Described is the preparation of antigenic liposomal nanoparticles and their use in stimulating B-cell activation in vitro and in vivo. Consistent and robust antibody responses led to the development of a new peanut allergy model. The protocol for generating antigenic liposomes can be extended to different antigens and immunization models.
Antibody responses provide critical protective immunity to a wide array of pathogens. There remains a high interest in generating robust antibodies for vaccination as well as understand how pathogenic antibody responses develop in allergies and autoimmune disease. Generating robust antigen-specific antibody responses is not always trivial. In mouse models, it often requires multiple rounds of immunizations with adjuvant that leads to a great deal of variability in the levels of induced antibodies. One example is in mouse models of peanut allergies where more robust and reproducible models that minimize mouse numbers and the use of adjuvant would be beneficial. Presented here is a highly reproducible mouse model of peanut allergy anaphylaxis. This new model relies on two key factors: (1) antigen-specific splenocytes are adoptively transferred from a peanut-sensitized mouse into a naïve recipient mouse, normalizing the number of antigen-specific memory B- and T-cells across a large number of mice; and (2) recipient mice are subsequently boosted with a strong multivalent immunogen in the form of liposomal nanoparticles displaying the major peanut allergen (Ara h 2). The major advantage of this model is its reproducibility, which ultimately lowers the number of animals used in each study, while minimizing the number of animals receiving multiple injections of adjuvant. The modular assembly of these immunogenic liposomes provides relatively facile adaptability to other allergic or autoimmune models that involve pathogenic antibodies.
Food allergy affects 8% of children in the United States, and has increased in prevalence over the past decade1. Allergy to peanut affects 1% of children and is not typically outgrown2. Although several promising clinical trials are underway for the treatment of food allergy, including oral immunotherapy (OIT), sublingual immunotherapy (SLIT), and epicutaneous immunotherapy (EPIT), there are currently no FDA-approved treatment strategies for desensitizing peanut-allergic individuals3,4,5,6,7,8. Therefore, allergic individuals must strictly avoid allergens to avoid anaphylaxis. Many questions remain regarding routes of sensitization and underlying mechanisms of food allergy development.
Mouse models are a valuable tool for studying the mechanisms of allergy as well as developing new tolerogenic and desensitization therapies9,10,11,12. This is particularly true because the major peanut allergen (Ara h 2; Ah2) in humans is also the dominant allergen in several described mouse models13,14. While mouse models of peanut allergy are invaluable in studying mechanisms of sensitization and tolerance, a drawback is that they can be variable and require the use of adjuvants. More potent immunogens would be one way to minimize the intrinsic variability of such models. Since B-cells are strongly activated by multivalent antigens, antigenic liposomes displaying the allergen are a good option because of their ability to potentially activate B-cells through the B-cell receptor (BCR) while also having the property of efficiently priming the T-cell compartment through being taken up non-specifically by antigen-presenting cells.
Here, we describe a detailed protocol for conjugating protein antigens to liposomal nanoparticles using a facile and modular strategy. Using a surrogate antigen, anti-IgM Fab fragment, we demonstrate how potent such antigenic liposomes can be in stimulating B-cell activation. Antigenic liposomes displaying Ah2 antigen were used to develop a new mouse model of conferred sensitivity. In this model, splenocytes from verified peanut allergic mice, containing peanut-specific memory B- and T-cells, are transferred into naïve congenic mice. Memory antibody responses are induced by injection of liposomes conjugated with Ah2 into the recipient mice, in order to induce antibodies against Ah2. Followed by only one boost with soluble Ah2, Ah2-specific antibodies give rise to a strong anaphylactic response when these mice are subsequently challenged with Ah2. As mice undergoing the allergic reaction respond in a highly uniform manner and have not received an adjuvant, this approach is a desirable peanut allergy model and the outcomes suggest that it may have utility in other mouse models driven by antigens directed at allergens and possibly autoantigens.
The general method of coupling protein to lipid and incorporating into liposomes is based largely on earlier work15. All animal procedures described below have been approved by the University of North Carolina at Chapel Hill Institutional Animal Care and Use Committee (IACUC). All mice used in the peanut allergy model are BALB/cJ females purchased from at 3 weeks of age. The University of Alberta Animal Care and Use Committee (ACUC) has approved experiments involving use of mouse spleens for ex vivo analysis from C57Bl/6 mice of at least 6 weeks of age.
