Вы находитесь на странице: 1из 8

From bloodjournal.hematologylibrary.org by guest onononononononon October 8, 2013. For personal use only.

October 8, 2013. For personal use only.

October 8, 2013. For personal use only.

October 8, 2013. For personal use only.

October 8, 2013. For personal use only.

October 8, 2013. For personal use only.

October 8, 2013. For personal use only.

October 8, 2013. For personal use only.

From <a href=bloodjournal.hematologylibrary.org by guest onononononononon October 8, 2013. For personal use only. From bloodjournal.hematologylibrary.org by guest From bloodjournal.hematologylibrary.org by guest From bloodjournal.hematologylibrary.org by guest From bloodjournal.hematologylibrary.org by guest From bloodjournal.hematologylibrary.org by guest From bloodjournal.hematologylibrary.org by guest From bloodjournal.hematologylibrary.org by guest October 8, 2013. For personal use only. October 8, 2013. For personal use only. October 8, 2013. For personal use only. October 8, 2013. For personal use only. October 8, 2013. For personal use only. October 8, 2013. For personal use only. October 8, 2013. For personal use only. 1993 82: 2031-2037 The role of hypoxia in the maintenance of hematopoietic stem cells MG Cipolleschi, P Dello Sbarba and M Olivotto Inf o rm at i o n about r ep r oduc in g t hi s a r t i c l e in pa r ts o r in i ts e n t ir ety m ay be f ou n d o nlin e a t: http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requests Information about ordering reprints may be found online at: http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprints Information about subscriptions and ASH membership may be found online at: http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtml Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036. Copyright 2011 by The American Society of Hematology; all rights reserved. " id="pdf-obj-0-56" src="pdf-obj-0-56.jpg">

1993 82: 2031-2037

The role of hypoxia in the maintenance of hematopoietic stem cells

MG Cipolleschi, P Dello Sbarba and M Olivotto

Information about reproducing this article in parts or in its entirety may be found online at:

Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.

Copyright 2011 by The American Society of Hematology; all rights reserved.

From <a href=bloodjournal.hematologylibrary.org by guest onononononononon October 8, 2013. For personal use only. From bloodjournal.hematologylibrary.org by guest From bloodjournal.hematologylibrary.org by guest From bloodjournal.hematologylibrary.org by guest From bloodjournal.hematologylibrary.org by guest From bloodjournal.hematologylibrary.org by guest From bloodjournal.hematologylibrary.org by guest From bloodjournal.hematologylibrary.org by guest October 8, 2013. For personal use only. October 8, 2013. For personal use only. October 8, 2013. For personal use only. October 8, 2013. For personal use only. October 8, 2013. For personal use only. October 8, 2013. For personal use only. October 8, 2013. For personal use only. 1993 82: 2031-2037 The role of hypoxia in the maintenance of hematopoietic stem cells MG Cipolleschi, P Dello Sbarba and M Olivotto Inf o rm at i o n about r ep r oduc in g t hi s a r t i c l e in pa r ts o r in i ts e n t ir ety m ay be f ou n d o nlin e a t: http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requests Information about ordering reprints may be found online at: http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprints Information about subscriptions and ASH membership may be found online at: http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtml Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036. Copyright 2011 by The American Society of Hematology; all rights reserved. " id="pdf-obj-0-123" src="pdf-obj-0-123.jpg">

The Role of Hypoxia in the Maintenance of Hematopoietic Stem Cells

By Maria Grazia Cipolleschi, Persio Dello Sbarba, and Massimo Olivotto

Bone marrow cell liquid

cultures were incubated at various

oxygen concentrations ranging

from 0% to 18% (air). The

total number of cells in culture (CT) at the end of a 6-day

incubationwas found to be directly proportional to the oxy- gen concentration. As compared with air-incubated con- trols, cells recovered from severely hypoxic (1 % oxygen) day-5 liquid cultures showed (1) the same day-7 colony- formation efficiency in semisolid culture (neutrophilic/ monocytic colonies) or in spleen; (2) a higher day-1 4 spleen colony-formation efficiency; (3) an enhanced radio-protec- tion ability; and (4) an increased marrow repopulation abil- ity, as measured by determining either total cell number in recipient marrow MRk,,, or the capacity of the latter of generating day-7 neutrophilic/monocytic colonies in sec-

ondary in vitro assays (MRA,,u.,M). Taking into account CT, the absolute numbers of progenitors in culture were also computed. The results showed that, with respect to

time 0, incubation in air

produced an increase in the num-

ber of day-7 CFUs and a decrease in the number of the

other progenitors, whereas in hypoxic cultures all types of progenitorsdecreased. However, as compared with air-in- cubated controls, all progenitors, except cells sustaining MR&FU.NM, were reduced in hypoxic cultures. The degree

of reduction paralleled the position of

the progenitor in the

hematopoietic hierarchy, being maximum for day-7 CFUs and null for cells sustaining MRACFU.NM,which, in fact, were better preserved in hypoxic cultures.

  • 0 1993 by The American Society of Hematology.

LTHOUGH developmental and cytokinetic properties A of hematopoietic stem cells have been intensely stud- ied, the in vivo organization of the hematopoietic tissue remains largely unknown. In particular, the mechanisms regulating the life-long maintenance of stem cells are still unexplained. According to a widely accepted view, the con- ditions for this maintenance are realized in physiologically segregated areas of bone marrow (BM) (“niches”), wherein stem cells are restrained from commitment to extensive pro- liferation and differentiation.

