CAG repeat number correlates with the rate of brainstem and cerebellar atrophy in Machado-Joseph disease

Machado-Joseph disease (MJD) is an autosomal dominant neurodegenerative disorder that is caused by an expansion of CAG trinucleotide repeats in a novel gene located at chromosome 14q32.1.¹

The CAG repeat size shows considerable individual variation among MJD patients and exhibits an inverse correlation with age at onset,^1,2 but it is not understood how the CAG repeat size is reflected by the clinical manifestations and pathologic changes.

In this study, we quantified atrophic changes of the brainstem and cerebellum by MRI and analyzed the correlation between the progression of their atrophy and CAG repeat size.

Methods and patients. The subjects were 30 Japanese patients who met the clinical criteria for MJD as defined by Lima and Coutinho,³ and who also were diagnosed by the expansion of the CAG repeat size of the MJD1 gene.^1,2 Age at MRI examination ranged between 22 and 78 years (mean 46.1 years). Age at onset was determined as the age when a patient first noticed any symptoms, as elucidated from information provided by the patient. The control group consisted of 32 healthy volunteers without any neurologic disorder, matched with the MJD group for age (mean 45.7 years; range 24 to 71 years; t = 0.202; p > 0.05).

Genomic DNA was extracted from peripheral blood samples of the MJD patients, and PCR amplification of the CAG repeats in the MJD1 gene was performed using a protocol previously described.¹ PCR products were separated by electrophoresis with an autoread sequencer(ALFred, Pharmacia, Uppsala, Sweden), and their sizes were determined by comparison with known repeat length standards or an M13 sequencing ladder.²

Brain MRI was performed once per patient with a MR scanner (Toshiba Visart, Tokyo, Japan) operating at 1.5 T. The section thickness was 5 mm. We measured the following parameters on the midsagittal T1-weighted spin echo images (time to repeat [TR], 500 milliseconds; time to echo [TE], 20 milliseconds): 1) the anteroposterior diameter of the pontine tegmentum(i.e., the minimum distance between the posterior surface of the pontine base and the floor of the fourth ventricle); 2) the anteroposterior diameter of the midbrain (i.e., the minimum distance between the anterior surface of the midbrain, at the nodal point with the outline of the pontine base, and the floor of the cerebral aqueduct); 3) the area of the pontine base; 4) the area of the vermis of anterior lobe of cerebellum; 5) the distance between the nasion and the inion; and 6) the area of the posterior fossa. These parameters were quantified on the computer monitor using the National Institutes of Health image version 1.60 software (Wayne Rasband, National Institutes of Health, Bethesda, MD). To adjust for individual variations in the size of the skull, the anteroposterior diameters of the pontine tegmentum and midbrain were expressed as a ratio to the distance between the nasion and the inion, and the areas of the pontine base and the vermis of anterior lobe of cerebellum were expressed as a ratio to the area of the posterior fossa. The ratio of the anteroposterior diameter of the pontine tegmentum to the distance between the nasion and the inion in each MJD patient was subtracted from the grand mean of the same ratio in all of the control cases. We defined this value as the degree of atrophy of the pontine tegmentum. We estimated the degree of atrophy of the midbrain in the same way. The ratio of the area of the pontine base to the area of the posterior fossa in each MJD patient also was subtracted from the grand mean of the same ratio in all of the control cases. We estimated the degree of atrophy of the vermis of anterior lobe of cerebellum in the same way. We analyzed the correlation of age at onset and the degree of atrophy with the CAG repeat length. Furthermore, we calculated the quotient of the degree of atrophy in each anatomic structure divided by age at the time of MRI examination and by duration from the age at onset, and compared these data to the CAG repeat length.

The relations of the age at onset, the degree of atrophy, the age-adjusted degree of atrophy, and the duration-adjusted degree of atrophy to the CAG repeat length were statistically analyzed using Pearson's correlation coefficient.

Informed consent for MRI studies and DNA diagnosis were established beforehand on all patients.

Results. The number of CAG repeats in these MJD patients ranged from 61 to 82, and the age at onset varied from 17 to 65 years. There was a high inverse correlation between the number of CAG repeats and the age at onset (r = -0.88, p < 0.0001,figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Correlation between age at onset and the CAG repeat number (Pearson's correlation coefficient, r = -0.88, p < 0.0001).

