Granulocyte-Macrophage Colony-Stimulating ... - Semantic Scholar

Granulocyte-Macrophage Colony-Stimulating Factor Is Not Responsible for the Correction of Hematopoietic Deficiencies in the Maturing op/op Mouse By S.K. Nilsson, G.J. Lieschke, C.C. Garcia-Wijnen, B. Williams, D. Tzelepis, G. Hodgson, D. Grail, A.R. Dunn, and I. Bertoncello Osteopetrotic (op/op) mice are characterized by an autosoin the absence mal recessive inactivating mutation resulting ofbiologicallyactive colony-stimulatingfactor-l (CSF-1). Consequently, young op/op mice have a severe deficiency of macrophages and osteoclasts resulting inexcessive bone formation, occlusion of the marrow cavity, and reduced marrow hematopoietic activity. Recently, we showed that the osteopetrosisandhematopoietic deficiencies evident in young op/op mice are not permanent but are progressively corrected with age. There are increases in osteoclast activity; bone resorption; femoral marrowspace; and marrowhematopoietic activity, cellularity, and macrophage content. In the present study we show that CSF-1”- granulocyte-macrophage colony-stimulatingfactor (GM-CSF)”--deficient mice also undergo the same pattern of hematopoietic correction

as the op/op mouse. Also, like theop/op mouse, the peritoneal cellularity and macrophage content of CSF-l/GM-CSFdeficient mice remains severely reduced. Our data show that the “knockout“ of GM-CSF does not change the op/op phenotype, and that GM-CSF is notessential for the correction of the hematopoietic deficiencies in the op/op mouse. Importantly, the data also show thatneither GM-CSF nor CSF-I is an absolute requirement forthe commitmentof primitive hematopoietic stem cells to themacrophage lineage or for the differentiationof at least some classes of macrophages. This finding suggests that an alternate regulatory factor can be involvedin macrophage and osteoclast commitment, differentiation, and function in vivo. 0 7995 b y The American Society of Hematology.

0

young op/op mice corrects the macrophage deficiencies but failsto resolve osteopetrosis.’”2oGM-CSF stimulates the survival, proliferation, differentiation, and function of myeloid cells and their precursors, particularly monocytes and macrophages, as well as granulocytes.” However, analysis of op/op mice indicatesthat, whereas induciblelevels of endogenous GM-CSF are elevated in some animals, this is not always the case.3 The most definitivemeans of assessing the roleof hematopoietic regulators in vivo is to study mice that are deficient in one or morehematopoietic growth factors. Recently, GMCSF-deficient mice were produced by gene targeting in embryonal stem cells.” GM-CSF-deficient mice were shown to be relatively unperturbed in terms of circulating hematopoietic cells and progenitor cell levels in hematopoietic organs. In the present study, we have analyzed mice deficient in both CSF-1 and GM-CSF (CSF-I”-GM-CSF” ) that were produced by interbreeding op/op mice with GM-CSF-deficient mice.” A thorough comparison of the hematopoietic system of young (5 weeks of age) and mature (40 weeks of age)CSF-1 -’-GM-CSF”- mice and theirwild-typelittermates was undertaken. The data showed that, for all the hematopoietic parameters measured, mice deficient in both of the op/op mouse. CSF- 1 and GM-CSF have the phenotype This finding extends toboth the deficiencies evident in young mice, as well as to the progressive, age-related corrections evident in maturing op/op mice. The data show that GMCSF is not essential for the correction of the hematopoietic deficiencies in op/op mice and suggestthat an alternate regulatory factor can be involved in macrophage and osteoclast commitment, differentiation, and function in vivo. This may be an as yet unidentified factor or may be a manifestation of plasticity amongst the known hematopoietic growth factors.

STEOPETROTIC (op/op) mice are characterized by an autosomalrecessive inactivatingmutation in the colony-stimulating factor- 1 (CSF- 1) gene, resulting in the absence of biologically active CSF- 1 which has a lineagespecific role in augmentingmacrophageproductionand As a consequence, young op/op mice function in a have impaired mononuclear phagocyte development and deficiency of both macrophages and osteoclasts, resulting in an increase in bone density and occlusion of marrow cavities.’-” Previous studies have indicated that the daily administrationof CSF-1to newborn op/op mice canprevent osteopetrosis and the accompanying hematopoieticdefects.”.” Our previous studie~’~”‘ have shown a naturally occurring, progressive,age-relatedsystemiccorrection in hematopoieticparameters in op/op mice.This finding suggests that the hematopoietic system has the capacity to use alternative mechanisms to compensate for the absence of CSF-I l 4 and raises the possibility of plasticity amongst previously regarded key growth factors.” It is interesting that exogenous administration of the hematopoietic growth factor granulocyte-macrophagecolony-stimulatingfactor(GM-CSF)to

From the Sir Donald and Lady Trescowthick Research Laboratories,Peter MacCallum Cancer Institute, Melbourne, Australia; and the Melbourne Tumour Biology Branch, Ludwig Institute for Cancer Research, The Royal Melbourne Hospital, Victoria, Australis. Submitted November 14, 1994; accepted January 30, 1995. The Anti-Cancer Council of Victoria supportedG.J.L. (John Maynard Hedstrom Research Scholarship). Address reprint requests to S.K. Nilsson, PhD, Sir Donald and Lady Trescowthick Research Laboratories, Peter MacCallum Cancer Institute, Locked Bag No I , A’Beckett Street, Melbourne, 3000, Australia. The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. section 1734 solely to indicate this fact. 0 1995 by The American Society of Hematology. 0006-4971/95/8601-0104$3.00/0 66

