Vaccination with mycobacterial lipid loaded nanoparticle leads to lipid antigen persistence and memory differentiation of antigen-specific T cells

Mycobacterium tuberculosis (Mtb) infection elicits both protein and lipid antigen-specific T cell responses. However, the incorporation of lipid antigens into subunit vaccine strategies and formulations has been underexplored, and the characteristics of vaccine-induced Mtb lipid-specific memory T cells have remained elusive. Mycolic acid (MA), a major lipid component of the Mtb cell wall, is presented by human CD1b molecules to unconventional T cell subsets. These MA-specific CD1b-restricted T cells have been detected in the blood and disease sites of Mtb-infected individuals, suggesting that MA is a promising lipid antigen for incorporation into multicomponent subunit vaccines. In this study, we utilized the enhanced stability of bicontinuous nanospheres (BCN) to efficiently encapsulate MA for in vivo delivery to MA-specific T cells, both alone and in combination with an immunodominant Mtb protein antigen (Ag85B). Pulmonary administration of MA-loaded BCN (MA-BCN) elicited MA-specific T cell responses in humanized CD1 transgenic mice. Simultaneous delivery of MA and Ag85B within BCN activated both MA- and Ag85B-specific T cells. Notably, pulmonary vaccination with MA-Ag85B-BCN resulted in the persistence of MA, but not Ag85B, within alveolar macrophages in the lung. Vaccination of MA-BCN through intravenous or subcutaneous route, or with attenuated Mtb likewise reproduced MA persistence. Moreover, MA-specific T cells in MA-BCN-vaccinated mice differentiated into a T follicular helper-like phenotype. Overall, the BCN platform allows for the dual encapsulation and in vivo activation of lipid and protein antigen-specific T cells and leads to persistent lipid depots that could offer long-lasting immune responses.


49
Vaccination is currently the most effective method for the prevention and eradication of 50 infectious diseases. These efforts have been enhanced by the development of subunit vaccine 51 formulations that include only immunodominant antigens from pathogens paired with select adjuvants.

