&p.1:Abstract Oncoproteins and tumour-suppressor proteinsare thought to possess an antagonistic function in theregulation of growth and differentiation processes duringembryonic and fetal development. In contrast, in theadult, tumour growth is associated with the overexpres-sion of oncoproteins or the malfunction of tumour-sup-pressor proteins. We examined the occurrence of the tu-mour proteins c-erb-B2 and c-fos and the tumour-sup-pressor protein p53 in 17 human embryos and fetuseswith the help of immunohistochemistry. C-erb-B2 wasdetected mainly in embryonic tissue that are not knownfor c-erb-B2-overexpression in tumours in the adult. Incontrast, c-fos was almost always located in fetal tissuescorresponding to its location in adult tumours. Stainingfor p53 was found in a wide variety of embryonic and fe-tal tissues. C-erb-B2 and p53 were localized in the sametissue structures of the developing skin, heart and mus-cle. In other tissues, e.g. muscle and bone, c-fos wasfound together with p53, suggesting an antagonistic ac-tion of these proliferative and antiproliferative factors.Furthermore, c-erb-B2, c-fos and p53 appear to be im-portant for growth and differentiation processes in hu-man development as the occurrence of these proteins wasnot only restricted to specific tissues but also to specificstages of development of these tissues.

&kwd:Key words C-erb-B2 · C-fos · p53 ·Immunohistochemistry · Human embryos and fetuses&bdy:

Introduction

Growth and differentiation processes are normally wellregulated. One way in which the cell employs these regu-latory mechanisms of proliferation is by balancing theeffects of oncogenes and antioncogenes, i.e. tumour-sup-

pressor genes (Merlino 1994). If their delicate interac-tion is disturbed, tumour growth can be initiated.

C-erb-B2 is a 185 kDa receptor-glycoprotein exhibit-ing an intrinsic tyrosine kinase activity (Akiyama et al.1986). Therefore, it is located in plasma membranes, aproperty regarded as a sign of specificity in routine his-tochemical evaluation (Van de Vijver et al. 1988; Slamonet al. 1989). C-erb-B2 can induce the expression of c-fos(Scott et al. 1991), the second oncoprotein investigatedin this study. Because gene amplification results in up-regulated mRNA, high levels of c-erb-B2 oncoproteinare detectable in a variety of human cancers, e.g. of thebreast (Parkes et al. 1990) or the ovaries (Slamon et al.1989). Also non-small-cell carcinoma of the lung (We-iner et al. 1990), adenocarcinoma of the stomach (Falckand Gullick 1989) and carcinoma of the liver (Voravud etal. 1989) show an increased c-erb-B2 concentration. C-erb-B2 has already been localised in human tissues dur-ing development, but the results are controversial. Quirkeet al. (1989) detected c-erb-B2 in a wide variety of epi-thelial and non-epithelial tissues of human fetuses. Incontrast, Press et al. (1990) could not find c-erb-B2 inany non-epithelial cell tissues.

C-fos is a 55 kDa protein, functioning as a nucleartranscription factor. Transcription factors are necessaryfor gene regulation (Latchman 1995). C-fos is thought toregulate the cell cycle, activating cells in G0 to the Sphase (Hesketh 1994), thereby promoting proliferativeactions. High levels of c-fos were reportd for osteosar-coma (Wu et al. 1990), hepatocellular carcinoma (Arbut-noth et al. 1991), renal cell carcinoma (Slamon et al.1984), colorectal adenomas and carcinomas (Magrisso etal. 1993) and carcinoma of the lung (Volm et al. 1992).C-fos has not been demonstrated in human embryos orfetuses. Up to the present time only c-fos mRNA hasbeen localised in human embryos during development ofthe long bones (Sandberg et al. 1988). Wang et al. (1992)established a c-fos-gene knockout mouse, which eventu-ally develops osteopetrosis. Whether c-fos oncoproteinplays a role in early human development other than inbone formation, remains to be elucitated.

