Liquiritigenin‐Rich Hydroalcoholic Extract of Brazilian Red Propolis Reduces Dyskinesia Induced by 3,4‐ Dihydroxyphenylalanine in Hemiparkinsonian Rats

ABSTRACT

The set‐up and progression of dyskinesia induced by 3,4‐dihydroxyphenylalanine (L‐DOPA) are strongly linked to oxidative stress and neuroinflammation. The aim of this work was to study and characterize the effects of the hydroalcoholic extract of Brazilian red propolis (HERP) on L‐DOPA–induced dyskinesia (LID) in hemiparkinsonian rats injected with 6‐hydroxydopamine (6‐OHDA) into the medial forebrain bundle (MFB). The abnormal involuntary movements (AIM) and spontaneous motor parameters were evaluated over 21 days of treatment, and then immunohistochemistry for glial fibrillary acidic protein (GFAP) and tyrosine hydroxylase (TH) was performed on rat brains. The presence of biochanin A, formononetin and the major compound liquiritigenin in HERP was confirmed by HPLC‐DAD. HERP presented a high antioxidant effect in vitro, while in vivo the locomotive, orolingual, limb and axial dyskinetic effects of L‐DOPA were counteracted with 10 mg/kg of HERP and 40 mg/kg of amantadine (AMAN). However, HERP alone did not reduce antiparkinsonian effects of L‐DOPA in behavioural assessment. Immunostaining showed that L‐DOPA increased GFAP expression, which was decreased by HERP and AMAN. HERP decreased the ipsilateral loss of TH expression, whereas HERP and AMAN increased contralateral expression of TH at the mesencephalon. HERP induced antidyskinetic effects in rats, with motor improvement, which is an advancement in comparison to standard medications, and these effects may be mediated by astrocyte‐related mechanisms.

Abbreviations

AIM

abnormal involuntary movements

AMAN

amantadine

AP

anteroposterior

BBB

blood–brain barrier

BDNF

brain‐derived neurotrophic factor

CONCEA

Brazilian College of Animal Experimentation

DAB

3,3‐diaminobenzidine

DAT

dopamine transporter

DPPH

2,2‐diphenyl‐1‐picrylhydrazyl

DV

dorsoventral

Eq.

equivalents

GA

gallic acid

GDNF

glial cell line–derived neurotrophic factor

GFAP

glial fibrillary acidic protein

HE

haematoxylin and eosin

HERP

hydroalcoholic extract from Brazilian red propolis

HPLC‐DAD

high‐performance liquid chromatography–diode array detector

IL

interleukin

iNOS

inducible nitric oxide synthase

ip

intraperitoneal

L‐DOPA

3,4‐dihydroxyphenylalanine

LID

L‐DOPA–induced dyskinesia

LL

lateral

MFB

medial forebrain bundle

NIH

National Institutes of Health

NMDA

N‐methyl‐D‐aspartate

nNOS

neuronal nitric oxide synthase

NO

nitric oxide

NOS

nitric oxide synthase

OF

open field

PBS

phosphate buffered saline

PD

Parkinson's disease

PTFE

polytetrafluorethylene

RB

rotational behaviour

RU

rutin

ROS

reactive oxygen species

sc

subcutaneous

SNpc

substantia nigra pars compacta

TBS

Trizma base buffer

TGF‐β

transforming growth factor beta

TH

tyrosine hydroxylase

TNF‐α

tumour necrosis factor alpha

TrkB

tropomyosin receptor kinase B

VC

vehicle

VMAT2

vesicular monoamine transporter 2

6‐OHDA

6‐hydroxydopamine

Summary.

The effects of the hydroalcoholic extract of Brazilian red propolis (HERP) on L‐DOPA–induced dyskinesia (LID) were studied in hemiparkinsonian rats injected with 6‐hydroxydopamine (6‐OHDA) into the medial forebrain bundle (MFB). HERP showed a high antioxidant effect in vitro. In vivo, the locomotive, orolingual, limb and axial dyskinetic effects of L‐DOPA were counteracted over 21 days of treatment with 10 mg/kg of HERP and 40 mg/kg of amantadine (AMAN). HERP alone did not reduce antiparkinsonian effects of L‐DOPA in behavioural assessment but led to antidyskinetic effects in rats, with motor improvement. This study highlights the advancements of HERP for the treatment of dyskinesia in comparison to standard medications, with effects potentially mediated by astrocyte‐related mechanisms.

1. Introduction

The most common strategy to manage patients bearing Parkinson's disease (PD) consists of dopaminergic replacement therapy by administering the 3,4‐dihydroxyphenylalanine (L‐DOPA) dopamine precursor. This therapy is efficient during the first years of administration because, in the early stages of PD, the remaining dopaminergic neurons are able to store exogenous dopamine and regulate its release, thus maintaining the physiological stimulation of dopamine receptors in the striatum [1]. As the disease progresses and dopaminergic neurons are lost, the levels of dopamine are dependent on exogenous L‐DOPA, which presents a very short plasma half‐life, resulting in a pulsatile stimulation of striatal receptors [2]. Thus, long‐term pharmacotherapy leads to disabling side effects [3], characterized by abnormal involuntary movements (AIM), which are known as L‐DOPA–induced dyskinesia (LID) [4, 5].

Relevant insights about the development of LID include maladaptive corticostriatal synaptic plasticity and increased glutamate release, which impact the responsiveness of striatal medium spiny neurons to dopamine and promote dyskinetic behaviours [6]. This hypothesis is supported by the fact that reducing striatal hyperactivity through the administration of N‐methyl‐D‐aspartate (NMDA) antagonists, such amantadine (AMAN), alleviates LID [7]. AMAN is the only FDA‐approved treatment for LIDs and is efficient against parkinsonian symptoms, such as motor fluctuations [8]. However, several serious adverse effects, e.g., confusion, hallucinations and depression, have been identified in patients treated with AMAN [9]. Thus, the need to develop alternative treatments is evident.

