P

Dp

Aa

b

c

a

ARR1AA

KGPB

1

clC(t2evoAditMvt

K

0h

International Journal of Pharmaceutics 465 (2014) 11–17

Contents lists available at ScienceDirect

International Journal of Pharmaceutics

journa l homepage: www.e lsev ier .com/ locate / i jpharm

harmaceutical Nanotechnology

NA complexed with TAT peptide and condensed using calciumossesses unique structural features compared to PEI polyplexes

bdulgader A. Baouma,b, C. Russell Middaughb, Cory Berklandb,c,∗

Department of Chemistry, King Abdulaziz University, Jeddah, Saudi ArabiaDepartment of Pharmaceutical Chemistry, The University of Kansas, Lawrence, KS 66047, USADepartment of Chemical and Petroleum Engineering, The University of Kansas, Lawrence, KS 66047, USA

r t i c l e i n f o

rticle history:eceived 29 September 2013eceived in revised form7 December 2013ccepted 29 January 2014vailable online 7 February 2014

eywords:

a b s t r a c t

Complexes of the TAT peptide with plasmid DNA (pDNA) show unusually high transfection efficiencywhen condensed via “soft” calcium cross links. In this study, we characterize the structure of pDNA withinTAT complexes and compare it with that of branched polyethylenimine (PEI, 25 kDa) complexes. Dynamiclight scattering (DLS), second derivative ultraviolet absorption spectroscopy, extrinsic dye fluorescence,Fourier transform infrared (FTIR) spectroscopy, circular dichroism (CD) spectroscopy and differentialscanning calorimetry (DSC) were employed to access various aspects of the structure of the pDNA. TATcomplexes showed the highest transfection efficiency at an N/P ratio of 25 in A549 cells. FTIR and CD

ene deliveryolyelectrolyte complexesiophysical characterization

spectra of complexes demonstrated that the pDNA remained in the B conformation when associatedwith TAT or PEI. DSC showed that both TAT and PEI stabilized all forms of pDNA, with TAT increasing themelting temperature of pDNA compared to PEI complexes. Second derivative ultraviolet spectroscopyof TAT complexes showed a substantial reduction in the absorbance and an increase in the wavelengthsof the peaks of pDNA at N/P > 13; however, a clear correlation between pDNA structure and transfectionefficiency was not readily apparent.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

While a variety of vectors currently exist for gene translocation,ell penetrating peptides (CPPs) have become one of the most popu-ar and efficient agents for intracellular delivery of genetic material.PPs have successfully achieved intracellular delivery of proteinsFawell et al., 1994), nucleic acids (Chiu et al., 2004), small moleculeherapeutics (Rothbard et al., 2000), quantum dots (Santra et al.,005), and even MRI contrast agents (Lewin et al., 2000). Highlyfficient translocation mediated by CPPs has been confirmed in aariety of cell lines with minimal toxicity, overcoming challengesften faced with other delivery methods (Nagahara et al., 1998).dditionally, CPPs can be covalently or noncovalently attached torugs effectively towing their cargo into cells with no loss of biolog-

cal activity. Once inside the cell, many of these CPPs are also ableo localize to the nucleus, with or without their cargo (Kueltzo and

iddaugh, 2003). A major problem with the use of CPPs as non-iral gene vectors, however, is their low level of induction of generansfection compared to viral vectors. The problem is compounded

∗ Corresponding author at: The University of Kansas, 2030 Becker Drive, Lawrence,S 66047, USA. Tel.: +1 785 864 1455; fax: +1 785 864 1454.

E-mail address: berkland@ku.edu (C. Berkland).

378-5173/$ – see front matter © 2014 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.ijpharm.2014.01.040

by the lack of structural information for these complexes. Previ-ous studies have defined the structure of nonviral gene deliverycomplexes by a combination of methods such as dynamic light scat-tering (DLS), fluorescence, Fourier transform infrared spectroscopy(FTIR), circular dichroism (CD) and differential scanning calorime-try (DSC) (Braun et al., 2003; Braun et al., 2001; Choosakoonkrianget al., 2001b; Deng et al., 2000; Kawaura et al., 1998; Lobo et al.,2003a; Lobo et al., 2002; Wiethoff et al., 2002, 2003; Wiethoff andMiddaugh, 2001). These methods potentially provide information-rich data that may be used to describe gene delivery complexes andserve as a basis to design next generation gene vectors.

CPPs possess the physical and chemical characteristics desiredof gene delivery agents (Fonseca et al., 2009). For example, theirpolycationic, sometimes amphiphilic, nature has been shownto overcome one or more of the major biological barriers totransfection (e.g. cell entry, endosome escape, and nuclear local-ization). In addition, many of these peptides can complex pDNA,although somewhat inefficiently. Cationic lipoplex vectors havebeen extensively investigated by various physicochemical meth-ods in previous studies, but detailed characterization studies of CPP

containing “polyplex” vectors remain scarce.