N. Miosge (✉) · W. Schneider · W. Götz · R. HerkenZentrum Anatomie, Abteilung Histologie, Universität Göttingen,Kreuzbergring 36, D-37075 Göttingen, GermanyTel.: 0551-397050; Fax: 0551-397067&/fn-block:

Anat Embryol (1997) 195:345–352 © Springer-Verlag 1997

O R I G I N A L A RT I C L E

&roles:Nicolai Miosge · Wolfgang Schneider · Werner GötzRainer Herken

The oncoproteins c-erb-B2, c-fos and the tumour suppressor protein p53in human embryos and fetuses

&misc:Accepted: 30 September 1996

P53 is a 53 kDa phosphoprotein with an extremelyshort half-life in the adult (Montenarh 1992; Zambettiand Levine 1993). P53 is thought to arrest the cell cycleof cells with a damaged genome to allow reparation pro-cesses (Lane 1992). P53 is also able to down-regulate thetranscription of c-fos after binding to a c-fos promotor(Ginsberg et al. 1991). P53 was found in a wide varietyof human tumours, e.g. of the breast (Moll et al. 1992),renal cell carcinoma (Bot et al. 1994) and colorectal can-cers (Starzynska et al. 1992). Nothing is known aboutthe occurrence of p53 in human embryos or fetuses.

Oncoproteins and tumor suppressor proteins presentduring human embryonic and fetal development have notbeen investigated comparatively. We localised c-erb-B2and c-fos oncoproteins and p53 tumor suppressor proteinduring the development of 17 human embryos and fetus-es at the light microscopical level with the help of perox-idase-anti-peroxidase (PAP) immunohistochemistry.

Materials and methods

Sources of tissues

Aborted human embryos were collected according to the regula-tions of the Ethics Committee of the Medical Faculty of the Uni-versity of Göttingen. The specimens (n=17) ranged from gestatio-nal week (gw) 4 to gw 20 (Table 1); eight of them were embryonic(gw 4–9) and nine fetal (gw 10–20). Ages were determined fromhistological data according to Carnegie stages (O’Rahilly andMüller 1987). Up to gw 12, tissues were chosen from abortionsbased on social indications; fetal tissues of later stages were takenfrom abortions based on maternal medical indications. No malfor-mations or abnormalities were observed in the specimens. Theabortion procedure leads to destruction of embryonic material.Our results, therefore do not include all organs and tissues of em-bryos and fetuses, but are limited to those that we were able to fol-low to completion.

Fixation and preparation of tissues

As in vitro experiments suggest that long-lasting hypoxia can leadto the induction of p53, the abortion material was transported tothe laboratory in histidine-tryptophan-ketoglutarate solution (Bret-schneider 1980) and fixed within 30 min by immersion in 4%paraformaldehyde in phosphate buffer, pH 7.2, at 4° C. After thematerial had been dehydrated in a graded series of ethanol from30% to 100%, it was embedded in paraplast. Serial sections of 5µm were cut with a Reichert’s microtome. Every fifth section wasstained with hematoxylin-eosine for topographical orientationwithin the anatomical regions examined.

Primary antibodies

The specificity of the monoclonal mouse antibody anti-c-erb-B2CB11(Medac, Hamburg, Germany) that is directed against the in-ternal domain of the c-erb-B2-protein was confirmed by Westernblots and immunoprecipitation (Corbett et al. 1990).

The polyclonal rabbit antibody anti-c-fos 456(Medac) wasraised against a bacterially expressed β-galactosidase-fos-fusion-protein. Its specificity was demonstrated by Western blots and im-munoprecipitation (Verrier et al. 1986).

The polyclonal rabbit antibody anti-p53 CM-1 (Medac) wasraised against human recombinant p53-protein. The specificity ofthe antibody was shown by immunoblot and immunoprecipitation(Midgley et al. 1992).

Immunohistochemistry

Sections were deparaffinized, rehydrated and rinsed for 10 min inTRIS-buffered saline (TBS, pH 7.4). Endogenous peroxidase wasblocked by incubation in 1% H2O2 dissolved in methanol for 40min in the dark. Incubation with the primary antibody that werediluted in 1% bovine serum albumin in TBS (1% TBS/BSA) fol-lowed. The anti-c-erb-B2-antibodies were employed at a dilutionof 1:40, the anti-c-fos-antibodies at a dilution of 1:500 and the an-ti-p53-antibodies at 1:50 dilution. Each of thse steps was followedby rinsing in TBS for 10 min. Incubation was performed overnightat 4° C. For c-erb-B2 a secondary peroxidase-conjugated anti-mouse IgG rabbit antibody (Medac) was incubated at 1:50 dilutionwith 1% TBS/BSA. For c-fos and p53 a bridge antibody (anti-rab-bit swine immunglobulin, Dakopats, Hamburg, Germany) wasused at 1:50 dilution followed by a PAP-complex (Dakopats,Hamburg, Germany), diluted 1:150. Each of these incubations wasperformed for 30 min at room temperature. Finally, the sectionswere treated with 3,3′-diamino-benzidine.