Other important aspects in the development and progression of LID include nitric oxide (NO)–mediated oxidative stress and neuroinflammation [10, 11]. In fact, dopamine is metabolized by neurons and astrocytes, and the chronic administration of L‐DOPA may lead to the formation and release of reactive oxygen species (ROS) [12]. These products activate microglial cells, which release proinflammatory mediators, e.g., tumour necrosis factor alpha (TNF‐α) [13] and interleukins 6 and 1β (IL‐6 and IL‐1β) [14], and induce an increased expression of inducible nitric oxide synthase (iNOS) [15]. Activated astrocytes also express iNOS and produce large amounts of NO [13] that may be toxic to neurons—the inhibition of NO production decreases LID in rodents [16].

It is noteworthy that AMAN has also demonstrated anti‐inflammatory effects in both in vitro and in vivo models of PD [17, 18], with a reduction in levels of TNF‐α, IL‐1β and NO. Thus, the antidyskinetic effects of AMAN may be partially attributed to a reduction of neuroinflammation. Also, the blockade of metabotropic glutamate receptor 5 concomitantly reduced LID and inflammatory marker levels in the brains of parkinsonian monkeys [19].

In the search for new approaches targeting neuroinflammation on LID, the hydroalcoholic extract obtained from the Brazilian red propolis (HERP), rich in flavonoids, has been reported as an antioxidant [20] and anti‐inflammatory compound [21, 22, 23, 24]. It would, therefore, be a potential therapeutic bioactive for LID. Among the biological actions of HERP and its chemical markers (i.e., formononetin, biochanin A, vestitol) [25], the reduction in the levels of TNF‐α, NO, IL‐1β and transforming growth factor beta (TGF‐β) [26, 27]; the decrease in the iNOS expression [28]; and the downregulation of nuclear factor κB transcription in inflammatory conditions [26] can be highlighted.

Regarding the central nervous system, HERP promoted functional recovery upon sciatic nerve injury in rodents, with the reduction of the neuroinflammatory profile in nerve tissue and the increase of the myelinated axon network [29]. Red propolis extract, in the form of an in situ gelling agent, has also been proposed to ameliorate and treat spinal cord injuries [30], as it can hinder the cascade of inflammatory events that lead to neuronal death and glial scar formation. In the present work, we evaluated and characterized the effects of HERP on LID in a PD rat model. To this end, we performed behavioural assessments of AIM and spontaneous motor performance in rodents treated with L‐DOPA concomitantly with the administration of HERP or AMAN. We also analysed the loss of dopaminergic neurons induced by 6‐OHDA and the effects of the treatments on astrocyte response (neuroinflammation) by using immunohistochemistry assay.

2. Methods

2.1. Chemicals, Reagents and Antibodies

If not specified otherwise, all chemicals were purchased from Sigma‐Aldrich (St. Louis, MO, USA).

2.2. Red Propolis Extraction

Red propolis (registration number in the National System of Genetic Heritage Management and of Associated Traditional Knowledge [SisGen] A7086AC, from the Genetic Heritage Management Council) was collected from a local apiary in the Brejo Grande region, Sergipe, Brazil, at coordinates 10°28′25″ S 36°26′12″ W. According to the classification of Brazilian propolis, the sample used in this study belongs to Group 13, whose botanical origin is Dalbergia ecastaphyllum (L.) Taub. To prepare the extract, red propolis samples (1 g) were treated with 70% ethanol (12.5 mL) at room temperature for 1 h in an ultrasound bath (Sonorex Super, Sigma‐Aldrich, St. Louis, MO, USA). After extraction, the product was filtered through a 0.45‐μm polytetrafluorethylene (PTFE) membrane (Millipore, HVHP, MA, USA), followed by the evaporation of the solvent to obtain HERP [29]. The yield of HERP was calculated as a percentage weight of the starting dried material, applying the following equation:

$$\text{yield}\left(\%\right)=\frac{\mathrm{DME}}{\mathrm{IM}}\times 100$$

where DME is the dry mass of the extract (g) and IM is the initial mass of the sample before extraction (g).

2.3. Characterization of HERP

2.3.1. Content in Phenolics and Flavonoids

A Folin–Ciocâlteu reagent was used for the quantification of total phenolic content [31]. Briefly, the HERP sample (250 μg/mL, in triplicate) was mixed with 2.5 mL of aqueous solution containing 10% Folin–Ciocâlteu and 1.75 mL of sodium carbonate at 7.5% (w/v) in an ultrasonic bath (Sonorex Super, Sigma‐Aldrich, St. Louis, MO, USA) at 45°C for 20 min. The absorbance was then recorded using a spectrophotometer (721G visible spectrophotometer, Infitek, Spokane, WA, USA) at 725 nm. Gallic acid was used as a standard to generate the calibration curve. The absorbance data were based on the following calibration equation: y = 0.0105x + 0.0315 (R ² = 0.9978). Total phenolic content was expressed as milligrams of gallic acid equivalents (Eq.) per gram (g) of extract (mg GA Eq./g E). The total flavonoid content was determined by pouring 500 mL of HERP in a test tube, and 0.1 mL of a solution with equal volumes of 10% aluminium chloride (AlCl₃) and 1.0 M of potassium acetate (CH₃CO₂K) was added to 4.3 mL of methanol [31]. Samples were homogenized and left in an ultrasonic bath (Sonorex Super, Sigma‐Aldrich, St. Louis, MO, USA) for 30 min. The absorbance was recorded in a spectrophotometer (721G visible spectrophotometer, Infitek, Spokane, WA, USA) at the wavelength of 425 nm. The calibration curve was built from 600 mg/mL of rutin (RU), based on absorbance, using the following calibration equation: y = 0.0042x − 0.0152 (R ² = 0.9988). The quantity of total flavonoids was expressed as the mass (mg) of RU equivalents (RU Eq.) per gram of extract (mg RU Eq./g E).

