Mitochondrial Metabolic Reprogramming of Cortical Neurons by Prenatal Exposure to Corticosterone: A Shift from ATP Synthesis to Membrane Potential Maintenance

Abstract

Mitochondrial bioenergetics plays a fundamental role in neuronal development and function. Prenatal exposure to corticosterone in rats (Corti.Pup) has previously been shown to cause delayed neurodevelopment and synaptic plasticity deficits, showing attention deficit hyperactivity disorder (ADHD) - like behaviors. However, the underlying mitochondrial metabolic adaptations remain unclear. This study investigated mitochondrial function and metabolic remodeling in prefrontal cortex neurons of Corti.Pups, focusing on oxidative phosphorylation, calcium handling, and redox balance. We assessed neuronal viability, reactive oxygen species (ROS) production, and oxygen consumption rate (OCR) through experiments conducted in both neuron-glia co-culture and neuron-only conditions. Furthermore, we evaluated electron transport chain (ETC) activity, mitochondrial membrane potential (MMP), and mitochondrial Ca²⁺ uptake in purified isolated mitochondria. In results, Corti.Pup neurons exhibited increased vulnerability to glutamate-induced excitotoxicity in the absence of glial support. Despite reduced ROS production, these neurons showed elevated mitochondrial OCR and proton leak, coupled with decreased non-mitochondrial OCR and ETC complex activity. Surprisingly, MMP remained elevated despite ETC dysfunction, and mitochondrial Ca²⁺ uptake was suppressed. These features indicate mitochondrial metabolic reprogramming, prioritizing MMP maintenance over ATP synthesis. The observed mitochondrial inefficiency and compensatory adaptations may impair energy production, contributing to delayed neuronal development in Corti.Pups. These findings suggest that mitochondrial dysfunction and metabolic remodeling play central roles in the pathogenesis of neurodevelopmental disorders such as ADHD.

INTRODUCTION

Recent advances in mitochondrial biology and clinical neuroscience have highlighted mitochondria as key therapeutic targets for a variety of neurological disorders. Mitochondrial functions are crucial for brain development and neurogenesis, both during embryonic stages and adulthood [1-3]. Consequently, the brain is highly susceptible to mitochondrial dysfunction. Growing evidence suggests that mitochondrial dysfunction plays a pivotal role in various brain disorders, including neurodevelopmental conditions like attention deficit hyperactivity disorder (ADHD), autism spectrum disorders, psychiatric illnesses like schizophrenia, and neurodegenerative diseases such as Alzheimer’s disease [4-6]. These disorders are commonly linked to mitochondrial dysfunction, leading to decreased energy metabolism, disrupted oxidation-reduction balance, and fragmented mitochondria. Previously, we reported that prenatal exposure to high corticosterone induced ADHD-like behaviors with cognitive deficits in rats, likely due to the downregulation of dopaminergic and phosphokinase A (PKA)-mediated signaling cascades during neurodevelopment, triggering the pathogenesis of psychiatric disorders like ADHD [7, 8]. In these studies, neurons prenatally exposed to elevated levels of corticosterone exhibited delayed synaptic development and impaired synaptic plasticity associated with learning and memory processes, without showing signs of cellular damage or apoptosis. These findings suggest the presence of compensatory mechanisms or reprogrammed metabolic regulations that may underlie the delayed neuronal development observed in these conditions. In neurodevelopmental disorders, metabolic dysregulation and mitochondrial dysfunction have already been highlighted by many researchers as key factors contributing to both the initiation and progression of the disease [9-13]. Therefore, it is necessary to investigate whether there is mitochondrial dysfunction in our models, particularly in the context of ADHD pathogenesis, where neurons exhibiting functional vulnerability or delayed development may reflect underlying metabolic abnormalities. To explore this hypothesis, we assessed mitochondrial function across multiple parameters, including oxygen consumption rate (OCR), mitochondrial membrane potential (MMP), calcium handling, and reactive oxygen species (ROS) production, with a focus on their potential contribution to cellular vulnerability and viability in neurons prenatally exposed to high corticosterone. These functional analyses were conducted in both intact neurons and isolated mitochondria to gain comprehensive insight into the metabolic status of our model showing ADHD symptoms.

MATERIALS AND METHODS

Materials

Minimal Essential Medium (MEM), Neurobasal medium, fetal bovine serum (FBS), B-27 serum-free supplement, glutamine and penicillin/streptomycin were purchased from Gibco BRL (Grand Island, NY, USA). 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) purchased from Amresco (Solon, OH, USA). Mito-SOX, Rhodamine123, Rhod-2am has been synthesized and become commercially available (e.g., Thermo Fisher, Waltham, MA USA). 2’,7’-Dichlorofuorescin diacetate (DCF-DA) and other reagents were obtained from Sigma-Aldrich (St. Louis, MO, USA) unless otherwise indicated.

Animals and drug administration

Sprague–Dawley (SD) rats used in this experiment were bred in the animal facility of the Medical School of Jeju National University. The environment of the breeding room was automatically controlled at a temperature of 23±3°C and humidity of 50±10%. The lighting system was set to turn on and off repetitively for 12 h a day. Food and water were provided ad libitum. One 7- to 8-week-old female rat and one 8- to 9-week-old male SD rat were mated in a cage for breeding. And then, the fertilization was confirmed by the presence of a vaginal plug, and we considered embryonic day E0. According to the previous paper reporting the concentration for the repetitive injection, corticosterone (Sigma, St. Louis, MO, USA) was dissolved in normal saline (0.1% Tween 80, 0.9% NaCl, Sigma, in DW, 20 mg/kg) [14] and administrated subcutaneously to pregnant rats (Corti.Moms) once daily during pregnancy immediately after confirming fertilization. Next, the rat pups born from them were defined as “Corti.Pups”. Rat pups born from rats which were injected with saline or not treated, were defined as “Nor.Pups”. All animal experiments and procedures were performed with approval from the Animal Care and Use Committee of Jeju National University.

