Abstract
While little is known about risk factors for cognitive impairment in early onset Parkinson disease (EOPD), postmortem studies have shown an association between dementia with Lewy bodies (DLB) and glucocerebrosidase (GBA) mutation. We compared Mini-Mental State Examination (MMSE) performance and self-reported cognitive impairment in 699 EOPD participants genotyped for mutations in parkin (PRKN), leucine-rich repeat kinase-2 (LRRK2), and GBA. Logistic regression was used to assess the association between reported cognitive impairment and MMSE score, as well as between GBA group membership and self-reported impairment and MMSE. GBA carriers reported more impairment, but MMSE performance did not differ among genetic groups. Detailed neuropsychological testing is required to explore the association between cognitive impairment and GBA mutations.
Keywords: Parkin, Leucine-rich repeat kinase-2, Glucocerebrosidase, Parkinson, Cognition, Mini-Mental State Examination, Genetics
INTRODUCTION
Early onset Parkinson disease (EOPD) is defined as disease onset before age 40 or 50 years (Butterfield, Valanis, Spencer, Lindeman, & Nutt, 1993; Kostic et al., 1994; Schrag & Schott, 2006). It has been estimated that up to 10% of PD patients have an age at onset (AAO) of ≤40 years (Schrag, Ben-Shlomo, Brown, Marsden, & Quinn, 1998). EOPD is probably a multifactorial disease caused by an interaction of environmental and genetic factors. Overall, the vast majority of EOPD patients do not carry a known genetic risk factor associated with PD (Macedo et al., 2009), but specific genetic mutations have been identified in a subset of patients. While in rare familial cases mutations in alpha synuclein (SNCA), DJ-1, and PINK-1 may be found, mutations in parkin (PRKN), leucine-rich repeat kinase-2 (LRRK2), and glucocerebrosidase (GBA) are more common, especially in selected populations—for example, LRRK2 and GBA mutations in Ashkenazi Jews. Mutations in the PRKN gene are found in approximately 10% of EOPD patients and are inversely associated with AAO (Clark et al., 2007a). Mutations have been described in the homozygous, compound heterozygous, and heterozygous states, but the role of heterozygous mutations remains controversial.
Although previous studies have shown that dementia is uncommon in EOPD with AAO ≤50 (Kostic et al., 1994) the pattern and extent of cognitive changes in EOPD and associated risk factors have not been described. While mutations in SNCA are associated with dementia (Schrag & Schott, 2006) it is unknown whether mutations more commonly seen in EOPD, in PRKN, LRRK2, or GBA, convey a risk for cognitive impairment as well.
A family-based study comparing PRKN carriers (n = 21) to EOPD patients without mutations did not find differences in neuropsychological performance (Lohmann et al., 2009). To our knowledge, the cognitive performance of EOPD with LRRK2 or GBA mutations has not been reported in large studies; however, we and others have reported that GBA mutations are associated with dementia with Lewy bodies (DLB) at autopsy (Clark et al., 2009; Neumann et al., 2009) and therefore may represent a risk for cognitive impairment.
The current study assessed in the largest sample of nondemented EOPD cases reported to date whether mutations in the PRKN, LRRK2, and GBA genes are associated with self-reported cognitive impairment. We further compared performance across the mutation groups on the Mini-Mental State Exam (MMSE), a commonly used cognitive screening test in PD centers (Dubois et al., 2007).
METHOD
Participants
Cases with AAO ≤50 (n = 699) participating in the Consortium On Risk for Early-Onset Parkinson Disease (CORE-PD) study were recruited from 13 sites (Clark et al., 2007b). Institutional review boards at all participating sites approved the protocols and consent procedures. PD cases were required to have an MMSE score >23 so that they could accurately provide their own medical history and family history of PD in their relatives. Data collected on each case included demographic information, AAO, the Unified Parkinson’s Disease Rating Scale (UPDRS) completed by a movement disorders specialist (Fahn, Elton, & Committee, 1987), MMSE (Folstein, Folstein, & McHugh, 1975), family history of PD using a previously validated interview, and a blood sample from which DNA was isolated and examined for mutations in PRKN, LRRK2, and GBA. Examiners were trained by a certified neuropsychologist (E.C.) in administrating the MMSE and were unaware of cases’ genetic status. Cases were subsequently divided into five groups based on mutation status: (a) noncarriers of mutations of any of the three genes, (b) PRKN heterozygotes, (c) PRKN homozygotes and compound heterozygotes combined, (d) LRRK2 mutation carriers, and (e) GBA mutation carriers.
