European Journal of Clinical Investigation ( 1983) 13, 289-296
`
`Angiogenic and growth factors in
`human amnio-chorion and placenta
`
`H:BURGOS, Blond McIndoe Centre, Queen Victoria Hospital, East Grinstead, Sussex, U.K.
`
`Received 1 1 August 1982 and in revised form 20 December 1982
`
`Abstract. Human amnio-chorionic membranes and
`placenta maintained in culture release factors with
`angiogenic and mitogenic capacities at concentrations
`corresponding to nanogram amounts of protein.
`Angiogenic activity of amnio-chorion and placenta-
`conditioned media was assessed by their ability to
`stimulate neovascularisation
`in the dorsal subcu-
`taneous fascia in the rat and the chorio-allantoic
`membrane in the chick embryo. Mitogenic character-
`istics were assessed by their ability to initiate DNA
`synthesis in cells at resting state, unstimulated periph-
`eral blood lymphocytes and serum-deprived 3T3 fibro-
`blasts. These growth promoting factors can be isolated
`from amnio-chorion and placenta-conditioned media
`mostly as factor-protein complexes of high molecular
`weight (higher than 100000 daltons) by gel filtration,
`and dissociated by magnesium chloride in components
`of low molecular weight including molecules readily
`diffusable through dialysis membranes of 2000 mole-
`cular weight cutoff. Presence of angiogenic and mito-
`genic factors in amnio-chorion suggests they might
`play a role in wound healing when amniotic mem-
`branes are used as biological dressings, besides the role
`they may play, in conjunction with placental factors, in
`embryonic and foetal development.
`
`Key words. Organ culture, conditioned medium, neo-
`vascularisation, cell division, growth substances,
`wound healing.
`
`Introduction
`Amniotic membranes have been used as dressings on
`ulcers, bums and other denuded areas [I]. Good
`clinical results were obtained by application of human
`amniotic membranes on chronic ulcers [2]. Formation
`of a highly vascularised granulation tissue after
`amnion dressing suggested presence of angiogenic
`factors in human amnion [3]. A culture system for the
`maintenance of viable human amnio-chorionic mem-
`
`Correspondence: Dr H. Burgos, Blond McIndoe Centre, Queen
`Victoria Hospital, East Grinstead, Sussex RH19 3DZ, U.K.
`00 14-2972/83/080O-0289$02.OO
`0 1983 Blackwell Scientific Publications
`
`branes was developed [4] which permitted the study of
`biological products released by these membranes.
`
`Methods and Materials
`
`Arnnio-chorionic and placental cultures
`Human amnio-chorionic membranes were collected
`aseptically from Caesarean sections in normal women
`at term. These membranes were thoroughly washed in
`phosphate buffered saline, pH 7-4 (Sigma London,
`U.K.) and cleaned of blood and maternal decidua.
`They were maintained for 1-8 days in modified tissue
`culture medium 199 (Gibco Europe Ltd., U.K.) or
`Hartmann’s solution (Travenol Laboratories Ltd.,
`U.K.) at pH 7.2-7.4 and 37°C in a humid atmosphere
`of 2.5% COz in air, following the procedures described
`in detail by Burgos and Faulk 141, except that serum
`was omitted throughout to prevent contamination by
`growth factors of serum origin notwithstanding the
`loss of optimal culture conditions. Supernatants were
`collected daily from the culture dishes and replaced by
`fresh medium. Collected media were individually
`centrifuged at 1800 g for 20 min at 4°C to remove any
`cells, concentrated 10-fold by ultra-filtration (Diaflo
`UM05, Amicon Corp., U.S.A.) and dialysed in Spec-
`trapor membranes with 1000 molecular weight cutoff
`(Spectrum Medical Inds., U.S.A.) against lo3 volumes
`of 0.15 mol/l sodium chloride for 24 h at 4°C. This
`preparation, called here amnio-chorion-conditioned
`medium (ACCM), was frozen or lyophilised, sterilised
`by gamma-irradiation (60000 R nominal dose) with a
`I3’Cs Gammacell (Isomedix Inc., U.S.A.), and stored
`at -20°C until use. Placental cotyledons were minced
`kith scissors and scalpels to yield fragments 1-3 mm3,
`and cultures and placenta-conditioned medium
`(PLCM) prepared as above. Tissue culture media
`submitted to identical procedures, including incuba-
`tion without placenta and amnio-chorion membranes,
`were used as control media (CTCM).