Primary culture with neurons of prefrontal cortex (PFC)

PFC neurons and glia were isolated from the brains of 1-day-old Corti.Pups or Nor.Pups. The PFC tissues were dissected from the neonatal brains and transferred to a plating medium containing MEM supplemented with 10% fetal bovine serum, 25 mM glucose, 1 mM sodium pyruvate, 25 mM glutamine, and 1% penicillin/streptomycin. The isolated tissues were then dissociated by gentle trituration using pipettes and plated onto poly-L-lysine-coated 96-well and 24-well plates at a density of 2.5×10⁴ cells/well and 8×10⁴ cells, respectively. After six hours, the plating medium was replaced with Neurobasal media supplemented with 2% B-27, 50 mM glutamine, and 1% penicillin/streptomycin, and half of the culture medium was replenished every four days. The neurons were incubated at 37°C in a humidified 5% CO₂ / 95% O₂ atmosphere and used in the experiments at days in vitro 8 to 12.

Measurement of cell viability and ROS production

Cell viability and ROS production were assessed using MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide] and DCF-DA [2’,7’-dichlorofluorescein diacetate] assays, respectively. Primary PFC neurons, cultured either alone or in co-culture with glial cells (DIV 8~12), were exposed to various concentrations of glutamate, rotenone, or KCl for 24 hr to induce excitotoxic stress. After treatment, the medium was replaced with a Ca²⁺-free external solution containing 25 µM DCF-DA or 2.5 µM MitoSOX, followed by incubation at 37°C for 20 min in the dark. Fluorescence intensity was measured at excitation/emission wavelengths of 485/535 nm (DCF-DA) and 510/580 nm (MitoSOX) using a SpectraMax iD3 microplate reader (Molecular Devices, San Jose, CA, USA). Following fluorescence measurement, the cultures were gently washed, and 0.4 mg/ml MTT solution was added to each well and incubated for 1 hr at 37°C. The supernatant was then removed, and the resulting formazan crystals were dissolved in dimethyl sulfoxide (DMSO). Absorbance was recorded at 550 nm using the same microplate reader. Cell viability was expressed as the relative absorbance ratio compared with untreated controls.

Preparation of pure mitochondria from PFC tissue

Mitochondria were isolated from the PFC tissue of rat brains, as previously described [15]. Briefly, the cortical tissues were separated from the brains of 15- to 20-day-old rats, placed in an EGTA-containing isolation buffer, and homogenized using a Dounce-type tissue grinder. Pestles A and B were used sequentially, with clearances of 0.0035~0.0065 mm and 0.0010~0.0030 mm, respectively. The isolation buffer contained 225 mM mannitol, 75 mM sucrose, 5 mM HEPES, 3 mM EGTA, and 0.1% BSA, adjusted to pH 7.4 with KOH. The homogenates were centrifuged at 600×g for 10 minutes, and the supernatant was then transferred to a new tube and centrifuged again at 12,000×g for 10 minutes. The mitochondrial pellet was resuspended in the isolation buffer without EGTA and homogenized using Dounce-type tissue grinders. The clearances of pestles A and B at this stage were 0.0028~0.0047 mm and 0.0008~0.0022 mm, respectively. The homogenates were centrifuged again at 12,000×g for 10 minutes. All procedures described above were conducted at 4°C. The isolated mitochondrial proteins were quantified using a Bio-Rad protein assay, and the oxygen consumption rate was measured. Electron microscopy was also performed to evaluate the metabolic activity and morphology of the isolated mitochondria.

Evaluating mitochondrial parameters in isolated mitochondria

After isolating pure mitochondria from the rat PFC tissue, the mitochondrial suspension was incubated in a mitochondrial assay buffer for 10 minutes at 37°C. This buffer contained Mito-SOX and Rhodamin1,2,3 Rhod-2am, which were used to measure mitochondrial reactive oxygen species and mitochondrial membrane potential, as well as mitochondrial calcium levels, respectively. The mitochondrial assay buffer consisted of 220 mM mannitol, 70 mM sucrose, 10 mM KH₂PO₄, 5 mM MgCl₂, 2 mM HEPES, 1 mM EGTA, and 0.2% fatty acid-free BSA, with a pH of 7.2. The mitochondria in the suspension were treated with rotenone and CaCl₂, along with other reagents, in 96-well plates. The mitochondrial suspension was plated at 100 μl/well. The measurements of MMP and mitochondrial ROS (mito-ROS) used a mitochondrial reaction buffer containing 100 mM sucrose, 100 mM KCl, 2 mM KH₂PO₄, 0.01 mM EGTA, 5 mM HEPES–KOH (pH 7.4), and 3 mM each of succinate, glutamate, and 0.1 mM ADP-K plus. The fluorescence intensities of Mito-SOX and Rh123 were measured at 510/580 and 505/560 nm excitation and emission wavelengths, respectively, using a fluorescence microplate reader. The fluorescence intensity of Rhod-2am (552/581 nm excitation and emission wavelengths) was used to measure calcium levels within the mitochondria. The mitochondria Ca²⁺ (mito-Ca²⁺) experiment used a K⁺-based medium containing 125 mM KC1, 0.5 mM MgCl₂, 3 mM KH₂PO₄, 10 mM Hepes-KOH, 5 mM succinic acid, 0.01 mM EGTA and CaCl₂ as needed.

Measurement of OCR in primary neurons

Primary cell culture was conducted as described previously. Briefly, 100 μl of cell suspension was seeded into each well of a poly-L-lysine coated 24-well Seahorse XF24 (Seahorse Bioscience) microplate, ensuring the cell density aligned with the specific growth requirements. The microplate was placed in a standard cell culture incubator for 1~3 hours to allow the cells to adhere evenly, preventing them from adhering to the sides of the wells. Once the cells had adhered, 400 μl of culture media was gently added to each well, bringing the final volume to 500 μl. Four wells were left unseeded for background correction. The instrument run protocol was designed according to the specific requirements of the experiment, including setting up the measurement cycles and any compound additions. The cells were carefully rinsed by removing 80% of the cultured medium and adding 500 μl of unbuffered DMEM medium supplemented with glucose, sodium pyruvate, and glutamine. Finally, the microplate was placed in a Seahorse XF analyzer, and the experiment was run according to the protocol design. This combined method allows for comprehensive measurement of OCR in both isolated mitochondria and primary neurons, providing valuable insights into mitochondrial function and cellular respiration (Supplementary Table 1).

