- Email: [email protected]
Age-related aneuploidy analysis of human blood cells in vivo by fluorescence in situ hybridization (FISH)
medlahmofqehg
and Mechanisms of Ageing and Development ELSEVIER
90 (1996) 145-156
Age-related aneuploidy analysis of human blood cells in vivo by fluorescence in situ hybridization (FISH) Asit B. Mukherjee
*, Jason Alejandro, Samuel Thomas
Shakira
Payne,
Department of‘ Biological Sciences. Fordham Univrrsit~~,Bronx, NY 10458, USA
Received 18 November 1995: revised 6 June 1996
Abstract All prior studies on human age-related chromosomal analysis were done using only metaphase figures derived from lymphocyte cultures in vitro. However, we believe that this procedure may provide only partial information, since the chromosomal abnormalities probably hidden in non-dividing and/or terminally differentiated leukocytes will not be detected by this method. We, therefore, have attempted to analyze the nature and extent of chromosome-specific aneuploidy at interphase by fluorescence in situ hybridization (FISH) in differentiated myeloid cells in vivo derived from various age-group people. Our data from an analysis of 30032 cells derived from 12 healthy donors indicate that there is an increase in the mean percentages of myeloid cells with chromosome-specific aneuploidy in the older groups as opposed to that of the younger groups. This increase is applicable to all the three cell types examined (promyelocytes, metamyelocytes and polymorphs). In both the younger and older females, the relatively higher mean frequencies of cells with aneuploidy were noted for chromosome nos. 4, 6 and the X, whereas the lowest mean frequencies of cells with aneuploidy were consistently observed for chromosome no. 3. In the younger and older male donors, similar to the female donors except the X chromosome the higher percentages of aneuploid cells were observed for chromosome nos. 4 and 6 whereas the lowest mean frequencies of aneuploid cells were noted for chromosome no. 3. Among the five autosomes studied, chromosome no. 3 consistently yielded the lowest frequency of aneuploidy in all the three cell types derived from both the younger and older groups of males and females. This *Corresponding
author. Tel.: + 1 718 8173663; fax: + I 718 8173645
0047-6374/96i$15.000
1996 Elsevier Science Ireland Ltd. All rights reserved
PII SOO47-6374(96)01762-9
146
A. B. Mukherjee et al. / Mechanisms of’ Age&
and Developmerzt 90 (1996) 145-156
could presumably mean that the maintenance of a very high frequency of cells with disomy for chromosome no. 3 might be more beneficial for the process of survival and/or differentiation of myeloid cells as compared to that of other autosomes such as chromosome nos. 4 and 6. Among the five autosomes studied, chromosome no. 1 exhibited the highest rate of increment in the aneuploidy values of both males and females with advancing age. Keywords:
Aneuploidy;
Blood cells; Aging: FISH
aneuploidy
analysis
1. Introduction An association between advanced chronological aging and increased chromosomal aneuploidy in human lymphocyte cultures was first reported by Jacobs et al. [1,2]. Although a number of subsequent studies, including a detailed report utilizing the G-banding technique [3], support this observation, a consistent correlation between chronological aging and chromosomal abnormality (numerical and/or structural) still remains to be fully explored. particularly after the development of molecular-cytogenetic techniques. For example, a previous study involving 493 subjects did not find any significant relationship of the frequency of any conventional aberration (structural) category to age with the sole exception of the dicentric chromosome, which showed a positive regression [4]. Also, all previous studies on the association between human chronological aging and chromosomal abnormality were done utilizing only metaphase figures derived from phytohemagglutinin-stimulated lymphocyte cultures in vitro. We, however, believe that the exclusive use of lymphocyte-derived metaphase chromosomes in the analysis of chromosomal aneuploidy as related to chronological aging may provide only partial information since the aneuploidy presumably hidden in nondividing and/or terminally differentiated leukocytes would remain undetected by this method. For this reason, we have now attempted to analyze the nature and extent of age-related chromosomal aneuploidy at interphase during human myeloid differentiation in vivo. The various types of terminally differentiated myeloid cells are non-dividing in peripheral blood and each cell type can be morphologically identified by the characteristic shape of its nucleus. In this study, we have used a powerful molecular-cytogenetic technique, called fluorescence in situ hybridization (FISH) [5]. We have utilized various chromosome-specific a-satellite DNA probes for accurate detection of chromosomal aneuploidy [6]. Alphoid DNA is a family of primate satellite DNAs located in the centromeric region of various chromosomes [791.
A.B. Mukherjee
et ul.
