Clinical Chemistry 53:3
392–398 (2007)
Molecular Diagnostics
and Genetics
Evaluation of the Risk for Tay-Sachs Disease in
Individuals of French Canadian Ancestry Living in
New England
Dianna C. Martin,1 Brian L. Mark,2 Barbara L. Triggs-Raine,1,3,4 and
Marvin R. Natowicz5*
Conclusions: We estimate the frequency of deleterious
TSD alleles in Franco-Americans to be 1:73 (95% confidence interval 1:55 to 1:107). These data provide a more
complete data base from which to formulate policy
recommendations regarding TSD heterozygosity screening in individuals of French Canadian background.
Background: The assessment of risk for Tay-Sachs
disease (TSD) in individuals of French Canadian background living in New England is an important health
issue. In preliminary studies of the enzyme-defined
carrier frequency for TSD among Franco-Americans in
New England, we found frequencies (1:53) higher than
predicted from the incidence of infantile TSD in this
region. We have now further evaluated the risk for TSD
in the Franco-American population of New England.
Methods: Using a fluorescence-based assay for ␤-hexosaminidase activity, we determined the carrier frequencies for TSD in 2783 Franco-Americans. DNA analysis was used to identify mutations causing enzyme
deficiency in TSD carriers.
Results: We determined the enzyme-defined carrier
frequency for TSD as 1:65 (95% confidence interval 1:49
to 1:90). DNA-based analysis of 24 of the enzymedefined carriers revealed 21 with sequence changes: 9
disease-causing, 4 benign, and 8 of unknown significance. Six of the unknowns were identified as
c.748G>A p.G250S, a mutation we show by expression
analysis to behave similarly to the previously described
c.805G>A p.G269S adult-onset TSD mutation. This putative adult-onset TSD c.748G>A p.G250S mutation has
a population frequency similar to the common 7.6 kb
deletion mutation that occurs in persons of French
Canadian ancestry.
© 2007 American Association for Clinical Chemistry
Tay-Sachs Disease (TSD)6 is an autosomal recessive lysosomal lipid storage disorder that is due to deficiency of
the lysosomal enzyme ␤-hexosaminidase A (HEXA) [reviewed in (1 )]. The HEXA isoenzyme is comprised of 1
␣-subunit encoded by HEXA7 and 1 ␤-subunit encoded by
HEXB. Two ␤-subunits can combine to form a 2nd isoenzyme with different but overlapping substrate specificities, ␤-hexosaminidase B (HEXB). Mutations in both alleles of HEXA result in a deficiency of HEXA, and
mutations in both alleles of HEXB result in deficiencies of
both HEXA and HEXB. Mutations at both alleles of either
gene locus lead to a deficiency of HEXA activity and an
accumulation of GM2 ganglioside, leading to similar lysosomal storage disorders, TSD and Sandhoff disease. There
are also mutations in these genes that result in a partial
deficiency of HEXA and lead to less severe forms of
disease. Heterozygotes for deleterious mutations in
HEXA and HEXB can be identified with an enzyme-based
activity assay.
To identify TSD carriers by use of an enzymatic assay,
the combined activities of HEXA and HEXB, as well as the
activity of HEXB after heat-inactivation of HEXA, are
determined. Heterozygotes for TSD have reduced HEXA
Departments of 1 Biochemistry and Medical Genetics, 2 Microbiology, and
Pediatrics and Child Health, University of Manitoba, Winnipeg, MB, Canada.
4
Manitoba Institute for Child Health, Winnipeg, MB, Canada.
5
Genomic Medicine Institute and Departments of Neurology, Pathology
and Laboratory Medicine and Pediatrics, Cleveland Clinic, Cleveland, OH.
* Address correspondence to this author at: Genomic Medicine Institute,
Cleveland Clinic Foundation NE-5, 9500 Euclid Avenue, Cleveland, OH 44195.
Fax 216-636-0009; e-mail [email protected].
Received November 6, 2006; accepted December 28, 2006.
Previously published online at DOI: 10.1373/clinchem.2006.082727
3
6
Nonstandard abbreviations: TSD, Tay-Sachs Disease; HEXA, ␤-hexosaminidase A; HEXB, ␤-hexosaminidase B; CI, confidence interval; HEXS,
␤-hexosaminidase S; SSCP, single-strand conformational polymorphism.
7
Human genes: HEXA, alpha polypeptide gene; HEXB, beta polypeptide
gene.
392
Clinical Chemistry 53, No. 3, 2007
activity (observed as a reduced percentage of HEXA
relative to the total ␤-hexosaminidase activities). This
enzyme-based assay is routinely used by clinical laboratories that screen for heterozygotes for TSD (2 ).
