| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
MRC Center for Transplantation, King’s College London, School of Medicine at Guy’s, King’s and St. Thomas’ Hospitals, London, UK
1Correspondence: MRC Center for Transplantation, Fifth Floor, Tower Wing, Guy’s Hospital, Great Maze Pond, London SE1 9RT, UK. E-mail: wilson.wong{at}kcl.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: antigen presentation • molecular exchange • surface molecules
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. As a consequence of these close interactions, surface molecules, including major histocompatibility complex (MHC) class I and II molecules, can be exchanged between cells (2
3
4
5
6
7)
. This transfer of MHC molecules between cells is a temperature- and energy-dependent process and is thought to occur either by an exchange of small sections of plasma membrane or by the release and capture of MHC-bearing exosomes or microvesicles (8)
. This has been demonstrated for T, B, natural killer (NK), and dendritic cells in vitro (2
3
4
5
6
7)
or, under contrived conditions, in vivo when dendritic cells (DCs) were injected into MHC-disparate hosts (7)
. Whether MHC transfer happens in vivo under normal physiological circumstances or whether it is purely an in vitro phenomenon, however, is still under intense debate. Despite its potential relevance to many fields that involve antigen presentation, such as infection or autoimmune disease, the study of MHC transfer in vivo under physiological conditions has not made much progress because all cells within the body express the same MHC haplotype, hampering the detection of MHC transfer. Transplantation and, to a lesser extent, blood transfusion are the only clinical situations in which foreign MHC molecules are deliberately introduced into patients. Donor MHC molecules can easily be tracked using donor-specific monoclonal antibodies. Therefore, transplantation lends itself to the study of MHC transfer. It has the added advantage of being a technique that is widely practiced in the clinic and therefore has potential implications for human health. During rejection of a transplanted organ, donor antigens can be recognized by recipient T cells either as intact whole MHC molecules on donor antigen-presenting cells (APCs) via the direct antigen presentation pathway or as processed peptides presented by self-MHC molecules on recipient APCs (indirect antigen presentation pathway). If donor MHC molecules, after transplantation, are transferred whole to the surface of recipient APCs, such as DCs, they will be recognized by recipient T cells via the direct antigen presentation pathway, despite being presented by recipient (as opposed to donor) APCs. These recipients will also express self MHC molecules that can present donor-derived peptides, which will be recognized by recipient T cells via the indirect pathway. Thus, this "semidirect" antigen presentation pathway can potentially link T cells of direct and indirect allospecificity to enable bystander activation or regulation (linked suppression).
Here, using mouse models of cardiac and renal allotransplantation, we investigate the phenomenon of MHC transfer between donor and recipient cells and demonstrate that there is extensive exchange of MHC molecules. Our data add a new facet to the T cell–APC interaction in a clinically relevant scenario and suggest that antigen presentation is more complex than traditionally thought.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 10)
.
Histological analysis
At various time points after transplantation, grafted organs and spleens from C57BL/6 recipients were harvested in optimal cutting temperature compound (Raymond A Lamb Ltd., Eastbourne, UK) and snap-frozen in liquid nitrogen. Then, 5-µm-thick sections were cut and stained with the appropriate antibodies: anti-I-Ab (clones AF6–120.1 and KH74), anti-I-Ad (clone AMS-32.1), anti-CD11c (clone N418), anti-CD68 (clone FA-11), anti-CD19 (clone 1D3), anti-CD40 (clone 3/23), and anti-CD86 (clone GL1). Antibodies were biotinylated (anti-I-Ab, anti-I-Ad) or biotinylated secondaries were used [polyclonal goat anti-rat (CD68, CD19, CD40, CD86), mouse anti-hamster cocktail (CD11c)], except anti-I-Ab clone KH74, which was conjugated to AlexaFluor 488. All antibodies were purchased from Pharmingen (San Diego, CA, USA) except anti-CD11c and anti-I-Ab clone KH74 (both Cambridge Bioscience, Cambridge, UK) and anti-CD68 (Serotec, Oxford, UK). Cells were labeled by addition of streptavidin conjugated to fluorescein, phycoerythrin (PE), or aminomethylcoumarin (AMCA; all Vector Laboratories, Burlingame, CA, USA), and excess biotin/streptavidin was blocked between each antibody using an avidin/biotin blocking kit (Vector Laboratories). Staining was visualized using a Leica microscope with 3 fluorescence filters (Leica Microsystems, Wetzlar, Germany) and digital overlay was carried out using ACT-1 and LUCIA G software (both Nikon Corporation, Tokyo, Japan). Positive cells were counted in 20 random high-power fields (HPFs, x400) of each sample by 2 independent observers masked to the experimental conditions. Results were expressed as means ± SE.