In a previous report, we proposed that these niches are

hypoxic areas, in which only oxygen-independent cells

are

able to survive.* This proposal was based on the following rationale. Arterial blood enters the highly branched network

of medullary sinuses only after circulating in the cortical

canalicular

sy~tem,~ so that a relatively desaturated blood

reaches the marrow. Indeed, average oxygen tension in BM blood is significantly lower than in other organs and tissues, being similar to that in the jugular vein! Moreover, it is

likely that cells crowding around sinuses strongly compete for the already scarce oxygen supplied, resulting in an even lower oxygen tension within the core of cell mass.5 It is highly plausible that in such areas the proliferation of stem cells is blocked without threatening their survival. In fact, while mitochondrial respiration seems a prerequisite for cells to enter the mitotic ~ycle,~-~ growth factors have been shown to sustain cell survival through the stimulation of glycoly~is.~-~’ Therefore, the deeply hypoxic areas of BM appear to be particularly suitable for the long-term mainte- nance of stem cells, whereas the better oxygenated areas would allow proliferation of more differentiated progeni- tors. In keeping with our hypothesis, hematopoietic progeni- tors actually appear to be aligned along the oxygen gradient

of the BM. Progenitors responsible for BM repopulation have in fact been shown to be more concentrated in the low oxygen areas of BM,” whereas fast cycling neutrophilic/ monocytic colony-formingunits (CFU-NM) and day-7 col- ony-forming units in spleen (CFU-S,) reside preferentially

in subendosteal areas, where

the blood enters the BM circu-

lation. Furthermore, the probability of progenitors to be recruited into the mitotic cycle seems inversely related to the distance from those areas.l3-I5

In our laboratory, the respiration-dependence of BM cells (BMC) was indirectly studied in vitro by treating short-term

liquid cultures with an excess of pyruvate or other oxidiz- able substrates. This excess, saturating the respiratory chain, produces the impairment of oxidative limiting steps of various metabolic pathways, thus mimicking the effects of inhibitors of mitochondrial respiration.16 The addition of pyruvate reduces colony formation from CFU-NM, with- out affecting their in vitro generation from stem cells, sug-

gesting differences between CFU-NM

and their progenitors

in the dependence on mitochondrial respiration.2 In this report, we present a study of the effects of reduced oxygen tension on mass liquid cultures of mouse BMC. We provide evidence that oxygen tension is a critical factor for BMC expansion in vitro and that the different types of he- matopoietic progenitors display a varying degree of sensitiv- ity to hypoxia. The lower this sensitivity, the higher the hier- archical level of the progenitor, so that incubation in severe hypoxia leads to a substantial concentration of some types of progenitors within the population and selectively pre- serves progenitors of the highest hierarchical level.

MATERIALS AND METHODS

Cell recovery and preparation.

BMC were obtained from

pooled 8- to 12-week-old CBA T6T6 mice by forced injection of RPMI-I640 medium (GIBCO Ltd, Paisley, UK) into femoral bone

From the Istituto di Patologia Generale, Universitd di Firenze,

Florence, Italy.

Submitted October 6, 1992; accepted June 2, 1993.

Supported by grants from Associazione Italiana per la Ricerca SUI

Cancro (AIRC), Consiglio Nazionale delle Ricerche (Project “Appli-

cazioni Cliniche della Ricerca Oncologica’), and Minister0 della

Universitd e della Ricerca Scientifica e Tecnologica (MURST,fondi

60% e 40%).

Address reprint requests to Massimo Olivotto, MD, Istituto di

Patologia Generale viale G.B. Morgagni 50, I-50134 Firenze, Italy.

The publication costs of this article were defrayed in part by page

charge payment. This article must therefore be hereby marked

“advertisement” in accordance with 18 U.S.C.section I734 solely to

indicate this fact.

  • 0 1993 by The American Society of Hematology.

0006-49 71/93/8207-00I0$3.00/0

2032

CIPOLLESCHI, DELL0 SBARBA, AND OLIVOTTO

shafts. Cells were then centrifuged (250g for 10 minutes), resus- pended in 0.87% NH4CI in H20 to lyse erythrocytes, washed, and plated in cell culture dishes in RPMI-1640 supplemented with 10% heat-inactivated horse serum (HS; Flow Laboratories, Irvine, UK). After 3 hours of incubation, nonadherent cells were recovered, counted, and transferred into liquid or semisolid cultures (see be- low). Cell counts were performed in a hemocytometer and cell via- bility estimated by the trypan blue exclusion test, diluting cell sus- pensions I : I with a 0.4% wt/vol trypan solution in 0.85% saline (Flow Laboratories). Spleen conditioned medium (SCM). Mouse SCM was prepared and tested for BMC growth-stimulatory activity as described else- where.’ In brief, 10’ mouse splenocytes were cultured in 50 mL of

RPMI- 1640 supplemented

with 5% pooled, heat-inactivated hu-

man serum and 0.4% pokeweed mitogen solution (GIBCO); after 7 days of incubation, culture medium was recovered, filtered, and frozen in small aliquots to be used without further manipulation.

Cell culture system.

