[CANCER RESEARCH57, 4123-4129, September 15, 19971

(plo7 and p130) results in enhanced proliferation (22). Inactivation ofboth Rb and p107 promotes proliferation of the neuroretina cells (23).Therefore, effects of Rb deficiency on specific cell types may dependon the activities of Rb-related genes in these cells. HomozygousRb' embryos die between El2.0 and El6.0 with major defects inneurogenesis, lens development, erythropoiesis, and myogenesis (24—27). However, chimeric mice with Rb' and wild-type cells areviable, and Rb@ cells variably contribute to all tissues examined,suggesting that the effect of Rb inactivation may be cell type specificor non-cell autonomous (28, 29).

All heterozygous Rb@' mice are predisposed to pituitary tumorsbut not retinoblastoma (24—26,30, 31). Inactivation of the wild-typeRb allele occurs prior to hyperplastic transformation (31). Chimericmice consisting of wild-type and Rb mutant cells invariably develop

pituitary tumors and die prematurely of metastatic tumors (28, 29).Therefore, the long-term effect of Rb inactivation cannot be followedin these chimeric mice. The transplantation of fetal liver cells wasundertaken to address the outcome of Rb' hematopoietic cells inboth short-term and long-term studies.

MATERIALS AND METHODS

Animals. Recipients of blood cell transplantation were 3—4-month-oldwild-type mice obtained from heterozygous Rb@' mouse crosses ( I 29/

C57BL). Hematopoietic stem cell donors were El l.5—El3.5embryos fromRb― heterozygotes crosses. The generation of the Rb@' strain has beendescribed (26). All animals were housed in microisolator cages and providedwith sterilized food and acidified water.

Transplantation of Embryonic Hematopoletic Stem Cells. Embryonicliver-derived cells were dissociated by gentle pipetting of wild-type and Rb'El l.5—El3.5livers in PBS. Cells were washed twice in PBS, and cell numberwas determined by hematocytometer prior to transplantation. Mice were given

myeloablative dose of gamma rays from a@ 37Cs source (I 100 cGy, 1 15.8

cGy/min) and divided into two groups. One group received iv. injection of

various numbers of embryonic hematopoietic stem cells as indicated within 8 hpostirradiation. The other group did not receive any transplantation and served

as the control group.In Vivo CFU Assay. The mice were sacrificed by cervical dislocation 7—10

days after transplantation, and the number of macroscopic and microscopiccolonies in the spleen were evaluated. Spleens were fixed in paraformalde

hyde, and paraffin sections were prepared for histological analysis.In Vitro Erythrocyte Progenitor Cell Assays. The procedure was similar

to that of Wu et aL (32). Briefly, individual fetal livers from El2.5 embryoswere removed into IMDM, dissociated, and passed through a metal mesh toobtain single-cell suspensions. After three washes with IMDM, aliquots of cellsuspension were diluted 1:20 in 2% acetic acid to lyse nonnucleated mature

erythrocytes and counted. A I.5 X l0@cells/ml cell suspension in IMDM and2% FBS was prepared. Cell suspensions from each liver (0.1, 0.3, and 0.6 ml)

were mixed with 3 ml of growth medium containing methylcellulose (StemCell Technologies Inc.), and 1.1 ml of the mixture was plated into 35-mmdishes in triplicate. Colony formation was assessed by benzidine staining forCFU-E and BFU-E after 2 and 8 days of incubation, respectively.

Flow Cytometry Analysis: Immunophenotyping of Mouse Blood Cells.Blood was drawn every 3—4weeks from the ocular vein of the recipient mice.The blood cell population was examined by Wright-Giemsa staining of theblood smear, as well as by FACS analysis. For FACS analysis, approximately

200—300 @.dof blood were collected into a Microtainer tube precoated with the

4123

Retinoblastoma Gene Deficiency Has Mitogenic but not Tumorigenic EffectsonErythropoiesis'Nanpin Hu, Margaret L. Gulley, John T. Kung, and Eva Y-H. P. Lee2Department ofMolecular Medicine/institute ofBiotechnology, The University of Texas Health Science Center at San Antonio, San Antonio. 7X 78245-3207 (N. H.. E. Y.H. P. Li:Depart,nent of Pathology, The University of Texas Health Science Center at San Antonio, San Antonio. Texas 78284-7750 (M. L G.J; The Audie Murphy Veteran ‘sHospital. SanAntonio, Texas 78284 (M. L G.J; and institute ofMolecular Biology, Academia Sinica. Taiwan, Republic of China (J. T. K.J

ABSTRACT

The retinoblastoma protein (Rb), an important ubiquitous cell cycleregulator, was Initially identified as the retinoblastoma tumor suppressor.To further address the activities of Rb in proliferation and tumorigenesisIn the hematopoletic lineage, we transplanted Rb@ fetal liver cells intosibling mice and assessed the outcome ofRb@ hematopoietic cells in bothshort-term and long-term studies. Rb@ hematopoletic cells rescued lethally Irradiated mice with an efficiency comparable to that of wild-typecells. In spleen colony-forming unit assays, proliferation rates oftbe Rb'cells were greater than those of the wild-type cells. Similarly, in vitroburst-forming unit-erythroid and colony-forming unlt-erythroid assaysshowed Increased erythroid colony numbers from Rb@ embryonic livers.Recipients of Rb' cells lived for more than 15—iSmonths, and mostblood cell lineages matured normally with the expected switch from fetalto adult hemoglobin. However, the continued presence of nucleated eryth

rocytes in the peripheral blood and extensive extramedullary erythropoiesis Indicated that the Rb@ erythrocytes were not completely normal. Noerythroleukemia developed during the 15—18month period followingti'ansplantation. These results demonstrate the mitogenic effect but nottumorigenic transformation in erythrocyte lineage in the absence of Rb,which Is distinct from the effect ofRb deficiency in neuroectodermal cells.

