A comparative study on the effect of Curcumin and Chlorin-p6 on the transport of the LDS cation across a negatively charged POPG bilayer: Effect of pH

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 173 (2017) 132–138

Contents lists available at ScienceDirect

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa

A comparative study on the effect of Curcumin and Chlorin-p6 on the transport of the LDS cation across a negatively charged POPG bilayer: Effect of pH G.K. Varshney, S.R. Kintali, P.K. Gupta, K. Das ⁎ Optical Spectroscopy & Diagnostic Lab, Laser Bio-Medical Applications & Instrumentation Division, Raja Ramanna Center for Advanced Technology, Indore, M.P. 452013, India

a r t i c l e

i n f o

Article history: Received 15 March 2016 Received in revised form 28 August 2016 Accepted 1 September 2016 Available online 03 September 2016 Keywords: Interfacial second harmonic generation Membrane transport

a b s t r a c t We report the use of interface selective Second Harmonic generation technique to investigate the transport of the LDS cation across POPG liposomes in the pH range of 4.0 to 8.0 in the presence and absence of two amphiphilic drugs, Curcumin and Chlorin-p6 (Cp6). Our results show that bilayer permeability of liposomes is significantly affected by the presence of the drugs and pH of the medium as evidenced by significant changes in the transport kinetics of the LDS. Studies carried out in the pH range 4.0–8.0 show that while Cp6 significantly enhanced the transport of LDS at pH 4.0, the transport of the cation was seen to increase with increasing pH, with maximum effect at pH 7.4 for Curcumin. The pH dependent bilayer localization of both the drugs was investigated by conducting steady state FRET studies using DPH labeled lipids as donors. The FRET results and the relative population of the various ionic/nonionic species of the drugs at different pH suggest that distance dependent interaction between the various ionic species of the drugs and polar head groups of the lipid is responsible for the observed pH dependence enhancement of the drug induced membrane permeability. Another interesting observation was that the stability of Curcumin in presence of POPG liposomes was observed to degrade significantly near physiological pH (7.4 and 8.0). Although this degradation did not affect the liposome integrity, interestingly this was observed to enhance the transport of the LDS cation across the bilayer. That the degradation products of Curcumin are equally effective as the drug itself in enhancing the membrane permeability lends additional support to the current opinion that the bioactive degradation products of the drug may have a significant contribution to its observed pharmacological effects. © 2016 Elsevier B.V. All rights reserved.

1. Introduction The interaction between a drug and a lipid bilayer can alter several important biophysical properties of membranes such as the membrane potential, fluidity and permeability [1–15]. Recently we have observed that transport (and hence membrane permeability) of a hemicyanine dye LDS-698 (LDS), an organic cation, across a POPG (1-palmitoyl-2oleoyl-sn-glycero-3-phospho-(1′-rac-glycerol) sodium salt) bilayer becomes significantly faster when Curcumin is present in the bilayer by using the interface selective Second Harmonic (SH) spectroscopic technique [16]. Subsequently we have investigated the membrane permeability effect of Curcumin and Chlorin-p6 (Cp6: a porphyrin based photosensitizer) on POPG liposomes by monitoring the transport kinetics of two organic cations; Malachite green at pH 5.0 and LDS at pH 7.4 using the SH spectroscopic technique [17]. Drug induced alteration in lipid membrane permeability are expected to depend on the chemical structure of the drug, and if the drug contains ionizable functional ⁎ Corresponding author. E-mail addresses: [email protected], [email protected] (K. Das).

http://dx.doi.org/10.1016/j.saa.2016.09.001 1386-1425/© 2016 Elsevier B.V. All rights reserved.

groups (present in Curcumin and Cp6), the pH of the medium. In this work we have investigated the role of pH in drug (Curcumin and Cp6) induced POPG membrane permeability using LDS cation as a probe over a pH range of 4.0 to 8.0 using the SH spectroscopic technique. Further, using fluorescently labeled POPG liposomes, steady state fluorescence resonance energy transfer (FRET) studies were conducted to find out the pH dependent bilayer localization of both the drugs. The results of SH and FRET studies were used to get an idea about how the membrane permeability depends upon the position of the drug in the POPG bilayer. In addition Curcumin in presence of POPG liposomes were observed to degrade, the rate being higher at basic pH. This degradation resulted in faster transport of LDS across the bilayer indicating that the degradation products of Curcumin are as effective as the drug itself to affect bilayer permeability. The chemical structures of the molecules and lipids used in this study are shown in Scheme 1. 2. Materials & Methods LDS (from Exciton) and Cp6 were a kind gift from Dr. N. Sarkar and Dr. A. Dube respectively, and they were used as received. Curcumin