BMC culture in liquid medium was per-

formed in gas leak-proof culture vessels that allowed us to establish, and maintain throughout the incubation, atmospheres composed of 5% C02and different proportions of oxygen (ranging from 0% to

10%)and nitrogen (95% to 85%).”*18BMC were plated

in 250-mL

glass flasks, in RPMI-1640 supplemented with 20% HS and 10% SCM (5 X IO6 cells in 50 mL per flask). After flushing the sterile-fil- tered gas mixture, the air-tight inlet and outlet were closed and the flasks transferred into an incubator at 37°C. Control cultures were established in the same type of flask, but kept in direct communica- tion with the atmosphere of the incubator: 95% air, 5% C02,water- saturated (incubation “in air”). The pH of the hypoxic cultures remained within physiologic limits (7.2 to 7.4) throughout the incu- bation. After 5 to 6 days of incubation in hypoxia or in air, cells were washed, counted, and diluted appropriately for the in vitro or in vivo assays, as indicated below. BMC recovered directly from ani- mals (“time-zero” of the cultures) were assayed under the same conditions in a larger set of separate experiments. In vitro clonal assay. To estimate the number of CFU-NM in BMC populations, semisolid cultures were established by plating IO4 viable cells per 3-cm petri dish in 1 mL of RPMI-I640 supple- mented with 20% HS, 10% SCM, and 0.3% (wt/vol) agar (Bacto Agar; DIFCO, Detroit, MI). Assays were performed in triplicate and always incubated in air. After 7 days ofincubation, the number

of colonies (aggregates of 50 or more cells)per dish was scored at 25

  • X magnification. Colony-formationefficiency (CFE) was expressed

as the ratio of the number of colonies developed to the number of

cells plated per dish.

In Vivo Assays

Cells were transplanted into pooled 8- to 12-week-old, syngeneic

mice that

had been lethally irradiated with a total dose of 10 Cy at a

rate of

1

Gy/min from a 6oCo, 1.2 MeV source

(Theratron 780;

Atomic Energy of Canada Ltd, Ottawa, Canada). Cell suspensions in serum-free culture medium were injected into the lateral tail vein

within 3 hours after irradiation. Irradiated mice injected with cul- ture medium only were used to determine the background level for each parameter studied.

Spleen colony-formation assays.

Cells were transplanted into 3

to 5 mice per group and spleens recovered from recipient mice 7

(CFU-S,) or 14 days (CFU-SI.,) after transplantation. Spleens were then fixed in Bouin’s solution, and macroscopic surface colonies counted at 2 X magnification. To determine the number of CFU- SI4,chosen as an assay for the “delayed” spleen colony-forming activity,lS2’various cell suspensionswere transplanted (2.5 X lo4to

5 x IO5 cells/0.2 mL). Counts of large day-14 spleen colonies, a possible source of inacc~racy,~~ were confirmed by determining spleen weight (not shown); CFE was expressed as the ratio of the colony number to the number of cells transplanted.

Radioprotection and marrow repopulation assays.

The radio-

protective ability (RPA) of BMC populations, ie, the capacity to

confer long-term survival to lethally irradiated mice, was measured

by transplanting various cell suspensions (2 X IO4 to 2 X

IO6 cells/

  • 0.2 mL), usually into IO mice per suspension. The fraction ofrecipi-

ent mice

surviving 30 days after transplantation was determined for

each suspension. RPA was expressed as the ratio of the percentage of surviving mice to the number of transplanted cells.z”6 BM repopulating ability (MU) of BMC populations was mea- sured basically as described by Hodgson et al;,,’’ by transplanting various cell suspensions (5 x io4to 2 x IO6 cellsj0.2 mL) into 3 to 5 mice per suspension. Femurs were recovered from recipient mice 14 days after transplantation and individually processed. The total number of nucleated cells per femur was determined and aliquots of cells were assayed for the presence of CFU-NM, as described above. MRA was expressed as the ratio of the total number of cells (MRAeII),or the number of CFU-NM (MRA,-m-NM), per recipient femur, to the number oftransplanted cells. MRA,.,, was chosen as an assay for MRA,-, .20,28s29 Data obtained with increasing dose of transplanted cells were plotted in log/iog scale and best fitted by linear regression, with the exclusion of plateau values, according to Hodgson et a1.28 Cells responsible for RPA or MRAel,,m-NM were referred to as RPA cells or MRAcell,CFU.NMce[ls.20.25.26.29

RESULTS

Dependence of BMC Growth on Oxygen Tension

Total number of cells (CT) was determined in BMC liq- uid cultures incubated for 6 days at different oxygen ten- sions. CT resulted directly proportional to oxygen concen- tration of the incubation atmosphere, increasing, as compared with the time-0 value, within an oxygen concen- tration range of 3% to 18% (Fig 1). There was no increase in cell number at 3% oxygen, whereas cell loss occurred at lower concentrations.

0

5

10

15

20

percentage of oxygen

Fig 1.

Effects of oxygen tension on BMC growth. CT in BMC

liquid cultures was determined at the end of 6 days of incubation.

Points are means standard errors of 11 separate experiments. Dashed line indicates CT value at time 0.

HYPOXIA PRESERVES HEMATOPOIETIC STEM CELLS

2033

The data indicated that, whereas BMC expansion in vitro is oxygen-dependent, a sizeable number of hematopoietic

cells is able to survive in severe hypoxia (1% to 3%). We

decided to characterize the composition

ofthis residual pop-

ulation with respect to several types of hematopoietic pro-

genitors and stem cells.

Incubation in 1% oxygen for 5 days,

providing sufficient numbers of viable cells for secondary in vitro or in vivo assays, was chosen on the basis of prelimi- nary tests (not shown) and used for all experiments reported

below.

