Effects of Monoclonal Antibody and Complement Treatment of Human Marrow on Hematopoiesis in Continuous Bone Marrow Culture1

[CANCER RESEARCH 45, 758-767, February 1985]

Effects of Monoclonal Antibody and Complement Treatment of Human Marrowon Hematopoiesis in Continuous Bone Marrow Culture1

Joel S. Greenberger,2 Lisa Rothstein, Paolo DeFabritiis, Marco Bregni, Robert Bast, Jr., Jerome Ritz,3

Lee M. Nadler, Jeffrey M. Lipton, and Mary Ann Sakakeeny

Joint Center for Radiation Therapy [J. S. G., L. R., M. A. S.], Department of Radiation Therapy, Harvard Medical School, and Departments of Medical Oncology[P. D., M. R, R B., J. R., L. M. N.] and Pediatrie Oncology [J. M. L], Tumor Immunology [P. D., M. B., R. B., J. R., L M. N.] Dana-Farber Cancer Institute,

Boston, Massachusetts 02115

ABSTRACT

Long-term bone marrow cultures were established from single-

cell suspensions of human bone marrow that had been treatedwith monoclonal antibodies and complement. Each treated cellsuspension was evaluated for production of hematopoietic stemcells over 20 weeks. Treatment with antibody to HLA-DR (la),

B1, J2, or J5 did not remove adherent cells including thosedifferentiating to adipocytes in 17-hydroxycorticosteroid. In con

trast, treatment with monoclonal antibody directed against human ftî-microglobulin reduced adipocyte numbers by 100-fold

and reduced the total adherent cell density over 70%. Cumulativetotal nonadherent cell and granulocyte-macrophage colony-forming units (GM-CFUc) production over 20 weeks was not significantly altered by one cycle of anti-la plus complement or up tothree cycles of treatment with complement and anti-J2, -J5, or-B1. However, one cycle of treatment with anti-ft-micro-globulindepressed production of both GM-CFUc and nonadherent cellsby over 100-fold compared to other treatment groups. While onecycle of treatment of anti-la and complement killed all detectablecells forming CFU-erythroid-granulocyte-megakaryocyte-macro-phage, blast-forming units (erythroid), and GM-CFUc, GM cluster-forming cells survived. Treatment of marrow with three cyclesof anti-la and complement removed all detectable GM colony-and GM cluster-forming cells; however, this marrow producedfewer cumulative la-positive GM-CFUc. Long-term bone marrow

cultures may prove to be an interesting system for in vitroanalysis of the effects of new immunotherapeutic agents including other monoclonal antibodies prior to clinical use.

INTRODUCTION

The association of specific cell surface antigenic determinantswith defined stages of differentiation of the pluripotential hema-

1Supported by NIH Research Grants CA25412, CA12662, and CA28740 andUnited States Department of Defense Contract DMD-17-82-C-2207.

2 Present address: Dept. of Radiation Oncology, University of Massachusetts

Medical School, 55 Lake Ave. N., Worcester, MA 01605. To whom requests forreprints should be addressed.

3 Scholar of the Leukemia Society of America.4The abbreviations used are: CFU-GEMM, multipotential erythroid-granulocyte-

megakaryocyte-macrophage colony-forming unit (hematopoietic progenitor cells);B2M, .¡j-macroglobulm; CALLA, common acute lymphoblastic leukemia antigen;GM-CFUc, granulocyte-macrophage colony-forming unit; BFUe, blast-forming unit(erythroid); CFU-F, fibroblast colonies; CPUs, pluripotential stem cell; CFU-mega,

megakaryocyte colonies.Received March 5,1984; accepted October 23,1984.

topoietic stem cell is well established (1-3, 12, 21, 24-26, 28,

50). Antigenic determinants have been described specific for theB-lymphocyte (25, 28), T-lymphocyte (3, 50, 52), monocyte-

macrophage (21, 22), or granulocyte pathway (21). Some antigenic determinants are shared between cells committed to asingle lineage and CFU-GEMM" hematopoietic progenitor cells

(21, 22). Other antigenic determinants may be associated withterminally differentiated cells of the neutrophile, granulocyte,eosinophile, or basophile mast cell lineage (2, 22).

The ease of correlating the distribution of specific antigenicdeterminants with the diverse hematopoietic cell types in themarrow has been aided by use of specific functional (34) andhistochemical (18) assays. Since leukemias and lymphomas canexpress cell surface antigens that are not found on the majorityof normal progenitor cells (1, 3,7, 21, 22, 24-26, 28, 36, 50, 52)

antibodies have been used to eliminate malignant cells fromhuman marrow (3,50). Monoclonal antibody reagents (3,12,36,50) have made available adequate quantities of reagent fortreating marrow in vitro prior to use in autologous marrowtransplantation (5, 29).

In contrast, the nonhematopoietic cellular elements of the bonemarrow (6, 35, 43, 49, 54), have not been as well characterized.There is uncertainty regarding the number of such cell lineagesthat are morphologically identified as fibroblasts, preadipocytes,reticular adventitial cells, endothelial cells, sinus-lining cells, andmacrophages (6-8, 10, 13, 14, 16, 20, 27, 35, 43, 46, 49, 54).

Furthermore, there are no generally accepted restrictive criteriafor defining each nonhematopoietic cell category (6-8, 10, 13,

14,16,20,27,35,43,46,49,54). Controversy exists concerningthe existence of a common progenitor cell for the pluripotentialhematopoietic stem cell and the nonhematopoietic lineages ofendothelial cell and bone marrow fibroblast (45).

Autologous marrow transplantion is now an accepted modalityin the treatment of hematopoietic cancers (3, 41 ) and some solidtumors (5, 48). Immunological removal of tumor cells by incubation of marrow in vitro with monoclonal antibodies and complement has been followed by reinfusion of marrow after administration of cytotoxic chemotherapy and/or total body irradiation(3). Successful transplant depends on the redistribution of engrafted marrow in vivo, proliferation of these cells, and theirappropriate differentiation. Current monoclonal antibody andcomplement reagents are tested for killing of hematopoietic stemcells by assays for GM-CFUc, BFUe, and CFU-GEMM. Since the

most primitive analogue of the pluripotential stem cell (CFUs)that can be tested in vitro with human marrow is the CFU-GEMM

(9), a monoclonal antibody and complement treatment that kills

CANCER RESEARCH VOL. 45 FEBRUARY 1985

758

Research. on September 21, 2015. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from

HEMATOPOIETIC EFFECTS OF MONOCLONAL ANTIBODIES

tumor cells (e.g., Burkitt's lymphoma cells) but spares CFU-

GEMM (and perhaps also BFUe and GM-CFUc) is considered to

be safe for clinical use.Recent data from allogeneic transplant protocols have pro

vided reason for renewed concern. These data indicate thatnonhematopoietic as well as hematopoietic cells of the donormarrow may engraft the recipient (27, 46). Whether donor nonhematopoietic cell engraftment is clinically relevant is unknown.As one approach toward understanding the effects of monoclonal antibody and complement treatment of human marrow,monoclonal antibody and complement reagents were tested foreffects on the establishment of long-term bone marrow culturesthat normally generate hematopoietic progenitor cells and gran-ulocytes in vitro for several months.