52
In contrast to attenuated and inactivated vaccines that contain diverse molecular components of 53 pathogens, subunit vaccines are often preferred due to their simplicity, safety, stability, and scalability 54 of production [1]. Yet the development of effective vaccines for several key endemic parasitic and 55 bacterial pathogens has remained elusive. Antigen selection for subunit vaccines has thus far focused Results 116 117 BCN morphology preserved after MA loading 118 We scalably assembled spherical BCN with and without loaded MA using flash 119 nanoprecipitation (FNP) as previously described [29]. Dynamic light scattering (DLS) showed that 120 both MA-BCN and BCN were consistent in size (364 ±19 nm and 354 ± 14 nm, respectively), and 121 were monodisperse based on their polydispersity indices (PDIs) of 0.24 ± 0.04 and 0.21 ±0.05, 122 respectively (Fig 1A). We next verified that the BCN formulation maintained its characteristic 123 interconnected aqueous channels using cryogenic transmission electronic microscopy (cryo-TEM) (Fig   124   1B) and negative staining transmission electron microscopy (TEM) (Fig 1C). Using small angle X-ray 125 scattering (SAXS) studies, we confirmed the internal cubic organization of BCN. Bragg peaks at the 126 √2, √4, and √6 ratios show that the primitive type of cubic internal organization was preserved between 127 MA-BCNs and BCNs (Fig 1D). Thus, MA encapsulation did not disturb the BCN architecture. We 128 also manufactured a PLGA nanocarrier formulation (PLGA-NP) with and without MA encapsulation. 7 and human (M11) MA-specific T cells in vitro and inducing IFN-g production (Fig 2A-D). In 143 particular, MA-BCN were significantly better at activating MA-specific T cells compared to free MA 144 at equivalent concentrations (Fig 2A, B and D). In comparison to MA-BCN, MA-PLGA was 145 significantly more effective at stimulating mouse MA-specific T cells in vitro at lower concentrations. 146 Interestingly, even in the absence of MA, PLGA-NP showed strong stimulation particularly for M11 T 147 cells. These results demonstrate that MA encapsulation within BCN enhances its ability to stimulate 148 antigen-specific T cells and reveal a considerable background, and thus difficult to control, stimulatory (LN) and lungs at 1 week post-vaccination (Fig 2E-G). The extent of cell proliferation induced by observed [34]. 170 Since BCN allow for the co-loading of lipid and protein antigens, we assessed the ability of 171 dual loaded Ag85B-MA-BCN to activate antigen-specific T cells as well as lead to antigen persistence. 172 We co-loaded Ag85B and MA in BCN and found Ag85B-MA-BCN to have a diameter of 380 nm and 173 PDI of 0.223, comparable to MA-BCN (Fig 4A). Encapsulation efficiency was also relatively high at 174 70%.Vaccination of hCD1Tg mice with Ag85B-MA-BCN activated and induced proliferation of both 175 p25 and DN1 T cells in the LN, lung, and spleen (Fig 4B-D) at one-week post-vaccination. In addition, 176 DN1 T cells were still able to be activated 6 weeks post-vaccination in the context of an Ag85B-MA-177 BCN vaccination, while p25 T cells did not show increased activation or proliferation compared to 178 blank BCN control at this time point (Fig 4E-H). Therefore, MA, but not Ag85B, appears to persist in 179 the context of an Ag85B-MA-BCN vaccination.  (Fig 6C). 225 We next tested whether MA likewise persisted primarily within AMs. 6-weeks after 226 vaccination, we enriched for AMs using a column-based magnetic cell isolation system containing 227 SiglecF antibody, which is highly expressed on AMs and a small population of eosinophils in the lung 228 (Fig 6D). We co-cultured enriched AMs or the flow through with or without hCD1Tg-expressing 229 BMDCs and DN1 T cells (Suppl Fig 6). While both enriched AM and flow through fractions from 230 MA-BCN vaccinated mice co-culture with BMDCs could activate DN1 T cells (Fig 6E), co-culture 231 with enriched AMs led to significantly higher DN1 T cell activation, suggesting AMs are the primary 232 location of MA persistence. The ability of flow through to likewise lead to DN1 T cell activation 233 suggests MA may also persist in other cell types or could be attributed to residual AMs in flow 234 through. Furthermore, AMs could not themselves present the MA to DN1 T cells, noted by the lack of used to define central memory T cells, particularly in the LN (Fig 7C and Supp Fig 7A). To 250 characterize the memory DN1 T cells, we performed RNA-seq analysis on sorted CD44 + CD62L + 251 (memory) and CD44 -CD62L + (naïve) DN1 T cells from LNs of MA-BCN vaccinated DN1-hCD1Tg 252 BM chimeric mice. We found that memory and naïve DN1 T cells clustered separately after principal 253 component analysis (PCA), despite these samples coming from the same animals (Suppl Fig 7B). A 254 total of 995 differentially expressed genes (DEGs) were identified of which 542 upregulated and 453 255 downregulated in the memory subset (Suppl Fig 7C). Next, we determined which T cell population 256 memory DN1 T cells most resembled. Toward this end, we obtained data from ImmGen database 257 (including terminally differentiated effector (Te), memory precursor (Tmp), central memory (Tcm), 258 effector memory (Tem), regulatory (Treg)) 42 , and two additional publications (follicular helper (Tfh) 43 , 259 exhausted (Texh) 44 ) and compared the respective DEG lists. Using PCA, we found that memory DN1 260 T cells clustered most closely to Tfh, Treg, and Texh (Fig 7D) and when this subset is isolated, most 261 closely to Tfh cells (Fig 7E). 262 Within the DEGs, we noted upregulation of key Tfh cell transcription factors BCL6 and TCF1

278
Research on the development of subunit vaccines containing lipid antigens is limited. This is in 279 part due to a lack of research into appropriate lipid vaccine formulations and their associated 280 properties. Some early evidence suggests that lipid vaccines could contribute to protection in Mtb 281 challenge in guinea pig model [38,39], which naturally express group 1 CD1 molecules [40].

