Transfection of epithelial cells is enhanced by combined treatment with mannitol and...

RESEARCH ARTICLE

Transfection of epithelial cells is enhanced bycombined treatment with mannitol andpolyethyleneglycol

Markus Duchler1*

Margret Pengg1

Sylvia Brunner2

Mathias Muller3

Gottfried Brem3

Ernst Wagner2

1Institute of Biotechnology in AnimalProduction, University of VeterinaryMedicine, IFA Tulln, KonradLorenzstrasse 20, A-3430 Tulln,Austria2Institute of Biochemistry, ViennaUniversity Biocenter, Dr Bohrgasse9/3, A-1030 Vienna, Austria3Institute of Animal Breeding andGenetics, University of VeterinaryMedicine, Veterinarplatz 1, A-1210Vienna, Austria

*Correspondence to: Dr. M Duchler,Institute of Biotechnology in AnimalProduction, University of VeterinaryMedicine, IFA Tulln, KonradLorenzstrasse 20, A-3430 Tulln,Austria.E-mail: [email protected]

Received: 22 May 2000

Revised: 30 October 2000

Accepted: 17 November 2000

Published online: 12 February 2001

Abstract

Background Gene transfer efficiency drops significantly when polarizedmammary epithelial cells are transfected instead of actively growing cells.However, fully differentiated cells are the targets for gene transfer in manyin vivo applications. Therefore, a simple and effective method for thetransfection of polarized mammary epithelial cells in confluent monolayerswas developed.

Methods Reporter gene plasmids were complexed with polyethyleniminewith an average molecular weight of 25 kDa (PEI 25), or other agents, totransfect confluent monolayers of ovine mammary epithelial cells (OMEC II)or human carcinoma cells (CaCo-2) in vitro. The improved techniqueincluded pretreatment of the cells with a hyperosmotic mannitol solution(7%) which caused a loosening of the tight contacts between the cells.Alternatively, the mannitol shock could be replaced by a short treatment withtrypsin or EDTA. In addition to the pretreatment, 12.5% polyethyleneglycolwith an average molecular weight of 8000 kDa (PEG 8000) was included inthe transfection mixture containing the DNA complexes.

Results The combined application of mannitol and PEG resulted in a veryreliable 5- to 30-fold increase in reporter gene expression in OMEC II andCaCo-2 cells, but not K562 cells (an example of another cell type). Theimproved technique can also be combined with other polymer-basedtransfection agents. The transfection rate was enhanced for confluentmonolayer cells with fully developed epithelial polarity but also forsubconfluent, growing epithelial cell cultures.

Conclusions A novel transfection protocol for epithelial cells is presented.The combined treatment of cells with mannitol and polyethyleneglycol resultsin substantial enhancement of in vitro transfection of epithelial cell lines.Copyright # 2001 John Wiley & Sons, Ltd.

Keywords gene transfer; polyethylenimine; confluent monolayer; osmoticshock; polyethyleneglycol; epithelial cells

Introduction

Somatic gene transfer requires safe and efficient techniques to introduce DNAinto mammalian cells. In many in vivo applications epithelial tissuesconsisting of well-differentiated cells are the targets for gene transfer. Suchtightly packed, polarized tissue cells are much harder to transfect thangrowing cells in tissue culture, probably because of their limited access for

THE JOURNAL OF GENE MEDICINEJ Gene Med 2001; 3: 115–124.DOI: 10.1002/jgm.171

Copyright # 2001 John Wiley & Sons, Ltd.

vector particles. Increasing the efficiency of gene transferinto differentiated cells is therefore of great importancefor the development of gene therapy protocols.Viral vectors are the most efficient gene transfer

vehicles but they also provoke serious concerns abouttheir application in vivo because of their potential safetyrisks [1–3]. To some extent all viruses also elicit ahost–immune response. Furthermore, their capacity tointroduce additional genes is limited.Besides viral vectors, cationic lipids [4] and polycations

[5] have been successfully employed for gene transfer.These polycationic agents efficiently condense theextended structure of DNA and neutralize its negativecharge. There is no size limitation for the DNA fragmentsthat can be complexed with polycations. In general, theirtoxicity is low [6–12] and they are easily prepared in largequantities.The cationic polymer, polyethylenimine (PEI), was

shown recently to mediate efficient gene transfer in vitro

as well as in vivo [13–17]. It exerts high pH-bufferingcapacity that is thought to prevent the DNA fromlysosomal degradation. In addition, PEI seems to facilitatenuclear transport while it remains associated with theDNA [18].Our approach to improve in vitro transfection of

confluent epithelial cells by PEI/DNA complexes was topretreat the cells with hyperosmotic mannitol solutions.This pretreatment resulted in a loosening of the cell–cellcontacts in the epithelial layer and a moderate increase intransfection efficiency. In addition we found that theinclusion of PEG in the transfection mixture furtherenhanced the transfection efficiency.

