## Introduction

As NASA prepares for exploration missions beyond low Earth orbit with no opportunities for resupply, pharmaceutical instability may present significant risk to the crew health due to the loss of efficacy over time. Astronauts suffer from sleep disruption during space flight that can affect their mental and physical health in performing routine work1,2,3,4. Therefore, mitigation against the effects of space radiation on pharmaceutical drugs is necessary for the success of long-term space missions. NASA is primarily concerned with the health risks for astronaut exposure to radiation and other flight infections. NASA estimates that a 30-month mission to Mars would expose astronauts to over 900 millisieverts (mSv) (1000 mSv = 1 gray), far beyond safe or acceptable Occupational Safety and Health Administration (OSHA) limits of 50 mSv in the United States5,6. Persistent radiation induces oxidative stress associated with the central nervous system, blood coagulation abnormalities, fibrosis, heart failures, gene expression changes, cancer as long-term effects and other physiological abnormalities7,8,9,10.

High energy radiation and oxidative stress due to UV and gamma radiation induce radiolysis of water generating highly reactive hydrogen (H•) and hydroxyl (OH•) radicals that diffuse to other molecules and react with particular chemical groups resulting in a loss of biochemical activity11. There are three mechanisms of radiation damage caused by OH• in the chemical structure of a compound: the abstraction of $$-H$$ by OH• leaving an unpaired electron on the carbon atom, the formation of a hydroxy derivative by breaking double bonds (e.g., $$-C=C-$$) and the electron transfer from anionic groups (e.g., $${{Cl}}^{-}$$) to OH•11. The production of reactive OH• has been demonstrated by chemical means such as UV photolysis of hydrogen peroxide (H2O2) and gamma irradiation using colbalt-60 (60Co) or cesium 137 (137Cs)12,13,14,15,16,17. The effects of radiation on a variety of pharmaceutical drug molecules have been studied using either gamma radiation or H2O2 hydrolysis by UV light at different doses that show linear proportionality of the decreasing amount of active substances with the dose of radiation18,19,20,21,22.

The objective of this work is to engineer a self-protected medication using a polymeric nanoparticle assembly of a biocompatible polymer, poly (lactic-co-glycolic acid) (PLGA), a free radical scavenger molecule (Trolox) and a linker poly-l-lysine (PLL) (Fig. 1a). Melatonin is used as a model relevant to space flight missions. In this design, melatonin was encapsulated in the core of PLGA polymer nanoparticles using an oil/water single emulsion solvent evaporation method. In a separate reaction, Trolox was conjugated to PLL via EDC/NHS chemistry for conjugating the carboxylates in Trolox with the amines in PLL (Fig. 1b). The residual amine groups on PLL-Trolox were further conjugated on the surface of melatonin-loaded PLGA nanoparticles via the same process of EDC/NHS reaction. It is hypothesized that Trolox conjugation on the surface of nanoparticles (PolyRad) protects the chemical stability of melatonin during spaceflight mimicking conditions that generate excessive radiation-induced free radicals through its potential for antioxidant capacity and free radical scavenging activity. We further hypothesize that the PolyRad approach offers great benefits by increasing the specific surface area and accelerating the dissolution velocities of drug molecules, which ultimately has a beneficial impact on the drug release and associated bioavailability.

## Materials and Methods

### Materials

All reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA) unless otherwise specified.

Melatonin-PLGA-PLL-Trolox (PolyRad) nanoparticles were prepared in three steps: entrapment of melatonin in PLGA (melatonin-PLGA), conjugation of Trolox with PLL (PLL-Trolox), and surface conjugation of the PLL-Trolox on melatonin-PLGA nanoparticles (melatonin-PLGA-PLL-Trolox). Melatonin (N-acetyl-5-methoxytryptamine) was encapsulated in PLGA polymer using water-in-oil (w/o) emulsion phase separation method (Fig. 1a)47,48,49. Briefly, 1 ml of each 1% (w/v) of melatonin dissolved in DMSO and 5% (w/v) of PLGA polymer dissolved in dichloromethane (DCM) were added to 20 ml of 1% (w/v) polyvinyl alcohol (PVA) solution using a syringe pump. The flow rate of 1 ml/h was maintained under sonication and homogenization. The nanoparticle suspension was stirred (350 rpm) overnight (for about 18 h) to remove DCM by evaporation.