MATERIALS AND METHODS Mice. Mice deficient in CSF-1 (op/op) and/or GM-CSF, as well aswild-typegenotypes,weregeneratedand raised as previously described.’4.22.23 Homozygous wild-type, heterozygous, and homozygous null genotypes at each locus are reported as +/+, +I-, and -/-. respectively. For these experiments, 5-week-old CSF-I ”GMBlood, Vol 86, No 1 (July I ) , 1995: pp 66-72

ROLE OF GM-CSF

IN

67

THE OPIOP MOUSE

CSF-” mice and their wild-type littermates were raised at the Ludwig Institute for Cancer Research, and the 5-week-old op/op mice and their wild-type littermates were raised at the Peter MacCallum Cancer Institute. All experimental work was performed at the Peter MacCallum Cancer Institute. Cell suspensions. Marrow cells from both CSF-l+’+GMCSF-’- and CSF-l+%M-CSF+” mice were collected by flushing femoral shafts with cold HEPES-buffered balanced salt solution (BSS) supplemented with 2% bovine calf serum iron-supplemented (BCS). Femurs from osteopetrotic CSF-l-/-GM-CSF+/+ and CSFl-/-GM-CSF-/- mice were routinely ground using a mortar and pestle, because less than 20% of marrow cells from osteopetrotic mice could be recovered by flushing a l ~ n e . ’ ~ . ’ ~suspensions The were gently vortexed to free any cells adhered to bone fragments, and the cell suspension was decanted and pooled, together with three washings of the bone fragments. Spleen cell suspensions were prepared by mincing the tissue in BSS supplemented with 2% BCS and by dispersion through a mesh sieve. Peripheral blood was collected from a throat bleed into an EDTAcoated tube and mixedwell. Excessive red blood cells contaminating the samples were lysed with 0.83% ammonium chloride (10 mL per 1 mL of red blood cell pellet) at 37°C for 5 minutes, and cells were then washed twice with BSS-2% BCS. Peritoneal cells were collected by injecting 10 mL ofBSS-2% BCS into the peritoneal cavity using a 23-gauge needle. The peritoneum was gently massaged, without removing the needle tip, to free loosely adherent cells before aspirating the solution. The cells were centrifuged at 1,000 rpm at 4°C for 5 minutes and resuspended in BSS-2% BCS. Cell counts were performed using a hemocytometer. Growthfactors. Partially purified pregnant mouse uterus extract was used as a source of CSF-1.24Recombinant human interleukin1 (IL-l: 2.5 x lo9 Ulmg protein) was a gift from Hoffman-La Roche Inc (Nutley, NJ). Recombinant murine IL-3 was prepared as conditioned medium from the genetically altered mouse mammary cell line (C127) expressing the mouse cDNA.’~ All factors were used at optimal concentrations.26 Hematopoietic progenitor cell assays. Marrow cell suspensions were assayed for low and high proliferative potential colony-forming cells (LPP-CFC and HPP-CFC), using a double-layer nutrient agar culture system exactly as previously described.26After 14 days of incubation, HPP-CFC typically generated colonies greater than 0.5 mm in diameter consisting of tightly packed cells. Functionally, these cells are defined as primitive cells by their relative resistance to 5-FU, their synergistic growth factor requirements, and their copurification with long-term reconstituting cells in v ~ v o . *LPP-CFC ~~~* are macrophage lineage-restricted cells responsive to CSF-1 alone and typically generate colonies measuring less than 0.5 mm in diameter.29 Monoclonal antibodies (MoAbs). A panel of rat MoAbs was used to investigate age-related changes in the expression of hematopoietic lineage cell surface antigens in each cell suspension. These antibodies included anti-714 (neutrophils, activated macrophages, and macrophage lineage-restricted progenitor^)^'; antGB220 (B cells)3’; anti-MAC-I3’ and a11ti-F4/80~~ (macrophage subsets); antiL3T4(CD4)and anti-Lyt-2 (CD8) (T cells)34;and anti-GR-l (neut r o p h i l ~ ) .Antibodies ~~ were diluted in phosphate-buffered saline (PBS) (pH 7.4, 290 mOsm) supplemented with 5% BCS (binding buffer) and used at predetermined optimal dilutions. Flow cytometric analysis. Cells were labeled using equal volumes of primary antibody and binding buffer for 30 minutes on ice and then washed three times and incubated with mouse serumadsorbed fluorescein isothiocyanate (FITC)-conjugated goat antirat IgG (H+L; Kirkegaard and Perry Laboratories, Gaithersburg, MD)