TAT complexes containing “soft” calcium cross links haverecently shown promise by our group as efficient gene deliv-ery agents (Baoum et al., 2009). A method to synthesize small

1 rnal o

(mwdapot

peuLTTbTiCTwFt

2

2

oEfBwfeHppwwwRMAFULLW

2

psb2a

2

lNc

2 A.A. Baoum et al. / International Jou

60–175 nm) and stable TAT complexes was achieved by the for-ation of a ‘loose’ complex between TAT peptide and pDNA thatas then condensed by calcium. The TAT-Ca system was shown toisplay high transfection efficiency even in the presence of serum,nd negligible toxicity in vitro and in vivo compared to branched PEIolyplexes. The transfection efficiency of TAT complexes dependsn several factors including the nitrogen/phosphate (N/P) ratio andhe concentration of CaCl2.

Despite extensive previous efforts investigating the potential ofolycations to deliver plasmid DNA, correlations between the prop-rties of polyplexes and their ability to transfect cells are still poorlynderstood (Braun et al., 2005; Choosakoonkriang et al., 2003a;obo et al., 2003b; Rexroad et al., 2006; Ruponen et al., 2006).hus, we have conducted a thorough physical characterization ofAT complexes with the objective of establishing a relationshipetween their physical properties by comparison to PEI polyplexes.he secondary structure of pDNA upon complexation with TAT wasnvestigated by using FTIR, UV second derivative absorption andD spectroscopies. The thermal stability of pDNA complexed toAT was studied using DSC. In addition, fluorescence spectroscopyas used to probe the extent of pDNA condensation in complexes.

inally, light scattering studies were performed to assess the par-icle size of the complexes.

. Materials and methods

.1. Materials

Plasmid DNA encoding firefly luciferase (pGL3, 4.8 kbp) wasbtained from Promega (Madison, WI, USA) and transformed intoscherichia coli DH5 ´̨ (Invitrogen, Carls-bad, CA). A single trans-ormed colony picked from an agar plate was cultured in LB Brothase (Invitrogen) liquid for plasmid DNA preparation. Plasmid DNAas purified with Plasmid Giga Kit 5 (Qiagen, Germantown, MD)

ollowing the manufacturer’s instructions. All pDNA had purity lev-ls of 1.8 or greater as determined by UV/Vis inspection (A260/A280).IV-1 TAT (49–57) peptide (RKKRRQRRR; Mw = 1338.85 Da) wasurchased from Biomatik (Cambridge, Ontario, Canada). Branchedolyethylenimine (PEI, 25 kDa) was obtained from Aldrich (Mil-aukee, WI). Calcium chloride (CaCl2·2H2O) and agarose mediumere purchased from Fisher Scientific (Pittsburgh, PA). A549 cellsere obtained from the American Type Culture Collection (ATCC,ockville, MD). The cell culture medium (Ham’s F-12 Nutrientixture, Kaighn’s modified with L-glutamine) and BCATM Protein

ssay were purchased through Fisher Scientic (Pittsburgh, PA).etal bovine serum (FBS) was purchased from Hyclone (Logan,T). Penicillin-streptomycin was purchased from MB Biomedical,LC (Solon, OH) and Trypsin-EDTA from Gibco (Carlsbad, CA). Theuciferase Assay System was obtained from Promega (Madison,

I).

.2. Preparation of complexes

Complexes were formed by addition of appropriate volumes ofDNA and TAT with or without CaCl2 (15 �L of 113 mM CaCl2) or PEIolutions calculated to produce the desired molar ratio, followedy a 20 min incubation at 4 ◦C prior to use (Baoum and Berkland,011). Complexes were freshly prepared before each individualnalysis.

.3. Particle size and zeta potential measurement

The effective hydrodynamic diameter of the complexes was ana-yzed using DLS employing a Brookhaven Instrument (Holtsville,Y) equipped with a 50 mW HeNe laser operating at 532 nm. Theomplexes were prepared at a constant pDNA concentration of

f Pharmaceutics 465 (2014) 11–17

100 �g/mL while the N/P ratios of the complexes were varied. Thescattered light was monitored at 90◦ to the incident beam. Foreach sample the data were collected continuously for three 1-minintervals. The mean diameter of the complexes was obtained fromthe diffusion coefficient by the Stokes-Einstein equation using themethod of cumulants. Zeta potential measurements were obtainedby phase analysis light scattering using a Brookhaven Zeta PALSinstrument. The electrophoretic mobility of the samples was deter-mined from the average of 10 cycles of an applied electric field. Thezeta potential was determined from the electrophoretic mobilityfrom the Smoluchowski approximation.

2.4. UV absorption spectroscopy

UV absorption spectra of pDNA were obtained employing aHewlett-Packard 8453 UV-visible diode-array spectrophotometer(Agilent, Palo Alto, CA). A pDNA concentration of 0.025 �g/�L wasused. Second derivatives of the absorption spectra from 240 nm to300 nm were generated using a 9-point filter and third degree poly-nomial then fit to a spline function with 99 points of interpolationusing software supplied with the instrument. Peak positions weredetermined using Origin® software from MicrocalTM.