MATERIALS AND METHODS

Monoclonal Antibody Reagents. Murine monoclonal antibodiestested included reagents directed against human lymphocyte antigens:J5 (CALLA) (26, 41); J2 (23); and B1 (47). A monoclonal antibodyprepared against the nonpolymorphic region of the a chain of human la,HLA-DR determinant, has been described previously (33). A monoclonal

antibody reagent prepared against B2M has also been described (33).The spectrum of cells expressing each antigenic determinant is shownin Table 1.

Treatment of Marrow with Antibody and Complement. Single-cell

suspensions of freshly removed intraoperative hip marrow were preparedfor long-term bone marrow culture according to published procedures(13). Briefly, single-cell suspensions were freed of bone spicules and

paniculate material by passage through a sieve. Suspensions weredrawn through successively smaller-gauge needles to a 30-gauge needlein McCoy's Medium 5A supplemented according to published procedures

(13) and containing 12.5% fetal calf serum and 12.5% horse serum (10,13). The marrow was then standardized to 108 to 109 cells in 50 ml

serum-free medium, washed by centrifugation, and then treated in a 5-

ml volume with each monoclonal antibody (1:100 dilution). Antibodytreatments were carried out at 4°for 30 min. Following each antibodytreatment, marrow preparations containing 108 to 109 cells per 5 ml were

then washed in serum-free medium by centrifugation, and the cells wereresuspended in 5.0 ml of rabbit complement at a 1:5 dilution at 37°for

45 min. The bone marrow preparations were then washed again andeither transferred to 40-sq cm plastic tissue culture flasks at 2 to 4 x107 cells/flask in 8.0 ml total volume (13) or subjected to 2 additional

cycles of antibody and complement treatment as above. Each humanmarrow specimen was used individually in a separate and unique experiment. No marrow samples were pooled.

For several studies, sequential treatment with different antibodies wascarried out. In each case, a full 30-min treatment with each monoclonalantibody was followed by a 45-min treatment with complement, followed

by an additional cycle with each of the other monoclonal antibodies.

Control cultures for each experiment were initiated and were treatedidentically by holding cultures at each temperature for each interval inserum-free medium alone (13).

Continuous Bone Marrow Cultures. Human long-term bone marrow

cultures were established and fed according to published methods (13).Single-cell suspensions, prepared as described above, were inoculatedinto 40-sq cm plastic flasks (Coming) with 2 to 4 x 107 cells/flask in 8.0ml complete medium, containing McCoy's Medium 5A, supplemented as

described previously (13); 12.5% horse serum, 12.5% fetal calf serum,and, in all experiments, 10~5 M hydocortisone were added freshly bi

weekly (17, 44).RBC were removed by first allowing all cultures to remain undisturbed

for 4 to 5 days, followed by removal of all nonadherent cells, ficoll-

Hypaque density gradient centrifugation to remove RBC, and then returnof the washed WBC fraction entirely to the culture flasks in fresh medium.This 5-day delay has been shown to be critical in facilitation of attachment

of the hematopoietic stem cell islands that contain the most primitivehematopoietic cells (13).

Cultures were fed by removal of all nonadherent cells and medium,centrifugation of this medium to remove the cells for specific assays, andthen refeeding of the cultures with an equivalent volume of fresh mediumcontaining fresh hydrocortisone. Cultures were fed biweekly unlessotherwise specified. Cultures were maintained at 33°in a high-humidity

incubator with 5% COi.Assays for Specific Hematopoietic Cell Lineages. Nonadherent cells

removed from the cultures biweekly were counted by hemocytometerand transferred to assays of multipotential progenitor cells, includingCFU-GEMM (9), BFUe (9), CFU-mega (9), and GM-CFUc (9,13). For theGM-CFUc assay, the source of colony-stimulating factor included phy-tohemagglutinin-lymphocyte-conditioned medium prepared according to

published methods (9), added in 10% volume, and 10% conditionedmedium from the Mo cell line generously provided by Dr. David Golde,UCLA Medical School (11). The use of both sources of conditionedmedium simultaneously allowed detection of the greatest number ofgranulocyte colony-forming cells in a harvest, compared to the resultsusing Mo-cell-conditioned medium or phytohemagglutinin-lymphocyte-

conditioned medium alone.Hematological and Histochemical Staining. Nonadherent cells re

moved from long-term bone marrow cultures were stained with Wright's-

Giemsa, and differential cell counts were performed according to published methods (13). Other cell smears were tested for superoxide-generating capacity by the nitroblue-tetrazolium dye reduction test (13,18) or were tested for specific myeloid esterase (ASD-chloroacetatesubstrate specific) (18), nonspecific esterase («-naphthol esterase sub

strate) (18, 37), myeloperoxidase (18), lysozyme (18), toluidine bluemetachromasia of basophil-mast cells (19), and benzidine stain for he

moglobin (19). Hematopoietic colonies forming in agar were staineddirectly in the combined method for specific and nonspecific esterase,and Luxol fast blue for eosinophils (37).

In some experiments, nonadherent cells from long-term marrow cultures were tested for antigen associated with the differentiating T-

Table 1Reactivity of mouse monoclonal antibodies with human hematopoietic cells measured by fluorescence-

activated cell sorter analysis and immunofluorescence

Lymphocytes Polymorpho-nuclear leuko-

NK T B Monocytes cytes GM-CFUc BFUe CFUe CFU-GEMM

B1 - - + - - ___B2M + + ++ + +++ +J5 (CALLA) --+"- + ___J2 --"+"- - -la - - + + - + + +

" Not expressed on surface immunoglobin-positive B-cells (26,41).J2 antigen (M, 26,000 glycoprotein) is not expressed on resting T-cells but is expressed on T-cells

following activation (23).

CANCER RESEARCH VOL. 45 FEBRUARY 1985

759

Research. on September 21, 2015. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from

HEMATOPOIETIC EFFECTS OF MONOCLONAL ANTIBODIES

lymphocyte pathway. For these studies, the phenotype of nonadherentcells was determined by indirect immunofluorescence and flow cytome-

try. The percentage of cells which reacted with monoclonal antibody T4or T8 was calculated according to published methods (23, 41, 42).