The study supports the prevalent model that loss of the ubiquitouslyexpressed tumor suppressor gene predisposes to only a limited spectrumoftumors.

INTRODUCTION

The human retinoblastoma susceptibility gene is one of the beststudied tumor suppressors that is mutated in multiple tumor types(reviewed in Refs. 1—3).Mutational inactivation of the retinoblastomagene is the rate-limiting step in Rb3 formation (4—8). Rb is alsomutated in other tumors, such as small-cell lung carcinoma and breastcarcinoma (9—i1). The heterogeneous expression of the Rb protein inthe breast carcinoma suggests that inactivation occurs during progression of the tumor (10, 12).

Rb plays an important regulatory role in G1-S progression (13). Itis a substrate for cydin-dependent kinases and an inhibitor of thetranscription factor E2F (reviewed in Refs. 1—3,14, and 15). Hypophosphorylated Rb inhibits G5-S progression, and phosphorylationof Rb by cyclin-dependent kinases abrogates the inhibitory activitiesof Rb (13, 16). The hypophosphorylatedform of Rb binds to E2F(17—19)and represses its transcription activator function, which isrequired for the expression of S-phase genes (reviewed in Ref. 15).

Two Rb-related genes, p107 and p130, have been identified (20,21). In chondrocytes, inactivation of these two Rb-related proteins

Received 5/I 2/97; accepted 7/17/97.Thecostsof publicationof thisarticleweredefrayedin partby the paymentof page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

I This work was supported by NIH Grants CA49649 and HD30625 (to E. Y-H. P. L.)

and in part by P. I. C. A. Coltman, Jr. Grant U0l-CA60432.2 To whom requests for reprints should be addressed, at Department of Molecular

Medicine/Institute of Biotechnology, The University of Texas Health Science Center atSan Antonio, 15355 Lambda Drive, San Antonio, TX 78245-3207.

3 The abbreviations used are: Rb, retinoblastoma protein; Rb, retinoblastoma gene;BFU, burst-forming unit; BFU-E, BFU-erythroid; CFU, colony-forming unit; CFU-E,CFU-erythroid; IMDM, Iscove's modified Dulbecco's medium; E, embryonic day.

Research. on January 27, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

RB IN ERYTHROPOIESIS

Variation in Proliferative Activity of Rb-deficient Hematopoietic Cells in Vivo. To further examine Rb' blood stem cells, wecompared their repopulation ability to those from the wild-type embryos in a CFU assay. CFUs were observed in the spleens of micetransplanted with wild-type, heterozygous, and homozygous cells onday 7 (Fig. 1, A and B) but not in control irradiated animals. Becauseno difference was found between the wild-type and heterozygousdonor cells in all assays performed, cells from mice of these twogenotypes will not be distinguished. On day 7, CFUs in mice receiving wild-type donor cells were uniform in size and included coloniesof multiple lineages (i.e., CFU-E, megakaryocytic, and myeloid).Larger CFU-Es were identified in mice receiving Rb' cells (Fig.IB). Because individual CFUs are clonally derived, variation in thesize of CFUs in recipients of Rb― cells suggests differences in theproliferation rate of their stem cells. On day 10, in contrast to thedistinct CFUs seen in the recipients of wild-type cells, there were nodistinct erythroid colonies in the spleens of mice receiving Rb'cells. Instead, the spleens were filled with immature erythroid cells atdifferent stages of maturation (data not shown). Differences in erythroid colonies between wild-type and Rb― recipients on days 7 and10 suggest that Rb@ erythroid cells have accelerated growth. Interestingly, no differences were found in megakaryocyte and myeloidcolonies (data not shown), consistent with previous findings in theRb@ embryos (24, 25, 26).

Variation in Proliferative Activity of Rb-deficient HematopoieticCellsin Vitro.Tofurtherdeterminewhichstage(s)oferythrocytedevelopment is affected by Rb deficiency, in vitro erythrocyte progenitor colony assay was performed. The number of CFU-Es, a rapiddividing cell that gives rise to an erythroblast colony of 8—64cells in2 days, and BFU-Es, a more immature, slowly dividing cell that gives

rise to larger colonies of erythroblasts in 7—10days, was determined.In Rb' hematopoietic cells, there were increased numbers of boththe rapidly dividing CFU-Es (Fig. 1C) and the more immature and lessfrequenfly dividing BFU-Es (Fig. 1D). This effect was most pronounced (—2-fold) in the 8-day culture of Rb' BFU-E. Theseresults are consistent with the in vivo finding of increased growth oferythrocyte progenitor cells.

Rb@ Donor Cells Persist in Bone Marrow of the Recipients.Because we were interested in the long-term effect of Rb inactivationin the hematopoietic cells, we tested whether Rb' donor cells weresustained in the recipients. A Neo probe that specifically hybridizes totranscripts from the mutant Rb allele but not the wild-type allele wasused for in situ hybridization. Six months after transplantation, allbone marrow cells but not the cartilage cells in the Rb' recipientswere shown to be derived from the donor (Fig. 2).

