close

Вход

Забыли?

вход по аккаунту

?

Clinical genetic analysis of Parkinson's disease in the contursi kindred.

код для вставкиСкачать
C h i d Genetic Analysis of Parkinson’s Disease
in the Contursi h d r e d
Lawrence I. Golbe, MD,* Giuseppe Di Iorio, MD,? Giuseppe Sanges, MD,? Alice M. Lazzarini, PhD,*
Salvatore La Sala, MD,$ Vincenzo Bonavita, MD,? and Roger C. Duvoisin, MD*
~~
~~
We performed a clinical genetic analysis of a kindred originating in the town of Contursi in Salerno province, Italy,
in which 60 individuals in 5 generations are known to have had Parkinson’s disease (PD). Two previously reported
autopsy cases showed typical PD with Lewy bodies. The inheritance pattern is apparently autosomally dominant with
a segregation ratio of 40.1% for kindred members aged 50 years and older. The mean age at PD onset is 45.6 years
(standard deviation, 13.48; range, 20-85) with a mean course to death of 9.2 years (standard deviation, 4.87; range,
2-20). Otherwise, clinical characteristics of PD in the kindred, including variance in onset age and incidence of tremor
and levodopa responsiveness, are similar to those of PD in the community. The presence of tremor tended to be
concordant in a e c t e d parent-child pairs, but there was no parent-child correlation for age at onset or intrasibship
clustering of tremor or onset age. A suggestion of anticipation disappeared after adjustment for age-related ascertainment
bias. The findings show that a presumably single mutation can produce a heterogeneous PD phenotype, even among
siblings. This is consistent with the hypothesis that PD in the community may in fact be caused by such a mutation,
but one producing a lower penetrance and older age at onset than those in this kindred.
Golbe LI, Di Iorio G, Sanges G, Lazzarini AM, La Sala S, Bonavita V, Duvoisin RC. Clinical genetic
analysis of Parkinson’s disease in the Contursi kindred. Ann Neurol 1996;40:767-775
The cause of Parkinson’s disease (PD) remains unknown, but the evidence increasingly incriminates genetic factors (1, 21. Three recent controlled univariate surveys reported increases of 4-fold, 6.5-fold, and
9.5-fold in the frequency of a family history of PD
among patients relative to that among control subjects
[MI. Case-control studies using multivariate methods
showed a family history of PD to be the most important independent contributor to PD risk [6-81. Reanalysis of a large twin study [9, 101 that had been
initially interpreted as providing evidence against a hereditary cause for PD found it to be inconclusive but
compatible with a heritability as high as 0.78 [ I l l .
Analyses of multiple small familial clusters showed autosomally dominant inheritance to be the most likely
explanation of familial aggregation of PD [ 12- 141.
Another approach to understanding the role of genetics in the cause of PD is to compare PD in large,
multiply affected kindreds to “sporadic” PD. Several
large families are known in which PD was confirmed
by autopsy showing neuronal loss and Lewy bodies in
a distribution consistent with PD [ 15-23]. Such families, in which a genetic mechanism of PD transmission
seems certain, may serve as a convenient “experiment
of nature” or genetic model of PD.
~
~
The Contursi kindred was the first large family with
PD to be reported with autopsy confirmation [15]. It
originated in the town of Contursi, a hill town with a
population of about 5,000 in the Salerno province of
the south-central region of Italy. Many of the kindred
members immigrated to New Jersey, where they were
initially recognized as members of a multiplex kindred.
We previously reported that autopsy of 2 affected individuals, an uncle and nephew, showed Lewy bodies
and other changes typical of PD [15]. Neurofibrillary
tangles, striatal damage, and other features atypical for
PD were absent.
One purpose of this report is to reevaluate critically
our assumption that PD in this familial aggregate is in
fact autosomally dominant rather than environmental
in origin. If that assumption is supported, we will compare the variance in its clinical phenotype with that of
PD in the community. This will help support or refute
the hypothesis that the clinical phenotype of PD in
the community is consistent with a primarily genetic
etiology, albeit with low penetrance.
We also sought to assess the presence of anticipation,
a tendency seen in some hereditary neurodegenerative
conditions for affected children to have the disease at
an earlier age of onset than their affected parent. This
~
From the *Department of Neurology, University of Medicine and
Dentistry of New Jersey-Robert Wood Johnson Medical School,
New Brunswick, NJ; t Isticut0 di Scienze Neurologiche, Seconds
Universiti degli Studi di Napoli, Naples, Italy; and SContursi, Salerno Province, Italy.
~~
Received Feb 12, 1996, and in revised form May 23. Accepted for
publication May 23, 1996.
Address correspondence to Dr Golbe, Deparrment of Neurology,
UMDNJ-Robert Wood Johnson
CN-19, New
Brunswick, NJ 08303.
Copyright 0 1996 by the American Neurological Association 767
phenomenon was reported in smaller kindreds with PD
[24, 251 a n d in a preliminary statistical analysis of t h e
Contursi kindred reported in abstract form [26].