! Mechanisms oj’ Ageing anti Dmelopment
90 (1996) 145-156
147
2. Materials and methods 2.1. Blood filtn preparation Peripheral blood samples were derived from six males (ages: 21, 22, 22, 50, 52 and 66 years) and six females (ages: 20, 21, 22, 61, 76 and 79 years), all chromosomally normal, healthy individuals representing both the younger and older groups. Heparinized peripheral blood sample from each individual was allowed to settle for 2-3 h at room temperature and was then centrifuged at 1500 rpm for 5 min. White blood cells were then collected from the buffy coat and, using a Pasteur pipette, 2 drops of buffy coat suspension were placed at the center of a clean microscope slide towards the frosted end. A second microscope slide was then placed directly above the first slide, allowing surface tension to spread the buffy coat suspension on the slides. After 3-5 s, the two slides were glided apart horizontally, allowed to sit for 5- 10 s and then placed in 3: 1 methanol:glacial acetic acid fixative for 45 min. The slides were then air-dried and stored at 4°C. This method of slide preparation provided cytoplasm-free cell nuclei in 98.6% of the cases. All cell samples were screened for aneuploidy analysis within one month of the initial preparation of blood films. 2.2. Fluorescence
in situ hyhridixtion
(FISH)
Chromosome-specific aneuploidy analysis in 3 types of differentiated leukocytes (promyelocyte, metamyelocyte and polymorphonuclear leukocyte or polymorph) obtained from 12 male and female individuals representing both younger and older groups was done by FISH [5]. Seven chromosome-specific n-satellite DNA probes labeled with biotin (chromosome nos. 1, 2, 3, 4, 6, X and Y) were utilized in this preliminary study. These centromeric DNA probes were obtained from ONCOR (Gaithersburgh, MD). Molecular hybridization and immunofluorescent detection of each DNA probe were performed according to the manufacturer’s protocol using ONCOR chromosome in situ kit. The same FISH protocol was used for all cell samples hybridized with various DNA probes. In the scoring procedure, only cell nuclei with compact and discrete FISH signal morphology were selected for aneuploidy analysis. In accordance with a previously published method of aneuploidy analysis in human blood cells, if a portion of a slide showed more than two nuclei without the FISH signals (hybridization regions), that part of the slide was considered to be incompletely hybridized and these nuclei were not scored for the chromosomal analysis [lo]. Additionally, as control for in vivo blood cell data, we evaluated the occurrence of chromosomespecific monosomy (hypodiploidy) as well as trisomy-tetrasomy (hyperdiploidy) in phytohemagglutinin-stimulated blood lymphocytes in vitro from normal individuals using the exact same FISH protocol of the in vivo study. In separate experiments, the incidence of monosomy varied from 7% to 9% and trisomy and tetrasomy was less than 1%. Our results are similar to that of other investigators who used several chromosome-specific DNA probes for FISH analysis [lo]. On the basis of our
148
A.B. Mukherjee et al. / Mechanisms of’ Ageing and Development 90 (1996) 145-156
results, we set 10% as the error limit for detection of monosomy and the scoring for aneuploidy was done accordingly. A total of 30032 cells (ranging from 1950 to 3116 cells/sample) derived from 12 individuals were screened blindly with a Leitz Orthomat fluorescent microscope and representative photographs were taken according to our previously published methods [l l- 131. For FISH analysis of each chromosome, a separate slide was used for individual samples. 2.3. Statistical
analysis:
Statistical analyses were performed using a l-tailed Wilcoxon signed-rank test to compare the frequencies of aneuploidy between younger and older age-group people of both sexes. Significance of values were determined by using a 0.025 probability.
3. Results During sequential identified identified
human myeloid differentiation, the various cell types are generated in order and these non-dividing (interphase) cells can be morphologically by the characteristic shapes of their nuclei. For example, promyelocyte is by its large round or oval nucleus (Fig. l), metamyelocyte by its
Fig. I. A promyelocyte nucleus derived from a 22 year old female with two brightly signals representing disomy for the X chromosome.
fluorescent
FISH
A.B.
Mukhecjee
el al. / Mechanisms
of Ageing
and Deuelopmenr
90 (1996)
Fig. 2. A polymorph nucleus derived from a 79 year old female with a single brightly signal representing monosomy for the X chromosome.
145-156
fluorescent
149
FISH
kidney-shaped or horse-shoe-shaped nucleus (Fig. 3) and polymorph by its nucleus with lobules of dense chromatin connected by one or more thin filaments (Figs. 2 and 4). In this study, only these three cell types with characteristic nuclear rnorphologies were analyzed for aneuploidy. Each chromosome-specific fluorescent signal in a particular cell nucleus produced by FISH represented the presence of a specific chromosome. Thus, a nucleus was considered to be monosomic when only one fluorescent signal was present (Fig. 2). Disomic nuclei showed two signals, (Figs. l-4) etc. In this study, only cell nuclei with one and/or three FISH signals/cell were taken into account in determining the total yield of aneuploidy for a particular chromosome. Since cell nuclei with zero or four FISH signals/cell were extremely rare (O-0.2%), they were not included in this chromosome analysis. Fig. 5 summarizes the aneuploidy data from the three types of terminally differentiated blood cells in vivo (promyelocyte, metamyelocyte and polymorph) at interphase involving five autosomes and the X chromosome derived from three younger and three older females. The mean age of the younger group was 21 years, whereas that of the older group was 72 years. In the younger females, the vast majority (mean% ranging from 96.5 to 99.1) of the three cell types showed diploidy (disomy) for the chromosomes under investigation. The mean percentages of cell nuclei with aneuploidy was variable from chromosome to chromosome (mean% ranging from 0.9-3.5) in the three cell types. There were no marked differences in
150
A. B. Mukherjee et al. / Mechani.sms of Ageing and Development 90 11996) i45- 156
the mean percentages of cell nuclei with aneuploidy between the three cell types (2.4% in promyelocyte, 2.0% in metamyelocyte and 2.2% in polymorph). Of the five autosomes, the higher mean percentages of nuclei in the three cell types were consistently obtained for chromosome nos. 6 (3.5%) and 4 (3.3%), respectively, whereas the lowest mean percentage of nuclei with aneuploidy was observed for chromosomes no. 3. In the older group of females, similar to the younger group, the vast majority (mean% ranging from 95.9 to 98.4) of the three cell types in each of the chromosome-specific FISH preparations exhibited diploidy (disomy). Again, the higher mean frequencies of cell nuclei with aneuploidy was noted for chromosome nos. 4 (4.1%) and 6 (3.9%), whereas the lowest mean frequency of cells with aneuploidy (1.6%) was consistently recorded for chromosome no. 3 in all the three cell types. There were no pronounced differences in the mean percentages of cells with aneuploidy among the three cell types (3% in promnyclocyte, 2.6O/oin metamyelocyte and 3.3% in polymorphs). However, in older females, there was a statistically significant increase in the mean frequencies of aneuploid cells involving all chromosomes under study as compared to that of the younger females (P < 0.025). Of the five autosomes, chromosome no. 1 showed the highest rate of increase in aneuploidy value (from 1.5% to 2.6%) with advancing age. Similar trend in the aneuploidy value was observed for the X chromosome in older females.
Fig. 3. A metamyelocyte nucleus derived from a 21 year old male with a single brightly signal representing the Y chromosome.
fluorescent
FISH
A.B. Mukhecjee
et al. ! Mechanisms
of Ageing and Deueloprrlent 90 (1996) 145-156
Fig. 4. Three polymorph nuclei derived from a 56 year old male each with two brightly signals representing disomy for chromosome no. 3.
fluorescent
151
FISH
Fig. 6 presents our data on chromosome-specific aneuploidy at interphase involving seven chromosomes (5 autosomes, X and Y) in the three types of myeloid cells in vivo derived from three younger and three older males. The mean age of the younger donors was 21.6 years whereas that of the older donors was 56 years.
