Add files using upload-large-folder tool

.gitattributes

@@ -150,3 +150,13 @@ fdf8a5d0779e58f715b3d1ec7a489af0ba19054e70d2d3212bfce606477b4049/preprint/images
 fdf8a5d0779e58f715b3d1ec7a489af0ba19054e70d2d3212bfce606477b4049/preprint/images/Figure_1.tif filter=lfs diff=lfs merge=lfs -text
 fdf8a5d0779e58f715b3d1ec7a489af0ba19054e70d2d3212bfce606477b4049/preprint/images/Figure_2.tif filter=lfs diff=lfs merge=lfs -text
 fdf8a5d0779e58f715b3d1ec7a489af0ba19054e70d2d3212bfce606477b4049/preprint/images/Figure_4.tif filter=lfs diff=lfs merge=lfs -text
+64c307dfebfa7db58f677f77308f404e8ab686d03b3d6be8689c328cc2b34807/preprint/images/Figure_1.tif filter=lfs diff=lfs merge=lfs -text
+64c307dfebfa7db58f677f77308f404e8ab686d03b3d6be8689c328cc2b34807/preprint/images/Figure_2.tif filter=lfs diff=lfs merge=lfs -text
+64c307dfebfa7db58f677f77308f404e8ab686d03b3d6be8689c328cc2b34807/preprint/images/Figure_3.tif filter=lfs diff=lfs merge=lfs -text
+e083fae6e5e22d3e2a4ae06dbcb295df3700e37243dca63ca11dacfdb074619a/preprint/images/Figure_5.tif filter=lfs diff=lfs merge=lfs -text
+e083fae6e5e22d3e2a4ae06dbcb295df3700e37243dca63ca11dacfdb074619a/preprint/images/Figure_2.tif filter=lfs diff=lfs merge=lfs -text
+e083fae6e5e22d3e2a4ae06dbcb295df3700e37243dca63ca11dacfdb074619a/preprint/images/Figure_1.tif filter=lfs diff=lfs merge=lfs -text
+e083fae6e5e22d3e2a4ae06dbcb295df3700e37243dca63ca11dacfdb074619a/preprint/images/Figure_4.tif filter=lfs diff=lfs merge=lfs -text
+e083fae6e5e22d3e2a4ae06dbcb295df3700e37243dca63ca11dacfdb074619a/preprint/images/Figure_3.tif filter=lfs diff=lfs merge=lfs -text
+e083fae6e5e22d3e2a4ae06dbcb295df3700e37243dca63ca11dacfdb074619a/preprint/images/Figure_6.tif filter=lfs diff=lfs merge=lfs -text
+ba3e5f15b7ab5215c8a0c3ad587f8294fe2d83ec05f39775525b7ce261c0cc99/preprint/images/Figure_5.tif filter=lfs diff=lfs merge=lfs -text

0132c0db112cd9ccf14cef4579c644ba3c6604295d29eb9fdfaf2eb10c346ad1/preprint/images_list.json

@@ -0,0 +1,50 @@
+[
+    {
+        "type": "image",
+        "img_path": "images/Figure_1.png",
+        "caption": "Overview of the technology scenarios and climate policy targets that our analysis spans. Colored bars show the direct aviation emissions under different technology pathways, while gray bars show emissions that are either directly mitigated or offset through DACCS.",
+        "footnote": [],
+        "bbox": [],
+        "page_idx": -1
+    },
+    {
+        "type": "image",
+        "img_path": "images/Figure_2.png",
+        "caption": "Cost to achieve CO2 and climate neutrality in the year 2050 under a scenario where synthetic fuels replace 100% of kerosene by 2050 (\"DACCU\") and under a scenario where fossil kerosene is used continuously, and emissions are offset through DACCS (\"DACCS\").",
+        "footnote": [],
+        "bbox": [],
+        "page_idx": -1
+    },
+    {
+        "type": "image",
+        "img_path": "images/Figure_3.png",
+        "caption": "Cost of achieving CO2 and climate neutrality by 2050 a) divided by abated emissions and b) divided by the installed units of DAC. Costs are shown for a scenario where synthetic fuels replace 100% of kerosene by 2050 (\"DACCU\") and for a scenario where fossil kerosene continues to be used and emissions are offset by DACCS (\"DACCS\").",
+        "footnote": [],
+        "bbox": [],
+        "page_idx": -1
+    },
+    {
+        "type": "image",
+        "img_path": "images/Figure_4.png",
+        "caption": "a) Total costs per flight per passenger and b) change in cost per flight per passenger relative to business as usual to achieve either CO2 or climate neutrality in 2050 for representative short-, medium-, and long-haul flights under the DACCU and DACCS pathways.",
+        "footnote": [],
+        "bbox": [],
+        "page_idx": -1
+    },
+    {
+        "type": "image",
+        "img_path": "images/Figure_5.png",
