Development 111, 683-690 (1991)Printed In Great Britain © The Company of Biologists Limited 1991

683

Mitogenic activity of chicken chorioallantoic fluid is temporally correlated to

vascular growth in the chorioallantoic membrane and related to fibroblast

growth factors

INGO FLAMME1, KLAUS SCHULZE-OSTHOFF2 and HEINZ JURGEN JACOB1

1 Institute of Anatomy, Ruhr-University Bochum, FRG^Institute of Experimental Dermatology, University of MUnster, FRG

Summary

The chorioallantoic membrane (CAM) is one of the mostvascularized tissues in the chicken embryo. Capillarygrowth proceeds until day 10 of development andthereafter abruptly regresses. As it is generally acceptedthat the formation of new blood vessel is regulated bygrowth factors, we have investigated the presence ofangiogenic and mitogenic factors in the chicken chorioal-lantois.

In the present study, we show that chorioallantoicfluid (CAF) contains angiogenic substances that areprobably synthesized in the CAM or the embryonickidney. When applied in the chorioallantoic membraneassay, CAF from 9 day chicken embryos elicits a strongangiogenic response. This angiogenic activity of CAF isassociated with pronounced mitogenic effects in vitro.Comparison of different embryonic fluids reveals thatmitogenic activity is particularly evident in the CAF butnot detectable in embryonic serum and amnion fluid.Expression of mitogenic activity is found to be tem-porally correlated with vascular growth in the CAM.High activity is detected in CAF prior to day 10 and then

sharply decreases, thus preceding termination of capil-lary growth by one day.

Heparin-sepharose affinity chromatography suggeststhat the biological activities of CAF probably correspondto the presence of acidic and basic fibroblast growthfactor (aFGF and bFGF). In Western blot analyses ofCAF, an immunoreactive bFGF-like protein of about17xlO3Mr is recognized by a monospecific anti-bFGFantiserum. This protein elutes at 2.4 M NaCl from theheparin-sepharose. The mitogenic activity of the CAFcan be specifically blocked by the anti-bFGF antibodyindicating bFGF to be the active mitogenic principle ofthe CAF.

These results strongly suggest that basic and probablyacidic FGF play an important role in the regulation ofchorioallantoic vascular growth.

Key words: chick embryo, angiogenesis, chorioallantoicmembrane, heparin-binding growth factors, basic fibroblastgrowth factor.

Introduction

The chorioallantois of the chicken embryo originatesfrom a small diverticulum protruding from the tail gutinto the chorion cavity at the second day of develop-ment. This diverticulum, the allantois, consists of anouter vascularized mesodermal and an inner endoder-mal layer. At day 4 the mesodermal layer fuses with thechorionic membrane lining the shell membrane and theallantois acquires an ectodermal coat and is sub-sequently called chorioallantois (Hamilton, 1965).

The choriollantois is essential during embryonicdevelopment. As it includes the chorioallantoic fluid(CAF) into which waste products are delivered, it canbe regarded as an 'in ovo urinary bladder'. Addition-ally, serving as the gas exchanging surface, it resemblesan 'in ovo-lung' (Hamilton, 1965; Romanoff, 1967).

The respiratory function is provided by the extensivecapillary network of the chorioallantoic membrane(CAM). Growth and differentiation of its capillaries areclosely associated with the development of the chorioal-lantois. Capillaries continue to grow up to day 11 andthen proliferation of endothelial cells ceases, coincidingwith an almost completed extension of the chorioallan-tois (Ausprunk et al. 1974).

Over the past decade, it has become evident thatdevelopmental events are controlled by polypeptidegrowth factors. Fibroblast growth factors (FGFs)constitute a family of strongly conserved heparin-binding mitogens which are supposed to be of majorimportance in angiogenic processes (for review seeBaird and Walicke, 1989). Indeed, basic and acidic FGFcould be identified in embryonic tissues of chicken brainand kidney and were shown to be temporally correlated

684 /. Flamme, K. Schulze-Osthoff and H. J. Jacob

with vascularization of these organs (Risau, 1986; Risauand Ekblom, 1986; Risau et al. 1988a). Both factorspossess potent mitogenic activity for endothelial cells aswell as a broad spectrum of mesodermal and neuroecto-dermal cells (for review see Gospodarowicz et al. 1986;Burgess and Maciag, 1989). A crucial developmentalrole of fibroblast growth factors is further substantiatedby their ability to induce mesoderm formation (Slack etal. 1987) and to regulate myoblast differentiation(Clegg et al. 1987). In this context, the presence ofbFGF within embryonic sceleton muscle cells is aninteresting phenomenon (Joseph-Silverstein et al.1989).

