X-Chromosome inactivation in healthy females: incidence of excessive lyonization with age and comparison of assays involving DNA methylation and transcript polymorphisms.
A prerequisite for all methods of clonality analysis based on X-inactivation is the ability to distinguish the paternally derived X-chromosome from the maternally derived one. The distinction is based on common polymorphisms that exist in the general population at different loci on the X-chromosome . A second prerequisite is the ability to distinguish the active from the inactive X-chromosome. 'Evidence for the underlying clonality of hematopoiesis was originally obtained by studying the glucose-6-phosphate dehydrogenase (G6PD) isoenzyme pattern, and its application was limited by the low number ********of informative females [3-5] (1).Currently, the most widely applicable technique makes use of the differences in DNA methylation between active and inactive chromosome loci such as phosphoglycerate kinase (PGK), hypoxanthine phosphoribosyl transferase kinase, the DXS255 locus (M27[beta]), and the human androgen receptor (HUMARA) [2,6-91. Recently, since only genes on the X-active chromosome are transcribed, clonality analysis based on transcript analysis has been developed X10-12]. One advantage of this technique is the possibility to study clonality in platelets and reticulocytes. Moreover, it avoids the problem of potential incomplete digestion of genomic DNA.
However, 21-31% of healthy women have a skewing pattern of X-inactivation (or lyonization) with techniques based on DNA polymorphism, leading to a potential limitation for X-inactivation clonality assays because constitutional skewing mimics clonal derivation of cells, rendering clonality results uninterpretable [7,13]. Different hypotheses have been suggested to explain the especially high rate of excessive lyonization in hematopoietic tissue. First, skewing of the X-chromosome inactivation pattern may be caused by the small stem-cell pool size at the time of X-chromosome inactivation . Patterns will therefore vary from tissue to tissue. Because X-chromosome inactivation takes place gradually over a period of time, as demonstrated by Tan et al., this will influence the patterns obtained; e.g., inactivation of the progenitor blood cell pool at an earlier development time would lead to considerably more skewing in blood cells than in gut cells, which are apparently one of the last cell populations to be inactivated . Second, it is possible that somatic cell selection occurs after random X-inactivation has been established and as a consequence of a selective pressure on blood cells X16-181. Finally, stem-cell depletion or the development of clonal hematopoiesis may also induce a nonrandom pattern of X-inactivation . Each of these acquired processes could result in highly skewed X-inactivation ratios and should be associated with an increased incidence of skewing with age. Fey et al. have reported data supporting a different incidence in skewing between elderly and younger females , although Gale et al. were unable to demonstrate an effect of age on skewing when the totality of published data for healthy females was considered (n = 100) . Recently, Busque et al. showed that incidence of excessive lyonization in healthy women increased significantly with age with the HUMARA and PGK genes. The incidence of skewing (allele ratios [greater than or equal to]3:1) was 8.6% (14 of 162) in neonates, 16.4% (11 of 67) in women 28 to 32 years old, and 37.9% (25 of 66) in women [greater than or equal to]60 years .
In addition to biological explanations, technical reasons may account for some of the discrepancies in the estimation of skewed lyonization in healthy females and possibly clonality analysis in various malignancies. This hypothesis is supported by the low frequency of skewed lyonization found by studying the expression of G6PD polymorphism at the protein level compared with the high frequency found with methylation analysis of the M27[beta] locus [3, 7,17]. Recently, new techniques based on transcript analysis have been described [10,12,19]. Thus, the aim of this study was to compare X-chromosome inactivation patterns by using two different PCR techniques. The first one is based on analysis of the difference in X-chromosome methylation between active and inactive chromosomes of the HUMARA gene . The second is based on analysis of polymorphic transcript genes, which are expressed in all peripheral blood fractions: genes for iduronate-2-sulfatase (IDS), P55, and G6PD [11,12]. These techniques were used to study the X-chromosome inactivation pattern in 123 samples from three age groups: female neonates, women ages 22-50 years, and women >50 years. We chose for comparison a group of patients with essential thrombocythemia (ET) in whom we already demonstrated the existence of clonal hematopoiesis in most cases .