Effects of Hypoxia on BMC Cloning Eficiency

Table 1 shows the effect of a 5-day incubation of BMC in 1% oxygen (hypoxia) or in air (aerobiosis) on cloning effi- ciency, as detected 7 days after replating in vitro (CFU-NM) or transplantation in vivo (CFU-S7).The in vitro clonal as- says were performed in both cases in air, to avoid differences in yield caused by the reported influence of oxygen tension on colony formation in semisolid Day-7 clon- ing efficiency, both in vitro and in spleen, was found to be significantly decreased, as compared with time-zero, inde- pendently of the incubation atmosphere. Furthermore, no significant effect on cloning efficiency was produced by in- cubation in 1% oxygen as compared with air. This indicated that hypoxia, although reducing CT (Fig l), did not alter the concentrations of CFU-NM and CFU-S,, and, therefore, that these progenitors are as sensitive to hypoxia as the over- all BMC population. On the contrary, BMC cloning efficiency in spleen, as detected 14 days after transplantation (CFU-S14),was signifi- cantly influenced by incubation in hypoxia. This is shown in Fig 2, in which the relationship between inoculum size and colony number is reported for all tested conditions. This relationship turned out to be linear in all cases, allow- ing for the calculation of the cloning efficiencies from the slopes of the lines. Cloning efficiency was found to be de- creased in day-5 cultures, as compared with time 0, indepen- dently of the incubation atmosphere. However, this reduc- tion was 5 times lower in hypoxia than in aerobiosis. Thus,

Table 1. Effects of BMC Incubation in Air or Hypoxia on Day-7 Cloning Efficiency in Culture (CFU-NM) or in Spleen (CFU-S)

Incubation

Colonies Per Dish

Colonies Per

Spleen

No (time 0)

34.9

?

1.63 (39;10)

15.2 k 0 75 (56; 11)

In air

21.2

?

5.39" (26; 7)

4.5

? 0.80t (27; 6)

In 1% oxygen

22 3 k

1.94" (19; 5)

6.8

?

0.89t

(1 7;

3)

BMC, recovered directly from donor mice (time 0) or from liquid CUI- tures incubated for 5 days in either air (control) or hypoxia (1% oxygen), were replated in semisolid medium (1O4 cells per dish) or transplanted into lethally irradiated syngeneic mice (1O5 cells per mouse), and colo-

nies

counted 7 days

later. Data are means

? intraexperiment standard

errors (total number of counts; number of separate experiments). Re- sults were assessed by one-way analysis of variance. Differences be- tween time 0 and day-5 cultures are defined by the symbols. Differences between hypoxic and control day-5 cultures were not significant.

P < .01 (71 degrees of freedom). t P < .01 (88 degrees of freedom).

HYPOXIA PRESERVES HEMATOPOIETIC STEM CELLS 2033 The data indicated that, whereas BMC expansion in vitro is

8

r

6

2

cn4

I

3

LL

02

0

1

2

3

cells

transplanted

4

5

(X~O-~)

Fig 2. Dose-response relationship between the number of BMC transplanted and the number of day-14 spleen colonies (CFU-S,J.

Cells to be transplanted were recovered directly from donor mice

(time

0; A) or from 5-day cultures incubated in hypoxia (1%oxygen;

0) or in air Points are

(0).The main comparison (air w hypoxia) was balanced. means ? standard errors of counts from three separate

and independent experiments. Lines were Wed by linear regres- sion on all experimental counts (time zero, 28; hypoxia, 32; air, 24). The slopes of the lines, corresponding approximately to colony-for-

mation efficiencies, were 13.28 ? 4.18; 4.90 -C 0.66; 0.97 -C

0.1 6, respectively. Differences according to the Student's t-test

between slopes were significant (time 0 wair P < .001,47 degrees

of freedom; time 0 w hypoxia P < .05,51 degrees of freedom; air w

hypoxia P < .001,46 degrees of freedom).

CFU-SI4appeared concentrated in hypoxic as compared with aerobic day-5 cultures, showing a substantial differ- ence between these progenitors and CFU-NM or CFU-s,.

Effects of Hypoxia on RPA

The percent survival of lethally irradiated mice is plotted in Fig 3 versus the inoculum size. To compare variously

treated cell populations, we considered intercepts of the lines best fitting the data with the number of transplanted cells required to obtain 50% survival. This number was esti- mated to correspond approximately to 3.5 X lo4 cells at

time zero, I 2 X 1O4 cells incubated in hypoxia, and 69 X 1O4

cells incubated in air. Thus, RPA was found decreased in

day-5 cultures, as compared with time 0, but this decrease

was 5.7 times lower in hypoxia than in aerobiosis, parallel-

ing the results obtained for CFU-Sl4.

