Alzheimer's disease (AD), the most common form of dementia, is an incurable and progressive neurodegenerative disease . One of the major problems in the management of AD is the lack of a definitive premortem diagnostic test. The traditional approach that is used for AD diagnosis is through analysis of symptoms, clinical history, and family history. However, AD's primary early symptom, short-term memory loss, is not unique to the disease. Another common approach is to use noninvasive imaging to identify areas of neurodegeneration. With magnetic resonance imaging (MRI), it is possible to quantify the volume of brain structures and it has been established that AD patients have reduced hippocampal and total brain volumes . More recent reports suggest that changes in regional brain volume may be present as early as 10 years before clinically diagnosable AD symptoms . These findings reinforce the need for earlier diagnostic paradigms. Brain atrophy caused by neurodegeneration is a critical source for symptoms and is considered irreversible.
The best standard for definitive diagnosis has been postmortem pathology. The two pathophysiological hallmarks of AD first identified by Alois Alzheimer over 100 years ago are amyloid plaques and neurofibrillary tangles. Both of these molecular pathology hallmarks predate and may contribute to neurodegeneration . However, detecting either of these hallmarks in vivo has proven difficult. The biggest focus has been on imaging amyloid plaques, which hold promise as a more specific diagnostic factor. The most notable work in plaque imaging has been the development of the Pittsburgh Compound B agent  for positron emission tomography, though various MRI approaches have also been suggested [6-9]. Most of the MRI approaches focused on either natural iron accumulation in the plaques  or MRI contrast agents [7-10] but, to date, these approaches typically understated the presence of plaques.
Macromolecular accumulation inclusive of amyloid beta, tau protein and likely additional proteins occurs well before plaque formation and neurodegeneration is believed to be a causative factor in AD . Therefore, imaging strategies that can detect early macromolecular accumulation before plaque formation and neurodegeneration would be extremely beneficial for both the diagnosis and monitoring of AD. Importantly, detection of preplaque macromolecular burden could lead to earlier detection of AD than any other previously mentioned approach. Previously, plasma [12, 13] and cerebrospinal fluid [14, 15] amyloid burden have been studied with mixed results. However, these are both indirect measurements of AD pathology. Furthermore, a recent study on familial AD patients found changes in the amyloid burden in cerebrospinal fluid as early as 25 years before symptoms and 10 years before amyloid plaque formation . Imaging macromolecular accumulation that occurs before plaque formation and neurodegeneration directly in the areas of the brain known to be affected earliest in AD such as the cortex  or the hippocampus [18, 19] may provide an strategy for early detection of AD.
Magnetization transfer contrast (MTC) is a MRI technique to specifically detect changes in macromolecule concentration and composition . Clinically, MTC is most commonly used to track changes in myelination as way to grade multiple sclerosis lesions . The technique uses the application of a radiofrequency pulse at a specific distance from the water resonance: the offset frequency. This radiofrequency pulse causes a loss of signal intensity proportional to macromolecular concentration. When combined with a reference image where the radiofrequency pulse is not applied, the percent of signal loss can be quantified in what is referred to as the magnetization transfer ratio (MTR). Specifically, MTC evaluates changes in semisolid macromolecules .
We hypothesized that the early accumulation of macromolecules in the Tg2576 mouse model of AD would have MTC effects. At the molecular level, both amyloid and tau begin as soluble monomers that eventually become insoluble deposits [19, 23]. The aggregation and eventual insolubility of these peptides suggests that the partially insoluble intermediates might provide an MTC effect. The focus of the study was on the two areas affected early in AD mentioned above: cortex and hippocampus.
In this work, we show that MTC MRI can detect AD-related macromolecular changes in the Tg2576 mouse model of AD. The Tg2576 mouse overexpresses a mutated form of amyloid precursor protein with the Swedish familial AD mutation and exhibits accumulation of detergent-insoluble amyloid as early as 6 months and eventual plaque formation as early as 10 months of age . This mouse model of AD was chosen because it does not present the advanced hallmarks of AD such as neurofibrillary tangles and neurodegeneration and is regarded as an “early” model of AD . However, phosphorylated tau has been observed in this model suggesting that some tau pathology is present [25-29]. We were able to observe an MTC signal increase before plaque formation in this mouse model. When the Tg2576 mouse model was combined with a treatment paradigm known to reduce amyloid accumulation and plaque formation, tau pathology, and learning and memory deficits [28, 30], the MTC signal went back to baseline. This imaging strategy has the potential to serve as an early imaging biomarker for AD before plaque development and neurodegeneration ensue.
- Top of page
Diagnosis of AD is still a critical area of need with most diagnoses given by process of elimination. While other imaging paradigms are sufficient to track disease progression, they are currently ineffective at detecting the earlier stages of AD. The goal of this study was to test if MTC MRI is sensitive to macromolecular changes that are seen early in AD. The results show that MTC MRI can detect differences in the Tg2576 mouse model well before plaque formation and learning and memory deficits. This is significant because no other MRI imaging modality has been reported to detect this earliest stage of AD . Importantly, the increased MTR in the Tg2576 mouse reversed back to baseline in the antioxidant treatment model. Therefore, we believe MTC MRI may be used in the Tg2576 model to examine response to treatment.
It will be important to identify the molecular source of the increased MTR. The two major molecular hallmarks of AD, and therefore likely contributing sources of the MTC signal, are amyloid beta and tau protein . However, there are other molecular pathologies that could also be involved including gliosis , vascular alterations [28, 40], and cytoskeletal rearrangements [28, 41]. It would be ideal to test each individual possibility individually to identify how much it contributes to changes in the MTC signal but that will be challenging to do in vivo. Therefore, verification of these results in the Tg2576 in other mouse models of AD will be necessary to ensure the result is not restricted to the Tg2576 model. For example, in the case of tau pathology, the Tg2576 is not the ideal model and better models of tau aggregation will be needed to evaluate the contributions of tau accumulation [42, 43]. The Tg2576 mouse model used here only has limited tau pathology and does not exhibit neurodegeneration unlike that which is observed in AD patients. Animal models that include all three pathological hallmarks (amyloid beta, tau, and neurodegeneration) may be more accurate predictors of the usefulness of MTC. Such additional models would allow for consolidation of our findings compared with the reported clinical MTC data for AD patients. Specifically, symptomatic AD patients (late stage) show a lower MTR than controls [44, 45]. This lowered MTR is likely due to neurodegeneration and subsequent loss of macromolecules. Additionally, as neurons die they are replaced by cerebrospinal fluid which has a much lower MTR.
In summary, we have shown that the MTR is significantly increased in the Tg2576 mouse model, a model of early AD that exhibits amyloid beta accumulation, tau hyperphosphorylation, and no neurodegeneration. These data demonstrate that MTR can detect macromolecular changes in a mouse model of early stage AD and that MTC imaging should be further evaluated in additional, more complex models of AD to better define the clinical potential.