Differential expression of a novel proline-rich homeobox gene (Prh) in human hematolymphopoietic cells

Differential Expression of a Novel Proline-Rich Homeobox Gene (Prh) in Human Hematolymphopoietic Cells

By Guidalberto Manfioletti, Valter Gattei, Emanuele Buratti, Alessandra Rustighi, Angela De luliis, Donatella Aldinucci, Graham H. Goodwin, and Antonio Pinto

Proline-rich homeobox (Prh) is a novel human homeobox- containing gene recently isolated from the CD34+ cell line KG-1A. and whose expression appears mainly restricted t o hematopoietic tissues. To define the pattern of Prh expression within the human hematopoietic system, we have analyzed its constitutive expression in purified cells obtained from normal hematopoietic tissues, its levels of transcription in a number of leukemia/lymphoma cell lines representing different lineages and stages of hematolymphopoietic differentiation, and its regulation during in vitro maturation of human leukemic cell lines. Prh transcripts were not detected in leukemic cells of T-lymphoid lineage, irrespective of their maturation stage, and in resting or activated normal T cells from peripheral blood and lymphoid tissues. In contrast, high levels of Prh expression were shown in cells representing early stages of B lymphoid maturation, being maintained up to the level of circulating and tissue mature B cells. Terminal B-cell differentiation appeared t o be conversely as-

EMATOPOIESIS is a highly regulated process by which a small population of self-renewing primitive

progenitors generates different lineages of increasingly differentiated end cells with specific functional activities.' Such a process appears to be mediated by the transcriptional modulation of different sets of genes regulating proliferation and maturation of hematolymphopoietic cell^.*^^ In turn, gene expression is regulated by the interaction of transcription factors with DNA sequences regulating the expression of specific target genes4 Proteins encoded by homeobox-containing genes represent a class of transcription factors' involved in the regulation of gene activity during the embry- onic development of several species, including humans."'

Recently, several lines of evidence have indicated that homeobox-containing genes play an important role in growth and differentiation control of normal and malignant human hematopoietic cells." Homeobox genes of the human HOX A, HOX B, and HOX C loci" are expressed in hematopoietic cells in a lineage-specific fashion, whereas HOX D genes seem not to be expressed within the hematopoietic system.'*"' In particular, expression of genes belonging to the HOX B cluster appears restricted to cells with an early erythroid phenotype,".' transcription of certain HOX A genes is confined to myelomonocytic cell types,I4 and that of HOX C4 to lymphoid cells." Other homeobox genes distantly related to the HOX class of genes are also transcribed in normal and leukemic hematopoietic cells. HB24 and HB9 genes are constitutively expressed by CD34+ bone marrow progenitors, but not in more differentiated hematopoietic c e ] l ~ , ' ~ - * ~ whereas the aberrant regulation of HOXl I expression has been implicated in the leukemic transformation of human T- cell In addition, changes in the expression of homeobox-containing genes have been reported in normal and leukemic hematopoietic cells upon stimulation with growth factors such as interleukin-3 (IL-3),I9 and during induction of cell maturation by chemical agents.14." A direct involvement of homeobox genes in the acquisition of the differentiated phenotype in hematopoietic cells, has also

H

Blood, Vol 85, No 5 (March l), 1995: pp 1237-1245

sociated with the deactivation of the gene, since preplasmacytic and plasmocytoma cell lines were found not to express Prh mRNA. Prh transcripts were also shown in human cell lines of early myelomonocytic, erythromegakaryocytic, and preosteoclast phenotypes. Prh expression was lost upon in vitro differentiation of leukemic cell lines into mature monocyte-macrophages and megakaryocytes, whereas it was maintained or upregulated after induction of maturation to granulocytes and osteoclasts. Accordingly, circulating normal monocytes did not display Prh mRNA, which was conversely detected at high levels in purified normal granulocytes. Our data, which show that the acquisition of the differentiated phenotype is associated t o Prh downregulation in certain hematopoietic cells but not in others, also suggest that a dysregulated expression of this gene might contribute to the process of leukemogenesis within specific cell lineages. 0 1995 b y The American Society of Hematology.

been shown by altering the expression of such genes with specific antisense oligomers and/or expression vec-

tion of homeobox gene expression or their structural alteration, eg, by specific chromosomal translocations, may lead to aberrant cell proliferation and contribute to the process of leukem~genesis.'"~~~

By using a pair of redundant oligonucleotides in conjunc- tion with a polymerase chain reaction approach, we have isolated and cloned a novel homeobox-containing gene named proline-rich homeobox (Prh) from chicken, mouse, and human cDNA libraries.30 The homeodomain encoded by the Prh gene is capable of sequence-specific DNA binding and shows only 46% identity with that of Antennapedia, and 56% and 54% with those encoded by human HB24 and HOXll genes, re- spectively.ls.22 Therefore, Prh represents a novel type of homeobox gene that cannot be assigned to a particular subfamily

tors.20.24-28 Furthermore, it has been suggested that deregula-

From the Dipartimento di Biochimica, Biojisica e Chimica delle Macromolecole, Universita di Trieste, Trieste, Italy; Institute of Can- cer Research, Chester Beatty Laboratories, London, UK; Unita Op- erativa Leucemie e Trapianto di Midollo, Divisione di Oncologia Medica, Centro di Riferimento Oncologico, INRCCS, Aviano, Italy.

Submitted July 28, 1994; accepted October 26, 1994. Supported by Associazione Italiana per la Ricerca sul Cancro

(AIRC), Milano, Italy; Consiglio Nazionale per le Ricerche (CNR), Progetto Finalizzato-Applicazioni Cliniche della Ricerca Oncolo- gica n.92.02347.PF39, Roma, Italy; Minister0 dell'universita e della Ricerca Scienti$ca e Tecnologica, Roma, Italy (40%), and Uni- versita degli Studi di Trieste, Italy (60%). A.R. is an AIRC fellow.

Address reprint requests to Guidalberto Manjoletti, DSci, Dipar- timento di Biochimica, Biojisica e Chimica delle Macromolecole, Via L. Giorgieri I , 34127 Trieste, Italy.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. section 1734 solely to indicate this fact. 0 1995 by The American Society of Hematology. 0006-4971/95/8505-00I8$3.00/0

1237

1238 MANFIOLElTI ET AL

and whose expression in chicken tissues appears strictly restricted to hematopoietic cells, lung, and liver.”

