innersurface of the silicone tube for bridging across a10-mm sciatic nerve gap in rats. Both in vitro and invivo experiments showed that PPy was useful togrowth of nerve cells. When the nerve tissue wascocultured with PPy, Schwann cells showed normaladhesion, survival, migration, and proliferation, andneurites showed normal extension. Whether PPymembrane is suitable for growth of nerve cells andSchwann cells or not is one of the standards for a good

Figure 9. Transmission electron microscopy photomicro-graph of regenerated nerve in the PPy-silicone tube. Theregenerated neurites and thin but high electron dense regen-erated myelin sheath were found and many regeneratedneurites were coated by a Schwann cell to form unmyeli-nated nerve fiber. Original magnification �18,000.

TABLE VComparison of Area of Regenerated Nerve and Number of Regenerated Axons Between the Two Groups (x� � SD)

Group

ARN (mm2) RRRN (%) NRA RRRA (%)

Proximal Distal Proximal Distal Proximal Distal Proximal Distal

PPy group 0.79 � 0.01 0.48 � 0.4 0.98 � 0.02 0.60 � 0.49 1999.5 � 197.27 841.0 � 69.30 0.95 � 0.99 0.41 � 0.04PS group 0.43 � 0.17 0.42 � 0.31 0.53 � 0.21 0.52 � 0.38 1554.0 � 63.64 581.5 � 415.07 0.74 � 0.03 0.28 � 0.21t Value 3 0.17 2.8 0.18 3.04 0.87 2.9 0.68p Value �0.05 �0.5 �0.05 �0.5 �0.05 �0.5 �0.05 �0.5

ARN, area of regenerative nerve; RRRN, recovery rate of regenerative nerve; NRA, number of regenerative axons; RRRA,recovery rate of regenerative axon; PS, plain silicone.

420 WANG ET AL.

Page 11: Evaluation of biocompatibility of polypyrrole in vitro and in vivo

candidate material of bridging and repairing periph-eral nerve gap. Particularly, it is essential thatSchwann cells and neurites of nerve cells grow well,because Schwann cells must proliferate a great deal,secrete many neurotrophic factors, form Bungner’sband, and induct neurites to regenerate and prolongin the nerve regeneration. It is obvious that PPy has anenhancement to nerve regeneration, the mechanism ofwhich has been discussed.22–26

Despite the reported chemical and electrochemicalmethods for PPy synthesis,27 it is worthy to mentionthat this is the first study to report electrochemicallydepositing PPy film onto the inner surface of the sili-cone tube. It is a very important structural foundationof guiding nerve regeneration in vivo to form a tube ofPPy membrane, because the structure of conduit ismore suitable for suturing and bridging across thenerve gap and adding the electric stimulation.

Moreover, because of the chemical and thermal sta-bility of PPy,28 there was still the tube of PPy mem-brane around the regenerated nerve tissue after 6months postoperation. It is possible that the tube ofPPy membrane would forever exist around nervoustissue in the body if used in vivo. Therefore, PPy mustbe examined for chronic toxicity, which will be donein our future study.

It is reasonable that electrical stimulation added tothe PPy membrane might lead to an enhanced growthof nerve tissue.11,29 To obtain a further understandingof the capability of PPy to repair defected nerve, it isnecessary that electric stimulation be included in fu-ture experiments, so that the electrical conductiveproperties of PPy can be utilized more effectively.

CONCLUSION

This study demonstrated that PPy extraction solu-tion had no evidence of acute and subacute toxicity,pyretogen, hemolysis, allergen, or mutagenesis. TheSchwann cell migration and neurite extension fromDRG were better on the surface of the PPy membranethan on the surface of bare glass in vitro. Moreover,there was only light inflammation during 6 months ofpostoperation after the PPy-silicone tube was bridgedacross the gap of rat sciatic nerve. The regeneratednerve grew slightly better in the PPy-silicone tubethan in the plain silicone tube by electrophysiologicaland histological tests. Therefore, this study concludesthat the PPy has a good biocompatibility with periph-eral nerve tissue.

References

1. Rosen JM. Concepts of peripheral nerve repair. Ann Plast Surg1981;7(2):165–171.

2. Heath CA, Rutkowski GE. The development of bioartificialnerve grafts for peripheral-nerve regeneration. Trends Biotech-nol 1998;16:163–168.

3. Kimura K, Yanagida Y, Haruyama T, Kobatake E, AizawaM. Electrically induced neurite outgrowth of PC12 cells onthe electrode surface. Med Biol Eng Comput 1998;36:493–498.

4. Ming G, Henley J, Tessier-Lavigne M, Song H, Poo M. Electri-cal activity modulates growth cone guidance by diffusiblefactors. Neuron 2001;29:441–452.

5. Diefenbach TJ, Guthrie PB, Kater SB. Stimulus history altersbehavioral responses of neuronal growth cones. J Neurosci2000;20:1484–1494.

