Species: Mouse
Modification: Conditional Knock-out
Disease Relevance: Parkinson's Disease
Strain Name: Ndufs2^tm1.1Job Slc6a3^{tm1.1(cre)Bkmn}/SurmJ
Genetic Background: Mixed 129, C57BL/6J, C57BL/6N, and SJL background.
Availability: Available through The Jackson Laboratory, Stock# 036313, Live and cryopreserved.

Modification Details:

Ndufs2fl/fl mice contain two loxP sites flanking exon 2 of the Ndufs2 gene, and DATIREScre/+ mice contain an internal ribosome entry site–linked Cre recombinase downstream (23 bp) from the stop codon of the endogenous Slc6a3 gene (The Jackson Laboratory).

Summary

The MCI-Park mice are compound mutant (Ndufs2fl/fl; DAT^IREScre/+) animals in which Ndufs2, a gene encoding a core subunit of the mitochondrial respiratory chain, is selectively inactivated in dopaminergic neurons (The Jackson Laboratory). The mice carry a homozygous flox mutation in the Ndufs2 gene and a heterozygous Cre transgene (DATIREScre) mutation driven by the Slc6a3 promoter that enables expression specifically in cells expressing the dopamine transporter (DAT). This leads to selective deletion of Ndufs2 in dopaminergic cells. Ndufs2 encodes a core subunit of the mitochondrial complex I (MCI) NADH:ubiquinone oxidoreductase, which contributes to the ubiquinone-binding site of MCI (The Jackson Laboratory; González-Rodríguez et al., 2021). Loss of Ndufs2 in DAT-expressing cells leads to MCI dysfunction and levodopa-responsive parkinsonism.

Unless otherwise specified, the ages of the mice in experiments described range from 20 to 120 days (González-Rodríguez et al., 2021).

Overall Health

MCI-Park mice exhibit progressive neural and behavioral decline, with mild deficits starting after around postnatal day 30 and becoming overt by day 100 (The Jackson Laboratory). Younger 6- to 8-week-old MCI-Park mice are considered to be in the prodromal state and have no overt motor dysfunction, while older 12- to 18-week-old mice are in the symptomatic parkinsonian state and do exhibit motor deficits (Summa et al., 2024; Patricia González-Rodríguez and D. James Surmeier 2024, personal communication).

At around 5 to 6 months of age, mice begin dying of unknown causes (González-Rodríguez et al., 2021). MCI-Park mice may have a low body temperature and may become scruffy in appearance around postnatal day 60 (The Jackson Laboratory).

Neuropathology

Around postnatal day 30, protein levels of tyrosine hydroxylase (TH) were decreased in the dorsal, but not ventral, striatum of MCI-Park mice compared to Dat-cre^−/−-Ndufs2^fl/fl controls (described as wild-type in the report) (González-Rodríguez et al., 2021). The number of TH-expressing neurons in the substantia nigra (SN) and ventral tegmental area (VTA) did not differ between controls and MCI-Park mice. By age 60 days, however, TH expression in the somatodendritic region of SN dopaminergic neurons in MCI-Park mice was decreased by about half compared to that of control mice. Nonetheless, no differences were observed in the degeneration of dopaminergic axons and cell bodies at 60 days of age using retrograde labelling with a striatal injection of Fluoro-Gold. Structurally, the dendritic arbors of SN dopaminergic neurons in MCI-Park mice at around 40 and 60 days of age did not differ from control mice. Thus, at 60 days of age, MCI-Park mice demonstrate a phenotypic downregulation in the absence of obvious neurodegeneration. By 120 to 150 days of age, however, about 40% of SN dopaminergic neurons are lost.

In 30- to 40-day-old MCI-Park mice, pacemaking activity of SN dopaminergic neurons was blunted, as measured using cell-attached electrophysiological recordings (González-Rodríguez et al., 2021). Using RiboTag to measure levels of mRNA that are in the process of being translated to protein, an increase was observed in the mRNA expression of potassium channels as well as dysregulation in other genes involved in pacemaking and current flow (e.g., calcium channels, hyperpolarization-activated cyclic nucleotide-gated channels), which might account for this functional change in pacemaking activity. Additionally, using whole-cell recording and Fura-2 calcium imaging, burst spiking in SN dopaminergic neurons was found to be of higher frequency and longer duration in MCI-Park mice versus control mice.