2.5. Transfection studies

The transfection efficiencies of the complexes were assessedusing plasmid DNA encoding firefly luciferase (pGL3, 4.8 kbp)in A549 human lung carcinoma epithelial cells. Cells weretrypsinized, counted and diluted to a concentration of approxi-mately 80,000 cells/mL. Then 0.1 mL of that preparation was addedto each well of a 96-well plate and the cells were incubated ina humidified atmosphere at 5% CO2 and 37 ◦C for 24 h. Immedi-ately before transfection, the cells were washed once with PBS and100 �L of sample (20% of complex to 80% of serum free cell cul-ture medium) was added to each well. Cells were incubated withthe complexes for 5 h. The transfection agent was then removed byaspiration and 100 �L of fresh serum medium was added followedby 24 h of incubation. The Luciferase Assay System from Promegawas used to determine gene expression following the manufac-turer’s recommended protocol. The light units were normalizedagainst protein concentration in the cellular extracts, which weremeasured using the BCATM Protein Assay. The transfection resultswere expressed as Relative Light Units (RLU) per mg of cellularprotein.

2.6. Fluorescence spectroscopy

The dye SYBR Green (Invitrogen, Carls-bad, CA) with an exci-tation wavelength of 497 nm and emission wavelength of 520 nmwas used in fluorescence studies. SYBR Green solution (concen-tration of 1 X dye) was added to the pDNA before complexformation. After a 30 min incubation, the complexes were preparedas described above. Fluorescence measurements were performedwith a QuantaMasterTM spectrofluorometer (PTI, Monmouth, NJ).The excitation and emission slits were set at 4 nm and spectra from480 to 590 nm were obtained by scanning at 1 nm intervals usinga 1 s integration time. The fluorescence intensity of dye alone inbuffer was subtracted from raw sample spectra.

2.7. Fourier transform infrared (FTIR) spectroscopy

FTIR spectra were obtained using a Nicolet Magna-IR 560 spec-

trometer equipped with a mercury cadmium telluride detector(Nicolet, Madison, WI). Samples were measured by an attenuatedtotal internal reflectance (ATR) method in which the sample in solu-tion was placed directly in a zinc selenide trough. Spectra were

rnal of Pharmaceutics 465 (2014) 11–17 13

ogae(wsv

2

cJqastFas

2

wcpmbttw

2

ohsta

3

3

t(

TTc

Fig. 1. The transfection efficiency of TAT and PEI (N/P 10) complexes as a function

A.A. Baoum et al. / International Jou

btained under dry air purge by accumulation of 256 interfero-rams at a final resolution of 4 cm−1. For ATR experiments, thessociation band of water near 2200 cm−1 was used as a refer-nce for subtraction of water from the spectra. Base-line correction1804 to 904 cm−1) and seven point Satvitsky-Golay smoothingere applied to all spectra. The final pDNA concentration of the

amples was 0.5 �g/�L while the polycations concentration wasaried.

.8. Circular dichroism (CD)

For CD measurements, complexes were formed at a final pDNAoncentration of 50 �g/mL. CD spectra were obtained using aasco J-720 spectropolarimeter (Easton, MD) in a 0.1 cm pathlengthuartz cuvette at 25 ◦C. Spectra were recorded from 350 to 200 nmt a scan rate of 20 nm/min and were analyzed as an average of threecans. The CD signal was converted to molar ellipticity [�] based onhe molar base concentration of the pDNA, smoothed with a Jascoast Fourier transform method, and then baseline adjusted to zerot 345 nm to correct for a small contribution by differential lightcattering.

.9. Differential scanning calorimetry (DSC)

Differential scanning calorimetry thermograms were obtainedith a VP-DSC (Microcal, Northampton, MA). Measurements

onsisted of a single scan from 0 to 120 ◦C at 1 ◦C/min under 3 atm ofressure. Samples were examined in triplicate and degassed beforeeasurement. Baselines were obtained by scanning with buffer in

oth the sample and reference cells. Data were analyzed by sub-racting the baseline from the sample thermogram, and convertinghe differential heat to molar heat capacity using the moleculareight and concentration of pDNA (0.4 �g/�L).

.10. Statistical analysis

Statistical evaluation of data was performed using an analysisf variance (one-way ANOVA). Newman–Keuls was used as a postoc test to assess the significance of differences. To compare theignificance of the difference between the means of two groups, a-test was performed; in all cases, a value of p < 0.05 was accepteds significant.

. Results

.1. Particle sizes and zeta potentials of TAT and PEI complexes

The mean hydrodynamic diameter and zeta potential ofhe complexes were determined by DLS and PALS, respectivelyTable 1). In general, adding 113 mM CaCl2 to TAT complexes at

able 1he diameter of TAT complexes as a function of N/P ratio with 113 mM CaCl2 con-entration. Results are presented as mean ± SD (n = 3).