Effects of Hypoxia on MRA

MRA,,, was calculated by plotting the number of cells recovered from recipient femoral marrow versus the num- ber of transplanted cells (Fig 4A). To compare variously treated cell populations, we considered intercepts of the lines fitting the data with a reference value of repopulation of recipient femoral marrow (2.5 X lo5cells, corresponding

to the dashed line in Fig 4A). It was estimated that, to obtain this value, one needs 8.7 X lo4 cells at time 0, 18.2 X lo4

2034

CIPOLLESCHL DELL0 SBARBA, AND OLIVOTTO

2.0. "1 A - a .- > 6.0. f h L a (I) i E 2
2.0.
"1 A
-
a
.-
>
6.0.
f
h
L
a
(I)
i
E
2
1.5.
5.5.
e
\
--
a
e
$
5.0.
2
-
UI
A
0
4.5
log,,
cells
transplanted
4.01
4
6
6
7
Fig 3. Dose-response relationship between the number of BMC
transplanted and the percentage of recipient mice surviving 30
log,,
cells
transplanted
days after transplantation (RPA). Cells to be transplanted were re-
covered directly from donor mice (time 0; A)or from 5-day cultures
4.0
r
incubated in hypoxia (1% oxygen;
0) or in air (0).The
main compari-
son (air v hypoxia) was balanced. Points represent percent survival,
calculated on groups of 10 mice per point. Lines were fitted by
linear regression, with the exclusion of plateau values. Dashed line
indicates 50% survival.
cells incubated in hypoxia, and 138 X lo4cells incubated in
air. Thus, again, MRA,, decreased in culture, but this de-
crease was 7.6 times lower in hypoxia than in aerobiosis, in
keeping with the results obtained for CFU-SI, and RPA.
The above procedure was applied to estimate
MRA,-,-,, (Fig 4B), assuming IO3 CFU-NM as a reference
value of repopulation of recipient femoral marrow (dashed
line in Fig 4B). This value was obtained with 14.8 X
IO4 cells
1.5 1
4
5
6
7
at time zero, 9.5 X lo4cells incubated in hypoxia, and 120 X
IO4 cells incubated in air. Strikingly, in hypoxic cultures,
MRA,,,, not only was much higher than in aerobic cul-
tures (12.6 times), but it also appeared unreduced, if not
enhanced, in comparison to time 0.
log,,
cells
transplanted
Efects of Hypoxia on the Absolute Number of Progenitors
in Culture
The effects of hypoxia on the absolute numbers of hema-
topoietic progenitors were tentatively quantified by taking
into account the changes in CT under the various culture
conditions (Table 2). It was shown that incubation in air for
5 days resulted in an increase in the number of CFU-NM
and CFU-S,, along with CT expansion. On the other hand,
all the other progenitors decreased: CFU-SI,, RPA cells,
and MRA,,, cells (progenitors sustaining RPA and MW,,,
respectively) to 26%. on the average, whereas MRACN-NM
cells (progenitors responsible for MRACN-NM) to about
50%. Incubation in hypoxia for 5 days reduced all types of
progenitors to 2 I %, on the average, the only exception being
MRA,,, cells, that were about 75% conserved.
Comparison of columns 2 and 3 (or 4 and 5) of Table 2
led to the conclusion that the enrichment of hypoxic cul-
tures with hematopoietic progenitors shown in Figs 2, 3,
Fig 4. Dose-response relationship between the number of BMC
transplanted and the number of nucleated cells (A) or CFU-NM (B)
recovered from each femur of recipient animals. Cells to be trans-
planted were recovered directly from donor mice (time 0; A)or from
5-day cultures incubated in hypoxia (1%oxygen; 0) or in air (0).The
main comparison (air v hypoxia)was balanced. Points are means -C
standard error of two to five separate and independent experi-
ments. Lines were fitted by linear regression, with the exclusion of
plateau values. Dashed lines indicate repopulation of recipient fe-
mur with 2.5 X 1O5 nucleated cells (A) or 1O3 CFU-NM (B). One-
way analysis of variance was performed on data normalizedfor a 5
x
1O5 cell inoculum as follows:
5
x
106
loglo( Cells Transplanted
where N is the number of cells (A) or CFU-NM (B) per femur. This
normalization gave (mean f SEM, total number of experimental
points, number of separate experiments) (A) time 0,3.19 f
0.07,
44, 8; hypoxia,
2.76 f 0.09,28,
7; air, 1.62
f
0.1 1, 29,
6; (B)
time0,3.44 -C 0.08.35.6; hypoxia,3.68 f 0.15,21,5;air,2.44
f 0.16, 12, 4. Differences between the above means were (A)
time 0 versus air,
P < .001; time 0 versus hypoxia, P < .01; air
versus hypoxia, P < .001; (B) time 0 versus air, P C ,001; time 0
versus hypoxia, not significant; air versus hypoxia, P < .001.

HYPOXIA PRESERVES HEMATOPOIETIC STEM CELLS

2035

Table 2. Computation of Total Number of Hematopoietic Progenitors in BMC Cultures

Type of Progenitor

t=O

Absolute Number of Cells in Culture Air

Hypoxia

Air

% oft = 0 Values

Hypoxia

~

 

~

 

_____

___

Culture no.

1

2

3

4

5

Cells (CT)

5 x lo6

21.2 x 106

2.4 X lo6

424.0

48.0

CFU-NM’

17,450

44,944

5,352

257.6

30.7

CFU-S,’

760

954

163

125.5

21.5

CFU-SI,’

664

206

118

31 .O

17.8

RPA cellst§

143

  • 31 20

21.7

14.0

MRA,,,

cells$§

57

  • 15 13

26.3

22.8

M ,,R,A,,

cells*§

34

  • 18 25

52.9

73.5

Abbreviations: t = 0 (time 0).BMC recovered directly from donor mice; air, 5-day cultures incubated in air; hypoxia, 5-day cultures incubated in 1 % oxygen.