Assays for Nonhematopoietic Cellular Components of Long-Term

Bone Marrow Cultures. Culture flasks were examined weekly for (a) thepercentage of cell surface area covered by adherent cells (13), (£>)thenumber of fat-containing adipocytes per flask (15), and (c) also thenumber of adipocytes per high-power field (15). Freshly treated cells

prior to establishment of the above marrow cultures in several experiments were seeded onto plastic Retri dishes in Dulbecco's modified

Eagle's medium containing 10% fetal calf serum. Fibroblast colonies

derived from single cells were scored 7 and 14 days later (16).The number of "cobblestone" areas (10, 13, 20) containing >50 cells

detected per culture flask was scored each week. These areas havebeen associated with the adherent component of hematopoiesis including those stem cells with a high self-renewal capacity in mouse long-

term marrow cultures (30) and with those hematopoietic cells that formcolonies with the greatest in vitro self-renewal following replating tosecond-assay cultures (7).

RESULTS

Effects of Monoclonal Antibody Treatment on Hematopoiesis in Long-Term Bone Marrow Cultures. Individual marrow specimens were treated with each monoclonal reagent usinga protocol identical to that used for clinical autologous marrowtransplantation as described under "Materials and Methods."

Weekly and cumulative production of nonadherent cells, weeklyand cumulative production of both 7- and 14-day GM-CFUc, and

the morphological appearance of the adherent cell compartmentwere quantitated for each experiment. In some experiments,BFUe and CFU-GEMM were quantified. A summary of the resultsin a representative experiment with each monoclonal antibody isshown in Table 2. Treatment with antibody to B1, J5, or J2 didnot significantly decrease CFU-GEMM, BFUe, CFU-mega, or GM-CFUc from marrow. Long-term bone marrow cultures establishedfrom these marrow samples produced CFU-GEMM, BFUe, CFU-mega, and GM-CFUc in vitro for several weeks (Table 2). Treatment of marrow with anti-la plus complement significantly re

duced CFU-GEMM, BFUe, CFU-mega, and GM-CFUc, while fewGM cluster-forming cells survived (Table 2). In long-term marrow

culture, however, this marrow reconstituted in vitro and generated CFU-GEMM, BFUe, CFU-mega, and GM-CFUc over the

next 2 to 4 weeks (Table 2). In contrast, marrow treated withantibody to B2M was devoid of detectable hematopoietic progenitor cells and did not produce any detectable progenitor cellsin vitro in the next 4 weeks (Table 2).

Long-term marrow cultures were next set up with marrow that

had been treated with each monoclonal antibody and complement and studied in greater detail. In one study, marrow wastreated with antibody to B1 alone, B1 + complement, J5 alone,J5 + complement, or complement alone. Several of the treatments reduced after 30 days the weekly production of nonadherent cells when compared to untreated marrow (Chart ~\A).

There was, however, no detectable decrease in cumulative non-

adherent cell production compared to control untreated marrow(Chart 16). There was also no detectable effect on cumulativeproduction of GM-CFUc scored as 7-day GM-CFUc colonies(Chart 2A) or 14-day GM-CFUc colonies (Chart 26). These data

with B1 and J5 confirm the data in Table 2.In contrast, in a second experiment with another marrow

specimen, treatment with monoclonal antibody to B2M andcomplement resulted in rapid and significant decrease in theweekly (Chart 3A) and cumulative (Chart 3B) production ofnonadherent cells. The decrease was detected within 2 weeksafter establishment of the cultures. Treatment with anti-B2M

alone or complement alone did not alter cell production. Asanother control for this experiment, culture of marrow treatedwith antibody to B1 + complement showed no decrease in cellproduction (Chart 3), confirming the data in Charts 1 and 2 by asecond experiment. Anti-B2M + complement treatment significantly decreased production of Gm-CFUc scored as Day 7

colonies (Chart 4A) or Day 14 colonies (Chart 46). Controlcultures or anti-B1 + complement-treated cultures producedGM-CFUc for over 8 weeks (Chart 4).

Effect of Monoclonal Antibody and Complement Treatmenton CFU-F Adipocyte Colony Formation, and Adherent Cell

Layer Function by Human Bone Marrow Cells in Vitro. An

Table 2Effect of monoclonal antibody and complement treatment of human marrow on removal of CFU-GEMM, BFUe, and GM-CFUc and a return of those cells in long-term

bone marrow cultures

immediate* posttreatmenfColonies/2 x

Experiment1231617TreatmentNone

Anti-B1 +complementNone

J5 +complementNone

Anti-la +complementNone

J2 +complementNone

Anti-B2M + complementCFU-GEMM/

10533345

04

340BFUe/105877178

3673

0NT=

NT230CFU-

mega/1058

1113

109

0NT

NT80GM-CFUc/GM

cluster/105110/31

97/2790/51

107/4093/23

0/16121/20

110/1891/27

0/13CFU-

GEMM2

(4f3(4)3

(12)2(12)3

(2)4(2)4

(5)3(5)5

(3)0 (1)BFUe18

(2)8(3)7(12)

18(12)19

(3)3(3)16

(5)8(5)18

(3)0 (1)CFU-mega3

(4)6(4)11

(12)10(12)6

(2)11(2)8

(5)10(5)6

(3)0 (1)GM-CFUc/GM

cluster151/21

(4)119/27(4)93/21

(12)110/17(12)38/14

(4)21/18(4)81/17

(5)71/21(5)94/18

(10)0/3 (4)

" GM-CFUc scored on Days 7 and 14 (results presented for Day 14) (13), BFUe, CFU-mega, and CFU-GEMM scored on Day 14 (9). Nonadherent cells from long-term

bone marrow cultures (13).6 Numbers in parentheses, week tested.c NT, not tested.

CANCER RESEARCH VOL. 45 FEBRUARY 1985

760

Research. on September 21, 2015. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from

HEMATOPOIETIC EFFECTS OF MONOCLONAL ANTIBODIES

20 40 60

DAYS IN CULTURE

80

DAYS IN CULTURE

Chart 1. A, weekly production of nonadherent cells by flasks of a single specimen of marrow prepared as described under "Materials and Methods' and treatedwith monoclonal antibody to B1, anti-B1 + complement (C'). complement alone.

anti-J5 alone, anti-J5 + complement, or untreated control. Results are the meannonadherent cell harvest per flask for at least 4 flasks per group, standardized to4 x 107 cells per flask after treatment. All treatments were one cycle. B, cumulative

nonadherent cell production of experiment shown in A over first 85 days ofexperiment. Results are as described in legend to A

analysis was next made of the effect of each monoclonal antibody and complement treatment on the number of >10 celladipocyte colonies in the adherent layer of the cultures at 4, 8,and 12 weeks; the percentage of surface area of the flaskcovered by adherent cells at 4, 8, and 12 weeks; and theformation of adherent fibroblast-like colonies of >50 cells byserial dilutions of single-cell marrow suspensions at 1 x 106, 1x 105, or 1 x 10" cells/4.0-ml Petri dish scoring at Days 7 or 14.