Fig. 5. Combined and polymorphs
mean% nuclei with chromosome-specific aneuploidy in promyelocytes, in younger (mean age = 21) vs. older (mean age = 72) females.
metamyelocytes
152
A.B. Mukherjee
et al. 1 Mechanisms
of Ageing and Development
90 (1996) 145-156
Fig. 6. Combined mean%) nuclei with chromosome-specific aneuploidy in promyelocytes, metamyelocytes and polymorphs in younger (mean age = 21.6) vs. older (mean age = 56) males.
Among all five autosomes, the highest (99%) and the lowest (96.9%) mean frequencies of diploid cells were observed for chromosome nos. 3 and 4, respectively. The highest level of aneuploidy was noted for chromosome no. 4 (3.1%), followed by chromosome no. 6 (2.9%) in this order. The relative proportions of cells with aneuploidy for chromosome nos. 1, 2 and the X were almost identical (1.4- 1.5%). In the younger males, similar to the younger females, the lowest mean percentage of cells (1.0%) with aneuploidy was consistently recorded for chromosome no. 3. The aneuploidy level of the Y chromosome was slightly higher (1.9%) than that of the X chromosome (1.5%). There were no marked differences in the mean frequencies of cells with aneuploidy among the three cell types (2.1% for promyelocyte, 1.8% for metamyelocyte and 2.0% for polymorph). In the older males, the highest (3.3%) and the lowest (1.3%) mean percentages of aneuploid cells were noted for chromosome nos. 4 and 3, respectively. The second highest level of aneuploidy was observed for chromosome no. 6, and chromosome no. 1 showed the highest rate of increment in aneuploidy value (from 1.4% in younger vs. 2.5% in older males) with advancing age, similar to that of older females. There were no striking differences in the mean frequencies of aneuploid cells among the three cell types (2.3% for promyelocyte, 2.6% for metamyelocyte and 2.4% for polymorph). However, with advancing age, there was a statistically significant increase in the aneuploidy values for all chromosomes under investigation (P > 0.025).
4. Discussion It is now well-recognized that the arrangement of chromosomes in plant and animal cells does not occur at random and specific chromosomes occupy specific
A.B. Mukherjee
et ul. / Mechanisms
of .dgeing and Development
90 (1996) 145-156
153
territories (called chromosomal domains) in the interphase nuclei [14,15]. More recently, numerous studies have shown that FISH with chromosome-specific centromeric DNA probes (cc-satellite) can be effectively utilized to detect specific chromosomes as fluorescent spots (FISH signals) in an interphase nucleus. This study has utilized the FISH technique to examine the extent of aneuploidy at interphase involving five automes and two sex chromosomes of terminally differentiated myeloid cells in the younger and older age-group people. Monosomy of a specific chromosome constituted the higher proportion of aneuploidy in all the three cell types derived from each individual. This finding is consistent with similar studies done with cultured lymphocytes [lo]. A comparative analysis of the average of the mean percentages of cell nuclei with each chromosome-specific aneuploidy under investigation indicates that, for any one particular age group in both sexes, there is no striking difference in the degree of aneuploidy in the three different cell types (i.e. promyelocytes, metamyelocytes and polymorphs) (Fig. 7). However, there is a clear trend in the increase of the mean percentages of cells with aneuploidy in the older groups of males and females as opposed to that of the younger groups (Figs. 5-7). This increase is applicable to all the three cell types studied. This finding is similar to our previous results on aneuploidy at metaphase in near-senescent human fibroblasts in vitro [16]. However, in the same study, we noted relatively lower frequencies of aneuploidy in completely senescent, non-dividing human skin fibroblasts in vitro as compared to the frequencies of aneuploidy in actively dividing young cells. As a possible explanation to this interesting finding, we proposed a selection hypothesis relevant to cellular survival at replicative
FEMALE
DONORS
Fig. 7. Average of the mean% of terminally differentiated older males and females involving all chromosomes under
MALE
DONORS
blood cells with aneuploidy investigation.
in younger
vs.
154
A.E. Mukherjee
et al. / Mechanisms
qf’ Aping
and Dewlopment
90 (1996) 14% 156
senescence of fibroblasts in vitro [16]. Since senescent cells cannot be distinguished from quiescent or terminally differentiated cells in tissues such as myeloid cells in peripheral blood used in this study, our previous results on aneuploidy levels in completely senescent fibroblasts in vitro is not directly comparable to the aneuploidy values of in vivo myeloid cells from older people. There is no convincing evidence that the non-replicating myeloid blood cells in vivo derived from older people necessarily undergo senescence in the same way, if at all, as we define in vitro senescence of hbroblasts (replicative senescence). Therefore, our previous report on decreased levels of chromosome-specific aneuploidy in completely senescent human fibroblasts in vitro is not in conflict with our present findings of increased levels of aneuploidy in myeloid cells in vivo with advancing age. Moreover, our present results on higher aneuploidy values with age is consistent with, previous report on the increased levels of aneuploidy in near-senescent but not completely senescent fibroblasts in vitro 1161. Among the five autosomes, the relatively higher mean frequencies of cells with aneuploidy were observed for chromosome nos. 4 and 6 in both the younger and older females and males. The lowest mean percentages of cells with autosomal aneuploidy were consistently noted for chromosome no. 3 in the younger as well as older male and female donors. In general, the same pattern of higher and lower chromosome-specific aneuploidy (autosomal) was observed in all the three cell types. These differential frequencies of chromosome-specific aneuploidy might indicate that, during myeloid differentiation and aging, the selection of a particular cell population to undergo differentiation and maturation with a particular type of chromosome-specific aneuploidy such as for chromosome nos. 4 and 6 might be less rigorous as compared to that of chromosome no. 3. Since the frequency of autosomal aneuploidy involving chromosome no. 3 was consistently found to be the lowest in all the three cell types derived from all individuals, it is conceivable that the retention of both copies of chromosome no. 3 (diploidy) as compared to that of chromosome nos. 4 and 6, for example, might be more beneficial for the process of survival and/or nuclear differentiation of myeloid cells. It was also noted that there was a notable increase in the mean frequency of the X chromosomal aneuploidy in older females and an increase in the frequency of Y chromosomal aneuploidy in the older males as compared to the corresponding values in the younger groups of females and males (Figs. 5 and 6). This increase was applicable to all the three types of differentiated leukocytes. Our present results on increased sex chromosomal aneuploidy in non-dividing, differentiated leukocytes in vivo as related to chronological aging are consistent with previous reports on cultured lymphocytes and bone marrow cells derived from various age-group people [3,17-201. In human fibroblasts, we have also reported increased sex chromosomal aneuploidy at interphase with in vitro aging of cells [21]. It appears that increased sex chromosomal aneuploidy with advanced chronological age is quite a common phenomenon in various types of human cells, both in vivo and in vitro. The genetic inactivation of one of the X chromosomes in females during early embryogenesis and the lack of much genetic information mented. It is conceivable
on the Y chromosome in males are now well docuthat, as they age, cells lacking the inactive X and/or the
A.B. Mukherjee et al. / Mechanisms qf Ageing and Development
90 (1996) 145-156
155
Y chromosome may develop greater survival capacity than normal cells in the context of selection against aneuploid cells. Our previous study clearly indicated that the proportion of interphase cells without the inactive X (i.e. the X-chromatin body) increases very significantly as human fibroblasts age in vitro [21]. The same study also showed that significant proportions of diploid male fibroblasts also lost the Y chromosome as in vitro aging progressed [21]. We believe, that during cellular aging, the numerical chromosome analysis at interphase both in vivo and in vitro is very likely to provide more accurate results as opposed to the analysis of only metaphase chromosomes, since the latter is completely dependent upon the divisional capacity of cells.
