Introduction

Phytoconstituents represent a huge group of therapeutically active compounds applicable to treat variety of diseases in humans (Sayed et al. 2019). Phytoconstituents such as quercetin, myricetin, and other flavonoids have been used in the treatment of neurodegenerative diseases (Singh et al. 2021); curcumin (CUR), quercetin, berberine, silymarin, resveratrol, and capsaicin in inflammatory diseases (Khan et al. 2021; Chen and Liu 2022); CUR, ombuine, indigtone, and anastatin in liver diseases (Ali et al. 2019); D-limonene, epigallocatechin gallate, thymoquinone, beta-lapachone, campothecin, genistein, taxanes, and vinca alkaloids in cancer (Nakonieczna et al. 2020; Asati 2022; Kaur et al. 2022); avarol, vindoline, catharanthine, berberine, and pinolene in diabetes (Alam et al. 2022; Behl et al. 2022); CUR, naringenin, quercetin, mimosine, diosgenin, catechin, and rutin in gynaecological disorders such as polycystic ovary syndrome and endometriosis (Balan et al. 2021; Temkar and Menon 2022; Chorosho et al. 2023; Luo et al. 2023); lycopene and carotene in aging-related diseases (Xue et al. 2021); kaempferol, myricetin, and silybin in infectious diseases (Kancherla et al. 2019; Joshi and Prabhakar 2020; Das et al. 2021); and naringin (NRG), hesperidin, and mangiferin in renal fibrosis (Xu et al. 2021). Albeit of desired therapeutic efficacy, challenges such as low solubility in water, insufficient absorption when taken orally, restricted stability, and a brief duration of action impede the molecules from being successfully introduced into the market (Suresh et al. 2014).

CUR (Fig. 1a) is a widely used phytochemical that is generally recognized as a safe (GRAS) molecule by the United States Food and Drug Administration (USFDA) (Ipar et al. 2019). CUR is a polyphenol derived from turmeric or Curcuma longa. CUR is been used as a spice, food additive, natural colouring agent, preservative, and pH-sensitive indicator in food industries (Jiang et al. 2021). It is used in the treatment of cardiovascular disease (Norooznezhad et al. 2020), diabetes (Bulboacă et al. 2019), cancer (Bolat et al. 2020; Mardani et al. 2020), neurodegenerative disorders (Yavarpour-Bali et al. 2019; Lin et al. 2020), polycystic ovary syndrome (Sohaei et al. 2019), and chronic disorders (Ghelani et al. 2019; Tagde et al. 2021). CUR has poor aqueous solubility (< 1 µg/ml) (Liu et al. 2020; Tabanelli et al. 2021) and undergoes P-glycoprotein (P-gp) efflux limiting its membrane permeability, while the poor oral bioavailability of CUR is due to extensive phase II metabolism in the presence of uridine 5′-diphosphate-glucuronosyltransferase (UGT) and β-glucuronidase enzymes (Han et al. 2023; Wdowiak et al. 2023). Strategies to enhance the solubility and oral bioavailability of CUR for better therapeutic effects have been attempted and are continuing with the advances in drug delivery technology (Yildiz et al. 2019).

Fig. 1
figure 1

a Structure of curcumin. Represents chemical structure of curcumin as per IUPAC nomenclature. b Structure of naringin. Represents chemical structure of naringin as per IUPAC nomenclature