`
`Microculture assay for mitogenesis
`Peripheral blood lymphocytes (PBLs) from healthy
`individuals were prepared by the flotation method of
`289
`
`MTF Ex. 1042, pg. 1
`
`

`

`290
`
`H. BURGOS
`
`Boyum [5] on Ficoll-Triosil (Nyegaard & Co., Oslo),
`washed twice and resuspended in Dulbecco’s mini-
`mum essential medium (DMEM) (Gibco Europe Ltd.,
`U.K.).
`PBLs (1-5x los) were plated in triplicate in flat
`bottom microtest tissue culture plates (Falcon 3040,
`Beckton, Dickinson & Co., U.S.A.) in 0.1-0.2 ml
`DMEM supplemented with 15% heat inactivated goat
`serum (Batch 207076, Sera Lab, U.K.). This serum was
`chosen among different batches of human, horse, calf
`and goat sera because of its low mitogenic activity on
`human PBLs. BALB/c 3T3 fibroblasts (Flow Labora-
`tories Ltd., U.K.) were maintained in DMEM supple-
`mented with 4% heat inactivated calf serum (Batch
`49SL3, Sera Lab, U.K.), and cell monolayers dispersed
`with 0.02% trypsin/0.02% versene (Difco Laboratories
`Ltd., U.K.), washed twiceand resuspended in DMEM.
`3T3 cells (1-5 x lo3) were plated in flat bottom micro-
`test tissue culture plates in 0.1-0.2 ml DMEM supple-
`mented with 0.5-1.0 pg ml-’ hydrocortisone, sodium
`succinate (Organon Laboratories Ltd., U.K.) and 2-4
`mg mi-’ crystalline bovine albumin (Koch-Light
`Laboratories Ltd., U.K.). Serum was ommited to
`allow 3T3 fibroblasts to reach a resting state in 2-3
`days culture, but albumin included for cell survival,
`and hydrocortisone for full expression of mitogenic
`activity in absence of serum [6]. Both cell-type cultures,
`PBL and 3T3, were incubated at 37°C in a humid
`atmosphere of 5% COz in air with no change of
`medium throughout the culture period. Varied con-
`centrations (0-05-50%) of ACCM, PLCM and CTCM
`were added at different times. Cell proliferation was
`assessed at sequential intervals by pulsing parallel
`cultures with 1 pCi 3H-thymidine (3HTdR) (TRA.310,
`Radiochemical Centre, Amersham, U.K.) for 16 h.
`After this period of isotopic labelling, 0.1 ml of a 1%
`aqueous solution of Nonidet P40 (BDH Chemicals
`Ltd., U.K.) was added to each culture and incubation
`extended for a further 60 min to ensure cell lysis. DNA
`was then adsorbed onto glass-fibre filters by means of a
`SAM-2 multiple culture harvester (Cryotech, U.K.),
`and incorporated radioactivity determined by a liquid
`scintillation counting system (LKB-Wallac 1216, LKB
`Instruments Ltd., U.K.).
`
`Bioassays for angiogenesis
`The rat air sac and chorio-allantoic membrane
`assays described by Folkman et al. [7] and Phillips and
`Kumar [8] were modified as follows:
`
`Subcutaneous implant (SCI) assay. Varied concen-
`trations (l-SO%) of ACCM, PLCM or CTCM in 0.1
`ml DMEM were mixed with an equal volume of 1%
`purified agar (Difco Laboratories Ltd., U.K.) liquefied
`by boiling and maintained at 56T, and the mixture
`was injected into empty pure gelatin capsules (Parke,
`Davis & Co., U.K.). These capsules were previously
`coated with ethylene/40% vinyl acetate copolymer
`(Aldrich Chemical Co., Inc., U.S.A.) that was exten-
`
`sively washed in absolute ethanol and dissolved in
`(Aldrich Chemical Co.,
`Inc.,
`dichloromethane
`U.S.A.). Sustained release of proteins embedded in this
`polymer without appearance of inflammatory reac-
`tions if exhaustively washed in ethanol has been
`previously reported [9]. The polymer coating became
`itself a capsule as the original gelatin capsule was
`dissolved by the injected hot agar-medium. A small
`incision was made in the skin covering the dorsum of
`rats under ether anaesthesia, and the polymer capsule
`containing the semi-solid mixture of test medium, agar
`and gelatine was implanted subcutaneously 2-3 cm
`away from the incision. An elastic bandage (Elasto-
`plast) (T.J.Smith & Nephew Ltd., U.K.) applied
`around the animal’s body maintained the implant in
`place without the need of surgical stitches. Rats were
`killed 4 days later, and the skin and fascia overlaying
`the implant removed and examined for presence of
`new vessels with the aid of a stereomicroscope. Evalu-
`ation of results was performed single-blinded in 3-6
`rats per sample. Randomly bred female rats of AS
`strain, 6-8 weeks old, and 90-1 10 g weight were used as
`recipients.