for a further 30 minutes onice. The cells were washed a further three times, resuspended inPBS supplemented with 0.25% BCS, and held at 4°C before flow cytometric analysis. Labeled cells were analyzed using a FACStarP’”scell sorter (Becton Dickinson, Mountain View, CA) equipped with a 5-W argon ion laser (Coherent Innova 90, Palo Alto, CA) running at 200 mW of power. Light-scatter signals were collected through a 488-nm band pass filter in the forward light scatter path. Background fluorescence (cell surface binding of FITC-conjugated goat antirat second antibody) was less than 5%. Preparation of microscopic sections. Tibiae removed from anesthetized (Penthrane; Abbott Laboratories, North Chicago, IL) mice were immersion fixed in 2% formaldehyde and 2% glutaraldehyde for 48 hours, washed in sucrose buffer, and decalcified in 10% EDTA for 2 weeks. Dehydration in graded ethanol was followed by infiltration and embedding in Polaron embedding medium (Biorad, Cambridge, MA). Three-micrometer sections were cut and stained with Giemsa (BDH Chemicals, Kilsyth, Victoria, Australia). Statistics. Analysis of accumulated data involved testing the difference between paired groups of wild-type mice (CSF-l+/+GMCSF’”) from different experiments using the Wilcoxon matched pairs test. Because no significant difference was detected, accumulated data were pooled and the differences between means of each age-matched genotype were evaluated by one-way analysis of variance (ANOVA). Significance was then tested using the Newmad Keul’s range test. RESULTS

Young (5 weeks old) CSF-l-”GM-CSF-’- mice weighed significantly less than their wild-type littermates, although, like the op/op (CSF-l”-GM-CSF+/+) mouse, by 40 weeks of age, no difference could be detected between the weights of CSF-l-”GM-CSF-” and either op/op mice or wild-type littermates (Table 1). Age-related changes in bone marrow cellularity and hematopoiesis. Young CSF-l-/-GM-CSF-/- mice had a severe deficiency in marrow cellularity similar to that observed in aged-matched op/op mice. By 5 weeks of age, the femoral marrow cellularity of wild-type mice had attained adult levels. In contrast, by 5 weeks of age, the femoral marrow cellularity of CSF-l-”GM-CSF-’- mice was only 3.9 X lo6 cells per femur, or 18% of the cellularity of wild-type femurs (Table 1). However, CSF-l-’-GM-CSF-/- mice subsequently demonstrated an age-related correction in femoral cellularity identical to that observed in the op/op mouse.I4 By 40 weeks of age, the femoral marrow cellularity of CSF1-/-GM-CSF-/- mice had becomecomparable to that of both op/op and wild-type littermate mice (Table 1). Flow cytometric analysis of the expression of the cell surface antigen F4/80 also showed a similar age-related correction of the marrow macrophage deficiency observed in young CSF- 1 -/-GM-CSF-” mice (Table 1) as that observed in maturing op/op At 5 weeks of age, the proportion of F4/80+ cells was approximately 75% that of wild-type marrow. By 40 weeks of age, the proportion of F4/80+ cells was comparable to that of wild-type. Because the femoral bone marrow cellularity in mature CSF- lPGM-CSF-/mice and wild-type mice was also comparable, the total numbers of F4/80+ cells per femur were also equivalent. The presence of equal proportions of macrophages in mature CSF-l-/-GM-CSF-“, op/op, and wild-type mice was

NILSSON ET AL

68

Table 1. Age-Related Changes in Bone Marrow 5-Week-Old Mice CSF-1 GM-CSF

Mouse weight (g) Nucleated cells/femur x106 Ag expression (% positive) F4/80t MAC-l GR-1 B220 CD4 CD8 CD41CD81

+I+ +I+

40-Week-Old Mice

-1-

-1-

-1-

+l+ +l+

+I+ -/-

-I-

+l+

+/+

-l-1-

17.5 5 0.4 23.3 5 1.7

13.8 5 0.4* 5.4 IT 0.6”

10.5 5 1.0* 3.9 i 1.2’

24.2 2 3.0 19.0 i 1.9

28.0 2 2.0 22.4 2 3.1

22.4 IT 2.3 20.0 5 4.0

20.1 i 2.6 17.4 2 3.1

44.1 33.4 34.3 43.1 8.6 2.0 10.8

22.1 5 3.2** 55.4 t 1.7’ 53.2 t 0.8* 20.5 t 3.4* ND ND 9.8 t 0.4

34.1 5 3.9’ 46.5 5 6.5’ 49.2 t 3.9* 15.2 5 5.8* 13.2 t 2.3 1.7 5 0.3 ND

46.9 5 4.4 55.5 5 1.8 46.2 i 8.4 19.4 2 1.9 11.3? 2.2 1.8 2 0.7 ND

50.3 2 2.1 50.7 5 2.7 52.7 -t 2.2 25.0 2 2.0 12.3 5 0.9 4.2 5 0.9 ND

46.1 60.2 67.0 14.6 10.3 3.5

49.1 2 2.8 62.0 2 3.5 58.1 i 3.8 18.6 2 2.1 12.3 ? 1.2 4.5 t 1.0 ND

5 2.5*

t 2.4 2 2.7

2 3.3 2 1.4 2 0.9 2 1.8

2 10.2

2 2.9 ? 2.8

2 1.8 5 1.9 2 1.6 ND

Values are the means -t SEM of analysis in at least three individual mice. Abbreviation: ND, not done. * P < .05. t For F4180, a forwardversus perpendicular light scatter window was set to exclude lymphocytic and granulocytic cells that may benonspecifically labeled.3g Values presented for the expression of F4180 antigen were obtained from both 3- and 5-week-old mice. § Anti-CD4 and anti-CD8 were used as an antibody cocktail.