Enzyme-based screening programs for HEXA deficiency were initiated in the 1970s for the Ashkenazi
Jewish and, later, French Canadian populations that were
each recognized to have an increased incidence of TSD
(3, 4 ). Subsequently, an increased prevalence of TSD was
discovered for individuals of Moroccan Jewish, Pennsylvania Dutch, and Southern Louisiana Cajun backgrounds
(5–7 ). These assays also led to the recognition of increased
TSD carrier frequencies among Franco-Americans living
in New England (8, 9 ), Iraqi Jews (10 ), and Irish Americans (11 ).
Elucidation of the molecular basis of TSD has made
apparent an alternative approach for heterozygosity
screening in some populations. In persons of Ashkenazi
Jewish background, DNA-based carrier screening is possible because 3 mutations account for ⬃95%–98% of
obligate TSD carriers (12, 13 ). In contrast, the molecular
basis of TSD in most non-Jewish populations is highly
heterogeneous, prohibiting DNA-based screening. Enzyme-based screening in the non-Jewish population is,
however, often supplemented by molecular analysis for
specific “pseudodeficiency” or benign mutations. These
benign mutations cause partial enzyme deficiency, but the
level of residual activity is adequate to prevent the
pathological accumulation of GM2 ganglioside (14 ).
In non-Jewish populations benign mutations account
for differences between the carrier frequencies determined by enzyme screening (1:167) compared with those
estimated on the basis of disease incidence (1:300)
(13, 15, 16 ). Through molecular analysis of enzyme-defined carriers, the apparent risk for TSD within a population can be modified by adjusting for the population
frequency of benign mutations. In the case of Irish Americans, such analyses lowered the carrier frequency from
1:25 to a maximum of 1:41 (11 ). In preliminary studies of
Franco-Americans living in Massachusetts (MA), accounting
for benign mutations reduced the risk from 1:57 to 1:66 (8 ).
Using a much larger study population and including
analyses for additional mutations, we have expanded our
studies of TSD carrier frequencies among persons of
French Canadian background living in New England. We
also characterize biochemical and structural aspects of a
mutation found to have an unexpectedly high frequency
in this population. Based on our analyses, we estimate the
maximal TSD carrier frequency to be 1:73 [95% confidence
interval (CI), 1:55 to 1:107].
Materials and Methods
samples
Samples were obtained from individuals referred to the
E.K. Shriver Center in Boston, MA for TSD carrier screening; all specimens were obtained after donors gave written informed consents. Individuals who self-identified as
393
being Franco-American or French Canadian, or who reported one or more grandparents of French Canadian
background, were eligible for the study except as noted
below. Individuals who were pregnant, had both Ashkenazi Jewish and French Canadian heritage, or indicated
that they had a relative who was a carrier for or had TSD,
were excluded from the study. The questionnaire used to
identify appropriate study participants was previously
published (8 ).
enzyme analysis
Heterozygosity for TSD was determined with the standard heat-inactivation assay (2, 17 ). The percentages of
serum HEXA activity used to identify heterozygotes and
non-carriers for TSD were as reported previously (8 ). The
diagnostic criteria for serum HEXA used to establish a
diagnosis of heterozygosity and non-carrier status for
TSD were: carrier, ⱕ55% HEXA and non-carrier, ⱖ61%
HEXA. The diagnostic criteria for leukocyte HEXA were:
carrier, ⬍55% HEXA and non-carrier, ⱖ62% HEXA. Each
case was individually evaluated on the basis of the serum
HEXA %, leukocyte HEXA %, and total HEX activity.
Samples with a percentage HEXA activity close to the
cutoff for the carrier range and without a mutation
detected by single-strand conformational polymorphism
(SSCP) in our original study (9 ) were reassayed by a 2nd
reference laboratory. Four samples were identified as
non-carriers by the 2nd reference laboratory, and they are
included as non-carriers in this analysis.
dna isolation and detection of mutations
From the 43 individuals that tested as TSD carriers,
leukocyte pellets or sonicates were available for 25; DNA
was successfully isolated from 24 samples by use of
previously described procedures (18 ). Nine enzymebased assay positive samples that were not part of our
previous analysis were tested for 4 DNA sequence variations, the 7.6 kb deletion (19 ), c.739C⬎T (15 ), c.805G⬎A
(20, 21 ), and c.748G⬎A (9 ), according to established protocols (9 ). For samples with enough remaining DNA (22
samples) and no initially identified mutation, SSCP analysis of each exon was performed as described previously
(9 ).
expression of hexa mutations
We used methods described previously (14, 22 ) to assess
the effect on HEXA activity of 2 novel amino acid substitutions, c.748G⬎A p.G250S and c.587A⬎G p. N196S.