Since the CAM represents one of the densest vascularnetworks in the chick embryo, it can be assumed thatangiogenic factors are involved in chorioallantoicvascular growth. We hypothesize that these putativefactors are at least partly released into the chorioallan-toic fluid (CAP). Since furthermore CAF is much moreeasily available than CAM extracts, we analyzed CAFfor presence of angiogenic and mitogenic activities. Thebiological activities are characterized and comparedwith the growth-promoting effects of other embryonicfluid compartments such as amnion fluid and embryonicserum. Furthermore, since angiogenesis of the CAM isstrongly correlated with certain developmental stages,the temporal distribution of growth activities is exam-ined.

Materials and methods

(1) Isolation and preparation of chorioallantoic fluid,amnion fluid and embryonic serumFertilized White Leghorn eggs were incubated at 37 -38 °C in60% relative humidity. Chorioallantoic fluid (CAF) wascollected from embryos at days 6 to 14 of incubation, amnionfluid and serum at day 9. After disinfection, the egg shell andshell membrane were removed over a wide area of thechorioaUantois avoiding any injury to the CAM. CAF wassoaked up using a syringe with a fine gauge. In the same way,amnion fluid was collected by puncturing the amnion cavitythrough the chorioaUantois. Serum was obtained by slitting alarge chorioallantoic vessel using an electrolytically sharpenedtungsten needle (Dossel, 1958) without injuring the CAM.Released blood was soaked up from the surface of the CAM.All fluids were cleared by centrifugation and supplementedwith antibiotics (Serva) (final concentrations: lOOOi.u.ml"1

penicilline, O.lmgml" streptomycine). In most of the CAFsamples contamination by serum was inevitable, but did notexceed 1 %. All serum samples were free of CAF.

(2) Chorioallantoic membrane assayChorioallantoic fluid (CAF) of 9 day embryos was concen-trated 15- to 20-fold using centricon-10™ microconcentrators(Amicon Corp., Danvers, USA). The concentrate was eithermixed with 4 % agar at 40 °C or with 1 % methylcellulose in aratio 1:1. The agar mixture was cast into a glass tube and, aftersolidification, the agar rod was cut into discs of equal size.Methylcellulose discs were prepared by pipetting 40^1 of themixture on sterile glass slides. Solidified discs were implantedonto the chorioallantoic membrane (CAM) of 9 day chickembryos. For control experiments PBS was applied instead ofCAF. Three days after reincubation, the experiments were

evaluated. In one group of experiments, the vascular systemof the embryos was injected with Indian ink (Christ et al. 1973)in order to make small blood vessels visible. The exposedareas were fixed in Serra's fluid, dehydrated in a graded seriesof alcohol and clarified in methylbencoate. A second group ofexperiments was processed for light microscopy. ExposedCAM areas were fixed in 4% glutaraldehyde in 0 .12 Mcacodylate buffer (pH 7.4) and embedded in Durcupan® afterdehydration in a graded series of acetone. Semithin sectionswere cut transversely through the whole exposed areas andstained in 1% methylene blue. In a third group ofexperiments, exposed areas were analyzed under a dissectionmicroscope for signs of angiogenesis. Angiogenic reactionswere evaluated according to the following criteria: positiveresponse: 'spoke wheel' pattern of blood vessels with newlyformed small blood vessels in its center; questionableresponse: 'spoke wheel' pattern of blood vessels withoutnewly formed blood vessels in its center; negative response:no reaction, edema and vascular deviations other than 'spoke-wheel'-like.

(3) Cell proliferation assayThe mitogenic activity of CAF, amnion fluid and serum wasdetermined using Swiss 3T3 cells as target cells. Cells wereseeded in 96-well plates at a density of 7.5 times 103 and grownin 0.1ml DMEM supplemented with 0.5% fetal calf serum(FCS) and antibiotics. After 16-24 h 25 /J of fluid samples wasadded. Following 24 h incubation, cells were labelled with[3H]thymidine (Amersham) at 0.25/iCi/well. After another24 h, cells were harvested on glass fiber filters. Incorporationof radioactivity was measured by scintillation counting anddetermined as average of 5 wells.