Recently, co-amorphous systems (CAM) combining two drugs, or drug and excipients, have been employed for solubility enhancement of poorly soluble drugs (Mizoguchi et al. 2019). CAM is the mixture of two or more low molecular weight drugs or excipients forming a single-phase homogenous system in the amorphous form (Yarlagadda et al. 2021). The solubility improvement by CAM is a result of the high energy of the amorphous form at the time of dissolution (Karagianni et al. 2018). Co-formers used in the preparation stabilize the drug by molecular mixing, molecular interactions such as hydrogen bonding, salt formation, and π-π interactions (Liu et al. 2021). NRG (Fig. 1b), categorized as a GRAS, is a flavonoid derived from numerous citrus fruits (Yadav et al. 2020). The interaction of flavonoids such as naringenin, naringin, hesperidin, nobiletin, and hesperetin with the drugs that are enzyme or P-gp substrates results in inhibition or induction of efflux proteins and metabolic enzymes leading to higher plasma drug levels and better absorption (Hwang et al. 2015). NRG is known to improve the metabolism and absorption of several drugs (Shilpa et al. 2023). NRG as a co-former is reported to enhance the permeability of fexofenadine, talinolol, atorvastatin, and pyrazinamide by P-gp efflux inhibition through the formation of supersaturated solution of drugs in the gut (Teja et al. 2015; Nair et al. 2020; Prajapati et al. 2022; Uppala et al. 2022). The structural similarity of NRG with hydrophobic polyphenols may be an advantage for applying NRG to develop an amorphous formulation due to improved miscibility between them (Hatanaka et al. 2022). NRG delivers potential supramolecular H-bonding sites in the glycan portion recognized as crucial to induce disorder in the co-amorphous phase owing to its strong glass-forming ability. Such co-amorphous phases display a higher rate of dissolution and solubility by virtue of high energy confined within them (Teja et al. 2015).

In this study, a CAM of CUR and NRG in two molecular ratios of 1:1 and 1:2 was developed by quench cooling technique. The solid-state characterisation of the prepared CAM was carried out by DSC, PXRD, and FT-IR to determine the amorphous state and compatibility of the drugs. A stability-indicating reverse phase HPLC (RP-HPLC) method was developed to simultaneously estimate CUR and NRG, and the method was validated as per ICH Q2 (R1) guideline. A bioanalytical method was also developed in gradient mode to simultaneously estimate CUR and NRG in plasma. The content of drugs in CAM, solubility, in vitro drug release, and the in vivo pharmacokinetic studies were carried out using the developed method.

The individual methods were scrutinized and noted down as per the literature survey to be adapted for developing a new analytical method to simultaneously estimate CUR and NRG. The literature presented several methods for analysing CUR and NRG as individual entities. An immense exploration of the literature revealed few HPLC methods reported for the estimation of NRG in nanocapsules (Budel et al. 2020), nanosuspensions, and nanoparticles (Bhandari et al. 2019). Quite a limited HPLC methods have been reported for the simultaneous estimation of NRG with drugs such as atorvastatin and naringenin (Sowmya et al. 2019; Nair et al. 2020). Limited methods have been reported for the analysis of CUR simultaneously with piperine (Khismatrao et al. 2018), paclitaxel (Praveena and Guru 2021), quercetin (Patil and Mahajan 2022), mesalamine (Awasthi et al. 2022), cyclosporine (Desai et al. 2018), chlorin (Jain et al. 2023), and thymoquinone (Jagtap et al. 2021). Only two analytical methods are available for the estimation of CUR simultaneously with magnolol (Han et al. 2021), and piperine (Wang et al. 2019) in CAM. To date, there is no HPLC method available in existing literature for the concurrent determination of CUR and NRG.

Consequently, there is space for the development of an effective, sensitive, and stable HPLC method to simultaneously estimate CUR and NRG with better resolution of drugs in the mixture. To our acquaintance, this is the foremost stability-indicating RP-HPLC method for the estimation of CUR and NRG simultaneously in CAM as well as in plasma. Therefore, the present study aimed to develop and validate HPLC analytical method to simultaneously estimate CUR and NRG in the CAM. Further, the developed method is validated for parameters like accuracy, precision, robustness, and stability according to ICH Q2 (R1) guidelines. The optimized method is utilized for quantifying CUR and NRG, analysing solubility, and in vitro dissolution samples of the CUR in CAM. The method with few modifications was used to analyse plasma samples obtained from pharmacokinetic studies to determine the oral bioavailability of CUR in the presence of NRG.

Materials and Methods

Reagents

NRG hydrate and CUR were purchased from TCI chemicals, India. Glacial acetic acid and hydrochloric acid (37%) were purchased from Merck Ltd., India. HPLC grade methanol (minimum 99.9% purity) was purchased from Finar Ltd., India. Hydrogen peroxide 30% AR was procured from Loba Chemie Pvt. Ltd., India. Sodium hydroxide pellets (97% purity) were procured from Molychem, Mumbai, India. Unless stated, additional chemicals of analytical grade were used.

