|
|
||||||||
Research Communications |
a Blood and Marrow Transplant Program, Division of Hematology/Oncology, University of Kentucky Medical Center, Lexington, Kentucky 40536-0093, USA; and
b Department of Physiological Chemistry, University of Groningen, Groningen, The Netherlands
| ABSTRACT |
|---|
|
|
|---|
Key Words: stem cells longevity cell proliferation aging quantitative trait locus
| INTRODUCTION |
|---|
|
|
|---|
To demonstrate that cell proliferation is indeed associated with organismic life span, genes that play a role in both processes have to be identified. As a first step toward this goal, we searched for quantitative trait loci (QTLs)1 that contribute to hemopoietic progenitor cell cycling and mouse life span in BXD and BXH recombinant inbred mice. These genetically mosaic mice are derived from C57BL/6 (low cycling stem cells, long life span) and DBA/2 or C3H/He mice (high cycling stem cells, short life span). Typing these strains for a phenotype of interest may result in a map position harboring the gene that causes the phenotypic variation between the two strains. We demonstrate a strong negative correlation between mouse life span and stem cell cycling in BXD mice, and report the chromosomal map position of two loci, mapping to the same intervals on chromosomes 7 and 11, that were associated with both stem cell cycling and mouse life span. Independent confirmation of the involvement of the locus on chromosome 11 was obtained from the evaluation of BXH mice.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Bone marrow cells were flushed from femora obtained from three mice, pooled, and used in the cobblestone area-forming cell (CAFC) assay.
CAFC assay
The CAFC assay, as described by Ploemacher et al. (14)
, was
slightly modified, as published earlier (15)
. This in vitro
assay allows quantification of primitive hemopoietic cell subsets of
various developmental stages. Cells of the stromal cell line FBMD-1
(16)
were seeded in 96-wells plates (Costar, Cambridge, Mass.) in
Dulbecco' modified Eagle's medium containing L-glutamine (Life
Technologies, Grand Island, N.Y.), supplemented with 5% horse serum
and 15% fetal bovine serum (both from Life Technologies),
10-4 mol/l ß-mercaptoethanol,
10-5 mol/l hydrocortisone (Sigma, St Louis,
Mo.), 80 U/ml penicillin and 80 µg/ml streptomycin (both from Life
Technologies), and 25 mmol/l NaHCO3. Plates were
incubated at 33°C in 5% CO2 and used 10-14
days after seeding. Bone marrow cells were harvested from the mice and
a nucleated cell count was performed. To measure the percentage of
cells in the S-phase of the cell cycle, we used a hydroxyurea suicide
technique, as described previously (5)
. To this end, two identical
samples were diluted to a concentration of 1 x
107 cells/ml. Hydroxyurea (Sigma) was added to
one sample at a concentration of 200 µg/ml and both samples were
incubated at 33°C for 1 h After incubation, both cell
suspensions were washed and a nucleated cell count was performed again.
Based on this cell count, the CAFC assay was initiated. Bone marrow
cells were seeded on the stromal layer in six dilutions, each cell
concentration threefold apart (81,000 cells/well being the highest cell
concentration, 333 cells/well the lowest). At this time the medium was
switched from 5% horse serum and 15% fetal bovine serum to 20% horse
serum. For each cell dilution, 40 replicate wells were used. After 1
wk, all wells were evaluated for the presence or absence of cobblestone
areas, defined as colonies of at least five small nonrefractile cells
that grow underneath the stromal cells. Colonies scored after 7 days
(i.e., CAFC day 7) have been shown to be derived from pluripotent
progenitor cells 3, 17)
. The fraction of cells killed by hydroxyurea
was calculated by dividing the estimated CAFC frequency in the
hydroxyurea-treated cell suspension by the control value.
Linkage analysis using BXD RI strains
The percentage of CAFC day 7 in S-phase was calculated as
described above, and the strain distribution pattern was used to search
genome-wide for QTLs that affect the cycling of this cell stage. This
was done by entering the values for all 26 BXD strains into the
software program MapManager QT (b15) (18)
, which computes regression
statistics between variation in phenotype and variation in alleles at
loci that have already been mapped. At present, some 1700 loci have
been mapped in the BXD strains. Similarly, mean life span data for the
BXD strains obtained from a study published by Gelman et al. (19)
were
used to search for QTLs associated with mean life span.
