BACKGROUND OF THE INVENTION

5 The present invention generally relates to the

isolation of precursor cells and their use in bone and

cartilage regeneration procedures and, more

particularly, is concerned with a direct method for

isolating precursor cells from a variety of body tissue

10 types utilizing cell surface antigen CD34 and other

precursor cell surface antigens on CD34+ cells.

Osteogenesis and chondrogenesis are highly complex

biological processes having considerable medical and

clinical relevance. For example, more than 1,400,000

15 bone grafting procedures are performed in the developed

world annually. Most of these procedures are

administered following joint replacement surgeries, or

during trauma surgical reconstructions. The success or

failure of bone grafting procedures depends largely on

20 the vitality of the site of grafting, graft processing,

and in the case of allografts, on immunological

compatibility between donor and host. Compatibility

issues can largely be negated as an important

consideration in the case of autologous grafting

procedures, which involve taking bone tissue from one

site of the patient for transplantation at another

site. While autologous bone grafts are generally

successful they do require additional surgery in order

to harvest the graft material, and not uncommonly are

accompanied by post-operative pain, hemorrhage and

infection.

Cartilage regeneration and replacement procedures

are perhaps even more problematic. Unlike osteogenesis,

chondrogenesis does not typically occur to repair

damaged cartilage tissue. Attempts to repair damaged

cartilage in any clinically meaningful fashion have met

with only limited success. In many cases, the most

effective treatment for cartilage damage is prosthetic

joint replacement.

These and other difficulties w th presently

available bone-grafting and cartilage regeneration

procedures have prompted intensive investigations into

the cellular and molecular bases of osteogenesis and

chondrogenesis. Some promising research to date has

been in the identif cation and isolation of bone and

cartilage precursor cells from marrow and other

tissues.

Early investigations into the complexity of bone

marrow demonstrated that lethally irradiated animals

could be rescued by marrow transplants, suggesting that

bone marrow contained a restorative factor having the

capacity to regenerate the entire hematopoietic system.

More recent experiments have shown that marrow also has

the capacity to regenerate bone and other mesenchymal

tissue types when implanted in vivo in diffusion

chambers. (See e.g. A. Friedenstein etal. "Osteogenesis

in transplants of bone marrow cells." J. Embryol. Exp.

Morph. 16, 381-390, 1960; M. Owen. " The osteogenic

potential of marrow." UCLA Symp. on Mol. and Cell.

Biol. 46, 247-255, 1987) Results of this nature have

led to the conclusion that bone marrow contains one or

more populations of pluripotent cells, known as stem

cells, having the capacity to differentiate into a wide

variety of different cell types of the mesenchymal,

hematopoietic, and stromal lineages.

The process of biological differentiation, which

underlies the diversity of cell types exhibited by bone

marrow, is the general process by which specialized,

committed cell types arise from less specialized,

primitive cell types. Differentiation may conveniently

be thought of as a series of steps along a pathway, in

which each step is occupied by a particular cell type

potentially having unique genetic and phenotypic

characteristics. In the typical course of

differentiation a pluripotent stem cell proceeds

through one or more intermediate stage cellular

divisions, ending ultimately in the appearance of one

or more specialized cell types, such as T lymphocytes

and osteocytes. The uncommitted cell types which

precede the fully differentiated forms, and which may

or may not be true stem cells, are defined as precursor

cells.

Although the precise signals that trigger

differentiation down a particular path are not fully

understood, it is clear that a variety of chemotactic,

cellular, and other environmental signals come into

play. Within the mesenchymal lineage, for example,

mesenchymal stem cells (MSC) cultured in vitro can be

induced to differentiate into bone or cartilage in vivo

and m vitro, depending upon the tissue environment or

the culture medium into which the cells are placed.

(See e.g. S Wakitani et al. "Mesenchymal cell-based

repair of large, full-thickness defects of articular

cartilage" J. Bone and Joint Surg, 76-A, 579-592

(1994); J Goshima, VM Goldberg, and AI Caplan, "The

osteogenic potential of culture-expanded rat marrow

mesenchymal cells assayed in vivo in calcium phosphate

ceramic blocks" Clin. Orthop. 262, 298-311 (1991) ; H

Nakahara etal "In vitro differentiation of bone and

hypertrophic cartilage from periosteal-derived cells"

Exper. Cell Res. 195, 492-503 (1991)) .

Studies of this type have conclusively shown that

MSC are a population of cells having the capacity to

differentiate into a variety of different cell types

including cartilage, bone, tendon, ligament, and other

connective tissue types. Remarkably, all distinct

mesenchymal tissue types apparently derive from a

common progenitor stem cell, viz. MSC. The MSC itself

is intimately linked to a trilogy of distinctly

differentiating cell types, which include

hematopoietic, mesenchymal, and stromal cell lineages.

Hematopoietic stem cells (HSC) have the capacity for

self-regeneration and for generating all blood cell

lineages while stromal stem cells (SSC) have the

capacity for self-renewal and for producing the

hematopoietic microenvironment .

It is a tantalizing though controversial prospect

whether the complex subpopulations of cell types

present in marrow (i.e. hematopoietic, mesenchymal, and

stromal) are themselves progeny from a common ancestor.

The search for ancestral linkages has been challenging

for experimentalists. Identifying relatedness among

precursor and stem cell populations requires the

identification of common cell surface markers, termed

"differentiation antigens, " many of which appear in a

transitory and developmentally-related fashion during

the course of differentiation. One group, for example,

has reported an ancestral connection among MSC, HSC,

and SSC, though later issued a partial retraction (S.

Huang & L. Terstappen. "Formation of haematopoietic

microenvironment and haematopoietic stem cells from

single human bone marrow stem cells" Nature, 360, 745-

749, 1992; L. Terstappen & S. Huang. "Analysis of bone

marrow stem cell" Blood Cells, 20, 45-63, 1994; EK

Waller etal. "The common stem cell hypothesis

reevaluated: human fetal bone marrow contains separate

populations of hematopoietic and stromal progenitors"

Blood, 85, 2422-2435, 1995) . However, studies by

another group have demonstrated that murine osteoblasts

possess differentiation antigens of the Ly-6 family.