(4) Heparin affinity chromatography10 ml CAF was loaded onto a heparin-sepharose column(0.5x5cm) equilibrated in 10mM Tris-HCl pH8.1. Afterextensive washing, bound proteins were eluted with a 30 minlinear gradient of 0-2.7 M NaCl in the same buffer. Fractionsof 1 ml were collected, equilibrated in DMEM and 0.5 % FCSand subjected to mitogenicity assays. For Western blotanalysis, proteins were eluted at steps of 1.2 and 2.4M NaCl.

(5) Western blot analysisFractions obtained by heparin-sepharose affinity chroma-tography, CAF and serum were desalted in 50 HIM ammoniumbicarbonate on Sephadex G-25 and lyophilized. Lyophilizedrecombinant human bFGF was obtained from Progen, FRG.Samples were redissolved in sample buffer (0 .1M Tris-HClpH6.8, 1% SDS, 10% glycerol, 5% mercaptoethanol) andseparated on 12.5-17.5% linear gradient polyacrylamidegels. Proteins were transferred electrophoretically on PVDFmembranes (Immobilon, Millipore) in 50 mM Tris-HClpH8.3. Blots were blocked for l h with 2% defatted milkpowder in Tris-buffered saline pH7.4 and 0.5% Tween 20(TBST) and incubated for 2h in anti-bFGF antiserum(Schweigerer et al. 1987) diluted 1:1000 in TBST. Boundantibody was detected with alkaline phosphatase-conjugatedsecond antibody (Dianova, Hamburg, FRG) using5-bromo-4-chloro-3-indole phosphate and nitroblue tetra-zolium as substrates.

(6) Immunoinhibition assayImmunoglobulin fractions from rabbit anti-bFGF antiserumand a non-immune serum were isolated by affinity chroma-tography on protein A sepharose (Pharmacia). Crude CAFwas preincubated with varying concentrations of the immuno-

Growth factors in chorioallantoic fluid 685

globulins for lh at 37 °C and then subjected to 3T3mitogenicity assay as described above.

Results

The rich vascularization of the CAM suggests thatangiogenic substances are involved in the regulation ofchorioallantoic vascular growth. It is supposed thatthese substances may also be detectable in thechorioallantoic fluid (CAF). Consequently, activitiesaccumulated in the CAF should reflect the temporalchanges in growth-promoting stimuli in the CAM andprobably correlate with changes of chorioallantoicvascular growth.

(1) Angiogenic activity of CAF in vivoAs a homologous (chick-chick) in vivo screening testfor angiogenic activity, we used the chorioallantoicmembrane assay (CAM assay). CAF of 9-day-oldembryos was applied to the CAM of 9-day-old embryos

2

and found to induce an angiogenic response of highreproducibility. After three days of reincubation, anangiogenic reaction was observed beneath the disk in78% of the experiments (n=32). The reaction com-prised 'spoke wheel' transformation of the chorioallan-toic vascular pattern with an extensively vascularizedcenter. As revealed by injection of Indian ink, thisvascularized center consisted of a sponge-like networkof newly formed small blood vessels, which weresupplied with blood and drained by vessels of the 'spokewheel' arrangements (Fig. 1A). In 19% of the exper-iments, angiogenic reactions were restricted to a 'spokewheel'-like transformation of the vascular patternwithout any formation of small vessels in the center.This type of response was subsumed under the category'questionable response'. A negative response wasobtained in only 3%. For comparison, 63% of thecontrol experiments (n=43) gave a negative (Fig. IB),24% a questionable and 13% a positive response.

Further analysis of angiogenic and negative responseswas obtained by light microscopy of semithin sections

Fig. 1. Representative resultsfrom chorioallantoic membraneassays three days after applicationof agar carriers containingchorioallantoic fluid concentrate(A) or PBS as control (B). Bloodvessels are made visible byinjection of Indian ink.Chorioallantoic fluid inducedformation of a radially arranged,spoke-wheel-like vascular patternbearing small blood vessels in itscenter (A). For comparison PBShad no effect on the vascularpattern (B). Arrows: edge of thecarrier. (60 times).Fig. 2. Representative semithinsections through chorioallantoicmembrane areas exposed tochorioallantoic fluid concentrate(A) or PBS as control (B) fixedthree days after test onset.(A) Corresponding to the newlyformed small blood vessels asshown in Fig. 1A, numerous small(arrows) and smallest(arrowheads) capillary bloodvessels are found in the exposedareas. Vessels lie embedded in afibroblast-rich, edematous stromawhich is infiltrated byinflammatory cells (open arrow).(B) In the control, wideinterstitial spaces indicatingedematous transformation of thechorioallantoic mesenchymepredominate. Only few,physiological blood vessels arevisible (arrows). The ectoderm(asterisks) is found to bethickened in both reactions. (600times).

i .