Extramedullary Erythropoiesis. Blood smear and flow cytometric analysis revealed the presence of nucleated erythrocytes in theperipheral blood of transplant mice receiving Rb' cells as early as3 weeks posttransplantation. To determine whether the abnormal

erythroid cells seen in the recipients were derived from Rb' donorcells, DNA was isolated from various tissues of two 8-month-oldrecipients. Southern blot analysis was carried out using a probespanning intron 19 of the Rb-I gene. A predominant band at 11.5 kbrepresenting the mutant Rb-i allele was seen in the spleens of micethat received Rb― cells (Fig. 3). In addition, Rb' cells were alsopresent in the liver and salivary gland. Extensive erythropoiesis wasdetected histologically by 6 months posttransplantation. Spleen, liver,salivary gland, and kidney of recipient mice of Rb' cells were filledwith immature erythroid cells. Wright-Giemsa stain indicated thatthese were likely intermediate and late stage normoblasts (data notshown). Some of the cells had abundant cytoplasm and clumpedchromatin similar to immunoglobulin-producing plasma cells. However, they stained negative by periodic acid-Schiff and positive by

anticoagulant EDTA (Becton Dickinson, San Jose, CA). The blood sampleswere vortexed and divided into 30—50@.daliquots in 12 X 75-mm plastic tubescontaining appropriate antibodies. After incubation in the dark for 20 mm, 1.5

ml of red cell lysis solution (FACS lysis solution, Becton Dickinson) wereadded, and the mixture was incubated for 5 rain. The blood cells were pelletedby centrifugation, washed once with PBS containing 1% sodium azide, andresuspended in 0.3 ml of PBS. Samples were analyzed immediately on a flowcytometer (FACScan, Becton Dickinson). The distribution of major blood celltypes and the percentage of undifferentiated cell populations were monitoredby a panel of antimouse monoclonal antibodies (PharMingen, San Diego, CA):

CD3 (pan T), CD4 (T-helper), CD8 (T-suppressor), B220 and slg (B cell),GR1 (granulocyte),andR-PE(erythrocyte).To furtheridentifylymphocytesatdifferent stages of differentiation, CD5, 6C3, heat-stable Ag, and CD43 wereused as the markers.

Histopathological Analysis. Mouse tissues, including spleen, liver, salivary gland, and kidney, were fixed in 4% paraformaldehyde in PBS, dehydrated, processed, and embedded in Paraplast (Fisher Scientific) followingstandard histological procedures. The paraffin tissue blocks were cut into 4—6pm thin sections, which were deparaffined and stained with H&E or WrightGiemsa stain. Blood smears were stained with SureStain Wright-Giemsa(Fisher Scientific) according to manufacturer's instructions. A 3'3'-dimethoxybenzidine staining was also performed on blood smears as described byMcLeod et al. (33). Photomicrographs were taken under a Zeiss Axiphotmicroscope with KOdakEktar 25 film (ASA 100).

Southern Blot Analysis. Genomic DNA was isolated from tail, spleen, liver,and salivarygland tissuesoftransplant recipientmice.All DNA was digestedwithEcoRI, fractionated on a 0.8% agarose gel, and transferred to nylon membranes

(Hybond, Amersham). Southern blots were hybridized with a 0.3-kb probe fromintron 19 of the mouse Rb-i gene, which distinguishes wild-type alleles by a10.0-kb band from mutated alleles with an I 1.5-kb band (26).

In Situ Hybridization. Digoxigenin-labeled antisense Nan 50-merprobe (5'-CCAAOCGGCCGGAGAACCTGCGTGCAATCCATCTfGTFCAATGGCCGATC-3') was prepared according to manufacturer's instructions(Genius 6 oligonucleotide tailing kit, Boehringer Mannheim). Six months aftertransplantation, bones from recipients were dissected, paraffin embedded, andcut in 4—6,.@msections. Hybridization and color development were accordingto the manufacturer's guide (Boehringer Mannheim).

Switching of Hemoglobln mRNA was extracted from wild-type and mutant livers at E13.5 and E14.5 and from the spleens of transplantation recipients, using Micro-Fast Track mRNA isolation kit (Invitrogen) according to themanufacturer's instructions. cDNA was synthesized from 100—200 ng ofmRNA using SuperScript II RNase H reverse transcriptase (Life Technologies, Inc.) with oligo(dT) as the primer. Oligonucleotide primer pairs wereused in PCR to amplify specific hemoglobin chains: 5'-primer, 5'-CTGAC

AGATGCTCTCUGGG-3' and 3'-primer, 5'-CACAA CCCCAGAAACAGACA-3' (@3-majorglobin);5'-primer, 5'-GGAG A GTCC A TFAAG AACCT AGACA A-3' and 3'-primer, 5'-CTGTG AATfC AUGC CGAAGTGAC-3' (€-globin,Y-2); and 5'-primer, 5'-GTGAC GAGGC CCAGAGCAAG AA-3' and 3'-primer, 5'-AGGGG CCGGA CTCAT CGTAC TC-3'($3-actin as internal control; Ref. 34). Briefly, each PCR reaction mixture

contained 2 @.dof cDNA mixture, reaction buffer containing 2 mMMg2@,0.2mM oligonucleotide primers, 200 nmi dNTPs, and 3 units of Taq DNApolymerase (Boehringer Mannheim). Reactions were cycled at 94°Cfor 30 s,55°Cfor 30 s, and 72°Cfor 60 s and for 40 cycles using a PTC-lOOProgrammable Thermal Controller (Mi Research, Inc.). For quantitative corn

parison, @3-actinwas used as control.Retransplantation of Blood Cells from Recipients of Rb@ Cells into

Nude Mice. The spleensof transplantationrecipientsof the Rb― cells wereremoved, and 5 x 106to 40 X 106dissociated spleen cells were injected intotail vein of four nude and three wild-type mice. Blood was withdrawn fromthese mice posttransplantation at 1-month intervals for FACS analysis, andblood smears were prepared and analyzed by light microscopy.