To understand whether Prh homeobox gene might be involved in the regulation of human hematopoiesis, we have analyzed its constitutive expression in purified normal cells of lymphoid and myeloid lineages, its levels of transcription in a number of leukemic cell lines representing different lineages and stages of hematolymphopoietic differentiation, and its regulation upon induction of differentiation of leukemic cells in vitro.

MATERIALS AND METHODS

Cell lines and culture conditions. K562 (early myeloerythroid), HEL (myeloblastic-erythroblastic), KG-l and KG-IA (early myeloblasts), HL-60 (intermediate myeloid-promyelocytes), U937 (early monoblasts), ML-3 (myelomonoblasts), THP-I (monoblasts), H9 (T cells), Molt-4 (common thymocyte phenotype., T cells), and Jurkat (postthymic phenotype, T cells) human leukemic cell lines were obtained through the American Type Culture Collection (Rockville, MD). BV-173 (early B lymphoblasts derived from a patient with lymphoid blast crisis of chronic myelogenous leukemia), Ri-l (non- Hodgkin’s lymphoma; intermediatelmature phenotype B cells), Ci- 1 and Sc-l (non-Hodgkin’s lymphoma; preplasmacytic phenotype B cells) cell lines were obtained by Dr K.H. Th’ng (Hammersmith Hospital, London, UK).3’ Peripheral T cell lymphoma cell lines HUT 78 and HUT 102 were kindly provided by Dr A. Colombatti (Centro di Riferimento Oncologico, Aviano, Italy) and the M07e (megakaryoblasts) cell line by Prof L. Pegoraro (University of Turin, Italy). The NB-4 [leukemic promyelocytes harboring the t( 15; 17)] cell line was obtained through Dr P.G. Pelicci (University of Pemgia, Italy). Nalm-6 (early pre-B cells), MN-60 (sIg+ B-cell acute lymphoblastic leukemia), U-266, LP-l (plasmocytoma), and Karpas-299 (CD30+ T-cell anaplastic lymphoma) cell lines were obtained through the German Collection of Microorganisms and Cell Cultures (Braunsch- weig, Germany). The osteoclast progenitor cell line FLG 29.1 has been previously established and characterized by some of us.32 Cell lineages, differentiation stages and phenotypic characteristics of cell lines used in the present study are summarized in Tables 1 and 2. All cell lines were cultured in RPM1 1640 medium (GIBCO, Paisley, Scotland) supplemented with 10% of fetal calf serum (FCS), with the exception of KG- 1 A, KG-l, HEL, and THP- l , which were maintained in Iscove’s modified Dulbecco’s medium (IMDM) (GIBCO) plus 20% FCS. M07e cells were cultured in IMDM plus 5% FCS supplemented with 10 ng/mL GM-CSF.

Induction of cell differentiation. HL-60 cells were induced to differentiate toward monocyte-macrophage by incubation with 1 X 10” molk. of 12-0-tetradecanoylphorbol-13-acetate (TPA) (Sigma Chemical CO, St Louis, MO) and toward granulocytes by exposure to l X IO-‘ m o m all-trans retinoic acid (ATRA) (Sigma). NB-4 cells were induced to mature into granulocytes by I X IO-’ molk. ATRA. K562 and Mo7e cells acquired a megakaryocytic phenotype after exposure to 1 X molk. TPA, whereas KG-I cells were induced to differentiate toward monocyte-macrophages by I X 10 ” molk. TPA. Differentiation of Nalm-6 from early to late pre-B cells was induced by TPA (1 x 10”.~”-9 mollL). Cells were collected at different time points after exposure to inducers, up to 72 hours for TPA and up to 6 days for ATRA. The occurrence of cell maturation in vitro was measured by morphologic assessment of May-Griinwald Giemsa stained cytospin preparations, by evaluation of cell adherence to plastic and by modifications of the surface antigenic phenotype with monoclonal antibodies (MoAbs). Induction of terminal osteoclast differentiation by TPA in FLG 29.1 cells was assessed by functional and phenotypic changes as described elsewhere.’2 Mor-

phologic and phenotypic changes of leukemic cell lines after exposure to differentiation inducers are summarized in Tables 1 and 2.

Cell phenotyping andjow cytometry. Surface phenotypes of leuke- miaAymphoma cell lines were analyzed by indirect and direct immuno- fl~orescence~~ with MoAbs recognizing the following differentiation antigens: CDIO, CD19, CD20, CD22, CD24, CD34, CD37, CDlq CD3,CD4, CD8,CDllb,CDllc,CD14,CD15, CD71,CD38,CD4la, CD61, PCA-l, anti-Ig, anti-K, anti-L, anti-IgM. Sources and specifici- ties of the MoAbs used in this study have been reported in detail elsewhere.”.” Expression of cytoplasmic Ig (cyIg) was determined on cytospin cell preparations. Viable, antibody-labeled cells were identified according to their forward and right-angle scattering, electronically gated, and assayed for surface fluorescence on a FACScan cytometer (Becton Dickinson Immunocytometry Systems, San Jose, CA).

Cell isolation and purijication. Peripheral blood buffy coats were obtained from leukapheresis of normal donors. Separation on a Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) gradient yielded the total mononuclear cell fraction; granulocytes were further purified by I .2% dextran sedimentation followed by lysis of contaminating red blood cells. The purity of granulocytes (93% to 96%) was verified by morphologic assessment of May-Griinwald Giemsa-stained cytospin preparations. Adherent monocytes were purified from the mononuclear fraction by a two-step negative selection using anti- CD2-and anti-CD19-conjugated immunomagnetic beads (Dyna- beads; Dynal, Oslo, Norway) as previously described,’3 followed by a 2-hour incubation at 37°C in tissue culture dishes. Peripheral T and B lymphocytes were recovered from the CD2’ and CD19’ positively selected cell fractions. Tissue T and B lymphocytes were purified by tearing out single cells from freshly excised tonsilb. After gradient separation mononuclear cells were T-cell depleted by negative selection with anti-CD2-immunomagnetic beads. Nonad- herent (Adh-) cells were collected after an overnight culture and positively selected by anti-CD19 immunomagnetic beads. The purity of positively and negatively selected cell fractions was assessed by flow-cytometry using anti-CD3 (Leu 4), anti-CD19 (Leu 12), anti- CD20 (Leu 16), and anti-CD14 (Leu M3) MoAbs (Becton Dickin- son). B (CD20+) and T (CD37 lymphocytes and monocyte (CD14’) cell fractions appeared more than 95% pure.