2.3.2. High‐Performance Liquid Chromatography–Diode Array Detector (HPLC‐Dad)

A chromatograph equipped with a DGU‐20A3 degasser, an SIL‐20A automatic display, two LC‐20ad pumps, a SPDM20Avp photodiode detector (DAD) and a CBM 20A controller and operated with the LC Solution data station software (Shimadzu, Tokyo, Japan) was used [28, 32]. Chromatographic separation was performed using an analytical C18 column, 250 × 4.6 mm (particle size 5 μL), with an inflow rate of 1 mL/min and an injection volume of 20 μL. Concentrated acetic acid (Solution A) and methanol (Solution B), HPLC grade, at 1% (v/v) were used. Elution started with 40% B (10 min), followed by 45%–50% B (for 10–15 min), 50%–55% B (for 15–20 min), 55%–65% B (for 35–40 min), 65%–75% B (for 40–45 min), 75%–85% B (for 45–50 min) and 85%–40% B (for 50–60 min), returning to initial conditions. The HERP sample (10 mg) was solubilized in 50 mL of methanol, yielding a solution of 0.2 mg/mL. This solution was filtered through a 0.45‐μm PTFE membrane (Millipore, HVHP, MA, USA). The analytical curve was recorded at 280 nm using the chromatographic standards of formononetin, daidzein, liquiritigenin and biochanin A (Sigma Aldrich, St. Louis, MO, USA).

2.3.3. Antioxidant Activity

The 2,2‐diphenyl‐1‐picrylhydrazyl (DPPH) radical scavenging method was used to assess the antioxidant capacity of HERP [20]. A 300‐μL aliquot of HERP was mixed with 2.7 mL of DPPH and incubated for 10 min under dark conditions. The reduction of free radical DPPH was then measured by absorbance reading at 517 nm in a microplate reader (Beckman Coulter Paradigm Multi‐Mode Microplate Reader, Walpole, MA, USA). The DPPH solution alone was used as the control. The antioxidant activity was determined using the following equation:

$$\text{antioxidant actitivity}\left(\%\right)=\frac{{\mathrm{Abs}}_{\text{control}}-{\mathrm{Abs}}_{\text{sample}}}{{\mathrm{Abs}}_{\text{control}}}\times 100$$

where ${\mathrm{Abs}}_{\text{control}}$ is the recorded absorbance of the control (DPPH solution alone) and ${\mathrm{Abs}}_{\text{sample}}$ the recorded absorbance of the sample test. The results were expressed as the IC₅₀, i.e., the amount of antioxidant required to reduce by 50% the initial concentration of radicals. The IC₅₀ values were determined using a nonlinear regression curve using the GraphPad Prism 7.0 software.

2.4. Biological Assay

2.4.1. Ethics Issues

The design of the in vivo study was approved by the Ethics Commission on Animal Use of Tiradentes University (Protocol Number 020518). Animal experiments were performed according to the Ethical Principles in Animal Research adopted by the Brazilian College of Animal Experimentation (CONCEA), implementing the National Institutes of Health (NIH) guidelines for the care and use of laboratory animals. The study was conducted in accordance with the Basic & Clinical Pharmacology & Toxicology policy (BCPT 2023 policy) for experimental and clinical studies [33].

2.4.2. Animals and Experimental Design

Thirty‐six adult male Wistar rats (250–300 g) were split in groups of three rats per cage and housed in a temperature‐controlled room (~22°C), under a 12/12‐h light/dark cycle with free access to food and water. The animals were randomly divided to receive a unilateral injection of saline solution (“saline” group) or 6‐hydroxydopamine (6‐OHDA, injured groups) into the medial forebrain bundle (MFB). After 19 days (time for dopaminergic depletion), the 6‐OHDA–injected animals were subjected to the apomorphine‐induced rotational behaviour test, and only animals with ≥ 3 rotations/min were selected to continue the study. This was performed to ensure that only animals with the same injury intensity (massive, effective) were used in the study and, thus, to avoid bias. Then, rats were randomly divided to receive daily treatments with the vehicle (VC), L‐DOPA, HERP or AMAN (n = 6–8 per group), as follows:

• “Saline” group: received saline instead of 6‐OHDA into the MFB and treated with the VC (0.5 mL of distilled water with 5% Tween 80, po);

• “6‐OHDA/VC” group: lesioned and treated with the VC (0.5 mL of distilled water with 5% Tween 80 po);

• “L‐DOPA/VC” group: lesioned and treated with L‐DOPA (20 mg/kg with benserazide 5 mg/kg po) + VC (0.5 mL of distilled water with 5% Tween 80, po);

• “L‐DOPA/HERP” group: lesioned and treated with L‐DOPA (20 mg/kg with benserazide 5 mg/kg) + HERP (10 mg/kg, in distilled water with 5% Tween 80, po); and

• “L‐DOPA/AMAN” group: lesioned and treated with L‐DOPA (20 mg/kg with benserazide 5 mg/kg) + AMAN (40 mg/kg diluted in saline, ip).

Doses were calculated by weight and diluted for each animal individually. Treatments (0.5 mL) were carried out daily for 21 days, with HERP and AMAN administered 30 min before the L‐DOPA solution. Doses of L‐DOPA and AMAN [34] or HERP [21, 28] were chosen based on previous works. During this period, the animals were evaluated through behavioural tests:

• Abnormal involuntary movements (AIM) as a measure of LID on Day 1 and Day 21 (i.e., at the beginning and the end of the assay);

• Open field (OF) as the assessment of spontaneous motor activity/antiparkinsonian effects of L‐DOPA, on Day 2 and Day 10 (as the treatment evolves with time);

• Rotational behaviour (run before starting the treatments to select lesioned animals only and then on Day 15 after starting the treatments to evaluate the dopaminergic receptors response); and

• Cylinder test (analysis of upper limb asymmetry on Day 22 and motor recovery on the last day).

Various tests were carried out on different days to avoid the habituation of the animals and to analyse multiple aspects related to LID. At the end of the behavioural tests, rats were euthanized by overdose of anaesthetics, i.e., ketamine hydrochloride (300 mg/kg) and xylazine hydrochloride (30 mg/kg), ip; the brains were removed; and the histological analysis was performed by counting the immunoreactive cells for tyrosine hydroxylase (TH+) in the substantia nigra pars compacta (SNpc) and the optical density of TH+ and glial fibrillary acidic protein (GFAP) expression at the striatum (Figure 1).

