Induction of angiogenesis by mixtures of two angiogenic proteins, angiogenin and acidic fibroblast growth factor, in the chick chorioallantoic membrane

Vol.

146,

August

No. 14,

BIOCHEMICAL

3, 1987

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

Pages 1122-1131

1987

INDUCTION OF ANGIOGENESIS BY MIXTURES OF TWO ANGIOGENIC PROTEINS, ANGIOGENIN AND ACIDIC FIBROBLAST GROWTH FACTOR, IN THE CHICK CHORIOALLANTOIC MEMBRANE James W. Fett,

J.

Lemuel

Bethune,

and

L.

Bert

Vallee*

Center for Biochemical and Biophysical Sciences and Medicine and Departments of Pathology and Radiology, Harvard Medical School and Brigham and Women's Hospital, Boston, Massachusetts 02115 Received June 3, 1987

The chick chorioallantoic membrane assay was employed to assess the angiogenic response induced by mixtures of human angiogenin with bovine heparin-binding acidic fibroblast growth factor. Statistical evaluation of data accumulated at several molar ratios of the two proteins indicate that the angiogenic activity observed is neither an additive nor a synergistic resultant of the activities of the proteins separately. The possibility exists, however, that at an -1:l mole ratio an apparent inhibitory effect can be observed. Mechanisms which could underlie such observed effects are discussed. 0 1987 Academic Press,

Inc.

the

Angiogenesis, a complex While

process

typical

endothelial

neovascularization cyclical

glaucoma, underlying recently to *

is

occurs

during

states and

tumor

such

the

and

minimal

uterine

as diabetic

growth

(2,31. are,

messengers

a hemovascular cell in

wound

such

processes

molecular

be isolated

of

of

of mammalian

turnover

development

pathological

development

normal

healing,

tissues

and

retinopathy,

mediating

(11,

embryonic

endometrium,

present,

in

certain

neovascular mechanisms

ill-defined angiogenesis

growth,

but have

begun

characterized.

Address correspondence to this author Biochemical and Biophysical Sciences Medical School, 2SO Longwood Avenue, 02115.

at the Center for and Medicine, Harvard Boston, Hassachusetts

Abbreviations: HBGF(s), heparin-binding growth factor(s): acidic fibroblast growth factor; bFGF, basic fibroblast factor; CAM, chick chorioallantoic membrane. 0006-291X/87 $1.50 Copyright All r&hts

is

differentiation.

The molecular at

network,

@ 1987 by Academic Press, of reproduction in any form

1~. reserved.

1122

aFGF, growth

Vol. 146, No. 3, 1987

One such from

molecule,

serum-free

tumor

cells

the is

a 14,124

with

elicits

of

the

dalton

has

been

of

the

been

detected

isolated

in

and

the

colon

in

vivo

classes

sequence,

are

et

and al.

acidic

found

(13) (~1-5)

primarily detected

and

non-neuronal

al?GF has

in

been

brain-derived sequencing

of

variety 8-10

acid for

bovine

classes

of

neural

tissue. of

2 HBGFs,

typified

and

types.

weights

bovine

15,000

bFGF (20),

and human

HBGFS are

of

known

(22)

weight

cells

by aFGF,

both

structure

to

17,000

they

have

neuronal bovine

human and cDNA

by bFGF, They

to

of

for

(17,18)

as well

HPLC [see

15,000

of

in

acid

however,

have

are

found

in

apparent

~1.5

17,000.

The amino

as the

cDNA sequences

bFGF have

been

to be mitogenic

in

1123

protein

heparin-

typified

as has that

cell

It

angiogenic

reversed-phase

The primary

protein

of (21)

tumor

on both

tissues

the

amino

1 HBGFs,

aFGF based Class

(8).

can be grouped

Recently,

human

has

a second

toward

molecular

(14).

molecular

sequence

in

(15,16),

of

and

of

are

weight,

determined

(19).

plasma

represent

affinity

Class

from

cells.

molecular

review].

origin

distinct

from

polypeptides

elution

lysates

human

which

molecules in

been

wide

for

normal

molecules

characteristic

Indeed,

an angiogenin

(HBGFs)

point,

Angiogenin

polypeptide

quite

Recently,

by their

isoelectric

is

factors

These

differentiated

Sepharose,

chain

as

possesses

adenocarcinoma

(9-12).