*

confirmed by light microscopy (data not shown). These cells were clearly distinguishable from other cell types by the presence of paracrystalline inclusions, which are characteristic of bone marrow macrophages in normal mice older than 60 days.36,” At 5 weeks of age, both young CSF-l-/-GM-CSF-/- and op/op mice had a significant increase in the proportion of MAC-1’ cells and Gr-l+ cells compared with that of their wild-type littermates (Table 1). This increase in the proportion of MAC-l+ cells was most likely caused by the increase in granulocytes, because the MAC-l antigen is expressed by both macrophages and granulocytes. Also, as previously described, there was a lower incidence of F4/80+ macrophages in both young CSF-l-/-GM-CSF-/- and op/op mice compared with wild-type littermates. There was also no evidence of the upregulation of MAC-l antigen expression, reflected by a shift in mean channel fluorescence (data not shown). However, the absolute number of GR-1’ cells per femur in both young CSF-l-/-GM-CSF-/- and op/op mice was markedly reduced compared with wild-type control values (1.9 t 0.6,2.9 f 0.4, and 6.9 5 1.3 X lo6,respectively). At 5 weeks of age, there was also a significant decrease in both the proportion (Table 1) and total numbers of B220’ lymphocytes in both CSF-l-/-GM-CSF-/- and op/op femurs compared with that of their wild-type littermates (0.5 t 0.2, 1. l -t 0.2, and 10.0 1.0 X lo6,respectively). Although, by 40 weeks of age, the proportion (Table 1) and total number of B220+ cells was comparable in CSF-l-/-GM-CSF-/-, op/ op, and wild-type mice. The hematopoietic potential of CSF-1”-GM-CSF-’- femoral bone marrow was evaluated by in vitro clonal assay of committed (LPP-CFC) and primitive (HPP-CFC) progenitors. There were no differences in the incidence of LPP-CFC or HPP-CFC in the marrow of CSF-l-”GM-CSF-/- and wild-type mice at anyage examined (data not shown). Therefore, the total femoral content of these progenitors (Table 2) reflected the age-related changes observed in marrow cellu-

larity (Table l). By 40 weeks of age, the femoral content in CSF-l-/-GM-CSF-”, op/op, and wild-type mice was comparable (Table 2). Age-related changes in splenic hematopoiesis. Like young op/op m i ~ e , ’ ~the , ’ diminished ~ space for marrow hematopoiesis and the reduced marrow hematopoietic activity in young CSF-i”-GM-CSF-” mice correlated with an increase in splenichematopoiesis. Although young CSF-1-/GM-CSF-I- mice weighed significantly less than their wildtype littermates (Table l), there was evidence of splenomegaly, with an average spleen cellularity 1.5 times (Table 3) and a spleen weight 1.3 times that of their wild-type littermates (data not shown). The primitive and committed progenitor cell incidence and content, as evaluated by the in vitro clonal assay of LPP-CFC and HPP-CFC, were also significantly elevated. Young CSF- 1-/-GM-CSF-/- mice had an average increase in the incidence of LPP-CFC and HPP-CFC 2- and 10-fold that of wild-type littermates, respectively (Table 4). Evidence of extramedullary hematopoiesis in the spleen of young CSF-l-/-GM-CSF-/- mice was confirmed by flow cytometric analysis (Table 3). At 5 weeks of age, spleens from CSF-l-/-GM-CSF-/- mice were characterized by an increase in granulocytopoiesis similar to that seen in the op/ op mouse. There was a 39%, 32%, and 76% increase in the

Table 2. Bone Marrow Hematopoiatic Activityin Mature Mice Femoral Content lx103/femur) CSF-l GM-CSF

+I+ +I+

+I+ -1-

-1-

LPP-CFC HPP-CFC

570 2 116 9.4 2 2.8

551 t 112 8.6 2 5.5

785 2 139 14.8 2 3.4

~~

+I+

-1-1-

544 2 76 11.2 2 3.6

~~~

Values are the means 2 SEM of three replicate dishes for three individual mice at 40 weeks of age. No significant difference was detected at the .05 level.

OF

ROLE

69

IN THE

OPIOP

MOUSE

Table 3. Age-Related Changes in Spleen ~~~

~~~~~

5-Week-Old Mice CSF-1 GM-CSF

+l+

Nucleated cellslspleen x10’ Ag expression (% positive) F4BO MAC- 1 GR-1 714 B220 CD4 CD8 CD4lCD8t

13.8 ? 1.2

22.1 ? 1.8

24.1 ? 4.5 9.8 ? 1.1 14.3 ? 2.9 5.5 ? 1.0 47.1 ? 2.9 20.3 ? 1.2 10.2 ? 1.6 23.0 ? 0.9

23.1 21.1 19.6 4.5 26.7

40-Week-Old Mice

-1+I+

+/+

2 1.0 2 2.2*

1.9 0.6 ? 2.1* ND ND 19.8 ? 3.2 ?

?

-1-1-

+l+ +l+

+I+ -1-

-1+I+

18.4 2 5.2

12.6 ? 3.5

13.3 ? 3.2

10.2 ? 1.2

55.1 ? 2.4* 11.5 2 2.0 27.9 ? 4.8 12.0 ? 2.3’ 13.2 ? 0.5* 14.6 -C 4.0 7.2 ? 1.1 ND

22.1 ? 1.8 9.4 c 0.3 16.3 ? 1.5 3.3 ? 0.5 30.4 t 5.9 30.8 t 1.0 14.5 ? 0.9 ND

20.9 t 3.4 10.3 2 1.1 14.8 2 1.8 6.2 2 1.8 27.8 c 2.8 32.5 2 1.6 15.6 ? 1.7 ND

20.0 ? 2.8 9.8 ? 0.3 18.8 2 1.0 9.9 2 0.2* 28.6 t 3.5 31.6 ? 1.9 15.1 ? 0.6 ND

-1-1-

8.7 2 2.2 19.9 8.7 18.4 8.9 21.4 39.1 16.5

2.3 t 1.1 ? 0.5 2 0.3’ c 3.5 ? 3.9 ? 1.7 ND ?