Controls

Each of the immunoreactions described above was accompaniedby specimens of either a c-erb-B2 overexpressing ovarian carcino-ma, a c-fos-positive eye lens or a p53-positive fetal kidney as posi-tive control.

Negative controls were performed applying normal serum(mouse or rabbit) diluted 1:100 in 1% TBS/BSA instead of the pri-mary antibody. No immunostaining was observed after this re-placement.

Results

C-erb-B2

Skin

The cells of the developing epidermis of human embryosand fetuses of gw 5–12 showed a positive reaction for c-

346

Table 1 Age of embryos and fetuses&/tbl.c:&tbl.b:

Gestational Internal registra- Carnegieweek tion numbers stages

4 D6 and E26 135 491/90 146 8133/90 167 5036/90 188 21207 and W8 20, 229 W2 and W10 −

10 31140 and 6635 −11 6050/91 and B13 −12 B14 −13 none −14 4400/90 −15 B4 −20 298/91 −

&/tbl.b:

Fig. 1 Skin, human embryo, gw 8 (arrow immunostaining for c-erb-B2 oncoprotein of plasma membrane regions of the develop-ing epidermis). No staining of the underlying dermis. Bar 25 µm.Inset: higher magnification of epidermis showing immunostainingfor erb-B2 of plasma membrane regions. Bar 10 µm&/fig.c:

Fig. 2 Heart, human embryo, gw 5 (arrow immunostaining for c-erb-b2 of plasma membrane zones of developing myocytes). Nostaining for erythrocytes (ery) or surrounding mesenchyme (mes).Bar 42 µm&/fig.c:

347

Fig. 3 Kidney, human embryo, gw 5 (arrow immunostaining forc-erb-B2 of plasma membrane regions of developing cells of thetubuli). No staining of the developing glomeruli (open arrow). Bar20 µm&/fig.c:

Fig. 4 Vertebra anlage, human fetus, gw 11 (arrow immunostain-ing for c-fos of nuclei of chondrocytes). No staining for the sur-rounding mesenchyme (mes). Bar 15 µm&/fig.c:

Fig. 5 Rib anlage, human fetus, gw 11 (arrow immunostainingfor c-fos of nuclei of osteocytes from the cortical region). Nostaining of the developing spongious bone (spo) or the surround-ing periosteum (open triangle). Bar 14 µm&/fig.c:

Fig. 6 Liver, human fetus, gw 10 (arrow immunostaining for c-fos of nuclei of a few hepatocytes). No staining for the majority ofcells or haemopoetic precursor cells (open triangle). Bar 12 µm&/fig.c:

erb-B2 in the plasma membrane regions (Fig. 1). Speci-mens up to gw 12, which still exhibited a peridermal-celllayer also showed a positive reaction. The developingdermis was not stained.

Heart

In human embryos of gw 4 the cells of the developingmyocardium showed a positive reaction for c-erb-B2 inplasma membrane regions. From gw 5–8, a strong reac-tion occurred in the developing myocardium, as well asin the plasma membrane regions of the developing epi-cardium (Fig. 2). Human embryos of gw 9 onwardsshowed no immunoreaction for c-erb-B2.

Muscle and bone

No immunoreaction was detectable for c-erb-B2 anti-body in developing skeletal structures in any of the hu-man embryos or fetuses investigated. In contrast, plasmamembranes of developing muscle cells of all human em-bryos and fetuses from gw 4 to 10 showed a positive re-action for c-erb-B2 antibody. From gw 11 onwards, theimmunoreaction ceased.

Liver

No immunoreaction was detectable within the liver ofthe specimens investigated.

Lung

In general, no c-erb-B2 immunoreactivity was seen inthe developing lung; only a few disseminated epithelialcells of the branching lung buds showed a weak reactionfor the c-erb-B2 antibody of plasma membrane regionsat gw 5.

Kidney

During meso- and metanephrogenic development, theplasma membrane zones of epithelial cells of the tubuliof embryos of tw 4–6 showed a positive reaction for c-erb-B2 antibody (Fig. 3). In contrast, no fetal develop-mental stages investigated showed c-erbB-2 immunore-activity.