6. McCaig CD, Rajnicek AM, Song B, Zhao M. Has electricalgrowth cone guidance found its potential? Trends Neurosci2002;25:354–359.

7. Borgens RB, Blight AR, Murphy DJ, Stewart L. Transecteddorsal column axons within the guinea pig spinal cord regen-erate in the presence of an applied electric field. J Comp Neurol1986;250:168–180.

8. Wong JY, Langer R, Ingber DE. Electrically conducting poly-mers can noninvasively control the shape and growth ofmammalian cells. Proc Natl Acad Sci USA 1994;91:3201–3204.

9. Garner B, Georgevich A, Hodgson AJ, Liu L, Wallace GG.Polypyrrole–heparin composites as stimulus-responsive sub-strates for endothelial cell growth. J Biomed Mater Res 1999;44:121–129.

10. De Giglio E, Sabbatini L, Colucci S, Zambonin G. Synthesis,analytical characterization, and osteoblast adhesion propertieson RGD-grafted polypyrrole coatings on titanium substrates.J Biomater Sci Polym Ed 2000;11:1073–1083.

11. Schmidt CE, Shartri VR, Vacamh JP, Langer R. Stimulation ofneurite outgrowth using an electrically conducting polymer.Proc Natl Acad Sci USA 1997;94:8949–8953.

12. Williams RL, Dohert PJ. A preliminary assessment of poly(pyr-role) in nerve guide studies. J Mater Sci Mater Med 1994;5:429–433.

13. Cui X, Wiler J, Dzaman M, Altschuler RA, Martin DC. In vivostudies of polypyrrole/peptide coated neural probes. Bioma-terials 2003;24:777–778.

14. Jiang X, Marois Y, Traore A, Tessier D, Tessier D, Dao LH,Guidoin R, Zhang Z. Tissue reaction to polypyrrole-coatedpolyester fabrics: an in vivo study in rats. Tissue Eng 2002;8:635–647.

15. Zhang Z, Roy R, Dugre FJ, Tessier D, Dao LH. In vitro biocom-patibility study of electrically conductive polypyrrole-coatedpolyester fabrics. J Biomed Mater Res 2001;57:63–71.

16. International Organization for Standardization. Internationalstandard (ISO 10993)-biological evaluation of medical devices.2nd ed. Printed in Switzerland; 1997.

17. ASTM F1748-82. Annual book of ASTM standards. Medicaldevices; 1983.

18. Meyer R, Claussen GC, Oh SJ. Modified trichrome stainingtechnique of the nerve to determine proximal nerve viability.Microsurgery 1995;16:129–132.

19. Keeley R, Atagi T, Aabelman E, Padilla J, Kadlcik S, Keeley A,Nguyen K, Rose J. Peripheral nerve regeneration across 14-mmgap: A comparison of autograft and entubulation repair meth-ods in the rat. J Reconstr Microsurg 1993;9:349–358.

20. Khor E, Li HC, Wee A. In situ polymerization of pyrrole inanimal tissue in the formation of hybrid biomaterials. Bioma-terials 1995;16:657–661.

21. Street GB. In: Skotheim TA, editor. Handbook of conduct-ing polymers. Vol. 1. New York: Marcel Dekker; 1986. p 265–301.

BIOCOMPATIBILITY OF POLYPYRROLE 421

Page 12: Evaluation of biocompatibility of polypyrrole in vitro and in vivo

22. Smith AB, Knowles CJ. Investigation of the relationship be-tween conductivity and protein-binding properties of polypyr-role. J Appl Polym Sci 1991;42:399–403.

23. Genoud F, Guglielmi M, Nechtschein M, Genies E, Salmon M.ESR study of electrochemical doping in the conducting poly-mer polypyrrole. Phys Rev Lett 1985;55:118–121.

24. Batz P, Schmeisser D, Gopel W. Electronic structure of poly-pyrrole films. Phys Rev B Condens Matter 1991;43:9178–9189.

25. Livache T, Bazin H, Mathis G. Conducting polymers on micro-electronic devices as tools for biological analyses. Clin ChimActa 1998;278:171–176.

26. Collier JH, Camp JP, Hudson TW, Schmidt CE. Synthesis andcharacterization of polypyrrole–hyaluronic acid composite

biomaterials for tissue engineering applications. J Biomed Ma-ter Res 2000;50:574–584.

27. Dicks JM, Hattori S, Karube I, Turner AP, Yokozawa T. Ferro-cene modified polypyrrole with immobilised glucose oxidaseand its application in amperometric glucose microbiosensors.Ann Biol Clin 1989;47:607–619.

28. Jiang X, Tessier D, Dao LH, Zhang Z. Biostability of electricallyconductive polyester fabrics: An in vitro study. J Biomed MaterRes 2002;62:507–513.

29. Kotwal A, Schmidt CE. Electrical stimulation alters pro-tein adsorption and nerve cell interactions with electri-cally conducting biomaterials. Biomaterials 2001;22:1055–1064.

422 WANG ET AL.