Mechanisms of Ageing and Development
92 (1996) 101 – 109

A longitudinal study of human age-related
ribosomal RNA gene activity as detected by
silver-stained NORs
Samuel Thomas, Asit B. Mukherjee*
Department of Biological Sciences, Fordham Uni6ersity, 441 East Fordham Road, Bronx,
NY 10458, USA
Received 1 October 1996; accepted 10 October 1996

Abstract
The relative frequencies of silver-stained nucleolar organizing regions (Ag-NORs) as a
function of age have been analyzed in skin fibroblasts derived from eight adult individuals
participating in the Gerontology Research Center (GRO) Longitudinal Study, NIA, Baltimore, MD. Since silver staining of NORs is correlated with rRNA gene activity, we used this
cytological method to examine the pattern of rRNA gene activity in specific individuals, each
at two different ages. Our results show that the average number of Ag-NORs/cell decreases
significantly with advancing age of each individual, presumably indicating a general pattern
of age-related decline/alteration in rRNA gene activity and this pattern is individual-specific.
The findings of our longitudinal study is consistent with the results of previous cross-sectional (population) studies on rRNA gene activity as detected by Ag-NORs. However, it
appears that the relative rate in the age-related decline of rRNA gene activity, as evidenced
by lower Ag-NOR frequencies with age, is variable from person to person. © 1996 Elsevier
Science Ireland Ltd.
Keywords: Age-related; rRNA genes; Longitudinal study

* Corresponding author: Tel.: + 1 718 8173663; fax: +1 718 8173645.
0047-6374/96/$15.00 © 1996 Elsevier Science Ireland Ltd. All rights reserved.
PII S 0 0 4 7 - 6 3 7 4 ( 9 6 ) 0 1 8 0 5 - 2

102

S. Thomas, A.B. Mukherjee / Mechanisms of Ageing and De6elopment 92 (1996) 101–109

1. Introduction
In humans, the 18s and 28s rRNA gene clusters have been localized in the
secondary constriction regions of five pairs of D and G group (acrocentric)
chromosomes [1 – 3]. These chromosomal locations are called NORs or nucleolar
organizing regions. Strehler and coworkers have previously reported an age-related
loss of rRNA genes in heart, skeletal muscle and brain tissues from beagle dogs,
human cardiac tissue and human cerebral cortex cells [4–9]. Their results indicate
that the actual loss of rRNA genes and/or alteration of rRNA gene activity might
be related to the aging phenomenon of mammalian species.
It is now well-established that a simple cytological method, called the silver
staining or N-banding method, can clearly detect the active NORs of five pairs of
human D and G group chromosomes at metaphase [10–15]. It is also known that
positive silver staining, as revealed by darkly-stained spots at NORs (Ag-NORs), is
indicative of transcriptionally active rRNA genes, whereas the negatively silverstained NORs represent transcriptionally repressed and/or lost rRNA genes
[10,14,16 – 18]. It is, therefore, possible to examine cytologically the degree of rRNA
gene activity in a given metaphase cell by scoring the number of silver-stained
NORs (Ag-NORs) of human acrocentric chromosomes.
Several cross sectional (population) studies previously indicated an age-related
decrease in the mean number of Ag-NORs/cell in normal individuals [19–21].
However, the Ag-NORs are quite variable from person to person and it is rather
difficult to interpret the cross-sectional data in relation to aging within a specific
individual at different ages. Since the number Ag-NORs/cell is fairly constant for a
given individual at a particular age, only longitudinal studies provide the opportunity to assess the pattern of rRNA gene activity as related to cellular aging of a
specific individual. For this reason, we have now carried out, for the first time, a
longitudinal study of age-related rRNA gene activity in normal adults, each at two
different ages.

2. Materials and methods

2.1. Cells
The human cells used in this study were obtained from the National Institute on
Aging (NIA) Cell Culture Repository at Camden, NJ. The cells were derived from
8 individuals (5 males and 3 females), each at two different ages (i.e. a total of 16
cell cultures). The skin biopsies for fibroblasts were taken from the same position
of the body and the cell cultures were obtained at the earliest population doubling
levels (PDLs) available. The specific cell strains and their corresponding Repository
numbers are listed in Table 1.
The cells were cultured in T-75 flasks at 37°C using MEM Alpha Medium
(Gibco) supplemented with 20% fetal bovine serum, 1% L-glutamine and antibiotics
(100 I.U./ml penicillin, 100 vg/ml streptomycin; Gibco). All culture conditions were

S. Thomas, A.B. Mukherjee / Mechanisms of Ageing and De6elopment 92 (1996) 101–109

103

kept constant for the entire experiment. Each cell strain was grown for a short
period just to obtain enough cells for chromosome preparation.