Materials and Methods
Data were gathered from family interviews (corroborated by
interviewing multiple family members whenever possible);
medical records: church and civic records of births, deaths,
and marriages in Contursi: review of family videotapes (2
subjects), and personal examination by one or more of us
(L. I. G., G. D. I., G. S., or R. C. D.). Seventeen of the
60 affected members were alive a t some time during this
study. All were of Generation VIII, IX, or X (Fig 1). Thirin the United States and 3
teen were examined by us-10
in Italy. O f the 4 who were not examined by us, 1 was
examined by other neurologists in New Jersey, 1 was diagnosed via a family videotape shown to us, and 2 (1 each in
Argentina and Germany) were diagnosed only through camily reports.
Criteria for a diagnosis of PD were two of the four
cardinal signs of rest tremor, bradykinesia, rigidity, and postural instability. For deceased individuals without medical
records, we accepted a report of a physician’s diagnosis of
PI) or a convincing report of relatives’ observations of the
cardinal signs. In assessing the presence of tremor, we accepted family reports as positive only if those who claimed
to remember the individual reached general agreement that
there was tremor.
We calculated segregation ratios by gender and by country
of origin. (The segregation ratio is the percentage of persons
at risk who are affected. “At risk” for an autosomally dominant disease is defined as having an affected parent or sibling.
For a hypothetical, fully penetrant, autosomally dominant
disorder, the segregation ratio is 50%.) Many of the assessments required an estimate of age at PD onset. We defined
it as the age at the initial appearance of symptoms that in
retrospect proved to be part of PD.
O u r test for anticipation was a statistically significant difference (>O) between the means for the parental and offspring generation. The anticipation interval is the parent’s
age at onset minus the child’s age at onset.
Some assessments, as specifically noted below, were repeated, including individuals with calculated (as opposed to
known) age at onset. We knew the onset age for 48 of the
60 affected individuals. For 8 of the other 12, we knew the
age at death but not the age at PD onset. For these, we
calculated the onset age by subtracting from the age at death
9 years, the approximate mean survival time afier symptom
onset in those for whom the appropriate dara were available.
For 4 affected individuals, we knew neither the onset age
nor the age at death. We did not attempt to include these
4 in any calculations involving onset age.
In addition. we knew the ages at death of 6 unaffected
obligate heterozygotes-46,
58, 69, 75, 85, and 94 years.
(Such individuals are those at risk who had affected offspring
but who died without evidence of PD in themselves, their
spouse, or their spouse’s parents or siblings.) W e did not
count them among the 60 affected individuals, but we did
include them i n some of our calculations of anticipation. For
that purpose, we arbitrarily considered their onset ages to be
768 Annals of Neurology
Vol 40
No 5
November 1996
r=
Fig 1. Contimi kindred pedigree as of Jnnuay I , 1996. The
2 individi~iilsin Generation I are the earliest known common
ancestors of alf 60 nffected individuals. Their years of birth
are unknown, but their 2 known children (Generation II)
were born in 1722 and 1726. There are no medical data for
Generations I through V. Generations X through H I lire not
shown. Generation X included 3 affected individuals. For
Generations VI through IX, open symbols represent absence
of parkinsonism by histoy or examination. The mean cIge of
individuals in Generation IX is approximately 50, suggesting
that that generation hm realized about half of its PD risk to
date.
identical to their ages at death, assuming that the probability
that they would have developed PD had they lived longer is
similar to the probability that we would have detected diagnostic signs of PD had we examined them antemortem.
For some analyses, we used only individuals in the two
most completely studied and ascertained generations, VII
and VIII (see Fig 1). We believed that the reliability of data
such as onset ages for earlier generations was less than optimal. Including Generations IX and younger, which have incompletely realized PD risk, would have downwardly biased
the apparent onset age of the affected members of those generations.
Table 1. Segregation Rdtios in the Co?iti~rsi
Kindred
Affected
Affected and
2 50 yr oldd or
deceased
At risk and
2 50 yr oldh
Segregation ratioL
~
Men
Women
Total
39
38
21
21
60
59
86
56
142
44.2%
37.5%
4 1.6%
~
’Difference from null hypothesis (male-famale ratio = 53:47 [ L ] ) :
= 0.28.
Note that at the time. 1 affected man was younger
than 50 and nor considered in this calculation.
bDifference from null hypothesis (male-female ratio = 1 : 1): p =
0.035, x”.At-risk individuals are those with an affected parenr and/
or sibling and, for purposes of this tabulation, age 50 or older.
‘Difference from null hypothesis (male = female): p = 0.54. x’.
p
x’.
For statistical analysis we used x’, paired and unpaired
Student’s t test, analysis of variance (ANOVA), Fisher’s exact
test, and Pearson correlation coefficients as appropriate.
cades past or in the historically impoverished area of
Italy where Contursi is located. This hypothesis is not
supported by the data, which are derived from the 45
deceased affected individuals for whom survival duration was known or calculated. Those who died before
the antibiotic era (i.e., before 1946) actually lived
longer (12.4 t 4.44 years, N = 9) than those who
died in later years (8.5 t 4.29 years, N = 36; p =
0.019). Those who died in Italy lived longer (10.4 2
4.58 years, N = 23) than those dying in the United
States (8.1 2 4.29 years, N = 22) but the difference
was not significant ( p = 0.078).