as well

(5,6).

indistinguishable

growth

at

ribonuclease.

however, (7).

adenocarcinoma

sequence

single

from

well-characterized

mediators

Lobb

which,

isolated

response

determined

angiogenin

isolated

been

colon

primary

to pancreatic

and biologically

from

of

two

been

enzyme

Heparin-binding type

have

that

pancreatic

chemically

Its

(pI>9.5)

activity

that

recently

an angiogenic

(4).

identity

RESEARCH COMMUNICATIONS

by human

vivo

gene

demonstrated

ribonucleolytic

is

in

cationic

35% sequence

has

conditioned

concentrations

structure

AND BIOPHYSICAL

angiogenin,

medium

and

femtomolar

it

BIOCHEMICAL

determined. vitro

for

Both a wide

a

BIOCHEMICAL

Vol. 146, No. 3, 1987

variety

of

cell

demonstrated

types

to

activity

of

are

the

assay

capacity

of

could

In

antagonistic

higher

mixing

of

binding

while

if

would these

inactive

the

stimulation

oppositely

activity or

attempt

possibilities.

to

of

to only

inhibition

explain For

by occupation

response

such

inhibitory there the

of

was competition protein

over

with

that

could

of

the

formation

yield complex

which

the

activity,

complex

in

that

or

specific

or the

case

an might no

be seen. above

all

physiologically if

with

on

no effect

discussed

example,

an additive

receptors

lower

one species,

last,

distinct

if

molecules

or

two proteins

the

inhibition,

would

two

other

Alternatively, charged

leading

The mechanisms not

potent

be observed.

complex

exhibit

equally

the

among

On the

example,

possessed

grossly

response. two

be observed for

discriminating

receptor

An apparent

and,

membrane

approach,

influence

seen.

receptor

are

one

and

antagonism,

the

of

Positive

a synergistic

might

affinity

both

controls between

is

angiogenin

these

same cell.

might

of

delineate

of

between

cell

angiogenic

the

mixing

interaction

in

response type

to

of

proteins

of

Within

mechanistic

on the

above

an investigation

synergy,

simplest

cooperativity

upon

one

from

activation

same responding

for

resultant

result

negative

no effect

has been

chorioallantoic

(26).

additivity,

i.e., could

chick

al.

from

the

mixtures

possible

the

of

types

is

receptors

cooperativity,

the

it

result

different

hand,

assay

agents:

response

any

To initiate

the

et

outcomes

these.

other

and angiogenic

several

employing

the

of

the

mitogenic

by which

of

Knighton

pharmacologic

two

heparin

unknown.

activity

of

obvious

none

Additionally,

among different

aFGF was assessed (CAM)

the

mechanisms

interactions

molecules,

both

RESEARCH COMMUNICATIONS

aFGF (12,24,25).

angiogenesis

possible

the

(13,23).

enhance

The precise induce

AND BIOPHYSICAL

are

angiogenin 1124

hardly

exhaustive

and do

relevant and aFGF act

on

Vol. 146, No. 3, 1987

separate

cell

anticipated steps It

egg is

populations above

that

here,

BIOCHEMICAL

lead

is

could

considered

no grading

negative.

Thus,

the if

to

it

would

make no difference

be positive.

proteins,

the

only

responses,

examination

scores

with would

of

typical

are

to

the if

the

as with

two

either assay

number

or agents agent

outcome,

of vessels the

were alone, all

were

only

individual

be the

same,

either

by qualitative

subjective,

There

positive

as would

the

a mixture still

of

a given

or not.

either

recorded

as employed Thus,

angiogenesis

vessels

can be assessed inevitably

subsequent

CAM assay,

combination

Similarly,

as numerous

differences the

as many blood

outcomes

the

questions.

they

any given

twice

possible upon

the

exhibit

responses;

yield

one-half

that

many mechanistic

to

would

three

RESEARCH COMMUNICATIONS

angiogenesis.