Acknowledgements
This investigation was supported in part by the New York State Education Department, Bureau of Professional Career Opportunity Programs.
References [l] P.A. Jacobs, W.M. Court-Brown and R. Doll, Distribution of human chromosome counts in relation to age. Nature, 191 (1961) 1178-l 180. [2] P.A. Jacobs, M. Brunton, W.M. Court-Brown, R. Doll and H. Goldstein, Changes in human chromosome count distributions with age: evidence for a sex difference. Nature. 197 (1963) 1080-1081. [3] SM. Galloway and K.E. Buckton. Aneuploidy and aging: chromosome studies on a random sample of the population using G-banding. Cytogenet. Cell Gent., 20 (1978) 78-95. [4] M.A. Bender, R.J. Preston, R.C. Leonard. B.E. Pyatt and P.C. Gooch, Chromosomal aberration and sister-chromatid exchange frequencies in peripheral blood lymphocytes of a large human population sample. 11. Extension of age range. Mutat. Res.. 21-7 (1989) 1499154. [5] D. Pinkel. T. Straume and J.W. Gray, Cytogenetic analysis using quantitative high sensitivity. fluorescence hybridization. Proc. Natl. Acad. Sci. USA. 83 (1986) 2934-2938. [6] H.F. Willard and J.S. Waye, Hierarchical order in chromosome-specific human alpha satellite DNA. Trends Genet., 3 (1987) 192- 198. [7] J.J. Maio. DNA strand reassociation and polyribonucleotide binding in the African green monkey, Cercopithecus aethiops. J. Mol. Biol., 56 (1971) 5799595. [S] D.M. Kurnit and J.J. Maio, Subnuclear distribution of DNA species in confluent and growing mammalian cells. Chromosoma, 42 (1973) 23 26. [9] L. Manuelidis, Chromosomal localization of complex and simple repeated human DNAs. Chromosoma. 6 (1978) 23332. [lo] D.A. Eastmond and D. Pinkel, Detection of aneuploidy and aneuploidy-inducing agents in human lymphocytes using fluorescence in situ hybridization with chromosome-specific DNA porves. Mutat. Res., 234 (1990) 303-318. [1 I] A.B. Mukherjee and N.Z. Parsa, Determination of sex Chromosomal constitution and Chromosoma1 origin of drumsticks. drumstick-like structures. and other nuclear bodies in human blood cells at interphase by fluorescence in situ hybridization. Chromosoma. 99 (1990) 432-435. [12] A.B. Mukherjee, V.V.V.S. Murty. E. Rodriguez, V.E. Reuter. C.J. Bosl and R.S.K. Chaganti, Detection and analysis of origin of i( 12p), a diagnostic marker of human male germ cell tumors by fluorescence in situ hybridization. Genes. Chrom. Cancer. 3 (1991) 300&307.
156
A.B. Mukherjee
et al. 1 Mechanisms
of’ Ageing and Development
90 (1996) 14% 156
[13] A.B. Mukherjee, V.V.V.S. Murty and R.S.K. Chaganti, Detection of cell cycle stage by fluorescence in situ hybridization: its application in human interphase cytogenetics. Cytogenet. Ccl/ Gent., 61 (1992) 91-94. [I41 D.E. Comings, Arrangement of chromatin in the interphase nucleus. Hum. Genet., 53 (1980) 131-143. [15] M.D. Bennett, The spatial distribution of chromosomes. In P.E. Brandham and M.D. Bennett (eds), Ken Chromosome Cotzjtirence ZZ, George Allen and Unwin. London, 1983, 71-79. [16] A.B. Mukherjee, S. Thomas and E. Schmitt, Chromosomal analysis in young vs. senescent human fibroblasts by fluorescence in situ hybridization: a selection hypothesis. Mech. Age. Dec., 80 (1995) I l-23. [I71 M.A. Abruzzo. M. Mayer and P.A. Jacobs, Aging and aneuploidy: evidence for the preferential involvement of the inactive X chromosome. Clytogenet. Cell Genet., 39 (1985) 275-278. [18] W.M. Court Brown, K.E. Buckton, P.A. Jacobs, I.M. Tough, E.V. Kuenssberg and J.D.E. Knox, Chromosome Studies on Adults, Cambridge University Press, London, 1966. [19] R.V. Pierre and H.C. Hagland, 45 X cell line in adult men: Loss of Y chromosome, a normal aging phenomenon. Mu~jo Clirr. Proc., 46 (1971) 52. [20] R.V. Pierre and H.C. Hoagland, Age-associated aneuploidy: Loss of Y chromosome from human bone marrow cells with aging. Cancer, 30 (1972) 889-894. [21] A.B. Mukherjee and K.C. Wallace, Changes in frequency and localization of human X- and Y-chromatin bodies at interphase during in t,itro cellular aging. Me&. Age. Deu., 53 (1990) 61-71.