C-fos

Skin

No c-fos immunoreactivity was detectable in the speci-mens investigated.

Heart

There was no staining for c-fos in the hearts in the speci-mens investigated.

Muscle and bone

Early embryos showed no reaction for c-fos antibody.From gw 9 to gw 15 the nuclei of chondrocytes, as wellas the nuclei of calcifying chondrocytes were positive forc-fos antibody (Fig. 4). The nuclei of osteocytes, espe-cially in the cortical regions of developing bones werepositive for c-fos antibody from gw 9 onwards (Fig. 5).Zones of enchondral ossification showed much weakerreactions. At gw 15, osteoclasts also showed staining ofc-fos. Some developing muscle cells showed a weak nu-clear staining for c-fos antigen from gw 9 onwards.

Liver

Human embryos showed no immunoreactivity for c-fosantibody in the liver neither in the hepatocytes nor withinhaemopoetic stem cells. There was a weak reaction in afew of the nuclei of hepatocytes in specimens of gw 10onwards. At gw 20, some of the nuclei of the hepa-tocytes in a subcapsular location were positive for c-fosantibody (Fig. 6).

Lung

Only weak immunoreactivity was found in human em-bryos of gw 5 in the nuclei of epithelial and mesenchy-mal cells in the developing lung. Older specimensshowed no staining for c-fos antigen.

Kidney

Staining for c-fos antigen was restricted to the nuclei ofsubcapsular cells of the metanephrogenic blastema ofhuman fetuses of gw 14 and 20 (Fig. 7). No other speci-men showed a reaction.

P53

Skin

From gw 6 onwards, the nuclei of the cells of the peri-dermal cell layer, as well as the nuclei of cells of the de-

348

Fig. 7 Kidney, human fetus, gw 20 (arrow immunostaining for c-fos of nuclei of the metanephrogenic blastema). No staining of thesurrounding cells of the developing parenchyme (par). Bar 10 µm&/fig.c:

Fig. 8 Skin, human embryo, gw 7 [arrow immunostaining for p53of nuclei of the developing epidermis and some nuclei of cells ofthe developing dermis (open arrow)]. Bar12 µm&/fig.c:

349

Fig. 9 Rib anlage, human embryo, gw 7 (arrow immunostainingfor p53 of nuclei of developing chondrocytes, open arrowimmu-nostaining of nuclei of cells of the periosteum). Bar 10 µm&/fig.c:

Fig. 10 Rib anlage, human embryo, gw 9 [arrow immunostainingfor p53 of nuclei of developing chondrocytes of the cortical zoneand of the spongiosa (spo), open arrowimmunostaining of nucleiof cells of the periosteum]. Bar 20 µm&/fig.c:

Fig. 11 Liver, human fetus, gw 20 (arrow immunostaining forp53 of nuclei of haemopoetic stem cells). No staining for hepa-tocytes (open triangle). Bar 5 µm&/fig.c:

Fig. 12 Kidney, human fetus, gw 20 (arrow immunostaining forp53 of nuclei of the subcapsular metanephrogenic blastema). Nostaining of adjacent cells of the developing parenchyme (par).Bar: 20 µm

veloping epidermal layer showed a marked immunoreac-tivity for p53 antigen. Cells of the dermis were alsostained (Fig. 8). From gw 10 onwards the immunoreac-tion of all three layers became weaker, but did not ceaseuntil gw 20.

Heart

All embryos showed a nuclear reaction for the p53 anti-body within all three layers of the developing heart. Incontrast, fetal heart tissues were not stained.

Muscle and bone

The nuclei of chondrocytes of the developing bones ofall developmental stages from gw 7 onwards werestained with the anti-p53-antibody. In early stages ofbone formation, the nuclei of the developing periosteumsurrounding the bone structures also showed a reactionfor p53 antibody (Fig. 9). During ossification, the nucleiof osteocytes of the cortical zones and the spongiosawere also stained, as were the nuclei of the cells of theperiosteum (Fig. 10). From gw 6 onwards, the nuclei ofdeveloping myocytes of various skeletal muscles withthe p53 antibdy.

Liver

A nuclear immunoreaction for p53 antibody was onlyfound sporadically in haematopoetic stem cells in fetusesof gw 14 and 20. At gw 20 these cells had a subcapsularlocation (Fig. 11).