Measurement of OCR and ATP in pure mitochondria

The OCR of isolated mitochondria from rat PFC tissue was measured using a Seahorse XF24 Extracellular Flux Analyzer. Two experimental protocols to assess mitochondrial function were performed based on the work by the Rogers team [16]. First, mitochondrial OCR was measured in a coupled state using 5 μg of brain mitochondria per well, with 10 mM succinate and 2 μM rotenone as substrates. Second, sequential electron flow through the electron transport chain (ETC) was examined, utilizing 5 μg of mitochondria per well with 10 mM pyruvate, 2 mM malate, and 4 μM Carbonyl cyanide 4-(trifluoromethoxy)phenyl-hydrazone (FCCP) to evaluate the activity of Complexes I~IV (CI~CIV). The OCR measurements were conducted using a Seahorse XF-24 extracellular flux analyzer, following the manufacturer’s protocol. Briefly, 5 μg of isolated mitochondria were suspended in 50 μl of assay medium containing 220 mM mannitol, 70 mM sucrose, 10 mM KH₂PO₄, 5 mM MgCl₂, 2 mM HEPES, 1 mM EGTA, and 0.2% fatty acid-free BSA, with the pH set at 7.2. One day prior to the experiment, a sensor cartridge was placed in the calibration buffer and incubated overnight in a non-CO₂ environment at 37°C. During the experiment, the following reagents were added sequentially according to the mitochondrial OCR protocol: ADP, oligomycin (an inhibitor of mitochondrial ATP synthase), FCCP (an ETC system accelerator), and rotenone and antimycin A (inhibitors of CI and CIII, respectively). The isolated mitochondria were evaluated for their respiratory parameters, including basal respiration, ATP-linked respiration, proton leak, maximal respiration, and spare respiratory capacity. For the sequential electron flow through the ETC protocol, the reagents were injected in series: rotenone (2 μM) and antimycin A (4 μM), 10 mM / 100 μM ascorbate / N,N,N’,N’-tetramethyl p-phenylenediamine. The OCR data recorded by the sensor cartridges were analyzed using the Seahorse XF-24 software according to the appropriate formula.

Mitochondria isolated from PFC tissue were also used for quantification of ATP production. After protein concentration was determined, each mitochondria sample was diluted to yield 30 μg of protein per well in a final volume of 100 μl. The concentration of the mitochondria suspension was adjusted so that it constituted one-tenth of the total reaction in each well. Standard reaction solutions were prepared according to the manufacturer’s instructions for the ATP Determination Kit (Thermo Fisher Scientific, A22066). ATP standard solutions were prepared and dispensed into the wells (10 μl per well), and 10 μl of the mitochondrial suspension was added to each well under the same conditions. Subsequently, 90 μl of the standard reaction solution was added to bring the total volume to 100 μl, followed by incubation at room temperature for 15 minutes. Luminescence generated by the luciferase-ATP reaction was measured at 560 nm using a white opaque 96-well plate. After subtracting the blank values, ATP concentrations were calculated from a standard curve generated using ATP standards ranging from 1 to 1,000 nm (n=3 replicates). Results were expressed as pmol ATP per μg of protein.

Statistics analysis

The data were presented as the mean value±SEM through at least 4 independent experiments. Statistical analysis was performed using Student t-test and one-way ANOVA. The differences between groups were considered to be statistically significant when p<0.05 or p<0.01.

RESULTS

Corti.Pup PFC neurons show increased vulnerability to glutamate-induced excitotoxicity

Previously, we observed delayed neurodevelopment in hippocampal neurons of Corti.Pups, which was associated with impairments in cognitive behavior, learning, and memory functions [7, 8]. These findings suggest that Corti.Pup neurons may exhibit underlying functional and structural vulnerabilities. To investigate this possibility, we assessed neuronal viability in response to neurotoxic stimulation by treating primary neurons with glutamate (0.1, 0.5, 1, 50, 100, or 500 µM), KCl (8, 10, or 15 mM) or rotenone (1, 100, 500, or 1,000 nM). MTT assays were performed under the conditions of co-culture with glia cells (Fig. 1A~C) or neuron-only culture (Fig. 1D~F) to evaluate their susceptibility to each treatment. In this experiment, when neurons were co-cultured with glial cells, the neurotoxic effects of glutamate, KCl and rotenone were significantly attenuated, resulting in a less pronounced reduction in neuronal viability in Nor.Pups and Corti.Pups without significant differences between two groups (Fig. 1A~C). However, the excitotoxic effect of glutamate on neuronal viability in Nor.Pup and Corti.Pup neurons exhibited significant differences between Nor.Pup and Corti.Pup neurons under the neuron-only culture condition (Fig. 1D). The results reveal those increasing concentrations of glutamate significantly reduced cell viability in both groups (at 50, 100 and 500 µM, respectively; Nor.Pup, 67.10±7.60, 57.70±6.52, 54.53±6.75%, n=6; Corti.Pup, 43.74±4.71, 43.60±6.48, 38.15±7.40%, n=6; **p<0.01, compared to 0 µM). Particularly, Corti.Pup neurons exhibited greater vulnerability to glutamate, showing a more pronounced reduction in viability than that observed in Nor.Pup neurons under the neuron-only culture condition (^†p<0.05, ^††p<0.01, compared to Nor.Pup at all doses). However, from both groups exhibited a similar, dose-dependent decrease in viability in response to KCl treatment, with no significant differences observed between the groups (Fig. 1E; Nor.Pup, 81.37±8.03, 63.73±6.34, 51.22±5.09%, n=6; Corti.Pup, 70.36±1.93, 68.83±2.16, 66.30±2.04; n=6; **p<0.01, compared to 0 mM). Rotenone, a mitochondrial complex I inhibitor, potentially reduced neuronal viability in both groups; however, no significant difference was also observed between Corti.Pup and Nor.Pup neurons except at 1 nM (Fig. 1F; at each dose, Nor.Pup, 33.88±1.98, 25.29±1.39, 26.95±1.90, 29.34±4.64%, n=6; Corti.Pup, 40.07±1.58, 24.27±2.59, 23.70±3.48, 28.92±2.66%, n=6; **p<0.01, compared to 0 nM). These results suggest that Corti.Pup neurons may be more sensitive to acute Ca²⁺ overload induced by glutamate compared to Nor.Pup neurons. These findings imply that glial cells exert a protective effect against excitotoxic stress, particularly shielding Corti.Pup neurons, which appear intrinsically more vulnerable, possibly due not only to mitochondrial dysfunction but also to a compensatory metabolic reprogramming for protection.

