Note: Descriptions are shown in the official language in which they were submitted.
--4--
INTRODUCI'ION
.
This invention relates to the production of
5 antibodies specific for connective tissue proteins and,
more particularly, ~o the production of monoclonal
antibodies by fused cell hybrids against human collagens
and enzymes involved in collagen degradation. Collagen is
by far the most prevalent human protein, constituting
almost half of the total body protein. The prolific
research in recent years in the area of collagen
~iochemistry has demonstrated that there are at least six
genetically distinct collagens and several related
collagen-degrading enzymes.
The collagen profile, i.e., the types of distinct
collagens and collagen-associated proteins present, their
distribution in the tissue, and the concentration ratios
among the distinct types, of any given tissue or body fluiG
sample varies with the tissue or fluid source. Moreover,
20 the collagen profile of a tissue or fluid sample also
varies with the physiological or pathological state of its
source. In fact, there are numerous connective tissue
disorders ana other pathological conditions in which
changes in the collagen profile occur, eventually resulting
in such large scale tissue alterations as to cause organ
impairment. Hence, a specific and reliable means for
detecting and/or quantitatively measuring changes in
collagen types and distribution in tissue and bo~y fluids
is extremely useful for diagnostic evaluations of the stage
of and specific organ involvement in certain diseases.
Furthermore, the detection and/or quantitative measurement
of different types of collagens and collagen associa.ed
enzymes in body fluids provides a means for monitoring
therapies that result in a release of collagens and
~ ,i
~2~;~'72
--5--
collagen-associated enzymes into body fluids upon the
eradication of cells, such as tumor cells, against which
the drug is targeted.
The use of monoclonal antibodies against
connective tissue proteins to establish the collagen
profile of histological, cytological and biological fluia
sample5 i5 a novel and advantageous approach to disease
10 diagnosis and therapy monitoring. Because of the high
specificity and sensitivity of monoclonal antibodies, early
detection of certain collagen-related pathological
conditions is possible as is early assessment of the
efficacy of certain therapeutic programs. To achieve these
goals, the invention provides: (1) a method for repeatedly
producing large quantities of monospecific antibodies
against distinct connective tissue proteins and (2)
procedures for using the monoclonal antibodies individually
or in combination as clinical probes for diagnosis and
20 therapy monitoring. The potential prognostic importance of
early and accurate disease diagnosis and determination of
the usefulness of certain therapies using the methoas of
this invention is highly significant.
2. BACKGROUND OF THE INVENTION
2.1. MONOCLONAL ANTIBODIES
Kohler and Milstein are generally credited with
30 having devised the technique that successfully resulted in
the formation of the first monoclonal antibody-producing
hybridomas [G. Kohler and C. Milstein, Nature 256:495-497
(1975); Eur. J. Immunol. 6:511-519 ~1976)]. The monoclonal
antibodies produced by hybridomas are highly specific
35 immunoglobulins of a single type. The single type of
immunoglobulin secreted by a hybridoma is specific to one
~ 5~
and only one antigenic determinant on an antigen, a complex
molecule having a multiplicity of antigenic determinants.
Hence, monoclonal antibodies raised against a single
5 antigen may be distinct from each other depending on the
determinant that induced their formation; but for any given
clone, all of the antihodies it produces are identical.
~ onoclonal methods are generally applicable and
have been used to produce antiboaies to antigens other than
the sheep red blood cells used by Kohler and Milstein. For
instance, it has been reported that monoclonal antibodies
haYe been raised against tumor cells lU.S~ Pat. No.
4,172,124~ and viruses lU~S- Pat. No. 4,196,265]. The
production of monoclonal antibodies against certain
collagens, procollagens (natural precursors of collagens)
and a collagen-associated gIycoprotein has also been
reported. Linsenmayer et al. reported using the cell
hybridi~ation technique to produce monoclonal antiboaies
against chick Type I collagen [Proc. Natl. Acad. Sci.
U.S.A. 76(8):3703-3707 (1979)~; Linsenmayer and Hendrix
later reported having produced a monoclonal antibody
specific for chick Type II collagen [Biochem. Biophys. Res.
Commun. 92:440-446 (1980)]. Both antibodies have been used
for biochemical and cytological studies of extracellular
matrices involved in the morphogenesis of the embryonic
chick. Walsh et al. lDev. Biol. 84:121-132 (1981~] have
reported producing a monoclonal antibody against human
fibronectin, a collagen~associated glycoprotein, as part of
3 an investigation to define human muscle surface antigens.
The biochemical and immunological characterization of
monorlonal antibodies specific for human collagens, Types
I, III and IV, and human procollagens Types I and III has
recently been reported ~N. SundarRaj et al., J. Cell Biol.
35 (Abstr~3 91(2):8028 (1981)]o Finally, a monoclonal
antibody against the collagen degrading enzyme elastase has
.~
~, ~
~25~2
--7--
been used to study the pathogenesis of inflammatory joint
disease [S. Gay et al., VIIIth Southeastern Meeting, Amer.
Rheum. Assoc., Abstr. 1, (1981)].
2.2. CONNECTIVE TISSUE PROTEINS
Information on the biochemistry of the
genetically-distinct collagen types and their role in
biological processes has grown prolifically in recent years
[P. Bornstein and H. Sage, AnnO Rev. Biochem. 49:957-1003
(1980); S. Gay and E. Miller, Collagen in the physiology
and pathology of connective tissue, Gustav Fischer Verlag,
New York (1978)]. Currently, the known collagens can be
subdivided into four categories based on their histological
distribution [S. Gay et al., Arthritis and Rheumatism
23(8):937-941 (1980)]. Each type of collagen has as its
biosynthetic precursor a procollagen molecule which differs
from the mature collagen molecule insofar as it has
additional amino acid sequences at the amino and carboxy
2 termini of each chain that are eventually cleaved by
specific processing enxymes.
The interstitial collagen molecules comprise the
majority of all the connective tissue proteins and account
25 for nearly all fibrillar tissue components. This class of
collagens represents four distinct molecular species: (1)
the Type I collagen molecule which exhibits the chain
composition [~l(I)]2 2(I). Fibers derived from Type I
collagen are found throughout the entire organism primarily
in supporting tissues which normally exhibit very little
distensibility under physical stress; (2) the Type I-trimer
collagen molecule which is comprised of three identical
~l(I) chains. This molecule has been described in certain
chondrocyte cultures and other experimental systems, but
its existence in normal tissue has not been firmly
~_~s~29
:
--8--
established; (3) the Type II collagen molecule which
- contains three ~l(II) chains. In most instances this
species forms relatively thin fibrils and displays a tissue
5 distribution restricted predominantly to cartilaginous
structures such as articular cartilage and nucleus pulposus
and to certain parts of the embryonic eye; and (4) the Type
III collagen molecule which is composed of three ~l(III)
chains. The fibrils formed by these molecules are usually
10 found in a reticular net~ork. The latter meshwork
apparently contains in addition to Type III molecules
certain quantities of a form of Type III procollagen
indicating that the conversion of Type III procollagen to
Type III collagen is incomplete. These procollagen
molecules participate in the formation of fine non-striated
filaments which are associated with the Type III fibrils.
The basement membrane collagens include at least
two distinct collagen chains, the ~l(IV) and ~2(IV), which
exhibit unique compositional features. The configuration
of these chains within native collagen molecules is
presently unknown. These collagens appear to be
universally distributed as components of the
morphologically distinct epithelial and endothelial
basement membranes.
The pericellular collagens commonly referred to
as Type V collagen contain three distinct chains, ~l(V),
~2(V) and ~3(V), which combine to ~orm a variety of
molecular species. Histologically, they are more
predominant in cells derived from the vascular system as
compared to other tissues and appear to form a
pericellular exocytoskeleton.