RESULTS

Efficient Rescue of Lethally Irradiated Mice by Rb-deficientHematopoietic Cells. One million to S million cells were found to besufficient to rescue lethally irradiated mice, regardless of the origin ofdonor cells.

4124

Research. on January 27, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

RB IN ERYTHROPOIESIS

benzidine (see below), confirming that they were immature erythrocytes. Thus, histological examination of Rb' recipient mice mdicated extensive extramedullary erythropoiesis.

Flow Cytometric Analysis Confirms the Presence of CirculatingImmature Erythrocytes. A systematic analysis of the hematopoietic cells in recipient mice by flow cytometric analysis using apanel of mouse monoclonal antibodies specific for different stagesof T-cells, B-cells, granulocytes, and erythrocytes revealed differences between wild-type and Rb' recipients (Table 1). At different times, 10—30weeks after transplantation, comparable populations of mature T lymphocytes (as indicated by staining of CD3,or CD5 or by double staining of CD4 and CD8) were present in therecipients. Similarly, the granulocyte numbers were comparable.Although there appeared to be a decrease in the mature B-cellpopulation, as indicated by 6C3 and B220 staining, such a decreasewas not consistently seen at 1 month posttransplantation (data notshown). On the other hand, an elevated percentage (36—38%) ofnucleated erythrocytes persisted in the blood of Rb@ recipients(Table 1). This phenomenon was confirmed morphologically byblood smear examination that showed 33—56%nucleated erythrocytes. The nucleated erythrocyte population persisted in all Rb'

recipients regardless of the stage of donor cells (El l.5—El3.5).Therefore, it appears to be an inherent property of the Rb' stemcells. On the basis of the flow cytometric and blood smear results,

we conclude that Rb deficiency specifically affects the erythrocytelineage.

Normal Hemoglobin Switching of Rb-deficient Erythrocytes.Despite the continued presence of peripheral nucleated erythrocytes, all mice transplanted with Rb― stem cells also harboredmorphologically normal red cells in peripheral blood, althoughthere was a propensity for increased polychromasia in the erythrocytes. To assess differentiation of Rb― erythrocytes, we determined the hemoglobin type of Rb― hematopoietic cells at different stages. Reverse transcription-PCR was carried out usingmRNAs isolated from embryonic livers and spleens of the transplant recipients. Both adult a-major and embryonic c-hemoglobinwere produced in the livers of El3.5 and El4.5 wild-type andmutant embryos (Fig. 4). In the spleens of recipients of wild-typeand Rb' cells, a complete switch to the adult a-major hemoglobin was observed (Fig. 4). Thus, consistent with the chimeraanalysis (28, 29), Rb-deficient erythrocytes undergo normal hemoglobin chain switching.

-@ _;

i@.. ; : • ‘@@ @j:.@@

. . .; .

n;

,.@ @—i,@;@

C10

U

I-

az

D8°

8

Fig. I. CFU analysis ofthe transplant recipients and proliferative activities ofRb@ erythrocyte lineage in vitro. A and B, CFUs in the spleen of recipients, 7 days posttransplantation.A, CFUs in the spleens of recipients of wild-type cells. The CFU-E (E) and myeloid CFU-M (M) are shown. Bar, 100 @m.B, CFUs in the recipients of Rb' cells. The CFU-Es areindicated. Bar, 100 @m.Erythrocyte progenitor cells from E12.5 wild-type and Rb-mutant fetal livers were cultured in growth medium for 2 and 8 days, respectively. The number ofCFU-Es (C) and BFU-Es (D) of Rb― erythroid cells are increased to —1.5-fold and 2-fold that of wild-type cells, respectively. C and D, columns, data from representative separateexperiments; bars, SD.

60

40

20

0

-@

az

6

4

2

Rb-I- Rb+/+ Rb-I- Rb+/+

Cell types

4125

Research. on January 27, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

RB IN ERYTI-IROPOIESIS

bri

.@@

Fig. 2. Prolonged replacement of Rb' cells in the bone marrow of the recipients. Six months after transplantation, in situ hybridization was performed on bone marrow sectionsof the recipient mice using Neo oligonucleotide probe. The Neo marker was negative in a recipient of wild-type cells (A), whereas it was positive in a recipient of Rb' cells (B).bn, bone. A. bar. 20 pm; B. bar, 12.5 @m.

5.

. V

I@

1@@ ‘

Mice Are Anemic after Transplantation of Rb-deficient Cells.The aberrant phenotypes in the recipients of Rb-deficient cells described thus far indicate deregulation of proliferation in the erythrocytes. It is also plausible that such deregulated erythropoiesis mani

fested by polychromasia and increased numbers of immatureerythroid precursors in the peripheral blood accompanied by widespread extramedullary hematopoiesis may be a physiological responseto anemia. To evaluate this, cell count and blood profile were performed in older transplant recipients. One year after transplantation, adecrease in hemoglobin amount from 10.8 g/dl in the wild-typerecipient to 6.8 g/dl in the Rb' recipient and a decrease in hema

tocrit from 3 1.9% in the wild-type recipient to I8.2% in the Rb'recipient was seen. On the basis of the blood analysis, it is concludedthat mice transplanted with Rb@ cells but not wild-type cells wereanemic.