`
`Chorio-allantoic membrane ( C A M ) assay. Briefly,
`small windows ( 1 x 1 cm) were cut along the horizontal
`axis of 8-10th day fertilised hen eggs, and egg-shell and
`shell-membrane removed. A minute perforation was
`made in the exposed CAM with a 30G needle, and 1-2
`mg of lyophilised ACCM, PLCM or CTCM placed
`over the hole with a thin spatula. The lyophilised
`powder absorbed humidity from the underlying tissues
`forming a small pellet that occluded the tiny hole.
`Shell-windows were then sealed with adhesive tape
`(Letraset Ltd., U.K.), and the eggs incubated at 37°C
`and 65% humidity. The CAM was examined daily for
`presence of new vessels by means of a stereomicros-
`cope, and evaluation of results performed single-
`blinded in 3-6 eggs per sample.
`
`Estimation of biological activity
`Mitogenic activities were quantified by the standard
`assay for growth factors described by Paul et al. [lo]
`with some modifications: 4 x lo3 3T3 fibroblasts were
`plated in flat bottom tissue culture plates in 0.1 ml
`DMEM supplemented with 0.2% heat inactivated calf
`serum (Batch 49SL3, Sera Lab, U.K.), and incubated
`in a humid atmosphere of 5% CO? in air for 72 h.
`Sedm-starved 3T3 cells maintained in cultures supple-
`mented with 0-2-0-4% serum show nil or negligible
`DNA synthesis after 2-3 days incubation [lo]. Two-
`fold serial dilutions of experimental and control
`samples were then added in a total volume of 0.1 ml per
`culture. Sample dilutions were carried out by standard
`titration technique in a liquid vehicle similar to that in
`which the samples were ultimately prepared (dialysing
`buffer or saline). Similar doubling dilutions of heat
`inactivated foetal calf serum (Batch 001 112, Sera Lab,
`U.K.) were used as standards for mitogenic activity.
`
`MTF Ex. 1042, pg. 2
`
`

`

`ANGIOGENIC AND GROWTH FACTORS IN PLACENTA
`
`291
`
`x 100
`
`Radioactive labelling was performed by adding 1 pCi
`3HTdR per culture after 24 h of sample addition, and
`incubation continued for a further 6 h. Determination
`of DNA incorporated isotope was carried out as
`described above. Results were calculated as percentage
`(units of activity) of the mitogenic activity expressed by
`cultures supplemented with 12.5% foetal calf serum
`[ 1 I] using the formula:
`Sample cpm - Background cpm
`FCS cpm - Background cpm
`where FCS represents the activity of 12.5% foetal calf
`serum, Background represents the activity in cultures
`supplemented with the dilution buffer or saline, and
`cpm the counts per minute given by the DNA incor-
`porated isotope.
`All sera were heat-inactivated at 56°C for 30 min,
`dispensed in 0.1-1.0 ml aliquots, and kept at -20°C
`until use.
`While mitogenic activity could be measured objecti-
`vely by radioisotope uptake, angiogenic activity was
`not amenable to quantitative measurements for lack of
`definite delimitations between degrees of angiogenic
`intensity. Although a distinction between weak and
`strong angiogenic responses could sometimes be done,
`counting new capillaries or capillary bifurcations and
`grading tortuosities and densities of vessels, as used by
`other workers [12, 131 was impossible to perform as
`positive responses were often profusely vascularised.
`
`Statistical analysis
`Results were analysed for significance with the
`Student’s t-test following the methods described by
`Lutz [14] and Schefler [15].
