Vol. 7, 961-967, July 1996 Cell Growth & Differentiation 961

Colonic Expression of c-myb Is Initiated in Utero andContinues throughout Adult Life1

Mark A. Rosenthal, M. Anne Thompson, Sarah Ellis,Robert H. Whitehead, and Robert G. Ramsa?

Ludwig Institute for Cancer Research, Tumor Biology Branch,Melbourne, Victoria, Australia 3050, [M. A. A., M. A. T., A. H. W.,A. G. A.], and Peter MacCallurn Cancer Institute, Locked Bag 1 , St.Andrews Place, East Melbourne, Victoria, 3002 [5. E., A. G. A.],Australia

Abstract

c-myb expression is an indicator of hemopoietic cellproliferation and its down-regulation occurs as an earlyevent in cellular differentiation in vitro. We havepreviously demonstrated c-Myb protein expression innormal adult colon and tumor-derived colonic cell linesand have also shown that the proliferation of coloncarcinoma cell lines is c-myb dependent. This studyused the techniques of RNase protection andimmunohistochemistry to define c-myb mRNA andprotein expression in the normal adult mouse andmouse embryos, with special focus on the colon.These experiments revealed wide-spread c-myb mRNAexpression in mouse embryos includingnonhemopoietic tissues, and developmental timecourse studies revealed alterations in organ-specificc-myb mRNA expression. Immunohistochemicalstudies confirmed the embryonal expression of c-Mybin concordance with the mRNA data as well as c-Mybexpression throughout the length of the adult mousecobonic crypt. In contrast, staining for the proliferatingcell nuclear antigen was confined to the lower onethird of the crypt. In addition, c-Myb staining extendedbeyond that of the proliferating cell nuclear antigenwithin the germinal centers of the spleen. These datasuggest that c-myb: (a) has a role in the embryonicdevelopment of nonhemopoietic organs; (b) plays aspecific role in the biology of the normal adult coboniccrypt; and (c) expression is not inextricably linked toproliferation in vivo.

IntroductionThe transcription factor c-Myb has a well-defined role in thedifferentiation and proliferation of immature hemopoieticcells (1 , 2). c-myb expression has been considered restrictedto hernopoietic tissues such as the bone marrow, thymus,

Received 2/1/96; revised 5/1/96; accepted 5/3/96.The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 1 8 U.S.C. Section 1 734 solely to mdi-cate this fact.1 This work was supported in part by a NH&MAC postgraduate scholar-ship (to M. A. A.) and a Research Fellowship (to A. G. A.).2 To whom requests for reprints should be addressed. Phone: 61 -3-96561863; Fax: 61-3-9656 141 1 ; E-mail: robr@petermac.unimelb.edu.au.

and spleen. However, c-myb mRNA or protein expressionhas also been noted in a number of human tumor types andcell lines, including breast, lung, and colon carcinomas, os-teogenic sarcomas, neuroblastomas, and mebanomas (3).

We have previously reported p80�m�� protein expressionin normal human cobonic crypt epithelium (4) and both c-mybmRNA and protein have been detected at elevated levels inpremalignant and malignant cobonic epithelium (2, 4-6). Thisraised the possibility that c-myb may participate in normalcobonic crypt biology as well as colon carcinogenesis. Fur-thermore, studies using c-myb antisense oligonucbeotides onhuman colonic carcinoma cell lines indicate that c-Myb may

regulate cobonic epithelial cell line proliferation, and there is

some evidence of down-regulation during differentiation (4,7). As a consequence, we were interested in exploring thepossible role of c-Myb in developing colonic epithelium aswell as its role in normal adult cobonic epithelium.

The developmental and functional role that c-Myb mayhave within nonhernopoietic organs is ill-defined. There isevidence that c-myb mRNA is expressed in mouse embry-onal cells (8) and is widely expressed throughout the chickenembryo (9). The crucial role of c-Myb in embryonic develop-ment was evident in the analysis of c-myb nullizygous mutantmice which revealed a failure of adult-type erythropoiesis,resulting in death at day 1 4.5 of embryogenesis (1 0). How-

ever, the mouse embryonal expression pattern of c-myb hasnot been extensively documented.

We have used RNase protection and immunohistochem-istry to study the spatial and temporal expression of c-myb in

the developing mouse embryo as well as across a range ofadult mouse tissues. Our results demonstrate extensivec-myb expression during embryonal gestation and expres-sion within the adult cobonic crypt. We suggest that c-mybexpression may play an important role throughout all stagesof colon development and function.