Confirmation of linkage using BXH RI strains
We tested all 12 BXH RI strains for progenitor cell cycling in a
manner similar to that described above. After an initial genome-wide
screen, Iapls2-37 on chromosome 10 was found to be linked to
the trait. Next, a composite interval mapping strategy was used to
search for additional loci with modifying effects on cell cycling. This
was done as recently described (20)
, by performing a second genome-wide
search, but now controlling for Iapls2-37. Thus, an analysis
is conducted to search for loci to explain remaining variation not
accounted for by the primary QTL (i.e., Iapls2-37). To
assess the effects of such secondary loci on the primary QTL, the
interval mapping procedure on chromosome 10 was repeated by correcting
for these modifiers.
Linkage statistics
To establish criteria for suggestive and significant linkage
(21)
, a permutation test was performed using MapManager QT (1000
permutations at 1 cM intervals (22)
. This test compares the peak LOD
score obtained for a given data set with the peak LOD scores obtained
for 1000 random permutations of the same data set (20)
. To assess
genome-wide significance, point-wise probabilities (P) for
both RI sets for the same interval were pooled using Fisher's
equation:
2 = -2(lnpBXD
+ lnpBXH), with 4 degrees of freedom (20)
.
| RESULTS |
|---|
|
|
|---|
|
Hemopoietic progenitor cell cycling is inversely correlated with
mean mouse life span
Our data as presented in Fig. 1
were compared with data obtained
by Gelman et al. (19)
. In this study, life span measurements of 22 BXD
strains (1020 mice per strain) were performed under carefully
controlled housing conditions. For example, sentinel animals were
present at all times to monitor for microbial pathogens. Histological
necropsies revealed a wide range of death causes without any clear
patterns. Figure 2
shows that, in young BXD mice, progenitor cell cycling and mean strain
life span were significantly inversely correlated
(P=0.0093), as we previously found in a survey of eight
commonly used inbred strains (6)
. Approximately 30% of the variation
in mean life span could be explained, and indeed predicted, by
differences in cycling (r2 = 0.29).
|
Mapping of loci associated with progenitor cell cycling and mouse
life span
The strain distribution patterns for progenitor cell cycling and
life span as shown in Figs. 1 and 2
were used to search for QTLs most
strongly associated with these traits using MapManager QT (b15)
software (18)
. A genome-wide search for linkage with progenitor cell
cycling revealed four putative loci; one major QTL on chromosome 11,
reaching a genome-wide threshold required for suggestive linkage 21, 22)
, and three weaker ones on chromosomes 4, 7, and 9 (Table 1
). Similarly, the data of Gelman et al. (19)
were used to search for
QTLs affecting life span. Four regions were identified; two major QTLs
on chromosomes 2 and 7, also reaching the genome-wide levels required
for suggestive linkage, and two weaker ones on chromosomes 4 and 11
(Table 1)
. These data were gathered during 19821986 and published in
1988, when the mouse genetic map was not dense and mapped traits in the
BXD set in particular were sparse. The QTL on chromosome 11 was not
detected in the original study because insufficient markers on that
chromosome were available at that time (19)
. In addition, the authors
had to omit data from BXD strains 31 and 32 from their analysis since
these strains had not been genotyped well enough at that time.
|
It is striking that of the four loci that we found in each individual
search, two were shared by both traits: one on chromosome 11 (near
D11Ncvs76) and one on chromosome 7 (near
D7Ncvs38). Subsequent interval mapping of both progenitor
cell cycling and mouse life span resulted in a remarkably similar
linkage pattern, with coincident peaks at chromosome 11, 31.0 cM
(Fig. 3
A) and at chromosome 7, 6.0 cM (not shown).
|
Figure 3B
shows how the BXD strains segregate into the two
extreme genotypes for these two loci on chromosomes 7 and 11. C57BL/6
alleles at both D7Ncvs38 and D11Ncvs76 were
associated with a longer life span (mean of 765 days) and low
progenitor cell cycling (9.2%), whereas the corresponding DBA/2
alleles predisposed animals to a 25% shorter life span (mean of 592
days) and high progenitor cycling (32.4%). Linkage to
D2Rik63 and Ms6hm, the other two life span QTLs
(Table 1)
, predisposed to a long life span in strains that had the D
alleles (i.e., opposite the parental phenotype), and these markers were
not associated with cell cycling.
