Departments of
* Microbiology & Immunology,
Pathology,
Nuclear Medicine, and
Internal Medicine, University of Texas Medical Branch, Galveston
Correspondence: Miles W. Cloyd, Ph.D., Department of Microbiology and Immunology, Rt. 1070, The University of Texas Medical Branch, Galveston, TX 77555-1070. E-mail: mcloyd{at}utmb.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: human • T lymphocytes • AIDS/HIV • cell trafficking
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
109) are eliminated per day in HIV+ individuals. Furthermore, researchers had to speculate as to a compensatory increase in production of new CD4 lymphocytes, otherwise depletion of CD4 lymphocytes would occur very rapidly, and AIDS would occur within weeks. In the last few years, studies examining the production of new CD4 lymphocytes have provided strong evidence that there is no major increase, but rather a slight decrease, in the production of new CD4 lymphocytes in HIV+ subjects [10
11
12
].
Other studies have shown that HIV has an effect on resting CD4 lymphocytes, which would be the predominant cells surrounding any productively infected lymphocytes present in lymphoid tissues (as 95–99% of all lymphocytes are resting) [13
]. Virus has been shown to bind (and signal upon binding) and enter resting lymphocytes and to reverse transcribe partial or complete DNA proviruses. These do not integrate, however [14
]. This is a type of abortive infection, with the unintegrated viral DNA having a reported half-life of 6 h to about 1 week [14
, 15
]. If this abortively infected cell is activated by antigen into the cell cycle within this period, the virus can complete its replication cycle and produce progeny virions. The cell is then productively infected. The ratio of the frequencies of productively infected cells (7 per 106) in blood and lymph nodes to abortively infected cells (approximately 7500 per 106 cells) [16
] shows that greater than 99% of all HIV-infected cells in the body of an infected individual are abortively infected, resting lymphocytes. This should be expected, as most CD4 lymphocytes are resting. The binding of HIV to its cellular receptors (CD4 and chemokine receptors) signals these cells, and we showed that this results in up-regulation of L-selectin, the receptor for homing to lymph nodes [17
]. This receptor stays elevated for approximately 3 days, and these cells display enhanced binding (12-fold increase) to high endothelial venules in sections of lymph nodes and enhanced homing when injected into the blood of severe combined immunodeficiency (SCID) mice [17
]. During the homing process, a large number (40–50%) of these cells are induced into apoptosis after they enter the lymph node, which appears to be a result of secondary signaling through any of several homing receptors (minimally L-selectin, CD44, CD11a) [18
]. Thus, a scenario has arisen from these studies that depicts resting CD4 lymphocytes in lymphoid tissues coming into contact with HIV virions, productively infected cells, or HIV-coated, follicular dendritic cells, resulting in induction of a partially activated phenotype, including up-regulation of L-selectin and Fas. This is maintained for many days. Because of normal lymph node/blood circulation in which most lymphocytes in lymphoid tissues migrate back to the blood within 2 days [19
], many of these cells will end up back in the blood at the time of maximum-induced expression of L-selectin. These cells would then home very rapidly back to lymph nodes. Following transendothelial migration and entry into the lymph nodes, approximately half of them would be induced into apoptosis, and they do not produce HIV [20
].
This scenario can explain many important observations in HIV+ patients. 1) As CD4 lymphocytes disappear in the blood, their numbers do not drop, and the CD4/CD8 ratios do not invert in lymph nodes until late in disease. 2) There is no increased frequency of dying cells in the blood, but there is in lymph nodes. (Note that during later stages of HIV disease, the blood CD4 cells possess a greater propensity for spontaneous or activation-induced apoptosis when cultured in vitro [1
, 2
, 21
], but this may be due more to their chronic immune (antigen-mediated) activation status, which occurs in infected patients, than to direct HIV infection/replication. This propensity is not significant at early disease stages when blood CD4 cell levels are above 300/mm3, yet more than half of the blood CD4 cells have disappeared.) 3) Cells that are dying in lymph nodes are not making HIV, and they are dying by apoptosis. 4) The early increase of CD4 cells in the blood following highly active antiretroviral therapy (HAART) treatment appears to be because of redistribution from tissues. This would be expected if the disappearance of blood CD4 cells is mainly a result of HIV-induced, enhanced migration out of the blood [22
]. 5) Steroids, which are known to down-regulate L-selectin and retard lymph node homing, have been shown to retard or stop the disappearance of CD4 cells in the blood of HIV+ patients [23
, 24
].