Conjugation of Trolox to PLL was carried out by activation of carboxyl groups in Trolox followed by reaction with the primary amine groups of PLL polymer resulting in the formation of amide bonds (Fig. 1b). Briefly, 4 mg of 1-ethyl-3-3-dimethylaminopropyl-carbodiimide (EDC) was directly added to 1 ml of 10 mg/ml Trolox dissolved in phosphate-buffered saline (PBS) of pH 7.4. Immediately, 6 mg of N-hydroxysuccinimide (NHS) was added to the reaction. The reaction solution was mixed gently, maintained at pH 7.4, and left at room temperature (R.T.; 25 °C) to react for 15 min. PLL (11.5 mg) was added to the solution that was incubated at R.T. for 2 h while shaking (450 rpm). The residual primary amine groups of PLL-Trolox were conjugated to the surface of melatonin-PLGA nanoparticles using EDC/NHS chemistry. The final reaction solution was filtered using 100 kDa Amicon Ultra-4 centrifugal filters to remove any excess reagents.

### PolyRad characterization: size, surface charge, and melatonin encapsulation efficiency

The size and surface charge of particles before and after PLL-Trolox conjugation on melatonin encapsulated PLGA nanoparticles were measured using dynamic light scattering (DLS) and zeta potential measurements, respectively using a NanoSeries Zetasizer ZS 90 (Malvern). Samples were prepared in deionized (DI) water. Encapsulation of melatonin in PolyRad nanoparticles was assessed by measuring the absorbance at 300 nm using a microplate reader (BioTek Synergy 2) and a melatonin standard curve (S.I. Figure 1). The percentage of melatonin encapsulation capacity was calculated using the following Eq. 1:

$$\% \,encapsulation\,efficiency=\frac{{M}_{0}-({M}_{u}-{M}_{s})}{{M}_{0}}\times 100$$
(1)

where, M0 = Initial melatonin mass, Mu = Melatonin mass in unwashed suspension and Ms = Melatonin mass in supernatant$$.$$

### Characterization of PLGA-PLL-Trolox formulation using Fourier Transform Infrared (FT-IR) Spectroscopy

FTIR spectroscopy was used to confirm the structural changes that occurred after PLL-Trolox conjugation with PLGA nanoparticles. Samples for FT-IR spectroscopy were prepared by mixing and grinding freeze-dried nanoparticles with FT-IR grade potassium bromide (KBr; Alpha Aesar) (w:w ratio of 1:100) to make pulverized pellets. FT-IR absorption spectra were obtained using a Thermo Nicolet NEXUS 470 FTIR instrument in the wavelength range from 4000 to 400 cm−1 at a resolution of 4 cm−1 using 32 scans per sample. The effect of background noise from the ambient air without a sample in place was subtracted from the sample spectra. All spectra were analyzed using EZ OMNIC E.S.P v.5.1 software (Thermo Scientific). The resulting FTIR spectra of PLGA-PLL-Trolox were compared to individual spectra of PLGA, PLL, and Trolox to characterize the conjugation reaction based on the intensity or shift of the vibrational bands.

### Determination of the composition of melatonin-PLGA-PLL-Trolox (PolyRad) nanoparticles using1H-NMR spectroscopy

For 1H-NMR spectroscopy, 10 mg of PolyRad nanoparticles, melatonin alone, PLGA alone, PLL alone, and Trolox alone were dissolved in 750 μl of DMSO-d6 except for PLGA, which was dissolved in d-Chloroform. Each solvent was used as the internal reference to determine chemical shifts (δ) in ppm. 1H-NMR spectra were then recorded using Bruker advanced III 400 MHz Liquid-State NMR instrument at R.T.