Values were obtained multiplying CT by the colony-formation efficiencies (see Table 1 and the legend to Fig 2). t Values are expressed in arbitrary units, 1 U corresponding to the number of transplanted cells supporting survival of 50%recipient mice after 30 days.

HYPOXIA PRESERVES HEMATOPOIETIC STEM CELLS 2035 Table 2. Computation of Total Number of Hematopoietic Progenitors in

$ Values are expressed in arbitrary units, 1 U correspondingto the number of cells to be transplanted to repopulate each femur of the recipient with

  • 2.5 x lo5 BMC (MRA,,,,) or lo3 CFU-NM (MRACFU.NM).

§ Values were obtained dividing CT by the number of transplanted cells correspondingto 1 U, calculated from Fig 3 (RPA), Fig 4A (MRA,,,,), or Fig 48

(MRAcFu-NM).

and 4A was essentially accounted for by the suppression in these cultures of the CT increase that occurs in air. MRACW-NM cells (Fig 4B) had the further advantage of be- ing better preserved in hypoxic cultures, rather than in aero- bic cultures.

DISCUSSION

The main information to emerge from this study was that hypoxic culture conditions clearly selected for certain types of hematopoietic progenitors in BMC cultures, as com- pared with aerobic cultures. This effect was maximal for MRACmsNMcells (Fig 4B). A clearer outline of the fate of various progenitors in aerobiosisor in hypoxia was obtained from the computation of their total number in culture (Ta- ble 2), showing that (1) the concentration of most progeni- tors in hypoxic cultures was essentially accounted for by

suppression of CT increase; and (2) MRApm-NM cells were better preserved in hypoxia than in aerobiosis. When total numbers of progenitors in hypoxic cultures (Table 2) were reported as percentages of the corresponding number in aerobic cultures (Fig 5), it was possible to iden-

respiration-dependent step in the mitotic cycle shown for CFU-NM2 and suggests the existence of a similar step in CFU-S,, that, like CFU-NM, are actively cycling commit- ted progenit~rs.~~,~~,~~ On the other hand, the oxygen-depen- dence of CFU-NM contrasts with the reported enhance- ment of colony-formation efficiency in semisolid cultures incubated in hypoxia, an effect attributed to a reduced oxy- gen toxicity on clonogenic progenitors.3G34However, sensi- tivity of these cells to oxygen toxicity is apparently lower in mass liquid cultures than in clonal assays in semisolid me- di~m.~’,~’ Thus, it is conceivable that in our system the posi- tive effects of oxygen on cell growth largely prevail over a low degree of oxygen toxicity. The low sensitivity to hypoxia of CFU-SI4, RPA cells, and MRA,” cells (Fig 5) is in keeping with the weak Rh-pos- itivity of group (2) cells, indicating a scarcity of active mito- chondria. It is worthwhile to recall that these progenitors are mainly q~iescent,~, but readily recruitable and committable by growth factors (“activated stem cells”) to generate higher

tify three main categories of

progenitors: (1) oxygen-depen-

dent progenitors, including CFU-NM and CFU-S,; (2)oxy- gen-independent progenitors (CFU-Sl4, RPA cells, and MR&,, cells); and (3) hypoxia-preserved progenitors (MRAcW-NM cells). The above categories fit very well with progenitor pheno- types identified on the basis of mitochondrial activity, as measured by rhodamine- 123 (Rh) (1) Rh-posi- tive cells (CFU-S, and persistent-CFU-S,,); (2) Rh-weakly positive cells (delayed-CFU-S,,, RPA cells, and MRA,, cells); and (3) Rh-dull cells (most MRApFU-NMce11s).19-21~24~29

n

cells

cells

cells

The finding that CFU-NM and CFU-S, depend on oxy- gen for survival and expansion in vitro is in keeping with the abundance of active mitochondria shown in group (1) cells by the Rh-positivity. This result is also consistent with the

Fig 5.

Effects of hypoxia on the absolute number of hematopoi-

etic progenitors in BMC cultures. Bars represent values referring to

hypoxic cultures (reported in column 3 of Table 2). expressed

as

percentages of those of control cultures in air (column 2 of Table 2). The latter are indicated by the dashed line (see text).

2036

CIPOLLESCHI, DELL0 SBARBA, AND OLIVOTTO

numbers of less immature pr~genitors.~~.~’ Therefore, acti- vated stem cells are likely to be depleted in our system as in any growth factor-stimulated culture. Indeed, in our experi- ments, these cells were found substantially decreased in both control and hypoxic cultures, pointing out the irrele- vance of cell respiration for their survival. Finally, within the limits that MW,,, cells (group 3) are actually Rh-dull, ie, lacking of active mitochondria, these progenitors could be regarded as anaerobically adapted. In keeping with this view, MRA cells have been found to be more highly concentrated in the low oxygen areas of BM.” Anaerobic adaptation of metabolism, usually accompanied by an enhanced cell sensitivity to oxygen toxicity:’ seems to be confirmed by our finding that