N/P ratio EffectiveDiameter (nm)

Zeta potentials(mV)

4 171.8 ± 3.9 15.7 ± 0.89 165.1 ± 3.7 18.2 ± 1.913 133.2 ± 4.3 17.5 ± 2.018 121.0 ± 1.9 20.3 ± 0.325 77.3 ± 9.6 19.0 ± 1.433 55.1 ± 7.5 23.6 ± 1.6Control complexesTAT N/P 25 (no CaCl2) 1000.8 ± 5.2 11.2 ± 0.5PEI N/P 10 (no CaCl2) 73.7 ± 1.4 15.8 ± 1.0PEI N/P 10 (CaCl2) 1230.1 ± 8.5 26.0 ± 0.7

of N/P ratio with 113 mM CaCl2 concentration. Results are presented as mean ± SD(n = 3), *p < 0.001 for the various N/P ratios of TAT complexes as compared with TAT(N/P 25) complexes.

N/P ratios between 4 and 33 induced a substantial decrease inthe particle size (60–175 nm) with relatively narrow polydisper-sity. In comparison, PEI complexes possessed a small particle size(∼74 nm) with zeta potential (16 mV) in the absence of calcium. Thezeta potential of TAT complexes produced particles with a positivesurface charge (>18 mV).

3.2. In vitro transfection of A549 cells

The transfection efficiencies of TAT and PEI complexes wereevaluated in the human lung carcinoma cell line A549. Gene expres-sion levels were dependent on the N/P ratio of the TAT complex(Fig. 1). The gene expression of TAT complexes was enhanced bythe inclusion of 113 mM CaCl2 for the various N/P ratios. The high-est transfection efficiency of TAT complexes was seen at N/P 25.Conversely, negligible expression was observed for TAT complexeswithout CaCl2. In comparison, PEI had excellent transfection effi-ciency in the absence of CaCl2.

3.3. Second derivative UV absorption spectroscopy

The UV spectrum of pDNA displayed an absorption maximum at260 nm. Plasmid DNA second derivative analysis revealed numer-ous absorption peaks between 240 and 300 nm arising from thepDNA bases. Five positive peaks at approximately 256, 268, 277,287 and 296 nm, and five negative peaks at 253, 259, 271, 282and 291 nm were found (Fig. 2A–E). The positions of these peaksare sensitive to the immediate environment of the correspondingresidues. In general, the second derivative peaks of pDNA com-plexed with polycations display a shift to higher wavelengths whenthe polycations are in molar excess (Rexroad et al., 2006; Mach et al.,1992). TAT complexes in the absence and presence of 113 mM CaCl2showed a reduction in the absorbance and an increase in the wave-lengths of these peaks. Above an N/P of 13, TAT complexes includingcalcium showed a reduction in the second derivative peaks ofpDNA. Conversely, PEI complexes produced only small red shiftsin the position of these peaks. Including calcium in the PEI com-plexes formulation revealed a marked decrease in the absorbanceof the peaks. Free TAT, PEI and CaCl2 had no significant peaks overthe range of wavelengths examined.

3.4. Fluorescence studies

The condensation of pDNA by the cationic materials in the pres-ence and absence of 113 mM CaCl2 was studied using SYBR green

14 A.A. Baoum et al. / International Journal of Pharmaceutics 465 (2014) 11–17

Fig. 2. Second derivative UV spectra of (A) free TAT and PEI and (B and C) TAT and PEI complexes as a function of N/P ratio with and without 113 mM CaCl2. The positive peakp ainst

didfwpqDaPflpct

3

so2It

ositions (D) and the negative peak positions (E) of the pDNA bases were plotted ag

ye (Mano et al., 2005). This fluorophore is virtually nonfluorescentn solution but forms highly fluorescent complexes when bound toouble-stranded DNA. As expected, the fluorescence intensity ofree pDNA probed with SYBR green was significantly higher thanhen condensed by polycations (Fig. 3A and B). It has been shownreviously that the addition of cationic carriers to dye labeled DNAuenched the fluorescence of the dye without displacing it from theNA (Wong et al., 2001). The addition of TAT or PEI to pDNA showedsubstantial decrease in the SYBR green fluorescence intensity withEI complexes superior to TAT complexes in quenching SYBR greenuorescence. This implies that the pDNA was more tightly com-acted by the PEI. Including 113 mM CaCl2 in the formulation of theomplexes resulted in a negligible reduction in the fluorescence ofhe complexes.

