Electronic Letters to:

Endocrine Care: Liora Sagi, Nehama Zuckerman-Levin, Aneta Gawlik, Lucia Ghizzoni, Atilla Buyukgebiz, Yardena Rakover, Tzvi Bistritzer, Osnat Admoni, Alessandra Vottero, Oshrat Baruch, Fuad Fares, Ewa Malecka-Tendera, and Ze’ev Hochberg Clinical Significance of the Parental Origin of the X Chromosome in Turner Syndrome J Clin Endocrinol Metab 2007; 92: 846-852 [Abstract] [Full text] [PDF]

eLetters: Submit a response to this article

Electronic letters published:

Reply: Parental Origin of the X chromosome in Turner Syndrome

Ze'ev Hochberg, Ada Tamir and Nehama Zuckerman-Levin   (13 August 2007)

Clinical significance of X-linked genomic imprinting

Carolyn A. Bondy, Vladimir K. Bakalov   (1 August 2007)

Reply: Parental Origin of the X chromosome in Turner Syndrome 13 August 2007
Ze'ev Hochberg, Division Chief, Pediatric Endocrinology Meyer Children's Hospital, Haifa, Israel, Ada Tamir and Nehama Zuckerman-Levin Send letter to journal: Re: Reply: Parental Origin of the X chromosome in Turner Syndrome z_hochberg{at}rambam.health.gov.il Ze'ev Hochberg, et al.	Bondy and Bakalov are disappointed to find differences between their own account (1, 2) and ours (3) on potential X-linked genomic imprinting by comparing phenotypic features in Turner syndrome patients divided into maternal X (Xm) vs. paternal X (Xp) groups. As pediatricians, we are never surprised to find differences between subjects with a mean age of 33 years and those with a mean age of 15 years. In the study of Turner syndrome, age has an even greater impact, as these women become overweight unless they are carefully advised about diet and physical activity. Bondy and Bakalov's patients had a mean BMI of 26 kg/m2, corresponding to ~+2 SDS, as compared to our group with a mean BMI SDS of +1.2. It is interesting that LDL levels in the two age groups were similar for Xp subjects, and the difference (possible imprinting or BMI-dependent) lie with higher levels in adults’ Xm group only. The results of these important studies suggest that the X-chromosome might imprint for lipoproteins levels that change over patients’ life span. We have recalculated our data, and reconfirm the total and LDL cholesterol findings as published. We disagree with the doubts raised whether ANCOVA can control for significant differences in age and BMI in a sample size of 78 patients. This sample size detects difference of 0.75 SD even when controlling for age and BMI. Restricted age group unpublished analysis supported the ANCOVA result. The difference in our respective findings with regard to renal malformations is difficult to explain. The types of malformations were similar in the two groups; the vast majority in both studies had double collecting systems or horseshoe kidneys. All other malformations comprised 1-2 cases each. Our finding of 0 cases in the Xp group is statistically very strong. Retrospective analysis of pooled published data of 208 cases (4-7), including ours (but not Bondy and Bakalov’s that were published later) supports our finding: 34 out of 152 Xm patients were affected, as compared to 5/51 Xp cases (p = 0.049). Whether this is “expected” or not depends on one’s expectations; our approach has been empirical. The eye disorders that we found to be imprinted are amblyopia, strabismus and ptosis. These disorders are pathogenetically related, and color blindness was not part of the Xp phenotype, as implied by Bondy and Bakalov. We certainly agree that research efforts should always be driven by biologically plausible and targeted hypotheses and executed by sound statistical methods. We would add humility in the face of empirical data and in postulating hypotheses on issues like X-chromosome imprinting, where knowledge is so limited. References 1. Bondy CA, Matura LA, Wooten N, Troendle J, Zinn AR, Bakalov VK 2007 The physical phenotype of girls and women with Turner syndrome is not X- imprinted. Hum Genet 121:469-474 2. Van PL, Bakalov VK, Zinn AR, Bondy CA 2006 Maternal X chromosome, visceral adiposity, and lipid profile. JAMA 295:1373-1374 3. Sagi L, Zuckerman-Levin N, Gawlik A, Ghizzoni L, Buyukgebiz A, Rakover Y, Bistritzer T, Admoni O, Vottero A, Baruch O, Fares F, Malecka-Tendera E, Hochberg Z 2007 Clinical significance of the parental origin of the X chromosome in turner syndrome. J Clin Endocrinol Metab 92:846-852 4. Ross JL, Feuillan P, Long LM, Kowal K, Kushner H, Cutler GB, Jr. 1995 Lipid abnormalities in Turner syndrome. J Pediatr 126:242-245 5. Mathur A, Stekol L, Schatz D, MacLaren NK, Scott ML, Lippe B 1991 The parental origin of the single X chromosome in Turner syndrome: lack of correlation with parental age or clinical phenotype. Am J Hum Genet 48:682-686 6. Lorda-Sanchez I, Binkert F, Maechler M, Schinzel A 1992 Molecular study of 45,X conceptuses: correlation with clinical findings. Am J Med Genet 42:487-490 7. Chu CE, Donaldson MD, Kelnar CJ, Smail PJ, Greene SA, Paterson WF, Connor JM 1994 Possible role of imprinting in the Turner phenotype. J Med Genet 31:840-842
