Note: Descriptions are shown in the official language in which they were submitted.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-1-
DETERMINING NUCLEIC ACID SEQUENCES IN A
BIOLOGICAL SAMPLE
TECHNICAL FIELD
This invention relates to methods and compositions used to assay a nucleic
acid burden in tissue or other body compartments. Specifically the invention
concerns the determination of HIV viral infections by determining viral
burden.
BACKGROUND
The HIV infection cycle begins with the entry of the virus into the target
cell. The human CD4 is believed to be the primary receptor on T cells
recognized by
HIV. The binding of the HIV envelope glycoprotein (env) to the CD4 receptor
results
in the fusion of virus and cell membranes, which in turn facilitates virus
entry into the
host. The eventual expression of env on the surface of the HIV-infected host
cell
enables this cell to fuse with uninfected CD4-positive cells, thereby
spreading the
virus. However, HIV can also enter other cells such as monocytes, B cells, and
dendritic cells, which can serve as viral reservoirs, even though they may not
express
CD4. Cytokines are known to affect HIV replication. Pro-inflammatory cytokines
promote HIV replication (Fauci, Nature 384:529-534, 1996), while ~3-chemokines
inhibit the replication of obligate CCRS utilizing viruses (Moore, et al., J.
Virol.
70:551-562, 1996), and enhance the replication of CXCR4 utilizing viral
isolates
(Dolei, et al., AIDS 12:183-190, 1998).
The normal intestinal tract is characterized by a low level of mild
inflammation, which is fueled by constitutive levels of locally secreted
chemokines
and cytokines (Shanahan and Anton, Gut Peptides, J. Walsh eds. (Raven Press,
Ltd,
New York, 1994, page 851; Schreiber et al., Gastroenterology 101:1020 (1991);
MacDermott et al., Inflammatory Bowel Diseases 4, 54 (1998); Luster, NEngl. J.
Med. 338:436 (1998)). In healthy controls, gastrointestinal lymphocytes are
known to
differ functionally and phenotypically from their peripheral blood
counterparts
(Allison et al., Gastroenterology 99:421 (1990); Jarry et al., Eur. J.
Immunol.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-2-
20:1097 (1990); McGowan et al., Neuroimmunomodulation 4:70 (1997)). Virtually
all mucosal CD4+ lymphocytes express activation markers and are of the CD45R0+
memory subset (Schieferdecker et al., J. Virol. 149:2816 (1992)). In the
setting of
HIV-1 infection, various phenotypic abnormalities of gut T lymphocytes have
been
described often associated with depletion of CD4+ lymphocytes (Schnieder et
al.,
Clin. Exp. Immunol. 95:430 (1994)).
Quantitative measurement of HIV RNA viral titers in plasma has quickly
become the mainstay of clinical management and the primary index of
therapeutic
efficacy in clinical trials of combination therapies for HIV. Reducing plasma
viral
load to an increasingly lowered threshold of detectability has been the
clinical target
the sensitivity of which is increasingly refined. Awareness of the persistence
of viral
activity in tissue (i.e., lymph nodes, the initial site of the observation)
and/or blood
cells in subjects with prolonged periods of undetectable levels of HIV in the
plasma
has fueled interest in the development of techniques that are as effective as
the plasma
1 S viral load kits commercially available for monitoring viral activity.
Reduction in plasma viral load indicates a reduced rate of viral replication.
However, the extent of viral replication throughout a subject's body is a
crucial
variable that cannot be inferred directly from the plasma viral load.
SUMMARY
The present invention is based upon the discovery that pathogens, such as
viruses (e.g., HIV and SIV) and bacteria can become established in tissue
compartments and immune cells deep within various tissues of the body shortly
after
infection. These deep tissues or compartments reflect the greatest percentage
of total
body pathogen burden. For example, the deep tissues or compartments reflect
the
greatest percentage of total body viral burden, and particularly the greatest
percentage
of HIV viral burden. As such, the currently available tests of viral burden
only
measure the concentration of virus in the patient's blood. One crucial
limitation of
these measurements is that even when virus is not detectable in blood, there
is still a
substantial viral burden and often ongoing replication in the patient's
tissues.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-,
Additionally, the present invention provides methods, compositions and
kits for maintaining and determining the quantity of a specific nucleic acids
in a
sample. Such methods utilize a "spiking" techniques in which the biological
sample
is spiked with an internal standard of a known quantity at the time of
obtaining the
biological sample in order to determine the natural degradation of the sample
over
time, such as during shipment.
In one embodiment the present invention provides a method of
determining a specific nucleic acid sequence in a non-fluid biological sample,
by
obtaining a biolgical sample from the subject and quantifying the amount of
the
specific nucleic acid present in the sample. The specific nucleic acid can be
related to
a pathologic condition, such as a viral or bacterial infection. The sample can
selected
from the group consisting of mucosal tissue, skin tissue, lymph tissue, ocular
tissue,
pulmonary tissue, and liver tissue.
In another embodiment, the present invention provides, a method for
determining the effect of a therapeutic treatment on whole body viral load in
a
subject, comprising determining the viral load in the plasma of the subject
and the
viral load in a tissue-biopsy of the subject compared to a standard whole body
viral
load, wherein a change in the viral load in the blood and the tissue is
indicative of an
effect.
In yet another embodiment, the present invention provides a kit for
obtaining a non-fluid biological sample from the skin, the kit having a
collection
device a known quantity of a nucleic acid; and a cell lysis buffer suitable of
preserving nucleic acids.
The details of one or more embodiments of the invention are set forth in
the accompanying drawings and the description below. Other features, objects,
and
advantages of the invention will be apparent from the description and
drawings, and
from the claims.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-4-
DESCRIPTION OF DRAWINGS
Figure 1 shows pre-amplification handling of tissue biopsy samples results
in a 5-10% RNA loss. Sero-negative samples were 'spiked' with 250 copies of
the
standard LTR sequence pre-extraction (left) and in a parallel sample post-
extraction
(right). Digitized quantification of 32P emission demonstrated a 5-10%
difference
between pre and post extraction additions of the same amount of LTR RNA.
Standards demonstrate an assay sensitivity of 10 copies.
Figure 2 shows the ability to detect and quantify levels of HIV RNA in
rectosigmoid biopsies from 4 patients with variable but detectable plasma
viral loads.
Samples had 10' to 103 copies HIV RNA in tissue per ~g total RNA added.
Quantified
values are corrected for percentage RNA loss and therefore reflect in vivo
levels.
Assay sensitivity is 10 copies.
Figure 3 demonstrates the reproducibility of the invention in patients with
detectable plasma viral load by showing that biopsies from the same subject in
1 S different quadrants at the same level in the rectosigmoid yield similar
amounts of HIV
RNA. Two representative subjects are shown with 3 biopsies (Patient #1) or 2
biopsies (Patient #2) taken at 30 cm and run in duplicate. For each subject,
there was
a 0.2 log SD between samples.
Figure 4 shows that HIV RNA can be successfully extracted from
subject's tissue (rectosigmoid biopsies) who have LTNdetectable plasma viral
load
(less than 400 copies/ml) in an assay with a sensitivity of 10 copies per
reaction.
Results can be expressed as actual copy numbers per reaction or standardized
as
number of copies per biopsy.
Figure 5 shows that HIV RNA can be reproducibly quantified from
different biopsies at the same level in the rectosigmoid colon in patients
with
undetectable viral load in the plasma (<400 copies per ml). These results in
15
subjects demonstrate HIV RNA yield over a 3 logo range with a sensitivity of
10
copies per reaction and an inter-sample variation of less than 0.2 loglo
copies per g
tissue RNA. These results confirm there is anatomical variation in tissue HIV
RNA
within the same 30 cm level in the rectosigmoid colon.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-5-
Figure 6 shows that HIV DNA can also be reliably quantitated from tissue
biopsies acquired from patients with undetectable plasma viral load.
Representative
samples from 4 subjects are shown with standard curve for HIV LTR to detect
HIV
DNA by direct PCR with a sensitivity of 10 copies per reaction. Samples were
standardized with an internal control (housekeeping gene), (3-globin. Results
are
reported in both actual copy numbers per reaction and calculated as number of
copies
per 2x106 ~3-globin copies (or 1x106 cells).
Figure 7 is a tabular presentation of results of 19 patients studied with
undetectable plasma viral load (<400 copies per ml) for >3 months who
underwent
tissue acquisition to determine tissue levels of HIV activity as expressed by
both HIV
RNA and HIV DNA. This demonstrates the utility of using a single biopsy to
quantify
both HIV RNA and DNA with the aim to establish an index to guide future
therapy.
There is a 3-logo range in results for HIV RNA. 88% of subjects retain
detectable
levels of tissue RNA despite persistent undetectable levels in plasma. 100% of
subjects have detectable HIV DNA.
Figure 8a and 8b show the ability to reliably extract and quantitate HIV
RNA and HIV DNA from tissue samples from subjects with undetectable plasma
viral
load using the 'ultrasensitive' assay (<40 copies/ml plasma). The results show
that
HIV RNA is detected in 67% of subject's tissue samples when undetectable in
plasma
(8a) and HIV DNA is found in 100% (8b).