.5. Infrared spectral properties of TAT and PEI complexes

FTIR spectroscopy was employed to investigate the secondarytructure of the pDNA component of complexes. The FTIR spectra

f pDNA and its complexes with PEI (N/P 10) and with TAT (N/P5) including 113 mM CaCl2 showed clear trends (Fig. 4A and B).

n the absence of cationic reagent, pDNA was in the B conforma-ion as indicated by the presence of the guanine/thymidine (G/T)

the N/P ratios.

carbonyl stretching band at 1711 cm−1, an asymmetric phosphatestretching vibration at 1217 cm−1 and a sugar-phosphate coupledvibration at 966 cm−1 (Braun et al., 2003; Choosakoonkriang et al.,2001a; Choosakoonkriang et al., 2001b). Addition of TAT or PEIto pDNA induced an increase in the peak frequency of the basecarbonyl stretching vibration and a decrease in the antisymmetricphosphate stretch (Fig. 4C–F). No change in the peak position of thesugar-phosphate coupled vibration was observed upon addition ofTAT to pDNA. In comparison, PEI induced a blue shift in the posi-tion of the vibration arising from the pDNA backbone. The positionof a pDNA ribose sugar vibration in both TAT and PEI complexes(1053 cm−1) shifted to lower frequency. Furthermore, no changein the peak position was observed for the symmetric phosphatevibration (∼1087 cm−1), which has been found to be independentof pDNA geometry (Choosakoonkriang et al., 2001a).

3.6. CD of TAT and PEI complexes

CD spectroscopy was also used to monitor the secondary struc-

ture of pDNA in TAT and PEI complexes. The polycations showedno intrinsic CD signal or ordered secondary structure within theUV region monitored. Therefore, the observed signals arose entirelyfrom the pDNA molecules. The CD spectrum of uncomplexed pDNA

A.A. Baoum et al. / International Journal o

Fa

dt(paeabiwstaec

3

empfwwitsdlb

pDNA remained in the B form although there were changes in

ig. 3. Extrinsic fluorescence spectra of TAT and PEI complexes in (A) the absencend (B) presence of 113 mM CaCl2.

emonstrated the characteristics of the native B form conforma-ion, a positive peak at 275 nm and a negative peak at 245 nmFig. 5). Upon TAT-Ca and PEI complexation, both regions of theDNA CD spectrum were altered. In general, the spectra of TAT-Cand PEI complexes showed a decrease in the value of the molarllipticity of the positive and negative bands, concomitant withred shift of the peak position of both the 275 nm and 245 nm

ands although not to the same degree. A decrease in the elliptic-ty and a shift in the position to higher wavelengths of both bands

ere observed after calcium was added to PEI complexes. Exclu-ion of calcium from the TAT complexes resulted in a change inhe overall shape of the spectrum, an increase in the 275 nm peaknd a decrease in the 245 nm band. Shifts in the peak position ofach band to higher wavelengths were also evident when excludingalcium.

.7. DSC of TAT and PEI complexes

The thermal stability of pDNA in TAT and PEI complexes wasvaluated using DSC (Fig. 6). Thermograms of pDNA displayed twoajor melting transitions. The first originated from supercoiled

DNA as a broad transition at ∼90 ◦C whereas the second aroserom the linear/open circular species as a series of small transitionsithin the 60◦-70 ◦C range (Lobo et al., 2002). All forms of pDNAere thermally stabilized when complexed to TAT (N/P 25) includ-

ng113 mM CaCl2 or to PEI (N/P 10). The 25 kDa PEI stabilized bothhe supercoiled (100 ◦C) and the linear/open circular (78 ◦C) pDNApecies at an N/P ratio of 10. The thermogram of TAT complexes

emonstrated stabilization of both the supercoiled (∼110 ◦C) and

inear/open circular form (∼95 ◦C) of pDNA with the thermal sta-ility of pDNA in TAT complexes superior to that of PEI complexes.

f Pharmaceutics 465 (2014) 11–17 15

4. Discussion

In our earlier studies, it was established that adding certainconcentrations of CaCl2 (e.g., 113 mM) to CPP complexes producedsmall nanoparticles (60–175 nm), leading to gene expression lev-els higher than those observed for the more commonly used PEIcomplexes (Baoum et al., 2009; Baoum and Berkland, 2011). Thetransfection efficiency of TAT complexes depended on the N/Pmolar ratio. Gene expression of TAT complexes formulated with113 mM CaCl2 was sustained for at least 10 days and was not influ-enced by the presence of serum. Moreover, CPPs showed negligiblecytotoxicity. The main goal of this study was to measure a widevariety of biophysical characteristics of pDNA in TAT complexesto better understand the structure of these particles and comparethem to pDNA in PEI complexes.

As observed in previous reports, pDNA remains in the B formwhen complexed with different amounts, molecular weights andforms of PEI (Choosakoonkriang et al., 2003a). The thermal stabil-ity of the supercoiled and the linear/open circular forms of pDNA,however, was observed to increase in the presence of a chargeexcess of branched PEI (25 kDa). A direct correlation betweenthe biophysical properties of PEI complexes and transfection effi-ciency was not, however, observed as had been previously reported(Choosakoonkriang et al., 2003a).

Six methods were selected for this study to probe various struc-tural aspects of the TAT and PEI complexes. DLS provided a directmeasurement of the size (Wiethoff and Middaugh, 2001). Thecondensation of pDNA induced by polycations was evaluated bydye-binding fluorescence spectroscopy (Ruponen et al., 2006). FTIRand CD have proven to be sensitive tools for characterizing thesecondary structure of pDNA in complexes (Braun et al., 2003;Choosakoonkriang et al., 2001a; Choosakoonkriang et al., 2003b;Choosakoonkriang et al., 2001b; Mach et al., 1992) with additionalstructural changes studied using UV second derivative absorptionspectroscopy (Mach et al., 1992). DSC was used to assess the ther-mal stability of pDNA within complexes (Braun et al., 2005).

A previous study has shown that the UV absorption spectra ofpDNA originates from the purine and pyrimidine bases since thesugar phosphate backbones of pDNA produce very little contribu-tion to its absorbance spectra above 200 nm (Mach et al., 1992).Nucleotide bases have very low symmetry and several unbondedelectrons. Due to the many different transitions that occur for eachbase in the spectral region from 200 to 300 nm (Freifelder, 1987),the second derivative spectra of pDNA is complex. The UV secondderivative spectra of polynucleotides is known to be dependenton the presence of secondary structure resulting in an exten-sively described hyper-chromic effect (Mach et al., 1992). Sincethe secondary structure does not change in this case, however, aninteraction between TAT or PEI and pDNA or a subtle change intheir internal environment is clearly indicated by the changes inthe position and the value of the absorbance of the pDNA secondderivative peaks. Despite these changes, the CD and FTIR resultslead us to conclude that the pDNA is maintained in the B formwhen complexed to various ratios of N/P TAT and PEI. Althoughthe B form was maintained, the pDNA is condensed into smallcomplexes based on the DLS data. Dye-based fluorescence studiesalso provide some indication that PEI compacted pDNA to a greaterextent than TAT. Surprisingly, the SYBR green fluorescence inten-sities were not perturbed upon the addition of 113 mM CaCl2 tothe TAT and PEI complexes although calcium clearly does cause asignificant decrease in the size of TAT complexes.

The FTIR spectra of TAT and PEI complexes showed that the

the position of the pDNA carbonyl and asymmetric phosphatevibrations. The increases in frequency of the base carbonyl vibra-tions suggested altered hydrogen bonding occurred within the

16 A.A. Baoum et al. / International Journal of Pharmaceutics 465 (2014) 11–17

F anda E) andc

bsifiwt

c

Fs

indicated a decreased interaction between bases in the complexes.

ig. 4. FTIR absorbance spectra of (A) PEI and (B) TAT complexes in solution withntisymmetric phosphate stretch (D), pDNA sugar-phosphate coupled vibration (omplexes.

ases. Also, the reduced frequency of the antisymmetric phosphatetretching vibration can be directly attributed to the electrostaticnteractions between polycations and pDNA. All these results con-rm interaction between polycation and pDNA where the pDNAas maintained in B form within the complexes but suggest induc-

ion of small changes in the structure of the pDNA.CD spectra of the TAT and PEI complexes also suggested some

hanges in the structure of pDNA upon complexation, as shown by

ig. 5. Effect of TAT and PEI in the absence and presence of 113 mM CaCl2 on the CDpectra of pDNA in the complexes.

without 113 mM CaCl2. The peak positions of the pDNA base carbonyl (C), pDNApDNA symmetric phosphate vibration (F) were plotted against the TAT and PEI

the changes in the pDNA CD peak positions and intensities. Previ-ous studies suggested that these CD changes were best explainedby limited local changes in the base/base interactions (Braun et al.,2005). Thus, the reduction of the value of the molar ellipticity likely

Moreover, such local structural perturbations may be due to a directinteraction between polycation and pDNA bases as observed inthe change seen in the carbonyl stretching region with FTIR. The

Fig. 6. Effect of TAT and PEI in the absence and presence of 113 mM CaCl2 on thethermal stability of pDNA in the complexes.

rnal o

sethfsatt((waf

olpoTatd

5

eshvpracsiwnpfftrc

R

B

B

B

B

B

A.A. Baoum et al. / International Jou

pectrum for TAT complexes including calcium supported suchxplanations, yielding a reduction in the molar ellipticity comparedo the spectrum for TAT complexes. In addition, a previous studyas shown that the CD spectra of Ca/pDNA complexes using dif-

erent concentrations of calcium, in general, were similar to thepectrum of pDNA alone with a slight decrease in the ellipticityt 275 nm (Patil et al., 2005). Further increase in calcium concen-ration beyond ∼25.9 mM did not affect the spectrum. Comparinghese results suggested that calcium interacted with both aminesTAT) and phosphates (pDNA), which we have described previouslyBaoum et al., 2009). The spectral changes seen are also consistentith absorption flattening but since they occur as the complexes

re becoming smaller, this does not seem a probable explanationor the results.

DSC studies of TAT and PEI complexes confirmed an effectf complexation on the thermal stability of the supercoiled andinear/open circular pDNA components. Complexation by botholycations increased the melting temperature of all the formsf pDNA suggesting stabilization of the various pDNA structures.AT complexes stabilized pDNA more than PEI complexes presum-bly as a result of the ability of calcium to selectively condensehe TAT complexes. Thus, all of the analyses suggest some discreteifferences in pDNA structure within TAT and PEI complexes.

. Conclusion

Complexes of CPPs such as TAT with pDNA were previouslystablished as a potent gene delivery vector when condensed intomall complexes using calcium. The simplicity of the formulation,igh levels of transfection efficiency and negligible effect on theiability of host cells inspired us to investigate the structure ofDNA within TAT complexes. Calcium concentration and N/P molaratio were previously varied to maximize transfection efficiencynd an optimized TAT complex was studied here. Extrinsic fluores-ence studies provided evidence of compaction of pDNA, with PEIeemingly inducing ‘tighter’ complexes. FTIR and CD spectroscopiesndicated that the secondary structure of pDNA was stabilized

ithin the complexes. Nevertheless, DSC studies of pDNA compo-ents in TAT and PEI complexes demonstrated an enhancement inDNA thermal stability. These data in combination with the trans-ection efficacy and the low cytotoxicity of TAT complexes supporturther exploration and perhaps new design schemes based onhis novel nonviral gene delivery vector. Nevertheless, the exactelationship between the structure of the complexes and their effi-iency as gene delivery vehicles remains elusive.

eferences

aoum, A., Xie, S.-X., Fakhari, A., Berkland, C., 2009. “Soft” calcium crosslinksenable highly efficient gene transfection using TAT peptide. Pharm. Res. 26,2619–2629.

aoum, A.A., Berkland, C., 2011. Calcium condensation of DNA complexed with cell-penetrating peptides offers efficient, noncytotoxic gene delivery. J. Pharm. Sci.100, 1637–1642.

raun, C.S., Jas, G.S., Choosakoonkriang, S., Koe, G.S., Smith, J.G., Middaugh, C.R.,2003. The structure of DNA within cationic lipid/DNA complexes. Biophys. J.84, 1114–1123.

raun, C.S., Kueltzo, L.A., Middaugh, C.R., 2001. Ultraviolet absorption and circu-

lar dichroism spectroscopy of nonviral gene delivery complexes. Methods Mol.Med. 65, 253–284.

raun, C.S., Vetro, J.A., Tomalia, D.A., Koe, G.S., Koe, J.G., Russell Middaugh, C., 2005.Structure/function relationships of polyamidoamine/DNA dendrimers as genedelivery vehicles. J. Pharm. Sci. 94, 423–436.

f Pharmaceutics 465 (2014) 11–17 17

Chiu, Y.-L., Ali, A., Chu, C.-y., Cao, H., Rana, T.M., 2004. Visualizing a correlationbetween siRNA localization, cellular uptake, and RNAi in living cells. Chem. Biol.11, 1165–1175.

Choosakoonkriang, S., Lobo, B.A., Koe, G.S., Koe, J.G., Middaugh, C.R., 2003a. Biophys-ical characterization of PEI/DNA complexes. J. Pharm. Sci. 92, 1710–1722.

Choosakoonkriang, S., Wiethoff, C.M., Koe, G.S., Koe, J.G., Anchordoquy, T.J., Mid-daugh, C.R., 2003b. An infrared spectroscopic study of the effect of hydration oncationic lipid/DNA complexes. J. Pharm. Sci. 92, 115–130.

Choosakoonkriang, S., Wiethoff, C.M., Anchordoquy, T.J., Koe, G.S., Smith, J.G., Mid-daugh, C.R., 2001a. Infrared spectroscopic characterization of the interaction ofcationic lipids with plasmid DNA. J. Biol. Chem. 276, 8037–8043.

Choosakoonkriang, S., Wiethoff, C.M., Kueltzo, L.A., Middaugh, C.R., 2001b. Charac-terization of synthetic gene delivery vectors by infrared spectroscopy. MethodsMol Med. 65, 285–317.

Deng, H., Bloomfield, V.A., Benevides, J.M., Thomas Jr., G.J., 2000. Structural basisof polyamine–DNA recognition: spermidine and spermine interactions withgenomic B-DNAs of different GC content probed by Raman spectroscopy. NucleicAcids Res. 28, 3379–3385.

Fawell, S., Seery, J., Daikh, Y., Moore, C., Chen, L.L., Pepinsky, B., Barsoum, J., 1994.Tat-mediated delivery of heterologous proteins into cells. Proc. Natl. Acad. Sci.USA 91, 664–668.

Fonseca, S.B., Pereira, M.P., Kelley, S.O., 2009. Recent advances in the use of cell-penetrating peptides for medical and biological applications. Adv. Drug Deliv.Rev. 61, 953–964.

