Originally published online as doi:10.1189/jlb.0204085 on July 26, 2004
Published online before print July 26, 2004
(Journal of Leukocyte Biology. 2004;76:835-844.)
© 2004
by Society for Leukocyte Biology
Antisecretory factor expression is regulated by inflammatory mediators and influences the severity of experimental autoimmune encephalomyelitis
Todd S. Davidson and
William F. Hickey1
Department of Pathology Borwell Building Dartmouth Medical School – DHMC Lebanon, NH 03756
1 Correspondence: Department of Pathology, DHMC Lebanon, NH 03756. E-mail: william.f.hickey{at}dartmouth.edu
 |
ABSTRACT
|
|---|
Antisecretory factor (ASF) was originally identified as a potent inhibitor of intestinal fluid secretion induced by a number of enterotoxins. In addition to its involvement in intestinal fluid secretion, ASF modulates the proliferation of memory/effector T cells and is expressed by cells of the immune system. This report describes the role of ASF in modulating immune responses and assesses the regulation of ASF during an in vivo immunological reaction. ASF expression was redistributed during adoptively transferred experimental autoimmune encephalomyelitis (EAE), and in response to other inflammatory stimuli. Administration of the anti-ASF antibody TLD-1A8A increased the clinical severity and duration of the disease. Consistent with these findings, addition of TLD-1A8A to T cell proliferation assays resulted in up-regulation of the proinflammatory cytokines IL-18 and IL-6 and in down-regulation of IL-10. Furthermore, we identified cytokines that regulated the expression of ASF at both the mRNA and protein level. ASF, therefore, appears to play a previously unappreciated and potentially important role in the regulation of immune responses.
Key Words: macrophage • T lymphocyte • cytokine • immunosuppression
|
INTRODUCTION
|
|---|
The development of inflammatory processes in the central nervous system involves a network of cells and molecules. Key central nervous system (CNS) resident cell types involved in these processes are perivascular cells (PVCs) and microglia; both of which are bone marrow-derived populations that serve as the immunological sentinels of the CNS [1
]. Among the many functions of these cells is their ability to present antigen to infiltrating T cells and thus control the initiation of an immunological response in situ [2
]. Although a number of molecules known to be involved in T cell-antigen presenting cell (APC) interactions have been demonstrated to be expressed by PVCs and microglia, we have undertaken studies to identify novel molecules expressed by these cells that participate in these interactions [3
, 4
]. The TLD antibodies are a panel of monoclonal antibodies generated by immunization of mice with living rat microglial cells [5
]. One such antibody from this panel, clone name TLD-1A8A, is specific for Antisecretory Factor (ASF) [6
].
ASF is a molecule that was originally isolated biochemically by Lönnroth et. al. and shown to be a potent inhibitor of intestinal fluid secretion [7
]. ASF is expressed by macrophages and is present in lymphoid organs, including gut-associated lymphoid tissue, lymph node, spleen, and thymus [6
]. Although cell surface expression of ASF on cultured microglia cells is low, immunohistochemistry nonetheless demonstrated considerable expression of the molecule by PVCs in the CNS. TLD-1A8A increased in vitro T cell proliferation, indicating a role for ASF in regulating T cell responses. Thus, the binding of the antibody to its ligand, be it a neutralizing or activating interaction, resulted in a measurable stimulatory change of an in vitro immunological response. However, it remained unclear what, if any, significance this may have in vivo. Furthermore, it remained unclear how ASF interacts with other immunological mediators. Specifically, what molecules may be affected by the expression of ASF, and conversely, what molecules may influence the expression of the ASF molecule itself.
Experimental autoimmune encephalomyelitis (EAE) is an inflammatory disease of the CNS that has served as an animal model of the human disease multiple sclerosis. EAE can be induced either actively, by immunization with CNS components, or by adoptive transfer of activated Th1 cells specific for CNS components. The pathological manifestations of the disease occur when activated T cells enter the CNS, recognize their cognate antigen in the context of an antigen presenting cell, and then recruit additional leukocytes to the CNS resulting in inflammation and tissue damage [8
]. Monocytes/macrophages are a critical element in the inflammatory infiltrate; their number and activation level are directly related to the severity of clinically evident disease. As is apparent, members of the monocyte/macrophages/microglia family, and their ability to interact with inflammatory T cells, are critical in the development of EAE. Thus, EAE can serve not only as a model of multiple sclerosis, but also as a more general model of T cell-mediated inflammatory responses in the nervous system.
|
MATERIALS AND METHODS
|
|---|
Immunohistochemistry
Animals were killed by anesthesia overdose. Organs were removed, embedded in Tissue-Tek OCT medium (Miles, Elkhart, IN), and snap-frozen. Sections (6-µm thick) were cut, mounted onto glass slides, fixed briefly with methanol, and put into a blocking solution of 10% FBS in 0.5M Tris, pH 7.6. The tissue sections were then covered with the antibody solution and incubated overnight at 4°C. The next day the sections were washed with blocking solution, then overlaid with rat-absorbed, biotinylated anti-mouse antibody (Vector Laboratories, Burlingame, CA) at room temperature for 45 min. Following the incubation, the sections were again washed with blocking solution and then placed into a solution of absolute methanol, 6% H2O2, for 15 min to remove endogenous peroxidase activity. The sections were then washed with Tris buffer, followed by blocking solution, then overlaid with ABC reagent (Vector) for 1–1.5 h at room temperature. Following washes in blocking solution, color was developed using 3,3'-diaminobenzidine in the presence of H2O2. The sections were then dehydrated through a series of ethanol solutions and xylene, cover-slipped, and then examined by light microscopy.
