Advancing maternal age has long been identified as the primary risk factor for human chromosome trisomy. More recently, altered patterns of meiotic recombination have been found to be associated with non-disjunction. We have used trisomy 21 as a model for human non-disjunction that occurs during the formation of oocytes to understand the role of maternal age and recombination. Patterns of recombination that increase the risk for non-disjunction of chromosome 21 include absence of any exchange, an exchange near the centromere or a single, telomeric exchange. Our recent work has shown that different susceptibility patterns are associated with the origin of the meiotic error and maternal age. For MI (meiosis I) errors, the proportion of oocytes with susceptible recombination patterns is highest among young mothers and decreases significantly in the oldest age group. In fact, the pattern of exchanges among the oldest age group mimics the pattern observed among normally disjoining chromosomes 21. These results suggest that oocytes of younger women, with functional meiotic apparatus and/or robust ovarian environment, are able to properly resolve all but the most susceptible exchange patterns. As women age, however, meiotic mechanisms erode, making it difficult to resolve even stable exchange events. Interestingly, our preliminary recombination results on MII errors reveal the opposite relationship with maternal age: susceptible pericentromeric exchanges occur most often in the older age group compared with the younger age group. If confirmed, we will have further evidence for multiple risk factors for non-disjunction that act at different times in the meiotic process.

Introduction

Improper chromosome segregation, or non-disjunction, during meiosis leads to genetically unbalanced eggs or spermatozoa. If fertilized, such gametes result in embryos with aneuploidy, having either one chromosome too many (trisomy) or too few (monosomy). In humans, aneuploidy is identified in at least 5% of all clinically recognized pregnancies, making it the leading known cause of pregnancy loss. One way to study the causes of non-disjunction in humans is to study individuals with trisomy 21 (or Down's syndrome), one of the few trisomic conceptuses that make it to term.

The ability to accurately determine the parental origin of the non-disjunction error and the type of error [MI (meiosis I), MII (meiosis II) or mitotic] has enhanced the study of risk factors for trisomy 21. The overwhelming majority of non-disjunction errors leading to trisomy 21 originate in MI. In fact, this is true for most human chromosome trisomies. This is not surprising: the first stage of female meiosis is initiated in the fetal ovary and ‘arrested’ in this stage until that oocyte is ovulated. Thus the first meiotic division is significantly protracted, taking at least 10–15 years and as many as 45–50 years to complete.

Maternal age

The age of the mother at the time of the conception of a fetus with trisomy 21 is, by far, the most significant risk factor for meiotic non-disjunction of chromosome 21. As a woman ages, her risk for having a conceptus with trisomy 21 significantly increases. This effect of an increased rate of Down's syndrome with advancing maternal age was noted by Penrose [1] in 1933. Although no specific explanation has come to the forefront, significant progress has been made towards characterizing this effect and understanding possible mechanisms.

Table 1 shows the mean maternal age at the time of the birth of a trisomy 21 infant for each type of chromosome error taken from a large population-based study of infants (Atlanta Down Syndrome Project 1989–2002, unpublished work). Important conclusions can be drawn. First, the maternal age effect is restricted to mothers in whom the non-disjunction error occurred. That is, an increased maternal age is not observed among mothers of fetuses that received the extra chromosome 21 through a non-disjunction error in spermatogenesis (paternal errors) or had a post-zygotic mitotic error. Secondly, advanced maternal age is a risk factor for both MI and MII maternal non-disjunction errors (e.g. Table 1, which updates results of [2]). This observation potentially ties together the MI and MII errors with respect to risk factors. In thinking about possible mechanisms to explain the maternal age effect, the most obvious include: (i) an accumulation of toxic effects of the environment during the arrested state of the oocyte; (ii) a degradation of meiotic machinery over time while in the arrested state, leading to a suboptimal resumption of MI and MII; (iii) a change in ovarian functioning due to suboptimal hormonal signalling; and (iv) degradation of the uterine environment. Most likely, several processes are affected by advanced maternal age and thus more than one of the various hypotheses proposed to explain this effect will be correct. Gaulden [3] and Eichenlaub-Ritter [4] provide excellent reviews of the hypotheses proposed to explain the maternal age effect. It is clear that such hypotheses need to focus on both the processes directly affecting oogenesis and the environment in which oocytes are formed in an aging woman.