Fig. 1.

Comparative analysis of cell viability in Nor.Pup and Corti.Pup PFC neurons under excitotoxic conditions. (A~C) Cell viability of PFC neurons co-cultured with glial cells after treatment with increasing concentrations of glutamate (A), KCl (B), and rotenone (C). (D~F) Cell viability of neuron-only cultures treated with the same agents: glutamate (D), KCl (E), and rotenone (F). In neuron-only cultures, Corti.Pup neurons showed significantly increased vulnerability to glutamate treatment compared to Nor.Pup neurons, whereas similar or slightly reduced sensitivity was observed for KCl and rotenone. Data represent mean±S.E and n=6 per group. * and **, p<0.05 and 0.01, respectively, compared to ‘0’; ^† and ^††, p<0.05 and 0.01, respectively, between Nor.Pup and Corti.Pup.

Corti.Pup neurons exhibit reduced ROS production in response to rotenone

In Fig. 2, the ratios of intracellular ROS or mito-ROS to cell viability were calculated by dividing the relative fluorescence units (RFU) values by the MTT optical density’s values. As shown in Fig. 1, MTT reduction was used to compare relative cell viability between groups under identical culture conditions, reflecting excitotoxic vulnerability rather than absolute mitochondrial metabolism. In subsequent fluorescence assays (e.g., DCF-DA), MTT values served solely as normalization references to adjust for cell density. Thus, throughout this study, MTT was treated as a relative indicator of neuronal viability, not as a direct measure of metabolic activity. Under co-culture conditions with glial cells, treatment with glutamate and rotenone led to slight but significant increases in intracellular and mitochondrial ROS levels in Nor.Pup neurons at certain doses. Interestingly, Corti.Pup neurons did not show such significant increase in ROS but also exhibited overall lower ROS production compared with Nor.Pup neurons (Fig. 2A, B, n=6, *p<0.05, **p<0.01, compared to control). This finding suggests that neuronal ROS production may be downregulated, or that additional antioxidant mechanisms mediated by glial cells may be contributing to this reduction. In the absence of glial cells, glutamate induced a comparable increase in both intracellular ROS and mito-ROS levels in Nor.Pup and Corti.Pup neurons. However, rotenone treatment markedly elevated ROS levels in Nor.Pup neurons, while Corti.Pup neurons remained unaffected, indicating a lower sensitivity to rotenone (Fig. 2C, D, n=6, *p<0.05, **p<0.01, compared to control). Although Corti.Pup neurons demonstrated greater vulnerability to glutamate in Fig. 1, they produced less ROS under similar excitotoxic conditions in this experiment. This discrepancy implies either reduced oxidative stress or the presence of enhanced antioxidant mechanisms. These findings suggest the existence of alternative protective mechanisms within the intracellular and/or mitochondrial compartments that may protect against oxidative signaling.

Fig. 2.

Comparison of intracellular and mitochondrial ROS levels in Nor.Pup and Corti.Pup neurons under excitotoxic and oxidative stress conditions. (A~D) ROS levels were measured and normalized to MTT values (ROS/MTT) in Nor.Pup and Corti.Pup neurons under co-culture with glial cells (A, B) or neuron-only culture (C, D). Both intracellular and mitochondrial ROS of Corti.Pup PFC neurons were not significantly increased by glutamate and rotenone treatment under co-culture with glial cells. In neuron-only culture, Corti.Pup neurons exhibited significantly less increase in ROS levels compared to normal pups, with notably reduced responsiveness to rotenone. Data represent mean±S.E. and n=6 per group. * and **, p<0.05 and 0.01, respectively, compared to ‘0’; ^† and ^††, p<0.05 and 0.01, respectively, between Nor.Pup and Corti.Pup.

Mitochondria in Corti.Pup PFC neurons reveal metabolic dysregulation and reduce ETC activity

Next, we focused on changes in metabolic properties by measuring OCR parameters (for details, refer figure legends and Supplementary Table 1) in both intact neurons and isolated mitochondria, as alterations in metabolic regulation are directly linked to cellular viability and oxidative signaling (Figs. 3, 4). In the OCR measurements of intact neurons (Fig. 3, Table 1), the presence of glial cells in the culture condition significantly influenced the differences in OCR parameters between Nor.Pups and Corti.Pups. Under co-culture conditions with glial cells, there were not significant differences in OCR parameters between the two groups. In particular, basal, maximal and ATP-linked OCR levels were comparable, and proton-leak values and coupling efficiency also showed no noticeable differences (Fig. 3A~C, Table 1).

Fig. 3.

