| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Department of Pathology*
and The
Washington Regional Primate Research Center,
University of Washington, Seattle, Washington
 |
Abstract |
|---|
 Top Abstract IntroductionStudy DesignMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionStudy DesignMaterials and MethodsResultsDiscussionReferences |
|---|
The pathogenesis of HIV-associated TMA is poorly understood. A central question remaining is whether the virus directly infects intrinsic renal cells in the course of this disease.2,14 To date, efforts to identify direct HIV infection of intrinsic renal cells in any disease process have been conflicting but generally have failed to provide convincing evidence of such infection. In vitro studies purportedly demonstrating such infection have used protocols in which viral transfection rather than infection seems to have occurred15 ; studies attempting direct infection of renal cells by viral isolates have generally failed to demonstrate productive infection.16 In vivo studies of human renal biopsies of patients with HIVAN have provided some evidence of direct HIV infection,17,18 but these studies are not convincing because of problems of reproducibility19,20 or questions concerning possible contamination of studied renal cell populations by infiltrating infected leukocytes.18
The situation concerning HIV infectivity is further complicated by limited knowledge concerning possible receptors for viral entry into specific renal cell types. Chemokine receptors have been identified as essential coreceptors for human immunodeficiency virus infection in mammalian cells.21-23 The chemokine receptor CCR5 serves as the major coreceptor together with CD4 for macrophage-tropic strains of HIV-1,24-28 whereas T lymphocyte-tropic strains of HIV-1 use the chemokine receptor CXCR4 as the principal coreceptor required for infection.29,30 The expression of chemokine receptors in HIV-associated renal disease is unknown.
In this study, we report on a nonhuman primate model infected by a well-characterized strain of HIV-2, HIV-2287.31,32 Pigtailed macaques (Macaca nemestrina) infected by HIV-2287 predictably and rapidly develop clinical features characteristic of human HIV-1 infections. A significant proportion of these animals develop renal and systemic TMA that is morphologically similar to TMA as it occurs in humans. These lesions develop despite an apparent lack of HIV infection of renal parenchymal cells, although systemic infection of hematopoietic cells can be readily demonstrated. Additional studies to identify the presence of chemokine coreceptors needed to support cellular infection by HIV-1 in humans suggest the absence of these receptors on renal cells,33,34 which may account for the difficulty in demonstrating the direct infection of such cells by strains of HIV-1. We believe the characterization of a model of HIV-associated TMA offers a major opportunity to understand this important disease process.
|
Study Design |
|---|
|
TopAbstractIntroductionStudy DesignMaterials and MethodsResultsDiscussionReferences |
|---|
We prospectively analyzed the renal pathology and function of 27 pigtailed macaques, experimentally infected intravenously with a virulent HIV-2 strain, HIV-2287. Necropsies were performed at 12 hours and at days 1, 2, 4, 6, 10, 14, 21, and 28 (n = 3, each time point) after infection. Renal pathology was evaluated at necropsy by light and electron microscopy. Renal function was assessed at necropsy by measurement of serum-creatinine, urea, and urinary protein excretion. Renal pathology and function were examined prospectively in four uninfected control macaques matched for age and sex. Furthermore, kidney tissues obtained from 39 pigtailed macaques experimentally infected with HIV-2287 that were involved in viral dissemination studies were analyzed retrospectively. HIV-2 RNA was detected in kidney tissues by an in situ hybridization technique. The expression of the HIV infection coreceptor CXCR4 in kidney tissues was analyzed by in situ hybridization.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionStudy DesignMaterials and MethodsResultsDiscussionReferences |
|---|
Twenty-seven pigtailed macaques, Macaca nemestrina (ages 1.2–3.1 years, median 2.1 years; 11 male, 16 female) were inoculated intravenously with 50 TCID50 (50% tissue culture infectious doses) HIV-2287. All macaques involved in this study were colony-born and weighed between 2.1 and 4.5 kg (median 3.0 kg) at necropsy. Before inoculation, each animal was determined to be clinically healthy by physical examination and complete blood cell count. All animals in this study were screened and determined to be negative for simian immunodeficiency virus (SIV) and simian retrovirus (SRV) coinfection before HIV-2 inoculation. This was determined by antibody, cell culture, and polymerase chain reaction (PCR) assays. The macaques were euthanized at 12 hours and on day 1, 2, 4, 6, 10, 14, 21, 28 (n = 3, each time point) after inoculation. The animals were sedated with an intramuscular injection of ketamine-HCl (10 mg/kg body weight) followed by an intravenous overdose of pentabarbital. Complete necropsy examinations were performed on all macaques. All study protocols and procedures were reviewed and approved by the Washington Regional Primate Research Center and the University of Washington Animal Care and Use Committee.