_ ^ - / • "

686 /. Flamme, K. Schulze-Osthoff and H. J. Jacob

(Fig. 2). The sponge-like bundle of small blood vesselscorresponded to a mesenchymal reaction that enclosedboth the augmentation of fibroblasts and the formationof numerous capillaries (Fig. 2A). Negative controlexperiments, on the other hand, exhibited edemawithout additional vessel growth (Fig. 2B). Moreover,in CAF, as well as in control experiments, thickening ofthe chorioallantoic ectoderm and endoderm, infil-tration of the interstitium by inflammatory cells andexulcerations beneath the carrier could be observed.Similar effects were previously described by Jakob et al.(1978) and Spanel-Borowski et al. (1988) as inflamma-tory reactions, which can be induced by different carriermaterials. These effects, however, were more pro-nounced in CAF than in control experiments.

(2) Mitogenic activity of CAF in vitroWe further analyzed whether the angiogenic activity ofCAF observed in vivo coincides with a mitogenicactivity in.vitro. CAF of day 9 was tested in 3T3 cellproliferation assays and compared to activities of serumand amnion fluid. As compiled in Fig. 3, both CAF andamnion fluid promote cell proliferation, with CAFbeing three times more potent than amnion fluid.Theactivity of both fluids was slightly enhanced at dilutionsbetween 5- and 10-fold and decreased at higherdilutions. In contrast, native serum of day 9 embryoshad no proliferative effects and even inhibited [3H]thy-midine incorporation.

Since capillary growth of the CAM progressivelyoccurs up to day 10 and thereafter decreases, wequestioned whether mitogenic activities can be corre-lated with the kinetics of vascular growth. Therefore,CAF from day 6 to 14 was collected and examined inmitogenesis assays. As pointed out in Fig. 4, highproliferative capacity is detected in CAF from day 6 to9. A maximum at day 7 is followed by a slight decrease

D : C A F• Amnion fluid

Serum

Native 1:5 1:10 1:50 1:100 1:200 1:1000 FCS PBS

Dilutions

Fig. 3. Comparative assessment of mitogenic activities ofchorioallantoic fluid (CAF), amnion fluid and serum from 9day old chicken embryos by end point dilution. Sampleswere tested on 3T3 cells for their ability to induce[3H]thymidine incorporation.

6 7 8 9 10 11 12

Days of incubation

13 14

Fig. 4. Temporal changes of mitogenic activity of thechorioallantoic fluid from day 6 to day 14 of incubation.3T3 cells were used as target cells.

to day 9. From day 9 to 10 biological activity sharplydecreases reaching a plateau or slight decline from day10 to 14. The sharp regression of mitogenic activityoccurs one day before vascular growth of the CAM isreduced.

(3) Characterization of the mitogenic activityFor further characterization of the mitogenic activity,CAF was separated by heparin-sepharose chroma-tography, a method suitable for the selective purifi-cation of fibroblast growth factors. Fractions collectedwere tested for their ability to stimulate 3T3 cellproliferation.

Indeed, as shown in Fig. 5, most of the bioactivematerial was adsorbed by the column. Maximumactivity was recovered in two distinct peaks eluted at0.8-1.6 and 1.6-2.6M sodium chloride, respectively.

2.5

2

1 5

1 G

05

0

1

cX1

c"i

;cts

•a4)

c

Q

CJ

p

10987654321

0

0.1

Ec

280

Q0

0

/

1 \ /P\\s \-

A /v \/ \ s r ^ ^

10 15 20 25 30 35 40 45

Fraction number

Fig. 5. Representative result from heparin-sepharoseaffinity chromatography: 10 ml of chorioallantoic fluid ofday 9 was applied to a heparin-sepharose column. Boundproteins were eluted using a 0-2.7 M NaCl linear gradientin 10 mM Tris-HCl pH8.1. Fractions were tested formitogenic activity on 3T3 cells.

Growth factors in chorioallantoic fluid 687

NaCI1.2M 2.4M

20

Hx1(T3

-92-67

-43

-30.5

-20.1

, Fig. 6. Western blot analysis of—14.4 CAF-fractions eluted from

~~"~~ heparin-sepharose at 1.2 and.+ ..^.-^™ 2.4M NaCI.