In dopaminergic neurons from the SN, MCI-Park mice at postnatal day 40 exhibited differences compared to wild-type mice in expression of genes (using the RiboTag approach) related to axonal growth and transport, synaptic transmission, dopamine synthesis/storage, and presynaptic regulation (González-Rodríguez et al., 2021). By postnatal day 60, the ventral striatum also exhibited a loss of axonal proteins associated with dopaminergic signaling.

Dopamine Deficiency

Dopamine release was measured in ex vivo brain slices using fast-scan cyclic voltammetry as well as with dLight, a genetically encoded optical dopamine sensor (González-Rodríguez et al., 2021). In dopaminergic axons in the dorsolateral striatum, electrical stimulation led to virtually no dopamine release in MCI-Park mice by age 30 days. Striatal dopamine release could be rescued by levodopa at 30 days of age, but not at 60 days of age. However, in the SN, somatodendritic dopamine release did not differ between controls and MCI-Park mice at age 30 days, although it did drop by 75 percent by 60 days of age.

Liquid chromatography and mass spectrometry analysis confirmed a drop in dopamine as well as DOPAC in the striatum at age 30 and 120 days in MCI-Park mice versus control mice. Levels of striatal acetylcholine did not differ from control mice at either of those ages, whereas serotonin was elevated at 120 days of age, but not at 30 days (González-Rodríguez et al., 2021).

Non-motor Impairment

Around postnatal day 30, MCI-Park mice were unable to perform an associative learning task (Y-maze test) that depends on dopamine-dependent striatal synaptic plasticity (González-Rodríguez et al., 2021). While levodopa was able to rescue mice from deficits in Y-maze test performance at 30 days of age, it was ineffective at 60 days of age.

MCI-Park mice exhibited significantly altered sleep-wake regulation and altered EEG activity patterns (Summa et al., 2024). In older mice (14-18-week-old), but not younger ones (6-8-week-old), the time spent awake was significantly increased compared to wild-type mice. Older MCI-Park mice also slept for a total shorter duration than wild-type mice. Moreover, when considered together, older and younger mice had a disrupted pattern of sleep, as they slept proportionally less during the light phase, the time when nocturnal animals sleep the most. Sleep fragmentation was also increased in younger and older MCI-Park mice compared to control mice, based on increases in the number of wake and non-rapid eye movement (NREM) bouts and a decrease in the duration of NREM bouts; the duration of wake bouts, however, did not differ between genotypes. REM sleep was also severely impaired in older (but not younger) MCI-Park mice, with decreased time, number of bouts, and relative proportion of time spent in REM sleep observed compared to control mice.

Motor Impairment

No gross impairments in motor function were observed at 30 days of age.

At 40 days of age, MCI-Park mice travelled the same total distance as control mice on the open-field test, but the mutant mice paused more frequently, and the pauses increased progressively with age (González-Rodríguez et al., 2021). By 60 days of age, MCI-Park mice started to show significant reductions in total distance travelled, which also worsened at 120 days of age. Levodopa improved performance on the open-field at all ages at which there were deficits compared to control mice.

MCI-Park mice showed perturbed rearing behavior in the cylinder test (González-Rodríguez et al., 2021). Compared to control mice, mutant mice exhibited fewer instances of rearing starting at 40 days of age, and by 60 days of age, the duration spent in the elevated posture was longer.

By 100 days of age, but not at 60 days of age, MCI-Park mice had splayed hindlimbs, and on a treadmill test, they demonstrated abnormal paw placement and alterations in stride (González-Rodríguez et al., 2021).

Striatal motor learning (adhesive removal test) was impaired in MCI-Park mice versus control mice at 30 days of age (González-Rodríguez et al., 2021). This dysfunction on the adhesive removal test progressively worsened at 40, 60, and 120 days of age. Unlike for the Y-maze test, levodopa was able to rescue performance on the adhesive removal test at all ages tested.