Resultsc-myb mRNA Expression in Embryonal Tissues and theAdult Mouse. c-myb expression has previously been dem-onstrated in a number of tissues and cell lines. The focus onexpression has centered on tissues containing immaturehemopoietic cells (2). Specifically, c-myb is expressed inprecursor myeloid and lymphoid cells and in the case of themouse, erythroid precursor cells, fetal liver, and yolk sac (1).Results from previous studies of c-myb expression in non-hemopoietic cell lines and tissues initiated our explorationinto the nature of c-myb mRNA expression within the mouseembryo and adult organs (3).

RNase protection assays were performed to allow quan-tification of c-myb mRNA expression in nonhemopoietic or-gans across a range of ages including the mouse embryo. Inparticular, we were interested in documenting an apparenttrend which suggested that specific organ expression of

AProbes Tissues

c-rnyb FR lB AB FC 7C AC FL 7L AL

� _,� �

;� �: i.:: ;a

Bi.o

0.8

0

� 0.6EN

.0 0.4>,E

C) 0.2

An

Pu! ,� ,I�

FD7A F D7ABrain Colon

Mouse Organ

Fig. 1 . A, analysis of c-myb expression in mouse brain (B), colon (C), andliver (L) in the mouse embryo (day 15, F), day 7 pups (7), and adult mice(A) by RNase protection oftissue RNA. c-myb and I3�M probes resulted inprotected fragments of 251 and a doublet at 120 bp, respectively. Left,radiolabeled ANA probes; right, control ANA lane shows RNase protectionof a sense riboprobe template generated from a munne c-myb cDNA. B,ratio of c-myb:�32M expression within brain, colon, and liver tissues acrossthe following age groups: embryos (day 15), 7-day-old pups, and adultmice.

c-myb becomes restricted during mouse organ ontogeny.

This interest developed from our in situ hybridization data

(data not shown) that demonstrated c-myb mRNA in nonhe-

mopoietic embryonal tissues known not to express c-myb

mRNA by adulthood (3).

The technique of RNase protection allowed a sensitive,

specific, and quantifiable assessment of c-myb mRNA ex-

pression across a range of mouse organs (brain, colon, and

liver) at three developmental stages: embryos (day 15),

7-day-old pups, and adults. Fig. 1A shows the level of c-myb

mRNA detected in 20 p.g of total RNA preparations obtained

from pools of mouse embryo organs and adult organs. The

expression of f32M3 mRNA served as an internal measure for

RNA recovery and integrity. c-myb expression resulted in a

predicted 251 -bp protected fragment, whereas �2M expres-

962 c-myb Expression in Colon

3 The abbreviations used are: �2M, �2-rnicroglobulin; PCNA, proliferatingcell nuclear antigen; PAS, periodic-acid Schiffs; TBS, Tris-buffered saline.

sion resulted in a predicted doublet at l20-bp. These data

Control were then used to quantitatively assess c-myb expression byRNA determining a c-myb:p2M ratio shown in Fig. lB. The use of

f32M expression as a control can be complicated by the

recognition that f32M expression may vary from tissue to

#{248}_4- c-myb tissue (3) and conceivably from one developmental stage toanother.

Embryonal (day 1 5) c-myb-specific mRNA expression was

, identified in brain, colon, and liver tissues. c-myb mRNA was

also identified in the brains, colons, and livers of 7-day-oldmice albeit at reduced levels in the brain and liver. In con-

trast, adult expression of c-myb mRNA in brain and liver

4- �M tissues fell to almost undetectable levels, whereas colon

expression of c-myb mRNA remained high. These findings

were consistent with our previous data obtained from RNase

protection assays of adult mouse tissue mRNA (3). Further-

more, we have also been able to demonstrate full-length

c-myb mRNA expression using Northern blot analysis in the

following fetal tissues: brain, colon, kidney, and liver (data

not shown).