Molecular genetic analysis
Parkin
Cases were screened for PRKN mutations by denaturing high-performance liquid chromatography (DHPLC; WAVE Transgenomic), as previously described (Pigullo et al., 2004). Amplicons were either directly sequenced (n = 126) or analyzed (n = 573) using a PRKN genotyping array (Clark et al., 2007a) in DNA samples with abnormal DHPLC elution profiles. Cycle sequencing was performed on the purified PCR product as per the manufacturer’s instructions (BigDye, Applied Biosystems). Products were processed on an ABI3700 genetic analyzer. Chromatograms were viewed using Sequencher (Genecodes), and sequence variants were determined. To identify genomic deletions and exon rearrangements in PRKN, semiquantitative multiplex polymerase chain reaction (PCR) was performed as previously described (Clark et al., 2006a).
LRRK2 and GBA
DNA samples from the majority of the cases (n = 506) were analyzed using a previously described genotyping chip (Clark et al., 2007a; Asper Biotech, Tartu, Estonia). For the remaining 193 cases, LRRK2 genotyping was performed using matrix-assisted laser desorption/ionization time of flight (MALDI-TOF) mass spectrometry (Sequenom) as previously described (Clark et al., 2006b). All samples analyzed by either the MALDI-TOF or the genotyping chip were examined for the LRRK2 mutations G2019S, R1441C, I2020T, and Y1699C. The rest of the GBA samples were genotyped by direct sequencing. All samples were tested for the N370S and L444P GBA mutations.
Statistical analyses
Self-report of cognitive impairment, MMSE performance, and demographic and disease characteristics of the five mutation status groups were compared using analysis of variance (ANOVA), Kruskal–Wallis, and chi square tests as appropriate. The effect size of MMSE performance was calculated using the probabilistic index comparing noncarriers to each mutation group separately. A value of .5 indicates a 50% chance of a score coming from a mutation carrier group to be higher than a score from the noncarrier group (i.e., no difference between the groups). Values closer to 0 indicate greater likelihood that noncarriers scored better than mutation carriers (Acion, Peterson, Temple, & Arndt, 2006).
The association between cases’ report of cognitive impairment, defined as a response of two (“moderate memory loss with disorientation and moderate difficulty handling complex problems”) or greater on Item I of the UPDRS Part I and MMSE performance, was tested using a logistic regression model where report of cognitive impairment was the dependent variable and MMSE performance the independent covariate.
Finally, we used a logistic regression model to test the association between GBA mutation status (dependent) and self-report of cognitive impairment adjusting for Jewish descent (defined as all four grandparents reporting Jewish ancestry), MMSE score, and self-reported depression, defined as a response of two (“sustained depression for >1 week”) or more on Item III on the UPDRS Part I (Levy et al., 2002). Self-reported depression was added to the model given that depression may affect one’s awareness or insight into their cognitive performance,
RESULTS
A total of 103 cases carried a genetic mutation: A total of 23 were PRKN heterozygotes, 20 were PRKN homozygotes/compound heterozygotes, 20 were G2019S LRRK2 carriers (19 heterozygotes, one homozygote; no other pathogenic LRRK2 mutations were found), and 37 were GBA carriers (12 L444P heterozygotes, 24 N370S heterozygotes, and 1 N370S homozygote). A total of 3 cases carried mutations in more than one gene (one PRKN heterozygote-LRRK2 carrier, one PRKN heterozygote GBA carrier, and one LRRK2-GBA carrier) and were excluded from the analyses. A total of 7 cases were administered the MMSE in Spanish. The median MMSE scores, the effect size of MMSE performance (in each mutation group compared to noncarriers), and disease characteristics of the genetic groups are presented in Table 1.
Table 1.