Table 2 shows some clinical data for 5 of our beststudied patients, all of whom were alive at the time of
writing. I t illustrates the range of clinical PD phenotype in the kindred. The response to levodopa was USLIally excellent at first, but in 1 patient, the drug caused
intolerable agitation. Many patients experienced doserelated wearing-off and a few developed choreiform
dyskinesias. Dementia of a highly variable sevcrity developed eventually in many patients. A few exhibited
mild depression.
Results
General Observations
Mechanism of Familial Clustering
While our original 1990 report [ I S ] was in press, we
confirmed our tentative conclusion that the two US
kindreds it described (both having immigrated from
Contursi) were related. T h e 41 persons then known to
be affected in 4 generations have now expanded to 60
in 5 generations (see Fig 1 ) .
All of the known affected individuals are descendants of a couple born in Contursi in approximately
1700, but medical data are available only for the most
recent 6 of the 12 known generations. PD was observed in Generations VI through X. Generation X, as
of January 1, 1996, included 2 10 individuals, of whom
3 were known to be affected and 26 were at risk. (The
number at risk will surely increase as more of their
parents [Generation 1x1 reach the age of maximum
risk and display signs of PD.) In addition, we know
of 45 members of Generation XI, of whom none was
affected and at least 2 were married. W e know of 1
member of Generation XI.
The illness in the Contursi kindred is atypical only
for its young average age at onset (Table I ) , 45.6 years
(standard deviation [ S D ] , 13.48; range, 20-85) and
for its relatively rapid mean duration from onset to
death of 9.2 years (SD, 4.87; range, 2-20, N = 39
for whom the ages at onset and death are known).
These compare to a mean onset age of 60 and a typical
survival time of 14 years in PD in the community [27].
T h e mean survival time among patients with PD in
the community whose onset occurs at age 45 is approximately 20 years [27].
We tested the hypothesis that the short mean survival time may be driven by poor medical care in de-
Informal inspection of the pedigree (see Fig 1 ) reveals
many instances of male-to-male transmission, ruling
out mitochondria1 inheritance as a mechanism. Sexlinked inheritance is ruled out by the identical segregation ratios for the sexes (see Gender below).
Autosomal<yrecessive inheritance with full penetrance
would produce a segregation ratio of 25%, not the
40.1% in the kindred overall or the 50% approached
in its better-studied branches [26] (see Table 1). Consanguinity, represented by a double horizontal line in
the pedigree (see Fig I), existed in three marriages that
involved a presumptive heterozygote and produced affected offspring. However, the 50% segregation ratio
among those offspring, 5 affected of 10 at risk, does
not differ significantly from that of the remainder of
the kindred, where the proportion was 44 (> age 50)
affected of 132 (> age 50) at risk (33%) ( p = 0.47,
Table 3). Any effect of consanguinity would be
confined to the offspring of the consanguineous marriages.
In the two best-ascertained generations (VII and
VIII, see Fig l ) , the segregation ratio among parents
of affected individuals (14/44,32%) is not statistically
different from that among their siblings ( I 5/34, 44%;
p = 0.38, x‘). This is the result predicted by an autosomal-dominant model but not by an autosomalrecessive model, which predicts a parental ratio far less
than the sibling ratio, which itself would be 25% assuming full penetrance.
An environmental cause with pseudodominant farnilial clustering must be considered. The segregation ratio
among kindred members over age 50 from sibships in
x’,
Golbe et al: PD i n the Contursi Kindred
769
Table 2. Illustrative Cases
Patient No.a
Sex
Onset age (yr)
Initial symptom
Rest tremor
Initial levodopa
response
Levodopa cornplications
50
48
55
56
51
M
M
40
RUE held flexed
F
42
Dragging FUE
F
55
General slowing
+
-
-
+
M
38
Postural tremor of
LUE
Excellent
Fair
Excellent
Excellent
DRWO; no AIMs
DRWO; chorea
mild at 2 yr,
DKWO; chorea mild
None
42
Action tremor of RUE
at
4 yr
severe at 8 yr
Cognitive data
MMSE = 29 at 6 yr;
HAM D = 5 at
10 yr
Unusual features
‘These
5
Unpredictable “on/
off”
75
MMSE = 23 at
12 yr; HAM
D=Oat
15 yr
Hallucinations, delusions at 7 yr; severe dementia at
9 Yr
NCSE
8 Y’
Mild vertical su-
Orthostatic hypotension requiring fludrocortisone
Fluent aphasia
pranuclear
gaze palsy
=
at
Intolerable agitation at low dosages
MMSE = 28 at
7 yr, 4 at 15
yr; moderate depression (“It’s
not worth it.”)
at 2 yr
Palilalia
patients are alive. Patient numbers tefer to Figure 1. Case descriptions of 2 autopsied patients appear in [15]
RLE = right lower extremity; RUE = right upper extremity; D R W O = dose-related wearing-offi AIMs = abnormal involuntary movements;
MMSE = Mini-Mental Status Examination score (range, 0-30; normal, 23-30); Ham D = Hamilton Depression Scale (range, 0 [normall52); NCSE = Neurobehavioral Cognitive Status Examination score (The Northern California Neurobehavioral Group, Inc, Fairfax, CA)
(range, 23-86; normal, 68-86).