out

either

of

the

dependent

to

be pointed answer

of

result

ultimately

should cannot

any

AND BIOPHYSICAL

all

positive. grading

such of

or by histological

specimens. EXPERIHENTAL

PROCEDURES

Angiogenin was isolated from normal human plasma by carboxymethylcellulose and high performance liquid chromatography as described (8). aFGF from bovine brain was purified by ammonium sulfate precipitation, CM-Sephadex CSO ion-exchange, and heparin-Sepharose affinity chromatography (27). Hitogenic activity of aFGF on Balb/c 3T3 cells was measured as described elsewhere (27 1. Angiogenic activity of angiogenin and aFGF, either separately or in combination, was assessed using the chick embryo CAM assay of Knighton et al. (26) as described in detail elsewhere (4,8). Briefly, 5-pL volumes of aqueous, salt-free samples were applied to Thermanox 15-mm disks (Miles Laboratories, Naperville, IL) and allowed to dry under laminar flow conditions. In the case of assays of combinations of angiogenic factors, the two proteins were mixed just prior to Although heparin has been demonstrated application to the disks. to bind strongly to and enhance the in vivo activity of aFGF (12), the angiogenic response elicited by angiogenin is not enhanced by the addition of heparin (unpublished observation). Therefore, heparin was not included in these assays. The dry, loaded disks were subsequently inverted and applied to the CAM surface of g-day-old embryos through windows which were cut on day 4. Eggs were viewed through a Nikon stereoscope and scored quantitatively for infiltration of blood vessels into the sample area and recorded as the number of positive angiogenic responses Statistical per number of eggs surviving per sample dilution. analyses were performed on data recorded at 68 + 2 h after sample A significance level of < 5% implantation as described (4,8). has to be attained for a sample to be considered active. 112.5

Vol. 146, No. 3, 1987

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

RESULTS AND DISCUSSION Our

laboratory

isolation

and

has been

characterization

promote

organogenesis

example

of

on the

the

of

the

heparin-binding paucity (4)

of

of

interactions

with

has

allowed

assay

Our to

induced

at

The assays

were

minimum

of

fifteen

percent

(2 S.E.M.)

angiogenin

third

given fraction

row,

probability for

are

at for

50, two

25,

a mixture and

and 12.5

each

of

0.7

in the 25,

0.3

in

+ 0.2.

and 0.3,

50 ng angiogenin

12.5

ng aFGF (J=2)

egg.

employing

a

CAM (Table

row for

at a 1.0

obtained ng per

Again, the

I).

assayed

first

12.5

in of

probabilities and

aFGF.

fraction

the

the

CAM

probabilities

mole to

of

1126

times

such

the

ng per

combinations

of

50,

or

(8)

here,

with

The average

a p (%) of fraction

reported

alone

the

of

employing

or three

example,

at

cells

plasma

studies

angiogenin

concentration

alone

human

for

either

applied

obtain

25 ng angiogenin

of

assayed

each

in

response

sample

tumor

However,

experiments,

the

angiogenin,

a mole

of

sample.

For

human

potential

suitable

for

of

2 (basic)

proteins.

calculated

fraction.

to

as of

angiogenic

per

angiogenin

The

from

present

amounts

of

three

structural/functional

is

either

number

assayed

averaged

into

combinations

eggs

of

(12-14,16,18,27,29,30).

series

repeated

of

angiogenin

class

concentrations

the mole

and

aFGF were

and aFGF in

J represents

1 (acidic)

angiogenic

by various and

termed

known

the

reported

properties

isolated

the

which

we have

one

as well

initial

in

as a specific

molecules:

angiogenin

assess

angiogenesis,

molecule

of

years

messengers

and biological

studies

that

several

Recently

originally

other

combination

were

factors

this

Angiogenin

for

growth

accumulation

was devised

mole

class

discovery

interactions.

any

others

extensive

relationships

and

particular.

angiogenic

angiogenin

precluded

recent

in

for

molecular

general

and chemical

types

(2,4-8,28),

of

in

former,

isolation

distinct

involved

egg in

(J=3)

the

average

and 25 ng aFGF and was 3.0

2 0.8.

Vol.

146,

No. 3. 1987

BIOCHEMICAL

Table mole

I.