Increased Fragility of Rb-deficient Erythrocytes. The efficientrescue of lethally irradiated mice by Rb' hematopoietic cells and

long life span of the transplant recipients suggested that@ erythrocytes are functional. However, the persistent anemia suggested thatthey are not completely normal. A plausible cause for anemic phenotype of Rb' recipients was revealed by blood smear. Increased lysisof the ‘ nucleated erythrocytes was consistently observed inblood smear preparations (Fig. 5A), suggesting elevated membranefragility. Addition of albumin prior to blood smear preparation greatlydecreased cell lysis (Fig. SB). Benzidine staining demonstrated thatincreased populations of nucleated blood cells were indeed immature

erythrocytes (Fig. SC). An additional osmotic fragility test also mdi

cated that Rb' nucleated erythrocytes are more fragile compared towild-type erythrocytes (data not shown).

Transplantation of Blood Cells from Recipients of Rb@ Cellsto Nude Mice Morphologically, the peripheral nucleated erythrocytes in recipients of Rb' cells do not resemble transformed erythroleukemia cells. To test whether peripheral nucleated erythrocytesseen in recipients of Rb' cells could continue to divide, we transplanted spleen cell preparations containing infiltrated hematopoieticcells from recipients of Rb' cells into nude mice. The Rb― cellsdid not persist in these nude mice. In addition, no nucleated erythrocytes or abnormal hematopoietic cells were seen in the peripheralblood of the nude mice (data not shown).

DISCUSSION

In this study, we have addressed the short-term and long-termeffects of Rb deficiency in hematopoietic cells. We demonstrated thatRb' cells could rescue lethally irradiated mice as efficiently aswild-type cells. The donor fetal liver cells were taken from El 1.5—E13.5 embryos, the stages at which little liver pathology was observedin Rb-deficient mice (26). Because a consistent phenotype was observed in Rb' recipients receiving a wide range of doses of donorcells (I X 106 to s.o x 106), it is likely that the abnormal phenotype

seen is intrinsic to the Rb' erythroblasts.

Fig. 3. Presence of Rb@' cells in multiple organs of the recipients. Genomic DNA was isolatedfrom tail, spleen, liver, and salivary gland tissues oftwo recipient mice (mice 440 and 470) that receivedRb' cells. A 10.0-kb band corresponding to the

wild-type Rb-I allele is shown in tail DNA. In thespleen DNA, a predominant I 1.5-kb band of themutant Rb-I allele is present. In liver and salivarygland DNA. both I0.0- and I I.5-kb bands are detected, indicating the presence of wild-type cells andRb@' cells.

4126

I ---#440 II #470?‘@

@@ @;:i@@ E-@@@

mRb—1

- 1 1 .5 kb — —@@ .@ —@@

1ftf@1 1_@@ . .@+ 1u.@_,1(0 ____@ aMp@ ::i:=

Research. on January 27, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

Table I Presence of nucleatederithrocvtes in the circulatingblood%

of nucleatederythrocytesintotalDonor

cellsAge (wk)nucleated

bloodcellsBy

FACS BymorphologyRb@'@10

20304

<13 <14<1Rb'10

2030NDa

3336 383856a

ND,notdetermined.

@..._@ww...—@

RB IN ERYTHROPOIESIS

Rb in the Acceleration of Erythrocyte Proliferation and Differentiation. Both spleen CFU assay in vivo and erythrocyte progenitorassay in vitro demonstrate accelerated growth of Rb ‘ erythrocyteprogenitor cells (Fig. 1). Because CFU-E in spleen was compared7—10days after transplantation, and colonies were scored 2 or 8 daysafter plating in the in vitro progenitor assays, the data indicate initialproliferation rates of the primary cells prepared from mouse embryos.These results differ from that of the extensively cultured cell lines inthat stable cell lines established after replacement of wild-type Rb

have growth rate comparable to that of the parental Rb-deficient cells(36). Similarly, the doubling time of Rb-deficient fibroblast cell linesis comparable to that of the wild-type fibroblasts despite a short@

phase in the former cells (37). The discrepancy may reflect a favorableselection of fast-growing cell population in culture upon continuouspassage.

Increased proliferation of Rb' cells seen in the spleen CFU-E

may reflect the intrinsic growth suppression function of Rb in theerythrocyte (1—3).On the other hand, there may be compound causesfor the anemic phenotype and persistent extramedullary erythropoiesis. Fragility testing indicated that the Rb@ erythrocytes were muchmore fragile than normal erythrocytes (Fig. 5 and data not shown).The molecular basis of the fragility is unknown, but it is likely thatderegulated expression of membrane-associated proteins may be involved. There are abundant examples in which Rb may activate orrepress the expression of downstream target genes through its interaction with transcription factors (2). Fragility of the Rb' nucleatederythrocytes may shorten their life spans and prevent further accumulation of mutations that promote their malignant transformation. Inaddition, the decreased hemoglobin levels in Rb' animals maystimulate erythropoiesis to maintain blood homeostasis. The anemicphenotypes seen in the recipients of Rb' cells may result from thesecombined factors.