G.K. Varshney et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 173 (2017) 132–138

133

Scheme 1. Chemical structures of LDS cation, Chlorin-p6 (Cp6), Curcumin, DPH-PC and POPG lipid.

were purified before use by methods as described earlier [17]. POPG from Avanti Polar Lipids and 2-(3-(diphenylhexatrienyl)propanoyl)-1hexadecanoyl-sn-glycero-3-phosphocoline (DPH-PC) from Molecular Probes were used as received. Unilamellar liposomes (for preparation details, see SI) were suspended in 20 mM phosphate buffer solution. The size and zeta potential of the liposomes (measured by dynamic light scattering technique) at different pH and with/without drugs are provided in the supporting information (Table S1). The molar ratio of DPH-PC:POPG was kept at 1:150. Steady-state absorption spectra were obtained on GBC UV–visible spectrophotometer with 1 nm resolution. Steady-state fluorescence spectra (for FRET efficiency measurements) were obtained on a Spex Fluorolog fluorimeter with a 4 nm band pass (excitation and emission) and corrected for lamp spectral intensity and detector response. Details of SH experimental set up were identical to our previous reports [16–17]. The average laser power used in the experiments was 600 mW and the generated SH light (at 400 ± 2 nm) was detected at

right angles with respect to the fundamental (800 nm) by a PMT using single photon counting technique. The SH signal integration time were kept at 1 s while the sample in the cuvette was constantly stirred during the measurement using a magnetic stirrer and the sample temperature was controlled by a Neslab circulating water chiller. Micro-liter aliquots from a concentrated stock solution of Curcumin were added to the liposome solution and incubated for few minutes for attaining equilibrium between the drug and the liposome. For Cp6 , following our previous study [17], the incubation time was kept at 30 min to ensure full equilibrium between the drug and the liposome. SH experiments were done as follows: First the signal from 2 mL of buffer containing 5 μM LDS was recorded followed by addition of 50 μL of liposome (with/without drug) solution at 50 s time point. The time corresponding to liposome addition were denoted as t = 0 in all the figures. The concentration of lipid was kept at 50 μM and the concentration of both Curcumin and Cp 6 were kept at 3 μM.

134

G.K. Varshney et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 173 (2017) 132–138

The electric field of the SH signal (E2ω) were obtained from the observed SH signal (I2ω) as: E2ω ðtÞ ¼

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Ið2ωÞdyeþliposome ðtÞ−Ið2ωÞbackground

LDS Cp6 Curc

ð1Þ

pH 7.4

3. Results Since the motivation of this work was to study the pH dependence of drug induced membrane permeability the probe molecule must be able to generate detectable SH signal (with our experimental setup) over a wide pH range. Malachite green which have been used in several studies [18–22] cannot be used as a probe as its pKa for inter-conversion between the charged and the neutral form is at pH 7.0. The electron donor group of LDS, (i.e. the –N(Me)2 group) is susceptible to protonation at acidic pH which would result in loss of electron delocalization thereby reducing its SH signal. Therefore an estimation of the pKa value for this group was necessary to determine the pH range at which the dye can act as a SH probe. Fig. 1 describes the absorption spectra of LDS over a wide range of pH. The absorption peak of LDS at 440 nm starts to decrease in intensity as the pH approaches 5.0 and at pH 3 a new peak appears at 360 nm. These changes in the absorption spectra are attributed to the protonation of the –N(Me)2 group of the dye which results in absorption at higher energy due to loss of electron delocalization. From the plot of the absorbance ratio at 440 and 360 nm versus pH, the pKa value of the –N(Me)2 group was estimated at 4.3. Therefore the lower limit of the pH for carrying out the transport of the dye across POPG liposomes was set at pH 4.0. Further, since several studies have showed that Curcumin degrades near physiological pH [23–25], the upper limit of pH was fixed at 8.0. Fig. 2 describes the changes in SH electric field (E2ω) of LDS (5 μM) before and after addition of POPG (50 μM) liposomes at different pH. Before the addition of liposomes the observed E2ω is due to hyper-rayleigh scattering from the dye. Addition of liposomes produces an instantaneous increase of E2ω due to adsorption of the cationic dye to the outer surface of the anionic head groups of POPG lipids. Subsequently E2ω decreases with time as LDS molecules transports from the outer to inner bilayer. The normalized E2ω values (E2ω values are normalized w.r.t. their maxima, i.e. immediately following liposome addition) are

0.40

OD440nm/OD360nm

3.0

pH decreasing Optical density

0.30 0.25

2.5 2.0 1.5

pKa = 4.3

1.0 0.5 0.0 1

2

3

4

5

6

7

8

9

pH

0.20 0.15 0.10 0.05 0.00 350

400

450

500

550

600

650

wavelength (nm) Fig. 1. Effect of pH on the absorption spectra of LDS (2.5 μM). The pH values are 8.5, 8.0, 7.4, 6.0, 5.0, 4.0 and 3.0. Inset: Plot of the ratio of absorbance values of the dye at 460 and 360 nm at different pH. The inflection point obtained from the sigmoidal fit gave a pKa of 4.3 for the –N(Me)2 group.