1. Conjugation of Protein Antigen to PEGylated Lipid
2. Liposome Preparation
3. Liposome Extrusion
4. Calcium Flux to Monitor B-cell Activation by Antigenic Liposomes
5. Preparation of Peanut Extract
6. Sensitization of Mice to Peanut
7. Challenge Mice with Peanut Extract
8. Isolation of Splenocytes from Allergic Mice and Adoptive Transfer
9. Injection of Mice with Ara h 2 Antigenic Liposomes
10. Boost and Challenge of Mice with Ara h 2
11. Quantification of Ara h 2-specific IgE and IgG1 by ELISA
Conjugation of the protein of interest with DSPE-PEG(2000) can be demonstrated by running a reducing showing an increase in molecular weight compared to the unconjugated protein. Figure 1A shows a representative gel of anti-mouse IgM F(ab) fragment conjugation to PEG-DSPE, which shows a 2–3 kDa bandshift for the denatured protein. Note that approximately 50% of the protein appears to be modified, which is expected given that 1:1 stoichiometry was achieved on the Fab fragment that is a heterodimer of the heavy and light chain. Figure 1B shows a representative gel of Ah2 conjugation to PEG-DSPE. To assess calcium flux of B-cells stimulated by antigenic liposomes, two things are crucial: (1) the instrument settings are tuned to see a difference in the ratio of Indo-1 fluorescence in the Ca2+ bound (violet) and unbound (blue) forms and (2) the proper gating strategy is used to assess B-cell activation. Figure 2A shows the gating scheme for the flow cytometry-based calcium flux assay. Live lymphocytes are gated in an SSC-A versus FSC-A (left panel), doublets are gated out in a FSC-W versus SSC-W plot (middle panel), and B200+CD5– B-cells are selected from a CD5-PE vs B220-PE/Cy7 plot (right panel). Figure 2B demonstrates the ratio of Indo-1 (violet) vs Indo-1 (blue) fluorescence over time as analyzed. Note that the total protein concentration of F(ab) and F(ab')2 in these assays were the same, demonstrating the superior ability of antigenic liposomes in stimulating B-cell activation. After preparing the peanut extract, run an aliquot on an SDS-PAGE gel to determine the relative quantities of Ara h 1, 2 and 3 within the extract. Figure 3 shows a representative gel of a peanut extraction that was run alongside purified Ah2. Figure 4 shows a schematic of the overall adoptive peanut allergy mouse model, including initial sensitization to peanut, challenge to peanut, splenocyte isolation and adoptive transfer, liposome injections, blood draw followed by Ah2 boost, and challenge to Ah2. Ah2-specific IgE and IgG1 ELISAs were run to quantify immunoglobulins in serum, as shown in Figure 5A and B. Mice with conferred sensitivity that have been boosted with Ah2 will have Ah2-specific IgE and IgG1 in their serum. Body temperatures recorded during the Ah2 challenge are shown in Figure 5C; allergic mice had decreased body temperatures following the challenge, while body temperatures in naïve mice remained consistent. Photos showing naïve mice compared to peanut-allergic mice during the challenge are shown in Figure 6.
Figure 1: Representative gel of anti-mouse IgM F(ab) and Ah2 conjugation to PEG-DSPE. (A, B) SDS-page analysis of (A) goat anti-mouse IgM F(ab) fragment and (B) Ah2 before and after conjugation to PEG-DSPE. Goat anti-mouse IgM and Ah2 have molecular weights of approximately 48 and 18 kDa, respectively. Following modification, a slightly larger molecular weight (~2.5 kDa) is clearly observed for both proteins. Please click here to view a larger version of this figure.
Figure 2: Representative gating strategy, calcium flux results, and analysis. (A) Cells were analyzed through the following gating strategy: live lymphocytes (FSC-A vs. SSC-A), single cells (FSC-W vs. SSC-W), and B-cells (B220+CD5–). (B) The Indo-1/Ca2+ flux response (violet vs. blue) of the B-cells. Shown is the calcium flux for anti-IgM Fab fragment liposomes and anti-IgM F(ab')2 at the same protein concentration (2.5 µg/mL) as well as buffer-stimulated cells as a control. Note that between 10-22 s is when the stimulation is added and, therefore, no data is acquired during this time. Please click here to view a larger version of this figure.
Figure 3: Representative gel showing peanut extract and Ah2. Peanut extract contains several proteins, including allergens Ara h 1, 2, 3 and 6, compared to the two isoforms that appear for Ah2. Please click here to view a larger version of this figure.