MRA,,,, cells were better maintained in hypoxia

than in

air, a striking difference from all of the other types of pro-

genitors. Interestingly, MRA,,,, cells are also mostly noncy~ling,~~~~~~~~ an essential attribute of the most imma- ture hematopoietic progenitors, the quiescent stem ~ells.4~ The latter are believed to be less easily depleted than acti- vated stem cells under the action of growth factors, as con- firmed by our data (see Table 2, column 4). In this context, one can hypothesize that anaerobic adap- tation of metabolism is a major feature distinguishing activated from quiescent stem cells. Assuming that the re- cruitment of stem cells to active hematopoiesis is respira- tion-dependent, the lack of active mitochondria in some of these cells would prevent their recruitment and limit the action ofgrowth factors to the support of survival.43In other words, the anaerobic orientation of a subset of stem cells would reduce their sensitivity to proliferation/differentia- tion stimuli and play a crucial role in their maintenance in Goand long-term conservation. On the whole, data presented seem to support the pro- posed hypothesis that quiescent stem cells survive in “hyp- oxic niches” of hematopoietic tissue, protected from com- mitment. An increase of number/activity of mitochondria may well be a prerequisite for the activation of stem cells. These activated stem cells could conceivably move to less hypoxic areas close to the niches, where the response to growth factors is no longer restrained and the generation of lineage-restricted progenitors is possible. The latter, in bet- ter oxygenated areas, would then undergo the extensive pro- liferation responsible for clonal expansion. Whatever the merit ofthese considerations, it is a fact that various subsets of hematopoietic progenitors are differently affected by oxygen tension. While this effect needs to be further explored by BMC fractionation procedures, incuba- tion in severe hypoxia or anoxia might prove to be a useful and “physiologic” tool to select quiescent stem cells from normal and leukemic BM.

ACKNOWLEDGMENT

The authors thank A. Fonnesu, MD, chairman of the Institute of General Pathology, Florence, for help and advice. Thanks are also due to L. Calorini, MD, Institute of General Pathology, Florence,

for technical assistance; A. Becciolini

and the Staff ofthe Radiother-

apy Unit, USL 10D, Florence, for advice and assistance to the irra-

diation of mice; V. Boddi, Institute ofGeneral Pathology, Florence,

and P. Urbano, MD,

Institute of Microbiology, Florence, for statis-

tical analysis of data; P.A. Bemabei, MD, Department of Hematol- ogy, USL IOD, Florence, for his helpful discussions and critical reading of the manuscript; and I.S. Hawkins, MD, for her revision of the English.

REFERENCES

1. Schofield R: The relationship between the spleen colony- forming cell and the haemopoietic stem cell-A hypothesis. Blood Cells 4:7, 1978

  • 2. Dello Sbarba P, Cipolleschi MG, Olivotto M: Hemopoietic

progenitor cells are sensitive to the cytostatic effect of pyruvate. Exp Hematol 15: 137, 1987

  • 3. Lichtman MA: The ultrastructure ofthe hemopoietic environ-

ment of the marrow: A review. Exp Hematol9:391, 1981

  • 4. Grant WC, Root WS: The relation of O2 in bone marrow

blood to post-hemorrhagic hemopoiesis. Am J Physiol 150:6 18,

1947

  • 5. Tannock IF: The relation between cell proliferation and the

vascular system in a transplanted mouse mammary tumour. Br J

Cancer 22:258, 1968

  • 6. Hams JW, Meyskens F, Patt HM: Biochemical studies of cy-

tokinetic changes during tumor growth. Cancer Res 30:1937, 1970

  • 7. Olivotto M, Paoletti F The role of respiration in tumor cell

transition from the non-cycling to the cycling state. J Cell Physiol 107:243, I98 1

  • 8. Olivotto M, Arcangeli A, Caldini R, Chevanne M, Cipolleschi

MG, Dello Sbarba P: Metabolic aspects of cell cycle regulation in

normal and cancer cells. Toxicol Pathol 12:369, 1984

  • 9. Whetton AD, Dexter TM: Effect of haematopoietic cell

growth factor on intracellular ATP levels. Nature 303:629, 1983

10.

Whetton AD, Bad1 GW, Dexter TM: Haemopoietic cell

growth factor mediates cell survival via its action on glucose trans-

port. EMBO J 3:409, 1984

11.

Dexter TM, Whetton AD, Bazill GW Haemopoietic cell

growth factor and glucose transport: Its role in cell survival and the

relevance of this in normal haemopoiesis and leukaemia. Differen- tiation 27:163, 1984

12.

Lord BI: The architecture of bone marrow cell populations,

in Murphy MJ Jr (ed): Concise Reviews in Clinical and Experimen- tal Hematology. Dayton, OH, AlphaMed, 1992, p 225

13.

Lord BI, Testa NG, Hendry JH: The relative spatial distribu-

tions of CFU-S and CFU-C in the normal mouse femur. Blood

46:65, 1974

14.

Shackney SE, Ford SS, Wittig AB: Kmetic-microarchitec-

tural correlations in the bone marrow of the mouse. Cell Tissue

 

%net 8:505, 1975

15.

Mason TM, Lord BI: Hendry JH: The development of spa-

tial distributions ofCFU-S and in-vitro CFU-C in femora of mice of

different ages. Br J Haematol 73:455, 1989

16.

Olivotto M, Caldini R, Chevanne M, Cipolleschi MG: The

respiration-linked limiting step of tumor cell transition from the non-cycling to the cycling state: Its inhibition by oxidizable sub- strates and its relationships to purine metabolism. J Cell Physiol

 

116:149, 1983

17.

Allalunis-Turner MJ, Koch CJ, Chapman JD: Growth of

murine bone marrow under various oxygen conditions in media

buffered with HEPES. Int J Cell Cloning 5:315, 1987

18.

Taylor WG, Camalier RF, Sanford KK: Density-dependent

effects of oxygen on the growth of mammalian fibroblasts in cul- ture. J Cell Physiol 95:33, 1978

19.