In certain tissues there are high molecular
weight aggregates which upon disulfide bond reduction and
~5~2~
denaturation are found to contain unique collagenous
subunits (Type VI collagen). These aggregates may serve
as structural polypeptides linking collagenous sequences
5 with noncollagenous sequences LD. Furuto and E. Miller, J.
Biol. Chem. 255(1):290-295 (1980); D. Furuto and E.
Miller, Biochem. 20:1635-1640 (1981)].
Procollagens and cross-linked collagen molecules
are susceptible to attack by specific collagen-degrading
enzymes collectively called collagenases; cleavage by such
enzymes yields procollagen peptides and collagen
peptides. For instance, elastase is a very distinct
collagen-degrading enzyme which selectively cleaves Type
5 III collagen, but not Type I collagen, and releases a
distinct trimer peptide, ~l(III) . [C. Mainardi et al.
J. Biol. Chem. 255(24):12006-12010 (1980)].
2.3 PATHOLOGICAL CONDITIONS INvOLVING
CONNECTIVE TISSUE PROTEINS
Pathological conditions involving connective
tissue proteins are numerous and can be grouped roughly
into three categories: conditions resulting from overt
trauma, heritable disorders9 and disorders commonly called
acquired diseases. The pathophysiology of connective
tissue that is characteristic of these disorders has been
reviewed by Gay and Miller [S. Gay and E. Miller, Collagen
in the physiology and pathology of connective tissue,
Gustav Fischer Verlag, New York (1978)].
Of the three categories of connective tissue
pathology, it is in the acquired connective tissue
disorders that changes in the collagen profile of
afflicted tissues most notably occur as the disease
progresses. The acquired disorders are pathophysiological
~5~
--10--
conditions in which large scale tissue alterations occur
as the result of an apparent lack of coordination between
collagen synthesis and de~radation. The disorders include
5 atherosclerosis, liver cirrhosis, lung fibrosis, bone
marrow fibrosis, systemic progressive sclerosis,
scleroderma, psoriasis, rheumatoid arthritis,
osteoarthrosis and certain benign and malignant tumors.
For the most part, these conditions arise through
fibroproliferative responses leading to an excessive
accumulation of collagen in affected tissues though some
disorders involve degenerative changes within previously
normal connective tissues.
The patterns of collagen deposition in three
different fibroproliferative disorders, atherosclerosis,
liver fibrosis (or cirrhosis) ~nd scl~roderma of skin are
quite similar and are illustrative of the types of changes
in collagen profile that occur in the acquired connective
tissue diseases. In the early stage of such pathological
2 conditions, an increase in basement membrane collagen
synthesis is first observed. The deposition of basement
membrane matrix containing Type IV collagen is followed by
a Type III collagen neosynthesis. The Type III collagen
thereby forms the reticular network of granulation
25 tissue. Finally the dense collagen fiber form~ the scar
tissue which is almost completely comprised of Type L
collagen molecules [S. Gay, Ital. J. Gastroenterol.
12:3~-32 (1980)]~
A number of tumors such as the various kinds of
fibromatoses elaborate matrices containing copious
quantities of fibrous collagen. Malignant tumors such as
the osteosarcomas or chondrosarcomas may also produce
large amounts of collagenous matrix In these disorders,
the collagen produced generally reflects the cellular
:
72g
--ll--
origin of the tumor. Thus, osteosarcoma cells produce a
matrix containing fibers derived from Type I molecules,
whereas the fibrous elements of chondrosarcomas are
5 derived from Type II molecules, [K. Remberger and S. Gay,
Z. Krebsforsch~ _:95-106 (1977)]. However, it is
possible that less differentiated tumors may synthesize a
number of different collagens. For instance, metastasized
neoplastic mammary epithelial cells of breast carcinomas
10 retain the ability to synthesiæe Type IV (basement
membrane) collagen [L.A. Liotta et al., The Lancet, July
21, 1979: 146-147].
Rheumatoid arthritis is an acquired disease
15 manifested by either fibroproliferative or degenerative
changes in the connective tissue of diarthrodial joints.
In the inflammatory-proliferative phase, rheumatoid
synovial tissue is characterized by the synthesis and
deposition of additional Type I and III collagens. The
20 blood vessels of the proliferating pannus tissue carry the
bulk of vascular-derived Type V collagen. The endothelial
basement membrane containing Type IV collagen often
appears irregular, discontinuous, and sometimes
multilamellated in the vessels of pannus tissue. The
25 altered basement membrane barrier is reflected by the
cellular synovial exudate. The exudate contains
phagocytes that exhibit inclusions of various collagens.
The presence of different collagens in phagocytes of the
synovial fluid is apparently due to degradation and
30 erosion of different parts of the joint due to proteolytic
activity on the part of collagenases. Phagocytosis of the
vessel-derived collagens such as Type IV collagen from
endothelium as well as Type V collagen surrounding smooth
muscle cells and pericytes may reflect at least in part
35 the degree of vascular necrosis. The presence of Type I
and III collagen within the exudate cells suggests the
~5~2g
-12-
destruction of the synovial matrix. However, the
demonstration of considerable amounts of Type III collagen
may also reflect collagen neosynthesis as observed in
5 other fibro-proliferative disorders. The existence of
Type II collagen in the synovial phagocytes undoubtedly
indicates the erosion of articular cartilage. From this
discussion it is clear that the collagen profile of
synovial exudate cells and synovial fluid can reflect the
10 nature and extent of initial joint damage and the progress
of the joint disease [S. Gay et al., Arthritis and
Rheumatism 23(8):937-941 (1980~3.
Osteoarthrosis is an example of a noninflammatory
15 joint disorder that involves degenerative loss of the
articular cartilage. During the early stages of
osteoarthrosis, articular cartilage is characterized by a
loss of proteoglycan aggregates, presumably due to the
release of unusually large amounts of degradative enzymes,
20 which results in demasking Type II collagen fibers on
fibrillated surfaces [S. Gay and R.K. Rhodes,
Osteoarthritis Symposium, pp. 43-44, Grune ~ Stratton,
Inc. (1981)]. In general, the fibrillated surface
persists and the initial clefts eventually extend into the
deeper layers of articular cartilage due to the
inefficient healing and repair capacity of cartilage
tissue. Chondrocytes do proliferate and form clusters
adjacent to the cartilage clefts. Although these
chondrocytes apparently retain their capacity to form new
30 proteoglycan aggregates, the capacity to synthesize new
cartilage specific Type II collagen molecules appears to
be greatly diminished or lost. Instead, a small
deposition of fibrocartilaginous material comprised of
collagen fibers derived from Type I molecules occurs. The
35 switch from Type II collagen synthesis to Type I collagen
synthesis appears to be an important step in the
5 ~7
-13-
pathogenesis of osteoarthrosis and hence the presence of
Type I collagen in biopsies can serve as an indicator of
the progress of the disease.
2.4 BIOCHEMICAL APPROACHES TO THE STUDY OF
COLLAGEN AND CONNECTIV~ TISSUE PATHOLOGY
Investigations on collagen in pathological states
10 have frequently taken the form of: solubility
(extractability) determinations in an effort to discern
the state or extent of cross-linking; analyses of tissue
hydroxyproline content as a measure of total collagen
content; and evaluations of the capacity for collagen
synthesis based on specific activity determinations in
both ln vivo and in vitro labeling experiments. Each of
these approaches is inherently limited and therefore has
several disadvantages. Thus the solubility or
extractability of the collagen in a given specimen is
heavlly dependent on the physical state of the specimen,
is often quite low, and most probably reflects the nature
rather than the extent of the collagen cross-links
prevalent within the tissue. Also, hydroxyproline
determinations may provide misleading values for total
25 collagen content due to the presence of other
hydroxyproline-containing proteins such as elastin or Clq,
as well as the presence of varying proportions of the
various collagens. With respect to the latter point, the
Type III collagen molecule contains about 30% more
30 hydroxyproline than the Type I collagen molecule.
