Leukemia Research Vol. 6, No 1. pp. 81 88. 1982. 0145-2126/82/010081-08503.00/0 Prmled in Great Britain © 1982 Pergamon Press Lid

P H Y S I C A L A N D K I N E T I C P R O P E R T I E S O F H A E M O P O I E T I C P R O G E N I T O R C E L L P O P U L A T I O N S

F R O M M O U S E M A R R O W D E T E C T E D IN F I V E

D I F F E R E N T A S S A Y S Y S T E M S

PAUL BAINES,* SIMON J. L. BOLt and MARTIN ROSENDAAL*

*Department of Anatomy and Embryology, University College London. Gower Street, London WC1E 6BT, U.K. and ?Radiobiological Institute TNO, 151 Lange Kleiweg, Rijswijk (ZH),

The Netherlands

(Receiz,ed 25 June 1981. Revised 17 September 1981. Accepted 28 September 1981)

Abstract--Properties of haemopoietic progenitor cells detected in several different assays have been compared in order to position them within the haemopoietic developmental lineage. The spleen colony-forming cell (CFUs), the high proliferation potential colony-forming cell (HPP-CFC) and two granulocyte-macrophage colony-forming cells (GM-CFC-1 and GM-CFC-2) have been studied.

Two experimental techniques were used: separation of cells on the basis of their buoyant density and comparison of the survival of haemopoietic cells after donor mice had been injected with the cytotoxic drug 5-fluorouracil (5-FU).

On linear BSA gradients the modal buoyant densities of CFU~, HPP-CFC and GM-CFC-1 were the same, 1.070 gcm- 3; the density of GM-CFC-2 was higher, 1.075 g cm- 3. GM-CFC-2 colonies were much smaller and contained far fewer cells than HPP-CFC or GM-CFC-1 colo- nies, even after prolonged culture, and this suggests that dense haemopoietic progenitors have less proliferation potential. This was confirmed by comparison of the size of colony formed, under identical culture conditons, by progenitors of different densities. Mean colony diameter was inversely related to the density of the progenitor cell.

With the exception of GM-CFC-1, low density progenitors were more resistant to the cytotoxic effects of 5-FU than high density precursor cells (GM-CFC-2). Consequently, the GM-CFC-1 could be distinguished from GM-CFC-2 on the basis of buoyant density and from the other low density populations on the basis of post-FU kinetics. The reasons why the GM-CFC-1 should be more sensitive to 5-FU than other low density progenitors are discus~ed and the relation of these low density precursors to one another in terms of their position withl.n the haemopoietic develop- mental lineage is elucidated.

Key words: GM-CFC, CFU~, buoyant density, 5-fluorouracil.

INTRODUCTION

SINGLE progenitor cells (granulocyte-macrophage colony-forming cells---GM-CFC-2) will form colonies of granulocytes and macrophages in semi-solid cultures of mouse marrow supplemented with pregnant mouse uterus extract (PMUE) [6]. After culture for 14 days, these colonies can contain up to 5 × 10 3 cells. Much larger colonies of granulo- cytes and macrophages, containing up to 5 × 104cells, will develop if the culture medium is supplemented with human or rat spleen conditioned medium in addition to PMUE [5, 1]. The cell giving rise to these large colonies has been called the high proliferative potential colony-forming cell (HPP-CFC).

Results of previous experiments have shown that the HPP-CFC can form the precur- sors (GM-CFC-2) of the smaller colonies [1]. This is consistent with a model of granulo-

Abbreviations: GM-CFC, granulocyte-macrophage colony-forming cell; HPP-CFC, high proliferative poten- tial colony-forming cell: CFUs, colony-forming unit in the spleen; 5-FU, 5-fluorouracil; HSCM, human spleen conditioned medium: PES, post-endotoxin serum: PMUE; pregnant mouse uterus extract.

81

82 PAUL BAINES et al.

cyte-macrophage differentiation/maturation suggested by Bol and Williams [4] who proposed: (a) that GM-CFC-2 are the progeny of GM-CFC-1 progenitors, which are cells capable of forming larger colonies in cultures supplemented with PMUE and post- endotoxin serum (PES) and (b) that the proliferative potential of GM-CFCs (i.e. the number of cells they can produce) is inversely related to their buoyant density. In other words, cells with least proliferative potential, and which form the smallest colonies, are characterised by high buoyant density.