T-cell activation. Activation of purified CD3’, CD19 , Adh T cells, obtained from peripheral blood by negative immunomagnetic selection with anti-CD1 9 MoAbs and 2 hours of incubation at 37”C,” was performed by exposing isolated T lymphocytes to TPA (10 ngl mL) and ionomycin A ( I .0 pg/mL) (Sigma), as described.” Periph- eral mononuclear cells (1 X 10‘ ImL) were stimulated in vitro upon incubation with 10 pglmL phytohemagglutinin-M (PHA) (GIBCO), and S0 pglml concanavalin A (Con A) (Sigma) as previously described.” Activation of Jurkat T cells was induced by TPA alone ( I x 10” mollL) or by incubating cells with PHA ( I pg/mL) and TPA (l X IO-’ mol/L).’R Cells were collected at different time points up to 72 hours for RNA extraction and phenotyping studies. T-cell activation was assessed by two-color flow cytometry with anti-CD3. anti-CD25, anti-CD38, anti-CD1 la, anti-CDS4, and anti-HLA-DR MoAbs”.” and by ’H-thymidine inc~rporation.‘~

RNA isolation and Northern blot analysis. Total RNA was extracted by the guanidine isothiocyanate method.” Twenty micro- grams of RNA per lane were size-fractionated on 1% agarose containing 6.7% formaldehyde and transferred onto nylon membranes (Genescreen TM, New England Nuclear, Boston, MA). Filters were then hybridized in 1 mol/L NaCl and 1% sodium dodecyl sulfate (SDS) at 68°C with I x 10‘ cpm/mL of random primed-labeled probes. After washings to a final stringency of 0.2 X standard sodium citrate and 0.1 % SDS at 65”C, filters were exposed to XAR-5 films (Eastman Kodak, Rochester, NY) at -80°C. Probes used were as follows: Prh, 1.2-kb Pst I-EcoRI cDNA fragment3’; c-myb, 0.7-kb EcoRI cDNA fragment (kindly provided by Dr M. Introna, Istituto

PROLINE-RICH HOMEOBOX IN HEMATOPOIETIC CELLS 1239

Table 1. Characteristics of Human Lymphoid Cell Lines U w d in the Study

PhenotvDic Markers'

Cell Line Differentiation Stage CD34 CD10 CD19 CD20 CD21 CD22 CD23 CD24 CD37 CD38 cylg Smlg

B lineage ~~

BV-173 Early pre-B + + + - - + Nalm-6 + Pre-B + + - + + + Nalm-6 + TPA Late pre-B ~ ? + + + + + - + +

Ri-l Mature B -

Ci-l Preplasmacytic - - + + + + + ? + + + -

- + - + - - - - - - - -

- - -

MN-60 Early B - c - + + - ND - + + - ND +t + + + + + +- +t +t - - -

sc- l Late preplasmacytic - + + + - - - - + + + + + + U-266 Plasma cell -

LP-1 Plasma cell -

- - - - - - ND - + + -* ? ND - - - - - - + ? -

CD7 CDla CD5 CD3 CD2 CD4 CD8 CD38

T lineage H9 T cell + + - - - - - ND Molt-4 Thymic + + - - t + + + Jurkat Postthymic + ? + + + + -

HUT 78 Mature T - - + + HUT 102 Mature T i + + + + Karpas-299 Mature T + - + - - + Abbreviations: CD, cluster of differentiation; Smlg, surface lg; ND, not done. * +, More than 50% of positive cells; 2, 25% to 50% of positive cells; -, less than 25% of positive cells. t IgM. * U-266 cells were also positive for the plasma cell-specific antigen, PCA-1.

- - + - - - - -

- - -

- ND

Mario Negri, Milano); glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 1.3-kb Pst I-Pst I cDNA fragment.

Densitometric measurements of autoradiographs were performed using a model 1650 Bio-Rad densitometer equipped with a C-R3A Shimadzu data module.

RESULTS

Characterization, stability and half-life of Prh mRNA. A major specific transcript of 1.7 kb was detected in total cellu-

lar RNA extracted from KG-1A cells upon Northern blot hybridization with the Rh cDNA probe (Fig 1A). In addition, one faint extra band of -4 kb was also detectable in some of the blots, most probably representing a Prh RNA precursor form. To establish whether inhibition of protein synthesis resulted in the alteration of Prh transcript levels, KG-1A cells were cultured for 1 and 2 hours in the presence of 10 pg/mL of cycloheximide (CHX). No changes in Prh mRNA levels were observed upon exposure of KG-l A cells

Table 2. Characteristics of Human Myeloid Cell Lines Used in the Study

Differentiation ~ ~ ~ ~ _ _ _ _

Cell Line Phenotype Inducer Differentiated Phenotype

KG-1A CD34' early myeloblasts - KG-l CD34' early myeloblasts TPA* Monocyte-macrophage (CD34-, CD1 lb', CDllc', Adh) HL-60 Myeloblasts-promyelocytes TPA Monocyte-macrophage (CDllb', CDllc', CD14', Adh)

-

ATRAt Granulocytes (CD15', CD1 lb', Morph*) NB-4 Promyelocytes; t(15; 17) ATRA Granulcotyes (CD15'. CDllb+, Morph*) U937 Early monoblasts ML-3 Myelomonoblasts

- -

THP-l Monoblasts K562 Early myeloerythroblasts TPA Megakaryocytes (CD4la'. CD61t, Morph§) HEL Myeloblasts-erythroblasts - Mo7e Megakaryoblasts TPA Megakaryocytes (CD4la+, CD61+, Morph§) FLG 29.1 Preosteoclasts TPA Osteoclasts (CD51'. CD61+. calcitonin-R+, bone resorption, Morphll)

- - - -

-

Abbreviations: Adh, adherent cells; Morph, morphological changes; calcitonin-R, calcitonin receptor. 1.0 x I O " mol/L for 72 hours.

t 1.0 X mol/L for 6 days. * Metamyelocytes, bands, and neutrophils. 8 Cytoplasmic enlargement, multinucleated cells. 11 Giant polykarion cells.

