Wound healing and bone regeneration.

This invention relates to healing wounds. Growth factors are polypeptide hormones which stimulate a defined population pf target cells. Examples of growth factors include platelet-derived growth factor (PDGF), insulin-like growth factor (IGF-I), transforming growth.factor béta (TGF-B), epidermal growth factor (EGF), and fibrobiast growth factor (FGF). PDGF is a cationic, heat-stable protein , found in the granules, of circulating platelets which is

Î0 known to stimulate in. vitro protein synthesis and collagen production by fibroblasts. It.is also known to act as an in vitro mitogen and chemotactic agent for fibroblasts, smooth.muscle celle, and glial cells.

It has been proposéd. to use PDGF to promote in it* vivo wound healing. For example, Groténdorst (1984) J.

Trauma.24:549-52 déscribes adding PDGF to Hunt-Schilling wire mesh chambers impregnated with a collagen gel and implanted in the backs of rats; PDGF was found to increase the amount of new collagen synthesized.

However, Leitzel et al. (1985) J. Dermatol. Surg. Oncol. 11.:617-22 were unable to accelerate normal wound healing in hamsters using PDGF alone or in combination with FGF and EGF.

; - -°⁹ * 59

- '. . - / . /: -/- 2 ' ' ./.?:: . ’

Michaèli, et àl. (1984) In Soft and Hard Tissue . Repair (Huât; T.K. et/al/. Eds),.-Praeger Publishérs, New ·'./ïotk', pp. 380-394, report thatapplication .of/â_:: _;partïally purifled préparation of PDGFôbtâinéd from'

- platelat-rich plasma stimulatéd angiogenesis when ·.

implanted in rabbit cprnéas. Becausè/PDGF is not/an angiogenic growth factor the investigators suggésted / that an unknown factor in -.jt-héir. partially pur if ied; PDGF , préparation was responsiblé for the angipgénic effeet.

₀ Summary of the -Invention' • In général, the invention, featurés, là oné/ ' ' aspect, healing an external wourid in. a mamrnal,e.g., a human patient,byapplyingtothewoundan effective _;amount of a'composition that: includes/purifled PDGF and purifled IGF-Ï. The composition aids/in/healing the wound,..-at- least in part, by promoting the growth qf épithelial andconnectivetissue and/thé 'synthesis of total protein andcollagen.Wound healing using the composition of the inveiition is mofé effective than that achieved inthe absence of treatment (i.e., withput applying expgênous agents)/or /by treatment :with; purifled

PDGF alone, or.'purlf-'ied/.iGÊ/l-.éhjné,.//·..///,///·//..'/-. / ' In. another aspect, the inventionfeatupes / < regeneràting bonë /of a mâmmal, e.g. , ’a human. patient, by 2$ administering to the patient, preferably by application to the area of injured pr depletéd bone, an effective amount of a composition that includes purifled PDGF and purifled IGF-I. The composition aids in regeriëration, atleast in part, by promoting the growth of connective 30 tissue, bone, and cementum/and :by stiînulâting protein and collagen synthesis .' /Régénération using thé composition of the invention is more effectivethah that achieved in the absence .of treatment ii. e, / without

<img file="OA9159A_D0001.tif" />

09159

- 3 applying exogénous agents) or by treatment with purified PDGF alone, or purified IGF-I alone.

In preferred embodiménts of both aspects of the invention, the'composition is prepared by combining, in a pharmaceutic.ally acceptable, carrier substance, e.g., commerciallÿ available inert gels or liquids (e.g., saline supplemented with albumin or méthyl cellulose), purified PDGF and IGF-I (both.of which are commerciallÿ available). Most preferably purified PDGF and IGF-i are .· combined in a weight-to-weight ratio of between 1:4 and 25:1, preferably between 1:2 and 10:1, and more preferably 1:1 or 2:1. The purified PDGF and IGF-I may be obtained from human platélets or by recombinant DNA technology. Thus, by the terms PDGF and IGF-I we mean both platelet-derived and recombinant materiâls of mammalian, preferably primate, origin; most preferably, the primate is a human, but can also be a c.himpanzee or other primate. ' Recombinant PDGF can be recombinant heterodimer, made by inserting into cultured prokarÿotic or eukaryotic cells DNA sequences encoding both . subunits, and then allowing the translated subunits to be processed by the cells to form heterodimér, or DNA encoding just one of the subunits (preferably the beta or 2 chain) can be inserted into cells, which then are

2$ cultured to produce homodimeric PDGF (PDGF-1 or PDGF-2 homodimer).

