Abstract

New mutations for Huntington disease (HD) arise from intermediate alleles (IAs) with between 29 and 35 CAG repeats that expand on transmission through the paternal germline to 36 CAGs or greater. Using single sperm analysis, we have assessed CAG mutation frequencies for four IAs in families with sporadic HD (IANM) and IAs ascertained from the general population (IAGP) by analyzing 1161 single sperm from three persons. We show that IANM are more unstable than IAGP with identical size and sequence. Furthermore, comparison of different sized IAs and IAs with different sequences between the CAG and the adjacent CCG tracts indicates that DNA sequence is a major influence on CAG stability. These studies provide estimates of the likelihood of expansion of IANM and IAGP to ≥36 CAG repeats for these individuals. For an IA with a CAG of 35 in this family with sporadic HD, the likelihood for siblings to inherit a recurrent mutation ≥36 CAG is ∼10%. For IAGP of a similar size, the risk of inheriting an expanded allele of ≥36 CAG through the paternal germline is ∼6%. These risk estimates are higher than previously reported and provide additional information for counselling in these families. Further studies on persons with IAs will be needed to determine whether these results can be generalized to other families.

Introduction

Huntington disease (HD), a hereditary progressive neurodegenerative disease, is caused by expansion of a CAG trinucleotide in a novel gene located on chromosome 4p16.3 (1). The gene for HD is one of five neurodegenerative disorders caused by amplification of a CAG trinucleotide repeat in the coding region of the gene (2). All new mutations for HD have been shown to arise from IAs containing between 29 and 35 CAG repeats (35). These CAG repeats expand on transmission through the paternal germline to 36 or more repeats. Intermediate alleles (IAs) are present on ∼1% of normal chromosomes of Caucasian descent (5). Affected individuals have an expanded allele of between 36–121 CAGs but incomplete penetrance has been shown for repeat lengths of 36–40 CAGs (68).

In an effort to gather information on the likelihood of recurrence of expansion of IAs in families with sporadic cases of HD as well as of IAs in the general population, we performed family analyses and directly compared CAG changes during transmission from parent to child in these families (5). We found that the IAs in families with sporadic cases were more unstable than IAs in the general population. However, this study was hampered by the small number of meioses examined (31 from families with sporadic HD and 62 in the general population). In view of the sample size, the estimate of stability of IAs in these situations could be inaccurate.

Figure 1.

DNA sequence between the CAG and CCG tract in the HD gene. DNA mutations can occur in this sequence resulting in loss of interruption of the CAG and CCG tracts.

Figure 1.

DNA sequence between the CAG and CCG tract in the HD gene. DNA mutations can occur in this sequence resulting in loss of interruption of the CAG and CCG tracts.

Single sperm analyses of trinucleotide repeats on affected chromosomes (9) as well as studies of affected families (10) have revealed that an increase in CAG repeat size is associated with marked elevation in the frequency of expansion. Therefore, CAG size alone represents a major contributor towards the likelihood of CAG expansion on HD chromosomes.

The sequence of the 12 bp segment between the (CAG)n and adjacent (CCG)n tracts in the HD gene may also significantly contribute to instability of the CAG repeats (5). The wild-type sequence is: (CAG)n CAA CAG CCG CCA (CCG)n. Change of the CAA to CAG and CCA to CCG leading to longer CAG and CCG tracts with no interruptions between them (Fig. 1), has been reported previously in one family with HD (5). Loss of the CAA and CCA interruption resulting in a perfect CAG repeat and CCG repeat are infrequent in HD, in contrast to spinocerebellar ataxia type I (SCA-1) where loss of interruption of the CAG repeat is seen on all affected alleles, and is a major determinant of instability on SCA-1 chromosomes (11,12). Loss of interrupting AGG trinucleotides within the CCG repeat of the FMR-I gene has also been shown to promote instability on fragile X chromosomes (13).

Single sperm analysis represents an important step forward in having the appropriate number of meioses to assess the range of changes in CAG size in persons with different sized IAs. Furthermore, single sperm analysis facilitates quantification of CAG mutation frequency and allows direct comparison of the role of both size and sequence changes on the likelihood of intergenerational change in the HD CAG repeat size.

In an effort to address these questions, we have collected sperm from persons from families with new mutations for HD and, in addition, specifically sought sperm from persons in the general population with equivalent sized IAs. Here we present the results of single sperm analyses of four different alleles. We have first assessed the mutation frequency in two IAs of the same size with wild-type sequence between the CAG and CCG repeat, one of which comes from a person in a family with sporadic HD, and the other ascertained from the general population. In addition, we have analyzed two alleles with the same sequence changes, one derived from an HD new mutation family and the other derived from the general population. These results confirm prior studies and show that IAs from HD new mutation families (IANM) do indeed show a significantly higher CAG mutation frequency than IAs from the general population (IAGP) with identical CAG sizes and sequence between the CAG and CCG repeats. Furthermore, we show that the IAs with changes in the sequence between the CAG and CCG repeats are significantly more unstable than those with the wild-type sequence and similar or greater CAG size.