Figure 9 shows novel clinical applicability of this invention used in a study
of gene therapy for treating HIV infection. Five subjects with detectable
plasma viral
HIV RNA were enrolled and received protocol-driven re-infusion of their own
apheresed and genetically altered cells. Impact on plasma and tissue viral
burden was
assessed at baseline, days 3, 7, and 14. Results show a trend toward
decreasing tissue
viral burden when no changes were appreciated in plasma. This contributed to
the
design of a mufti-centered trial of gene-therapy for HIV-infected subjects
with
undetectable plasma viral load (<40 copies/ml for >3 months) for which change
in
tissue viral burden using this invention were the primary endpoints.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-6-
Figure 10 a, b and c show novel clinical applicability of this invention to
use tissue samples to quantify viral burden and subsequently determine
patterns of
genotypic and phenotypic viral resistance using the same biopsy samples.
Importantly, because resistance testing requires at least 1.2 kb intact
sequences and
grow the viruses for phenotypic testing, these results demonstrate that the
HIV
detected is replication-competent HIV RNA, not simply short fragments of HIV
RNA. Figure 10a shows the clinical features of the subjects studied including
present
medications to which viral resistance had developed as evidenced by
incompletely
suppressed, break-through HIV RNA in plasma. Tissue HIV RNA levels are
detectable in these subjects as inferred from Figure 4 and 5. Figure l Ob
shows the
patterns of genotypic and phenotypic resistance in contemporaneous plasma and
the
tissue (gut) samples from these subjects with detectable plasma viral load.
There is a
high concordance in sensitivity and resistance profiles to 14 antiviral
medications.
Figure l Oc shows tabular results of drug-resistance patterns in patients with
undetectable plasma viral load. These results are in the process of being
finalized but
demonstrate the ability to use tissue samples to first quantity the amount of
virus and
subsequently determine drug resistance patterns in this population of
patients. This
information will provide logical and scientific guidance for intensifying
therapy to
further decrease total body viral burden in those without detectable plasma
levels.
Figure 11 shows the incorporated ability to trace recoverability of RNA
from the point of acquisition during the patient's procedure. By including an
internal
control separate from the tissue sample, one is able to trace the
loss/recovery ratios
throughout the extraction and amplification process and subsequently correct
detected
amounts for process-related degradation. These results show gel and tabulated
data
from 8 subjects sero-negative for HIV (and therefore without HIV LTR sequence
in
their tissue) with known amounts of HIV LTR added to the frozen sample pre-
assay.
Recovery was nearly 100% regardless of whether the sample was fully or
partially
utilized for RNA quantitation. For HIV-positive subjects, similar experiments
have
been performed using cyclophilin RNA as internal control; a variety of non-
human
RNA sequences can be utilized to provide the same results.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
_7_
Figure 12 shows similar recoverability and capacity to target loss of DNA
during acquisition, extraction and amplification. Results in gel and tabular
form from
2 subjects (2 biopsies each) demonstrate the expected and actual recovery of
an
added, non-human (firefly luciferase DNA) gene product to tissue samples
(rectal
biopsies). Recovery regardless of whether the sample was fully utilized or
divided
(with part later used for RNA quantitation) resulted in >90% recovery. As ~i-
globin is
usually run in parallel, these percentages can be used to quantify yield per
corrected
number of cells.
Figure 13 shows CCRS receptor expression on CD4+ lymphocytes from
blood and from gut mucosa. Flow cytometry scatter plots (A, B) demonstrate
lymphocyte subset analysis to quantify percentages of cells expressing CCRS
and/or
CD4 in a representative subject for blood (A) and gut (B). The number on the
upper
right quadrant of each plot indicates the percentage of CD4+ lymphocytes in
that
subject that expressed CCRS. (C) The individual data points for the six
subjects; data
from the blood and gut of each subject. The gut samples of all six subjects
had a
greater percentage of CCRS+CD4+ cells compared with the blood (P=.03);
differences range from 2.0-5.4-fold.
Figure 14 shows the number of CCRS receptors per cell on CD4+
lymphocytes from blood and from gut mucosa. Flow cytometry histograms (A,B)
demonstrate quantitation of CCRS expression on CD4+ lymphocytes of one of the
six
subjects for blood (A) and gut (B). The rightward shift of mean fluorescence
index in
the CCRS+ CD4+ mucosal cells illustrates the increased numbers of CCRS
receptors
per CD4+ lymphocyte. The number above the bars in A and B indicates the number
of molecules of CCRS expressed per CCRS+ CD4+ lymphocyte in the blood and gut
of that individual. (C) The gut samples of all six subjects had higher
expression of
CCRS compared with the blood (P=.03); differences range from 1.4 to 3.5-fold.
Symbols for each person are the same as those used in Figure 13.
Figure 15 shows the number of picograms of p24 produced by MMC and
PBMC after infection with HIVSX or HIV~,_3. Line graphs indicate the p24
production (picograms of p24 per 104 CD4+ lymphocytes) at 18, 72, and 130
hours
WO 01/00871 CA 02377972 2001-12-24 pCT/[JS99/14366
_g-
after a 3-hour infection with either M-tropic HIVSX (A) or T-tropic
HIV~,,~,4_3 (B). After
72 and 130 hours the supernatants from the MMCs cultured in the presence of 20
IU/mL of IL-2 (~) contained greater concentrations of p24 than the
supernatants from
either PBMC grown with (~) or without (~) 20 IU/mL of IL-2. The greater p24
production from the cultured mucosal cells suggests that they are more
susceptible
than PBMC to replication of M-, or T-tropic HIV-1.
Figure 16 shows the 3 day/IL-2 culture for isolation of mucosal
mononuclear cells yields increased numbers of CD45+,CD3+ CD4+ and CD8+ cells
as compared to conventional collagenase/dispase digestion. The mononuclear
cell
populations isolated by each technique do not appear to differ significantly
in their T
cell subset make-up.
Figure 17 shows that the isolation process does not alter relevant receptor
expression. Flow diagrams of peripheral blood mononuclear cells (PBMC) stained
directly [upper panels] with antibodies to CD4, CDB, CCRS or CXCR4 as
identified
on the horizontal and vertical axes. Lower panels show results of parallel
staining of
the same individual's PBMC following exposure to the isolation process used
for
mucosal mononuclear cells.
Figure 18 shows increased mucosal compared to blood CCRS expression
on CD4+ T cells are detected in normal, inflammatory and HIV-infected samples.
Mean percentages of CD4+ and CCRS+ double-stained cells are shown from
seronegative healthy controls (n=6), inflammatory controls (n=4) and HIV-
infected
(n=8). P values under the subjects category on the x-axis identify the
significance
between blood and gut cells within the clinical group. P values at the top of
the graph
identify significance levels between the CCRS-expressing CD4 T cells in the
mucosal
compartment between clinical groups.
Figure 19 shows that CCRS+ CD4:CD8 ratios in blood and gut decline in
IBD and HIV. The left panel shows relative CCRS expression in blood from
healthy,
sero-negative controls, seronegative inflammatory controls and subjects with
stable
HIV infection. The changing ratios of CCRS-expressing CD4+ T cells to
WO 01/00871 CA 02377972 2001-12-24 PCT/L1S99/14366
-9-
CCRS-expressing CD8+ T cells are boxed underneath. Similar presentations for
mucosal lymphocytes are shown on the right.
DETAILED DESCRIPTION
It must be noted that as used herein and in the appended claims, the
singular forms "a", "and", and "the" include the plural unless the context
clearly
dictates otherwise. Thus, for example, reference to "a cell" includes a
plurality of
such cells.
Unless defined otherwise, all technical and scientific terms used herein
have the same meaning as commonly understood to one of ordinary skill in the
art to
which this invention belongs. Although any methods, devices and materials
similar
or equivalent to those described herein can be used in the practice or testing
of the
invention, the preferred methods, devices and materials are now described.
All publications mentioned herein are incorporated herein by reference in
l5 full for the purpose of describing and disclosing the cell lines,
antibodies, and
methodologies which are described in the publications which might be used in
connection with the presently described invention. The publications discussed
above
and throughout the text are provided solely for their disclosure prior to the
filing date
of the present application. Nothing herein is to be construed as an admission
that the
inventors are not entitled to antedate such disclosure by virtue of prior
invention.
The present invention provides a method for determining the nucleic acid
burden (e.g., the viral burden) in a subject. Such methods detect, for
example, the
presence or absence of viral nucleic acids. For example, quantitiating tissue
levels of
specific RNA and/or DNA sequences that accurately reflect in vivo status. More
particularly, the present invention utilizes both RNA and DNA to detect viral
burden
in tissue compartments. The invention provides a method for determining viral
burden in a subject wherein the subject does not present with plasma viral
load. Such
methods are advantageous in determining the best therapeutic route or
treatment for
modulating or further reducing the viral load in a subject or for confirming
new
CA 02377972 2001-12-24
WO 01/00871 PCT/US99/14366
-10-
infection. Furthermore, the methods of the invention are useful for
determining the
best therapy for affecting viral load in a subject.
The present invention is applicable to the detectin of specific nucleic acid
seequences shown to be associated with hujan pathology including oncogenic or
heritable disorders as well as infections from a variety of organisms,
including, for
example, bacterial infections, viral infections (e.g., retroviral infection
such as HIV
and SIV) as well as other type of infections known to those of skill in the
art. For
example, retroviruses are RNA viruses wherein the viral genome is RNA. When a
host cell is infected with a retrovirus, the genomic RNA is reverse
transcribed into a
DNA intermediate which is integrated very efficiently into the chromosomal DNA
of
infected cells. The integrated DNA intermediate is referred to as a provirus.