Values are the means ? SEM of analysis in at least three individual mice. Abbreviation: ND, not done. P < .05. t Anti-CD4 and anti-CD8 were used as an antibody cocktail.

proportion of GR-l+, MAC-l+,and F4/80+ cells, and a 75% reduction in the proportion of B220+ cells, respectively (Table 3). This increase in both spleencellularity and the proportion of GR-l+, MAC-I+,and F4/80+ cells was reflected by a significant increase in the total numbers of these cell types in CSF- 1-/-GM-CSF-/- mice compared with theirwild-type controls (4.6 2 0.6,2.0 t 0.4, and 10.2 2 3.0 X IO7, respectively, compared with 2.3 2 0.5, 1.4 2 0.2, and 3.8 2 1.0 X lo7, respectively). Similarly, there was a decrease in the total number of B220+ spleen cells in CSF-I-/-GM-CSF-/mice compared with their wild-type littermates (2.4 t 0.7 and6.6 2 0.9 X lo7, respectively)..Comparedwiththeir wild-type littermates, young CSF-.l-’-GM-CSF-“ mice were also characterized by a significant increase in the pioportion (Table 3) and total number of 7/4+ cells (19.8 2 2.1 and 7.9 2 2.0 X IO6, respectively), an antigen expressed on neutrophils and activated macrophages. With the age-related correction of bone marrow hematopoietic parameters, there was asimilar resolution of the splenomegaly, spleen cellularity, and a return to normal values ofthemyeloidandlymphoid cell proportions in CSF-1-/GM-CSF-I- mice (Table 3) as that previously described in the op/op mouse.I6 In addition, the incidence (Table 4) and content of splenic LPP-CFCandHPP-CFC decreased to become comparable to that of wild-type littermates. By 40 weeks of age, there were no differences in the measured

Table 4. Age-Related Changes in Splenic Progenitor Cell Incidence Progenitors/2,5M)Cells 5-Week-Old Mice CSF-1 GM-CSF

LPP-CFC HPP-CFC

+l+ +l+

3.7 ? 0.5 0.03 ? 0.03

-1-I-

7.9 ? 1.3* 0.09” 0.3

40-Week-Old Mice

+l+

+I+

-1-1-

2.7 ? 1.1 0.06 5 0.06

02 0

5 ? 2.2

Values are the means ? SEM of three replicate dishes for three individual mice. P < .05.

hematopoietic characteristics between the spleens of CSFl-/-GM-CSF-/-, op/op, and wild-type mice, except for the continued increase in both the proportion (Table 3) and total number of7/4+ cells in CSF-l-I-GM-CSF-” and op/op mice compared with their wild-type littermates (7.8 2 2.1, 10.2 2 1.4, and 4.6 2 1.7 X lo6). Age-related changes in peripheral blood immunophenotypes. A panel of MoAbswas also used to analyze the peripheral blood cell types (Table 5). No differences were seen in thePropomonofF4/80+ cells, in eitherCSF-1-/GM-CSF-” or op/op mice compared with wild-type mice at either 5 or after 40 weeks of age (Table 5). This observation is in agreement with peripheral blood differential counts, in which no diffbrences were seen in the number of monocytes (data not shown). At 5 weeks of age, both young CSF-I-/-GM-CSF”- and op/op mice had a significant increase in the proportion of Gr-l+ cells as well as a decrease in the proportion of B220’ lymphocytes compared withthat of theirwild-typelittermates (Table 5). However, although this difference was reflected in the total numbers of these cell types (1 1.l 2 2.2, 12.8 2 1.4, and 5.8 2 0.9; and 5.9 t 0.7, 7.0 2 0.04, and 9.4 2 1.7 X 105/mL,respectively), the changes were not significant. By 40 weeks of age, the proportionof expression (Table 5) and total number of all surface antigens examined were comparable in CSF-l -/-GM-CSF-/-, op/op, and wildtype mice. Age-related changes in peritoneal cell immunophenotypes. Significantly fewer cells wererecovered from the peritoneum ofboth CSF-l-l-GM-CSF-” and op/op mice compared with that of their wild-type littermates at both 5 and 40 weeks of age (Table 6). Consequently, there were significantly less cells of all types analyzed within theperitoneal lavages of both CSF-l-”GM-CSF-/- and op/op mice compared with their wild-type littermates at both ages.Analysis of peritoneal macrophages showed an additional significant decrease in the proportion of F4/80+ cells in both CSF-l-”GM-CSF-’- and op/op mice compared with their wild-type littermates at both 5 and 40 weeks of age. No age-

