| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Published online before print October 4, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,1
Departments of
* Microbiology and Immunology and
Pharmacology and
Center for Substance Abuse Research, Temple University School of Medicine, Philadelphia, Pennsylvania
1 Correspondence: Department of Microbiology and Immunology, Temple University School of Medicine, 3400 North Broad Street, Philadelphia, PA 19140. E-mail: tke{at}temple.edu
ABSTRACT
We have previously shown that abrupt withdrawal (AW) from morphine induces greater than 80% immunosuppression in murine spleen cells, as assessed by the capacity to mount an in vitro plaque-forming cell response to sheep red blood cells. Present studies about the mechanisms of immunosuppression following AW showed that addition of highly enriched (CD11b+) splenic macrophages (obtained by cell sorting or magnetic separation) from AW mice to cultures of normal, unfractionated spleen cells suppressed immune responses. Further, addition of highly enriched (CD19+) B cells (but not T cells) from AW mice to normal cells was also immunosuppressive. B cells from AW mice were also able to inhibit the proliferative response of normal spleen cells to concanavalin A but not to lipopolysaccharide. Overall, the data suggest that immunosuppression by AW spleen cells is a result of active suppression by macrophages and B cells.
Key Words: neuroimmunology • opioids • plaque-forming cell • suppressor cells • Con A
INTRODUCTION
The morphine abstinence syndrome, as it pertains to the nervous system, has been well characterized pharmacologically and biochemically in a substantial literature [1
2
3
4
5
6
7
8
]. In contrast, there are only a few publications that have examined the effects of precipitated withdrawal [9
, 10
] or abrupt withdrawal (AW) [10
11
12
13
] from morphine on parameters of the immune system. In regard to AW, Bhargava et al. [11] showed that murine splenic cytotoxic T cell activity, production of interleukin (IL)-2, and B cell proliferation in response to treatment with anti-immunoglobulin M (anti-IgM)/IL-4 were inhibited significantly at 8 h post-AW. A deficit in splenic immune function was also reported by West et al. [13], who showed that rats subjected to AW had depressed proliferative responses to T cell mitogens, production of interferon-
(IFN-), and natural killer (NK) cell activity. Suppression of these immune functions was maximal at 12 h postwithdrawal with a return to normal levels by 24 h, except for NK cell activity, which was still suppressed at 48 h postwithdrawal. Our laboratory was the first to study the effects of AW from morphine on splenic antibody responses in mice [10
]. We showed that AW, by removal of morphine pellets from dependent animals, resulted in profound immunosuppression, which is maximal at 24–48 h postpellet removal and is still present at 144 h. In human studies, Govitrapong et al. [12] showed that AW from heroin induced perturbations in the (peripheral blood) CD4/CD8 ratio and percentages of B cells and NK cells, with an inability to regain normal cellular levels until 3 years postwithdrawal. The above findings suggest that AW can lead to suppression of a number of parameters of the immune system with the possibility of a protracted effect.
Our previously published study documented the effects of withdrawal on the splenic plaque-forming cell (PFC) response [10 ] but did not address the mechanisms mediating the immunosuppression. It is interesting that none of the other studies (mentioned above) has addressed the immune mechanisms, which may mediate the inhibition of normal responses following withdrawal or the potential factors that contribute to the sustained immunosuppressive effects of AW from morphine. In the present study, we used the PFC assay to determine whether cells from mice subjected to AW mediate immunosuppression by activation of a population of suppressor cells or production of a suppressor factor. It was observed that macrophages and B cells, taken from mice undergoing AW, inhibit immune responses by normal spleen cells in vitro.
MATERIALS AND METHODS
Mice
Six-week-old C3HeB/FeJ female mice were purchased from Jackson Laboratories (Bar Harbor, ME) and were housed in sterilized cages with mouse chow and water provided ad libitum. All mice were acclimated for a minimum of 1 week prior to being used in experiments.