### Fluorometric Hydroxyl Radical Scavenging Capacity (HOSC) assay

A fluorometric HOSC assay was applied to determine the mechanisms of highly reactive free radical absorption by PLGA-PLL-Trolox composite and Trolox alone was used as a standard50. OH• radicals were generated using Fe (III)/H2O2 Fenton-like reaction system in a pH 7.4 sodium phosphate buffer51. Assay reactions were carried out in Costar black 96-well polystyrene plates and analyzed using a microplate reader (BioTek Synergy 2) with an excitation wavelength of 485 nm, emission wavelength of 530 nm, and 0.1 s read time for each well with each plate read once per minute for 3 h. The reaction mixture contained 170 μl of 9.28 × 10−8 M fluorescein (FL) prepared in 75 mM sodium phosphate buffer (pH 7.4), 30 μl of blank or antioxidant, 40 μl of 0.1990 M H2O2 and 60 μl of 3.43 mM FeCl3 added in that order50. Trolox concentrations of 20, 40, 60, 80, and 100 μM were used. Relative fluorescence intensity of each Trolox standard concentrations were calculated by dividing individual fluorescence readings by the initial fluorescence reading at time equal to zero.

To examine the resistance to photochemical degradation of melatonin, 2 mg of PolyRads containing 2.5 μg melatonin were dissolved in 1 ml 40% acetonitrile-60% water (v/v) solvent and spiked with 0.75 mM hydrogen peroxide (H2O2) followed by UV radiation at 254 nm and 39 W spectral output for 0, 1, 2, 4, 6 and 8 h. H2O2 is known to be decomposed into hydroxyl radical (OH•) and superoxide ions ($${O}_{2}^{-{\boldsymbol{.}}}$$) when exposed to UV radiation. Pure melatonin powder was used as a control and were treated with 0.75 mM H2O2 followed by UV irradiation at every 20 min for up to 2 h. The reaction progress was monitored by determination of melatonin concentration and its two degradant products (N1-acetyl-N2-formyl-5-methoxykynuramine, AFMK and 6-hydroxymelatonin) using Agilent 1260 Liquid Chromatography (LC) High-Performance Liquid Chromatography (HPLC) apparatus equipped with Waters C18 column of 4.6 × 250 mm (particle size 5 μm). The system consisted of a vacuum degasser, a binary pump unit, an autosampler for sample injections, a thermostat column compartment, and a photodiode array detector (DAD) for peak analysis at 304 nm. The mobile phase used was acetonitrile −25 mM ammonium acetate (40:60, v/v). The system was run at ambient temperature (~22 °C) with a flow rate of 1.0 ml/min. Sample injections of 30 μl were taken and observed for characteristic peaks over 8 min intervals between injections. Quantification of melatonin and degradation products were carried out using calibration curves obtained by repeated injections of standard solutions of known concentration.

The effects of gamma irradiation on the chemical structure of melatonin were carried out using radioactive cobalt 60 (60Co) isotope of activity 200 mCi. Two sample groups of melatonin alone and PolyRad were prepared in 1 ml of 10 mg/ml aqueous solutions and irradiated at 1, 5 and 10 Gray (Gy) doses. Non-radiated (0 Gy) melatonin and nanoparticles were used as controls. The samples were lyophilized and dissolved in DMSO-d6. The effects of gamma irradiation on the structure of melatonin alone and melatonin in PolyRad nanoparticles were investigated using 1H-NMR spectroscopy (Bruker instrument 400 MHz). The spectra were calibrated and normalized relative to DMSO-d6 at 25 °C (δ = 0 ppm). After 1H-NMR measurements, the relative area under the curves were calculated with respect to non-radiated melatonin standards and plotted as a function of radiation doses.