Mutations were introduced into the ␣-subunit cDNA with
oligonucleotide primers 5⬘ ACG GCT CCG GAG TAT
CCG TGT GC 3⬘ for c.748G⬎A, and 5⬘CAT GGC GTA
CAG TAA ATT GAA CG 3⬘ for c.587A⬎G. The presence
of only the desired change in the cDNA was confirmed by
sequencing. Mutated cDNAs were subcloned into pSVL
vectors to create ␣pSVLG748A and ␣pSVLA587G, again
following the same approach previously used in our
laboratory (14 ). Vectors with mutations associated with
394
Martin et al.: Tay-Sachs Carriers in New England
Table 1. Carrier frequency for TSD in the New England states.
Total Tested
TSHg
TSH Frequency
a
Massachusetts;
b
MAa
NHb
CTc
RId
VTe
MEf
Total
2072
36
1/57
542
5
1/108
76
1
1/76
15
0
47
1
1/47
31
0
2783
43
1/65
New Hampshire;
c
Connecticut;
d
Rhode Island;
e
Vermont; f Maine;
normal (␣pSVL), TSD (␣pSVLC508T), adult-onset TSD
(␣pSVLG805A), and pseudodeficiency (␣pSVLC739T)
phenotypes were used as described in previous studies
(14, 22 ). These vectors were transfected into COS-7 cells,
alone or in combination with the ␤-subunit cDNA
(pCD43), and the HEXA (␣␤ dimer) and ␤-hexosaminidase S (HEXS) (␣␣ dimer) activities, and ␣-subunit protein
levels were monitored according to previously described
procedures (14 ).
statistical analysis
We used a binomial distribution to calculate 95% CIs for
the carrier frequencies for a proportion (NCSS Software).
Results
determination of the tsd enzyme-defined
carrier frequencies
Our prospective analysis of heterozygosity for TSD in
individuals of French Canadian background living in
New England included samples collected between 1989
and 2001, from 2783 nonpregnant donors. Samples were
obtained from individuals from 6 New England states; the
majority of the samples were from Massachusetts and
New Hampshire residents (Table 1). In total, 43 TSD
carriers were identified, giving an enzyme-based TSD
carrier frequency in this population of 1 in 65 (95% CI,
1:49 to 1:90).
detection of known common tsd mutations
For 24 of the 43 TSD carriers (above), DNA was available
for molecular analysis of the HEXA gene. We first used
direct analysis to test for common mutations in the
samples; if the result was negative, we then used SSCP to
screen for other mutations in the 14 HEXA exons. All
variations were confirmed by sequencing. The results of
the molecular analyses are reported in Table 2. In total,
g
Tay-Sachs disease heterozygote.
mutations were identified in 21 of 24 enzyme-defined
carriers.
We categorized the identified variations on the basis of
their associated phenotype— disease-causing, benign, or
unknown (Table 2). In total, 9 of 21 samples had known
disease-causing mutations and 4 of 21 samples had
known benign mutations (c.739C⬎T or c.745C⬎T). Three
variations of unknown significance, c.748G⬎A p.G250S, c.
587A⬎G p.N196S, and c.1146 ⫹ 18A⬎G, accounted for
the remaining 8 alleles. Six persons had the c.748G⬎A
p.G250S variation.
functional assessment of novel amino acid
substitutions
To assess the functional significance of the 2 novel amino
acid changes, c.587A⬎G p.N196S and c.748G⬎A p.G250S,
these changes were introduced into the ␣-subunit cDNA
and transfected into COS-7 cells, as were several control
vectors. Transfection of the ␣-subunit vectors alone to
generate HEXS (␣␣) showed that the N196S variant had
only a slight impact on ␣-subunit function, reducing the
HEXS activity to 87% of the wild-type activity (Fig. 1A).
This reduction was significantly less than the well-characterized benign mutation c.739C⬎T p.R247W, which
reduced the HEXS activity to 37%. However, c.748G⬎A
p.G250S reduced the activity of HEXS to only 4% of the
activity of the wild-type ␣-subunit cDNA after the background is subtracted, an activity level similar to that of the
c.805G⬎A p.G269S mutation that results in a late- or
adult-onset form of TSD. These findings suggest that the
c.748G⬎A encoded G250S substitution severely impacts
the function of the ␣-subunit, whereas the N196S substitution encoded by c.587A⬎G has little effect.
To further investigate the functional effects of these
changes, the normal and mutant ␣-subunit vectors
(␣pSVL, ␣pSVLC508T, ␣pSVLG805A, ␣pSVLC739T,
Table 2. Mutations in enzyme-defined TSD carriers.