Figure 4: Schematic of adoptive transfer protocol. Naïve BALB/cJ mice were sensitized with peanut extract (PN) and cholera toxin (CT), and subsequently challenged to PN. Splenocytes from confirmed allergic mice were isolated and transferred into naïve mice. Mice were later primed with immunogenic Ara h 2 liposomes or PBS, then boosted with soluble Ah2. Finally, mice were challenged with Ah2 to monitor anaphylaxis. Blood collected on days 45 and 60 were used to quantify Ah2-specific immunoglobulins. Please click here to view a larger version of this figure.
Figure 5: Immunogenic Ara h 2 liposome boosts conferred-memory allergic responses. Serum was isolated pre (day 45) and post (day 60) boost with PBS or 200 µL of 300 µM immunogenic Ah2 liposome to measure Ah2-sepcific IgE (A) and IgG1 (B). Individual mice are represented with lines indicating medians. Mice that received allergic splenocytes and were administered Ah2 liposomes on day 32 had significantly higher levels of Ah2-specific IgE and IgG1. Anaphylaxis was measured in the different treatment groups by recording body temperatures after Ah2challenge (C). Mean body temperatures depicted with SEM. Naïve: PBS (Naïve splenocytes and PBS prime), Allergic: PBS (confirmed allergic splenocytes and PBS prime), Allergic: Ah2 (confirmed allergic splenocytes and Ah2immunogenic liposome prime). * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001 determined by unpaired 2-tailed Student's t-test. Note that these results are representative of two independent experiments. Please click here to view a larger version of this figure.
Figure 6: Anaphylactic symptoms observed during Ah2 challenge. Naïve mice remain active and have pink complexion on their feet (A, C). Allergic mice have decreased activity, are often hunched up, have labored breathing and experience cyanosis, indicated by darker purple complexion on their feet (B, D). Please click here to view a larger version of this figure.
MW | 790 | 387 | 2900 | 3000 | 48000 | |
DSPC | Cholesterol | PEG-DSPE | excess PEG-DSPE | aIgM-PEG-DSPE | TOTAL | |
Molar ratio | 57 | 38 | 3.9 | 1 | 0.1 | 100.00 |
mass (mg) | 0.56 | 0.18 | 0.14 | 0.04 | 0.06 | 0.98 |
mmol | 0.71 | 0.47 | 0.05 | 0.01 | 0.00 | 1.25 |
mL | 112.50 | 45.93 | 70.64 | 0.00 | 525.97 | 755.03 |
Conc. (mg/mL) | ||||||
DSPC | 5 | |||||
Cholesterol | 4 | |||||
PEG-DSPE | 2 | |||||
aIgM-PEG-DSPE | 0.114 |
Table 1: Calculations for creating a 1.25 µM total lipid concentration liposome. Example of calculation table for goat anti-mouse IgM F(ab) fragment liposomes.
The methods outlined here are a general protocol for the conjugation of a protein to a lipid which enables the display of the protein on liposomal nanoparticles. For very large multi-subunit proteins, this protocol may have limited utility. The ideal method would be the introduction of a site-specific tag that enables a biorthogonal chemical linking strategy to be used. If expressing the protein recombinantly, this can be possible using available site-specific strategies17, and a wide array of functional groups at the end of the PEGylated lipid that are commercially available. Accordingly, this protocol is geared towards the linking of protein isolated from natural sources and is accessible to a wide-range of scientists not necessarily familiar with chemical and biochemical transformations.
A key aspect of running a calcium flux experiment that should to be emphasized is that the cells should not be activated before running them on the flow cytometry. If sterile technique is not rigorously used, this will account for a high background signal from activated cells (Indo-1 fluorescence skewed towards its emission in the violet channel). Similarly, cells should be kept on ice prior to running the assay to avoid abnormal levels of background signal in the Indo-1 channels. It is also important to ensure that the cells are warmed for 2–3 min prior to acquiring by flow cytometry and throughout stimulation during data acquisition. If the cells are not maintained at 37 °C during data acquisition, it can be expected that cells will not respond maximally. Since antigen-specific B-cells are present in mice at very small numbers, using a surrogate antigen to stimulate all B-cells from mice is ideal. Although numerous transgenic mouse strains are available that express antigen-specific B-cells, the purpose of this technique is to enable the average user to be able to validate their ability to make antigenic liposomes using wild-type mice. To accomplish this objective, Fab fragments of anti-mouse IgM were linked to lipids and formulated into antigenic liposomes that cross-link the B-cell receptor (BCR) in a polyclonal fashion. It is noteworthy that it was previously demonstrated that this method can be used to stimulate human B-cells10 using the appropriate anti-human IgM Fab fragment, which is significant because it is not possible to access significant numbers of antigen-specific human B-cells. In these studies, the potency of antigenic liposomes to stimulate B-cell activation was illustrated by their enhanced ability to induce calcium flux compared to equal amounts of anti-mouse IgM F(ab')2 fragments.