That finding is significant in the present context

because the Ly-6 antigens are also expressed by cells

of the murine hematopoietic lineage. (M.C. Horowitz

etal. "Expression and regulation of Ly-6

differentiation antigens by murine osteoblasts"

Endocrinology, 135, 1032-1043, 1994) Thus, there may

indeed be a close lineal relationship between

mesenchymal and hematopoietic cell types which has its

origin in a common progenitor. A final answer on this

question must await further study.

One of the most useful differentiation antigens

for following the course of differentiation in human

hematopoietic systems is the cell surface antigen known

as CD34. CD34 is expressed by about 1% to 5% of normal

human adult marrow cells in a developmentally, stage-

specific manner [CI Civin etal. "Antigenic analysis of

hematopoiesis. Ill . A hematopoietic progenitor cell

surface antigen defined by a monoclonal antibody raised

against KG-la cells. J Immunol, 133, 157-165, 1984] .

CD34+ cells are a mixture of immature blastic cells and

a small percentage of mature, lineage-committed cells

of the myeloid, erythroid and lymphoid series. Perhaps

1% of CD34+ cells are true HSC with the remaining

number being committed to a particular lineage. Results

n humans have demonstrated that CD34+ cells isolated

from peripheral blood or marrow can reconstitute the

entire hematopoietic system for a lifetime. Therefore,

CD34 is a marker for HSC and hematopoietic progenitor

cells .

While CD34 is widely recognized as a marker for

hematopoietic cell types, it has heretofore never been

recognized as a reliable marker for precursor cells

having osteogenic potential m vivo. On the contrary,

the prior art has taught that bone precursor cells are

not hematopoietic origin and that bone precursor

cells do not express the hematopoietic cell surface

antigen CD34 (MW Long, JL Williams, and KG Mann

"Expression of bone-related proteins in the human

hematopoietic microenvironment" J. Clm. Invest. 86,

1387-1395, 1990; MW Long etal. "Regulation of human

bone marrow-derived osteoprogenitor cells by osteogenic

growth factors" J. Clm. Invest. 95, 881-887, 1995; SE

Haynesworth et al. "Cell surface antigens on human

marrow-derived mesenchymal cells are detected by

monoclonal antibodies" Bone, 13, 69-80, 1992) .

To date, the most common sources of precursor

cells having osteogenic potential have been periosteum

and marrow. Many researchers use cells isolated from

periosteum for in vitro assays (See e.g. I Binderman et

al. "Formation of bone tissue in culture from isolated

bone cells" J. Cell Biol. 61, 427-439, 1974) . The

pioneer of the concept of culturing bone marrow to

isolate precursor cells for studying bone and cartilage

formation is A.J. Friedenstein. He developed a culture

method for isolating and expanding cells (CFU-f) from

bone marrow which can form bone (AJ Friedenstein et al.

"The development of fibroblast colonies in monolayer

cultures of guinea pig bone marrow and spleen cells"

Cell Tiss. Kinet. 3, 393-402, 1970) . Others have used

Friedenstein' s culture system extensively to study the

origin of osteoblasts (See e.g. M. Owen, "The origin of

bone cells in the postnatal organism" Arthr. Rheum. 23,

1073-1080, 1980) . Friedenstein showed that CFU-f cells

from marrow will form bone, cartilage, and fibrous

tissue when implanted, though CFU-f cells cultured from

other sources such as thymus, spleen, peripheral blood,

and peritoneal fluid will not form bone or cartilage

without an added inducing agent. Friedenstein recently

discussed the pos ^' sible clinical utility of CFU-f and

pointed out some obstacles that must be overcome, such

as the need for culturing for several passages and

developing a method for transplanting the cells (AJ

Friedenstein "Marrow stromal fibroblasts" Calcif. Tiss.

Int. 56(S) : S17, 1995) .

Similarly, the most common sources of cartilage

precursor cells to date have been periosteum,

perichondrium, and marrow. Cells isolated from marrow

have also been used to produce cartilage in vivo (S

Wakitani et al . "Mesenchymal cell-based repair of

large, full-thickness defects of articular cartilage"

J. Bone and Joint Surg, 76-A, 579-592 (1994) .

Periosteal and perichondral grafts have also been used

as sources of cartilage precursor cells for cartilage

repair (SW O'Driscoll et al. "Durability of regenerated

articular cartilage produced by free autogenous

periosteal grafts in major full-thickness defects in

joint surfaces under the influence of continuous

passive motion" J. Bone and Joint Surg. 70A, 1017-1035,

1986; R Coutts et al. "Rib perichondral autografts in

full-thickness articular defects in rabbits" Clin.

Orthop. Rel. Res. 275, 263-273, 1992) .

In a series of patents, Caplan etal. disclose a

method for isolating and amplifying mesenchymal stem

cells (MSC) from marrow. (U.S. Patents 4,609,551;

5,197,985; and 5,226,914) The Caplan method involves

two basic steps: 1) harvesting marrow and 2) amplifying

the MSC contained in the harvested marrow by a 2 to 3

week period of in vitro culturing. This method takes

advantage of the fact that a particular culture medium

favors the attachment and propagation of MSC over other

cell types. In a variation on this basic method, MSC

are first selected from bone marrow using specific

antibodies against MSC prior to in vitro culturing.

(Caplan and Haynesworth; WO 92/22584) The in vitro

amplified, marrow-isolated MSC may then be introduced

into a recipient at a transplantation repair site. (A.

Caplan. "precursor cells cells" J. Ortho. Res. 9, 641,

1991; S.E.Haynesworth, M.A.Baber, and A.I. Caplan.