Chromatographic System and Conditions

The HPLC comprised of Shimadzu LC 2010CHT system equipped with autoinjector, column oven, and LC Solution 5.57 system control software. The separation of analytes was attained on a 4 µm Genesis C18 column with a 4.6 mm × 100 mm dimension. Mobile phase comprised of 48:52 and 75:25 v/v gradient ratio of 0.1% acetate buffer, adjusted to pH of 3.8 and methanol. MilliQ water was filtered through a 0.45-µm membrane filter before use. Flow rate was set to 0.5 ml/min for pumping of mobile phase through the column. The injection volume was 20 µl. The analytes were analysed at a single wavelength of 289 nm. MilliQ water to prepare buffer and sample was obtained from Millipore Elix® Advantage 3 purification system. The temperatures of column oven and autosampler were set at 25 and 4 °C, respectively.

Preparation of Standard and Sample Solutions

For the preparation of the stock solution, 10 mg each of CUR and NRG was dissolved in 10 ml methanol to attain 1000 µg/ml concentration. Subsequent dilutions of samples with a concentration between 0.125 and 16 µg/ml were prepared using the mobile phase.

Validation of Analytical Method

ICH Q2 (R1) guidelines were followed to validate the analytical method for specificity, system suitability, linearity, precision, accuracy, and robustness (ICH I. Q2 (R1) 2005).

Specificity

Three replicates of standard solution (CUR-NRG) in 10 µg/ml were injected to check the specificity.

System Suitability

Six replicates of standard solution (CUR-NRG) in 10 µg/ml were injected to check system suitability. The constraints, such as peak area, retention time, theoretical plate, and tailing factor, were analysed for acceptability.

Accuracy and Precision

Four replicates of standard solution (CUR-NRG) in 8, 10, and 12 µg/ml were injected to determine inter- and intra-day precision. The % RSD should be < 2% for the validity of the method. The % recovery of CUR and NRG at 8, 10, and 12 µg/ml concentrations indicates accuracy, which should be 98–102%.

Robustness

By presenting slight variations in the flow rate (± 0.1-unit), injection volume (20 to 19 µl and 21 µl), temperature (25 to 24 °C and 26 °C), and detector wavelength (289 to 288 and 290 nm), any variation in peak area, retention time, tailing factor, and theoretical plate count was observed. A standard solution (CUR-NRG) of 1 µg/ml was injected in triplicate to determine the robustness. In order for the method to be robust, % RSD should be < 2%.

Linearity

Suitable dilutions of 0.125, 0.5, 1, 2, 4, 8, and 16 µg/ml were prepared from the stock solution of CUR and NRG (100 µg/ml) using the mobile phase. The peak area of CUR and NRG against the respective concentration was plotted to obtain calibration curves.

Limit of Detection (LOD) and Limit of Quantification (LOQ)

LOD and LOQ calculations were done by means of the y-intercept standard deviation of regression lines and slope obtained from the calibration curve (54).

Bench-Top Stability

The stability was evaluated by injecting precision samples (1 µg/ml) in triplicates after 24 h together with fresh samples (1 µg/ml). The following equation was used to calculate the similarity index:

$$\mathrm{Similarity\;index}= \frac{{\mathrm{peak area}}_{\mathrm{old\;std}}\times {{\text{amount}}}_{\mathrm{new\;std}}}{{\mathrm{average\;peak\;area}}_{\mathrm{new\;std}}\times {{\text{amount}}}_{\mathrm{old\;std}}}$$

where peak areaold std and peak areanew std are the peak areas of sample injected after 24 h and freshly prepared samples, respectively. Amountold std and amountnew std correspond to the amount of precision sample and freshly prepared sample, respectively (Mutalik et al. 2021).

Stability Testing by Forced Degradation Studies

The sample of CUR-NRG (1 µg/ml) was subjected to acid-induced (0.1 and 1 N HCl), alkali-induced (0.1 and 1 N NaOH), thermal, and photolytic, in addition to oxidative (3% H202) degradation.

Preparation of Acid-Induced and Alkaline-Induced Degradation Samples

The methanolic standard solution of CUR-NRG in 1000 µg/ml was treated with 3 ml of 0.1 N HCl and 1 N HCl distinctly. The acidic mixture was subjected to heating up to 24 h and 12 h, respectively, at 60 °C, following which they were neutralized by 0.1 and 1 N NaOH, respectively, and were diluted further and injected in triplicate.

For alkaline-induced hydrolysis, about 3 ml of 0.1 and 1 N NaOH were separately added to 1000 µg/ml standard solution of CUR-NRG. The alkaline mixture was heated for 24 h and 12 h, respectively, at 60 °C. They were neutralized by 0.1 and 1 N HCl, respectively, and were diluted further and injected in triplicate for analysis.

Preparation of Oxidation-Induced Degradation Product

The standard solution of CUR-NRG in 1000 µg/ml after treatment with 3% v/v H2O2 was placed in the dark for 24 h. The solution was injected in triplicate following suitable dilutions.

Preparation of Photolytic Degradation Product

The standard solution of CUR-NRG (1000 µg/ml) after 24 h of exposure to direct sunlight was analysed for photolytic degradation of drugs. The samples were suitably diluted and injected in triplicates to the column of HPLC.

Preparation of Thermal Degradation Product

The standard solution of CUR- NRG (1000 µg/ml) was exposed to 80 °C heat for a period of 24 h to study the thermal degradation of drugs. The samples were suitably diluted and injected in triplicates for analysis.

Chromatographic Conditions and Analytical Technique for the Simultaneous Estimation of CUR and NRG in Plasma

The same chromatographic conditions as that of the analytical method were followed for bioanalytical method development except for the gradient method: pH 3.5 0.1% acetate buffer and methanol in the ratio 45:55%v/v up to 5 min, 75:25%v/v from 7 to 13 min, and 45:55%v/v up to 17 min. The injection volume was set to 40 µl. The protein precipitation method was followed to extract the drug from plasma using chilled methanol. Resveratrol was selected as an internal standard. The samples were centrifuged at 10,000 rpm for 15 min at 4 °C, and the obtained supernatant was analysed by HPLC. The peak area of CUR and NRG along with internal standard against the respective concentration was plotted to obtain calibration curves.

Preparation of CAM

CAM of CUR and NRG was prepared by quench cooling. A total of 500 mg of CUR and NRG in 1:1 and 1:2 molar ratios were taken in a silica crucible and melted in a heating mantle at a temperature near the melting point of drugs and held for a minute, ascertaining the absolute melting of both the drugs. Further, the crucible was transferred into a pre-cooled glass desiccator and stored at − 80 °C for 3–4 h, following which the sample was scrapped gently and pulverized using a mortar and pestle to get a fine powder. The formulation was stored in a vacuum desiccator for further characterisation (Nair et al. 2020).

Characterisation of CAM

Differential Scanning Calorimetry (DSC)

The thermal characteristics were determined by DSC. In the analysis, pure drug samples, physical mixtures, and CAM were loaded separately in aluminium cups, placed inside DSC-60 (Shimadzu, Japan), and scanned at 30 to 300 °C. The nitrogen flow and heating rate were maintained at 40 ml/min and 10 °C/min, respectively, using Shimadzu TA- 60WS thermal analyser.

Powder X-ray Diffraction (PXRD)

The PXRD analysis of the pure drugs, physical mixtures, and CAM was performed using Rigaku X-ray diffractometer (Rigaku Co., Tokyo, Japan). A current of 15 mA and 40 kV voltage were maintained. The instrument was equipped with a monochromator made of graphite and a detector. About 10–20 mg of sample was analysed at 5–40° 2θ to obtain the diffractograms.

Fourier Transform-Infrared Spectroscopy (FT-IR)

The FT-IR spectra of CUR, NRG, physical mixtures, and CAM were obtained using Alpha II compact attenuated total reflectance FT-IR (ATR-FTIR) spectrometer (Bruker, USA). After the blank correlation, a small amount of the sample was placed on the ATR crystal, and the anvil was pressed down gently, ensuring sample contact with the crystal. The spectra were obtained upon scanning at a wave number ranging from 400 to 4000 cm−1 (Pisay et al. 2022).

Scanning Electron Microscopy (SEM) Analysis

SEM analysis using EVO MA18 coupled with Oxford EDS (Germany) was performed to visualize surface morphology of pure drugs and CAMs. The CUR, NRG, PM, and CAMs were placed individually on aluminium stub, and gold sputtering technique was followed (Kara et al. 2023).

Quantification of CUR and NRG in CAM

The drug content, solubility, and in vitro drug release of prepared CAM of CUR and NRG were estimated using the validated method. Five-milligram equivalent of CUR from CAM was dissolved in 10 ml methanol and diluted with mobile phase to obtain a 1 µg/ml concentration. Triplicates of CAM were injected to estimate CUR and NRG simultaneously, and the results were equated with standard CUR-NRG (1 µg/ml).

Solubility Analysis

The equilibrium solubility of CUR and the CAM was determined in the presence of USP phosphate buffer pH 6.8. An excessive amount of each sample was taken in an Eppendorf containing 1 ml of buffer in triplicates and placed for 24 h in a tube rotator at 50 rpm maintaining the room temperature. The samples were centrifuged at 10,000 rpm for 15 min, and the separated supernatant was analysed by HPLC.

In Vitro Drug Release Study

The release of pure CUR, CUR in physical mixtures, and CAM was performed in USP 50 mM phosphate buffer pH 6.8 and hydrochloric acid of pH 1.2. Drug release was monitored at 37 °C with 250–300 rpm in 20 ml buffer. One thousand microliters of samples at 5, 10, 15, 30, 60, 90, and 120 min were withdrawn, and the sink condition was maintained. Collected samples were centrifuged at 10,000 rpm for 15 min, and the supernatant obtained was analysed by HPLC (Navya et al. 2022).

Pharmacokinetic Studies

As per the CPESCA guidelines and upon approval by the Institutional Animal Ethics Committee (IAEC/KMC/45/2022), Manipal Academy of Higher Education (MAHE), the pharmacokinetic studies were carried out. The Sprague Dawley rats weighing 200–220 g were divided into CUR-, PM-, and CAM (1:2)-treated groups. Each group comprised three rats, and parallel study design was followed. The rats were subjected to fasting for 12 h prior to the studies with water accessibility. The CUR-treated group received 500 mg/kg of CUR in 0.5% w/v carboxyl methyl cellulose (CMC) solution. PM- and CAM-treated groups received 500 mg equivalent of CUR in 0.5% w/v CMC. The suspension (2 ml) was orally administered in rats by means of oral gavage. The blood samples were collected in heparinized tubes by retro-orbital bleeding method at a predetermined time interval of 1, 2, 4, 6, 8, 10, 12, and 24 h. The collected samples were centrifuged at 10,000 rpm, 4 °C for 10 min, to obtain the plasma and kept at − 80 °C till further study. GastroPlus™ software (version 9.7, Simulation Plus Inc. Lancaster, CA-USA) was applied to compute the pharmacokinetic parameters.

Results and Discussion

RP-HPLC Method Development

To begin with, acetonitrile and phosphate buffer, acetonitrile, and acetate buffer were tried in several ratios for the separation of individual drugs and the combination. The peak area of CUR was less in these trials, along with peak splitting. Further, methanol substituted acetonitrile was used with 0.1% acetate buffer in various ratios to analyse individual drugs and the combination. Acetate buffer (0.1%; pH adjusted to 3.8) showed symmetric peaks and better separation of the analytes. Combination of CUR and NRG, 10 µg each per 1 ml, was injected into the HPLC column after diluting it with mobile phase: methanol and acetate buffer (0.1% acetic acid; pH adjusted to 3.8) at various ratios. Gradient mode of elution with 0.5 ml/min flow rate showed better resolution of peaks. Both the peaks had a tailing factor of less than 2%. Chromatographic parameters along with a typical chromatogram of drugs are presented in Table 1 and Fig. 2, respectively.

Table 1 Chromatographic parameters along with gradient programme for the simultaneous estimation of CUR and NRG
Fig. 2
figure 2

Typical chromatogram representing (a) blank (b) NRG and CUR. Mobile phase was methanol-pH 3.8 0.1% acetate buffer with 0.5 ml/min a flow rate in gradient mode

Validation of the Analytical Method

ICH Q2 R1 guidelines were followed for validation of developed method.

System Suitability and Specificity

NRG and CUR eluted at 5.162 and 11.135 min, respectively, with high resolution and % RSD < 2%, as shown in Fig. 2.