Confocal microscopy was performed using 15 µm thick sections. A Nikon Eclipse TE2000-U microscope (Nikon Corporation) with an argon (wavelength 488 nm) and a helium neon laser (wavelength 543 nm). Conventional images and z stacks were taken with EZ-C1 3.10 software (Nikon Corporation).
Flow cytometric analysis
Single-cell suspensions were prepared from spleens harvested from kidney allograft recipients 9 days after transplant. After lysing of red blood cells with Pharmlyse (Pharmingen), used according to manufacturer’s instructions, cells were stained with a biotinylated anti-I-Ad (clone AMS-32.1; Pharmingen) followed by streptavidin-PE (Pharmingen) and AlexaFluor 488-conjugated anti-I-Ab (clone KH74, Cambridge Bioscience). Cells were acquired using a FACScan flow cytometer (Becton Dickinson; Oxford, UK) and analyzed using Cellquest software version V3.3 (Becton Dickinson).
Preparation of lysed splenocytes
Single-cell suspensions of DBA/2 splenocytes were lysed by freeze-thawing 3x using dry ice in absolute alcohol to freeze the cells, followed by thawing at room temperature. After cell lysis, the suspension was centrifuged at 1100 rpm for 10 min at 4°C, and the supernatant equivalent to one whole spleen was then injected into each mouse intravenously 4 days before their spleens were harvested.
Fluorescent in situ hybridization (FISH)
Ten-micrometer-thick sections were cut from spleens of C57BL/6 recipient mice of BALB/c kidneys (n=2) 4 days after transplantation and hybridized with a probe directed against donor MHC class I mRNA (H-2d). The probe (digoxigenin conjugated, sequence CGCCGCCATGTCCGCCGCCGTCCACGTTTTCAGGTCTTCGTTCAGGGC; nucleotides 837–844 within the coding sequence of L29190) was designed and synthesized by GeneDetect (Auckland, New Zealand). Hybridization was carried out overnight at 37°C, and unbound probe was then removed by washes of decreasing stringency according to the manufacturer’s instructions. A fluorescein isothiocyanate-anti-digoxigenin antibody (Roche Applied Science, Burgess Hill, UK) was applied for probe detection. This was followed by staining with an antibody against donor or recipient MHC class II (I-Ad or I-Ab, respectively), carried out as above.
Statistics
Student’s t test was used to test for differences in percentages of cells staining positive for various markers.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. We therefore tracked the presence of allogeneic donor MHC class II+ cells in the spleens of MHC-mismatched mouse kidney and heart allograft recipients. After transplantation of DBA/2 kidneys or DBA/2 hearts, donor MHC class II I-Ad-positive cells could be detected within the spleens of C57BL/6 recipient mice as early as day 1 after transplantation (Fig. 1
A). Up to twice the number of cells were present in the recipient spleen after kidney compared with heart transplantation at day 8 after transplantation. DBA/2 kidney allografts were spontaneously accepted by C57BL/6 recipients (12
, 13)
, enabling donor cells to be tracked long term after transplantation. The number of donor leukocytes present in the spleen rose to a maximum of 28.4 ± 4.8 cells/HPF on day 14 and declined thereafter, although cells could still be detected up to 180 days after transplant (Fig. 2
).