In control cultures, we scored 50 ±7 adipocyte colonies atWeek 4, 73 ±9 at Week 8, and 115 ±18 at Week 12 (mean ±S.E. of at least 3 flasks at each point). Cultures derived frommarrow treated with antibody to B1, J5, J2, or la with or withoutcomplement treatment showed no detectable decrease in thisparameter. In contrast, cultures derived from marrow treatedwith anti-B2M + complement (but not B2M alone) showed 3 ±

1 adipocyte colonies/flask at 4 weeks and no detectable adipocyte colonies at Week 8 or 12, a significant decrease.

In control cultures, we scored 7 ±1 CFU-F/105 cells plated to

Petri dishes in 4.0 ml complete medium, scoring at 7 days; and18 ±3 colonies/dish scoring at Day 14. Only colonies of >50

20 40 60

DAYSINCULTURE:

80

10'

101

B

101

ControlAnti-B1Anti-BI-t-C'C'

J5J5K'

20 40 60 80

DAYS IN CULTURE

Chart 2. A, Cumulative GM-CFUc per 1 x 105 nonadherent cells harvested eachweek, shown in Chart 1. Cells were plated in Mo cell-conditioned medium andphytohemagglutinin-lymphocyte-conditioned medium as a source of colony-stimulating factor as described under "Materials and Methods." Results are the mean ofat least 3 plates at 1 x 105 cells per plate per group. S.E. was <10% of the mean.GM-CFUc cultures were scored on Day 7. B, cumulative GM-CFUc per 1 x 10s

nonadherent cells harvested each week, as shown in Chart Mi. Results arepresented as described in the legend to A but scoring occurred on Day 14. C'.

complement.

cells/colony were counted. In contrast, cultures derived frommarrow treated with anti-B2M + complement showed a detectable decrease at Day 7 (3 ±1 per 106 cells plated) and Day 14

(1, O, O colonies detected in triplicate plates).The percentage of surface area covered by adherent cells was

100% at 4 weeks for control and all monoclonal antibody groupsexcept those treated with anti-B2M + complement. Here, 8% of

the surface was covered.Effect of Monoclonal Antibody and Complement Treatment

on "Cobblestone Island" Formation by Human Marrow in

Continuous Marrow Cultures. The adherent cell layer of culturesderived from marrow treated with antibody to B2M and complement showed significant decrease in "cobblestone" areas by 6

weeks (<1/flask; control, 50 to 63/flask) and no detectableproduction of hematopoietic cells past 6 weeks. In other monoclonal antibody-treated cultures with or without complement

treatment, there was no decrease compared to control culturesin the number of "cobblestone" areas and cultures continued to

CANCER RESEARCH VOL. 45 FEBRUARY 1985

761

Research. on September 21, 2015. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from

HEMATOPOIETIC EFFECTS OF MONOCLONAL ANTIBODIES

I£

ÃŽ

k,

15S

£^

2CT 40

DAYS IN CULTURE

60

10"

Contro!*—«Anti-BttC'

a—AC1

«—»Anti-B2Mo—OAnti-B2M+C'

20 40

DAYS IN CULTURE

60

Charts. A, weekly production of nonadherent cells from flasks of a singlespecimen of marrow prepared as described under 'Materials and Methods" treatedwith monoclonal antibody to B1 + complement (C'), complement alone, antibody

to B2M alone, or anti-B2M + complement. Results are presented as described inthe legend to Chart 1A In this experiment, there were 2 x 107 cells/flask after

treatment. The line for O stops early due to the absence of detectable cells in theflask at that time. B, cumulative nonadherent cell production for experiment shownin A over first 58 days of experiment. Results are presented as described in thelegend to Chart 1S.

produce over 10" cells/flask after 6 weeks (Charts 3 and 4). No

significant toxic effect on cumulative cell production or GM-CFUc

production was detected using one cycle treatment with antibodyto J2 (Chart 5) or HLA-DR (la) (Chart 6) (Table 2).

Effects on Hematopoiesis in Long-Term Marrow Cultures

of Multiple Cycles of Treatment with One Monoclonal Antibody or Sequential Treatment Cycles with Different Monoclonal Antibodies. To determine whether sequential treatmentwith 3 different monoclonal antibodies would produce a toxiceffect on production of hematopoietic stem cells in long-term

marrow cultures that was not observed using one antibody alone,a single specimen of >1010 human marrow cells was treated

sequentially with each of 3 monoclonal antibodies and comple-

IO20 40

DAYS IN CULTURE

60

104

10»

•—•Control<—»Anli-BUC1

i—»Anh-B2Mo--<Anti-B2M + C'

If

§ 10Q 20 40

DAYS IN CULTURE

60

Chart 4. A, cumulative Day 7 scored GM-CFUc from the nonadherent cellsharvested from the experiment shown in Chart 3. Results are presented asdescribed in the legend to Chart 2A. B, cumulative GM-CFUc per nonadherent cellharvest. Results are as described in the legend to Chart 2B scoring on Day 14.C', complement.

ment (B1, J2, and J5) prior to establishment of long-term marrow

cultures. As shown in Table 3, this treatment did not detectablydepress cumulative cell production or GM-CFUc production bycultures initiated with 2.5 x 107 nucleated cells/flask adjusted at

the last treatment compared to control cultures. The lack oftoxicity was comparable to that of cultures treated with onemonoclonal antibody and complement cycle as shown above.These data indicate that 3 treatments are not necessarily moretoxic than one treatment using monoclonal antibodies to antigensthat are not on normal hematopoietic stem cells.

To determine whether 3 treatments with a monoclonal antibody directed toward an antigen that is present on normalhematopoietic stem cells would have a greater toxic effectcompared to one treatment, several experiments were carriedout with monoclonal antibody to HLA-DR (la). First, 3 successive

treatments with monoclonal antibody and complement werecarried out, and the cultures were established. As shown inTable 3, there were no detectable GM-CFUc surviving this treatment regimen compared to some survival of GM cluster-forming

CANCER RESEARCH VOL. 45 FEBRUARY 1985

762

Research. on September 21, 2015. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from

HEMATOPOIETIC EFFECTS OF MONOCLONAL ANTIBODIES

S 10'•Conn*OJ2-C'

•C'