Samples
⌬ 7.6 (19 )
ⴙ4 bp (40 )
739C>T (15 )
745C>T (16 )
748G>A (9 )
587A>G (9 )
Misca
Total
24b
Disease-causing
Benign
Unknownc
5
5
2
2
2
2
6
1
3
2
2
2
6
1
1
21
9
4
8
Mutations found in only 1 sample included the disease-causing mutations, c.508C⬎T, and c.409C⬎T, and a variant mutation of unknown significance, c.1146 ⫹
18 A⬎G.
b
Total number of DNA samples analyzed from enzyme-defined nonpregnant carriers.
c
The effect of the mutation on the clinical phenotype is unknown.
a
395
Clinical Chemistry 53, No. 3, 2007
Fig. 1. Expression analyses in COS-7
cells.
Expression of HEXS (␣␣) activity in COS-7
cells (A). Normal and mutant ␣pSVL constructs were transfected, and the percentage HEXS activity was calculated by taking
the vector expressing the wild-type (WT)
␣-subunit, ␣pSVL, as 100%. The other values were determined as a proportion of
␣pSVL activity ⫻ 100 (n ⫽ 2). Expression of
HEXA (␣␤) and HEXS (␣␣) activities in
COS-7 cells (B). Normal and mutant ␣pSVL
(6.5 ␮g) were cotransfected with the ␤-subunit cDNA, pCD43 (2.3 ␮g). HEXA- and
HEXS-specific activities were determined by
using the fluorescent substrate, 4-methylumbelliferyl-6-sulfo-␤-N-acetylglucosaminide. For each of 3 experiments, the average
wild-type activity from 3 replicates was determined and defined as 100%. The average activity for each mutant was then determined as a percentage of the wild-type
activity for each of 3 experiments, and the
average mean (SEM) is plotted in the bar
graph. c.739C⬎T, pseudodeficiency mutation; c.805G⬎A. adult-onset mutation;
pSVL, vector control; c.587A⬎G, unknown
sequence variant; c.748G⬎A, unknown sequence variant.
␣pSVLG748A, and ␣pSVLA587G) were cotransfected
with the ␤-subunit cDNA to generate both the HEXS (␣␣)
and HEXA (␣␤) ␤-hexosaminindase isoenzymes in COS-7
cells. This approach has been used previously and is
known to amplify residual activity associated with the
mutant ␣-subunit, allowing the differentiation of mutations associated with complete and partial HEXA deficiencies (14, 23, 24 ). Using this approach, we noted similar findings to that for the transfection of the ␣-subunit
cDNAs alone (Fig. 1, Panel B and Fig. 2). The levels of
HEXA and HEXS activities as well as mature ␣-subunit
protein associated with the c.587A⬎G variation are
clearly greater than that associated with the benign mutation (c.739C⬎T). In contrast, the c.748G⬎A mutation
reduced the activities of HEXA and HEXS to levels
comparable to the c.805G⬎A adult-onset mutation, although above that of the c.508C⬎T mutation associated
Fig. 2. Western blot analysis of transfected cell extracts.
Cell extracts (60 ␮g protein) were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis and detected with an anti-HEXA antibody. Lane 1,
mock; lane 2, ␣pSVL; lane 3, ␣pSVLC508T (control infantile mutation); lane 4,
␣pSVLG805A (adult-onset mutation); lane 5, ␣pSVLC739T (pseudodeficiency
mutation); lane 6, ␣pSVLA587G (unknown); lane 7, ␣pSVLG748A (unknown).
The precursor and mature forms of ␣ and ␤ subunits of HEXA are indicated. The
X denotes an unidentified antigenically cross-reacting band.
with an infantile form of TSD. The levels of HEXA protein
were also reduced by the c.748G⬎A mutation to a level
similar to that of the adult-onset TSD mutation c.805G⬎A
(Fig. 2).
Discussion
The assessment of the risk for TSD among individuals of
French Canadian background in the New England states
is an important health issue because up to 10.3% of the
residents of these states declared French Canadian heritage in the 2000 census and a much larger proportion
reported French ancestry (25 ); the latter, in turn, almost
certainly relates to the heavy migration of French Canadians to the New England states (26 ). The risk for TSD in
this group, however, cannot be easily predicted by comparison to the TSD carrier frequency found in French
Canadians living in Quebec because the latter varies
between regions, from 1:13 to 1:300 (4 ). In our study of
2783 individuals, we found an enzyme-defined TSD carrier frequency of 1:65, before considering the molecular
basis of HEXA deficiency in this group. This risk is lower
than that of the Ashkenazi Jewish population (⬃1:25), but
much higher than the enzyme-defined carrier frequency
in the general non-Jewish population (1:167).