Fig. lB illustrates the temporal nature of organ-specific

change in c-myb mRNA expression levels. With increasing

age, the level of c-myb expression falls in relation to the levelof �32M expression within particular organs as determined byquantitative densitometry using a Phosphorlmager. Although

the fall in c-myb expression is pronounced in the brain andliver over the time course studied, a constantly high level of

F D7 A c-myb mRNA is expressed within the colon from the embryoLiver to adulthood.

c-Myb Expression Revealed by Immunohistochemis-try. To expand on the observations made for c-myb mRNA

expression, we performed immunohistochemical analyses

on paraffin-embedded embryonal sections. These immuno-

histochemical studies revealed a high level of concordance

between c-Myb expression and our mRNA expression data.Within the mouse embryo, c-Myb expression is widespread

and includes the brain, gastrointestinal tract, and liver. In

particular, immunohistochemistry of embryonal (El 8) colon

revealed high levels of c-Myb expression within the nuclei of

crypt cells (Fig. 2A). In addition, c-Myb expression levels

were still high within some embryonal (day 18) liver cells (Fig.

2B). Morphological examination of cells expressing c-Myb

within the embryonal liver revealed that some of them were

epithelial in nature. Fig. 2, C and D, represent matchedsections of embryonal colon and liver, respectively, that were

exposed to nonimmune serum.c-Myb Expression within the Adult Mouse Colon. We

decided to explore further the nature of c-Myb expression

within the colon. Having defined that p75�mYb protein was

expressed within mouse cobonic crypts using Western blot-

ting (data not shown), we elected to identify the specific

expression pattern of c-Myb within the crypt with the tech-

nique of immunohistochemistry.

Immunohistochemical studies of adult mouse colon sec-

tions consistently revealed c-Myb expression throughout in-

dividual crypts from the base of the crypts to the luminal

surface (Fig. 2E). This indicated that the epithelial cell ex-pression of c-Myb was not confined to the recognized regionof proliferation which occupies the bottom third of the crypt.

.

B�* �.�.

,t.

‘

.�

�.

�e �.;��t:� #{149}#{149}

,b.�

�

0’-#{176}.’

.e

.1

S’s

a �

�: � .� � �

�. � .. � *.� 4�

Cell Growth & Differentiation 963

.�

,‘. v� , � :�., �� #{176}� # * �.

z� � � .

#{176} �, - ,, ‘�‘:‘#{149}� ..

S � �

., .::�‘ � .

. �, � P. � . � �

‘� . ., . . �‘* � . � 1�.� � �

..:#{176}�. ::���‘i �

D#{149}��

.#{176}�

#{149}.. * I.

.�.,‘ �

.� ... :

�#{149}�%

I’

,1&

.� 4.�

a

� � �.

p:”� #{149}

�Ea:�IE

, :�%, . �

�

� ‘�

� �

J. � i-a�. 4�.

44.j.

Fig. 2. A-D, immunohistochemical staining for c-Myb protein in normal mouse embryonal tissues. Paraffin sections of normal mouse embryo colon (A andC) and liver (B and D) were incubated with anti-Myb rabbit serum (A and B) and nonimmune rabbit antiserum (C and D) revealing c-Myb within the nucleiof cells found in the embryonal gut and liver. E-G, immunohistochemical analysis of adult murine colonic crypts. Sequential paraffin sections were incubatedwith anti-Myb rabbit serum (E), nonimmune rabbit serum (F), and anti-PCNA mouse monoclonal antibody (G) revealing c-myb expression throughout thelength of the crypt in comparison to the expression of PCNA which is limited to the bottom third.

964 c-myb Expression in Colon

In accordance, PCNA staining of parallel longitudinal sec-tions confirmed that this proliferative zone was limited to thebottom third of the crypt (Fig. 2G). A preimmune controlsection is shown in Fig. 2F, indicating the specificity of theantisera used. In addition, this section contains a number ofdarkly staining mitotic figures within the lumen of the crypts,further highlighting the relationship between the region of

PCNA staining and cell cycling. Staining of these sectionswith mouse ascites as a control also produced a negligiblesignal. These data suggest that c-Myb expression extendsbeyond the proliferative zone within the colonic crypt.

It was noted that not all crypt cell nuclei expressed c-Mybantigen. This was best demonstrated by examination of cryptcross-sections stained with c-Myb, hematoxylin, and PASreagent, which detects mucins associated with intestinalgoblet cells (Fig. 3A). Morphological examination of sections

suggest that c-Myb is expressed in columnar cells. In con-trast, intensely staining PAS-positive cells with the morpho-logical appearance of fully differentiated goblet cells in gen-oral do not appear to express c-Myb. A higher powermagnification (x 1000) emphasizing this point is shown in Fig.3B.