CAM

AND

Assay

BIOPHYSICAL

of

RESEARCH

Angiogenin

and

COMMUNICATIONS

aFGFa

b

fraction

angiogenin

aFGF

JC

p wd

S.E.M.e

3

0.3

yo.

2

0.82

[50]

0.18

t12.51

1

0.7

0.70

[ 1

0.30

1 1

2

3.0

50.8

3

4.3

24.0 20.6

50 25

[ 1 50 25 12.5

0.53 0.36

t

0.22

[12.5]

[ 1 50 25 12.5

0.47

1

12.5 25

25 12.5

0

0.64

i 25 3 50

2

1.5

0.78

I501

1

2.3

3

3.4

[ 1 50 25 12.5

1.00

+2.3

aProteins were implanted onto the CAM either alone or &n combination as described in Experimental Procedures. Mole fractions for each protein combination assayed were calculated based on molecular weights for angiogenin and aFGF of 14 kD and 16 kD, respectively. The absolute amounts of each protein in nanograms per egg assayedcare given alongside the mole fraction in brackets (see text). The J value represents the number of different combinations at each male fraction for which individual probabilities were calculated. The average probabilities (in percent) were calculated from each individual probability (the

number

of

each

calculated mole fraction.

All

or

which,

statistically,

limits

of

the

upon

at

conclusion

an

an at

separately

mixing

the

the

least

inhibiting

neutral must

and

active p

be

(“a)

nor

two

proteins

at

range

interaction

they

possess

viewed

with

from

1127

0.3

is

concentrations.

of

between charges, since

4.3

the effect

i.e., ratio

to

within

several

opposite caution

in < 5%)

synergistic

combination, fraction

was

examined

Therefore,

additive

turn

a particular

probabilities

values

no

mole

in

at

aFGF

(i.e.,

equivalent.

charge pH

which

angiogenin

active

approximate

J value) assayed mean.

the are

the

combination error of the

are

CAM assay,

Interestingly,

reflect

of

Furthermore,

I).

evident

to

protein

Standard

(Table

since

esch

concentrations

combination

occurs

corresponding

for

the

p(%)

= 4.3, This

1:l. the

two

proteins

although S.E.M.

could

this is

+4.0.

BIOCHEMICAL

Vol. 146, No. 3, 1987

Figure

1.

of

identifying

as employed,

overall

Only

However, angiogenin

molecular

positive

by gross

although

angiogenesis,

precise

response.

(Figure

mechanisms

versus

observation

qualitative

1).

the

between

of

the

CAM is underway

of

the

activities

this

angiogenic

responses

induced

by mixtures

not

appear

additive,

in this

assay,

that

either

protein

angiogenesis

other

does not

influence

this

indeed

Whether

pathological proteins

is

state involved

the

or is

not

the

are at present

with

to physical

1128

able

to discern evaluation

the

two proteins

do

hypothesis

is

once one of

a given

response,

unknown.

indistin-

and that

case in a developmental

due only

by

Since

simplest

angiogenesis,

cellular

is

level.

by interaction this

recorded.

any differentiation

of

the

are

histologic

whether

to

induced

is

responses,

at

itself

responses

CAM assay

a gross

in the

or separately

can be revealed

them initiates

lend

angiogenesis

to determine

can induce

for

involved

negative the

Since

differences

allowing

does not

and aFGF in combination

guishable

RESEARCH COMMUNICATIONS

CAM assay depicting a positive (right-hand panel) and negative (left-hand panel) angiogenic response. The positive response was induced by 25 ng of angiogenin. A control disk (left-hand panel) contained 5 ,uL of water applied near the black spot. There is no gross visual difference between positive angiogenic responses induced by angiogenin, aFGF or combinations of the two.

The CAM assay evaluation

AND BIOPHYSICAL

and vice

cell,

the

versa.

or

properties

of

the

Vol.

146,

No. 3, 1987

The involve

BIOCHEMICAL

non-additivity competition

cell.