Mitochondrial respiration assessed by OCR in PFC neurons from Nor.Pup and Corti.Pup. (A~C) OCR profiles were measured in PFC neurons co-cultured with glial cells. (A) Representative OCR traces showing sequential addition of oligomycin (OM), FCCP, and rotenone/antimycin A (ROT/AA). (B) Quantitative analysis of mitochondrial parameters including non-mitochondrial respiration (Non-Mito. Res.), basal respiration (Basal Res.), maximal respiration (Max. Res.), proton leak, ATP-linked respiration (ATP-linked Res.), and spare respiratory capacity (Spare Cap.). (C) Comparison of spare capacity and coupling efficiency (Coupling Eff., %). (D~F) OCR measurements in cortical neurons cultured without glial cells. (D) Representative OCR traces showing changes after compound injections. (E) Quantification of metabolic parameters reveals significant increases in basal respiration, maximal respiration, proton leak, and ATP-linked respiration in Corti.Pup neurons compared to Nor.Pup. (F) Percentage analysis of spare capacity and coupling efficiency shows reduced coupling efficiency in Corti.Pup neurons under neuron-only conditions. All OCR parameters were calculated according to the Seahorse XF equation (Supplementary Table 1). Data represent mean±S.E. *p<0.05; **p<0.01, between Nor.Pup and Corti.Pup.

Fig. 4.

Analysis of mitochondrial OCR and ETC complex activity in isolated mitochondria from PFC tissue of Nor.Pups and Corti.Pups. (A) Representative OCR traces from isolated mitochondria showing changes in respiration after sequential addition of ADP, oligomycin (OM), FCCP and rotenone/antimycin A (ROT/AA). (B) Quantification of basal respiration (Basal Res.), ADP-induced respiration (ADP Res.), ATP-linked respiration (ATP-linked Res.), and maximal respiration (Max. Res.), proton leak, and spare respiratory capacity (Spare Cap.) in isolated mitochondria. Corti.Pup mitochondria exhibited significantly increased basal, ADP-induced, ATP-linked, and maximal respiration, as well as elevated proton leak, while spare capacity was reduced compared to Nor.Pup. (C) OCR profile of ETC complex-specific activity. Rotenone (ROT), succinate (SUC), antimycin A (AA), and ascorbate/TMPD (ASC/TMPD) were sequentially applied to dissect the contributions of each complex to mitochondrial respiration. (D) Quantified ETC complex activities showing that CI, CII and CIII activities were significantly lower in Corti.Pup mitochondria, while CIV activity showed no significant difference. (E) Quantification of ATP levels in isolated PFC mitochondria showing lower ATP content in Corti.Pup compared with Nor.Pup. Data represent mean±S.E. *p<0.05; **p<0.01, between Nor.Pup and Corti.Pup. n.s., not significant.

Table 1.

Comparison of OCR parameters between with and without glia culture conditions

Parameter	With glia				Neuron-only
	Nor.Pup (n=17)	Corti.Pup (n=18)	p value		Nor.Pup (n=29)	Corti.Pup (n=20)	p-value
Non-Mito. Res.	70.53±6.06	83.37±5.64	n.s		34.86±1.33	25.39±1.10	0.025
Basal Res.	238.11±18.78	288.67±29.37	n.s		125.16±1.80	190.36±1.71	7.3E-7
Max. Res.	361.27±42.47	425.31±33.15	n.s		167.47±2.28	221.22±1.05	1.6E-4
Proton Leak	56.47±8.31	67.69±7.25	n.s		25.03±0.79	80.43±0.70	2.6E-23
ATP-Linked Res.	183.87±13.48	233.12±22.22	n.s		100.13±1.69	113.88±0.78	0.040
Spare Res. Cap.	123.16±26.40	136.64±9.72	n.s		42.38±2.02	30.85±2.58	n.s
Spare Res. Cap. (%)	145.03±8.58	152.79±4.71	n.s		133.15±0.67	125.44±1.97	n.s
Coupling Eff.	78.38±2.25	81.63±1.32	n.s		79.63±0.20	66.35±1.49	0.033

Non-Mito. Res., Non-mitochondrial respiration; Basal Res., Basal respiration; Max. Res., Maximal respiration; ATP-Linked Res., ATP-linked respiration; Spare Res. Cap., Spare respiration capacity; Coupling Eff., Coupling efficiency; n.s., not significant. Data are presented as means±S.E.

However, under culture conditions without glial cells, all OCR parameters, including basal, maximal, and ATP-linked OCR, were significantly higher in neurons from Corti.Pups compared to those from Nor.Pups, with the exception of non-mitochondrial OCR (Fig 3C, D, Table 1). Nevertheless, the ratios of spare capacity and coupling efficiency were observed to be comparable and even significantly reduced, respectively (Fig. 3F). These data provide evidence that, in the absence of glial support, Corti.Pup neurons exhibited elevated metabolic activity characterized by increased oxygen consumption showing less efficiency, suggesting a greater demand for metabolic energy to support cellular survival or development. Conversely, it is also possible that this excessive metabolic activity reflects abnormal mitochondrial respiration in Corti.Pup neurons. To further investigate this possibility, we next examined additional metabolic activities from isolated mitochondria, such as OCR, ETC activity, MMP, mito-Ca²⁺ and mito-ROS.

In Fig. 4 and Table 2, we compared several metabolic parameters of OCR in purified mitochondria isolated from Corti.Pup and Nor.Pup PFC tissues. Mitochondria from Corti.Pup exhibited significantly elevated basal respiration and ADP-indued respiration compared to those from Nor.Pups, indicating higher mitochondrial activity and an enhanced response to ADP stimulation (Fig. 4A, B, Table 2). ATP-linked respiration was also markedly increased in Corti.Pup mitochondria, indicating a potentially greater capacity for ATP production. However, maximal respiration measured after FCCP treatment did not differ significantly between two groups, suggesting that the peak respiratory capacity remains comparable (Fig. 4A, B, Table 2). Notably, mitochondria from Corti.Pup tissue showed a slight but statistically significant increase in proton leak, implying a degree of inefficiency in coupling respiration to ATP synthesis (Fig. 4A, B, Table 2). Despite this, the absolute spare respiratory capacity was not altered, consistent with the unchanged maximal respiration. These findings suggest that, under basal conditions, mitochondria of Corti.Pup PFC tissue do not exhibit structural or functional limitations that impair their metabolic capacity (Fig. 4B, Table 2).

Table 2.