These studies represent the first study of single sperm from persons with IAs (CAG 29–35) and provide more accurate estimates of the likelihood of expansion to ≥36 CAGs of an IANM compared with an IAGP in these families. These results also provide further insights on the crucial role of DNA sequence for the stability of the CAG repeat and show that the sequence between the CAG and CCG repeat in the HD gene may be an important influence on CAG stability.

Results

Mutation frequency of IAs derived from new mutation families—effect of CAG repeat length and sequence

Single sperm with IAs known to undergo mutation to ≥36 CAGs and manifest with the clinical symptoms of HD (IANM34(36) and IANM35) were analyzed (Table 1). By convention with primers used in predictive testing which only encompass the pure CAG tract and overlaps neighboring sequence including the CAA CAG repeat (10). In individual #2, the new mutation IA has 34 CAG and the IA from the general population has 31 CAG. However, mutations in the sequence between the CAG and CCG tract have effectively resulted in a pure tract of 36 CAGs (IANM34(36)) and 33 CAGs (IAGP31(33)) respectively, in this person (Fig. 1).

For both new mutation IAs, similar proportions of sperm showed expansions of the CAG repeat to ≥36 CAGs (Tables 1 and 2; Fig. 2a and b). In contrast, there were marked differences in the stability of these IAs in sperm. For example, 63.8% (134/210) of the IANM35 sperm showed no change in CAG size (compared with lymphocyte CAG size), whereas only 38.4% (56/146) of the IANM34(36) sperm were stable (P <10−6). Importantly, the difference between IANM35 and IANM34(36) is also reflected in the percentage of sperm that show expanded CAGs [20.5% (43/210) versus 39.7% (58/146)] respectively (P = 0.0001) (Tables 1 and 2).

DNA haplotype analyses (Tables 1 and 3) revealed no differences between IANM34(36) and IANM35 at two intragenic trinucleotide polymorphisms, the CCGn immediately downstream of the CAG repeat (14) and the GAG polymorphism at residue 2642 (15). DNA sequence analysis, however, revealed significant differences. IANM34(36) is from a large family where loss of the CAA and CCA interruptions between the CAG and CCG repeat tracts was demonstrated previously (5). The IANM in this family was cloned and sequenced eight times in three different individuals, and in all instances confirmed the loss of the CAA and CCA interruptions. This suggests that loss of the CAA and CCA interruptions was the major factor in the greater instability of IANM34(36) compared with IANM35, although both IAs are of similar size.

Figure 2.

CAG size distribution of single sperm with intermediate alleles (IAs) which have been shown to be associated with sporadic HD (IANM). CAG repeats in sperm that have expanded to ≥36 CAGs are marked as black bars. The number of sperm is indicated above each bar. (a) IANM35 with the sequence (CAG)35 CAA CAG CCG CCA (CCG)7. (b) IANM34(36) with the sequence (CAG)34 CAG CAG CCG CCG (CCG)7.

Figure 2.

CAG size distribution of single sperm with intermediate alleles (IAs) which have been shown to be associated with sporadic HD (IANM). CAG repeats in sperm that have expanded to ≥36 CAGs are marked as black bars. The number of sperm is indicated above each bar. (a) IANM35 with the sequence (CAG)35 CAA CAG CCG CCA (CCG)7. (b) IANM34(36) with the sequence (CAG)34 CAG CAG CCG CCG (CCG)7.

Table 1.

Single sperm analysis on three persons comprising four intermediate alleles, two derived from NM families and two from the general population

Mutation frequency of IAs in the general population (IAGP)—effect of repeat length and sequence

Two IAGP derived from chromosomes in the general population, i.e. which have not been shown to expand to 36 CAGs or greater, were analyzed. These IAs (IAGP31(33) and IAGP35) have CAG sizes in lymphocytes of 31 and 35 respectively. For IAGP35, 152 of 199 sperm (76.4%) showed no change from the lymphocyte CAG size while 25 sperm (12.6%) had an expanded CAG size and 22 sperm (11%) showed a contraction (Fig. 3a and b).

Table 2.

Comparison of the mutation frequency between different intermediate alleles associated with HD or seen in the general population

Table 3.

Comparison of haplotypes on IANM, HD affected and control chromosomes

Surprisingly, IAGP31(33) was less stable than IAGP35 (46.9% versus 76.4% unchanged from lymphocyte CAG sizes; P <10−6), and displayed a higher frequency of expansions (36.1% versus 12.6%; P <10−6) (Tables 1 and 2; Fig. 3a and b). Sequence analysis revealed that, as with IANM34(36), IAGP31(33) lacks CAA and CCA interruptions between the CAG and CCG tracts. This IAGP was sequenced eight times, using DNA from six different family members and all showed the same uninterrupted CAG and CCG tracts. This strongly suggests that, for this IAGP, loss of the CAA and CCA interruptions results in greater liability to expansion than increased CAG size. The instability of IAGP31(33) was also reflected by pedigree analysis, where four siblings of this individual inherited the same IA from their mother but with a repeat size of 30.

The significance of loss of interruptions for allele instability is highlighted further by the observation that IAGP31(33), although from the general population and thus not known to lead to new mutations for HD, is also even more unstable than IANM35 which has been associated with new mutations for HD (36.1% versus 20.5% of sperm demonstrating expansions; P = 0.0009) (Tables 1 and 2).