More direct evidence is now needed to show whether this scenario actually occurs in HIV+ patients. The studies reported herein were initiated to examine the rates of migration (homing) of blood CD4 lymphocytes in vivo.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
View this table: [in a new window] |
Table 1. CD4 Cell Counts and Viral Loads of HIV+ Volunteers
|
The cells were then delivered to the Department of Nuclear Medicine (University of Texas Medical Branch, Galveston) for 111In labeling and injection. The saved autologous plasma was centrifuged at 2450 RPM for 20 min to produce platelet-poor plasma. The purified CD4 lymphocytes were resuspended in 6 ml 0.9% saline, and the solution was drawn up gently into a syringe and dispensed back into the tube. This process was repeated until the button of CD4 cells was completely dispersed. 111In oxine (1 mCi) was added drop by drop to the CD4 cells suspension, and the mixture of CD4 cells and 111In oxine was incubated for 30 min at room temperature. The mixture was gently agitated three to four times during the incubation period, and the CD4 cell/111In oxine mixture was brought up to a volume of 15 ml with an appropriate volume of platelet-poor plasma. The suspension was centrifuged at 1400 RPM for 5 min. After spinning, the supernatant was withdrawn from the tube using a syringe and spinal needle without disturbing the CD4 cell pellet. The supernatant, which contained the unincorporated 111In oxine, was discarded. Platelet-poor, autologous plasma (8 ml) was added to the CD4 cell platelet, and the cells were resuspended. An aliquot was analyzed on a 
counter to determine the efficiency of labeling.111In-labeled CD4 cells (500 µCi) were dispensed into a syringe and injected intravenously (i.v.; anticubital fosa) into the original donor. Thus, each donor received the same amount of radioactivity, even though the absolute numbers of labeled CD4 cells injected differed. In this manner, the amount of radioactivity accumulating in any given body site can be directly compared between different study subjects.
Camera and scanning
For the total body scintiphotos, the subjects were imaged, 1, 3, and 24 h post-injection with a dual-head camera (Vertex, ADAC Laboratories, Milpitas, CA), equipped with medium-energy, general-purpose collimators. Flood correction was done with In intrinsic floods, and two 20% energy windows at 173 and 250 KeV were used. For static planar view, 600 s/frame were collected on the chest and pelvic area. The matrix size was 256 x 256 x 6. The scan speed for the total body was 5 cm/min, and the matrix size was high-resolution, eight deep. Two experienced nuclear medicine physicians visually interpreted total body and planer scans from a computer display. Images were analyzed using the Pegasys processing terminals; irregular regions of interest (ROI) in each organ were drawn, duplicated, mirrored, and positioned on the sites of interest. Average counts in the ROIs were obtained as counts/pixel.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
camera, 1, 3, and 24 h after infusion, and localization of the CD4 cells was determined and quantitated.
Figure 1
shows whole-body scinto photos of one representative uninfected and one representative HIV+ volunteer at various time points post-infusion. The radioactivity detected by the camera was computer-digitized and is presented as colors (red>orange>yellow>green; blue is negative). The actual quantitative data, however, are presented in Table 2
. The lungs contain the greatest percentage of labeled cells at 1 and 3 h post-infusion in all subjects, but that is expected as the lung is the first major organ where the infused, labeled cells will travel, and a large percentage of them will stay in the lung for the first few hours. Another obvious finding is the increased intensity of labeled CD4 T cells in the vertebral and iliac bone marrow and cervical lymph nodes in the HIV+ patient compared with the control at 1 and 3 h. This demonstrates that more CD4 T cells migrated to these areas per unit time in the HIV+ subject compared with the uninfected subject. The two other HIV+ subjects tested displayed similar enhanced accumulation of CD4 T cells in the bone, compared with the two other control subjects (Fig. 2
Fig. 3
, Table 2
). It was also obvious that most of the labeled CD4 T cells in normal subjects migrated to bone marrow, and this was almost exclusively to vertebral and iliac marrow and not marrow in the long bones in both types of subjects. Also, the axillary lymph nodes of the HIV+ patient contained more labeled cells than the control subjects at 1 and 3 h post-infusion (observed best in Fig. 3
), and this was a statistically significant difference (P<0.01). By 24 h, the labeled cells appeared to have distributed homogenously, and there were no significant differences between control and HIV+ subjects at that time point.
 View larger version (110K): [in a new window] |
Figure 1. Whole body camera scinti photos localizing 111In-labeled CD4 lymphocytes in an uninfected and HIV-infected subject. Freshly isolated blood CD4 lymphocytes were labeled with 111In (500 µCi) in vitro, rinsed, and then i.v. reinfused back into the original donor. The subject was then scanned by a camera at 1 (A, D), 3 (B, E), and 24 (C, F) h post-infusion. The intensity of the signals in organs corresponds to the number of radioactive lymphocytes that migrated from the blood within the indicated time points. Computer digitization of the radioactivity is presented as the following colors: red > yellow > green > blue (negative). The images presented are 600-s exposures.