The term purified as used herein refers to PDGF or IGF-I which, prior to mixing with the other, is 95% qr greater, by weight, PDGF or IGF-I, i.e., is substantially free of other proteins, lipids, and carbohydrates with which it is naturally associated.

- ' 4L

-/ .

09159

2$ &lt; A puf if ied protein. préparât ion will geriérally . yield a single major band on a polyacrylamide gel for · ' each subunit of PPGF^;.qr ÎGFr-i. Most/jprefèrablÿ, thé -purified PDGF oriGF-ï used in-thecompositions of the invention is pure as ’ judged by aminô-tetminal amino acid seguericé analysis . J . · - ''/ · .

Thé compositiph pf/ the invention provides a fast, effective method/for healing/exteinal wounds of mammals, e.g.-, bèd sores, làcetatiôhs and burns. ' The composition enhances connective/tissue formation compared. to nàtural healing (i.e. no exogenous agents i added) or pure PDGF or /IGF-I.-àloftè. fJhlike /pure PDGF alone, thé composition promotes about a 250% increase in new connective tissue andaboutaS5% increasein the growth of ·épithélial tissûéi The epithélial layer y obtained is thiçker than/that çreatedby/ hatural \ héaling, and also contains moré épithelial projections connecting it to the new connective tissée; itis thus more firmly bound and: pfôteçtivé. ïn addition, sçar formation is minimized.

The composition of/the invention also provides a fast, effective method for régénération, of connective tissue. and bone of mammals, e. g ·, humans, with a hi story of peridontâl disease. The composition enhances / connective tissue and boné formation compared to nàtural healing (i.e. no exogenous agents added) or /pure,PDGF or IGF-I alone. ’ · /-·.- : -./&lt;Other featurés-âncl advahtagés of the invention will be apparent from the following description of the preferred embodiments:théréof, and from thé claims

<img file="OA9159A_D0002.tif" />

0915δ . . . - 5 - ' '

Description of the Preferred Embodiments We now desçribe preferred embodiments of the invention.

External wounds, e.g., bed sorés and burns, are treated, and bone and connective tissue regenerated, according to the invention, with. PDGF/IGF-I mixtures prepared by combining pure PDGF and IGF-I. IGF-Σ is commercîally availabié from Amgen Corporation (Thousand Oaks, CA) and Kabi (Sweden). -Purified recombinant PDGF and purified PDGF derived from human platelets are commercîally available from PDGF, ïnc. (Boston, MA), collaborative Research (Waltham, MA), and Amgen Corp, . (Thousand Oaks, CA). Purified .PDGF can also be prepared as follows:

Five hundrèd to 10.00 unit s of washed human ’ platelet pellets are suspended in IM NaCl &lt;2ml per platelet unit) and heated at 100°C for 15 minutes. The supernatant is then separated by centrifugation and the précipitât© extracted twice with the IM NaCl. ‘

The extracts are combined and dialyzed against 20 0.08M NaCl-O.OlM sodium phosphate buffer (pH 7.4) and mixed overnight at 4^eC with CM-Sephadex C-50 equilibrated with the buffer. The mixture is then poured into a column (5 x 100 cm), washed extensively with 0.08M NaCl-O.OlM sodium phosphate buffer (pH 7,4), and eluted with IM NaCl while 10 ml fractions are collected.