FIGURE 1.

Experimental design of biological tests. D19–D22 indicate the time of procedures in days in relation to 6‐OHDA injection (captions: AIM, abnormal involuntary movements; OF, open field; RB, rotational behaviour).

All behavioural and histological analyses were performed by two trained researchers, calibrated with the level of intra‐ and inter‐examiner agreement (kappa index > 0.8), both blinded to the treatments.

2.4.3. 6‐OHDA Lesion of Medial Forebrain Bundle (MFB)

The animals were pre‐treated intraperitoneally with desmethylimipramine (inhibitor of the high‐affinity noradrenaline transport system, 25 mg/kg, ip) (Sigma‐Aldrich, St. Louis, MO, USA) for 30 min, prior to 6‐OHDA injection. Animals were then anaesthetized using a mixture of xylazine (100 mg/kg, ip) and ketamine (10 mg/kg, ip), followed by the administration of a single injection of 6‐OHDA (3 μL of saline solution containing 16 μg of 6‐OHDA and 0.05% ascorbic acid) administered in the right MFB: anteroposterior (AP) at −4.4, lateral (LL) at +1.2 from the bregma and dorsoventral (DV) at −8.2 from the skull. The infusion rate was 1 μL/min, and the cannula remained in the spot for 3 min before withdrawal. Rats received veterinary broad‐spectrum antibiotics (0.2 mL/kg, im, Zoetis, São Paulo, Brazil) and were kept warm under a heating lamp for 1 h and then returned to their cage after recovery from surgery [13]. Control (named “saline” group) animals received 3 μL of saline solution containing 0.05% ascorbic acid.

2.5. Behavioural Analysis

All behavioural analyses were performed by two trained and calibrated researchers, blinded to the treatments. Rats were filmed with a video camera at the time of testing, and then the behaviours were quantified visually.

2.5.1. Abnormal Involuntary Movements (AIM)

Rats were monitored for AIM using a rat dyskinesia scale [35]. Three weeks after the injection of 6‐OHDA, animals were submitted to a chronic L‐DOPA administration (20 mg/kg plus benserazide 5 mg/kg, orally) for a period of 21 days. Axial, limb, orolingual and locomotive (increased locomotion with contralateral side bias) AIM subtypes were scored for 1 min every 20 min, for a total of 180 min post L‐DOPA administration [16]. Each of these four subtypes was scored by applying the severity scale ranging from 0 to 4 (where 0 = absent, 1 = occasional dyskinesia displayed for < 50% of observed time, 2 = frequent and sustained displayed for > 50% of observed time, 3 = continuous but interrupted by sensory distractions and 4 = continuous dyskinesia not interrupted by any external stimuli) [36]. The sum of these scores and the intensity of the AIM (amplitude scores during the observation period, from 0 to 4) was reported. Limb, axial and orofacial AIM scores were evaluated as one item (integrated AIM), while locomotive scores were evaluated as a separate parameter.

2.5.2. Open Field (OF)

For the OF test, rats were placed in an arena, which consisted of a quadrilateral white wood cage with walls of 34.5 cm high and area of 4830.25 cm², with the base split into 16 quadrants and open on the top. Locomotor activity was evaluated for 5 min, during which the numbers of crossings (i.e., number of quadrants crossed) and rearing (standing with the forepaws raised in the middle of the arena or against the wall) were recorded [16].

2.5.3. Rotational Behaviour

Rotational behaviour was consecutively measured for 45 min in a circular arena (20‐cm diameter, 12‐cm height), 5 min after a subcutaneous (sc) injection of 0.5‐mg/kg apomorphine (in distilled water). The total number of contralateral rotations was counted. Animals were allowed to accommodate and acclimate to the environment for 15 min before being subjected to the tests [16].

2.5.4. Cylinder Test

To examine the use of spontaneous forelimb, rats were individually placed inside a glass cylinder (16‐cm diameter, 30‐cm height), with mirrors located behind, to allow a 360° observation. Rats were immediately video recorded with an Olympus C‐7070 camera, coupled to an Olympus CX31 microscope (Olympus, Tokyo, Japan), for 5 min without a habituation period. Paw touches were assessed using freeze‐frame analysis of the recordings, and the number of contacts with fully extended digits (touches) performed with the forelimb ipsilateral and/or contralateral to the lesion was counted. Simultaneous paw contacts were excluded from the analysis. Data were expressed as a percentage of contralateral touches, calculated as follows [37]:

$$\frac{\text{contralateral touches}}{\text{ipsilateral touches}+\text{contralateral touches}}\times 100$$

2.6. Tissue Processing

The rats were euthanized by an overdose of anaesthetics, i.e., ketamine hydrochloride (300 mg/kg) and xylazine hydrochloride (30 mg/kg, ip), on the 22nd day after treatments. Brains were immediately removed; fixed in 10% paraformaldehyde for 7 days; and then dehydrated, diaphanized and paraffin‐embedded for subsequent immunohistochemistry procedure. Four serial sections of 5 μm were obtained through the striatum and SNpc, using a microtome (Minux S700, RWD, San Diego, CA, USA).