Comparison of OCR parameters between isolated mitochondria of PFC

Parameter	Nor.Pup (n=8)	Corti.Pup (n=8)	p-value
ADP-Induced Res.	269.95±19.67	370.15±4.93	2.1E-4
Basal Res.	208.58±9.31	260.64±6.75	9.9E-4
Max. Res.	451.33±44.71	508.19±10.87	n.s
ATP-Linked Res.	277.54±25.72	425.52±17.95	9.6E-4
Proton Leak	101.88±11.12	130.02±3.81	0.048
Spare Res. Cap.	259.24±38.26	245.54±9.11	n.s

ADP-Induced Res., ADP-induced respiration; Basal Res., Basal respiration; Max. Res., Maximal respiration; ATP-Linked Res., ATP-linked respiration; Spare Res. Cap., Spare respiration capacity; n.s., not significant. Data are presented as means±S.E.

To further investigate the source of mitochondrial inefficiency, we assessed the activity of individual ETC complexes as well as ATP production. The results revealed a significant reduction in the activities of CI, CII and CIII in Corti.Pup compared to Nor.Pup mitochondria (Fig. 4C, D, Table 3). Consistent with these findings, direct quantification of ATP levels in isolated mitochondria also confirmed significantly lower ATP production in Corti.Pup (Fig. 4E; Nor.Pup=1.01±0.28; Corti.Pup=0.44±0.07 nmol/μg mit, n=4, p=0.036). Although CIV activity was also lower in Corti.Pup than in Nor.Pup, this difference did not reach statistical significance. These results suggest that, despite increased oxygen consumption and ATP-linked respiration, Corti.Pup mitochondria exhibit intrinsic deficits in ETC function and ATP synthesis, particularly in CI~CIII. This mismatch between respiratory activity and ETC efficiency may underlie the reduced coupling efficiency and limited metabolic reserve, ultimately reflecting a compensatory, yet energetically inefficient mitochondrial state in Corti.Pup neurons.

Table 3.

Comparison of ETC activity

Parameter	Nor.Pup (n=8)	Corti.Pup (n=8)	p-value
CI	118.07±5.46	78.44±8.96	1.2E-4
CII	264.60±11.56	173.96±13.10	1.2E-4
CIII	267.96±11.23	189.23±13.39	1.5E-5
CIV	274.69±12.84	249.34±12.57	n.s

CI-IV, Complex I – IV; n.s.; not significant. Data are presented as means±S.E.

Mitochondrial metabolism is redirected toward the maintenance of MMP and mito-Ca²⁺ rather than ATP production

To further evaluate mitochondrial function in Corti.Pup PFC tisuse, we assessed MMP, Ca²⁺ uptake, and ROS production (Figs. 5, 6). MMP was significantly higher in Corti.Pup compared to Nor,Pups under both control and stimulated conditions with FCCP, indicating a greater electrochemical gradient across the mitochondrial membrane (Fig. 5A, B, Nor.Pup vs. Corti.Pup; basal, 1.09±0.02 vs. 1.34±0.04, ^†p<0.05; FCCP, 1.36±0.04 vs. 1.57±0.02, ^††p<0.01, **p<0.01, compared to control). This suggests a sustained ability to maintain mitochondrial polarization, potentially linked to enhanced mitochondrial capacity. Despite the elevated MMP, Corti.Pup mitochondria exhibited a blunted mito-Ca²⁺ uptake response compared to Nor.Pups. At baseline, Nor.Pup showed greater mitochondrial Ca²⁺ levels than Corti.Pup (Fig. 5C, D; 5.29±0.09 vs. 4.80±0.08, ^††p<0.01).

Fig. 5.

Analysis of MMP and mito-Ca²⁺ uptake in Nor.Pup and Corti.Pup PFC mitochondria. (A) Time course analysis of MMP in isolated mitochondria using TMRE fluorescence (RFU), measured under control and FCCP-treated conditions. Corti.Pup mitochondria exhibited higher MMP levels compared to Nor.Pup mitochondria. (B) Quantification of MMP at 10 minutes shows significantly elevated MMP in Corti.Pup mitochondria under both control and FCCP conditions. (C) Time-dependent changes in mito- Ca²⁺ uptake following exposure to different extracellular Ca²⁺ concentrations. Corti.Pup mitochondria showed lower uptake levels than Nor.Pup under all tested concentrations. (D) Quantified mito- Ca²⁺ levels at 10 minutes showing significantly reduced Ca²⁺ accumulation in Corti.Pup mitochondria compared to Nor.Pup at each concentration of Ca²⁺. Data represent mean±S.E and n=8 per group. **p<0.01, compared to ‘control’ or ‘0’; ^† and ^††, p<0.05 and 0.01, respectively, between Nor.Pup and Corti.Pup.

Fig. 6.

Mitochondrial ROS production in response to rotenone and Ca²⁺ stimulation in isolated mitochondria from Nor.Pup and Corti.Pup PFC tissues. (A) Time course analysis of ROS levels (RFU) over 10 minutes in isolated mitochondria from Nor.Pup (left) and Corti.Pup (right). Mitochondria were treated with rotenone (ROT) and Ca²⁺. Corti.Pup mitochondria consistently showed lower ROS generation than Nor.Pup mitochondria across all conditions. (B) Quantitative comparison of ROS levels at 10 minutes across treatments. Rotenone and Ca²⁺ stimulation significantly elevated ROS in both Nor.Pup and Corti.Pup mitochondria. However, the increase in ROS production in Corti.Pup mitochondria was smaller than that observed in Nor.Pup mitochondria. Data represent mean±S.E and n=8 per group. * and **, p<0.05 and 0.01, respectively, compared to ‘control’; ^††p<0.01, between Nor.Pup and Corti.Pup.