This elution profile strongly indicated the presence offibroblast growth factors. To confirm this assumptionfractions eluted from heparin-sepharose at 1.2 and2.4 M NaCI were separated by SDS-PAGE andanalyzed in Western blots with a monospecific anti-serum against bFGF. (An immunological detection ofaFGF is in preparation). Western blot analysis with theanti-bFGF antiserum revealed an immunoreactiveprotein of about YlxV&MT in the 2.4M fraction(Fig. 6). This protein was more clearly detectable innative CAF (Fig. 7). It is somewhat larger thanrecombinant human bFGF (Progen, FRG) migrating atabout 16xlO3Mr. The bFGF-like protein was confinedto CAF and not detectable in embryonic serum, which -as shown above - lacked any mitogenic potential. In

67-

43-

30.5-

20.1-

14.4-

Fig. 7. Western blot analysis ofchorioallantoic fluid (lane b) andembryonic serum (lane c) forbFGF using a monospecificbFGF antiserum crossreactingwith human recombinant bFGF(lane a). On lane b protein from4 ml chorioallantoic fluid wasloaded. An equal amount ofprotein was loaded on lane c.

cX

7

<P

15-

10-

5-anti-bFGFnon-immune IgG

0 2.5 25

IgG concentration (//gml"1)

250

Fig. 8. Results from immunoinhibition assay. Crude CAFwas incubated either with anti-bFGF immunoglobulin ( • )or with immunoglobulins from a nonimmune serum (^ )for control at increasing concentrations. Mitogenic activitywas tested on 3T3 cells.

addition, both CAF and serum contained a proteinband of about 60x 103 MT. This immunoreactive proteinis probably nonspecifically recognized by the antiserum(Fig. 7).

To determine whether mitogenic activity of the CAFis caused by the bFGF-like immunoreactive protein, animmunoinhibition assay was conducted. Native CAFwas incubated with the immunoglobulin fraction fromthe rabbit anti-bFGF antiserum and a control immuno-globulin at varying concentrations and was then testedfor mitogenic activity on 3T3 cells. Incubation with theanti-bFGF immunoglobulin remarkably blocked themitogenic activity of the CAF in a dose-dependentmanner. At highest immunoglobulin concentrations,activity was reduced for more than 50 %. The controlimmunoglobulin had no effect on mitogenic activity(Fig. 8).

Discussion

In the present study, chorioallantoic fluid (CAF) ofchick embryos was examined for the presence ofangiogenic and mitogenic activities. Four main con-clusions can be drawn from our results. First, mitogenicactivities in CAF are probably due to fibroblast growthfactors. This is suggested from chromatographic be-havior of CAF after heparin-sepharose chromatogra-phy. The two heparin-binding activities obtained inCAF clearly correspond to elution profiles reported foraFGF and bFGF, respectively (Lobb et al. 1986; Risauet al. 1988a). The identity of bFGF in CAF was furthersubstantiated by Western blot analysis with a mono-specific antiserum. (An immunological detection ofacidic FGF is currently under investigation.) Immu-noinhibition assays with the anti-bFGF antibody clearlydemonstrated that bFGF is a main reason for mitogenic

688 /. Flamme, K. Schulze-Osthoff and H. J. Jacob

activity in the CAF. Second, comparison of differentembryonic fluids by biological and immunologicalcriteria revealed that bFGF is accumulated in the CAFbut not shifted from serum into the CAF. Third,temporal expression of CAF mitogenic activities strik-ingly corresponds to vascular growth patterns observedin the CAM (Ausprunk et al. 191 A). Fourth, CAF isstrongly angiogenic in vivo. This result from chorioal-lantoic membrane assay is in Line with previous studiesthat demonstrate that FGFs elicit strong angiogenicresponses when applied to various in vivo systems(Gospodarowicz et al. 1979; Risau, 1986; Thompson etal. 1989).

Fibroblast growth factors have been isolated from avariety of adult tissues. Whereas acidic FGF is generallyrestricted to neural tissue, basic FGF seems to be morewidespread (for review see Gospodarowicz et al. 1986;Burgess and Maciag, 1989). FGFs possess a molecularrange of about 16-18xlO3Mr. Recent studies revealevidence for the presence of additional high molecularmass forms in the range of 22-25xld* MT. Differencesin molecular weights are presumably due to differenttranslational initiation or truncation of N-extendedforms (review by Rifkin and Moscatelli, 1989). It hasbeen reported that commonly applied acidic conditionsmay account for N-terminal cleavage of FGF (Klags-brun et al. 1987). As these conditions, however, wereavoided in the present study, it is conceivable that theUxlO'M, immunoreactive bFGF-like protein rep-resents a native and untruncated form. This protein wasnot detectable in embryonic serum.