Cell culture
Peritoneal exudate cells (PECs) were obtained by peritoneal lavage with ice-cold PBS of animals that had been challenged five days previously by intraperitoneal injection of 3 ml of aged 10% brewer’s thioglycolate (Difco, Detroit, MI). Resident peritoneal macrophages were obtained by peritoneal lavage with ice-cold PBS of animals that had received no previous manipulation. Splenocytes were obtained by mechanical disruption of aseptically recovered spleens from naive animals followed by extensive washing with ice-cold PBS. NR8383 cells were obtained from the American Type Culture Collection (Manassas, VA) and maintained in Ham’s F12 medium supplemented with 15% FBS and 5% NCTC-109, 5 x 10–5M 2-mercaptoethanol, 2 mM glutamine, 100 µg/ml penicillin, 100 µg/ml streptomycin, and 100 µg/ml amphotericin B.
Quantitative real-time PCR
Total RNA was obtained from cells or tissues using the Trizol® reagent (Life Technologies) according to the manufacturer’s directions. RNA (500 ng) was reversed transcribed using an oligo-d(T)16 primer and MuLV reverse transcriptase at 42°C for 15 min, followed by heat inactivation of the enzyme at 99°C for 5 min. Quantitative real-time PCR was then performed on a Perkin Elmer 7600 using the SYBR® Green QPCR master mix kit from Stratagene. Actin was amplified using the forward primer 5'-TACAACCTCCTTGCAGCTCC-3' and reverse primer 5'-GGATCTTCATGAGGTAGTCTGTC-3'. Antisecretory Factor was amplified using the forward primer 5'-CATCTGGCTCTGAAGCACCG-3' and reverse primer 5'-CGAATGGCTTCATTGTTGGG-3'. After 10 min at 95°C, the sample was amplified for 40 cycles using the following conditions: 95°C for 30 s, 55°C for 1 min, 72°C for 1 min 30 s; followed by 3 min at 72°C. The threshold value Ct, at which fluorescence became detectable, was calculated for each sample. The fold change (FC) in gene expression was calculated using the formula FC = 2-
Ct where Ct = CtExperimental – CtControl and Ct = CtASF – CtActin. Depending upon the logistical concerns of the experiment, samples were analyzed in duplicate or triplicate.
Flow cytometry
Cells for flow cytometric analysis were washed with ice-cold PBS, resuspended in 10% FBS-PBS at 1 x 106 cells/ml, dispensed into Falcon 2059 tubes at 5 x 105 cells per test, and incubated on ice for 30–45 min. For intracellular analysis, 0.2% saponin was included. The cells were then centrifuged, resuspended in the appropriate antibody solution, and incubated on ice for 30–45 min. Antibodies were in the form of exhausted culture supernatants that were diluted with an equal volume of 10% FBS-PBS (including 0.2% saponin for intra-cellular analysis, as appropriate). Controls consisted of no antibody, isotype control primary plus secondary antibody, and secondary antibody alone. The cells were then washed twice with PBS, resuspended in 200 µl of a 1:200 dilution of goat anti-mouse FITC (Sigma, St. Louis, MO) diluted in 10% FBS-PBS (including 0.2% saponin for intra-cellular analysis, as appropriate), and incubated on ice for 30–45 min. The cells were then washed three times with PBS, fixed in 2% paraformaldehyde, and stored in the dark at 4°C until analyzed. Cells were analyzed on a FACScan (Becton Dickinson). Five thousand events per sample were collected.
Confocal microscopy
Staining was performed as for flow cytometry, with the inclusion of 0.2% saponin to permeabilize the cells. After washing and fixing, 10 µl of sample was pipeted onto a microscope slide and allowed to dry. Cover slips were mounted using Anti-fade mounting medium (Molecular Probes, Eugene, OR) and allowed to set. Images were obtained on a Zeiss confocal microscope and analyzed using Adobe Photoshop.