CS O 20 40 60 80

Oars IN CULTUFf

Charts. A, weekly production of nonadherent cells by flasks of a single specimen of human marrow prepared as described under "Materials and Methods"treated with monoclonal antibody to J2 + complement (C'), complement alone, or

untreated control. Results are presented as described in the legend to Chart i A.Here, control cultures had 5 x 107cells while anti-J2 + complement-treated culturesstarted with 3 x 107 cells. B, cumulative nonadherent cell production per flask by

the cell harvest shown in A over the first 85 days of the experiment. Results areas described in the legend to Chart 1B. Control cultures had twice as many startingcells in this experiment; therefore, the line comparison is between A and •

cells after one cycle of treatment (Tables 2 and 4). Thus, a 3-cycle treatment was more effective at killing GM cluster-forming

cells than was one treatment cycle. The inoculum used toestablish the cultures after 3 successive monoclonal antibodytreatments was standardized to 2.5 x 107 nucleated cells/flask

and contained no detectable GM-CFUc (Tables 3 and 4). Non-adherent cells removed after 1 week from long-term marrowcultures established from this marrow contained 3 to 5 per 105GM-CFUc and 14 to 15 GM cluster-forming cells per 10s cellsassayed (Table 3). At 2 weeks, these GM-CFUc were inhibitedby antibody to la (Table 3). The cultures from 3-cycle anti-la +complement-treated marrow did not produce as many cells as

did control cultures (Chart 7). The appearance of the adherentcompartment in anti-la-treated cultures was quantitated weekly.

As shown in Fig. 1, numerous cobblestone areas (50 to 76/flask)were scored at Week 8 in cultures derived from marrow thatwas treated with 3 sequential cycles of antibody to la andcomplement in the experiment in Table 3. These data estabishthat the cells forming cobblestone areas were not removed byup to 3 cycles of anti-HLA-DR monoclonal antibody and comple

ment treatment. Cell proliferation continued in these culturesalthough at a decreased level compared to untreated controls(Chart 7; Fig. 1). In 2 separate experiments (Tables 3 and 4),there was a cumulative decrease in GM-CFUc production andtotal cell production in long-term marrow cultures following 3

cycles of treatment with monoclonal antibody to la and complement.

Mrs IN CULTURE

•Control•C'

30 50

DAYS IN CULTURE

70

Chart 6. A, weekly production of nonadherent cells by flasks of a single specimen of human marrow prepared as described under "Materials and Methods,"treated with monoclonal antibody to HLA-DR (la) + complement (C'), complementalone, or untreated control. Results are for control cultures initiated with 7 x 107nucleated cells/flask and antibody- or complement-treated cultures started at 3 x107 nucleated cells/flask. The true comparison is between •and A. B, cumulative

nonadherent cell production by flasks shown in A for the first 75 days of theexperiment. Results are as described in the legend to Chart 1B. Control cultureshad twice as many cells per flask at initiation.

DISCUSSION

Monoclonal antibody reagents currently used to prepare forautologous transplant of marrow from patients with CALLA-positive acute lymphoblastic leukemia (42), T-cell leukemia (50),

or myeloid leukemia (2) have been tested in assays for killing ofnormal multipotential stem cells (CFU-GEMM) and committedhematopoietic progenitor cells BFUe and GM-CFUc (9,21). Monoclonal antibodies prepared for specificity against lymphocyte-

specific lineage cells have usually been found to spare cellsforming CFU-GEMM and GM-CFUc in vitro (21). Thus, the suc

cessful engraftment and tumor cell reduction in several patientswhose whole marrow was purged with J5 + complement mightbe solely attributable to selective sparing of the donor stem cellsthat successfully seeded the host.

The role if any in vivo of donor marrow nonhematopoietic cellsthat are known to facilitate self-renewal of hematopoietic pro

genitors in vitro (13) is not well understood. Recent data show adonor cell origin of some fibroblastic cells that appear in long-

term bone marrow cultures derived from the marrow of successfully engrafted allogeneic transplant recipients (27, 46). Thesedata have stressed the need to understand the possible phys-

CANCER RESEARCH VOL. 45 FEBRUARY 1985

763

Research. on September 21, 2015. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from

Effect of re

HEMATOPOIETIC EFFECTS OF MONOCLONAL ANTIBODIES

Table3i/emenior

hematopoiesisin long-term marrow cultures

Experiment Treatmentwith1

NothingJ2 + complement, J5 + complement,

anti-B1 + complement(sequential)2

NothingAnti-la + complement (3 cycles)Anti-la -t-complement (1 cycle) after

harvest from long-term bone marrowcultures*

NA, nonadherent; NT, not tested.0 Mean ±S.E.c Also <1 CFU-GEMM/2 x 105;<1 CFU-mega/2 xInitial

posttreat-mentmarrow

GM-CFUc/GMcells

(Day14)98±6/31 ±2"

79 ±5/41 ±371

±8/18±30C/0C10s;

and <1 BFUe/2Hematopoietic

longevityGM-CFUc/GM

cluster/1x 10s(Day 14) NA"cellsWk

1 Wk28±1/3±1

17 ±17 ±1/20 ±18±15

±1/5 ±2 10 ±3 ±.5/14±2 8±NTx

105.2/5

±21/18±11/8

±22/6 ±10/0Wk36

±1/31 ±23±1/17±29±

1/8 ±2NTNTCumulative

NAGM-CFUc/flaskDay

71017

14705210

4117Day

145210

49214210

3900Wk

production of>104NAcells/flask12

1113

9

iological role of nonhematopoietic cells in the donor marrow.There are as yet no specific markers for identifying each of themorphological categories of these nonhematopoietic cells of themarrow.

In the present studies, we have quantified the ability of marrowto establish a functioning long-term bone marrow culture (13)

after treatment with monoclonal antibody and complement.There was no detectable toxicity of monoclonal antibodies B1,J2, or J5 alone or in combination to the establishment of thelong-term marrow culture system. In this regard, CALLA hasbeen found on cultured marrow fibroblasts in vitro (4); however,no CALLA was detected by immunoperoxidase staining offreshly removed human hematopoietic tissues (31). Thus, CALLAmay be expressed during tissue culture on some nonhematopoietic cells that normally do not express this antigen in situ. Ourdata showing no detectable damage to long-term marrow cul

tures by prior J5 treatment support clinical transplant data whereJ5 treatment of marrow resulted in successful engraftment (42).In contrast, anti-B2M and complement treatment was severelytoxic to the establishment of long-term bone marrow cultures as

evidenced by rapid decline of hematopoiesis in culture flasksstandardized to control cultures for a similar number of viablecells posttreatment. The mechanism of the toxicity is not knownbut may be attributable to death of nonhematopoietic cells inaddition to hematopoietic progenitor cells, since both populationswere reduced as measured by colony assays.