|
|
Identification of cells positive for both donor and recipient MHC class II
To demonstrate the transfer of MHC molecules between neighboring cells in vivo after transplantation, flow cytometry was used initially. Spleens of recipients were harvested 9 days after transplantation for flow cytometry and stained for donor I-Ad+ and recipient I-Ab+ cells. Figure 3
shows that 6.39 ± 0.97% of cells present within spleens of recipients were positive for donor I-Ad molecules, whereas 3.7 ± 0.94% of cells within spleens of recipients stained positive for both donor and recipient MHC class II molecules.
|
To directly visualize donor and recipient MHC class II double-positive cells, we performed immunohistochemical studies to provide independent and more convincing proof of intercellular MHC transfer in vivo after organ transplantation. Spleens of DBA/2 kidney allograft recipients were stained for both donor I-Ad and recipient I-Ab MHC class II molecules using fluorescent immunohistochemistry followed by digital overlay. Cells positive for both donor and recipient class II MHC molecules could clearly be seen (Fig. 1B-D
). The percentage of donor MHC class II-positive cells that also stained positive for recipient class II MHC molecules peaked at 26 ± 1.8% (45 days after transplantation) of all donor I-Ad-positive cells (Fig. 4
).
|
To determine whether this phenomenon is restricted to the DBA/2 to C57BL/6 strain combination that demonstrates spontaneous acceptance of renal, but not cardiac, allografts, we transplanted BALB/c donor kidneys, which are normally rejected (13)
, into C57BL/6 recipients. Figures 2
and 4
show that donor MHC class II-positive passenger leukocytes could again be seen in the spleens of recipients, and double-positive cells could also be seen. This indicates that MHC transfer is not a rare phenomenon but a common occurrence and takes place in at least 2 donor/recipient strain combinations during both rejection and spontaneous acceptance. Double-positive cells could also be seen in the spleens of DBA/2 kidney recipients that had survived long term, although the percentage of these cells declined with time, suggesting that MHC transfer is more frequent during the early immune response against the graft.
Location of the MHC class II molecules
The transfer of MHC class II molecules signifies an alternative pathway of antigen presentation. When a recipient APC acquires whole donor MHC/peptide complexes, they will be recognized by recipient T cells via the direct antigen presentation pathway, despite their position on recipient APCs. This would only be the case, however, if the acquired MHC molecules were on the cell surface, rather than intracellular, to allow contact with T-cell receptors on T cells. Therefore, confocal microscopy was used to determine the distribution of I-Ad and I-Ab on double-positive cells within the spleens of recipients. Again, I-Ad was detected with PE and I-Ab with fluorescein. The digital overlay shown in Fig. 1E
shows a cell positive for both donor and recipient class II MHC molecules. The ability of confocal microscopy to take images through a tissue section provides a 3D picture of the cell. The upper panel shows a cross section of the cell, and the lower panel is a transverse section taken at the position of the red line. Together, these images show staining of both MHC class II molecules on the surface of the cell. MHC class II staining was never seen within the cytoplasm of cells, which suggests that the colocalization of donor and recipient MHC class II on the surface of cells was not simply the first step in the uptake of these foreign molecules before processing and presentation via the indirect pathway.
To provide further evidence that the double-positive cells seen were not the result of the antibodies used recognizing processed MHC peptides rather than the intact MHC molecule, C57BL/6 mice were injected with lysed DBA/2 splenocytes. The lysate, containing DBA/2 MHC molecules, would presumably be taken up by recipient APCs, processed into peptides, and presented via the indirect pathway. No double-positive cells were seen in spleens from these mice (n=2) 4 days after injection (data not shown). This result suggests that the antibody against the donor MHC class II molecule (I-Ad) does not recognize processed peptides presented by recipient MHC molecules and that the epitope that it recognizes is not preserved during antigen processing, at least under the conditions used here. Three-color staining was also performed using an antibody against donor I-Ad molecules in combination with 2 different clones of antibodies recognizing recipient I-Ab molecules. All donor I-Ad-positive cells that were positive for recipient I-Ab clone AF6–120.1 were also positive for recipient I-Ab clone KH74 (Fig. 1F-I
). Because some of these triple-positive cells would be donor cells that had acquired recipient MHC class II molecules (see Direction of MHC transfer), this result suggests the transfer of whole, intact molecules.