Materials and methods

Materials

Dulbecco’s modified Eagle’s medium (DMEM), fetal calfserum (FCS), glutamine, penicillin, streptomycin and10rtrypsin/EDTA solution were supplied by Gibco/BRLLife Technology Inc., Rockville, MD. Polyethylenimineswith an average molecular weight of 25 kDa (PEI 25)and 800 kDs (PEI 800) were purchased from Fluka,Milwaukee, WI (50% aqueous solution) and used afterneutralization with hydrochloric acid as a 0.9 mg/mlaqueous solution. Polyethyleneglycol with an averagemolecular weight of 8000 kDa (PEG 8000) was obtainedfrom J. T. Baker (Deventer, The Netherlands) anddissolved in serum-free medium (SFM) at 25% (w/v).The procedures for synthesizing transferrin-polylysine

(Tf-pLys) conjugates, streptavidin-polylysine (STAV-pLys)conjugates and biotinylated 8-MOP-inactivated adeno-virus particles have been described elsewhere [19–21].GeneGrip vector DNA was purchased from Gene

Therapy Systems Inc. (San Diego, CA, USA).The cationic lipid formulations, DOSPER and Lipo-

fectamine, were obtained from Roche Molecular Bio-chemicals, Mannheim, Germany and Gibco/BRL Life

Technology Inc., respectively. All other chemicals werepurchased from Sigma (St Louis, MO, USA).

Cell culture

OMEC II is an ovine mammary epithelial cell line that wasestablished from primary mammary epithelial cells in ourlaboratory [22]. K562 and CaCo-2 cells were obtainedfrom ATCC, Rockville, MD. OMEC II cells were cultured inDMEM containing 10% FCS, 2 mM glutamine andantibiotics (penicillin 100 U/ml, streptomycin 0.1 mg/ml) (=TCM) in a humidified atmosphere at 37uC and 5%CO2. K562, a human chronic erythroid leukemia cell line,was cultured in RPMI 1640 with 10% FCS, 2 mMglutamine and antibiotics. CaCo-2 (ATCC HTB-37), acolorectal epithelial adenocarcinoma cell line, was grownin DMEM (high glucose), 20% FCS heat inactivated (hi)plus 1% non-essential amino acids.

Transfection

Epithelial cells were plated at 35 000 cells/well in 24-wellplates the day before transfection, or at 60 000 cells/well2 days before transfection to obtain 50 or 100% confluentmonolayers, respectively. Suspension cells (K562) wereplated at a density of 8r104 cells/well on the day oftransfection.

DNAThe plasmids used were pCLuc in which a CMV promoter/enhancer controls expression of luciferase [23] andpEGFP from Clontech (Palo Alto, CA, USA).

Mannitol treatmentCell culture supernatant was replaced by a 7% mannitolsolution in H2O and the plates were incubated at 37uC.After 1 h the mannitol solution was replaced by thetransfection mixture. Alternatively, cells were treatedwith 0.2% EDTA for 4 min, or with 1rtrypsin or trypsin/EDTA for 2–3 min. For OMEC II cells grown on permeablefilter supports a 9%mannitol solution was applied for 1 h.Complexes were formed by mixing 1.6 mg plasmid DNA

in 100 ml DMEM without FCS and without antibio-tics (SFM, serum-free medium) with 1.44 mg PEI 25(or PEI 800) in 100 ml SFM at room temperature. After20 min 200 ml SFM were added to the mixture which wasthen distributed to two wells of the 24-well plate. ForPEG-enhanced transfection, complexes were formed inSFM and a 25% PEG solution in SFM was used to dilutethe complexes after 20 min to give a final concentration of12.5% PEG. After 4 h at 37uC the medium was replacedwith fresh TCM.Cationic lipids or hTf-pLys were used in the same way as

PEI 25 at the following concentrations: DOSPER, 5 mg/mgDNA; Lipofectamin, 3 mg/mg DNA; hTf-pLys, 1 mg/mgDNA. For the adenovirus-enhanced transfection (AVET),three incubations at 20 min at room temperature werecarried out: first, 0.5 mg streptavidin-pLys in 100 mlHEPES buffered saline (HBS) were incubated with 10 ml

116 M. Duchler et al.

Copyright # 2001 John Wiley & Sons, Ltd. J Gene Med 2001; 3: 115–124.