and Mechanisms of Ageing and Development ELSEVIER
90 (1996) 145-156
Age-related aneuploidy analysis of human blood cells in vivo by fluorescence in situ hybridization (FISH) Asit B. Mukherjee
*, Jason Alejandro, Samuel Thomas
Shakira
Payne,
Department of‘ Biological Sciences. Fordham Univrrsit~~,Bronx, NY 10458, USA
Received 18 November 1995: revised 6 June 1996
Abstract All prior studies on human age-related chromosomal analysis were done using only metaphase figures derived from lymphocyte cultures in vitro. However, we believe that this procedure may provide only partial information, since the chromosomal abnormalities probably hidden in non-dividing and/or terminally differentiated leukocytes will not be detected by this method. We, therefore, have attempted to analyze the nature and extent of chromosome-specific aneuploidy at interphase by fluorescence in situ hybridization (FISH) in differentiated myeloid cells in vivo derived from various age-group people. Our data from an analysis of 30032 cells derived from 12 healthy donors indicate that there is an increase in the mean percentages of myeloid cells with chromosome-specific aneuploidy in the older groups as opposed to that of the younger groups. This increase is applicable to all the three cell types examined (promyelocytes, metamyelocytes and polymorphs). In both the younger and older females, the relatively higher mean frequencies of cells with aneuploidy were noted for chromosome nos. 4, 6 and the X, whereas the lowest mean frequencies of cells with aneuploidy were consistently observed for chromosome no. 3. In the younger and older male donors, similar to the female donors except the X chromosome the higher percentages of aneuploid cells were observed for chromosome nos. 4 and 6 whereas the lowest mean frequencies of aneuploid cells were noted for chromosome no. 3. Among the five autosomes studied, chromosome no. 3 consistently yielded the lowest frequency of aneuploidy in all the three cell types derived from both the younger and older groups of males and females. This *Corresponding
author. Tel.: + 1 718 8173663; fax: + I 718 8173645
0047-6374/96i$15.000
1996 Elsevier Science Ireland Ltd. All rights reserved
PII SOO47-6374(96)01762-9
146
A. B. Mukherjee et al. / Mechanisms of’ Age&
and Developmerzt 90 (1996) 145-156
could presumably mean that the maintenance of a very high frequency of cells with disomy for chromosome no. 3 might be more beneficial for the process of survival and/or differentiation of myeloid cells as compared to that of other autosomes such as chromosome nos. 4 and 6. Among the five autosomes studied, chromosome no. 1 exhibited the highest rate of increment in the aneuploidy values of both males and females with advancing age. Keywords:
Aneuploidy;
Blood cells; Aging: FISH
aneuploidy
analysis
1. Introduction An association between advanced chronological aging and increased chromosomal aneuploidy in human lymphocyte cultures was first reported by Jacobs et al. [1,2]. Although a number of subsequent studies, including a detailed report utilizing the G-banding technique [3], support this observation, a consistent correlation between chronological aging and chromosomal abnormality (numerical and/or structural) still remains to be fully explored. particularly after the development of molecular-cytogenetic techniques. For example, a previous study involving 493 subjects did not find any significant relationship of the frequency of any conventional aberration (structural) category to age with the sole exception of the dicentric chromosome, which showed a positive regression [4]. Also, all previous studies on the association between human chronological aging and chromosomal abnormality were done utilizing only metaphase figures derived from phytohemagglutinin-stimulated lymphocyte cultures in vitro. We, however, believe that the exclusive use of lymphocyte-derived metaphase chromosomes in the analysis of chromosomal aneuploidy as related to chronological aging may provide only partial information since the aneuploidy presumably hidden in nondividing and/or terminally differentiated leukocytes would remain undetected by this method. For this reason, we have now attempted to analyze the nature and extent of age-related chromosomal aneuploidy at interphase during human myeloid differentiation in vivo. The various types of terminally differentiated myeloid cells are non-dividing in peripheral blood and each cell type can be morphologically identified by the characteristic shape of its nucleus. In this study, we have used a powerful molecular-cytogenetic technique, called fluorescence in situ hybridization (FISH) [5]. We have utilized various chromosome-specific a-satellite DNA probes for accurate detection of chromosomal aneuploidy [6]. Alphoid DNA is a family of primate satellite DNAs located in the centromeric region of various chromosomes [791.
A.B. Mukherjee
et ul.
! Mechanisms oj’ Ageing anti Dmelopment
90 (1996) 145-156
147
2. Materials and methods 2.1. Blood filtn preparation Peripheral blood samples were derived from six males (ages: 21, 22, 22, 50, 52 and 66 years) and six females (ages: 20, 21, 22, 61, 76 and 79 years), all chromosomally normal, healthy individuals representing both the younger and older groups. Heparinized peripheral blood sample from each individual was allowed to settle for 2-3 h at room temperature and was then centrifuged at 1500 rpm for 5 min. White blood cells were then collected from the buffy coat and, using a Pasteur pipette, 2 drops of buffy coat suspension were placed at the center of a clean microscope slide towards the frosted end. A second microscope slide was then placed directly above the first slide, allowing surface tension to spread the buffy coat suspension on the slides. After 3-5 s, the two slides were glided apart horizontally, allowed to sit for 5- 10 s and then placed in 3: 1 methanol:glacial acetic acid fixative for 45 min. The slides were then air-dried and stored at 4°C. This method of slide preparation provided cytoplasm-free cell nuclei in 98.6% of the cases. All cell samples were screened for aneuploidy analysis within one month of the initial preparation of blood films. 2.2. Fluorescence
in situ hyhridixtion
(FISH)
Chromosome-specific aneuploidy analysis in 3 types of differentiated leukocytes (promyelocyte, metamyelocyte and polymorphonuclear leukocyte or polymorph) obtained from 12 male and female individuals representing both younger and older groups was done by FISH [5]. Seven chromosome-specific n-satellite DNA probes labeled with biotin (chromosome nos. 1, 2, 3, 4, 6, X and Y) were utilized in this preliminary study. These centromeric DNA probes were obtained from ONCOR (Gaithersburgh, MD). Molecular hybridization and immunofluorescent detection of each DNA probe were performed according to the manufacturer’s protocol using ONCOR chromosome in situ kit. The same FISH protocol was used for all cell samples hybridized with various DNA probes. In the scoring procedure, only cell nuclei with compact and discrete FISH signal morphology were selected for aneuploidy analysis. In accordance with a previously published method of aneuploidy analysis in human blood cells, if a portion of a slide showed more than two nuclei without the FISH signals (hybridization regions), that part of the slide was considered to be incompletely hybridized and these nuclei were not scored for the chromosomal analysis [lo]. Additionally, as control for in vivo blood cell data, we evaluated the occurrence of chromosomespecific monosomy (hypodiploidy) as well as trisomy-tetrasomy (hyperdiploidy) in phytohemagglutinin-stimulated blood lymphocytes in vitro from normal individuals using the exact same FISH protocol of the in vivo study. In separate experiments, the incidence of monosomy varied from 7% to 9% and trisomy and tetrasomy was less than 1%. Our results are similar to that of other investigators who used several chromosome-specific DNA probes for FISH analysis [lo]. On the basis of our
148
A.B. Mukherjee et al. / Mechanisms of’ Ageing and Development 90 (1996) 145-156
results, we set 10% as the error limit for detection of monosomy and the scoring for aneuploidy was done accordingly. A total of 30032 cells (ranging from 1950 to 3116 cells/sample) derived from 12 individuals were screened blindly with a Leitz Orthomat fluorescent microscope and representative photographs were taken according to our previously published methods [l l- 131. For FISH analysis of each chromosome, a separate slide was used for individual samples. 2.3. Statistical
analysis:
Statistical analyses were performed using a l-tailed Wilcoxon signed-rank test to compare the frequencies of aneuploidy between younger and older age-group people of both sexes. Significance of values were determined by using a 0.025 probability.