Four uninfected, untreated, clinically healthy pigtailed macaques, M. nemestrina (ages 1.4–2.6 years; three male, one female) were euthanized and served as control animals.
Kidney tissue specimens were obtained from 39 additional macaques experimentally infected by HIV-2287. These animals were involved in previous studies investigating HIV-2 dissemination kinetics and pathogenesis. Protocols of these studies have been published previously.31,32 All 39 macaques (ages 0.3–6.4 years, median 1.9 years; 16 male, 23 female) were colony-born and weighed between 0.4 and 9.3 kg (median 2.9 kg) at necropsy. The macaques were euthanized between 63 and 556 days (median 165 days) after HIV-2 infection.
Virus
HIV-2287 was derived by serial passage of HIV-2EHO (originally obtained from Dr. Luc Montagnier, Pasteur Institute) in M. nemestrina. Animal F87265 was inoculated intravenously with 106 TCID (tissue culture infective dose) cell-free HIV-2EHO as well as 107 autologous peripheral blood mononuclear cells (PBMCs) infected in vitro. Ten milliliters of whole blood obtained from macaque F87265 44 weeks after infection was transfused into animal F89071, and then 10 ml of blood from animal F89071 at 20 weeks after infection was inoculated into F89287. The HIV-2287 challenge stock was derived from coculture of lymph node mononuclear cells of animal F89287 with fresh stimulated allogenic macaque PBMCs. Virus stocks were prepared as clarified supernatants (3000 x g for 20 minutes at 4°C) and aliquoted and stored at -80°C until use. In the prospective part of this study, all 27 macaques were inoculated intravenously with 50 TCID50 of the virus stock solution.
Detection of HIV-2-Infected PBMC
HIV-2 infected PBMC were detected by a quantitative coculture assay as described.31,35 Briefly, freshly isolated macaque PBMCs or lymphocytes were serially diluted in triplets, starting with 106 cells, and cocultivated with fresh human CD8-positive T-cell-depleted phytohemagglutinin-activated PBMCs. Cultures were incubated for 14 days, and the presence of virus was detected using an HIV-2 p27 antigen capture assay. Titers were calculated as the maximum dilution of cells that gave positive cultures and reported as numbers of positive cells/106 PBMCs.
Hematological and Serological Parameters
Samples of serum, EDTA-plasma, and urine were obtained from each macaque before experimental euthanasia. Complete blood count, serum-creatinine, urea, and urinary protein excretion were measured using standard procedures. The CD4-positive T-lymphocyte subset was measured as described previously.36 Briefly, lymphocytes were stained with PE-conjugated anti-CD4 antibody (LEU 3a; Becton-Dickinson, San Jose, CA) and analyzed with a flow cytometer (FACScan/FACSort; Becton-Dickinson).
Histopathology
Tissue samples of kidney, liver, lung, brain, adrenal gland, skin, gut, and lymph nodes were fixed in 10% phosphate-buffered formalin, embedded in paraffin, and sectioned. Sections of kidney tissue were routinely stained with hematoxylin and eosin, periodic acid-Schiff (PAS), and Jones's silver methenamine reagents, respectively. Tissue sections of all other organs were routinely stained for hematoxylin and eosin, PAS, and elastic van Gieson, respectively.
Electron Microscopy
Tissue for electron microscopy was fixed in half-strength Karnovsky's solution (1% paraformaldehyde and 1.25% glutaraldehyde in 0.1 mol/L sodium cacodylate buffer, pH 7.0). After fixation, tissue was postfixed in 1% osmium tetroxide for 2 hours, dehydrated in graded ethanols, and embedded in epoxy resin. Thin sections were stained with uranyl acetate and lead citrate and examined with a Phillips 410 electron microscope (Phillips Export BV, Eindhoven, the Netherlands). Kidney tissue examination included at least two and usually three or more glomeruli per animal, as well as a survey of cortical and medullary interstitium, tubules, and blood vessels.