(approximately 1010 particles) biotinylated 8-MOP-inac-tivated adenovirus particles. In a second step, 6 mg DNA in150 ml HBS were added and after another 20 min 6 mghTf-pLys in 150 ml HBS were added. The resulting 500 mlwere diluted with DMEM containing 2% FCS to 1 mlwhich was distributed to two wells. The transfectionmixture was replaced with fresh TCM after 4 h.Reporter gene expression was examined 20–24 h after

transfection. For the luciferase assay cells were lysed in100 mM Tris pH 8.0; 1% NP40 (150 ml/well) andluciferase activity was quantified from one-tenth of thelysate from one well using a Lumat LB 9507 luminometer(Berthold, Bad Wildbad, Germany). Each experiment wasrepeated several times, and luciferase values are themeans of duplicates from one representative experiment.Luciferase values are expressed as light units integratedover 10 s per microgram of cell protein.Expression of green fluorescent protein was monitored

using a Nikon HFX-DX fluorescence microscope, andphotographs were taken from representative areas of thecell monolayers.

Determination of transepithelialelectrical resistance

OMEC II cells were seeded onto permeable filter supports(‘Transwell’; Corning Costar Corp., Cambridge, MA).After they had reached confluency they were cultured foranother 24 h. Transelectrical resistance was measuredusing the Millicell-ERS (Millipore, Bedford, MA).

Measurement of particle sizes oftransfection complexes by laser lightscattering

The size of DNA particles complexed with PEI 25 wasdetermined by quasi-elastic (dynamic) light scattering[24] using a BI90 particle size analyzer (Brookhaven;Holtsville, NY, USA). DNA (4 mg/ml) was mixed with3.6 mg PEI 25 in a final volume of 800 ml H2O. After a20-min incubation optionally PEG was added and theparticle sizes of complexes were measured, followed byfurther measurements after 20, 50 and 80 min.

Particle binding to cells

Vector DNA labeled with Rhodamine (GeneGrip) wasused to analyze DNA/PEI25 complex binding to ovinemammary epithelial cells. Complexes were prepared asusual in SFM except that the DNA concentration was6 mg/ml. Cells were transfected for 4 h, then the celllayers were washed six times with ice-cold phosphate-buffered saline (PBS), fixed in 2% paraformaldehyde andwashed again twice with cold PBS. Complexes associatedwith the cells were monitored using a Nikon HFX-DXfluorescence microscope, and photographs were takenfrom representative areas of the cell monolayers.

Results

Enhanced transfection efficiency afterpretreatment of confluent monolayerswith mannitol

OMEC II, an ovine mammary epithelial cell line, wastransfected with reporter gene constructs condensed bypolyethylenimine (PEI 25). A standard DNA concen-tration of 4 mg/ml was used for all the experiments. Asexpected, a strong decrease in transfection efficiency wasobserved when confluent cells as opposed to growing cellswere transfected (compare Figure 1A and B). In anattempt to make confluent cells more competent for DNAuptake, confluent monolayers were treated with ahyperosmotic mannitol solution immediately before theaddition of transfection complexes onto the cells. Thispretreatment resulted in a three-fold increase in thenumber of cells expressing GFP after transfection (datanot shown).

Addition of PEG further enhancestransfection efficiency

The enhancing effect of the mannitol treatment could befurther improved by the inclusion of PEG into thetransfection mixture. Various concentrations of PEGwere tested (Figure 2) and 12.5% was found to beoptimal. The combined mannitol/PEG treatment gavevery reproducible improvements in transfection efficiencyby a factor of 4–6 (Figures 1C and 3) when PEI 25 wasused for complexation of DNA. The new technique couldalso be combined with other transfection agents. Themost dramatic increase in transfection efficiency wasobserved when hTf-pLys was used for DNA complexation(Figure 3). Mannitol pretreatment alone also improvedcationic lipid mediated gene transfer, but the addition ofPEG strongly reduced the uptake of DNA/lipid complexes.Application of the mannitol or the PEG solutions alone

sometimes resulted in a decrease in transfection effi-ciency. We do not have a simple explanation for thisphenomenon. Prolonged mannitol treatment clearly isharmful for the cells.In order to show that the mannitol and PEG treatment

also worked for cells with fully developed epithelialpolarity, OMEC II cells were grown on permeable filtersupports (Transwells). Conditions for the mannitolpretreatment had to be adjusted: a 9% mannitol solutionhad to be applied to give optimal enhancement(Figure 3).

