Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02385187 2002-05-07
1
METHOD FOR OBSERVING HUMAN ENDOMETRIAL CELLS
IN LIVING ANIMALS
BACKGROUND OF THE INVENTION
(a) Field of the Invention
The invention relates to a method for observing human endometrial cells or
tissue in a living animal and more particularly for visualizing human
endometriotic
lesion implantation, progression and regression in live animals. The invention
also concerns genetically modified human endometrial cells or tissue, animal
models and methods for the study of endometriosis and for evaluating candidate
drugs or gene products potentially useful for the treatment of endometriosis.
(b) Description of Prior Art
Endometriosis is a gynecological disease characterized by the growth of
endometrial tissue at extra-uterine sites. Although benign, it is an
aggressive and
invalidating disease, linked to several symptoms such as intense, and
sometimes
chronic, pelvic pain, infertility, deregulated and/or abundant menstrual
bleeding
(dysmenorrhea), painful intercourse (dyspareunia), diarrhea and/or
constipation,
in 10 to 15% of women of reproductive age. It is postulated that endometriosis
arises from dissemination of endometrial cells in the peritoneal cavity
through
retrograde menstruation. The exact reasons why, in some individuals, these
endometrial cells then adhere to the peritoneal wall, or other organs found
within
the abdominal cavity, proliferate, induce intense vascularization, and form
endometriotic lesions are not well understood. Nonetheless, immunological,
hormonal, genetic and environmental factors are suspected to play a role in
establishment and maintenance of the disease.
Direct and well-designed in vivo studies on human endometriosis are
hampered by practical and ethical considerations. On the other hand, in vitro
studies cannot reproduce the three-dimensional organization of endometrial
tissue and do not take into account the peculiarity of the peritoneal
environment,
the major site of endometriotic lesion implantation. Hence, some investigators
have used primate animal models for the study of endometriosis, essentially
CA 02385187 2002-05-07
2
based on spontaneous or chemically-induced formation of endometriotic lesions
in the peritoneum of monkeys. Although clinically relevant, these approaches
have been limited by both the extensive length of time required for lesions to
occur and their prohibitive costs. Therefore, other models, mainly based on
the
implantation of autologous endometrium tissue, have been developed and
validated in rodents such as rabbits, rats and mice. However, these systems
are
of limited value for the study of the human disease because of the significant
histological and biochemical (hormone and cytokine responsiveness, for
instance) differences between human and animal endometria. Therefore, the
transplantation of human endornetrium into immunodeficient animals (for
instance
Severe Combined Immunodefic;iency "SCID" or nude mice) has been undertaken
in many laboratories. This system was shown to faithfully reproduce the
behavior
of endometriotic lesions in humans (Bruner et al., 1997. J Clin Invest 99:2851-
2857), is amenable to drug testing, and meets practical considerations such as
low cost, reliability, easy handling and ethics. Notwithstanding the above,
this
human-mouse xenograft model still has some caveats. In fact, endometriotic
lesions are small (1-2 mm2), not easily distinguishable from the surrounding
tissues and their detection is arduous. More importantly, identification and
quantification of endometriotic lesions require killing of the animal for each
experimental point. This renders this type of model inconvenient for dynamic
studies on lesion development or regression after pharmacological treatment in
a
given animal and multiplies the number of animals to be used. Development of
new means to observe human endometriotic lesions in live animals is thus most
needed, as they would make of the human-mouse xenograft model the approach
of choice for the study of endometriosis and potential treatments.
Until recently, non-invasive imaging techniques have been mainly
developed for the visualization of tumor cells in animals. These models are
based
on complex and expensive instrumentation, not particularly suited for small
animals, and often requiring injection of chemical probes to the animals. For
instance, tumor cells expressing the luciferase gene can be non-invasively
detected in living mice after intraperitoneal injection of luciferine to
animals,
photon capture by an intensified charge coupled device camera and image
CA 02385187 2002-05-07
3
processing (see U.S. Patent 5,650,135). Positron emission tomography has also
been used for the same purposes, but this system requires that a positron-
emitting isotope be specifically retained inside the tumor cells. Furthermore,
the
resolution routinely achieved with this method is around 6 mm3, too imprecise
for
the detection of endometriotic lesions.
