Dapagliflozin ameliorates Lafora disease phenotype in a zebrafish model

Abstract

Lafora disease (LD) is an ultra-rare and still incurable neurodegenerative condition. Although several therapeutic strategies are being explored, including gene therapy, there are currently no treatments that can alleviate the course of the disease and slow its progression. Recently, gliflozins, a series of SGLT2 transporter inhibitors approved for use in type 2 diabetes mellitus, heart failure and chronic kidney disease, have been proposed as possible repositioning drugs for the treatment of LD. With this in mind, we tested dapagliflozin (50 µM), canagliflozin (2.5 µM) and empagliflozin (200 µM) in our epm2a^−/− zebrafish model, investigating their effects on pathological behaviour. In the case of dapagliflozin, we also investigated the possible mechanisms of action. Overall, the gliflozins reduced or rescued neuronal hyperexcitability and locomotor impairment. Dapagliflozin also reduced spontaneous seizure-like events in epm2a^−/− larvae. At the biochemical and molecular level, dapagliflozin was found to slightly reduce glycogen content, and suppress inflammation and oxidative stress. It also ameliorates autophagic homeostasis and improves lysosomal markers. In conclusion, our preclinical study showed that dapagliflozin was able to ameliorate part of the pathological phenotype of epm2a^-/- zebrafish larvae and could potentially be a suitable drug for repurposing in LD. However, since our model does not present Lafora bodies (LBs), at this early disease stage at least, it would be important to use mouse models in order to ascertain whether it is able to prevent or reduce LB formation.

Graphical Abstract

Highlights

• •Dapagliflozin improves neuronal excitability, locomotor activity, and reduces seizure-like events in the epm2a −/− zebrafish model.
• •Treatment decreases total glycogen content, as well as inflammatory and oxidative stress markers in epm2a −/− larvae.
• •The compound ameliorates autophagy homeostasis and normalizes lysosomal markers, restoring conditions closer to those observed in wild-type larvae.

1. Introduction

Lafora disease (LD) is a genetic neurodegenerative disorder with a prevalence of 4 patients per 1,000,000 individuals worldwide [1]. It is linked to mutations in two genes, EPM2A [2], which encodes laforin, a glucan phosphatase, and EPM2B/NHLRC1 [3], which encodes malin, an E3-ubiquitin ligase that regulates glycogen metabolism. The clinical manifestations of LD include a progressive myoclonic epilepsy, motor impairment and cognitive decline, the latter having onset in adolescence [1]. The disease inevitably leads to early death, within 10 years of the appearance of the first symptoms [4]. To date, there is no effective therapy able to modify the natural course of LD. Pending the development of targeted therapies for testing in patients, repurposing drugs already approved for other clinical conditions may be a faster way forward, to at least try to slow down the disease progression and improve patients’ quality of life.

Recently, gliflozins, which are the sodium glucose co-transporter 2 (SGLT2) inhibitors, have been proposed as possible treatments for LD [5]. These drugs have been approved for the treatment of type 2 diabetes mellitus (DMT2) because they reduce blood sugar levels without insulin intervention, i.e., by blocking the SGLT2 transporters expressed on the proximal tubule of the nephron [6]. Because they do not increase the risk of hypoglycaemia, gliflozins are well tolerated compared with other antidiabetic drugs [7]. The use of gliflozins in LD is supported by literature evidence [5]. First, the relationship known to exist between brain metabolism and seizure occurrence supports the repositioning of metabolic drugs, such as metformin, in epilepsy [8], [9]. In this regard, a clinical trial is ongoing to test gliflozins as a potential new treatment option for epilepsy (https://clinicaltrials.gov/). By inhibiting SGLT2 transporters, which are also present in the brain [10], gliflozins may reduce glucose entry into neurons and astrocytes, and in parallel sodium load, and thus reduce glycogen storage and dampen neuronal excitability. Gliflozins may have a metabolic potential similar to that of the ketogenic diet, which is known to lessen seizure frequency and severity in childhood refractory epilepsy [11]. Furthermore, some studies showed neuroprotective effects, obtained by modulating different biochemical pathways including mTOR, cytokines and mitochondria, of these drugs in animal models of Alzheimer’s disease and stroke [10], [12]. Finally, dapagliflozin (DAPA) and empagliflozin (EMPA) have been shown to prevent glycogen accumulation in the kidney in a mouse model of glycogen storage disease type 1b [13], to reduce glycogen accumulation in an animal model of glycogen storage disease type XI [14], attenuate renal proximal tubulopathy in patients [15] and reduce neutrophil dysfunction-related symptoms in affected patients [16] and to improve autophagy in several disease models [17], [18].

