Molecular Human Reproduction, Vol. 6, No. 2, 197-198,
February 2000
© 2000 European Society of Human Reproduction and Embryology
Letters to the editor |
Hypothesis testing by X chromosome inactivation patterns may be more informative with lineage-specific cells
Associate Professor of Medicine, University of Queensland, Department of Medicine, Mater Adult Hospital, South Brisbane, Queensland 4101, Australia Research Officer, Leukemia Laboratory, Mater Adult Hospital, South Brisbane, Queensland 4101, Australia
Dear Sir,
Fisk et al. (1999) have postulated that twin–twin transfusion syndrome (TTTS), which only occurs in monochorionic diamniotic twin pregnancies, could arise from asymmetry of the splitting of the inner cell mass which forms the twins. The timing of the splitting of the inner cell mass occurs around the same time as the X chromosome inactivation process (Lyonization). In their interesting paper, Fisk et al. used X chromosome inactivation patterns to test their hypothesis that asymmetric splitting predisposes to TTTS. However, they found that TTTS was not significantly associated with skewing of the X chromosome inactivation pattern, which was interpreted as evidence against the hypothesis that Lyonization precedes monochorionic twinning, and they concluded that X chromosome inactivation patterns could not be used to study the symmetry of inner cell mass splitting.
We believe that their conclusions could be unduly pessimistic. Skewing of the X chromosome inactivation pattern has classically been used to study clonality in uniform cell populations (Elliot et al., 1992). Skewing (as well as asymmetry of the skew between twin pairs, i.e. asymmetry of the direction of the skew towards either greater maternal X or greater paternal X inactivation) of the X chromosome inactivation patterns derived from semiquantitative polymerase chain reaction (PCR)-amplified DNA from cell lineages which make up a tissue, could be masked if there were more than one X chromosome inactivation pattern in the different cell lineages of that tissue.
The critical determinants of the X chromosome inactivation pattern will depend on the timing of Lyonization, and the number of progenitor cells from which cell lineages are derived (McLaren, 1972
) with skewing in a random process being more likely if the number of progenitor cells are small (Gale et al., 1994) and conversely, skewing being unlikely with a larger population of progenitor cells. X chromosome patterns are lineage specific because random X chromosome inactivation of the primitive ectoderm (Rastan et al., 1980) does not occur simultaneously but proceeds gradually in the subpopulations of the embryonic ectoderm and mesoderm (Tan et al., 1993), when cells are no longer pluripotent.
So, the X chromosome inactivation pattern of tissue from 5–10 cm of umbilical cord of each twin (Fisk et al., 1999) will depend on the number of progenitor cells at the time of Lyonization from which each cell lineage is derived to make up that tissue, and the mixing, number and size of the different cell lineages which make up that tissue.
If Fisk et al. were to sort lineage-specific cells from their fetal tissue and then study inactivation patterns using similar methodology on pure cell lineages, then we would predict that, if there is asymmetry of the inner cell mass split and twinning post-dates Lyonization, they should be able to detect different magnitudes of skew with the twin from the smaller split more likely to have greater skewed lineages.
If the splitting of the inner cell mass occurred before Lyonization, then the magnitude of the skew of the X-inactivation patterns for each twin pair would tend to be similar for a random process (Nance, 1990) but not symmetric. However, significant skewing, which could occur if Lyonization happened very early in embryogenesis, would be predicted in both twins and might also be symmetric if there were a substantial delay in splitting to permit development of large progenitor cell populations.
References
Elliott, S., Taylor, K. et al. (1992) Proof of differentiative mode of action of all-trans retinoic acid in acute promyelocytic leukemia using X-linked clonal analysis. Blood, 15, 1916–1919.
Fisk, N., Howard, C., Ware, M. and Bennett, P. (1999) X-chromosome inactivation patterns do not implicate asymmetric splitting of the inner cell mass in the aetiology of twin–twin transfusion syndrome. Mol. Hum. Reprod., 5, 52–56.
Gale, R., Wheadon, H., Boulos, P. and Linch, D. (1994) Tissue specificity of X-chromosome inactivation patterns. Blood, 83, 2899–2905.