The
family Retroviridae are enveloped single-stranded RNA viruses typically infect
mammals, such as, for example, bovines, monkeys, sheep, and humans.
Retroviruses
are unique among RNA viruses in that their multiplication involves the
synthesis of a
DNA copy of the RNA which is then integrated into the genome of the infected
cell.
The Retroviridae family consists of three groups: the spumaviruses (or
foamy viruses) such as the human foamy virus (HFV); the lentiviruses, as well
as
visna virus of sheep; and the oncoviruses (although not all viruses within
this group
are oncogenic). The term "lentivirus" is used in its conventional sense to
describe a
genus of viruses containing reverse transcriptase. The lentiviruses include
the
"immunodeficiency viruses" which include human immunodeficiency virus (HIV)
type 1 and type 2 (HIV-l and HIV-2) and simian immunodeficiency virus (SIV).
In
the absence of effective therapy, most individuals infected with a human
immunodeficiency virus develop acquired immune deficiency syndrome (AIDS) and
succumb to either opportunistic infections and malignancies resulting from
either the
deterioration of the immune system or the direct effects of the virus. The
oncoviruses
are further subdivided into groups A, B, C and D on the basis of particle
morphology,
as seen under the electron microscope during viral maturation. A-type
particles
represent the immature particles of the B- and D-type viruses seen in the
cytoplasm of
infected cells. These particles are not infectious. B-type particles bud as
mature
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-11-
virions from the plasma membrane by the enveloping of intracytoplasmic A-type
particles. At the membrane they possess a toroidal core of ~75 nm, from which
long
glycoprotein spikes project. After budding, B-type particles contain an
eccentrically
located, electron-dense core. The prototype B-type virus is mouse mammary
tumor
virus (MMTV). No intracytoplasmic particles can be observed in cells infected
by C-
type viruses. Instead, mature particles bud directly from the cell surface via
a
crescent 'C'-shaped condensation which then closes on itself and is enclosed
by the
plasma membrane. Envelope glycoprotein spikes may be visible, along with a
uniformly electron-dense core. Budding may occur from the surface plasma
membrane or directly into intracellular vacuoles. The C-type viruses are the
most
commonly studied and include many of the avian and murine leukemia viruses.
Bovine leukemia virus (BLV), and the human T-cell leukemia viruses types I and
II
(HTLV-I/II) are similarly classified as C-type particles because of the
morphology of
their budding from the cell surface. However, they also have a regular
hexagonal
morphology and more complex genome structures than the prototypic C-type
viruses
such as the murine leukemia viruses (MLV). D-type particles resemble B-type
particles in that they show as ring-like structures in the infected cell
cytoplasm, which
bud from the cell surface, but the virions incorporate short surface
glycoprotein
spikes. The electron-dense cores are also eccentrically located within the
particles.
Mason Pfizer monkey virus (MPMV) is the prototype D-type virus.
Retroviruses are defined by the way in which they replicate their genetic
material. During replication the RNA is converted into DNA. Following
infection of
the cell a double- stranded molecule of DNA is generated from the two
molecules of
RNA which are carried in the viral particle by the molecular process known as
reverse
transcription. The DNA form becomes covalently integrated in the host cell
genome
as a provirus, from which viral RNAs are expressed with the aid of cellular
and/or
viral factors. The expressed viral RNAs are packaged into particles and
released as
infectious virions.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-12-
The retrovirus particle is composed of two identical RNA molecules. Each
genome is a positive sense, single-stranded RNA molecule, which is capped at
the 5
end and polyadenylated at the 3' tail. The diploid virus particle contains the
two RNA
strands complexed with gag proteins, viral enzymes (pol gene products) and
host
tRNA molecules within a ' core' structure of gag proteins. Surrounding and
protecting
this capsid is a lipid bilayer, derived from host cell membranes and
containing viral
envelope proteins. The env proteins bind to the cellular receptor for the
virus and the
particle typically enters the host cell via receptor-mediated endocytosis
and/or
membrane fusion.
After the outer envelope is shed, the viral RNA is copied into DNA by
reverse transcription. This is catalyzed by the reverse transcriptase enzyme
encoded
by the pol region and uses the host cell tRNA packaged into the virion as a
primer for
DNA synthesis. In this way the RNA genome is converted into the more complex
DNA genome.
The double-stranded linear DNA produced by reverse transcription may,
or may not, have to be circularized in the nucleus before integration into the
host cell
genome. The provirus now has two identical repeats at either end, known as the
long
terminal repeats (LTR). The junction between the two joined LTR sequences
produces the site recognized by a pol product - the integrase protein - which
catalyzes
integration, such that the provirus is always joined to host DNA two base
pairs (bp)
from the ends of the LTRs. A duplication of cellular sequences is seen at the
ends of
both LTRs, reminiscent of the integration pattern of transposable genetic
elements.
Integration is thought to occur essentially at random within the target cell
genome.
Transcription, RNA splicing and translation of the integrated viral DNA is
mediated by host cell proteins. Variously spliced transcripts are generated.
In the case
of the human retroviruses HIV-1/2 and HTLV-I/II viral proteins are also used
to
regulate gene expression. The interplay between cellular and viral factors is
important
in the control of virus latency and the temporal sequence in which viral genes
are
expressed.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-13-
Retroviruses can be transmitted horizontally and vertically. Efficient
infectious transmission of retroviruses requires the expression on the target
cell of
receptors which specifically recognize the viral envelope proteins, although
viruses
may use receptor-independent, nonspecific routes of entry at low efficiency.
In
addition, the target cell type must be able to support all stages of the
replication cycle
after virus has bound and penetrated. Vertical transmission occurs when the
viral
genome becomes integrated in the germ line of the host. The provirus will then
be
passed from generation to generation as though it were a cellular gene. Hence
endogenous proviruses become established which frequently lie latent, but
which can
become activated when the host is exposed to appropriate agents.
The oncoviruses (often called the RNA tumor viruses) have been
subdivided into two groups of pathogens, namely the acutely transforming and
slow
transforming retroviruses.
Acutely transforming retroviruses can transform cultured cells and can
cause disease rapidly in susceptible animals. These viruses usually carry an
oncogene
(v-onc) within the viral genome, which is directly responsible for their
tumorigenicity, and which is different in each type of virus. The viral
oncogenes have
been derived from cellular genes that the viruses have acquired, probably as a
result
of the inclusion of cellular RNA within a viral particle. Subsequent
recombination
between viral and cellular RNA during reverse transcription leads to the
incorporation
of the cellular sequences into the viral genome and delivery of this novel
unit into the
host cell DNA. If the transduced gene normally has a central role in control
of cellular
growth and differentiation, the changes in coding sequence and/or control of
expression that it undergoes on incorporation into the viral genome can render
it
oncogenic. Such cellular proto-oncogenes (c-onc) may become oncogenic by being
placed under novel, virally determined transcriptional control (both
quantitatively and
temporally), and/or by sustaining critical mutations to the coding sequence.
However,
full cellular transformation usually requires the expression of v-one in
conjunction
with other genetic and epigenetic changes within the target cell.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-14-
The slow transforming retroviruses typically do not contain a 'classical'
oncogene. The mechanism of transformation is believed rather to involve the
insertion
of provirus near, or in, the coding region of a cellular proto-oncogene,
called
insertional mutagenesis. The strong promoter and enhancer sequences within the
viral
LTRs can exert transcriptional effects from distances of up to several
kilobase pairs
from the proto-oncogene. The normal regulation of expression of the cellular
gene is
disrupted, and over-expression or inappropriately timed expression can
contribute to
transformation.
HTLV-I is a slow-transforming virus, causally associated with adult T-cell
leukemia (ATL), but it probably promotes T-cell transformation by a different
pathway involving virally encoded regulatory proteins, especially p40tax,
which
transactivate expression of cellular proto-oncogenes. HIV-l and 2 have also
been
implicated in both the direct and indirect promotion of various types of
malignancy
(such as Kaposi's sarcoma) which present much more frequently in AIDS patients
than in the general population. However, the direct role of HIV in malignant
transformation remains doubtful as many patients who are immunosuppressed as a
result of other infections or treatments (e.g. transplant recipients) also
develop
tumours at increased rates.
The D-type viruses are not aetiologically associated with malignancy,
although MPMV was initially associated with a mammary tumor in a rhesus
monkey.
D-type viruses cause immune suppression in simian primates but by an unknown
mechanism. Immune suppression is also a feature of infection by the
lentiviruses (e.g.
HIV and SIV) and variant strains of feline leukemia virus (FeLV). In infection
with
HIV and FeLV large amounts of unintegrated proviral DNA have been observed,
which may be related to the pathogenesis.
The lentiviruses, including HIV-1/2 and visna virus of sheep, are
associated with slow progressive disease leading to immune suppression and
neurological disorders. HIV is the widely recognized causative agent of the
acquired
immunodeficiency disease syndrome (AIDS). The pharmaceutical compositions and
methods of the invention, including antibodies, peptides, peptidomimetics,
chemical
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-15-
compositions, etc., are all useful for treating subjects either having or at
risk of having
an immunodeficiency virus (e.g., HIV) related disorder. AIDS and ARC are
preferred
examples of such disorders. HIV-associated disorders have been recognized
primarily in "at risk" groups, including homosexually active males,
intravenous drug
users, recipients of blood or blood products, and certain populations from
Central
Africa and the Caribbean. The syndrome has also been recognized in
heterosexual
partners of individuals in all "at risk" groups and in infants of affected
mothers.
Retroviruses have been linked to a wide range of diseases, including
anaemia, neurological disorders, immune suppression, and malignancy. HTLV-I,
for
example, is associated with tropical spastic paraparesis, a condition similar
in some
respects to multiple sclerosis.
As used herein, "tissue" means any tissue in which pathologic changes,
such as with an infection can exist. The term tissue also encompasses the
cells of a
tissue, for example, a mucosal tissue is composed of mucosal cells. Such
tissues and
1 S cells include, for example, circulating, isolated and/or cultured cells,
gastro-intestinal
tissues (e.g., the small intestine, the large intestine, the rectum), uro-
genital tissue
(e.g., vaginal tissue, penile tissue, urethra), nasal-larynx tissue (e.g.,
nasal tissue,
larynx tissue), liver tissue and hepatic cells, and skin (including
keratinocytes,
fibroblasts) to name a few. Other tissues, including mucosal tissues, are
known and
easily identifiable by one of skill in the art.
The inventors have discovered that viral infections (e.g.,
immunodeficiency viral infection such as HIV infections) are an inflammatory
condition present in the tissue of a subject (e.g., mucosal tissue). Most
reports have
emphasized a state of lymphopenia or "anti-inflammation" in the mucosa
paralleling
that seen progressively in the blood of infected subjects. Because mucosal
tissue is
populated by an increased number of activated, memory, co-receptor expressing
CD4+ T cells in healthy uninfected individuals, the vulnerability to infection
is high.
In a typical response to infection, mucosal immune cells (most likely CD8+ T
lymphocytes and macrophages) secrete increased levels of pro-inflammatory
chemokines and cytokines with the intent of recruiting additional T-
lymphocytes to
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-16-
the mucosal site. This heightened response or ''inflammatory response",
although
instigated with the intent of limiting infection, serves to provide
significantly
increased numbers of potential new targets for infection, favoring spread of
the virus
and maintaining the viral burden in these tissue compartments.
For example, in SIV-infected macaques, the gastrointestinal tract is the
major site of early CD4+ lymphocyte depletion and viral replication to such an
extent
that it has been suggested that SIV infection may primarily be a disease of
the
mucosal immune system (Veazey et al., Science 280:427 (1998); MacDonald and
Spencer, Gastrointestinal and Hepatic Immunology, R.H. Heatley, Ed. (Cambridge
University Press, 1994). In humans, HIV infection also involves the mucosal
immune
system and infectious viral particles have been recovered directly from
mucosal
samples and in situ studies have demonstrated that lamina propria T
lymphocytes are
among the first cells that are infected (Koteler et al., Am. J. Pathol.
139:823 (1991);
Heise et al., J. Infect. Dis. 169:1116 (1994); Heise et al., Am. J. Pathol.
142:1759
(1993); Smit-McBride et al., J. Virol. 72:6646 (1998); Clayton et al.,
Gastroenterology 103:919 (1992); Ellakay et al., Am. J. Clin. Pathol., 87:356
(1987);
Jarry et al., Histopathology 16:133 (1990); Lacner et al., Am. J. Pathol.
153:481
(1998). Moreover, the mucosal lining of the rectosigmoid colon is a primary
site for
viral introduction during anal-insertive intercourse (Patterson et al., Am. J.
Pathol.
153:481 (1998)).
The intestine, even in healthy HIV-uninfected patients, maintains a state of
low-level physiologic inflammation that is necessary to protect the interior
milieu
from the bath of potential pathogens that contact its surface. The great
majority of
lymphocytes and macrophages that compose this infiltrate are therefore
stimulated or
activated. The primary target of HIV, in which the virus most effectively
replicates, is
the stimulated CD4+ cell. This type of cell fills the gastrointestinal mucosa.
Increased
viral replication results in greater spread of HIV throughout the mucosa and
higher
mucosal HIV viral loads. The predominant aim of anti-HN therapy is to decrease
the
ability of HIV to replicate and therefore spread amongst CD4+ cells which are
eventually destroyed by the virus.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-17-
The gastrointestinal mucosa is one element of this lymphoid tissue and
increasing evidence suggests that HIV involves the mucosa at all stages of
disease.
Not only is the gastrointestinal tract the route of transmission for the
majority of
patients, but it is the largest lymphoid organ (e.g., a gut-associated
lymphoid tissue).
As mentioned above, the gastrointestinal mucosa is characterized by a state of
low-level physiologic inflammation, and the majority of its lymphocytes are
activated. The naturally high concentration of pro-inflammatory cytokines that
are
present in the mucosa appear to enhance HIV replication in this site,
resulting in a
high mucosal HIV viral load and successive rounds of infection of new target
gastrointestinal CD4+ cells, regardless of the route of infection.
The inventors have found that the majority of gastrointestinal CD4+ T
cells express the chemokine receptors that are necessary for HIV entry. The
vast
majority of lymphocytes of the gastrointestinal mucosa express both CCRS and
CXCR4. In the case of CCRS, it also appears that the mucosal mononuclear cells
1 S (MMCs) express higher levels of this receptor than blood derived monocytes
on a per
cell basis. The inventors have found that the mucosal cells are more
susceptible to
HN than are peripheral blood cells in vitro. Accordingly, the gastrointestinal
tissue
harbors a majority of the virus. During infection, active virus replicates and
is shed
by infected cells into the surrounding cellular milleu and is eventually
transported
into the blood stream. Current techniques measure the presence of the virus
through
assays on a subject's plasma. Thus, therapeutic treatment of a subject may
reduce the
amount of replication or activiation of the virus resulting in a concommitant
reduction
in the amount of viral burden in the plasma. However, this reduction is not
indicative
of the whole body viral burden as a majority of the virus exists in the tissue
compartments of the subj ect.
HIV nucleic acids can be found in the mucosa of the majority of
HIV-infected individuals; Kotler et al. detected HIV DNA by PCR using gag-
specific
primers in 70% of 20 patients he investigated. The inventors have found that
even
patients with undetectable plasma viral loads have replicating virus in their
mucosa. A
high SIV viral load is seen in the gastrointestinal mucosa whether the macaque
is
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-18-
infected via the gut or via the parenteral route suggesting that the mucosa
has a high
intrinsic susceptibility to SIV or HIV. After infection, these macaques
exhibit a
profound early (within 7 to 21 days) loss of gastrointestinal mucosal CD4+
cells. This
sign of vigorous HIV activity was not mirrored in other lymphoid sites. In
humans,
mucosal CD4+ cell depletion has been described in the colon and duodenum
during
both the early asymptomatic phase of chronic infection and after the onset of
clinical
AIDS.
Highly, active antiretroviral therapy (HAART) has the potential to drive
plasma viral load (PVL) to below detectable levels in some patients as defined
by
current levels of detection, as well as future definitions of plasma
detectability. Yet
substantial amounts of HIV remain in these patient's tissues and this viral
population
eventually undercuts the effectiveness of HAART. As a result, it is widely
acknowledged that HAART as it is currently practiced cannot cure AIDS.
The limitations of plasma viral load as discussed above, have two
ramifications. First advances in the treatment of HIV disease will require
assays that
reflect viral parameters in tissue compartments. The lowest limits of
detection in
PVL tests are attainable by current treatment strategies, although these
strategies fail
to suppress HIV completely. Second, since many of the processes underlying the
disease occur in tissue compartments, an understanding of the disease will
require the
ability to access viral dynamics in these compartments. Accordingly, the
present
invention provides methods and compositions useful in assaying the viral loads
in
these compartments. When used in combination with therapeutic agents such
methods and compositions are hopeful in further affecting the control of HIV
replication, activation and ultimately reduce the incidence of AIDS.
Since the limits of PVL tests are reached by treatments that do not cure the
disease, these tests are unable to determine the effect therapeutics might
have on the
viral load in various tissue compartments outside of the plasma and blood.
Accordingly, the present invention provides a method for assaying viral
load in these tissues in order to better quantitate the viral load (e.g.,
whole body viral
load) as well as the effect of therapeutic treatments in reducing viral load
in tissues
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-19-
outside of the blood and plasma as well as viral load in the whole body during
therapeutic treatment. Such methods include the potential to determine whole-
body
viral burden or the total amount of viral load (e.g., HIV viral load) in a
subject. By
"subject" is meant any mammal, preferably a human. Such determination will
utilize
measurements from various samples (e.g., tissue or cells) obtained from a
subject.
For example. tissue, cell or other sources useful in the present invention
include for
example, circulating, isolated and/or cultured cells, gastro-intestinal
tissues (e.g., the
small intestine, the large intestine, the rectum), uro-genital tissue (e.g.,
vaginal tissue,
penile tissue, urethra), nasal-larynx tissue (e.g., nasal tissue, larynx
tissue), liver tissue
and hepatic cells, and skin (including keratinocytes, fibroblasts) to name a
few. The
tissue will then be assayed to determine the viral load based upon the
presence of
viral proteins or nucleic acids (e.g., DNA and RNA).
While highly active anti-retroviral therapy (HAART) may reduce viral
load to undetectable levels in the plasma, the vast majority of patients will
suffer a
rebound increase in plasma viremia when therapy is halted, suggesting that
reservoirs
of virus exist beyond the bloodstream. One viral reservoir has been discovered
in
lymphoid tissue where the majority of the body's lymphocytes reside (98%).
Since the
gastrointestinal mucosa contains the majority of the body's lymphocytes (40-
65%), it
likely represents the largest reservoir of HIV, and should be considered a
primary
target for anti-HIV therapy.
The use of HAART by the majority of HIV-infected patients has resulted
in prolonged life expectancy of HIV-infected patients. Unfortunately, despite
a
reduction of plasma load to undetectable levels, multiple studies have shown
continued HIV replication in lymphoid organs. Studies have shown that 88% of
patients with undetectable plasma viral load have quantifiable HIV nucleic
acid in
their gastrointestinal mucosa. For example, mucosal biopsies can be used for
mucosal
mononuclear cell (MMC) isolation, Rnase protection assay (RPA), PCR for HIV
RNA and proviral DNA as well as quantitative image analysis (QIA).
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-20-
Also provided are methods for determining the effect of a therapeutic on
whole body-viral load. For example, subjects with plasma detectable or
undetectable
HIV, having undergone baseline biopsies are treated with a therapy. At various
times
during the therapy, the subjects will undergo repeat endoscopic biopsies and
repeat
phlebotomy to obtain PBMCs and biomolecular experiments performed to
quantitiate
various parameters in the sample, including the presence or absence of
particular
nucleic acids (e.g., viral DNA or RNA), particular proteins, polypeptides, or
antigens
(e.g., env protein of HIV) as well as others commonly used to determine the
presence
or absence of various viral markers. In addition, and as an example, one
biopsy at
each time point will be utilized for analysis of the patients tissue viral
load and
plasma obtained by phlebotomy at each time point will be analyzed by the Roche
ultrasensitive assay (detection level(40 copies of HIV RNA). RT- and DNA-PCR
are
performed to analyze changes between baseline and post-therapy tissue and
plasma
viral load. For statistical evaluation, the solitary index of efficacy will be
a decrease
in tissue viral load without a concomitant increase in plasma viral load.
Accordingly, the present invention provides a method for assaying viral
load in these tissues in order to better quantitate the total body viral load
as well as the
effect of therpaeutic treatments in reducing viral load in tissues outside the
blood and
plasma. Such methods include determining whole-body viral burden or the total
amount of HIV in a subject. By "subject" is meant any mammal, preferably a
human.
Such determination will utilize measurements from various tissue sources in a
subject.
For example, tissue sources useful in the present invention include gastro-
intestinal
tissues (e.g., the small intestine, the large intestine, the rectum), uro-
genital tissue
(e.g., vaginal tissue, penile tissue, urethra), nasal-larynx tissue (e.g.,
nasal tissue,
larynx tissue) to name a few. The tissue will then be assayed to determine the
viral
load based upon the presence of viral proteins or nucleic acids (e.g., DNA and
RNA).
Thus, in one embodiment, the present invention provides a method for
determining the viral load in a sample obtained from a subject. The sample may
be
obtained by any number of methods including, for example, endoscopically or
percutaneously, as well as other methods known in the art. In an endoscopic
WO 01/00871 CA 02377972 2001-12-24 PCT/CTS99/14366
-21 -
embodiment a flexible tube (e.g., an endoscope) is inserted into an orifice,
such as the
rectum, which allows a technician, physician or nurse to examine the rectum
and use
a forceps to obtain a biopsy sample. The biopsy sample is typically taken at a
distance of about 30 cm and is about 8 mm in diameter. The tissue biopsy is
then
processed, using standard techniques known to those of skill in the art, to
extract
nucleic acids (e.g., RNA and DNA). The RNA is converted to DNA by reverse
transcriptase and amplified by PCR in order to quantify the presence of a
virus (e.g.,
HIV by detecting HIV RNA) in the sample. In another embodiment, the quantity
of
HIV nucleic acids in the tissue sample are compared to a standard sample or
directly
amplifed by PCR to quantify DNA. Alternatively, the quantity of viral nucleic
acids
in the tissue can be compared to the amount of viral nucleic acids in a
corresponding
plasma sample obtained from the same subject. This later embodiment allows for
the
determination of the effect of antiviral treatment on tissue viral burden,
blood and
rectal biopsy specimens in order to determine the effect of therapy on the
various viral
loads in each sample. An additional embodiment is the quantitiative monitoring
using
internal controls of nucleic acid decay in each sample from the time of
acquisition
through the assay in order to ensure that detected levels accurately correlate
with the
in vivo status of the sample
In another embodiment the invention provides a kit for obtaining samples
from a subject comprising a collection device, such as a forceps or scapel, a
cell lysis
buffer suitable for preserving nucleic acids in the biological sample, and a
spiking
agent, such as a nucleic acid (e.g., RNA or DNA). By "spiking agent" is meant
a
DNA or RNA sample of known amount that degrades at a known rate, such that one
skilled in the art can determine the rate of change in any particular sample
or in any
particular molecule within a sample. Such a kit may also include a carrier
means
being compartmentalized to receive in close confinement one or more containers
such
as vials, tubes, and the like, each of the containers comprising one of the
separate
elements to be used in the method. If present, a second container may comprise
a
lysis buffer. The kit may also have containers containing nucleotides for
amplification of or hybridization to a specific nucleic acid sequence which
may or
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-22-
may not be labeled, or a container comprising a reporter, such as a biotin-
binding
protein, such as avidin or streptavidin, bound to a reporter molecule, such as
an
enzymatic, florescent, or radionuclide label.
EXAMPLES
Tissue-based immune cells can be easily and safely obtained and isolated
from the mucosal lining of the gut, a renewable tissue source. As studies of
HIV-1
pathogenesis increasingly focus on tissue compartments for both persistence
and
transmission studies, techniques to easily and safely access lymphoid tissue
are
essential. Lymph node and tonsillar resection/aspirations have been the most
commonly reported methods. Studies using these approaches have already
provided
illuminating, concept-changing findings including evidence that HIV-1 activity
persists in lymphoid tissue when plasma levels are stable or undetectable.
These
tissue sources reveal the biologic events that occur in secondary, organized
lymphoid
structures during HIV-1 infection, but require invasive surgical support for
tissue
acquisition. In contrast, the gut mucosal lymphoid tissue is abundant, easily
accessible, quickly healing, self replenishing, and directly visible.
Endoscopic
biopsies are safe, quick, painless, and provide access to the lymphoid
compartment
with 100% of samples containing lymphocytes. The biopsies maintain
architectural
orientation and can be examined histologically, or can be dissociated for flow
cytometric and tissue culture evaluation as described here.
Several examples have been presented in Figure form to demonstrate
representative clinical utilities of this invention. The invention has been
used to
quantify tissue viral burden (RNA and DNA) in (i) a study of 20 subjects with
persistently suppressed plasma viral burden (<400 copies/ml plasma) every
three
months for 1 year; (ii) a study of 5 subjects with detectable plasma viral HIV
RNA
undergoing a trial of gene therapy in which tissue sampling was utilized to
quantitate
changes in tissue viral burden; (iii) a study of 40 subjects with undetectable
plasma
viral burden (<40 copies per ml plasma) undergoing gene therapy for their HIV
infection in which gene therapy was used to detect changes in tissue HIV RNA
and
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
- 23 -
DNA as primary endpoints; (iv) a group of 8 subjects with detectable plasma
viral
HIV RNA to determine patterns of tissue genotypic and phenotypic resistance
based
on tissue-extracted nucleic acids; (v) a group of 10 subjects with
undetectable (<40
copies per ml plasma) to determine tissue levels of HIV RNA and HIV DNA as
well
as tissue patterns of genotypic and phenotypic resistance.
For all studies, informed consent (approved by the UCLA IRB) was
obtained prior to enrollment. Routine flexible sigmoidoscopy was performed and
a
standardized site of 30 cm in the rectosigmoid colon was routinely used for
all
sampling to avoid potentially confounding inflammation resulting from
traumatic or
infectious proctitis. Biopsies were collected using 3.3 OD forceps (8mm open
span).
For nucleic extraction studies, samples are removed from the forceps in to pre-
labeled
cryovials and immediately suspended in liquid nitrogen (<15 seconds from
acquisition to freezing) for transport to the laboratory. At that point,
samples are
transferred into -80°C freezer and data entered into a 4-dimensional
relational
database.
For biopsies intended for isolation assays to obtain mononuclear
suspensions of mucosal cells, acquired biopsies are maintained at room
temperature.
Informed consent was obtained prior to undergoing elective endoscopy for a
history
of blood in stool or routine polyp screening. No subjects had diarrhea
symptoms or
history of intestinal inflammatory or infectious disorders. Hematoxylin and
eosin
stained biopsies taken in the same area as study biopsies revealed no
pathology and
were all normal appearing when reviewed in a blinded fashion by a
gastrointestinal
surgical pathologist. The study was approved by the UCLA Human Subjects
Protection Committee. A site of 30 cm in the rectosigmoid colon was used
routinely
for all sampling to avoid potentially confounding inflammation resulting from
traumatic or infectious proctitis. Mucosal mononuclear cells (MMC) were
isolated
from four endoscopic biopsies from each donor. Biopsies were collected using
3.3
OD forceps into 15m1 of tissue culture medium (RPMI 1640, Irvine Scientific).