Linearity

The calibration curve (area of CUR and NRG peaks versus the corresponding concentration) was plotted, and the linearity was observed in the concentration range of 0.125 to 16 µg/ml for both drugs. As shown in Table 2, the equations for linearity of CUR and NRG were observed to be y = 24,715x + 10,443 and y = 61,289x + 2556.5, respectively, with regression coefficient 1, indicative of the method to be valid and acceptable.

Table 2 Summary of parameters of the HPLC method

Accuracy and Precision

Known concentrations of combined solution of CUR and NRG at low (8 µg/ml), middle (10 µg/ml), and high (12 µg/ml) levels were injected to determine accuracy and precision. The % drug recovery ranged between 90 and 110%, and the % RSD was < 2.

LOD and LOQ

LOD and LOQ were observed to be 0.063 and 0.125 µg/ml, respectively.

Robustness of the Method

Developed analytical method was deliberately assessed for robustness by varying chromatographic conditions. Varying the flow rate, temperature, injection volume, and wavelength resulted in slight variations in elution time, tailing factor, and theoretical plate count without significantly impacting peak area. The % RSD of < 2 designated the method robustness.

Stability Studies of CUR and NRG by Forced Degradation

CUR and NRG were subjected to forced degradation conditions to establish the potential of the method in autonomously resolving and measuring drug content to a wide range of stress-related conditions. Distinct chromatograms and a few degradation peaks at slightly shifted retention time were observed for both drugs. At 0.1 N HCl, CUR recovery was 41.94%, and NRG recovery was 63.29%. At 1 N HCl, CUR recovery was − 27.44%, and NRG recovery was 38.15%. At 0.1 N NaOH, CUR recovery was − 28.08%, and NRG recovery was 19.625%. At 1 N NaOH, CUR recovery was − 27.73%, and NRG recovery was 4.26%. At oxidative degradation, CUR recovery was 81.59%, and NRG recovery was 101.117%. Photolytic degradation of the CUR and NRG showed 87.42% and 91.89% recovery, respectively. The benchtop stability of the CUR and NRG showed 102.13% and 87.29% recovery, respectively. Thermal exposure of CUR and NRG to 80 °C showed 56.83 and 81.27% recovery, respectively.

Bioanalytical Method Development

Gradient mode of elution with 0.5 ml/min flow rate showed better resolution of peaks. All the peaks had a tailing factor of less than 2%. USFDA guidelines were followed for developing bioanalytical method by HPLC. The method displayed a retention time of 5.2 min and 15.14 min for NRG and CUR, respectively, while the internal standard resveratrol, showed a retention time of 7.7 min (Fig. 3). The method had linearity from 100 to 3200 ng/ml (R2 = 0.9586 for NRG; R2 = 0.9869 for CUR) with the lower limit of quantification (LLOQ) found to be 100 ng/ml. The chromatograms of QC samples are provided in Supplementary data.

Fig. 3
figure 3

Chromatogram representing NRG and CUR (100 ng/ml) eluting at 5.2 min and 15.10 min, respectively, internal standard (IS) resveratrol (400 ng/ml) at 7.7 min

Preparation and Characterisation of CAM

CAM of CUR and NRG (1:1 and 1:2 molar ratios) was prepared by quench cooling technique. The solid-state characterisation of CAM followed by drug content analysis and in vitro studies were carried out.

DSC

CUR, NRG, PM, and CAM were analysed by DSC to study their thermal characteristics. CUR presented a sharp endothermic peak at 182.44 °C, conforming to its crystallinity. NRG showed a broad peak corresponding to its semi-crystalline state; the crystalline PM exhibited endothermic peaks at 157–158 and 180 °C, corresponding to the melting points of NRG and CUR, respectively. The CAM (1:1 and 1:2) showed no endothermic peaks at 182.44 °C and 157–158 °C, respectively, confirming the conversion of CUR to an amorphous state during quench cooling as shown in Fig. 4.

