The prion protein M129V polymorphism

Abstract

The PRNP gene encodes the cellular isoform of prion protein (PrP^c). The M129V polymorphism influences the risk of prion diseases and may modulate the rate of neurodegeneration with age. We present the first study of the polymorphism among Polish centenarians. In the control group (n = 165, ages 18 to 56 years) the observed M129V genotype frequencies agreed with those expected according to the Hardy-Weinberg equilibrium (MM, MV, VV): 43%, 44%, 13% (HWE p > 0.05). Among centenarians (n = 150, ages 100 to 107) both homozygotes were more common than expected and HWE was rejected: 46%, 37%, 17% (expected 42%, 46%, 13%; HWE p = 0.025). This finding is consistent with a higher mortality rate among heterozygotes. However, the observed allele and genotype frequencies did not differ significantly between the oldest-old and the young controls. The genotypic frequencies were not related to severe cognitive impairment among the centenarians.

Keywords: :

• PRNP
• longevity
• codon 129 polymorphism
• cognitive impairment
• centenarians
• M129V
• selective mortality
• genetic epidemiology

The PRNP gene encodes the cellular isoform of the prion protein (PrP^c: normal conformation; PrP^Sc: misfolded). There is ample reason to consider inherited variation in the protein as relevant to prion disease: Experimental data implicate PrP^c in anti-apoptotic activities, cell signaling & adhesion, synaptic function and protection against oxidative stress (for a review, see ref. 1), although exact physiological mechanisms for normal and altered forms remain elusive. Clearly, misfolding to PrP^Sc plays a central causal role in the pathogenesis of Creutzfeldt-Jakob disease (CJD) and other spongiform encephalopathies.² Rare inherited mutations in PRNP cause familial forms of prion diseases. The common M129V methionine to valine polymorphism at codon 129 influences the risks of sporadic and acquired iatrogenic CJD: the homozygous MM and VV genotypes are both over-represented as compared to the heterozygous MV genotype.^3,4 All pathologically confirmed cases of variant CJD reported to date are homozygous MM.⁵

Beyond prion diseases, the M129V polymorphism appears to constitute a part of the genetic background for other neurodegenerative disorders. The VV genotype has been associated with cognitive impairment (MMSE score <24) and the V allele has been associated with greater brain amyloid deposition.^6,7 The codon 129 polymorphism was proposed as a risk factor for Alzheimer disease (AD) (MM conferring elevated risk), mild temporal lobe epilepsy and primary progressive aphasia.^8-11 Its influence on survival in AD, clinical presentation of idiopathic Parkinson disease and clinical course of neurological Wilson disease has also been reported.^12-14 This raises the possibility of selective mortality depending on codon 129 genotype. Moreover, in healthy individuals a link between the M129V polymorphism and long-term memory and cognitive performance has been described.^15,16 However, not all studies find associations.^17-20

These reported studies concern middle-aged individuals or elderly persons at most in their 80s or 90s, not the oldest-old, centenarians, who have selectively survived, avoided, fatal conditions. In this paper we present the first genetic association study of the PRNP codon 129 polymorphism among Polish centenarians (n = 150) comparing genotypic frequencies to those found for a sample of younger Poles aged 18 to 56 years (n = 165). Among the centenarians, genotype frequencies are contrasted for subjects with (n = 82) and without severe cognitive impairment (n = 33).

We found that the genotypic frequencies in the control group of younger Poles were expectedly consistent with Hardy-Weinberg equilibrium: MM: 43%; MV: 44%; VV: 13% (HWE p > 0.05). These frequencies are in the range of other reported results. The distribution of the codon 129 genotypes found in different European populations was shown to vary between 35−52% for MM, 39−56% for MV and 9−17% for VV genotypes, placing Poland in the middle of these ranges.^7,21-24

Unlike the control subjects, in our centenarian population both homozygous genotypes were found more frequently than expected (MM, MV, VV): 46%, 37%, 17% (expected 42%, 46%, 13%), i.e., HWE was rejected (p = 0.025). This is consistent with selectively better survival among persons who carry the homozygous genotypes to the age of 100 years.

Using the more general hypothesis of unspecified association, directly comparing genotypic frequencies for the two groups of subjects did not provide evidence that centenarians differed from younger Poles in terms of codon 129 polymorphism (Table 1A).

Table 1. Comparison of PRNP codon 129 polymorphism frequencies between analyzed groups.

Centenarians vs. young controls
	Centenarians n =150	Controls n=165	P
Genotype			0.472
MM	69 (46.0)	71 (43.0)
MV	56 (37.3)	72 (43.7)
VV	25 (16.7)	22 (13.3)
Allele			0.962
M	194 (64.7)	214 (64.8)
V	106 (35.3)	116 (35.2)
Centenarians with vs. without severe cognitive impairment (s.c.i.)
	Centenarians with s.c.i. n=82	Centenarians without s.c.i. n=33	P
Genotype			0.318
MM	38 (46.3)	15 (45.5)
MV	35 (42.7)	11 (33.3)
VV	9 (11.0)	7 (21.2)
Allele			0.420
M	111 (67.7)	41 (62.1)
V	53 (32.3)	25 (37.9)

Figures in brackets are percentages.

No trends were demonstrated for the number of M as opposed to V alleles (0, 1, 2) comparing the two groups (Cochran-Armitage test for trend, p = 0.999). This test has been recommended as useful in samples not consistent with HWE.²⁵

We then considered whether codon 129 polymorphism formed part of the genetic background of severe cognitive impairment among centenarians (Table 1B). At this sample size (82 subjects with and 33 without severe cognitive impairment), there was no statistically significant evidence that genotypic frequencies differed for the two groups. However, it is noteworthy that heterozygotes tended to be more common among the severely impaired (43% vs. 33%). No statistically significant trends in the number of M and V alleles were identified (data not shown).