1240

CHX ActD

A

Prh -. c 28s

+ 18s

B

- 100

c E 60 W

.-

t 1/2 = 25 min

J " I

0 l 2 3 4

Time (hours)

Fig 1. Effects of actinomycin D and cycloheximide on Prh expression in KG-1A cells. (A ) Cells were treated with cycloheximide (10 pglmL) for 1 and 2 hours and with actinomycin D (5 pg/mL) for 0.5, 1, 2, and 4 hours. Total RNA was extracted and 20 pgllane were loaded and analyzed by Northern blotting for Prh expression (upper panel). Ethidium bromide staining is shown in the lower panel. (B) Diagram derived from densitometric quantitation of actinomycin D lanes shown in (A). Quantitation was performed by assuming the control sample (no actinomycin D) as 100% accumulation of Prh mRNA. Half-life was estimated by assuming an exponential rate of decay.

to CHX, which caused more than 90% inhibition of protein synthesis with respect to untreated cells (Fig IA). To deter- mine the half-life of Prh mRNA in KG-l A cells, total RNA synthesis was blocked by more than 95% upon addition of actinomycin D (5.0 pg/mL), and the levels of' Prh transcript were analyzed at various time intervals up to 4 hours (Fig IA). The estimated half-life of Prh mRNA, evaluated by densitometric scanning of the autoradiographs, was =25 minutes (Fig IB). Equal amounts of RNA were loaded as quantified by ethidium bromide staining (Fig IA).

Expression of the Prh gene in normal mature hematolym- phoid cells. The pattern of Prh gene expression in human normal hematopoietic cells is shown in Fig 2. A specific band of hybridization with the Prh cDNA probe was detected in purified CD19+ B cells from peripheral blood and tonsils of two different donors (a,b), and in circulating granulocytes.

MANFIOLETTI ET AL

In contrast, Prh transcripts were not detected in purified CD2' T lymphocytes from both peripheral blood and tonsils, and in adherent CD2-, CD19-, and CD14' monocytes recovered from peripheral blood (Fig 2).

Expression of the Prh gene in human leukemicdlymphoma cells of Iymphoidphenotype. Constitutive levels of mRNA encoded by the Prh gene were analyzed by Northern blot analysis of human leukemidlymphoma cell lines (Figs 3 and 4) representing different stages of B and T-cell differentiation (Table l ) .

In leukemidlymphoma, B cells the expression of the Prh gene appeared to be maintained up to the pre-plasma cell stage (Fig 3A). High levels of Prh transcripts were detected in cell lines of early pre-B (BV-173 ; CD34', CD38'. CD37-, CDIO', CD19', CD20-, cyIg-) and pre-B (Nalm- 6; CD34-, CD38', CD37-, CDIO', CD19', CD20-, cyIg') phenotypes. The triggering of phenotypic differentiation of Nalm-6 to late pre-B cells by TPA ( 1 .O X IO"' mol/L), as shown by the induction of CD20 and CD22, the increased expression of CD 19, and the significant reduction of surface CD10 and CD38 determinants (Table l ) , did not modify Prh mRNA levels (Fig 3B). Although more effective in inducing phenotypic differentiation and growth inhibition of Nalm-6

Peripheral blood Tonsils

- 28s

18s

Fig 2. Expression of the Prh gene in normal mature hematopoietic cells. Adherent (Adh'), CD2-, CD19-, CD14' monocytes were purified from peripheral blood by a two-step immunomagnetic selection. Nor- mal T and B lymphocytes were purified from peripheral blood and tonsil tissues of two different donors (a,b) by positive immunomagnetic selection with anti-CD2 and anti-CD19 MoAbs. Granulocytes were recovered by dextran sedimentation followed by erythrocyte lysis. In all cases, 20 pgllane of total RNA extracted from greater than 95% pure cell populations was probed by Northern blotting for Prh expression (upper panels). Total RNA from K562 cells was included as a positive control. Staining of ribosomal RNA on ethidium bromide gels confirms integrity of RNA and shows comparable RNA loading in single lanes (lower panels).


cells, exposure to increasing concentrations of TPA ( 1 .O X IO-* and 1.0 X IO-' mom) again did not result in significant modifications of Prh gene expression (not shown). Similarly, levels of Prh mRNA comparable with those present in immature B cells, were still observed in the early B phenotype cell line MN-60 (CD19', CDIO', CD20', CD37'. slgM', slgC-) and in the non-Hodgkin's lymphoma (NHL) cell line Ri-l, showing a mature B-cell phenotype (CD19'. CDIO-, CD20', CD21-, CD23-, CD24'. CD37', CD38', SmIg', cylg'). In contrast, terminal Bcell maturation appeared associated with the progressive deactivation of the Prh gene. Prh mRNA levels showed a significant reduction in the preplasmacytic cell line

A Prh -.

(U (D

Y m

Fig 4. Expression of the Prh gene in human lymphoid cells of T- lineage phenotype. Total RNA (20 pgllane) from human T-cell lines and activated normal T cells was analyzed by Northern blotting with a Prh probe. Jurkat T cells were activated by exposure to TPA (10" mol/L) or t o TPA 110" mollL) plus PHA (1 pg/mL). Normal peripheral

C- 28s T lymphocytes (CD3*, CD19-, Adh-) were activated by incubation with TPA (10 ng/mL) and ionomycin A (1.0 pg/mL) for 72 hours. In all of the experiments, the same blots were also probed with a - 1 8s GAPDH cDNA fragment, and total RNA from K562 cells was included as a positive control for Prh transcripts.

7 'R=-

1 Prh- W-=- a

B Nalm-6+TPA

cu

GAPDH -m

Fig 3. Expression of the Prh gene in human lymphoid cells of B- lineage phenotype. (A) Total RNA (20 pg/lane) from human Bcell lines was analyzed by Northern blotting with a Prh probe. (B) Expres-

Ci-l (CD19', CD20', CD21'. CD23', CD24', CD37'. CD38', SmIg-, cylg'), being no longer detectable in the late preplasma cell line Sc-l (CD19', CD20', CD21', CD23'. CD24-, CD37'. CD38+, SmIg-, cylg') and in two plasmocytoma cell lines U-266 and LP-l (CD19-, CD20-, CD24-, CD37-, CD38', PCA-I' [U-2661, SmIg-, cylg').