E2ω (normalized)

where I(2ω)dye+liposome(t) is the SH signal detected at 2ω at time t and I(2ω)background represents the contributions from the factors other than the SH field generated by the dye adsorbed on the lipid bilayer.

0.35

pH 8.0

1.00 0.75 0.50 0.25 0.00

pH 6.0

pH 5.0

pH 4.0

0

500

1000

1500

2000

time (second) Fig. 2. Effect of pH on the SH electric field of 5 μM LDS before and after addition of 50 μM POPG liposomes at 25 °C. Liposomes are added at t = 0 time point. The black curves denote only liposomes; red and green curves denote liposomes containing 3 μM of Cp6 and Curcumin respectively. All the curves are normalized with respect to the maximum E2ω value. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

plotted for a reason that will be discussed later and also due to the fact that normalization provides a better clarity on the effect of pH as well as the drugs on the E2ω characteristics of LDS. The decays of E2ω [18–22] are fitted to exponential decay functions of the following form: E2ω ðt Þ ¼ A0 þ

X

ai expð−t=τi Þ

ð2Þ

The observed decays of E2ω were consistently observed to be bi-exponential and the fitting parameters are provided in Table 1. The driving force for the transport of the LDS cation from the outer to the inner bilayer of the POPG liposomes originates from the potential difference created between the outer and inner bilayer due to electrostatic adsorption of the LDS cation on the outer bilayer surface [18–22]. The observed bi-exponential decays of E2ω could arise due to initial rapid transport of the cation from the outer bilayer to the inner bilayer (~100 s, Table 1) and as the inner bilayer gets populated with the LDS cation the potential difference decreases causing the transport rate of the LDS cation to be slower (~1000 s, Table 1). As observed earlier [19–22] a notable feature of the decay characteristics of E2ω is the observation that it levels off at a value that is significantly greater than zero (corresponding to the term A0 in Eq. (2)). This has been attributed to the finite population difference in the dye molecules adsorbed on the outer (Nout) and inner (Nin) surface of the liposome and the term 1-A0 represents the Nin/Nout ratio. The transport kinetics of LDS after addition of POPG liposomes containing either Curcumin or Cp6 (3 μM each) changes significantly. In presence of the drugs, the decays of E2ω (from LDS) depend both on the pH of the medium and the drug (Curcumin/Cp6) present in the bilayer. It is clear from Fig. 2 that at acidic pH (4.0) bilayer permeability

G.K. Varshney et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 173 (2017) 132–138

a2

τ2

τav

Nin/Nout

4.0 +3 μM Cp6 −2 ≈ 6.3 Cp−1 6 :Cp6 +3 μM Curc 5.0 +3 μM Cp6 Cp6−1:Cp6−2 ≈ 0.6 +3 μM Curc 6.0 +3 μM Cp6 −2 ≈ 0.1 Cp−3 6 :Cp6 +3 μM Curc 7.4 +3 μM Cp6 −2 ≈ 2.5 Cp−3 6 :Cp6 +3 μM Curc Curc0:Curc−1 ≈ 4.0 8.0 +3 μM Cp6 −2 Cp−3 ≈ 10 6 :Cp6 + 3 μM Curc Curc0:Curc−1 ≈ 2.5

0.32 0.08

0.42 0.81

170 29

0.26 0.11

2600 390

750 65

0.68 0.92

0.40 0.45 0.19

0.19 0.37 0.31

90 199 44

0.41 0.18 0.50

1143 2220 494

486 464 262

0.60 0.55 0.81

0.23 0.42 0.12

0.47 0.34 0.19

69 242 84

0.296 0.24 0.69

790 6752 1947

266 1713 1351

0.77 0.58 0.88

0.22 0.34 0.18

0.18 0.30 0.30

79 79 57

0.6 0.36 0.52

235 3470 1166

155 1280 629

0.78 0.66 0.82

0.10

0.67

47

0.23

184

74

0.90

0.14 0.21

0.22 0.26

54 57

0.64 0.53

2120 1016

1379 557

0.86 0.79

0.14

0.63

81

0.23

418

147

0.86

a The fitted values represent the decays of the SH electric field of LDS under various conditions where the experimental time windows vary considerably. For example, in the presence of only POPG liposomes decays of E2ω were recorded on a much longer time scale (see Fig. S1 in SI) to get an accurate estimation of the time constants. All the time constants are in seconds. b The ratio of the abundant ionic species of the drugs at different pH (obtained from their corresponding pKa values) are provided in grey background.