Both aFGF and bFGF bind tightly to heparin and,probably more important, to heparan sulfate. Indeed,recent biochemical and immunohistological data havedemonstrated that FGFs are closely associated with theextracellular matrix of their cellular sources (Vlodavskyet al. 1987; Folkman et al. 1988; Gordon et al. 1989;Presta et al. 1989). Hence, it is rather surprising that inthe CAF the FGFs are apparently present in freesolution.

However, the presence of FGFs, probably related toaFGF and bFGF, in the CAF, is analogous to thepresence of both factors in the vitreous of the chickembryo, which is also an acellular fluid compartment.In this compartment, FGFs may have their function ininduction of the proliferation of lens epithelial cells(Mascarelli et al. 1987).

Since genes of aFGF and bFGF do not encode atypical signal sequence, the growth factors are probablynot released by conventional secretory pathways (forreview see Burgess and Maciag, 1989). In tissue repairand inflammatory processes in which FGFs are presum-ably implicated, it has been suggested that growthfactors are released by cell lysis or hydrolytic enzymesfrom extracellular structures (Gajdusek and Carbon,1989; Schulze-Osthoff et al. 1990). In the chick embryo,however, FGF release cannot be explained by celldeath. Moreover, it is possible that growth factors arereleased into the chorioallantoic fluid by specializedtransport mechanisms.

The expression of FGFs in CAF closely coincides

with the progression of vascular growth in CAMindicating the role of these growth factors in bloodvessel formation. In this respect, it is interesting that theobserved decrease in CAF mitogenic activity pervolume strikingly corresponds to the regression ofvascular growth in the CAM (Ausprunk et al. 1974). InWestern blot analyses, we found that the decrease inmitogenicity is probably not due to differential syn-thesis of inhibitors, but to a decrease in bFGFimmunoreactive protein during development (data notshown). Changes in mitogenic activity and immuno-reactivity, however, do not correlate with the changesin total CAF volume, which attains a maximum at day13 of development (Romanoff, 1967). The time intervalbetween decrease of mitogenicity and regression ofcapillary growth may reflect the period over whichreleased FGFs remain active.

The correlation between changes in the production ofFGF and vascular growth activity, which we describehere, is to our knowledge the first in vivo correlationbetween decrease of FGF expression and regression ofcapjllary growth.

In contrast, previous studies in embryonic develop-ment mainly describe a correlation between increasinglevels of FGF and induction of vascular growth,including studies on central nervous system (Risau,1986; Risau et al. 1988a), embryonic kidney (Risau andEkblom, 1986) and embryoid bodies (Risau et al.19886). In these studies, both angiogenesis, i.e.formation of new blood vessels by immigration ofendothelial cells in primarily avascular organs, andvasculogenesis, which means in situ differentiation of avascular plexus from presumptive endothelial cells,were correlated to the expression of aFGF and bFGFrelated mitogens.

The contribution of FGFs in embryonic angiogenesiswas further supported by immunohistological means.Hanneken et al. (1989) showed that newly formedcapillaries of the bovine fetal retina stained positivelywith bFGF-specific antibodies. In our immunohisto-chemical investigations, we were not able to identify acellular source of the bFGF-like immunoreactiveprotein of the CAF. CAM itself as well as theembryonic kidney which contributes to the chorioallan-toic fluid (Boyden, 1924) and recently was shown toproduce a FGF-related factor (Risau and Ekblom,1986), are potential sources of the aFGF- and bFGF-related activities.

Whether FGFs are the only agents in the native CAFinducing angiogenesis can not be decided from ourexperiments. As angiogenesis is thought to be a processof multifactorial genesis (Folkman and Klagsbrun,1987), other diffusible angiogenic factors may bereleased into the CAF additionally to FGFs.

Recently, further FGF-homologous oncogenes havebeen identified, which, as far as presently established,display similar biological activities as aFGF and bFGF,but, most remarkably, contain functional secretorysignal sequences (for review see Burgess and Maciag,1989). Despite a still limited number of investigations,the expression of these additional members of the FGF

Growth factors in chorioallantoic fluid 689

family seems to be confined to the early embryonicdevelopment whereas no transcripts can be detected inadult tissues. K-FGF was found to be expressed inembryonic stem cells (Heath et al. 1989). The ex-pression of the int-2 proto-oncogene has recently beenstudied in the early mouse embryo and its pattern oflocalization suggests possible roles in cell migrationduring gastrulation and neurulation (Wilkinson et al.1988).