+        "caption": "Impacts of the local variation in important input parameters on the 2050 difference between a) DACCU and DACCS pathway to achieve CO2 neutrality and b) DACCU and fossil jet fuels assuming a 100% dominance of the respective fuel types by 2050. Blue cells indicate when DACCU becomes cheaper than fossil kerosene, while red cells are the opposite.",
+        "footnote": [],
+        "bbox": [],
+        "page_idx": -1
+    },
+    {
+        "type": "image",
+        "img_path": "images/Figure_6.png",
+        "caption": "Impact of varying assumptions on different policies on the difference in cost by 2050 of a) DACCS and DACCU pathways to reach CO2 neutrality and b) DACCU and fossil jet fuels assuming a 100% dominance of each fuel type by 2050. The row with 0% represents the standard assumption about how the policy is implemented, namely a price on CO2 emissions by 100\u20ac/tCO2, a price on aviation climate impacts by 100 \u20ac/tCO2eq, a subsidy to DACCS by 100\u20ac/tCO2, a subsidy to DACCU by 33 \u20ac/t synthetic fuel, or a restricted use of excess electricity of a price by 0.003\u20ac/kWh. The other rows represent variation of this input assumptions on the policy value (e.g., by -70% the price on CO2 emissions will be 30\u20ac/tCO2, while by +400% it will be 500\u20ac/tCO2).",
+        "footnote": [],
+        "bbox": [],
+        "page_idx": -1
+    }
+]

0132c0db112cd9ccf14cef4579c644ba3c6604295d29eb9fdfaf2eb10c346ad1/preprint/preprint.md

@@ -0,0 +1,239 @@
+# Abstract
+
+Mitigating the impact of aviation on climate change faces significant challenges due to growing demand and limited scalable mitigation options. As a result, direct air capture (DAC), a novel technology, may gain prominence due to its versatile applications as either an emissions offset (DACCS) or a synthetic fuel production technology (DACCU). Through a comprehensive analysis of cost-effectiveness, life-cycle emissions, energy consumption, and technology scale-up, we explore the conditions under which synthetic fuels from DACCU can become competitive with an emit-and-offset strategy. We find that DACCU is competitive with an emit-and-offset strategy once we explicitly include non-CO₂ climate impacts and under favorable conditions such as low electricity and high fossil fuel prices and emissions pricing. By highlighting strategic interventions that favor these conditions and thus enhance the competitiveness of DACCU in the aviation sector, our results provide valuable insights into how policymakers could move the aviation sector away from fossil fuels.
+
+Earth and environmental sciences/Environmental social sciences/Climate-change mitigation  
+Earth and environmental sciences/Environmental sciences/Environmental impact  
+Earth and environmental sciences/Environmental social sciences/Energy and society/Energy economics
+
+# Introduction
+
+Aviation has historically contributed to approximately 4% of anthropogenic climate warming<sup>1</sup>. About two-thirds of this is attributed to non-CO<sub>2</sub> effects, such as contrail cirrus cloud formation or indirect effects due to nitrous oxide emissions<sup>2–5</sup>. While aviation’s historical contribution to climate change may appear small, its role in the future could be significant due to the expected growth of the sector and the challenges of mitigating its emissions<sup>6–9,9–11</sup>. The effects of viable decarbonization options, such as operational improvements and efficiency gains, are currently jeopardized by rising demand<sup>12–14</sup>, and the switch to biofuels is constrained by biophysical limits, such as the availability of sustainable biomass, which is also in demand for other mitigation purposes<sup>15–17</sup>. While some mitigation technologies, such as hydrogen and electric aircraft, could theoretically curb all emissions, they are not yet technically feasible, especially for long-haul flights, and would require a complete renewal of the global aviation fleet<sup>18–21</sup>.