|
|
View this table: [in a new window] |
Table 2. Quantitation of 111In-Labeled CD4-T Cells in Various Organs at 1 and 3 h Post-Infusion
|
 View larger version (123K): [in a new window] |
Figure 2. Higher resolution of the hip/pelvic area of the three uninfected and three HIV-infected individuals at 3 h post-infusion. (A–C) Three uninfected subjects; (D–F) three HIV+ subjects. (A, D) The rectangles indicate the areas of bone marrow (BM) that were used for quantitation (data of Table 2
).
|
 View larger version (146K): [in a new window] |
Figure 3. Higher resolution of the chest/neck area. (A–C) Three uninfected subjects at 3 h post-infusion; (D–F) three HIV+ subjects at 3 h. (A, D) Circled areas indicate the axillary lymph nodes (ALN) and cervical lymph nodes (CLN) that were used for quantitation in the data of Table 2
.
|
As we compared HIV-infected to -uninfected subjects, we next examined if CD4 lymphocytes displayed enhanced migration in another viral (non-HIV) infection to ascertain whether enhanced, CD4 T cell homing is a common feature of other viral infections. It should not, as antigen activation of T cells induces shielding of L-selectin, and less-than-normal lymph node homing from the blood would be expected [26 ]. Two patients with chronic hepatitis B virus (HBV) infection volunteered for this part of the study. The results, shown in Table 2 , demonstrate that CD4 lymphocytes in the blood of these HBV-infected subjects migrated slightly less to bone marrow and considerably less to cervical and axillary lymph nodes in comparison with control subjects, as expected. We also did not see significantly increased accumulation of CD4 cells in the liver, which was surprising. Thus, the enhanced migration to lymph nodes and bone observed in HIV+ subjects appears somewhat specific for HIV infection with active viral replication, as determined by detectable virus on quantitative HIV RNA assays.
To further confirm this, we retested two of the HIV+ volunteers after they had been on HAART since their first scans. The virus load of volunteer #1 at the initial test had been 27,694 copies/ml and at repeat testing after 3 months on HAART, was <400 copies/ml. Volunteer #3’s virus load before HAART was 25,427 copies/µl and after 10 months on HAART, was 1050 copies/µl. Figure 4 shows the results and demonstrates that the CD4 T cells in the blood of both volunteers were now migrating at rates similar to those of uninfected subjects. The blood CD4 counts of volunteer #1 had also gone up from 503 to 692 cells/µl, and those of volunteer #3 rose from 594 to 707 cells/µl. Thus, the enhanced homing of CD4 lymphocytes in HIV+ patients appears to correlate with the presence of replicating HIV.
 View larger version (91K): [in a new window] |
Figure 4. Quantitation of CD4 T cell migration in two HIV+ patients before and after 3–10 months of HAART treatment. The radioactive counts per 460 pixels on 600-s scans are presented for bone and axillary lymph nodes at 1 h post-infusion of each uninfected (Uninf) subject and each HIV-infected subject before HAART and two of the HIV-infected subjects after 3–10 months on HAART.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Many recent studies examining the CD4 lymphocytes in HIV-positive subjects on HAART have shown that the first increase of CD4 lymphocytes is predominantly comprised of memory cells and not newly made naïve cells [22 ] and has been termed the "redistributive phase." This indicated that HIV was causing CD4 lymphocytes to be sequestered in or to migrate to nonvascular compartments. This strongly supported our earlier studies showing that HIV contact could directly induce resting CD4 lymphocytes to home to peripheral lymph nodes and the theory that the disappearance of CD4 lymphocytes from the blood was actually a result of their migration out of the blood [17 , 18 , 33 ]. The present study further supports this concept.
The situation now known to occur in HIV+ individuals where their blood CD4 lymphocytes are accumulating at about twofold increased rates in peripheral lymph nodes and bone marrow is likely because of the HIV-induced signaling of resting CD4 cells, which have come into contact with HIV-producing cells, virus-coated follicular dendritic cells, or cell-free virions [17
]. These effects are much more frequent on a cell-to-cell basis than the frequencies of productively infected cells (1000-fold) [16
]. Specifically, although 1 in 100,000 CD4 cells is productively infected making significant amounts of new progeny virus, between 0.8% and 1% of the CD4 lymphocytes possess unintegrated HIV proviruses [16
], and these are most likely resting cells that have come into contact with virus. It has been amply documented that when a resting CD4 lymphocyte is bound by HIV, the virus does enter (as well as signal through its cellular receptors) and reverse transcribes partial or complete provirus; however, this remains unintegrated [14
]. Thus, studies quantitating the frequencies of CD4 lymphocytes possessing unintegrated HIV proviral DNA are actually also quantitating the frequencies of resting CD4 cells that have been signaled by HIV contact. Therefore, the twofold, enhanced accumulation rate of CD4 T cells in the peripheral lymph nodes and axial bone marrow is presumably a result of the behavior of about 1% of the cells. A mathematical model of this HIV-induced, enhanced migration process affecting approximately 1% of the CD4 cells on a continuing basis has revealed that gradual depletion of CD4 would occur over an 8- to 10-year period and closely mimics the rate of CD4-cell depletion in AIDS patients [34
].