Reagents
Rat anti-mouse CD11b-phycoerythrin (PE), CD3
-peridinin chlorophyll protein (PerCP), and CD19-allophycocyanin (APC) were purchased from PharMingen (San Diego, CA). Concanavalin A (Con A) and lipopolysaccharide (LPS; Escherichia coli O127:B8) were purchased from Sigma Chemical Co. (St. Louis, MO).
Surgery and experimental design
Mice were anesthetized with isoflurane, and an area of the back was shaved. A 1-cm incision was made in the skin, and mice were implanted subcutaneously (s.c.) with a 75-mg morphine pellet [National Institute on Drug Abuse (NIDA), Rockville, MD] or a cellulose placebo pellet (placebo controls) encased in a 1 x 1-cm square nylon mesh bag to permit easy removal, according to standard procedures in our laboratories. Other control/normal mice had no surgery and received no drugs. Based on published results [10
, 14
], it was established that mice were tolerant to the immunosuppressive effects of morphine by 96–120 h postmorphine pellet implantation. In all experiments, pellets were left in place for 96 h before initiation of AW. For AW, morphine pellets were surgically removed from dependent mice, which were killed at 24 h after pellet removal, and the number of antibody-forming cells was determined using the PFC assay (see below).
Primary in vitro antibody response
Normal mice and animals undergoing withdrawal were killed, and the number of antibody-forming cells was determined using a PFC assay [15
]. Four mice per group were killed, and their spleens were removed. Spleen cells from two mice in each group were pooled to obtain two spleen pools per treatment. Immune function was assessed using an in vitro PFC assay, described previously, which measures the capacity of spleen cells to mount a primary antibody response to sheep red blood cells (SRBCs). A single-cell suspension of spleen cells was obtained by pushing the spleen through nylon mesh bags in RPMI supplemented with 5% fetal calf serum (FCS). RBCs were lysed by hypotonic shock with sterile water. Cells were washed twice and resuspended in tissue-culture medium consisting of RPMI 1640 supplemented with 10% FCS, 2 mM L-glutamine, 50 µg/ml gentamicin, 1 mM nonessential amino acids, 1 mM sodium pyruvate, 10 µg/ml adenosine, uridine, cytosine, and guanosine, and 0.05 mM 2-mercaptoethanol. The cells were counted using a particle counter (Model Z1, Beckman Coulter, Fullerton, CA), resuspended to 1.0 x 107 cells/ml, and dispensed into flat-bottom 24-well tissue-culture plates. SRBCs (Rockland, Gilbertsville, PA) were washed three times in sterile saline, and 3.5 x 106 cells were added to each well in 50 µl. For each cohort, a minimum of triplicate cultures was tested, each with its own negative control (i.e., splenocytes without SRBCs). Cultures were incubated in an atmosphere consisting of 7% O2, 10% CO2, and 83% N2 at 37ºC. Twenty-four hours later, cultures were fed with nutrient cocktail consisting of RPMI 1640 supplemented with 
30% FCS, 5.5 mM L-glutamine, 3 mM each essential amino acids, and nonessential amino acids, 5.5 mg/ml dextrose, 0.61% sodium bicarbonate, and 40 mg/ml each adenosine, uridine, cytosine, and guanosine. On Day 5, cells were harvested and washed in RPMI, and the number of direct PFCs (cells producing IgM antibodies against SRBCs) was quantitated using the Cunningham and Szenberg [16] modification of the Jerne hemolytic plaque assay. Data from the PFC assay were calculated as the mean number of PFC/107 cells. Each datum point represents a minimum of two spleen pools, and each pool was tested in at least triplicate. Many points are the average ± SE of data obtained from multiple experiments with several pools in each group in each experiment. For cocultures, spleen cells from mice undergoing withdrawal were coincubated with spleen cells from normal mice in a 1:5 ratio as described previously [14
]. Briefly, 2.5 or 5 x 106 spleen cells from AW mice (1x107 cells/ml) were added to 1 x 107 normal spleen cells. AW spleen cells were left unfractionated or separated into specific populations of macrophages, B cells, and T cells by magnetic separation or cell sorting (Fig. 1
).