+
+This led to the emergence of two additional mitigation strategies: offsetting aviation emissions with carbon removals<sup>22–26</sup> and deploying renewable Fischer-Tropsch synthetic fuels from air-captured CO<sub>2</sub> and green hydrogen<sup>12,27–29</sup>. To ensure scalability, both solutions could rely on direct air capture (DAC), as this technology has relatively small land and water footprints and does not require biomass<sup>17,30–33</sup>. DAC can be used either in combination with CO<sub>2</sub> storage to offset aviation emissions (as direct air carbon capture and storage [DACCS]) or to produce synthetic fuels via Fischer-Tropsch synthesis (as direct air carbon capture and utilization [DACCU]). In addition to its potential for scalability, especially if deployed in remote areas<sup>31,34</sup>, the use of DAC to tackle aviation’s climate impacts could benefit climate mitigation in a larger sense; bearing the high initial costs of this technology can be seen as an equitable strategy<sup>35</sup> to overcome the steepest segment of its learning curve<sup>36–39</sup> and realize its economic viability for other applications. Financing improvements in DAC via increases in ticket prices would indeed fall most heavily on middle-to-high income consumers and households<sup>40,41</sup> but provide long-term benefits for the entire world by making the technology ready for large-scale carbon removal<sup>37,38,42</sup>, which will be necessary to remedy overshoots of a Paris-aligned carbon budget<sup>43,44</sup>.
+
+On this background, we explore the use of DAC for medium-term mitigation of the aviation sector’s climate impacts and investigate the conditions under which the use of DACCU-based synthetic fuels could be cost-effective than offsets via DACCS. Previous techno-economic assessments have concluded that DACCS is a more cost-effective option for achieving CO<sub>2</sub>-neutral aviation globally<sup>22,45</sup>. However, they also noted that these cost benefits may not materialize because they are based on uncertain assumptions<sup>45</sup> and that DACCS offers fewer co-benefits, such as potential mitigation of non-CO<sub>2</sub> impacts<sup>2,46,47</sup> and alignment with fossil fuel phase-outs<sup>45</sup>. The only study that compared the deployment of DACCS and DACCU to achieve climate neutrality concluded that it is unrealistic to rely entirely on DACCU-based fuels for European aviation fuel consumption if green hydrogen production is to take place only in Europe<sup>26</sup>.
+
+In this study, we aim to broaden the discussion by offering a global perspective on DAC deployment to achieve CO<sub>2</sub> and climate neutrality in aviation. The global focus is justified by emerging trends in countries such as Chile, Saudi Arabia, Australia, and Morocco, which are positioning themselves as producers of cheap renewable energy and exporters of green hydrogen thanks to their abundant land and renewable energy resources<sup>48,49</sup>. In addition, recognizing the imperative to emancipate aviation from fossil entanglements<sup>50</sup> and societal preferences for DACCU over DACCS<sup>51</sup> and, more generally, for direct emissions reductions over the offsets<sup>52–54</sup>, we set out to identify the conditions under which DACCU can become cost-competitive with DACCS and even with fossil fuels. By examining the drivers of future costs and policy implications, we present a comprehensive analysis that contributes to the knowledge base and provides decision-makers with actionable insights to enable DACCU to take off.
+
+# Results
+
+## Scenarios and framework
+
+Our study examines two key technology scenarios for achieving CO₂ and climate neutrality in the global aviation sector by 2050. In the DACCU scenario, synthetic fuels produced from green hydrogen and CO₂ captured by DAC lead to a gradual substitution of fossil fuels, eventually replacing conventional jet fuels entirely by 2050. This substitution follows an S-shaped curve, according to technology diffusion theories⁵⁵–⁵⁹. Conversely, the DACCS scenario focuses on the incremental DACCS-based offsetting of continued fossil jet fuel use. To ensure comparability, the share of emissions offsets follows the same S-shaped curve of DACCU deployment, reaching 100% by 2050.