Leveraging on these findings, we decided to test multiple gliflozins, namely DAPA, canagliflozin (CANA) and EMPA, in a zebrafish model of laforin deficiency that we recently characterized [19]. After screening their effects on the behavioural phenotype of our LD model, we investigated the possible mechanisms of action of DAPA in epm2a-ko (epm2a^−/−) larvae. This gliflozin was chosen because, among gliflozins approved by the Food and Drug Administration (FDA) and European Medicines Agency (EMA) for paediatric use, it seemed to have the best outcome on the mutant behavioural phenotype, and because it is already known to reduce glycogen accumulation in other glycogen storage diseases [13], [14].

2. Methods

2.1. Zebrafish care and maintenance

Adult wild-type (WT) AB and mutant epm2a^−/− strain fish were maintained at the Neurobiology and Molecular Medicine facility at the IRCCS Stella Maris Foundation (Pisa, Italy) according to standard procedures [20]. Zebrafish eggs were obtained from the natural spawning of 8-month-old adults. Once collected, fertilized eggs were incubated at 28°C in petri dishes (Ø 10 cm) filled with 50 mL of egg water (60 mg of “Instant Ocean®” sea salt added to 1 L of distilled water), with an initial stock density of 100 eggs per dish. Sea salt used for the preparation of the egg water was purchased from Spectrum Brands (Blacksburg, VA). Handling of zebrafish complied with the guidelines of the Animal Care and Use Committee of the Stella Maris Foundation. Experiments were performed under the supervision of this same committee and in accordance with European Directive No. 63 of 22/09/2010 on the protection of animals used for scientific purposes. Every effort was made to minimise both animal suffering and the number of animals needed to collect reliable scientific data.

2.2. Pharmacological treatments

Fifty mutant and fifty WT 4hpf embryos were randomly placed in 60 mm × 15 mm Petri dishes containing different doses of gliflozins diluted in egg water. Gliflozin stock solutions were prepared in DMSO (Cayman Chemical, Ann Arbor, MI) and diluted in egg water to obtain the final administered concentrations. Drug doses were chosen based on preliminary results. The concentration chosen as the working dilution was 50 µM for DAPA (Cayman, CAS-No. 461432–26–8), 2.5 µM for CANA (Cayman, CAS-No. 842133–18–0) and 200 µM for EMPA (Cayman, CAS-No. 864070–44–0). Experimental solutions were freshly prepared in standard fish medium, which consisted of osmotic water with 60 mg per litre of “Instant Ocean®” sea salt (Spectrum Brands, Blacksburg, VA), and were refreshed daily up to 5 dpf.

2.3. Analysis of larval morphology

Live zebrafish were mounted on glass depression slides with 1 % low-melting agarose. Images were obtained using a Leica M205FA stereomicroscope (Leica Microsystems, Wetzlar, Germany). Zebrafish body length was measured using Danioscope software (Noldus Information Technology, Wageningen, The Netherlands) [21].

2.4. Locomotor behaviour

The behavioural endpoints considered were: i) general locomotor activity, measured as the total distance moved (mm) and the velocity (mm/s) of movement over the whole area of the well during each minute of the test procedure, and ii) burst activity during 30 s of the test procedure, expressed as a %. Locomotor behaviour (distance travelled and velocity) was measured in homozygous epm2a^−/− F2 larvae and WT larvae at 5 dpf using the DanioVision device (Noldus Information Technology). Briefly, single larvae were transferred into 96-well plates containing 300 μL of E3 medium per well. Then, one plate at a time was placed in the DanioVision system, and larval locomotor activity was recorded for 30 min and analysed using EthoVision XT software (Noldus Information Technology). Coiling behaviour (burst activity) was measured in homozygous epm2a^−/− and WT embryos at 30 hpf; specifically, the number of tail flicks was measured in 30-s time frames using Danioscope software (Noldus Information Technology).