Molecular Probes
HIV-2
A 1.7-kb sequence of DNA coding for HIV-2 gp120 (env) (nucleotides 6480–8335 of the HIV-2287 sequence) was subcloned into pCR II (Invitrogen) (kindly provided by Bristol-Myers Squibb, Seattle, WA), linearized with XhoI, and transcribed with Sp6 RNA polymerase for the antisense probe, or linearized with BamHI and transcribed with T7 RNA polymerase for the sense probe. Detailed protocols for the transcription reaction and the characterization of the specificity of this riboprobe have been described previously.33
CXCR4
A 1.1-kb sequence of DNA coding for human CXCR4 was subcloned into pcDNAI/amp (Invitrogen) (obtained through the AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases, National Institutes of Health, originally provided by Dr. Nathaniel Landau),24,37 linearized with HindIII, transcribed with Sp6 for the antisense probe or linearized with XbaI, and transcribed with T7 for the sense probe. Specificity of the CXCR4 antisense riboprobe has been demonstrated previously by Northern analysis and by in situ hybridization using a series of cell lines transfected with different members of the family of chemokine receptors.34
In Situ Hybridization
HIV-2 RNA and CXCR4 mRNA were detected in tissue sections with in situ hybridization techniques following protocols that we have previously used.33,38,39 Four-micron sections of formalin-fixed, paraffin-embedded tissue samples were rehydrated through xylene and graded ethanols, washed with 0.5x standard saline citrate (SSC) (1x SSC = 150 mM NaCl, 15 mM Na citrate, pH 7.0), and digested with proteinase K (5 µg/ml; Sigma) in Tris buffer for 30 minutes at 37°C. Several 0.5x SSC washes were followed by prehybridization for 2 hours in 100 µl of prehybridization buffer (0.3 mol/L NaCl, 20 mmol/L Tris (pH 8.0), 5 mmol/L EDTA, 1x Denhardt`s solution, 10% dextran sulfate, 10 mmol/L dithiothreitol). The hybridizations were started by adding 500,000 cpm of35 S-labeled riboprobe in 50 µl of prehybridization buffer and allowed to proceed overnight at 50°C. After hybridization, sections were washed with 0.5x SSC, and treated with RNase A (20 µg/ml, 30 minutes at 37°C), washed in 2x SSC (2 x 2 minutes), followed by three high-stringency washes in 0.1x SSC/0.5% Tween 20 (Sigma) for 40 minutes each at 50°C and several 2x SSC washes. After the tissue was dehydrated and air-dried, it was dipped in NTB2 nuclear emulsion (Kodak, Rochester, NY) and exposed in the dark at 4°C for 2 weeks (HIV-2) or 4 weeks (CXCR4). After developing, the sections were counterstained with hematoxylin and eosin, dehydrated, and coverslipped.
|
Results |
|---|
|
TopAbstractIntroductionStudy DesignMaterials and MethodsResultsDiscussionReferences |
|---|
Infection with HIV-2287 in this animal model results
in a rapid and predictable decline in CD4-positive cells in the
peripheral blood lymphocyte population. Blood samples obtained at
necropsy demonstrated a decrease in CD4-positive cells below 350
cells/µl in five of six animals 21 days and 28 days after infection
(Figure 1A)
. HIV-2-infected PBMCs were
detectable in PBMC cocultures in 12/12 animals after 10 days of
infection (Figure 1B)
. None of the 27 macaques involved in this study
developed other features of clinical AIDS.
|
Pathological examinations of renal tissue from the 27 macaques
acutely infected with HIV-2 revealed TMA lesions in six animals. Renal
TMA was detectable in two out of three animals as early as 12 hours
after HIV-2 infection, and four out of nine macaques had developed TMA
after 14 days of infection. None of the macaques euthanized between 1
and 10 days after HIV-2 infection demonstrated features of TMA at
necropsy. Table 1
summarizes the
histological and ultrastructural features of all 27 macaques.
|
. These lesions were identical
to those observed in human HIV-associated TMA (Figure 2C)
. Whereas four
animals had focal TMA lesions involving approximately 10–25% of the
glomeruli available for examination, one animal showed a very global
pattern of TMA involving more than 60% of the glomeruli. As indicated
in Table 1
, several macaques revealed variable degrees of interstitial
inflammation. These lesions characteristically consisted of discrete
foci of accumulations of mononuclear inflammatory cells, principally
lymphocytes and macrophages. Tubular atrophy in these areas was
typically minimal to mild. Areas of inflammation generally comprised
less than 5% of the renal parenchyma examined histologically in each
affected animal.
|
Ultrastructural examination confirmed the diagnosis of TMA in all five
cases. A frequent ultrastructural finding in the glomeruli of all
animals with histological appearance of TMA was areas of mesangiolysis
with accumulation of electron-lucent material admixed with cellular
debris (Figure 3A)
. Another frequent
finding in some glomeruli of all TMA cases was detachment of
endothelial cells from the glomerular basement membranes, with
subendothelial accumulation of electron-lucent material (Figure 3B)
.