The
biopsies were maintained at room temperature on a rotating platform until
isolation
(roughly 20-60 minutes) then removed to a 1Ox35mm petri-dish containing
phosphate
WO 01/00871 CA 02377972 2001-12-24 pCT~S99/14366
-24-
buffered saline (PBS) with 1mM EDTA and SOmM 2- mercaptoethanol and the
samples teased apart using 18G needles. The disrupted tissue was incubated at
37°C
for 20 minutes in a shaking water bath. Following centrifugation, the tissue
samples
were digested with a mixture of collagenase and dispase (Boehringer Mannheim #
269638; 0.1 mg/mL in RPMI) for 1 hour at 37°C. Further disruption was
achieved by
sample passage through syringes with a series of decreasing needle gauges.
Debris
was removed using a 70 micron cell strainer (Falcon # 2350). Resulting cells
were
resuspended in RPMI containing 10% fetal calf serum. Mononuclear cells, which
included primarily epithelial cells and leukocytes, were counted visually
using a
hemocytometer and the proportion of mononuclear cells that were leukocytes was
estimated. About 20% of the mononuclear cells were leukocytes from a yield of
mean 1.3 x 106 ~ 1.1 x 106 S.D. (n = 6) per four biopsies. Viability,
determined by
the exclusion of trypan blue, was > 90%. Blood from the donors was collected
in
EDTA and was stained using the whole blood staining method.
Among the multiple co-receptors that HIV-1 is able to utilize, CCRS and
CXCR4 play a major role, with CCRS-tropic viruses predominating during initial
infection, CXCR4-tropic viruses becoming more prevalent with advanced disease
and
heterozygosity of CCRS contributing to longer survival. Differential
expression of
CCRS on mucosal CD4+ T lymphocytes could contribute to preferential
transmission
of M-tropic viruses. In order to compare co-receptor expression on mucosal
versus
circulating lymphocytes, mucosal mononuclear cells (MMC) were isolated from
rectosigmoid endoscopic biopsies and obtained unstimulated phlebotomy samples
from HIV-1 seronegative healthy individuals. We quantified co-receptor
expression
on CD4+ cells by flow cytometry.
In agreement with published studies, a median of 23% (interquartile [i.q.]
range 18-30%) of all CD4+ lymphocytes in blood expressed CCRS. As shown in
Figure 15, a median of 71 % (i.q. range 50-87%) of the CD4+ lymphocytes in the
gut
expressed CCRS, a 2.8-fold greater percentage than in the blood (P=0.03).
Mucosal
CD4+ lymphocytes also expressed significantly more CCRS receptors per cell
than
did their CCRS-expressing CD4+ lymphocyte blood counterparts, further
extending
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
- 25 -
the compartmental difference. As shown in Figure 16, the median CCRS receptor
number per CD4+ mucosal lymphocyte was 6,946 molecules (i.q. range
6,306-10,416) compared to approximately 3,841 (i.q. range 3,259-4,441) CCRS
receptors per CD4+ blood lymphocyte, a 2.2-fold increase (P=.03). Taken
together,
this translates into a 6.2-fold increase in total expressed CCRS receptors
potentially
available for viral access on CD4+ lymphocytes in the gut compared to the
blood.
These findings suggest that mucosal CD4+ lymphocytes may be much more
vulnerable to infection by M-tropic HIV-1 than their blood counterparts.
Nearly all (97%) of the CCRS expression on CD4+ lymphocytes in both
the blood and gut was on cells of the memory CD45R0+ phenotype. In agreement
with published studies, we found an increased proportion of CD45R0+ memory
cells
among gut CD4+ lymphocytes (median 95%; i.q. range 90-97%) compared to blood
(median 46%; i.q. range 38-53%). Our findings are based on isolated viable
cells
using a method found to preserves cell surface expression of CCRS and CXCR4.
In order to evaluate the susceptibility of mucosal mononuclear cells to
HIV-1 infection, we subjected isolated MMC from endoscopic biopsies and
isolated
PBMCs from healthy HIV-seronegative volunteers to in vitro HIV infection.
Infection of mucosal cells with laboratory strains of HIV (M-tropic HIVSX or
T-tropic HIVNL4-3) was performed in the presence of 20 IU of interleukin-2 (IL-
2),
as data had shown that IL-2 was required to maintain viability of mucosal cell
populations. As a control, since IL-2 is known to upregulate CCRS and could
enhance viral replication, PBMCs were also infected from the same patient both
with
and without IL-2. Infection was quantified at 18 hours, 72 hours and 130 hours
by
p24 production in the supernatant and expressed in terms of pg of p24 produced
per
104 CD4+ lymphocytes. Mucosal mononuclear cells were able to support vigorous
viral replication in culture compared to PBMC with or without IL-2 as shown in
Figure 17 in a representative experiment (one of two). PBMC infected in the
absence
of IL-2 could not support HIV replication by either HIVSX or HIVNL4-3. When
compared with similarly cultured PBMCs in the presence of IL-2, mucosal cells
were
markedly more susceptible than PBMC to M-tropic and T-tropic HIV. For example,
WO 01/00871 CA 02377972 2001-12-24 pCT~S99/14366
-26-
at 72 hours, supernatant p24 levels of the M-tropic HIVSX in the MMC culture
was
164 pg/ml per 104 CD4+ lymphocytes compared with 51 pg/ml in the PBMC culture.
For T-tropic HIVNL4-3, viral growth accelerated over time and at 130 hours,
supernatant p24 levels in the MMC cultures was 1194 pg/ml per 104 CD4+
lymphocytes compared to undetectable levels in cultures of PBMC. These data
indicate that the enhanced vulnerability to infection suggested by the mucosal
CD4+
lymphocyte co-receptor phenotype renders them functionally infectable in vitro
by
both M and T-tropic strains of HIV-1.
These studies is that CCRS expression is markedly increased on human
mucosal CD4+ lymphocytes, both as a percentage of total CD4+ lymphocytes and
on
a per cell basis compared to peripheral blood cells. These mucosal T cells
support
higher levels of viral replication than CD4+ lymphocytes from blood (Figure
17). As
CCRS is the co-receptor most associated with HIV-1 in early infection, and the
gastrointestinal tract is one of the most common sites of transmission as well
as being
the body's major lymphoid organ, the amount of detected CCRS expression
carries
important implications for transmission, primary infection, ongoing local
spread and
treatment. Projections of T cell infectability by M-tropic HIV-1, based on
CCRS
expression in blood, would lead to a dramatically different mathematical
modeling of
disease progression. When CD4 was not a limiting factor, a minimum of 700-2000
CCRS receptors per cell was adequate for maximal susceptibility to infection.
By our
calculations, the number of CCRS receptors on blood T cells (median of
approximately 3000 per cell) would be within this range of in vitro
infectability. The
mucosal levels (median of approximately 7000 receptors per cell) far exceed
this
minimal range. Factors including b-chemokine production levels and presence of
cellular activation factors in the MMC could also impact the capacity of these
cells to
support replication. Cytokines including TNF-a and IL-2 are found at high
levels in
gut mucosa and may also contribute to enhanced viral replication at this site.
Our results show that tissue biopsies of gut mucosa can be used to obtain
quantitative information on immunologic and virologic determinants that may
influence HIV-1 transmission and pathogenesis. The potential vulnerability to
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
- 27 -
primary and persistent HIV-1 infection of the gastrointestinal mucosa, a
sexually-exposed and easily traumatized lining, is dramatic. This
vulnerability
includes the predominance of activated memory CD4+ lymphocytes at this mucosal
site. Our results also show that both CCRS and CXCR4 are highly expressed on
mucosal CD4+ lymphocytes from the gastrointestinal tract in healthy,
HIV-1-seronegative individuals, and these mucosal cells are highly susceptible
to
infection in vitro, much more so than cells from the blood.
To ensure that the collagenase/dispase isolation process used on the gut
biopsies did not degrade nor strip surface antigens, peripheral blood
mononuclear
cells (PBMC) were isolated by Ficoll-Hypaque separation and then either
stained and
analyzed directly by flow cytometry for percent CD45, CD4, CDB, CCRS and
CXCR4 expression or processed through the mucosal isolation procedure
(collagenase/dispase treatment) and then stained and analyzed for the
antigens.
PBMC routinely isolated and those exposed to mucosal isolation enzymes showed
no
discernible differences in quadrant percentages for all antibodies studied.
Thus, the
observation of increased percentages and expression of CCRS on cells from the
gut
compared to those from the blood does not result from the isolation process
since
CCRS expression was not increased by treatment with collagenase/dispase.
MMCs from each healthy, HIV-seronegative volunteer were isolated from
four endoscopic mucosal biopsies after mechanical disruption followed by 3-day
culture in Iscove's DMEM medium supplemented with 10% human serum, containing
10 mg/ml gentamycin, penicillin, streptomycin and glutamine. Interleukin-2 (IL-
2,
Amgen) was added at 20 IU per mL. A total of 105 mucosal mononuclear cells and
PBMCs were plated in a 96-well plate in 100 microliters of medium after a 3-
hour
infection with 50 mg of HIVSX or HIVNL4-3. Prior to plating, the cells were
washed
twice to remove free virus and adherent p24. Thirty microliters of supernatant
was
sampled at each time point 18 hours, 3 days (72 hours), and 5 days (130 hours)
for
p24 measurement by ELISA (Coulter). CD4+ percentages were determined by flow
cytometry and were used to determine the number of CD4+ lymphocytes in the
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-28-
cultures. Co-receptor expression on the gut cells was not tested on these
donors
because cells yields were not sufficient.