We also surveyed c-Myb expression within the germinalcenters of adult mouse spleen. Fig. 3C and E demonstratesc-Myb expression within a typical germinal center and showsthat the signal is intense and extends beyond the center ofthe germinal center. In contrast, Fig. 3D and F demonstratesthe extent of PCNA expression within a matched section. Inthis case, as in the cobonic crypts, the extent of PCNA ex-pression appears to be more confined in comparison toc-Myb expression.

DiscussionThe role of c-myb in the development and function of non-

hemopoietic organs has not been well defined. This con-trasts with the substantial research into the role played byc-myb with respect to hemopoietic cell proliferation anddifferentiation. Our interest in nonhemopoietic organs stemsfrom evidence that c-Myb protein is expressed in normal

human cobonic crypt epitheliurn as well as being overex-pressed in malignant cobonic epithelium (4).

We have examined c-myb expression in developing non-hemopoietic organs through the techniques of RNase pro-tection and immunohistochemistry, focusing on the expres-sion of c-Myb within murine cobonic crypts. These modalitiesdemonstrate that c-myb mRNA and protein are widely ex-pressed in the developing mouse embryo. This expands onthe in situ hybridization data of Ess et a!. (1 1), which focused

on c-myb expression within the embryonal thymus and liver,and Sitzmann et aL (1 2), which demonstrated c-myb expres-sion in nonhemopoietic embryonal tissues.

The expression of c-myb is not static during mouse do-vebopment, indeed mRNA expression in at least two organs

(brain and liver) falls dramatically with age to the point that

expression levels are very low by adulthood. We also haveevidence suggesting that the same fall in c-myb expressionoccurs in the kidney (data not shown). This progressive fall inexpression suggests that c-myb may be important in thedeveloping mouse embryo but, in most organs (except co-

Ion), its expression is less critical or unimportant by adult-hood.

These data extend our knowledge of c-myb expressionpatterns. Furthermore, these observations suggest that theneed for c-myb expression is more extensive than indicated

by the analysis of c-myb nullizygous mutant mice which dieby day 14.5 in utero from an apparent failure of fetal livererythropoiesis (1 0). Intriguingly, there was no overt nonhe-mopoietic organ abnormality noted in these mice suggestingthat c-myb is not crucial in the development of nonhemo-poietic organs, at least up to day 14.5 of embryonic life.However, it is noteworthy that the embryo may not requirefunctioning gut for viability until or after birth (13). Indeed,extensive gastrointestinal modeling and nervous system pro-gramming occur during the last 4-5 days of murine fetalgestation (1 3). In contrast, fetal liver erythropoiesis is abso-lutely critical for embryonic viability.

We suggest that c-myb may play an important role inembryonal organogenesis beyond the role in hemopoiesisimplied by the c-myb nullizygous mutant mice. These miceclearly require c-Myb to attain adult-type erythropoiesis. The

apparent lack of nonhemopoietic organ disturbance mayreflect that such an event arises after day 14.5. As a conse-quence, a critical gut abnormality may not have had theopportunity to manifest itself in the c-myb nullizygous mutantmice.

We also believe that c-myb may play an important role inthe function of some adult nonhemopoietic organs. This is

suggested by the high level of c-myb mRNA and proteinexpression in those organs that have rapidly proliferatingepithelium such as within cobonic crypts. In contrast, the highmRNA expression levels within the rapidly proliferating em-bryonal brain, kidney, and liver fall dramatically by adulthood.This parallels the low level of proliferation that occurs withinthese organs during adult life.

Our immunohistochemical studies of mouse cobonic cryptcells suggest that c-Myb is expressed along the entire lengthof the crypt. Interestingly, this expression pattern extendsbeyond the commonly recognized proliferative zone occu-pying the lower third of the crypt as confirmed by PCNAstaining and indicated by the presence of mitotic figures.This pattern of expression was unexpected and intriguingbecause it suggests that c-Myb expression is not absolutelylinked to cellular proliferation within the cobonic crypt. Inaddition, c-Myb expression appears to be maintained withinsome differentiated cells at the top of the crypt, despite thegenerally held view that proliferation has ceased at this point(13).

Similarly, our immunohistochemistry data suggest that c-Myb expression within the germinal centers of the spleen ismore extensive than that of PCNA. Therefore, we proposethat c-Myb is required for proliferation of colonic crypt epi-thelial cells and for cells within germinal centers of thespleen, and for some crypt and spleen cell types, c-Myb

expression remains compatible with differentiation. Thus,there appears to be a discordance between c-Myb expres-sion and proliferation in both the cobonic crypt and the ger-minal centers of the spleen.