2.2. Chromosome preparation and sil6er staining
For metaphase chromosome preparation, the experimental cell cultures were
treated with colcemid (0.05 vg/ml) for 2–3 h before being trypsinized. The
trypsinized cells were then centrifuged for 10 min at 1000 rpm. The cells were
exposed to hypotonic solution (complete medium:distilled water, ratio 1:3) for 15
min and were prefixed at the end of hypotonic treatment. The cells were then
centrifuged for 10 min at 1000 rpm and fixed in 1:3 fixative (glacial acetic
acid:absolute methanol) for 1 h. The cells were fixed at least 3 more times for 30
min each, before being dropped onto cold, wet slides and air-dried. The slides were
stored for a week before being used for silver staining of NORs. The slides were
processed for silver staining using the gelatin–silver nitrate method [14,15].
Fifty to sixty complete metaphase figures/sample were screened blindly under a
Leitz photomicroscope and representative photographs were taken. The slides were
coded and the D and G group chromosomes were examined for presence of
Ag-NORs.

2.3. Statistical analysis
Statistical analyses were performed using a 1-tailed Wilcoxon signed-ranks test to
compare the relative frequencies of silver-stained NORs/cell at younger and at older
ages of each individual of both sexes. Significance of values were determined by
using a 0.005 probability.

Table 1
Sixteen skin fibroblast cultures derived from 5 males and 3 females (each at two different ages) utilized
in the longitudinal study
Individual No.

Sex

Skin fibroblasts obtained at
Younger age (years)

1
2
3
4
5
6
7
8

M
M
M
M
M
F
F
F

Older age (years)

42
55
59
63
76
47
73
87

51
63
63
69
85
53
79
92

(AG05416)
(AG05192)
(AG04353)
(AG06881)
(AG04144)
(AG05837B)
(AG06952)
(AG05247E)

(AG11364)
(AG09844)
(AG05804B)
(AG11247)
(AG09558A)
(AG11248A)
(AG11020)
(AG09602)

M, male; F, female. The numbers in parentheses indicate the Cell Repository numbers for specific cell
cultures.

104

S. Thomas, A.B. Mukherjee / Mechanisms of Ageing and De6elopment 92 (1996) 101–109

Table 2
Frequency distribution of Ag-NORs in D and G group chromosomes at metaphase derived from
firoblast cultures of eight individuals at younger ages
Individual No.

Sex

Age

Mean Ag-NORs/cell

NOR +ve chromosomes/cell
D group (mean)

1
2
3
4
5
6
7
8

M
M
M
M
M
F
F
F

G group (mean)

7.0
7.5
7.8
7.4
7.6
6.9
7.5
7.3

4.5
4.7
5.1
4.8
5.3
4.6
5.3
5.3

2.5
2.8
2.7
2.6
2.3
2.3
2.2
2.0

62.7

Mean

42
55
59
63
76
47
73
87

7.37

4.95

2.42

M, male; F, female.

3. Results
Tables 2 and 3 present the mean number of Ag-NORs/cell and the mean number
of Ag-NORs in D and G group chromosomes/cell for 8 individuals each at younger
and at older ages, respectively. At younger ages of various individuals (ranging
from 42 to 87 years), the range of mean Ag-NORs/cell lies between 6.9–7.8,
whereas that at older ages is found to be between 5.3–6.4. For each individual, the
average number of Ag-NORs/cell decreases significantly with advancing age (compare Tables 2 and 3) (PB0.005). Both the D and G group chromosomes show a
decrease in the mean number of Ag-NORs/cell with advancing age in each
Table 3
Frequency distribution of Ag-NORs in D and G group chromosomes at metaphase derived from
firoblast cultures of eight individuals at older ages
Individual No.

Sex

Age

Mean Ag-NORs/cell

NOR +ve chromosomes/cell
D group (mean)

1
2
3
4
5
6
7
8
Mean
M, male; F, female.