Table 3. Efect of Parental Consanguinity on Ofjpring
Segregation Ratio in the Contursi Kindred
Parents consanguineous”
Parents nonconsanguineous
Offspring
Affected
Offspring
Unaffected
5
44
5
88
p
=
0.47, xz
’Three consanguineous marriages (third cousins in 2, second cousins
in 1). These data apply only to offspring over age 50. See text for
explanation of results.
whom full data are available and who spent their entire
lives in Italy was 15 (27%) affected of 55 at risk.
Among such persons who emigrated from Italy before
PD onset or who were born outside of Italy, the proportion was 27 (39%) of 69. This difference was not
significant ( p = 0.23, x’). Furthermore, a 1995 telephone survey of all five physicians in Contursi failed
to elicit a report of any patient in their practices who
had PD but was not a member of the kindred. The
sole possible exceptions were 2 elderly brothers in Contursi with long-standing action/ postural tremor whose
surname is shared with some affected members of the
kindred. Examination of 1 of the brothers by three of
us (L. I. G., G. D. I., and G. S.) revealed no other
770 Annals of Neurology Vol 40
No 5
November 1996
signs of PD and our genealogical investigations were
unable to delineate any connection between the parkinsonian kindred members and these 2 brothers. W e conclude that they had familial essential tremor unrelated
to the PD in the Contursi kindred.
Of the 60 known living and deceased affected individuals, 26 were born in Italy and did not emigrate, 6
were born in Italy and emigrated to the United States
before PD onset, 3 were born in Italy and emigrated
elsewhere before PD onset (1 to Germany, 2 to Argentina), and 25 were born in the United States and remained there.
These geographic observations strongly suggest that
an environmental influence peculiar to Contursi cannot explain this familial cluster.
Gender
The 60 known affected individuals comprised 39 men
(65%) and 21 women (see Table 1). This gender ratio
does not differ statistically from the male-female ratio
of 53 :47 from a study of PD mortalicy covering all of
Italy [28] ( p = 0.23, x’) or from the approximately
3 :2 ratio for living patients with a younger age at onset
referred to a typical tertiary movement disorder referral
center ( p = 0.71, x’) [29]. We calculated the segregation ratio, a more rigorous approach to the question
of genetic risk, separately for men and for women (see
7able 4. PD Onset Age in the Contursi k'iTidred
Known ~iffecre@
(as of Januaiy I , 1996)
Onset age (vr):
Known oiisct nges only
N
Mean
Standard deviarion
Known and calculated
onset ages
N
Men
Wonieii
'l'otal
p Value
39
21
60
0.14
(x')
-nr
2s
$
20
U
.c
8
15
cn
J
31
46.8
17
43.4
48
45.6
K
10
A
8
0.42 (unpaired
2
5
Q)
13.56
13.86
13.61
t test)
n
20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74
75+
PD Onset Age
Mcan
35
46.0
21
44.6
56
45.5
Standard dcviation
13.54
14.23
13.69
I
Contursi kindred,
known onset age
0.70 (unpaired
t rest)
'See Table 1 for sex-specific segregarion ratios.
Table 1). T o avoid obscuring the calculations with individuals with little risk of having developed PD, we
confined this analysis to individuals age 50 and older.
Among this group, there were 86 men at risk, of whom
38 (44%) were themselves affected. For women, the
proportion was 21 affected of 56 at risk (37.5%). The
similarity of these segregation ratios ( p = 0.54, x')
suggests that men and women in this kindred are
equally at risk of acquiring PD.
The asymmetrical sex ratio of individuals at risk (86
men, 56 women) does not quite reach statistical significance ( p = 0.095 for the difference from the 1 : 1
ratio, x')but its trend is not readily explained. A review of the family histories of 50 consecutive patients
age 50 or older of one of us (L. I. G.), with no known
affected first- or second-degree relatives, revealed a sex
ratio among patients' siblings age 50 or older close to
the 1 : 1 ratio expected (70 brothers, 68 sisters). The
tendency of the Contursi kindred to have fewer female
members who survive to age 50 is therefore not shared
by families of those with PD in the community.
Onset Age
Onset age was known for 48
of the affected members; the means are 46.8 years for
men, 43.4 years for women, and 45.6 overall. The
number of subjects included and standard deviations
appear in Table 4. The difference between men's and
women's mean age at onset is not significant (twotailed unpaired Student's t test).
The Contursi kindred's mean age at onset is approximately 15 years younger than that of PD in the community, but its standard deviation, 13.61 years, is similar or larger [30, 311. The distribution of onset ages
can be compared with that of one community series
from Italy [31] (Fig 2). The similarity of the shapes of
the distributions (despite the difference in the means)
is apparent.
DESCRIPTIVE STATISTICS.
I
Contursi kindred
calculated OnSelsge
i
PD in community
(Ferrara, Italy)
Fig 2. Onset age distributions in the Contirrsi kindred mid
in the community of Ferrara, /tab [31]. The Contursi kindred data are derived ,from 48 individuals with known onset
ages and 8 calculated onset ages (see text). The distribution is
similar in shape and width to tbat of PD in the commirnity.
PARENT-OFFSI'RING
CONCORDANCE
FOR
O N S E T AGE.