Therefore, the total collagen content of a given specimen
cannot be related to hydroxyproline content unless a
reasonably accurate estimate of the proportions of these
collagens in the tissue is available. And finally, the in
vivo as well as ln vitro labeling experiments are often
difficult to interpre~ since rates of collagen ~egradation
5~'7
-14-
and pool sizes are not commonly evaluated. At best, then,
these biochemical approaches provide only limited insight
into the possible alterations in collagen chemistry and
5 biosynthesis in diseased tissues. Moreover; they offer
virtually no information with respect to the prevalence or
disposition of the various collager.s in such tissues, and
hence are of limited or no diagnostic use.
2.5 I~MUNOLOGICAL APPROACHES TO THE STUDY OF
COLLAGEN AND CONNECTIVE TISSUE PATHOLOGY
Even though the genetically distinct types of
collagens are very similar to one another in their
15 macromolecular structure, they are sufficiently different
in their amino acid sequence to allow the production of
specific antibodies. Antibodies can be raised against
antigenic determinants located in five identifiable
regions of collagen or procollagen molecules,
specifically, the globular amino and carboxy termini of
procollagen molecules, the non-helical termini of mature
collagen molecules, the helical portion of collagen and
procollagen molecules, and the central amino acid
sequences of individual ~-chains, obtained by denaturing
collagen molecules. Thus, the antigenic regions in
collagen consist both of sequential and conformational
determinants.
Despite the weak antigenicity of collagen
30 molecules, antibodies (in conventional antisera) have been
successfully raised against distinct collagens,
procollagens and collagen-associated proteins [Timpl et
al., J. Immunol. Methods 18:165-182 (1977), U.S. Pat. No.
4,312,853; J. Risteli et al., Fresenius Z. Anal. Chem.
35 301:122 (1980)] and have proved to be exceedingly useful
reagents in elucidating the precise distribution of the
~'7
-15-
various collagens in tissues ana body fluids as well as in
determining the capacity of certain cells to synthesize
the various collagens. In fact, the information on the
5 changes in collagen profiles which occurs during the
connective tissue disorders discussed in S~ction 2.3 was
obtained primarily through the use of immunohistological
techniques and radioimmunoassays based on antibodies to
the genetically distinct collagens.
While antibodies against collagens have been usea
mostly in connection with biological and biomedical
research, the use of antibodies has also been suggested as
a means for early clinical recognition of certain
5 collagen-related diseases and other pathological
conditions. A raaioimmunoassay for Type III procollagen
and Type III procollagen peptide has been reported by
Timpl for the purpose of measuring these antigens in
blood. Detection may indicate the presence of such
possible disease states as liver cirrhosis or hepatitis
[U.S. Pat. No. 4,312,853]~ which, a~ early stages, are
often accompanied by the release of procollagen and
procollagen peptide Type III into the serum and other body
fluidsO An immunohistochemical method for detecting Typ~
IV (basement membrane) collagen-producing cells was
reportedly used for the localization of single metastatic
cells (which produce Type IV collagen and which could not
be detected otherwise) in sections of lymph-nodes of
breast cancer patients [L.A. Liotta et al., The Lancet,
30 July 21, 1979:146]. Radioimmunoassays for two basement
membrane proteins, 7S collagen (non-Type IV) and the
non-collagenous protein laminin have been reported by J.
Risteli et al. lFresenius Z. Anal. Chem. 301:122 (1980)].
The proposed use was for monitoring basement membrane
disorders (such as the microangiopathic lesions of
diabetes mellitus) by measuring the amount of these
.~.
-16-
proteins circulating in the human blood stream.
Enzyme-linked immunoadsorbent assays have been developed
for types I, II, III, and IV collagen and for laminin ana
5 fibronectin by using antibodies prepared in rabbits and
goats ~S.I. Rennard et al , Anal. Biochem.
04:205-214~1980)~.
Notably, all the antibodies used to detect the
10 presence of collagens and collagen-associated prcteins in
body fluids and tissues and to study collagen distribution
during pathological states have been polyclonal antibodies
produced by conventional means. Since various levels of
cross-reacting antibodies may occur in the antisera, the
15 specificity of such antisera must be increased by
time-consuming immuno-adsorption procedures. [Timpl et
al., J. Immunol. Methods 18:165-182 (1977)].
Adaptation of monoclonal techniques to the
production of highly specific antibodies against the
genetically distinct collagens and other connective tissue
proteins for use in ln vltro diagnostics and chemotherapy
monitoring represents a clear improvemen~ over previous
immunological approaches to the detection of collagen-
2 related pathological conditions. The fused cell hybridsmade with these me~hods produce a single kind of antibody
specific for the collagen antigen of interest. Higher
titers of identical immunoglobulins are available in
essentially limitless supply since the antibody-proaucing
hybridomas can be cultured indefinitely in vitro or
propagated in mice or other laboratory animals.
Conventional methods fo~ pro~ucing antibodies result in
preparations of less specific polyclonal antisera which
have to be purified extensively prior to use and can never
35 be reproduced identically. The monoclonal approach,
however, permits the quantitatively large-scale yet
~ 7~
inexpensive production of highly specific antibodies,
requiring minimal purification, if any, in small-scale
culture vessels or laboratory animals.
3. SUMMARY OF THE INVENTION
Prior to the presen~ invention, applicant
believes there has been no report of a clinically useful
10 preparation of monoclonal antibodies specific for all the
known genetically distinct types of human collagens,
collagen-associated enzymes and collagen peptide fragments
resulting from enzymatic cleavage. Because collagen
profiles of human body tissues and fluids change ouring
15 certain pathological conditions and during therapeutic
regimens and because the changes can be detected by
immunohistological and immunoserological techniques, the
monoclonal antibodies of this invention represent a new in
vitro means of early and accurate disease or cancer
diagnosis and monitoring of drug therapyO
The present invention provides a method for
producing monoclonal antibodies against human collagens
Types I through VI, the collagen degrading enzyme elastase
and the al(III) peptide cleaved from Type III collagen
by elastase. The monoclonal antibodies may be used in
standard radioimmunoassays or enzyme-linked immunosorbent
assays for the quantitative measurement of the spectrum of
connective tissue proteins in a given sample of body
30 fluid, thereby permitting non-invasive diagnosis of
certain pathologocal states and the monitoring of
therapies that result in release of connective tissue
proteins into sera and other biological fluids. The
monoclonal antibodies may be tagged with compounds which
35 fluoresce at various wavelengths so that the aistribution
of collagens in tissue biopsies can be determined by
~S~L'7~29
-18 -
immunohistological techniques. Radioimmunoassays and
immunohistological me~hods employing the monoclonal
antibodies of this invention can be used to detect and
5 follow the pathogenesis of diseases, such as: genetic
disorders affecting skeleton, skin and muscles; formation
of excessive scar tissue; and deposition of pathological
amounts of connective tissue in bod~ organs, including
kidney, intestines and heart, and in liver by liver
10 cirrhosis, in skin by scleroderma; in lung by lung
fibrosis; in bone marrow by leukemia; in blood vessels by
atherosclerosis; and in joints by rheumatic diseases. The
methods involving monoclonal antibodies can also be used
to detect changes in the neosynthesis of collagens that is
15 indicative or suggestive of the malignant state of cells
derived from such tumors as breast carcinomas.