These proposals, however, were based on data drawn from cultures of only short duration and it is possible that, given time. high density precursor cells may form colo- nies as large as those formed by low density precursors. If this were so high buoyant density would correlate with a low proliferative rate rather than a low proliferative potential.

In the work described here, we have assayed the cellularity of colonies during serial transplantations through fresh culture conditions and demonstrate that in cultures of short duration, colony size reflects the proliferative potential of the colony4orming cell. not simply its proliferative rate. We have determined the buoyant densities of HPP-CFC grown in cultures supplemented with human spleen conditioned medium (HPP-CFC-hcm) or rat spleen conditioned medium (HPP-CFC-rcm). and PMUE. We have compared the buoyant densities of HPP-CFC. CFU~, GM-CFC-1 and GM-CFC-2 and have related this property to the size of colony formed by each progenitor popula- tion. Further, in cultures supplemented with both PMUE and human spleen conditioned medium (HSCM), we have compared colony size with the buoyant density of the colony- forming cell. Finally, we have compared the sensitivity of all these progenitor popula- tions to the cytotoxic drug 5-fluorouracil which, it is proposed [10], spares only non- cycling, high proliferative potential progenitor cells. We have correlated these data and have been able to relate these progenitor populations to each other and to their position within the haemopoietic developmental lineage.

MATERIALS AND METHODS

Animals

C57B1/10 Sc Sn × BALB/c F~ mice were reared in the animal unit, Department of Anatomy and Embryo- logy, University College London. C57BI/Rij × C3H/LawF1 mice were reared in the animal unit at TNO. Rijswijk. Mice were routinely used at 10-12 weeks of age.

Preparation of cells Femoral cells were flushed into balanced salt (BSSt solution using a syringe and 25-gauge needle.

In vitro colony-forming cell assays HPP-CFC-hcm [5] were cultured in alpha medium (Flow) containing 25~o foetal calf serum (Flow Lab.l.

This medium was supplemented with 20 mg/l gentamycin. The stimulants for HPP-CFC-hcm cells were 6-10°~, v/v pregnant mouse uterus extract (PMUE) and 6-100~, W'V human spleen conditioned medium. HPP-CFC-rcm cells [ l ] were cultured in alpha medium containing 25% human serum with 6°~, PMUE and 6Q~, rat spleen conditioned medium as the stimulants. GM-CFC-2 were cultured in Dulbecco's modified Eagles medium (Gibco Biocult) containing 20°J o of a mixture of 2 vol of horse serum ~non-commercial batch) and I vol of foetal calf serum (Flow Lab.). This medium was supplemented with 2 mg L-asparagine, 56mg L-glutamine. 370 mg NaHCOa, 7.5 mg DEAE dextran, l04 I.U. Penicillin and 10mg streptomycin per 100ml (single strength). The stimulants for GM-CFC-2 and GM-CFC-I, were, respectively, 10~o PMUE (v/v of whole culture) and 10~o v/v PMUE + 10~o v/v of a 1:8 dilution of 18 h post-endotoxin serum (PES) i-2].

In all cultures, double strength medium was prepared and diluted to single strength, shortly before adding cells, by addition of equal volumes of 0.6~/O liquid agar (Bacto Agar Difco). In some cultures (HPP-CFC-hcm and HPP-CFC-rcm) a double layer technique was used. Double strength medium was diluted with an equal volume of 1% agar for the underlayers.

HPP-CFC were routinely cultured in duplicate in 60 mm diameter dishes (Sterilin and Corning) and GM-CFC-I and GM-CFC-2 were cultured in triplicate in 30ram diameter dishes (Sterilin and Corning). HPP-CFC cultures and GM-CFC cultures were routinely harvested after 12 and 8 days, respectively, of incubation in a fully humidified atmosphere containing 107/o CO2, 50/0 02 and 85°0 N 2.

Physical and kinetic properties of haemopoietic progenitors 83

PMUE was prepared as described by Bradley et al. [6] and rat spleen conditioned medium as described by M etcalf and Johnson [12]. Human spleen conditioned medium (HSCM) was prepared by incubating cells from normal human spleens or spleens of patients with carcinoma at 2 × 10~/ml in alpha medium + 5'~<, FCS. Supernates from these cultures were harvested after seven days and used undiluted to supplement cultures. Post-endotoxin serum (PESt was prepared as reported by Bol et al. [2 3.