2.7. Immunohistochemistry

Histological sections were deparaffinized in xylene and dehydrated with increasing concentrations of alcohol (two cycles, 15 min each, 100° and 90° GL). Antigen retrieval was done using 0.1 M of citrate buffer, pH 6.0, in a high‐powered microwave (two cycles, 5 min each, 95°C). A 1% H₂O₂ solution in phosphate buffered saline (PBS, 0.1 M) was used for 7 min to block endogenous peroxidase. Sections were incubated overnight at 4°C, with the primary antibodies (TH, clone TH‐16, 1:500, Batch No. T2928, Sigma‐Aldrich; GFAP, clone S206A‐8, 1:100, Batch No. SAB5201104, Sigma‐Aldrich) in PBS containing 5% bovine albumin serum. Biotinylated secondary antibody (DAKO LSAB 2 kit, CiteAb, Bath, UK) and then streptavidin–peroxidase (DAKO LSAB 2 kit, CiteAb, Bath, UK) were used for detection of the positive reactions (30 min for each step; the sections were completely covered with reagents). The reaction was revealed by the addition of Trizma base buffer (TBS, 1 M, pH 7.4) containing 3,3‐diaminobenzidine (DAB), in 0.3% H₂O₂ solution (20 min). Finally, the slices were counterstained with haematoxylin and mounted on coverslip slides for microscopic observations (Bioptika B60 microscope, Colombo, Paraná, Brazil).

2.8. Imaging Analysis

All the imaging analyses were performed on both sides of the brain. The stained neurons in the SNpc were counted to estimate the number of TH+ cells using a computerized image analysis system, through a 100× objective using the Bioptika B60 professional biological microscope (Colombo, Paraná, Brazil) and digitized with a video camera. Within the striatum (Str), three rostrocaudal levels were checked for TH+ and GFAP expression, with measurements made in the dorsomedial, dorsolateral and ventrolateral quadrants. Integrated relative optical density values (calculated by multiplying the mean grey value and area) of the corresponding sectors in each section per hemisphere (contralateral and ipsilateral to the lesion) were determined as the mean. The images were converted to 8‐bit grayscale, and a mean grey value of the stain was expressed in arbitrary greyscale units where the scale ranges from 0 to 255 (most intense labelling was scored 0 as it represents the darkest) to form one density measurement performed in the sections. Background (lateral ventricle) was subtracted from all subsequent measurements. The counts and measurements were performed with the NIH ImageJ 1.43 software in 4 ± 1 section slides, and then the mean value per rat was calculated. No stereological analysis was performed.

2.9. Statistical Analysis

Values are presented as mean ± SD. The Shapiro–Wilk normality test was performed to assess the Gaussian distribution of the data. In the AIM analysis, parametric statistics was performed because scores of LID represent the extent of time with behavioural response development (continuous data). This was determined using two‐way repeated‐measures ANOVA with Bonferroni's multiple‐comparison post hoc test. Data from other behavioural evaluation and GFAP expression were compared by the Kruskal–Wallis test and then Dunn's post‐test. The means of TH+ labelling different groups were compared by one‐way ANOVA followed by the Tukey post‐test (p < 0.05). The software GraphPad Prism (Dotmatics, Boston, MA, USA) Version 5.0 was used.

3. Results

3.1. Characterization of the HERP

A yield of 30.28% (v/v) was obtained for HERP. The total phenolic content and flavonoids in HERP were 468.7 ± 73.5 mg of GA Eq./g E and 201.1 ± 9.4 mg of R Eq./g E, respectively, and the IC₅₀ was 45.72 ± 7.06 μg/mL.

The yield and antioxidant activity of HERP were similar to those reported previously for red propolis extracts, but its phenolic and flavonoid content was notably higher [20, 28]. These differences can be attributed to the seasonality and geographic origin of the vegetal material, but the fact that flavonoid amounts in the samples can increase the biological activity of the extracts is noteworthy [38].

HPLC analysis (Figure 2) revealed the presence of Northeastern Brazilian red propolis chemical markers [25] quantified per gramme of extract, namely, biochanin A (retention time [RT]: 28.97 min; 7.2 mg/g), formononetin (RT: 24.19 min; 6.3 mg/g) and daidzein (RT: 12.84 min; 3.1 mg/g), as well as the major compound found in the sample, liquiritigenin (RT: 26.84 min; 10 mg/g). Important biological effects of these flavones in the central nervous system include the attenuation of neuroinflammatory response [39, 40, 41], antioxidant‐related neuroprotective action [42], inhibition of microglial activation [43] and the increased production of neurotrophic factors [44, 45]. These compounds may be related to the antidyskinetic actions observed in this study.

FIGURE 2.

Chromatogram from the HPLC analysis of the HERP showing the presence of the main chemical markers.

3.2. Biological Assay

3.2.1. Behavioural Assessment

Rotational behaviour was observed in all animals that received 6‐OHDA injection (total number of rotations: 291.07 ± 55.52) one day before the beginning of the treatments. None of the saline‐microinjected rats presented rotations. This rotational behaviour test is used to evaluate the effectiveness of 6‐OHDA–induced lesion and only occurs in rats with up to 80% dopamine depletion [46].

Once the effectiveness of the dopaminergic lesion procedure is confirmed, the assessment of LID was performed and the results showed that L‐DOPA/HERP– and L‐DOPA/AMAN–treated groups presented a significant decrease in locomotion (p = 0.0049) and integrated (orofacial + limb + axial, p = 0.0006) AIM compared to the L‐DOPA/VC–treated group on Day 21. These effects were not observed on the first day (p > 0.05); therefore, the chronic (but not acute) treatment with HERP and AMAN decreased LID (Figure 3). Rats from saline and 6‐OHDA/VC groups did not show AIM.

FIGURE 3.

L‐DOPA–induced dyskinesias. (A, B) Global/integrated AIM on Day 1 and Day 21, respectively; (C, D) locomotive AIM on Day 1 and Day 21, respectively. The 6‐OHDA–lesioned rats were treated with vehicle (VC), hydroalcoholic extract of red propolis (HERP) or amantadine (AMAN) concomitant to L‐DOPA. The p‐values refer to significant differences from the L‐DOPA/VC group (***p < 0.001; **p < 0.01); two‐way repeated‐measures ANOVA with the Bonferroni post‐test indicates differences between treatments: axial, limb and orofacial: F(2, 19) = 21.14, p < 0.0001; locomotive: F(2, 19) = 6.394, p = 0.0075.