3. Results During sequential identified identified
human myeloid differentiation, the various cell types are generated in order and these non-dividing (interphase) cells can be morphologically by the characteristic shapes of their nuclei. For example, promyelocyte is by its large round or oval nucleus (Fig. l), metamyelocyte by its
Fig. I. A promyelocyte nucleus derived from a 22 year old female with two brightly signals representing disomy for the X chromosome.
fluorescent
FISH
A.B.
Mukhecjee
el al. / Mechanisms
of Ageing
and Deuelopmenr
90 (1996)
Fig. 2. A polymorph nucleus derived from a 79 year old female with a single brightly signal representing monosomy for the X chromosome.
145-156
fluorescent
149
FISH
kidney-shaped or horse-shoe-shaped nucleus (Fig. 3) and polymorph by its nucleus with lobules of dense chromatin connected by one or more thin filaments (Figs. 2 and 4). In this study, only these three cell types with characteristic nuclear rnorphologies were analyzed for aneuploidy. Each chromosome-specific fluorescent signal in a particular cell nucleus produced by FISH represented the presence of a specific chromosome. Thus, a nucleus was considered to be monosomic when only one fluorescent signal was present (Fig. 2). Disomic nuclei showed two signals, (Figs. l-4) etc. In this study, only cell nuclei with one and/or three FISH signals/cell were taken into account in determining the total yield of aneuploidy for a particular chromosome. Since cell nuclei with zero or four FISH signals/cell were extremely rare (O-0.2%), they were not included in this chromosome analysis. Fig. 5 summarizes the aneuploidy data from the three types of terminally differentiated blood cells in vivo (promyelocyte, metamyelocyte and polymorph) at interphase involving five autosomes and the X chromosome derived from three younger and three older females. The mean age of the younger group was 21 years, whereas that of the older group was 72 years. In the younger females, the vast majority (mean% ranging from 96.5 to 99.1) of the three cell types showed diploidy (disomy) for the chromosomes under investigation. The mean percentages of cell nuclei with aneuploidy was variable from chromosome to chromosome (mean% ranging from 0.9-3.5) in the three cell types. There were no marked differences in
150
A. B. Mukherjee et al. / Mechani.sms of Ageing and Development 90 11996) i45- 156
the mean percentages of cell nuclei with aneuploidy between the three cell types (2.4% in promyelocyte, 2.0% in metamyelocyte and 2.2% in polymorph). Of the five autosomes, the higher mean percentages of nuclei in the three cell types were consistently obtained for chromosome nos. 6 (3.5%) and 4 (3.3%), respectively, whereas the lowest mean percentage of nuclei with aneuploidy was observed for chromosomes no. 3. In the older group of females, similar to the younger group, the vast majority (mean% ranging from 95.9 to 98.4) of the three cell types in each of the chromosome-specific FISH preparations exhibited diploidy (disomy). Again, the higher mean frequencies of cell nuclei with aneuploidy was noted for chromosome nos. 4 (4.1%) and 6 (3.9%), whereas the lowest mean frequency of cells with aneuploidy (1.6%) was consistently recorded for chromosome no. 3 in all the three cell types. There were no pronounced differences in the mean percentages of cells with aneuploidy among the three cell types (3% in promnyclocyte, 2.6O/oin metamyelocyte and 3.3% in polymorphs). However, in older females, there was a statistically significant increase in the mean frequencies of aneuploid cells involving all chromosomes under study as compared to that of the younger females (P < 0.025). Of the five autosomes, chromosome no. 1 showed the highest rate of increase in aneuploidy value (from 1.5% to 2.6%) with advancing age. Similar trend in the aneuploidy value was observed for the X chromosome in older females.
Fig. 3. A metamyelocyte nucleus derived from a 21 year old male with a single brightly signal representing the Y chromosome.
fluorescent
FISH
A.B. Mukhecjee
et al. ! Mechanisms
of Ageing and Deueloprrlent 90 (1996) 145-156
Fig. 4. Three polymorph nuclei derived from a 56 year old male each with two brightly signals representing disomy for chromosome no. 3.
fluorescent
151
FISH
Fig. 6 presents our data on chromosome-specific aneuploidy at interphase involving seven chromosomes (5 autosomes, X and Y) in the three types of myeloid cells in vivo derived from three younger and three older males. The mean age of the younger donors was 21.6 years whereas that of the older donors was 56 years.