Scorinq colonie+s The abbreviations HPP-CFC-hcm and HPP-CFC-rcm refer, not necessarily to different populations, but to

different culture conditions. In cultures supplemented with P M U E and human spleen conditioned medium, colonies were scored as HPP-CFC-hcm colonies if they achieved a diameter of greater than 1.5 ram. In cultures supplemented with P M U E and rat spleen conditioned medium, colonies were scored as HPP-CFC-rcm if they exceeded 2 mm diameter. Colonies in either set of culture conditions were routinely counted at days 10-12 of culture and usually exhibited typical HPP-CFC- colony morphology, i.e. a core with a more diffuse periphery of cells.

In cultures supplemented with human spleen conditioned medium and PMUE, colonies are compact + a prerequisite for accurate measurement of colony diameter.

GM-CFC-1 numbers were calculated by subtracting the numbers of colonies developing in the presence of P M U E alone, from the number of colonies developing in the presence of P M U E + PES.

All colonies were counted on an Olympus binocular microscope using a calibrated eyepiece graticule to measure colon? diameter.

Estimation o/colony eelhdarity Colonies were removed from agar by suction, using a Pasteur pipette and disrupted by vigorous pipetting

into BSS. Cells were counted on a Neubauer counting chamber.

Serial transplants oJ colonie~ fi'om cultures o[ normal marrow supplemented with P M U E and human ,spleen conditioned medium

Individual colonies of defined colony size were removed from cultures of normal marrow after 9-12 days, by suction. Whole colonies were then placed intact into fresh nutrient agar. containing P M U E and HSCM, in such numbers as to keep the total number of cells transplanted per dish the same for each size category of colony transplanted.

Spleen colony-forming cells assay The spleen colony-forming cell (CFUs) assay was performed as reported by Till and McCulloch [14]. Mice

were given 800 rads whole body irradiation using a Marconi X-irradiation source (230kV, 15 mA, filtered through 1 mm Cu + 1 mm AI, 100-200rads/min).

Equilibrium density centriluqation ~[ cells on BSA gradients This was performed in the manner described by Bol et al. [3].

Drug treatments

5-Fluorouracil (Sigma) was dissolved (7.5 mg/ml) in saline and injected i.v. at 150 mg/kg body wt.

RESULTS

1. Colony size and cellularity The cellularities of colonies of size categories >1.5, 0.5-1.0 and < 0 . 5 m m were

measured after various times of culture. Irrespective of duration of culture, cellularity was directly proportional to colony diameter. Colonies continued to increase in diameter until day 14 of culture. Cellularities of colonies after 14 days of culture were: (a) for colonies > l . 5 m m diameter 7.2-4-0.6 × 10+; (b) for colonies 0.5-1ram diameter 1,5 + 0.3 × 10+; (c)for colonies <0.5 mm diameter 1.1 × 10 3. The cellularity of colonies transplanted, intact, into fresh culture conditions after 9-12 days of culture continued to increase well beyond 14 days of culture which suggests that a deterioration of culture conditions was responsible for cessation of growth in the original cultures (Table 1). Colony growth in serially transplanted cultures ceased after 2-3 transplants, presumably because the full proliferative potential of the original precursor cell under these conditions had been realised. Importantly, the results of serial transplan- tation show that small colonies never attained the cellularity of colonies formed by HPP-CFC.

84 P.a, UL BAINES e t ~11.

TABLE 1. CELLULARITY OF COLONIES REMOVED INTACT FROM ORIGINAL

CULTURES AND SERIALLY TRANSPLANTED A T WEEKLY INTERVALS

THROUGH SEMI-SOLID CULTURES SUPPLEMENTED WITH P M U E AND

HUMAN SPLEEN CONDITIONED MKDIUM

Retransplant Colony size Cells,colony ( x l0 "~l No. category (mini* Expt I t Expt 2.{.

0~ > 1.5 6 _+ 0.5 4.6 4- 0.7 0.5 1 1.1 _+ 0.3 0.8 4- 1.5

<0.5 N.D. 0 .04

> 1.5 28.6 + 2.6 14.0 4- 1.6 0.5-i 8.6 4- 0.9 2.2 4- 0.4

<0.5 N.D. 0.13

> 1.5 32.8 + 3.9 N.D. 0.5-1 4.9 4- 1.2 3.2 + 0.7

<0.5 N . D 0.13 _+ 0.06

> 1.5 N.D. 24.8 4- 2.7 0.5 1 1.9 + 0.2 1.0 4- 0.5

<0.5 N.D. 0.16 + 0.05

*On removal from original culture. iColonies harvested from original cultures at day 9. ~Colonies harvested from original cultures at day 12. §Cellularity at time of harvest of original culture ti.e. days 9 and 12

for Expts 1 and 2 respectively). IIColonies pooled. N.D. = not done.

