Does Gadolinium Deposition Lead to Metabolite Alteration in the Dentate Nucleus? An MRS Study in Patients with MS

DISCUSSION

Our findings support the hypothesis that Gd deposition could be related to metabolite alterations in the DN. Despite evidence of metabolite changes among the studied groups, these findings were minor and their clinical importance should be investigated in further studies. To elaborate, we did not detect any significant alteration in the levels of NAA and Cho among the studied groups, suggesting the absence of neuronal tissue damage in the DN with visually detectable Gd deposition. Nonetheless, the levels of mIns, Cr, lactate, and lipid showed a significant increase in the case group, which can be interpreted as a change in cellular metabolism as discussed below.

Two studies by McDonald et al^22,23 used inductively coupled plasma mass spectrometry and detected Gd deposition in the endothelial wall, neural tissue interstitium, and nuclei of neurons in the absence of any gross histologic changes. The concept of Gd crossing the blood-brain barrier and being deposited in neural nuclei raises concern about the potential cytotoxicity of Gd. To this end, preclinical in vitro studies provided insight about the time- and dose-dependent cytotoxic and neurotoxic mechanism of Gd through disturbance of mitochondrial function and oxidative stress.^19,24-26 The authors speculated that Gd, as a calcium antagonist due to their similar atomic radius, can interrupt the mitochondrial calcium metabolism leading to cellular death.²⁷

Feng et al¹⁹ observed that Gd exposure is associated with mitochondrial membrane depolarization, caspase-3 activation, cytochrome C release, lactate dehydrogenase increase, intracellular reactive oxygen species increase, adenosine triphosphate synthesis decrease, and subsequent DNA fragmentation. All the aforementioned cellular mechanisms indicate mitochondrial dysfunction and oxidative stress leading to neuronal cell apoptosis. The cellular studies of GBCA exposure in humans could help to predict its potential long-term clinical consequences. However, there is only 1 study using PET/CT in human subjects that has investigated the metabolic activity in Gd-deposition regions.²⁸ The authors found 16% and 27% lower [¹⁸F] FDG uptake in the DN and globus pallidus, respectively, of individuals who received GBCAs.²⁸ Most interesting, a recent study used untargeted mass spectroscopy–based metabolomic analyses to investigate the plasma metabolite alterations after Gd administration.²⁰ Compared with healthy controls, patients with Gd-deposition disease showed differences in 45 biochemicals, mostly related to mitochondrial function, similar to our findings in the brain.²⁰

We propose that the observed increased levels of Cr in case participants might indicate a disturbance in cellular energy homeostasis. Cr transmission through the blood-brain barrier is minimal, and most Cr is produced in the brain using the arginine:glycine amidinotransferase and guanidinoacetate methyltransferase enzymes.²⁹ Cr is the essential component of high-energy phosphate metabolism (рCr+ADP↔Cr+ATP) and plays a vital role in channeling energy into the cytosol to maintain cellular energy homeostasis.³⁰ Therefore, because MRS measures both Cr and creatine phosphate, compromised cellular energy production due to Gd accumulation could be caused by the responsive up-regulation of Cr as the substrate to compensate for an altered cellular energy system.

Lactate is the product of anaerobic glycolysis and increases in stroke, encephalopathies, lactic acidosis, neonatal hypoxia, and mitochondrial myopathies.¹⁸ Neuronal metabolism appears to be mostly oxidative, and astrocytic metabolism is glycolytic according to the hypothesis of astrocyte-neuron lactate shuttle (Fig 3).^31-33 Astrocytes take up glucose through the glucose transporter 1 and metabolize it to lactate. Lactate, then, is conveyed to the outside of the astrocytes and is captured by neurons via monocarboxylate transporters. Neurons oxidize intracellular lactate to pyruvate and metabolize it through the oxidative phosphorylation pathway in the mitochondria.^31-33 Impaired mitochondrial function and the subsequent oxidative phosphorylation in neurons can result in accumulation of lactate. Therefore, the observed increased lactate in the case group might indicate the existence of an impaired mitochondrial energy environment.

• Download figure
• Open in new tab
• Download powerpoint

FIG 3.

The astrocyte-neuron lactate shuttle and glutamate-glutamine cycles. After traveling through the endothelial cells, glucose (Glc) enters the astrocytes via Glc transporter 1 (GlcT1). It is metabolized to pyruvate and then lactate by lactate dehydrogenase (LDH5). The lactate is carried outside the astrocyte and inside the neuron via the monocarboxylate transporters (MCTs). The lactate inside the neuron is metabolized in the oxygen pathway. Glutamate (Glu) molecules released from presynaptic neurons are transported into the astrocytes through Na⁺-dependent channels. In astrocytes, the Glu is transformed into Glutamine (Gln) or α-ketoglutarate to perform oxygen metabolism in the Krebs cycle. ATP indicates adenosine triphosphate; ADP, adenosine di-phosphate; P, phosphate; GluR, glutamate receptor; OH/HCO₃, hydroxide/bicarbonate; K+, potassium; GLT, glutamate transporter; PGK, phosphoglycerate kinase; ATPase, adenosine triphosphatase.