T cell proliferation assay and gene expression profiling
Short-term antigen specific T cell lines were generated as described previously [3
]. The encephalitogenic peptide, EP, of guinea pig myelin basic protein GP68-88 (Y G S L P Q K S Q R S Q D E N P V V H) [9
] was dissolved in PBS and emulsified with an equal amount of complete Freund’s adjuvant supplemented with 5 mg/mL Mycobacterium tuberculosis strain H37RA at a final concentration of 1 mg/ml EP. The emulsion was injected intradermally into the footpad and at the base of the tail of female Lewis rats at 200 µg EP per animal. On Day 9, the animals were killed via anesthesia overdose, and the draining popliteal and inguinal lymph nodes were aseptically removed. Single cells were isolated from the lymph nodes by mechanical disruption and placed into culture with 50 µg/ml EP in RPMI-1640 medium supplemented with 1% rat serum, 5% NCTC-109, 5 x 10–5M 2-mercaptoethanol, 2 mM glutamine, 100 µg/ml penicillin, 100 µg/ml streptomycin, and 100 µg/ml amphotericin B (T cell initiation/proliferation medium). On the third day of culture, dead cells were removed using Histopaque 1.077 (Sigma) and antigen specific blasts were recovered, washed with PBS, and returned to culture without additional antigen in similar medium with the exception of 10% FBS (HyClone Laboratories, Logan, UT) replacing the 1% rat serum as well as supplementation of 5% exhausted medium from ConA stimulated splenocytes (T cell expansion medium). The cells were split as necessary and allowed to return to resting phase. Antigen-specific T cells were then recovered by centrifugation, washed with PBS, and plated with irradiated (1500 rads) lymph node cells to serve as APC, in T25 flasks (Falcon, Lincoln Park, NJ). Antigen was added at 50 µg/ml, and antibodies were added at 20 µg/ml. At 48 h, total RNA was obtained from the cells using the Trizol® reagent (Life Technologies) according to the manufacturer’s directions. Expression profiles were determined by using the SuperArray GEArrayTM Q series mouse common cytokine kit. The probes used in this kit have been determined by the manufacturer to be cross-reactive with rat. Briefly, 1 µg of total RNA was reversed transcribed in the presence of [
-33P]-dCTP (NEN) at 42°C for 25 min. The probe was then denatured at 94°C for 5 min and hybridized to the membrane at 60°C overnight with continuous agitation. Following washes, the signal was detected with a phosphor imager and analyzed with densitometry.
Induction of EAE by adoptive transfer
EAE was induced in female Lewis rats by tail-vein injection of EP-specific T cells that had stimulated for 72 h in the presence of 5 µg/ml Concanavalin A (Sigma) and irradiated (1500 rads) syngeneic splenocytes. T cells were derived as described for T cell proliferation assays and carried for at least two restimulations with antigen before stimulation with ConA [10
]. Activated T cells were injected at 10–25 x 106 cells per animal in a volume of 2 ml of PBS. Antibodies, Protein G (Sigma) purified TLD-1A8A or MOPC-31 (isotype control; Sigma) were suspended at a concentration of 0.5 mg/ml in PBS and 2 ml per animal (1 mg Ab per animal) were injected into the tail vein at least 1 h before injection of the cells on Day 0 and again on Day 2. The animals were weighed daily and scored for clinical symptoms of EAE according to the following scale: 0 = no symptoms of disease, 1 = flaccid tail, 2 = complete hind-limb paralysis, 3 = total paralysis/moribund [10
]. Each experiment consisted of a minimum of three animals per group. Clinical scores are expressed as a mean score of all three animals during the course of the experiment. Weights are expressed as the average percentage of weight lost compared with the average weight of the animals within the group on Day 0.
|
RESULTS
|
|---|
ASF expression changes in time in the CNS during EAE
Lewis rats were given adoptively transferred EAE as described in Materials and Methods and killed at 24 h intervals starting one day following the injection of T cells. Spinal cords, spleens, and submandibular lymph nodes were collected for immunohistochemical analysis. Spinal cords examined on Day 1 post-transfer, clinical score 0, only expressed ASF in the perivascular area (Fig. 1a
). Spinal cords examined on Day 2, also clinical score 0, showed the same pattern of staining. By Day 3, although clinical signs of disease were still absent, the staining intensity of perivascular cells appeared to increase, while infiltrating macrophages expressing ASF also began to appear. Spinal cords examined on Day 4 from rats with clinical scores of 1 to 1.5, showed a greatly increased number of infiltrating macrophages expressing ASF. In addition, perivascular cells showed elevated ASF expression relative to uninflammed tissue, and staining on microglia became apparent. Spinal cords from Days 5 and 6 from rats with clinical scores of 2 and 2.5, respectively, showed similar staining patterns. However, by Day 7, as EAE was resolving, perivascular cells had maintained their high expression levels of ASF, while infiltrating macrophages expressing the molecule were greatly reduced in number, though still present. A similar staining pattern was apparent on Day 8 from rats with a clinical score of 1. By Day 9, the animals had fully recovered. Remarkably, perivascular cells from these spinal cords maintained their high expression levels of ASF, while only an occasional ASF -positive infiltrating macrophage could be observed.