`
`Biochemical methoak
`Preliminary procedures for isolation and purifica-
`tion of active components were carried out by gel
`filtration and cellulose chromatography. Gel filtration
`of ACCM and CTCM (10-fold concentrated or lyo-
`philised) was performed through a column (1-6 x 90
`cm) of Sephacryl S-300 (Pharmacia U.K. Ltd.) pre-
`pared and equilibrated in 50 mmol/l Tns/HCI buffer
`pH 7 containing 0.1 mol/l NaCI, which was also used
`as elution buffer. The column was operated at 4°C with
`a flow rate of 12 ml h-I maintained by a LKB 2120
`peristaltic pump (LKB Instruments Ltd., U.K.). Frac-
`tions were collected in 3 ml volumes and monitored at
`280 nm.
`DEAE-cellulose chromatography of ACCM and
`CTCM diffusates through dialysis membranes of 2000
`molecular weight cutoff (Sigma London Chemical Co.,
`U.K.) was carried out in a column (1.5 x 20 cm) of
`DE-52 (Whatman Biochemicals Ltd., U.K.) prepared
`and equilibrated in 10% propan-2-01 (BDH Chemicals
`Ltd., U.K.), which was also used for elution. The
`column was operated as above.
`
`Results
`
`Angiogenic activity
`All conditioned media ( 18 different amnio-chorions)
`expressed angiogenic activity in both the SCI and
`CAM assays. Positive results were obtained in the rat
`SCI assay with 0.1 ml original supernatant or 0.010 ml
`concentrated medium in a total volume of 0.2 ml
`DMEM containing a final concentration of 0.5%
`purified agar. Higher concentrations of agar prevented
`free diffusion of active components. Original and
`concentrated media gave similar angiogenic patterns
`at the concentrations mentioned above. Smaller
`volumes of conditioned media failed in general to elicit
`a good response. On the other hand, larger concentra-
`tions yielded stronger angiogenic reactions with a
`clinical picture somewhat obscured because of profuse
`neovascularisation accompanied sometimes by minute
`haemorrhagic spots. Control media did not produce
`these types of responses. Inflammatory reactions were
`sometimes present but they showed a pathological
`picture with characteristics of acute inflammation and
`readily distinguishable from a positive angiogenic
`response. Fig. 1 shows typical positive reactions with
`ACCM from a 3rd day culture. Positive angiogenic
`response was demonstrated by newly formed capil-
`laries migrating towards the periphery of the im-
`planted capsule. In contrast, implants containing
`CTCM did not show neovascular formation.
`Positive angiogenic activity was also demonstrated
`in the CAM assay by radial migration of new capil-
`laries towards the pellets containing ACCM while
`those carrying CTCM did not draw out this pattern.
`Fig. 2 shows typical results in this assay 48 h after
`application of 2 mg lyophilised ACCM and CTCM on
`the chorio-allantoic membranes of 10-day-old chick
`embryos.
`Similar angiogenic responses were also found with
`conditioned media from placental cotyledon cultures
`(six different placentae).
`
`Mitogenic activity
`Ability of ACCM to stimulate proliferation of cells
`at resting state was demonstrated with normal unsti-
`mulated peripheral blood lymphocytes. Fig. 3 shows
`results obtained in a typical experiment with 1 x lo5
`cells in 0.2 ml DMEM supplemented with 15% goat
`skrum. Pilot studies showed the necessity of serum for
`survival of human PBLs under experimental condi-
`tions. Addition of ACCM and CTCM at 2% concen-
`tration (volume/volume) at time of plating, with no
`further addition, gave no significant (P < 0.05) uptake
`of 3HTdR between the 1st and 5th day of culture. A
`5-fold rise in radioactive incorporation was afterwards
`present in experimental cultures supplemented with
`ACCM over the control cultures supplemented with
`CTCM, and steadily maintained to the end of culture
`with a level of significance between experimental and
`
`MTF Ex. 1042, pg. 3
`
`

`

`292 H. BURGOS
`
`Figure 1. Subcutaneous implants of POlYVinYl CaPSUkS in rats. (A)
`Capsule containing amnio-chorion conditioned medium shows
`neovascularisation 4 days post-implantation. (B) Capsule contain-
`ing control tissue culture medium showing negative response.
`
`Figurt 2 Chorio-allantoic membranes of IZday-old chicks. (A)
`Neovascular response to lyophilixd amnibchorion conditioned
`medium placed 48 h previously. (B) Membrane containing control.
`medium showing a negative response.