Mannitol pretreatment seems to causeloosening of tight junctions

OMEC II cells develop a polarized epithelial phenotype inconfluent cultures and tight junctions are formedbetween individual cells [22]. Mannitol treatment of

Epithelial Cells Transfection with mannitol and Polyethyleneglycol 117


confluent monolayers converts the tightly packed cell

layer into a looser one, where contacts between cells are

diminished (compare Figure 4A and B). This process is

fully reversible for the specified mannitol concentration

and time of the treatment.Addition of PEG to the transfection mixture accelerated

the process of reformation of tight contact between cells.

Thirty minutes after the addition of 12.5% PEG the

monolayer appeared tightly closed again (Figure 4D)

whereas cell layers where the mannitol solution had

been replaced with SFM still showed residual gaps(Figure 4C).These observations were further confirmed by mea-

surements of transepithelial electrical resistance (TER).OMEC II cells were grown on permeable filter supportsand the TER was determined for Transwells without cellsor with confluent monolayers to obtain the minimal andmaximal control values, respectively (Figure 5, leftsection). When the tissue culture medium was replacedby a 9% mannitol solution, TER showed a constantdecrease over time (Figure 5). Replacement of themannitol solution with a 12.5% PEG solution resultedin a rapid increase in TER.

Mannitol-enhanced transfection isconfined to adherent cells

To investigate whether the loosening of tight cell–cellcontacts is essential for the observed enhancement bymannitol treatment we also subjected CaCo-2, anotheradherent, epithelial cell line, and K562, a suspension cellline, to transfection with or without mannitol/PEGtreatment. As presented in Figure 6, pretreatment ofK562 cells with mannitol resulted in a decreasedtransfection rate irrespective of the presence of PEG.The sole presence of PEG gave only slightly (two times)higher transfection rates.In contrast, transfection of the adherent CaCo-2 cell

line showed a similar enhancement as was observed withOMEC II cells. Applying the mannitol/PEG protocol fortransfection of confluent CaCO-2 cells gave a 14-foldhigher luciferase expression rate as compared to simpletransfection with PEI 25 (Figure 6).

Mannitol can be replaced by otheragents which cause tight junctionloosening

We next examined whether the loosening of cell contactswould be sufficient to obtain an enhancing effect. Wetherefore used EDTA, trypsin or a combination of both toloosen up the tight monolayer of confluent cells. Asshown in Figure 7, the same enhancing effect wasobtained with the alternative treatments thus confirmingits independence of the chemical properties of mannitol.

Effect of mannitol/PEG on thetransfection of growing epithelial cellcultures

The effect of PEG, mannitol or a combination of bothagents was also tested on subconfluent, growing mono-layers of OMEC II cell cultures (Table 1). Mannitoltreatment of growing cells resulted in a decrease intransfection rate. Addition of PEG (without mannitol)gave increased transfection rates. The combination ofDNA/PEI25 complexes with mannitol and PEG gave

Figure 1. Fluorescence microscope images of OMEC II cells24 h after transfection with pEGFP and PEI 25. (A) 50% sub-confluent or (B) 100% confluent OMEC II monolayers weretransfected with PEI 25/DNA complexes. (C) 100% confluentOMEC II monolayers were treated with 7% mannitol for 1 hbefore transfection and 12.5% PEG was added to the PEI25/DNA transfection mixture. Bar=40 mm



reporter gene expression far higher than that obtainedwith any other transfection method tested (see Table 1).

PEG increases the size of thetransfection particles

To explore the effects that PEG has on DNA complexes,we determined the size of the particles in the presence orabsence of PEG. Table 2 shows that the addition of PEGresulted in a seven-fold increase of the size of transfectionparticles. The increase seems to occur immediately afterPEG addition, and the size was stably maintained duringprolonged incubation of the mixture up to 80 min. It haspreviously been shown that complexes with increased sizeshow enhanced binding to cells and higher transfectionactivity [24].

Influence of PEG on binding of DNAcomplexes to OMEC II cells

We next examined the influence of PEG on the association

of DNA/PEI 25 complexes with ovine mammary epithelial

cells. Rhodamine-labeled DNA (GeneGrip) was used for

the formation of transfection complexes. Confluent

monolayers of OMEC II cells were transfected and DNA

associated with the cells was monitored using fluores-

cence microscope analysis.After transfection without PEG some spots containing

large complexes alternated with large areas devoid of any

complexes. Inclusion of PEG resulted in a more regular

distribution of complexes over the cell layer and in a

significantly enhanced association of DNA with OMEC II

cells (Figure 8).