View larger version (55K):
[in this window]
[in a new window]
|
Figure 1. Kinetics of ASF Expression During EAE. Lewis rats were given encephalitogenic T cells intravenously and were killed at one-day intervals starting one day post-injection. Tissues were collected and processed for immunohistochemistry (a) or quantitative real-time PCR (b). a) TLD-1A8A staining of spinal cord from the indicated day, 20x objective. Sections stained with an isotype control antibody showed no detectable staining (data not shown). b) QPCR analysis of ASF mRNA levels, results are expressed as fold change relative to d0 levels, P = 0.0008, ANOVA.
|
|
Quantitative real-time PCR analysis (QPCR) was performed on RNA derived from spinal cords of animals with EAE. ASF mRNA levels were normalized to actin mRNA levels, and the ASF mRNA level at Day 1 was defined as 1. As shown in Fig. 1b
, on Day 2 ASF mRNA levels had dropped twofold, recovered to baseline levels by Day 3 and peaked at nearly fourfold above baseline levels by Day 5, which coincides to the peak of disease. Subsequently, ASF message levels steadily decreased back to baseline levels of expression by Day 9. No identifiable changes were detected in the immunohistochemical distribution of ASF in spleens or lymph nodes throughout the course of disease (data not shown).
Inflammatory stimuli lead to redistribution of ASF
To determine whether other proinflammatory stimuli would alter the distribution of ASF in the spleen and lymph node, Lewis rats were injected in the hind foot pads and at the base of the tail with a pathogenic antigen (encephalitogenic peptide) emulsified in Freund’s complete adjuvant. On Day 7 post-injection, the animals were killed, and draining popliteal and inguinal lymph nodes and spleens were collected and analyzed via immunohistochemistry. In the spleens of naïve animals, ASF expression was primarily confined to large, stellate cells in the marginal sinus as well as scattered macrophages in the red and white pulp (Fig. 2
). In contrast, in the spleens of immunized animals, ASF expression was found on a greatly increased number of macrophages in the red pulp relative to naïve controls in addition to staining the same population of large stellate cells in the marginal sinus.
View larger version (73K):
[in this window]
[in a new window]
|
Figure 2. Redistribution and up-regulation of ASF upon immunization. Lewis rats were immunized in the hind footpads and at the base of the tail with encephalitogenic peptide emulsified in Freund’s complete adjuvant and were killed seven days later. Spleens and lymph nodes were collected from the immunized animals and from unimmunized control animals and were processed for immunohistochemistry (a) or flow cytometry (b). a) TLD-1A8A immunohistochemical staining of spleen or lymph node from immunized or naive animals, 20x objective. Sections stained with an isotype control antibody showed no detectable staining (data not shown). b) Single-cell suspensions of spleen or lymph node from immunized or naïve animals were stained with TLD-1A8A and analyzed by flow cytometry.
|
|
In the lymph nodes of naive animals, ASF was expressed by a population of large cells with dendritic morphology in the germinal centers, as well as by macrophages scattered throughout the T cell zones. However, in the draining lymph nodes from immunized animals, ASF expression was found primarily on a population of smaller cells throughout the parenchyma, whereas the larger cells in the germinal centers were no longer readily apparent.
Increased expression of ASF in the spleen and draining lymph nodes from immunized animals was confirmed with flow cytometry. Single-cell suspensions of splenocytes and lymph node cells from animals that had been immunized seven days previously were compared with naive controls. As shown in Fig. 2
, splenocytes and lymph node cells stained more intensely for ASF than did cells from naïve controls.
Based on these data, we hypothesized that ASF expression might be linked directly to the activation state of members of the monocyte/macrophage/microglial group. To test this hypothesis, the relative expression levels of both intracellular and extracellular ASF were compared by flow cytometry on activated and resting macrophages. Thioglycolate elicited peritoneal exudate cells (PECs), and resident peritoneal macrophages represented those two functional states, respectively. While the number of cells expressing intracellular ASF was comparable between the two groups, a measurable increase was noted in the number of cells expressing plasma membrane-associated ASF on PECs relative to resident peritoneal macrophages (Fig. 3a
). This experiment also indicates that the majority of ASF is localized intracellularly. Next, 24 h culture of PECs, which causes their activation, resulted in significant increases of both surface associated and intracellular ASF (Fig. 3b)
. Consistent with this, splenocyte cultures increased their ASF expression in a time-dependent manner. Furthermore, addition to the splenocyte cultures of the T cell specific mitogenic lectin concanavalin A greatly increased the speed with which ASF expression was enhanced (Fig. 3c)
.