`
`MTF Ex. 1042, pg. 4
`
`

`

`ANGIOGENIC AND GROWTH FACTORS IN PLACENTA
`
`293
`
`5
`
`4
`
`3
`
`2
`
`I
`
`C P y
`1110
`
`i'
`
`/
`
`T
`
`14
`
`I2
`
`10
`
`8
`
`6
`
`4
`
`C PM
`Ix 103
`
`2
`
`7
`
`-
`
`
`
`~
`
`I
`2 i '6 i Ibli
`DAYS IN CULTURE
`Figure 3. Effect of amnio-chorion-conditioned medium on 3H-thy-
`midine incorporation in normal. unstimulated peripheral blood
`lymphocytes.
`Experimental cultures supplemented with 2% amnio-chorion
`conditioned medium at time of plating (0-0). Control cultures
`supplemeuted with similar concentration of control tissue culture
`medium (0- - -0). Each point represents mean 2 S D of triplicate
`cultures. Increase in DNA isotope incorporation was measured in
`counts per minute (cpm).
`
`control values increasing from P>O.OI at 6 days to
`P > 0.00 1 at 12 days.
`Mitogenic activity of ACCM was also demonstrated
`in 3T3 fibroblast cultures. Dose response studies
`carried out to find effective ACCM concentration in
`serum-free cultures showed a progressively increasing
`3HTdR incorporation with increasing concentration
`of ACCM between 0.05 and 12% (volume/volume).
`Control cultures supplemented with similar concentra-
`tions of CTCM did not show incremental isotope
`uptake. Presence of 0-5-1 pg/ml hydrocortisone was
`indispensable for full mitogenic expression of ACCM,
`and presence of 2-4 mg/ml albumin for cell survival in
`absence of serum, as reported by others [6]. Fig. 4
`illustrates some of these results. 1 x lo3 3T3 fibroblasts
`were plated in 0.1 ml DMEM supplemented with 1
`pg/ml hydrocortisone and 4 mg/ml crystalline albu-
`min, replicate cultures having no supplementing hyd-
`rocortisone/albumin. Doubling concentrations of
`ACCM and CTCM (0.0516%) were added to parallel
`groups of cultures after 0,24,48 and 72 h incubation.
`Preliminary studies showed that neither media had any
`effect on attachment of 3T3 cells to the bottom of the
`culture wells.
`Cultures were incubated for 24 h after ACCM and
`CTCM addition, and labelled with 1 pCi 3HTdR for 16
`h afterwards, their radioactive uptake being deter-
`mined as described in Methods. Experimental cultures
`supplemented with hydrocortisone/albumin and
`ACCM at time of plating (0 h group: experimental in
`presence of HCT/BSA) showed a progressively in-
`cremental uptake of isotope from 0.05 to 8% ACCM
`concentration, reaching a plateau or deppression at
`
`I
`16
`
`I
`
`2
`
`4
`
`8
`
`aos OJ (u 05
`'la A C C M
`V l V
`Figure 4. Effect of increasing concentrations of amnio-chorion
`conditioned medium on 3H-thymidine incorporation in 3T3 fibro-
`blasts.
`Experimental cultures supplemented with amnio-chorion-condi-
`tioned medium at 0 h (time of plating) in presence (0-0)
`and
`of hydrocortisone/albumin, and at 72 h in
`absence (A-A)
`of hydrocortisone/albu-
`presence (0-0) and absence (A-A)
`min. Control cultures supplemented with control tissue culture
`medium in similar conditions showed no significant (P < 0.05)
`and
`differences between them, and only maximum (O---O)
`minimum (O---O)
`values are shown. Each point represents the
`mean SEM of triplicate cultures, the ordinate, amount of radioac-
`tivity incorporated in DNA (counts per minute), and the abscissa,
`concentration of conditioned medium.