|
CD3-PE followed by incubation with anti-PE microbeads. The cells to be separated were then loaded onto a magnetic separation unit, composed of a column held in place by a MACS® magnet. The column was then washed three times with cold PEB, and the magnetically labeled cells were retained in the column, which was removed from the unit, and the positively selected cells were eluted off with 5 ml PEB onto another column held in place by the magnet (for a second enrichment step). The column was washed in PEB; positively selected cells were again collected. For isolation of highly enriched populations of B and T cells (see Fig. 3
), CD11b– cells were first collected by negative selection using CD11b microbeads followed by positive selection with B or T cell-specific microbeads (see Fig. 1
). Negatively selected cells (washed off the separation column while it is held in place by the magnet) and positively selected cells were adjusted to 1 x 107 cells per ml using the particle counter and were used in cocultures in 24-well tissue-culture plates. Purity of the fractionated cells was assessed by flow cytometry (see Table 1
).
|
|
CD11b-PE), T cells (CD3-PerCP), and B cells (CD19-APC). After incubation, the cells were washed two times with PBS-BSA and resuspended in 300 µl PBS-BSA. For each cell-surface marker, cells were sorted (at a rate of 2.5x106 cells/h) on a MoFlo cell sorter (Cytomation Inc., Fort Collins, CO) at the Wistar Institute (Philadelphia, PA). Purity was checked by flow cytometry and was >99%. Sorted cells were collected in 12 x 75 mm snap-cap polypropylene tubes (Falcon, Becton Dickinson), filled with culture medium, supplemented with 50% FCS. Cells were then washed, resuspended in tissue-culture medium, and then used for cocultures in 24-well tissue-culture plates.
Proliferative response to mitogens
Spleen cells were cultured in a volume of 100 µl RPMI (without serum, supplemented with 2 mM L-glutamine) in flat-bottom 96-well plates (Costar, Corning, NY). Normal cells or cells from mice undergoing AW were plated at 8 x 105 or 1.0 x 106 cells/100 µl. For cocultures, 2 x 105 CD19+ B cells from mice undergoing AW for 24 h, purified by magnetic separation, were added to 6 x 105 or 8 x 105 normal cells. Con A (0.1 µg/well) was added in 50 µl RPMI. All wells were brought to a final volume of 200 µl with medium. Cells were incubated for 48 h in 5% CO2, and 3H-thymidine (0.5 µCi in 50 µl) was then added. Eighteen hours later, cells were harvested onto glass fiber filters (Packard, Downers Grove, IL) using a Packard multichannel harvester. Thymidine uptake was determined by counting filters, which were placed in liquid scintillation solution (Cytoscint, MP-Biomedical, Irvine, CA) using a Packard liquid scintillation analyzer. Data are from triplicate cultures expressed as mean counts per minute (cpm) ± SE, corrected for background by subtraction of cpm in the absence of mitogen.
Statistical analysis
Plaque counts and mitogenicity data were analyzed using a mixed model ANOVA followed by pair-wise group comparisons using the Dunn-Bonferoni method adjusting for multiple comparisons and random experimental effects. Differences were considered significant for P values 
0.05. NS = comparisons that were not significant.
RESULTS
AW spleen cells can suppress normal immune responses
Spleen cells taken from mice at 24 h post-AW were suppressed profoundly in capacity to mount an in vitro PFC response to SRBCs (Fig. 2
). The hypothesis was tested that immunosuppression following withdrawal could result from activation of a population of suppressor cells, which can inhibit immune function. To test this possibility, spleen cells from mice undergoing AW were cocultured with spleen cells from normal mice. Addition of 2.5 or 5 x 106 AW cells to normal cells led to suppression of normal PFC responses, with a marked effect at the highest concentration of AW cells (Fig. 2)
. Coculture with 5 x 106 AW cells led to a decrease in the PFC response, which was comparable with PFC levels observed from AW cells alone. Addition of equal numbers of normal cells did not have a similar effect. These results suggest that AW cells contain a population of suppressor cells. In the following studies, the contribution of macrophages, B cells, or T cells to the immunosuppression was investigated using highly enriched or purified cell subpopulations from spleens of withdrawn mice.