Our results suggest selective survival related to the combination of PRNP codon 129 alleles that have been inherited: the observed deviation from HWE in the centenarian group may result from a higher mortality rate among heterozygotes. The centenarians were not related, so inbreeding as a potential source of the observed deviation from HWE can be excluded. Our finding, of course, needs to be verified in other samples. However, it is remarkable that any single common polymorphism demonstrates an effect on longevity. As single factors, APOE ε2 and the absence of APOE ε4 have previously been shown to influence average longevity, explaining perhaps 1% of lifespan.^26,27

On the other hand, in our study the allele and genotype frequencies do not differ significantly between the oldest-old and the young controls (general association). This suggests that the influence of M129V polymorphism on susceptibility for a spectrum of disorders is too weak or the diseases are too rare—as it is in the case of prion diseases—to result in elimination of particular genotype carriers from long-living individuals. Further studies in large groups are necessary to finally elucidate the influence of the PRNP polymorphism on the life-span.

Concerning neurodegeneration, the relevance of M versus V allele needs to be considered more carefully in terms of age, age at onset, duration of disease and method of case acquisition. The results published thus far have been divergent, pointing at VV^6,28,29 as well as MM genotype⁸ as a risk factor for cognitive impairment. The discrepancies may depend on the selection criteria applied for the analyzed groups. This study considered severe cognitive impairment regardless of its origin. These findings indicate that prevalence-incidence bias may operate for codon 129 variation and account for differing implications of reported studies. Given the findings of this study, larger studies might demonstrate that codon 129 variation continues to be a determinant of neurodegeneration and cognitive impairment at very advanced ages. This is in contrast to the situation for APOE where the ε4 allele is most relevant at relatively younger ages.²⁶

It was proposed that the oldest individuals may be divided into three groups: the “escapers” who reached the age of 100 free from typical age-related diseases, the “survivors” who live long despite suffering from age-associated diseases and the “delayers”, affected by age-related diseases with an unusually late age of onset.³⁰ The centenarians with cognitive impairment may belong to the two latter groups. Irrespective if they represent the delayers or the survivors, their molecular background of cognitive impairment may be different than in populations analyzed in average AD studies, i.e., usually 20−30 years younger than centenarians. A healthy survival past the age of 100 years may be influenced by a pattern of genetic as well as environmental factors which still remains to be unraveled.

Materials and Methods

We investigated 150 centenarians (130 women and 20 men) aged 100−107, mean age 101.1 ± 1.1 years, who participated in the multicenter scientific project: “Genetic and Environmental Factors of Longevity of Polish Centenarians” (PolStu), coordinated by the International Institute of Molecular and Cellular Biology in Warsaw. The individuals were not related. The recruitment of the participants and their general clinical evaluation are described by Mossakowska et al.³¹

In the present study we used data on cognitive performance extracted from the existing database. Cognitive status was known for 115 centenarians. In 82 individuals severe cognitive impairment had been defined according to the following criteria:

1. clinical diagnosis of dementia done by a neurologist with the use of the established international criteria: DSM IV,³² NINCDS/ADRDA,^33,34 ICD 10 ³⁴, Hachinski Ischemic Scale³⁵ or

2. the CDR score of 1 or above or

3. the MMSE score of 60% or below of a maximum possible score (if clinical diagnosis or CDR score was not available).

In centenarians the maximum MMSE score was established for every person separately, considering his/her age-associated disabilities, such as vision problems. The maximum possible scores differed between the analyzed individuals, so we used the percentages instead of absolute cut-off values.

The criteria of dementia—listed in points 1 and 2—were satisfied for 45 centenarians (35 AD patients, 1 vascular dementia and 9 cases of mixed dementia). The subgroup classified according to point 3 included 37 subjects.

A determination of no severe cognitive impairment was made for 33 centenarians known to have CDR score of 0 or 0.5 or MMSE score of 80% or above of a maximum possible score (when CDR not available). Thirty five centenarians remained unclassifiable. Among them, 19 subjects did not meet any of the above criteria, and in 16 cases, the cognitive status was not assessed due to several typical reasons: deafness, neither cognitive testing nor interviews were possible, there was refusal to participate in testing or the participant was otherwise uncooperative.

As controls, we tested DNA samples from 165 blood donors aged 18−56, mean age 27.8 ± 9.1 years; 91 women and 74 men, obtained from the Department of Paediatrics, Oncology, Haematology and Diabetology, Medical University of Lodz. These subjects are assumed to represent the same source population as the centenarians, but with little selection for mortality.

The study was performed according to accepted ethical standards and formal approval from the Human Subjects Review Board.

Genotype for the PRNP codon 129 was determined using an established PCR-RFLP method.³⁶ We investigated Hardy-Weinberg Equilibrium (HWE) for the centenarians and for the control subjects. Genotype frequency distributions, and allele frequencies, found for the two subject groups were contrasted using chi-square test. Additionally, Cochran-Armitage test for trend in allele dose (0, 1, 2 alleles) with the use of Monte Carlo method (performed using the XLStat software) was carried out. Statistical significance was assumed at P ≤ 0.05 (two-sided).

Acknowledgements

The study was supported by the State Committee for Scientific Research (KBN) grants: No. PBZ-KBN-022/PO5/1999 (collection of samples and clinical characterization of centenarians), No. 2P05 B08430 (genotyping) and funds of Medical University of Lodz, No 503/1-034-04/503-01. Prof. Shirley Lindenbaum (Graduate Center City University of New York) is kindly acknowledged for assistance with preparation of the manuscript.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.