As opposed to B cells, no expression of the Prh gene was found in leukemidymphoma cell lines of T lineage, irrespective of their relative maturation stages (Table l ) , encom- passing T (H9), thymic (Molt-4), postthymic (Jurkat) and mature T cell (HUT 78, HUT 102, Karpas-299) phenotypes (Fig 4). Prh transcripts remained undetectable in leukemic Jurkat T cells upon in vitro activation for 12, 36, and 72 hours with either TPA or TPA plus PHA"." (Fig 4, and data not shown). Similarly, TPA and ionomycin A's stimulation did not induce Prh transcripts in purified peripheral CD3', CD19-, Adh- normal T cells, as shown by Northern blots performed at 12, 36, and 72 hours after activation (Fig 4, and data not shown). Additional experiments in which normal peripheral T lymphocytes were activated in vitro by PHA and Con A'" further confirmed that Prh mRNA remained undetectable by Northern blotting up to 72 hours after T-cell activation (data not shown). Activation of Jurkat and normal T cells was monitored by 'H-thymidine incorporation and by changes in expression of CD25. CD38. CD1 la. CD54. CD71 and anti-HLA-DR sur-

sion of Prh transcripts during TPA (lo" mol/L)-induced differentia- face antigens (not shown). t ion of Nalm-6 from early t o late p reB cells. In all of the experiments, the same blots were also probed with a GAPDH cDNA fragment, and In Of RNA were loaded as total RNA from K562 cells was included as a oositive control for Prh quantified by hybridization with a GAPDH probe (Figs transcripts. and 4).

1242 MANFIOLETTI ET AL

Prh-

7

.c - 28s

+- 18s

Fig 5. Expression of the Prh gene in human cells of myeloid lineage. Total RNA (20 pg/lane) from human myelomonocytic, erythroid, megakaryocytic, and osteoclast cell lines was extracted and analyzed by Northern blotting with a Prh probe. In all of the experiments, the same blots were also probed with a GAPDH cDNA fragment.

Expression of the Prh gene in human leukemic myeloid cells. As opposed to lymphoid cells, Prh expression in myeloid leukemic cell lines did not show any lineage restriction. As indicated by Northern blotting experiments (Fig 5). cell lines of myelo-granulocytic (KG-l A, KG- I , HL-60, NB-4). erythro-megakaryocytic (K562, HEL. Mo7e) and osteoclas- tic (FLG 29. I ) phenotypes (Table 2), displayed overall comparable constitutive levels of Prh mRNA. Among monocytic cell lines, THP-I expressed the lowest amount of Prh transcripts as compared with U937 and ML-3 (Fig 5). To analyze changes in Prh gene expression during myelopoietic differentiation, KG-l, HL-60, NB-4, K562, and Mo7e cells were exposed to maturation inducers (Table 2) and total cellular RNA was obtained at different time intervals after induction (Fig 6). TPA-induced differentiation of KG-l and HL-60 cells toward mature monocytic-macrophage cells, as shown by increased cell adherence, reduction of CD34 and or CD7 I expression (KG-l, HL-60) and enhanced expression of CD1 Ib, CD1 IC, and CD14 determinants (HL-60), was accompanied by a time-dependent reduction in Prh transcripts (Fig 6A). In contrast, induction of granulocytic differentiation by ATRA, as evidenced by morphologic changes and upregulation of CD15 and CD1 1 b antigens, did not cause any reduction of Prh mRNA levels in NB-4 promyelocytes, and even resulted in a slight increase of Prh transcripts in HL-60 cells (Fig 6B). Maturation of K562 and Mo7e cells toward a megakaryocytic phenotype, as shown by the induction of the CD4IdCD61 (IIb-IIIa) complex of surface glyco- proteins and morphologic changes, resulted in a dramatic reduction of Prh expression (Fig 6C). Finally, TPA-induced acquisition of a mature osteoclast phenotype (bone resorption. expression of functionally active calcitonin and vitronectin receptors) by FLG 29.1 preosteoclast cells, did not result in any change of Prh transcripts levels, which remained as high as in uninduced cells (data not shown).

In all cell lines, induction of in vitro differentiation by both TPA and ATRA was accompanied by a concurrent decrease of c-myh proto-oncogene transcripts (Fig 6). Ex- pression of this latter gene, known to be negatively regulated during maturation of myeloid cells,3x.'" was tested on the same filters used for Prh detection.

DISCUSSION

It has been shown that more than 20 different homeobox- containing genes are expressed within the hematopoietic tissues."'"' Several of these genes are active within the same

A KG-l +TPA HL-60 +TPA

a...

B HL-6O+ATRA NB-4 + ATRA

O N * W u u u

c-myb - - GAPDH-

- c

Mo7e +TPA

c-myb - .I Fig 6. Changes in Prh gene expression during in vitro-induced

differentiation of human myeloid cell lines. (A) KG-l and HL-60 cells were exposed to TPA (10" mol/L) to induce monocytic maturation. (B) HL-60 and NB-4 cells were exposed to ATRA (lo-' mol/L) to induce granulocytic maturation. (C) K562 and M07e cells were exposed t o TPA (10" mol/L) to induce the appearance of a megakaryocytic phenotype. Total RNA was extracted from uninduced and induced cells at the indicated times, and 20 p g of RNA were loaded in each lane and analyzed by Northern blot. The same blots were sequentially probed with Prh, c-myb, and GAPDH.


cell types and/or cellular lineages, often as multiple alterna- tively spliced transcript^.""^ Therefore, to gain additional insights into the functional significance of homeodomain genes redundancy in hemopoietic development, it appears crucial to define the exact pattern of homeobox gene expression in different blood cell lineages.