+
+Our analysis includes two different 2050 goals for the aviation sector. The first is to achieve CO₂ neutrality, that is, to reduce CO₂ emissions to net-zero by 2050. In the DACCS pathway, this means offsetting CO₂ emissions only. In the DACCU pathway, fuel substitution is assumed to fully eliminate CO₂ emissions (except for indirect emissions, cf. Methods). Since DACCU-based fuels are expected to burn cleaner⁴⁶,⁴⁷, this pathway also achieves a partial mitigation of the non-CO₂ effects. Therefore, the climate benefits of the two pathways are not equal under a CO₂ neutrality target. The second target, climate neutrality, on the other hand, includes non-CO₂ effects and thus enables a more balanced comparison of the two technology pathways. In fact, to achieve climate neutrality both pathways must neutralize any residual non-CO₂ effect with the deployment of DACCS. A schematic of how the different pathways and a business-as-usual with fossil kerosene achieve different targets is shown in Fig. 1.
+
+Our analysis combines these different technologies and climate target scenarios while assuming rising aviation demand (cf. Methods). This comprehensive framework enables a holistic comparison of DACCU, DACCS and conventional aviation based on fossil kerosene in terms of costs, energy use, and climate impacts.
+
+## Emit-and-offset is cheaper under a CO neutrality target, but not under a climate neutrality target
+
+We first calculate the costs of the two technology pathways to achieve CO₂ and climate neutrality under our standard input assumptions (see Methods and Supplementary Tables 1–3). For CO₂ neutrality, the DACCS pathway is significantly less costly than the DACCU pathway, which it outperforms by about €200 billion in 2050 (Fig. 2) and €120 billion in 2060 (see Supplementary Fig. 4). This cost difference is mainly due to the high electricity and capital costs of electrolysis in the DACCU pathway, which is essential for synthetic fuel production. The cost comparison under CO₂ neutrality does not capture the full benefits of DACCU-based fuels because the reduction in non-CO₂ impacts due to cleaner synthetic fuels is not reflected in the cost (see Supplementary Fig. 1). Both the DACCS and DACCU pathways achieve substantially higher costs than a business-as-usual scenario with continued fossil jet fuels use, which is cheaper than the DACCU scenario by over €500 billion.
+
+Under climate neutrality, where the climate impacts of the two pathways are identical, the DACCU pathway has significant cost advantages over DACCS, which it outperforms by over €280 billion in 2050. The higher cost of the DACCS pathway is mainly attributable to the higher carbon removal rates required to offset non-CO₂ emissions, which are higher than in the DACCU pathway (see Supplementary Figs. 1–2). The large offset requirements are due to the sustained demand growth assumed in the analysis. However, assuming no growth of the sector still results in a competitive advantage of the DACCU pathway (see Fig. 5b). Despite its economic advantage, the DACCU pathway results in higher electricity consumption due to energy-intensive electrolysis (cf. Supplementary Fig. 3). This limits its scaling potential to regions with abundant and affordable renewable energy. Finally, both DACCS and DACCU pathways are more expensive alternatives compared to the continued use of fossil kerosene, highlighting the role of policy interventions to propel these pathways forward.
+
+## Emit-and-offset is more expensive than synthetic fuels on a cost-per-avoided-emissions basis, but is more efficient in scaling DAC
+
+Looking at the total costs for abated emissions relative to the business-as-usual (Fig. 3a), the resulting picture is almost opposite than the one drawn when looking at absolute yearly costs (Fig. 2). Under the CO₂ neutrality target, the DACCS pathway has the highest costs per emissions abated, reaching abatement costs over €500/tCO₂e compared to less than €200/tCO₂e for the DACCU pathway. This difference arises because DACCS only includes costs associated with reducing CO₂ emissions. Conversely, in the DACCU pathway, the abatement extends to non-CO₂ emissions, thereby increasing the total volume of abated emissions over which the costs are distributed. Under the climate neutrality target, where both technology pathways abate the same level of emissions, DACCU again emerges as more cost-effective because of the smaller amounts of carbon removals required to offset the remaining non-CO₂ effects.