In an attempt to increase the yield of mucosal mononuclear cells for
functional, infectivity and flow studies, an alternative isolation method was
undertaken. Freshly collected endoscopic biopsies are minced directly into l
Oml of
Iscoves Medium supplemented with 20 units/ml of IL-2 in a 100x300mm petri-dish
and cultured for 3 days in 5%C02 at 37 degrees. Cells are harvested through a
70qm
cell strainer and the total mononuclear cell yield enumerated visually by
hemocytometer. The yield of CD45+, CD3+, CD4+ and CD8+ cells was determined
using TruCount beads and was compared to the yield from biopsies collected
from the
same individual isolated using the conventional collagenase/dispase protocol.
There is
a 6-fold increase in the yield of mucosal lymphocytes. (Figure 18).
Quantitation of HIV-1 in tissue.
Extraction of RNA from rectosigmoid biopsies results in >95% recovery
of tissue RNA.
We have developed a quantitative RT PCR assay for tissue RNA and
quantitative PCR for tissue DNA, adapted from that previously described by the
Chen
laboratory. Our initial results show that HIV-1 RNA can be quantitated by RT
PCR to
levels as low as 10 copies (Figure 1).
Samples are immediately homogenized (using Powergen 125 tissue
homogenizer)or pulverized using a mortar and pestle from the frozen state,
Trizol-extracted with separation of RNA and DNA containing phases. RNA is
further
extracted using an Rneasy column. Quality control studies have confirmed
minimal
RNA degradation (agarose gel electrophoresis) and no DNA contamination (PCR of
RNA template). The number of HIV RNA copies is quantitated using an adaptation
of
the rTTH RNA PCR kit (Perkin-Elmer) with HIV LTR specific primers 667/AA55
designed to capture unspliced/multiply spliced HIV RNA. A linear standard
curve is
generated using a 127 by sequence recognized by the 667/AA55 primer pairs. DNA
is
isolated by ethanol precipitation with at least 2 washes in O.1M sodium
citrate/10%
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-29-
ethanol buffer. For HIV DNA, the same primer pairs are used for PCR
amplification
(667/AA55). Linear standard curves have been generated using ~3-globin
primers.
In an effort to standardize our approach and have resultant yields most
closely reflect in vivo amounts of HIV RNA, known quantities of HIV LTR RNA
were added to sero-negative biopsies both pre and post nucleic acid isolation.
A linear
standard curve was generated using purified HIV LTR RNA diluted in O.Sug/ul of
Hela-cell total RNA and RT-PCR was performed as previously described. The
difference between the pre and post LTR- supplemented samples was quantified
and
found to be minimal (>95% recovery).
As DNA and RNA are extracted from the sample (and same total sampel
voluem) and equally divided with internal tracer controls (e.g., luciferase
DNA and
cylcophilin RNA), the amount of cells in the DNA sampel can be determined
using
the housekeeping gene (3-globin, corrected for loss (via luciferase tracking
("spiking")); the number of cells supporting the RNA assay can then be
indirectly
1 S inferred and corrected for RNA loss.
For quantitative assessment of recovery, known quantities of luciferase
DNA, a bacterial sequence with no known human homology, to quantitate tissue
DNA recovery (usually >75%); sero-negative samples receive a known quantity of
LTR HIV sequence to quantitate RNA recovery (>95%).
HIV-RNA is reproducibly detected in rectosi~rnoid biopsies from subjects with
undetectable plasma viral load.
Efforts were made to demonstrate the replicability of results from one
biopsy by comparison with others concurrently obtained at the same
circumferential
level (30 cm) from the same patient. Single biopsies (10 mg each) from
subjects with
undetectable plasma viral loads were frozen, RNA extracted and amplified using
LTR-specific primers 667/AA55, as described above. Each biopsy yielded an
average
of 25 gg RNA of which usually 1/100 was used for quantitation. Results in
Figure 2
demonstrate the reproducibility of quantitated RNA viral burden using
rectosigmoid
biopsies in a sensitive assay. The data demonstrates the tissue HIV RNA viral
load
WO 01/00871 CA 02377972 2001-12-24 pCT/US99/14366
-30-
from 2 biopsies obtained during the baseline sigmoidoscopies for subjects with
undetectable plasma viral loads. These individuals reported undetectable
plasma viral
load for >1 year. On average, there is < 0.2 1og10 difference between samples
within
the same subject. These data show the reproducibility and minimal sample to
sample
variation in using biopsies to quantitate tissue viral load. Equally important
is the
demonstration of detectable levels of tissue HIV RNA (usually 102 to 103 per
~g
RNA) in subjects with undetectable plasma HIV RNA.
HIV-DNA is re~roduciblv detected in rectosi~moid biopsies from subjects with
undetectable plasma viral load.
In a separate group of subjects with undetectable plasma viral load, HIV
DNA was amplified using quantitative PCR with specific 667/AA55 primers for
the
LTR region and b-globin specific primers used for internal linear standard.
Triplicates
were assayed for the b-globin specific primers; duplicates of the HIV proviral
DNA
quantitation are shown in Figure 3. Although lower limit of detection is 3
copies, 10
copies were used as our lower cut-off point. The figure shows both the actual
copy
number quantitated and the calculated number of copies based on a
b-globin-dependent cell count. These results show our technique can detect
copies of
proviral DNA as low as 10 in subjects with undetectable plasma viral load.
Nucleic Acid Isolation.
Total RNA and DNA is simultaneously isolated from a single endoscopic
biopsy by directly homogenizing the tissue using a Powergen 125
homogenizer(Fisher Scientific). Tissue samples are placed into 140q1 of urea
lysis
buffer and 701 of DEPC water in a SOmI conical centrifuge tube (Falcon 2070).
Following complete homogenization of the tissue SOOuI of saturated phenol
(Fisher #
1750-400) is added. An alternative approach to extract RNA and DNA from the
same
sample is achieved by pulverizing frozen tissue using a mortar and pestle
(Fisher
Scientific). Equal amounts are then used for DNA and RNA quantitation having
been
derived from the same piece of heterogeneously mixed tissue.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-31 -
For total RNA isolation 355.1 of the homogenate is added to lml of Trizol
reagent
(Gibco BRL #15596-56) in an Rnase and Dnase free 2m1 micro-centrifuge tube
(Intermountain Scientific # 3260-1 ) incubated at room temperature for 5
minutes prior
to the addition of 200u1 of chloroform. The upper phase is separated to a
fresh 2m1
micro-centrifuge tube and diluted with an equal volume of 70% ethanol. The
sample
was then applied to an Rneasy column ( Qiagen#74104) for further extraction as
per
the manufacturers protocol. RNA is eluted in DEPC treated water to a final
volume of
SO~I.Samples are subsequently treated with Dnase (Promega # M610A) for
40minutes at 37 degrees in a volume of 200q1 of DEPC water containing lmm DTT,
1 Omm MgClz, l Omm Tris and 2 units of Dnase. RNA is extracted by adding an
equal
volume of a 1:1 phenol/chloroform mixture, separation of the upper phase by
centrifugation (repeated x1) followed by a single chloroform extraction and
precipitation with 2.5x volume of 100% ethanol and 1/10 volume of SM NaCI at -
20
degrees for 1-2 hours.The precipitated RNA is pelleted, air-dried for 20-30
minutes
and re-dissolved in 30-SOqI of DEPC water. The yield of RNA is determined
spectrophotometrically at 260nm. Agarose gel electrophoresis confirms the RNA
to
contain both 28s and 18s species of RNA with little or no evidence of
degradation.
PCR amplification of the RNA template with Taq Polymerase confirms no DNA
contamination. Similar methods have been utilized extracted mRNA using
Pharmacia
kits (see Figure 12).
DNA is isolated from the remainder of the urea lysis buffer homogenate
by a series of phenol/chloroform (SOO~lx2) extractions using a 1:1
phenol/chloroform
mix followed by a single chloroform (5001) extraction. DNA is precipitated at -
20
degrees from the final extract by the addition of 251 of SM NaCI and 8501 of
100%
ethanol following a minimum incubation time of 30 minutes. The DNA is pelleted
by
spinning at maximun speed for 15 minutes, washed once in 70% ethanol and air
dried
for 30-40 minutes. The DNA pellet is redissolved by repeated pipetting to a
volume of
30-501 in water followed by incubation at 65 degrees for approximately 60
minutes
to ensure complete solubilization of the sample. DNA yield is determined
spectrophotometrically.
WO 01/00871 CA 02377972 2001-12-24 pCT/US99/14366
-32-
Determination of % Recovery; ability to trace loss.
In order to assess any losses in RNA or DNA during the isolation
procedure a series of experiments were conducted in which HIV negative control
biopsies were ''spiked" with increasing copy numbers of a known RNA (HIV LTR)
and DNA (firefly luciferase) sequence prior to homogenization. The recovery of
these
sequences in the final RNA and DNA was determined by PCR using the relevant
oligonucleotide primer pairs and expressed as a percentage of the copy numbers
added prior to extraction. A series of experiments were conducted in which the
external control sequences were added in inceasing quantity in order to
demonstrate
recovery efficiency at different starting concentrations. Since the
simultaneous
isolation process involves the splitting of the initial sample homogenate,
recovery
experiments were also undertaken in which the sample was either divided
equally for
RNA and DNA extraction or else was processed without division. Recovery
experiments demonstrated that both RNA and DNA can be extracted with high
efficiency (83-100%) and that division of the homogenate does not favor the
isolation
one form of nucleic acid over the other (potential problem if the sample was
not
homogeneous). These techniques are now being modified to enable detection of
loss
and subsequent correction during handling/shipping by the use of prepackaged
vials
from the inventor's laboratory.