"Cell surface antigens on human marrow-derived

mesenchymal cells are detected by monoclonal

antibodies," Bone, 13, 69-80, 1992)

The current methods used to isolate precursor

cells have a number of drawbacks to consider. First,

the methods require that bone marrow or other tissues

be harvested. Harvesting bone marrow requires an

additional surgical procedure with the appendant

possibility of complications from anesthesia,

hemorrhage, infection, and post-operative pain.

Harvesting periosteum or perichondrium is even more

invasive. Second, the Caplan method requires a

substantial period of time (2 to 3 weeks) for in vitro

culturing of marrow-harvested MSC before the cells can

be used in further applications. This additional cell

culturing step renders the method time-consuming,

costly, and subject to more chance for human error.

Consequently, a need exists for a quicker and

simpler method for identifying and isolating precursor

cells having osteogenic and chondrogenic potential

which can be used for in vivo bone and cartilage

regeneration procedures.

SUMMARY OF THE INVENTION

It is an object of this invention to provide a

method for isolating precursor cells having the

potential to generate bone or cartilage from a variety

of hematopoietic and non-hematopoietic tissues.

It is also an object of this invention to provide

a method for isolating precursor cells having the

potential to generate bone or cartilage from peripheral

blood, marrow, or adipose tissue based on binding by a

reagent to cell surface antigen CD34 or other surface

antigens on CD34+ cells.

It is another object of this invention to provide

a method for isolating precursor cells having the

potential to generate bone or cartilage from adipose

tissue based on sedimentation differences in the cells

comprising the tissue.

It is a further object of the present invention to

provide a method for in vivo bone and cartilage

regeneration involving transplantation with CD34+

precursor cells isolated from peripheral blood, marrow,

or adipose tissue.

It is a still further object of the present

invention to provide a direct, single-step method for

in vivo bone or cartilage regeneration involving the

isolation of CD34+ precursor cells from peripheral

blood, marrow, or adipose tissue and immediate

implantation at a connective tissue site needing repair

without the need ^' for in vitro culturing of precursor

cells .

It is yet another object of the present invention

to provide a method to enhance the implantability of

bone prosthetic devices.

It is still another object of the present

invention to provide an improved bone implantation

prosthetic device in which the device is seeded with

precursor cells having osteogenic potential isolated

from a patient's peripheral blood, bone marrow, or

adipose tissue.

These and other objects are provided by the

present invention.

The ability to isolate autologous precursor cells

having osteogenic and chondrogenic potential has far

reaching clinical implications for bone and cartilage

repair therapies, either alone or in conjunction with

prosthetic devices. The present invention provides a

simple method for isolating precursor cells having the

potential to generate bone or cartilage from a variety

of tissue types including peripheral blood, marrow, and

adipose tissue. The precursor cells are isolated using

reagents that recognize CD34 or other markers on the

surface of CD34+ precursor cells, for example CD33,

CD38, CD74, and THYl. Significantly, the present

invention does not require in vitro culturing of

isolated precursor cells before the cells can be used

in further in vivo procedures. Indeed, precursor cells

isolated by the present mvention may be transplanted

in vivo immediately for bone or cartilage regeneration.

Thus, the 2 to 3 week time delay required by other

methods for in vitro culturing of progenitor cells is

eliminated making the method economical, practical and

useful for the clinical environment.

Accordingly, the present invention relates to a

method for isolating precursor cells havmg the

potential to generate bone or cartilage directly from

hematopoietic and non-hematopoietic tissues, including

peripheral blood. The method includes steps of

collecting tissue samples, contacting the sample with

an antibody or other reagent that recognizes antigen

CD34 or other antigens on CD34+ precursor cells, and

separating the reagent-precursor cell complex from

unbound material, by for example, affinity

chromatography. Precursor cells isolated by the present

method may be used immediately for bone and cartilage

regeneration in vivo.

In one aspect, the present invention is a method

for isolating precursor cells having the potential to

generate bone or cartilage from peripheral blood,

marrow or adipose tissue.

In another aspect, the present mvention is a

method for isolating precursor cells having the

potential to generate bone or cartilage based on

selecting cells from hematopoietic and non-

hematopoietic tissues that carry cell surface marker

CD34.

In yet another aspect, the present invention is a

method for bone or cartilage regeneration which

utilizes CD34+ precursor cells isolated from peripheral

blood, marrow, or adipose tissue.

DESCRIPTION OF THE PREFERRED EMBODIMENT(S)

Terms used throughout this disclosure are defined

as follows:

Adipose Tissue

A complex tissue containing multiple cell types

including adipocytes and microvascular cells. Adipose

tissue is one of ^' the most convenient sources of

precursor cells n the body. As used herein the term

"adipose tissue" is intended to mean fat and other

sources of microvascular tissue in the body such as

placenta or muscle. The term specifically excludes

connective tissues, hematologic tissues, periosteum,

and perichondrium.

Chondrogenic

The capacity to promote cartilage growth. Th s

term is applied to cells which stimulate cartilage

growth, such as chondrocytes, and to cells which

themselves differentiate into chondrocytes. The term

also applies to certain growth factors, such as TGF-β,

which promote cartilage growth.

Connective Tissue

Any of a number of structural tissues in the body

including bone, cartilage, ligament, tendon, meniscus,

and joint capsule.

Differentiat on

A biological process in which primitive,

unspecialized, cells undergo a series of cellular

divisions, giving rise to progeny havmg more

specialized functions. The pathway to terminal

differentiation ends with a highly specialized cell

having unique genetic and phenotypic characteristics.

The conventional wisdom of the past taught that

differentiation proceeded in one direction only - from

less specialized to more specialized. This dogma is now

being challenged by new results which suggest that in

fact the pathway may be bi-directional. Under certain

conditions more specialized cells may in fact produce

progeny which effectively reverse the flow toward

greater specialization.

Hematopoietic Stem Cell

Primitive cell having the capacity to self-renew

and to differentiate into all blood cell types.

Mesenchymal Stem Cell

Primitive cell type having the capacity for self-

regeneration and for differentiating through a series

of separate lineages to produce progeny cells having a

wide variety of different phenotypes, including bone,

cartilage, tendon, ligament, marrow stroma, adipocytes,

dermis, muscle, and connective tissue.