2.5. Local field potential recordings and analysis

Electrophysiological forebrain recordings were performed in 5dpf zebrafish larvae to characterise the “epileptic” phenotype. The larvae were placed on a drop of 1.2 % low-melting agarose and local field potentials (LFPs) were recorded with an AgCl electrode inside a glass micropipette backloaded with 2 M NaCl. Electrophysiological signals were amplified 500-fold (EXT-02F, NPI, Tamm, Germany), band pass filtered (0.3–1300 Hz), and digitised at a rate of 5 kHz (Axon Digidata 1550B, Clampex 10.7.0.3, Molecular Devices, Berkeley, CA). The microelectrode was positioned by advancing its tip until it punctured the skin, after which it was carefully advanced into the forebrain [22]. Data analyses of the LFP recordings were performed using a suite of custom MATLAB scripts. Burst activity analysis was performed on the LFP signals filtered in the 30–95 Hz band as already described [23]. The root mean square (RMS) power of the signal was computed on the recording on rolling windows of 250 ms and in steps of 50 ms. The distribution of the logarithm of the RMS power measurements was fitted with a single Gaussian distribution. In some recordings, the distribution was asymmetrical due to a high-energy population. Asymmetrical distributions were fitted with double Gaussians, and each ROC curve and area under the curve were calculated; values > 0.9 were considered acceptable and events were extracted as previously described [19]. For normally distributed recordings, bursts were identified by extracting all events with a power higher than three standard deviations from the mean. Violin plots were produced using the online-available function ‘violin.m’ [24]. Statistical analysis was performed using GraphPad Prism version 9 software (GraphPad Software, Inc., San Diego, CA).

2.6. Reactive oxygen species (ROS) analysis

ROS analysis was conducted following the protocol outlined by Naef et al. [25]. ROS levels were assessed using an in vivo carboxy-H2DCFDA fluorescent probe (Abcam, Cambridge, MA) at 30 hpf according to Schindelin et al. [26]. To this end, a lateral image of each larva was captured using a fluorescence microscope, and the fluorescence intensity in the selected region of interest (ROI) was quantified using ImageJ 64 software (https://imagej.nih.gov/ij/).

2.7. Glycogen assay

Glycogen was quantified in zebrafish larvae (10 per experiment), at 5 dpf, using the Glycogen Assay Kit II (Abcam, Cambridge, UK) according to the manufacturer’s instructions and as previously described [27].

2.8. Live staining imaging

LysoTracker Green DND-26 (L7528, Thermo Fisher Scientific; Waltham, MA, USA,1:1000) was used to stain lysosomes and other acidic organelles in live zebrafish larvae. Zebrafish larvae at 5 dpf were incubated with the dye for 30 min in the dark. Following this staining step, larvae were rinsed three times with fresh egg water. All images were acquired using a Leica M205FA stereomicroscope (Leica Microsystems), and fluorescence analysis was performed using Image-J software v.1.46, calculating the total-body fluorescence intensity. The ROIs used for the analysis were selected using LysoTracker and they were located in the brain and in the tail.

2.9. RNA isolation and quantitative reverse transcription polymerase chain reaction (qRT-PCR)

Total RNA was extracted from 30 larvae per group at 5 dpf using the Quick RNA Miniprep Kit (ZymoResearch, Irvine, CA) according to the manufacturer’s instructions. cDNA synthesis and qRT-PCR analysis were performed as described elsewhere [26]. Relative mRNA expression was quantified using the Mic Real-Time PCR System (Bio Molecular Systems, Upper Coomera, Australia) and the comparative ΔCt method. The results obtained in at least three independent experiments were normalised to the expression of the housekeeping gene, β-actin (ENSDARG00000037746). The expression of each gene transcript in mutant larvae was calculated setting the mean of the controls at one, and the p-value was calculated using GraphPad Prism 9 software (GraphPad Software, Inc.). Supplementary Table S1 lists the sequences of the primers used.

2.10. Western blotting

Larvae collected at 5 dpf (n = 60 per sample, three samples per group) were lysed in triplicate in RIPA buffer (Cell Signaling Technology Inc., Danvers, MA) and Western blotting was performed as described elsewhere [28], [29]. ImageJ software (https://imagej.nih.gov/ij/) was used for densitometry analysis. The following antibodies were used: anti-LC3 (NB100–2220, Novus Biologicals, Centennial, CO; 1:1000), anti-LAMP1 (ab24170), Abcam, 1:500), and anti-GAPDH (GTX100118, GeneTex, Irvine, CA; 1:5000).

2.11. Immunohistochemistry staining of whole-mount zebrafish larvae

To prevent the development of pigmentation, embryos were treated with 0.005 % phenylthiourea from 24 hpf. Whole-mount immunohistochemistry was performed in 5dpf larvae fixed in 4 % paraformaldehyde overnight and stored in methanol as described in Kani et al. [30]. The primary and secondary antibodies used were rabbit anti-LAMP1 (ab24170, Abcam, 1:500) and peroxidase-conjugated anti-rabbit (Cell Signalling Technology Inc., Danvers, MA), respectively. Images were acquired using a confocal microscope (Zeiss LSM 700, Jena, Germany). A ROI in the brain containing the telencephalon, the optic tectum and the anterior part of the hindbrain was selected. All images were acquired using a 40X objective (zoom 2). The stacks were composed of 30 slices of 1 micron each. LAMP-1 staining was automatically quantified with Fiji (ImageJ).