Figure 3. Transfection of 100% confluent monolayers of OMEC II cells. Transfection complexes consisted of luciferase plasmidDNA and various transfection agents: PEI 25, PEI 800, DOSPER, Lipofectamin (LA), hTf-pLys, and adenovirus-enhanced genetransfer (AVET) are shown. Open bars represent luciferase light units obtained after transfection without additional treatment;hatched bars, after mannitol treatment; grey bars, in the presence of 12.5% PEG; and solid bars, the combination of mannitoland PEG. PEI 25 was also used to transfect confluent monolayers grown on Transwells (PEI 25/TW). Luciferase activity 24 hafter transfection is shown

Figure 2. Enhancement of transfection by polyethyleneglycol. OMEC II cells were pretreated with mannitol and transfectedwith PEI 25/CMV-Luc complexes in the presence of PEG 8000 at the indicated concentrations. Luciferase activity was deter-mined 20 h after transfection and is expressed as relative light units per microgram of protein



Discussion

Most of the well-established gene transfer techniques

promote transport of DNA into subconfluently growing

cells much more efficiently than into confluent quiescent

cells. For many somatic gene transfer applications,

however, differentiated cells such as present in epithelial

tissues are the target. In the present study we tried to

improve in vitro transfection of confluent epithelial cells.Initially we used complexes consisting of reporter

Figure 4. Mannitol treatment causes a visible loosening of tight contacts between cells. Confluent monolayers of OMEC II cellsare shown before (A) and after (B) treatment with a 7% mannitol solution for 1 h. Consecutively, the mannitol solution wasreplaced by normal medium (C), or medium containing 12.5% PEG (D), and photographs were taken after 30 min. Bar=40 mm

Figure 5. Transepithelial electrical resistance (TER) is reduced by mannitol. The controls, 100% confluent OMEC II cells andTranswells without cells gave the maximal and minimal electrical resistance values, respectively (left section, hatched bars).Tissue culture medium was replaced by a 9% mannitol solution and the TER was measured every 3 min. After 75 min thelower chamber was filled with culture medium (TCM) and a 12.5% PEG solution in transfection medium (SFM) was added tothe upper chamber. TER measurement was continued for another 10 min (solid bars). The minimal (no cells) and maximal(confluent monolayer) electrical resistance values for Transwells containing the PEG solution are shown in the right most sec-tion of the figure (hatched bars)



plasmid DNA and PEI 25 to transfect OMEC II, an ovine

mammary epithelial cell line, in vitro. Our new method

included the pretreatment of the cells with hyperosmotic

mannitol solutions to increase their competence for DNA

uptake. Osmotic shock to improve gene transfer has

already been described [25–27], however these protocols

are different in that glycerol or DMSO is given to the cells

after transfection. More recently, Budker et al. [28]

described a ten-fold increase in transfection efficiency

when plasmid DNA in a 20% mannitol solution was

introduced in vivo into hepatocytes as compared with

isotonic solution.In our in vitro studies, confluent epithelial cell

monolayers were incubated with hyperosmotic mannitol

solutions prior to transfection, resulting in a 3–4-fold

enhancement in transfection efficiency.

During the mannitol treatment loosening of the tightly

packed epithelial layer was observed. Applying optimized

mannitol concentration for an optimized time, trans-

parent gaps became apparent between the cells, but no

cells were detached from the surface by the treatment.

The transepithelial electrical resistance measured in

OMEC II monolayers cultured on Transwells constantly

declined during the mannitol treatment.This process was fully reversible and 2–4 h after

replacement of the hyperosmotic solution with normal

growth medium the tightly packed layer had reformed.

The presence of 12.5% PEG in the medium shortened the

time required for reformation to a few minutes

(Figure 5).Basically, the loosening of tight junctions could cause

an increase in accessible surface or the exposure of

Figure 6. Transfection of CaCo-2 and K562 cells. The adherent epithelial carcinoma cell line, CaCo-2, and the erythroleukemiasuspension cell line, K562, were transfected with complexes containing 4 mg/ml luciferase plasmid DNA and 3.6 mg/ml PEI 25.Mannitol pretreatment and PEG addition are indicated. Luciferase activity 20 h after transfection is shown