We sought to further assess the risk for TSD in this
group through a mutational analysis of HEXA heterozygosity. Of the 24 enzyme-defined carriers for whom DNA
was available, we identified 9 individuals with known
disease-causing mutations, 4 with known benign mutations, and 8 with mutations of unknown significance. The
3 individuals for whom mutations were not identified
may have had mutations that were not detectable by
SSCP, because this method does not detect 100% of
mutations. Furthermore, general strategies to detect large
396
Martin et al.: Tay-Sachs Carriers in New England
duplications or insertions that could also account for the
enzyme deficiency were not performed. In these 3 individuals, however, the relatively high percentages of
HEXA activity in the serum (59.8%, 58%, 54.2%) and
white blood cells (49.5%, 53.4%, 54.1%) suggest that the
results may be false positives of the enzyme assay.
The 9 known disease-causing mutations that were
identified in this study were largely representative of
those that have previously been identified in individuals
of French Canadian background. Most of these individuals (5 of 9) had the 7.6 kb deletion, a mutation unique to
the French Canadian population that originated in the
region surrounding the Bas-St.-Laurent-Gaspésie region
of Quebec and the Madawaska Valley in the province of
New Brunswick (27 ). A 2nd mutation unique to French
Canadians, c.805 ⫹ 1G⬎A (28 ), was not identified in these
samples. The c.805 ⫹ 1G⬎A mutation originated in the
Saguenay-Lac St-Jean region of Quebec (28 ), and its
absence from our samples suggests a different geographic
origin for the Franco-Americans that we examined. Two
of the other known disease-causing mutations that were
identified, c.1287insTATC and c.508C⬎T p.R170W, have
previously been identified in French Canadians or FrancoAmericans (29, 30 ) and c.409C⬎T p.R137X has been described in a cohort of European subjects (31 ).
The known benign variants, c.739C⬎T p.R247W and
c.745C⬎T p.R249W, accounted for 4 of the mutations in
enzyme-defined TSD carriers in this study. These mutations have been well characterized and when taken into
account reduce the estimated frequency of heterozygosity
for deleterious HEXA alleles in the non-Jewish population from ⬃1:67 to ⬃1:167. When we accounted for these
known benign mutations in individuals with enzymedefined heterozygosity, the number of carriers was reduced from 43 to 39.
We further considered the 3 mutations of unknown
significance in our population, c.748G⬎A p.G250S,
c.587A⬎G p.N196S, and c.1146 ⫹ 18A⬎G, to try to
determine their impact on the enzyme-defined TSD carrier frequency. We found these sequence changes to be
absent both in 50 enzyme-defined non-carriers (9 ) and in
an additional 100 controls (data not shown) and therefore
consider them likely to be the cause of enzyme-deficiency
in these individuals. In addition, although the c.1146 ⫹
18A⬎G change was not predicted to have a major impact
on splicing (data not shown), no other variant was identified by denaturing HPLC analysis of the HEXA exons
(data not shown). The c.587A⬎G p.N196S was identified
only once among our enzyme-defined carriers, although
we had identified it in an earlier study in a carrier who
was pregnant (9 ). This substitution did not significantly
reduce the HEXA and HEXS activity and may be a benign
mutation that leads to apparent HEXA deficiency. If
c.587A⬎G p.N196S is included as a benign sequence
variant, the number of enzyme-defined carriers is reduced
to 38, giving a maximal deleterious Tay-Sachs carrier
frequency of 1:73 (95% CI 1:55 to 1:107).
We were especially interested in analyzing the
c.748G⬎A p.G250S mutation because it was identified in
6 of our enzyme-defined TSD carriers but, to our knowledge, has never been identified in a TSD patient. This
suggested to us that it may be another pseudodeficiency
mutation. Surprisingly, our expression analysis showed
the G250S substitution decreased HEXA and HEXS activities as well as ␣-subunit protein to levels well below that
of a known benign mutation and similar to that of an
adult-onset TSD mutation. Based on our expression system and functional and immunochemical assays, the
G250S substitution has the potential to cause a late-onset
form of TSD.
Interestingly, 3 different mutations have been identified at the G250 position. The first, G250D, was found to
cause a juvenile form of TSD in a Lebanese Marronite
family (32 ). A G250V change was subsequently found
among enzyme-defined TSD carriers of Iraqi Jewish origin (33 ). The G250S change was identified among FrancoAmerican enzyme-defined TSD carriers (9 ) and in this
study. The risk for TSD associated with the G250V and
G250S changes is difficult to assess because neither has
been detected in an individual with any form of TSD or
pseudodeficiency state. Nonetheless, as described above,
the behavior of the G250S-containing ␣-subunit is similar
to that of the adult-onset G269S mutation and has a more
severe impact on function than the pseudodeficiency
mutation R247W. Thus, our expression analyses suggest
the G250S mutation is more likely to be disease-causing
than benign.