### In vitrodrug release kinetics

In vitro release of melatonin from non-irradiated and irradiated PolyRad was carried out in PBS at pH 7.4. Each 10 mg of PolyRad and melatonin powder were put in separate 1.5 ml centrifuge tubes and exposed to 0.75 mM H2O2 followed by UV irradiation for 2 h. PolyRad was suspended in 10 ml PBS and kept in a 37 °C water bath. PBS solution of 500 μl was withdrawn at 0, 5, 10, 15 and 30 min and then at 1, 2, 4, 8, 12, 24, 48 and 72 h. The solution was centrifuged at 1000 rcf for 1 min. The supernatant (150 μl) was put in a 96 well assay plate in triplicates. PBS was used as a blank. Melatonin concentrations were measured using absorbance at 300 nm using the microplate reader and melatonin standard curve. The drug release kinetics data were plotted as log cumulative drug release versus log time (Eq. 2).

$$\log ({M}_{(t-\tau )}/{M}_{\infty }-b)=m\,\log (t-\tau )+\,\log \,k$$
(2)

where $${M}_{(t-\tau )}$$ is the cumulative drug release at time $$t$$, $${M}_{\infty }$$ is the cumulative drug release at infinite time, $${M}_{(t-\tau )}/{M}_{\infty }$$ is the fractional drug release at time $$t$$, $$\tau$$ is the lag time, $$b$$ is the fractional drug burst release, $$k$$ is a kinetic constant that measures the drug release rate characteristic of the drug/polymer system, and $$n$$ is the release exponent which characterizes the drug release mechanism52,53,54,55,56,57,58.

### Melatonin cytotoxicity

To determine and compare the cytotoxicity of melatonin on endothelial cells, human umbilical vein endothelial cells (HUVEC) were treated with free melatonin, melatonin-loaded PolyRad (carrier control), PolyRad, PolyRad treated with 10 Gy of gamma radiation, and PolyRad treated with 0.75 mM of H2O2 followed by 1 h of UV radiation. HUVEC were seeded in Costar 96 well assay plate (sterile, black, clear-bottom with lid) at a cell density of 10,000 cells/well. The cells were incubated in PromoCell medium with 10% FBS and 1% penicillin-streptomycin (10,000 units/ml) at a 37 °C and 5% CO2 incubator. Cells were allowed to grow overnight. The cells were incubated for 72 h with melatonin solution control and PolyRad at melatonin concentrations of 0.01, 0.1, 1 and 10 mg/ml. PBS was used as a negative control. Untreated viable (live) HUVEC were used as a positive control. After 72 h, cells were rinsed with PBS and cellular viability were measured using MTT assay (Invitrogen). The reduction of MTT to formazan by viable HUVEC was quantified measuring absorbance at 570 nm. Absorbance of treated cells were measured and calculated by subtracting the mean background level of wells containing the medium. The % cell viability was calculated as absorbance ratio of treated cells to live cells as shown in Eq. (3):

$$\% \,{\rm{cell}}\,{\rm{viability}}=\frac{{A}_{570{\rm{of}}{\rm{sample}}}-{A}_{570{\rm{of}}{\rm{medium}}}}{{A}_{570{\rm{of}}{\rm{live}}{\rm{cells}}}-{A}_{570{\rm{of}}{\rm{medium}}}}\times 100$$
(3)

To visualize cells under fluorescence microscope, live and dead cells were stained using 2 μM calcein AM and 4 μM of EthD-1, respectively following the manufacturer’s protocol. Images were acquired using a 10X Plan Fluor (NA = 0.3) objective equipped with a Carl Zeiss Axio Observer Z1 microscope at excitation/emissions of 470 ± 40/525 ± 50 and 545 ± 25/605 ± 70, respectively.