In the open‐field test, the 6‐OHDA induced a decrease in crossing (6‐OHDA: p = 0.0097; L‐DOPA/VC: p = 0.0029; L‐DOPA/HERP: p = 0.0107; L‐DOPA/AMAN: p = 0.0164) and rearing (6‐OHDA: p = 0.0220; L‐DOPA/VC: p = 0.0061; L‐DOPA/HERP: p = 0.0192; L‐DOPA/AMAN: p = 0.0110) compared to saline on Day 2. The 6‐OHDA (crossing and rearing with p < 0.0001 from saline) and L‐DOPA/AMAN (crossing: p = 0.0256, rearing: p = 0.0038 from saline) groups still presented lower values, while L‐DOPA/VC and L‐DOPA/HERP groups showed increased means of crossing (p = 0.0098 and 0.0343, respectively) and rearing (p = 0.0200 and 0.0456, respectively), compared to 6‐OHDA on Day 10 (Kruskal–Wallis test followed by Dunn's post hoc test). Thus, all lesioned rats presented significant impairment in crossing and rearing on Day 2. After 10 days of treatment, the 6‐OHDA/VC– and L‐DOPA/AMAN–treated groups maintained lower locomotor scores, while L‐DOPA/VC– and L‐DOPA/HERP–treated groups showed increased means of crossing and rearing, compared to 6‐OHDA/VC (Figure 4A–D).

FIGURE 4.

Results of the behavioural analysis. (A–D) Results of the open‐field tests showing the mean number of crossings (A, Day 2; and B, Day 10) and rearing (C, Day 2; and D, Day 10). (E, F) Results of rotational behaviour performed before and after (15 days) the beginning of the treatments with vehicle (VC), hydroalcoholic extract of red propolis (HERP) or amantadine (AMAN) concomitant to L‐DOPA, in 6‐hydroxydopamine (6‐OHDA)–lesioned rats and saline group. (G) Results of cylinder test (22 days). Significant difference with ****p < 0.0001, ***p < 0.001, **p < 0.01 and *p < 0.05 (Kruskal–Wallis test followed by Dunn's post hoc test).

In the rotational test performed after 2 weeks of treatments, only animals of the L‐DOPA/AMAN group showed increased rotational behaviour compared to 6‐OHDA/VC (p = 0.0176, Figure 4F) (6‐OHDA/VC: 350.7 ± 30.5; L‐DOPA/VC: 439 ± 64.5; L‐DOPA/HERP: 505.67 ± 54.7; L‐DOPA/AMAN: 602.5 ± 52.7). All 6‐OHDA–injected rats exhibited locomotor asymmetry in the cylinder test (Figure 4G, 6‐OHDA/VC: 16.7 ± 6.9; L‐DOPA/VC: 18.9 ± 9.6; L‐DOPA/HERP: 9.6 ± 6.5; L‐DOPA/AMAN: 12.5 ± 4.6 of correct alterations). A significant difference (p = 0.0059, p < 0.01, p < 0.0001, p = 0.0003 and p = 0.0020, respectively) from saline microinjected group (48.5 ± 4.2) was found, without differences between each other (p > 0.05; Kruskal–Wallis test, followed by Dunn's, p = 0.0005 overall). Values refer to the total number of rotations recorded in the rotational test.

3.2.2. Histological Analysis

An increase in the mean density of GFAP striatal expression was found in 6‐OHDA/VC– and L‐DOPA/VC–treated groups compared to saline/VC. The values decreased in L‐DOPA/HERP– and L‐DOPA/AMAN–treated groups, compared to both 6‐OHDA/VC– and L‐DOPA/VC–treated groups (Figure 5).

FIGURE 5.

Immunohistochemistry for GFAP: representative photomicrographs of the histological analysis of saline (A), 6‐OHDA/VC (B), L‐DOPA/VC (C), L‐DOPA/HERP (D) and L‐DOPA/AMAN (E) groups (400× magnification). (A) The sections of the control group (saline) presented a few astrocytes (brown marked cells). (B, C) In 6‐OHDA/VC and L‐DOPA/VC groups, an increase in the number of marked cells and morphological changes such as hypertrophy (reactive astrocytes) can be observed. (D, E) In treatments with HERP and AMAN, the histological profile is similar to the saline group (a few cells with typical morphology). (F) The quantitative analysis expresses the increased GFAP expression after 6‐OHDA and L‐DOPA counteracted by HERP and AMAN. Significant difference at ****p < 0.0001, ***p < 0.001, **p < 0.01 and *p < 0.05. The 6‐OHDA (p < 0.0001) and L‐DOPA/vehicle (p = 0.0402) groups presented an increase in the mean density of GFAP striatal expression compared to saline. This effect was decreased by L‐DOPA/HERP (p < 0.0001 from 6‐OHDA and p = 0.0062 from L‐DOPA/VC) and L‐DOPA/AMAN (p < 0.0001 from 6‐OHDA/VC and p = 0.0005 from L‐DOPA/VC) (Kruskal–Wallis test followed by Dunn's post hoc).

The quantitative analysis of TH+ neurons revealed a massive loss of these neuronal cells in the ipsilateral SNpc after 6‐OHDA, with the exception of the L‐DOPA/HERP–treated group, which showed remaining cells (qualitative analysis because there were no positive cells in many groups). Moreover, in the contralateral side, L‐DOPA/HERP– and L‐DOPA/AMAN–treated groups presented an increased number of TH+ cells (Table 1 and Figure 6). The analysis of TH+ optical density at the striatum revealed a massive ipsilateral loss in all lesioned rats, without differences between groups regarding the contralateral side, F(4, 40) = 0.02458, p = 0.9988 (Figure 6).

TABLE 1.

Mean number of TH+ neurons in SNpc.