We have recently isolated a novel homeobox-containing gene named Prh from a cDNA library of the human early progenitor (CD34+) cell line KG-lA.30 Prh was subsequently located on human chromosome and sequence analysis has shown that its homeodomain is distinct from other non- class l homeodomain proteins such as HB24 (56% homology) and HOXl1 (54% homology), by lacking basic residues at position 3 and 5 in the minor groove binding N-terminal arm and by the presence of an arginine at position 7. These differences could confer Prh protein the ability to bind to and regulate a different set of target genes3'

As shown in the present study, Prh gene behavior within the human hematopoietic system was characterized by the lack of expression in T cells as detected by Northern blotting, and by a pattern of regulated expression during the maturation sequence of the other cellular lineages. Prh mRNA was shown in immature cell lines of myelomonocytic, erythromegakaryocytic, and B-lymphoid lineages. In particular, Prh transcripts were detected in cell lines representing early stages of B-lymphocyte development and were maintained up to the level of mature B cells. Conversely, preplasmacytic and plasmocytoma cell lines were found no longer to express Prh mRNA. In addition, we have shown that Prh homeobox expression was maintained or even increased with progres- sion of myeloid cell lines to mature granulocytes, whereas differentiation of leukemic cell lines into macrophages was accompanied by the deactivation of the gene. Accordingly, the lowest levels of Prh expression among monocytic cell lines were detected in those (THP-1) displaying the more mature phen~type.~' The Prh expression pattern seen with the cell lines was confirmed by the analysis of purified normal hematopoietic cells, indicating that the presence of Prh mRNA in leukemidlymphoma cell lines did not result from a dysregulated gene expression caused by neoplastic transformation. High levels of Prh transcripts were detected in mature normal B cells and in granulocytes, whereas T lymphocytes and monocytes were found not to express the gene. Therefore, at variance with genes such as HB24 and HB9,19.2" it appears that the deactivation of Prh does not represent an absolute prerequisite for lymphomyeloid differentiation to occur.

The absence of Prh expression in human normal and leukemic T cells is in agreement with previous results showing that both the chicken Prh homeobox gene" and its murine homologue hex:* were not detectable in leukemic T-cell lines, peripheral T lymphocytes, and thymic tissues. This finding represents a point of divergence with HB24 and HB9 because these latter genes have been reported to be transcribed in activated normal T lymphocytes and in subsets of malignant T cell^.'^^^^^^' In contrast, we have shown here that activation of Jurkat T cells and peripheral T lymphocytes by different stimuli did not result in any induction of Prh mRNA. However, the use of more sensitive techniques, such

as reverse-transcriptase polymerase chain reaction or RNase protection assays, is required to rule out the presence in T cells of small amounts of Prh mRNA undetectable by North- ern blotting.

The consistent expression of Prh in B cells and monocyte progenitors along with its deactivation once terminal maturation has occurred provides a further biologic link between B lymphoid and monocytic lineages. Within the hematopoietic system, the supposed early divergence between cells committed to myelomonocytic and B-cell differentiation has been rechallenged by several evidences. Extensive immuno- phenotyping studies of lymphoid leukemias have shown that the expression of myelomonocytic antigens is mostly restricted to acute and chronic leukemias of B-cell derivation.44,45 More intriguingly, the existence of early bipotent progenitors capable of generating both B cells and macro-

and of leukemic and normal B committed precursors coexpressing monocytic antigens4' has been shown in mice and humans. Therefore Prh may contribute to the physi- ologic development of both these cell lineages and its dereg- ulated expression may be relevant to the process of leukemogenesis in B cells and monocytes.

The differential regulation of Prh during monocytic and osteoclast differentiation also appears of interest. Although osteoclasts have been suggested to derive from hematopoietic progenitors closely related to the monocyte-macrophage lineage and show several common features with tissue macrophage~,'",~~ little is known about genes regulating osteoclast-specific functions. Our findings that Prh expression is lost in mature monocyte-macrophages and maintained in differentiated osteoclasts, provide an additional tool to separate these related cell types.

The downregulation of Prh during terminal differentiation of cells of B-lymphoid, monocytic, and megakaryocytic lineages may be related to the recent finding that Prh may act as a transcription repressor (Jayaraman et al, manuscript in preparation). Thus, one possibility is that Prh represses genes required for terminal maturation to occur in cells of the above lineages, and therefore, its downregulation is necessary for such cells to differentiate. Conversely, a sustained expression of Prh might be needed for terminal maturation of granulocytes. As a consequence, an abnormally maintained expression of Prh may contribute to the process of leukemogenesis within certain hematopoietic lineages, whereas its repression may be associated to neoplastic transformation in others, as suggested for different homeodomain

Even though the target genes for Prh in human cells have not been presently identified, by using its DNA binding consensus sequence,3" a preliminary search among genes that could be potential targets for Prh homeodomain protein has been performed. A number of genes turned out to contain in their promoter a potential binding site for Rh. Interestingly, among these there are molecules able to regulate the biologic behavior of neoplastic hematopoietic cells and whose expression in leukemias and lymphomas has been shown to be of clinical rele~ance.~" In particular, consensus sequences able to bind Prh were found in the promoter regions of B-cell antigen genes regulating ion fluxes like CD20,5' surface molecules implicated in cell adhesion5* such as CD1 1b,53,54

1244 MANFIOLETTI ET AL

CD53, CD56/N-CAM,''.'0 CD106N-CAM- 1 ,57 and cytokines like IL-4""'" and IL-5.5y~.h0 Experiments aimed at evalu- ating the effects of enforced Prh expression on the biologic properties of hematopoietic cells and a detailed survey of Prh expression in human hematologic malignancies are currently ongoing.

ACKNOWLEDGMENT

We thank Dr M. lntrona for the c-nzyh cDNA probe; physicians of the Department of Otorhinolaryngology of the Children Hospital Burlo-Garofolo (Trieste, Italy) for providing tonsil tissues: col- leagues of Blood Transfusion Centers (Trieste Hospital and CRO, Aviano, Italy) for kindly providing peripheral blood apheresis prod- ucts; and Cinzia Borghese for excellent technical assistance.