M
M
M
M
M
F
F
F

G group (mean)

51
63
63
69
85
53
79
92

5.5
6.0
5.7
5.5
6.4
5.3
6.2
5.6

3.5
4.0
4.6
4.0
4.3
3.6
4.4
4.5

2.0
2.0
1.1
1.5
2.1
1.7
1.8
1.1

69.4

5.77

4.11

1.66

S. Thomas, A.B. Mukherjee / Mechanisms of Ageing and De6elopment 92 (1996) 101–109

105

Fig. 1. Fibroblast-derived metaphase figures showing variable numbers of Ag-NOR/cell in individual
number 7 at two different ages (Table 1): (a) eight Ag-NORs at age 73 (arrows); and (b) five Ag-NORs
at age 79 (arrows).

individual and no significant difference is observed in the mean number of
Ag-NORs/cell between males and females. The average of the mean numbers of
Ag-NORs/cell in all 8 individuals declines from 7.37 at younger ages to 5.77 at
older ages (compare Tables 2 and 3). Fig. 1(a) and (b) show eight and five
Ag-NORs/metaphase derived from individual no. 7 (Table 1) at ages 73 and 79
years, respectively. Fig. 2 graphically presents a comparison of the frequency
distributions of Ag-NORs/metaphase cell derived from 8 individuals, each at
younger and at older age.
Table 4 analyzes the possible relationship between a specific time period (years)
spent in one’s life span and the corresponding decline in the mean number of
Ag-NORs/cell derived from 8 different individuals. Our results indicate that there is
no consistent pattern relating a specific timespan spent in one’s longevity and the

106

S. Thomas, A.B. Mukherjee / Mechanisms of Ageing and De6elopment 92 (1996) 101–109

Fig. 2. Frequencies of Ag-NORs in metaphase chromosomes of fibroblasts derived from 8 individuals,
each at younger vs. older ages.

corresponding decrease in the mean number of Ag-NORs/cells, presumably reflecting reduced rRNA gene activity. For example, individual number 1, in a timespan
of 9 years, exhibits a reduction of 1.5 Ag-NORs/cell (mean) whereas individual
number 3, in a time period of 4 years, shows a larger decline of 2.1 Ag-NORs/cell
(mean). All 8 individuals exhibit their unique patterns in the relationship between
a specific timespan spent in one’s longevity and its corresponding decline in the
mean number of Ag-NORs/cell, i.e. the degree of reduction in the rRNA gene
activity (Table 4).

Table 4
Relationship between specific time-span in longevity and corresponding decrease in mean Ag-NORs/
cell in 8 inviduals
Individual No.

Sex

Timespan between younger
and older ages (years)

Amount of decrease in mean
Ag-NORs/cell between younger
and older ages

1
2
3
4
5
6
7
8

M
M
M
M
M
F
F
F

9
8
4
6
9
6
6
5

1.5
1.5
2.1
1.9
1.2
1.6
1.3
1.7

M, male; F, female.

S. Thomas, A.B. Mukherjee / Mechanisms of Ageing and De6elopment 92 (1996) 101–109

107

4. Discussion
Our results clearly indicate that, during aging of skin fibroblasts in each individual under study, there is a gradual reduction in the mean number of AgNORs/cell.
This observation presumably indicates that there is a general phenomenon of
alteration/reduction in the activity of rRNA genes with advancing age in at least
certain tissues such as human skin fibroblasts. The results of our longitudinal study
on age-related Ag-NOR frequencies in human skin fibroblasts are consistent with
previous cross-sectional findings on age-related Ag-NOR frequencies in human
lymphocytes and fibroblasts [19–21]. Moreover, this longitudinal study unlike a
cross-sectional study, avoids the inter-individual genetic variations as a possible
modulating factor in the display of differential degrees of Ag-NOR frequencies as
related to human aging. For example, individual number 2 at age 55 displays the
same mean number of Ag-NORs/cell (7.5) as compared to that of individual
number 7 at age 73 (7.5) and individual number 1 at age 42 shows a lower value in
the mean member of Ag-NORs/cell (7.0) as opposed to that of individual number
5 (7.6) at age 76 (Table 2). Since there is inter-individual variation in the expression
of rRNA genes as revealed by silver staining of NORs, only a longitudinal study,
and not a cross-sectional study, can provide the most meaningful interpretation of
rRNA gene activity in a specific individual at a specific age. Our study also shows
both intra- and inter-individual variations in the age-related frequencies of AgNORs/cell in human skin fibroblasts.
Strehler and coworkers have reported an age-related loss of rRNA genes in
certain tissues of human and other mammalian species [5,6,8,9]. However, the same
investigators reported no significant differences in rRNA gene dosage as a function
of age in tissues such as liver, kidney and spleen of beagle dogs [6]. In order to
support the view that rRNA genes are being lost during aging of nonreplenishable
cells, Strehler has noted that there is a loss of nucleolar organizing regions (NORs,
or sites of rRNA gene transcription) that correlated with the measured loss of
rRNA genes [4]. Other investigators have found that rRNA gene dosage remained
the same for brain, liver, spleen and kidney tissues throughout the major part of the
adult lifespan of C57BL/6J mice but a striking reduction in the rRNA gene
hybridization was noted after 800 days (i.e. about 60% of lifespan potential) [22,23].
Although the mechanisms that might be responsible for these observations remain
unclear, the hypothesis of age-dependent genetic loss from the chromosomal DNA
has received considerable prominence [5,6]. It might also be possible that rRNA
genes are not actually lost from the chromosomal DNA but some genes might not
be available to hybridization probes after they are covered by protein or other
cross-linkers [23]. Although there is some experimental evidence in support of this
notion [23,24], no clearcut evidence has yet been provided. It is also known that
gene methylation might contribute in the age-related loss of NORs in rodent
post-mitotic cells [25], and genomic instability during aging of post mitotic mammalian cells has been extensively documented [26].
Our longitudinal study is in agreement with the basic premise that rRNA gene
sites/activities are altered during aging of human cells and that silver specifically