Because the monoclonal antibodies are produced by
hybridoma techniques, the present invention provides
20 theoretically immortal cell lines capable of consistently
producing high titers of single specific antibodies
against the distinct connective tissue proteins. This is
a distinct advantage over the traditional technique of
raising antibodies in immunized animals where the
25 resulting sera contain multiple antibodies of different
specificities that vary in both type and titer with each
animal, ana, in individual animals, with each immunization.
The invention contemplates the extension of the
30 hybridoma technique to the production of monoclonal
antibodies to other genetically distinct collagens and
collagen-associated proteins and enzymes as they become
known and their use in the in vitro diagnosis of disorders
and cancers involving connective tissue proteins.
-
~5~'729
--19--
The invention further contemplates the use of
monoclonal or polyclonal antibodies against connective
tissue proteins for ln vivo diagnostic and therapeutic
5 purposes. Antibodies produced by either conventional
methods or the monoclonal techniques of this invention can
be labelled with radioactive compounds, for instance,
radioactive iodine, and administered to the patient. The
antibodies localize in areas of active collagen
10 neosynthesis such as certain malignant tumors or other
tissues undergoing pathological changes involving
collagen. The localization of the antibodies can then be
detected by emission tomographical and radionuclear
scanning techniques such detection is of diagnostic
value. In addition, monoclonal or polyclonal antibodies
against connective tissue proteins can be conjugated to
certain cytotoxic compounds (radioactive compounds or
other therapeutic agents) and can be used for therapeutic
purposes, for instance, cancer therapy. The antibodies,
targeted for malignant cells expressing the appropriate
~0
collagen antigen, localize on or in the vicinity of the
individual cells or tumor at which point the conjugated
cytotoxic compound takes effect to eradicate the malignant
cells.
4. DESCRIPTION OF THE INVENTION
4.1. THE ANTIGENS
The genetically distinct types of collagens and
other connective tissue proteins can be derived from a
variety of tissue sources throughout the human body.
Purification of the collagens has been described in the
literature [E. Miller and R. Rhodes, Structural and
35 contractile proteins, in: L. Cunningham and D. Frederiksen
(editors), Methods in Enzymology, Academic Press, New York
~1981)].
'7
-20-
Depending on the antibody desired, any one of
these distinc~ connective tissue proteins is a suitable
antigen with which to immunize animals, such as mice or
5 rabbits, to obtain antibody-producing somatic cells for
fusion. The choice of animal can influence the type of
antibody obtained vis a vis the determinant on the antigen
against which the antibody is directed. For example, if
antibodies directed toward amino or carboxy terminal
10 determinants are desired, rabbits should be immunized.
When rats or mice are immunized, antibodies proaucea
against determinants in the more stable helical portion of
the various collagen molecules are usually the result.
4.2. SOMATIC CELLS
Somatic cells with the potential for producing
antibody and, in particular, B cells, are suitable for
fusion with a B-cell myeloma line. Those antibody-
20 producing cells that are in the dividing plasmablast stagefuse preferentially. Somatic cells may be derived from
the lymph no~es and spleens of primed animals.
Once-primed or hyperimmunized animals can be used as a
source of antibody-producing lymphocytes. Mouse
25 lymphocytes give a higher percentage of stable fusions
with the mouse myeloma lines described in Section 4.3.
However, the use of rabbit, human and frog cells is also
possible.
4.3. MYELOMA CELLS
Specialized myeloma cell lines have been
developed from lymphocyte tumors for use in
hybridoma-producing fusion procedures ~G. Kohler and C.
35 Milstein, Europ. J. Immunol. 6:511-519 (1976); M. Shulman
et al., Nature 276:269-270 (1978)].
L'729
Several myeloma cell lines may be used for the
production of fused cell hybrids, including X63-Ag8,
NSI-Ag4/1, MPC11-45.6TGl.7, X63~Ag~.653, Sp2/0-Agl4, FO,
5 and S194/5XXO.BU.l, all derived from mice, including
MOPC-21 mice, 210.RCY3.Agl.2.3 deri~ed from rats ana
U-226AR, and GM1500GTGAL2, derived from rats and
U-226~R, and GM1500GTGAL2, derived from humans. [G.J.
Hammerling, U. Hammerling and J.F. Xearney (editors),
Monoclonal antiboaies and T-cell hybridomas in: J.L. Turk
(editor) Research Monographs in Immunology, Vol. 3,
Elsevier/North Holland Biomedical Press, New York (1981)].
4.4. FUSION
Methods for generating hybrids of antibody-
producing spleen or lymph node cells and myeloma cells
usually comprise mixing somatic cells with myeloma cells
in a 2:1 proportion (though the proportion may vary from
about 20:1 to about 1:1), respectively, in the presence of
an agent or agents that promote the fusion of cell
membranes. It is preferred that the same species of
animal serve as the source of the somatic and myeloma
cells used in the fusion procedure. Fusion methods have
been described by Kohler and Milstein [Nature 256:495-497
(1975) and Eur. J. Immunol. 6:511-519 (1976)], and by
Gefter et al. [Somatic Cell Genet. 3:231-236 (1977)].
The fusion-promoting agent used by those investigators
were Sendai virus and polyethylene glycol (PEG),
respectively. The fusion procedure of the example of the
present invention is a modification of the method of
Gefter et al. [supra]; PEG is added to the mixture of
mouse spleen and myeloma cells to promote the formation of
fused cell hybrids. Dimethyl sulfoxide ~DMSO), another
agent affecting cell membranest may also be included, in
addition to PEG, in the fusion mixture.
~2~
-22-
4-5O ISOLATION OF CLONES AND ANTIBODY DETECTION
Fusion procedures usually produce viable hybrids
5 at very low freguency. The frequency of heterokaryon
formation using state-of-the art technigues with PEG as
fusing agent is generally lx10 2 Ensuing nuclear
fusion a~d formation of synkaryons has a frequency of
lx10 3. Thus only one in 105 fused cells under
10 optimal conditions will yield a viable hybrid cell line.
This frequency, when multiplied by the average frequency
of the specific plaque-forming cells in spleen ~lx10 3)
yields an overall success exp~ctation of about
lx10 8 Therefore, one immune mouse spleen, containing
2x108 cells, should yield at least one specific
hy~ridoma clone [~. J. Hammerling et al., supral.
.,
Because of the low frequency of obtaining viable
hybrids, it is essential to have a means of selecting ~he
fused cell hybrids from the remaining unfused cells,
particularly the unfused myeloma cells. Generally, the
selection of fused cell hybrids is accomplished by
culturing the cells in media that support the growth of
hybridomas but prevent the growth of the myeloma cells
which normally would go on dividing indefinitely. In the
example of ~he present invention, myeloma cells lacking
hypoxanthine phosphoribosyl transferase (~PRT ) are
used. These cells are selected against in
hypoxanthine/aminopterin~ thymidine (HAT) medium, a medium
in whic~ the fused cell hybrids survive due to the
HPRT-positive genotype of the spleen cells. The use of
myeloma cells with different genetic aeficiencies (e.gO,
other enzyme deficienciest drug sensitivities, etc.) that
can be selected against in media supporting the growth of
genotypically competent hybrids is also possible.
,~
~,~
~2 ~
-23-
Generally, around 3% of the hybrids obtained
produce the desired antibody, although a range of from 1
to 30% is not uncommon. The detection of
5 antibody-producing hybrids can be achieved by any one of
several standard assay methods, including enzyme-linkea
immunoassay and radioimmunoassay techniques which have
been described in the literature [R. Kennet, T. McKearn
and K. Bechtol (editors), Monoclonal antibodies,
hybridomas: a new dimension in biological analyses, pp.