Materials and Methods
Eighty-three peripheral blood samples and 40 cord blood samples were obtained from unrelated healthy females after informed consent. The samples were classified in three age groups: (a) neonates (cord blood taken instead of peripheral blood), n = 40; (b) women ages 22 to 50 years, n = 49; (c) women ages 50 to 80 years, n = 34. Forty-eight ET patients were also studied. All samples wre collected after informed consent. Clinical and laboratory data of patients have already been reported in detail .
Peripheral blood was collected on citrate or EDTA and fractionated as previously described . Cord blood was diluted in an equal volume of physiological saline before fractionation.
DNA AND RNA ISOLATION
DNA was extracted from granulocyte and mononuclear cell fractions with the Isoquick kit (Orca Research), and from whole blood with the Gnome Kit (Bio 101). RNA was extracted from cell fractions, including platelets, according to Chomczynski and Sacchi .
GENOTYPING AND CLONALITY ASSAYS
These were performed on nucleated cells: granulocytes or total blood and in platelets. T lymphocytes were also studied in females with a skewed pattern of lyonization or when results were discordant.
Clonality analysis involving differences in DNA methylation: HUMARA trinucleotide repeat polymorphism. This method has previously been described in detail . Briefly, DNA was amplified by using two primers flanking the STR in the HUMARA gene. One primer was labeled at the 5' end with fluorescein. The products were analyzed and quantified with an automated DNA sequencer (ALF, Pharmacia-Biotech). The allele ratio was defined as the ratio between the intensity of PCR products from the two HUMARA alleles in a given sample, after digestion with methylation-sensitive enzymes (a' /a' +b', a' being the smaller allele after digestion). The corrected ratio (CR) was defined as the allele ratio of the precut DNA sample (a'/a'+b') divided by the allele ratio of the nonprecut sample (a/a+b) specimen. This ratio compensates for potential preferential amplification of one of the two alleles (in general, the smaller one). Excessive skewing or clonality of hematopoiesis was defined as CR [greater than or equal to]3:1, which corresponds to the finding of one allele present on the same X-inactive chromosome in [greater than or equal to]75% of cells .
Genotyping and clonality studies involving IDS, P55, and G6PD transcript polymorphism: PCR-restriction fragment length polymorphism (RFLP) technique. Genotyping and clonality analysis by PCR-RFLP technique has been described in detail in previous reports . Briefly, DNA was tested for a silent exonic polymorphism at nucleotides 1311, 438, and 358. In heterozygous females, after reverse transcription by using random priming, cDNAs were amplified, digested, and analyzed on ethidium bromide-stained agarose gel. After migration, two bands were seen when X-chromosome inactivation was random, whereas only a single band was seen in females with skewed inactivation.
Relative quantification of mRNAs by fluorescent primer extension assay. This was performed by primer extension assay with a fluorescent primer close to the polymorphic site for IDS, P55, and G6PD genes (Fig. 1) [21-23].
PCR products were obtained by using primers located on two different exons for each polymorphism: IDS 3 and IDS 4B; P55 1B and P55 4; G6PD 10B and G6PD 1]. IDS 4B, P55 1B, and G6PD 10B were end-labeled with biotin according to the instructions of the manufacturer. Single-strand DNA was produced with 10 [micro]L of M280 Dynabeads (Dynnl) and 20 [micro]L of amplified products. Then 5 [micro]L of single-strand PCR product was mixed with 1 [micro]L of hybridization buffer (1 mol/L Tris-HCI pH 7.4, 100 mmol/L Mg[Cl.sub.2],) and 0.5 pmol of the purified and fluorescent primer: IDS 4F, P55 3F, or G6PD 11F for 15 min at 37 [degrees]C and 10 min at room temperature. Sequenase (0.5 U) (USB) and then termination mixture (5 [micro]L) were added and the mixture was incubated for 3 min at 37 [degrees]C. This latter mixture contained 50 mmol/L NaCl; 40 mmol/L Tris-HCI pH 7.4; and 500 [micro]mol/L dATP, 500 [micro]mol/L dCTP, 500 [micro]mol/L dGTP, and 500 [micro]mol/L ddTTP (dideoxythymidine triphosphate) for IDS polymorphism. For P55 and G6PD genes the termination mixture contained ddCTP (with dATP, dGTP, and dTTP) and ddGTP (with dATP, dCTP, and dTTP) respectively. After migration on an automated sequencer (ALF, Pharmacia), allele-specific products were detected as fluorescent peaks and the area under the curve was determined with the fragment manager software (Pharmacia). In females with a nonrandom inactivation pattern, only one peak was obtained (in addition to the peak corresponding to the fluorescent primer), whereas two peaks were detected in cases of random X-inactivation. Excessive skewing or clonality of hematopoiesis was defined as an allelic ratio [greater than or equal to]3:1. Table 1 shows the sequence and the position of all primers, as well PCR conditions.