2.12. Statistics

All data in the manuscript represent three or more independent experiments. Statistical analysis was performed using GraphPad Prism 9 (GraphPad Software, Inc., La Jolla, CA). All quantitative variables were analysed using either parametric or non-parametric methods, depending on the distribution shown by the Shapiro-Wilk test. For comparisons between two different groups, the analysis was performed using the t-test for normally distributed data; instead, the Mann-Whitney test was applied in the case of non-normally distributed data. For multiple comparisons, Dunn’s test was performed after the Kruskal-Wallis test, since the data of the various groups examined did not show Gaussian distribution. Statistical significance is reported as follows: * p ≤ 0.05, * * p ≤ 0.01, * ** p ≤ 0.001, or * ** * p ≤ 0.0001. The figure legends report the specific test applied for each analysis.

3. Results

3.1. Beneficial effects of gliflozins on behavioural phenotype of epm2a^−/− mutants

The pathological phenotype of LD zebrafish larvae includes neuronal hyperexcitability and impaired locomotor activity [19]. Here, we screened different concentrations of gliflozins to assess their effect on the behavioural phenotype of mutant embryos and larvae (Supplementary material Figure S1). The best concentrations were found to be 50 µM for DAPA, 2.5 µM for CANA and 200 µM for EMPA. We treated zebrafish embryos and larvae from 4 hpf to 5 dpf, refreshing the drugs daily (Fig. 1A). As shown in Fig. 1B, DAPA 50 µM and CANA 2.5 µM rescued the hyperexcitability of epm2a^−/− embryos as assessed using the tail coiling assay at 30 hpf, whereas EMPA 200 µM reduced the difference compared to wt, but did not reduce neuronal hyperexcitability significantly compared to the untreated epm2a group. Conversely, all the drugs showed beneficial effects on locomotor impairment; indeed, they rescued the locomotor deficits of epm2a^−/− larvae, restoring distance moved and velocity travelled to control levels (Fig. 1C).

Fig. 1.

(A) Schedule of treatment with gliflozins. (B) Analysis of the effects of gliflozins on coiling behaviour of epm2a^−/− and WT embryos at 30 hpf. The experiments were performed in triplicate. The number of larvae in each group ranges from a minimum of 40 to a maximum of 300. The graphs represent the median and standard error of the mean (SEM). (C) Analysis of the effects of gliflozins on locomotion (distance travelled and velocity) in epm2a^−/− and WT larvae at 5 dpf. The experiments were performed in triplicate. The number of larvae in each group ranges from a minimum of 40 to a maximum of 300. The graphs represent the median and SEM. Statistical analysis was performed using the Kruskal-Wallis test. Dunn’s test was used to perform post-hoc analysis for multiple comparisons after the Kruskal-Wallis test. (**** p ≤ 0.0001; ***p ≤ 0.001; **p ≤ 0.01; *p ≤ 0.05). Non-significant comparisons are not reported.

3.2. DAPA reduced spontaneous seizure-like events in epm2a^−/− larvae

Using LFP recordings, we previously observed that epm2a^−/− larvae showed an epileptic phenotype characterised by a higher frequency of high-power electrophysiological events compared with WT siblings at 5 dpf [19]. In particular, the duration and power of these electrophysiological events were significantly increased in epm2a^−/− larvae versus controls, this likely accounting for the spontaneous seizures observed in the former [19]. As previously highlighted, both experimental groups exhibited frequent events characterized by low duration and low power, likely associated with normal brain activity. However, only the mutants showed events associated with multiple high-power, long-duration bursts (>1 s), which are identified as seizure-like events [19]. The presence of these high-power spontaneous events is an important pathological sign of spontaneous epileptic seizures, given that the larvae were recorded under normal conditions and were not exposed to pro-epileptic stimuli. Here, we showed that treatment with DAPA (50 µM) was able to significantly reduce the duration (Fig. 2A) and power (Fig. 2B) of seizure-like events in epm2a^−/− larvae at 5 dpf, attenuating their “epileptic” phenotype and normalizing the events compared to WT controls. Supplementary Table S2 shows the descriptive statistical parameters related to LFP recordings across the 3 groups: WT, epm2a^−/−, and DAPA-treated epm2a^−/− larvae.

Fig. 2.