376-384, Plenum Press, New York (1980)]. The detection
method used in the example of the present invention was an
enzyme-linked immunoassay employing an alkaline
phosphatase-conjugated anti-mouse immunoglobulin.
4. 6 . CELL PROPAGATION AND ANTIBODY PRODUCTION
Once the desired fused cell hybrids have been
selected and cloned into individual antibody-producing
cell lines, each cell line may be propagated in either of
two standard ways. A sample of the hybridoma can be
injected into a histocompatible animal. The injected
animal develops tumors secreting the specific monoclonal
antibody produced by the fused cell hybrid. The boay
fluids of the animal, such as serum or ascites fluid, can
be tapped to provide monoclonal antibodies in high
concentration. Alternatively, the individual cell lines
may be propagated 1n vitro in laboratory culture vessels.
4 . 7 . IN VITRO DIAGNOSTIC USES FOR ~lONOCLONAL
ANTIBODIES TO CONNECTIVE TISSUE PROTEINS
In Section 2.3., supra, pathological conditions
involving connective tissue proteins were enumerated and
the changes in collagen profiles of affected tissues and
body fluids that occur as the diseases progress were
,
,
2 ~'7
-24-
discussed. To illustrate how monoclonal antibodies
against specific collagens can be used to diagnose
pathological conditions in humans, the following three
5 examples involving l) non-invasive serological diagnosis
of disease, 2) histological diagnosis of disease and 3)
cancer detectisn are offered.
Rheumatoid arthritis and osteoarthrosis synovial
10 fluid may be withdrawn from the knee which can be
subjected to radioimmunoassays (and immunofluorescent
assays for any cell which may be present in the fluid) in
which monoclonal antibodies against the various types of
collagen are used. If Type II collagen, for instance, is
15 detected in the synovial fluid, this may indicate the
destruction of articular cartilage which is characteristic
of osterarthrosis, but which also occurs in other erosive
joint disorders. On the other hand, if Types I and III
collagens are detected, this may be more indicative of the
inflammatory-proliferative phase of rheumatoid arthritis.
The ability to diagnose early lesions of articular
cartilage can effect the choice of the appropriate therapy
with which to treat the patient.
The application of monoclonal antibodies
conjugated to fluorophores that fluoresce at variable
wavelengths to various tissue sections represents a
sensitive means of detecting changes in the collagen
distribution within biopsied tissue samples. For
30 instancef if liver cirrhosis is suspected, part of the
affected tissue can be immunohistologically stained with
monoclonal antibodies against Types I, III and IV
collagens. Normally, liver contains very little collagen;
thus a lack of significant fluorescent staining would
35 indicate a healthy liver. On the other hand, if Type IV
collagen was detected in the sample, this would suggest
5 ~'7~9
-25-
the early stages of cirrhosis. Similarly, if the
monoclonal antibodies against Types III and I collagens
detected the deposition of such collagens, this would
5 suggest the more advanced stages of the fibrotic disease.
[It should be noted that fibrotic aiseases affecting the
liver and the connective tissue of the organs such as skin
and bone can also be detected by serological (i.e.,
non-biopsy) means utilizing monoclonal antibodies on serum
sarnples.]
~0
One of the most important applications of
monoclonal antibodies against connective tissue proteins
is for the purpose of early and accurate cancer
15 diagnosis. For example, malignant epithelial cells of
breast carcinomas actively produce basement membrane (Type
IV) collagen. The production of this collagen continues
as the cell metastasizes to other locations, such as the
lymph nodes surrounding the breast area. With monoclonal
20 antibodies against Type IV collagen, the presence of a
single metastasized cell can be detected
immunohistologically in a lymph node biopsy. An early
diagnosis o~ infiltrating cells and lymph node metastasis
as judged on the basis of as little as one basement
25 membrane collagen-synthesizing tumor cell is of
significant prognostic importance.
Other types of malignancies may also be diagnosed
by detecting the neosynthesis of collagens. For instance,
30 monoclonal antibodies against collagens may also prove
useful for locating malignant cells in cytological samples
such as in Pap smears taken to diagnose cervical and/or
uterine cancers. Monoclonal antibodies may also be used
in immunohistological differential diagnoses to
distinguish, for example, malignant melanomas from benign
naevi.
' :'
~.~5~"t7~5~
4 . 8 . THERAPY MONITORING USING
MONOCLONAL ANTIBODIES AGAINST
CONNECTIVE TISSUE PROTEINS
Monoclonal an~ibodies against connective tissue
proteins may be used to monitor th~ effectiveness of
antifibrotic drug therapies. They provide the
immunoserological, immunohistological and
10 immunocytological means to detect an inhibition or
suppression of collagenous connective tissue neosynthesis,
the resulting diminution in the accumulation o the
collagenous matrix, and hence, the antifibrotic effect of
the drug.
Similarly, monoclonal antibodies may be used to
monitcr the effectiveness of certain chemotherapies aimea
at eraaicating malignant tumor cells. For example, tumor
cells present in bone marrow malignancies produce the
enzyme elastase which selectively cleaves one fourth of
the Type III collagen molecule to yield a peptide
fragment. If such cells are successfully destroyed by
chemotherapeutic means, both elastase and the Type III
peptide fragment are released and enter the serum.
Detection of this enzyme and peptide in serological
samples using monoclonal antibodies provides a sensitive
and non-invasive means for monitoring the efficacy of
anti-tumor drug therapies.
4.9. ANALYTICAL MET~ODS
4.9.l. RADIOIMMUNOASSAY
A radioactively labeled connective tissue protein
35 is mixed with monoclonal antibodies specific for that
particular protein as antigen and with a serological
!''~, -
~,
'7
-27-
sample containing an unknown amount of unlabeled
connective tissue protein. The labeled and unlabeled
antigen compete for binding with the monoclonal antibody.
5 The more unlabeled connective tissue protein there is in
the serological sample, the less labeled antigen binds
with antibody to form an insoluble complex. By measuring
the amount of radioactivity associated with either the
insoluble or soluble fractions of the reaction mixture and
10 comparing the values obtained with an appropriately
constructed calibration curve (wherein known amounts of
unlabeled and labeled antigen were reacted with antibody),
the amount of connective ~issue protein in the sample can
be accurately quantitated.
4 . 9 ~ 2 ENZY~E-LINKED I~UNOSORBENT ASSAY
.
Connective tissue proteins in serological samples
can be measured by a variation of the en~yme-linked
immunosorbent assay (ELISA) used to screen hybrids for
antibody production (see Section 4.5). Enzyme
immunoassays (EIA) are based on the principle of
competitive binding as described in Section 4.9.1 for
radioimmunoassay (RIA). The procedures differ in that an
25 enzyme is used as the ~label" in EIA as opposed to a
radioisotope as in RIA.
4.9.3. DMMUNOHISTOLOGICAL AND
IMMUNOCYTOLOGICAL STAINING
-
Slides containing cryostat sections of frozen,
unfixed tissue biopsy samples or cytological smears are
air dried and incubated with a single monoclonal an~ibody
preparation. The slides are then layered with a
preparation of antibody directed against the monoclonal
antibody. This anti-monoclonal antibody immunoglobulin is
5 ~2
-28-
tagged with a compound that fluoresces at a particular
wavelength for instance rhodamine. If it i5 desirable to
immunohistologically (or immunocytologically) stain for
5 more than one type of connective tissue protein in a given
sample, the slide is then layered with another coating of
a second type of monoclonal antibody. This is followed by
the application of a second anti-monoclonal antibody
immunoglobulin tagged with a compound that fluoresces at a
different wavelength than the first fluorophore, such as
fluorescein isothiocyanate, and so on until all the
connective tissue proteins have been stained. The
localization of the connective tissue proteins within the
sample is then determined by fluorescent light microscopy
15 and optionally photographically recorded.