is significantly increased by Cp6 whereas near physiological pH bilayer permeability is markedly enhanced by Curcumin. To get an idea about how the bilayer transport of LDS is affected by the drugs over the pH range used in this study we have compared the relative changes of two transport parameters: average transport time constant (τav; where, τav = a1τ1 + a2τ2) and the relative changes of the Nin and Nout values. The Nin/Nout ratios under different experimental conditions are listed in Table 1. However it is pertinent to note that the transport kinetics of LDS across the bilayer reached equilibrium at ~ 500 s in the presence of 3 μM Cp6 at pH 4.0 (or 3 μM Curcumin at pH 7.4) whereas in the absence of the drugs at least an 1 h is needed to attain equilibrium (see Fig. S1, supporting information). Therefore the Nin/Nout ratio of LDS at equilibrium obtained by exponential fitting actually represents the Nin/Nout values at widely different time points. On the other hand, following the addition of liposomes at t = 0, E2ω(t) will be proportional to [Nout(t) − Nin(t)]EωEω at any time t. Therefore if the E2ω values are normalized to unity at t = 0, then under similar experimental conditions the drug induced relative changes of the quantity: [Nout(t) - Nin(t)] or ΔN(t) can be compared by simply taking the ratio of E2ω values at that particular time t. It is important to note that by comparing the different E2ω(t) i.e. ΔN(t) values a fair idea about the transport process can be gained at any time without the need to fit the experimental data. The smaller the value of ΔN(t) will correspond to a faster rate of transport across the bilayer. Fig. 3 shows the relative changes of τav and ΔN1000s (at t = 1000 s) values at different pH experimental conditions. Both the drugs used in this study have functional groups that can be ionized depending upon the pH of the medium. This ionization is expected to affect the hydrophobic/hydrophilic balance of the drug thereby affecting its bilayer localization. It is evident from Figs. 2 and 3 that pH plays an important role in the drug induced membrane permeability of LDS. Therefore, to get an idea about how membrane permeability depends upon the bilayer localization of the drug, FRET studies were carried out using POPG liposomes containing DPH labeled PC phospholipid. Since the fluorescence spectra of DPH overlaps nicely with the absorption spectra of both the drugs (see Fig. S2, SI) in the

Cp 6

16

Curcumin

1.2 1.0

12 0.8 8 0.6 4

0.4

Relative change: N1000s

τ1

Relative change: τav

a1

∇

a0

1.4

20

Table 1 Fitted parameters for the decay of SH electric field of LDS after addition of different POPG liposomes at different pHa. pHb

135

0.2

0 4

5

6

7

8

pH Fig. 3. Relative changes in: i) average transport time constant τav (solid data points connected by solid lines) and ii) ΔN1000s (hollow data points connected by dashed lines) of LDS. Relative changes are calculated by normalizing with the corresponding value in the absence of the drugs.