For this calculation, we used only the 12 dually affected parent-child pairs in which the parent was in
Generation VII, the child was in Generation VIII, and
onset ages of both members were known. The Pearson
correlation coefficient ( r ) for these pairs was -0.12,
not a significant correlation ( p = 0.71).
For this
calculation, we used the 13 sibships that included at
least 2 affected members with known onset age. We
performed ANOVA to test the hypothesis that the intrasibship variance for onset age was not different from
that among all 35 members of such sibships. The variance within sibships was 161.9 years and that of the
entire group was 152.5 years, not a significant difference (F = 0.94, p = 0.53).
INTRAWSIBSHIP C O N C O R D A N C E FOR O N S E T AGE.
There were 46 dually affected parentchild pairs for whom we had a known or calculated
onset age in both members. Among these pairs, the
mean anticipation interval was 3.6 years ('Table 5 ) , a
statistically significant difference ( p = 0.03). However,
when we confined the analysis to those 25 pairs in
which both members' onset ages were known rather
than calculated, the anticipation interval declined to
5.4 years, a nonsignificant difference ( p = 0.25). Considering only the pairs from the most fully ascertained
generations, VII and VIII, gave an anticipation interval
of 8.7 years ( p = 0.055) for the 26 pairs with either
known or calculated onset ages and an interval of -5.0
years ( p = 0.31) for the 12 pairs with known onset
ages. The last result denotes a nonsignificant trend
against anticipation.
We found no effect of the sex of the transmitting
parent on the degree of anticipation, regardless of
whether the analysis considered all parent-child pairs
or only those with known onset age (Table 6).
AN'I'ICIPATION.
Golbe et al: PD in the Contursi Kindred 771
Tuble 5. Cornparism of‘ Pawntul and Ofipring Onset Ages in
the Contursi
Kindred
N
Mean
(yr)
Standard
Deviation
(yr)
46
46
56.3
46.7
15.87
14.03
0.03
25
25
50.9
45.5
10.46
12.34
0.25
26
26
57.0
48.3
18.37
12.10
0.055
12
12
45.8,’
508
9.49
12.02
0.3 1
p Value
(Paired t Test)
Dually affected parent-child pairs, all generations, both parental genders
Onset age known or calculated i n one or both members (see text)
Parent’s onset age
Child’s onset age
Onset age known in both members
Parent’s onser age
Child’s onser age
Dually affected parent-child pairs, generations VII/VIII (Fig 1) only,
both parental genders
Onset age known or calculated in one or both members (see rext)
Parent’s onset age
Child’s onset age
Onset age known in both members
Parent’s onset age
Child’s onset age
-
~
’Note a trend opposite that of anriciparion
Tuble 6 hrect of Gender ofAjFected Pizretit on Ofipring PD Oiiset Age in the Contimi Kindred: Antiripirtion IntervaL
(Parent; Onset Age Minus Child; Onset Age)
~ _ _ _ _ _ _ _ _
N
Dually affected parent-child pairs with known onser age only
Father-child pairs
Mother-child pairs
1)ually affected parent-child pairs with known or calculated onset age
Father-child pairs
Mother-child pairs
iMean
Standard
Deviation
p Value (Unpaired
(yr)
(yr)
Student’s t Test)
15
10
5.6
5.0
15.15
19.19
0.93
27
19
10.5
10.8
21.43
19.10
0.96
PARENT-CHII.D TREMOR (TINCORDANCE.
T h e presence
of tremor tended to be concordant in the 14 dually
affected parent-child pairs, with 10 concordant for the
presence of tremor and 4 discordant ( p = 0.023, Fisher’s exact test).
able (Table 7). Of the 25 affected individuals in such
sibships, 14 were members of a majority (with regard
to tremor status) within their sibships, 1 was in a minority, and 10 were in sibships that included equal
numbers with and without tremor. T h e null hypothesis
that tremor does not cluster within sibships implies
that the sum of the number of individuals who are
members of intrasibship majorities plus half the number who are in ties should equal the sum who are in
minorities plus half of those in ties. We observed 19
in the “majority plus ties” group and 6 in the “minority plus ties” group. This does not differ significantly
from the 12.5: 12.5 expected under the null hypothesis
( p = 0.1 1, x’), suggesting that tremor does not cluster
within sibships t o a statistically significant degree in
this kindred.
INTRASIHSHIP TREMOR CONCORDANCE.
There were 3
sibships with 2 or more affected members for whom
data o n the presence or absence of tremor were avail-
Discussion
A young onset age and rapid rate of disease progression
distinguish PD in the Contursi kindred from that in
Tremor
Despite observations to the contrary in our initial report [26], tremor occurred frequently in this kindred.
Of the 31 affected individuals for whom data on rest
tremor were reliably obtained, 18 (58.1%) had tremor
and 1 3 (41.9%) did not. Onset age did not differentiate these two groups, with the mean onser ages of 47.2
years for the group with tremor and 45.2 years for the
group without tremor ( p = 0.67, two-tailed Student’s
t test).