Active fractions are pooled and dialyzed against 0.3M NaCl-O.OlM sodium phosphate buffer (pH 7.4), centrifuged, and passed at 4°C through a 2.5 x 25 cm column of Blue Sepharose (Pharmacie) equilibrated with 0.3M NaCl-O.OlM sodium phosphate buffer (pH 7.4).

£ '· :/.-/ . , 0OO

- 6 -^; ' · . .«· ' ' : The column’is then washed W'ith/ the? buffer .and partially purified PDGF éluted wit-h/a 1:1 solution of 1M-NaCÎ and ethylene glycol.

. The partially. purif ied PDGF fractions are diluted (1:1) with lM NaCl,/diàlyzed against 1M açetic acid, and lyophilized./ The lyophilized samplés are dissolved' in 0,8M NaÇl-0 / 01M sodium· phosphate? buffet ( pH 7.4) and passed through a 1.2 x 40 cm cdlumn of Œ-Sephadex C-50 equilibrated with the' buffet. PDGF is -then eluted with a NaCl, gradient (0.08\to IM) .

The active fractions ate combinedi/ dialyéed against IM acetic acid, lyophilized, and dissolved in a small volume of IM acetic acid. 0,5 ml portions are applied to a 1.2 x 100 cm column of Bipgel P-150 (100 to 200 mesh) equilibratéd with. 1M acetic acid. -The PDGF is then eluted with IM acetic acid while 2 ml fractions are collected. '^:

Each active fraction-containing 100-to'200 mg of protein is lyophilized, dissolved in 100 ml of 0.4% trifluoroacétic acid, and subjiectèd to reverse phase high performance liquid chromatography on a phényl Bondapak column (Wa.ters)à Êlution with a linear acetonitrile gradientL(0 tp 80%l yields pure PDGF.

PDGF made by recombinant DNA'tephnology càn be prepared as follows:

Platelet-derived growth factor (PDGF) detived from human plateléts çontains two polypeptide, séquences (PDGF—1 and PDGF-2 polypeptides; Antoniades, H.N. and Hunkapiller, M. (1983) Science 220;963-965). PDGF-1 is encoded by a gene local ized in chromosome 7 (Betsholtz, C. et al., Nature 320:695-699)and PDGF-2 is/encoded by the sis oncogene (Doolittle-, R. et al. ( 1983) science 221;275-277) îocalized in chtômosome 22 (Dalla-Favera, R, (1982) Science 218:686-688). The sis gene encodes

0151 ^; - 7 — . . .

the transforming protein of the Simian Sarcoma Virus (SSV) which is closely related to PDGF-2 polypeptide. The human cellular c-sis also encodés the PDGF-2 chain ·. (Rao, C.D, et al. (1986) Proc, Natl. Acad. Soi; USA

83:2392-2396). Because thé two polypeptide chains of

PDGF are coded by two different genes.localized in separate chromosomes, the possibility ..exists that human PDGF consists of .a disulfide-linked heterodimer.of · PDGF-1 and PDGF-2, ôr a. mixture of the two homodimers

W (homodimer of PDGF-1 and homodimer of PDGF-2), or a mixture of the heterodimer and the two homodimers.

Mammalian cells in. culture infeqted with the Simian Sarcoma Virus, which contains the gene encoding the PDGF-2 chain, were shpwn to synthesize the PDGF-2 polypeptide and to process it into à disulfide-linked homodimer (Robbins, K. et al..(1983) Nature 1 305:605-608). In addition, PDGF-2 homodimer reacts with antisera raised against human PDGF. Furthermore, the functional. properties of the secreted PDGF-2 homodimer are similar to those of platelet-derive'dPDGF in that it stimulâtes DNA synthesis in cultured fibroblasts, it induces phosphorylation at the tyrosine residue of a 185 kd cell membrane protein, and it is capable of competing with human (¹²⁵i)-pdgf for binding to spécifia cell surface PDGF receptors (Owen, A. et al. (1984) Science 225:54-56). Similar properties were shown for the sis/PDGF-2 gene product derived from cultured normal· human cells (for example, human arterial endothélial cells), or from human malignant cells expressing the sis/PDGF-2 gene (Antoniades, H. et al. (1985) Cancer Cells 3:145-151).