Upon stimulation with 10 mM Ca²⁺, the increase in mito-Ca²⁺ was significantly more robust in Nor.Pup than in Corti.Pups (Fig. 5C, D; 10.34±0.37 vs. 8.61±0.09, ^††p<0.01). This attenuated Ca²⁺ uptake in Corti.Pup mitochondria may represent a protective mechanism, minimizing the risk of calcium overload and mitochondrial dysfunction under excitotoxic conditions. Consistent with this, Corti.Pup mitochondria also produced less mito-ROS compared to Nor.Pups across multiple conditions (Fig. 6A, B). Under control conditions, mito-ROS levels were significantly lower in Corti.Pups than Nor.Pups (1.69±0.12 vs. 1.06±0.06, ^††p<0.01). After exposure to 1 μM rotenone, ROS levels increased in both groups yet remained lower in Corti.Pups versus Nor.Pups (3.15±0.19 vs. 2.45±0.29, ^††p<0.01, **p<0.01, compared to control). Similarly, under Ca²⁺ stimulation, Corti.Pups again exhibited reduced ROS production (10 mM Ca²⁺; 3.75±0.59 vs. 3.30±0.15, ^††p<0.01; **p<0.01, compared to control). These findings suggest that Corti.Pup mitochondria appear to activate protective mechanisms such as enhanced antioxidant responses or suppression of oxidative stress, potentially as a compensatory strategy to prevent damage caused by inefficient oxygen utilization despite their elevated metabolic activity and MMP.

DISCUSSION

In neurons of central nervous system, mitochondria play a crucial role in neuronal energy metabolism, calcium regulation, and redox balance, all of which are essential for structural and functional neurodevelopment and synaptic functions. In this study, we investigated the mitochondrial bioenergetics and metabolic adaptations in PFC neurons of Corti.Pups, which are previously reported to exhibit delay of cognitive functions and incomplete synaptic plasticity for learning and memory [7, 8]. The metabolic state of Corti.Pup neurons was characterized by low cell viability against excitotoxicity, low ROS production, high mitochondrial OCR, high proton leak, low non-mitochondrial OCR, low ETC activity, high MMP and low mito-Ca²⁺ levels. Our findings suggest that neuroendocrine dysregulation of Corti.Pup neurons induce metabolic reprogramming of mitochondria, prioritizing MMP maintenance and survivals instead of ATP synthesis. This cellular process seems to affect cellular viability and neurodevelopment.

Previously, we reported that Corti.Pups exhibit ADHD-like behaviors, including attention deficits, impulsivity, and hyperactivity, along with delayed cognitive functions in behavioral tests [7]. These behavioral phenotypes were associated with impaired long-term potentiation of hippocampal circuits, characterized by delayed development of postsynaptic components [8]. Particularly, low expression of postsynaptic density 95 and abnormal composition of N-methyl-D-aspartate receptor subunits in postsynaptic sites of CA1 neurons provide evidence that cellular factors to regulate neuronal development may be insufficient in Corti.Pup neurons. Previously, neuroendocrinal dysregulation by prenatal hypoxic stress can induce long-term changes in brain development, resulting in mitochondrial dysfunction, oxidative stress, and apoptosis in the developing brain, ultimately contributing to the pathogenesis of neurodevelopmental disorders [17]. Furthermore, it has been well established that neurons, astrocytes, and oligodendrocytes are particularly vulnerable to damage under various pathogenic conditions during neonatal life [18]. Therefore, we specifically investigated mitochondrial metabolism in Corti.Pup neurons exhibiting delayed development, given the critical roles of mitochondrial energy production in sustaining normal neurodevelopment.

Dysregulation in neuronal development critically increases neuronal vulnerability to excitotoxicity, leading to the activation of apoptotic signaling [19-21]. In this study, we investigated neuronal viability under two conditions: one in which glial cells were co-cultured with neurons and another where neurons were cultured along (Fig. 1). Unexpectedly, our findings revealed that Corti.Pup neurons were not significantly more vulnerable than Nor.Pup neurons when glial support was present. Under co-culture conditions with glial cells, neuronal viability did not differ significantly between the two groups upon exposure to glutamate, KCl, or rotenone, all of which induce excitotoxicity. Notably, under rotenone treatment, Corti.Pup neurons even tended to show slightly higher viability than Nor.Pup neurons, with a statistically difference observed only at 100 nM. In contrast, when neurons were cultured alone with glial cells, Corti.Pup neurons became markedly more vulnerable to glutamate toxicity, while responses to KCl and rotenone remained comparable between groups. These findings indicate that glial cells play a crucial protective role, enabling Corti.Pup neurons, which are intrinsically more susceptible to excitotoxic stress, to maintain viability under supportive co-culture conditions. This is consistent with previous studies reporting the importance of glial support for neuronal survival under various conditions of neuronal diseases [22-25]. Furthermore, the fact that Corti.Pup neurons did not show increased vulnerability to KCl, or rotenone suggests that their membrane potential regulation and ROS defense mechanisms are not significantly different from those of normal neurons. Thus, Corti.Pup neurons may possess intrinsic mechanisms to maintain their essential functions or employ additional self-protective strategies in response to neurodevelopmental delay, because developmentally impaired neurons exhibit markedly reduced defenses against excitotoxic stress [19, 20]. Moreover, the finding that rotenone exerted completely different effects on neuronal viability depending on the presence or absence of glial cells suggests that an additional extrinsic antioxidant mechanism mediated by glia may contribute to the protection of Corti.Pup neurons. This issue should be further addressed in future study.