Fig. 4
figure 4

DSC thermograms of crystalline CUR (186.29 °C), NRG, PM (NRG 157–158 °C and CUR 180 °C), and CAM of CUR and NRG

PXRD

The X-ray diffractograms confirmed the formation of CAM. As per the PXRD data presented in Fig. 5, CUR showed characteristic peaks at 8.83°, 14.5°, 17.28°, 18.15°, 23.77°, 25.47°, and 28.89° representing its crystalline nature. NRG did not show any discrete peaks indicating its presence in the semi-crystalline state. The PM (1:1) showed characteristic peaks at 8.39°, 14.18°, 14.76°, 15.5°, 16.74°, 20.67°, 22.45°, 25.47°, and 28.83° whereas, PM (1:2) showed characteristic peaks at 8.59°, 14.16°, 14.74°, 15.39°, 16.75°, 17.78°, 19.06°, 21.38°, and 22.35° corresponding to crystalline CUR. The CAM of CUR and NRG displayed a halo spectrum confirming the amorphous phase of CAM, thus confirming the formation of CAM.

Fig. 5
figure 5

XRD diffractograms of crystalline CUR, NRG, physical mixtures, and CAM of CUR and NRG (1:1 and 1:2)

FT-IR

Any possible intermolecular interactions between the drugs in the CAM were identified by comparing CUR, NRG, and physical mixture spectra with the FT-IR spectra of CAM. The spectra of CUR, NRG, PM, and CAM of CUR and NRG are presented in Fig. 6. CUR showed characteristic peaks at 3496.56, 2358.22, 1625.35, 1596.89, 1423.40, and 1260.31 cm−1 corresponding to phenolic -OH stretching, O = C = O stretching, carbonyl stretching, C = C benzene ring, and C = C aromatic stretching. Distinctive peaks for NRG at 3344.84, 2922.10, 1640.41, 1511.07, and 1361.26 cm−1 indicated phenolic OH stretching, C-H aliphatic stretching, -C = O carbonyl stretching, aromatic bending, and -OH bending, respectively. The spectra of physical mixtures of CUR and NRG were referred to study the intermolecular interactions in the spectra of CAM. The O = C = O stretching in CUR observed at 2358.22 cm−1 shifted to 2362.75 cm−1 in CAM. A shift in the C = O carbonyl stretching was observed from 1625.35 to 1624.97 cm−1 in CAM. In the CAM (1:2), the O = C = O stretching was observed at 2362.75 cm−1, and the C = O carbonyl stretching shifted from 1625.35 to 1628.45 cm−1. The phenolic -OH stretching in CUR at 3496.56 cm−1 is not observed in the spectra of CAM. The C-H stretching peak in NRG at 2922.10 cm−1 is not evident in CAM. It can be concluded that hydrogen bonding between phenolic -OH in CUR and the C-H aliphatic group in NRG is likely to be a driving force for the formation of CAM. Other characteristic peaks of the drugs were either shortened or broadened in CAM.

Fig. 6
figure 6

FTIR spectra of crystalline CUR, NRG, physical mixtures, and co-amorphous formulations of CUR and NRG

Scanning Electron Microscopy (SEM)

SEM analysis was done to visualize the surface morphology of pure drugs and CAMs. CUR exhibited a spherical structure (Fig. 7a), while NRG showed a sheet-like structure (Fig. 7b) as observed by J. Li et al. (Li et al. 2022). The PM showed characteristics of both CUR and NRG (Fig. 7c). The CAMs showed irregularly shaped blocks representing the presence of drugs in their amorphous form (Fig. 7d and e). The structural irregularity in CAMs compared to CUR and PM may be associated with the thermodynamic due to quench cooling (Li et al. 2021). The irregular structures could increase surface area thereby enhancing solubility of CUR (Mohapatra et al. 2021).

Fig. 7
figure 7

SEM images of a CUR, b NRG, c PM, d CAM 1:1, and e CAM 1:2

Quantification of CUR and NRG in CAM

The developed method was able to estimate CUR and NRG simultaneously. The CAM prepared by the quench cooling technique was evaluated for drug content. The CAM showed 64.42 and 92.24% of CUR and NRG, respectively.

Solubility Studies

Solubility studies of CUR, physical mixtures, and CAM were carried out in USP pH 6.8 phosphate buffer. The samples were analysed by HPLC. The solubility of CUR and NRG was found to be 1.35 and 13.8 µg/ml, respectively, and that of CUR in physical mixtures (1:1 and 1:2) was 22 and 41 µg/ml, respectively. The solubility of CUR in CAM (1:1 and 1:2) was 81.5 and 252.5 µg/ml, respectively, indicating significant improvement in the solubility of CUR compared to its crystalline counterpart.