## Results

We synthesized melatonin-encapsulated PLGA nanoparticles with an average diameter of 325 ± 21 nm (Fig. 2a), polydispersity index (pDI) of 0.07 ± 0.05, and a zeta potential of about −23 ± 8.5 mV (Fig. 2b) as confirmed by the DLS analysis. The PLL-Trolox shell attributed to the increase in particle size due to the deposition of its complex onto the surface of melatonin-PLGA nanoparticles. The average diameter of PLL-Trolox conjugated PolyRad nanoparticles were approximately 90% larger (622 ± 114 nm; Fig. 2c) than melatonin-PLGA control nanoparticles without PLL-Trolox shell as seen by the shift in the average peak intensity to the right-hand side. The pDI and zeta potential increased to 0.23 ± 0.12 and 29 ± 10 mV (Fig. 2d), respectively, indicating the successful conjugation of positively charged PLL on the surface of nanoparticles. The encapsulation efficiency of melatonin inside nanoparticles was determined as 44 ± 12% (Table 1).

### Fourier-Transform Infrared Spectroscopy (FT-IR) for the determination of PLL and Trolox conjugation to PLGA nanoparticles

PLL-Trolox conjugation to PLGA nanoparticles was accomplished by reacting amine groups of PLL with carboxyl groups (-COOH) in PLGA. FTIR studies were carried out to characterize the related structural modifications (Fig. 3). PLGA spectra’s absorbance peaks appear strong at 1,765–1,720 cm−1 for ester carbonyl and carboxyl C=O stretch, ether C-O-C stretch absorbance peak at 1,290–1,180 cm−1, and secondary alcohol C-O stretch at 1,124–1,087 cm−1 (Fig. 3a). Broad medium absorbance peak of PLGA alkyl C-H stretch at 3,000–2,840 cm−1 and alcohol O-H at 3,550–3,200 cm−1. PLL’s spectra present strong amide C=O stretch peak at 1,680–1,640 cm−1, a broad, strong primary and secondary N-H stretch peak at 3,500–3,250 cm−1 and 3,350–3,310 cm−1, respectively (Fig. 3b). Also, a broad medium bend of PLL alkyl C-H stretch peak and amine N-H stretch appear correspondingly at 3,000–2,840 cm−1 and 1,650–1,540 cm−1. Trolox spectra shows strong phenol O-H peak at 3,550–3,200 cm−1 and carboxylic C=O stretch peak at 1,720–1,706 cm−1 (Fig. 3c). Numerous ether C-O-C peaks at 1,300–1,000 cm−1 and a broad, medium carboxylic O-H stretch peak at 3,300–2,500 cm−1 were observed. PLGA-PLL-Trolox composite spectrum displays strong secondary N-H, medium alkyl C-H, and amide C=O stretch peaks from PLL and strong phenol O-H, ether C-O-C stretch peaks from Trolox, and ester carbonyl C=O stretch and secondary alcohol C-O stretch peaks from PLGA, confirming the presence of three blocks in the synthesized composite (Fig. 3d).