PrCR Ouantitation: HIV RNA
Since tissue may contain many potentially inhibitory molecules, we
performed a series of experiments in which a known copy number (previously
ascertained to be within the linear amplification range) of Armored
cyclophilin RNA
(Ambion) were added to increasing amounts of tissue RNA and subjected to RT-
PCR
using the Thermo-stable rTth Reverse Transcriptase RNA PCR kit (Perkin-Elmer).
The
downstream cyclophilin primer is end labeled with P-32; radio-labeled PCR
products
are visualized and quantitated following polyacylamide gel electrophoresis
exposure to
a phospho-image plate and analysed using Image-Quant software (Molecular
Dynamics).
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
- 33 -
Such experiments demonstated that RNA concentrations of up to 261ng/ql
did not affect the amplification of the external control (> 100%) with only a
slight
diminution in efficiency at concentrations up to 561ng/ul (95%). An upper
limit of 250ng
of RNA is added to the RT-PCR to avoid any potential inhibitory interactions.
HIV RNA is quantitated using the Thermostable rTth Reverse Transcriptase
RNA PCR kit (Perkin-Elmer) in conjunction with the oligonucleotide primer pair
667/AA55 specific for the R/LJS region of the LTR (detects both spliced and
unspliced
HN RNA). 667 was end-labeled with P-32 using T4-Kinase (Gibco-Life
Technologies),
was subjected to 30 cycles of PCR (94 deg/1 min 65 deg/2 mins) followed by an
8
minute elongation at 72 degrees along with a previously established linear
standard curve
prepared from a stock of purified HIV LTR RNA. PCR products were subjected to
polyacrylamide gel electrophoresis and the radio-label incorporated in the
specific140bp
LTR product is determined by phosphoimage analysis. Samples are quantitated
with
reference to the linear portion of the standard curve by importing the raw
data into an
1 S excel spreadsheet. HIV RNA copies were expressed per qg of tissue RNA.
Samples can
also be expressed as a calculated result (number of copies per 106 cells) by
back-
calculation to determine the number of cells from which the RNA was derived.
As DNA
and RNA are extracted from the same sample (and same total sample volume) and
equally divided with internal tracer controls (e.g. luciferase DNA and
cyclophilin RNA),
the amount of cells in the DNA sample can be determined using the housekeeping
gene
~3-globin, corrected for loss (via luciferase tracking); the number of cells
supporting the
RNA assay can then be indirectly inferred and corrected for RNA loss.
PCR uantitation: HIV DNA
1-2~g of DNA is amplified using the oligonucleotide primer pair
AA55/667 recognizing the R/LTS region of the LTR and Taq Polymerase. As an
internal standard 100-200ng of DNA is amplified using oligonucleotide primers
LA1
and LA2 specific for a 11 Obp product of the ~-globin gene. Both reactions are
subjected to 30 cycles of PCR (94 deg/lmin 65deg/2mins) along with DNA
standard
curves for both products that are within a previously established linear
amplification
WO 01/00871 CA 02377972 2001-12-24 pCT/US99/14366
-34-
range. By end labeling the upstream oligonucleotide primer with P-32, radio-
labeled
product is resolved and quantitated by phospho-image analysis. Samples are
quantitated with reference to the linear portion of the respective standard
curves using
an excel spreadsheet. HIV DNA copies are expressed per 2x106 copies of ~3-
globin.
S
CCRS co-receptor expression on mucosal CD4 T cells.
Isolation of mucosal mononuclear cells does not alter phenotypic
expression of CD4, CDB, CCRS and CXCR4.
Samples were isolated from healthy, seronegative controls' blood
(peripheral blood mononuclear cells: PBMC) and intestinal mucosa (mucosal
mononuclear cells: MMC) to establish baseline CCRS and CXCR4 expression in
both
compartments. Figure 19 demonstrates that our isolation procedure neither
strips
relevant receptors (CD4, CDB, CCRS, CXCR4) nor alters their surface
expression.
Mucosal expression of CCRS on CD4 T cells is greatly increased
compared with PBMC.
Isolated mucosal mononuclear cells and peripheral blood mononuclear
cells were obtained from healthy, sero-negative control subjects and evaluated
to
determine the relative percentages of CD4 T lymphocytes in each compartment
expressing CCRS receptors. The 2D7 CCRS antibody is conjugated in a l :l ratio
with
phycoerythrin; the flow cytometry instruments used are calibrated to detect 44
phycoerythrin molecules per RFI channel (based on a standardized CD4
expression
and number of antibodies bound per cell). Consequently, the number of anti-
CCRS
antibodies bound per cell can be translated to number of receptors per cell,
assuming
monovalent binding of antibody to receptor.
The percentages of CCRS-expressing CD4+ T cells is significantly
increased in the gut (87%) compared to the blood (11%) (p=0.0019).
Further enhancing the vulnerability to HIV infection, shows mucosal CD4
T cells also express significantly more receptors per cell (mean of 8500)
compared to
blood CD4 T cells (mean 2700)(p=0.007).
CA 02377972 2001-12-24
WO 01/00871 PCT/US99/14366
-35-
CCRS expression on mucosal CD4 T cells is nearly exclusively on the
memory subset.
Blood and mucosal samples from the same seronegative, healthy controls
were counter stained with CD45R0 antibody as an indicator of the memory subset
to
determine the relative distribution of CCRS staining. After gating on CD4+
fluorescence, 91 % of CD4+ CD45R0+ mucosal cells express CCRS compared to
24% of a matched group in the blood (p=0.017).
CCRS expression on mucosal CD4 T cells remains increased compared to PBMC in
HIV-infected and inflammator~ples.
Having ascertained preliminary baselines of CCRS expression in mucosal
and blood CD4 T cells in healthy, seronegative subjects, the expression on CD4
T
cells in the setting of chronic HIV infection was evaluated to test the
hypothesis that
CCRS expression would remain increased in the mucosa compared to blood,
favoring
HIV replication. Inflammatory controls were included to discern changes not
directly
related to HIV-infection. Inflammatory controls were well-characterized
subjects with
inflammatory bowel disease (IBD), specifically, ulcerative colitis. Subjects
were
clinically in remission (maintained but controlled mucosal inflammation) on 5-
ASA
anti-inflammatory agents only, no steroids or immunosuppressive medications
were
used. HIV-infected individuals had peripheral CD4 counts between 200-700
cells/mm3 with a range of plasma viral loads (undetectable by ultrasensitive
assay:
n=2; plasma viral load between 200-2000 copies/ml: n=2; plasma viral load
between
20,000-40,000 copies/ml: n=4).
The differential expression of CCRS between mucosal and blood CD4 T
cells observed in seronegative normal controls was maintained in inflammatory
controls (p=0.012) and HIV-infected subjects (p=0.04)(Figure 18). In agreement
with
our hypothesis, there is a trend toward significance identifying a decrease in
mucosal
CCRS expression on CD4 T cells in HIV compared with normal controls as shown
in
this figure.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-36-
The CD4~CD8 ratio of CCRS-expressing T cells decreases in HIV and IBD in both
blood and -g,ut.
CCRS receptor on CD4 T cells are also expressed on CD8+ T cells. To
further evaluate if there was a true decrease in CCRS expression on mucosal
CD4 T
cells, evaluation of the relative distribution of CCRS receptors between CD4
and
CD8 T lymphocytes in both compartments in the three clinical conditions.
Figure 19
demonstrates the dramatic downward shifts in the CD4:CD8 ratios of CCRS
expressing cells, decreasing in blood and gut samples by roughly 50% in IBD
and
70-90% in HIV. This may represent a protective down regulation of CCRS to
inhibit
HIV spread. Given the similar trend in the inflammatory controls, the primary
stimulus for decreased surface expression likely relates to the inflammatory
milieu.
Samples are being processed to confirm that b-chemokine levels are elevated in
both
conditions, but even more so in HIV than in IBD samples. Supporting our
hypothesis,
these findings would suggest an extremely active inflammatory mucosal state in
HIV
(as defined by chemokine activity) despite the histological reports of
relative
lymphopenia.
Mucosal mononuclear cells (MMC) are significantly more infectible in
vitro than PBMC.
Initial observations demonstrated increased vulnerability to HIV infection
of mucosal cells due, in part, to increased co-receptor expression. This
hypothesis was
tested in vitro using isolated MMC and PBMC from the same individuals and
incubated with M-tropic HIVSX for 2 hours, washed and cultured for 3-10 days.
Aliquots of supernatant were collected at the demonstrated times and assayed
fro p24
production as evidence of infection. The p24 production at the first time
point (3
days) was markedly increased compared to concurrently incubated PBMC (4000 ng
p24/ml in MMC compared to 550 ng p24/ml in PBMC). These pilot data support our
hypothesis that the increased co-receptor expression on mucosal CD4 T cells
enhances infectivity by HIV.
WO 01/00871 CA 02377972 2001-12-24 PCT/US99/14366
-37-
A number of embodiments of the present invention have been described.
Nevertheless, it will be understood that various modifications may be made
without
departing from the spirit and scope of the invention.