70

NILSSON ET AL Table 5. Age-Related Changes in Peripheral Blood 40-Week-Old Mice

5-Week-Old Mice

CSF-l GM-CSF

+l+ +I+

-1+l+

Nucleated celldmL x106 Ag expression (% positive) F4/80 GR-1 B220 CD4lCD8t

2.0 ? 0.1

2.9 2 1.8 0.5

19.8 29.0 46.1 14.2

? 3.0

2 3.0 ? 5.9 ? 2.1

14.2 % 3.1 44.0 ? 5.0* 24.0 ? 0.2 ND

-l-

+/+ +l+

-12 0.4

25.2 ? 5.1 62.8 % 2.3* 31.1 2 7.9 12.9 ? 2.1

-1+l+

+l+ -1-

-1 -1-

1.7 3.5 22.1

2 0.4

3.0

0.7

1.8 2 0.2

28.6 ? 1.5 42.3 i 5.5 34.6 f 4.2 20.2 ? 4.3

31.0 ? 5.2 41.0 ? 3.9 32.9 2 5.1 24.7 ? 5.7

23.0 52.5 36.8 18.3

? 1.6

18.1 ? 0.8 52.0 ? 9.0 22.5 f 1.8 31.0 t- 8.3

? 12.8

2 6.3

t- 1.1

Values are the means ? SEM of analysis in at least three individual mice. Abbreviation: ND, not done. * P < .05. t Anti-CD4 and anti-CD8 were used as an antibody cocktail.

related increase in this population was detected in either CSF-l-I-GM-CSF-” or op/op mice (Table 6). The proportion of B220+ peritoneal cells was significantly increased in both 5-week-old CSF-l-’-GM-CSF-” and op/ op mice compared with wild-type littermates (Table 6). Although, because of the significant reduction in peritoneal cellularity, the total number of B220+ cells remained significantly lower in both 5-week-old CSF-l-/-GM-CSF-/and op/op mice compared with wild-type littermates (2.7 ? 0.4, 1.1 2 0.1, and 5.3 -t 1.6 X lo5,respectively). Similarly, although no differences in the proportion of B220+ peritoneal cells could be detected by 40 weeks of age, the total number of peritoneal B220+ cells in CSF-l-/-GM-CSF-/and op/op mice compared with wild-type littermates remained significantly reduced (1.2 ? 0.9, 2.1 ? 0.5, and 50.0 -t 16.3 X lo5, respectively), because no age-related correction in peritoneal cellularity was observed. The proportion of CD4+ andor CD8+ cells was significantly increased in 5-week-old CSF-l”-GM-CSF-/- mice compared with wild-type littermates (Table 6). However, as a consequence of the significant reduction in peritoneal cellularity in 5-week-old CSF-l-/-GM-CSF-’- mice compared with wild-type littermates (Table 6), there was no significant difference in the total number of CD4+ andor CD8+ cells (1.4 ? 0.1 and 1.4 ? 0.3 X lo5, respectively). By 40 weeks of age, the proportion of CD4+ andor CD8+

cells in both op/op and CSF-l-/-GM-CSF-/- mice remained significantly higher than wild-type littermates, but, because there was no increase in peritoneal cellularity, no significant difference was detected between the total numbers of CD4’ and/or CD8+ cells in CSF-l-”GM-CSF-’- and op/op mice compared with wild-type littermates (1.5 ? 0.7, 3.2 ? 1 . 1 , and 17.2 2 7.5 X lo5, respectively). DISCUSSION

This studydemonstratesthatyoung CSF-l-/-GM-CSF-” mice have the same hematopoietic characteristics and deficiencies as wepreviouslydescribedforthe op/op mouse.14-lh Young CSF-l -/-GM-CSF-/- mice had a severe deficiency in bone marrow cellularity and macrophage content. During this period of suppressed marrow hematopoiesis, the spleen played a primary hematopoietic role. The CSF-l-”GM-CSF-/- mouse also undergoes a similar age-related correction of the hematopoietic deficiencies evident in the young mouse as that described in the op/op mouse.14-16 All measured hematopoietic parameters, except for peritoneal cellularity and macrophage content, became comparable to that of wild-type littermates by 40 weeks of age. Therefore, as with the op/op mouse, the osteopetrosis and hematopoietic deficiencies in young CSF-l-/-GMCSF-/- mice are not permanent, but undergo a progressive age-related correction.

Table 6. Age-Related Changes in Peritoneal Cells ~

~~

5-Week-Old Mice

CSF-1 GM-CSF

+l+ +l+

Nucleated cellsllavage x106 Ag expression (% positive) F4/80 GR-1 B220 CD4lCD8t

1.5 ? 0.2 0.4 56.9 24.3 t- 2.1 2 5.9 18.6 18.6 34.2 c 2.9 9.4 ? 2.2

-I+I+ ? 0.06* 0.6

+-33.1 1.9* ? 3.6

54.0 ? 5.2” ND

40-Week-Old Mice -/-1? 0.1* 10.8 ? 33.6 6.0* 32.3 ? 7.4 48.1 2 4.4* 22.9 +- 3.6’

Values are the means t SEM of analysis in at least three individual mice. Abbreviation: ND, not done. P < .05. t Anti-CD4 and anti-CD8 were used as an antibody cocktail.

+l+ +l+ %4.4 2.7

18.2 It 5.2 31.6 17.4 f 8.0 54.1 f 2.6 14.4 2 2.6

+l+ -I? 1.0*

5.9 It 4.2 ? 5.8 55.2 t 4.6 12.2 2 2.0

-1+l+

0.7

?