View larger version (19K):
[in this window]
[in a new window]
|
Figure 3. Up-regulation of ASF upon cellular activation. a) Peritoneal exudate cells and resident peritoneal macrophages were isolated from rats that had either been treated intraperitoneally with thioglycolate or had previously undergone no treatment, respectively. Cells were stained with TLD-1A8A in the presence or absence of 0.2% saponin for intracellular or extracellular staining, respectively; residents compared with PECs stained for intracellular ASF, P = 0.0004; residents compared with PECs stained for cell surface ASF, P = 0.14. b) Peritoneal exudate cells were obtained as before and stained either immediately or were cultured for 24 h before staining; fresh compared with cultured stained for extracellular ASF, P < 0.0001; fresh compared with cultured stained for intracellular ASF, P = 0.0001. c) Unfractionated splenocytes were cultured for the indicated time periods in the presence or absence of 5 µg/ml concanavalin A, then harvested and stained with TLD-1A8A to measure surface expression of ASF; ASF expression on ConA stimulated compared with control splenocytes, P = 0.0001, 0.003, 0.2, 0.1 for 24, 48, 72 and 96 h, respectively. d) NR8383 macrophages were cultured for the indicated time periods in the presence of 10 ng/ml LPS, 50 U/ml IFN- , the combination of the two, or saline control, harvested, and stained with TLD-1A8A to measure surface expression of ASF.
|
|
To further dissect the specific signals resulting in ASF up-regulation, the rat derived NR8383 macrophage cell line was treated with various cytokines and stimulants, and the effect on ASF expression measured by flow cytometry. Of the stimulants tested, only IFN- and LPS increased the expression levels of ASF. As shown in Fig. 3d
, LPS stimulation did not result in significant up-regulation of ASF surface expression until 72 h post exposure, while IFN- stimulated cells did not alter surface levels of ASF until 96 h following stimulation. The combination of the two was not additive and followed kinetics nearly identical to LPS alone.
Confocal microscopy of NR8383 cells stimulated with LPS, IFN- or the combination of the two, demonstrated that not only do these stimulants result in up-regulation of ASF, but intracellular redistribution of the molecule as well. ASF in untreated cells was located intracellularly primarily in a perinuclear area (Fig. 4
). In LPS treated cells this distribution remained the same out to 48 h. By 72 h, however, LPS stimulated cells appeared to have the majority of their ASF located diffusely throughout the cytoplasm. At the 96 h time point, the molecule was mainly located at the cell surface. A similar shift in intracellular distribution is seen in the IFN- treated cells. The same also held true of LPS/IFN- treated cells, although ASF was redistributed to the cell surface by 72 h and subsequently disappeared from the cell surface by 96 h. In marked contrast to LPS and IFN-, TNF- stimulation resulted in a measurable depression of ASF surface expression as determined by flow cytometry (Fig. 5
). The effect was dose-dependent (Fig. 5a)
and reached a maximum by 24 h (Fig. 5b)
.
View larger version (15K):
[in this window]
[in a new window]
|
Figure 5. Regulation of ASF expression following cytokine stimulation. a) NR8383 macrophages were treated for 24 h with the indicated concentrations of TNF-, harvested, and stained with TLD-1A8A to measure surface expression of ASF. Only living cells, based upon forward vs. side scatter, were included in the analysis, P = 0.0003. b) NR8383 macrophages were treated with 10 ng/ml TNF- for the indicated time periods, harvested, and stained with TLD-1A8A to measure surface expression of ASF. Only living cells, based upon forward vs. side scatter, were included in the analysis, P = 0.01 Chi-squared. c) NR8383 macrophages were treated with the indicated concentrations of TNF- or saline control, mRNA was harvested at 24 h, and relative ASF levels were measured by quantitative real-time PCR. ASF levels were normalized against actin levels, and the ASF levels in the saline control samples were defined as 1, P < 0.0001. d) NR8383 macrophages were treated with 10 ng/ml LPS, 50 U/ml IFN-, the combination of the two, or saline control. mRNA samples were obtained at the indicated time points, and relative ASF levels were measured by QPCR. ASF levels were normalized to actin levels. Units are expressed as fold change relative to saline control for the indicate time point. ANOVA analysis for LPS, IFN-, and LPS plus IFN-, P < 0.0001. No statistically significant difference was apparent between the groups: LPS compared with IFN-, P = 0.05, LPS compared with LPS plus IFN-, P = 0.08, IFN- compared with LPS plus IFN-, P = 0.06.
|
|
QPCR analysis for ASF message levels of TNF- stimulated cells mirrored the results seen at the protein levels; decreasing levels in a dose-dependant manner with a nearly twofold drop at the maximum TNF- dosage (Fig. 5c)
. Conversely, while the ASF protein levels were maximum between 72 and 96 h when stimulated with LPS and IFN-, QPCR analysis showed that message levels of ASF were maximum at 48 h and then dropped back down to baseline levels by 72 h (Fig. 5d)
. While there was a trend suggesting that the most potent stimulus for up-regulating ASF mRNA levels was IFN-, followed by LPS, and that the combination of the two was less effective than either stimulus alone, differences in the effectiveness of the various different treatments was statistically insignificant.
The anti-ASF antibody TLD-1A8A increases the severity of adoptively transferred EAE
Female Lewis rats were given adoptively transferred EAE in conjunction with either TLD-1A8A or an isotype matched control antibody and monitored for clinical symptoms of disease. Animals that received TLD-1A8A consistently experienced a more severe course of disease than did animals that received the control antibody. As shown in Fig. 6
, the two groups developed clinical signs of EAE at the same time. However, TLD-1A8A treated animals consistently developed greater maximum clinical scores and had a slightly delayed recovery compared with control animals, P < 0.0001. In addition, TLD-1A8A treated animals lost more weight than control animals, P < 0.0001.