In the field of endometriosis, the use of external fluorescent labels for the
detection of endometriotic lesions in animals has been attempted but
fluorescence is very weak and is limited to very short time periods due to
rapid
diffusion of the probe out of the cells (Tabibzadeh et al., 1999. Front Biosci
4:C4-
C9; Yang et a1.,1996. Am J Ob:~tef Gynecol 174:154-160).
There is therefore a long felt need for a method which allows a precise
detection of endometriotic lesions in living animal models.
There is also a need for a non-invasive whole-body imaging method which
permits to monitor endometriotic lesion formation in living animals, in a
dynamic
way.
There is also a need fc~r human endometrial tissue or cells genetically
modified so as to express a detectable level of a light-emitting protein since
direct
fluorescent protein production within endometrial tissue has never been
achieved.
There is also a need for an animal model for the study of endometriosis which
comprises such genetically modified human endometrial tissue.
There is a need also for a method wherein human endometrial cells,
genetically modified so as to express a detectable level of a fluorescent
protein,
are administered to a living animal so as to permit an easy and sensitive
visualization of the human cells.
SUMMARY OF THE INVENTION
The invention relates to cells and tissue, animal model and methods for
observing human endometrial tissue in a living animal and more particularly
for
visualizing human endometriotic lesion implantation, progression and
regression
in live animals.
CA 02385187 2002-05-07
4
According to a first aspect, the invention relates to isolated or purified
human endometrial cells, or fragments of endometrial tissue, genetically
modified
so as to express a detectable level of a bioluminescent protein.
According to another aspect, the invention relates to an animal model for
;i the study of endometriosis, comprising isolated or purified human
endometrial
cells, or fragments of endometriaf tissue as defined previously. This animal
model
can be used for a dynamic visualization of endometriotic lesion implantation,
development and regression, through whole-body imaging. This animal model is
very useful as a tool for drug screening and target validation in pre-clinical
studies
1 CI of human endometriosis.
A further aspect of the present invention concerns a method for observing
human endometrial cells in a living animal. The method comprises the steps of:
- administering sub-cutaneously or intra-peritoneally to a living animal a
plurality
of human endometrial cells genetically modified so as to express a detectable
15. level of a fluorescent protein; and
- observing per-cutaneously fluorescence emitted by said human cells.
According to a more particular aspect, the invention concerns a method for
monitoring human endometrial implantation, progression and regression in a
living animal. The method comprises the steps of:
20 a) providing a plurality of human endometrial cells, either isolated or in
the form
of whole endometrial tissue;
b) introducing into these human endometrial cells a nucleotide sequence
capable of expressing a detectable level of a fluorescent protein into these
human endometrial cells;
25 c) administering sub-cutaneously or intra-peritoneally a plurality of the
cells of
step (b) to a living animal;
d) allowing the administered cells to form endometrial lesions) in the
abdominal
cavity of the living animal;
e) exposing the living animal to a light source capable of stimulating or
30 intensifying fluorescence emission by the fluorescent protein expressed in
the
human fluorescent cells; and
CA 02385187 2002-05-07
f) observing per-cutaneously fluorescence emitted by the human fluorescent
cells.
According to a further aspect, the invention relates to a method for
assaying a compound efficacy for the treatment of endometriosis. The method
:5 comprises the steps of: 1) administering a compound to an animal model as
defined previously; and 2) visualizing biological effects of the administered
compound onto the human fluorescent endometrial cells of the animal.
An advantage of the present invention is that it provides for the first time
human endometrial cells exprEasing a light-emitting protein, thereby
conferring a
1!) long-term and intense fluorescence to the cells. Furthermore, the
invention allows
for the first time a per-cutaneous observation of human endometrial cells in
living
animals, an approach which is much less invasive than what is commonly used in
the art. The cells, animal model and methods of the invention are therefore
very
useful for the study of endometriosis and for evaluating candidate drugs or
gene
1;i products potentially useful for the treatment of endometriosis.
BRIEF DESCRIPTION OF THE DRAWINGS
Figures 1 A, 1 B, 1 C, 1 D and 1 E are pictures showing GFP expression in
fragments of human endometrial biopsies after adenoviral infection.
2(1
Figures 2A and 2B are graphs showing a flow cytometric analysis of GFP
expression in human endometrial cells after adenoviral infection at different
multiplicities of infection (M01).
2:i Figures 3A and 3B are bar graphs showing that treatment of endometrial
tissue
with proteases before infection improves efficacy of human endometrial cells
adenoviral infection.
Figures 4A and 4B are pictures showing in vivo imaging of GFP-expressing
3Ci endometriotic lesions established sub-cutaneously in a living nude mouse.
CA 02385187 2002-05-07
6
Figures 5A, 5B and 5C are pictures showing in vivo imaging of an intra-
peritoneal
GFP-expressing endometriotic lesion in a living nude mouse.
DETAILED DESCRIPTION OF THE INVENTION
.5 Tissues and cells
The first aspect of thE: invention concerns isolated or purified human
endometrial cells, or fragments of endometrial tissue, genetically modified so
as
to express a detectable level of a bioluminescent (fluorescent) protein.
The isolated or purified human endometrial cells of the invention includes,
1 t) but is not limited to, primary, transformed or immortalized endometrial
cells,
whole or fragments of endomEarial biopsies, purified endometrial cell subsets,
or
any human cell giving rise to endometriotic lesions in vivo. Endometrial
tissue
obtained from endometrial biopsies at any stage of the menstrual cycle,
endometrial cell lines or primacy cultures of endometrial cells, transformed
or not,
15 or other cells able to produce endornetriotic-like lesions in animals can
be used
for the practice of the present invention.
The human endometrial cells of the invention are genetically modified so
as to express a detectable level of a fluorescent protein. Such modification
may
be done by the means of well-known methods such as transfection or viral
2(1 infection of a vector encoding the fluorescent protein, be it either
extrachromosomal or integrated into - the genome. The invention also
encompasses techniques wherein the incorporation of bioluminescent proteins,
or
the DNA sequences coding such proteins, directly into the cells by the means
of
methods such as receptor-mediated internalization, liposomes or micro-
injection.
2Ci In a preferred embodiment, the cells are infected with an adenovirus with
a
nucleotide sequence corresponding to the green fluorescent protein (GFP) from
the bioluminescent jellyfish Aequorea victoria. However, the fluorescent
protein
could be any detectable light emitting protein suitable for tracking
endometrial
cells non-invasively in a live vertebrate, including but not limited to near-
infrared
3Ci probes, proteins with an enhanced fluorescence emission spectrum (e.g.
enhanced green fluorescence protein or EGFP), and proteins with other emission
CA 02385187 2002-05-07
7
wave lengths or fluorescent-fusion proteins encompassing features similar to
GFP or EGFP.
According to another embodiment, the human endometrial cells further
comprise a potentially anti-proliferative exogenous nucleotide sequence, i.e.
a
;i sequence potentially capable of reducing, inhibiting, blocking
proliferation of the
human endometrial cells when expressed in these cells. Examples of exogenous
nucleotide sequences include but are not limited to antisense molecules,
complete or partial gene candidates (e.g. tyrosine kinase) and any similar
molecules with known or potent anti-proliferative or apoptotic activities. The
1 CI biological activity of the exogenous nucleotide sequence (or of a gene
product
encoded by the same), and proliferation of the endometrial cells may be
monitored using well known methods.
Animal model
Another aspect of the invention concerns an animal model that may be
1 ~~ used for the study of endometriosis and which comprises isolated or
purified
genetically modified human endometrial cells as defined previously.
Such an animal is obtained by administering thereto, into diverse locations
such as sub-cutaneous or into intra-peritoneal spaces, a plurality of
genetically
modified human endometrial cells according to the invention. The
transplantation
2G of the fluorescent endometrial cells may be done by any suitable means,
preferably by the means of injection or surgery. Preferably, these human
endometrial cells form endometrial lesions in the abdominal cavity of the
animal
and implantation, progression and regression of these endometrial lesions may
be monitored by observing, through the animal's skin, the fluorescence emitted
25 by the human fluorescent cells.
Preferably, the living animal of the invention consists of an immuno-
compromised animal or a syngeneic animal. More preferably, the living animal
consists of a rodent, including but not limited to mice, rats, hamsters,
guinea pigs,
rabbits, or a non-human primate.
CA 02385187 2002-05-07
8
Visualization of fluorescent cells
The invention allows a non-invasive detection of fluorescent endometrial
cells and endometriotic lesian(s) in live animals by visualizing light emitted
by
these cells or lesion(s).
Preferably, this is achieved by exposing the living animal with fluorescent
human endometrial cells to a light source capable of stimulating or
intensifying
fluorescence emission by the fluorescent protein. Of course the light source
must
provide an adequate wave-length for the excitation and subsequent fluorescent
emission of the fluorescent protein. Results can be viewed by a naked eye, or
by
using a more sensitive detector. The results may also be recorded through a
charge-coupled device camera or through any other image recording system.
Dynamic and real-time visualization of fluorescent endometriotic tissue
growing in
animals can be repeated many times, for instance for studying responses after
drug treatment (see hereinafter). More preferably, the animal is shaved or
hairless in order to facilitate light-emission visualization.
Drug screening
The animal model and cells of the invention can be used as a tool for drug
screening and target validation in pre-clinical studies of human
endometriosis.
Therefore, according to another aspect, the invention provides a method for
assaying a compound efficacy for the treatment of endometriosis (e.g. a
candidate drug or a gene product potentially useful for the prevention or
treatment of endometriosis). The method comprises the steps of: 1 )
administering
a compound for which efficacy is to be assayed to an animal model as defined
previously, and 2) visualizing biological effects of the administered compound
onto the fluorescent endometrial cells of this animal. For instance,
visualization
may consist of observing implantation, progression and regression of
endometrial
lesions in the animal, or may cansist of measuring fluorescence levels emitted
by
the human fluorescent cells. Of course, a much simpler method would consists
of
contacting in vitro the cells of the invention with the compounds) to be
assayed.
CA 02385187 2002-05-07
9
The present invention will be more readily understood by referring to the
following example. This example is illustrative of the wide range of
applicability of
the present invention and is not intended to limit its scope. Modifications
and
variations can be made therein without departing from the spirit and scope of
the
invention. Although any methods and materials similar or equivalent to those
described herein can be used in the practice for testing of the present
invention,
the preferred methods and materials are described.
EXAMPLE 1: In vivo visualization of implantation, pros~ression and
rearession of GFP-human endometriotic lesions in mice
Material and methods
Tissues and cells
Human endometrial biopsies were cut in small fragments of 1-2 mm3
with a surgical blade and are kept in culture medium [Dulbecco's modified
Eagle's
medium/Ham's F12 (Mediatech, Herndon, VA) containing 10% fetal calf serum
(FCS), 100 IU/ml penicillin, 100 pg/ml streptomycin, 250 ng/ml Amphotericin B
(Mediatech), and possibly 10 nM 17~-estradiol (Sigma-Aldrich, Oakville, ON,
Canada) or other cytokines or hormones]. In some cases, endometrial fragments
were incubated for 20 min at 37°C in either 0.05% trypsin, 0.53 mM EDTA
(Wisent), 1 mg/ml collagenase I;type 1A, Sigma) or 1 mg/ml collagenase, 1
mg/ml
hyaluronidase (Sigma), in order to slightly disrupt the tissue and favor
adenoviral
infection. Enzymatic digestion was stopped by the addition of 10% fetal calf
serum and extensive washing in culture medium. Ten fragments of endometrial
tissue were cultured per well of 48-well plates.
Adenoviral infection and GFP expression
In a well of a 48-well plate, ten endometrial fragments were incubated in
100 NI culture medium, alone or in the presence various amounts of purified
recombinant adenoviral particlE~s containing the GFPq gene under the CMVS
promoter (AdGFP) (Q-BIOgene, Montreal, QC, Canada). Cultures were incubated
for 24 h with continuous orbital agitation at 37°C, 5% C02 in a
humidified
CA 02385187 2002-05-07
incubator. GFP expression in whole endometrial fragments was verified by
fluorescence microscopy (Leica Canada, Willowdale, ON, Canada) or with a blue
light (470 nm) illumination system (Lightools Research, Encinitas, CA).
Quantification of the proportion of GFP-expressing cells was done by flow
5 cytometry (see hereafter).
Mice
Five to eight-week old ovariectomized female nude mice were purchased
from Harlan-Sprague Dawley (Indianapolis, IN) and kept under specific pathogen
10 free conditions. These immuno-compromised mice were chosen because they
are tolerant human tissue xenographs and hairless, facilitating light-emission
visualization. Other immuno-compromised mice or syngeneic mice in the case of
mouse tissue transplantation can also be used. Twenty-four to 96 h before
endometrial tissue transplantation, mice were anesthetized with 2% isofurane
and
sterile 60-d release capsules containing 1.5 mg 17~i-estradiol (Innovative
Research of America, Mendelein, IL) were inserted subcutaneously at a site
just
below the scapula, in order to maintain constant steroid levels during the
whole
experiment and optimize endometriotic lesion formation. Mice were always
manipulated in sterile conditions.
Infection of human endometrial tissue and non-invasive assessment of lesion
formation
After the 24 h culture step, endometrial fragments were washed
extensively in sterile PBS/2% FCS before transplantation to nude mice. Animals
were injected either subcutaneously or intraperitoneally with 10 human
endometrial fragments in 200 p1 PBS/2% FCS, with a 18-gauge needle. At
different time points after injection, live mice were examined under a blue
light
source (Lightools Research) for non-invasive (whole body imaging) assessment
of lesions. Alternatively, two to three weeks after transplantation, mice were
sacrificed and their peritoneum and visceral organs were examined, under blue
light illumination, for the removal of endometriotic-like lesions. Lesions
were
immediately placed in ice-cold culture medium for later analysis by flow
CA 02385187 2002-05-07
11
cytometry. Images were acquired with a Nikon Coolpix 990T"" digital camera
(Nikon Canada, Ville St-Laurent, QC, Canada).
Flow c~rtometric analysis
!i Human endometrial fragments infected 24h in vifro with AdGFP were
digested with 0.05% trypsin, 53 mM EDTA and 1 mg/ml collagenase in order to
obtain single cell suspensions. Cells were extensively washed and resuspended
in PBS. GFP fluorochrome was excited at 488 nm and fluorescence was detected
at 525 nm. Analysis was performed on a Coulter XLT"" flow cytometer with the
XL2T"" software (Coulter Electronics, Ville St-Laurent, QC, Canada).
Results
Fluorescence of endometrial fragments cultured for 24h with adenoviral
particles
carrying the GFPg giene
1 ~~ Development of endometriotic lesions in the mouse is better achieved by
the transfer of whole human endometrial fragments, rather than isolated cells,
into the peritoneal cavity of animals. In order to preserve the three
dimensional
structure of the tissue while achieving high levels of fluorescence, in a
slowly
cycling tissue, adenoviral-mediated gene transfer was used. Furthermore, we
chose a "bright" GFP variant, which exhibits 100-fold brighter fluorescence
than
wild-type GFP and placed it under the control of a strong modified
cytomegalovirus promoterlenhancer (CMVS).
A first experiment was carried out to verify the fluorescence of endometrial
fragments after infection with GFPq-encoding adenoviral particles. Endometrial
fragments were incubated for 24h with different doses of adenoviral particles
containing the GFP gene (AdGFP). Typical aspects of GFP-expressing
endometrial fragments and cells are shown in Figure 1. Under a fluorescent
microscope, intensity and homogeneity of GFP expression at the surface of
endometrial biopsies appears to vary between samples, even when identical
preparation and dose of AdGFP particles were used {compare Figures 1A, 1 B
and 1 C). On the other hand, cells having detached from endometrial fragments
and adhering to the bottom of the well appear more homogeneously infected
CA 02385187 2002-05-07
12
(Figure 1 D). Also, a macroscopic view of GFP-expressing endometrial
fragments,
under blue-light illumination is shown in Figure 1 E.
Quantification by flow-cytometry of GFP expression in endometrial cells after
adenoviral infection at different multiplicities of infection (M01)
To evaluate the proportion of GFP-expressing cells in whole endometrial
fragments, we quantified GFP fluorescence by flow cytometry, which analyzes
fluorescence at the single cell level. Figure 2A shows the widespread GFP
expression in cells obtained from a pool of 5 endometrial fragments incubated
with 500x106 AdGFP particles. In this case, 26% of the cells were GFP-
positive,
with expression levels extending over three logs of fluorescence. The
percentage
of GFP-positive cells varied from one experiment to another, probably due to
the
structure of the target tissue, ranging from 10 to 30% for a viral load of
500x106
particles/well for instance (Figure 2B).
Improvement of adenoviral transduction efficiencv by slight aroteases
treatment
of endometria) tissue prior to infection
fn order to obtain more efficient and homogeneous adenoviral infection, we
submitted endometrial fragments to mild enzymatic treatment, thus slightly
loosening the tissue and favoring contacts between viral particles and target
cells.
Fragments were treated by a mixture of either trypsin/EDTA/collagenase (TC) or
hyaluronidase/collagenase (HC) for 20 min at 37°C, before adenoviral
infection.
Such treatments did improve infection efficiencies (Figure 3A), and increased
the
number of GFP-expressing cellls from 15%, without treatment, to 35% and even
50% after HC or TC treatment, respectively. Interestingly, not only more cells
expressed GFP after enzymatic treatment, but also higher levels of GFP
fluorescence were achieved (Figure 3B). We later confirmed that enzymatic
treatment was not deleterious to the further implantation of endometrial
tissue in
the mouse peritoneal cavity. -This pre-treatment of endometrial tissue should
represent a valuable approach for increasing both the proportion of cells
expressing GFP and the global fluorescence of infected fragments, allowing an
easier visualization of endometriotic lesions in living animals.
CA 02385187 2002-05-07
13
In vivo imaging of GFP-expressing endometriotic lesions established sub-
cutaneously in a living nude mouse
We determined whetheir GFP levels achieved after adenoviral infection
would be sufficient to observe endometriotic lesion formation through the
skin.
Mice were injected subcutaneously (s.c.) with ten endometrial fragments
infected
with 1000x106 viral particles for 24h. It has been shown previously that
endometrial fragments injected s.c. in the nude mouse can grow and develop as
endometriotic implants on the outer face of the peritoneal membrane, and form
little blebs under the skin of the animal. Four days after s.c. injection, a
little bulge
was readily detectable through the skin of the animal, and appeared
fluorescent
when observed under a blue light source and through an orange filter,
providing
adequate excitation and emission filters for GFPq (Figure 4A). After removal
of
the skin, fluorescence was even more intense (Figure 4B). Of note, although
adenoviral DNA does not integrate into the hast genome, expression of the
foreign GFP gene remained elevated for a sufficient period of time {at least 3
weeks), allowing establishment and subsequent visualization of endometriotic
lesions. This experiment shows that endometrial fragments transduced with the
GFP gene by adenoviral infection can survive and implant in vivo, while
remaining sufficiently fluorescE~nt in order to be seen through the skin of
the
animal.
In vivo imaging of an intra-peritoneal GFP-expressing endometriotic lesion in
a
living nude mouse
To assess whether GFP-expressing endometrial tissue injected
intraperitoneally could also form lesions visible through the skin of the
animal, ten
fragments of endometrium cultured with 1000x106 viral particles for 24h were
injected. Mice were observed 12 days later. Without exposure to the blue
light, no
external sign of lesion formation was observed. However, when the mouse was
illuminated, a small but distinctive lesion was visible in the pelvis (Figure
5A).
Fluorescence became more evident as the skin (Fig. 5B) and the peritoneum
(Fig. 5C) were removed.
CA 02385187 2002-05-07
14
While several embodiments of the invention have been described, it will be
understood that the present invention is capable of further modifications, and
this
application is intended to cover any variations, uses, or adaptations of the
invention, following in general the principles of the invention and including
such
departures from the present disclosure as to come within knowledge or
customary practice in the art to which the invention pertains, and as may be
applied to the essential features hereinbefore set forth and falling within
the
scope of the invention or the limits of the appended claims.