[FIGURE 1 OMITTED]
To study the reliability and the sensitivity of the technique, we performed mixing experiments. We added clonal cells b (C allele expression for IDS gene) to polyclonal cells a at different ratios with a total of 8 million cells in each experiment. The different ratios were: 8 a; 7:1; 6:2; 5:3; 4:4; 3:5; 2:6; 1:7; 8 b. After RNA extraction and reverse transcription, PCR and primer extension assay were carried out for IDS polymorphism.
VALIDATION OF PRIMER EXTENSION ASSAY
To polyclonal cells with two peaks of 50% each (a, Fig. 2A), we added clonal cells b at different ratios with an arithmetic progression of 12.5%. A strong linearity was observed between allele ratios and the percentage of cells expressing the same X chromosome (Fig. 2B). The peak of the C allele increased 11%, 24%, 38%, 50%, 64%, 74%, 90%, and 100% when we added 12.5%,25%,37.5%,50%,62.5%, 75%, 82.5%, and 100% of clonal cells with C polymorphism, respectively. A clonal population is then detected according to our criteria (allelic ratio [greater than or equal to]3:1) when it represents more than or equal to half of the population cells.
AGE VARIATION OF X-INACTIVATION PATTERN IN HEALTHY CONTROLS
By the difference in DNA methylation: All DNA samples were tested for HUMARA gene polymorphism. The incidence of excessive lyonization by age group was: 9 of 38 (24%) in neonates, 0 of 46 in women <50 years, and 6 of 29 (21%) in women >50 years of age (Table 2).
By analysis of RNA polymorphism: Percentage of X-skewed lyonization was 3%, 0%, and 21% in the different groups respectively (Table 2). The incidence of skewing in neonates by using HUMARA gene polymorphism (24%) compared with transcript analysis (3%) was statistically different ([chi square] = 3.83), but the number of samples was small.
Comparison between clonality assays involving difference of methylation on DNA and trancript analysis. We used four polymorphic genes, which enabled us to study all except two of 125 samples, 87 of them being informative for at least one DNA and one RNA polymorphism.
Results were concordant in 82 of 87 cases (94%) and discordant in five cases. Among the 27 subjects informative for both techniques in neonates, we noticed a nonrandom pattern in four samples using HUMARA polymorphism and a random pattern by analysis of RNA transcripts (P = 0.15, Fisher test) (Table 3a, Fig. 3). The opposite was observed in one elderly woman (Table 3c, Fig. 3).
X-inactivation pattern was the same in the different hematopoietic lineages obtained from each individual: granulocytes, platelets, and T lymphocytes.
RESULTS OF CLONALITY ASSAYS IN ET PATIENTS
The overall results have been published previously . Briefly, 74% and 68% of the patients showed a nonrandom pattern in granulocytes with DNA and mRNA analysis respectively. Among 48 patients studied, 28 were informative for both techniques with HUMARA polymorphism and at least one RNA polymorphism. Results were concordant in 26 of 28 cases (93%) and discordant in two patients (Table 3d, Fig. 3). In these two patients, a random pattern of X-inactivation was obtained in nucleated cells by using HUMARA polymorphism, whereas a clonal hematopoiesis was observed by transcript analysis in all fractions except T lymphocytes in one patient, and a skewed lyonization in the other patient. In two patients, no conclusion could be drawn as the same nonrandom lyonization was observed in all fractions.