All animals with the histological TMA diagnosis showed variable degrees
of intracapillary thrombi composed of amorphous material, platelets,
cellular debris, and, in some cases, fibrin (Figure 3C)
. Visceral
epithelial cells were generally well preserved and did not display
features of retraction and/or prominent effacement of foot processes.
However, at sites of severe injury disruption of the podocyte network
was identified (Figure 3C)
. Immune complex-type electron-dense deposits
were undetectable in all observed glomeruli. Electron microscopic
examination of renal tissue was performed in two additional cases that
showed no significant pathological abnormalities by light microscopy
but revealed mild proteinuria at necropsy (Table 1
and Figure 5B
). One
animal (K95267, Table 1
) had clear features of TMA ultrastructurally,
including mesangiolysis, endothelial detachment from basement membranes
with accumulation of subendothelial electron-lucent material, and
platelet thrombi in glomerular capillary loops. However, the second
proteinuric macaque showed neither histological nor ultrastructural
abnormalities. Visceral epithelial cells and foot processes were well
preserved in both cases.
|
|
) was indeterminate for TMA. This
animal lacked unequivocal microthrombi on histological and
ultrastructural examination, but had some features that were suggestive
of such a process, including the histological appearance of congested,
ischemic glomeruli and the ultrastructural presence of a thrombus in
one small artery.
None of the four uninfected control macaques demonstrated histological
or ultrastructural features of renal TMA (Table 1)
. Histological
examination revealed variable degrees of interstitial inflammation
(focal accumulation of mononuclear inflammatory cells) in three of four
uninfected control macaques. Ultrastructural examination did not detect
features of mesangiolysis, endothelial cell injury, or thrombi in
glomerular capillary lumina in any of the four uninfected control
animals. One animal (J96086, Table 1
) demonstrated ultrastructural
features of a glomerular basement abnormality. Thickening and
lamination of the basement membrane similar to those seen in human
hereditary nephritis/Alport's syndrome were present without any
additional identifiable abnormalities. Immune complex-type
electron-dense deposits were undetectable. One animal (F96092, Table 1
)
demonstrated predominantely mesangial immune complex-type
electron-dense deposits with focal small accumulations. Similar
deposits were occasionally present in subendothelial portions of
peripheral capillary walls and rarely present in intramembranous and
subepithelial locations. No additional identifiable ultrastructural
abnormalities were present. Endothelial cells, mesangial cells, and
visceral epithelial cells were well preserved in all cases.
Microvascular thrombi remained undetectable.
TMA Is a Systemic Disease in HIV-2-Infected Macaques
We analyzed tissue sections of liver, lung, brain, adrenal gland,
skin, and gut (including stomach, duodenum, ileum, jejunum, and colon)
for the presence of microthrombi in three animals with diagnosed renal
TMA. Focal microthrombi were detected in the gastrointestinal
submucosa, the adrenal gland, and in the periadrenal connective tissue
in these animals that had developed renal TMA. Microthrombi were
undetectable in all other organs. We subsequently analyzed tissue
sections of gut and adrenal (n = 4 tissue
sections per animal) of all 27 macaques acutely infected with HIV-2.
The analysis was performed on PAS-stained tissue sections by an
observer unaware of the presence of renal TMA in these animals.
Extrarenal thrombi were identified in five of 27 macaques (Figure 4
, Table 1
). Four animals had developed
focal microthrombi in the gastrointestinal submucosa, three in the
periadrenal connective tissue, and one in the adrenal medulla. The
lesions were generally focal, involving less than 5% of the analyzed
tissue, and inflammatory cell infiltrates were absent at sites of the
thrombi. All five animals with extrarenal thrombi had diagnosed renal
TMA lesions. Extrarenal microthrombi were undetectable in all animals
that had not developed renal TMA.
|
None of the 27 macaques that had been infected with
HIV-2287 had serological evidence of renal failure at
necropsy. Figure 5
illustrates the renal
functional parameters of the six macaques acutely infected with HIV-2
that had developed renal TMA. At necropsy all six animals were
clinically healthy and indistinguishable from the animals without TMA.