LFP recordings in WT (n = 14), epm2a^−/− (n = 14) and DAPA-treated epm2a^−/− (n = 14) animals. Violin plots of the duration (A) and power (B) of the seizure-like events detected in all epm2a^−/− (n = 14) and DAPA-treated epm2a^−/− (n = 14) fish. The red line represents the mean, and the black lines represent SEM. Statistical analysis was performed using the Kruskal-Wallis test. Dunn’s test was used to perform post-hoc analysis for multiple comparisons after the Kruskal-Wallis test. (****p ≤ 0.0001; ***p ≤ 0.001; **p ≤ 0.01; *p ≤ 0.05).

3.3. Dapagliflozin slightly reduced glycogen content and attenuated the inflammatory and oxidative stress markers

We previously found epm2a^−/− larvae to exhibit a smaller body length at 4 dpf and increased glycogen content at 5 dpf compared with controls [19]. Here, we investigated the possible effects of DAPA on body length and glycogen storage. DAPA treatment was not found to affect the body size of our LD zebrafish larvae (Fig. 3A). On the contrary, it slightly reduced glycogen content in these epm2a^−/− larvae (Fig. 3B). This latter result is in line with a previous study in which DAPA reduced glycogen accumulation in an animal model of glycogen storage disease type XI [14] and prevented kidney glycogen accumulation in a mouse model of glycogen storage disease type 1b [13].

Fig. 3.

(A) Lateral view photographs of zebrafish larvae (epm2a^−/−, n = 15, and DAPA-treated epm2a^−/−, n = 13). Body length was compared between groups using Student’s t-test. Three independent experiments were performed for each experimental group (epm2a^−/− and DAPA-treated epm2a^−/− larvae). (B) Glycogen concentration, expressed in µg/mL, measured in epm2a^−/− and DAPA-treated epm2a^−/− larvae at 5 dpf. The values are expressed as mean ± standard deviation. Three independent experiments were performed for each experimental group (epm2a^−/− and DAPA-treated epm2a^−/− larvae). Statistical analysis was performed using Student’s t-test. (C) qRT-PCR analysis of inflammatory genes. Three independent experiments were performed for each experimental group (WT controls, epm2a^−/− and DAPA-treated epm2a^−/− larvae). Statistical analysis was performed using the ANOVA test. The values are expressed as mean ± standard error of the mean (SEM). (D) Representative fluorescence images of reactive oxygen species (ROS) production in controls (n = 21), epm2a^−/− (n = 24) and DAPA-treated epm2a^−/− (n = 31) larvae. Three independent experiments were performed for each experimental group (WT controls, epm2a^−/− and DAPA-treated epm2a^−/− larvae). Quantitative analysis of ROS production showed a significant increase in epm2a^−/− larvae compared with controls and with DAPA-treated epm2a^−/− larvae. Data are represented as individual values (lines as median ± SEM). Statistical analysis was performed using the Kruskal-Wallis test. Dunn’s test was used to perform post-hoc analysis for multiple comparisons after the Kruskal-Wallis test. (**** p ≤ 0.0001; ***p ≤ 0.001; **p ≤ 0.01; *p ≤ 0.05).

Our epm2a^-/- larvae showed upregulation of key inflammatory genes such as tnfα (tumor necrosis factor alpha) and il-1β (interleukin-1 beta), in line with what we previously described [19], as well as other genes such as il-6 (interleukin-6) and nlrp3 (NLR family pyrin domain containing 3). The latter is involved in the inflammasome pathway. The presence of inflammation has been found in several LD models [19], [31], [32], [33]. Here, we showed that DAPA was able to rescue inflammation by restoring the expression of inflammatory genes to the levels of WT animals (Fig. 3C). We then assessed ROS production in WT, epm2a^−/− and DAPA-treated epm2a^−/− specimens, finding an increase in oxidation stress in epm2a^−/− versus WT larvae at 24 hpf. The probe we used provides a general readout of ROS levels in cells rather than pinpointing the exact ROS species involved [34]. DAPA treatment reduced oxidative stress by restoring ROS in epm2a^−/−-treated larvae to control levels (Fig. 3D).

3.4. Dapagliflozin ameliorate autophagic markers in epm2a^−/− larvae

As in many neurodegenerative disorders, autophagy is impaired in LD, and this includes our epm2a^−/− zebrafish model [19]. Here, we explored the potential effects of DAPA on autophagic gene expression (by means of qRT-PCR analysis) and on protein levels (by Western blotting). We found that DAPA was able to reduce or rescue the upregulation of most autophagic genes related to autophagosome production [35], including tfeb, atg5, atg12 and lc3, and to rescue the levels of mtor (an autophagy inhibitor) and p62 (a ubiquitinated protein that provides an index of autophagic degradation) [35] (Fig. 4A). Furthermore, the LC3B-II/LC3I ratio, which was increased in mutants, was normalised (returned to control levels) by DAPA treatment (Fig. 4B).