MMSE performance and disease characteristics of 696 probands stratified by genetic group1
| Non- carriers (n=596) |
Parkin heterozygotes (n=23) |
Carriers of 2 parkin mutations (n=20) |
LRRK2 carriers (n=20) |
GBA carriers (n=37) |
Sig.2 | |
|---|---|---|---|---|---|---|
| Self report of cognitive impairment |
4.2% (22)a |
10.0% (2)a | 0a | 0a | 21.9% (7)b |
<0.001 |
| Median MMSE score |
30 | 29 | 29 | 30 | 30 | NS |
| MMSE performance effect size3 |
0.40 | 0.40 | 0.42 | 0.50 | ||
| Age (years) | 52.0 (8.6)a |
48.8 (8.0)a,b | 46.1 (13.4)b | 54.9 (7.9)a |
54.4 (4.9)a |
<0.001 |
| AAO (years) | 41.9 (6.3)a |
38.9 (9.2)a | 28.7 (9.3)b | 43.6 (4.8)a |
42.3 (6.8)a |
<0.001 |
| Disease duration |
10.2 (7.3)a |
9.9 (9.4)a | 17.7 (13.0)b | 11.3 (7.6)a |
12.1 (7.3)a |
<0.001 |
| UPDRS-III | 19.9 (11.4) |
21.6 (12.9) | 22.6 (13.0) | 19.8 (14.7) |
23.8 (11.3) |
NS |
| Jewish ancestry |
7.4% (44) |
0 | 0 | 65.0% (13) |
35.1% (13) |
<0.001 |
| Levodopa daily dose (mg) |
469 (470) |
276 (334) | 667 (911) | 420 (407) |
559 (347) |
NS |
| Education (years) |
15.6 (2.8) |
14.8 (3.5) | 13.9 (2.8) | 16.0 (3.2) |
15.7(3.0) | NS |
Values are means and standard deviations (in parentheses), except for MMSE (median) and Jewish ancestry where values are in percent and numbers (in parentheses). Values with different superscript letters differ significantly on post-hoc testing
Excluding 3 probands with more than one genetic mutation
Statistical significance was calculated using ANOVA, except for MMSE performance (Kruskal Wallis), and Jewish ancestry (chi square).
Effect size represents the likelihood of members of the genetic group to obtain a higher MMSE score than the non-carrier group
The UPDRS Part I was available for 617 cases. A larger proportion of GBA carriers reported cognitive impairment (score of 2 or more) than all other genetic groups (21.9%, 7/32, p < .001). Subjective cognitive impairment on the UPDRS Item I, which was reported by 5% of cases (n = 31) was associated with a lower score on the MMSE. For each 1-point decline on the MMSE the odds of reporting cognitive impairment increased by 38% (odds ratio, OR = 0.62; 95% confidence interval, CI: 0.51–0.76, p < .001, logistic regression).
A total of 57% (396/699) of participants obtained a score of 30 on the MMSE. The mean score was 29.2, the median 30. Mistakes were most often made on the long-term memory items (19.1% of all cases made at least one error in recall), working memory (10.3% of all cases made at least one error), naming the county (8.9%), and visuospatial orientation (8.6% made an error copying intersecting pentagons). There were no significant differences among the genetic groups in overall MMSE performance (Table 1) or on individual item performance.
In a logistic regression model (Table 2) including GBA carriers and noncarriers, report of cognitive impairment on the UPDRS-I and Jewish ancestry predicted GBA mutation group membership after adjusting for reported depression, but MMSE performance did not.
Table 2.
Logistic regression model predicting GBA mutation group membership versus non-carriers including 560 cases
| Variable | OR | 95% confidence interval | p value |
|---|---|---|---|
| Self reported cognitive impairment | 5.9 | 2.0–17.8 | 0.001 |
| MMSE performance | 1.05 | 0.77–1.42 | 0.78 |
| Self reported depression | 0.46 | 0.1–2.2 | 0.33 |
| Jewish Ancestry | 7.9 | 3.4–18.2 | <0.001 |
DISCUSSION
This is the largest study reported to date assessing self-reported cognitive impairment and MMSE performance in EOPD. Individuals with EOPD are usually highly functional—for example, working and driving; consequently the identification of risk factors for cognitive impairment in this group is vital (Schrag et al., 1998). Here, we did not find an association between MMSE performance and the most frequent genetic mutations found in EOPD; however, GBA carriers’ self-report of more cognitive complaints than any other mutation carrier group or noncarriers may be viewed, in the context of postmortem studies suggesting an association between GBA mutations and DLB (Clark et al., 2009; Neumann et al., 2009), as a possible early marker for cognitive decline.