Microvascular Cell

Cells comprising the structure of the microvasculature

such as endothelial, smooth muscle, and pericytes.

Osteogenic

The capacity to promote or to generate the

production of bone. The term may be applied to

osteoblasts which have the capacity to promote bone

growth, or to cells which themselves are able to

differentiate into osteoblasts. The term would also

apply to growth factors having the capacity to promote

bone growth.

Precursor Cell

A cell with the potential to differentiate to

perform a specific function.

Stem Cell

Pluripotent precursor cell having the ability to

self-renew and to generate a variety of differentiated

cell types.

The present invention is premised upon two

surprising discoveries. First, that precursor cells

having the potential to form connective tissue in vivo

can be isolated from a variety of hematopoietic and

non-hematopoietic tissue sources, including peripheral

blood, and adipose tissue. And second, that cell

surface marker CD34, a heretofore unrecognized

identifier for connective tissue precursor cells, may

be used as a marker for precursor cells having the

potential to form bone and cartilage in vivo.

The inventors have discovered two convenient, new

sources for precursor cells (viz. peripheral blood and

adipose tissue) , and a source from marrow which does

not require in vitro culture. Unlike prior methods,

which have used bone marrow or periosteum as the source

for osteogenic and chondrogenic precursor cells, the

present invention enables isolation of these cells from

more conveniently harvested tissues, such as peripheral

blood and adipose tissue. The ability to isolate

osteogenic and chondrogenic precursor cells from

tissues other than marrow and periosteum lends

considerable convenience and simplicity to an otherwise

complicated method.

In one embodiment, the present invention is an

affinity method enabling the isolation of precursor

cells m humans having the potential to generate

connective tissue based on expression of antigen CD34

and other cell surface markers on CD34+ cells. Some

examples of other markers on CD34+ cells would include

CD33, CD38, CD74, and THYl, which list is not intended

to be exclusive. In another embodiment, precursor cells

are isolated from adipose tissue based on differential

sedimentation properties. Significantly, unlike

previous methods, the present invention enables the

immediate use of isolated precursor cells for bone and

cartilage regeneration procedures without the need for

in vitro culturing. As a consequence, the present

method is quicker and easier to implement than

previously described procedures.

I . Isolating Precursor Cells

In one embodiment, the present method for

isolating precursor cells involves collecting a body

tissue sample, contacting the sample with an antibody

or other reagent that recognizes and binds to an

antigen on the surface of the precursor cells, and then

separating the precursor cell-reagent complex from

unbound material by, for example, affinity

chromatography. The method can be applied to peripheral

blood, marrow, or other tissues, including adipose

tissue. For ease and simplicity of isolation, however,

blood is the preferred source material since surgical

procedures are not required,

(a) Peripheral blood as the source of precursor cells

By way of example, about 1 unit of blood is taken

by any suitable means, for example by syringe

withdrawal from the patient's arm. A particularly

attractive method in the clinical environment is

apheresis, which has the added advantage of removing

red cells. Removal of red cells is not essential,

although it does enhance the performance of the method

and is preferred. Red cells may be removed from the

sample by any suitable means, for example, lysis,

centrifugation, or density gradient separation. It is

preferred that the sample also be anticoagulated by,

for example, treatment with citrate, heparin, or EDTA.

The yield of precursor cells is expected to be

about 0.1% to 0.5% of the population of nucleated blood

cells. Yields may vary, depending upon the health and

age of the donor, and on the freshness of the sample.

The yield may be dramatically increased by

administering drugs or growth factors to the patient

before blood collection. Although the method will work

on samples which have been stored under refrigeration,

fresh samples are preferred.

A critical step in the procedure of isolating

precursor cells from peripheral blood involves

contacting the blood sample with a reagent that

recognizes and binds to a cell surface marker on CD34+

cells. Any reagent which recognizes and binds to CD34+

cells is within the scope of the invention. Suitable

reagents include lectins, for example soy bean

agglutinin (SBA) , antibodies and attachment molecules

such as L-selectm.

In the preferred embodiment the sample is

contacted with an antibody against CD34. Either

monoclonal (mAb) or polyclonal antibodies may be used.

Methods for preparing antibodies directed against CD34

and other cell surface antigens on CD34+ cells are well

known to those skilled m the art. Suitable human

antibody preparations directed against CD34 and other

cell surface markers on CD34+ cells may be obtained

commercially from Cell Pro, Inc., Bothell, WA, or

Becton-Dick son, Mountain View, CA.

Suitable cell surface antigens on precursor cells

include CD34 and other antigens on CD34+ cells, for

example THYl, CD33, CD38, and CD74. The preferred cell

surface marker is CD34. It is expected that the

procedure will be successful using other cell surface

antigens on CD34+ cells as markers for precursor cells.

Following a ^' brief incubation of the sample with

the antibody to enable binding, the precursor cell-

antibody complex is recovered by any suitable method

such as, for example, affinity chromatography, magnetic

beads, and panning. In the preferred embodiment,

recovery is by affinity chromatography. (see, e.g. RJ

Berenson etal . "Positive selection of viable cell

populations using avidin-biotin immunoadsorption" J.

Immunolog. Meth. 91, 11-19, 1986)

Briefly, the affinity recovery method utilizes a

biotin-avidin coupling reaction in which the antibody

is coupled to biotin by any suitable method. The

antibody-biotin labeled precursor cell complex is

separated from unbound materials by passing the

reaction mixture through a column packed with an avidin

labeled matrix. Unbound materials are removed from the

column by washing. A useful commercially available cell

separation kit includes biotin-labeled human anti-CD34

and a column packed with an avidin labeled matrix

("CEPRATE7LC" available from CellPro, Inc. Bothell,

WA) .

Indirect labelling methods are also within the

scope of the invention. For example, the primary

antibody could be directed against a precursor cell

surface marker and a secondary antibody, labelled with

biotin, directed against the primary antibody.