Fig. 4.

(A) qRT-PCR analysis of the autophagy-lysosomal pathway. Three independent experiments were performed per experimental group (WT controls, and epm2a^−/− and DAPA-treated epm2a^−/− larvae). Statistical analysis was performed using the ANOVA test. The values are expressed as mean ± standard error of the mean (SEM). (B) Three independent larval homogenates, from controls (n = 50), epm2a^−/− larvae (n = 50) and DAPA-treated epm2a^−/− larvae (n = 50), were tested by Western blotting for the expression of LC3 protein. The protein levels were normalized to GAPDH. Statistical analysis was performed using the ANOVA test. The values are expressed as mean ± SEM. (**** p ≤ 0.0001; ***p ≤ 0.001; **p ≤ 0.01; *p ≤ 0.05).

3.5. Dapagliflozin enhanced lysosomal markers

We then explored lysosomal function to investigate potential alterations at this level that could affect autophagy. To determine lysosome function in epm2a^-/- larvae, we used a LysoTracker probe that stains acid organelles. As shown in Fig. 5A, the LysoTracker fluorescence was reduced in epm2a^−/− larvae compared to controls at 5 dpf, indicating a decrease in the acidic environment or number of lysosomes. Treatment with DAPA significantly increased LysoTracker fluorescence in both brain and muscles, restoring it to control levels (Fig. 5A). Furthermore, we investigate LAMP1 protein, observing reduced levels in mutants compared to controls, a feature indicating impaired lysosome number [36].LAMP1 protein levels were restored by DAPA treatment (Fig. 5B). Immunostaining confirmed lower levels of LAMP1 in epm2a^−/− versus control larval brains at 5 dpf, and DAPA treatment increased these levels by approximately 40 % (Fig. 5C). Similar results were shown by Yu et al. demonstrating that DAPA accelerates autophagosome clearance by reducing lysosomal destruction in cardiomyocyte primary culture [37].

Fig. 5.

(A) In vivo LysoTracker staining in 5dpf WT, epm2a^−/− and DAPA-treated epm2a^−/− larvae with statistical analysis. Three independent experiments were performed in each group. The values are expressed as median ± standard error of the mean (SEM). Statistical analysis was performed using the Kruskal-Wallis test. Dunn’s test was used to perform post-hoc analysis for multiple comparisons after the Kruskal-Wallis test. Scale bar = 100 μm. (B) Three independent larval homogenates from controls (n = 50), epm2a^−/− (n = 50) and DAPA-treated epm2a^−/− larvae (n = 50) were tested by Western blotting for the expression of LAMP1. The protein levels were normalized to ß-actin or GAPDH. The values are expressed as mean ± SEM. Statistical analysis was performed using the ANOVA test. (C) LAMP1 whole-mount immunostaining in WT (n = 28), epm2a^−/− (n = 35) and DAPA-treated epm2a^−/− (n = 23) larvae at 5 dpf. Images were acquired with a confocal microscope with a magnification of 20x/zoom 2x. Scale bar 5 µm. On the right: quantification of LAMP1 puncta density ratio. The graph represents the median and SEM. Statistical analysis was performed using the Kruskal-Wallis test. Dunn’s test was used to perform post-hoc analysis for multiple comparisons after the Kruskal-Wallis test. (**** p ≤ 0.0001; ***p ≤ 0.001; **p ≤ 0.01; *p ≤ 0.05).

4. Discussion

LD is a rare and still incurable neurodegenerative disorder. As with other rare diseases, the search for new drugs to treat this condition is a costly and time-consuming process. Major pharmaceutical companies show little interest in developing orphan drugs, because the high investment costs are not justified by the limited market return. In this scenario, drug repurposing is a good strategy to reduce the costs and time involved in developing new treatments [38]. Recently, gliflozins, SGLT2 transporter inhibitors approved for DM2 treatment, have been proposed as possible repositioning drugs for the treatment of LD [5]. Here, we demonstrate for the first time that gliflozins improved the pathological phenotype in a zebrafish model of LD. DAPA, EMPA and CANA were indeed able to reduce the motor impairment in epm2a^-/- larvae, and DAPA and CANA also rescued neuronal hyperexcitability in epm2a^-/- embryos. It should be emphasised that we have explored neuronal hyperexcitability with behavioural tests in zebrafish embryos at 24 hours post-fertilisation, so we cannot exclude the possibility that over time administration of EMPA may also be able to reduce hyperexcitability. Based on behavioural tests, the best-performing gliflozin approved for paediatric age was DAPA, that also mitigated the “epileptic” phenotype, reducing the spontaneous seizure-like events observed in our LD model.