Cell Growth & Differentiation 965

�A

� Ir � �� ��

...�.

,$

M��,#{149}I

I‘S

� �,#{176}�

� ; � f� �

� � � �

� �

�:(� 4

C

S:�4,

, �*

(�‘� �)

4

C #{149} . �, � . .. ii� #{149}�� �

:.�,. . : . ..c ...�.

..�

� . . ..� .�. �;�:�:

;..#{176}.

� � �.‘

#{149}. ,�* . .-‘,‘.., #{176}�

#{176}p. � � � � . .3� � . - � .. ‘. e��’ �

E � . � #{176}, . .4.)#{176}� �

� I. � �::&‘ .�

� *� � �

�_T � S,��;f�?S� 4$� �� 41 � � �, S � ‘ *�#{176} � S #{149}�� . � � *“�‘�� �.- .w �

� � �..,.� � - ‘. �:

t �147’� � � 0� �i�? � �

� .�#{216};

t S � � � .�,#{149} �

f -� . S � � � #{176}-: �, � , ,� ��$ )c � S � -

S. �;t�-� . � S �

.p...� �

., it�Y�

� .5 . �#{149}a��‘; ,,

:: b’� � ;i��t4�S.. #{176}� � �‘ � � � ‘ S� �

4iLX� � #{176}‘4f�.S.� � -!� � : � � �;. “�f�’’�t �

F’ � � S �

�‘. #{149}#{149}.�#{176} .� ...4’. C.

� � ,v. -� � #{149}�i��? .�, � � :.�

* ,:c�b� � :i: �‘.,i� � �:,�;#{176}�‘St�.:

. �:. � � .

. . ;,_5eoi.*4.�’.�P#{149}�. �t.p*.. -‘S � #{149} �

Fig. 3. A and B, immunohistochemical analysis of adult mouse colonic crypt cross-sections incubated with anti-c-Myb rabbit serum at x400 (A) andx 1 000 (B). Both A and B demonstrate c-Myb within columnar-like cells but not goblet cells which are PAS positive (arrows). C-F, immunohistochemicalanalysis of spleen germinal centers. Sequential sections incubated with anti-c-Myb rabbit serum (C) and anti-PCNA mouse monoclonal antibody (D) at x 100and x200 (E and F). Sections show more extensive c-Myb expression than that seen for PCNA.

c-myb Expression in Colon

How are these data reconcilable? We suggest that in colon(and spleen), c-Myb expression is not inextricably linked toproliferation in vivo. First, c-Myb may play a specific but asyet undefined biological role within the murine cobonic crypt.Second, a dichotomy of c-Myb expression may occur withinthe cells of the cobonic crypt. Specifically, the high levels ofc-Myb expression in columnar cells throughout the crypt arein stark contrast to the absence of expression by most ma-

ture goblet cells. Indeed, this dichotomy of expression hasbeen documented in a bi-potential hemopoietic cell line(K562) in which induced differentiation requires down-regu-lation of c-myb in erythroid but not megakaryocytic cells (14).

Third, there is some evidence that the strong relationship

between c-myb expression and proliferation does have itsexceptions. For instance, stimulation of fully differentiatedperipheral T lymphocytes results in the proliferation of ma-ture lymphocytes and a corresponding increase in c-mybexpression (15-i 7). It also appears that down-regulation ofc-myb is only required for the differentiation of some hemo-poietic cell lines during part of the differentiation-induction

phase. Consequently, after this time c-myb expression canbe reinitiated without perturbing programmed cell differenti-ation (1 8). Finally, it is noteworthy that a number of c-mybtarget genes encode products indicative of end cell functionin cells that no longer divide (3).

The data presented here suggest that c-myb may play arole in nonhemopoietic organ development. Within the co-Ionic epithelium, c-myb mRNA and protein are highly ox-pressed from the embryo to adulthood, and c-Myb is ox-pressed in murine colonic crypt cells beyond the zone ofcellular proliferation. This raises the possibility that c-Mybmay play an important role in normal colon cell biology. In

addition, we have shown discordance between c-Myb ex-pression and proliferation occurring within two distinct or-gans: the cobonic crypt and the spleen. This suggests thatprevious in vitro data regarding the relationship betweenc-Myb expression and proliferation may in fact not representtheir in vivo relationship.