772 Annals of Neurology
Vol 40
No 5
November 1996
Table Z Does Tremor Cluster Within Multiply Affected
Sibships in the Contursi
Affected Members of Sibship
(numbers refer to Fig 1)
With
Tremor
Without
Tremor
4,5
13, 14
15-18
19-22
23, 24
25, 26
36-40
47,48
55. 56
'Included are data from sibships with 2 or niore affected individuals
for whom data on tremor arc available. T h e tendency for tremor
to cluster within sibships was not significant ( p = 0.11, x'). See
text for details.
the community. But P D in the Contursi kindred shows
an onset age variance, gender-related risk, tremor prevalence, and pathological picture similar to those in PD
in the community. Most affected kindred members
whose response to levodopa was observed responded
well to the drug and exhibited its usual motor complications. Some degree of cognitive impairment occurred
eventually in most patients and a few developed depression. T h e hereditary pattern in the kindred is clearly
autosomally dominant. T h e most important conclusion
from these data is that a single genetic defect can produce a condition with the features and variability of
PD. This is consistent with the hypothesis that PI1 in
the community is itself caused by a single, autosomally
dominant genetic defect.
Such a model avoids the necessity to explain the
wide phenotypic variation of P D in the community as
the result of differential exposure to one or more exogenous factors. Our data, however, do not address the
possibility that exogenous exposures influence the phenotypic expression of PD in the community to some
extent [32]and do not exclude the possibility that they
may in fact d o so in the Contursi kindred.
Nor do our data address the issue of whether a putative single gene mutation causing PD in the community is the same mutation at work in the Contursi kindred, or whether etiological or genetic heterogeneity
exists in PI1 in the community. T h e older onset-age
distribution and lower apparent penetrance of PD in
the community suggest that, in fact, the disease gene(s)
locus could be different from that in the Contursi kindred. However, is it possible for both (or the several)
mutations to be allelic-that
is, to reside in the same
gene? O n e precedent is muscular dystrophy, where a
mutation in the dystrophin gene that completely prevents the production of normal dystrophin causes the
rapidly progressive, early-onset Duchenne dystrophy
and a less severe mutation in the same gene causes
only partial deficiency of dystrophin and the slowly
progressive and somewhat later-onset Becker dystrophy [ 3 3 ] . Another example is pallido-ponto-nigral
degeneration (PPND) and disinhibition-dementia-parkinsonism-amyotrophy complex (DDPAC, the Wilhelmsen-Lynch syndrome) (Z. K. Wszolek, personal
communication, 1995). These two non-Lewy-body,
autosomally dominant, multiple-system degenerations
have easily distinguishable but overlapping clinical and
pathological pictures and each is related to defects in
the region of 17q21-22 [34-361. While the precise loci
of the two mutations remain unknown, it seems likely
that they reside in the same gene.
A mechanism for phenotypic variation in an illness
arising from a mutation at a single locus is variation
in the length of a tandem repeat at that locus. T h e
most frequent such repeat, the trinucleotide cytosineadenosine-guanine (CAG), varies in length in Huntington's disease [37],Machado-Joseph disease [38],
and other neurodegenerative conditions. In some of
these conditions, the repeat lengths (within a certain
range) inversely correlate with age at symptom onset.
Progressive repeat lengthening during gametogenesis
helps explain the anticipation that is well documented
in such kindreds. In the Contursi kindred, a screen for
linkage to a CAG repeat [39] has proved negative [40].
That result is consistent with the present data showing
no convincing evidence for anticipation in the kindred.
T h e reasons for the early onset age and short survival
time are obscure. W e saw no atypical (for I'D) occurrence of postural instability or dysphagia, each of which
can shorten survival in neurological disease. Nor did
we see a preponderance of any particular nonneurological cause of death such as premature cancer, diabetes,
arteriosclerosis, or suicide. T h e short survival time cannot be attributed to poor medical care in the past or in
the historically medically underdeveloped area of Italy
where Contursi is located.
T h e evidence for autosomal dominance as the mechanism of familial aggregation in the Contursi kindred
is compelling. W e observed similar segregation ratios
among siblings and parents of affected individuls and
among offspring of consanguineous and nonconsanguineous marriages. There was no evidence for sexlinked or mitochondria1 inheritance, a result in accord
with previous analyses of series of smaller kindreds [ 12,
14, 411. An environmental factor is unlikely to link
the temporally and geographically disparate generations
and branches of the family. This would require us to
postulate a most remarkable exogenous intoxication
caused by a subtle cooking style or other cultural practice that survived immigration, assimilation, and succession of generations in the United States, Germany,
and Argentina. O u r attempts to uncover such a prac-
Golbe et al: PD in the Contursi Kindred 773
tice by closely interviewing family members in Contursi and in the United States were fruitless. Furthermore, the absence of PD among living nonkindred
residents of Contursi speaks strongly against an environmental cause.