Figure 7. Loosening of cell–cell contacts by alternative pretreatment. 100% confluent monolayers of OMEC II cells were trans-fected with luciferase plasmid and PEI 25. Pretreatment with trypsin solution was done for 2 min; with 0.2 mM EDTA for4 min, or with both agents for 2 min. PEG was included in the transfection mixture where indicated



receptors that are otherwise hidden. Alternatively, theobserved enhancement of transfection efficiency could beexplained by the inclusion and engulfment of DNA/PEIparticles during the process of tight junction reformation.Mannitol treatment had no effect on the transfection

efficiency of the suspension cell line, K562, but treatmentof CaCo-2, another epithelial cell line which forms tightmonolayers, also resulted in enhanced reporter geneexpression. This indicates that the loosening of thecontacts between the cells is crucial for the observedeffects.To further test the view that not mannitol itself but the

tight junction loosening is responsible for the observedeffects, confluent monolayers of OMEC II cells weretreated with EDTA, trypsin or both. The same looseningwas observed (though after shorter time periods), andalso the enhancement of transfection efficiency wasmaintained (Figure 7). EDTA and trypsin clearly do notexert osmotic forces on the cells, confirming thesignificance of the physical change in the cell monolayer.Transfection rates were further improved by adding

PEG to the transfection mixture giving a very reliable5–6-fold, and sometimes up to 30-fold, enhancement oftransfection rates (Figure 3). PEG seems to exert morethan one effect, acting on both the cells and thetransfection complexes. It has been described previouslythat improved sedimentation of DNA/PEI complexes ontothe cells can strongly enhance gene transfer [29].

Suspecting that PEG addition might accelerate sedimen-tation of the DNA complexes, we measured the size ofDNA/PEI 25 complexes in the presence or absence of

PEG. Immediately (1–2 min) after the addition of PEG thesize of the complexes was seven-fold larger than withoutPEG (Table 2), and no further growth was observed. This

increase in particle size indeed caused a ’precipitation’ ofthe complexes onto the cell monolayer as judged byvisual examination.PEG is used in yeast transfection protocols where it

probably enhances binding of DNA to the cell membraneand exerts some fusogenic activity [30,31]. We thereforeexamined whether PEG would also enhance the associa-

tion of DNA/PEI 25 complexes with mammalian cells.Association of DNA complexes with OMEC II cells wasvisualized by transfection with Rhodamin-labeled plas-

mid DNA. In the absence of PEG, large areas free of anyDNA/PEI 25 complexes were seen interrupted by spotswith a high concentration of transfection complexes

(Figure 8). The presence of PEG significantly enhancedthe amount of DNA that was associated with the cells,and the complexes were evenly distributed all over the

cell layer.Different transfection agents were compared in combi-

nation with the mannitol and PEG treatments. Whencationic lipids were used for DNA condensation and

transfection, mannitol treatment alone resulted in theexpected enhancement but addition of PEG failed tofurther improve transfection efficiency. Conversely,

addition of PEG dramatically diminished gene transfer(Figure 3). We also used adenovirus-enhanced genetransfer [20] in combination with PEG treatment. The

DNA complexes contain inactivated adenovirus particlesthat are known to efficiently support escape fromendosomes where they accumulate after receptor-

mediated transfection. Combining AVET with mannitoltreatment resulted in a two-fold enhancement, probablydue to better accessibility of receptors. Addition of

PEG did not show significant effects when combinedwith AVET.The mannitol/PEG method was also tested on growing

cultures (Table 1). Mannitol treatment alone usually

resulted in a decreased transfection rate. Growing cells

seem to be more sensitive to the relatively harsh mannitol

treatment, resulting in the observed decrease in reporter

gene expression. Conversely, PEG addition alone showed

Table 1. Subconfluent monolayers of OMEC II cells (50–70% con-fluency) were transfected with complexes consisting of luciferaseplasmid DNA and PEI 25 or PEI 800 at standard concentrations,or with other transfection agents. Mannitol pretreatment,addition of PEG or a combination of both are indicated. Theresults shown are the average luciferase values [with standarddeviation (SD) given as a percentage value] obtained for opti-mized DNA complex conditions for each method

RLU/mg protein (SD as a percentage)

PEI 25 1 650 500 (0.1)+Mannitol pretreatment 582 000 (8.6)+12.5% PEG 4 947 000 (0.1)+Mannitol/PEG 5 595 000 (0.4)

PEI 800 175 000 (5.1)+Mannitol pretreatment 56 000 (8.9)+12.5% PEG 2 515 000 (11.1)+Mannitol/PEG 3 278 000 (2.0)

DOSPER 290 000 (9.0)Lipofectamin 99 000 (0.4)HTf-pLys 80 (12.5)