108

S. Thomas, A.B. Mukherjee / Mechanisms of Ageing and De6elopment 92 (1996) 101–109

stains active rRNA gene sites (NORs) in five pairs of human D and G group
chromosomes. Although the relative frequencies of active NORs (Ag-NORs)/cell
could not be determined in the very early stages of individual lifespans in this
longitudinal study due to unavailability of suitable cell cultures from the Cell
Repository, the present investigation seems to indicate that the loss of rRNA gene
activity with aging is gradual, individual-specific and can occur throughout one’s
lifespan. It is quite possible that an irreversible loss/change in rRNA gene activity
takes place with a concomitant decline in the production of rRNA and this
interpretation is consistent with the findings of Strehler et al. [8,9]. Also, as pointed
out by Denton et al. [20], there must be some degree of rRNA gene repression/alteration at all ages, because all five pairs of human D and G group chromosomes
seldom exhibit ten Ag-NORs/cell in any specific individual. However, we show that,
in a particular individual, there is a specific pattern of age-related decline in rRNA
gene activity as evidenced by the relative frequencies of Ag-NORs/cell.
Our study also indicates that there is no consistent and generalized pattern
relating a specific timespan (years) spent in one’s longevity and the corresponding
decrease in the mean number of Ag-NORs/cell, reflecting reduced rRNA gene
activity. For example, individual number 3, in a period of 4 years, shows a
reduction of 2.1 Ag-NORs/cell (mean), whereas individual number 5, in a timespan
of 9 years, exhibits only 1.2 Ag-NORs/cell (mean) (Table 4). It appears then, that
the relative rate in the age-related decline of rRNA gene activity, as revealed by
Ag-NOR frequencies, in tissues such as human skin fibroblasts is different from
individual to individual.

References
[1] A.S. Henderson, D. Warburton and K.C. Atwood, Location of ribosomal DNA in the human
chromosome complement. Proc. Natl. Acad. Sci. USA, 69 (1972) 3394 – 3398.
[2] H.J. Evans, R.A. Buckland and M.L. Pardue, Location of gene coding for 18s and 28s ribosomal
RNA in the human genome. Chromosoma, 48 (1974) 405 – 426.
[3] T.C. Hsu, S.E. Spirito and M.L. Pardue, Distribution of 18s and 28s ribosomal genes in
mammalian genomes. Chromosoma, 53 (1975) 25 – 36.
[4] B.L. Streheler, Genetic instability as the primary cause of human aging. Exp. Geront., 21 (1986)
283–319.
[5] R.W. Johnson and B.L. Strehler, Loss of genes coding for ribosomal RNA in aging brain cells.
Nature, 240 (1972) 412–414.
[6] R.W. Johnson, C. Chrisp and B.L. Strehler, Selective loss of ribosomal RNA genes during the aging
of post-mitotic tissues, Mech. Age. De6., 1 (1972) 183 – 198.
[7] L.K. Johnson, R.W. Johnson and B.L. Strehler, Cardiac hypertrophy, aging and changes in cardiac
ribosomal RNA gene dosage in man. J. Mol. Cell Cardiol., 7 (1975) 125 – 133.
[8] B.L. Strehler, M. Chang and L.K. Johnson, Loss of hybridizable ribosomal DNA from human
post-mitotic tissues during aging: I. Age-dependent loss of human myocardium. Mech. Ageing De6.,
11 (1979) 371–378.
[9] B.L. Streheler and M. Chang, Loss of hybridizable ribosomal DNA from human post-mitotic
tissues during ageing. II. Age-dependent loss in human cerebral cortex – hippocampal and somatosensory cortex comparison. Mech. Ageing De6., 11 (1979) 379 – 382.