These findings that CD4 lymphocytes in HIV+ subjects migrate at enhanced rates to peripheral lymph nodes and bone marrow lend strong support to the hypothesis that a major pathogenic mechanism of HIV is its ability to signal resting CD4 cells and induce them to home to lymph nodes and die there. This hypothesis stems from previous studies, which show that HIV binding to resting CD4 lymphocytes induced increased expression of L-selectin and enhanced lymph node homing as judged following placement in the blood of SCID mice [17 ]. Other studies showed that many of these cells induced to home to lymph node are also induced into apoptosis via signals received during the homing process, and they do not make HIV [18 ]. This is identical to what is observed in HIV patients. (There is an increase of dying CD4 cells in lymph nodes, and they are not making virus; they have been termed "bystander" cells.) Thus, building on that background, it was important to show whether CD4 lymphocytes in HIV+ patients are actually migrating at enhanced rates or not, and that was the reason for this study. What is happening in the bone marrow of these patients remains unknown. Perhaps, they are induced into apoptosis, as they appear to be in lymph nodes [18 ]. Enhanced homing of CD4 cells from the blood into peripheral lymph nodes would not occur during normal antigen-mediated activation of CD4 cells, as that induces shedding of L-selectin, and these cells migrate to other locations [32 ]. It was not surprising, then, when we did not observe enhanced lymph node homing of CD4 cells in the HBV-infected subjects. Thus, our finding of increased lymph node homing of blood CD4 cells in HIV-infected individuals speaks against the "global immune activation and attendant activation-induced cell death" theory of HIV depletion of CD4 cells, as it would be expected that less-than-normal homing from blood to lymph nodes would occur in that situation. Furthermore, these new data do not support one of the other major competing theories on how HIV causes depletion of CD4 cells, which states that the virus is causing reduced production of new CD4 cells. Fewer new cells means fewer naïve cells, and naïve cells rapidly migrate (home) from the blood into lymph nodes. If that theory were true, one would expect to observe less-than-normal rates of blood CD4-cell homing into lymph nodes in HIV+ subjects.
Further studies testing the hypothesis that HIV causes depletion of CD4 cells via its effect on resting CD4 lymphocytes could examine whether inhibitors of lymph node homing or homing receptor induction of apoptosis would prevent depletion of CD4 lymphocytes in patients. Steroids, which down-regulate L-selectin expression on T cells, have been shown to stop CD4-cell depletion in the blood of patients [24 ]. However, as steroids have many effects, this correlation is not definitive. Newer compounds that are very specific for lymph node homing may prove to be informative. Also, it may be worthwhile to consider targeting these events for therapy.
Received November 17, 2001; revised March 25, 2002; accepted March 26, 2002.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. Ji and M. W. Cloyd HIV-1 binding to CD4 on CD4+CD25+ regulatory T cells enhances their suppressive function and induces them to home to, and accumulate in, peripheral and mucosal lymphoid tissues: an additional mechanism of immunosuppression Int. Immunol., March 1, 2009; 21(3): 283 - 294. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Ji, J. J-Y. Chen, V. L. Braciale, and M. W. Cloyd Apoptosis induced in HIV-1-exposed, resting CD4+ T cells subsequent to signaling through homing receptors is Fas/Fas ligand-mediated J. Leukoc. Biol., January 1, 2007; 81(1): 297 - 305. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. K. Sanghavi and T. A. Reinhart Increased Expression of TLR3 in Lymph Nodes during Simian Immunodeficiency Virus Infection: Implications for Inflammation and Immunodeficiency J. Immunol., October 15, 2005; 175(8): 5314 - 5323. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Mondal, C. A. Williams, M. Ali, M. Eilers, and K. C. Agrawal The HIV-1 Tat Protein Selectively Enhances CXCR4 and Inhibits CCR5 Expression in Megakaryocytic K562 Cells Experimental Biology and Medicine, October 1, 2005; 230(9): 631 - 644. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Foley, C.-R. Yu, R. Solow, M. Yacobucci, K. W. C. Peden, and J. M. Farber Roles for CXC Chemokine Ligands 10 and 11 in Recruiting CD4+ T Cells to HIV-1-Infected Monocyte-Derived Macrophages, Dendritic Cells, and Lymph Nodes J. Immunol., April 15, 2005; 174(8): 4892 - 4900. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. K. Choi, B. A. Fallert, M. A. Murphey-Corb, and T. A. Reinhart Simian immunodeficiency virus dramatically alters expression of homeostatic chemokines and dendritic cell markers during infection in vivo Blood, March 1, 2003; 101(5): 1684 - 1691. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