IANM is more unstable than IAGP of identical size and sequence

A comparison of the mutation frequency in sperm with IAs of same size and sequence derived either from sporadic HD families (IANM35) or from the general population (IAGP35) reveals that IAGP35 is significantly more stable than IANM35 (76.4% versus 63.8% unchanged; P = 0.008) (Tables 1 and 2). This data confirmed the result of a prior analysis using a much smaller number of meioses from family studies (5). The number of expansions was higher for IANM35 (P = 0.04), again suggesting that IAs in families with sporadic HD have increased rates of expansion compared with similar sized IAs in the general population. Sequence analyses of IANM35 and IAGP35 revealed identical sequences of (CAG)35 (Figs 2a and 3a).

DNA haplotype analysis

The increased instability seen in IANM35 compared with IAGP35, both of which have the same CAG size and sequence between the CAG and CCG tracts, suggests that other factors acting in cis might be responsible for the greater instability of IANM35. Haplotype analysis of the (CCG)n and Δ2642 glutamic acid loci showed that the two IAs differ at the Δ2642 locus (Table 1). In addition, haplotype analysis of 13 IAs from new mutation families revealed that IANM chromosomes have a significantly higher frequency of the glutamic acid Δ2642 deletion (B-allele), compared with control chromosomes (Table 3). As expected, the B-allele is also over-represented on HD chromosomes, supporting the notion that other factors acting in cis may mediate instability and expansion of the HD CAG repeat on these chromosomes.

Mutation frequency of normal alleles (<29 CAG repeats)

A total of 459 sperm, containing normal alleles of identical size (19 CAGs in lymphocytes) from two unrelated individuals (IANM35 and IAGP35), were analyzed (Table 1). The vast majority of these meioses were stable (95.3 and 98.0% respectively). The normal allele of IANM35 showed expansion in 2/215 sperm (0.9%) of 3–6 repeats and contractions in 8/215 sperm (3.7% range 1–7 repeats) (Fig. 2a). The normal allele of IAGP35 contracted in 3/244 sperm (1.2%) and expanded in 2/244 sperm (0.8%) (Fig. 3a). These expansion mutation frequencies are similar to that previously reported for intergenerational CAG expansions in parent-child pairs (3/440 or 0.68%). However, analysis of family data has not previously documented contractions of normal alleles (10), and the basis for contractions in these two particular alleles is unknown.

Discussion

Single sperm analysis provides a unique opportunity for determining mutation frequencies of the CAG repeat in the HD gene. In this study, we have analyzed single sperm carrying both IAs which are known to have expanded to ≥36 CAGs and result in a sporadic case of HD (IANM), and IAs derived from the general population (IAGP35). We identified three informative persons for this study, matching as well as contrasting persons with IAs from new mutation families with IAs derived from the general population. This provides us with new insights concerning the likelihood of expansion to 36 CAGs or greater in both these situations. We have reported previously, based on analysis of 22 meioses in families with sporadic HD, that the likelihood of expansion to ≥36 CAGs is ∼4.5% per meiosis and we proposed that at present, for genetic counselling purposes, the appropriate risk of inheriting a CAG repeat length ≥36 CAGs for a sibling of a sporadic case could be ∼2.25% (5). However, this present study suggests that the risk estimates need to take into account both the size and the sequence of the IA.

Figure 3.

CAG size distribution of single sperm with intermediate alleles (IAs) derived from chromosomes not known to be associated with sporadic HD (IAGP). CAG repeats in sperm that have expanded to ≥35 CAGs are marked as black bars. The number of sperm is indicated above each bar. (a) IAGP35 with the sequence (CAG)35CAA CAG CCG CCA (CCG)7. (b) IAGP31(33) with sequence (CAG)31CAG CAG CCC CCG (CCG)10.

Figure 3.

CAG size distribution of single sperm with intermediate alleles (IAs) derived from chromosomes not known to be associated with sporadic HD (IAGP). CAG repeats in sperm that have expanded to ≥35 CAGs are marked as black bars. The number of sperm is indicated above each bar. (a) IAGP35 with the sequence (CAG)35CAA CAG CCG CCA (CCG)7. (b) IAGP31(33) with sequence (CAG)31CAG CAG CCC CCG (CCG)10.

Analysis of 210 sperm with IANM35, which has a CAG size of 35 and regular DNA sequence with interruptions between the CAG and the CCG tracts, revealed that ∼20% of sperm had expanded to 36 CAGs or greater. Therefore, for offspring of this male, the likelihood of inheriting an expansion mutation would be as high as 10%, assuming no deleterious effect on the capacity of such sperm for fertilization. CAG size plays a major role in determining the likelihood of expansion to ≥36 CAGs. IAs of a smaller CAG size but similar sequence would be expected to have a lower frequency of expansion to ≥36 CAGs. Clearly, additional allele-specific information from single sperm analyses in different families is needed to determine whether risk estimates from these studies can be generalized to other families.

Our results also suggest that the sequence between the CAG and CCG repeat may play a critical role in determining stability of the CAG repeat. Both IANM34(36) and IAGP31(33) show conversion of the CAA and the CCA trinucleotides between the CAG and CCG repeats to a CAG and a CCG respectively, thus resulting in a pure CAG and CCG tracts of larger size. Comparison of IANM35, which has the CAA and CCA interruptions, and IANM34(36) indicates that IANM34(36) was less stable (38.4% versus 63.8% unchanged in size) and had a higher proportion of expansions (39.7% versus 20.5%).

A more dramatic indication of the influence of the sequence between the CAG and CCG tracts on instability is provided by comparison of IAGP31(33), which has uninterrupted CAG and CCG tracts, with IAGP35, which has the CAA and CCA interruptions between the two tracts. IAGP31(33), despite a smaller CAG size, had a significantly greater percentage of expansions compared with IAGP35 (36.1% versus 12.6%) and was less stable (46.9% versus 76.4% unchanged in size; P >10−6). This clearly indicates that an uninterrupted sequence between the CAG and CCG tract has a major influence, greater than CAG size, on increasing the intergenerational stability of these CAGs.

However, it should be noted that loss of the CAA and CCA interruptions on new mutation IA chromosomes is not common. It has occurred in only one of 12 IANM that we have sequenced. Nonetheless, to provide the most accurate estimate of the likelihood of expansion for IAs, sequencing of the CAG and the CCG tracts would provide important additional information for risk assessment.

IAs are relatively frequent in the general population. One large world-wide study of CAG sizes on 2400 normal chromosomes determined the frequency of IAs (29–35 CAGs) to be 0.93% (5). This is equivalent to 1.9% of the general population carrying an HD gene with a CAG repeat size in the iA range. An important question in this situation is whether this particular finding has any clinical relevance in terms of risk to HD offspring. We previously have studied the susceptibility of these IAGP to expand to ≥36 CAGs (5). In that study, the 62 meioses that were assessed showed no expansions to 36 CAGs or greater. The conclusion from that small series was that for persons in the general population found to have IAs, the risk of inheriting a repeat size ≥36 CAGs in offspring was negligible.

This study provides a more quantitative risk assessment for an IAGP with a CAG size close to 36 CAGs. For this IAGP of 35 CAGs in this family, the likelihood of expansion to ≥36 CAGs is expected to be ∼12.6%, with a corresponding 6% risk of transmission to offspring. Incomplete penetrance for HD has been described for individuals with 36–40 CAGs (7) and, therefore, the risk of developing signs and symptoms of HD with an expanded allele in this range is not clear. It is important, however, to note that this information only applies to offspring of fathers with IAs, and that there is very little information concerning offspring of mothers with IAs. What is clear is that the likelihood of expansion from female meiosis is less. In our previous study which analyzed 36 maternal meioses in the general population, 2/36 (5.6%) meioses were unstable (+1 CAG and-2 CAGs) while 3/26 (11.5%) paternal meioses in the general population were unstable (±1 CAG) (5). None of these expansions were up to 36 CAGs or greater. In addition, studies of 18 families with new mutations resulting in HD reveal no instance of a new mutation arising from IAs of mothers (4,5,1620). More detailed family studies of mothers with IAs in the general population will have to be conducted in an effort to reach more precise estimates. Furthermore, the risk estimates for expansion to 36 CAGs or greater provided for IAGP35 as well as for IANM35 might be influenced by possible PCR artifacts which may particularly cause changes of one repeat. Although we considered errors due to PCR artifacts similar for all IAs in the study, the probability of expansion to≥36 CAGs due to artifacts is higher for a larger repeat size. Risk estimates for IAGP35 and IANM35 may, therefore, be somewhat lower than 6 and 10% respectively.

These findings also provide further support for a model of expansion based on DNA structure. The fact that new mutations for HD derive from chromosomes with an intermediate size of the CAG repeat and not from chromosomes with smaller CAG size suggests that length dependence is critical in influencing the likelihood of instability. If polymerase slippage were the major contributor to CAG instability, however, we would expect an equal number of expansion and contraction events. However, as demonstrated in previous studies of HD families (10), and in this and another study of single sperm from families with HD (9), the number of mutations associated with expansion is significantly greater than that associated with contractions. Furthermore, as previously pointed out, slippage alone would not explain the protective effect of sequence interruption in the HD gene as seen in this study (21).

A model for instability that includes DNA structure as a critical element in the mechanism for expansion is the hypothesis that the CAG trinucleotides are capable of forming hairpin structures of high stability (22). A change in DNA sequence causes a shift in the threshold energy necessary to form hairpin structures which has the effect of increasing the likelihood of expansion. Furthermore, the threshold for this instability may also be crossed by increasing the CAG copy number, which again provides some explanation for the increased instability of IAs as CAG size increases (3,22).

However, clearly CAG size and DNA sequence between the CAG and CCG tract do not explain completely the different mutation frequencies on different chromosomes. This point is highlighted in the comparison of the levels of instability of the 35 CAG tract of an IA associated with sporadic HD (IANM35) with that of an IA ascertained through the general population (IAGP35). Here two IAs of the same size and sequence have significantly different likelihoods of instability, with IANM35 demonstrating significantly increased likelihood for expansion. This provides additional support for the hypothesis that a cis-acting sequence (or sequences) in the vicinity of the CAG repeat also may contribute to DNA structure and thus influence the likelihood of expansion. Furthermore, other factors which might influence the likelihood of CAG instability, such as the timing of initial instability in the germline, have not been assessed in this study. The hypothesis of cis-acting factors influencing the likelihood of expansion is strengthened further by the observation of an over-representation of the B-allele of the Δ glutamic acid polymorphism at residue 2642 on new mutation chromosomes as well as on HD chromosomes compared with controls (Table 3). Further studies are needed to define these cis-acting sequences to allow a more comprehensive understanding of mechanisms influencing instability of the HD CAG repeat.

Materials and Methods

Sperm collection

Sperm samples were collected from three unaffected and unrelated individuals carrying four IAs. Individual #1, 51 years of age, has a paternally derived IA of 35 CAG (IANM35) which has led to a new mutation for HD in his sibling. He has wild-type sequence between the CAG and CCG repeat.

Individual #2, 53 years of age, has two IAs, one allele of 34 CAG (IANM34(36)) derived from the unaffected father which has given rise to HD in two siblings and one allele of 31 CAG (IAGP31(33)) maternally derived from the general population. This IAGP allele was found by chance in this family and has not been associated with expansion to 36 CAGs or greater. We have explored further the mother's family and found a similar sized IA in one of her siblings. Both the IANM34(36) and the IAGP31(33) in individual #2 have changes in the sequence between the CAG and CCG tracts. These sequence changes have occurred independently as they are each present in unrelated family members in different sides of the family. By convention using primers as described for predictive testing (10), these persons would be recognized as having CAG 34 and 31 on the IANM and IAGP respectively. However, sequence analysis reveals that, in both instances, these changes have effectively added two CAGs to the prior assessment of CAG size for these two alleles. These alleles were included in the study of Goldberg et al. (5) as 33 CAG and 30 CAG. Later, repeat analysis of this individual revealed that the CAG length had been underestimated by one CAG repeat when compared against CAG repeat tracts of known DNA sequence. Therefore, the correct CAG size based on PCR analysis is 34 and 31 CAG.

Individual #3, 48 years of age, has an IA of 35 CAG (IAGP35) derived from the general population with wild-type sequence between the CAG and CCG tract.

Sperm samples were washed three times in phosphate-buffer saline (PBS), frozen in 4% dimethylsulphoxide (DMSO) and stored at −70°C.

Single sperm isolation and preparation

Single sperm cells were deposited in 96-well pCR trays by cell sorting. Before separation in the flow cytometer, the sperm was stained with the DNA dye Hoechst 33258 (Sigma St. Louis, MO). Frozen sperm samples were thawed at room temperature and diluted 1:2 with Dulbecco's PBS (pH 7.1). An aliquot of 1µl of Hoechst solution (200µg/ml) was added for each 100µl volume of diluted sperm sample. Typically, 300µl of sperm sample was more than enough to adjust the instrument and put single sperm into each well of ten 96-well V-bottom PCR plates. Stained samples were incubated for 1 h at 4°C to allow for dye uptake by the cells.

The sperm was separated with a modular flow sorter (23,24). This instrument is equipped with an XY-stepper table that positions the 96-well trays under the sort stream. The sort-electronics puts one cell in a well and then signals the table to proceed to the next position. The table signals when the new position has been reached. This cycle is repeated until all wells are filled. For these experiments, an Argon-ion single laser was tuned to all lines UV (∼364 nm) at 110 mW. Hoechst fluorescence was detected with a photo-multiplier shielded with a 425 nm long-pass glass absorption filter (Schott) and a 2.5 neutral density filter.

The flow sorter was calibrated with YG fluorescent beads (Polysciences Inc., Warrington, PA). A test target was placed on the stepper table. The stepper table was adjusted so that deflected drops landed exactly at the center of the target circles. After this alignment, the system was checked again by sorting beads into a 96-well V-bottom PCR tray. The tray was inspected with a fluorescent microscope to assure that the beads fell in the tip of the conical tray wells. Only if >90% of the wells had a bead at the very tip, would a sperm sort be performed. After the experiment, the alignment of the instrument was tested in the same manner to assure that the alignment had not drifted during the experiment.

The PCR plates were irradiated with UV light (Stratagene) shortly before each sort. Single sperm were sorted into individual wells containing 5 µl of lysis buffer as described (9).

After sorting, wells were overlayered with 20 µl of liquid wax (Chill-out 14, MJ Research, Inc.). The PCR plates were then incubated at 65°C for 20 min to lyse the cells and cooled at room temperature for 5 min before adding 5µl of neutralization buffer as described (9). The plates were then stored at −20°C until DNA analysis.

Single sperm analysis

A nested PCR strategy was employed to amplify the HD gene CAG and CCG tracts from single sperm. This allowed us to distinguish the two different IAs in individual #2 as the IANM34(36) allele had seven CCGs and the IAGP31(33) was associated with 10 CCGs. Allele sizes were determined by size comparison with CAG repeats of known sequences since commercially available sequence ladders are unreliable and can lead to incorrect assessment of CAG size (data not shown).