The main rationale for using gliflozins in LD is that they can prevent glucose from entering neurons and astrocytes by inhibiting SGLT1/2 transporters. This, in turn, could reduce glucose storage, decreasing polyglucosan accumulation (Lafora body formation) and slowing the disease progression. Here, we showed that DAPA, a gliflozin approved by the FDA and EMA for paediatric use, is able to slightly reduce glycogen content in epm2a^−/− larvae, confirming what has been observed in other models of glycogen storage disease [13], [14]. We did not differentiate the localization of glycogen between muscle and brain, as we tested glycogen content in whole larvae as previously described [19], [27]. However, the action of DAPA at the brain level is well documented [39], as well as its effectiveness in various models of neurodegenerative diseases of the central nervous system [10], [12]. Additionally, we did not test the effect of DAPA on glycogen content in WT controls. However, this molecule has already been tested in various animal models of glycogen storage diseases, demonstrating its specificity in reducing glycogen accumulations [13], [14]. However, since our LD model does not present Lafora bodies (LBs), at this early disease stage at least, we have no way of ascertaining whether DAPA reduces their formation and accumulation.

Inhibition of SGLT1/2, and therefore of inward currents of sodium across the cell membrane, could reduce sodium load in the brain, as well as the uptake of glucose that causes membrane depolarisation, thereby reducing neuronal hyperexcitability [40]. In line with this suggestion, we observed a reduction in neuronal hyperexcitability when administering gliflozins, and in the case of DAPA, also a mitigation of the seizure-like events in mutants.

Although the main effect of DAPA, and the one for which it is used in DM2, is blood glucose concentration reduction, several studies have provided new insights into its mechanisms of action. In different models of neurodegenerative disorders [41], myocardial injury [37] and diabetic kidney diseases [42], as well as in patients with DM2, DAPA has shown the ability to alleviate inflammation and oxidative stress [43], [44], both dependently on and independently of blood glucose levels [42]. In our model of LD, too, the treatment with DAPA rescued inflammation and oxidative stress. DAPA reduced il-1β, il-6 and tnf-α expression, probably altering macrophage/microglia polarisation and inhibiting production of inflammatory cytokines [45]. In addition, SGLT-2 inhibitors may reduce the production of NLRP3/ASC inflammasome, thereby suppressing inflammatory responses [46]. Indeed, in our model, too, we observed a reduction in nlrp3 levels.

Dapagliflozin also seemed to target autophagy [17], [18], [37], a pathway involved in many neurodegenerative disorders, including LD [19], [47], [48], [49]. Studies on different LD models have produced conflicting results on the nature of the autophagic impairment [47], [48], [49], [50], [51], [52], [53], [54], [55], that remain poorly understood. While some studies hypothesised a deficit in the synthesis of autophagosomes [53], subsequent findings support an impairment of the later stage of the proteolytic process [48]. In our model, we observed an upregulation of mTOR, consistent with a previous study on a LD mouse model [53] suggesting that laforin deficiency blocks autophagy by increasing the mTOR pathway. However, some of our results suggest a more complex regulation of autophagy in our LD model. For instance, we observed both an increase in mTOR, known for its inhibitory effect on autophagy, and an increase in Beclin-1, known to promote autophagy and autophagosome formation. It is likely that laforin deficiency leads to the upregulation of mTOR, which inhibits autophagy, and that the increased expression of Beclin-1 in our model represents an attempt to restore autophagy, likely through mTOR-independent pathways [56], [57]. Therefore, this dual dysregulation ultimately disrupts the clearance of autophagic substrates, contributing to cellular dysfunction. Upregulation of beclin-1, atg5, atg12 and lc3, together with increased levels of LC3B-II/LC3B-I and ATG5 (see also [19]), could indicate an increase in autophagosome formation to compensate the pathological conditions, as in similar neurodegenerative conditions like Batten disease [58]. As an alternative hypothesis, the increased in LC3-II/LC3-I ratio might indicate a block in fusion or degradation [59], followed by a compensatory increase upstream. Interestingly, we observe a reduction of LAMP1 and lysotracker fluorescence in epm2a^-/- mutants that may suggest a potential lysosome deficiency and so a reduction in autophagosome degradation [60]. Therefore, taken together, these data could indicate a possible impairment in the clearance of autophagosomes in the LD zebrafish model [60]. On the other hand, we also observed an upregulation of p62, a ubiquitinated protein that provides an index of autophagic flux. Unfortunately, we can’t test its protein levels for the lack of antibodies working against zebrafish-p62. Our p62 finding, together with previous studies on LD mouse models demonstrating constantly an increased in the protein levels of P62 [19], [48], [53], [61], [62], may lead to hypothesize a reduction in autophagosome degradation [61].