|
AW spleen cells contain populations of suppressor macrophages
We further examined whether the suppressive activity of spleen cells from AW mice is mediated by macrophages. For these experiments, spleen cell fractions from mice undergoing withdrawal were titrated into cocultures with spleen cells from normal mice. Addition of AW CD11b+ macrophages, purified by magnetic separation (Fig. 4A
) or by cell sorting (Fig. 4B)
to spleen cells from normal mice, suppressed PFC responses. The data show that normal macrophages are not suppressive at any of the concentrations tested, indicating that the magnetic bead separation did not activate the cells. Magnetically purified macrophages gave dose-dependent immunosuppression, with the highest observed in cultures of AW cells alone (Fig. 4A) . Normal CD11b+ cells fractionated by flow cytometry were also not suppressive. Flow-sorted CD11b+ cells from AW mice gave 60% suppression, which is comparable with the level of suppression observed in cocultures containing unfractionated AW cells (Fig. 3)
. These results support the conclusion that populations of macrophages mediate the immunosuppression of antibody responses by AW spleen cells.
|
|
|
DISCUSSION
This study is the first to show that spleen cells from AW mice mediate immunosuppression by the activity of suppressor macrophages and B cells. Splenic populations of CD11b+ macrophages, CD19+ B cells, and CD90+ or CD3+ T cells were isolated by magnetic cell sorting (90% purity) and flow cytometric cell sorting (99% purity). The results suggest that the inhibition of IgM antibody responses is not a result of the presence of suppressor T cells, as neither CD90+ nor CD3+ cells from AW mice were able to inhibit normal PFC responses. In contrast, addition of unfractionated spleen cells or purified populations of AW CD11b+ macrophages or CD19+ B cells to whole spleen cells from control mice led to suppression of antibody responses. In addition, AW CD19+ B cells suppressed normal mitogenic responses to Con A but not to LPS.
Macrophages, under various activating conditions, can be potent suppressors of immune responses [17 ]. In the literature, suppressor macrophages have been shown to secrete several inhibitory factors including nitric oxide (NO), prostaglandins, and transforming growth factor-ß (TGF-ß), which has been identified as a suppressor factor following acute exposure of human peripheral blood mononuclear cells to morphine in vitro [18 ], and NO has been identified as the suppressor factor in rat spleens after a single s.c. injection of morphine [19 , 20 ]. We have carried out preliminary studies to explore the mechanism by which macrophages from AW mice suppress normal spleen cells. We found that NG-monomethyl-L-arginine, an inhibitor of NO synthase, reversed the suppression observed in cocultures of normal and AW cells (data not shown). However, in extensive experiments, NO was not detected in the culture supernatants containing suppressor cells. We are continuing to explore the reason for the discrepancy in the results using the inhibitor and by direct measurement of NO. We also tested the effect of a monoclonal anti-TGF-ß in reversing the immunosuppression induced by adding unfractionated AW cells to normal cells in coculture and found no effect of the antibody at concentrations up to six times the neutralizing dose 50%, as recommended by the supplier (data not shown).