MHC transfer within the graft
To investigate whether MHC transfer also occurs within donor grafts, tissue sections from kidney allografts were stained for recipient and donor MHC class II molecules. Although both donor and recipient MHC class II single-positive cells could be easily identified, no double-positive cells could be seen at 1, 2, or 4 days after transplantation (data not shown).
Direction of MHC transfer
To determine the direction of MHC class II transfer in our model, whether it was from donor to recipient cells, or vice versa, FISH was performed on spleen sections from C57BL/6 recipients of BALB/c kidneys. Because mRNA is strictly intracellular, it would not be transferred between cells. It is also unstable and therefore unlikely to survive following phagocytosis. A probe complementary to donor MHC class I messenger (m)RNA was used to identify donor cells (Fig. 1J, M
) in conjunction with an antibody against either donor (Fig. 1K
) or recipient (Fig. 1N
) MHC class II molecules. Probe-positive cells that were also positive for recipient MHC class II (i.e., donor cells that have acquired recipient MHC class II molecules) could be seen (Fig. 1O
). In addition, cells could be seen that were negative for donor MHC class I mRNA but positive for donor MHC class II molecules (Fig. 1L
), representing recipient cells that have taken up donor MHC class II molecules. The transfer of MHC molecules after transplantation is thus bidirectional. The fact that recipient cells can acquire donor MHC class II molecules suggests that the direct and indirect antigen presentation pathways can be physically linked, because recipient cells can also process donor antigens for presentation via the indirect pathway. Probe-negative cells accounted for 19.5 ± 4.1% of the donor-MHC class II-positive population, whereas 16.6 ± 3% of probe-positive cells were also positive for recipient MHC class II. So, MHC class II appears to be transferred in both directions in similar proportions, with no direction predominating.
Cell types involved in MHC uptake
Transfer of cell-surface molecules has been demonstrated for several cell types in vitro. But the exact cell types that are able to take part in this process in real-life, clinically relevant situations in vivo is unknown. In the case of transplantation, it is possible that certain cell types, which may be capable of acquiring cell surface molecules, may not in vivo be in the appropriate activation state or have access to foreign MHC class II-bearing cells due to their localization within the body. The type of cell positive for both donor and recipient MHC class II is extremely important regarding the influence of these double-positive cells. Only antigen-presenting cells would be able to interact with T cells and also possess the other attributes necessary to stimulate a full immune response. Therefore, we investigated which cell types carried out MHC transfer in our model. Because expression of MHC class II molecules is normally restricted to antigen-presenting cells, i.e., DCs, macrophages, and B cells, we used markers of these cells (CD11c, CD68, and CD19, respectively) to determine the lineages of the double-positive cells. An example of donor MHC class II, recipient MHC class II, and CD68 triple staining is shown in Fig. 1
(P–S). Following a BALB/c kidney transplant, the majority (62.5±7.3%) of the double-positive cells within the recipient spleen were DCs, with macrophages and B cells making up the rest (Fig. 5
A). Following transplantation of DBA/2 kidneys, double-positive cells were more evenly distributed among the 3 different cell types (Fig. 5B
). CD68 is also present at low levels on DCs, which may explain the high levels of CD68+ cells. Furthermore, each cell type surface marker was stained for on different sections of spleen, which may account for the percentage of each cell type not totaling 100%. These results, however, must be interpreted with caution because the possibility of intercellular transfer of these cell surface markers cannot be excluded.