Lung

In early human embryos of gw 7 and 8, especially nucleiof mesenchymal cells and some of the epithelial cells ofthe branching lung buds were positive for p53. In fetalstages these immunoreactions became weaker. Smoothmuscle cells showed immunoreactions for p53 antigen inembryos of gw 8 and fetuses of gw 10–12.

Kidney

From gw 13 onwards, the nuclei of mesenchymal cells ofthe subcapsular – located metanephrogenic blastemashowed strong staining for p53 antigen (Fig. 12).

Discussion

We have demonstrated that c-erb-B2, c-fos and p53 areexpressed during embryonic and fetal development inman.

It is striking that c-erb-B2 was found mainly in devel-oping tissues in the embryo that are not known for an as-sociation of c-erb-B2 in tumour growth in the adult. Itwas located in the developing epidermis, muscle, heartand kidney mainly during embryonic stages from gw 4 to9 of development (Table 2). In line with Quirke et al.(1989), but in contrast to Press et al. (1990), we alsofound c-erb-B2 in non-epithelial tissues. In contrast tothe involvement of c-erb-B2 in the pathogenesis of lungor liver cancers these two organs seem to develop duringtheir embryonic and fetal phases largely without the in-fluence of c-erb-B2. One can speculate that upon activa-tion of the c-erb-B2 receptor by a ligand that is uncom-mon in physiological growth and differentiation process-es during human embryonic or fetal development, prolif-

350

Table 2 C-erb-B2, c-fos andp53 location (+positive immu-noreaction, − negative immuno-reaction; only tissue sites withpositive reactions of at leasttwo consecutive gestationalweeks are included in the table&/tbl.c:&tbl.b:

Gestational weeks

Embryonic Fetal

4–6 7–9 10–15 16–20

Tissue sites of c-erbB2Skin: developing epidermis + + + −Heart: developing myocardium + + − −Heart: developing epicardium + + − −Muscle: developing myocytes + + − −Kidney: developing epithelium + − − −

Tissue sites of c-fosMuscle: developing myocytes − − + +Bone anlage − − + +Liver: developing hepatocytes − − + +

Tissue sites of p53Skin: developing epidermis − + + +Skin: developing dermis − + + +Heart: all developing layers + + − −Bone anlage − + + +Muscle: developing myocytes − + + +Kidney: developing epithelium − − + +

&/tbl.b:

erative pathways that lead to tumour development areinitiated. C-erb-B2 and p53 were localised in the devel-oping skin, heart and muscle during the same develop-mental period. Further studies have to elucidate whetherthe localisation of the two oncoproteins in the same tis-sues at the same stages of development points to an an-tagonistic function of the two. The appearance of c-erb-B2 was found mainly during the embryonic and not thefetal period of development. The embryonic period ischaracterized by cell proliferation, cell differentiationand migration, whereas in the fetal period, growth, i.e.proliferation, is the predominant process. The localisa-tion of c-erb-B2 in the plasma membranes can be takenas an indication that this oncoprotein is involved in cell-cell recognition mechanisms, e.g. as a receptor.

C-fos was found mainly in those tissues that are relat-ed to c-fos overexpression during tumour growth in post-natal life, e.g. bone or liver. One can only speculate thatthe re-occurrence of c-fos in the adult, leading to tumourgrowth, is due to the erroneous activation of early embry-onic proliferative processes. In general, c-fos was mainlydetected in tissues during fetal development (Table 2)and, therefore, does not seem to be involved in growth ordifferentiation processes of the human embryo. C-fosoncoprotein was first detected in the developing skeletalsystem in human embryos from gw 9. As it was not foundin earlier developmental stages, one can predict that c-foshas no function in the blastema stage of bone develop-ment. However, it is mainly located in regions of desmalossification of cortical bone formation predominantlyduring fetal life. Staining for c-fos was very sparse in en-chondral ossification centres. Wang et al. (1992) statedthat mice developing without c-fos were found to exhibita diminished cortical bone zone. Viewing these results to-gether, c-fos seems to be essential in the regulation pro-cesses of desmal ossification in the cortical bone.