In the control cases, no brainstem or cerebellar atrophy was found by MRI, and there was no significant correlation between age and the size of the anatomic structures of concern. The ratios of the anteroposterior diameters of the pontine tegmentum and midbrain to the distance between the nasion and the inion in the controls were 0.030 ± 0.002 (mean ± SD) and 0.067 ± 0.003, respectively, and the ratios of the areas of the pontine base and the vermis of anterior lobe of cerebellum to the area of the posterior fossa in the controls were 0.123 ± 0.006 and 0.140 ± 0.014, respectively. The degree of atrophy in MJD patients was 0.011 ± 0.004 in the pontine tegmentum, 0.013 ± 0.007 in the midbrain, 0.033± 0.016 in the pontine base, and 0.033 ± 0.018 in the vermis of anterior lobe of cerebellum.

The quotient of the degree of atrophy of each anatomic structure in the brainstem and cerebellum divided by age at the time of examination was compared with CAG repeat length. The results of this analysis are shown infigure 2. There were significant correlations between the CAG repeat size and the age-adjusted degree of atrophy in the pontine tegmentum (r = 0.768, p < 0.001), midbrain(r = 0.641, p < 0.0001), pontine base (r= 0.501 p < 0.001), and vermis of anterior lobe of cerebellum(r = 0.448, p < 0.02). On the other hand, there were no significant correlations of either the duration-adjusted degree of atrophy or the degree of atrophy without correcting for age and duration with the CAG repeat size in any of the anatomic structures.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. The relation between the quotient of the degree of atrophy of each anatomic structure divided by age at the time of examination and the CAG repeat number. There were significant correlations between the CAG repeat number and the age-adjusted degree of atrophy in pontine tegmentum (r = 0.768, p < 0.0001) (A), midbrain (r = 0.641, p < 0.0001) (B), pontine base (r = 0.501, p < 0.001) (C), and vermis of anterior lobe of cerebellum (r = 0.448, p < 0.02) (D).

Discussion. Our results demonstrate that the quotient of the degree of the brainstem and cerebellar atrophy as assessed by MRI divided by age at examination is strongly correlated with the CAG repeat number of the MJD1 gene. In contrast, either the quotient of the degree of atrophy divided by duration from the onset of the disease or the degree of atrophy without correcting for age and duration was not correlated with the CAG repeat size. These findings suggest that the rate of development of atrophy in the brainstem and cerebellum, which are the most severely affected in MJD, strongly depends on the size of the CAG repeats. In other words, numerous CAG repeats induce faster atrophy of the brainstem and cerebellum. Furthermore, this progression of atrophy does not develop after the onset of disease, but is likely to develop even in the presymptomatic stage, maybe linearly from birth.

This type of strong relation between CAG repeat length and the quotient of phenotypes divided by age at examination is similar to our previous demonstration of a correlation between the clinical severity divided by age and CAG repeat size in spinal and bulbar muscular atrophy.^4,5 Similarly, in Huntington's disease, the rate of progression of clinical severity has been correlated with CAG repeat size,⁶ and Furtado et al. and Penney et al. recently have found that there is a high degree of correlation between the CAG repeat number and the quotient of the degree of cell loss in the striatum divided by age at death.^7,8 However, in MJD, there are few reports concerning the correlations between CAG repeat length and the rate of disease progression.⁹

Whether the disease progresses linearly from birth, or develops in a multistage process with a long latent phase and then a rapid progression of the dysfunctional stage, remains to be solved until sufficient data from presymptomatic patients have been accumulated.

Our current results combined with other reports on spinal and bulbar muscular atrophy^4,5 and Huntington's disease^6-8 indicate that CAG repeat length in the responsible gene strongly influences the rate of disease progression in CAG repeat diseases. This would be a common phenomenon among the trinucleotide repeat diseases.

Acknowledgments

This work was conducted in collaboration with Drs. N. Yuasa and S. Terao(Aichi Medical University); Dr. N. Sakurai (Ichinomiya Municipal Hospital); Dr. Y. Watanabe (Ohgaki Municipal Hospital); Dr. M. Uchida (Tousei General Hospital); Dr. S. Ito (Komaki Municipal Hospital); Dr. T. Takegami (National Nagoya Hospital); Dr. A. Yamashita (National Higashi Nagoya Hospital); Dr. K. Yasui (Tokai Kinen Hospital); and Dr. T. Yanagi (Nagoya Daini Red Cross Hospital). The authors are grateful to these doctors for providing samples and clinical information.