|
Activation status of cells involved in MHC uptake
To determine if the donor and recipient MHC class II double-positive cells could provide the costimulation required for T-cell activation, the expression of CD40 and CD86 was also investigated. An example of donor MHC class II, recipient MHC class II, and CD86 triple staining is shown in Fig. 1
(T–W). The percentage of double-positive cells also positive for CD40 was 54.8 ± 3.8 and 52.9 ± 8.4% following BALB/c and DBA/2 kidney transplants, respectively, which was slightly, but not significantly, higher than the percentage of CD40 positive cells in the total donor class II-positive population (42.9±10 and 36.9±5.4% for BALB/c and DBA/2, respectively; Fig. 6
A). CD86 staining showed similar results (Fig. 6B
). After BALB/c kidney transplantation, 68.8 ± 5.8% of double-positive cells were CD86 positive, compared with 48.8 ± 7.7% of the overall donor MHC class II-positive population. After a DBA/2 kidney transplant, the percentages were 59.9 ± 10.2 and 53.2 ± 10.4%, respectively. This result is in keeping with the notion that mature cells carry out MHC transfer more efficiently (although APCs are thought to acquire MHC molecules in the steady state) (14)
, a theory that is also supported by the higher frequency of MHC transfer in the presence of inflammation early after transplantation. The high percentage of double-positive cells expressing costimulatory molecules also suggests that these cells would be capable of acting as effective APCs.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We have also stained C57BL/6 recipient spleen sections with an antibody against H-2 Kb (clone AF6–88.5). Because of the expression of this molecule on all recipient cells, the widespread fluorescence made it difficult to identify donor-derived cells not expressing this molecule. Purely for this technical reason, we chose to investigate the transfer of MHC class II molecules. Nonetheless, it is very likely that MHC class I molecules were transferred between donor and recipient cells in our model.
It is difficult to provide further proof in vivo that the intact MHC molecule is being transferred, but in vitro it has previously been convincingly demonstrated, using Western blotting to determine the molecular weight of the acquired molecule, that the size of the transferred molecule matches the size of the original molecule rather than the much smaller weight of a peptide (6)
. The functional abilities of acquired molecules in vitro (7
, 15
16
17
18)
lends further weight to the theory that whole molecules are transferred.
MHC transfer is similar after DBA/2 and BALB/c kidney transplantation, not only in the proportions of cells undergoing MHC transfer but also in the lineages of these cells and their activation states. This similarity occurs despite the fact that DBA/2 kidney allografts are usually accepted by C57BL/6 mice spontaneously, with no immunosuppression, as previously shown by us and others (12
, 13)
, whereas the majority of BALB/c kidney allografts are acutely rejected by these recipients (13)
. This result lends weight to the theory that MHC transfer is not a feature of one particular type of immune response but a natural consequence of any situation involving close contact of cells, and therefore any immune response.
The lack of MHC transfer within the graft was an unexpected finding. Although donor passenger leukocytes migrate out of the graft within several hours of transplantation, as evident from CD11c, CD19, and CD68 staining (unpublished observation), we were still able to observe donor MHC class II expression within allografts, presumably from endothelial or epithelial cells which are known to up-regulate MHC class II molecule expression during stress. The lack of double-positive cells within the allografts may be because only the splenic environment is conducive to this process. Alternatively, it is possible that only certain cell types are capable of "donating" MHC molecules in vivo, even though in vitro, transfer of MHC class II molecules from epithelial to DCs has been demonstrated (7)
.
The presence of both donor and recipient MHC class II molecules on the cell surface would allow these molecules to be recognized by their specific T-cell receptor to initiate an immune response. One double-positive cell could therefore interact with T cells that recognize donor MHC class II through the direct pathway and also with T cells of a different specificity, recognizing donor peptides presented by recipient MHC class II. This provides a physical route that can link the direct and indirect pathways of antigen presentation through a single double-positive cell, which could lead to bystander activation or linked suppression. The presence of donor and recipient MHC class II molecules on antigen-presenting cells and the expression of costimulatory molecules on these cells suggests that they would be able to provide both signals 1 and 2 to elucidate a full T-cell response.