Fig. 5. Combined and polymorphs
mean% nuclei with chromosome-specific aneuploidy in promyelocytes, in younger (mean age = 21) vs. older (mean age = 72) females.
metamyelocytes
152
A.B. Mukherjee
et al. 1 Mechanisms
of Ageing and Development
90 (1996) 145-156
Fig. 6. Combined mean%) nuclei with chromosome-specific aneuploidy in promyelocytes, metamyelocytes and polymorphs in younger (mean age = 21.6) vs. older (mean age = 56) males.
Among all five autosomes, the highest (99%) and the lowest (96.9%) mean frequencies of diploid cells were observed for chromosome nos. 3 and 4, respectively. The highest level of aneuploidy was noted for chromosome no. 4 (3.1%), followed by chromosome no. 6 (2.9%) in this order. The relative proportions of cells with aneuploidy for chromosome nos. 1, 2 and the X were almost identical (1.4- 1.5%). In the younger males, similar to the younger females, the lowest mean percentage of cells (1.0%) with aneuploidy was consistently recorded for chromosome no. 3. The aneuploidy level of the Y chromosome was slightly higher (1.9%) than that of the X chromosome (1.5%). There were no marked differences in the mean frequencies of cells with aneuploidy among the three cell types (2.1% for promyelocyte, 1.8% for metamyelocyte and 2.0% for polymorph). In the older males, the highest (3.3%) and the lowest (1.3%) mean percentages of aneuploid cells were noted for chromosome nos. 4 and 3, respectively. The second highest level of aneuploidy was observed for chromosome no. 6, and chromosome no. 1 showed the highest rate of increment in aneuploidy value (from 1.4% in younger vs. 2.5% in older males) with advancing age, similar to that of older females. There were no striking differences in the mean frequencies of aneuploid cells among the three cell types (2.3% for promyelocyte, 2.6% for metamyelocyte and 2.4% for polymorph). However, with advancing age, there was a statistically significant increase in the aneuploidy values for all chromosomes under investigation (P > 0.025).
4. Discussion It is now well-recognized that the arrangement of chromosomes in plant and animal cells does not occur at random and specific chromosomes occupy specific
A.B. Mukherjee
et ul. / Mechanisms
of .dgeing and Development
90 (1996) 145-156
153
territories (called chromosomal domains) in the interphase nuclei [14,15]. More recently, numerous studies have shown that FISH with chromosome-specific centromeric DNA probes (cc-satellite) can be effectively utilized to detect specific chromosomes as fluorescent spots (FISH signals) in an interphase nucleus. This study has utilized the FISH technique to examine the extent of aneuploidy at interphase involving five automes and two sex chromosomes of terminally differentiated myeloid cells in the younger and older age-group people. Monosomy of a specific chromosome constituted the higher proportion of aneuploidy in all the three cell types derived from each individual. This finding is consistent with similar studies done with cultured lymphocytes [lo]. A comparative analysis of the average of the mean percentages of cell nuclei with each chromosome-specific aneuploidy under investigation indicates that, for any one particular age group in both sexes, there is no striking difference in the degree of aneuploidy in the three different cell types (i.e. promyelocytes, metamyelocytes and polymorphs) (Fig. 7). However, there is a clear trend in the increase of the mean percentages of cells with aneuploidy in the older groups of males and females as opposed to that of the younger groups (Figs. 5-7). This increase is applicable to all the three cell types studied. This finding is similar to our previous results on aneuploidy at metaphase in near-senescent human fibroblasts in vitro [16]. However, in the same study, we noted relatively lower frequencies of aneuploidy in completely senescent, non-dividing human skin fibroblasts in vitro as compared to the frequencies of aneuploidy in actively dividing young cells. As a possible explanation to this interesting finding, we proposed a selection hypothesis relevant to cellular survival at replicative
FEMALE
DONORS
Fig. 7. Average of the mean% of terminally differentiated older males and females involving all chromosomes under
MALE
DONORS
blood cells with aneuploidy investigation.
in younger
vs.
154
A.E. Mukherjee
et al. / Mechanisms
qf’ Aping
and Dewlopment
90 (1996) 14% 156
senescence of fibroblasts in vitro [16]. Since senescent cells cannot be distinguished from quiescent or terminally differentiated cells in tissues such as myeloid cells in peripheral blood used in this study, our previous results on aneuploidy levels in completely senescent fibroblasts in vitro is not directly comparable to the aneuploidy values of in vivo myeloid cells from older people. There is no convincing evidence that the non-replicating myeloid blood cells in vivo derived from older people necessarily undergo senescence in the same way, if at all, as we define in vitro senescence of hbroblasts (replicative senescence). Therefore, our previous report on decreased levels of chromosome-specific aneuploidy in completely senescent human fibroblasts in vitro is not in conflict with our present findings of increased levels of aneuploidy in myeloid cells in vivo with advancing age. Moreover, our present results on higher aneuploidy values with age is consistent with, previous report on the increased levels of aneuploidy in near-senescent but not completely senescent fibroblasts in vitro 1161. Among the five autosomes, the relatively higher mean frequencies of cells with aneuploidy were observed for chromosome nos. 4 and 6 in both the younger and older females and males. The lowest mean percentages of cells with autosomal aneuploidy were consistently noted for chromosome no. 3 in the younger as well as older male and female donors. In general, the same pattern of higher and lower chromosome-specific aneuploidy (autosomal) was observed in all the three cell types. These differential frequencies of chromosome-specific aneuploidy might indicate that, during myeloid differentiation and aging, the selection of a particular cell population to undergo differentiation and maturation with a particular type of chromosome-specific aneuploidy such as for chromosome nos. 4 and 6 might be less rigorous as compared to that of chromosome no. 3. Since the frequency of autosomal aneuploidy involving chromosome no. 3 was consistently found to be the lowest in all the three cell types derived from all individuals, it is conceivable that the retention of both copies of chromosome no. 3 (diploidy) as compared to that of chromosome nos. 4 and 6, for example, might be more beneficial for the process of survival and/or nuclear differentiation of myeloid cells. It was also noted that there was a notable increase in the mean frequency of the X chromosomal aneuploidy in older females and an increase in the frequency of Y chromosomal aneuploidy in the older males as compared to the corresponding values in the younger groups of females and males (Figs. 5 and 6). This increase was applicable to all the three types of differentiated leukocytes. Our present results on increased sex chromosomal aneuploidy in non-dividing, differentiated leukocytes in vivo as related to chronological aging are consistent with previous reports on cultured lymphocytes and bone marrow cells derived from various age-group people [3,17-201. In human fibroblasts, we have also reported increased sex chromosomal aneuploidy at interphase with in vitro aging of cells [21]. It appears that increased sex chromosomal aneuploidy with advanced chronological age is quite a common phenomenon in various types of human cells, both in vivo and in vitro. The genetic inactivation of one of the X chromosomes in females during early embryogenesis and the lack of much genetic information mented. It is conceivable
on the Y chromosome in males are now well docuthat, as they age, cells lacking the inactive X and/or the
A.B. Mukherjee et al. / Mechanisms qf Ageing and Development
90 (1996) 145-156
155
Y chromosome may develop greater survival capacity than normal cells in the context of selection against aneuploid cells. Our previous study clearly indicated that the proportion of interphase cells without the inactive X (i.e. the X-chromatin body) increases very significantly as human fibroblasts age in vitro [21]. The same study also showed that significant proportions of diploid male fibroblasts also lost the Y chromosome as in vitro aging progressed [21]. We believe, that during cellular aging, the numerical chromosome analysis at interphase both in vivo and in vitro is very likely to provide more accurate results as opposed to the analysis of only metaphase chromosomes, since the latter is completely dependent upon the divisional capacity of cells.