The validity of our observations on the occurrence
of PD in the Contursi kindred may suffer from incomplete or even mistaken reports by family members and
from the possible presence of clinical phenocopies (i.e.,
family members with parkinsonism not caused by the
genetic defect that causes PD in their relatives). With
a lifetime incidence of PD in the population of approximately 2% 1421, a family of this size is likely to include
at least 1 such person. The more common error, however, is probably false-negative observations produced
by age-related nonpenetrance. With the oldest observed
onset age being 85, it is likely that some individuals
in the kindred who died without developing symptoms
of PD actually carried the gene. This phenomenon is
illustrated by the 6 individuals who had affected children and an affected parent and died themselves unaffected at ages 46 through 94. We would have failed to
recognize as heterozygotes such individuals who had
no affected children. The sensitivity and specificity of
”F-dopa positron emission tomography [43] or other
presymptomatic tests for identifying such persons have
not been adequately measured and were not applied in
this study.
Although this kindred includes by far a greater number of affected members with PD than any other reported kindred, there remains the possibility of type B
error, in which statistical tests can give negative results
solely because of an insufficient number of subjects. In
particular, our results regarding the relation of gender
to segregation ratio (see Table I), of parental consanguinity to offspring segregation ratio (see Table 3), and
the relationship of parental gender to anticipation interval (see Table 6) must be interpreted in this light.
Markopoulou and colleagues [44] recently reported
a Greek-American kindred with clinically typical PD
in 16 affected individuals in 3 generations. It shares
with the Contursi kindred the appearance of autosomally dominant inheritance, mean onset age in the 40s
with a wide variance, mean survival time of 9 years,
and the presence of tremor and a good levodopa response in the majority of affected persons. With origins
in southern Italy and Greece, respectively, it seems possible that the Contursi kindred and that of MarkopouIOU and colleagues [44] share a common ancestor.
Molecular genetic analysis of the Contursi kindred
is in progress. Once linkage is found, presymptomatic
members of the kindred may be identified with acceptable and measurable specificity. This will permit identification of additional heterozygotes, which will hasten
the positional cloning of the gene itself. That, in turn,
will permit a definitive evaluation of the extent to
774 Annals of Neurology
Vol 40
No 5
which PD in the community is caused by abnormalities
in the same gene that is defective in the Contursi kindred.
This work was supported in part by a center grant from the American Parkinson’s Disease Association.
November 1396
We are graceful for the assistance of William G. Johnson, MD,
Ronald Cody, EdD, Oscar S. Gershanik, MD, J. William Langston,
MD, and Peter Vieregge, MD.
References
1. Duvoisin RC. The genetics of Parkinson’s disease: a review.
Adv Neurol 1993;60:306-315
2. G o b e LI.The genetics of Parkinson’s disease: a reconsideration. Neurology 1990;4O(suppl 3):7-14
3. Payami H , Karsen K, Bernard S, Nutr J. Increased risk of Parkinson’s disease in parents and siblings of paricnts. Ann Neurol
1994;36:659-66 1
4. Vieregge P, Heherlein I. Increased risk of Parkinson’s disease
in relatives of patients. Ann Neurol 1995;37:685
5. Semchuk KM, Love EJ, Lee RG. Parkinson’s disease: a test
of the mu~tifactoria~
etiologic hypothesis. Neurology 1993;43:
1173-1 180
6. Alonso ME, Otero E, D’Regules R, Figucroa HH. Parkinson’s
disease: a genetic study. Can J Neurol Sci 1986;13:248-251
7. Hubble JP, Cao T, Hassanein RES, et al. h s k factors for Parkinson’s disease. Neurology 1994;43:1693- 1697
8. De Michele G, Filla A, Volpe G, et al. Environmental and
genetic risk factors in Parkinson’s disease: a case-control study
in southern Italy. Mov Disord 1996;11:17-23
9. Ward CD, Duvoisin RC, Ince SE, et al. Parkinson’s disease
in 65 pairs of twins and in a set of quadruplers. Neurology
1983;33:815-824
10. Duvoisin RC, Eldridge R, Williams A, et al. Twin study of
Parkinson’s disease. Neurology I98 1;31:77-80
11. Johnson WG, Hodge SE, Duvoisin RC. Twin studies and the
genetics of Parkinson’s disease-a
reappraisal. Mov Disord
1990;5: 187- 194
12. Lazzarini AM, Myers RH, Zimmerman TR, et al. A clinical
genetic study of Parkinson’s disease: evidence for dominant
transmission. Neurology 1994;44:499- 506
13. Bonifati V, Fabrizio E, Vanacore N , et al. Familial Parkinson’s
disease: a clinical genetic analysis. Can J Neurol Sci 1995;22:
272-279
14. Maraganore DM, Harding AX, Marsden CD. A clinical and
generic study of familial Parkinson’s disease. Mov Disord 1990;
6~205-211
15. (hlbe L1, Di Iorio G, Bondvita V, et al. A large kindred with
autosomal dominant Parkinson’s disease. Ann Neurol 1990;27:
276-282
16. G o b e LI, Lazzarini AM, Schwarz KO, et al. Autosomal dominant parkinsonism with benign course and typical Ley-body
pathology. Neurology 1993;43:2222-2227
17. Waters C H , Miller CA. Autosomal dominant LKY~body parkinsonism in a four-generation family. Ann Neurol 1994;35:
59-64
18. Bhatia KP, Daniel SE, Marsden CD. Familial parkinsonism
with depression: a clinicopathological study. Ann Neurol 1993;
34:842-847
19. Spellman GG. Report of familial cases of parkinsonism. JAMA
1962;179:160- 162
20 Muenter MD, Howard FM, Okazaki H , et al. A familial parkinson-dementia syndrome. Neurology 1386;36(suppl 1):115
(Abstract)
21. Perry TL, Wright J M , Berry K. er al. Dominantly inherited
apathy, central hypoventilation, and Parkinson’s syndrome:
clinical, biochemical. and neuropathologic studies of 2 new
cases. Neurologv 1990;40:1882- 1887
22. Purdy A, Hahn A, Barnett HJ, et al. Familial fatal parkinsonism with alveolar hypoventilation and mental depression. Ann
Neurol 1979;6:523-531
23. Inose T, Miyakawa M. Miyakawa K, et al. Clinical and neuropathological study of a familial case of juvenile parkinsonism.