[FIGURE 2 OMITTED]
Because most techniques used for X-inactivation studies are based on variations of gene methylation, the aim of this study was to compare the methylation pattern of DNA with the expression pattern of genes present on the X-chromosome. In contrast to a previous report , we were not able to reliably detect HUMARA gene expression in hematopoietic cells; we therefore developed clonality assays based on analysis of polymorphic transcripts of other genes located on the X-chromosome.
Methylation represents the final step of the X-chromosome inactivation process . Whether or not this methylation is complete at birth and whether it varies with age is still debated [7,17,18]. For this reason, we compared X-inactivation by using DNA and RNA polymorphisms in different age groups, from neonates to elderly women, as well as in a group of ET patients, a majority of whom are known to show clonal hematopoiesis.
[FIGURE 3 OMITTED]
Our techniques used to study HUMARA gene polymorphism and quantification of mRNAs by fluorescent primer give quantitative results that are therefore directly comparable. Results were concordant in most cases in healthy females and patients. However, some discordances were observed. First, we found a statistically significant difference in skewing rate in neonates using DNA polymorphism (24%) compared with RNA polymorphisms (3%). One explanation could be that methylation was incomplete at birth on one allele, paternal or maternal. Indeed, the opposite pattern was observed in one elderly woman and two ET patients. In the case of the elderly woman, the discrepancy could be due to hypermethylation of DNA on the active X-chromosome resulting in incomplete digestion and consequently a random pattern, while basically X-inactivation was skewed. Alternatively, the occurrence of a mutation on one allele of the IDS gene may lead to the absence of its expression in females with a random pattern of X-inactivation. In patients, it may be due to hypermethylation of DNA as already described in tumors [2[, especially because of the clear clonality feature obtained by transcript analysis in one patient (clonality in granulocytes and platelets with polyclonality in T lymphocytes).
Comparison between the DNA and RNA polymorphisms showed some interesting results that may help clarify the interpretation of excessive lyonization in healthy females. Both techniques showed an increase in skewing rate between females <50 years and elderly ones. This suggests that methylation does not increase with age, but that depletion of stem cells, clonal selection, or the appearance of clonal hematopoiesis might occur with age. Furthermore, using RNA polymorphisms, we confirmed the findings of others that incidence of skewed lyonization is very low at birth. This low incidence of excessive skewing in neonates also suggests that the estimated number of stem cells present at the time of X-inactivation could be higher than previously reported.
Our results are not in agreement with those of other authors who found a low rate of nonrandom X-inactivation at birth using HUMARA polymorphism , but our group of neonates was much smaller (40 vs 162).
Our study showed that clonality assays involving DNA and RNA polymorphisms are usually concordant except in neonates. We also confirmed that the same pattern of X-chromosome inactivation is observed in all hematopoietic lineages in healthy females, as they all derive from the same pluripotent stem cells, in contrast to the results in ET patients with clonal hematopoiesis. The results of this study further emphasize that appropriate control tissue should be used for each individual female when conducting an X-inactivation clonality assay. A somatic tissue not involved in the disease process and embryologically related to the sample must be chosen to eliminate excessive lyonization or acquired skewing.
We express our gratitude to B.J. Paniel and J. Gour from Maternity Department (Centre Hospitalier Intercommunal de Creteil) for providing cord blood samples, and P. Salesses, H. Plana, and E. Poulet from HLA laboratory (Centre de Transfusion Sud Est Francilien, Creteil) for their participation in cell fractionation. This work was supported in part by Association Claude Bernard and la Fondation pour la Recherche Medicale.
Received April 21, 1997; revision accepted August 29, 1997.
[1.] Lyon MF. Gene action in the X chromosome of the mouse. Nature 1961;190:372-3.
[2.] Vogelstein B, Fearon ER, Hamilton SR, Feinberg AP. Use of restriction fragment length polymorphisms to determine the clonal origin of human tumors. Science 1985;227:642-5.
[3.] Fialkow PJ. Primordial cell pool size and lineage relationships of five human cell types. Ann Hum Genet 1973;37:39-48.