282
Vaccination with a formulation composed of total Mtb lipid encapsulated within liposomes with either 283 dimethyl-dioctadecyl-ammonium (DDA) and/or QS-21 adjuvants showed lower bacterial burden and 284 pathology in lung and spleen compared to unvaccinated controls [38]. In another study, two Mtb lipids 285 (diacylated sulfoglycolipids and the phosphatidylinositol dimannosides) encapsulated into liposomes 286 formulated with DDA and trehalose 6,6'-dibehenateled (TDM) to a decreased bacterial burden in the 287 spleen and overall improved pathology in lung and spleen [39]. This work motivated our current study, 288 where we used hCD1Tg mice that express human group 1 CD1 molecules, including CD1b and MA-289 specific (DN1), as well as p25-specific T cells to investigate the use of nanocarriers coloaded with both 290 Mtb lipid and protein antigens during vaccination (Fig 8). 291 We found that PEG-b-PPS BCN could effectively encapsulate MA as well as the protein 292 antigen Ag85B to elicit activation of antigen-specific T cells in vaccinated mice (Fig 4). To our 293 knowledge, this is the first time a dual protein and lipid loaded nanocarrier has been shown to 294 effectively stimulate both lipid and peptide antigen-specific T cells simultaneously in vivo. PEG-b-PPS 295 BCN can encapsulate both hydrophobic and hydrophilic molecules, making it an excellent platform for 296 subunit vaccine delivery as both antigens and adjuvants can be co-loaded and delivered to the same 297 cellular targets. Prior work on TB vaccines has shown that the inclusion of several protein antigens 298 spanning the Mtb life cycle along with adjuvants with multiple boosts is paramount in the development 299 of a vaccine that can provide effective protection against Mtb [30,41]. As we have established that 300 PEG-b-PPS BCN can easily incorporate a variety of payloads, future work will involve determining 301 the optimal antigen and adjuvant combination for providing protection in a Mtb challenge experiment.

302
. CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. ;https://doi.org/10.1101https://doi.org/10. /2023 As several formulations of PLGA-NP have been approved for clinical uses [42], we compared 303 MA encapsulated by either BCN or PLGA-NP and found that while PLGA appeared to offer greater, 304 yet less controllable stimulation in vitro (Fig 2), BCN exhibited significantly greater efficacy in vivo 305 (Fig 2). The result is not completely unexpected as a more fragile PLGA-NP could allow for fast 306 intracellular release and antigen presentation in vitro while also leading to significant antigen loss prior 307 to APC interaction in vivo (Fig 3) (Fig 2), allowing their immunomodulation to be dictated  [48]. We found that in the pulmonary route of vaccination, BCN primarily were taken up and remained 320 within alveolar macrophages in the lung (Fig 6). In fact, the ability of nanocarriers to target alveolar 321 macrophages or other APCs may contribute to their distinct capacity for inducing antigen persistence, 322 which was not seen with cellular carriers such as MA-pulsed BMDCs (Fig 5). Alveolar macrophages 323 are the first line of defense of the lungs, responsible for keeping alveoli sterile by taking up inhaled 324 particles and pathogens through phagocytosis [49][50][51]. Targeting these cells may be particularly 325 beneficial in a TB vaccine design since they may also be the main site of Mtb latency [52]. While MA 326 remained within alveolar macrophages in the lung, T cell activation appeared to occur prominently 327 within the draining lymph nodes (Fig 2), suggesting that either MA was transported from the lungs to 328 lymph nodes where the presentation took place or T cells activated in the lung migrated to the lymph 329 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. ; https://doi.org/10.1101/2023.03.07.531489 doi: bioRxiv preprint nodes. The first option seems most likely, as CD1b-expressing DCs are more prevalent in the lymph 330 nodes [53]. Furthermore, in Mtb infection models, it has been previously shown that apoptotic 331 macrophage and vesicles could deliver mycobacterial antigens to DCs that then activate cognate T 332 cells [54,55]. Similarly, we showed that alveolar macrophages from MA-BCN vaccinated mice could 333 only lead to T cell activation in the presence of BMDCs from hCD1Tg mice (Fig 6). 334 We demonstrated MA persistence in the lung 6 weeks post-vaccination using an in vivo antigen antigen archiving by LECs [11,12]. Similar to our findings, antigen-specific T cell activation required  (Fig 6). The lack of persistence for Ag85B could be due to its leakage out from the 350 aqueous pores of BCN, which we have observed for other proteins [29]. Peptide antigens may be 351 modified to include a hydrophobic tail for stable retention within BCN and therefore allow for peptide 352 antigen persistence as well. MA's extreme hydrophobic surface, which correlates with its low 353 permeability as a part of Mtb cell wall, may also contribute to its prolonged persistence in vivo [62].