The recombinant PDGF-2 homodimer is obtained by the introduction of cDNA clones of c-sis/PDGF-2 gene into mouse cells using an expression vector, .The /&lt; ;/?-:;/ OÎW

- 8/-/ ' · · :/.

tô c-sis/PDGF-2 cloné used' fo'r the exprès si dn vasobtairiëd / from’ normal human. cùïtuted endotheliai cells .^Colline;

T., et al ./(1985) 'Nature 216:748-750). /

Wound Healing '. ' ;,//- .. . ·

To détermine-the;_:effectiveness of//PDGF/ÎGF-I / mixtures inpromôtihg/ wound healing/ the following ^Λ. expérimente wëre- pèrformed//

- Yoûng white /ïorkshira piçjs (Pafson'a Farm, / Hadley, MA) weighing between &lt; lG..and 15 kg were fasted for at least 6 hours priôr to surgeryand thêri/_ . anesthetized. Under aseptie conditions,, the backand thoracic areas Were clipped/ shaved, and washed with ; mild soap and: water . The area. tp/be wounded was/then

- àisinfected with 70£ àlcohol. ' .

Wounds mëasurincf 1 cm x 2 cm were induced ata / . depth of 0.5 mm using a/modified Castroviëjà. . electrokeratpme (Stdrz, St./ Louis, MO, as modified by · /Brownells,/ϊησ.). /. The wounds resuîtëd /ixx complété; ./ removal of the epithelium, às well as a /port ion ; of the &lt; -, underlying dermis (comparable to a second degree burn/: injury). Indiyidual/ wounds were/ sepàratéd byat least 15 mm of unwounded skin. wounds/tëceiving /idëntical treatment were organized as /a group» /ând separated: f rom other groups by at lëast/3 cm. //Wounds/ reçëdving/ no;/:/ growth factor treatment were separated from wounds receiving such treatment by.at/least' 10 cm. v

The wounds were treated direçtly with a single . /application of the following growth factors/ suspénded in biocompatible gelr 1) 500 ng purehumanPDGF (purif ied; by high performance liquid: chromâtogfaphy) or . / recombinant PDGF- àlone; 2) 500 ng pure PDGF in // ; combination with each of the following: a) 500 ng EGF; b) 500 ng EGF plus 500 ng IGF-I;c) 500 ng IGF-i. ;;

09159 stôrëd in cold media

The biopsy 10

Following wounding, biopsy specimens were takeir on days 3 through 10. .Biopsy specimens for histologie, évaluation were taken as wedges approximately 3 mm deep and placed in 10% formalin. Specimens for biochemicàl analysis and autoradiography wéré obtained using an electrokeratome. The final dimérisions of.the specimens· were 1.5 mm x 10 mm x 1.5 mm. Three specimens per. wound were collected for biochemicàl analysis, while two specimens per wound collected for autoradiography. . Following collection., the specimens were Eagle's Modified Essential Medium (EMEM) supplemented with 10% fêtai calf sérum, specimens were analyzed as follows.

Autoradiography

Biopsy specimens were incubatéd Eagle's Modified Essential Medium (EMEM) calf sérum containing 15 pCi/ml of H-thymidihe for one hour; the specimens were then washed .twice with 10% formalin. Four micron sections were made from the specimens using standard paraffin imprégnâting and embedding techniques. They were thendéparaifinized, dipped in Nuclear Tract émulsion ntb-2 (Kodak), and exposed for two weeks. Subséquently, they were developed and stained with hematoxylin and eosin.

Autoradiograms of the stained specimens were recorded and scored at equally distributed points by counting the total number of labelled cells (5 grains or more per cell) versus the total number of cells in the area.