The primer pair HD319F (GCGACCCTGGAAAAGCTGATGA) and HD512R (TGAGGCAGCAGCGGCTGT) were used in the first round of amplification. Reactions were performed in a final volume of 50 µl, each containing 1× exo(−) Pfu polymerase buffer [20 mM Tris-HCl (pH 8.75), 10 µM KCl, 10 mM (NH4)2SO4, 2 mM MgSO4, 0.1% Triton X-100, 0.1 mg/ml bovine serum albumin (BSA)], 0.2 µM of each primer, 0.25mM of each dNTP, 10% DMSO and 0.025 U/ml of exo−) Pfu polymerase (Stratagene, La Jolla, CA). Reactions were subjected to an initial denaturation at 98°C for 2min, followed by 25 cycles of denaturation at 98°C for 45 s, annealing at 60°C for 1min and elongation at 72°C for 1 min, with elongation time increasing by 4 s per cycle.

Two µl aliquots of first round PCR product were transferred to fresh reaction tubes containing 48 µl of reagents resulting in a final concentration of 20 mM Tris-HCl (pH 8.75), 10 mM KCl, 10 mM (NH4)2SO4, 2mM MgSO4, 0.1% Triton X-100, 0.1mg/ml BSA, 0. 25 mM of each dNTP, 10% DMSO, 0.025 U/ml of exo(−) Pfu polymerase and 0.2 µM of the primers HD344F-Fam (Fam-CCTTCGAGTCCCTCAAGTCCTTC) and HD483R (CGGCTGAGGAAGCTGAGGAG). Secondary PCR was performed as above with the exception that elongation in the first cycle was for 1min 30 s, increasing by 4 s for each subsequent cycle.

To amplify only the CCG tract, primer HD424F-Hex (Hex-AGCAGCAACAGCCGCC) was used instead of HD344F-Fam in the secondary reaction, using a separate aliquot of the primary reaction product.

Nested PCR products were analyzed on agarose minigels to check for successful amplification and to distinguish normal alleles from IAs. One-fifth of the reaction products were loaded onto 2.5% agarose, 0.5x TBE gels containing 0.1 µg/ml ethidium bromide (or propidium iodide), then electrophoresed at 7.5 V/cm for 45min prior to UV visualization.

For precise sizing of alleles, 0.5 µl of each sample was added to 0.5 µl of Genescan size marker (ABI) and 2 µl of a 5:1 formamide:loading buffer mix.

After denaturation at 94°C for 2min, sample mixes were electrophoresed across a 5% polyacrylamide, 7m urea gel in an ABI 377 fluorescence gene scanner for 2–3 h. Collected data were analyzed using the Genotyper 1.1 software (ABI).

In order to control for any PCR contamination, we included four no-sperm control wells on each 96-well microtiter plate. Contamination was evident in a control well only very rarely, and no data from the particular plate was included. However, it cannot be determined if the normal sized alleles (17–19 CAG) observed in the sperm of individual #2 represent true contractions from the IAs, or are in fact contaminants which did not affect the negative controls.

Any well with more than one allele was also excluded from further analysis. Such multiple alleles can arise if there is an error such as PCR slippage during the initial 1–4 PCR cycles, contamination or multiple sperm in the same well. We considered the impact of a potential low level contamination or errors due to PCR artifacts to be similar for all four alleles studied here, thus allowing comparisons of allele size changes.

Assessment of CAG size

The size of each IA was first determined on DNA derived from lymphocytes (10) and standardized against markers of known sequence and with CAG sizes of 15, 24, 30, 35, 40, 47 and 64 repeats. The mutation frequency and direction of the mutation (expansion or contraction) was assessed by comparing the size of single sperm PCR products with the allele size from the same lymphocyte DNA (Table 1). The affected range is defined as CAG size seen in persons clearly diagnosed with the clinical manifestations of HD (6,7) (Table 1). A total of 1160 sperm were analyzed. For individual #2, CCG analysis was also done on single sperm to distinguish sperm carrying the CCG7 size (IANM34(36)) from sperm with CCG10 size (IAGP31(33)) (Table 1) (14).

Sequence analysis

PCR products spanning both the CAG and CCG repeats were cloned into Invitrogen PCRII-vector and subjected to automated sequence analysis using an ABI 373A DNA sequencer using dye Terminator cycle sequencing.

Haplotype analysis

Two intragenic polymorphic markers, Δ2642 (15) and CCG (14), were used to construct haplotypes where phase could be determined by pedigree studies.

Acknowledgements

This work was supported by the MRC Canada, the Canadian Genetic Diseases Network, the Huntington Society of Canada, MRC Sweden and the Sweden-America Foundation. Dr M.R. Hayden is an established investigator of the B.C. Children's Hospital.