A monoclonal antibody that is selectively toxic for nonhematopoietic cells was not available for the present studies and willbe helpful if available in the future. The data with anti-HLA-DR(anti-la) and complement-treated marrow are of interest with

regard to a previously published report claiming that humanpluripotential stem cells do not contain (la) HLA-DR (32). Theprevious study with a 2-step culture system established long-

term marrow cultures using first an inoculum of untreated marrow and then a "recharged" addition to the flasks 3 to 4 weeks

later of a second inoculum of anti-la and complement-treatedmarrow that was free of detectable GM-CFUc or BFUe (32).Since the anti-la and complement-treated cultures generatedGM-CFUc and BFUe with detectable surface la, it was concluded

that primitive human multipotent stem cells had escaped antibody and complement killing and therefore were la negative (32).It is now known that the first inoculum is itself a vast reserve of

10*O 10 20 30 40 50 60

DAYS IN CULTURE

70

108

10

106

, -O- —-O O —-o- _ o- —-o —O —-O- ~*

•—•Control

o--o Anti-Io + C'

13 Cycles)

4,62 x 10 Cells/Flask4.21 x W7Cells/Flask

10 30 40 50 60 70

DAYS IN CULTUREChart 7. A, weekly production of nonadherentcells by flasks of a single marrow

specimen treated as described under "Materials and Methods" with 3 cycles ofantibody to HLA-DR(la)+ complement (C') and standardizedfor 4 x 107nucleated

cells/flask. 6, cumulative nonadherent cell production by cultures established asdescribed in the legend to A.

CANCER RESEARCH VOL. 45 FEBRUARY 1985

764

Research. on September 21, 2015. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from

HEMATOPOIETIC EFFECTS OF MONOCLONAL ANTIBODIES

«1oiiS</>1.5IÕ¿duster-fa^i3AÕ•BSSss•*

oco^3*H

5iS•51«dIi§1'S1i1V

co81I0OtE1I0_>,iVi0)coVi81.Cmi.5co!Co•v<5"Si

•Q•oS1^jU-90B^ìS1mIi9ut

Ãœr->,SCo?S

S5"9Èu_

O8ÃŽ,-,.CM

IOoCOìoOo(0OS1CO80sBozMgIiliiL°,+1

-Hi-COCMCM

T-+1

+1+1COcoÇJi_ÇO_l—

J—CO i~ ZZ+1

+1r-r*CMO+1inT-

O OO5" o" o"o"+\•^CMCO+1Es.0.0,T-

0 00+1inCM+1í^o'o'o"KNo

in H t-cÑTí"+1

+1coco00+1

mle&Slo000+1i81",-j.CM1CDCOÛo00oCMCDO*"•èIoI+1^"o"+1coCM+100

O

00IOco+1

1-

^i+1i-CO8"+l

+ico"

o"CM+1inCM+1

E°-i-O^.linco+1^"5CO+ICMCM"^"+1

+1COCM

00in+1

inc-Tcò"

•H+1I

°8 IIì~°£~

|~

co oco0Z.O.CM

COCO CMCM CMSIiQc31o1I.£li.B1iI2§COv>coInr^P<DÕedC0)S£3CO0

COÃŽ3oVi

0)SOoo5i(Do*1Iiii0)

11aA

¿~0

«|Z

0|1

s118«1•sì•¿=

0) "3e/) O•*-fc

co SHU)-Coilsu

1

hematopoietic stem cells (7, 8,10,13, 39).In the present studies, one cycle of la + complement treatment

did not remove all GM cluster-forming cells, while 3 cycles of

treatment that did remove all proliferative hematopoietic cellsresulted in a significantly decreased cumulative hematopoieticcell production in long-term culture. Whether the decrease in

hematopoiesis in vitro after 3 cycles is attributable to selectiveremoval of other HLA-DR-positive nonhematopoietic cells suchas endothelial cells (38) or to efficient removal of enough CFU-

GEMM and possibly the human pluripotential progenitors is notyet known. Three cycles of treatment itself did not account forthe toxicity since cultures from marrow treated with J5, J2, andanti-B1 in succession were not detectably altered. The return ofla-positive GM-CFUc in long-term culture after 3 cycles of treatment with anti-la + complement suggests that la-positive cellsthat form GM-CFUc may evolve from la-negative treated marrow

in culture. However, the possibility exists that the data areexplained by: (a) very small numbers of residual la-positive true

human CPUs which can repopulate the whole in vitro system (2,40); or (b) that human la-positive CFUs could be protected fromanti-la and complement killing due to residence within a nonhe

matopoietic protective cell membrane that is itself la negative.Adventitial reticular cells that line the marrow sinuses could besuch protective cells and could function as do the murine thymic"nurse" cells described by Werkele (53) and recently identified in

human thymus (51). Further studies will be required to resolvethese possibilities.

ACKNOWLEDGMENTS

We thank Thomas Novak for technical assistance; Drs. Thomas Thornhill,Clement Sledge. William Thomas, and Frederick Ewald for obtaining marrowspecimens; and Briana Walker for typing the manuscript.

REFERENCES

1. Andrews, R. G., Torok-Storb, B., and Bernstein, I. D. Myeloid-associated

differentiation antigens on stem cells and their progeny identified by monoclonalantibodies. Blood, 62: 124-132,1983.

2. Ball, E. D., Bernier, G. R., Cornwall. G. G., Ill, Mclntyre, D. R., O'Donnell. J.

F., and Farger. M. W. Monoclonal antibodies to myeloid differentiation antigens:in vivo studies of those patients with acute myelogenous leukemia. Blood 62.1203-1210, 1983.

3. Bast, R. C.. Jr., Ritz, J., Upton, J. M., Feeney, M., Sallan, S. E., Nathan. D.G., and Schlossman, S. F. Elimination of leukemic cells from human bonemarrow using monoclonal antibody and complement. Cancer Res., 43: 1389-

1394, 1983.4. Braun, M. P., Martin, P. J., Ledbetter, J. A., and Hansen, J. A. Granulocytes

and cultured human fibroblasts express common acute lymphoblastic leu-kemic-associated antigens. Blood, 67: 718-725. 1983.

5. Capone, P. M., Rapsiderio, L. D., Groghan, G. A., and Chu, T. M. Experimentaltumoricidal effects of monoclonal antibody against solid breast tumors. Proc.Nati. Acad. Sci. USA, 80: 7328-7332, 1983.

6. Castro-Malaspina, H., Gay, R. E., Jharwar, S. C.. Hamilton, J. A., Chiarier, D.R., Meyers, P. A., Gay, S., and Moore, M. A. S. Characteristics of bone marrowfibroblast colony forming cells (CFUf) and their progeny in patients withmyeloproliferative disorders. Blood, 59. 1046-1054, 1982.

7. Coulombel, L., Eaves, A. C., and Eaves, C. J. Enzymatic treatment of longterm human marrow cultures reveals the preferential location of primitivehematopoietic progenitors in the adherent layer. Blood, 62: 291-297, 1983.