The staining pattern of p53 was nuclear, as would beexpected of a nuclear protein (Montenarh 1992). P53was found in a wide variety of tissues, generated from allthree germ layers (Table 2). In general, p53 was found inlarger amounts in embryonic tissues, e.g. those of theskin, kidney or heart and only occasionally seen in fetaldevelopmental stages. During embryonic development,the amount of wild-type p53 and/or its half-life seem tobe greatly increased compared to adult tissue. A down-regulation of c-fos through the action of p53, as found invitro by Ginsberg et al. (1991), does not seem to be aphysiological regulatory mechanism in the human em-bryo. If these in vitro results hold true for the in vivo sit-uation in the human, p53 should occur at a specific tissuesite and developmental stage with a weak immunoreac-tion for c-fos or none at all. We did not find tissue siteswith any such coordinated c-fos and p53 occurrence. Incontrast, there are some tissues, like those of the kidneyor bone, where p53 and c-fos are detectable during thesame developmental stages. This might indicate an an-tagonistic role for proliferation processes of the two pro-teins, with p53 down- and c-fos up-regulating growth, asproposed by Merlino (1994). The normal development of

p53 knockout mice before birth (Donehower et al. 1992)might suggest an only minor role for p53 during murineembryonic and fetal development. However, in contrast,to the results on mouse embryos, the wide occurrence ofp53 in the developing human embryo might well indicatesome influences of the protein during human embryonicdevelopment.

In conclusion, our results indicate that c-erb-B2 mightbe involved in tumour growth in the adult, involving apathway that is not active in embryonic or fetal develop-ment. C-fos was located in tissues, e.g. bone, kidney orliver, that are generally linked to c-fos overexpressionduring tumour development. It is, therefore, possible,that malignant growth is initiated through reactivatedmechanisms functioning normally mainly in fetal life. C-fos and p53, predominantly found in the nuclei of tissuesof human fetuses, seem to control cell proliferation pro-cesses as transcription factors.

&p.2:Acknowledgements Parts of the work were taken from the doc-toral thesis of Wolfgang Schneider. We would like to thank CyrillaMaelicke for editing the manuscript.

References

Akiyama T, Sudo C, Ogawara H, Toyoshima K, Yamamoto T(1986) The product of the human c-erbB-2 gene: a 185-kilo-dalton glycoprotein with tyrosine kinase activity. Science 232:1644–1646

Arbuthnot P, Kew M, Fitschen W (1991) c-fos and c-myc expres-sion in human hepatocellular carcinomas. Anticancer Res 11:921–924

Bot FJ, Godschalk JC, Krishnadat KK, Van-der-Knast TH, Bos-man FT (1994) Prognostic factors in renal-cell carcinoma: im-munohistochemical detection of p53 protein versus clinico-pathological parameters. Int J Cancer 57:634–637

Bretschneider HJ (1980) Myocardial protection. Thorac Cardio-vasc Surg 28:295–302

Corbett IP, Henry JA, Angus B, Watchorn CJ, Wilkinson L, Hen-nessy C, Gullick WJ, Tuzu NL, May FEB, Westly BR, HorneCHW (1990) NCL-CB11, a new monoclonal antibody recog-nizing the internal domain of the c-erbB-2 oncogene proteineffective for use on formalin-fixed, paraffin-embedded tissue.J Pathol 161:15–25

Donehower LA, Harvey M, Slagle BL, McArthur MJ, Montgom-ery CA Jr, Butel JS, Bradley A (1992) Mice deficient for p53are developmentally normal but susceptible to spontaneous tu-mours. Nature 356:215–221

Falck VG, Gullick WJ (1989) c-erbB-2 oncogene product stainingin gastric adenocarcinoma. An immunohistochemical study. JPathol 159:107–111

Ginsberg D, Mechta F, Yaniv M, Oren M (1991) Wild-type p53can down-modulate the activity of various promoters. ProcNatl Acad Sci USA 88:9979–9983

Hesketh R (1994) The Oncogene Handbook. Academic Press,London

Lane DP (1992) p53, guardian of the genome. Nature 358:15–16Latchman DS (1995) Transcription-factor mutations and disease.