REFERENCES

I . Dexter TM. Spooncer E: Growth and differentiation in the hemopoietic system. Ann Rev Cell Biol 3:423, 1987

2. Sachs L: The molecular control of blood cell development. Science 238:1374, 1987

3. Lubbert M, Herrmann F. Koeffler HP: Expression and regulation of myeloid-specific genes in normal and leukemic myeloid cells. Blood 77:909, 1991

4. Sachs L: The molecular regulators of normal and leukemic blood cells. Proc R Soc Lond 231:289, 1987

S. Levine M, Hoey T: Homeobox proteins as sequence specitic transcription factors. Cell 55537, 1988

6. Gehring WJ, Hiromi Y: Homeotic genes and the homeobox. Ann Rev Genet 20: 147, 1986

7. Scott MP, Tamkun JW, Hartzell GW 111: The structure and function of the homeodomain. Biochem Biophys Acta 989:25, 1989

8. Laughon A: DNA binding specificity of the homeodomain. Biochemistry 30: I 1357, 1991

9. Mavilio F: Regulation of vertebrate homeobox-containing genes by morphogens. Eur J Biochem 212:273, 1993

10. Lawrence HJ, Largman C: Homeobox genes in normal hematopoiesis and leukemia. Blood 80:2445, 1993

1 I , Scott MP: Vertebrate homeobox gene nomenclature. Cell 71:551. 1992

12. Shen WF, Largman C. Lowney P, Corral JC. Detmer K, Hauser CA, Simonitch TA, Hack FM, Lawrence HJ: Lineage-restricted expression of homeobox-containing genes in human hematopoietic cell lines. Proc Natl Acad Sci USA 86:8536, 1989

13. Mathews CHE, Detmerk K, Boncinelli E, Lawrence HJ. Larg- man C: Erythroid-restricted expression of homeobox genes of the human HOX 2 locus. Blood 78:2248, 1991

14. Vieille-Grosjean, Roullot V, Courtois G: Lineage and stage specific expression of HOX 1 genes in the human hematopoietic system. Biochem Biophys Res Commun 183:1124, 1992

15. Kongsuwan K, Webb E, Housiaux P, Adams JM: Expression of multiple homeobox genes within diverse mammalian haemopoi- etic lineages. EMBO J 7:2131, 1988

16. Magli MC, Barba P. Celetti A, De Vita G, Cillo C, Boncinelli E: Coordinate regulation of HOX genes in human hematopoietic cells. Proc Natl Acad Sci USA 88:6348, 1991

17. Lawrence HJ, Stage KM, Mathews CHE, Detmer K, Sciebien- ski R, MacKenzie M. Migliaccio E, Boncinelli E, Largman C: Ex- pression of HOX C homeobox genes in lymphoid cells. Cell Growth Diff 4565. 1993

18. Deguchi Y, Moroney JF, Wilson GL, Fox CH, Winter HS, Kehrl JH: Cloning of a human homeobox gene that resembles a diverged Drosophila homeobox gene and is expressed in activated lymphocytes. New Biol 3:353, 1991

19. Deguchi Y, Kehrl JH: Selective expression of two homeobox genes in CD34-positive cells from human bone marrow. Blood 78:323, 1991

20. Deguchi Y, Kirschembaum A, Kehrl JH: A diverged homeobox gene is involved in the proliferation and lineage committment of human hematopoietic progenitors and highly expressed in acute myelogenous leukemia. Blood 79:2841, 1992

21. Hatano M, Roberts CWM, Minden M, Crist WM, Korsmeyer SJ: Deregulation of a homeobox gene, HOX I I , by the t( I O : 14) in T cell leukemia. Science 253:79, 1991

22. Kennedy MA, Gonzalez-Sarmiento R. Kees UR, Lampert F. Dear N, Boehm T, Rabbitts TH: HOXl I , a homeobox-containing T-cell oncogene in human chromosome 10q24. Proc Natl Acad Sci USA 88:8900, 1991

23. Lu M, Gong Z , Shen W, Ho AD: The tcl-3 proto-oncogene altered in chromosomal translocation in T-cell leukemia codes for a homeobox protein. EMBO J 10:2905, 1991

24. Shen WF, Detmer K, Mathews CHE. Hack FM, Morgan DA. Largman C, Lawrence HJ: Modulation of homeobox gene expression alters the phenotype of human hematopoietic cell lines. EMBO J I I :983. 1992

25. Takeshita K, Bollekens JA, Hijiya N. Ratajcazak M, Ruddle FH, Gewirtz A: A homeobox gene of the Antennapedia class is required for human adult erythropoiesis. Proc Natl Acad Sci USA 9 0 3 3 5 . 1993

26. Blatt C, Lotem J, Sachs L: Inhibition of specific pathways of myeloid cell differentiation by an activated Hox-2.4 horneobox gene. Cell Growth Differ 3:671. 1992

27. Deguchi Y, Thevenin C. Kehrl J H: Stable expression o f HB24, a diverged human homeobox gene, i n T lymphocytes induces genes involved in T cell activation and growth. J Biol Chern 2673222, 1992

28. Deguchi Y. Agus D, Kehrl JH: A human homeobox gene, HB24. inhibits development of CD4+ T cells and impairs thymic involution in transgenic mice. J Biol Chem 268:3646. 1993

29. Maulbecker CC, Gruss P: The oncogenic potential of deregu- lated homeobox genes. Cell Growth Differ 4:431. 1993

30. Crompton MR. Bartlett TJ. MacGregor AD, Manfioletti G. Buratti E, Giancotti V, Goodwin GH: Identification of a novel homeobox gene expressed in haematopoietic cells. Nucleic Acids Re\ 20566 I . 1992

3 l . Th'ng KH, Garewal G, Kearney L, Rassool F, Melo JV, White H, Catovski D, Foroni L, Luzzato L, Goldman JM: Establishment and characterization of three new malignant lymphoid cell lines. In t J Cancer 39:X9, 1987

32. Gattei V, Bernabei PA, Pinto A, Bezzini R, Ringressi A. Formigli L, Tanini A. Attadia V, Brandi ML: Phorbol ester induced osteoclast-like differentiation of a novel human leukemic cell line (FLG 29.1). J Cell Biol I16:437, 1992

33. Carbone A, Gloghini A, Volpe R, Attadia V, Gattei V. Zago- ne1 V. Pinto A: Follicular lymphoma of compartmentalized small cleaved center cells and mantle zone lymphocytes. Evidences for a common derivation. Am J Clin Pathol 98:437, 1992

34. Pinto A, Gloghini A, Gattei V. Aldinucci D, Zagonel V. Carbone A: Expression of the c-kit receptor in human lymphomas is restricted to Hodgkin's disease and CD30+ anaplastic large cell lymphomas. Blood 83:785, 1994