Mitochondrial abnormalities

By postnatal day 20, mitochondria from MCI-Park mice were in an oxidative phosphorylation deficit and consumed more ATP than they produced (González-Rodríguez et al., 2021). This finding was supported by an experiment where the adenine nucleotide transporter (ANT) inhibitor carboxyatractyloside was applied to dopaminergic neurons in ex vivo brain slices: in wild-type mice, the inner mitochondrial membrane potential rose in response to ANT blockade, while in MCI-Park mice, the potential dropped. At baseline, there were no differences in membrane potential (as measured with tetramethylrhodamine dye) between control and MCI-Park mice.

Mitochondria from dopaminergic neurons of the SN were examined using transmission electron microscopy (González-Rodríguez et al., 2021). While the density of somatic mitochondrial was normal at age 35 days, the shape of the mitochondrial cristae was significantly altered in MCI-Park mice compared to control mice. Mitochondrial density was similar between MCI-Park and wild-type mice at 60 days of age, and there were no differences in mitochondrial shape at this age.

Additionally, the expression of genes involved in mitochondrial function was altered in MCI-Park mice (González-Rodríguez et al., 2021). mRNA from SN dopaminergic neurons was sequenced, and evidence of metabolic reprogramming was observed. Namely, a Warburg-like effect was present in MCI-Park samples, with an upregulation of genes involved in glycolysis and a downregulation of genes involved in oxidative phosphorylation. A functional experiment confirmed this reprogramming to a predominantly glycolytic state, whereby MCI-Park mice were sensitive to inhibition of glycolysis (with 2-deoxyglucose) but not to inhibition of mitochondrial complex V, which is a key step in oxidative phosphorylation. Control mice, on the other hand, were predominantly sensitive to inhibition of oxidative phosphorylation with oligomycin.

Phenotype Characterization

Q&A with Model Creator

Q&A with Patricia Rodriguez-Gonzalez and D. James Surmeier

What would you say are the unique advantages of this model?

The model relies upon a very specific, targeted disruption of mitochondrial function in dopaminergic neurons that mimics a deficit that has been observed in humans with Parkinson’s disease (PD). The model also reproduces the progressive, axon-first pathology in dopaminergic neurons widely thought to be occurring in PD.

What do you think this model is best used for?
The mouse models progressive PD but does so on an experimentally reasonable time scale of weeks and months rather than years. The progressive nature of the model allows for study, not only of “prodromal” PD, but also the progressive changes associated with clinical PD.

What caveats are associated with this model?
Although loss of mitochondrial complex I (MCI) function in dopaminergic neurons is a hallmark of idiopathic PD, there is no evidence that this loss is due to a disruption in Ndufs2 function; targeting Ndufs2 may lead to a more rapid and profound loss of MCI function than that occurring in humans. Furthermore, the deletion of Ndufs2 in the MCI-Park model occurs in the late embryonic period, raising the possibility that there are developmental adaptations to the insult that are not relevant to adult-onset PD. Another caveat is that it relies upon a dopamine transporter promoter/enhancer sequence to control Cre recombinase expression and Ndufs2 deletion. This reliance could affect the timing and extent of the Ndufs2 deletion in a way that does not mimic the acquired deficit hypothesized to occur in humans. Another caveat is that the pathology in this model is restricted to dopaminergic neurons, whereas in human PD there is pathology in a variety of other locations in the brain. Lastly, there are reasons to think that the mechanisms responsible for human PD are heterogeneous; the MCI-Park mouse provides a model of mitochondrially driven pathology.

COMMENTS / QUESTIONS

No Available Comments

Make a comment or submit a question

To make a comment you must login or register.

References

Paper Citations

1. González-Rodríguez P, Zampese E, Stout KA, Guzman JN, Ilijic E, Yang B, Tkatch T, Stavarache MA, Wokosin DL, Gao L, Kaplitt MG, López-Barneo J, Schumacker PT, Surmeier DJ. Disruption of mitochondrial complex I induces progressive parkinsonism. Nature. 2021 Nov;599(7886):650-656. Epub 2021 Nov 3 PubMed.

External Citations

1. The Jackson Laboratory
2. The Jackson Laboratory
3. The Jackson Laboratory, Stock# 036313
4. The Jackson Laboratory

Further Reading