The x-ray crystal structure of a functionally mature
glycosylated form of human lysosomal HEXA was recently determined (34 ). As shown in Fig. 3, the ␣- and
␤-subunits of HEXA are kidney-shaped, 2-domain proteins that form dimers. A number of disease-causing
mutations in the ␣-subunit of HEXA have been mapped to
the interface between domains I and II, including the
G250S variation discussed here. Additional notable mutations occurring at this interface include G250D (35 ),
R166G (36 ), and E482K (37 ). In each case, the interface is
disrupted by either loss of a salt bridge or a hydrogen
bond, as suggested for R166K and E482K, respectively, or
overpacking and burying a polar residue, as suggested by
G250D (34 ). An amino acid substitution of G250 cannot be
accommodated, because the added side chain cannot pack
efficiently against the ␣-helixes of domain I (Fig. 3). It
appears that the larger and more polar the substituted
residue at this position, the more detrimental it is to the
proper folding and catalytic function of HEXA. Interestingly, the benign mutations R247W and R249W appear to
have a much less dramatic effect on folding and function
because these residue positions occur on an ␣-helix of
domain II such that a bulky substitution is accommodated
by pointing the side chains toward the solvent; thus, they
do not appear to seriously affect packing or electrostatic
interactions (Fig. 3).
397
Clinical Chemistry 53, No. 3, 2007
Fig. 3. The HEXA crystal structure in
complex with the mechanistic inhibitor
NAG-thiazoline (NGT) (PDB 2GK1).
The ␣-subunit is drawn as a ribbon diagram
with domain I colored red and domain II
colored blue. The ␤-subunit, which by way of
dimerization is rotated ⬃120° toward the
viewer relative to the ␣-subunit, is colored
white. NAG-thiazoline bound into the active
site of each subunit is colored by element
and drawn as a space-filled molecule. The
inset to the left of the HEXA structure
highlights the G250S/D/V mutation hot
spot and the 2 benign mutations R247W
and R249W.
A retrospective analysis of the incidence of TSD in
northern New England suggested the carrier frequency
for TSD in this population is lower than that found in the
Ashkenazi Jewish community (i.e., ⬃1:25), although a
specific frequency statistic was not determined (38 ). Because that study ascertained cases of TSD by analysis of
death records of pediatric individuals from the 3 most
northern New England states, any pediatric individuals
who were not correctly diagnosed or who did not receive
a correct ICD disease code, as well as all individuals with
late-onset phenotypes of TSD, would not have been
identified. In addition, using an estimate of 1:73 for
heterozygosity for deleterious alleles at the HEXA locus
based on the data from this work, that retrospective
analysis would likely not have had adequate statistical
power to detect the predicted number of cases of TSD.
Consequently, the data from this study augments that of
the retrospective analysis.
Current policy recommendations regarding TSD heterozygosity testing in persons of French Canadian background specify that screening for carrier status should be
offered to such individuals before pregnancy if one or
both members of a couple are of French-Canadian or
Cajun descent (39 ). Molecular genetic testing of the TSD
gene locus affords the ability to detect alleles associated
with pseudodeficiency, lethal infantile-onset, and juvenile- and adult-onset phenotypes. This information must
be incorporated both in personal decision-making when
heterozygosity screening or prenatal testing is done and
in the formulation of policy recommendations. Overall,
the data from the current work add to earlier studies and
provide a considerably more robust database with which
public health policy recommendations concerning TSD
carrier screening can be formulated.
This work was supported by grants form the Manitoba
Medical Service Foundation, the Canadian Institutes for
Health Research and The Manitoba Institute for Child
Health (BT-R), and from the Department of Mental Retardation of the Commonwealth of Massachusetts (MRN).
DCM was the recipient of salary support from the Manitoba Institute for Child Health and the Canadian Genetic
Diseases Network (CGDN). We gratefully acknowledge
the meticulous work and devotion of all of the staff of the
Tay-Sachs Disease Prevention Program of the Shriver
Center for Mental Retardation, Boston, MA. We also
dedicate this manuscript to the memory of a friend and
colleague, Dr. Eugene Grebner, an important clinician and
researcher in the Tay-Sachs disease community.
References
1. Sandhoff K. The GM2-gangliosidoses and the elucidation of the
␤-hexosaminidase system. Adv Genet 2001;44:67–91.
2. Kaback MM, Shapiro LJ, Hirsch P, Roy C. Tay-Sachs disease
heterozygote detection: a quality control study. Prog Clin Biol Res
1977;18:267–79.
3. Kaback MM. Thermal fractionation of serum hexosaminidases:
applications to heterozygote detection and diagnosis of Tay-Sachs
disease. Meth Enzymol 1972;28:862–7.