N Engl J Med 334:28–33Magrisso IJ, Richmond RE, Carter JH, Pross CB, Gilfillen RA,

Carter HW (1993) Immunohistochemical detection of RAS,JUN, FOSand p53 oncoprotein expression in human colorek-tal adenomas and carcinomas. Lab Invest 69:674–681

Merlino G (1994) Regulatory imbalances in cell proliferation andcell death during oncogenesis in transgenic mice. Cancer Biol5:13–20

351

Midgley CA, Fisher CJ, Bartek J, Vojtesek B, Lane D, Barnes DM(1992) Analysis of p53 expression in human tumours: an anti-body raised against human p53 expressed in Escherichia coli.J Cell Sci 101:183–189

Moll UM, Riou G, Levine AJ (1992) Two distinct mechanisms al-ter p53 in breast cancer: mutation and nuclear exclusion. ProcNatl Acad Sci USA 89:7262–7266

Montenarh M (1992) Biochemical propertis of the growth sup-pressor/oncoprotein p53. Oncogene 7:1673–1680

O’Rahilly R, Müller F (1987) Developmental stages in human em-bryos. Including a revision of Streeter’s “Horizon” and a sur-vey of the Carnegie Collection (Carnegie Institution of Wash-ington, Publication 637). Carnegie Institution, Washington

Parkes HC, Lillycrop K, Howell A, Craig RK (1990) C-erbB2mRNA expression in human breast tumors: Comparison withc-erbB2 DNA amplification and correlation with prognosis. BrJ Cancer 61:39–45

Press F, Cordon-Cardo C, Slamon DJ (1990) Expression of theHER/neu proto-oncogene in normal human adult and foetaltissues. Oncogene 5:953–962

Quirke P, Picles A, Tuzi NL, Mohamdee O, Gullick WJ (1989)Pattern of expression of c-erbB-2 oncoprotein in human fetus-es. Br J Cancer 60:64–69

Sandberg M, Vuorio T, Hirvonen H, Alitalo K, Vuorio E (1988)Enhanced expression of TGF-β and c-fos mRNAs in thegrowth plates of developing human long bones. Development102:461–470

Scott GK, Dodson JM, Montgomery PA, Johnson RM, Sartup JC,Wong WL, Ullrich A, Shephard HM, Benz CC (1991)p185HER2 signal transduction in breast cancer cells. J BiolChem 266:14300–14305

Slamon DJ, De Kernion JB, Verma IM, Cline MJ (1984) Expres-sion of cellular oncogenes in human malignancies. Science224:256–262

Slamon DJ, Godolphin W, Jones LA, Holt JA, Wong SG, KeithDE, Levin WJ, Stuart SG, Udove J, Ullrich A, Press MF(1989) Studies of the HER2/neu proto-oncogene in humanbreast and ovarian cancer. Science 244:707–712

Starzynska T, Bromley M, Ghosh A, Stern PL (1992) Prognosticsignificance of 53 overexpression in gastric and colorectal car-cinoma. Br J Cancer 6:558–562

Van de Vijver MJ, Peterse JL, Mooi WJ, Wisman P, Lomans J,Dalesio O, Nusse R (1988) Neu-protein overexpression inbreast cancer. Association with comedo-type ductal carcinomain situ and limited prognostic value in stage II breast cancer. NEngl J Med 319:1239–1245

Verrier B, Müller D, Bravo R, Müller R (1986) Wounding a fibro-blast monolayer results in the rapid induction of the c-fospro-to-oncogene. EMBO J 5:913–917

Volm M, Efferth T, Mattern J (1992) Oncoprotein (c-myc, c-erb-B1, c-erb-B2, c-fos) and suppressor gene product (p53) ex-pression in human squamous cell carcinoma of the lung. Anti-cancer Res 12:11–20

Voravud N, Foster CS, Gilbertson JA, Sikora K, Waxman J (1989)Oncogene expression in cholangiocarcinoma and in normalhepatic development. Human Pathol 20:1163–1168

Wang Z-Q, Ovitt C, Grigoriadis AE, Möhle-Steinlein U, Rüther U,Wagner EF (1992) Bone and haematopoetic defects in micelacking c-fos. Nature 36:741–745

Weiner DB, Nordberg J, Robinson R, Nowell PC, Gazdar A,Greene MI, Williams WV, Cohen JA, Kern JA (1990) Expres-sion of the neu gene-encoded protein (p185neu) in humannon-small cell carcinomas of the lung. Cancer Res 50:421–425

Wu JX, Carpenter PM, Grescens C, Keh R, Niman H, Morris JW,Mercola D (1990) The proto-oncogene c-fos is over-expressedin the majority of human osteosarcomas. Oncogene 5:989–1000

Zambetti GP, Levine AJ (1993) A comparison of the biological ac-tivities of wild-type and mutant p53. FASEB J 7:855–865

352