[4.] Fialkow PJ, Faguet GB, Jacobson RJ, Vaidya K, Murphy S. Evidence that essential thrombocythemia is a clonal disorder with origin in a multipotent stem cell. Blood 1981;58:916-9.
[5.] Gaetani GF, Ferraris AM, Galiano S, Giuntini P, Canepa L, d'Urso M. Primary thrombocythemia: clonal origin of platelets, erythrocytes and granulocytes in A G[d.sup.B]/[Gd.sup.Mediterranean] subject. Blood 1982;59:76-9.
[6.] Gilliland DG, Blanchard KL, Levy J, Perrin S, Bunn HF. Clonality in myeloproliferative disorders: analysis by means of the polymerase chain reaction. Proc Natl Acad Sci U S A 1991;88:6848-52.
[7.] Fey MF, Peter HJ, Hinds HL, Zimmermann A, Liechti-Gallati S, et al. Clonal analysis of human tumors with M27[beta], a highly informative polymorphic X chromosomal probe. J Clin Invest 1992;89:143844.
[8.] Tsukamoto N, Morita K, Maehara T, Okamoto K, Sakai H. Clonality in chronic myeloproliferative disorders defined by X-chromosome linked probes: demonstration of heterogeneity in lineage involvement. Br J Haematol 1994;86:253-8.
[9.] Allen RC, Zoghbi HY, Moseley AB, Rosenblatt HM, Belmont JW. Methylation of Hpall and Hhal sites near the polymorphic CAG repeat in the human androgen-receptor gene correlates with X chromosome inactivation. Am J Hum Genet 1992;51:1229-39.
[10.] Prchal JT, Guan YL. A novel clonality assay based on transcriptional analysis of the active X-chromosome. Stem Cells 1993;11: 62-5.
[11.] El Kassar N, Hetet GLY, Briere J, Grandchamp B. Clonal analysis of haematopoietic cells in essential thrombocythaemia. Br J H aem atol 1995;90:131-7.
[12.] El Kassar N, Hetet G, Briere J, Grandchamp B. Clonality analysis of hematopoiesis in essential thrombocythemia: advantages of studying T-lymphocytes and platelets. Blood 1997;89:128-4.
[13.] Busque L, Gilliland DG. Clonal evolution in acute myeloid leukemia. Blood 1993;82:337-42.
[14.] Young NS. The problem of clonality in aplastic anemia: Dr Dameshek's riddle, restated. Blood 1992;79:1385-92.
[15.] Tan SS, Williams EA, Tam PPL. X-chromosome inactivation occurs at different times in different tissues of the post-implantation mouse embryon. Nature Genet 1993;3:170-4.
[16.] Luzzatto L, Usanga EA, Bienzle U, Esan GFJ, Fusuan FA. Imbalance in X-chromosome expression: evidence for a human X-linked gene affecting growth of hemopoietic cells. Science 1979;205:141820.
[17.] Gale RE, Wheadon H, Boulos P, Linch DC. Tissue specificity of X chromosome inactivation patterns. Blood 1994;83:2899-905.
[18.] Busque L, Mio R, Mattioli J, Brais E, Blais N, Lalonde Y, et al. Nonrandom X-inactivation patterns in normal females: lyonization ratios vary with age. Blood 1996;88:59-65.
[19.] Prchal JT, Liu Y, Prchal JF, Hoffman R, Tushinski R. Are normal stem cells present in myeloproliferative syndrome? [Abstract]. Blood 1994;84:207.
[20.] Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156-9.
[21.] Lee JS, Anvret M. Identification of the most common mutation within the porphobilinogen deaminase gene in Swedish patients with acute intermittent porphyria. Proc Natl Acad Sci U S A 1991;88:10912-5.
[22.] Porcher C, Malinge MC, Picat C, Grandchamp B. A simplified method for determination of specific DNA or RNA copy number using quantitative PCR and an automatic DNA sequencer. Biotechniques 1992;13:106-13.
[23.] Gouya L, Deybach JC, Lamoril J, Da Silva V, Beaumont C, Grandchamp B, Nordmann Y. Modulation of the phenotype in dominant erythropoietic protoporphyria by a low expression of the normal ferrochelatase allele. Am J Hum Genet 1996;58:292-9.