Alternatively, the secondary antibody may be coupled to

a suitable solid support material.

Negative selection schemes are also intended to be

withm the scope of the invention. Using a negative

selection, the antibody, or other reagent, would be

directed against a cell surface marker which is absent

on CD34+ cells.

(b) Bone marrow as the source of precursor cells

The method disclosed above for isolating precursor

cells from blood may be applied m essentially the same

fashion to bone marrow. Bone marrow is collected by any

suitable fashion, for example illiac crest aspiration.

In the preferred embodiment the marrow is treated with

an anticoagulant such as EDTA, heparin, or citrate and

nucleated cells are separated from non-nucleated cells

by any suitable means, for example by hemolysis or by

density gradient centrifugation.

Precursor cells that express the CD34 cell surface

antigen are isolated from marrow using a reagent that

recognizes and binds to CD34 or to some other antigen

on the surface of CD34+ cells. Suitable reagents

include antibodies, lectins, and attachment molecules.

Bound cells are separated from unbound cells by

affinity chromatography, magnetic beads, or by panning.

In the preferred embodiment, an antibody directed

against CD34 is used in the binding reaction and bound

cells are separated from unbound cells by affinity

chromatography, as disclosed more fully in the examples

which follow.

(c) Adipose tissue as the source of precursor cells

As defmed at the beginning of this section,

"adipose tissue" is used throughout this disclosure

a generic sense to mean fat and other tissue types

(excluding connective tissues, hematologic tissues,

periosteum, and perichondrium) which contam

microvascular cells. Microvascular tissue, from which

capillaries are made, is an integral part of the blood

transport system and, as such, is ubiquitous throughout

the body. Microvascular tissue is composed of at least

three cell types - endothelial, pericytes, and smooth

muscle. Early investigations suggested that

microvascular tissue might play an important role in

bone metabolism. ^" A key observation was that

microvascular cells and tissue arose de novo and

proliferated at sites of bone repair and new bone

growth. Such observations led to speculation that

endothelial cells, pericytes, or both may be

osteoprecursor cells, or alternatively, that

microvascular cells exert a mitogenic effect on bone

precursor cells. (See e.g. C Brighton et.al. "The

pericyte as a possible osteoblast progenitor cell"

Clin. Orthop. 275, 287-299, 1992) A more recent study

using in vitro cultured cells suggests both progenitor¬

like cell proliferation and mitogenic effects by

microvascular cells. (AR Jones et.al. "Microvessel

endothelial cells and pericytes increase proliferation

and repress osteoblast phenotype markers in rat

calvarial bone cell cultures" J. Ortho. Res. 13, 553-

561, 1995) . Thus, within the microvascular cell

population are precursor cells having osteogenic and

chondrogenic potential.

The method of the present invention, as applied to

adipose tissue, has two embodiments. In the first

embodiment, the tissue is contacted with a reagent that

recognizes CD34 or other surface antigen on CD34+

cells. As with peripheral blood and marrow, suitable

binding reagents for use with adipose tissue include

lectins, antibodies, and attachment molecules. The

affinity binding method, as applied to adipose tissue,

differs from the method as applied to blood and marrow

by requiring a step to produce a single-cell suspension

before incubation with the antigen binding reagent. Any

suitable dissociation enzyme such as, for example,

collagenase may be used. Cells that bind the reagent

can be removed from unbound cells by any suitable

means, for example affinity chromatography, magnetic

beads, or panning.

In the preferred embodiment of the invention as

applied to adipose tissue, a sedimentation method is

utilized to obtain a fraction of cells that is enriched

for precursor cells having osteogenic and chrondrogenic

potential. Following harvest of the tissue and

digestion with an enzyme to form a single-cell

suspension, the cells are separated by gravity

sedimentation on the bench top, or by centrifugation.

By way of example, fat could be secured by

liposuction or any other suitable method. About 10 cc

to 30 cc of fat tissue is digested with enough

dissociation enzyme (e.g. collagenase) to produce a

single-cell suspension. Suitable reaction conditions

for enzyme digestion will vary depending on the enzyme

used, as known to those skilled in the art. Following

enzyme digestion, the adipocytes are separated from

other cell types by centrifugation. Adipocytes float to

the surface while denser cells, which include precursor

cells, collect on the bottom and are separable

thereafter by any suitable means. After washing the

harvested precursor cells they can be mixed with a

suitable carrier and immediately implanted in vivo at a

site needing repair.

II. In Vivo Mesenchymal Tissue Regeneration

The precursor cells recovered by the present

procedure are useful for a variety of clinical

applications. For example, they may be transplanted

without further processing to a connective tissue site

in a patient to promote the repair or regeneration of

damaged bone or cartilage.

Unlike previous methods, the present invention

does not require in vitro culturing in order to obtain

a suitable cell type or an adequate quantity of

precursor cells to be of use for in vivo application.

The present invention takes advantage of the unexpected

finding that osteogenic and chondrogenic precursor

cells may be isolated from a variety of hematopoietic

and non-hematopoietic body tissues such as peripheral

blood and adipose tissue. This finding has created a

heretofore unappreciated reservoir of precursor cells

that can be drawn from conveniently to provide enough

cells for in vivo applications without an additional

time-consuming step of amplifying cell numbers by in

vitro culturing. This aspect of the invention saves

time and money with less risk of complication and pain

for the patient.

By way of example only and in no way as a

limitation on the invention, the precursor cells

isolated by the present method from any suitable tissue

source may be implanted at any connective tissue site

needing bone or cartilage regeneration. Suitable

implanting procedures include surgery or arthroscopic

injection.

While the factors that determine biological

differentiation are not fully understood, it is known

that precursor cells will differentiate into bone or

cartilage if transplanted to a site in the body needing

repair. Precursor cells isolated by the present method

can be implanted alone or premixed with growth factors

such as TGF-β. It is preferred that the cells be mixed

with a suitable carrier material, well known to those

skilled in the art, so as to impede the dislodgement of

implanted cells. A non-exclusive list of suitable

carriers would include, for example, proteins such as

collagen or gelatin; carbohydrates such as starch,

polysaccharides, saccharides, methylcellulose, agar, or

algenate; proteoglycans; synthetic polymers; ceramics;

or calcium phosphate.