Freifelder, D., 1987. Microbial Genetics. Jones and Bartlett, Boston.Kawaura, C., Noguchi, A., Furuno, T., Nakanishi, M., 1998. Atomic force microscopy

for studying gene transfection mediated by cationic liposomes with a cationiccholesterol derivative. FEBS Lett. 421, 69–72.

Kueltzo, L.A., Middaugh, C.R., 2003. Nonclassical transport proteins and peptides:an alternative to classical macromolecule delivery systems. J. Pharm. Sci. 92,1754–1772.

Lewin, M., Carlesso, N., Tung, C.-H., Tang, X.-W., Cory, D., Scadden, D.T., Weissleder,R., 2000. Tat peptide-derivatized magnetic nanoparticles allow in vivo trackingand recovery of progenitor cells. Nat. Biotechnol. 18, 410–414.

Lobo, B.A., Koe, G.S., Koe, J.G., Middaugh, C.R., 2003a. Thermodynamic analysis ofbinding and protonation in DOTAP/DOPE (1: 1): DNA complexes using isother-mal titration calorimetry. Biophys. Chem. 104, 67–78.

Lobo, B.A., Rogers, S.A., Choosakoonkriang, S., Smith, J.G., Koe, G., Middaugh, C.R.,2002. Differential scanning calorimetric studies of the thermal stability ofplasmid DNA complexed with cationic lipids and polymers. J. Pharm. Sci. 91,454–466.

Lobo, B.A., Vetro, J.A., Suich, D.M., Zuckermann, R.N., Middaugh, C.R., 2003b. Struc-ture/function analysis of peptoid/lipitoid: DNA complexes. J. Pharm. Sci. 92,1905–1918.

Mach, H., Russell Middaugh, C., Lewis, R.V., 1992. Detection of proteins and phenol inDNA samples with second-derivative absorption spectroscopy. Anal. Biochem.200, 20–26.

Mano, M., Teodósio, C., Simoes, S., Pedroso, d.L.M., Paiva, A., 2005. On the mech-anisms of the internalization of S413-PV cell-penetrating peptide. Biochem. J.390, 603–612.

Nagahara, H., Vocero-Akbani, A.M., Snyder, E.L., Ho, A., Latham, D.G., Lissy, N.A.,Becker-Hapak, M., Ezhevsky, S.A., Dowdy, S.F., 1998. Transduction of full-lengthTAT fusion proteins into mammalian cells: TAT-p27Kip1 induces cell migration.Nat. Med. 4, 1449–1452.

Patil, S.D., Rhodes, D.G., Burgess, D.J., 2005. Biophysical characterization of anioniclipoplexes. Biochim. Biophys. Acta Biomembr. 1711, 1–11.

Rexroad, J., Evans, R.K., Middaugh, C.R., 2006. Effect of pH and ionic strength on thephysical stability of adenovirus type 5. J. Pharm. Sci. 95, 237–247.

Rothbard, J.B., Garlington, S., Lin, Q., Kirschberg, T., Kreider, E., McGrane, P.L., Wender,P.A., Khavari, P.A., 2000. Conjugation of arginine oligomers to cyclosporin A facil-itates topical delivery and inhibition of inflammation. Nat. Med. 6, 1253–1257.

Ruponen, M., Braun, C.S., Middaugh, C.R., 2006. Biophysical characterization of poly-meric and liposomal gene delivery systems using empirical phase diagrams. J.Pharm. Sci. 95, 2101–2114.

Santra, S., Yang, H., Stanley, J.T., Holloway, P.H., Moudgil, B.M., Walter, G., Mericle,R.A., 2005. Rapid and effective labeling of brain tissue using TAT-conjugated CdS:Mn/ZnS quantum dots. Chem. Commun., 3144–3146.

Wiethoff, C.M., Gill, M.L., Koe, G.S., Koe, J.G., Middaugh, C.R., 2002. The structuralorganization of cationic lipid-DNA complexes. J. Biol. Chem. 277, 44980–44987.

Wiethoff, C.M., Gill, M.L., Koe, G.S., Koe, J.G., Middaugh, C.R., 2003. A fluorescencestudy of the structure and accessibility of plasmid DNA condensed with cationicgene delivery vehicles. J. Pharm. Sci. 92, 1272–1285.

Wiethoff, C.M., Middaugh, C.R., 2001. Light-scattering techniques for characteriza-

tion of synthetic gene therapy vectors. Methods Mol. Med. 65, 349–376.

Wong, M., Kong, S., Dragowska, W.H., Bally, M.B., 2001. Oxazole yellow homodimerYOYO-1-labeled DNA: a fluorescent complex that can be used to assess structuralchanges in DNA following formation and cellular delivery of cationic lipid DNAcomplexes. Biochim. Biophys. Acta Gen. Subjects 1527, 61–72.