0.2* 0.3

3.8’ t10.6 2.4* 8.5 11.4 ? 3.0 29.6 t- 1.8’ 53.4 ? 2.9*

-1-1?

0.1*

? 1.3 47.9 ? 3.5 44.8 c 6.5’

ROLE OF GM-CSF IN THE

OHOP

MOUSE

It is interesting to note that there was a comparable incidence and content of peripheral blood monocytes in CSFl”-GM-CSF-”, op/op, and wild-type mice at all ages examined. In addition, there was no age-related correction of the deficiencies in peritoneal cellularity or macrophage content in either CSF-l-”GM-CSF-/- or op/op mice by 40 weeks of age, suggesting an essential requirement for CSF-1 for this macrophage subpopulation. These findings are in agreement with the data recently published by Cecchini et a l , 3 8 whose observations also indicate differences in the dependency of various subpopulations of mononuclear phagocytes on CSF- l. CSF-l-/-GM-CSF-/- mice have a pulmonary pathology often complicated by infectionz3that may have an impact on the hematopoietic profile we have described. For example, the increase in splenic 7/4+ cells can be explained byan increase in activated macrophages, perhaps resulting in part from the pulmonary pathology. This finding is consistent with the large increase in F4/80+ splenic cells, both proportionally and numerically. Other studies have indicated that the daily administration of CSF-1 to newborn op/op mice is required to prevent osteopetrosis and the accompanying hematopoietic deficien~ i e s . ” , ’Our ~ previous studies have shown that the hematopoietic system has the capacity to use alternative mechanisms to compensate for the absence of biologically The present study demonactive CSF-1 in op/op mi~e.’~‘’~ strates conclusively that one of the growth factors that might have been suspected to ameliorate the hematopoietic deficiencies involved, GM-CSF, is not essential for the correction of the hematopoietic deficiencies in the op/op mouse, particularly in macrophage development. The data also show that neither GM-CSF nor CSF-1 is an absolute requirement for the commitment of primitive hematopoietic stem cells to the macrophage lineage or for the differentiation of at least some classes of macrophages. This finding suggests that an alternate, regulatory growth factor can be involved in macrophage production and function in vivo. This may be an as yet unidentified factor or may be a manifestation of plasticity among the known hematopoietic factors.” ACKNOWLEDGMENT

WeacknowledgeKallyYuen analysis of the data.

for her advice on the statistical

REFERENCES l. Lane PW: Research news. Mouse Newsletter 6050, 1979 2. Gisselbrecht S,Sola B, Fichelson S, Bordereaux D, Tambourin P, MatteiM-G,SimonD,Guent J-L The murine CSF-l gene is localized on chromosome 3. Blood 73:1742, 1989 3. Wiktor-Jedrzejczak W, Bartocci A, Ferrante AW Jr, AhmedAnsari A, Sell KW, Pollard J W , Stanley E R Total absence of colony-stimulating factor 1 in the macrophage-deficient osteopetrotic (op/op) mouse. Proc Natl Acad Sci USA 87:4828, 1990 4. Yoshida H, Hayashi S, Kuniscada T, Ogawa M, Okamura H, Sudo T, Schultz LD,Nishikawa S: The murinemutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345:442, 1990 5. Stanley ER, Guilbert W, Tushinski RI, Bartelmez SH: CSF-