View larger version (13K):
[in this window]
[in a new window]
|
Figure 6. TLD-1A8A exacerbates the clinical signs of EAE. Groups of three female Lewis rats were given either TLD-1A8A or an isotype-matched control antibody intravenously, allowed to rest for 1 h, then given 25 x 106 encephalitogenic T cells intravenously. Animals were weighed and scored daily for clinical signs of EAE. Clinical scores are described in Materials and Methods. Clinical scores (a) TLD-1A8A compared with control P < 0.0001, Chi-squared. Percentage weight loss (b) TLD-1A8A compared with control P < 0.0001, Chi-squared. Weights are indicated as percentage of weight lost as compared with d0 weight. Results are representative of seven separate experiments.
|
|
TLD-1A8A administration increases proinflammatory cytokine and decreases anti-inflammatory cytokine expression in vitro
RNA isolated at 48 h from in vitro T cell proliferation assays to which either TLD-1A8A or an isotype matched control antibody had been included was analyzed for cytokine gene expression using the GEArrayTM Q Series cytokine expression array from SuperArray. Previous investigations in this laboratory have shown that TLD-1A8A increases T cell proliferation [6
]. Since the T cell proliferation assays were pulsed with 3H-thymidine at 48 h, this time point was chosen for RNA isolation to correlate with our previous findings. Table 1
shows transcripts that were found to be either up-regulated or down-regulated greater than twofold. Of note, several proinflammatory molecules, including IL-18, IL-6 and CD40L, were up-regulated by TLD-1A8A. Also upregulated were the growth factors FGF-12a, thrombopoietin, erythropoietin and HGF. Notably, IL-10 an anti-inflammatory gene known to be involved in the remission phase of EAE was down-regulated.
|
DISCUSSION
|
|---|
Two general conclusions can be drawn from these data. First, the expression levels and distribution of ASF are markedly altered during an immunological response in vivo. Second, ASF in turn appears to have an effect on the immune response. While we originally cloned ASF from a rat microglial cDNA library, the function of the molecule is clearly not CNS specific. Indeed, ASF expression is not limited to microglia, but is expressed by several members of the macrophage family in addition to a number of other cell types [6
, 11
, 12
]. While macrophages constitutively express ASF, i.e., even in the absence of an overt immunological stimulus, the level of expression and distribution of the molecule is certainly altered noticeably by a proinflammatory stimulus.
For example, in uninflammed CNS, ASF expression is limited to cells in the perivascular area. However, during EAE, ASF expression not only increases on these cells, but also appears in parenchymal microglia and infiltrating leukocytes. These infiltrating leukocytes morphologically appear to be macrophages. The increased expression of ASF as observed by immunohistochemistry generally correlated with the severity of disease; reaching its maximum at peak disease severity and subsequently tapering off. Upon resolution of disease, no expression of ASF was observed in the parenchyma, as is the case in the CNS before inflammation. However, perivascular cells clearly stained more intensely for ASF after disease resolution than before the inflammatory insult. Quantitative real time PCR analysis of ASF mRNA levels mirrored the results seen with immunohistochemistry. As the stimuli that lead to the initial increase in ASF expression levels are presumably no longer present upon resolution of disease, it is curious that ASF levels remain high on perivascular cells. Also, it is interesting to note that QPCR analysis demonstrated a twofold drop in ASF mRNA levels immediately preceding the onset of clinical symptoms.
No overt changes were apparent in either expression levels or distribution of ASF in peripheral organs, such as the spleen and lymph nodes, during the course of disease despite the accumulation of ASF positive leukocytes in the CNS. Since ASF levels are at almost undetectable levels on circulating monocytes, this suggests a mechanism whereby ASF is upregulated on these cells at the area of inflammation upon macrophage activation and/or as a result of interaction with activated T lymphocytes.
This stands in contrast to the observation that footpad immunization leads to redistribution of ASF in the draining lymph nodes as well as in the spleen. While in naïve spleen ASF expression was primarily limited to a population of large, stellate cells of the marginal sinus in the germinal center, spleens responding to an immunological stimulus had a greatly increased number of ASF positive cells in the red pulp. In naïve lymph nodes, ASF is expressed almost exclusively by a population of large, round cells in the germinal centers. However, lymph nodes responding to an immunological stimulus redistributed their ASF expression such that this population of cells was no longer readily apparent, while a much smaller population of cells that was distributed more homogenously throughout the parenchyma of the organ was expressing the molecule. While we have not exhaustively phenotyped the ASF positive cells, we can conclude from the drastic reorganization of the molecule that the cells that do express it regulate it in response to an immunological stimulus.