Serum BUN (Figure 5A)
and creatinine levels (data not shown) showed no
significant difference in HIV-2-infected macaques that had developed
TMA compared to HIV-2-infected macaques without TMA or uninfected,
normal control macaques (Figure 5A)
. However, three animals had
developed a mild proteinuria at necropsy (74–111 mg protein/dl; Figure 5B
). Whereas two of these three HIV-2-infected macaques demonstrated
the presence of renal TMA, the third proteinuric macaque (J96174, Table 1
) showed no pathological abnormalities by either histological or
ultrastructural examination.
Comparison of the hematological parameters in the HIV-2-infected
macaques that had developed TMA with HIV-2-infected macaques that had
not developed TMA showed no significant correlation between the
incidence of TMA and the red blood cell count (Figure 6A)
, platelet count (Figure 6B)
, white
blood cell count (data not shown), or CD4 cell count (data not shown),
respectively. None of the 27 HIV-2-infected animals had evidence of
thrombocytopenia at the time of necropsy (Figure 6B)
.
|
. Retrospective Analysis of Kidneys from HIV-2-Infected Macaques
In the second component of this study, we retrospectively examined
renal tissue samples obtained from 39 additional macaques that had been
infected with HIV-2287. Seven of these 39 animals
demonstrated typical histological and ultrastructural features of renal
TMA lesions. Table 2
summarizes the
clinical and hematological data of the seven macaques that had
developed renal TMA. All seven macaques had a focal distribution
pattern of the TMA lesions involving 10–25% of the glomeruli. As in
the macaques acutely infected with HIV-2, lesions characteristic of
human HIVAN remained undetectable. Tubulointerstitial inflammatory
infiltrates were detectable in some specimens, but these lesions were
generally mild, were of focal distribution, and were not associated
with the TMA lesions of affected glomeruli or blood vessels.
|
Serum and urine samples obtained at necropsy were additionally available in three of seven macaques with renal TMA in the retrospectively analyzed group. No decrease in renal function or elevation of urinary protein excretion was demonstrable in these three animals.
HIV-2 RNA Is Undetectable in Kidneys of HIV-2-Infected Macaques
Direct infection of renal parenchymal cells has been implicated in
pathogenetic mechanisms underlying HIV-associated renal disease. We
analyzed sections of renal tissue samples from all 66 HIV-2-infected
macaques for the presence of HIV-2 RNA by in situ
hybridization. HIV-2 RNA was completely absent in kidney tissue
sections of 65/66 HIV-2-infected macaques. All 13 macaques that had
developed renal TMA lesions showed no detectable HIV-2 RNA by this
technique (Figure 7C)
. In one animal
acutely infected with HIV-2 (K96183, Table 1
), a few HIV-2-infected
cells were demonstrable (Figure 7D)
. HIV-2 RNA expression in this
animal was localized to infiltrating mononuclear leukocytes at sites of
mild, focal tubulointerstitial inflammation (Figure 7D)
. HIV-2 RNA
expression was completely absent in renal parenchymal cells of the
glomerular, tubular, or vascular compartments.
|
illustrates several HIV-2
RNA-expressing mononuclear cells within a mesenteric lymph node
obtained from a macaque 10 days after intravenous HIV-2 infection.
HIV-2-infected cells were detected in lymph node tissue samples of all
macaques after 10 days of HIV-2 infection (Eitner et al, manuscript in
preparation). Chemokine Receptor CXCR4 Expression in Kidneys of HIV-2-Infected Macaques
Chemokine receptors have been identified as coreceptors for HIV
infection in mammalian cells. Our preliminary results indicate that
HIV-2287 preferentially utilizes the chemokine receptor
CXCR4, rather than CCR5, as coreceptor (M.-A. Rey-Cuille and S.-L. Hu,
personal observations). We studied the cellular sites of synthesis of
CXCR4 in kidneys of all 27 macaques acutely infected with
HIV-2287, in the seven additional macaques that had
developed TMA, and in four uninfected normal macaques. The CXCR4
antisense probe used in this study was originally generated against
human CXCR4. We were able to show that this riboprobe also detects
nonhuman primate CXCR4. In situ hybridization demonstrated
the presence of CXCR4-expressing cells in lymph node tissue from
HIV-2-infected macaques (Figure 8A)
and
from uninfected macaques (data not shown). Numerous CXCR4
mRNA-expressing lymphocytes were detectable in the macaque lymph nodes
(Figure 8A)
. Several of the analyzed kidneys showed the presence of a
few CXCR4 mRNA-expressing cells by in situ hybridization.