Table 2. Plasmid DNA (3.2 mg) in 200 ml H2O (or SFM) were mixed with 2.9 mg PEI 25 in 200 ml H2O (or SFM), and after 20 min400 ml H2O (or SFM) containing 0, 12.5 or 25% PEG were added and complex size was measured with QELS immediately (1 min)or after the indicated time points

Time (min)

Influence of PEG on the size of DNA/PEI 25 complexes in nm (SD)

Water Water+6.25% PEG Water+12.5% PEG SFM SFM+12.5% PEG

1 53 (1) 157 (4) 344 (10) 252 (2) 1844 (58)20 50 (1) 159 (2) 342 (18) 344 (3) ND50 52 (6) 156 (9) 350 (4) 396 (13) ND80 53 (5) 160 (7) 369 (5) ND ND

ND, Not done; SD, standard deviation.



up to three-fold improvement, and surprisingly combi-

ning mannitol/PEG treatment with PEI 25 mediated

transfection gave higher transfection rates than obtained

by any other method tested. An explanation for the

improvement observed in growing cultures might be that

epithelial cells grow in small colonies rather than as single

cells. Mannitol treatment would loosen the tight junctions

between the cells in these colonies.Our current hypothesis is that PEG accelerates sedi-

mentation of DNA/PEI complexes onto the cells and into

the intercellular gaps resulting from the mannitol

treatment. In addition, PEG accelerates reformation of

tight junctions enhancing the uptake of the transfection

complexes. The enhanced association of the complexed

DNA with the cells in the presence of PEG may be an

additional contribution to the overall improvement.In summary, we have developed a simple and

convenient gene transfer technique for quiescent epithe-

lial cell monolayers. The transfection efficiencies that can

be obtained using the described technique approach or

even exceed those of much more expensive methods. The

improved technique can be combined with various

transfection agents, however its application is restricted

to adherent cells. The introduction of a new principle,

namely pretreating mammalian cells to increase their

competence for DNA uptake, may stimulate the develop-

ment of similar approaches in the future.

Acknowledgement

This work was supported by the Austrian FWF, project number

507 405-MOB.

References

1. Andreadis ST, Roth CM, Le Doux JM, et al. Large-scaleprocessing of recombinant retroviruses for gene therapy.Biotechnol Prog 1999; 15: 1–11.

2. Bueler H. Adeno-associated viral vectors for gene transfer andgene therapy. Biol Chem 1999; 380: 613–622.

3. Kochanek S. High-capacity adenoviral vectors for gene transferand somatic gene therapy. Hum Gene Ther 1999; 10:2451–2459.

4. Clark PR, Hersh E. Cationic lipid-mediated gene transfer:Current concepts, current opinion. Mol Ther 1999; 1: 158–176.

5. Mahato RI, Smith LC, Rolland A. Pharmaceutical perspectives ofnonviral gene therapy. Adv Genet 1999; 41: 95–156.

6. Behr JP. Gene transfer with synthetic cationic amphiphiles:prospects for gene therapy. Bioconjug Chem 1994; 5: 382–389.

7. Budker V, Hagstrom JE, Lapina O, et al. Protein/amphipathicpolyamine complexes enable highly efficient transfection withminimal toxicity. Biotechniques 1997; 23: 142–147.

8. Lee RJ, Huang L. Lipidic vector systems for gene transfer. CritRev Ther Drug Carrier Syst 1997; 14: 173–206.

9. Li S, Wu SP, Whitmore M, et al. Effect of immune response ongene transfer to the lung via systemic administration of cationiclipidic vectors. Am J Physiol 1999; 276: L796–L804.

10. Ropert C. Liposomes as a gene delivery system. Braz J Med BiolRes 1999; 32: 163–169.

11. Templeton NS, Lasic DD. New directions in liposome genedelivery. Mol Biotechnol 1999; 11: 175–180.

12. Zauner W, Brunner S, Buschle M, et al. Differential behaviour of

Figure 8. Influence of PEG on association of DNA/PEI 25 complexes with OMEC II cells. 100% confluent monolayers of OMECII cells were transfected with Rhodamin-labeled DNA complexed with PEI 25 in the absence (C) or presence (D) of 12.5% PEG.Four hours after the addition of transfection complexes the cells were washed six times with ice-cold PBS, fixed with 2% parafor-maldehyde and examined for associated complexes using a fluorescence microscope. Phase-contrast photographs of the identi-cal sectors as in (C) and (D) are shown in (A) and (B), respectively. Bar=40 mm



lipid based and polycation based gene transfer systems intransfecting primary human fibroblasts: a potential role ofpolylysine in nuclear transport. Biochim Biophys Acta 1999;1428: 57–67.