4.9.4. IMMUNOELECTRONMICROSCOPY
Under some circumstances it may be necessary to
use immunoelectronmicroscopy to detect the presence of
collagen and other connective tissue proteins in
histological samples. [See, e.g., D. Engel et alO~ Archs
oral Bio. 25:283-296 (l980)].
5 EXAMPLES
5.10 CONSTRUCTION OF HYBRIDOMAS SECRETING
MONOCLONAL ANTIBODIES TO
CONNECTIVE TISSUE PROTEINS
The cell hybridization techniques of this
invention are adopted from the protocol of Drs. J.
Kearney, A. Anderson and P. Burrows, of the Cellular
Immunobiology Unit, 224 Tumor Institute, University of
Alabama in BirminghamO [G.J. Hammerling, U. Hammerling
and J.F. Kearney (editors), Monoclonal antibodies and
","~,.....
9 ~l9
~L~ f A~
-29-
T-cell hybridomas in- J.L. Turk (editor), Research
Monographs in Immunology, Vol. 3, Elsevier/North Holland
Biomedical Press, New York (1981)].
5.1.1. PURIFICATION OF CONNECTIVE_TISSUE PROTEINS
Methods for the preparation of the individual
types of collagens have been described extensively by
10 ~iller and Rhodes [Structural and contractile proteins,
in: L. Cunningham and D. Frederiksen (editors~, Methods in
Enzymology, Academic Press, New York (1981)].
The method used to isolate and purify the
15 collagen-degrading enzyme, elastase, is a modification of
the procedure of Mainardi et al. lJ. Biol. Chem 255(24):
12006-12010 (1980)].
- 5.1.2. IMMUNIZATION SCHEDULES
At 5 to 6 weeks of age, e.~., BALB/c female mice
(Jackson Laboratories) are immunized with 200 ug of a
purified connective tissue protein as antigen. The
antigen is delivered in 0.5 ml of complete Freund's
25 adjuvant by subcutaneous inoculation. An immunization
schedule is followed wherein the mice are boosted
intraperitoneally with a similar amount of antigen 21 days
after the initial priming. Only a single boost is
administered, though other immunization schedules with
30 multiple boosts may be used with similar success. The
spleens and lymph nodes are removed 4 days after the
booster inoculation following standard techniques
[Llnsenmayer, T.F., Hendrix, M.J.C. and Little, C.D.,
Proc. Natl. Acad. Sci. U.S.A. 76:3703-3707 (1979)~.
~2~7~g
-30-
5.1.3. SPLEEN CELL PREPARATION
Spleens of immunized BALB/c mice are removed
5 under sterile conditions and washed in serum-free RP~I
1640 medium (Seromed, Munchen, F.R.G.). The spleens are
macerated through cheesecloth and are then resuspended in
serum-free RPMI 1640 medium and centrifuged; this washing
procedure is performed three times at 4C. After the
10 final washing, the cells are resuspended in the same
medium in a 50 ml sterile tube. The number of cells in
the preparation is determined microscopically before
mixing with myeloma cells for fusion (see Section 5.4).
-
5.1.4. MYELOMA CELL PREPARATION
A variant subclone of the mouse myeloma cell lineP3-X63-Ag8, isolated by Kearney, et al. [J. Immunol.
123(4):1548-1550 (1979)] and designated X63-A98.653, is
20 maintained in Dulbecco's MEM or RPMI 1640 medium (Seromea,
Munchen, F.R.G.) supplemented with 15% fetal calf serum,
2 mM glutamine, 50 uM 2-mercaptoethanol (Merck, Darmstadt,
F.R.G.), 100 units/ml penicillin, 100 ug/ml streptomycin,
and 0.25 ug/ml Fungizone (Flow Laboratories, Bonn,
F.R.G.)(hereinafter called "complete mediumn). Like its
parent, X630-Ag8.653 is a hypoxanthine/aminopterin/
thymidine-(HAT)-sensitive cell line. However, unlike its
parent, X63-Ag8.653 has lost immunoglobulin expression
entirely and does not synthesize 1 or K chains of X63
30 origin upon fusion with antibody-forming cells. The
myeloma cells are cultured in complete medium and
harvested during the exponential phase of growth.
Harves~ed cells are transferred to 50 ml sterile tubes and
are washed three times in serum-free RPMI 1640 medium at
35 4C The cells are counted microscopically prior to
fusion with spleen cells.
~5~ 2
-31-
5 . 1. 5 . FUS ION PROCEDUR2
Spleen cells and X63-Ag8.653 myeloma cells are
5 combined in a ratio of 2:1 or 1:1 ~spleen cells:myeloma
cells) and washed once in serum-free RPMI 1640 medium at
37C~ The cell mixture is centrif~ged at room temperature
at 1,000 rpm for 7 minutes. The pellet fraction is
carefully aspirated to leave it as dry as possible. Next,
the pellet is loosened by gentle tapping and is
~esuspended with gentle agitation in l.D to 1.5 ml of
PEG-4000 ~polyethylene glycol) solution at 37C. The
PEG-400 solution is prepared by autoclaving 20 gm PEG 4000
in a lDO ml bottle, cooling and adding 28 ml of sterile
15 phosphate buffered saline. After approximately 30
seconds, the cell mixture is slowly ailuted dropwise to a
volume of roughly 20 ml with serum-free RPMI 1640 medium
at 37C; the tube is then filled to 50 ml with the same
medium. The cells are centrifuged at room temperature and
resuspended at 37C in HAT medium, which is selective for
fused cells. HAT medium is prepared by adding 1 ml of a
stock solution (lOOX) of hypoxanthine (~) and thymidine
(t) and 1 ml of a stock solution (lOOX) of aminopterin (A)
to 100 ml of complete medium. The ~T stock contains 272.2
25 mg hypoxanthine and 7.75 mg thymidine in 200 ml distilled
water~ Because the hypoxanthine does not dissolve well,
the p~ of the solution is adjusted to p~ 8.1-8.5 with 1-2
drops of lN NaO~. The solution is sterilized through a
0.45 u Millipore filter and stored at 4C. The A stock
contains 3.52 mg aminopterin in 200 ml distilled water.
It is also sterilized by Millipore filtration and stored
at 4C.
The cell mixture is resuspended in HAT medium at
a concentration of about 2 to 5 x 105 spleen cells/ml.
Care is taken not to break up cell clumps. Peritoneal
29
-32-
exudate feeder cells are then added (roughly, the
peritoneal washout of one normal, non-immunized mouse per
100 ml of ~AT/fused-cell suspension) and 1 ml of the cell
5 suspension is added per well of a 24 well macroti~er plate
(Costar, Cambridge, Massachusetts).
5.1.6. OUTGROWTH AND SELECTION
After 4 or 5 days in the selective HAT medium,
~the cells are observed with an inverted microscope to
check for myeloma cell death. (The X63-Ag8.653 cell line
is HAT-sensitive and thus, unfused myeloma cells cannot
survive in this medium; unfused spleen cells naturally die
15 out of the culture.) Contamination of the wells is also
checked for and any contaminated wells are killed with a
copper sulfate solution.