The presence of morphologically normal mature red cells and thenormal switching of hemoglobin genes indicate that Rb' erythrocytes undergo normal differentiation processes. This is consistent with

the finding from chimera analysis (28, 29). However, the anemicphenotype seen in the recipients of Rb― cells suggests a suboptimalfunction of Rb― erythrocytes. The lack of such a phenotype in the

Effects of Rb Deficiency in Various Blood Cell Lineages. Rbappears to have predominant effects on erythrocyte lineage based onFACS analysis, blood smear, and histopathological studies. Consistent with earlier work on chimeric mice by injection of Rb― embryonic stem cells into RAG-2-deficient blastocysts (35) in which all Tand B cells derived from Rb― embryonic stem cells are normal (35),there is no gross abnormality in T and B cells in the recipient mice. Itis plausible that p107 and p130 complement Rb function in T and B

cells. Indeed, the neuroretinal defect seen in the p1O7Rb@' miceis consistent with overlapping functions between these two genes invivo (23). Erythrocyte lineage in chimeric mice of Rb@ and wildtype embryos appeared normal, although transient presence of nudeated erythrocytes was noticed (28, 29). The lack of a persistentabnormality in the erythrocyte lineage in the chimeric mice may bedue to short life spans caused by metastatic pituitary tumors and thepresence of wild-type erythrocytes in these chimeric animals (seebelow).

It is well acceptedthat myeloid cells, including platelets,neutrophils, and erythrocytes, are all derived from common precursor cells.In these studies, the platelets appear normal in Rb@ cell recipients(data not shown). Additionally, no aberrant phenotype was observedin the neutrophil lineage. Therefore, defects due to Rb deficiency aremainly limited to the CFU-E compartment. Both morphology andhistological staining suggested that the peripheral nucleated erythrocytes were normoblasts at intermediate and late stages of differenti

ation, indicating that no single specific stage of erythrocyte development was preferentially affected.

Fig. 4. Embryonic and adult hemoglobin expression in wild-type and Rb-mutant mice. mRNA wasextracted from livers of El3.5 and El4.5 wild-typeand mutant embryos and from spleens of wild-typeand mutant cell recipients. Reverse transcriptionPCR analysis was performed using specific primerpairs to amplify adult (a-major) and embryonic (e)hemoglobin and @-actinas control.

4127

Liver SpleenE14.5

-I-- +1+ —I--+1+ —1-+1+E13.5 adult

f'—major

C

@actin

Research. on January 27, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

RB IN ERYTHROPOIESIS

A o@@@ O(@;)&@ C):

,.I@@ c;@@@@ 0@ U@@@@@ 0@ 0@@@

—, c@ @:-@@@ 0G5@ a@ •:s@ e

‘ : Vofl:@ @o:@@

@ :-@ )oc@0n9fl@@ ui

0@@@@

.@ 0 ,@ (- @)

<, - S@@@@@@ @-@

@@ -- . .\ y.. ;@@

@ :@‘@@ @-@@@ ,c @—@:@

C D

‘z-@ I-(in

Bn1)' m

.(

.- -@ .†-̃ S

‘fl@I@'

ne...

@i,

0‘@@-@i@ .

Fig. 5. Fragility of the Rb' nucleated erythrocytes. A, blood smears prepared from Rb' transplant recipient. lne, lysed nucleated erythrocyte; I, lymphocyte; m, monocyte; n,neutrophils. B, same as A except that I drop of bovine albumin was added prior to preparation of blood smear. ne, nucleated erythrocyte. C, nucleated erythrocytes and mature RBCsare benzidine stain positive. D, blood smear of wild-type transplant recipient. Bar. 20 @m.

hematopoietic lineages in the chimera may be in part due to the factthat fully functional erythrocytes and Rb' erythrocytes coexist inthe chimera. On the other hand, recipients having complete replacement of hematopoietic cells in our study had only functionally subnormal Rb@ hematopoietic cells that may have led to the anemicphenotype described. Roles for Rb in differentiation have been demonstrated and implicated in several cell types. Neuronal differentiationis incomplete in the absence of Rb (38). Studies of muscle cells andadipocytes have also shown that Rb is required for their differentiation(39, 40). Rb-deficient mice partially rescued by a low level of Rbexpression encoded by a transgene were shown to have shorter myotubes with fewer myofibrils and reduced muscle fibers (27). Therefore, Rb is required for full differentiation in many cell lineages.Although Rb@ erythrocytes are morphologically normal and expressf3-globin, the level of j3-globin may not be optimal, similar to that of

the decreased myosin in Rb@ muscle.Mitogenic but not Tumorigemc Effects of Rb in Erythrocyte

Lineage. The absence of erythroleukemia despite persistent presence ofperipheral immature nucleated erythrocytes and increased proliferation inCFU-E assay in vivo and in vitro in Rb― recipients is intriguing.Although mutation of the Rb gene has been found in many tumor types,only specific cancer risks (i.e., retinoblastoma and osteosarcoma) arehighly elevated in the Rb heterozygotes (1). It is plausible that Rbmutation may facilitate tumor progression in multiple cell types, but itsrole in tumor initiation may be cell type specific. Such specific tumorpredisposition is also true in heterozygous Rb@ mice in that they are

100% predisposed to pituitary tumors of the intermediate lobe (30).Inactivation of genes in the Rb pathway have been shown to be tumorigenic (reviewed in Ref. 41). Deregulation of cell proliferation does notalways lead to tumorigenesis. Underlying mechanisms of the cell-typespecific effects and the different consequences of the dysfunction of a cellcycle regulator in a given cell type are yet to be addressed.

ACKNOWLEDGMENTS

We thank Dr. Y-W. Kan for advice and for reviewing the histopathologicalslides and Drs. Steve Skapek and Gopal Dasika for critical reading of the

manuscript.

REFERENCES

1. Riley, D. J., Lee, E. Y-H. P., and Lee, W-H. Retinoblastoma gene: more than a tumorsuppressor. Annu. Rev. Cell. Biol., 10: 1—29,1994

2. Skapek, S. X., Qian, Y-W., and Lee, E. Y-H. P. The retinoblastoma protein: morethan meets the eyes. Prog. Retinal Eye Res., 4: 591—626, 1997.