Groups	SNpc ipsilateral	SNpc contralateral
Saline	54.17 ± 4.94	50.54 ± 2.57
6‐OHDA/VC	0.00 ± 0.00	43.25 ± 6.32
L‐DOPA/VC	0.00 ± 0.00	43.60 ± 10.96
L‐DOPA/HERP	9.00 ± 0.76	77.47 ± 10.51*,**
L‐DOPA/AMAN	0.00 ± 0.00	77.57 ± 4.0*,**

* indicates significant difference from the 6‐OHDA/VC group (L‐DOPA/HERP, p = 0.0017; L‐DOPA/AMAN p = 0.0245), and ** indicates significant difference from the L‐DOPA/VC group (L‐DOPA/HERP, p = 0.0008; L‐DOPA/AMAN p = 0.0148), one‐way ANOVA, F(4, 52) = 8.037, p < 0.0001.

FIGURE 6.

Immunohistochemistry for TH+: (A–E) Representative photomicrographs of the histological analysis of saline (A), 6‐OHDA/VC (B), L‐DOPA/VC (C), L‐DOPA/HERP (D) and L‐DOPA/AMAN (E) groups in the ipsilateral (A′, B′, C′, D′, E′) and contralateral sides of the medial forebrain bundle 6‐OHDA injection (100× magnification). (F, F′) Means of optical density at striatum (in percentage from the same side in saline group), from the contralateral and ipsilateral sides of rat brains, respectively (p > 0.05); (G, G′) neuron counts of immunolabelled cells at substantia nigra pars compacta, contralateral and ipsilateral to stereotaxic surgery, respectively (***p < 0.001, **p < 0.01 and *p < 0.05, one‐way ANOVA followed by Tukey's post hoc test). Columns represent the means and bars the ± SD.

4. Discussion

In this work, we observed that subchronic HERP, in a similar way of AMAN (the gold standard strategy to counteract LID), reduced the AIM induced by L‐DOPA in 6‐OHDA–lesioned rats. The treatment with HERP did not modify the OF motor parameters that were improved by L‐DOPA, i.e., its antiparkinsonian effect, and decreased brain neuroinflammation as highlighted by GFAP immunolabelling. In the HERP, liquiritigenin, formononetin, daidzein and biochanin A were identified as the main chemical markers.

Both HERP and AMAN reduced AIM (orofacial, axial, limb and locomotive) after 21 days of treatment. Similar results were found by others, investigating the effects of anti‐inflammatory compounds on LID [47, 48, 49]. As mentioned earlier, pulsatile oscillations in brain dopamine levels with anomalous stimulation of postsynaptic dopamine receptors in striatal neurons and increased glutamate levels are typical conditions underlying LID [1]. This neurotransmitter imbalance in the basal ganglia leads to neuroinflammation, which is considered another critical hallmark of LID [12].

Thus, the mechanism by which HERP reduced LID may be linked to the biological actions of HERP and its main compounds, mainly related to the reduction of proinflammatory cytokine release, such as IL‐1β and TNF‐α [26, 27], and microglial activation [39]. Moreover, formononetin (100 μM) ameliorated dyskinesia and protected dopaminergic neurons through Nrf2 signalling pathway activation in the MPP(+)‐induced Caenorhabditis elegans PD model, reducing ROS accumulation [50].

Another interesting result concerns the fact that the antiparkinsonian effect of L‐DOPA was not significantly counteracted by HERP in the open‐field test; i.e., HERP did not alter L‐DOPA benefits to restore movement. At the beginning of the experiments (Day 2), all lesioned groups presented significant decrease (compared to saline group) in crossing and rearing, as expected [36], while both parameters were increased by L‐DOPA/VC and L‐DOPA/HERP after 21 days. In the L‐DOPA/AMAN group, on the other hand, there was no improvement in motor activity, when compared to the 6‐OHDA/VC group.

A decrease in exploratory/motor activity induced by AMAN was previously reported in doses such as 40 mg/kg, owing to a decrease in dopamine receptors 2 and 3 (D2/3R) binding in the first 15 min post AMAN [51]. In turn, although there is no direct evidence about the role of liquiritigenin, formononetin or biochanin A on dopamine receptors, the dopamine level may be modulated (i.e., increased) by several flavonoids that seem to act as monoamine oxidase competitive inhibitors and/or dopamine transporter (DAT) inhibitors [52]. In any case, this result must be seen carefully, since AMAN effects vary among studies, depending on doses, time after administration and the duration of the test [51].

The fact that HERP did not alter L‐DOPA antiparkinsonian effects indicates that antidyskinetic outcomes are not due to a reduction in general motor activity and also points to a possible advantage for use of HERP. On the other hand, repetitive use of AMAN in rats can lead to a decrease in its effects, both antiparkinsonian and antidyskinetic [53].

In the rotational test performed 2 weeks after the beginning of the treatments, only the L‐DOPA/AMAN–treated group showed increased rotational behaviour compared to the 6‐OHDA/VC–treated group. There is evidence that AMAN is capable of increasing the availability of dopamine derived from L‐DOPA [51, 54]. This event, associated with hypersensitivity of postsynaptic receptors in hemiparkinsonian rats and the provision of the L‐DOPA precursor and the apomorphine agonist, could lead to an increase in the number of contralateral turns to the lesion, which was not observed in the L‐DOPA/HERP group. These high rotational scores can also reflect, at least in part, a more sustained action of AMAN during the testing interval.

The cylinder test assesses the spontaneous forelimb lateralization, making use of the animals' natural exploratory instinct in a new environment. This test was implemented at the end of the treatments (22 days of L‐DOPA/VC, L‐DOPA/HERP or L‐DOPA/AMAN), and all 6‐OHDA–injected rats presented locomotor asymmetry. Although no differences were found from the latter group (lesioned), there was a significant difference from the saline‐treated group. The lower mean of correct alternations in lesioned groups indicates a dopaminergic cell loss of at least 40% [37], which supports the homogeneity of the lesion between 6‐OHDA groups. The absence of an antiparkinsonian effect of L‐DOPA on spontaneous forelimb use can be explained by the time point of the test (at the end of experiments), when the severe AIM may affect the limb and trunk regions [35].