The present studies also show that CD19+ B cells from mice undergoing AW are able to suppress normal PFC responses. In the literature, suppressor B cell activity has been shown to inhibit antibody responses to SRBCs [21 22 23 ], proliferative responses to Con A [21 ], delayed-type hypersensitivity responses to LPS and SRBCs [24 , 25 ], and IL-2 consumption in spleen cells from Mycobacterium bovis bacillus Calmette-Guerin-infected mice [26 ]. The mechanism of induction and action of these suppressor B cells was never defined. However, Cuff et al. [22] showed that primary and secondary T cell-dependent antibody responses are inhibited by splenic suppressor B cells and suggested that suppressor B cells may regulate T cell activity directly. In the present study, coculture of normal spleen cells with B cells from AW mice led to suppression of proliferative responses to the T cell mitogen Con A but not to the B cell mitogen LPS. These observations on responses to mitogens support the existence of a B cell population with suppressor activity, as cocultures of B cells from AW spleens suppressed responses of normal spleen cells to Con A. At present, we do not have a mechanism for the suppressor activity of the B cells and are unaware of any mechanisms that have been postulated in the literature. The suppression does not seem to be a simple dilution effect of nonresponsive AW B cells on the normal spleen cultures, as the AW B cells constitute only 20–25% of the total cells in the culture, but the responses are inhibited by 64% (Fig. 6 ). The fact that the AW suppressor B cells inhibit responses to Con A but not to LPS perhaps suggests that the B cells are acting on accessory cells. Mitogen responses to LPS do not require macrophage cooperation, but responses of T cells to Con A do need an antigen-presenting cell [27 ]. A consideration is whether the suppressor activity of the B cell population could be a result of a contaminating population of suppressor macrophages. We have tried to rule out this possibility by purifying the AW B cells by two different methods (see Fig. 1 ) and by flow cytometric cell sorting. The sorted B cell population is 99% pure and is still suppressive. In addition, fluorescein-activated cell sorter analysis of the various populations (Table 1) shows that the T cell fractions, obtained by magnetic bead purification, contain a similar level of macrophage contamination as the B cells, but the T cells are not immunosuppressive.
Taken together, the results presented in this paper provide strong evidence for the existence of populations of suppressor macrophages and B cells. It should be noted that other observations from our laboratory indicate that AW spleen cells, freshly isolated at 24 h postwithdrawal, have depressed levels of mRNA and protein for proinflammatory cytokines such as tumor necrosis factor and IL-1, and macrophage cytokine production is deficient [28
]. Addition of normal spleen cells or IL-1 or IFN- to AW cells reversed the immunosuppression [28
]. A problem for future research is to reconcile these observations with the results of the present paper. Our current hypothesis for reconciling what appear to be conflicting results is that coculture of AW cells with normal cells in some way activates them. Future experiments need to compare the activation state of AW cells alone with AW cells in coculture, which will require a way to label, retrieve, and examine the AW cells once they have been in contact with normal cells [28
].
In the present study, we have examined the immune mechanisms that mediate suppression of murine splenic antibody responses by spleen cells from AW mice. The results show that at 24 h post-AW, immunosuppression is mediated by activation of a population of suppressor macrophages and the activity of suppressor B cells. Previously, our laboratory has shown that AW mice have a slow rate of recovery from withdrawal-induced immunosuppression [10 ]. Results from the present study suggest that following AW in vivo, the restoration of normal immune responses may be inhibited by the activity of populations of suppressor cells. A delay in the rate of recovery of immune responses following the morphine-abstinence syndrome could be deleterious to the host by providing a window for enhanced susceptibility to infection or an opportunity for reactivation of latent infections.
ACKNOWLEDGEMENTS
This work was supported by NIDA Grants DA14223, DA11134, and DA13429. We thank Dr. John P. Gaughan for assistance with statistical analyses and Alistaire S. Acosta and Jeffrey S. Faust for operation of flow cytometric and cell-sorting equipment.
Received March 3, 2004; revised August 22, 2005; accepted August 30, 2005.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or normal (
) CD11b+ macrophages isolated by magnetic separation. *P < 0.05 coculture with AW CD11b+ cells (2.5x106) versus coculture with normal CD11b+ cells (2.5x106). (B) Cocultures contained 1 x 107 normal cells with addition of 2.5 x 106 normal (shaded bar) or AW CD11b+ macrophages (hatched bar) isolated by cell sorting. Antibody responses are expressed as the mean number of PFC/107 cells ± SE. Each point represents data from three experiments, with each experiment done in triplicate. *P < 0.05 coculture with AW CD11b+ cells (hatched bar) versus coculture with normal CD11b+ cells (shaded bar). *P < 0.001 AW alone or cocultures with AW CD11b+ cells (hatched bar) versus N (solid bar).