The data presented in Table 2 demonstrate the

operability of the invention for in vivo applications.

The rat calvarial model used in these studies

demonstrated that CD34+ cells isolated from marrow

using a monoclonal antibody were as effective at

promoting bone growth in an in vivo environment as were

the positive controls (autologous graft) . The data also

show that the antibody itself can affect the outcome of

the results probably via interaction with the

complement system. For example, cells bound by mAb 5E6

did not stimulate bone growth in the rat calvarial

model. Although both antibodies tested recognize CD34

and are IgM isotypes, 5E6 binds complement effectively

while 2C6 does not.

III . Prosthetic Devices

A variety of clinically useful prosthetic devices

have been developed for use in bone and cartilage

grafting procedures, (see e.g. Bone Grafts and Bone

Substitutions. Ed. M.B.Habal & A.H. Reddi, W.B.

Saunders Co., 1992) For example, effective knee and hip

replacement devices have been and continue to be widely

used m the clinical environment. Many of these devices

are fabricated using a variety of inorganic materials

having low immunogenic activity, which safely function

in the body. Examples of synthetic materials which have

been tried and proven include titanium alloys, calcium

phosphate, ceramic hyroxyapatite, and a variety of

stainless steel and cobalt-chrome alloys. These

materials provide structural support and can form a

scaffolding into which host vascularization and cell

migration can occur.

Although surface-textured prosthetic devices are

effectively anchored into a host as bare inorganic

33 structures, their attachment may be improved by seeding

with osteogenic precursor cells, or growth factors

which attract and activate bone forming cells. Such

"biological-seeding" is thought to enhance the

effectiveness and speed with which attachment occurs by

providing a fertile environment into which host

vascularization and cell migration can occur.

The present invention provides a source of

precursor cells which may be used to "seed" such

prosthetic devices. In the prefered embodiment

precursor cells are first mixed with a carrier material

before application to a device. Suitable carriers well

known to those skilled in the art include, but are not

limited to, gelatin, collagen, starch, polysaccharides,

saccharides, proteoglycans, synthetic polymers, calcium

phosphate, or ceramics. The carrier insures that the

cells are retained on the porous surface of the implant

device for a useful time period.

A more complete understanding of the present

invention can be obtained by referring to the following

illustrative examples of the practice of the invention,

which examples are not intended, however, to be unduly

limitative of the invention.

EXAMPLE 1

Animal model for bone regenera ting capaci ty of

precursor cells

A rat calvarial model was used to test the

operability of the invention for in vivo applications.

The model consisted of monitoring the ability of

various test samples to promote bone growth in

calvarial defects which had been surgically introduced

into the rat skull. Calvarial defects were introduced

into 6 month to 9 month old Fisher rats having

bodyweights in the range of about 300 g to 500 g

according to the following procedure. Animals were

anesthesized by intramuscular injection using a

Ketamine- Rompun (xylazine)- Acepromazine (acepromazine

maleate) cocktail, and surgical incisions made in the

calvarial portion of the skull. After peeling back the

skin flap, a circular portion of the skull measuring 8

mm in diameter was removed using a drill with a

circular trephine and saline irrigation. An 8 mm

diameter disk of "GELFILM" was placed in each defect to

separate the exposed brain from the test material and

to maintain hemostasis. The calvarial defects produced

in this fashion were then packed with a test sample

consisting of an isolated cell population. For some

experiments the test samples were mixed with a carrier

material consisting of rat tail collagen or "AVITENE"

bovine collagen before introduction into the calvarial

defect. The positive control consisted of an autograft

while the negative control consisted of a tricalcium

phosphate (TCP) carrier only implant. After surgical

closure of the wound site, treated animals were

returned to their cages, maintained on a normal food

and water regime, and sacrificed 28 days after surgery.

The effectiveness of a test sample to induce bone

growth in calvarial defects was assessed by estimating

new bone formation at the site of the defect by

measuring the closure in the linear distance between

cut bone edges or noting islands of bone growth in the

central portion of the defect. The scoring criteria are

shown in Table 1. The results are summarized in Table

2.

EXAMPLE 2

Isola tion of an enriched nuclea ted cell popula tion from

ra t bone marrow.

Rat bone marrow was isolated from the

intramedullary cavities of 6 femurs taken from male

Fisher rats between 8 to 10 weeks of age. Prior to

sacrifice the animals had been maintained on a normal

food and water diet. The marrow was extracted from

excised femurs by flushing into a test tube containing

approximately 5 ml of ACD buffer. Buffer ACD in the

neat state consists of 2.2g Na3Citrate.2H20, 0.8g

citric acid, and 2.4g dextrose dissolved in 100 ml

distilled water. Unless otherwise noted, buffer ACD was

diluted to a concentration of 15% in PBS. The extracted

marrow cells were gently suspended into the buffer

solution by pipetting. In order to separate red cells

from white cells, the marrow cell suspension was

underlaid with approximately 4 ml of Ficoll-Hypaque

with a specific gravity of 1.09 (Sigma Chemical Co.,

St.Louis, MO) and centrifuged at 1200 x g for 20

minutes. After centrifugation the interface layer

containing the nucleated cells was removed by

pipetting. The cells were washed in 5 ml of ACD and

centrifuged at 250 x g for 6 to 7 minutes. The pellet

was washed twice more in 1% BSA/PBS (bovine serum

albumin, phosphate buffered saline; supplied with

CEPRATE LC kit) /All PBS was Ca+2 and Mg+2 free to

prevent clotting.

EXAMPLE 3

Isola tion of CD34+ cel ls from ra t bone marrow using a

monocl onal an tibody and affini ty chroma tography and

their use for in vivo bone regenera tion in ra t

ca lvarial model .