Astrogliosis and a subacute increase in GFAP expression occur after 6‐OHDA lesions as responses to neuronal injury [55]. Moreover, increased GFAP expression was associated with LID in rats, which was blocked by a nitric oxide synthase (NOS) inhibitor [13]. Thus, the increased GFAP expression on 6‐OHDA/VC and L‐DOPA/VC groups was observed, as expected. Since HERP and AMAN decreased both AIM and GFAP expression by Day 21, we hypothesized that the antidyskinetic effects of HERP and AMAN on LID may be mediated by mechanisms involving glial cells.

It is well documented that chronically activated astrocytes are expected to promote the development of LID as active counterparts in neuronal networks [13]. Also, L‐DOPA–treated dyskinetic rats exhibit an enhanced striatal expression of GFAP in reactive astrocytes, which is expected to result in excessive toxic products, e.g., NO by iNOS and neuronal nitric oxide synthase (nNOS) enzymes [13, 16].

The capacity to decrease GFAP expression has not been well described before, neither for HERP nor for AMAN, but data from memantine, a similar non‐competitive NMDA receptor antagonist [56], showed that it reduced GFAP expression in the hippocampus of depressive rats [57] and decreased NO production in cerebral ischaemia/reperfusion in mice [58].

HERP interacts with the NO system in the inflammatory process [6, 7, 22, 23]. Liquiritigenin itself presented anti‐inflammatory activity as a consequence of the proinflammatory cytokine production and inhibition of NF‐kappa B–dependent iNOS [59].

On the other hand, HERP is a complex mixture, and both biochanin A and formononetin may act against microglial inflammatory responses [39, 40, 41, 43, 44, 45]. Since the HERP compounds promoted behavioural, cellular and biochemical effects after oral, intragastric and intraperitoneal administrations [39, 40, 41, 43, 44, 45, 46], it is reasonable to assume that they cross the blood–brain barrier (BBB) and may be responsible, at least in part, for the antidyskinetic effect of HERP. In fact, it was demonstrated that liquiritigenin [60], formononetin [61] and other isoflavones [62] cross the BBB, whose permeability is increased in damaged parkinsonian brains [63].

The results obtained for the quantitative analysis of TH+ neurons are suggestive that HERP and AMAN participate in compensatory mechanisms (such as neuronal plasticity) with increased expression of TH+ in the contralateral side, possibly to increase dopamine production. Corroborating this hypothesis, AMAN was reported as a neurotrophic compound that induced expression of the glial cell line–derived neurotrophic factor (GDNF) in astroglia [64] and restored neurotrophic factor expression and dendritic arborization [65]. HERP showed neuroprotective and regenerative effects on sciatic nerve injury, increasing the number of myelinated axons [29]. Liquiritigenin increased the number of cells that differentiated into neurons in brain‐derived progenitor cell cultures [66] and ameliorated memory and cognitive impairment by cholinergic and BDNF pathways in the mouse hippocampus [45]. This same compound was protective against lipopolysaccharide administration through an anti‐inflammatory way and the BDNF/TrkB signalling pathway [45, 46] and counteracted glutamate‐induced apoptosis in hippocampal neuronal cells [67]. There are other interesting reports with flavonoids and dopamine systems, indicating how they could promote an increase in TH+ expression in contralateral SNpc. For example, baicalin increased mRNA and protein levels of TH+ [68], and puerarin induced differentiation and proliferation of dopamine‐producing cells in PD animal models [46]. However, more studies are needed to effectively measure the levels of neurotrophic factors and also inflammatory mediators after treating LID with HERP.

Results from the histological analysis (TH+) also showed that all 6‐OHDA groups presented approximately 90% of dopaminergic cell loss ipsilateral to lesion, which is consistent with a massive dopaminergic lesion, as expected and related by others using the same doses of 6‐OHDA [13, 69]. A few remaining TH+ cells were found in the SNpc of HERP‐treated animals (about 85% of dopaminergic loss). Although the results indicate that HERP and isoflavones promote regenerative effects on sciatic nerve injury, improve memory tests and increase neurotrophic factors [45, 46], a possible role of HERP in promoting proliferation of neurons must be seen carefully. In the present work, rats performed rotational behaviour induced by apomorphine 19 days after lesion, and it is well known that hypersensitivity occurs only after most of the dopaminergic neurons in the SNC (approximately 90%) have been eliminated [70].

5. Conclusions

Our data show that the high antioxidant HERP, having liquiritigenin as the main chemical marker, presented antidyskinetic effects on LID in rats without the loss of L‐DOPA antiparkinsonian properties. These effects were similar to those of AMAN and may be attributed, at least in part, to astrocyte‐mediated mechanisms. Despite the numerous important biological actions already reported for red propolis, this is the first evidence of the hydroalcoholic extract action in brain disorders. Also, most of the studies focus on formononetin and vestitol as chemical markers in HERP, while we found liquiritigenin as the main compound. Thus, our study offers a new perspective on the therapeutic potential of liquiritigenin found in red propolis, mainly for the management of PD.

Author Contributions

All authors contributed to the study conception and design. Sheilla da Silva Barroso, Lorenna E.S. Lopes, Ana M.G. dos Santos, Reinaldo V.B. Neto, Bruno dos Santos Lima, Adriano A. Souza Araújo and Margarete Z. Gomes contributed to the methodology, validation, formal analysis, investigation, supervision and project administration. Sheilla da Silva Barroso, Lorenna E.S. Lopes, Juliana C. Cardoso, Patricia Severino, Eliana B. Souto and Margarete Z. Gomes contributed to the writing (original draft preparation). Juliana C. Cardoso, Patricia Severino, Eliana B. Souto and Margarete Z. Gomes contributed to the writing (review and editing), validation of scientific results and funding acquisition. All authors have made a substantial contribution to the work and have read and approved the final version of the manuscript.

Ethics Statement

The experiments were performed according to the Ethical Principles in Animal Research adopted by the CONCEA, in accordance with the NIH guidelines for the care and use of laboratory animals. The study was approved by the Ethics Commission on Animal Use of Tiradentes University (Protocol Number 020518) and implemented according to the BCPT 2023 policy for experimental and clinical studies.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The datasets generated during and/or analysed during the current study are available from the corresponding authors upon request.