3. Weinberg, R. A. The retinoblastoma protein and cell cycle control. Cell, 81: 323—330,1995.

4. Knudson, A. G. Mutation and cancer: statistical study of retinoblastoma. Proc. Nail.Acad. Sci. USA, 68: 820—823, 1971.

5. Cavenee, W. K., Diyja, T. P., Phillips, R. A., Benedict, W. F., Godbout, R., Gallie, B. L.,Murphiee, A. L., Strong, L C., and White, R. L Expression of recessive alleles bychromosomalmechanismsin retinoblastoma.Nature (Lond.),305: 779-784, 1983.

6. Friend, S. H., Bernards, R., Rogelj, S., Weinberg, R. A., Rapaport, J. M., Albert,D. M., and Dryja, T. P. A human DNA segment with properties of the gene thatpredisposes to retinoblastoma and osteosarcoma. Nature (Lond.), 323: 643—646,1986.

4128

Research. on January 27, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

RB IN ERYTHROPOIESIS

7. Fung, Y. K. T., Murphree, A. L, rAng, A., Qian, J., Hinrichs, S. H., and Benedict,W. F. Structural evidence for the authenticity of the human retinoblastoma gene.Science (Washington DC), 236: 1657—1661,1987.

8. Lee, W-H., Bookstein, R., Hong, F., Young, L-J., Shew, J-Y., and Lee, E. Y-H. P.Human retinoblastoma susceptibility gene: cloning, identification, and sequence.Science (Washington DC), 235: 1394—1399,1987.

9. Lee, E. Y-H. P. Tumor suppressor genes and their alterations in breast cancer. Semin.Cancer Biol., 6: 119—125,1995.

10. Spandidos, D., Karaiossifidi, H., Malliri, A., Linardopoulos, S., Vassiloaros, S.,Tsikkinis, A., and Field, J. Expression of Ras, Rbl and p53 in human breast cancer.Anticancer Rca., 12: 81—90,1992.

11. Hensel,C.H.,Hsieh,C.L.,Gazdar,A.F.,JOhnSOO,B.E.,Sakaguchi,A.Y.,Naylor,S.L.,Lee, W-H., and Lee, E. Y-H. P. Altered stnsctureand expressionof the human retinobbstumasuscepilbility gene in small cell lung cancer. CancerRes, 50: 3067-3072, 1990.

12. Varley, J. M., Armour, J., Swallow, J. E., Jeffreys, A. J., Ponder, B. A. J., T'Ang, A.,Fung, Y-K. T., Brammar, W. J., and Walker, R. A. The retinoblastoma gene isfrequently altered leading to loss of expression in primary breast tumors. Oncogene,4:725—729,1989.

13. Goodrich, D. W., Wang, N. P., Qian, Y-W., Lee, E. Y-H. P., and Lee, W-H. TheretinoblastomageneproductregulatesprogressionthroughG1phaseofthe cellcycle.Cell, 67: 293—302,1991.

14. Sherr, C. J., and Roberts, J. M. Inhibitors ofmammalian G1cyclin-dependent kinases.GenesDev., 9:1149—1163,1995.

15. Nevin, J. R. A link between the Rb tumor suppressor protein and viral oncoproteins.Science (Washington DC), 258: 424—429, 1992.

16. Ewen,M. E., Sluss,H. K.,Sherr,C. J., Matsushime,H.,Kato,J-y.,andLivingston,D. M. Functional interactions of the retinoblastoma protein with mammalian D-typecyclins. Cell, 73: 487—497,1993.

17. Hem, K., Harlow, E., and Fattaey, A. Inhibition of E2F-l transactivation by directbinding of the retinoblastoma protein. Mol. Cell. Biol., 13: 6501—6508,1993.

18. Kaelin, W. G., Krek, W., Sellers, W. R., DeCaprio, J. A., Ajchenbaum, F., Fuchs,C. S., Chittenden, T., Li, Y., Farnham, P. J., Blanar, M. A., Livingston, D. M., andRemington, E. K. Expression cloning of a cDNA encoding a retinoblastoma-bindingprotein with E2F-like properties. Cell, 70: 351—364,1992.

19. Shan, B., Thu. X., Chen, P-L., Durfee, T., Yang, Y., Sharp, D., and Lee, W-H.Molecular cloning of cellular genes encoding retinoblastoma-associated proteins:identificationof a genewithpropertiesof the transcriptionfactorE2F.Mol.Cell.Biol.,12:5620—5631,1992.

20. Ewen, M. B., Xing, Y., Lawrence, J. B., and Livingston, D. M. Molecular cloning,chromosomal mapping, and expression of the cDNA for p107, a retinoblastoma geneproduct-related protein. Cell, 66: 1155—1164, 1991.

21 . Hannon, G. J., Demetrick, D., and Beach, D. Isolation of the Rb-related p130 throughits interaction with CDK2 and cyclins. Genes Dev., 7: 2378—2391,1993.

22. Cobrinik, D., Lee, M-H., Hannon, G., Mulligan, G., Bronson, R. T., Dyson, N.,Harlow, E., Beach, D., Weinberg, R. A., and Jacks, T. Shared role ofthe pRB-relatedp130andplO7proteinsin limbdevelopment.GenesDcv.,10: 1633—1644,1996.