As far as these members of the FGF family areconcerned no data exist about their expression inchicken embryos. Studies of their distribution inmammalian species, however, suggest these factors aresuitable candidates for inductive or cell specificationprocesses. Further comparative studies on the temporaland spatial expression of several FGF members shouldtherefore be promising in elucidication of embryonicblood vessel growth.

The authors wish to thank Dr L. Schweigerer, Heidelberg,for kindly providing the anti-bFGF antibody.

References

AUSPRUNK, D. H., KNIGHTON, D. R. AND FOLKMANN, J. (1974).Differentiation of vascular endothelium in the chickchorioallantois: A structural and autoradiographic study. DeviBiol. 38, 237-248.

BAIRD, A. AND WAJJCKE, P. A. (1989). Fibroblast growth factors.British Medical Bulletin 45, 438-452

BOYDEN, E. A. (1924). An experimental study of the developmentof the avian cloaca, with special reference to a mechanical factorin the growth of the allantois. J. exp. Zool. 40, 437—472.

BURGESS, W. H. AND MACIAG, T. (1989). The heparin-binding(fibroblast) growth factor family of proteins. A. Rev. Biochem.58, 575-606.

CHRIST, B., JACOB, H. J. AND JACOB, M. (1973). Ober einVerfahren zur GefaBdarstellung beim Huhnerembryo. DerPrUparator 19, 1/2

CLEGG, C. H., LINKHART, T. A., OLWTN, B. B. AND HAUSCHKA, S.

D. (1987). Growth factor control of skeletal muscledifferentiation: commitment to terminal differentiation occurs inGl phase and is repressed by fibroblast growth factor. J. CellBiol. 105, 949-956.

DOSSEL, W. E. (1958). Preparation of tungsten micro-needles foruse in embryonic research. Lab. Invest. 7, 171-173.

FOLKMAN, J. AND KLAGSBRUN, M. (1987). Angiogenic factors.Science 235, 442-447.

FOLKMAN, J., KLAGSBRUN, M., SASSE, J., WADZINSH, M., INGBER,D. AND VLODAVSKY, I. (1988). A heparin-binding angiogenicprotein - basic fibroblast growth factor - is stored withinbasement membrane. Am. J. Pathol. 130, 393-400.

GAJDUSEK, C. M. AND CARBON, S. (1989). Injury-induced releaseof basic fibroblast growth factor from bovine aortic endothelium./. cell. Physiol. 139, 570-579.

GORDON, P. B., CHOI, H. U., CONN, G., AHMED, A., EHRMANN,B., ROSENBERG, L. AND HATCHER, V. B. (1989). Extracellularmatrix heparan sulfate proteoglycans modulate the mitogeniccapacity of acidic fibroblast growth factor. /. cell. Physiol. 140,584-592.

GOSPODAROWICZ, D . , BlALECH, H . AND THAKRAL, T . K. (1979).

The angiogenic activity of the fibroblast and epidermal growthfactor. Expl Eye Res. 28, 501-514.

GOSPODAROWICZ, D., NEUFELD, G. AND SCHWEIGERER, L. (1986).Molecular and biological characterization of fibroblast growthfactor, an angiogenic factor which also controls the proliferation

and differentiation of mesoderm and neuroectoderm derivedcells. Cell Differentiation 19, 1-17.

HAMILTON, H. L. (1965). Lillie's Development of the Chick. 3rded., Holt, Rinehart and Winston, NY.

HANNEKEN, A., LUTTY, G. A., MCLEOD, D. S., ROBEY, F. ,

HARVEY, A. K. AND HJELMELAND, L. M. (1989). Localization ofbasic fibroblast growth factor to the developing capillaries of thebovine retina. / . cell. Physiol. 138, 115-120.

HEATH, J. K., PATERNO, G. D., LINDON, A. C. AND EDWARDS, D.

R. (1989). Expression of multiple heparin-binding growth factorspecies by murine embryonal carcinoma and embryonic stemcells. Development 107, 113-122.

JAKOB, B. W., JENTZSCH, K. D., MAUERSBERGER, B. AND HEDER,

G. (1978). The chick embryo chorioallantoic membrane as abioassay for angiogenesis factors, reactions induced by carriermaterials. Expl Path. 15, 241-249.

JOSEPH-SILVERSTEIN, J., CONSIGLI, S. A., LYSER, K. M. AND PAULT,

C. V. (1989). Basic fibroblast growth factor in the chick embryo,immunolocahzation to striated muscle cells and their precursors.J. Cell Biol. 108, 2459-2466.