presence of POPG liposomes it is expected that FRET will occur between DPH and the drugs. The R0 values for the FRET pairs DPH-Cp6 and DPHCurcumin were estimated to be 34.9 ± 0.8 and 42.5 ± 1.0 Å respectively (for details see SI). Fig. 4 shows the FRET efficiency between DPH-Cp6 and DPH-Curcumin with increasing concentrations of the drugs at various pH calculated from the steady state fluorescence intensity of the donor DPH. The FRET efficiency curves increases with increasing concentration of the drugs and then tends to saturate. While the FRET efficiency between DPH and Cp6 were observed to significantly decrease with increase in the pH of the medium, at acidic pH range (4.0 to 6.0) the FRET efficiency between DPH and Curcumin were observed to marginally decrease with increase in the pH of the medium. Several studies have shown that in aqueous medium, free Curcumin is unstable in neutral and basic pH but its stability improves significantly in the presence of macro-molecular assemblies like micelles, proteins and liposomes made from saturated lipids [14–15,23–28]. In this study the effect of pH on drug induced membrane permeability were carried out over a pH range of 4.0 to 8.0. Since pKa of the phenolic – OH group of Curcumin is reported to be ~8.4 [23,25] a substantial population of the drug is expected to be in aqueous phase in liposomal solutions whose pH are closer to 8.0. Indeed we have observed significant degradation of Curcumin in presence of POPG liposomes at pH 7.4 and 8.0 which prevented us to conduct FRET studies between DPH and Curcumin at these pH values. Instead we have studied the degradation kinetics of the drug in presence of POPG liposomes and consequently, how this degradation affects the transport of LDS across the POPG bilayer. The degradation kinetics of Curcumin has been assessed by monitoring the time dependent absorption spectra of the drug in presence of POPG liposomes at pH 7.4 and 8.0 and compared to that at pH 5.0. Fig. 5 shows the normalized changes in the optical density of the drug at its absorption maxima (~ 422 nm) monitored over a period of more than 2 h in the presence of POPG liposomes along with their fits. The fitted parameters are provided in supporting information. It is clear from Fig. 5 that while the degradation of the drug is nominal at pH 5.0, it is significant at pH 7.4 and 8.0. Fig. 6 describes the effect of Curcumin degradation on the transport characteristics of LDS across the POPG bilayer. At pH 7.4 and 8.0 the transport of LDS becomes distinctly faster when liposomal Curcumin solutions are aged for 4 h compared to liposomal Curcumin solutions that were prepared fresh. However when this experiment is done at pH 5.0 the transport characteristics of the cation remains quite similar (Fig. S3, Supporting information) which is consistent with the results obtained in Fig. 5. Finally we have also checked the effect of Curcumin degradation on the membrane integrity. The size and zeta potential of liposomes in presence of

136

G.K. Varshney et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 173 (2017) 132–138

0.6

0.4

pH 8.0 At 0 hour: τav = 110s

60

At 4 hour: τav = 70s

40 0.2

20

ESHG

FRET efficiency

80

pH 4.0 pH 5.0 pH 6.0 pH 7.4 pH 8.0

0.0

0 80

0.0

0.5

1.0

1.5

2.0

2.5

Chlorin-p6 (μM)

FRET efficiency

0.8

At 0 hour: τav = 60s

60

pH 4.0 pH 5.0 pH 6.0

1.0

pH 7.4

3.0

At 4 hour: τav = 40s

40

20 0.6

0 0

0.4

200

400

600

800

1000

time (second) 0.2 0.0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

Curcumin (μM) Fig. 4. pH dependent FRET efficiency (λex = 360 nm) between DPH-PC and Cp6 (top) and between DPH-PC and Curcumin (bottom). FRET efficiency (defined as: 1-FDA/FD; where FDA and FD is the fluorescence intensity of DPH in the presence and absence of Curcumin or Cp6, respectively) was obtained by monitoring the fluorescence maxima of DPH-PC at 432 nm. In these experiments the lipid concentration was kept at 150 μM (DPHPC:POPG = 1:150).

Fig. 6. SH electric field of 5 μM LDS before and after addition of 50 μM POPG liposomes containing 3 μM of Curcumin at pH 8.0 (top) and pH 7.4 (bottom). Liposomes are added at t = 0 time point. The black curves denote experiments conducted immediately after Curcumin addition to the liposomes and the red curves denote experiments where liposomes containing Curcumin were aged for 4 h and then used. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Curcumin (lipid: drug = 50: 3) were monitored up to 24 h and no significant changes were observed (see SI). Therefore on the basis of the data presented in Figs. 5 and 6 it can be concluded that Curcumin as well as its degradation products remarkably affects the membrane permeability of POPG liposomes against the organic cation, LDS.

1.0

OD422 nm (normalized)

4. Discussion 0.8

0.6

pH 5.0 pH 7.4 pH 8.0

0.4

0

50

100

150

time (minute) Fig. 5. Decay of Curcumin absorption maxima at 422 nm at different pH in presence of POPG liposomes (Curcumin:Lipid = 1:10) and the corresponding fits (line). The values shown are normalized with respect to zero time. The data for pH 5.0 and 7.4 can be fitted satisfactorily with linear fits while the data for pH 8.0 is best fitted with a single exponential decay function. The fitted parameters are provided in the supporting information.