+
+Apart from mitigating the aviation sector, both options could also serve as a means of scaling up DAC. This rationale is rooted in the potential role that the aviation sector could play as a niche for the initial deployment of DAC, as the sector is bound to face significant costs in mitigating its emissions due to the lack of affordable alternatives. This perspective results in a picture opposite to that of cost-effective abatement. We find that as the volume of DAC installations increases, the DACCS pathway consistently offers a lower cost per DAC unit than the DACCU pathway (Fig. 3b). DACCU incurs higher costs due to the production of green hydrogen. This has a significant impact on the cost per unit of DAC installed.
+
+## The price difference for a CO₂-neutral flight with DACCS and DACCU is small
+
+We further assess the increase in price per flight per passenger to achieve CO₂ and climate neutrality via the DACCS and DACCU pathways. In the context of CO₂ neutrality, offsetting aviation CO₂ emissions with DACCS proves to be more economical than fueling the same flight with DACCU-based synthetic fuels. However, the cost difference per passenger is modest, ranging from approximately €20–55 for long-haul flights (London-New York and London-Perth) to only €4 for a short-haul flight from London to Berlin. While the overall cost per passenger increases to achieve climate neutrality, DACCU becomes cheaper than DACCS, saving about €35–100 per passenger on long-haul flights and €6 on short-haul flights.
+
+We also assessed the impact on the cost of flying relative to the expected future cost of flying in a business-as-usual scenario with continued use of fossil fuels. The projected increase in ticket prices for flights in 2050 ranges between 15–30% for DACCU and 8–20% for DACCS to achieve CO₂ neutrality, rising to up to 40% (DACCU) and 60% (DACCS) to achieve climate neutrality. However, the increase in price is not the same for all flights, since the contribution of fuel costs to ticket prices varies for different routes, as the price is adjusted to demand and to endure competition. While the increases in price due to a complete neutralization of the climate effects of a flight may seem substantial, they lie well below the range of current variance in prices. Indeed, the difference in price between buying a ticket two weeks or two months in advance is, on average, 400% for the London-Berlin route, over 100% for the London-New York route, and 70% for the London-Perth route⁶⁰.
+
+## Cheaper electricity and high fossil jet fuel prices can make DACCU cheaper than DACCS (and even business-as-usual) even under CO₂ neutrality
+
+To understand the conditions under which DACCU-based fuels could be economically competitive in the less-advantageous CO₂-neutrality scenario with an emit-and-offset strategy via DACCS and even with the business-as-usual with continued use of fossil jet fuel, we perform local sensitivity analyses on the most influential parameters (see Supplementary Table 1–3).
+
+Figure 5a shows that DACCU can become more cost effective than DACCS when electricity prices fall below 0.015 €/kWh. This threshold is well below the 2023 price of the cheapest renewable energy sources, onshore wind⁶¹, but not unachievable in the future through technology learning, optimal siting, or in moments of excess production of renewable electricity, for example on sunny summer days in grids with a high share of solar PV⁶²,⁶³. In contrast, even when powered by free electricity, DACCU is still not competitive with the business-as-usual.
+
+Conversely, rising fossil fuel prices prove transformative: DACCU becomes cost-competitive with DACCS at a fossil fuel price of €0.9/L and with the business-as-usual scenario at €1.8/L. Such high costs would not only make DACCU a more economical option, but would also discourage demand. However, doubling the current price of fossil jet fuel would require dedicated political ambition.
+
+Accelerated technological learning and steeper learning curves benefit both DACCU and DACCS scenarios. Thus, even a learning rate of 50% - higher than has been observed historically for fast-learning technologies such as solar PV - cannot close the gap between the DACCU and DACCS pathways.