`
`16% concentration. This represents a progressive
`increment between 2- and 5-fold higher than the
`uptake in control cultures supplemented with CTCM
`in presence of hydrocortisone/albumin (0 h group:
`control in presence of HCT/BSA). The degree of
`confidence for these values increased from a level of
`significant (P> 0.02) at 0.1% concentration to highly
`significant (P> 0.001) at 0.2-8% concentration. Exper-
`imental cultures added ACCM in absence of hydrocor-
`tisone/albumin (0 h group: experimental in absence of
`HCT/BSA) showed a much reduced incorporation of
`iiotope between 1 and 8% ACCM concentration. This
`was only 1.5-2.2 times higher than in control cultures
`(0 h group: control in absence of HCT/BSA) with a
`significant level of confidence (P> 0.01). Similar
`results were obtained if ACCM addition was delayed
`for 4 and 30 h after plating. After this time, however,
`the longer the delay in ACCM addition, the lower the
`incorporation of 3HTdR. Nevertheless, incorporation
`was still present in cultures supplemented with hydro-
`cortisone/albumin and pulsed with ACCM after 3 days
`incubation (72 h group: experimental in presence of
`
`MTF Ex. 1042, pg. 5
`
`

`

`294
`
`H. BURGOS
`
`HCT/BSA). This uptake was only between 1.6 and 3.2
`higher than in control cultures (72 h group: control in
`presence of HCT/BSA) at 1 and 8% concentration
`respectively, with degrees of confidence increasing
`from probably significant ( P > 0.05) to significant
`(P> 0-01). Experimental and control cultures incu-
`bated for 3 days in absence of hydrocortisone/albumin
`(72 h group: experimental and control in absence of
`HCT/BSA) did not show any significant ( P c 0.05)
`uptake of isotope 24 h after addition of ACCM and
`CTCM.
`Fig. 5 illustrates the results of a kinetic study to
`demonstrate the ability of ACCM to initiate DNA
`synthesis in serum-deprived 3T3 fibroblasts at resting
`state. One-thousand cells were plated in 0-2 ml DMEM
`supplemented with hydrocortisone/albumin as in the
`previous experiment. A concentration of 1% ACCM
`(from a 3rd day culture) added after 3 days incubation
`was enough to initiate 3HTdR incorporation 24 h later.
`This was maintained by additional pulses made every
`other day. Control cultures supplemented with similar
`concentrations of CTCM showed no rise in isotope
`incorporation. Differences between experimental and
`control mean values were statistically significant
`( P > 0-05-P> 0.01). To determine if mitogenic activity
`was limited to chorion or amnion, these membranes
`were separated by blunt dissection and cultured inde-
`pendently. Culture supernatants from the three anato-
`mical entities, chorion, placental amnion and reflected
`amnion, were screened without further preparation or
`concentration for presence of mitogenic capacity using
`the doubling dilution assay described in Methods.
`Table 1 shows the results obtained. All supernatants
`expressed mitogenic activities although their potency
`diminished with time of culture. Similar mitogenic
`capacities were also present in placenta(coty1edon)-
`conditioned media. Fig. 6 illustrates results of a typical
`experiment using the assay for estimation of mitogenic
`5.
`
`I
`6 1
`
`-
`
`Table 1. EKect of amnion and chorion supernatants on 'H-thymi-
`dine incorporation in 3T3 fibroblasts
`
`Supernatant
`
`Chorion
`
`1st
`2nd
`3rd
`1st
`Reflected amnion 2nd
`3rd
`1st
`2nd
`3rd
`1st
`2nd
`3rd
`
`Placental amnion
`
`Control medium
`
`Background
`
`Supernatant concentration
`
`Day or
`culture 50%
`
`25%
`
`12.57;
`
`6.25%
`
`23018 20013
`19172 11194
`18802 12705
`15025 14918
`11728
`7323
`9591
`8610
`12145 11052
`10820
`5311
`8302
`4704
`1679
`2159
`1293
`1187
`I323
`1170
`263
`176
`
`18491
`9343
`7379
`13644
`6087
`4909
`8840
`3331
`4568
`2156
`1550
`1633
`625
`
`24178.
`25598
`18729
`12550
`8400
`3273
`13323
`15662
`5228
`1992
`1935
`1466
`871
`
`Mean counts per minute of triplicate cultures. Standard
`deviations vaned between 47 and 6375 cpm.
`
`activity. Placenta-conditioned supernatants showed as
`good or better mitogenic activity than 12.5% foetal calf
`serum under assay conditions. Samples stored over 1
`year, as described in Methods, showed no detectable
`loss of activity.
`Placenta and amnio-chorion extracts in normal
`saline, and aqueous alcohol and acid solutions also
`expressed strong mitogenic and angiogenic activities
`(in preparation).
`
`UNITS OF
`ACTIVITY
`
`C Phi
`la103
`
`DAYS IN CULTURE
`Figure 5. Effect of amnio-chorionconditioned medium on 'H-thymi-
`dine incorporation in serumdeprived 3T3 fibroblasts.
`supplemented with O.S% amnio-
`Experimental cultures (0-0)
`chorion conditioned medium on days 3, S and 7 (arrows). Control
`cultures (0- - -0) supplemented with similar concentrations of
`control tissue culture medium. Other symbols and lines are as in Fig.