KLAGSBRUN, M., SMITH, S., SULLIVAN, R., SHING, Y., DAVIDSON,

S., SMITH, J. A. AND SASSE, J. (1987). Multiple forms of basicfibroblast growth factor: amino-terminal cleavages by tumor cell-and brain cell-derived acid proteinases. Proc. natn. Acad. Sci.U.S.A. 84, 1839-1843.

LOBB, R., SASSE, J., SULLIVAN, R., SHING, Y., D'AMORE, P.,

JACOBS, J. AND KLAGSBRUN, M. (1986). Purification andcharacterization of heparin-binding endothelial cell growthfactors. J. biol. Chem. 261, 1924-1928.

MASCARELU, F., RAULAIS, D., COUNIS, M. F. AND COURTOIS, Y.

(1987). Characterization of acidic and basic fibroblast growthfactors in brain, retina and vitreous chick embryo. Biochem.biophys. Res. Com. 146, 478-486.

PRESTA, M., MAIER, J. A. M., RUSNATI, M. AND RAGNOTTI, G.

(1989). Basic fibroblast growth factor is released fromendothelial extracellular matrix in a biologically active form. / .cell. Physiol. 140, 68-74.

RIFKIN, D. B. AND MOSCATELLI, D. (1989). Recent developmentsin the cell biology of basic fibroblast growth factor. J. Cell Biol.109, 1-6.

RISAU, W. (1986). Developing brain produces an angiogenesisfactor. Proc. natn. Acad. Sci. U.S.A. 83, 3855-3859.

RISAU, W. AND EKBLOM, P. (1986). Production of a heparin-binding angiogenesis factor by the embryonic kidney. J. CellBiol. 103, 1101-1107.

RISAU, W., GAUTSCHI-SOVA, P. AND BOHLEN, P. (1988a).

Endothelial cell growth factors in embryonic and adult chickbrain are related to human acidic fibroblast growth factor.EMBO J. 7, 959-962.

RISAU, W., SAJUOLA, H., ZERWES, H. G., SASSE, J., EKBLOM, P.,

KEMLER, R. AND DOETSCHMAN, T. (19886). Vasculogenesis andangiogenesis in embryonic-stem-cell-derived embryoid bodies.Development 102, 471-478.

ROMANOFF, A. L. (1967). Biochemistry of the Avian Embryo.Wiley, NY.

SCHULZE-OSTHOFF, K., RlSAU, W . , VOLLMER, E . AND SORG, C.

(1990). In situ detection of basic fibroblast growth factor (bFGF)by highly specific antibodies. Am. J. Pathol. (in press).

SCHWEIGERER, L., MALERSTEIN, B , NEUFELD, G. AND

GOSPODAROWICZ, D. (1987). Basic fibroblast growth factor issynthesized in cultured retinal pigment epithelial cells. Biochem.biophys. Res. Com. 143, 934-940.

SLACK, J. M. W., DARLINGTON, B. G., HEATH, J. K. AND

GODSAVE, S. F. (1987). Mesoderm induction in early Xenopusembryos by heparin-binding growth factors. Nature 326,197-200.

SPANEL-BOROWSKI, K., SCHNAPPER, U. AND HEYMER, B. (1988).

The chick chorioallantoic membrane assay in the assessment ofangiogenic factors. Biomed. Res. 9, 253-260.

THOMPSON, J. A., HAUDENSCHILD, C. C , ANDERSON, K. D.,

DIPIETRO, J. M., ANDERSON, W. F. AND MACIAG, T. (1989).

Heparin-binding growth factor 1 induces the formation of

690 /. Flamme, K. Schulze-Osthoff and H. J. Jacob

organoid neovascular structures in vivo. Proc. natn. Acad. Sci. WILKINSON, D. G., PETERS, G., DICKSON, C. AND MCMAHON, A.U.S.A. 86, 7928-7932. P. (1988). Expression of the FGF-related proto-oncogene int-2

VLODAVSKY, I., FOLKMAN, J., SULLIVAN, R., FRIDMAN, R., ISHAI- during gastrulation and neurulation in the mouse. EMBO J. 7,MICHAELI, R., SASSE, J. AND KLAGSBRUN, M. (1987). Endothelial 691-695.cell-derived basic fibroblast growth factor: synthesis anddeposition into subendothelial extracellular matrix. Proc. natn.Acad. Sci. U.S.A. 84, 2292-2296. (Accepted 19 November 1990)