4. Andermann E, Scriver CR, Wolfe LS, Dansky L, Andermann F.
Genetic variants of Tay-Sachs disease: Tay-Sachs disease and
Sandhoff’s disease in French Canadians, juvenile Tay-Sachs disease in Lebanese Canadians, and a Tay-Sachs screening program
in the French-Canadian population. Prog Clin Biol Res 1977;18:
161– 88.
5. Bach G, Navon R, Zeigler M, Beyth Y, Porter B, Cohen MM.
Tay-Sachs disease in a Moroccan Jewish family: a possible new
mutation. Isr J Med Sci 1976;12:1432–9.
6. Kelly TE, Chase GA, Kaback MM, Kumor K, McKusick VA. TaySachs disease: high gene frequency in a non-Jewish population.
Am J Hum Genet 1975;27:287–91.
7. McDowell GA, Mules EH, Fabacher P, Shapira E, Blitzer MG. The
presence of two different infantile Tay-Sachs disease mutations in
a Cajun population. Am J Hum Genet 1992;51:1071–7.
8. Prence EM, Jerome CA, Triggs-Raine BL, Natowicz MR. Heterozygosity for Tay-Sachs and Sandhoff diseases among Massachusetts residents with French Canadian background. J Med Screen
1997;4:133– 6.
398
Martin et al.: Tay-Sachs Carriers in New England
9. Triggs-Raine BL, Richard M, Wasel N, Prence EM, Natowicz MR.
Mutational analyses of Tay-Sachs disease: studies on Tay-Sachs
carriers of French Canadian background living in New England.
Am J Hum Genet 1995;56:870 –9.
10. Karpati M, Gazit E, Goldman B, Frisch A, Colombo R, Peleg L.
Specific mutations in the HEXA gene among Iraqi Jewish TaySachs disease carriers: dating of founder ancestor. Neurogenetics 2004;5:35– 40.
11. Branda KJ, Tomczak J, Natowicz MR. Heterozygosity for Tay-Sachs
and Sandhoff diseases in non-Jewish Americans with ancestry
from Ireland, Great Britain, or Italy. Genet Test 2004;8:174 – 80.
12. Triggs-Raine BL, Feigenbaum AS, Natowicz M, Skomorowski MA,
Schuster SM, Clarke JT, et al. Screening for carriers of Tay-Sachs
disease among Ashkenazi Jews. N Engl J Med 1990;323:6 –12.
13. Kaback M, Lim-Steele J, Dabholkar D, Brown D, Levy N, Zeiger K.
Tay-Sachs disease– carrier screening, prenatal diagnosis, and the
molecular era. An international perspective, 1970 to 1993. The
International TSD Data Collection Network. JAMA 1993;270:
2307–15.
14. Cao ZM, Petroulakis E, Salo T, Triggs-Raine BL. Benign HEXA
mutations, C739T(R247W) and C745T(R249W), cause ␤-hexosaminidase A pseudodeficiency by reducing the ␣ subunit protein
levels. J Biol Chem 1997;272:14975– 82.
15. Triggs-Raine BL, Mules EH, Kaback MM, Lim-Steele JS, Dowling C,
Akerman B, et al. A pseudodeficiency allele common in non-Jewish
Tay-Sachs carriers: implications for carrier screening. Am J Hum
Genet 1992;51:793– 801.
16. Cao Z, Natowicz M, Kaback M, Lim-Steele J, Prence E, Brown D, et
al. A second mutation associated with apparent ␤-hexosaminidase A pseudodeficiency: identification and frequency estimation.
Am J Hum Genet 1993;53:1198 –205.
17. Prence EM, Natowicz MR, Zalewski I. Unusual thermolability
properties of leukocyte ␤-hexosaminidase: implications in screening for carriers of Tay-Sachs disease. Clin Chem 1993;39:
1811– 4.
18. Hoar DI, Haslam DB, Starozik DM. Improved direct molecular
diagnosis and rapid fetal sexing. Prenat Diagn 1984;4:241–7.
19. Myerowitz R, Hogikyan ND. A deletion involving Alu sequences in
the ␤-hexosaminidase ␣-chain gene of French Canadians with
Tay-Sachs disease. J Biol Chem 1987;262:15396 –9.
20. Navon R, Proia RL. The mutations in Ashkenazi Jews with adult
GM2 gangliosidosis, the adult form of Tay-Sachs disease. Science
1989;243:1471– 4.
21. Paw BH, Kaback MM, Neufeld EF. Molecular basis of adult-onset
and chronic GM2 gangliosidoses in patients of Ashkenazi Jewish
origin: substitution of serine for glycine at position 269 of the
␣-subunit of ␤-hexosaminidase. Proc Natl Acad Sci U S A 1989;
86:2413–7.