Visser WM, de Vries P Isolation of spleen-colony forming

HYPOXIA PRESERVES HEMATOPOIETIC STEM CELLS

2037

cells (CFU-s) using wheat germ agglutinin and rhodamine 123label- ing. Blood Cells 14:369, 1988

  • 32. Ishikawa Y, Ito T: Kinetics of hemopoietic stem cells in a

hypoxic cultures. Eur J Haematol40:126, 1988

 

20.

Ploemacher RE, Brons NHC Cells with marrow and spleen

  • 33. Lu L, Broxmeyer HE Comparative influences of phytohe-

repopulating ability and forming spleen colonies on day

16, 12, and

moagglutinin-stimulated leukocyte conditioned medium, hemin,

are sequentially ordered on the basis ofincreasing rhodamine retention. J Cell Physiol 136531, 1988

8

123

prostaglandin E, and low oxygen tension on colony formation by erythroid progenitor cells in normal human bone marrow. Exp He-

 

2 1. Ploemacher RE, Brons

NHC In vivo proliferative and dif-

matol 13:989, 1985

ferential properties of murine bone marrow cells separated on the

  • 34. Broxmeyer HE, Cooper S, Lu L, Miller ME, Langfeld CD,

basis of rhodamine 123 retention. Exp Hematol 16:903, 1988

Ralph P: Enhanced stimulation of human bone marrow macro-

 

22.

van Zant G Studies of hematopoietic stem cells spared by

phage colony formation in vitro by recombinant human macro-

 

5-fluorouracil. J

Exp Med 159:679, 1984

phage colony-stimulatingfactor in agarose medium and at low oxy-

 

23.

Necas E, Znojil V: A comparison of stem cell assays using

gen tension. Blood 76323, 1990

early or late spleen colonies. Cell Tissue %net 22: 11 I, 1989

  • 35. Johnson LV, Walsh ML, Bockus BJ, Chen LB: Monitoring

 

24.

Mulder AH,

of relative mitochondrial membrane potential in living cells by fluo-

sis of bone marrow

Visser JWM: Separation and functional analy- cells separated by rhodamine-123 fluorescence.

rescence microscopy. J Cell Biol 88:526, 1981

Exp Hematol 15:99, 1987

  • 36. Bertoncello I, Hodgson GS, Bradley TR Multiparameter

 

25.

Ploemacher RE, Brons NHC: Isolation of hemopoietic stem

analysis of transplantable hemopoietic stem cells: I. The separation

cell subsets from murine bone marrow: I. Radioprotectiveability of purified cell suspensions differing in the proportion of day-7 and

and enrichment of stem cells homing to marrow and spleen on the basis of rhodamine- 123 fluorescence. Exp Hematol 13:999, 1985

day-12 CFU-S. Exp Hematol 16:21, 1988

  • 37. Hodgson GS, Bradley TR In vivo kinetic status of hemato-

 

26.

Ploemacher RE, Brons NHC Isolation of hemopoietic stem

poietic stem and progenitor cells as inferred from labeling with bro-

cell subsets from murine bone marrow: 11. Evidence for an early

modeoxyuridine. Exp Hematol 12583, 1984

precursor of day-12 CFU-S and cells associated with radioprotec-

  • 38. Lin H, Hsu S: Modulation of tissue mononuclear phagocyte

tive ability. Exp Hematol 16:27, 1988

clonal growth by oxygen and antioxidant enzymes. Exp Hematol

 

27.

Hodgson GS, Bradley TR: Properties of haematopoietic

14:840, 1986

stem cells surviving 5-fluorouracil treatment:

Evidence for a pre-

  • 39. Balin AK, Fisher AJ, Carter DM: Oxygen modulates growth

CFU-S cell? Nature 281:381, 1979

of human cells at physiologic partial pressures. J Exp Med 160: 152,

 

28.

Hodgson GS, Bradley TR, Radley JM: The organization of

1984

hemopoietic tissue as inferred from the effects of 5-fluorouracil.

  • 40. Spangrude GJ, Johnson GR Resting and activated subsets

Exp Hematol 10:26, 1982

of mouse multipotent hematopoietic stem cells. Proc Natl Acad Sci

 

29.

Ploemacher RE, Brons NHC: Separation of CFU-S from

USA 87:7433, 1990

primitive cells responsible for reconstitution of the bone marrow

  • 41. Visser JWM, van Bekkum DW: Purification of pluripotent

hemopoietic stem cell compartment following irradiation: Evi-

hemopoietic stem cells: Past and present. Exp Hematol 18:248,

dence for a pre-CFU-S cell. Exp Hematol 17:263, 1989

1990

 

30.

Bradley TR, Hodgson GS, Rosendaal M: The effect of oxy-

  • 42. Oberley LW, Oberley TD, Buettner GR: Cell differentiation

gen tension on haemopoietic and fibroblast cell proliferation in vitro. J Cell Physiol 97517, 1978 3 1. Rich IN, Kubanek B: The effect of reduced oxygen tension on colony formation of erythropoietic cells in vitro. Br J Haematol 52579, 1982

and cancer: The possible role of superoxide and superoxide dismu- tases. Med Hypotheses 6:249, 1980

  • 43. Potten CS, Schofield R, Lajtha LG: A comparison of cell

replacement in bone marrow, testis and three regions of surface

epithelium. Biochim Biophys Acta 561:28 1, 1979