McLaren, A. (1972). Numerology of development. Nature, 239, 274–6.[Medline]
Nance, W.E. (1990) Do twin Lyons have larger spots? Am. J. Hum. Genet., 46, 646–648.[Web of Science][Medline]
Rastan, S., Kaufman, M., Handyside, A. and Lyons, M. (1980) X-chromosome inactivation in the extraembryonic membranes of diploid parthenogenetic mouse embryos demonstrated by differential staining. Nature, 288, 172–173.[Medline]
Tan, S., Williams, E. and Tam, P. (1993) X-chromosome inactivation occurs at different times in different tissues of the post-implantation mouse embryo. Nat. Genet., 3, 170–174.[Web of Science][Medline]
Department of Maternal and Fetal Medicine, Imperial College School of Medicine, Queen Charlotte's & Chelsea Hospital, Goldhawk Road, London W6 OXG, UK
Dear Sir,
We thank Florin and Taylor for their interest in our study. Although we consider our hypothesis sound, we acknowledged in our report that use of X-inactivation to investigate inner cell mass (ICM) splitting in twin–twin transfusion syndrome (TTTS) was speculative (Fisk et al., 1999). This was necessarily based on assumptions relating to the timing of X-inactivation in humans, and to ICM splitting in twins. Our finding that TTTS was not associated with increased skewing of inactivation patterns is consistent with more recent evidence that X-inactivation occurs 3–4 cell divisions before ICM splitting in monochorionic twins (Monteiro et al., 1998; Chitnis et al., 1999). Monteiro et al.'s finding of reduced skewing in monochorionic twins compared to dichorionic twins may explain the trend in our data towards reduced median percentage skewing in both TTTS and non-TTTS monochorionic twins [for HhaI, 18.8 (range 2.0–87.4), and 12.0 (range 1.0–47.6) respectively] compared with singleton controls [34.6, (range 4.0–65.8)]. If X-inactivation is now considered to precede monochorionic twinning, then reports of discordant X-linked disease in monozygous female `carrier' twins attributed to skewed inactivation must presumably be confined to monozygous dichorionic twins.
Measuring the distribution of X-inactivation skewing has been used to estimate the number of progenitor cells giving rise to an individual tissue (Puck, 1998). Florin and Taylor criticize us for not studying lineage specific cells. Our aim, however, was not to study X-inactivation patterns in individual tissues, but to estimate global skewing in the ICM at around the time of splitting. Accordingly, we studied umbilical cord tissue containing a mixture of endodermal and mesodermally-derived elements. Simulation data suggesting that the number of cells at the time of human X-inactivation is only 5–16 may explain reports of large patch sizes in single tissues such as arterial media and haemopoietic cells (Gale et al., 1994; Chung et al., 1998). Human X-inactivation thus appears to occur earlier than in the mouse. In our paper we acknowledged mouse data suggesting differential timing of inactivation in different tissues (Tan et al., 1993), but referred to the limited human data showing similar patterns in cord, amnion and chorion (Bamforth et al., 1996), and now also buccal mucosa (Monteiro et al., 1998), trophoblast and placental stroma (Looijenga et al., 1999).
References
Bamforth, F., Machin, G. and Innes, M. (1996) X chromosome inactivation is mostly random in placental tissues of female monozygotic twins and triplets. Am. J. Med. Genet., 61, 209–215.[Medline]
Chitnis, S., Derom, C., Vlietinck, R. et al. (1999) X chromosome-inactivation patterns confirm the late timing of monoamniotic-MZ twinning. Am. J. Hum. Genet., 65, 570–571.[Web of Science][Medline]
Chung, I., Schwartz, S. and Murry, C. (1998) Clonal architecture of normal and atherosclerotic aorta: implications for atherogenesis and vascular development. Am. J. Pathol., 152, 913–923.[Abstract]
Fisk, N.M., Howard, C., Ware, M. and Bennett, P. (1999) X-chromosome inactivation patterns do not implicate asymmetric splitting of the inner cell mass in the aetiology of twin–twin transfusion syndrome. Mol. Hum. Reprod., 5, 52–56.
Gale, R., Wheadon, H., Boulos, P. et al. (1994) Tissue specificity of X-chromosome inactivation patterns. Blood, 15, 2899–2905.
Looijenga, L., Gillis, A.J., Verkerk, A.J. et al. (1999) Heterogeneous X inactivation in trophoblast cells of human full-term female placentas. Am. J. Hum. Genet., 64, 1445–1452.[Medline]
Monteiro, J., Derom, C., Vlietinck, R. et al. (1998) Commitment to X inactivation precedes the twinning event in monochorionic MZ twins. Am. J. Hum. Genet., 63, 339–346.[Web of Science][Medline]
Puck, J.M. (1998) The timing of twinning: More insights from X inactivation. Am. J. Hum. Genet., 63, 327–328.[Web of Science][Medline]
Tan, S.S., Williams, E.A. and Tam, P.P. (1993) X-chromosome inactivation occurs at different times in different tissues of the post-implantation mouse embryo. Nature Genet., 3, 170–174.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||