Table 1
Origin of chromosome 21 non-disjunction error and mean maternal age

Unpublished results from the Atlanta Down Syndrome Project, 1989–2002.

Parent of origin Meiotic error Number of cases Percentage of error type Maternal age (mean±S.D.) 
Maternal MI 240 MI/(MI+MII)=240/311=77.2% 30.98±6.81 
 MII 71 MII/(MI+MII)=71/311=22.8% 31.44±7.60 
   Maternal/All=311/348=89.4%  
Paternal MI 12 PI/(PI+PII)=12/22=54.5% 28.50±7.51 
 MII 10 PII/(PI+PII)=10/22=45.5% 26.00±5.39 
   Paternal/All=22/348=6.3%  
Mitotic  15 Mitotic/All=15/348=4.3% 29.73±5.06 
Controls  493  27.50±6.11 
Parent of origin Meiotic error Number of cases Percentage of error type Maternal age (mean±S.D.) 
Maternal MI 240 MI/(MI+MII)=240/311=77.2% 30.98±6.81 
 MII 71 MII/(MI+MII)=71/311=22.8% 31.44±7.60 
   Maternal/All=311/348=89.4%  
Paternal MI 12 PI/(PI+PII)=12/22=54.5% 28.50±7.51 
 MII 10 PII/(PI+PII)=10/22=45.5% 26.00±5.39 
   Paternal/All=22/348=6.3%  
Mitotic  15 Mitotic/All=15/348=4.3% 29.73±5.06 
Controls  493  27.50±6.11 

Recombination

Aside from maternal age, there is only one other factor that has been shown to be associated with an increased susceptibility of maternal non-disjunction, namely altered recombination patterns. Warren et al. [5] provided the first evidence to suggest that a proportion of maternal non-disjunction errors was associated with reduced recombination along chromosome 21. Further examination has shown that, in addition to the absence of an exchange along the non-disjoined chromosome 21, the placement of an exchange is an important susceptibility factor for non-disjunction [6].

Briefly, examination of recombination along the maternal non-disjoined chromosome 21 has suggested three susceptibility exchange patterns: (i) no exchange leads to an increased risk of MI errors, (ii) a single telomeric exchange leads to an increased risk of MI errors and (iii) a pericentromeric exchange leads to an increased risk of so-called MII errors. These patterns are similar to those observed in model organisms: absent or reduced levels of recombination, along with suboptimally placed recombinant events, increase the likelihood of non-disjunction [713]. Exchanges too close to the centromere or single exchanges too close to the telomere seem to confer the most instability.

The association of maternal MII errors with a specific recombination pattern suggests that at least some proportion of MII errors are initiated in MI. Perhaps, the presence of a pericentromeric exchange increases the likelihood of chromosome ‘entanglement’ or premature sister chromatid separation at MI, with the resulting disomic gamete having identical centromeres; such an error would be scored as originating at MII even though the precipitating event occurred at MI. Thus, at least for chromosome 21, maternal ‘MII’ errors may have their genesis in MI. Instead of changing nomenclature, we will simply use the designation ‘MII’ to indicate this suggestion and refer to the ‘type’ of meiotic error instead of the ‘stage’ of meiotic error.

Recombination and maternal age

The link between recombination and human non-disjunction prompts the obvious question: what is the association between altered recombination and the only other known predisposing factor for trisomy, increasing maternal age? That is, does either the frequency or the location of exchanges vary with the age of the mother? Previous studies of trisomy 21 failed to identify any association; however, the sample size was relatively small with respect to the amount of variation in recombination along this small chromosome [14]. An examination of chromosome 15 non-disjunction provided the first evidence of an age–recombination relationship: Robinson et al. [15] found that, among maternal MI-derived errors, the age of the mother was significantly increased among cases with multiple recombinants compared with those having zero or only one detectable recombinant. From this, Robinson et al. [15] suggested that cases with multiple recombinants might be more resistant to non-disjunction because of increased stability of the bivalent over time. Similarly, an analysis of maternal non-disjunction of the X chromosome showed that the mean maternal age of cases with recombination was significantly older than that of cases with no recombination [16]. This same pattern was observed for trisomy 18, although the difference was not statistically significant [16].