We also observed lower Glx levels in the case group. Glx is a mixture of similar amino acids that contribute to excitatory-inhibitory neurotransmission processes. Glutamate is released from the presynaptic neuron to stimulate glutamate receptors on the postsynaptic neuron. It enters the astrocytes through the synaptic gap via sodium (Na)-dependent excitatory amino acid activating transporters. Then, it is metabolized into glutamine (by the glutamine synthase enzyme or α-ketoglutarate by glutamate dehydrogenase) to process further oxygen metabolism in the tricarboxylic acid cycle (Krebs cycle) in the mitochondria. Glutamine, then, is transported to neurons to complete further glutamate production using phosphate-activated glutaminase (Fig 3).^31,33 Impairment of mitochondria functioning would lead to a compromised glutamate-glutamine cycle, which is dependent on the glutaminase enzyme in the mitochondria. In addition, impairment of the tricarboxylic acid cycle results in less use of α-ketoglutarate and, accordingly, impairment of the transformation of glutamate to α-ketoglutarate. Taken together, a glutamate/glutamine decrease can also be interpreted as a marker for metabolic impairment of neuronal cells.

Lipid levels were also significantly higher in the case group than in healthy controls. A recent study also detected elevated levels of 7-dehydrocholesterol, the cholesterol precursor, in patients with Gd deposition.²⁰ One of the degradation products of the leucine, 3-methylglutarate/2-methylglutarate, also showed increased levels after Gd deposition due to nicotinamide adenine dinucleotide phosphate/nicotinamide adenine dinucleotide phosphate–dependent enzyme impairment.²⁰ Sterol synthesis and leucine degradation are connected via the mevalonate shunt, leading to increased sterol production due to 3-methylglutarate/2-methylglutarate accumulation.^34,35 Hence, the authors believe that mitochondrial metabolism impairment due to Gd deposition leads to elevated levels of sterol biosynthesis through the mevalonate shunt.

Myo-inositol and its transporters may provide neuroprotection during or following brain ischemia³⁶ and are often increased in cerebral infarction. Impairment of mitochondrial function and oxidative phosphorylation can simulate a relatively hypoxic condition that leads to the accumulation of mIns. The H⁺-mIns cotransporter is mainly expressed in the brain and is strongly stimulated by a decrease in pH.³⁷ Subsequently, brain exposure to a substantial lactate load—as discussed before—and a favorable H⁺ gradient result in a significantly enhanced mIns uptake by the brain. Although mIns is known as a glial cell marker, studies have observed its elevated levels in neurodegenerative disorders without glial cell involvement, including Alzheimer disease, Huntington disease, and ataxia.¹⁸ Taken together, the aforementioned mIns observations and the associated clinical conditions raise concerns regarding the long-term sequelae of Gd deposition.

Apart from the proposed notions, there is another potential mechanism behind the accumulation of these metabolites. There is a common understanding that mIns and Cr are the main osmolytes in the brain, which undergo alteration under chronic osmotic changes.³⁸ For instance, sodium-myo-inositol cotransporter-1 in the cortical astrocytes of rats was observed to up-regulate under a chronic hyperosmolar situation.³⁹ Therefore, chronic Gd deposition in the neural tissue interstitium could result in a hyperosmolar condition that induces the production of these osmolytes. Such hyperosmolarity has not been observed at clinical levels to date.

Our findings provide preliminary insight about a potential neurochemical alteration in patients with Gd deposition. Although we observed some degree of metabolite change in association with cellular metabolism and mitochondrial function, there were no significant changes in NAA and choline that would indicate the existence of any major neuronal damage. The authors posit that Gd deposition interrupts the mitochondrial function and results in some minor metabolite changes that are kept at the minimum level through the regulatory and compensatory cellular mechanisms. Nonetheless, the Gd deposition in the DN and evidence of malfunction of the mitochondrial energy pathway shown in animal and in vitro studies raise the specter of subclinical neurotoxicity effects of Gd deposition. Some animal studies have shown loss of motor coordination, including tremor, seizure, ataxia, and stereotyped movements and myoclonus.^40,41 In a recent study, patients with Gd-deposition disease reported symptoms similar to those with known mitochondrial-related diseases.²⁰ Although 2 other clinical studies reported lower verbal fluency in patients with MS with repetitive GBCA exposure,^42,43 there has not been a comprehensive study to indicate whether there are major adverse clinical effects attributed to multiple GBCA administrations in patients with MS.^6,27 We recommend that clinicians limit CE-MR imaging to the necessary indications and conform with the latest guidelines from such reputable bodies as the Consortium of Multiple Sclerosis Centers,⁴⁴ especially in those with chronic CNS diseases requiring repeat imaging.

Our study has several limitations. Our results indicate some statistically weak associations (.05 <P < .10), but we cannot ensure that they would remain the same if the sample size were larger. Second, although we enrolled patients with RRMS in similar stages of the disease, our case and control groups were not matched for age and sex. Several studies have hypothesized that MS therapeutic agents could alter the brain metabolites, but we did not include the treatment regimens in this study because we were interested in any potential harm of Gd, even if it was caused by an interaction with medications in patients with MS. In addition, our assessment did not include clinical evaluations of the patients and was limited to a specific macrocyclic GBCA. Moreover, this study requires more in-depth comparison of background information between the case and control groups in addition to age and sex, including, but not limited to, extensive medical history, smoking history, environmental exposures, and other neurologic conditions. We highly recommend conducting randomized controlled trials that incorporate information on the suggested background conditions that could help reveal the potential clinical significance of gadolinium deposition.