Histologie Evaluation

Histologie specimens were prepared. using standard paraffin imprégnating and.embedding techniques. Four micron sections were made and stained using filtered Harris hemotoxylin and alcoholi,c eosin;

I

<img file="OA9159A_D0003.tif" />

ΙΟ:

they were then: observed_;under ; a microsçopé . ; Ail / 'spécimens were scored, blindiy by two: invéstigators at equally distributed points throughout the/sections:, .The widths of the epithélial and;.connective/tissue, layers were scored using/a grid. placed within the ocular: of the microscope; the méasuremèiïtwas then/converted into ' . millimeters/using a micrometer viewed under tiie/same . conditions. ' .· / / -, ^:

In Vitro. Ketabolic Labellinq

Biopsy specimens weretransferredto individual : tubes containing 0,3 ml df EMEM. plus io%:fetàl -calf/ sérum, 15 pCi/ml H-Thymidine,and5 jiÇi/ml / ^C-leucine, and incubated for one' ftour at :37°Ç;. At the end ôf the hour, the/ medium: was: removed and t-issue · . metabo.lism stopped by adding 0.5 ml cold 5% pérçhloric/ acid. The spécimens were then f inely mineed and washed; three times; -the resuiting precipitate was resùspended in 0.5 ml 14.8 M ammonium hydroxide:and sônicated./ / Follawing sônicatiôn, the spécimens-,were incubated in ?

séalsd tubes at 45^eÇ for 16^-24 hours,/ résonicated, and the radioactivity measured usihg standard liguid / / scintillation méthode with/cross: çhannel çounting .: DNA and Protein Détermination &lt;

DNA’ détermination/ was/ per f ormed Using/a φ modification ôf the method of Labafca et al. (1980) Anal. Bïochem. 120:344-52. A 50 μί/àliguot of Tissüe extract in concentrated ammonium hydroxide was added to 400 μΐ of a buffer solution containing 1 :M sodium phosphate and 2M sodium chïoridé /(pFf 7/01; thé pH of the jû resulting solution was adjusted to 7.4 :using HCl..

Afterwards, the f inal solution volume was broUght· to 500 μΐ, while maintaining the pH ât :7.4. / ^ then addèd to 2.5 ml of a bufféred solution' (0/05 M / sodium phosphate, 2M sodium/chloride, pH =&gt;/7.4) of

09159

Hoesht dyë (1.14 mg/ml). Fluorescence was inducéd at an excitation wavelength of 352’ nm and émission meàsured at 454 nm. Calf thymus DNA prepared by identical treatment was used.to develpp standard curves.

Protein content of the tissue extract in.

j concentrated ammonium hÿdroxide was meàsured by the

Bradford method (Bradford (1976) Anal. Biochém. 72:248-54), with bovine sérum albumin.as a standard. Results

The results from histologie évaluation indicated that wounds treated with the combination of purified human PDGF Or recombinànt PDGF and IGF-I had thicker connective tissue and épithélial layers, and more extensive épithélial projections connecting these layers, than wounds receiving no treatment, pure IGF-I, pure PDGF, or PDGF in combination with other growth factors. There was no evidence of scar formation up to 6 weeks following treatment. Autoradiogràphy revealed an merease in the percentage of H-thymidine-labelled cells in PDGF/IGF-I-treated wounds' than wounds receiving no treatment, PDGF alone, or IGF-I alone. Metabolic labelling results showed that the uptaké of °H-thymidine and ^x^C-leucine was greater for wounds treated with purified PDGF/IGF-I préparations than for wounds receiving no treatment; the PDGF/IGF-I-treated wounds similarly had greater DNA and protein contents. The recombinant purified PDGF-2 homodimer, prepared as described above, was alsô tested in partial thickness skin wounds alone and in combination with. IGF-I. Histological analysis of the six day old wounds indicate that the application of PDGF-2 or IGF-I alone resulted in no significant différences from controls in connective tissue or epidermal morphology. However, when PDGF-2 was combined with IGF-I, the combination ' . 4;.-/ .ν' ' · . ~¹² - ' '

- produced a 2.0 fold incféase: in the width of the new .⁷connective tissue l.ayèr àhd a 20% increase. in. êpidërmal thi.ckness at six dayscpo^tôperatively. hiriè days/ /;

PDGF-2 alône resulted in \à 22% inçrëasé in both new .

connective tissue and .epidermal thicknéss. PDGF-2 in .combination with IGF-1 resulted in 75% increases in both the new connective tissue and epidermal layers. ⁷ The &lt; Oônnectiye .tissue:of ^;PDGF-2/IGF-Itreated woünds had definite areas of polarizaticnof liqht ihdicating these.