Jpn Psychiarr Neurol 1988;42:265-276
24. Bonihti V, Vanacore N,Meco G . Anticipation of onset age
in familial Parkinson’s disease. Neurology 1994:44: 1978- I979
25. Payami H , Bernard S, Larsen K. et al. Generic anticipation in
Parkinson’s discase. Neurology 1995:45: 135-138
26. Golbe LI, Di Iorio G. Bonavita V, Duvoisin RC. A large kindred with Parkinson’s disease: onset age, segregation ratios and
anricipation. Mov Disord 1992;7:24 (Abstract)
27. Hughes AJ, Daniel SE, Blankson S. Lees AJ. A clinicopathologic study of 100 cases of Parkinson’s disease. Arch Neurol
1993;50:140-148
28. Chi6 A. Magnani C, Tolardo G , Schiffer D. Parkinson’s disease mortality in Italy, 1951 through 1987. Arch Neurol 1993;
50: 149-1 53
29. Hoehn MM, Yahr MD. Parkinsonism: onset, progression and
mortality. Neurology 1967; 17:427-442
30. Mutch WJ, Dingwall-Fordyce I, Downe AW, et al. Parkinson’s
disease in a Scottish city. BMJ 1986;292:534-536
31. Granieri E, Carrrras M , Casetta I. et al. Parkinson’s disease in
Ferrara. Italy, 1967 through 1987. Arch Neurol 1991;48:854857
32. Taylor CA, Myers RH, Carver A, et al. Co-aggregarion of family history and environmental exposure in Parkinson’s disease.
Mov Disord 1995;10:693 (Abstract)
33. Rowland LP. Dystrophin: a triumph of reverse genetia and
the end of the beginning. N Engl J Med 1‘)88;318: 1392- 1394
34. Wilhelmsen KC. Lynch T. Nygaard TG. Localization ofdisin-
hibition-~~‘~rl\insotiism-amyorro~~li~
coinplcx to 17q21-22. Am
J H u m Genet 1994;55:1 159-1 165
35. Lynch T, Sano M, hhrder KS. et al. Cliniccd characrerisrics of
36.
37.
38.
39.
40.
41.
42.
43.
44.
a family with chromosome 17-linked disinhibition-dernentiaparkinsonism-amyotrophy complex (DDPAC). Neurology
1994:44:1878-1884
Wszolek ZK. Pfeiffer RF, Hhart M H , et A. kipidly progressive autoaomal dominant parkinsonism and dementin with
pallido-ponto-nigral degeneration. Ann Neurol 1 ‘192;32:312320
Andrew SE, Goldberg YP,Kremer B, et al. The relationship
benvcen trinucleotide (CAG) repeat length and clinical features
of Huntington’s disease. Nature Genet 1993;it:398-403
Maciel P, Gaspar C . DeStefano AL, et al. Correlation hetwecn
CAG repeat length and clinical features i n Machado-Joseph
disease. Am J Hum Genet 1995;57:54-61
Schalling M, Hudson TJ, Buerow KH, Housman DE. 1)irect
detection of novel expanded trinuclcotide repeats i n the human
genome. Nature Genet I993;4: 135- 139
Carrero-Valenzuela R, Lindblad K, I’ayami H, et al. No evidence for association of familial Parkinson disease with the
CAG repeat expansion. Neurology 1395;45: 1760-1 763
Zweig RM. Singh A, Cardillo JE, Langston JW. ’The familial
occurrence of Parkinson’s disease: lack of evidence for maternal
inheritance. Arch Neurol 1992;49:1205-1207
Kurland LT. Epidemiology: incidence, geographic disrribution
and genetic considerations. In: Fields WS, ed. Pathogenesis
and treatment of parkinsonism. Springfield, IL: Charles C
Thomas, 1958:5-49
Sawle GV. Playford ED. Burn DJ. et al. Separating Parkinson’s
disease from normality: discriminant function analysis of fluorodopa F 18 positron emission tomography d m . Arch Neurol
1994:5 1237-243
Markopoulou K, Wszolek ZK, I’feiffer RF. A Greek-American
kindred with autosomal dominant, levodopa-responsive parkinsonism and anticipation. Ann Neurol 1995;38:373-378
Golbe ct al: I’D in the Conrursi Kindred
775
Документ
Категория
Без категории
Просмотров
3
Размер файла
845 Кб
Теги
kindred, clinical, analysis, disease, parkinson, genetics, contursi
1/--страниц
Пожаловаться на содержимое документа