The main objective of this work was to study the pH dependence of Curcumin and Cp6 on the transport of the LDS cation across a POPG bilayer. Since the pKa of the donor group of LDS (− NMe2) is at 4.3, the population of LDS monocation is roughly 33% and 83% at pH 4.0 and 5.0 respectively. Since the observed E2ω signal comes exclusively from LDS monocation, the Nin/Nout ratio and τav values at pH 4.0 and 5.0 reflects the effect of changing populations of the LDS monocation. The decay characteristics of the SH electric field of LDS after addition of different POPG liposomes (Fig. S1, SI) and the corresponding fitted parameters (especially the Nin/Nout ratio) provided in Table 1 shows that the transport of the cation increases with increasing pH. The Nin/Nout value approaches ~ 0.90 as the pH of the medium is increased to 8.0. This roughly corresponds to the ratio of the area of the inner leaflet to the outer leaflet for the POPG liposomes, (size ~ 200 nm and bilayer thickness ~5 nm) which indicates the completion of the transport process. For both the drugs, the Nin/Nout value was observed to be ~0.90 at certain pH indicating completion of the transport process. The pH dependence of drug induced bilayer permeability was assessed by monitoring the relative changes in the two calculated

G.K. Varshney et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 173 (2017) 132–138

parameters: average transport time constant (τav) and ΔN1000s values. The observed trends in the relative τav and ΔN1000s values show that the permeability of the POPG bilayer containing Curcumin (drug:lipid = 0.06) against the LDS cation increases remarkably as the pH is increased, it peaks at 7.4 (~17× decrease in τav with corresponding ΔN1000s ≈ 0.3; Fig. 3) and then decreases significantly at pH 8.0. The decrease in Curcumin induced bilayer permeability at pH 8.0 could be explained by the fact that at this pH the amount of neutral Curcumin population (which is expected to be in the bilayer) decreases by 29% (Table 1). Interestingly an opposite trend is observed when relative τav and ΔN1000s values are compared for POPG liposomes in presence of similar amount of Cp6 (drug:lipid = 0.06). Cp6 induced bilayer permeability was observed to be maximum at pH 4.0 (~ 10 × decrease in τav with corresponding ΔN1000s ≈ 0.5; Fig. 3), a further increase in the pH of the medium decreased the drug induced membrane permeability and any further increase in pH did not affect the bilayer permeability significantly. Curcumin and Cp6 are amphiphilic in nature due to the presence of ionizable functional groups (−COOH groups in Cp6 and –OH groups in Curcumin). The pKa values of some of these functional groups are either within the pH range used in this study or closer to it. Therefore depending upon the pH of the medium these drugs can be charged or neutral which is expected to affect their lipophilicity. The first and second pKa of the –COOH groups of Cp6 are at 7.0 and 4.8 [29–31]. Therefore Cp6 −2 can exist as Cp−1 and Cp−3 depending upon the pH of the medi6 , Cp6 6 um. The ratios of the various forms of Cp6 present at different pH are shown in Table 1. So, at pH 4.0 where Cp−1 6 species (more hydrophobic) is the majority, the bilayer permeability gets significantly enhanced species (more hydrophilic) is abundant whereas at pH 8.0 where Cp−3 6 the change in the bilayer permeability is modest. The average center-to-center distance (dx) between DPH and the drugs were estimated from the FRET efficiency and corresponding R0 values (see SI). The dx values of DPH and Cp6 in the POPG bilayer increases from 34.6 ± 0.5 to 43.6 ± 0.8 Å as the solution pH increases from 4.0 to 8.0. The dx values of DPH and Curcumin in the POPG bilayer increases from 33.9 ± 0.4 to 36.4 ± 0.5 Å as the solution pH increases from 4.0 to 6.0. It can be seen from Table 1 and Fig. 3 that the increase in the dx values has a marked effect on the transport characteristics of the LDS ion. For example an increase in the dx value of Cp6 from 34.6 ± 0.5 at pH 4.0 to 41.6 ± 0.6 at pH 6.0 results in a 9 fold decrease while an increase in the dx value of Curcumin from 33.9 ± 0.4 at pH 4.0 to 36.4 ± 0.5 at pH 6.0 results in a 7 fold increase in the relative transport time constant of the cation. It therefore appears that with increasing pH, the outward movement of Curcumin and Cp6 (with respect to the bilayer center) affects the membrane permeability in an opposite way. In order to search for an explanation for this opposite effect we note that: i) the length (long axis) of both the drugs are comparable to the length of a POPG molecule and, ii) it is reasonable to assume that in the bilayer the ionizable group/groups of the drugs should be closer to the liposome-water interface. Earlier reports have indicated that interaction of the functional groups of Curcumin (phenolic –OH and enolic –OH) with the polar head groups of the choline moiety of the phosphatidyl choline lipids (DMPC and DPPC) plays a major role in modulating the bilayer organization [14–15]. Based on FRET results, the proximity between the enolic –OH group of the drug and the polar head groups of POPG is likely to get enhanced with increasing pH, which might cause the significant enhancement the bilayer permeability by Curcumin. One way to test this hypothesis would be to investigate with Curcumin derivatives where the enolic –OH group of the molecule is replaced by some inert group say, − OMe. Cp6, on the other hand, have the three –COOH groups at one end of the molecule. With an increase in solution pH the outward movement of the drug is likely to reduce the proximity between the –COOH group of the drug and the polar head groups of POPG as suggested by the FRET study. Therefore Cp6 induced bilayer permeability gets significantly reduced with increase in solution pH.