8. Coulombel, L., Kalousek, D. K., Eaves, C. J., Cupta, C. M., and Eaves, A. C.Long term marrow culture reveals chromosomally normal hematopoietic progenitor cells in patients with Philadelphia chromosome-positive chronic mye-logenus leukemia. N. Engl. J. Med., 208: 1493-1498, 1983.

9. Fauser, A. A., and Messner, H. A. Identification of megakaryocyte, macrophgeand eosinophils in colonies of human bone marrow containing neutrophilicgranulocytes and erythroblasts. Blood, 53: 1023-1027, 1979.

10. Gartner, S., and Kaplan, H. S. Long term culture of human bone marrow cells.Proc. Nati. Acad. Sci. USA, 77: 4756-4758, 1980.

CANCER RESEARCH VOL. 45 FEBRUARY 1985

765

Research. on September 21, 2015. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from

HEMATOPOIETIC EFFECTS OF MONOCLONAL ANTIBODIES

11. Golde, 0. W., Bersch, N., Quan,S. G., and Lusis, A. J. Productionof erythroid-potentiating activity by a human T-lymphocyte cell line. Proc. Nati. Acad. Sci.USA. 77: 593-596,1980.

12. Greaves, M. F., Harris, G., Newman, R. A., Sutherland, D. R., Ritter, M. A.,and Ritz, J. Selectiveexpression of the common acute lymphoblasticleukemia(GP100)antigen on immature lymphoid cells and thier malignantcounterparts.Blood, 6:628-638, 1983.

13. Greenberg, H. M., Newburger, P. E., Parker, L. M., and Greenberger, J. S.Generation of physiologically normal human peripheral blood granulocytes incontinuous bone marrow culture. Blood, 58: 724-731,1981.

14. Greenberg, H. M., Parker, L. M., Newburger, P. E., Said, J., Cohen, G. I., andGreenberger, J. S. Corticosteroid dependence of continuous hemopoiesis invitro with murine or human bone marrow. In: R. Neth and R. C. Gallo (eds.).Modern Trends in Human Leukemia IV, Vol. 26, pp. 289-293. Berlin: SpringerVerlag, 1981.

15. Greenberger, J. S. Sensitivity of corticosteroid-dependent, insulin-resistantlipogenesis in marrow preadipocytes of mutation diabetic-obese mice. Nature(Lond.),275: 752-754, 1978.

16. Greenberger, J. S. Corticosteroid dependent differentiation of human bonemarrow preadipocytes. In Vitro (Rockville),15: 823-828,1979.

17. Greenberger,J. S., Davisson, P. B., and Gans, P. J. Murine sarcoma virusesblock corticosteroid-dependent differentiation of bone marrow preadipocytesassociated with long-term in vitro hemopoiesis.Virology, 95: 317-333,1979.

18. Greenberger, J. S., Newburger, P. E., Karpas, A., and Moloney, W. C.Constitutive and ¡nducibtegranulocyte-macrophage functions in mouse, ratand human myeloid leukemia-derivedcontinuous tissue culture lines. CancerRes., 38: 3340-3348, 1978.

19. Greenberger, J. S., Sakakeeny, M. A., Humphries, K. C., Eaves, C. G., andEckner, R. J. Demonstration of permanent factor-dependent multipotential(erythroid/neutrophil/basophil)hematopoietic progenitor cell lines. Proc. Nati.Acad. Sci. USA,80: 2931-2935,1983.

20. Greenberger, J. S., Sakakeeny, M. A., and Parker, L. M. In vitro proliferationof hematopoietic stem cells in long term marrow cultures: principles in mouseapplied to man. Exp. Hematol., 7 (Suppl. 5V 135-148,1979.

21. Griffin, J. B., Larcom, P., and Schlossman, S. F. Use of surface markers toidentify a subset of acute myelomonocyticcells with progenitor cell properties.Blood, 62: 1300-1303, 1983.

22. Hancock, L. W., Zola, H., and Atkins, R. C. Antigenic heterogeneity of humanmononuclear phagocytes: immunohistotopic analysis using monoclonal antibodies. Blood, 62: 1271-1279,1983.

23. Hercend, T., Nadler, L. M.. Pesando, J. M., Reinherz, E. L., Schlossman, S.F., and Ritz, J. Expression of a 26,000 dalton glycoprotein on activated humanT cells. Cell. Immunol.,64:192-199,1981.

24. Hoang, T., Gilmore, D., Metcalf, D., Cobbold, S., Wetz, S., Clark, M., Furth,M., and Waldman, H. Separation of hematopoietic cells from adult mousemarrow by use of monoclonalantibodies. Blood, 61: 580-588, 1983.

25. Hoffman, G. M., Yanagihara,E., Byrne, B., Billing, R., Baird, S., Frisman, D.,and Taylor, C. R. Analysisof B-cell antigens in normal reactive lymphoidtissueusing four B-cell monoclonal antibodies. Blood, 62: 775-783,1983.

26. Hokland, P., Rosenthal, P., Griffin, J. D., Nadler, L. M., Daley,J., Hokland, M.,Schlossman, S. F., and Ritz, J. Purification and characterization of fetalhematopoietic cells that express the common acute lymphoblastic leukemiaantigen (CALLA).J. Exp. Med., Õ57:114-129, 1983.

27. Keating, A., Singer, J. W., Killen, P. D., Striker, G. E., Salo, A. C., Sanders,J.,Thomas, E. D., Thoming, P., and Fialkow, P. J. Donor origin of the In vitrohematopoietic macroenvironment following marrow transplantation in man.Nature (Lond.),298: 280-282, 1982.

28. Knowles, D. M., II, Tolidjian, B., Marboe, C. C.. Halper, J. P., Azzo, W., andWang, C. Y. A new human B-lymphocyte surface antigen (BL2) detectable bya hybridoma monoclonalantibody: distribution on benignand malignantlymphoid cells. Blood, 62: 191-199, 1983.

29. Mach,J. P.,Chatel,J-F., Lumbroso,J-D., Buchegger,F., Formi,M., Rittschard,J., Berche, C., Pouillard, J. Y., Carrel, S., Helyn, M., Steplewski, Z., andKoprowski, H. Tumor localization in patients by radiolabelled monoclonalantibodies against colon carcinoma. Cancer Res., 43: 5593-5600,1983.

30. Mauch, P., Greenberger,J. S., Botnick, L. E., Hannon, E. C., and Hellman,S.Evidencefor structured variation in self-renewalcapacity within long-termbonemarrow cultures. Proc. Nati. Acad. Sci. USA, 77: 2927-2930, 1980.

31. Metzgar, R. S., Bonowitz, M. J., Jones, N. H., and Dowell, B. L. Distributionof common acute lymphoblastic leukemiaantigen in nonhematopoietictissues.J. Exp. Med., 754: 1249-1254, 1981.