The fused cells are allowed to incubate in HAT
medium for two weeks at which time 0.5 ml of the
supernatant of each well is discarded and replaced with
O.5 ml of complete (non-HAT) medium. This medium
replenishment is repeated daily for another week. Two to
three days after the last medium replenishment, which i
enough time to allow for sufficient production of
antibodies for testing, the macrotiter plates are scored
for hybrid growth and assayed for antibody activity by the
ELISA method described in Section 5.1.7.
5.1.7. I~MUNOLOGICAL CHARACTERIZATION OF
HYBRID-PRODUCED MONOCLONAL ANTIBODIES
The identification of those hybrids synthesizing
antibodies which recognize the connective tissue protein
3 used as antigen is accomplished using a modification of
the enzyme-linked immunosorbent assay (ELISA) [Engvall, E.
'7~
and Perlman, P., lmmunochem. 8:871-876 (1971)] as detailed
by Kearney et al. [J. Immunol. 123:1548-1550 (1979)~.
The wells of a 96-well polyvinyl microtiter plate
are coated with 100 ul/well of 1 mg/ml solution of
collagen antigen in borate saline. The plate is incubated
for four hours at 25~C or overnight and 4C. The plates
are then blocked with 1~ bovine serum albumin (BSA) in
10 borate buffered saline (BS-BS~) and incubated for one hour
at 25C. The wells of ~he microtiter plate are washed
twice with saline, after which the supernatants
(containing monoclonal antibodies) from the wells of the
macrotiter plates used for outgrowth and selection of
15 fused hybrids are added to the microtiter wells. The
plate is incubated for four hours at 25C (or overnight at
4C) and washed two to three times with saline. To each
well, 100 ul of alkaline phosphatase-labeled antibodies
(yoat antimouse-immunoglobulin) diluted 1:500 in BS-BSA is
added. After incubating for four hours at 25C or
overnight at 4C, the wells are washed 4-5 times with
saline and 200 ul of substrate (p-nitrophenylphosphate) is
added per well. The reaction is stopped by the addition
of 50 ul 3 N NaOH to each well. The absorbance of the
25 fluid in the wells i5 then determined
spectrophotometrically.
In those wells to which monoclonal antibodies
from the culture supernatants bind and to which the
30 enzyme-linked goat antimouse-immunoglobulin subsequently
bind, the alkaline phosphatase converts colorless
p-nitrophenylphosphate into yellow p-nitrophenol. The
colorometric reaction permitts the easy identification of
those culture supernatants containing collagen-specific
35 antibodies and hence the identifi~ation of the desired
fused hybrids. This step is performed to exclude from
~5~'7~:~
-34-
further analysis those hybrids that do not produce
immunoglobulin and those that synthesize antibodies not
specific for the collagen protein antigen.
5.l.8. CLONING OF ~YBRIDS
The extent of hybrid cell growth in the wells of
the macrotiter plates (see Section 5~lo6~) is determined
10 3~4 weeks after the initial plating in HAT medium. The
cell suspensions were agitated gently and 2-5 ul are
diluted from each well into 30 rnl of complete medium
containing peri~oneal exudate feeder cells. Into each
well of a 96-well costar microtiter plate, 200 ul of the
15 diluted cell suspension are distributed. This suspension
is diluted ~urther by delivering l0 ml in 20 or 30 ml of
medium containing feeder cells and 200 ul aliquots are
added to each well of another microtiter plate. Further
dilutions of the cell suspension can be performed if
necessary. This method is used to insure that the wells
of at least one plate contain clones derived from a single
cell. Samples of cells from the original
hybridoma-containing macrotiter wells are frozen for
safekeeping.
After a sufficient time for growth of the
hybridoma cells (clones), the supernatants of the
microtiter wells are rescreened for monoclonal antibody
production using the ELISA assay described in Section
30 5.1.7.
5.1.9. STABILITY OF PHENOTYPE DETERMINATION
Those hybrids identified to be specific antibody
35 producers are transferred to new Costar plates at low cell
density (approximately 5 cells/well~. Surviving hybrids
'7
-35-
are screened and those continuing to demonstrate antiboay
production are recloned to insure that the antibodies
produced arise from a single fused hybrid and hence are
5 monospecific.
5.1.10. ~ETERMINATION OF MONOCLONAL ANTIBODY SPECIFICITY
_
The culture media from hybrids that survived tws
10 successive clonings and that continued to exhibit a stable
phenotype are screened for cross-reactivity against the
other types and individual molecular forms of collagen as
well as other connective tissue proteins using the ELISA
assay of Section 5.l.7. Instead of using the antigen
15 against which the monoclonal antibody was raised to coat
the wells of the Costar plates, the other individual
collagens and connective tissue proteins are used in the
ELISA assay. Only those monoclonal antibodies exhibiting
no cross-reactivity are used in the procedures for
detecting connective tissue proteins in body fluids and
tissue samples described in Sections 5.2, 5.3, and 5.
below.
5.l.ll. PROPAGATION OF HYBRID CELLS
AND ANTIBODY PRODUCTION
Hybrids which synthesized antibodies of the
desired specificity are amplified in cell culture and
stored in liquid nitrogen so that an adequate supply of
30 cells producing identically monospecific antibodies are
available. To propagate the hybrids, samples of the fused
cells are injected intraperitoneally into BALB/c mice
(106 cells/mouse) resulting in the subsequent induction
of palpable tumors within a few weeks. The tumors
generally produce ascites fluid (approximately 2 ml per
mouse) containing antibody amounts significantly greater
5 ~
-3~-
(as high as 60 mg per mouse) than those obtained by in
vitro cell culture techniques. Sera samples from the
inoculated mice contain antibody titers comparable to tbat
of ascites fluid. The mouse hybridoma-produced monoclonal
antibodies are purified by subjecting sample~ of ascites
fluid, sera, or media to immunoadsorption chromatography.
5.2. DETECTION AND MEASUREMENT OF
CONNECTIVE TISSUE PROTEINS IN
BIOLOGICAL FLUIDS WITH MONOCLONAL
ANTIBODIES
_ .
5 . 2 .1. RADIOIMMUNOASSAY
Iodinated connective tissue protein antigens are
prepared as described by Rohde et al. and are used in a
modification of the radioimmunoassay described by the same
authors lJ. Immunol. Meth. 11:135-145 (1976)]. Antibody
titrations are carried out by diluting the monoclonal
antibody preparation with PBS. Duplicate tubes containing
0.1 ml monoclonal antibody preparation (ascites fluid or
tissue culture flui~), 0.1 ml labeled antigen, and 0.2 ml
1~ BSA dissolved in PBS are incubated for 24 hours at
25 4C. After mixing with 0.5 ml antiserum to mouse Ig, the
incubation is continued for an additional 24 hours at
4C. Insoluble material is collected by centrifugation
and the precipitate is washed three times with cold
PBS/BSA prior to counting. Non-specific precipitation of
labeled antigen is determined by replacing the monoclonal
antibody preparation by non-immune Ig. Antigen binaing
capacity of the monoclonal antibody preparation is
calculated according to Minden and Farr [D.~. ~eir,
(editor) Handbook of Experimental Immunology, Blackwell,
Oxford, England, p. 151].
-37-
In the competition assay, sufficient monoclonal
antibody is used to bind 80% of the labeled antigen.
However, the monoclonal antibody preparation is first
5 incubated with a sample containing unlabeled connective
tissue protein at 4C for 24 hours and then the labeled
antigen is added to the reaction, ollowed by incubation
and finally addition of and incubation with anti-mouse Ig
as above. Precipitable counts are measured, also as above
in a Beckman Gamma 300 counter.