Given that freshly isolated macrophages become activated in culture, it is noteworthy that culturing these cells leads to enhanced ASF expression levels even without the addition of any further stimulation. Splenocyte cultures also enhanced ASF expression in a time dependent manner, while the addition of the mitogenic lectin concanavalin A to the cultures increased the rate at which this enhanced expression occurred. This observation is important in two regards. First, it was consistent with the observation that adoptive transfer of activated, encephalitogenic T cells increased expression of ASF by endogenous parenchymal cells as well as recruitment of ASF positive cells to the CNS. Since ASF is not expressed by T cells, this indicates that activation of T cells in turn leads to upregulation of ASF by macrophages.
To circumvent the effects of culture on ASF expression by primary macrophages, the rat macrophage cell line NR8383, which expresses ASF at a constant level, was used. LPS and IFN- increased ASF expression on these cells while TNF- decreased the expression of the molecule. The kinetics involved in this regulation are perhaps the most interesting aspect of this observation. It is not uncommon for stimuli such as LPS, IFN- and TNF- to alter the expression levels of genes within hours. However, the expression level of ASF was not perceptibly modified until several days following stimulation. While the significance of this relatively delayed response has yet to be borne out, it could be that ASF functions in the latter stages of an immune response, such as the resolution phase, rather than at the beginning stages of an immune response, such as at the priming stage.
Confocal microscopy demonstrated that stimulation with LPS and IFN- results in redistribution of ASF within the cell. Before stimulation, the majority of ASF is located intracellularly. Following stimulation, the molecule is gradually redistributed to the cell surface. By 72 h, the molecule appears dispersed evenly throughout the cytoplasm, and by 96 h is almost entirely localized to the plasma membrane. From the flow cytometry experiments, we know that there is, in fact, an increased level of ASF at the surface by 72 h. This result indicates that the majority of ASF is still localized inside the cell at this time point.
The ability of the anti-ASF antibody TLD-1A8A to affect the clinical course of EAE suggests an unrecognized role for this molecule in vivo. In previous work, we have shown that TLD-1A8A has the capacity to increase T cell proliferation in vitro. This observation correlates with the ability of the antibody to exacerbate EAE. EAE was induced in these experiments by adoptive transfer of primed, activated T cells suggesting that TLD-1A8A’s effects were most probably at the level of the cell’s effector function in the target organ, as opposed to the priming or initiation phases. Also, there were no changes in the immunohistochemical distribution of ASF in the spleen or lymph nodes during EAE suggesting that the increased clinical severity of disease induced by TLD-1A8A administration is due to an in situ effect, rather than a more systemic effect.
Our data demonstrates that the anti-ASF antibody accentuated the weight loss in rats with EAE as compared with controls. In addition, TLD-1A8A produced a more dramatic effect on weight loss than in augmenting the clinical signs of paralysis. While an explanation for these phenomenon cannot be stated with certainty, it is well documented that EAE in rats is comprised of distinct, independently controlled, features. In an immunogenetic study of the loci related to EAE in rats, Gasser, et al. [13
] demonstrated that the neurological signs of ascending paralysis, the weight loss and the presence of histological inflammation in the CNS segregated independently. It was concluded that these three features classically associated with EAE were controlled by distinct linkage groups. In light of this, it is possible that TLD-1A8A, and by implication ASF itself, may exert a greater effect on the pathways associated with weight loss than with those of paralysis or CNS inflammation. However, the weight loss exhibited by the rats receiving TLD-1A8A cannot be attributed to some pathophysiological imbalance produced by the antibody alone, but unrelated to its effects on EAE. Rats given the anti-ASF antibody intravenously, but in which EAE was not induced, demonstrated no abnormal change in their body weight compared with animals given a control antibody (n=3 rats each of TLD-1A8A or isotype control); they continued to exhibit the daily, incremental weight gain normally seen in Lewis rats (data not shown).
The mRNA expression profiles of T cell proliferation assays treated with either TLD-1A8A or a control antibody were analyzed to address how TLD-1A8A increases the severity of EAE and the in vitro proliferation of T cells. As expected, a number of proinflammatory molecules were up-regulated, such as IL-18, IL-6, and CD40L, while a critical anti-inflammatory cytokine, IL-10, was down-regulated. Although this analysis is certainly not exhaustive of the effects of TLD-1A8A, it provides key insight into how this molecule functions in terms of its interactions with other molecules relevant to the immune system.
The data concerning ASF are consistent with the hypothesis that it has a role in down-regulating immune responses. TLD-1A8A does not affect NO production by macrophages in response to LPS stimulation, nor does it induce calcium signaling in these cells, which suggests that TLD-1A8A is an ASF antagonist (data not shown). We have also shown previously that TLD-1A8A can block the effect of soluble recombinant ASF in the ligated loop model of intestinal fluid secretion, indicating that the antibody functions by preventing association of the molecule with its receptor. Thus, the effects of TLD-1A8A on T cell proliferation and EAE were most likely due to blocking the effects of ASF as opposed to acting as an ASF agonist. The anti-inflammatory effects of ASF help make sense of the delayed kinetics of ASF up-regulation following stimulation with LPS and IFN-, as well as the increased levels of ASF on perivascular cells. Presumably, ASF expression under these conditions serves an anti-inflammatory function. The increased expression of ASF during the course of EAE can therefore be interpreted as a means of counteracting the pro-inflammatory environment and limiting tissue damage. The ability of TLD-1A8A to increase proinflammatory cytokine messages, and decrease anti-inflammatory cytokine messages, further supports this conclusion.