### Confirmation of the composition of PolyRad nanoparticles using1H NMR Assay

Formation of melatonin-PLGA-PLL-Trolox nanoparticles was evaluated using 1H-NMR (Fig. 4). The NMR results depicted a successful conjugation of PLL-Trolox on melatonin encapsulated PLGA nanoparticles to form PolyRad nanoparticles compared to their independent control NMR peaks (melatonin, PLGA, PLL, and Trolox alone). In Fig. 4, proton peaks at 2.5 and 7.25 ppm represent solvent background signals from DMSO-d6 and d-Chloroform, while the broad peak at 3.25 ppm represents residue water proton peak. Proton peaks from PLL, denoted by a, b, c and d which correspond to -CH-, -NH-, -CH2-, and -NH2 groups occurred at 1.7, 3.5, 4.75 and 5.2 ppm, respectively. Also, melatonin distinctive peaks denoted by letters ‘e-n’ correspond to the -CH3-, the two -CH2-, -CH3O-, the three phenyl -CH-, the cyclic -CH-, and the two -NH- respectively. These peaks were recorded at 1.8, 2.75, 3.25, 3.75, 6.75–7.25, 7.25, 7.8 and 11.6 ppm, respectively. For Trolox, distinctive peaks were lettered ‘o-v’. Trolox 1H-NMR peaks showed propionic acid -CH3- (o) and -OH- (p) protons at 1.5 and 1.75 ppm respectively, cyclic -CH2- (q and r) at 1.9–2.1 ppm, phenyl methane protons (s, t and u) at 2.0–2.2 and 2.6 ppm, and the phenyl -OH- (v) protons at 7.25 ppm. Proton peaks for PLGA were labeled from ‘w-z’ and then ‘#’. PLGA proton peaks displayed lactide -CH3- (w), -CH- (y), and glycolide -CH2- (x) protons at 1.5, 6.2 and 5.5 ppm respectively. However, the lactide end and the glycolide end -OH- protons (z and # respectively) in PLGA were missing probably by being lost to neighboring protons. Comparing with the 1H-NMR of the individual components, 1H-NMR of nanoparticles displayed virtually all the peaks that were shown in the individual components as shown in Fig. 4. However, the chemical shift of protons (b) in PLL formed an amide bond with the lactide end -OH- protons (z) of PLGA, hence we had that this amide bond occurred around 3–3.5 ppm in the NPs’ 1H-NMR. Likewise, proton peak at 5.25 ppm in the NPs’ 1H-NMR corresponds to the -NH- group of the corresponding amide bond (CONH) between primary amino group of PLL (d) and the carboxyl group of Trolox (p).

### Scavenging capacity of Trolox using fluorometric HOSC assay

Trolox serves as a potent scavenger of OH• radicals, efficiently blocks the effects of highly reactive free radicals even at low concentration and therefore, is tested to prevent the photochemical degradation of melatonin in PolyRad nanoparticulate form. Fluorometric Hydroxyl Radical Scavenging Capacity (HOSC) was performed to analyze the OH• scavenging capacity of Trolox in its free solution. Results showed that an increase in Trolox concentration in the HOSC medium increased the free radical scavenging capacity of Trolox. Figure 5 shows the fluorescence decay curve of fluorescein with time in the presence of different Trolox concentrations. In the absence of Trolox (Blank), the fluorescence intensity of fluorescein was quenched rapidly with more than 50% reduction in the relative fluorescence intensity within 10 min, indicating the reaction of fluorescein with OH• radicals caused by the Fenton like Fe3+/H2O2 system. Trolox worked as OH• radical scavenging agent that delayed the decay in fluorescence intensity of fluorescein by diminishing the quenching rate significantly in a dose-dependent manner. As the Trolox concentration was increased from 20 to 100 μM, the capacity for scavenging OH• radicals increased. The ~100% fluorescence intensity retention time were 10, 50, 80, 100 and 120 min using 20, 40, 60, 80 and 100 μM Trolox, respectively.