Alternatively,

second

site

--

is

so-called

aFGF

of

unable

of

results

populations

responding

could

if

occur

common

from

being

could

also

after or

of

angiogenin

mole

and

ratios

possibility result

in

protein

potentially the

1:l.

exists

the

that

of

is

each

result

in

thereby

affecting

account

for

It induce

for is,

occur

diffusion,

This

the

due

response

but

at

of

this

observation.

however,

known vivo,

is

that

that

both

proteins

can

that

observed

this

that

molar

ratio.

in

interplay

total

activity could etc.,

may certainly

to

angiogenin

could

catabolism,

allocated

being

the

For

molecules

the

at

each

complexing

the

1129

for

A dynamic such

mixed

plausible

present

differ

are

charged

angiogenin

systems

combination

seem

mechanisms

be unambiguously

in

been

cell

molecular

other

a result,

angiogenesis

above

active

solubility,

activity.

assay

the

proteins

at

Additionally,

vitro

basis

than

apparent

could

interaction

the

lower

observations.

the in

is

such

Currently, several

two

of

for

cannot

as yet

prevent

may then

oppositely

observed

that

angiogenesis,

least

the

these

in

such

the

of

diminished.

not

proteins.

for

combination

most

proteins

changes

It

which being

two

known

As mentioned

two

could

occupied

synergistic.

approaching

account

between

which

when

interaction

the

the

a specific

distinct

pathway

aFGF occurs

separately,

example,

to

above,

a response

has

from

as noted

Interestingly,

cells,

cells

arise

is

is

investigation.

binding

additive

it

responding where

site

between

a single

could

on the

first

endothelial

separately was

site

the

distinction further

mixtures

Although

such

await

COMMUNICATIONS

may be involved

when

with any

there

steps

sites

with

angiogenin

must the

receptor

respond

vitro

Hence,

possibilities

same

RESEARCH

to

cooperativity.

in

demonstrated.

to

to

BIOPHYSICAL

responses

different

interact

interaction

the

the

negative

can

above,

for

AND

with

apparent any

single

process.

aFGF utilizing

investigated

and aFGF, several

inhibition

to

elucidate

each

physical-chemical

able

to

Vol. 146, No. 3, 1987

properties.

BIOCHEMICAL

The amino acid

proteins

are distinct.

and its

activity

displays

Moreover,

of

these

has been demonstrated wide

variety

of

proliferation yet

aFGF binds

to interact types

two distinct

observed

via

to

of endothelial

receptors

heparin

above, with

induces

both

cells.

Anqioqenin

Therefore, interact

two

anqioqenin

specific

and specifically

assay

the

aFGF, as mentioned

molecules

in the

of

strongly

while

properties.

been so characterized.

these

effects

cell

RESEARCH COMMUNICATIONS

and sequence

by heparin,

and locomotion

has not

which

composition

is enhanced

neither

AND BIOPHYSICAL

a

as

the mechanisms to produce

system employed

here

by

the

can only

be

surmised. At present, are unknown Therefore,

precise

nor have it

is not

any physiological However, answer involved

studies

their

sites

of

known whether

circumstances such as those

such questions in modulating

and pathologic

physiological

as well the

roles synthesis these

complex

interact

should

the molecular

processes

proteins

been defined.

angioqenesis here

as define

these

molecules

to induce initiated

for

involved

under

in vivo. begin

to

mechanisms in normal

anqioqenesis. ACKNOWLEDGMENT

This work was supported agreements with Harvard

by funds University.

from

the Monsanto

Co. under

REFERENCES 1. 2. 3. 4. 5. 6. 7.

Denekamp, J. (1984) in Progress in Applied Microcirculation (Hammersen, F. & Hudlicka, O., Eds.), Vol. 42, pp. 28-38, Karqer, Basel. Vallee, B.L., Riordan, J.F., Lobb, R.R., Higachi, N., Fett, J.W., Crossley, G., Biihler, R., Budzik, G., Breddam, K., Bethune, J.L., Alderman, E.M. (1985) Experientia 41, 1-15. Folkman, J. & Klaqsbrun, M. (1987) Science 235, 4=-447. Fett, J.W., Strydom, D.J., Lobb, R.R., Alderman, E.M., Bethune, J.L., Riordan, J.F., & Vallee, B.L. (1985) Biochemistry 24, 5480-5486. Strydom, D.J., Fett, J.W., Lobb, R.R., Alderman, E.N., Bethune J.L., Riordan, J.F., & Vallee, B.L. (1985) Biochemistrv 24. 5486-5494. Kurachi, K.: Evie, E.W., Strydom, D.J., Riordan, J.F., & Vallee, B.L. (1985) Biochemistry 24, 5494-5499. Shapiro, R., Riordan, J.F., & Valse, B.L. (1986) Biochemistry 25, 3527-3532. 1130

Vol.