ÎO wounds çohtainèd more itiature connective tissue than either wounds receiving PDGF-2 alone or no treatment.

Thus, application of recombinant PDGF-2 to ' _; wound healing using: thé animal modël desoribed above, : produces résulta similar to those with purified human.

. PDGF when combined with/recombinant IGF-Γ. Thé combination of recombinant PDGF-2 and IGF-X prgducès . dramatic increases in ths number df new fibroblasts and the rate of collagen synohesis, accompanied by. hyperpïasia of the dermi i and epidermis /(2.5-fold total _2Ô ’ increase) compared to th ; control animal in the absence

... of treatment or by treatiient with recombinant PDGF-^2or . IGF-I alone. /

These resuits indicate that;recombinânt PDGF“2 . homodimer and native purified human PDGF both inter act 2§ synergistically with IGF-t when applied topicaily to wounds. · ' ..

Dosage ·

To détermine the appropriàte dosage of purified/ PDGF, the above^described experiments were repëated / . èxcept that the wounds were treated with 2.5 ng^ 5rÔ ng/ . and 10 ng PDGF équivalents of purified PDGF; per square millimeter of wound dispërsed in 30μ1 of biocômpatible.

<img file="OA9159A_D0004.tif" />

· · · .

09150

<img file="OA9159A_D0005.tif" />

gel. The results shôwed that· optimum'effects were produçed when the PDGF content was 5.0 ng/mm² or higher.

To détermine the appropriate. dosage of pure

PDGF plus IGF-I, combinations in which the weight .to weight ratio of PDGF to IGF-I. rànged from 1:10 to 25:1 were evaluated as described above. Optimum results were achieved with a ratio of between 1:1 and 2:1.

Bone Régénération φ To détermina the effeetiveness of PDGF/IGF-I préparations in promoting periodontium and/or bone growth, the following experiments were performed.

Seagle dôgs with naturally occurring · periodontal disease were selected on the basis of an initial radiographie examination. The teeth which exhibited 30% to 80% bone loss were initially scaled using ultrasonic.instruments. Surgical flaps and root planing techniques were then performed, and the experimental teéth were treated with a composition containing purified PDGF and IGF-I in a pharmaceutically acceptable carrier substance, e.g., commercially available inert gels, e.g., méthyl cellulose. Teeth in the remaining quadrants received control gel alone, or pure PDGF or IGF-I alone. Block biopsies of the teeth and bone were taken at two weeks postsurgery and prepared for histologie évaluation using standard démineraiizing and processing techniques.

Results

Results of histologie.analysis of periodontal and bone spécimens indicated that, adjacent to the roo.t surfaces of experimental specimens (i.e., those treated with the PDGF/IGF-I combination), distinct areas of new bone formation were présent and a deposit res.embling // / - ' ’ ' ./ / τ' . . · · . / cementum was présent on the root surface./adjacent to the new bone. New bone was/also présent/on /the; periosteal surface of the specimens, and areasofankylosis had / occurred within. the apical^extent of the//ligament. A . /

5. dense layer of osteoblast-like celle, and connective tissue lined the /newly formed bone with/ newly formed; collagen fibers inserting into the newly formed cementum.

In the control/specimens there was no évidence of new bone formation, an absence of new cementurn-like deposits, and connective tissue. :was/priente^/ / perpendicular to the bdny surfâce/appearing/to form a ...cap over :the original bone. The results thus indicate that the PDGF/ÏGF-I çompositibn of the .invention . enhances ostéogénie and/ connective, tissue responses. -,/