In Vitro Drug Release Studies

The release studies were performed at sink conditions to determine the behaviour of crystalline CUR and CUR in CAM in an aqueous medium. The release profile of CAM was compared with the physical mixtures of CUR and NRG. It was deceptive that the release of CUR in both acidic and basic buffers (Fig. 8a and b, respectively) increased with time, attaining a steady state after 2 h. The CAM (1:2) displayed a maximum rate of dissolution of CUR compared to CAM (1:1). The physical mixtures exhibited a significantly lesser extent of release of CUR.

Fig. 8
figure 8

a Drug release profiles of (a) crystalline CUR, CUR in physical mixtures (b) 1:1 and (c) 1:2, and CUR in CAM (d) 1:1 and (e) 1:2 in pH 1.2 HCl. b Drug release profiles of (a) crystalline CUR, CUR in physical mixtures (b) 1:1 and (c) 1:2, and CUR in CAM (d) 1:1 and (e) 1:2 in USP pH 6.8 phosphate buffer

Pharmacokinetic Studies

It is necessary to render the advantage of solubility into oral bioavailability to measure the effectiveness of CAM. CUR is a P-gp substrate and undergoes phase II metabolism, CAM improves the oral bioavailability of CUR due to its presence in an amorphous state. The two possible reasons for improving oral bioavailability of CUR could be the improved solubility and P-gp efflux inhibition in the presence of NRG, abetting rapid transit into the blood (Yarlagadda et al. 2023). During pharmacokinetic study, CUR demonstrated a maximum plasma concentration of 45.37 µg/ml at 6 h. Based on in vitro studies, CAM 1:2 showed better efficacy compared to CAM 1:1 and was chosen for in vivo pharmacokinetic study. CUR in PM and CUR in CAM exhibited a maximum concentration in plasma of 59.76 µg/ml at 8 h and 113.76 µg/ml at 6 h, respectively (Fig. 9). The pharmacokinetic parameters were determined using GastroPlus™ software, employing non-compartmental and one-compartmental model. Table 3 shows the pharmacokinetic parameters of CUR, PM, and CAM (1:2). As per the non-compartmental and one-compartment analysis, CAM showed 14.9- and 7.8-fold increase in the AUC than the pure CUR and PM, respectively. The Cmax of CUR in CAM was increased by 2.75 and 1.46-fold compared to pure CUR and PM, respectively. The enhanced plasma concentration exhibited by CAM of CUR and NRG demonstrates CUR’s improved solubility and oral bioavailability.

Fig. 9
figure 9

Plasma concentration and time profile (a) CUR, (b) CUR in PM, and (c) CUR in CAM 1:2

Table 3 Pharmacokinetic parameters after administering a single dose of CUR equivalent to 500 mg/kg orally (n = 3)

Conclusion

CUR has been widely used in food industries, cosmetic preparations, and pharmaceutical arena. NRG and CUR are used in the development of functional food that aids in the treatment and management of various ailments. In this work, sensitive, robust, stability-indicating analytical method is developed and validated for the quantification of CUR and NRG simultaneously. Both the drugs eluted within 17 min minimising the analysis time and solvent usage resulting in a cost-effective method. Lower limits of detection and quantification with a wide linearity range represent method sensitivity. Forced degradation analysis in the presence of various stress factors such as acid, alkali, heat, oxidation, and light concluded that both the drugs are stable. CAM of CUR and NRG was prepared by quench cooling. DSC and XRPD confirmed the conversion of CUR to an amorphous form. FT-IR revealed hydrogen bonding between CUR and NRG. The developed method was applied in quantifying CUR and NRG, solubility studies, and in vitro drug release studies. The solubility and drug release of CUR in CAM were significantly enhanced compared to crystalline CUR. The increase in the release of CUR from CAM (1:2) could be due to improved solubility and rate of dissolution of amorphous forms than crystalline counterpart. NRG is a glass-forming agent due to the presence of abundant hydroxyl groups in the glycan part that offers possible supramolecular hydrogen bonding sites, vital for inducing disorder in the co-amorphous phases. The profound effect of co-former NRG on the amorphous state of CUR, its solubility, and oral bioavailability is thereby evident from the study. However, further characterisation and evaluation studies are required to evaluate the efficacy of the formulations.