Ma terials and Methods .

Mouse IgM monoclonal antibodies 2C6 and 5E6 were

raised against rat CD34 present on the surface of a

subpopulation of rat hematopoietic cells. The CD34

mAb' s used in these experiments were the gift of Dr.

Othmar Forster and were prepared in a manner well-known

to those skilled in the art. Anti-mouse IgM:FITC, used

for fluorescence sorting of cells bound with mAb's 2C6

and 5E6, was obtained from Boehringer Mannheim, Cat. #

100807. Avidin: FITC also used in fluorescence sorting

was obtained from Boehringer Mannheim, Cat. # 100205.

CD34+ cells labeled with mAb 2C6 or 5E6 were separated

from unbound cells using an affinity column method. A

useful, commercially available affinity cell separation

kit, "CEPRATE LC, " may be obtained from CellPro

(CellPro, Inc. Bothell, WA 98021) . Anti-mouse

IgM:biotin was purchased from Southern Biotech,

Birmingham, AL, Cat. # 1022-08.

Cells carrying the CD34 surface antigen were

isolated from rat marrow as follows. The rinsed

nucleated cells, isolated in the manner described in

Example 2, were resuspended in about 0.5 ml of

1%BSA/PBS (from CellPro kit) . Then, a volume of mAb

ranging in concentration from about 1 μg/ml to 40 μg/ml

was added and the mixture incubated for about 1 hour at

room temperature with occasional, gentle agitation.

Following incubation the mixture was brought to 10 ml

with 1%BSA/PBS and the mixture centrifuged at 250 x g

for 6 minutes. The pellet was gently resuspended and

rinsed two additional times in 10 ml 1%BSA/PBS and spun

as before. After another resuspension and

centrifugation, the final cell pellet was resuspended

in 2 ml 1%BSA/PBS for incubation with a biotinylated

anti-mouse IgM.

About 10 μl of Goat anti-mouse IgM:biotin (0.5

mg/ml before dilution) was added to the resupended mAb-

cell pellet obtained at the previous step. The mixture

was incubated at room temperature for about 30 minutes

with gentle agitation, after which the cells were

rinsed twice by centrifugation and resuspension in

BSA/PBS, as previously described. The final cell pellet

was resuspended to about 100 x 106 cells/ml in 5% BSA

in a volume of 1 ml to 4 ml for loading onto an avidin

column.

Antibody-labeled and unlabeled cells were

separated on the "CEPRATE LC" avidin column using the

conditions recommended by the manufacturer (Cell Pro,

Inc., Bothell, WA) . Briefly, the column contained a bed

of PBS- equilibrated avidin matrix. Prior to loading

the sample, about 5 ml of 5% BSA was run through the

column. The pre-diluted cell sample was then layered

onto the top of the gel matrix and the sample

thereafter allowed to run into the matrix gel.

Unlabeled cells were washed from the column with about

3 ml to 5 ml of PBS. The mAb-labeled cells were then

released from the matrix and collected into a small

volume of 5% BSA by gently squeezing the column so as

to agitate the matrix while washing the column with

PBS. Small aliquots were saved from the bound and

unbound fractions for cell counting and flow cytometry.

For implantation experiments the cells were washed 2

times in PBS/BSA and once in PBS only.

.Resul ts .

Each experiment generated about 10 to 20 x 106

adherent cells of which about half this number were

implanted into a calvarial defect. Cell fractions taken

from the column were tested for viability by trypan

blue cell counts using a hemacytometer and found to be

in the range of about 85% to 97% viable. The adherent

cell population appeared to be a group of small blast

cells. FACS was used to determine the purity of CD34+

cells isolated on the column. The adherent cell

population contained about 50% of the original number

of CD34+ cells at a purity of about 50%.

CD34+ cells were implanted into rat calvarial

defects with or without a suitable carrier material.

Two carriers were tried in these experiments, "AVITENE"

and rat tail collagen, both of which were found to be

useful. Rat tail collagen is preferred, however, since

it showed the least inflammatory response. About 50 mg

of collagen was dissolved in 1 ml of PBS at 60°C and

equilibrated to 37°C prior to mixing with cells. In

some experiments pellets containing collagen and cells

were formed by mixing 100 μl of collagen solution with

a cell pellet and cooling the mixture to 4°C prior to

implantation mto a calvarial defect. Surgical

implantations were performed as described in Example 1

with sacrifice of recipient animals at 28 days post-

surgery.

Histology scoring for bone formation was assessed

according to the scheme shown in Table 1.

Discussion .

The finding that CD34+ cells isolated by mAb 5E6

failed to stimulate bone regeneration vivo may be

explained by the ancillary observation that this

antibody is a more effective activator of the

complement system than mAb 2C6 (data not shown) .

EXAMPLE 4

(a) Bone regenera tion m ra t calvaria l model using

Fi coll -separa ted whole blood.

The rat calvarial model described in Example 1 was

used to determine the bone regenerating capacity of

Ficoll-separated whole blood. Approximately 2.5 ml of

donor blood was used for each recipient calvarial

defect. Donor animals were 8 to 10 week old male F344

strain rats. Recipients were 6 to 8 months old. Donors

were bled into 3 cc syringes, which contained about 0.5

cc of ACD solution to inhibit coagulation.

ACD Stock Solution ACD Working Solution

2.2 g Na3Citrate.2H2015 ml ACD Stock Solution

0.8 g citric acid.lH2θ 100 ml PBS (Ca++/Mg++ free)

2.4 g dextrose

100 ml distilled water

Blood was placed into 15 ml conical tubes and

brought up to 5 ml with ACD working solution. The

samples were underlaid with 4 ml of Ficoll-Hypaque and

centrifuged at 1200 xg at room temperature for 20

minutes. After centrifugation, the white cell layer was

removed from each tube by pipet.

Ficoll-separated blood cells were used for

implantation experiments, either directly or after

mixing with a carrier material. For direct

implantation, the cell pellet was washed twice in 10 ml

of PBS and the final pellet, containing roughly 5 to 10

x 106 cells, delivered neat into a calvarial defect.