+
+In summary, extremely optimistic changes in fossil fuel or electricity prices are required to make DACCU cost-competitive with DACCS or business-as-usual by varying a single parameter. However, a synergy of lower electricity prices with either rising fossil fuel costs or higher technological learning could accelerate a scenario where DACCU outperforms DACCS or even fossil jet fuels under optimistic but possible conditions (see Supplementary Figs. 6–8).
+
+## Pricing aviation climate impacts or limiting DACCU operation to times when excess electricity is available is sufficient to make DACCU cheaper than DACCS
+
+Given the observed sensitivity of DACCS and DACCU performance to highly uncertain input assumptions, we examine the potential impact of different policies that affect these assumptions. Figure 6 shows the cost difference of DACCU compared to DACCS (Fig. 6a) and fossil jet fuel (6b) under different policies affecting some of the key input variables (see Supplementary Table 4).
+
+Pricing emissions internalizes the impact of continued fossil jet fuel emissions and thus acts similarly to increasing the price of fossil fuels, while also internalizing the environmental costs of life-cycle emissions for both DACCS and DACCU. Conversely, pricing CO₂ emissions alone cannot make DACCU cost-competitive with DACCS since, under CO₂ neutrality, it applies only to indirect emissions, which are higher in the DACCU pathway (see Supplementary Fig. 1). On the other hand, pricing all aviation-related climate impacts can significantly favor the DACCU pathway, which already becomes more cost-effective than the DACCS pathways at €30/tCO₂e*. Pricing emissions is also crucial to make DACCU economically competitive with fossil jet fuels. However, the prices on emissions need to be extremely high, starting at €500/tCO₂ for CO₂ emissions alone and at least €100/tCO₂e* for all aviation-related impacts.
+
+In contrast to direct subsidies based on synthetic fuel production, which are not sufficient to make DACCU competitive with DACCS even at €500/tfuel (which corresponds to about €1600/tCO₂ for DACCU-based fuels), a strategic approach is to leverage cheap electricity (below €0.01/kWh). Policies, such as seasonal restrictions aligned with periods of electricity surplus, could achieve this by limiting DACCU-based synthetic fuel production to periods of significantly cheaper surplus electricity. However, this approach comes with the constraint of limiting the volume of DACCU-based synthetic fuels that can be produced. While limiting the number of operating hours could increase the weight of capital expenditures per DACCU output, and thus lead to potential cost increases not accounted for in our modelling⁶⁴, it could also reduce the deterioration, and thus extend the lifetime, of costly components of the electrolyzers and DAC, namely the stack and adsorbent.
+
+# Discussion
+
+In this study, we investigate the conditions under which aviation mitigation via DACCU-based synthetic fuels becomes cost-competitive with an emit-and-offset strategy via DACCS. We found that these conditions are realized by either (1) ambitious climate targets for the aviation sector that consider the non-CO₂ impacts of aviation, or (2) policies that internalize the cost of unabated emissions or limit DACCU to the use of excess, cheap electricity. In addition, our analysis highlights that achieving CO₂ neutrality through DACCU increases flight ticket prices only slightly relative to the DACCS pathway and even relative to a business-as-usual pathway. This small price difference for consumers sheds light on the attractiveness of DACCU, which has a lower cost per avoided emissions and is consistent with broader societal goals of climate mitigation and fossil fuels phase-out.
+
+These findings mark a departure from previous studies<sup>22,26</sup>, which favored DACCS due to conservative assumptions about future electricity prices (which exceed current wind and solar PV prices) and carbon-intensive energy mixes, resulting in higher lifecycle emissions of DACCU. Furthermore, due to their regional focus on Europe, where land availability is scarce, Sacchi et al. concluded that the land use of the energy-intensive DACCU pathway is a bottleneck under a scenario of continued demand growth for the aviation sector. While their regional land availability constraint does not apply to our global analysis, spatial considerations may indeed affect the cost at which the DACCU pathway could be realized due to the spatial distribution of electricity costs and the potential need for additional transportation infrastructure from remote locations.