137

According to a recent study, the degradation of Curcumin (which is itself an antioxidant) proceeds via autoxidation which occurs mainly through the breaking of the conjugated alkyl chain [33]. The major degradation product has been identified as a bicyclopentadione derivative of the drug. Therefore the various pharmacological activities of the drug are surprising given its low bioavailability due to its degradation. It has been suggested that the bioactive degradation products of the drug may have a significant contribution to the observed pharmacological effects [34,35]. Therefore it is also important to explore the effect of drug degradation in presence of POPG liposomes and its effect on the transport kinetics of LDS at various pH used in this study. The degradation of the drug near physiological pH is indeed an issue as demonstrated in Fig. 5. This is most likely caused by the ionization of the phenolic OH group of Curcumin whose pKa is estimated to be around 8.4 [23, 25]. The percentage of the ionic species (which initiates the degradation process) of the drug (Curc−1; Table 1) present at pH 7.4 and 8.0 is ~20% and ~ 30% respectively. From the fitted parameters of the degradation kinetics (provided in SI), it can be estimated that after 4 h the amount of liposomal Curcumin (which is Curc0) will decrease by 17, 70 and 80% for pH 5.0, 7.4 and 8.0 respectively. However the decay of E2ω of LDS (Fig. 6) becomes faster by 1.6 to 1.5 times (at pH 7.4 and 8.0) when liposomal Curcumin solutions are used which were aged for 4 h. Since size and zeta potential of the liposomes in presence of Curcumin shows no significant changes till up to 24 h of incubation it is fair to say that the degradation products of the drug does not affect the integrity of the liposomes. Our experimental results thus lends to further support of the notion that degradation products of the drug may have a significant contribution to the observed pharmacological effects in the sense that it may alter the membrane permeability significantly. In conclusion we have studied the transport kinetics of the LDS cation across POPG liposomes in the pH range of 4.0 to 8.0 in the presence and absence of two amphiphilic drugs, Curcumin and Cp6. Our results show that transport (and consequently bilayer permeability) of LDS across the bilayer is governed by the nature of the drug and pH of the medium. While Cp6 significantly enhanced the transport of LDS at pH 4.0, Curcumin gradually enhanced the transport of the cation as the pH is increased which become maximum near physiological pH (7.4) and decreased a bit at pH 8.0. FRET studies confirmed that the bilayer localization of Cp6 is pH dependent, due to the presence of the carboxyl groups. Combining the results of FRET studies and the relative population of the various ionic/nonionic species of Cp6 at different pH leads to the suggestion that the reduced distance between the carboxyl group of Cp6 and polar head groups of the lipid might be responsible for the observed enhanced permeability of LDS at acidic pH. Understanding how pH plays a role in Curcumin induced bilayer permeability is more complex as the drug itself gets degraded near physiological pH. However the results obtained from the FRET and degradation studies suggest the possibility that interaction between anionic species of the drug (Curc−1) and the polar head groups of the lipids might result in significant modulation of the bilayer organization. Finally, one of the interesting findings of this study reveals that degradation products of Curcumin formed around physiological pH are equally effective in enhancing the membrane permeability of the LDS cation.

Appendix A. Supplementary data Detail procedures for preparation and characterization (with respect to their size and zeta potential) of liposomes. SH intensity profiles of LDS under various conditions. Absorption spectra of Curcumin and Cp6 and emission spectra of DPH-PC in presence of POPG liposomes, FRET calculation details and fitting parameters for Curcumin degradation kinetics in presence of POPG liposomes at various pH. Supplementary data associated with this article can be found in the online version, at doi: http:// dx.doi.org/10.1016/j.saa.2016.09.001.