The fact that cellular strategies for maintaining viability under excitotoxic conditions differ has been reported to reflect underlying changes in metabolic characteristics, such as ROS production and antioxidant defenses [26-29]. In this study, we observed that rotenone-induced cytosolic and mitochondrial ROS levels were lowered in Corti.Pup neurons than in Nor.Pups, suggesting either reduced ROS production capacity or enhanced antioxidant signaling in Corti.Pup neurons (Fig. 2). Notably, although rotenone typically inhibits CI in the ETC, leading to increased mitochondrial ROS production in normal neurons, isolated mitochondria from Corti.Pup neurons produced significantly less ROS in response, suggesting functional alteration in CI activity [30]. Interestingly, in the absence of glial cells, Corti.Pup neurons exhibited increased vulnerability to glutamate toxicity (Fig. 1). However, the concomitant reduction in mitochondrial ROS production and lower mito-Ca²⁺ levels suggest that glutamatergic excitotoxicity may proceed independently of mitochondrial function, or alternatively, that mitochondrial metabolism could be reprogrammed for compensatory regulation to protect the neurons from excitotoxic stress. Notably, our previous study has indicated that Corti.Pup neurons show a higher dependence on NR2B-containing NMDA receptors, implying a greater susceptibility to glutamate-induced Ca²⁺ influx [8]. Thus, intracellular Ca²⁺-mediated signaling may be excessively or differentially activated under these conditions, suggesting that further multifaceted experimental approaches are needed to elucidate these adaptive or pathological processes. Taken together, the reduced ROS production observed in Corti.Pup mitochondria suggests a lower dependency on CI for ETC function, which may also reflect alteration in oxygen consumption dynamics associated with ATP production.

Measuring OCR is one of the most approaches for evaluating mitochondrial function related to cellular energy metabolism, as OCR variations are frequently observed accompanied by mitochondrial dysfunction in a range of metabolic and neurodegenerative diseases [31-34]. Interestingly, in our results of OCR measurement shown in Fig. 3, Corti.Pup neurons exhibited only slightly elevated basal OCR and maximal OCR under co-culture conditions with glial cells without any statistical significances, compared to Nor.Pup neurons. However, in the absence of glial cells, Corti.Pup neurons showed a significant increase in mitochondrial OCR, accompanied by a relative reduction in non-mitochondrial OCR. Under these conditions, basal, maximal, and ATP-linked OCR were significantly elevated, a pattern that was consistently reproduced in isolated mitochondria, indicating that the observed metabolic enhancement is intrinsic to the mitochondria (Fig. 4). This suggests that as previously reported, although glial cells contribute to metabolic homeostasis, the increased OCR observed in Corti.Pup neurons likely reflects a cell-intrinsic metabolic adaptation, rather than merely a response of glial modulation [35].

In normal neurons, an increase in OCR is typically associated with enhanced ETC activity, leading to elevated oxidative stress and increased ROS production, while proton leak is reduced [36, 37]. However, in Corti.Pup neurons exhibiting decreased ROS production capacity and increased OCR, we observed an elevation in proton leak (Fig. 3). This suggests that mitochondrial oxygen consumption is not being efficiently coupled to ATP production, reflecting a distinct form of metabolic adaptation. It is well established that inefficient coupling of oxygen consumption to ATP production can lead to excessive mitochondrial OCR as a compensatory mechanism to maintain metabolic stability [38-40]. Particularly, this mitochondrial inefficiency in energy metabolism may contribute to the delayed development of Corti.Pup neurons, despite the presence of compensatory adaptations within the mitochondria. Our results are consistent with previous reports indicating that metabolic impairments contribute to neurodevelopment deficits [41, 42]. Furthermore, since the increase in proton leak is associated with the downregulation of ETC activity in mitochondria, it is important to determine whether the reduced activity of ECT complexes contributes to the inefficiency of mitochondrial oxygen consumption in Corti.Pup neurons [36, 43]. As shown in Fig. 2C and D, the fact that rotenone-induced CI inhibition did not lead to a significant increase in ROS production suggests that Corti.Pup neurons may already have a reduced dependency on CI for energy metabolism. This observation is further supported by previous studies reporting that CI dysfunction is associated with mitochondrial metabolic reprogramming, potentially shifting the focus from ATP synthesis toward alternative mechanisms aimed at maintaining mitochondrial stability [44-46]. Furthermore, In the PFC mitochondria of Corti.Pup, which exhibited lower ATP levels, OCR measurements revealed that not only CI, but also CII and CIII activities were significantly reduced (Fig. 4C~E). This broad impairment of ETC function, together with the lower ATP levels, may underlie the increased proton leak observed in these neurons, as inefficient electron flow can disrupt coupling between respiration and ATP synthesis [47, 48]. As a result, mitochondrial energy metabolism may become less efficient, prompting a compensatory increase in oxygen consumption, a form of metabolic reprogramming aimed at maintaining cellular energy balance despite impaired ETC activity. Additionally, as shown in Fig. 5, mitochondria of Corti.Pup PFC exhibited elevated MMP despite reduced ETC activity. Since an increase in MMP typically results from high ETC activity, this finding under conditions of impaired ETC function suggests abnormal regulation of mitochondrial metabolism (Fig. 5A, B) [49]. This may indicate that mitochondria are redirecting their metabolic focus toward maintaining MMP rather than producing ATP as a protective mechanism, or that reverse electron transport is occurring [50, 51]. In neurons, excessive mitochondrial Ca²⁺ uptake has been reported to trigger ROS production and the opening of MPTP, ultimately leading to apoptotic signaling [52]. To prevent this, Corti.Pup mitochondria may have strongly suppressed Ca²⁺ influx, showing low mito-Ca²⁺ (Fig. 5C~F). While this suppression may serve as a protective mechanism, it could also compromise mitochondrial energy metabolism and reduce ATP production. Under these conditions, mitochondria may have undergone metabolic reprograming, prioritizing the maintenance of MMP over inefficient ATP synthesis in order to ensure cellular survival.

In conclusion, Corti.Pup neurons may exhibit developmental delays due to ATP deficiency resulting from inefficient energy metabolism. This inefficiency is likely linked to ETC dysfunction, which appears to trigger metabolic reprogramming as a compensatory survival mechanism. During this process, antioxidant pathways may be activated to minimize ROS production. ETC impairment may also be associated with downregulated Ca²⁺ signaling in neurons, although this remains to be clarified in future investigations. Neurons employ various survival strategies in response to metabolic stress, many of which are orchestrated by mitochondrial regulation. The interplay between these compensatory responses and their consequences - such as neuronal developmental delays and impaired synaptic plasticity - may underlie the pathogenesis of neurodevelopmental disorders, including ADHD.