Analysis of variation in hemopoietic progenitor cell cycling in BXH
strains
C3H/He mice show an even more pronounced progenitor cell
proliferation than do DBA/2 mice 4-6)
, and therefore we also
evaluated variation in cell cycling in BXH recombinant inbred mice
(derived from C57BL/6 and C3H/He). Figure 4
A depicts the strain distribution pattern for the 12 BXH
strains. No life span data for BXH strains are available, so we could
only perform a linkage analysis for the cycling trait. As described
above, a genome-wide search was carried out using the data presented in
Fig. 4A
. Significant genome-wide linkage (LOD=3.7,
point-wise P=7.3 x 10-5,
genome-wide P=0.007) was found with a locus on chromosome
10, Iapls2-37, mapping ~60 cM from the centromere (Fig. 4B
). This is very close to the Steel locus encoding the
hemopoietic cytokine c-kit ligand (also known as stem cell factor or
mast cell growth factor). Since it was obvious from the BXH phenotypes
that more than one locus was involved in the variation in cell cycling
and because we had obtained sufficient statistical power that allowed
us to suspect the involvement of a locus on chromosome 10, we used the
composite interval mapping approach, as recently described by Williams
et al. (20)
. In such an analysis, a second search is initiated for loci
associated with the phenotypic variation that is unaccounted for by the
primary QTL. To do this we performed a second genome-wide search using
all BXH data, but now controlling for Iapls2-37. Using this
approach, linkage was observed with loci on chromosome 11 (Iapls1-74,
Scya3, and Xmv42, peak LOD score 2.8, point-wise P value
0.00036), bracketing the peak shown in Fig. 3A
obtained in
the BXD set. Thus, we obtained independent evidence of the involvement
of a locus on chromosome 11 in progenitor cell cycling (the combined
LOD score was 7.34, point-wise P value = 8.37 x
10-7, genome-wide P<0.001). When we
controlled for the loci on chromosome 11, the LOD score for the locus
on chromosome 10 in the BXH set greatly improved (to LOD = 6.45,
point-wise P=4.93 x 10-8,
genome-wide P<0.001, Fig. 4B
), again confirming
the veracity of the chromosome 11 QTL. We propose to name the QTL on
chromosome 7 `stem cell proliferation 1' (Scp1), the locus
on chromosome 11 Scp2, and the locus on chromosome 10,
mapped in the BXH set, Scp3.
|
| DISCUSSION |
|---|
|
|
|---|
Our earlier studies have shown that if highly cycling hemopoietic stem cells of short-lived DBA/2 genotype are put in competition with slowly cycling stem cells of C57BL/6 genotype in the common environment of embryo aggregation chimeras, mature blood cells of the DBA/2 genotype disappear from the circulation at a time roughly equivalent to the maximal life span of that strain (7) . These and other results have led us to argue that true self-renewal of individual hemopoietic stem cells may not exist and that the number of stem cells is finite (23) . The aging of stem cells (i.e., decline of stem cell quality) would be dependent on the rate of cycling of these cells, which we have shown to be highly strain dependent and inversely correlated with organismal life span. Stem cells of long-lived C57BL/6, whether measured by long-term repopulating ability 24, 25) , cell surface phenotype determined by flow cytometry (26) , or using the CAFC assay 5, 27) , have not been shown to be detrimentally affected by aging, whereas stem cell activity of shorter lived CBA/J (25) and DBA/2 5, 27) strains clearly declines with age.
To map loci that contributed to this strain-dependent variation in
proliferative activity of stem cells, we used BXD and BXH recombinant
inbred mice. The analytical power of using these recombinant inbred
strains resides in the fact that the genotype is fixed, so that one can
go back to study a different trait in the same genetic background. This
is clearly emphasized by our current study, where we compared a strain
distribution pattern for cell cycling with life span data obtained 15
years ago (19)
. This advantage has been underscored by Plomin et al.
(28)
, who have established a BXD recombinant inbred-QTL Cooperative
Data Registry that enables comparison of longevity data to other traits
reported in the literature concerning the BXD series (28)
. To date,
they have not been able to correlate any of some 137 BXD-phenotypes
with longevity (29)
. In this report we present the first such
significant correlation. Our analysis resulted in the identification of
four putative intervals associated with cell cycling. It is worth
noting that when these were compared with those obtained by searching
for life span loci, two of four independently mapped QTLs associated
with these traits in BXD mice mapped to the same intervals. The
individual QTLs reached the statistical criteria required for
suggestive linkage, and combining BXD and BXH data by using composite
interval mapping strategies resulted in the identification of a
significantly linked QTL on chromosome 11 20, 21)
. The strain
distribution pattern for stem cell cycling that we present in Fig. 1
is
almost an exact mirror image of the pattern reported by Muller-Sieburg
(30)
, where the frequency of long-term culture-initiating cells in BXD
strains was measured. At present we have no clear explanation for this
finding.