22. Wicklow BA, Ivanovich JL, Plews MM, Salo TJ, Noetzel MJ, Lueder
GT, et al. Severe subacute GM2 gangliosidosis caused by an
apparently silent HEXA mutation (V324V) that results in aberrant
splicing and reduced HEXA mRNA. Am J Med Genet A 2004;127:
158 – 66.
23. Petroulakis E, Cao Z, Clarke JT, Mahuran DJ, Lee G, Triggs-Raine
B. W474C amino acid substitution affects early processing of the
␣-subunit of ␤-hexosaminidase A and is associated with subacute
GM2 gangliosidosis. Hum Mutat 1998;11:432– 42.
24. Brown CA, Mahuran DJ. ␤-Hexosaminidase isozymes from cells
cotransfected with ␣ and ␤ cDNA constructs: analysis of the
␣-subunit missense mutation associated with the adult form of
Tay-Sachs disease. Am J Hum Genet 1993;53:497–508.
25. Brittingham A, de la Cruz GP. Ancestry: 2000. http://www.census.
gov/prod/2004pubs/c2kbr-35.pdf (Accessed May, 2006).
26. Brault GJ. The French-Canadian Heritage In New England.
Hanover, NH: University Press of New England, 1986.
27. DeBraekeleer M, Hechtman P, Andermann E, Kaplan F. The French
Canadian Tay-Sachs disease deletion mutation: identification of
probable founders. Hum Genet 1992;89:83–7.
28. Hechtman P, Boulay B, DeBraekeleer M, Andermann E, Melancon
S, Larochelle J, et al. The intron 7 donor splice site transition: a
second Tay-Sachs disease mutation in French Canada. Hum
Genet 1992;90:402– 6.
29. Hechtman P, Kaplan F, Bayleran J, Boulay B, Andermann E,
DeBraekeleer M, et al. More than one mutant allele causes
infantile Tay-Sachs disease in French-Canadians. Am J Hum Genet
1990;47:815–22.
30. Fernandes M, Kaplan F, Natowicz M, Prence E, Kolodny E, Kaback
M, et al. A new Tay-Sachs disease B1 allele in exon 7 in two
compound heterozygotes each with a second novel mutation.
Hum Molec Genet 1992;1:759 – 61.
31. Akli S, Chelly J, Lacorte JM, Poenaru L, Kahn A. Seven novel
Tay-Sachs mutations detected by chemical mismatch cleavage of
PCR-amplified cDNA fragments. Genomics 1991;11:124 –34.
32. Trop I, Kaplan F, Brown C, Mahuran D, Hechtman P. A glycine250
to aspartate substitution in the ␣-subunit of hexosaminidase A
causes juvenile-onset Tay-Sachs disease in a Lebanese-Canadian
Family. Hum Mutat 1992;1:35–9.
33. Karpati M, Peleg L, Gazit E, Akstein E, Goldman B. A novel
mutation in the HEXA gene specific to Tay-Sachs disease carriers
of Jewish Iraqi origin. Clin Genet 2000;57:398 – 400.
34. Lemieux MJ, Mark BL, Cherney MM, Withers SG, Mahuran DJ,
James MN. Crystallographic structure of human ␤-hexosaminidase A: interpretation of Tay-Sachs mutations and loss of GM2
ganglioside hydrolysis. J Mol Biol 2006;359:913–29.
35. Hechtman P, Boulay B, Bayleran J, Andermann E. The mutation
mechanism causing juvenile-onset Tay-Sachs disease among
Lebanese. Clin Genet 1989;35:364 –75.
36. Peleg L, Meltzer F, Karpati M, Goldman B. GM2 gangliosidosis B1
variant: biochemical and molecular characterization of hexosaminidase A. Biochem Mol Med 1995;54:126 –32.
37. Proia RL, Neufeld EF. Synthesis of ␤-hexosaminidase in cell-free
translation and in intact fibroblasts: an insoluble precursor ␣
chain in a rare form of Tay-Sachs disease. Proc Natl Acad Sci
U S A 1982;79:6360 – 4.
38. Palomaki GE, Williams J, Haddow JE, Natowicz MR. Tay-Sachs
disease in persons of French-Canadian heritage in northern New
England. Am J Med Genet 1995;56:409 –12.
39. ACOG committee opinion. Number 318, October 2005. Screening
for Tay-Sachs disease. Obstet Gynecol 2005;106:893– 4.
40. Myerowitz R, Costigan FC. The major defect in Ashkenazi Jews
with Tay-Sachs disease is an insertion in the gene for the ␣-chain
of ␤-hexosaminidase. J Biol Chem 1988;263:18587–9.