The mechanisms responsible for the MHC transfer seen in our in vivo model here is unclear. At least in vitro, however, there appears to be more than one mechanism responsible for transfer of cell surface molecules (8)
. Transfer of cell surface molecules after binding to their ligand on neighboring cells is well documented and appears to occur most often after synapse formation; it has been shown that T cells acquire molecules much more efficiently from syngeneic APCs, that is, where synapses can be formed (19)
. Both the TcR and CD28 can acquire their respective ligands within a synapse (3
, 4
, 20)
, and acquisition of mouse Dd MHC class I molecules by NK cells has been shown to be dependent on the MHC class I receptor Ly49A (6)
. Cells without receptors for MHC molecules, however, are also able to acquire these molecules. DCs have been shown to acquire membrane from surrounding cells, not through formation of an immunological synapse but by "nibbling," a process that may involve the class A scavenger receptor present on DCs (21
, 22)
. Molecules can also be transferred between cells by the release of segments of membrane from one cell and the uptake of these by another cell. These membrane fragments come in 2 forms: exosomes, which are contained within multivesicular bodies within the cell and are secreted when the multivesicular body fuses with the cell membrane (23)
, and microvesicles, which are formed when sections of the plasma membrane bud off the cell (24)
. Both exosomes and microvesicles can travel between cells along nanotubular networks (25
, 26)
.
The functional consequences of MHC transfer in vivo are unknown. One obvious example is the linking of the direct and indirect antigen presentation pathway (7)
. Because regulatory T cells are thought to mostly recognize antigens through the indirect pathway (27)
, it is tempting to suggest that this phenomenon provides a physical link between regulatory T cells and alloreactive T cells of the direct pathway.
Another possible function of this phenomenon is that cells may act to spread processed antigen from invading organisms by acquiring MHC/peptide complexes from infected cells. Although we have used transplantation models to demonstrate intercellular MHC transfer here, it is quite possible that this process also takes place under normal physiological circumstances. In the same way that donor-derived APCs were shown here to travel to the spleen to exchange MHC molecules with recipient APCs, activated APCs from an infected organ, site of vaccination or tumor could travel to secondary lymphoid organs. They could then either present viral peptides, antigenic peptides from the vaccine, or tumor-specific peptides directly to T cells or transfer them to other APCs, thus amplifying the process of T-cell priming. Indeed, increased antigen-presenting capability has been demonstrated in B cells following antigen acquisition (2
, 3)
. In addition, for T cells, it is thought that MHC class I uptake can render the T cells sensitive to killing by surrounding T cells with the same specificity, that is, fratricide (3)
. In contrast, the transfer of MHC class I molecules to NK cells at the site of infection could serve to protect them from neighboring NK cells, which kill cells with low or absent self class I MHC molecules (missing self hypothesis).
Whereas there are potential benefits of MHC transfer, whether this process is regulated is still unknown (8)
. One of the ways this may happen is through the stability of expression of the transferred molecules. Whereas MHC molecules have been found to remain on recipient cells for as long as 2 days following transfer (15)
, costimulatory molecules such as CD80 and OX40L seem to be degraded much more quickly, with expression lasting up to 4 and 12 h, respectively (4
, 28)
. After loss of costimulatory molecules, cells that have acquired MHC molecules will no longer be able to stimulate T cells and may actually down-regulate the immune response.
Our results demonstrate the exchange of MHC molecules between donor and recipient cells after transplantation. The high percentages of cells expressing both donor and recipient MHC molecules and their expression of costimulatory molecules suggest that they may play an important role in the immune response. Transplantation is a obvious model to study this phenomenon because donor and recipient MHC molecules are different, but it is likely that MHC transfer also takes place under normal physiological conditions and has implications in the fields of infection, vaccination, and tumor immunology.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 12, 2008. Accepted for publication June 19, 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. Atkinson, J. C. Varela, and S. Tomlinson Complement-Dependent Inflammation and Injury in a Murine Model of Brain Dead Donor Hearts Circ. Res., November 20, 2009; 105(11): 1094 - 1101. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