The use of antigenic liposomes has enabled the development of a method for sensitizing BALB/cJ mice to peanut, transferring splenocytes from allergic mice into naïve mice, injecting the host mice with Ah2 liposomes, and challenging them with Ah2. Antibody levels and anaphylactic responses within groups of mice are reproducible in this model. It has been well-established that memory B- and T-cell responses are critical in the development of peanut-specific antibodies in mice. Moreover, a recent study has shown that memory B-cells repopulate plasma cells in mice, which maintain elevated antigen-specific IgE levels; therefore allergen-specific memory B-cells are an attractive target for therapeutic intervention18. The model described here now allows an opportunity to directly target Ah2-specific memory B cells, as opposed to an immunosuppressive approach using bortezomib to deplete the entire plasma cell repertoire19. Another approach to combat allergies is to induce tolerance in the T-cell compartment. One potential strategy is to encapsulate tolerance-inducing adjuvants within the liposome. Previously, nanoparticles encapsulating rapamycin were used to induce Tregs and diminish autoimmune disease20. This type of approach could be employed to develop tolerogenic therapies to treat peanut allergy. Overall, manipulation of the allergen-specific liposome by conjugating molecules to the surface allows for tailored immune responses, and encapsulating drug-like molecules enables delivery of drug to the allergen-specific cell.
General limitations of this mouse model of peanut allergy are that sensitization to peanut with Cholera toxin and subsequent peanut challenge via i.p. injection is not entirely representative of peanut allergy in humans (e.g., humans have reactions upon oral ingestion). However, this allergy model is the current standard in the field11,18,21, and allows us to investigate molecular mechanisms of allergic disease and develop novel therapeutic approaches. Compared to these conventional approaches, this adoptive transfer model offers three primary advantages. The first is that it enables memory B- and T-cells specific to the allergen to be studied more directly. Indeed, evidence now strongly implicates memory B-cells as the major source of long-term allergies18. Second, because the same number of memory B- and T-cells are implanted into each recipient mouse, the allergic responses are minimally variable. Third, this model provides the opportunity to study allergic responses to individual allergens, which is useful in the context of testing new immunotherapies. One intriguing application of this model is to use the general approach in the context of a humanized mouse model, where allergic responses have been studied in immunized models22. Applied in the context of an adoptive transfer model, a humanized mouse model could be a powerful approach to expanding and in vivo testing of B- and T-cells from allergic patients.
Antibody disease pathogenesis does not only occur in allergic models, but also in B-cell mediated autoimmune diseases23. Adoptively transferring sensitized splenocytes from mice given an autoantigen would ultimately be very useful in lowering the number of animals used, minimize the number of animals receiving multiple injections of adjuvant, and increase the reproducibility of responses between cohorts of mice in multiple disease indications. One example would be in an experimental autoimmune model of myasthenia gravis (EAMG) in which susceptible mouse strains (C57BL/6, SJL, and AKR) are immunized multiple times with acetylcholine receptor (AChR) in adjuvant to develop pathogenic autoantibodies. These autoantibodies ultimately lead to immunopathological features present in humans such as muscle fatigue/weakness, deposit of immunoglobulins, and complement components on neuromuscular junctions and in severe cases morbidity/death24. In EAMG, the disease incidence varies widely between 50–70%, so to account for this variability large animal cohorts are needed to reach statistical significance25,26. The method described here would ultimately reduce the number of animals used by decreasing the variability in disease incidence. As previously mentioned, EAMG is induced by multiple injections with AChR + adjuvants (Complete Freund's Adjuvant and Incomplete Freund's Adjuvant) to trigger auto-antibody responses. This method would reduce the number of animals that receive multiple injects with adjuvant. Finally, due to the nature of multiple immunizations and boosts, it is relatively difficult to define appropriate therapeutic windows for prophylactic and therapeutic treatments. This method would help synchronize the antibody response and disease pathogenesis window, making it a more predictable and reproducible. This same logic can be applied to many other mouse models of antibody mediated autoimmune diseases, such as Rheumatoid arthritis27.
The authors have nothing to disclose.
This research was supported by grants from the Department of Defense (W81XWH-16-1-0302 and W81XWH-16-1-0303).