`5.
`
`*I*
`Figure 6. Mitogenic activity of increasing concentrations of placenta
`conditioned supernatants on quicscent 3T3 fibroblasts incubated for
`3 days in serum-starving conditions.
`Placentasonditioned supernatant from 1st day (0-0)
`and 3rd
`day ( O d ) cotyledon cultures. Mitogenic capacities are shown as
`percentage (units of activity) of the mitogenic activity expressed by
`12.5% foetal calf serum. Each point represents the mean+SEM of
`triplicate cultures, and the abscissa, concentration of conditioned
`supematants.
`
`MTF Ex. 1042, pg. 6
`
`

`

`ANGIOGENIC AND GROWTH FACTORS IN PLACENTA
`
`295
`
`Preliminary biochemical studies
`Components of different molecular size were iso-
`lated from 4 ml 10-fold concentrated or lyophilised
`(and reconstituted) conditioned medium supernatants
`by gel filtration through Sephacryl S-300 as described
`in Methods. Fig. 7 shows a typical pattern at 280 nm.
`Fractions under discernible peaks were pooled and
`their angiogenic and mitogenic activities determined.
`Angiogenic and mitogenic activities were both present
`mainly in the firstly eluted fractions. Fractions under
`
`The diffusate was applied to a DE-52 column and
`fractionated as described in Methods. Fig. 8 shows a
`typical optical density pattern at 280 nm. Angiogenic
`and mitogenic activities were present in the first peak
`(fractions 9-17) with a protein content of 25 pg/ml,
`and only negligible activities in the second peak
`(fractions 18-27) with a protein content of 53 pg/ml. In
`contrast, stronger activities (3-8 fold) were found in
`the area corresponding to fractions 71-80 showing no
`discernible peak at 280 nm, a maximum optical density
`of 0.02 (fraction 76) at 206 nm but undetectable
`protein content by Lowry’s assay [17].
`
`0.3 4
`
`0 0
`280 n n
`
`1.0-
`
`0 . 8 -
`0.6 -
`
`0 . 4 -
`
`0 . 2 1
`
`O D
`28Onn
`
`FRACTION NUMBER
`x 3 nl
`Figure 7. Optical density profile obtained by filtration of4ml 10-fold
`concentrated amnio-chorion conditioned medium through Sephac-
`ryl S-300.
`Increase in optical density (OD) was measured at 280 nm. Elution
`areas or catalase (mol. wt. 232000) and albumin (mol. wt. 67000).
`marked CAT and ALB respectively, are shown as reference for
`molecular weights.
`Angiogenic and mitogenic activities were present mostly in
`fractions under peak I, following fractions showing low or negligible
`activities.
`
`the first peak showed these activities at much higher
`degree (5-10-fold) than fractions under consecutive
`peaks, the last eluted peaks being weakly positive or
`negative. Corresponding results were obtained after
`normalizing pool volumes by appropriate concentra-
`tion to rule out dilution effects. Relative and total
`protein contents were not correlated with their potency
`(41, 76, 69, 50, 43, 49 and 35 pg/ml protein for peaks
`1-7 respectively). Dissociation of proteins by MgCh
`[ 161 yielded free diffusable components that were
`separated by dialysis and anion-exchange chromato-
`graphy. Angiogenic and mitogenic activities were
`present in Visking cellulose diffusates (in distilled H20
`or 150 mmol/l NaCl for 24-48 h) while retentates were
`devoid of detectable activities. Positive activities were
`also found in diffusates (in distilled H2O or 10%
`propanol) of dialysis for 2, 4, 16 and 48 h in
`benzoylated membranes with 2000 molecular weight
`cutoff, their potency progressively increasing with
`dialysis time. Retentates in these cases, however, also
`showed positive activities. DEAE-cellulose chromato-
`graphy of 48 h diffusates yielded discrete separation of
`low molecular size components with angiogenic and
`mitogenic activities. Four millilitres of 10-fold concen-
`trated ACCM were mixed with MgC12 6HtO at 2 M
`concentration, and dialysed (membrane with 2000
`pore size) in 16 ml 10% propan-2-01 for 48 h at 4°C.
`
`FRACTION NUMBER
`a 3 mI
`Figure 8. Optical density pattern obtained by elution of 16 ml 107;
`propanol diffusate of amnio-chorion conditioned medium from a
`DEAE-cellulose chromatography column.