Clinical significance of X-linked genomic imprinting 1 August 2007
Carolyn A. Bondy NICHD, NIH, Bethesda, MD, Vladimir K. Bakalov Send letter to journal: Re: Clinical significance of X-linked genomic imprinting bondyc{at}mail.nih.gov Carolyn A. Bondy, et al.	Sagi et al. (1) investigated potential X-linked genomic imprinting by comparing phenotypic features in Turner syndrome patients, divided into maternal X (XM) vs. paternal X (XP) groups. Their finding of an apparent excess of renal anomalies in the XM group (P = 0.03) is unexpected since if monosomy for XM caused renal anomalies, then normal XM males should demonstrate the same renal problems seen in Turner syndrome. X-linked genomic imprinting promotes sex dimorphism in certain aspects of behavior (2) and metabolism (3) because X-linked imprinted genes are differentially expressed in males, who are monosomic for XM, and females, who are XM/XP mosaics. Thus, a gene selectively expressed from XP would be expressed in ~50% of cells in a female (assuming random inactivation) but in no male cells. Because of this asymmetry, non-dimorphic traits are not expected to be regulated by X-linked imprinting. In contrast to Sagi et al. (1), a recent prospective study found renal anomalies in ~25% of both XM and XP groups (4), confirming earlier work (5). Sagi et al. (1) found eye anomalies (color blindness, ptosis, strabismus, amblyopia) more frequent in the XP group (P = 0.017). These ocular abnormalities are not pathogenetically related, so it is unclear why subjects with any one of these were grouped together. Furthermore, the inheritance of X-linked color blindness is well understood and is not imprinted; nor is there sex dimorphism in the other eye abnormalities that would support genomic imprinting. Since there was no correction for multiple comparisons in this study, Type I errors likely explain the puzzling renal and eye findings. Sagi et al. (1) compared lipid levels in XM vs. XP patients aged 6 mo to 39 yr drawn from 8 centers in Israel, Turkey, Italy and Poland. Important parameters that affect lipid metabolism such as age, BMI, ethnicity, growth hormone and estrogen treatment were not defined in the two groups. Total and LDL cholesterol were higher in the XP group, but the statistical analysis in the legend to Table 4 states: “ANCOVA to age and BMI for the total cholesterol, P = 0.014, and for the LDL cholesterol, P = 0.035”, while in the text it states that “ANCOVA to age and BMI gave P = 0.003 for the total cholesterol and P = 0.045 for the LDL cholesterol.” In any case it is doubtful that ANCOVA can actually control for significant differences in age and BMI in such small groups. We compared lipid profiles in larger groups of XM vs. XP women similar in age, BMI, and ethnicity that were off HRT during this prospective study (3). In this analysis LDL-cholesterol was higher in the XM group (P = 0.005) - consistent with our hypothesis that monosomy for XM contributes to the more atherogenic lipid profile seen in men vs. women. Turner syndrome offers a unique model to study X-chromosome gene imprinting. The relative rarity of this condition warrants cooperative efforts from multiple centers as Sagi et al. (1) suggest. However, the research efforts should always be driven by biologically plausible and targeted hypotheses and executed by sound statistical methods. References 1. Sagi L, Zuckerman-Levin N, Gawlik A, Ghizzoni L, Buyukgebiz A, Rakover Y, Bistritzer T, Admoni O, Vottero A, Baruch O, Fares F, Malecka-Tendera E, Hochberg Z. 2007. Clinical Significance of the Parental Origin of the X Chromosome in Turner Syndrome. J Clin Endocrinol Metab 92:846-852 2. Skuse DH, James RS, Bishop DV, Coppin B, Dalton P, Aamodt-Leeper G, Bacarese-Hamilton M, Creswell C, McGurk R, Jacobs PA. 1997. Evidence from Turner's syndrome of an imprinted X-linked locus affecting cognitive function. Nature 387:705-708 3. Van PL, Bakalov VK, Zinn AR, Bondy CA. 2006. Maternal X Chromosome, Visceral Adiposity, and Lipid Profile. JAMA 295:1373-1374 4. Bondy CA, Matura LA, Wooten N, Troendle J, Zinn AR, Bakalov VK. 2007. The physical phenotype of girls and women with Turner syndrome is not X-imprinted. Hum Genet 121:469-474 5. Chu CE, Donaldson MD, Kelnar CJ, Smail PJ, Greene SA, Paterson WF, Connor JM. 1994. Possible role of imprinting in the Turner phenotype. J Med Genet 31:840-842