| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Published online before print June 22, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Vaccine and Infectious Disease Organization (VIDO), University of Saskatchewan, Saskatoon, Canada
1 Correspondence: Vaccine and Infectious Disease Organization (VIDO), University of Saskatchewan, 120 Veterinary Rd., Saskatoon, SK, S7N 5E3. E-mail: philip.griebel{at}usask.ca
ABSTRACT
We have previously shown that bovine polymorphonuclear cells (PMNs) have an impressive capacity to passively acquire membrane lipids and proteins from apoptotic cells. The present study used confocal microscopy to analyze the interaction between PMNs and a variety of donor cells, and assays were used to determine if passively acquired membrane proteins altered PMN biology. Confocal microscopy revealed that direct cell–cell contact and microparticles shed by donor cells may be a source of passively acquired membranes and integral membrane proteins, which then integrate into the PMN plasma membrane. Donor cells expressing green fluorescent protein in their cytoplasm were also used to demonstrate the transfer of cytoplasmic proteins from donor cells to PMNs. The functional consequences of passive membrane protein acquisition by PMNs were then investigated using two distinct systems. First, PMNs were incubated with membranes isolated from an adenovirus-permissive cell line, and this passive transfer of cell membranes significantly increased adenovirus infection of PMNs. Second, major histocompatibility complex (MHC) class II molecules were passively transferred from ovine B cells to bovine PMNs, and PMNs with ovine MHC class II on their surface were able to induce a proliferative response and increased cytokine gene expression in alloreactive bovine T cell lines. In conclusion, passively acquired membrane proteins integrated into the plasma membrane of bovine PMNs and altered the functional capacity of these cells.
Key Words: antigen presentation • MHC class II • neutrophils
INTRODUCTION
Research continues to reveal that each type of leukocyte may perform diverse, immununological functions, and this functional diversity at the cellular level creates extensive redundancy and a complex communication network within the immune system. The most important role of polymorphonuclear neutrophils (PMNs) in innate immunity may be the clearance of immune complexes, phagocytosis of opsonized particles, and the release of inflammatory mediators. Recent research has revealed, however, the possibility that PMNs are functionally much more diverse than previously imagined. For example, when stimulated in an appropriate manner, PMNs may be able to acquire some of the functional characteristics of an antigen-presenting cell (APC) [1 ]. The present investigation examined the functional potential of bovine PMNs following the passive acquisition of integral membrane proteins from necrotic or apoptotic cells.
Although investigating the possibility that bovine PMNs might present antigen, our previous research confirmed that integral membrane proteins were transferred rapidly and efficiently to bovine PMNs from a variety of apoptotic and necrotic cells [2 ]. Specifically, we observed that PMNs rapidly acquired proteins such as major histocompatibility complex (MHC) class II and CD3 from a variety of syngeneic, allogeneic, and xenogeneic cell types. Such acquisition occurred within 40 min of PMN coculture with isolated peripheral blood mononuclear cells (PBMCs) and was not inhibited at 4°C. The transfer of MHC class II molecules from murine splenocytes to bovine PMNs precluded the possibility of endogenous protein expression. We also demonstrated the transfer of fluorescently labeled membrane lipids along with integral membrane proteins. Collectively, these observations supported the hypothesis that the transfer of membrane proteins was mediated by direct cell–cell contact or the release of microparticles (MPs) from dead or dying cells.
The phenomenon of membrane/protein transfer between cells is not a novel observation, and several examples of such transfer occur between leukocytes [3 4 5 6 7 8 9 ]. T cells, for example, can acquire immune complexes and a variety of membrane proteins from APCs in an activation-dependent process [7 ]. An exchange of membrane proteins has also been observed with B cells [10 , 11 ], dendritic cells (DC) [4 ], natural killer (NK) cells [8 , 9 ], macrophages [12 ], and basophils [4 ]. Furthermore, passively acquired membrane proteins may alter the function of the recipient cells. For example, T cells that acquire MHC class II/peptide complexes can inhibit [3 , 5 6 7 ] or activate antigen-specific T cell responses [13 ]. Thus, the functional consequences of passive membrane protein acquisition requires further investigation, especially as this phenomenon may have implications beyond the regulation of immune responses. For example, the passive acquisition of the CD21 protein by NK cells can confer susceptibility to Epstein-Barr virus infection [14 ]. Thus, integral membrane proteins can retain their function following transfer from one cell to another.
Several mechanisms have been proposed by which proteins may be transferred between leukocytes. Transfer can involve an intimate interaction between cells at an "immunological synapse", where tight junctions are formed between adhesion molecules. The formation of a CD8 T cell/target cell immunological synapse resulted in the formation of a physical bridge between cells that facilitated the lateral diffusion of proteins and lipids between fused membranes [15 ]. DC use a different mechanism to physically sample extracellular or intracellular membrane/protein fragments from live cells [4 ], but other cells have also been shown to acquire membrane proteins from dead or dying cells [4 ]. Our investigations revealed that bovine PMNs acquired membrane proteins from apoptotic or necrotic cells, and this transfer appeared to require close cell contact. We hypothesized that direct cell–cell contact or the release of MPs from dying cells might mediate the transfer of membrane proteins to PMNs. MPs are shed from the plasma membranes of virtually all cell types in response to a variety of stress conditions and apoptosis or following cellular activation [16 ]. Furthermore, MPs exhibit increased adherence to cells [17 ,18] and can transfer integral membrane proteins, which alter the function of recipient cells [19 , 20 ]. For example, MP-mediated transfer of CC chemokine receptor 5, a coreceptor for human immunodeficiency virus (HIV), conferred susceptibility to HIV infection [12 ], and MP-mediated transfer of oncogenes induced tumor formation [21 ]. Thus, MPs can interact with adjacent and distant cells and may function as vectors for the intercellular exchange of biologic information.
The aim of the present investigation was to characterize the cellular interactions mediating the passive transfer of membrane lipids and protein to bovine PMNs and to determine the functional consequences of such passive acquisition. Confocal microscopy was used to examine the interaction between PMNs and a variety of donor cell populations, and a variety of assays were then used to determine if proteins passively acquired were functional and conferred new biological properties to recipient PMNs.
MATERIALS AND METHODS
Culture media and reagents
We previously identified culture conditions, which enhanced bovine PMN survival in vitro. Costimulation of PMNs with lipopolysaccharide (LPS) and 10 ng/ml recombinant bovine granulocyte macrophage-colony stimulating factor or interferon-
(IFN-) maintained over 50% cell viability during a 72-h culture period. Supplementation of fetal bovine serum (FBS; Gibco-BRL, Burlington, Ontario, Canada) from 5% to 20% also maintained PMN viability at 80 ± 8.6% (mean±1 SD; n=5) during a 24-h culture and 46 ± 6.8% (mean±1 SD; n=5) during a 48-h culture. Thus, AIM-V serum-free lymphocyte medium (Gibco-BRL), supplemented with 20% (v/v) heat-inactivated FBS and 50 µM 2-mercaptoethanol (Bio-Rad Laboratories, Mississauga, Ontario, Canada), was used to maintain PMN viability during membrane transfer, adenovirus infection, and alloreactive T cell stimulation studies. This medium is referred to as "complete AIM-V medium" unless otherwise stated, and LPS and cytokines were not added to these cocultures.
Purification of blood leukocytes
Blood was collected from the jugular vein of male or female Holstein cattle (age >12 months) using 0.3% EDTA (Sigma-Aldrich, Oakville, Ontario, Canada) as an anticoagulant, and PBMCs and PMNs were purified as described previously [2
]. Cells isolated by these methods were consistently >98% viable as determined by trypan blue dye exclusion and contained less than 4% mononuclear cells. No attempt was made to differentiate between PMN and eosinophilic granulocytes, but the latter were present in low numbers (range=2–10%).
Cell lines
An ovine B cell line (Clone 2; Cl2) was cultured as described previously [22
], and green fluorescent protein (GFP)-expressing B cells (Cl2-GFP) were generated by using a retroviral vector to transduce the GFP gene [23
]. The 293 cell line (ATCC #CRL-10852) was cultured in Dulbecco’s minimal essential medium supplemented with 10% FBS (Gibco-BRL), and 293 cells were split 1:4 and passaged every 3–4 days.
Xenoreactive T cell populations
Bovine PBMCs (10x106) were cocultured 10:1 with -irradiated (11000 rads) ovine Cl2 cells in each well of a six-well plate (Corning Inc., NY). After 5 days, PBMC cultures were split 1:3 into fresh media supplemented with 2 ng/ml recombinant human interleukin-2 (IL-2; R&D Systems, Minneapolis, MN). Cultures were split again 1:3 after 3 days and transferred to fresh media supplemented with 2 ng IL-2/ml. Following this second, 3-day culture period, the xenostimulation procedure, using -irradiated Cl2 cell stimulation and IL-2-dependent T cell expansion, was repeated once more before performing a mixed lymphocyte-proliferative response (MLR) assay. Following the second xenostimulation, cells in the culture were >99% bovine CD3+ T cells, and these cells displayed a strong lymphoproliferative reaction [stimulation index (SI)>100] following the addition of -irradiated Cl2 cells.
Biotin, PKH, and oxidative burst fluorescent labeling
To monitor the transfer of cellular membranes, donor cells were labeled with the lipophilic PKH-26 Red fluorescent cell linker kit (Sigma-Aldrich) according to the manufacturer’s instructions. PKH is a lipophilic fluorescent dye, which stably inserts into the plasma membrane of viable cells and can be used to visualize the formation and transfer of MPs. Plasma membrane proteins on viable cells were labeled using EZ-Link sulfo-normal human serum-biotin (Pierce, Rockford, IL) in accordance with the manufacturer’s instructions. The transfer of biotinylated proteins from donor cells to PMNs was monitored through the surface labeling of PMNs with streptavidin-fluorescein isothiocyanate (SA-FITC). The biotinylation of surface proteins on donor cells was confirmed prior to performing coculture by analyzing SA-FITC binding with flow cytometry [2
] and examining plasma membrane fluorescence with confocal laser-scanning microscopy (CLSM). The compound 2',7'-dichlorofluorescein-diacetate (DCFH-DA; Sigma-Aldrich) was used to fluorescently label PMNs. The nonfluorescent, nonpolar dye DCFH-DA is able to cross PMN membranes, and within the cell, it is transformed by esterases into a polar molecule, which cannot diffuse out of the cell. In the presence of H2O2, this polar DCFH is then converted to DCF, which fluoresces green and provides a PMN-specific marker. PMNs (1x107 cells/ml) were incubated with 0.1 mM DCFH-DA for 20 min and then washed twice with Ca++Mg+-free phosphate-buffered saline (PBS). This procedure labeled over 99% of PMNs with an intense green fluorescence. Bovine PBMCs were labeled with PKH-26 or biotin-SA-FITC prior to coculture with PMNs, which were unlabeled or stained with PKH or DCFH-DA.
Transfer of cell membranes and proteins to bovine PMNs
To achieve a rapid and consistent transfer of membrane protein from 293 cells and Cl2 B cells to bovine PMNs, the donor cells were lysed prior to incubation with PMNs. This method of membrane-protein transfer also eliminated the potential contamination of PMNs with the donor cell population. Donor cells (1x107 cells/ml) were sonicated for 2 x 7 s using a microtip vibrating at 20 kHz and amplitude of 40% (VibraCellTM sonicator, Betatek Inc., North York, Ontario, Canada). Lysed cells were pipetted vigorously to disperse any clumps and then centrifuged for 3 min at 216 g to remove intact cells and cell debris. To achieve a more consistent transfer of membrane proteins, the bovine PMNs were incubated for 24 h with lysed cell membranes using an approximate ratio of 1 PMN:3 "lysed cell equivalents".
Construction of GFP-expressing bovine adenovirus (BAdV) and infection of cells
A recombinant BAdV Type 3 (BAdV-3) vector was constructed and contained the GFP gene in the E3 region (BAV304) [24
]. Expression of GFP in this vector did not require viral replication. BAV304 was added to 293 cell cultures, PMNs alone, PMNs cocultured with lysed 293 cell membranes, or PMNs cocultured with bovine PBMCs. Cocultures were incubated for 48 h before GFP expression was quantified with flow cytometry. The detection of GFP expression in PMNs following BAV304 infection was restricted to viable PMNs by using propidium iodide (2.5 µg/ml) staining to exclude dead cells.
Flow cytometry
The monoclonal antibody (mAb) specific for a conserved epitope on the 
chain of the bovine ortholog of human leukocyte antigen-DR (Clone TH14B) was from VMRD (Pullman, WA). FITC-conjugated, isotype-specific goat antimouse immunoglobulin antibodies were purchased from Southern Biotechnology (Birmingham, AL). The level of specific mAb binding was quantified by subtracting the percentage of cells that reacted with an isotype-matched, irrelevant mAb (Cedar Lane Laboratories Ltd., Hornby, Ontario, Canada). The irrelevant mAb was used at the same concentration (5 µg/ml) as the relevant mAb. Cells were fixed in 2% paraformaldehyde and stored in the dark at 4°C until data acquisition was performed with a FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA), and data analysis was performed using the CellQuest program. This cell-labeling and analysis protocol was used to confirm that detectable levels of ovine MHC class II had been acquired by bovine PMNs following coculture with lysed Cl2 cell membranes.
MLR assays
MLR assays were used to assess responses of the xenoreactive T cell lines, developed with Cl2 B cells as stimulators, and to determine if PMNs could function as APCs following passive acquisition of MHC II class proteins from Cl2 membranes. Xenoreactive T cells (5x104 cells/well) were cocultured with -irradiated Cl2 B cells [1x103 cells/well; responder/stimulator (R/S)=50:1], -irradiated heterologous bovine PBMCs (11,000 rads; 5x104 cells/well; R/S=1:1), or autologous or heterologous bovine PMNs incubated with or without Cl2 membrane fragments (5x104 cells/well; R/S=1:1). Stimulator cells were added to triplicate cultures of each T cell line in 96-well plates (Corning Inc.), and cultures were incubated for 90 h at 37°C in a humidified atmosphere with 5% CO2 before the addition of 0.4 µCi/well [methyl-3H] thymidine (Amersham Pharmacia Biotech, Piscataway, NJ) during the final 6 h of culture. Cultures were harvested onto unifilter glassfiber microplates (Perkin Elmer, Woodbridge Ontario, Canada), and incorporation of 3H-thymidine was measured using a microplate scintillation and luminescence counter (Top Count, Canberra Packard, Montreal, Canada). Data are expressed as a SI = counts per minute (cpm), and Cl2 B cells or other stimulator cells/cpm for T cells were cultured in medium alone, and all assays were performed with xenoreactive T cell lines generated from six animals.
Analysis of cytokine gene expression in T cell lines
Xenoreactive T cells lines (5x106 cells/well), specific for Cl2 B cells, were plated in a 12-well plate (Corning Inc.) and incubated for 2 h, 4 h, 8 h, or 12 h with 5 x 105 -irradiated Cl2 cells (R/S=10:1), 1 x 106 syngeneic PMNs (R/S=5:1), or 1 x 106 syngeneic PMNs (R/S=5:1), which had been incubated for 24 h with membranes from 3 x 106 lysed Cl2 cells. T cell cultures were collected at each time-point, washed twice with PBS, and then lysed with 1 ml Trizol (Gibco-BRL). Total RNA was extracted as described previously [25
]. RNA concentration was determined spectrophotometrically by measuring absorbance at 260 nm, and purity was assessed using an Agilent 2100 bioanalyzer with RNA 6000 Nano kits (Agilent Technologies Canada Inc., Mississauga, Ontario). The quality of total RNA was assessed by comparing the ratio of the area under the ribosomal peaks for 28S and 18S. Quantitative real-time polymerase chain reaction (qRT-PCR) was conducted using SuperScriptTM III Platinum® two-step qRT-PCR kit with SYBR® Green (Invitrogen Canada, Burlington, Ontario) on the Bio-Rad iCycler (Bio-Rad Laboratories), as indicated in the manufacturer’s protocol. Primers for sense and antisense strands of selected genes (Table 1
) were designed using Clone Manager 7 (Scientific and Educational Software, Cary, NC), and PCR products were sequenced to confirm the specificity of the primers (data not shown). The qRT-PCR analysis for each cytokine was performed in triplicate using RNA isolated from three T cell lines and a melt curve was performed to ensure that product detected by the iCycler was specific to the desired amplicon. The comparative threshold cycle (Ct) values for actin varied less than one Ct (18.1±0.9; mean±1 SD) among all T cell cultures. Cytokine gene expression data were, therefore, normalized relative to actin and expressed as relative fold change using the comparative Ct method [26
]. The 
Ct value for each cytokine gene was calculated at each time-point relative to T cells cocultured with autologous PMNs.
|
RESULTS
PBMCs and PMNs interact through cell–cell contact and MP transfer
Confocal microscopy was used to visualize interactions, which might mediate the transfer of membrane lipids and proteins between PBMCs and PMNs. Our previous observations had suggested that direct cell–cell contact was required for this transfer to occur and that the transfer of membrane proteins could occur rapidly after coculture [2
]. A combination of cellular staining techniques was used to identify donor and recipient cells and allow us to observe interactions between bovine PBMCs and PMNs immediately after mixing the two cell populations. It was apparent in all views (10 fields/sample with 25–30 cells/field) that during coculture, the PBMCs associated closely with PMNs (Fig. 1A
and 1C
). It was also evident, however, in many of these fields that PBMCs could form MPs (Fig. 1A)
and release the MPs (Fig. 1B)
, and MPs could then attached to the surface of PMNs (Fig. 1C)
. It was possible to watch the formation and release of MPs from individual cells over a 1-min interval. Thus, confocal microscopy revealed that direct cell–cell contact and the transfer of MPs might play a role in the transfer of membrane lipids and proteins from PBMCs to PMNs.
|
35% of PMNs (green) had red fluorescence within their plasma membranes. These observations support the conclusion that there was a time-dependent transfer of integral membrane proteins and membrane lipids from the surface of PBMCs to the plasma membrane of PMNs. These observations also suggested that this transfer may occur by a direct fusion of cell membranes or MPs with the PMN plasma membrane and possibly through phagocytosis of MPs.
|
-irradiation to induce apoptosis, numerous GFP-containing MPs (green) were observed (five fields for one sample) to be released from Cl2-GFP cells (Fig. 3A
). Following coculture of PMNs with Cl2-GFP cells for 24 h, it was possible to observe individual PMNs within each field (five fields for one sample), which contained discreet foci of GFP and diffuse green fluorescence within their cytoplasm (Fig. 3B)
. The discreet foci of green fluorescence may indicate phagocytosis of GFP containing MPs, but diffuse green fluorescence within the cytoplasm suggests a release of GFP from within a phagosome or GFP release into the PMN cytoplasm following direct fusion of MPs with the PMN plasma membrane.
|
|
; TNF) and TH2 (IL-4; IL-13) cytokines was analyzed. IL-2 expression was also analyzed, as this cytokine contributes to T cell-proliferative responses, and IL-10 expression was analyzed, as this cytokine regulates bovine T cell cytokine expression [30
]. This analysis revealed that PMNs, following passive acquisition of ovine MHC class II, specifically induced alloreactive T cell expression of TH1 and TH2 cytokines (Fig. 6
). The onset, amplitude, and duration of gene expression for each cytokine, with the exception of TNF, varied significantly when T cell stimulation by PMNs was compared with -irradiated Cl2 B cells. RT-PCR did not amplify detectable levels of bovine cytokine genes when RNA was isolated from Cl2 B cells (data not shown), but increased expression levels were observed for all genes following stimulation of alloreactive T cells by PMNs, which had passively acquired ovine MHC class II. T cells, stimulated by PMNs, which passively acquired ovine MHC class II, expressed significantly less IL-2 than T cells stimulated with Cl2 B cells, which was consistent with the previously observed differences in the T cell-proliferative response assay (Fig. 5)
. Thus, passive acquisition of Cl2 B cell membranes by bovine PMN induced increased expression of a broad range of cytokine genes by xenoreactive bovine T cells.
|
|
There are increasing data to support the conclusion that the passive transfer of membranes and integral membrane proteins among leukocytes is a general phenomenon, and this exchange of biological material can have important functional consequences [3 4 5 6 7 8 9 ]. Few studies have explored the mechanisms of membrane/protein transfer in depth, although passive acquisition of protein from dying cells has been reported previously [4 ]. The present study shows that fluorescently labeled integral membrane proteins and lipids can integrate into the PMN plasma membrane (Fig. 2) . Confocal microscopy reveals that this process can occur by direct cell–cell contact or MP attachment to the cell surface. Although our previous observations were most consistent with MP attachment and fusion with the PMN plasma membrane [2 ], the present observations suggest that multiple mechanisms may mediate such transfer. It is important to note, however, that all short-term PMN and PBMC cocultures were incubated at 20–22°C prior to confocal microscopy. This may be significant, as membrane fluidity and fusion may be reduced at this temperature when compared with body temperature [31 ]. Thus, passive acquisition and fusion of membranes and integral membrane proteins with the PMN plasma membrane may be even more efficient at body temperature. It is also interesting to note that during confocal studies, there was no evidence of reciprocal membrane/protein transfer from PMNs to PBMCs, even following a 24-h coculture at 37°C (Fig. 2D) . Finally, the present investigation also provided evidence that cytoplasmic proteins may be transferred between cells (Fig. 3) . In conjunction with previous data, which confirmed the transfer of oncogenes between cells [21 ], we must now consider the possibility of RNA and cytoplasmic protein transfer to PMNs. Further studies will be required to determine if the magnitude of this cytoplasmic transfer is sufficient to alter the biology of the recipient cell.
Confocal microscopy provided substantial evidence that MPs were shed by donor cells and could attach to the surface of PMNs with subsequent integration of passively acquired membranes and membrane proteins (Figs. 1 and 2) . A previous study demonstrated that MPs shed from leukocytes attached to the surface of epithelial tumor cells, but the acquired proteins remained in segregated "rafts" rather than diffusing throughout the recipient plasma membrane [20 ]. In contrast, PKH-labeled lipids and biotin-labeled proteins from donor leukocytes appeared to diffuse throughout the plasma membrane of bovine PMNs (Fig. 2D) . These observations were consistent with a full integration of membranes/proteins within the plasma membrane of the bovine PMN, with no apparent restriction in their subsequent distribution. Furthermore, time-course studies indicated that this integration can occur relatively rapidly. The attachment of MPs to the surface of bovine PMNs may explain a previous report that goat PMNs can passively acquire MHC class II, but detection of MHC class II was dependent on the cell isolation method [32 ]. Centrifugation of goat PMNs through a density gradient removed detectable MHC class II, and we observed a similar phenomenon with bovine PMNs [2 ]. Based on present observations, we propose that centrifugation may remove leukocyte-derived MPs, which are loosely attached to the PMN surface. Thus, reports that surface molecules can be removed during a cell isolation procedure may need to be re-examined for the potential passive acquisition of surface proteins from cells shedding MPs.
The passive acquisition of integral membrane proteins by bovine PMNs was of sufficient amplitude to significantly alter recipient cell biology (Figs. 4 5 6) . Our research has shown that the passive acquisition and integration of functional membrane proteins altered viral infection of PMNs and enabled PMNs to activate antigen-specific T cells. Increased viral infection of bovine PMNs was observed following their exposure to lysed cell membranes from 293 cells, which are highly permissive to BAV304 infection (Fig. 4) . PMNs are not permissive to BAdV replication (data not shown), and thus, increased GFP-transgene expression cannot be interpreted as increased viral replication. This assay primarily provided a method to monitor viral uptake by PMNs and therefore, provides indirect evidence for the passive transfer of a viral receptor protein from 293 cells to PMNs. These observations were, however, consistent with previous reports that viral receptor proteins can be passively acquired by leukocytes and subsequently alter viral tropisms [12 , 14 ].
We provided direct evidence that PMNs acquired MHC class II passively, and this conferred antigen-presentation capabilities to the PMNs. In our experiments, the passive acquisition of ovine MHC class II (Fig. 5B)
resulted in bovine PMNs inducing a significant MLR with antigen-specific T cell lines (Fig. 5A)
. This response was by primed T cells, but specific T cell activation should still require functional MHC class II with antigen bound in its groove [28
, 29
]. The requirement for costimulation signals may, however, be reduced for activated T cells, and in this assay, a functional cross-reactivity between ovine and bovine costimulatory molecules may not have been necessary. Passive acquisition of ovine MHC class II by bovine PMNs significantly induced lower levels of T cell proliferation than Cl2 B cells, and this difference in T cell responses was observed despite using 50-fold more stimulator PMNs than -irradiated Cl2 B cells. Decreased T cell activation may reflect the 44-fold lower density of passively acquired MHC II molecules on bovine PMNs [mean fluorescence intensity (MFI)=13.8] than on donor Cl2 B cells (MFI=617; Fig. 5B
). Furthermore, the level of passively acquired MHC class II decreases with time (data not shown), and this decline may also explain the diminished, proliferative response and shorter duration of cytokine gene expression (Fig. 6)
. Alternatively, decreased PMN-induced T cell responses may reflect inadequate costimulatory molecule transfer or limited cross-species activity of passively transferred ovine costimulatory molecules. The present assay was designed to discriminate clearly between potential endogenous MHC class II expression and passively acquired MHC II molecules, but this assay may not be optimal for exploring the full potential of bovine PMNs to function as APCs following the passive acquisition of membrane proteins.
It has been postulated that the transfer of membrane fragments to leukocytes may serve to expand and/or regulate an immune response [33 ]. For example, macrophage antigen presentation is reduced following phagocytosis of apoptotic cell bodies, but antigen presentation is enhanced following phagocytosis of necrotic cell bodies [34 ]. Thus, we examined in more detail the antigen-presentation capacity of PMNs following passive acquisition of ovine MHC class II. The relative fold change in gene expression at each time-point revealed that for most genes, there was a ten- to 100-fold difference in expression level, and the onset of increased cytokine gene expression was also delayed when T cells were stimulated with PMNs, which had passively acquired ovine MHC class II. Thus, although gene expression analysis did not reveal a qualitative deficiency in cytokine gene expression when PMNs functioned as APCs, the magnitude and kinetics of gene expression were reduced significantly when T cells were stimulated with PMNs + Cl2 B cell membranes versus intact Cl2 B cells (Fig. 6) . Clearly, PMNs did not function as potent APCs, but one implication of the current in vitro studies may be that PMNs, which passively acquire membrane proteins at a site of inflammation, could function to maintain the stimulation of activated T cells. The relatively large number of PMNs, which can infiltrate sites of inflammation, may compensate for their relatively poor antigen presentation by providing T cells with an increased opportunity to interact with antigen. The efficiency and consequences of T cell activation by PMNs may also depend on which costimulatory molecules are acquired passively through the uptake of MPs. Further studies will be required to determine if the cotransfer of MHC class II and costimulatory molecules has a significant impact on the efficiency of PMN antigen presentation.
In summary, our studies confirmed that bovine PMNs rapidly acquired membrane fragments and their associated proteins through a variety of interactions with dying cells. In vitro studies also confirmed that these passively acquired membrane proteins retained their function and conferred novel biological properties to recipient PMNs. The observation that passively acquired MHC class II can be presented to activated T cells provides a novel mechanism by which PMNs may play a role in innate and adaptive immune responses.
ACKNOWLEDGEMENTS
L. A. B. holds a Canada Research Chair in Vaccinology, and this research was funded by grants from the Canadian Institutes of Health Research. This manuscript is published with permission of the Director of VIDO as Journal Series #417. We thank the Animal Care staff at VIDO for care and handling of animals and Terry Beskorwayne for excellent assistance with cell culture. Confocal laser microscopy was carried out at the National Water Research Institute, Environment Canada (Saskatoon, Saskatchewan), by Dr. George Swerhone.
Received February 2, 2006; revised April 11, 2006; accepted April 12, 2006.
REFERENCES
B Oncogene 22,4531-4542[CrossRef][Medline] treatment on the constitutive expression of MHC class I and class II molecules on goat neutrophils Vet. Immunol. Immunopathol. 70,19-32[CrossRef][Medline]Related Article
This article has been cited by other articles:
|
|
 |
|
  B. E. Potts and S. K. Chapes Functions of C2D macrophage cells after adoptive transfer J. Leukoc. Biol., March 1, 2008; 83(3): 602 - 609. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hao, J. Yuan, and J. Xiang Nonspecific CD4+ T cells with uptake of antigen-specific dendritic cell-released exosomes stimulate antigen-specific CD8+ CTL responses and long-term T cell memory J. Leukoc. Biol., October 1, 2007; 82(4): 829 - 838. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
). RNA was isolated from T cell cultures following coculture for the indicated time intervals, and data presented are the mean and 1 SD of values from xenoreactive T cell lines derived from three animals. **, P < 0.01; *, P < 0.05.