Together with our finding that, in young animals from eight standard inbred strains of mice, progenitor cell cycling and life span are inversely correlated (6) , this in vivo study suggests strongly that genes affecting the intrinsic rate of cell cycling also affect the rate of aging of mammalian organisms. The loci we have mapped contain genes that regulate cell cycling in at least one, but potentially more, critical renewing tissues. The nature of these genes remains to be elucidated and is under study. The genomic segment encompassing D11Ncvs76 contains the mouse cytokine cluster (i.e., genes for interleukins 3, 4, 5, 13, and granulocyte-macrophage colony-stimulating factor map here) and is syntenic to human 5q31. Perhaps not coincidentally, we have recently mapped a QTL significantly associated with hemopoietic stem cell (CAFC day 35) frequency to a region of mouse chromosome 18 that is also syntenic to human 5q (6) . Deletions of this chromosomal region in humans are associated with the 5q- myelodysplastic syndrome 31-37) .
Aging is clearly a multifactorial process 19, 38-42) , but our study suggests that, like recent advances made in this field in Caenorhabditis elegans and Saccharomyces cerevisiae 43, 44) , it is open to genetic analysis in mammals. Our findings support a model in which the rate of aging of an organism is partly the result of proliferation kinetics of certain critical tissues/organs. Inherent differences in cell cycle kinetics may determine the speed at which Hayflick's mitotic clock ticks 11, 45) and telomeres shorten 12, 13, 46) , but may also alter the rate at which potentially fatal DNA damage accumulates 39, 47, 48) . Genetic and/or environmental variations affecting either of these parameters may alter potential life span.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
1 Abbreviations: CAFC, cobblestone area-forming cell;
QTLs, quantitative trait loci; Scp1, stem cell
proliferation 1 locus on chromosome 7; Scp2, locus on
chromosome 11; Scp3, locus on chromosome 10 mapped in
the BXH set. ![]()
Received for publication September 29, 1998.
Revision received November 23, 1998.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. van Os, L. M. Kamminga, A. Ausema, L. V. Bystrykh, D. P. Draijer, K. van Pelt, B. Dontje, and G. de Haan A Limited Role for p21Cip1/Waf1 in Maintaining Normal Hematopoietic Stem Cell Functioning Stem Cells, April 1, 2007; 25(4): 836 - 843. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Henckaerts, J. C. Langer, J. Orenstein, and H.-W. Snoeck The Positive Regulatory Effect of TGF-{beta}2 on Primitive Murine Hemopoietic Stem and Progenitor Cells Is Dependent on Age, Genetic Background, and Serum Factors J. Immunol., August 15, 2004; 173(4): 2486 - 2493. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Henckaerts, J. C. Langer, and H.-W. Snoeck Quantitative genetic variation in the hematopoietic stem cell and progenitor cell compartment and in lifespan are closely linked at multiple loci in BXD recombinant inbred mice Blood, July 15, 2004; 104(2): 374 - 379. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. de Haan, L. V. Bystrykh, E. Weersing, B. Dontje, H. Geiger, N. Ivanova, I. R. Lemischka, E. Vellenga, and G. Van Zant A genetic and genomic analysis identifies a cluster of genes associated with hematopoietic cell turnover Blood, August 28, 2002; 100(6): 2056 - 2062. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Geiger, J. M. True, G. de Haan, and G. Van Zant Age- and stage-specific regulation patterns in the hematopoietic stem cell hierarchy Blood, November 15, 2001; 98(10): 2966 - 2972. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. de Haan, S. J. Szilvassy, T. E. Meyerrose, B. Dontje, B. Grimes, and G. Van Zant Distinct functional properties of highly purified hematopoietic stem cells from mouse strains differing in stem cell numbers Blood, August 15, 2000; 96(4): 1374 - 1379. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. de Haan and G. Van Zant Dynamic Changes in Mouse Hematopoietic Stem Cell Numbers During Aging Blood, May 15, 1999; 93(10): 3294 - 3301. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |