Hanine Zakaria,
Riham El Kurdi and
Digambara Patra*
Department of Chemistry, American University of Beirut, Beirut, Lebanon. E-mail: dp03@aub.edu.lb; Fax: +961 1365217; Tel: +961 1350000 ext. 3985
First published on 4th October 2022
The main purpose of this paper is to design curcumin loaded PLGA nanocapsules for the selective detection of dopamine using fluorescence spectroscopy. In the present work curcumin loaded PLGA nanocapsules were synthesized using a solid-in-oil-in water (s/o/w) emulsion technique. The prepared nanocapsules were coated with a poly(diallyldimethylammonium)chloride (PDDA) polymer to increase the entrapment of curcumin into the core of PLGA polymer. PLGA-Cur-PDDA nanocapsules were characterized using different microscopic and spectroscopic techniques. Unlike free curcumin, the formed CUR-PLGA-PDDA NCs were established as nanoprobes for the selective detection of dopamine molecules. The selectivity and specificity of nanocapsules toward dopamine was achieved by measuring the fluorescence emission spectra of the NCs in the presence of other interference molecules such as tryptophan, melamine, adenine, etc. It was noticed that increasing the concentration of the different molecules had no significant change in the fluorescence signal of the nanocapsules. These results confirm the strong quenching between dopamine and curcumin in the nanocapsules. Hence, this fluorescence emission technique was found to be selective, easy and fast with low cost for the determination of dopamine in a concentration range up to 5 mM with a detection limit equal to 22 nM.
In fact, curcumin is one of the most used drugs in the biomedical field. Curcumin is a bioactive polyphenol derived from the rhizome of the Curcuma longa-a turmeric plant.5 Besides its therapeutic benefits such as its antioxidant and anti-inflammatory effects,6 antimicrobial and anticarcinogenic activities,7 curcumin is being combined to nanomaterials to design effective nanosensors for the detection of specific analytes. For example, liposomal curcumin nanocapsules in the presence of PDDA polymer were used to detect ATP molecule.8 In addition, a study conducted by Bechnak et al. has proven the efficiency of curcumin/F-108 polymeric nanocapsules in the detection of nucleic acid.9 However, curcumin suffers from one major problem which is its poor bioavailability.6 Also, other obstacles are associated with curcumin like low water solubility, crystalline nature and alkaline degradation.10 Nevertheless, many methods had been studied to enhance the stability and the bioavailability of curcumin like the addition of agents like piperine, and the complexation in novel drug carriers such as liposomes, nanoparticles and micelles.11
Poly lactic-co-glycolic acid (PLGA) is a copolymer synthesized through a random ring opening copolymerization of the cyclic dimers of the lactic and glycolic acids.12 PLGA is a member of the FDA approved biodegradable polymers.13 PLGA has been used as a biomaterial because of its biocompatibility and biodegradability.12 In addition, PLGA has extensively used in drug delivery applications this due to its mechanical and processing properties.14 Also, it possesses a promising application as a platforms for tissue engineering.13 PLGA has involved in nanotechnologies and it showed successful diagnostic and therapeutic effects.12 Many researches had been done on the incorporation of curcumin into PLGA based nanocapsules. Studies showed that CUR-PLGA nanocapsules have a spherical shape.15 Hence, it was confirmed that the loading of curcumin in PLGA could increase the bioavailability of curcumin.16 Due to the biodegradable nature of PLGA, it enhances the biocompatibility of curcumin.17 Also, the solubility of curcumin is greatly enhanced upon encapsulation.17 The antioxidant properties do not alter with long term storage of CUR-PLGA nanocapsules.18
Poly(dimethyldiallylammonium chloride) (PDDA) is known as a cationic polyelectrolyte polymer, that can be used as a surfactant in order to increase the stability of the prepared nanocapsules. PDDA is known to be essential in biomedical applications, particularly in drug delivery applications, due to its biocompatibility. It is also easily administered and its method of synthesis is simple and cheap.19 Henceforth, PDDA polymer was proven by Othman et al., to be an effective polymer in increasing the stability of curcumin inside DMPC liposomes by decreasing thereby its release.20
Dopamine is a catecholamine neurotransmitter in nervous, cardiovascular and hormonal systems, plays an important role as an extracellular chemical messenger.21,22 Many diseases result from disorder in the level of dopamine. Low level of dopamine causes sleeping disorders, schizophrenia, Huntington's and Parkinson's diseases.22 Whereas, high dopamine level leads to cardiotoxicity which is accompanied with hypertension, drug addiction, increased heart rate and heart failure.23 Developing techniques for the detection of dopamine play a significant role in increasing the efficiency of finding suitable treatments. Many methods have been developed for the determination of dopamine including surface plasmon resonance, fluorescence, chemiluminescence, photoelectrochemical sensor, high-performance, liquid chromatography, in addition to electrochemical methods21,24 and the calorimetric method.25 These methods suffer from low selectivity, the necessity of big amount of samples, long time manipulation, etc. For this reason, it was necessary to find a simple method to detect dopamine with high selectivity and sensitivity.
In this work, curcumin molecules were entrapped into PLGA polymer, and coated with poly diallyl dimethyl ammonium chloride (PDDA) polymer in order to increase its stability and bioactivity. Hence, the produced nanocapsules were found to be efficient as fluorescent probe for the sensitive and selective detection of dopamine. The prepared nanocapsules were able to detect dopamine in a wide concentration range till 5 mM.
Furthermore, the effect of the temperature on the interaction between the nanocapsules and dopamine molecule was studied at 45 °C using a thermostat attached to the spectrofluorometer and with the aid of an external thermometer.
All the experiments were done by keeping the concentration of PLGA-Cur-PDDA NCs constant in a total volume of 3 mL.
Fig. 1 SEM image of (A) CUR-PLGA-PDDA NCs and (B) pure curcumin and (C) zeta potential analysis of CUR-PLGA-PDDA NCs. |
Furthermore, the surface charge of CUR-PLGA-PDDA nanocapsule was investigated through zeta potential analysis. Interistingly, the formed nanocapsules were positively charged with a value equal to + 10.07 mV (see Fig. 1C). This is mainly due to the presence of ammonium NH4+ in the PDDA molecules which was adsorbed on the outer layer of the CUR-PLGA NCs causing their stabilization and thus leading to the production of positively charged CUR-PLGA-PDDA NCs.
Moreover, the UV-visible spectrum was established for curcumin and the produced nanocapsules as shown in Fig. 2A. Curcumin absorbs prominently in the UV-visible region around 426 nm (S0 → S1 transition). However, absorption of CUR-PLGA-PDDA NCs appears at ∼469 nm. Thus, the identification of absorption ∼469 nm makes it easier for establishing the formation of CUR-PLGA-PDDA NCs in the solution. In addition, it is remarkable that a sharp and strong absorption peak was obtained for CUR-PLGA-PDDA NCs. Meaning that, all the curcumin was encapsulated into the core shell of the PLGA polymer, thereby the absence of free curcumin in the formed NCs. Furthermore, the successful production of nanocapsules was verified through the fluorescence emission spectra. As shown in Fig. 2B, a blue shift was occurred when curcumin is being entrapped in the core shell of the PLGA polymer. In fact, the nanocapsules emits at λ = 510 nm and curcumin emits at λ = 555 nm. This shift is due to the incorporation of curcumin inside the PLGA forming smaller nanocapsules. Also, this can be related to the hydrophobic environment caused by the PLGA polymer.
Fig. 2 (A) UV-visible spectra and (B) fluorescence spectra of pure curcumin and CUR-PLGA-PDDA NCs excited at λex = 425 nm. |
To further establish the physical characteristic of the nanocapsules, CUR-PLGA-PDDA NCs and curcumin were analyzed by X-Ray Diffraction (XRD) technique. The diffractograms of curcumin and nanocapsules are illustrated in Fig. 3A. The main characteristic peaks of curcumin appeared at diffraction angles of 2θ equal to 8.06°, 9.20°, 12.46°, 14.95°, 17.75°, 19.8°, 23.7°, 24.6°, and 26.5° revealing the crystalline form of curcmin.28 Yet, as it is shown in the diffractogram of the nanocapsules, these peaks were completely absent. Hence, this confirms the encapsulation of curcumin inside the core shell of the PLGA polymer, inducing the formation of amorphous nanocapsules. Finally, thermogravimetric analysis was performed to assess the stability of the prepared nanocapsules. As shown in Fig. 3B, around 200 °C the weight loss pattern of raw curcumin occurs, where it loses around 65% of its mass between 200 and 560 °C. Hence, a gradual decrease in the mass of curcumin was obtained within the increase of the temperature. Consequently, for CUR-PLGA-PDDA NCs no weight loss was observed around 100 °C, this assures that the synthesized NCs are dehydrated. However, the weight loss of CUR-PLGA-PDDA NCs was observed around 300 °C, after which it shows a sharp weight loss that ends at ∼400 °C. In this temperature range, 86% weight loss was observed. This difference in the degradation pattern means that the temperature over which curcumin is stable has been increased. Therefore, curcumin gained extra stability when encapsulated inside the nanocapsules. Same degradation pattern of PLGA was obtained with Su et al.29 Accordingly, we can conclude that CUR-PLGA-PDDA NCs follows the same degradation pattern of PLGA.
Fig. 3 (A) X-ray diffractogram and (B) thermogravimetric analysis of pure curcumin and CUR-PLGA-PDDA NCs. |
In fact, in the absence of PDDA the encapsulation efficiency was equal to 75%. This value had increased to 91% when adding the PDDA layer on the surface of the PLGA-Cur NCs. The enhancement of the EE percentage is due to the increase in the layer added, where it boosts the encapsulation of curcumin into the core of PLGA polymer. The high encapsulation efficiency value obtained in our case was similar to the EE value calculated by Gao et al.30
Furthermore, the drug release for both nanocapsules was investigated. For this, in the beginning a given amount of nanocapsules was kept for a specific time in given volume of double distilled water, then separated by centrifugation and concentration of the supernatant was measured. In the next step, again a given volume of double distilled water was added to the pallet of nanocapsules and kept for another specific time followed by centrifugation to measure concentration. The procedure was repeated till saturation in curcumin release was obtained. To find out total amount of curcumin release for a particular time interval, additive concentration and additive time were estimated by adding concentration and time from the previous measurements. Thus, additive concentration, total concentration curcumin release, increased with time. Interestingly, it was found that faster release of the drug is obtained when the nanocapsules surface was not coated by any PDDA layer (see Fig. 4). However, the rate of release decreases by adding PDDA layers. Such layer by layer assembly improves the stability of the nanocapsules.31
First of all, CUR-PLGA-PDDA NCs were mixed with 1 mM of dopamine and the emission intensity of was measured. Interestingly, a remarkable decrease was noticed accompanied with a red shift (see Fig. 5A). Based on this, several solutions were prepared with different dopamine's concentration in the range from 10 μM to 5 mM. As presented in Fig. 5B, the emission intensity of the nanocapsules decreases within the increase in dopamine's concentration. However, at high concentration of dopamine, higher than 1 mM, a remarkable red shift occurred from 504 nm to 531 nm, meaning that curcumin is being presented in the polar region. Thus, when varying the concentration from 10 to 500 μM, the emission intensity decreases slightly. This decrease can be due to the binding of the negatively charged dopamine, to the positively charged NCs. However, the enhancement of dopamine's concentration leads to an effective fluorescence quenching because at higher concentration of quencher normally quenching sphere action is operative. This is because of fact that nanocapsules helps to bring dopamine closer to curcumin encapsulated in the nanocapsules.
Fig. 5 Emission intensity at λex = 425 nm of CUR-PLGA-PDDA NCs in the presence of (A) 1 mM of dopamine and (B) several concentration of dopamine in the range of 10 μM to 5 mM at 25 °C. |
During fluorescence intensity reduction, the quenching efficiency can be symbolized as (I0/I)/I0, where I0 and I represent the fluorescence intensity of curcumin in the NCs in the absence and presence of dopamine. Hence, the quenching constant (Ksv) can be found using Stern–Volmer equation:
I0/I = Ksv × Cdopamine + 1 |
The I0/I at the maximum wavelength was plotted versus the concentration of dopamine as shown in Fig. 6A and B. The I0/I ratio signal is linear to dopamine concentration in the ranges 0.01–0.2 and 0.5–5 mM. The linear equations for the two concentration ranges are I0/I = 1.37047[dopamine] + 1.09967 with a correlation coefficient of 0.99726 and I0/I = 0.94765[dopamine] + 1.15409 with a correlation coefficient of 0.99484. The limit of detection is found to be 22 nM referring to 3σ/s criteria, where σ is the standard deviation of the measurements and s is the slope of the calibration curve.31 The efficiency of the method was compared to previous study in the literature (see Table 1).
Fig. 6 Linear correlation of I0/I of CUR-PLGA-PDDA vs. concentration of dopamine (A) from 0.01 to 0.2 mM; and (B) from 0.5 to 5 mM. |
Method used | Concentration range | LOD | Interference |
---|---|---|---|
FRET between acridine orange and CuO NPs32 | 1–40 μM | 40 nM | Ascorbic acid, uric acid, glucose, tryptophan and acetaminophen |
A simple and convenient fluorescent strategy based on graphene quantum dots33 | 0.5–120 μM | 0.16 μM | NaCl, KCl, CaCl2, glucose, cysteine, ascorbic acid, epinephrine |
Calorimetric detection using microfluidic paper34 | 0.5–4.75 μM | 0.37 μM | Ascorbic acid, uric acid |
Calorimetric detection using gold nanoparticles35 | 0.5–10 μM | 1.85 μM | Amino acids, glucose, ascorbic acid, uric acid |
FRET between PLGA-Cur-PDDA (our work) | 0.01–0.2 mM 0.5–5 mM | 22 nM 1.89 μM | Tryptophan, adenine, uracil, guanine, cytosine, melamine, glutathione, cystine, kreatinine, tyrosine, silymarin, ascorbic acid, uric acid, L-tyrosine, tymarine, L-dopamine |
In this case, the plot between I0/I and the concentration of dopamine showed a good linear relationship (with R2 = 0.9951) in a wide concentration range from 0 to 5 mM. Moreover, the Stern–Volmer equation can be fitted as: I0/I = 0.94765 × Cdopamine + 1, while Ksv was found to be 0.94765 mM, equal to the slope value of the linear fit. Consequently, the Ksv value is considered large which confirms the binding between dopamine and curcumin.
To prove the efficiency and the role of the nanocapsule, a control experiment was handled in the presence of curcumin alone and dopamine in aqueous solution. Interestingly, I0/I of curcumin remain constant in the presence of different dopamine's concentration (see Fig. 7). These results reveal the role of the encapsulation of curcumin into the core of PLGA polymer in the presence of PDDA layer. In fact, the negatively charged curcumin cannot interact with the dopamine having also a negative surface charge. Hence, the presence of PDDA polymer increases the surface charge of the nanocapsule; and consequently, facilitates the interaction between dopamine and the nanocapsules. Thus, bringing dopamine and curcumin close enough to have fluorescence quenching.
To understand the effect of temperature on the interaction between CUR-PLGA-PDDA NCs and dopamine, fluorescence spectra were recorded at an elevated temperature (45 °C). Similar to the previous results, fluorescence intensity showed a decrease with increase in dopamine concentration (see Fig. 8A & B). However, this decrease in fluorescence intensity saturated exponentially as shown in Fig. 8B at higher concentration suggesting a quenching sphere-of-action. Therefore Stern–Volmer equation was applied to lower concentration of dopamine as shown in Fig. 8C. The Stern–Volmer constant was estimated, and it was found that the Ksv value changed from 0.947656 mM−1 to 11.0019 mM−1 as temperature increased from room temperature to 45 °C. Temperature effect can tell whether the system is undergoing a static or a dynamic quenching.33 The obtained result confirms that static quenching is taking place. The increase in temperature leads to a decrease in the stability of the formed complex, which would decrease the static quenching efficiency.33 In contrast, in the present case increase in temperature increased the quenching rate (Ksv) and at higher temperature increase in concentration made quenching sphere-of-action operative. This may be because at higher temperature curcumin is more displaced from the polar to the non-polar region, thus, bringing more dopamine in close contact with the curcumin resulting higher quenching rate. This is further confirmed from the fact that a red shift in fluorescence wavelength maximum was observed only when increasing the concentration of dopamine at 25 °C, this red shift was totally absent when increasing the temperature to 45 °C. Furthermore, the assessment of the selectivity and specificity of the PLGA-Cur-PDDA NCs toward dopamine detection was attained by measuring the fluorescence emission of the NCs in the presence of other interference molecules such as L-tyrosine, tyramine, L-dopamine, tryptophan, melamine, adenine, etc. These molecules were selected because they own comparable structures to that of dopamine, so they can typically interfere in dopamine detection. Thus, in Fig. 9 it is obvious that increasing the concentration of the different molecules had no significant change in the fluorescence signal of the nanocapsules. These results confirm the strong quenching between dopamine and curcumin in the nanocapsules.
Moreover, the stability of the proposed system was done by measuring the fluorescence emission intensity within 1 hour in the absence of dopamine and in the presence of dopamine (C = 2 mM). The obtained results are normalized with respect to the ratio obtained at zero time. Hence, within 1 hour I0/I remain constant revealing the stability of the proposed nanoprobe in the detection of dopamine (see Fig. 10).
Finally, to test the applicability, the analytical recovery of three unknown samples was estimated by using the obtained fitted calibration curve. The obtained results where summarized in Table 2. The percent of recovery of dopamine was obtained to be between 98.75 and 100.5% (n = 3).
Theoretical concentration (mM) | Experimental concentration (mM) | Recovery (%) | |
---|---|---|---|
Sample 1 | 0.03 | 0.0298 | 99.3 |
Sample 2 | 0.4 | 0.402 | 100.5 |
Sample 3 | 4 | 3.95 | 98.75 |
This journal is © The Royal Society of Chemistry 2022 |