+
+The efforts needed to enable CO₂ neutral and, especially, climate neutral flying may not be feasible. In fact, more than 2 GtCO₂ of DAC would need to be installed by 2050 to achieve CO₂ neutrality, rising to 7 GtCO₂ if the goal is to offset fossil jet fuel emissions to achieve climate neutrality. These amounts of DAC far exceed the projections of novel CDR methods by 2050 in Integrated Assessment Models simulations consistent with < 2°C targets<sup>31,42,43,65</sup>. However, the assumed growth rate up to 2050 (roughly 50 to 60% annually) is in line with that assumed by Integrated Assessment Models for the years between 2040–2080<sup>31</sup> and with that observed historically for solar PV<sup>66</sup>. On the other hand, by 2050, the DACCU pathway will require over 15 PWh of electricity to produce the amount of synthetic fuels necessary to fully meet global aviation demand if this continues to grow. Given that in 2021 the global renewable energy produced amounted to 8 PWh<sup>67</sup>, this energy demand would require a massive scale-up of renewable energy. However, DACCU’s renewable energy requirements are compatible with estimates of the total technical renewable energy potential (170–270 PWh according to Angliviel de La Beaumelle et al., 2023).
+
+The superiority of DACCU in our results also hinges on uncertain variables, particularly the effectiveness of DACCU-based synthetic fuels in mitigating non-CO₂ impacts. While early empirical evidence is consistent with this trend<sup>46,47,69</sup>, the limited number of studies evaluating the impacts of synthetic fuels, coupled with the inherent uncertainty surrounding aviation’s non-CO₂ effects, introduces a degree of uncertainty. Notably, our analysis explicitly accounts for these uncertainties, and while they could significantly alter the absolute costs of DACCS and DACCU, their relative merit mostly remains unchanged.
+
+By shedding light on the conditions that make DACCU cost-competitive, our analysis can guide policymakers in designing strategies to facilitate the competitiveness of DACCU with both a emit-and-offset pathway relying on DACCS and a business-as-usual scenario. These strategic policy interventions could be justified based on the drawbacks of the DACCS pathway associated with its reliance on fossil jet fuels and the climate mitigation benefits of DACCU fuels.
+
+# Methods
+
+In this study, we combined techno-economic modelling with life cycle assessment to compare the costs of mitigating the aviation sector by either compensating aviation emissions with DACCS or by replacing the whole volume of jet fuel with DACCU-based synthetic fuels, as shown in Fig. 7.
+
+## Demand and fuel scenarios
+
+All scenarios are based on the same demand for jet fuel, which is derived from a combination of historical data<sup>5</sup> from 1990 to 2018 with estimates of future demand until 2060. These are based on the assumptions of full recovery to pre-covid levels by 2024–2025 and on a 2% growth from 2024 to 2060, which are consistent with projections from various studies<sup>10,70–75</sup>. In addition to the total fuel demand, we also project the total annual distance flown by applying a 2% increase in efficiency, consistent with the International Civil Aviation Organization’s target<sup>76</sup>, to the historical relationship between distance flown and amount of fuel burned<sup>5</sup>. While this relationship may change in the future due to an increase in long-haul flights<sup>9,77</sup> that burn more fuel per kilometer<sup>78</sup>, its effect would not significantly alter the results of our analysis, as shown in our sensitivity analysis (see Fig. 5a and Supplementary Fig. 9).
+
+As detailed in the “Scenarios and Framework” section, we consider two different mitigation pathways for aviation, one based on continued reliance on fossil jet fuel and offsetting through DACCS, and the other based on the gradual substitution of fossil fuels with DACCU-based synthetic fuels. Although the American Society for Testing Material D7566 standard<sup>79</sup> currently allows only up to 50% synthetic fuels blends, we assume that aircraft will operate on 100% DACCU-based synthetic fuels by 2050, being expected that blends up to 100% will be certified in due course, so that planes can fully run on synthetic fuel. Similarly, we model an upscaling of DACCS that enables full offsetting of aviation emissions by 2050, simplistically assuming no constraints on the rate of adoption of this technology.