Model 2110 Fraction Collector | BioRad | 7318122 | |
Cholestrol | Sigma | C8667 | Sigma grade 99% |
SPDP | Thermo Fisher Scientific | 21857 | |
DSPC | Avanti | 850365 | |
DSPE-PEG 18:0 | Avanti | 880120 | |
DSPE-PEG Maleimide | Avanti | 880126 | |
Extruder | Avanti | 610000 | 1mL syringe with holder/heating block |
Filters 0.1 µm | Avanti | 610005 | |
Filters 0.8 µm | Avanti | 610009 | |
10mm Filter Supports | Avanti | 6100014 | |
Glass Round Bottom Flask | Sigma | Z100633 | |
Turnover stoppers | Thermo Fisher Scientific | P-301398 | |
Tubing | Thermo Fisher Scientific | P-198194 | |
Leur Lock | Thermo Fisher Scientific | k4201634503 | |
Sephadex G50 Beads | GE Life Sciences | 17004201 | |
Sephadex G100 Beads | GE Life Sciences | 17006001 | |
Heat Inactivated Fetal Calf Serum | Thermo Fisher Scientific | 10082147 | |
HEPES (1M) | Thermo Fisher Scientific | 15630080 | |
EGTA | Sigma | E3889 | |
Penicillin-Streptomycin (10,000 U/mL) | Thermo Fisher Scientific | 15140122 | |
1x RBC lysis Buffer | Thermo Fisher Scientific | 00-4333-57 | |
Indo-1 | Invitrogen | I1203 | |
CD5-PE | BioLegend | 100608 | |
B220-PE-Cy7 | BioLegend | 103222 | |
HBSS | Thermo Fisher Scientific | 14170112 | without calcium and magnesium |
MgCl2 | Sigma | M8266 | |
CaCl2 | Sigma | C4901 | |
Fab anti-mouse IgM | Jackson ImmunoResearch | 115-007-020 | |
F(ab')2 anti-mouse IgM | Jackson ImmunoResearch | 115-006-020 | |
Peanut flour | Golden Peanut Co. | 521271 | 12% fat light roast, 50% protein |
Animal feeding needles | Cadence Science | 7920 | 22g x 1.5", 1.25 mm – straight |
Microprobe thermometer | Physitemp | BAT-12 | |
Rectal probe for mice | Physitemp | Ret-3 | |
Cholera toxin, from vibrio cholera | List Biological Laboratories, Inc. | 100B | Azide free |
BCA Protein Assay Kit | Pierce | 23225 | |
Carbonate-bicarbonate buffer | Sigma | C3041 | |
TMB Stop Solution | KPL | 50-85-06 | |
SureBlue TMB Microwell Peroxidase Substrate | KPL | 5120-0077 | |
96 well Immulon 4HBX plate | Thermo Scientific | 3855 | |
Purified soluble Ara h 2 | N/A | N/A | purified as in: Sen, et al., 2002, Journal of Immunology |
HSA-DNP | Sigma | A-6661 | |
Mouse IgE anti-DNP | Accurate Chemical | BYA60251 | |
Sheep anti-Mouse IgE | The Binding Site | PC284 | |
Biotinylated Donkey anti-Sheep IgG | Accurate Chemical | JNS065003 | |
NeutrAvidin Protein, HRP | ThermoFisher Scientific | 31001 | |
Mouse IgG1 anti-DNP | Accurate Chemical | MADNP105 | |
HRP Goat anti-mouse IgG1 | Southern Biotech | 1070-05 | |
1 mL Insulin Syringes | BD | 329412 | U-100 Insulin, 0.40 mm(27G) x 16.0 mm (5/8") |
Superfrost Microscope Slides | Fisher Scientific | 12-550-14 | 25 x 75 x 1.0 mm |
ACK Lysing Buffer | gibco by Life Technologies | A10492-01 | 100 mL |
RPMI 1640 Medium | Thermo Fisher Scientific | 11875093 | 500 mL |
Cell Strainer | Corning | 352350 | 70 μm Nylon, White, Sterile, Individually packaged |
NuPAGE 4-12% Bis-Tris Protein Gels | Invitrogen | NP0322BOX | 10 gels |
NuPAGE LDS buffer, 4X | Invitrogen | NP0008 | 250 mL |
SeeBlue Plus2 Pre-stained standard | Invitrogen | LC5925 | 500 µL |
NuPAGE MES/SDS running buffer, 20X | Invitrogen | NP0002 | 500 mL |
GelCode Blue Stain | Thermo Scientific | 24590 | 500 mL |