References

1
Huntington's Disease Collaborative Research Group
A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington's disease chromosomes
Cell
 , 
1993
, vol. 
72
 (pg. 
971
-
983
)
2
Ross
C.A.
When more is less: pathogenesis of glutamine repeat neurodegenerative diseases (Minireview)
Neuron
 , 
1995
, vol. 
15
 (pg. 
493
-
496
)
3
Goldberg
Y.P.
Kremer
B.
Andrew
S.E.
Theilmann
J.
Graham
R.K.
Squitieri
F.
Telenius
H.
Adam
S.
Sajoo
A.
Starr
E.
Heiberg
A.
Wolff
G.
Hayden
M.R.
Molecular analysis of new mutations for Huntington's disease: intermediate alleles and sex of origin effects
Nature Genet.
 , 
1993
, vol. 
5
 (pg. 
174
-
179
)
4
Myers
R.H.
MacDonald
M.E.
Koroshetz
W.J.
Duyao
M.P.
Ambrose
C.M.
Taylor
S.A.M.
Barnes
G.
Srinidhi
J.
Lin
C.S.
Whaley
W.L.
Lazzarini
A.M.
Schwarz
M.
Wolff
G.
Bird
E.D.
Vonsattel
J.-P.G.
Gusella
J.F.
De novo expansion of a (CAG)n repeat in sporadic Huntington's disease
Nature Genet.
 , 
1993
, vol. 
5
 (pg. 
168
-
173
)
5
Goldberg
Y.P.
McMurray
C.T.
Zeisler
J.
Almqvist
A.
Sillence
D.
Richards
F.
Gacy
A.M.
Buchanan
J.
Telenius
H.
Hayden
M.R.
Increased instability of intermediate alleles in families with sporadic Huntington disease compared to similar sized intermediate alleles in the general population
Hum. Mol. Genet.
 , 
1995
, vol. 
4
 (pg. 
1911
-
1918
)
6
Kremer
B.
Goldberg
P.
Andrew
S.E.
Theilmann
J.
Telenius
H.
Zeisler
J.
Squitieri
F.
Lin
B.
Bassett
A.
Almqvist
E.
Bird
T.D.
Hayden
M.R.
A worldwide study of the Huntington's disease mutation
The sensitivity and specificity of measuring CAG repeats. N. Engl. J. Med.
 , 
1994
, vol. 
330
 (pg. 
1401
-
1406
)
7
Rubinsztein
D.C.
Leggo
J.
Coles
R.
Almqvist
E.
Biancalana
V.
Cassiman
J.-J.
Chotai
K.
Connarty
M.
Craufurd
D.
Curtis
A.
Curtis
D.
Davidson
M.J.
Differ
A.-M.
Connarty
M.
Craufurd
D.
Curtis
A.
Curtis
D.
Davidson
M.J.
Differ
A.-M.
Dode
C.
Dodge
A.
Frontali
M.
Ranen
N.G.
Stine
O.C.
Sherr
M.
Abbott
M.H.
Franz
M.L.
Graham
C.A.
Harper
P.S.
Hedreen
J.C.
Jackson
A.
Kaplan
J.-C.
Losekoot
M.
MacMillan
J.C.
Morrison
P.
Trottier
Y.
Novelletto
A.
Simpson
S.A.
Theilmann
J.
Whittaker
J.L.
Folstein
S.E.
Ross
C.A.
Hayden
M.R.
with 36–39 repeats
Am. J. Hum. Genet.
 , 
1996
, vol. 
59
 (pg. 
16
-
22.
)
8
Nance
M.A.
Huntington disease-another chapter rewritten
Editorial review. Am. J. Hum. Genet.
 , 
1996
, vol. 
59
 (pg. 
1
-
6
)
9
Leeflang
E.P.
Zhang
L.
Tavar
S.
Hubert
R.
Srinidhi
J.
MacDonald
M.C.
Myers
R.H.
de Young
M.
Wexler
N.S.
Gusella
J.F.
Arnheim
N.
Single sperm analysis of the trinucleotide repeats in the Huntington's disease gene: quantification of the mutation frequency spectrum
Hum. Mol. Genet.
 , 
1995
, vol. 
4
 (pg. 
1519
-
1526
)
10
Kremer
B.
Almqvist
E.
Theilmann
J.
Spence
N.
Telenius
H.
Goldberg
Y.P.
Hayden
M.R.
Sex-dependent mechanisms for expansions and contractions of the CAG repeat on affected Huntington disease chromosomes
Am. J. Hum. Genet.
 , 
1995
, vol. 
57
 (pg. 
343
-
350
)
11
Chung
M.-Y.
Ranum
L.P.W.
Duvick
L.A.
Servadio
A.
Zoghbi
H.Y.
Orr
H.T.
Evidence for a mechanism predisposing to intergenerational CAG repeat instability in spinocerebellar ataxia type 1
Nature Genet.
 , 
1993
, vol. 
5
 (pg. 
254
-
258
)
12
Chong
S.S.
McCall
A.E.
Cota
J.
Subramony
S.H.
Orr
H.T.
Hughes
M.R.
Zoghbi
H.Y.
Gametic and somatic tissue-specific heterogeneity of the expanded SCA1 CAG repeat in spinocerebellar ataxia type 1
Nature Genet
 , 
1995
, vol. 
10
 (pg. 
344
-
350
)
13
Eichler
E.E.
Holden
J.J.A.
Popovich
B.W.
Reiss
A.L.
Snow
K.
Thibodeau
S.N.
Richards
C.S.
Ward
P.A.
Nelson
D.L.
Length of uninterrupted CGG repeats determines instability in the FMR1 gene
Nature Genet.
 , 
1994
, vol. 
8
 (pg. 
88
-
95
)
14
Andrew
S.E.
Goldberg
Y.P.
Theilmann
J.
Zeisler
J.
Hayden
M.R.
A CCG repeat polymorphism adjacent to the CAG repeat in the Huntington disease gene: implications for diagnostic accuracy and predictive testing
Hum. Mol. Genet.
 , 
1994
, vol. 
3
 (pg. 
65
-
67
)
15
Ambrose
C.M.
Duyao
M.P.
Barnes
G.
Bates
G.P.
Lin
C.S.
Srinidhi
J.
Baxendale
S.
Hummerich
H.
Lehrach
H.
Altherr
M.
Wasmuth
J.
Bucklar
A.
Church
D.
Housman
D.
Berks
M.
Micklem
G.
Durbin
R.
Dodge
A.
Read
A.
Gusella
J.
MacDonald
M.E.
Structure and expression of the Huntington's disease gene: evidence against simple inactivation due to an expanded CAG repeat
Somat. Cell Mol. Genet.
 , 
1994
, vol. 
20
 (pg. 
27
-
34
)
16
ZYhlke
C.
Riess
O.
Shrsder
K.
Siedlaczck
I.
Epplen
J.T.
Engel
W
Thies
U.
Expansion of the (CAG)n repeat causing Huntington's disease in 352 patients of German origin
Hum. Mol. Genet.
 , 
1993
, vol. 
2
 (pg. 
1467
-
1469
)
17
De Rooij
K.E.
De Koning Gans
P.A.M.
Straastad
M.I.
Belfroid
R.D.M.
Vegter-Van Der Vlis
M.
Roos
R.A.C.
Bakker
E.
Van Ommen
G.-J.B.
Den Dunnen
J.T.
Losekoot
M.
Dynamic mutation in Dutch Huntington's disease patients: increased paternal repeat instability extending to within the normal size range
J. Med. Genet.
 , 
1993
, vol. 
30
 (pg. 
996
-
1002
)
18
Davis
M.B.
Bateman
D.
Quinn
N.P.
Marsden
C.D.
Harding
A.E.
Mutation analysis in patients with possible but apparently sporadic Huntington' disease
Lancet
 , 
1994
, vol. 
344
 (pg. 
714
-
717
)
19
Legius
E.
Cuppens
H.
Dierick
H.
Van Zandt
K.
Dom
R.
Fryns
J.-P.
Evers-Kiebooms
G.
Decruyenaere
M.
Demyttenaere
K.
Marynen
P.
Cassiman
J.-J.
Limited expansion of the (CAG)n repeat of the Huntington gene: a premutation (?)
Eur. J. Hum. Genet
 , 
1994
, vol. 
2
 (pg. 
44
-
50
)
20
McGlennen
R.C.
Allison
P.S.
Hagen
V.L.M.
Parker
T.L.
Lovell
M.A.
Kelly
T.E.
Evidence of an unstable paternal 27 CAG repeat allele in the Huntington gene giving rise to clinically overt Huntington disease in a patient with the genotype (17/38)
Am. J. Hum. Genet
 , 
1995
, vol. 
57
 pg. 
A1423
 