IO0-

~o

g

~ I0O

g

i >, g

~ 50-

(o)

i / / ."

' . . i f :

\ \

(b)

11060 I 'o7"0 I 08o 1090

DensJty, gem "3

FIG. I. (a) Density distr ibut ions C F U , (Ok G M - C F C - ! (at) and G M - C F C - 2 (A) from normal marrow separated on l inear BSA gradients" (b) density distr ibut ions o f HPP-CFC-hcm (O) and HPP-CFC-rcm (O) f rom normal marrow separated on BSA gradients. (One of three similar

e x p e r i m e n t s . )

Phvsical and kinetic properties of haemopoietic progenitors 85

2. Physical characterization of haemopoietic progenitor cells from normal marrow on linear BSA gradients

Fractionation of normal marrow on BSA gradients gave results consistent with earlier findings [4, 2]. Cells forming colonies in the presence of PMUE alone showed a broad density distribution peaking at 1.075-1.078 g cm -3 (Fig. la). In work reported earlier [2] it was noted that combinations of some batches of PMUE and serum could stimulate not only the GM-CFC-2 population (modal density 1.075 g cm -3) but also the GM-CFC-3 population (modal density 1.078 gcm 3) which normally only grows in cultures supple- mented with both PMUE and rat erythrocyte lysate. It seems likely that the broad profile seen here includes the profiles of both these progenitor populations.

GM-CFC-1, HPP-CFC-hcm, HPP-CFC-rcm and CFUs populations all showed nar- row density distributions peaking at 1.070 g cm-3 (Figs. la, b). These results were consist- ent over three experiments and confirm, in part, the results of others [-4, 9].

In cultures supplemented with PMUE and human spleen conditioned medium, colo- nies were sorted into size categories for scoring. The data (Fig. 2) show that the smallest colonies (< 0.5 mm diameter) were formed by progenitors of highest density (peak modal density 1.078gcm-3). It seems probable that these high density precursors are GM-CFC-3. Precursors of colonies of intermediate size (0.5 1.5 mm) showed a broad density distribution and again, probably GM-CFC-2 and GM-CFC-3 progenitors both contribute to colonies in this range. The largest colonies (> 1.5 mm) were formed by low density precursors (peak modal density 1.070 g cm-3). Since there is a positive corre- lation between colony diameter and cellularity (see Section 1) this means that progeni- tors of greater density produce fewest progeny and therefore possess least proliferative potential.

Supplementing cultures with both PMUE and human spleen conditioned medium therefore enables colony formation by a variety of haemopoietic progenitor populations ranging from the high density, low proliferative potential GM-CFC-3 to the low density, high proliferative potential HPP-CFC.

3. Decline and recovery of haemopoietic progenitor cell populations in the marrow of mice pre-treated with 5-.fluorouracil (5-FU)

The drug 5-FU is proposed [10] to be most toxic to low proliferative potential, cycling cells. The data indeed show (Fig. 3) that the GM-CFC-2, a cycling population, was very sensitive to 5-FU whereas the HPP-CFC-rcm, a predominantly non-cycling population [1] was much less sensitive. Furthermore, in cultures supplemented with both PMUE

,oo- • ~

° i-" .>:: °° .°

K

• . . o ~ . . •" -° .

~*" y

o

~" , ;+o ~ b ,o , ;8o ,dg0 o

Dens~l"y, gcm -3

F](;. 2. Density distr ibut ions of the precursors of colonies of > 1.5 mm (@), 0.5-1 mm (Zh) and <0.5 mm ('q) diameter in normal marrow cultures supplemented with P M U E and human spleen

conditioned medium.