`Four millilitres of 10-fold concentrated amniochorion condi-
`tioned medium mixed. with MgClz to dissociate protein complexes.
`were dialysed against 16 ml 10% propanol in membranes with 2000
`molecular weight cutoffand thediffusate fractionated in thecolumn.
`Increase in optical density (OD) was measured at 280 nm.
`Angiogenic and mitogenic activities were present mainly in
`fractions 71-80, and fractions under peak I. and only negligible
`activities in fractions under peak 2.
`
`Discussion
`This work reports for the first time angiogenic and
`mitogenic activities in normal human amnio-chorionic
`membrane and placenta cultures. A distinction
`between angiogenic and mitogenic activities is made in
`this report because of the specific characteristics
`demonstrated by the angiogenic and mitogenic assays,
`not because these capacities could be separated as
`independent activities. Presence of angiogenic and
`mitogenic factors in amnio-chorion suggests that they
`might play a role in the production of a healthy
`granulation tissue in ulcers, bums and other wounds
`following biological dressing with amniotic mem-
`branes, in addition to the role they may play together
`with placental factors in embryonic and foetal de-
`velopment.
`Similarity between amnio-chorionic and placental
`angiogenic factors and angiogenic factors from other
`sources is not unlikely but has not been established.
`Tumour angiogenic factor (TAF) and endothelial cell
`stimulatory factor (ESAF) are also present as factor-
`carrier complexes of high molecular weight (about
`100000 daltons) in extracts and unpurified prep-
`arations [7, 16, 181. TAF and ESAF complexes have
`
`MTF Ex. 1042, pg. 7
`
`

`

`296
`
`H. BURGOS
`
`been dissociated by MgC12 yielding a compound of
`200-800 molecular weight with angiogenic activity [ 16,
`191. Presence of angiogenic and mitogenic activities in
`2, 4, 16 and 48 h ACCM/MgC12 diffusates through
`dialysis membranes with 2000 molecular weight cutoff
`suggests a component of very low molecular size.
`Mitogenic activity of ACCM and PLCM shows
`similar characteristics to that found in epidermal
`growth factor (EGF), fibroblast growth factor (FGF),
`and other growth promoting factors of this kind.
`ACCM and PLCM are effective on different cell types,
`and lack species specificity. These are characteristics
`shared by EGF and FGF [20,21]. ACCM and PLCM
`are released as high molecular weight complexes that
`can be dissociated in active components of low
`molecular weight. This characteristic is also found in
`EGF [20, 221. For maximum DNA synthesis in
`serum-free 3T3 cultures, ACCM and PLCM need the
`presence of hydrocortisone, a characteristic found in
`FGF as well [6]. ACCM and PLCM are active even at
`very low concentration, in the order of 0.05-1%. This
`would correspond to activity at nanogram concentra-
`tions of protein found with other growth promoting
`factors [20, 21, 221. However, no correlation between
`degree of activity and protein content was found in
`ACCM fractions, this being a reason for relating
`activity to volume rather than protein content in the
`present report. Concentrations of ACCM and PLCM
`over saturation levels show a repressing mitogenic
`effect. This is a characteristic also found with FGF and
`EGF f21, 231. Amnion and trophoblast have been
`found to express high EGF activity in the mouse [24].
`EGF receptors have been reported in human placenta
`125, 261, and FGF in bovine placenta [6]. Colony
`stimulating factors (CSFs) have been isolated from
`human placenta-conditioned medium [27l. The mito-
`genic effect of ACCM and PLCM on peripheral blood
`lymphocytes suggests presence of factors with similar
`activities to lymphocyte growth factors in amnio-chor-
`ion and placenta. Very recently, while revising the
`present report, the production of Interleukin 1 by
`placental mononuclear phagocytes has been published
`[28f. Accumulated information indicates that placenta
`and amnio-chorion may be rich sources for isolation
`and purification of diverse growth promoting factors.
`
`Acknowledgments
`This work was supported in part by grants from the
`Medical Research Council and The East Grinstead
`Research Trust. The author wishes to thank Mr G.
`Warriner for technical assistance.
`
`References
`1 Eldad A, Stark M, Anais D, Golan J, Ben-Hur W. Amniotic
`membranes as a biological dressing. S Afr Med J 1977;51:272-5.
`2 Bennett JP, Matthews RN. Faulk WP. The treatment ofchronic
`ulceration of the legs with human amnion. Lancet 19801: 11 53-6.