Defining the precise autophagic alterations in our LD model was not the primary focus of this study; however, we acknowledge its relevance. Our investigation into autophagy is not robust enough to draw definitive conclusions. This is mainly due to the limitations in testing the protein levels of several key autophagic players, as well as the absence of experiments using autophagosome-lysosome fusion blockers to assess the functional consequences on autophagy. Based on our findings, we can state that the epm2a-/- larvae exhibit alterations in autophagic homeostasis, as evidenced by changes in the gene or protein levels of key autophagy molecules, along with a reduction in lysosomal markers. Moreover, we observed that DAPA was able to improve autophagic homeostasis and normalize lysosomal markers in our LD model, restoring conditions closer to those observed in wild-type larvae. Regarding the increase in LC3-I protein and decrease in LC3-II in DAPA-treated epm2a^−/− larvae, this observation may indicate an enhancement of autophagosome formation. DAPA appears to improve the defective autophagy process, leading to increased turnover of LC3-II during autophagosome maturation and degradation [62]. The accumulation of LC3-I could result from the ongoing replenishment of non-lipidated LC3 needed to sustain autophagosome formation [63]. On the other hand, the reduction in LC3 mRNA levels might represent a feedback mechanism, linked to the restoration of autophagic flux by DAPA bringing the levels of LC3 transcription closer to those of WT larvae. It worth to underline that the observed reduction in atg5/atg12 expression upon DAPA treatment may reflect their active utilization during enhanced autophagic activity, as indicated by the upregulation of Beclin 1, suggesting a dynamic regulation of autophagy components. These results suggest that DAPA plays a beneficial role in normalizing autophagy dysregulation in epm2a^−/− larvae.

Our study provides some preclinical evidence in vivo for repurposing DAPA in LD. We found that DAPA: i) improves neuronal excitability, locomotor activity, and seizures in the epm2a-/- zebrafish model; ii) slightly reduces total larvae glycogen content, and inflammatory and oxidative stress markers; and iii) improves autophagic homeostasis and normalizes lysosomal markers. However, it is important to highlight the main limitations of this study. Regarding glycogen metabolism, we did not differentiate between the localization of glycogen in muscle and brain, and our model does not exhibit LB in the early stages of the disease. Therefore, further testing in mouse models would be necessary to determine whether DAPA can prevent or reduce LB formation, in line with what has been observed in other glycogen storage diseases. Another limitation in the study of autophagy is the inability to verify protein levels of several key players due to the lack of specific antibodies compatible with zebrafish. Furthermore, we did not use an agent to block autophagosome-lysosome fusion, which could have strengthened the reliability of our findings. For these reasons, further studies are needed to investigate the therapeutic potential and mechanisms of action of DAPA in LD.

Funding

This research project is funded by the Italian Ministry of Health, Ricerca Corrente 2024 and 5 × 1000 to FMS. S.D.V. is the holder of Telethon grant GSA22B005 and M.M. is the holder of Telethon grants GSA23C003 and GGP20011. S.D.V. holds a PhD fellowship in neuroscience, University of Florence, Italy; M.M. is the holder of the Telethon Career Award.

CRediT authorship contribution statement

Paola Imbrici: Visualization, Validation, Methodology. Stefania Della Vecchia: Writing – review & editing, Writing – original draft, Methodology, Funding acquisition, Data curation, Conceptualization. Rosario Licitra: Visualization, Validation, Methodology, Investigation, Formal analysis. Devid Damiani: Writing – review & editing, Visualization, Validation, Methodology, Investigation, Formal analysis. Valentina Naef: Visualization, Validation, Methodology, Investigation, Formal analysis. Antonella Liantonio: Visualization, Validation, Methodology. Maria Marchese: Writing – review & editing, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation. Sara Bernardi: Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation. Filippo Maria Santorelli: Writing – review & editing, Visualization, Validation, Supervision, Funding acquisition, Conceptualization.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary material

Supplementary material

Data availability

Data will be made available on request.