Materials and Methods

Ribonuclease Protection Assays. Total cellular ANA was isolated using

a guanidmne thiocyanate method (1 9). Aibonuclease protection experi-ments were carried out with sense and antisense nboprobes preparedfrom pGEM3Z (Prornega) containing a murine c-myb insert consisting of

a 25i-bp Nael-Pstl fragment (nucleotides 858-i 109) of a full-length c-myb cDNA (20). The antisense riboprobe was created by plasmid linear-ization with EcoRl followed by in vitro transcription in the presence of

r2P]UTP. The j32M cDNA was cloned into pGEM4Z and linearized withSnaBI. A total of 20 �g of total ANA in 30 p.1 of hybridization buffer (40 m�PIPES at pH 7.5, 0.4 M NaCI, 1 rnui EDTA, and 20% formamide) containing5 x 10� cpm of antisense riboprobe was heated to 85#{176}Cfor 5 mm and then

transferred to a 45#{176}Cwater bath for 15 h. Following hybridization, 300 �lof nbonuclease digestion buffer (10 m� Tris-HCI at pH 7.5, 300 rn� NaCI,and 5 rnui EDTA) containing ribonuclease A (40 �g/rnl; Sigma) and ribo-nuclease Ti (2 pg/mI; United States Biochemical) were added, and themixture incubated at 30#{176}Cfor 60 mm. Digestion reactions were terminated

by adding 20 p1 of 10% SDS and 5 j.d proteinase K (10 mg/mI; Bcehringer

Mannheim). The samples were phenol/chloroform extracted, ethanol pre-cipitated, and separated using electrophoresis on a 6% polyacrylamide/6M urea gel at 1 000 V for 2 h and the gel dried. Autoradiography was

performed using a Molecular Dynamics Phosphorlmager(Sunnyvale, CA).Anti-Myb AntIbody and Immunohistochemistry. Antibody to c-Myb

was provided as a gift from Diane Favierand Tom Gonda(IMVS, Adelaide,

Australia) and was raised in rabbits using Escherichia coil-expressed

recombinant murlne CT3 c-Myb conjugated to keyhole limpet hemocya-nm. It was assessed for c-Myb-specific activity, purified on protein A-Sepharose, and used at a 1 :250 dilution.

Anti-PCNA monoclonal antibody (PC-b) was used at a 1 :100 dilutionaccording to the manufacturer’s instructions (Santa Cruz Biotech). Tissuewere fixed in Methacam (60% chloroform, 30% methanol, and 10%glacial acetic acid) for 24 h, paraffin embedded, sectioned to 4-7 p.musing a cryostat, and placed on glass slides. Sections were dewaxed,

rehydrated, and immersed in 1 x TBS, 0.1 % Tween 20 (Sigma), and 10%

nonfat skim milk protein, and left at room temperature for at least 4 h. Thesections were immersed in 1 x TBS, 0.1 % Tween 20, and 1% nonfat skimmilk protein containing the anti-c-Myb rabbit antibody for at least 1 2 to18 h at room temperature. The next day a series of three 20-mm washes

were performed with 1x TBS and 0.1 % Tween 20 followed by an appli-cation of a secondary antirabbit antibody conjugated with horseradishperoxidase used at a 1:250 dilution for 30 mm (Bio-Rad). Another series ofthree washes was performed. The chromogen diaminobenzidene withnickel enhancement was used for light microscopic visualization of thesignal (Pierce). The sections were counterstained with PAS and hematox-ylin and mounted with coverslips.

AcknowledgmentsWe thank Val Feakes and Fiona Clay (Ludwig Institute for Cancer Re-search) for assistance and advice with respect to the immunohistochem-istry experiments, Diane Favier and Tom Gonda (IMVS) for their generousgift of the anti-c-Myb antibody, and Ashley Dunn for critical advice.

References1 . Shen Ong, G. L The myb oncogene. Biochim. Biophys. Acts, 1032:

39-52, 1990.

2. Weston, K. M. The myb genes. Semin. Cancer Biol., 1: 371-382, 1990.

3. Thompson, M. A., and Aarnsay, A. G. Myb: an old oncoprotein withnew roles. Bioessays, 17: 341-350, 1995.

4. Ramsay, A. G., Thompson, M. A., Hayman, J. A., Reid, G., Gonda, T.

J., and Whitehead, A. H. Myb expression is higher in malignant humancolonic carcinoma and premalignant adenomatous polyps than in normalmucosa. Cell Growth & Differ., 3: 723-730, 1992.