Recently, we updated our original reports using a larger population of 400 maternal MI non-disjoined cases and a more refined genetic analysis [17]. Cases were subdivided into three groups based on the age of the mother at the time of birth of their offspring with Down's syndrome: mothers younger than 29 years of age (n=126), mothers from 29 to 34 years of age (n=138) and mothers 35 years of age or older (n=136). Even with this increased sample size, the frequency distributions of the number of exchanges within each age group were not significantly different from each other. For example, 52% of bivalents with no exchange were observed in the youngest group, 29% in the middle-aged group and 45% in the oldest group. Because the variation in number of exchange events along chromosome 21 is small, more data are needed to determine whether there is any association with maternal age and the frequency of exchange along chromosome 21.

Interestingly, patterns differed significantly among age groups with respect to the location of the exchanges. The proportion of cases with susceptible exchanges (pericentromeric or single telomeric) was highest among the young group of mothers and lowest among the older group. In fact, the pattern of exchanges among the oldest age group began to mimic the pattern observed among normally disjoining chromosomes 21. For example, among the youngest age group, nearly 80% of the single exchanges occurred in the most telomeric interval compared with 33 and 14% among the middle and oldest age groups respectively. These distributions were significantly different from the normally disjoining sample for all three trisomic age groups. However, the level of significance declined with increasing age.

One plausible explanation for these findings suggests that multiple pathways lead to non-disjunction, some age dependent and others age independent. In a young woman, meiotic machinery (spindle function, sister chromatid adhesive proteins, microtubule motor proteins etc.) functions optimally and, as a result, can correctly segregate all but the most susceptible exchange configurations. For young women then, the greatest risk factor for MI non-disjunction is the presence of a susceptible exchange pattern in the oocyte. As a woman ages, her meiotic machinery is exposed to an accumulation of environmental and age-related insults, becoming less efficient/more error-prone. Suboptimal exchange patterns still increase susceptibility to non-disjunction, but now even bivalents with correct exchanges are at risk. Over time, the proportion of non-disjunction due to normal exchange configurations increases as age-dependent risk factors exert their influence. As a result, the most prevalent exchange profile of non-disjoined oocytes shifts from susceptible to non-susceptible patterns with age of the oocyte.

If ‘MII’ errors are initiated in MI, exchange patterns among maternal age groups with ‘MII’ errors are predicted to be similar to those observed for MI errors. Preliminary results suggest that this is not the case (S.L. Sherman, N.E. Lamb and E. Feingold, unpublished work). In the limited study sample (∼40 cases in each age group), the amount of recombination significantly decreased with increasing maternal age for ‘MII’ errors (P<0.01). For example, the mean age of women with an ‘MII’ error and one observed recombinant was 32.8 years, whereas the mean maternal age of those with two or more recombinants was 28.2 years. Moreover, the proportion of susceptible exchanges increased with age, the opposite pattern to that observed in MI. Whether these results are robust remains to be seen. If these results are confirmed, the apparent difference in MI and ‘MII’ maternal age-associated exchange patterns may provide insight into the non-disjunction mechanism and associated risk factors.

Summary

Significant progress has been made in understanding non-disjunction of chromosome 21, the most common cause of Down's syndrome. Over the past 10 years, the genetic tools available have increased the ability to accurately separate cases into discrete groups based on the parent of origin and type of non-disjunction error. Clearly, maternal age and altered recombination remain the only well-established risk factors for non-disjunction of chromosome 21. Nevertheless, additional risk factors for this multifactorial trait will be identified and progress towards understanding the effect of maternal age on the meiotic process will be made given the advances in technology, publicly available genomic resources and interdisciplinary approaches to these important studies.

Meiosis and the Causes and Consequences of Recombination: Biochemical Society Focused Meeting held at University of Warwick, U.K., 29–31 March 2006. Organized by R. Borts (Leicester, U.K.), D. Charlesworth (Edinburgh, U.K.), A. Eyre-Walker (Sussex, U.K.), A. Goldman (Sheffield, U.K.), G. McVean (Oxford, U.K.), D. Monckton (Glasgow, U.K.), G. Moore (John Innes Centre, U.K.), J. Richards (Roslin Biocentre, U.K.) and M. Stark (Glasgow, U.K.). Edited by D. Monckton.