Those cells appeared to be leukocytes and were localized in the
interstitium at sites of focal mononuclear cell infiltration (Figure 8D)
. By in situ hybridization CXCR4 mRNA expression was
completely absent in intrinsic renal cells of the glomerular, tubular,
interstitial, or vascular compartments (Figure 8C)
. Endothelial cells
of the renal vasculature showed no detectable CXCR4 mRNA hybridization
signal. Chemokine receptor CXCR4 expression was undetectable at sites
of TMA lesions in these animals.
|
|
Discussion |
|---|
|
TopAbstractIntroductionStudy DesignMaterials and MethodsResultsDiscussionReferences |
|---|
We recognize that the best animal model for human HIV infections would be one in which HIV-1 infects and induces an AIDS-like disease. Previous investigation has shown that chimpanzees are susceptible to infection with HIV-1, but to date only one animal has progressed to clinical AIDS.41 In addition, availability, financial considerations, and the protected status of chimpanzees as an endangered species make the chimpanzee an unsuitable candidate for extensive animal model development. Similarly, although pigtailed macaques (M. nemestrina) are susceptible to HIV-1 infection, the infection appears to be of limited duration, with low virus levels and restricted pathogenic features.36,42 We have now demonstrated that one strain of HIV-2, HIV-2287, originally derived from a human AIDS patient, is not only readily infectious for M. nemestrina but also highly pathogenic, progressing to a clinical AIDS syndrome within an accelerated time frame of 6–12 months.31,32 Although HIV-2 is closely related to simian immunodeficiency virus (SIV), the HIV-2 viruses remain distinct from all known strains of SIV,43,44 thereby presenting a unique opportunity for the study of a macaque model of human AIDS, using a virus adapted from a known human pathogen.
Evaluation of the renal pathology of 66 pigtailed macaques infected with HIV-2287 demonstrated TMA in 13 animals. One additional HIV-2-infected animal had features of a focal vasculitis and crescentic glomerulonephritis. Characteristics of human HIVAN (FSGS, hyperplasia/proliferation of visceral epithelial cells, tubular microcysts) were undetectable in all animals. The TMA lesions in the HIV-2-infected macaques were virtually identical with TMA lesions that can be seen in some HIV-infected humans. Interestingly, none of the HIV-2-infected macaques with TMA had developed clinical features typical for the manifestation of this disease in humans. Neither renal failure nor thrombocytopenia, frequently detectable in HIV-associated TMA in humans, was seen in the macaques with the pathological diagnosis of TMA. The subclinical course of TMA in this animal model could be explained by the scheduled euthanasia of the animals relatively early after HIV-2 infection. None of the macaques involved in this study had to be euthanized because of the development of severe AIDS manifestations.
Several hundred cases of HIV-associated TMA have been reported in the literature, with particularly increased recognition during the last several years (reviewed in2-7 ). Compared to several million HIV-1-infected humans worldwide, the percentage of TMA-affected HIV-1-positive patients might seem low. However, some authors consider HIV-associated TMA to be among the most common forms of TMA currently affecting adults,7 with HIV-associated TMA accounting for up to 30% of hospitalized cases of TMA affecting adults in some institutions.2 We hypothesize that TMA lesions might be underdiagnosed in HIV-infected humans, presumably because of a lack of suggestive clinical symptoms that would lead to renal biopsies in this patient group.
Because of the design of this study, we cannot completely rule out the presence of preexisting TMA lesions in these animals independent of the HIV-2 infection. However, nephropathological examinations of four uninfected control macaques involved in this study did not reveal TMA lesions, indicating that these lesions are not commonly encountered in macaque populations. Further historical controls in our laboratory included additional uninfected normal macaques (n = 5), HIV-1-infected macaques (n = 9), SIVMne-infected macaques (n = 83), and SRV-infected macaques (n = 40). TMA lesions similar to those seen in the HIV-2-infected macaques were not identified in any of these 137 animals (Ref. 45 and our unpublished observations), indicating that TMA is a consequence of infection with certain retroviral species, such as HIV-2287 in this case, and not a preexisting condition. These findings further indicate that induction of an immunodeficiency state by such retroviral infections does not provide a sufficient basis for the development of TMA.