5. Torelli, G., Venturelli, D., Cob, A., Zanni, C., SoIled, L, Moretti, L,Calabretta, B., and Torelli, U. Expression of c-myb protooncogene and

other cell-cycle related genes in normal and neoplastic human colonicmucosa. Cancer Res., 47: 5266-5269, 1987.

6. Alitalo, K., Wmnqvist, A., Un, C. C., de Ia Chapelle, A., Schwab, M., andBishop, J. M. Aberrant expression of an amplified c-myb oncogene in twocell lines from colon carcinoma. Proc. NatI. Acad. Scm. USA, 81: 4534-

4538, 1984.

7. Melani, C., Aivoltini, L, Parmiani, G., Calabretta, B., and Colombo, M.P. Inhibition of proliferation by c-myb antisense oligodeoxynucleotides in

colon adenocarcinoma cell lines that express c-myb. Cancer Res., 51:2897-2901, 1991.

8. Dyson, P. J., Poirler, F., and Watson, A. J. Expression of c-myb inembryonal carcinoma cells and embryonal stem cells. Differentiation, 42:24-27, 1989.

9. Queva, C., Ness, S. A., Grasser, F. A., Graf, T., Vandenbunder, B., andStehelin, D. Expression patterns of c-myb and of v-myb Induced mye-bid-i (mim-1) gene during the development of the chick embryo. Devel-opment, 1 14: 125-133, 1992.

10. Mucenski, M. L, McLain, K., Kier, A. B., Swerdlow, S. H., Schreiner,

C. M., Miller, T. A., Pietryga, D. W., Scott, W. J. J., and Potter, S. S. Afunctional c-myb gene is required for normal murine fetal hepatic hema-topoiesis. Cell, 65: 677-689, 1991.

11 . Ess, K. C., Whitaker, T. L, Cost, G. J., Witte, D. P., Hutton, J. J., andAronow, B. J. A central role for a single c-Myb binding site in a thymiclocus control region. Mol. Cell. Biol., 15: 5707-5715, 1995.

12. Sitzmann, J., Noben-Trauth, K., and Klempnauer, K. Expression ofmouse c-myb during embryonic development. Oncogene, 11:2273-2279,1995.

Cell Growth & Differentiation 967

1 3. Gordon, J. I., and Hermiston, M. L Differentiation and self-renewal inthe mouse gastro-intestinal epithelium. Curr. Opin. Cell Biol., 6:795-803,1994.

14. Rosson, D., and O’Brien, T. G. Constitutive c-myb expression in K562cells inhibits induced erythroid differentiation but not tetradecanoyl phor-bol acetate-induced megakaryocytic differentiation. Mol. Cell. Biol., 15:772-779, 1995.

15. Lipsick, J. S., and Boyle, W. J. c-myb protein expression is a lateevent during T-Iyrnphocyte activation. Mol. Cell. Biol., 7: 3358-3360,1987.

16. Churllla, A. M., Braciale, T. J., and Braciale, ‘I. L Regulation of Tlymphocyte proliferation. Interleukin 2-mediated induction of c-myb geneexpression is dependent on T lymphocyte activation state. J. Exp. Med.,170: 105-121, 1989.

i 7. Nelson, P. J., Geller, A. L, and Bach, F. H. Gene expression In CD8

and CD4 T-ceII populations following activation with the calcium iono-phore A23187. Scand. J. Immunol., 35: 611-619, 1992.

18. Richon, V. M., Ramsay, A. G., Rifkind, A. A., and Marks, P. A. Mod-ulation of the c-myb, c-myc and p53 mRNA and protein levels during

induced murine erythroleukemia cell differentiation. Oncogene, 4: 165-

173, 1989.

19. Chomczynski, P., and Sacchi, N. Single step method of ANA isolationby acid guanidiniurn thiocyanate-phenol-chloroform extraction. AnalyticalBiochemistry, 162: 156-159, 1987.

20. Gonda, T. J., Gough, N. M., Dunn, A. A., and de Blaquiere, J. Nude-otide sequence of cDNA clones of the murlne myb proto-oncogene.

EMBO J., 4: 2003-2008, 1985.