71

1-A mononuclear phagocyte lineage-specific hemopoietic growth factor. J Cell Biochem 21:151, 1983 6. Hume DA, Pavli P, Donahue RE, Fidler U: The effect of human recombinant macrophage colony-stimulating factor (CSF-l) on the murine mononuclear phagocyte system in vivo. Immunol141:3405, J 1988 7. Sherr CJ: Regulation of mononuclear phagocyte proliferation by colony-stimulating factor-l. Int J Cell Cloning 8:46, 1990 8. Marks SC Jr, Lane PW Osteopetrosis, anew recessive skeletal mutation on chromosome 12 of the mouse. J Hered 67: 11,1976 9. Wiktor-Jedrzejczak WW, Ahmed A, Szczylik C, Skelly RR: Hematological characterization of congenital osteopetrosis in op/op mouse. Possible mechanism for abnormal macrophage differentiation. J Exp Med 156:1516, 1982 10. MarksSC Jr: Morphologicalevidence of reduced bone resorption in osteopetrotic (op) mice. Am J Anat 163:157, 1982 11. Marks SC Jr: Osteopetrosis-multiple pathways for the interception of osteoclast function. Appl Pathol 5:172, 1987 12. Felix R, Cecchini MC, Fleisch H: Macrophage colony stimulating factor restores in vivo bone resorption in the op/op osteopetrotic mouse. Endocrinology 127:2592, 1990 13. Wiktor-Jedrzejczak W, Urbanowska E, Aukerman SL, Pollard J W , Stanley ER, RalphP, Ansari AA,Sell KW, Szperl M: Correction by CSF-1 of defects in the osteopetroticop/op mouse suggests local, development, and humoral requirementsfor this growth factor. Exp Hematol 19:1049, 1991 14. Begg SK, Radley JM, Pollard,J W , Chisolm OT, Stanley ER, Bertoncello I Delayed hematopoietic development in osteopetrotic (op/op) mice. J Exp Med 177:237, 1993 15.Begg SK, Bertoncello I: The hematopoieticdeficiencies in osteopetrotic (op/op) mice are not permanent, but progressivelycorrect with age. Exp Hematol 21:493, 1993 16. Nilsson SK, Bertoncello I: Age-related changes in extramedullary hematopoiesis in the spleenof normal and perturbed osteopetrotic (op/op) mice. Exp Hematol 22:377, 1994 17.MetcalfD:Hematopoieticregulators:Redundancy or subtlety? Blood 82:3513, 1993 18. Wiktor-Jedrzejczak W, Urbanowska E, Szperl M: GM-CSF corrects macrophagedeficienciesbutnotosteopetrosisin CFS-l deficient op/op mouse. Exp Hematol20769, 1992 (abstr 246) 19. Wiktor-Jedrzejczak W In vivo role of macrophage growth factors as delineated using CSF-1 deficient op/op mouse. Leukemia 7:S117,1993 20. Wiktor-Jedrzejczak W, Urbanowska E, Szperl M: Granulocyte-macrophage colony-stimulatingfactor corrects macrophage deficiencies, but not osteopetrosis, in the colony-stimulating factor-ldeficient op/op mouse. Endocrinology 134:1932, 1994 21. Gasson JC: Molecular physiology of granulocyte-macrophage colony-stimulating factor. Blood 77:1131, 1991 D, Hodgson G, Gall 22. Stanley E, Lieschke GJ, Grail D, Metcalf JAM,Maher DW, Cebon J, Sinickas V, Dunn AR: Granulocytemacrophage colony-stimulating factor-deficient mice show no major perturbation of hematopoiesis butdevelop a characteristicpulmonary pathology. Proc Natl Acad Sci USA 91:5592, 1994 23.LieschkeGJ,StanleyE,Grail D, Hodgson G, Sinickas V, Gall JAM, Sinclair RA, DUM AR: Mice lacking both macrophageand granulocyte-macrophage colony-stimulatingfactor have macrophages and co-existent osteopetrosis and severe lung disease. Blood 84:27, 1994 24. Bradley TR, Stanley ER, Sumner MA: Factors from mouse tissues stimulating colony growth of mouse bone marrow cells in vitro. Aust J Exp Biol Med Sci 49:595, 1971 25. Karasuyama H, Melchers F: Establishment of mouse cell lines which constitutively secrete large quantities of interleukins 2, 3, 4,

72

or 5, using modified cDNA expression vectors. Eur J Immunol 18:97, 1988 26. Bartelmez SH, Bradley TR, Bertoncello I, Mochizuki DY, Tushinski RJ, Stanley ER, Hapel AJ, Young IG, Kriegler AB, Hodgson GS: Interleukin I plus interleukin 3 plus colony-stimulating factor 1 are essential for clonal proliferation of primitive myeloid bone marrow cells. Exp Hematol 17:240, 1989 27. McNiece IK, Bertoncello I, Kriegler AB, Quesenbeny PJ: Colony-forming cells with high proliferative potential (HPP-CFC). Int J Cell Cloning 8:146, 1990 28. Bertoncello I: Status of high proliferative potential colonyforming cells in the hematopoietic stem cell hierarchy. Curr Top Microbiol Immunol 177:83, 1992 29. Bradley TR, Hodgson GS: Detection of primitive macrophage progenitor cells in mouse bone marrow. Blood 54: 1446, 1979 30. Hirsch S , Gordon S : Polymorphic expression of a neutrophil differentiation antigen revealed by monoclonal antibody 7/4. Immunogenetics 18:229, 1983 3 1. Coffman R L : Surface antigen expression and immunoglobulin gene rearrangement during mouse pre-B cell development. Immuno1 Rev 69:5, 1982 32. Springer T, Galfre G, Secher DS, Milstein C: Mac-l: A macrophage differentiation antigen identified by monoclonal antibody. Eur J Immunol 9:301, 1979 33. Austyn JM, Gordon S: F4/80, a monoclonal antibody directed

NILSSON ET AL

specifically against the mouse macrophage. Eur J Immunol I1:805, 1981 34. Cobbold SP, Jayasuriya A, Nash A, Prosper0 TD, Waldmann H: Therapy with monoclonal antibodies by elimination of T-cell subsets in vivo. Nature 312:548, 1984 35. Hestdal K, Ruscetti F W , Ihle JN, Jacobsen SE, Dubois CM, Kopp WC, Longo DL, Keller JR: Characterization and regulation of RB6-8C5 antigen expression on murine bonemarrow cells. J Immunol 147:22, 1991 36. Hudson G, ShortlandJR:Crystalloidmaterial in marrow macrophages of specific pathogen-free mice. Acta Anat 87:404, 1974 37. Sasaki K, Matsumura G, Ito T: Crystalloid inclusion-containing macrophages in the bone marrow and red pulp of the mouse, with particular relation to age, sex and hydrocortisone administration: Qualitative and quantitative electron microscopy. Arch Histol Jpn 46:381, 1983 38. Cecchini MG, Dominguez MG, Mocci S, Wetterwald A, Felix R, Fleisch H, Chisholm 0, Hofstetter W, Pollard JW, Stanley ER: Role of colony stimulating factor-l in the establishment and regulation of tissue macrophages during postnatal development of the mouse. Development 120:1357, 1994 39. Hume DA: Murine eosinophil granulocytes bind the macrophage-monocyte specific monoclonal antibody F4/80. J Leukoc Biol 51517, 1992 (letter)