354
After infection and vaccination, peptide antigen persistence has been shown to occur through 355 archiving by binding to CD21 receptors on follicular dendritic cells (FDCs) in lymph nodes [3,4]. MA 356 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. ; https://doi.org/10.1101/2023.03.07.531489 doi: bioRxiv preprint antigen persistence may likewise be a natural phenomenon that can occur after infections, although its 357 detection may be difficult given the lack of available assays to detect very small amounts of lipid 358 antigen. T cells may, in fact, be more sensitive than any commercially available tool. While we 359 determined that MA persists within AMs following pulmonary vaccination, this localization is likely 360 dependent on the vaccination route. In the case of SC and IV routes, significant exposure of AMs to 361 MA-BCN is unlikely to occur. Consequently, the exact localization of MA after vaccination via SC, 362 IV, or attenuated Mtb remains to be determined, and it is plausible that other types of macrophages 363 may be involved in this process. Additionally, we found that different forms of encapsulation (such as 364 BCN, MC, attenuate Mtb) can induce antigen persistence, DC internalization alone was not sufficient 365 to replicate this phenomenon (Fig 5). Thus, the question still remains regarding the necessary was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. ; https://doi.org/10.1101/2023.03.07.531489 doi: bioRxiv preprint attributed to limited research or the administration of low doses of TDM. It is important to highlight 384 that while MA play a role in immune evasion, MA without additional functional groups has not been 385 shown to elicit similar toxic effects as TDM [62]. 386 The use of memory T cells in the setting of antigen persistence may be debatable, given that 387 memory response has historically referred to the adaptive immune response in the setting of antigen 388 clearance [72]. However, given that the persistence of MA occurs even in the setting of attenuated Mtb 389 vaccination, Mtb lipid-specific long-term adaptive immunity will likely naturally exist in an antigen 390 persistent environment. We found that DN1 T cells in this setting expressed key markers of Tfh cells 391 such as CXCR5, PD-1, BCL6, and TCF1 (Fig 7). T cells differentiate into Tfh cells through the  was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. ; https://doi.org/10.1101/2023.03.07.531489 doi: bioRxiv preprint  Subcutaneous vaccination was administered between the shoulders over the neck portion of the mouse.

518
. CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023.    was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023.    was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023.  . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. ;https://doi.org/10.1101https://doi.org/10. /2023   was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. ;https://doi.org/10.1101https://doi.org/10. /2023  . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. ;https://doi.org/10.1101https://doi.org/10. /2023  was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. ;https://doi.org/10.1101https://doi.org/10. /2023  was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023.    was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. ; https://doi.org/10.1101/2023.03.07.531489 doi: bioRxiv preprint  was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made

Supplementary Figure Legends
The copyright holder for this preprint (which this version posted July 13, 2023. ; https://doi.org/10.1101/2023.03.07.531489 doi: bioRxiv preprint (CD11c + SiglecF + ), DCs (CD11c + ), monocytes (CD11b + CD11c -), B cells (CD19 + ), T cells (CD3 + ), NK 28 cells (NK1.1 + ), and eosinophils (CD11c -SiglecF + ).  . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023.  was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023.  was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023.  was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023.  was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted July 13, 2023. ; https://doi.org/10.1101/2023.03.07.531489 doi: bioRxiv preprint