23. Lee, M-H., Williams, B. 0., Mulligan, G., Mukai, S., Bronson, R. T., Dyson, N.,Harlow, E., and Jacks, T. Targeted disruption of plO7: functional overlap betweenp107 and Rb. Genes Dcv., 10: 1621—1632,1996.

24. Clarke, A. R., Maandag, E. R., van Roon, M., van der Lugt, N. M. T., van der Valk,M., Hooper, M. L., Berns, A., and te Ride, H. Requirement for a functional Rb-l genein murine development. Nature (Lond.), 359: 328—330,1992.

25. Jacks, T., Fazeli, A., Schmitt, E. M., Bronson, R. T., Goodell, M. A., andWeinberg, R. A. Effects of an Rb mutation in the mouse. Nature (Lond.), 359:295—300,1992.

26. Lee, E. Y-H. P., Chang, C-Y., Hu, N., Wang, Y-C. J., Lai, C-C., Herrup, K., Lee,W-H., and Bradley, A. Mice deficient for Rb are nonviable and show defects inneurogenesisandhaematopoiesis.Nature(Lond.),359: 288—294,1992.

27. Zacksenhaus, E., Jiang, Z., Chung, D., Marth, i. D., Phillips, R. A., and Gallie, B. L.pRb controls proliferation, differentiation, and death of skeletal muscle cells and otherlineages during embryogenesis. Genes Dev., 10: 3051—3064, 1996.

28. Williams, B. 0., Schmitt, E M., Remington, L, Bronson, R. T., Albert, D. M., Weinberg,R. A., and Jacks, T. Extensive contribution of Rb-deficient cells to adult chimeric micewith limited histopathological consequences. EMBO J., 13: 4251-4259, 1994.

29. Maandag, E. C. R., van der Valk, M., Vlaar, M., Feltkamp, C., O'Brien, J., van Roon,M., van der Lugt, N., Berns, A., and te Ride, H. Developmental rescue of anembryonic-lethal mutation in the retinoblastoma gene in chimeric mice. EMBO J.. 13:4260—4268, 1994.

30. Hu, N., Gutsmann, A., Herbert. D. C., Bradley, A., Lee, W-H., and Lee, E. Y-H. P.Herterozygous Rb-l @‘-mice are predisposed to tumors of the pituitary gland withnearlycompletepenetrance.Oncogene,9: 1021—1027,1994.

31. Nikitin, A. Y., and Lee, W-H. Early loss of the retinoblastoma gene is associated withimpaired growth inhibitory innervation during melanotroph carcinogenesis in Rb@mice. Genes Dcv., 10: 1870—1879,1996.

32. Wu, H., Liu, X., Jaemsch, R., and Lodish, H. F. Generation of committed erythroidBFU-E and CFU-E progenitors does not require erythropoietin or the erythropoietinreceptor.Cell,83: 59—67,1995.

33. McLeod, D. L., Shreeve, M. M., and Axelrad, A. A. Improved plasma culture systemfor production of erythrocytic colonies in vitro: quantitative assay method for CFU-E.Blood, 44: 517—534,1974.

34. Weiss, M. J., Keller, G., and Orkin, S. H. Novel insights into erythroid developmentrevealed through in vitro differentiation of GATA-l-embryonic stem cells. GenesDcv., 8: 1184—1197,1994.

35. Chen, J., Gorman, J. R., Stewart, V., Williams, B., Jacks, T., and Alt, F. W.Generation of normal lymphocyte populations by Rb-deficient embryonic stem cells.Cuff.Biol.,3: 405—413,1993.

36. Chen, P. L., Chen, Y., Shan, B., Bookstein, R., and Lee, W-H. Stability of retinoblastoma gene expression determines the tumorigenicity of reconstituted retinoblastoma cells. Cell Growth & Differ., 3: 119—125,1992.

37. Herrera, R. E., Sah, V. P., Williams, B. 0., Makela, T. P., Weinberg, R. A., and Jacks,T. Altered cell cycle kinetics, gene expression, and G1 restriction point regulation inRb-deficient fibroblasts. Mol. Cell. Biol., 16: 2402—2407, 1996.

38. Lee, E. Y-H. P., Hu, N., Yuan, S-S. F., Cox, L. A., Bradley, A., Lee, W-H., andHen-up, K. Dual roles ofthe retinoblastoma protein in cell cycle regulation and neurondifferentiation. Genes Dcv., 8: 2008—2021, 1994.

39. Gu, W., Schneider, J. W., Condorelli, G., Kaushal, S., Mabdavi, V., and NadalGinard, B. Interaction of myogenic factors and the retinoblastoma protein mediatesmuscle cell commitment and differentiation. Cell, 72: 309—324, 1993.

40. Chen, P-L., Riley, D. J., Chen, Y., and Lee, W-H. Retinoblastoma protein positivelyregulates terminal adipocyte differentiation through direct interaction with CIEBPs.Genes Dcv., 10: 2794—2804, 1996.

41. Lin,S-C.J., Skapek,S. X.,andLee,E. Y-H.P. Genesin the RBpathwayandtheirknockout mice. Semin. Cancer Biol., 7: 279—289,1997.

4129

Research. on January 27, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

1997;57:4123-4129. Cancer Res   Nanpin Hu, Margaret L. Gulley, John T. Kung, et al.   Tumorigenic Effects on ErythropoiesisRetinoblastoma Gene Deficiency Has Mitogenic but not

  Updated version

  http://cancerres.aacrjournals.org/content/57/18/4123

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

  .pubs@aacr.orgDepartment at

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

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/57/18/4123To request permission to re-use all or part of this article, use this link

Research. on January 27, 2021. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from