13. Boussif O, Lezoualc’h F, Zanta MA, et al. A versatile vector forgene and oligonucleotide transfer into cells in culture and invivo: polyethylenimine. Proc Natl Acad Sci U S A 1995; 92:7297–7301.

14. Abdallah B, Hassan A, Benoist C, et al. A powerful nonviralvector for in vivo gene transfer into the adult mammalian brain:polyethylenimine. Hum Gene Ther 1996; 7: 1947–1954.

15. Goula D, Benoist C, Mantero S, et al. Polyethylenimine-basedintravenous delivery of transgenes to mouse lung. Gene Ther1998; 5: 1291–1295.

16. Goula D, Remy JS, Erbacher P, et al. Size, diffusibility andtransfection performance of linear PEI/DNA complexes in themouse central nervous system. Gene Ther 1998; 5: 712–717.

17. Ringenbach L, Bohbot A, Tiberghien P, et al. Polyethylenimine-mediated transfection of human monocytes with the IFN-gammagene: an approach for cancer adoptive immunotherapy. GeneTher 1998; 5: 1508–1516.

18. Pollard H, Remy JS, Loussouarn G, et al. Polyethylenimine butnot cationic lipids promotes transgene delivery to the nucleus inmammalian cells. J Biol Chem 1998; 273: 7507–7511.

19. Wagner E, Cotten M, Mechtler K, et al. DNA-binding transferrinconjugates as functional gene-delivery agents: synthesis bylinkage of polylysine or ethidium homodimer to the transferrincarbohydrate moiety. Bioconjug Chem 1991; 2: 226–231.

20. Wagner E, Zatloukal K, Cotten M, et al. Coupling of adenovirusto transferrin-polylysine/DNA complexes greatly enhancesreceptor-mediated gene delivery and expression of transfectedgenes. Proc Natl Acad Sci U S A 1992; 89: 6099–6103.

21. Cotten M, Saltik M, Kursa M, et al. Psoralen treatment ofadenovirus particles eliminates virus replication and transcrip-

tion while maintaining the endosomolytic activity of the viruscapsid. Virology 1994; 205: 254–261.

22. Duechler M, Schmoll F, Pfneisl F, et al. OMEC II: a new ovinemammary epithelial cell line. Biol Cell 1998; 90: 199–205.

23. Plank C, Zatloukal K, Cotten M, et al. Gene transfer intohepatocytes using asialoglycoprotein receptor mediated endo-cytosis of DNA complexed with an artificial tetra-antennarygalactose ligand. Bioconjug Chem 1992; 3: 533–539.

24. Ogris M, Steinlein P, Kursa M, et al. The size of DNA/transferrin-PEI complexes is an important factor for gene expression incultured cells. Gene Ther 1998; 5: 1425–1433.

25. Kluxen FW, Lubbert H. Maximal expression of recombinantcDNAs in COS cells for use in expression cloning. Anal Biochem1993; 208: 352–356.

26. Li S, Thacker J. High-efficiency stable DNA transfection usingcationic detergent and glycerol. Biochem Biophys Res Commun1997; 231: 531–534.

27. Lopata MA, Cleveland DW, Sollner-Webb B. High level transientexpression of a chloramphenicol acetyl transferase gene byDEAE-dextran mediated DNA transfection coupled with adimethyl sulfoxide or glycerol shock treatment. Nucleic AcidsRes 1984; 12: 5707–5717.

28. Budker V, Zhang G, Knechtle S, et al. Naked DNA deliveredintraportally expresses efficiently in hepatocytes. Gene Ther1996; 3: 593–598.

29. Boussif O, Zanta MA, Behr JP. Optimized galenics improvein vitro gene transfer with cationic molecules up to 1000-fold.Gene Ther 1996; 3: 1074–1080.

30. Maggio B, Ahkong QF, Lucy JA. Poly(ethylene glycol), surfacepotential and cell fusion. Biochem J 1976; 158: 647–650.

31. Lentz BR, McIntyre GF, Parks DJ, et al. Bilayer curvature andcertain amphipaths promote poly(ethylene glycol)-inducedfusion of dipalmitoylphosphatidylcholine unilamellar vesicles.Biochemistry 1992; 31: 2643–2653.



Date post:	19-Nov-2023
Category:	Documents
Upload:	independent
View:	0 times
Download:	0 times

Transfection of epithelial cells is enhanced by combined treatment with mannitol and...

Documents