The strengths of our study lie in its large size, which is important given that most EOPD individuals are not mutation carriers. Limitations include the tools used to assess cognitive dysfunction and self-report cognitive impairment. While the MMSE is perhaps the most commonly used screen for cognitive decline, its lack of sensitivity is well documented. Our novel use of Item I of the UPDRS questionnaire as a measure of subjective cognitive changes may also represent a weakness in the study; however, subjective memory complaint assessed via a single question (“Do you have complaints about your memory?”) has been found to be predictive of incident Alzheimer’s disease with an average follow-up of 3.2 years in healthy elders in whom no cognitive impairment was identified (Geerlings, Jonker, Bouter, Ader, & Schmand, 1999). Furthermore, self-report of cognitive impairment was associated with lower performance on the MMSE, demonstrating the validity of the UPDRS questionnaire. Finally, the participants included likely represent a restricted sample in that participants scoring < 24 on the MMSE were excluded.
We have shown that self-reported cognitive impairment is associated with GBA mutations and with lower performance on the MMSE. MMSE performance was not significantly different among mutation groups, either because there are no cognitive differences among the mutation groups, or because the MMSE was not sensitive enough to detect the GBA carriers’ cognitive impairment. Future studies utilizing comprehensive neuropsychological assessment measures would more fully characterize the nature of these cognitive complaints.
Acknowledgments
This study was funded by NIH NS36630, UL1 RR024156 (K.M.), NS050487, NS060113 (L.N.C.), and the Parkinson’s Disease Foundation (K.M., S.F., and L.N.C.). RNA was supported by a Parkinson’s Disease Foundation H. Houston Merritt Fellowship in Movement Disorders. The authors thank Adam Brickman, Paul Greene, and Diana Ruiz for their assistance.
Footnotes
Disclosure: The authors report no conflicts of interest.
REFERENCES
- Acion L, Peterson JJ, Temple S, Arndt S. Probabilistic index: An intuitive non-parametric approach to measuring the size of treatment effects. Statistics in Medicine. 2006;25(4):591–602. doi: 10.1002/sim.2256. [DOI] [PubMed] [Google Scholar]
- Butterfield PG, Valanis BG, Spencer PS, Lindeman CA, Nutt JG. Environmental antecedents of young-onset Parkinson’s disease. Neurology. 1993;43(6):1150–1158. doi: 10.1212/wnl.43.6.1150. [DOI] [PubMed] [Google Scholar]
- Clark LN, Afridi S, Karlins E, Wang Y, Mejia-Santana H, Harris J, et al. Case-control study of the parkin gene in early-onset Parkinson disease. Archives of Neurology. 2006a;63(4):548–552. doi: 10.1001/archneur.63.4.548. [DOI] [PubMed] [Google Scholar]
- Clark LN, Haamer E, Mejia-Santana H, Harris J, Lesage S, Durr A, et al. Construction and validation of a Parkinson’s disease mutation genotyping array for the Parkin gene. Movement Disorders. 2007a;22(7):932–937. doi: 10.1002/mds.21419. [DOI] [PubMed] [Google Scholar]
- Clark LN, Kartsaklis LA, Wolf Gilbert R, Dorado B, Ross BM, Kisselev S, et al. Association of glucocerebrosidase mutations with dementia with Lewy bodies. Archives of Neurology. 2009;66(5):578–583. doi: 10.1001/archneurol.2009.54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Clark LN, Ross BM, Wang Y, Mejia-Santana H, Harris J, Louis ED, et al. Mutations in the glucocerebrosidase gene are associated with early-onset Parkinson disease. Neurology. 2007b;69(12):1270–1277. doi: 10.1212/01.wnl.0000276989.17578.02. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Clark LN, Wang Y, Karlins E, Saito L, Mejia-Santana H, Harris J, et al. Frequency of LRRK2 mutations in early- and late-onset Parkinson disease. Neurology. 2006b;67(10):1786–1791. doi: 10.1212/01.wnl.0000244345.49809.36. [DOI] [PubMed] [Google Scholar]