35. Lane P, Traunecker A, Hubele S, lnui S. Lanzavecchia A. Gray D: Activated human T cells express a ligand for the human B- cell-associated antigen CD40 which participates in T cell-dependent activation of B lymphocytes. Eur J Immunol 22:2573. 1992

36. Maio M, Tessitori G , Pinto A, Temponi M. Colomhatti A. Ferrone S: Differential role of distinct determinants of intercellular adhesion molecule-l in immunologic phenomena. J Immunol 143:IXI. 1989


37. Chomczynsky P, Sacchi N: Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156, 1987

38. Emilia G, Donelli A, Ferrari S, Torelli U, Selleri L, Zucchini L, Moretti L, Venturelli D, Ceccherelli G, Torelli G: Cellular levels of mRNA from c-myc, c-myb and c-fes oncogenes in normal myeloid and erythroid precursors of human bone marrow: An in situ hybrid- isation study. Br J Haematol 62:287, 1986

39. Kastan M, Slamon DJ, Civin CI: Expression of proto-oncogene c-myb in normal human haematopoietic cells. Blood 73: 1444, I989

40. Hromas R, Radich J, Collins S: PCR cloning of an orphan homeobox gene (prh) preferentially expressed in myeloid and liver cells. Biochem Biophys Res Comm 195:976, 1993

41, Abrink M, Gob1 AE, Huang R, Nilsson K, Hellman L: Human cell lines U-937, THP-l and Mono Mac 6 represent relatively immature cells of the monocyte-macrophage cell lineage. Leukemia 8:1579, 1994

42. Bedford FK, Ashworth A, Enver T, and Wiedemann LM: HEX: A novel homeobox gene expressed during haematopoiesis and conserved between mouse and human. Nucleic Acids Res 21:1245, 1993

43. Deguchi Y, Yamanaka Y, Theodossiu C, Najfeld V, Kehrl JH: High expression of two diverged homeobox genes, HB24 and HB9, in acute leukemias: Molecular markers of hematopoietic cell immaturity. Leukemia 7:446, 1993

44. Drexler HG, Thiel E, Ludwig WD: Review of the incidence and clinical relevance of myeloid antigen-positive acute lymphoblastic leukemia. Leukemia 5:637, 1991

45. Pinto A, Zagonel V, Carbone A, Del Vecchio L: Expression of myeloid associated antigens in chronic lymphocytic leukemia. Br J Haematol 82: 173, 1992

46. Cumano A, Paige CJ, Iscove NN, Brady G: Bipotential precursors of B cells and macrophages in murine fetal liver. Nature 536:612, 1992

47. Uckun FM: Regulation of B-cell ontogeny. Blood 76:1908, I990

48. Kurihara N, Suda T, Miura Y, Nakauchi H, Kodama H, Hakeda Y, Kumegawa M: Generation of osteoclasts from isolated hematopoietic progenitor cells. Blood 74: 1295, 1989

49. Udagawa N, Takahashi N, Akatsu T, Tanaka H, Sasaki T, Nishijima T, Koga T, Martin TJ, Suda T: Origin of osteoclasts: Mature monocytes and macrophages are capable of differentiating into osteoclasts under a suitable microenvironment prepared by bone

marrow-derived stromal cells. h o c Natl Acad Sci USA 87:7260, 1990

50. Pinto A, Gattei V, Soligo D, Parravicini C, Del Vecchio L: New molecules burst at the leukocyte surface. Leukemia 8:347, 1994

51. Bubien JK, Bell PD, Frizzell RA, Tedder TF: CD20 directly regulates transmembrane ion flux in B lymphocytes, in Knapp W, Dorken B, Gilks WR, Rieber EP, Schmidt RE, Stein H, Von Dem Borne AEGKr (eds): Leucocyte Typing IV. White Cell Differentia- tion Antigens. Oxfrod, UK, Oxford University, 1989, p51

52. Hochman J, Levy E, Mador N, Gottesmann MM, Shearer GM, Okon E: Cell adhesiveness is related to tumorigenicity in malignant lymphoid cells. J Cell Biol 99:1282, 1984

53. Bradstock K, Matthews J, Benson E, Page F, Bishop J, Aus- tralian Leukemia Study Group: Prognostic value of immunopheno- typing in acute myeloid leukemia. Blood 84:1220, 1994

54. Mittelman M, Karcher DS, Kammerman LA, Lessin LS: High Ia (HLA-DR) and low CD1 lb (Mol ) expression may predict early conversion to leukemia in myelodysplasic syndromes. Am J Hematol 43:165, 1993

55. Vidriales MB, Orfao A, Gonzalez M, Hernandez JM, Lopez- Berges MC, Garcia MA, Canizio MC, Caballero MD, Macedo A, Landolfi C, San Miguel J F Expression of NK and lymphoid-associated antigens in blast cells of acute myeloblastic leukemia. Leukemia 7:2026, 1993

56. Kern WF, Spier CM, Hannemann EH, Miller TP, Matzner M, Grogan TM: Neural cell adhesion molecule-positive peripheral T-cell lymphoma: A rare variant with a propensity for unusual sites of involvement. Blood 79:2432, 1992

57. Freedman AS, Munro JM, Morimoto C, McIntyre BW, Rhy- nart, Lee N, Nadler LM: Follicular non-Hodgkin’s lymphoma cell adhesion to normal germinal centers and neoplastic follicles involves very late antigen-4 and vascular adhesion molecure-l. Blood 79:206, 1992

58. Punnonen J, Aversa G, de Vries JE: Human pre-B cells differentiate into Ig-secreting plasma cells in the presence of interleukin- 4 and activated CD4+ T cells and their membranes. Blood 82:2871, 1993

59. Masuda M, Hoshino S, Motoji M, Oshimi K, Mizoguchi H: Effects of various cytokines on proliferation of acute lymphoblastic leukemia cells. Leukemia Res 14:533, 1989

60. Touw I, Donath J, Pouwels K, van Bitenen C, Schipper P, Santini V, Hagmejer A, Lowenberg B, Delwel R: Acute myeloid leukemias with chromosomal abnormalities involving the 21q22 re- gion identified by their in vitro responsiveness to interleukin-5. Leu- kemia 5:687, 1991

Documents

Differential expression of a novel proline-rich homeobox gene (Prh) in human hematolymphopoietic cells