8. 9.

10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

30.

146,

No. 3, 1987

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

Shapiro, R., Strydom, D.J., Olson, K.A., 6 Vallee, B.L. (1987) Biochemistry (in press). Gospodarowicz, D., Cheng, J., Lui, G-H., Baird, A., & Bijhlen, P. (1984) Proc. Natl. Acad. Sci. U.S.A. 81, 6963-6967. Shing, Y., Folkman, J., Sullivan, R., Butterfield, C., Murray, J., & Klagsbrun, M. (1984) Science 223, 1296-1299. Thomas, K.A., Rios-Candelore, M., Gimenez-Gmego, G., DiSalvo, J., Bennett, C., Rodkey, J., & Fitzpatrick, S. (1985) Proc. Natl. Acad. Sci. U.S.A. 82, 6409-6413. Lobb, R.R., Alderman, E.M., & Fett, J.W. (1985) Biochemistry 24, 4969-4973. Lobb, R-R., Sasse, J., Sullivan, R., Shing, Y., D'Amore, P., Jacobs, J., & Klagsbrun, M. (1986b) J. Biol. Chem. 26, 1924-1928. Lobb, R.R., Rybak, S.M., St. Clair, D.K., & Fett, J.W. (1986) Biochem. Biophys. Res. Commun. 139, 861-867. Esch, F., Ueno, N., Baird, A., Hills, F., Denoroy, L., Ling, N Gospodarowicz, D., & Guillemin, R. (1985b) Biochem. BiAphys. Res. Commun. 133, 554-562. Strydom, D.J., Harper, J.W., & Lobb, R.R. (1986) Biochemistry 25, 945-951. Gimenez-Gallego, G., Conn, G., Hatcher, V.B., & Thomas, K.A. (1986) Biochem. Biophys. Res. Commun. 138, 611-617. Harper, J.W., Strydom, D.J., & Lobb, R.R. (1986) Biochemistry 25, 4097-4103. Jaye, M., How% R. Burgess, W., Ricca, G.A., Chiu, I.-M., Ravera, N.W., O'Brien, S.J., Modi, W.S., Maciag, T., & Drohan, W.N. (1986) Science 233, 541-544. Esch, F., Baird, A., Ling, N., Ueno, N., Hill, F., Denoroy, L Klepper, R., Gospodarowicz, D., Bohlen, P., & Guillemin, RI'(1985a) Proc. Natl. Acad. Sci. U.S.A. 82, 6507-6511. Abraham, J.A., Mergia, A., Whang, J.L. Tumolo, A., Friedman, J Hjerrild, K.A., Gospodarowicz, D., & Fiddes, J.C. (;;86b) Science 233, 545-548. Abraham, J.A., Whang, J.L. Tumolo, A., Mergia, A., Friednz-i, Gospodarowicz, D., h Fiddes, J.C. (1986a) The EMBO J Johrnal 2, 2523-2528. Gospodarowicz, D., Neufeld, G., & Schweigerer, L. (1986) Cell Differentiation 19, l-17. Thornton, S.C., Mueller, S.N., & Levine, E.M. (1983) Science 222, 623-625. midge, A. & D'Amore, P. (1986) Microvasc. Res. 31, 41-53. Knighton, D., Ausprunk, D., Tapper, D., & Folkman, J. (1977) Br. J. Cancer 35, 347-356. Lobb, R.R. & Fett, J.W. (1984) Biochemistry 23, 6295-6299. Palmer, K.A., Scheraga, H.A., Riordan, J.F., & Vallee, B.L. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 1965-1969. Lobb, R.R., Strydom, D.J., & Fett, J.W. (1985) Biochem. Biophys. Res. Commun. 131, 586-592. Lobb, R.R., Harper, J.W., & Fett, J.W. (1986) Anal. Biochem. g, 1-14.

1131