86 PAUL BAINES et al.

%

c~

o

qO00

100

ro

i

I 2 5 5 8

Time a f t e r in jec t ion of 5 - FU, doys

FIG. 3. Toxicity of 5-FU to H P P - C F C - h c m (O), HPP-CFC-rcm (O), G M - C F C - I (A) and GM -C F C -2 (A) progenitor populations. (One of two similar experiments.)

and human spleen conditioned medium, the precursors of the category of largest colonies (HPP-CFC-hcm) proved to be considerably less sensitive to 5-FU than the precursors of smaller colonies developing in the same cultures (Table 2). This also means that the HPP-CFC-rcm population may therefore appear more sensitive to 5-FU than the less numerous HPP-CFC-hcm population, because it contains a mixture of HPP-CFC-hcm and precursors of smaller colonies.

The above data suggest that rapidly-cycling, low proliferative potential progenitors are more sensitive to 5-FU than non-cycling, high proliferative potential progenitors. Furthermore, we have demonstrated elsewhere [8, 13] that 5°FU does not in fact deplete the largely non-cycling CFU~ population nearly as markedly as was reported earlier [10]. Nevertheless, GM-CFC-1 progenitors were sensitive to 5-FU despite evidence that few of these cells are cycling [7; Visser, personal communication].

TABLE 2. DECLINE AND RECOVERY OF H P P - C F C - h c m ( > l . 5 m m COLONIES) AND THE PRECURSORS OF SMALLER COLONIES tN CULTURES OF

5 - F U PRli-TREATED MARROW SUPPLEMENTED WITH P M U E AND HUMAN

SPLEEN CONDITIONED MEDIUM*

Days post-5-FU

% Surviving colony-forming cells generating colonies of each size category

> 1.5"1" 0.2-1.5 < 0.2

Nbm 100 100 100 I 21 2.2 3.5 2 15 0.8 0.7 3 21 2.0 0.9 5 37 3.0 3.6 8 153 40 4.7

*One of two similar experiments. 4-Colony diameter in mm.

Physical and kinetic properties of haemopoietic progenitors 87

DISCUSSION

The experiments described concern the relationships between various haemopoietic progenitor cells which can be detected in different assay systems. Two developmental stages of granulocyte-macrophage progenitor cells (GM-CFC-I and GM-CFC-2), high proliferation potential colony-forming cells (HPP-CFC) cultured under two sets of con- ditions and spleen colony forming cells (CFU,) were compared with respect to prolifera- tive potential, buoyant density and sensitivity to in vivo injections of 5-FU.

Firstly. serial transplantation of 9-12-day-old colonies through fresh culture conditions induced more proliferation in colonies already recorded as large colonies, than in smaller colonies. So, although cultures were routinely harvested after a limited incubation period, colony size after 9-12 days of culture reflects proliferation potential and not simply proliferative rate. Compared with HPP-CFC colonies, smaller colonies contained a lower proportion of granulocytes and a larger proportion of macrophages--at any time of culture (unpublished datal. This supports the proposal [-4] that decreasing proliferative potential is paralleled by a shift from granulocyte to macrophage differentiation such that the switch to macrophage production occurs earlier in smaller colonies.

Buoyant density fractionations show that CFUs, GM-CFC-1 and HPP-CFC popula- tions consist of low density cells capable of rapid and extensive proliferation. The GM-CFC-2 population, however, is composed of cells of higher density with lower proliferative potential and which, therefore, form smaller colonies. Similarly, in cultures supplemented with both PMUE and HSCM and harvested after 10-12 days of incuba- tion, colony diameter was inversely related to the buoyant density of the colony-forma- tion cell. These data are consistent with the suggestion [-4] that high proliferative potential progenitor cells are characterized by tow buoyant density.

Few of these low density cells appear to be cycling as judged by their sensitivity to the phase specific drug hydroxyurea and tritiated thymidine. Nine per cent of HPP-CFC-rcm progenitors were killed by a single exposure to hydroxyurea [1] compared to 10~o for CFU, [11 : unpublished results]. On the basis of cycling state, it seems probable that the low density GM-CFC-1 and the high density GM-CFC-2 represent similar populations to those which respond to mouse lung conditioned medium (CSF-mlcm) and human urine CSF (CSF-hu) respectively [7]. Using hydroxyurea these authors obtained kills of 17°,, for the low density GM-CFC population which formed colonies in the presence of CSF-mlcm and 44°o for the high density GM-CFC population which proliferated in the presence of CSF-hu. This is similar to our own observations on cell kill by tritiated thymidine for GM-CFC-1 and GM-CFC-2 (unpublished results). GM-CFC-2 therefore are actively cycling whereas GM-CFC-1 more closely resemble the CFU~ and HPP-CFC in cycling state.