5.2.2. ENZYME-LINKED IMMUNOSORBENT ASSAY
Monoclonal antibodies directed against connective
15 tissue proteins are conjugated to alkaline phosphatase by
the method of Hammerling et al. [Monoclonal antibodies and
T-cell hybridomas in: J.L. Turk (editor) Research
Monographs in Immunology, Vol. 3, Elsevier/North ~olland
Biomedical Press, New York (1981)]. Dialysis tubing is
boiled for 20 minutes in deionized water. Alkaline
phosphatase (1.5 mg as an ammonium sulfate-precipitated
slurry) is centrifuged at 4C for 2-3 minutes at 12,000xg
and the supernatant is discarded. The pelleted enzyme is
dissolved in buffer (Dulbecco's PBS with magnesium and
calcium cations, DP~S) containing an appropriate amount of
monoclonal antibody, in a volume of approximately 0.2 ml.
The antibody-enzyme mixture is dialyzed against 100 ml of
DPBS overnight at 4C. The contents of the dialysis
tubing are washed out into a graduated glass tube and the
30 volume is adjusted to 0.5 ml with DPBS. Next, 25
glutaraldehyde is added to a final concentration of 0.2
(4 ul for 0 5 ml). The mixture is gently agitated on a
Vortex mixer and is incubated for 2 hours at room
temperature. After dialyzing overnight against DPBS at
3 4C, the enzyme-coupled antibody is diluted to 10 ml with
5% BSA 0.05 M Tris buffer, which serves as a stock
solution.
~ t7~ ~
-38-
Enzyme-linked monoclonal antibodies thus prépared
are mixed with serological samples containing unknown
amounts of the specific connective tissue protein being
5 assayed. The mixtures are transferred to the welis of
microtiter plates pre-coated with the appropriate antigen
and the enzyme activity of the conjugated alkaline
phosphatase is measured as described in Section 5.1.7.
5.3. IMMUNO~ISTOLOGICAL APPLICATION OF
MONOCLONAL ANTIBODIES AGAINST
CON~ECTIVE TISSUE PROTEINS
5. 3 . l. IMMUNOFL~ORESCENT STAINING OF
BIOPSIED TISSUE SECTIONS
Sections of tissues 4-6 um thick are prepared
from frozen, unfixed biopsy samples by cryostat
sectioning. The air-dried sections are incubated with a
20 particular monoclonal antibody. For controlst sections
are incubated with immunoglobulin (Ig) from pre-immune
serum. After 30 minutes of incubation in a humidifiea
chamber at room temperature, the sections are rinsed three
times with phosphate-buffered saline (PBS, pH 7.4) and, in
25 a second step, layered with a 1:30 dilution of
fluorescein-isothiocyanate conjugated (FITC) rabbit
anti-mouse Ig for 30 minutes. Finally, the slides are
washed exhaustively to remove nonspecifically associated
reagents and are sealed with a solution of 90%
30 glycerol/10% PRS unaer a coverslip. The localization of
staining is observed and photographed using a
Leitz-fluorescence microscope equipped with a K2 filter
system for FITC.
~L2~ 9
-39-
5.3.2. IMMUNOELECTRONMICROSCOPY
Immunoelectronmicroscopy is performed according
5 to [R. Fleischmajer et al., J. Invest. Dermat. 75:189-191
(1980)] and [Gay et al., Collagen Rel. Res. 1:370-377
(1981)]. Tissues, for instance kidney, are fixed in
phosphate buffered 4% paraformaldehyde at 4C for 2 hours
with one change . Tissues are then washed for 36 hours in
10 PBS with 4% sucrose at 4~C with multiple changes. I'he
last wash is performed in PBS with 4% sucrose and 5~
glycerol for 1 hour. Tissues are then placed in OCT
freezing medium with a cork or plastic backing to hold
them and quickly frozen by immersing them in a jar of
15 methylbutane (isopentane) placed in a small chamber of
liquid nitrogen. The frozen tissues are then wrapped in
aluminum foil and stored in a closed container at -20C.
Frozen sections 8 um thick are cut and placed in albumin
coated slides and air-dried for at least 5 minutes.
20 Slides are then placed in a solution of ice-cold NaBH4
(10 mg/100 ml) in PBS for 1 hour with one change.
Following this procedure, the slides are washed at 4C in
PBS, 3 changes for 30 minutes each.
Tissue sections are reacted with the appropriate
monoclonal antibody in a moist chamber overnight at 4C or
at room temperature for 2 hours. Slides are washed
thoroughly with PBS and then incubated an additional 2
hours with secondary antibody (goat or rabbit anti-mouse
30 Ig). This is followed by washing with cold PBS and a
third antibody treatment with Fab-peroxidase-
anti-peroxidase tFab-pAp) for 3 hours. The Fab-PAP
solution is removed by washing with PBS and the tissue
sections are incubated in 150 ml of 0.1 M Tris, pH 7.6,
35 containing 40 mg of 3,3-diaminobenzidine
tetrahydrochloride and 15 ul of 5% H2O2 for 15-18
minutes. 51ides are then washed with cold PBS and stained
~ ,g t ~r ~
7~J
-40-
with 1% osmium tetroxide for l hour at room temperature.
The stained slides are again rinsed with cold PBS,
dehydrated in acetone, embedded in MaraglasR (70%) and
5 ultra-thin sections are made for examination using a Zeiss
EM 10 electron microscope.
5.4. IMMUNOCYTOLOGICAL APPLICATION OF
MONOCLONAL ANTIBODIES AGAINST
CONNECTIVE TISSUE PROTEINS
. .
To determine the production of collagens and the
type of collagen synthesized by cells such as skin
fibroblasts which can be cultivated ln vitro by standard
cell culture techniques, the following procedure is used.
Anchorage-dependent cells which have grown to confluent
monolayers on solid supports are detached by exposure to
trypsin and are replated in the Dulbecco-Vogt modification
of Eagle's medium containing 104 fetal calf serum in 35 x
10 mm Falcon plastic tissue culture dishes. The dishes
are incubated at 37C under a 5~ CO2/95~ air
atmosphere. About 6 hours later, the medium in each dish
is replaced with fresh media which in some cases contained
50 ug/ml of newly dissolved ascorbic acid. These media
are replaced every day. At various times after plating
the cells, dishes are taken for analysis, the media are
removed, and the dishes are rinsed at least four times
with 0.15 M NaCl, 0.05 M Tris-HCl pH 7.4.
The air-dried culture dishes are rinsed with
acetone and allowed to dry. Purified monoclonal
antibodies dissolved in 0.15 M NaCl, 0.02 M sodium
phosphate, pH 7.4, are added to the dishes and allowed to
react for 2 hours at 20C. Controls are run to assess the
nonspecific associations of reagents with the cells. Such
control experiments indicated that the nonspecific
5 ~7
41-
association of label is negligible. Subsequently, the
dishes are rinsed three times with 0.15 M NaCl, 0.02 M
sodium phosphate, p~ 7.4, and are layered with 1 ml of a
5 1:32 dilution of fluorescein-isothiocyanate-conjugated
rabbit antimouse Ig. When cell samples are simultaneously
stained for two antigens, the dishes are first exposed to
one type of monoclonal antibody against a connective
tissue protein and then to the fluorescein-isothiocyanate-
10 conjugated rabbit antimouse Ig. Subsequently, the aishesare exposed to monoclonal antibodies against a different
type of connective tissue protein, washed, and then
reacted with rhodamine-conjugatea rabbit antimouse Ig.
Finally, all dishes are washed extensively to remove
15 adventitiously associated reagents and sealed from the air
with a solution of 90~ glycerol/10% saline under a cover
slip. The localization of fluorescent stains on the
dishes is observed in a Zeiss Universal fluorescence
microscope and recorded photographically.