Acknowledgements
This investigation was supported in part by the New York State Education Department, Bureau of Professional Career Opportunity Programs.
References [l] P.A. Jacobs, W.M. Court-Brown and R. Doll, Distribution of human chromosome counts in relation to age. Nature, 191 (1961) 1178-l 180. [2] P.A. Jacobs, M. Brunton, W.M. Court-Brown, R. Doll and H. Goldstein, Changes in human chromosome count distributions with age: evidence for a sex difference. Nature. 197 (1963) 1080-1081. [3] SM. Galloway and K.E. Buckton. Aneuploidy and aging: chromosome studies on a random sample of the population using G-banding. Cytogenet. Cell Gent., 20 (1978) 78-95. [4] M.A. Bender, R.J. Preston, R.C. Leonard. B.E. Pyatt and P.C. Gooch, Chromosomal aberration and sister-chromatid exchange frequencies in peripheral blood lymphocytes of a large human population sample. 11. Extension of age range. Mutat. Res.. 21-7 (1989) 1499154. [5] D. Pinkel. T. Straume and J.W. Gray, Cytogenetic analysis using quantitative high sensitivity. fluorescence hybridization. Proc. Natl. Acad. Sci. USA. 83 (1986) 2934-2938. [6] H.F. Willard and J.S. Waye, Hierarchical order in chromosome-specific human alpha satellite DNA. Trends Genet., 3 (1987) 192- 198. [7] J.J. Maio. DNA strand reassociation and polyribonucleotide binding in the African green monkey, Cercopithecus aethiops. J. Mol. Biol., 56 (1971) 5799595. [S] D.M. Kurnit and J.J. Maio, Subnuclear distribution of DNA species in confluent and growing mammalian cells. Chromosoma, 42 (1973) 23 26. [9] L. Manuelidis, Chromosomal localization of complex and simple repeated human DNAs. Chromosoma. 6 (1978) 23332. [lo] D.A. Eastmond and D. Pinkel, Detection of aneuploidy and aneuploidy-inducing agents in human lymphocytes using fluorescence in situ hybridization with chromosome-specific DNA porves. Mutat. Res., 234 (1990) 303-318. [1 I] A.B. Mukherjee and N.Z. Parsa, Determination of sex Chromosomal constitution and Chromosoma1 origin of drumsticks. drumstick-like structures. and other nuclear bodies in human blood cells at interphase by fluorescence in situ hybridization. Chromosoma. 99 (1990) 432-435. [12] A.B. Mukherjee, V.V.V.S. Murty. E. Rodriguez, V.E. Reuter. C.J. Bosl and R.S.K. Chaganti, Detection and analysis of origin of i( 12p), a diagnostic marker of human male germ cell tumors by fluorescence in situ hybridization. Genes. Chrom. Cancer. 3 (1991) 300&307.
156
A.B. Mukherjee
et al. 1 Mechanisms
of’ Ageing and Development
90 (1996) 14% 156
[13] A.B. Mukherjee, V.V.V.S. Murty and R.S.K. Chaganti, Detection of cell cycle stage by fluorescence in situ hybridization: its application in human interphase cytogenetics. Cytogenet. Ccl/ Gent., 61 (1992) 91-94. [I41 D.E. Comings, Arrangement of chromatin in the interphase nucleus. Hum. Genet., 53 (1980) 131-143. [15] M.D. Bennett, The spatial distribution of chromosomes. In P.E. Brandham and M.D. Bennett (eds), Ken Chromosome Cotzjtirence ZZ, George Allen and Unwin. London, 1983, 71-79. [16] A.B. Mukherjee, S. Thomas and E. Schmitt, Chromosomal analysis in young vs. senescent human fibroblasts by fluorescence in situ hybridization: a selection hypothesis. Mech. Age. Dec., 80 (1995) I l-23. [I71 M.A. Abruzzo. M. Mayer and P.A. Jacobs, Aging and aneuploidy: evidence for the preferential involvement of the inactive X chromosome. Clytogenet. Cell Genet., 39 (1985) 275-278. [18] W.M. Court Brown, K.E. Buckton, P.A. Jacobs, I.M. Tough, E.V. Kuenssberg and J.D.E. Knox, Chromosome Studies on Adults, Cambridge University Press, London, 1966. [19] R.V. Pierre and H.C. Hagland, 45 X cell line in adult men: Loss of Y chromosome, a normal aging phenomenon. Mu~jo Clirr. Proc., 46 (1971) 52. [20] R.V. Pierre and H.C. Hoagland, Age-associated aneuploidy: Loss of Y chromosome from human bone marrow cells with aging. Cancer, 30 (1972) 889-894. [21] A.B. Mukherjee and K.C. Wallace, Changes in frequency and localization of human X- and Y-chromatin bodies at interphase during in t,itro cellular aging. Me&. Age. Deu., 53 (1990) 61-71.