The low density progenitor cells defined by the assays used here--CFU~, GM-CFC-1, HPP-CFC-hcm and HPP-CFC-rcm--therefore all represent developmentally "early" populations. Although there are no obvious differences among these cell types with respect to buoyant density and cycling state, they do differ in their sensitivities to in vivo injections of 5-FU. GM-CFC-1 are more sensitive to 5-FU than HPP-CFC (Fig. 3) or CFU, [-13, 8]. This may indicate differences in metabolic activity.

In conclusion we note that the HPP-CFC is capable of generating GM-CFC-2 among its progeny [1], The GM-CFC-1 too may be derived from the HPP-CFC and may represent a population intermediate between the HPP-CFC and GM-CFC-2 since it resembles the former in density but resembles the latter in sensitivity to 5-FU. We are currently designing experiments to test this possibility.

A~'knowledyements- T h i s work was supported by grants from the Medical Research Council and the Cancer Research Campaign. Dr Bol is a fellow of the Netherlands Foundat ion for Medical Research ( F U N G O I which

88 PAUL BAINES et al.

is subsidized by The Organization for the Advancement of Pure Research. Dr Baines would like to thank the Radiobiological Institute TNO for their hospitality. We are grateful to Ms. V. van Slingerland. Ms. K. Rodwdl. Ms. S. Villa and Ms. D. Frendo for their assistance.

REFERENCES

1. BAINES P., BOL S. J. L. & ROSENDAAL M. (1981) Characterization of a developmentally early granulocyte: macrophage progenitor found in normal mouse marrow. Br. J. Haemat. 48, 147.

2. BOL S., V1SSER J. & VAN DEN ENGH, G. J. (1979) The physical separation of three subpopulations of granulocyte/macrophage progenitor cells from mouse bone marrow. Expl. Haemat. 7, 541.

3. BOL S., VISSER J., WILLIAMS N. 8,~ VAN DEN ENGH G. J. (1977) Physical characterization of haemopoietic progenitor cells by equilibrium density centrifugation. In Cell Separation Methods (BEOEMENDAL H., Ed.). p. 39. Elsevier North Holland, Amsterdam.

4. BOL S. J. L. ~,~ WILLIAMS N. (1980) The maturation state of three types of granulocyte macrophage progenitor cells from mouse bone marrow. J. Cell. Physiol. 102, 233-243.

5. BRADLEY T. R. & HODGSON G. S. (1979) Detection of primitive macrophage progenitor cells in mouse bone marrow. Blood 54, 1446.

6. BRADLEY T. R., TELFER P. A. & FRY P. (1971) The effect of erythrocytes on mouse bone marrow colony development in vitro. Blood 38, 353.

7. BYRNE P. V., HEIT W. & KUBANEK B. (1979) The proliferation states of density subpopulations of granulo- cyte-macrophage progenitor cells. Expl. Haemat. 7, 105.

8. DIXON R. x, ROSENDAAL M. Contrast between the response of the mouse haemopoietic system to 5-fluor- ouracil and irradiation. Blood Cells (in press).

9. HASKILL J. S., McNEIL T. A. & MOORE M. A. S. (1970) Density distribution analysis of in civo and in citro colony-forming cells in bone marrow. J. Cell. Physiol. 75, 167.

10. HODGSON G. S. & BRADLEY T. R. (1979) Properties of haemopoietic stem cells surviving 5-fluorouracil treatment: evidence for a pre-CFU~ celt? Nature, Lond. 281, 381.

11. HODGSON G. S., BRADLEY T. R., MARTIN R. F., SUMNER M. & FRY P. (1975) Recovery of proliferating haemopoietic progenitor cells after killing by hydroxyurea. Cell TiNs. Kinet. 8, 51.

12. METCALF D. 8¢, JOHNSON G. R. (1978) Production by spleen and lymph node cells of conditioned medium with erythroid and other haemopoietic colony stimulating activity. J. Cell. Physiol. 96, 31.

13. ROSENDAAL M., DIXON R. & PANAY1 M. Haemopoietic stem cells: possibility of toxic effects of 5-fluorourao cil on spleen colony formation. Blood Cells (in press).

14. TILL J. E. • McCuLLOCH E. A. (1961) A direct measurement of the radiation sensitivity of normal mouse bone marrow cells. Radiat. Res. 15, 213.