[24.] Busque L, Zhu J, DeHart D, Griiffith B, Willman C, Carroll R, et al. An expression based clonality assay at the human androgen receptor locus (HUMARA) on chromosome X. Nucleic Acids Res 1994;22:697-8.
[25.] Simmler MC. L'inactivation du chromosome X chez les mammiferes. Med Sci 1992;8:972-8.
Nahed El Kassar, * Gilles Hetet, Jean Briere, And Bernard Grandchamp
INSERM U409, Association Claude Bernard, Faculte de Medecine Bichat, BP 416, F-75870 Paris cedex 18.
* Author for correspondence. Fax 33 1 42 26 46 24; e-mail U409@bichat.inserm.fr.
(1) Nonstandard abbreviations: G6PD, glucose-6-phosphate dehydrogenase; PGK, phosphoglycerate kinase; HUMARA, human androgen receptor; IDS, iduronate-2-sulfatase; ET, essential thrombocythemia; CR, corrected ratio; and RFLP, restriction fragment length polymorphism.
Table 1. Primers and PCR conditions used for primer extension assays. Annealing temp., Gene Primer Sequence 5'~3' Exon [degrees]C IDS IDS 3 TGT ACG ACT TCA ACT 3 47 CCT ACT GGA IDS 4B AGA CCA GCT ATA CGG 4 47 AGA ATC ACC IDS 4F GGG ATA TCT TCT AAC 4 37 CAT P55 P55 1B GAG AGT GGG GGC AGC 1 61 ATG CAC AC P55 4 CAT CGC CTT CTG CAG 4 61 CTG ATC CAC P55 3F GCA CTG TTT TTC ATT 3 37 CAG CTT C G6PD G6PD 10B CCT GAC CTA CGG CAA 10 54 CAG ATA CAA G6PD 11 AGT GCA TCT GGC TCC 11 54 CGC AGA A G6PD 11F GAC GTC CAG GAT GAG 11 37 GCG Size of [Mg[C1.sub.2]], PCR Gene Cycles mmol/L product IDS 35 1.5 158 bp 35 1.5 158 bp P55 35 1.5 387 bp 35 1.5 387 bp G6PD 35 1.5 92 bp 35 1.5 92 bp B, biotinylated primers used in the PCR reactions; F, fluorescently labeled primers used in the extension reaction. Table 2. X-chromosome inactivation pattern by RNA and DNA polymorphisms in different age groups. DNA Random Nonrandom Neonates (n = 40) 29/38 (76%) 9/38(24%) 22-50 years (n = 49) 46/46(100%) 0/46 (0%) 50-93 years (n = 34) 23/29 (79%) 6/29(21%) RNA Random Nonrandom Neonates (n = 40) 28/29 (97%) 1/29 (3%) 22-50 years (n = 49) 39/39(100%) 0/39 (0%) 50-93 years (n = 34) 23/29 (79%) 6/29(21%) Table 3. Comparison of the results of X-chromosome inactivation pattern between clonality assays involving RNA and DNA polymorphisms in different age groups. RNA Random Nonrandom (a) Neonates (n = 27) DNA Random 22 0 Nonrandom 4 1 (b) 22-50 year-old women (n = 36) DNA Random 36 0 Nonrandom 0 0 (c) 50-93 year-old women (n = 24) DNA Random 18 1 Nonrandom 0 5 (d) Patients (n = 28) DNA Random 5 2 Nonrandom 0 21 Subjects were healthy females and ET patients who were informative for the HUMARA polymorphism and at least one of the three RNA polymorphisms. The X-chromosome inactivation pattern was scored as random or nonrandom and the results obtained with RNA and DNA polymorphisms are compared.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Molecular Pathology and Genetics|
|Author:||El Kassar, Nahed; Hetet, Gilles; Briere, Jean; Grandchamp, Bernard|
|Date:||Jan 1, 1998|
|Previous Article:||Cardiac troponin I measurement with the ACCESS[R] immunoassay system: analytical and clinical performance characteristics.|
|Next Article:||Molecular diagnosis of medium-chain acyl-CoA dehydrogenase deficiency by oligonucleotide ligation assay.|