It is noteworthy that TMA lesions can be identified in some macaques as early as 12 hours after intravenous HIV-2 infection. At this time point HIV-2-infected cells were undetectable in PBMCs, in lymphoid tissues, and in the kidney. The presence of viral proteins capable of inducing TMA in the kidney without renal infection might be one explanation for the high incidence of TMA 12 hours after experimental intravenous HIV-2287 infection. Because of the design of this study, we cannot exclude the possibility that TMA developed in numerous animals very early after experimental infection, and that after an initial recovery, only a few animals continued to exhibit the disease. Future studies, including the inoculation of macaques with inactivated, noninfectious HIV-2287 preparations, might help in the further characterization of mechanisms involved in the pathogenesis of HIV-associated TMA.
TMA can be seen in macaques after intravenous and mucosal infection. Most animals that developed TMA in the present study had been infected intravenously, but this might reflect the fact that the majority of macaques involved in this study had been infected intravenously.
Finally, as all 13 macaques that had developed TMA were colony-born animals, we sought to determine whether a common familiar background might be associated with the TMA development in these animals independently of the HIV-2 infection, as sometimes happens in human kidneys with familial TMA because of abnormalities of clotting proteins. Macaques with TMA lesions could not be identified as members of one inbred family. Comparison of all known ancestors (up to five generations) showed that although some animals were related to the same grandparents, as expected for colony-born animals, several macaques clearly demonstrated absent familial relationships with all other animals that had developed TMA.
Pathogenic mechanisms leading to endothelial cell injury and platelet deposition in the microvasculature of HIV-associated TMA are poorly understood.2,11,14 Several studies have suggested the potential for HIV infection of renal tissue.2,11 Direct endothelial cell infection by HIV-1 has been reported both in vivo and in vitro by several groups (reviewed in Ref. 2 ), but direct evidence for a pathogenic role of this infection in TMA is lacking. Published data on HIV infection of endothelial cells in HIV-associated TMA is very limited. Del Arco et al detected HIV-1 p24 antigen by immunohistochemistry in some bone marrow endothelial cells from a single case of thrombotic thrombocytopenic purpura in a woman infected with HIV.46 However, HIV-1 immunostaining in this case was not associated with any evidence of local tissue damage in the analyzed bone marrow specimen. The presence of HIV-1 protein or RNA has never been reported in renal microvasculature endothelial cells of HIV-1-infected humans that developed TMA.
In the present study we sought to detect HIV-2 RNA in kidney sections of HIV-2-infected macaques by in situ hybridization. Renal expression of HIV-2 RNA was absent in all 13 animals that had developed TMA lesions. Furthermore, HIV-2 RNA was not detectable in renal parenchymal cells comprising the glomerular, tubular, or vascular compartments of all 66 HIV-2-infected macaques. A few HIV-2 RNA-expressing infiltrating leukocytes were detectable in a single animal at sites of focal interstitial inflammation. Despite the presence of mild, focal interstitial inflammation in several additional HIV-2-infected macaques, HIV-2 RNA remained undetectable in these cases. (Our in situ hybridization procedures detected numerous HIV-2 RNA-expressing cells in PBMCs or lymphoid tissues of the infected macaques.) We recognize that the in situ hybridization technique without PCR amplification has a limited sensitivity and that there are likely to be some infected cells in which the number of RNA copies is below the threshold of our detection system. However, in situ hybridization offers the important advantage of clearly colocalizing the RNA expression to cellular phenotypes.
Chemokine receptors have been shown to serve as coreceptors together with CD4 for HIV-1 and HIV-2 strains. Some HIV-2 isolates have been described that use CXCR4 for virus entry in a CD4-independent fashion.47,48 This raises the possibility that HIV may be able to infect CD4-negative cells that express CXCR4 or perhaps other chemokine receptors. Some preliminary data indicate that HIV-2287 is most likely a CXCR4-tropic virus (M.-A. Rey-Cuille and S.-L. Hu, personal observations), although we cannot completely rule out the possibility that it is also using some unknown monkey receptor. In this study, CXCR4 expression was undetectable in intrinsic renal cells of HIV-2-infected macaques.