- Dubois B, Burn D, Goetz C, Aarsland D, Brown RG, Broe GA, et al. Diagnostic procedures for Parkinson’s disease dementia: Recommendations from the movement disorder society task force. Movement Disorders. 2007;22(16):2314–2324. doi: 10.1002/mds.21844. [DOI] [PubMed] [Google Scholar]
- Fahn S, Elton R . Committee, M. o. t. U. D. The Unified Parkinson’s Disease Rating Scale. In: Fahn S, Marsden CD, Calne DB, Goldstein M, editors. Recent developments in Parkinson’s disease. Vol. 2. Florham Park, NJ: Macmillan Healthcare Information; 1987. pp. 153–163.pp. 293–304. [Google Scholar]
- Folstein MF, Folstein SE, McHugh PR. “Mini-mental state.” A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research. 1975;12(3):189–198. doi: 10.1016/0022-3956(75)90026-6. [DOI] [PubMed] [Google Scholar]
- Geerlings MI, Jonker C, Bouter LM, Ader HJ, Schmand B. Association between memory complaints and incident Alzheimer’s disease in elderly people with normal baseline cognition. American Journal of Psychiatry. 1999;156(4):531–537. doi: 10.1176/ajp.156.4.531. [DOI] [PubMed] [Google Scholar]
- Kostic VS, Filipovic SR, Lecic D, Momcilovic D, Sokic D, Sternic N. Effect of age at onset on frequency of depression in Parkinson’s disease. Journal of Neurology, Neurosurgery, and Psychiatry. 1994;57(10):1265–1267. doi: 10.1136/jnnp.57.10.1265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Levy G, Tang MX, Louis ED, Cote LJ, Alfaro B, Mejia H, et al. The association of incident dementia with mortality in PD. Neurology. 2002;59(11):1708–1713. doi: 10.1212/01.wnl.0000036610.36834.e0. [DOI] [PubMed] [Google Scholar]
- Lohmann E, Thobois S, Lesage S, Broussolle E, du Montcel ST, Ribeiro MJ, et al. A multidisciplinary study of patients with early-onset PD with and without parkin mutations. Neurology. 2009;72(2):110–116. doi: 10.1212/01.wnl.0000327098.86861.d4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Macedo MG, Verbaan D, Fang Y, van Rooden SM, Visser M, Anar B, et al. Genotypic and phenotypic characteristics of Dutch patients with early onset Parkinson’s disease. Movement Disorders. 2009;24(2):196–203. doi: 10.1002/mds.22287. [DOI] [PubMed] [Google Scholar]
- Neumann J, Bras J, Deas E, O’Sullivan SS, Parkkinen L, Lachmann RH, et al. Glucocerebrosidase mutations in clinical and pathologically proven Parkinson’s disease. Brain. 2009 doi: 10.1093/brain/awp044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pigullo S, De Luca A, Barone P, Marchese R, Bellone E, Colosimo A, et al. Mutational analysis of parkin gene by denaturing high-performance liquid chromatography (DHPLC) in essential tremor. Parkinsonism and Related Disorders. 2004;10(6):357–362. doi: 10.1016/j.parkreldis.2004.04.012. [DOI] [PubMed] [Google Scholar]
- Schrag A, Ben-Shlomo Y, Brown R, Marsden CD, Quinn N. Young-onset Parkinson’s disease revisited—clinical features, natural history, and mortality. Movement Disorders. 1998;13(6):885–894. doi: 10.1002/mds.870130605. [DOI] [PubMed] [Google Scholar]
- Schrag A, Schott JM. Epidemiological, clinical, and genetic characteristics of early-onset parkinsonism. Lancet Neurology. 2006;5(4):355–363. doi: 10.1016/S1474-4422(06)70411-2. [DOI] [PubMed] [Google Scholar]