To study the effect of photochemical oxidation on the structure of melatonin in PolyRad formulation, irradiation experiments were performed on PolyRad suspensions using H2O2 in 40% acetonitrile-60% water solvent followed by UV light exposure. Bare melatonin was used as a control. This process relies on the production of OH• through the irradiation of H2O2 with UV radiation: $${H}_{2}{O}_{2}\mathop{\to }\limits^{h\nu }2{\rm{OH}}\bullet$$ that degrades melatonin12,15. The influence of H2O2 and UV on melatonin degradation was analyzed with time using HPLC (Fig. 6a). The melatonin peak did not change by more than ~20% by UV treatment only for 2 h (Fig. 6a; hatched column). UV irradiation in the presence of H2O2 influenced the disappearance of the melatonin peak by ~80% within 2 h (Fig. 6a; filled column). Pure melatonin control showed only one characteristic peak with a retention time of 5.67 min eluted from the HPLC column (Fig. 6b). The photochemical degradation of melatonin (N-acetyl-5-methoxytryptamine) formed two major degradant peaks (B and C) at 4.7- and 1.98-min retention times for N1-acetyl-N2-formyl-5-methoxykynuramine (AFMK) and 6-hydroxymelatonin, respectively4,59,60,61. The measured changes of AFMK and 6-hydroxymelatonin concentrations increased as a function of different UV exposure time in the presence of H2O2 (SI Figs. 2–4). Also, a peak at ~ 3.95 min, likely due to the melatonin degradation product, N-acetyl-5-methoxykynuramine (AMK) is seen. The overall reaction followed a first-order reaction: $$-{r}_{{Melatonin}}=k{C}_{{melatonin}}$$, where $$-{r}_{{Melatonin}}$$, $$k$$ and $${C}_{{melatonin}}$$ represent the rate of disappearance of melatonin, reaction rate constant and concentration of melatonin, respectively (SI Fig. 5). The reaction rate constant ($$k$$) is determined as $$(8\times {10}^{-5}){{\min }}^{-1}$$ (SI Fig. 5a). In contrast to bare melatonin (Fig. 6c, hatched column), a remarkable reduction of melatonin degradation was observed using PolyRad nanoparticles (Fig. 6c, filled column). PolyRad protected ~75, 65, 55 and 15% of the active structure of melatonin after UV irradiation only, and 2 h of UV followed by 0.75, 1, and 5 mM of H2O2 exposure, respectively indicating that nanoparticles scavenged H2O2 in a concentration-dependent manner. Encapsulation of melatonin using PolyRad nanoparticles reduced the formation of AFMK by ~50% (SI Fig. 6).

### 1H-NMR spectrometry studies on molecular structure of melatonin after gamma radiation

Based on the 1H NMR spectrum of melatonin as presented in Fig. 7a and Table 2, major melatonin peaks are observed at 1.9 (NHCOCH3), 2.9 (CH2), 3.5 (CH2NH), 3.9 (OCH3) and 6.9–7.4 ppm (benzene aromatic rings and NHCHO). Free melatonin is degraded into other compounds that appear at 3.2, 3.6 and 3.78 ppm and predominantly at 10.6 (OH•) and 10.9 (NHCO) ppm. Melatonin in PolyRad nanoparticles did not show any significant differences between the non-radiated control and gamma irradiated PolyRad samples. The percentage of active melatonin remaining was calculated using the integral ratio of peaks of groups I through V (Table 2) based on the non-radiated melatonin control (Fig. 7b). The structural integrity of melatonin alone was completely degraded to 25.7, 10.8 and 5.5% at 1, 5 and 10 Gy, respectively compared to non-radiated free melatonin. The nanoparticle plays a critical role in the protection of melatonin by >98% most likely by scavenging of free radicals generated from gamma radiation62.

### In vitrodrug release kinetics

The melatonin release kinetics from PolyRad is shown in Fig. 8a before and after radiation. A total of 70% melatonin was released from non-radiated nanoparticles after 72 h at pH 7.4. Notably, more than 50% of the drug was released after only 2 h, suggesting a spontaneous rapid degradation of solid nanoparticles under physiological conditions. Exposure to radiation elicited higher drug release, reaching >90% drug release in 72 h. The drug release profile follows a biphasic pattern, indicating a controlled drug release mechanism. The data from in vitro experiments were fitted to a power-law model to investigate the kinetics of melatonin release (Fig. 8b and Table 3). The cumulative release is linear with respect to time, $$t$$ ($${R}^{2} > 0.97$$) where diffusion exponent $$n$$ varies between 0.1–0.3, indicating a Fickian diffusion profile. The rate constant ($$k$$) varies between 0.3–0.4 suggesting a slow melatonin release from PolyRad.