Abbreviations

     
  • MI

    meiosis I

This work was supported by NIH (National Institutes of Health; Bethesda, MD, U.S.A.) R01 HD38979 and by NIH/NCRR (National Center for Research Resources) M01 RR00039.

References

References
1
Penrose
L.S.
J. Genet.
1933
, vol. 
27
 (pg. 
219
-
224
)
2
Yoon
P.W.
Freeman
S.B.
Sherman
S.L.
Taft
L.F.
Gu
Y.
Pettay
D.
Flanders
W.D.
Khoury
M.J.
Hassold
T.J.
Am. J. Hum. Genet.
1996
, vol. 
58
 (pg. 
628
-
633
)
3
Gaulden
M.E.
Mutat. Res.
1992
, vol. 
296
 (pg. 
69
-
88
)
4
Eichenlaub-Ritter
U.
Environ. Mol. Mutagen.
1996
, vol. 
28
 (pg. 
211
-
236
)
5
Warren
A.C.
Chakravarti
A.
Wong
C.
Slaugenhaupt
S.A.
Halloran
S.L.
Watkins
P.C.
Metazotou
C.
Science
1987
, vol. 
237
 (pg. 
652
-
654
)
6
Lamb
N.E.
Feingold
E.
Savage
A.
Avramopoulos
D.
Freeman
S.
Gu
Y.
Hallberg
A.
Hersey
J.
Karadima
G.
Pettay
D.
, et al. 
Hum. Mol. Genet.
1997
, vol. 
6
 (pg. 
1391
-
1399
)
7
Koehler
K.E.
Boulton
C.L.
Collins
H.E.
French
R.L.
Herman
K.C.
Lacefield
S.M.
Madden
L.D.
Schuetz
D.
Hawley
R.S.
Nat. Genet.
1996
, vol. 
14
 (pg. 
406
-
413
)
8
Ross
L.O.
Maxfield
R.
Dawson
D.
Proc. Natl. Acad. Sci. U.S.A.
1996
, vol. 
93
 (pg. 
4979
-
4983
)
9
Sears
D.D.
Hegemann
J.H.
Shero
J.H.
Hieter
P.
Genetics
1995
, vol. 
139
 (pg. 
1159
-
1173
)
10
Rasooly
R.S.
New
C.M.
Zhang
P.
Hawley
R.S.
Baker
B.S.
Genetics
1991
, vol. 
129
 (pg. 
409
-
422
)
11
Moore
D.P.
Miyazaki
W.Y.
Tomkiel
J.E.
Orr-Weaver
T.L.
Genetics
1994
, vol. 
136
 (pg. 
953
-
964
)
12
Krawchuk
M.D.
Wahls
W.P.
Genetics
1999
, vol. 
153
 (pg. 
49
-
55
)
13
Zetka
M.C.
Rose
A.M.
Genetics
1995
, vol. 
141
 (pg. 
1339
-
1349
)
14
Lamb
N.E.
Freeman
S.B.
Savage-Austin
A.
Pettay
D.
Taft
L.
Hersey
J.
Gu
Y.
Shen
J.
Saker
D.
May
K.M.
, et al. 
Nat. Genet.
1996
, vol. 
14
 (pg. 
400
-
405
)
15
Robinson
W.
Kuckinka
B.D.
Bernascoi
F.
Brondum-Neilsen
K.
Christian
S.
Horsthemke
B.
Langlois
S.
Ledbetter
D.
Michaelis
R.
Petersen
M.
, et al. 
Hum. Mol. Genet.
1998
, vol. 
7
 (pg. 
1011
-
1019
)
16
Thomas
N.S.
Ennis
S.
Sharp
A.J.
Durkie
M.
Hassold
T.J.
Collins
A.R.
Jacobs
P.A.
Hum. Mol. Genet.
2001
, vol. 
10
 (pg. 
243
-
250
)
17
Lamb
N.E.
Yu
K.
Shaffer
J.
Feingold
E.
Sherman
S.L.
Am. J. Hum. Genet.
2005
, vol. 
76
 (pg. 
91
-
99
)