138

G.K. Varshney et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 173 (2017) 132–138

References [1] C. Peetla, A. Stine, V. Labhasetwar, Mol. Pharm. 6 (2009) 1264–1276. [2] A. Grancelli, A. Morros, M.E. Cabanas, O. Domenech, S. Merino, J.L. Vazquez, M.T. Montero, M. Vinas, J. Hernandez-Borrell, Langmuir 18 (2002) 9177–9182. [3] A. Berquand, N. Fa, Y.F. Dufrene, M.P. Mingeot-Leclercq, Pharm. Res. 22 (2005) 465–475. [4] F. Barcelo, J. Prades, S.S. Funari, J. Frau, R. Alemany, P.V. Escriba, Mol. Membr. Biol. 21 (2004) 261–268. [5] H.Y. Cheng, C.S. Randall, W.W. Holl, P.P. Constantinides, T.L. Yue, G.Z. Feuerstein, Biochim. Biophys. Acta, Biomembr. 1284 (1996) 20–28. [6] Y. Corvis, W. Barzyk, G. Brezesinski, N. Mrabet, M. Badis, S. Hecht, E. Rogalska, Langmuir 22 (2006) 7701–7711. [7] A.V. Agasosler, L.M. Tungodden, D. Cejka, E. Bakstad, L.K. Sydnes, H. Holmsen, Biochem. Pharmacol. 61 (2001) 817–825. [8] A.A. Hidalgo, W. Caetano, M. Tabak, O.N. Oliveira, Biophys. Chem. 109 (2004) 85–104. [9] S.S. Feng, K. Gong, J. Chew, Langmuir 18 (2002) 4061–4070. [10] A. Preetha, N. Huilgol, R. Banerjee, Colloids Surf., B 53 (2006) 179–186. [11] A. Preetha, N. Huilgol, R.J. Banerjee, Membr. Biol. 219 (2007) 83–91. [12] W.C. Hung, F.Y. Chen, C.C. Lee, Y. Sun, M.T. Lee, H.W. Huang, Biophys. J. 94 (2008) 4331–4338. [13] Y. Sun, C.-C. Lee, W.-C. Hung, F.-Y. Chen, M.-T. Lee, H.W. Huang, Biophys. J. 95 (2008) 2318–2324. [14] J. Barry, M. Fritz, J.R. Brender, P.E. Smith, D.K. Lee, A.J. Ramamoorthy, Am. Chem. Soc. 131 (2009) 4490–4498.

[15] A. Perez-Lara, A. Ausili, F.J. Aranda, A. Godos, A. Torrecillas, S. Corbalan-Garcia, J.C.J. Gomez-Fernandez, Phys. Chem. B 114 (2010) 9778–9786. [16] G.K. Varshney, R.K. Saini, P.K. Gupta, K. Das, Langmuir 29 (2013) 2912–2918. [17] R.K. Saini, G.K. Varshney, A. Dube, P.K. Gupta, K. Das, J. Mol. Struc. 1074 (2014) 22–26. [18] A. Srivastava, K.B. Eisenthal, Chem. Phys. Lett. 292 (1998) 345–351. [19] E.C.Y. Yan, K.B. Eisenthal, Biophys. J. 79 (2000) 898–903. [20] Y. Liu, E.C.Y. Yan, K.B. Eisenthal, Biophys. J. 80 (2001) 1004–1012. [21] J. Liu, X.M. Shang, R. Pompano, K.B. Eisenthal, Faraday Discuss. 129 (2005) 291–299. [22] J. Liu, M. Subir, K. Nguyen, K.B.J. Eisenthal, Phys. Chem. B 112 (2008) 15263–15266. [23] M. Bernabe-Pineda, M.T. Ramirez-Silva, M. Romero-Romob, E. Gonzalez-Vergara, A. Rojas-Hernandez, Spectrochim. Acta A 60 (2004) 1091–1097. [24] M.H.M. Leung, H. Colangelo, T.W. Kee, Langmuir 24 (2008) 5672–5675. [25] Z. Wang, M.H.M. Leung, T.W. Kee, D.S. English, Langmuir 26 (2010) 5520–5526. [26] M.H.M. Leung, T.W. Kee, Langmuir 25 (2009) 5773–5777. [27] G. Began, E. Sudharshan, K. Udaya Sankar, A.G. Appu Rao, J. Agric. Food Chem. 47 (1999) 4992–4997. [28] A. Kunwar, A. Barik, R. Pandey, K.I. Priyadarsini, Biochim. Biophys. Acta 1760 (2006) 1513–1520. [29] A. Datta, A. Dube, B. Jain, A. Tiwari, P.K. Gupta, Photochem. Photobiol. 75 (2002) 488. [30] R.K. Saini, P.K. Gupta, K. Das, Phys. Chem. B 116 (2012) 4199–4205. [31] K. Das, B. Jain, A. Dube, P.K. Gupta, Chem. Phys. Lett. 401 (2005) 185–188. [33] O.N. Gordon, C. Schneider, Trends Mol. Med. 18 (2012) 361–363. [34] L. Shen, H.-F. Ji, Trends Mol. Med. 18 (2012) 138–144. [35] L. Shen, H.-F. Ji, Trends Mol. Med. 18 (2012) 363–364.