Thus, multiple lines of evidence suggest that TLD-1A8A’s interaction with ASF is antagonistic. However, alternate mechanisms may account for the observations presented here. TLD-1A8A may act by directly inducing a signaling cascade in the ASF bearing cell. If this is indeed the case, then ASF may serve a role in augmenting T cell responses, presumably by augmenting the function of the APC, which in turn would lead to an exacerbated course of EAE. Yet another hypothesis is that T regulatory cells express an ASF receptor and that engagement of this receptor by ASF activates the suppressive functions of these cells. In this scenario, the augmented severity of EAE would be explained by eliminating the effect of these regulatory cells.
While the ligands with which ASF interacts to achieve these effects remain unknown, as are the signaling cascades initiated by this interaction, it is clear that investigations along these lines are warranted. The apparent participation of ASF in EAE, as well as the regulation of ASF by numerous inflammatory stimuli underscore the potential immunological importance of this molecule.
|
ACKNOWLEDGEMENTS
|
|---|
Flow cytometry and confocal microscopy was performed at Dartmouth Medical School in The Herbert C. Englert Cell Analysis Laboratory, which was established by equipment grants from the Fannie E. Rippel Foundation, the NIH Shared Instrument Program, and Dartmouth Medical School and is supported in part by the Core Grant (CA 23108) from the National Cancer Institute to the Norris Cotton Cancer Center. The authors would like to thank James D. Gorham for helpful discussions and critical reading of this manuscript. This work was supported by NIH Grant N547360-01 (WFH).
Received February 11, 2004;
revised June 2, 2004;
accepted June 25, 2004.
|
REFERENCES
|
|---|
- Hickey, W. F., Kimura, H. (1988) Perivascular microglial cells of the CNS are bone marrow-derived and present antigen in vivo Science 239,290-292[Abstract/Free Full Text]
- Aloisi, F. (2001) Immune function of microglia Glia 36,165-179[CrossRef][Medline]
- Williams, K. C., Zhao, W., Politopoulou, G., Male, D., Hickey, W. F. (2000) Inhibition of experimental allergic encephalomyelitis with an antibody that recognizes a novel antigen expressed on lymphocytes, endothelial cells, and microglia Lab. Invest. 80,313-326[Medline]
- Male, D., Rahman, J., Linke, A., Zhao, W., Hickey, W. (1995) An interferon-inducible molecule on brain endothelium which controls lymphocyte adhesion mediated by integrins Immunology 84,453-460[Medline]
- Flaris, N. A., Densmore, T. L., Molleston, M. C., Hickey, W. F. (1993) Characterization of microglia and macrophages in the central nervous system of rats: definition of the differential expression of molecules using standard and novel monoclonal antibodies in normal CNS and in four models of parenchymal reaction Glia 7,34-40[CrossRef][Medline]
- Davidson, T. S., Hickey, W. F. (2004) Distribution and immunoregulatory properties of antisecretory factor Lab. Invest. 84,307-319[CrossRef][Medline]
- Lonnroth, I., Lange, S. (1986) Purification and characterization of the antisecretory factor: a protein in the central nervous system and in the gut which inhibits intestinal hypersecretion induced by cholera toxin Biochim. Biophys. Acta 883,138-144[Medline]
- Hickey, W. F. (1999) Leukocyte traffic in the central nervous system: the participants and their roles Semin. Immunol. 11,125-137[CrossRef][Medline]
- Mannie, M. D., Paterson, P. Y., U?Prichard, D.C., Flouret, G. (1985) Induction of experimental allergic encephalomyelitis in Lewis rats with purified synthetic peptides: delineation of antigenic determinants for encephalitogenicity, in vitro activation of cellular transfer, and proliferation of lymphocytes Proc. Natl. Acad. Sci. USA 82,5515-5519[Abstract/Free Full Text]
- Hickey, W. F., Cohen, J. A., Burns, J. B. (1987) A quantitative immunohistochemical comparison of actively versus adoptively induced experimental allergic encephalomyelitis in the Lewis rat Cell. Immunol. 109,272-281[CrossRef][Medline]
- Lange, S., Jennische, E., Johansson, E., Lonnroth, I. (1999) The antisecretory factor: synthesis and intracellular localisation in porcine tissues Cell Tissue Res. 296,607-617[CrossRef][Medline]
- Lange, S., Lonnroth, I. (2001) The antisecretory factor: synthesis, anatomical and cellular distribution, and biological action in experimental and clinical studies Int. Rev. Cytol. 210,39-75[Medline]
- Gasser, D. L., Goldner-Sauve, A., Hickey, W. F. (1990) Genetic control of resistance to clinical EAE accompanied by histological symptoms Immunogenetics 31,377-382[CrossRef][Medline]
TOP
ABSTRACT
INTRODUCTION