Cell samples pre-mixed with a carrier material were

combined with rat tail collagen prior to implantation.

About 50 mg of rat tail collagen (obtained from Sigma,

St. Louis, MO; Cat.# C-8897) was heated to 60°C in 500

μl PBS to dissolve the collagen protein. The collagen

solution was equilibrated to 37°C prior to mixing with

the cell pellet. About 60μl of collagen solution was

mixed with the cell pellet and the entire cell-collagen

mixture implanted into a calvarial defect.

EXAMPLE 5

Isola tion of CD34+ cells from ra t blood using a

monoclonal an tibody and affini ty chroma tography.

(1) Hemolysis Buffer - IPX Stock Solution

Dissolve the following in 1 L distilled water, adjust

pH to 7.3, filter sterilize and store at 2 - 8°C.

83 g NH4C1

10 g NaHC03

4 g Na2EDTA

(2) Phosphate Buffered Saline (PBS) Ca2+ and Mq2+

Free

Dissolve in 1 L distilled water, adjust pH to 7.2,

filter sterilize, and store at 2 - 8°C.

8 g NaCl

1.15 g Na2HP04

0.2 g KH3P04

0 . 2 g KCl

(3) PBS + Bovine Serum Albumin

Dissolve lg BSA in 100 ml PBS.

(a) Approximately 100 ml of whole blood was

collected by cardiac puncture from 17 male F344 rats 8

to 10 weeks old and heparinized by standard procedures.

Red cells were lysed by mixing the whole blood with 300

ml of IX hemolysis buffer at 37°C and allowing the

mixture to sit for about 3 minutes. Then 100 ml of

PBS/BSA washing solution was added and the mixture

centrifuged at 170 xg for 10 minutes. The resulting

supernatant was aspirated without disturbing the cell

pellet. The pellet was washed two more times by gently

resuspending in PBS/BSA followed by centrifugation. The

final pellet was brought up to 2 ml in PBS/BSA in

preparation for incubation with the mAb, and a small

aliquot removed for cell counting and FACS analysis,

(b) The cell pellet, resuspended in 2 ml PBS/BSA

as in step (a) , was incubated with 3 ml of neat mAb 2C6

in order to bind CD34+ cells. The mAb-cell mixture was

incubated at 4°C for 45 minutes and the cells gently

agitated once to resuspend during incubation. Following

the incubation period the volume was brought up to 10

ml with PBS/BSA and the sample washed twice as in step

(a) . The washed pellet was resuspended in 2 ml PBS/BSA

and 15 ml of goat anti-mouse IgMrbiotin was added for a

30 minute incubation at 4°C with one gentle agitation

during incubation to resuspend cells. The cells were

rinsed twice in PBS/BSA, as described in step (a) , and

the final pellet resuspended in 10 ml of 5% BSA. 5 ml

of the resuspended pellet were used for each of two

"CEPRATE LC" column sorts, as described in Example 3.

Antibody-bound cells were released from the column as

described in Example 3 and the released cells washed

twice in PBS/BSA, and once in PBS. The final cell

pellet was mixed on a glass slide with 60ml of rat tail

collagen (100 mg/ml) at 37°C, and the mixture of

collagen and cells placed briefly on ice to form a

solid pellet. The cell-containing pellet was then

transplanted immediately into a rat calvarial defect,

as described in Example 1.

EXAMPLE 6

Isola tion of microvascular cells from ra t epididymal

fa t pads .

Two epididymal fat pads were removed by dissection

from a male Fisher F344 rat, minced with scissors under

sterile conditions, and incubated in 10 ml PBS/1%BSA

the presence of 8 mg/ml collagenase (Type II Crude,

273U/mg; Worthington Laboratories) for 45 minutes at

37°C with gentle shaking. After digestion the sample

was centrifuged at 250 xg for 4 minutes and the low

density fat at the top of the tube removed by

aspiration. The pellet, which contained the precursor

cells, was washed twice in PBS/1%BSA and once in PBS.

The washed pellet was mixed with 50 ml rat tail

collagen at 37°C, placed briefly on ice to gel, and

implanted into a rat calvarial defect.

It is thought that the method for isolating and

using bone and cartilage precursor cells by the present

invention and many of its attendant advantages will be

understood from the foregoing description and it will

be apparent that various changes may be made in the

form, construction, and arrangement of the elements

thereof without departing from the spirit and scope of

the invention or sacrificing all of ts material

advantages, the form hereinbefore described being

merely a preferred or exemplary embodiment thereof.

Table 1

Bone Formation Scoring

Site Score Description

Defect 0 No net gain in bone; either less

formation than resorption or no

formation at all.

1 Less that 5% of linear distance

between cut bone edges is bridged by

new bone.

2 About 5% to 33% of the defect is

bridged by new bone, or there is an

island of bone in the central portion

of the defect.

3 About 33% to 66% of the defect is

bridged by new bone.

4 Greater than 66% of the defect is

bridged by new bone.

5 Complete bridging of the defect by new

bone.

Table 2.

RBRA Tissue/Cell Type N (Mean ± S.D.)

Autologous Graft (positive control) 1 14422 2.4± 0.7

TCP (negative control) 105 l.O± 0.9

Marrow 30 2.5± 1.1

Marrow Ficoll 18 2.3± 0.8 Marrow/Avitene 9 1.8± 0.4

Blood Ficoll 11 1.3± 0.5

Blood/RTC Ficoll 16 1.4+ 0.5

2C6+ cells 12 1.8± 0.4

2C6- cells 12 0.7± 0.5 5E6+ cells 12 1.3+ 0.6

5E6- cells 12 1.5+ 0.5

SBA+ cells 12 l.β±l.l

SBA- cells 18 1.4+0.7

RBRA: Relative bone regeneration activity

N: Number of experiments

S.D.: Standard deviation

2C6 and 5E6 cells were isolated from marrow

SBA: Soy Bean Agglutinin