32. Moore, M. A. S., Broxmeyer, H. E., Sheridan,A. P. C., Meyers, P. A., Jacobsen,N., and Winchester, R. J. Continuous human bone marrow culture, la antigen

characterization of possible pluripotential stem cells. Blood, 55: 682-690,1980.

33. Nadler, L. M., Stashenko, P., Hardy, R., Pesando, J. M., Yunis, E. J., andSchlossman, S. F. Monoclonal antibodies defining serologically distinct HLA-D/DR related la antigens in man. Hum. Immunol., 7: 77-81,1981.

34. Newburger, P. E., Robinson, J. M., Pryzwansky, K. B., Rosoff, P. M., Greenberger, J. S., and Tauber, A. I. Neutrophil dysfunction with giant granulesanddefective activation of the respiratory burst. Blood, 67: 1247-1257,1983.

35. Patt, H. M., Maloney, M. A., and Flannery, M. L. Hematopoietic Microenviron-ment transfer by stromal fibroblasts derived from bone marrow varying incellularity. Exp. Hematol., 10: 738-742, 1982.

36. Perri, R. T., Royston, I., Lebier, T. W., and Kay, N. E. Chronic lymphocyticleukemia progenitor cells carry the antigen T65, BA-1 and la. Blood, 67: 871-884, 1983.

37. Phillips, P. G., Chikkapa, G., and Brinson, P. S. A triple stain to evaluatemonocyte, neutrophil and eosinophil proliferation in soft agar cultures. Exp.Hematol., 77: 10-17,1983.

38. Pober, J. S., Collins, T., Gimbrone, M. A., Cotran, R. S., Gitler, J. D., Fiers,W., Clayberger,C., Krensky, A. M., Burakoff, S. J., and Reiss, C. S. Lymphocytes recognize human vascular endothelial and dermal fibroblast la antigeninduced by recombinant immune interferon. Nature (Lond.), 305: 726-729,1983.

39. Potter, C. G., Rowell, A. C., and Weatherall,P. J. Continuous long-termcultureof human bone marrow. Clin. Lab. Hematol., 3: 245-255,1981.

40. Reincke, V., Hannon, E. C., Rosenblatt, M., and Hellman, S. Proliferativecapacity of murine hematopoietic stem cells in vitro. Science(Wash. DC),275:1619-1622,1982.

41. Ritz, J., Pesando, J. M., Notis-McConarty, J., Lazarus, H., and Schlossman,S. F. A monoclonal antibody to human acute lymphoblastic leukemia antigen.Nature (Lond.), 283: 583-585, 1980.

42. Ritz, J., Sallan,S. E., Bast, R. C., Jr., Upton, J. M., Clavell, L. A., Feeney,M.,Hercend,T., Nathan, D. G., and Schlossman, S. F. Autotogous bone marrowtransplantation in CALLA-positive acute lymphoblastic leukemia after in vitrotreatment with J5 monoclonal antibody and complement. Lancet, 2: 60-63,1982.

43. Roodman, G. D., Horadam, V. W., and Wright, T. L. Distribution of erythroidcolony formation by autologous marrow adherent cells from patients with theanemiaof chronic disease. Blood, 62: 406-412, 1983.

44. Sakakeeny, M. A., and Greenberger, J. S. Longevity of granulopoiesis incorticosteroid-supplemented continuous bone marrow culture varies significantly between 28 mouse strains and outbred stocks. J. Nati. Cancer Inst.,68:305-317,1982.

45. Singer, J. W., Keating, A., and Fialkow, P. J. Evidencesuggesting a commonprogenitor for hematopoietic and stromal cells. Exp. Hematol., 77 (Suppl. 14):4, 1983.

46. Singer, J. W., Keating, A., Ramberg, R., McGuttin, R., Sanders, J. E., Sale,G., Fialkow, P. J., and Thomas, E. D. Long term stable hematopoieticchimer-ism following marrow transplantation for acute lymphoblastic leukemia:a casereport with in vitro marrow culture studies. Blood, 62: 869-872,1983.

47. Stashenko, P., Nadler, L. M., Hardy, R., and Schlossman, S. F. Characterization of a human B lymphocyte-specificantigen. J. Immunol., 725:1678-1685,1980.

48. Takvorian, T., Parker, L. M., Hochberg, F. H., and Caneltos,G. P. Autologousbone marrow transplantation: host effects of high dose BCNU.J. Clin. Oncol.,7:610-620,1983.

49. Toun, I., and Lowenberg, B. No stimulative effect of adipocytes on hemato-poiesis in long-term human bone marrow cultures. Blood, 67: 770-774,1983.

50. Vallera,B. A., Ash, R. C., Zanjani, E. P., Kersey,J. H., LeBain,T. W., Beverley,P. L. L., Neville, D. M., Jr., and Yorie, R. J. Anti T-cell reagents for humanbone marrow transplantation: ricin conjugate of three monoclonal antibodies.Science(Wash. DC),222: 512-515, 1983.

51. Vand de Wijngaert, F. P., and Rademakers, L. H. P. M., Schuurman, H-J.,deWeger, R. A., and Kater, L. Identification and in situ localization of the"thymic nurse cell" in man. J. Immunol., 730: 2348-2351,1983.

52. Vodinelich, L., Tax, W., Pegram, B. S., Capel, P., and Greaves, M. F. Amonoclonal antibody (WT1) for detecting leukemias of T-cell precursors (T-ALL). Blood, 62: 1108-1113,1983.

53. Werkele, H., Ketelsen, V. W., and Ernst, M. Thymic nurse cells. Lymphoepi-thelial cell complexes in murine thymuses: morphological and serologicalcharacterization.J. Exp. Med., 757: 925-935,1980.

54. Zuckerman, K. S., and Wicha, M. S. Extracellular matrix production by theadherentcells of long-term murinebone marrow cultures. Blood, 67:540-547,1983.

Fig. 1 Morphologicalappearanceof "cobblestone" hematopoietic islands among a flask derived from marrow that was treated with 3 cycles of anti-la + complement.A, 8 weeks in culture; B, 12 weeks in culture. Arrows, cobblestone areas, x 1000.

CANCER RESEARCH VOL. 45 FEBRUARY 1985

766

Research. on September 21, 2015. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from

*

•û"

' .91

È _&. ^ * • «T ' M

•

o

767

Research. on September 21, 2015. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from

1985;45:758-767. Cancer Res   Joel S. Greenberger, Lisa Rothstein, Paolo DeFabritiis, et al.   CultureHuman Marrow on Hematopoiesis in Continuous Bone Marrow Effects of Monoclonal Antibody and Complement Treatment of

  Updated version

  http://cancerres.aacrjournals.org/content/45/2/758

Access the most recent version of this article at:

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  [email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  [email protected] at

To request permission to re-use all or part of this article, contact the AACR Publications

Research. on September 21, 2015. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from