Two recent studies demonstrated chemokine receptor CXCR4 expression in endothelial cells.49,50 CXCR4 mRNA and protein expression was detected in cultured endothelial cells isolated from human umbilical veins, human coronary arteries, human brain microvasculature, bovine aorta, bovine pulmonary arteries, and the endothelial cell layer in rabbit thoracic aortic tissue.49,50 Both studies showed identical CXCR4 expression patterns on the transcript and protein levels, indicating that CXCR4 mRNA is subsequently translated into protein. We have previously reported that CXCR4 mRNA was undetectable by in situ hybridization in renal microvascular endothelial cells in normal, mature human kidney and in human allograft nephrectomies undergoing severe allograft rejection.34 In the present study, CXCR4 mRNA expression was undetectable in endothelial cells of the renal vasculature in uninfected and HIV-2-infected macaques. This observation may provide additional evidence against a role for direct HIV-2 infection of intrinsic renal cells as a proximate cause of injury in this disease model. However, we cannot rule out the possibility that other chemokine receptors that could act as coreceptors for HIV-2 infection might be expressed on renal vascular endothelial cells.
The HIV-2287 M. nemestrina model may be useful in the further study of the pathogenesis of HIV-associated TMA lesions. This study does not support a role for productive viral infection of intrinsic renal cells in the pathogenesis of this disease. We believe that our failure to detect specific viral RNA and chemokine receptor CXCR4 mRNA in renal parenchyma of macaques infected with HIV-2 provides additional insight into the difficulty encountered in attempts to demonstrate direct infection of intrinsic renal cells by immunodeficiency viruses. The most attractive hypothesis remaining to explain the high incidence of TMA in both human and nonhuman primates infected by HIV involves injury to the endothelium, with particular susceptibility of the glomerular and arteriolar vascular beds. Studies to determine whether the endothelium is damaged or assumes a procoagulant phenotype in HIV-associated TMA are currently under way.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported in part by grants DK 49514, DK 47659, and RR 00166 from the National Institutes of Health.
Accepted for publication April 11, 1999.
|
References |
|---|
|
TopAbstractIntroductionStudy DesignMaterials and MethodsResultsDiscussionReferences |
|---|
, MIP-1ß receptor as a fusion cofactor for macrophage-tropic HIV-1. Science 1996, 272:1955-1958[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  Z. V. Popovic, R. Sandhoff, T. P. Sijmonsma, S. Kaden, R. Jennemann, E. Kiss, E. Tone, F. Autschbach, N. Platt, E. Malle, et al. Sulfated Glycosphingolipid as Mediator of Phagocytosis: SM4s Enhances Apoptotic Cell Clearance and Modulates Macrophage Activity J. Immunol., November 15, 2007; 179(10): 6770 - 6782. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. I. de Silva, F. A. Post, M. D. Griffin, and D. H. Dockrell HIV-1 Infection and the Kidney: An Evolving Challenge in HIV Medicine Mayo Clin. Proc., September 1, 2007; 82(9): 1103 - 1116. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L. Hudkins, D. G. Gilbertson, M. Carling, S. Taneda, S. D. Hughes, M. S. Holdren, T. E. Palmer, S. Topouzis, A. C. Haran, A. L. Feldhaus, et al. Exogenous PDGF-D Is a Potent Mesangial Cell Mitogen and Causes a Severe Mesangial Proliferative Glomerulopathy J. Am. Soc. Nephrol., February 1, 2004; 15(2): 286 - 298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. L. Kimmel, L. Barisoni, and J. B. Kopp Pathogenesis and Treatment of HIV-Associated Renal Diseases: Lessons from Clinical and Animal Studies, Molecular Pathologic Correlations, and Genetic Investigations Ann Intern Med, August 5, 2003; 139(3): 214 - 226. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Segerer, F. Eitner, Y. Cui, K. L. Hudkins, and C. E. Alpers Cellular Injury Associated with Renal Thrombotic Microangiopathy in Human Immunodeficiency Virus-Infected Macaques J. Am. Soc. Nephrol., February 1, 2002; 13(2): 370 - 378. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Eitner, Y. Cui, G. Grouard-Vogel, K. L. Hudkins, A. Schmidt, T. Birkebak, M. B. Agy, S.-L. Hu, W. R. Morton, D. M. Anderson, et al. Rapid Shift from Virally Infected Cells to Germinal Center-Retained Virus after HIV-2 Infection of Macaques Am. J. Pathol., April 1, 2000; 156(4): 1197 - 1207. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