21
McMurray
C.T.
Mechanisms of DNA expansion
Chromosoma
 , 
1995
, vol. 
104
 (pg. 
2
-
13
)
22
Gacy
A.M.
Goellner
G.
Juranic
N.
Macura
S.
McMurray
C.T.
Trinucleotide repeats that expand in human disease form hairpin structures in vitro
Cell
 , 
1995
, vol. 
31
 (pg. 
533
-
540
)
23
van den Engh
G.
Stokdijk
W.
Parallel processing data acquisition system for multi-laser flow cytometry and cell sorting
Cytometry
 , 
1989
, vol. 
10
 (pg. 
282
-
293
)
24
van den Engh
G.J.
Farmer
C.
Photo-bleaching and photon saturation in flow cytometry
Cytometry
 , 
1992
, vol. 
13
 (pg. 
669
-
677
)
25
Squitieri
F.
Andrew
S.E.
Goldberg
Y.P.
Kremer
B.
Spence
N.
Zeisler
J.
Nichol
K.
Theilmann
J.
Greenberg
J.
Goto
J.
Kanazawa
I.
Vesa
J.
Peltonen
L.
Almqvist
E.
Anvret
M.
Telenius
H.
Lin
B.
Napolitano
G.
Morgan
K.
Hayden
M.R.
DNA haplotype analysis of Huntington disease reveals clues to the origins and mechanisms of CAG expansion and reasons for geographic variations of prevalence
Hum. Mol. Genet.
 , 
1994
, vol. 
3
 (pg. 
2103
-
2114
)
26
Almqvist
E.
Spence
N.
Nichol
K.
Andrew
S.E.
Vesa
J.
Peltonen
L.
Anvret
M.
Goto
J.
Goldberg
Y.P.
Hayden
M.R.
Ancestral differences in the distribution of the 2642 glutamic acid polymorphism is associated with varying CAG repeat lengths on normal chromosomes: insights into the genetic evolution of Huntington disease
Hum. Mol. Genet.
 , 
1995
, vol. 
4
 (pg. 
207
-
214
)