S. Thomas, A.B. Mukherjee / Mechanisms of Ageing and De6elopment 92 (1996) 101–109

109

[10] C. Goodpasture and S.E. Bloom, Visualization of nucleolar organizer regions in mammalian
chromosomes using silver staining. Chrornosoma, 53 (1975) 37 – 50.
[11] S.E. Bloom and C. Goodpasture, An improved technique for selective silver staining of nucleolar
organizer regions in human chromosomes. Hum. Genet., 34 (1976) 199 – 206.
[12] T.E. Denton, W.M. Howell and J.V. Barrett, Human nucleolar organizer chromosomes: satellite
associations. Chromosoma, 55 (1976) 81 – 84.
[13] T.E. Denton, W.R. Brooke and W.M. Howell, A technique for simultaneous staining of both
nucleolar organizer regions and kinetochores of human chromosomes with silver. Stain Technol., 52
(1977) 311–313.
[14] W.M. Howell and D.A. Black, Controlled silver staining of nucleolar organizing regions with a
protective colloidal developer: a 1-step method. Experimentia, 36 (1980) 1014 – 1015.
[15] J.R. Gold and J.R. Ellison, Silver staining for nucleolar organizing regions of vertebrate chromosome. Stain Technol., 58 (1983) 51 – 55.
[16] D. Warburton and A.S. Henderson, Sequential silver staining and hybridization in situ on nucleolus
organizing regions in human cells. Cytogenet. Cell Genet., 24 (1979) 168 – 175.
[17] C.C. Morton, J.A. Brown, W.M. Holmes, W.E. Nance and B. Wolf, Stain intensity of human
nucleolus organizer region reflects incorporation of uridine into mature ribosomal RNA. Exp. Cell
Res., 145 (1983) 405–413.
[18] E. Julian, M. Hartung, A.M. Vagner and A. Stahl, Association of ribosomal genes in the human
oocyte at meiotic prophase: cytogenetic consequences. Chrom. Today, 9 (1987) 301 – 309.
[19] C.H.C.M. Buys, J. Osinga and G.J.P.A. Anders, Age-dependent variability of ribosomal RNA-gene
activity in man as determined from frequencies of silver staining nucleolus organizing regions of
metaphase chromsomes of lymphocytes and fibroblasts. Mech. Ageing De6., 11 (1979) 55 – 75.
[20] T.E. Denton, S.L. Liem, K.M. Cheng and J.V. Barrett, The relationship between aging and
ribosomal gene activity in humans as evidenced by silver staining. Mech. Ageing De6., 15 (1981)
1–7.
[21] B.C. Das, R. Rani and U.K. Luthra, The number of silver-staining NORs in lymphocytes of
newborns and its relationship to human development. Mech. Ageing De6., 36 (1986) 117 – 123.
[22] J. Gaubatz, N. Prashad and R.G. Cutler, Ribosomal RNA gene dosage as a function of tissue and
age for mouse and human. Biochim. Biophys. Acta, 418 (1976) 358 – 375.
[23] J. Gaubatz and R.G. Cutler, Age-related differences in the number of ribosomal RNA genes of
mouse tissues. Gerontoloy, 24 (1978) 179 – 207.
[24] R.G. Cutler, Cross linking hypothesis of aging: DNA adducts in chromatin as a primary aging
process. In K.C. Smith (ed.), Aging, Carcinogenesis and Radiation Biology. The Role of Nucleic Acid
Addition Reactions, Plenum, New York, 1976, p. 443.
[25] K. Swisshelm, C.M. Disteche, J. Thorvaldsen, A. Nelson and D. Salk, Age-related increase in
methylation of ribosomal genes and inactivation of chromosome-specific rRNA gene clusters in
mouse. Mutat. Res., 237 (1990) 131 – 146.
[26] J.W. Gaubatz, Genomic instability during aging of post mitotic mammalian cells. In A. MacieiraCoelho (ed.), Molecular Basis of Aging, CRC Press, New York, 1995, pp. 71 – 135.