+
+## Emissions and offsets
+
+To calculate the amount of direct emissions from fossil jet fuel combustion, we apply the relationships between fossil jet fuel and CO<sub>2</sub>, water vapor, sulfur dioxide, soot, and NO<sub>x</sub> emissions reported by Lee et al.<sup>5</sup>. Contrail cirrus formation was calculated using the relationship between the distance flown and contrail length, also reported in Lee et al.<sup>5</sup>. To calculate the emissions and contrail clouds formation of DACCU-based fuels, we follow the approach described in Brazzola et al.<sup>25</sup> (see their Supplementary Table 2) and propagate their uncertainty ranges throughout the analysis.
+
+The direct flight emissions then drive the demand for carbon removals via DACCS to offset their climate impact. The amount of removals is further determined by (1) the specific climate target chosen (i.e., CO<sub>2</sub> or climate neutrality, see Fig. 1), and (2) the lifecycle emissions of each technology pathway. First, to achieve CO<sub>2</sub> neutrality, we simplistically assume that we can fully compensate the climate impact of one ton of CO<sub>2</sub> by removing an equivalent amount via DACCS, neglecting the uncertainties of this relationship<sup>80,81</sup>. To achieve climate neutrality, we compensate for the non-CO<sub>2</sub> effects with DACCS based on the GWP* metric following the approach of Brazzola et al.<sup>25</sup> and using their ‘Gold’ definition of climate neutrality<sup>25</sup>. Thereby, we use and propagate throughout the analysis the uncertainties in the relationship between non-CO<sub>2</sub> emissions and their effective radiative forcing reported in Lee et al.<sup>5</sup>.
+
+Finally, we also offset through DACCS the lifecycle emissions due to the material and energy footprint of the two pathways. We calculate lifecycle emissions for both fossil fuels, DACCU-based fuels and DACCS. For fossil jet fuels, we considered the well-to-tank emissions from Moretti et al.<sup>82</sup>, which reflect European averages. Future reductions in oil refining emissions are based on the oil industry decarbonization prospects<sup>83</sup>, leading to a progressive decrease in well-to-tank emissions for fossil jet fuels. Material footprints are based on values for the production of required adsorbents and DAC modules by Deutz and Bardow<sup>84</sup>; values for electrolysers by Delpierre et al.<sup>85</sup>; values for CO electrolysis production units from Adnad and Kibria<sup>86</sup>. In addition, we calculated the energy requirements of all technologies involved and applied an electricity carbon footprint for an average global electricity grid<sup>84</sup>, assuming high decarbonization efforts over time leading to near net-zero emissions in 2060. As the synthesis of DACCU-based fuels is a multi-functional unit process with by-products, notably diesel, we assume the production of 0.82 tons of diesel per ton of jet fuel<sup>22</sup>. The lifecycle inventory of the unit processes up to the Fischer-Tropsch unit was then allocated to jet fuel by means of mass allocation (resulting in a 54.5% share for jet fuel).
+
+## Techno-economic assessment of DACCS and DACCU pathways
+
+Finally, we calculate the energy consumption and capital costs of each technology and fuel included in the DACCS and DACCU pathways from 2020 to 2060. This includes the cost of fossil jet fuel, electricity and heat consumption, CO<sub>2</sub> transport and storage, and the capital costs of DAC, CO<sub>2</sub> reduction, electrolysis, and Fischer-Tropsch synthesis.
+
+For both pathways, we consider a low-heat solid-sorbent DAC system. While high-temperature liquid-solvent DAC may be more energy-efficient for the production of DACCU-based fuels, there are currently no plants that operate completely without burning natural gas<sup>87</sup>. As a result, using liquid-solvent DAC to produce jet fuel may result in net CO<sub>2</sub> emissions. We moreover assume a fixed cost of 20 €/tCO<sub>2</sub> for CO<sub>2</sub> transport and storage as in Becattini et al.<sup>22</sup>, based on the assumption that DACCS would be optimally located next to storage sites.
+
+For the production of DACCU-based synthetic fuels, we introduce some variance by considering four different combinations of two water electrolysers (either polymer membrane or alkaline electrolysers) and two CO<sub>2</sub> reduction methods (electrochemical CO<sub>2</sub> reduction and reverse-water-gas-shift). While we also calculate total costs for each technology configuration (cf. Supplementary Fig. 9), in the main results we use an average of the costs of all four possible configurations since we cannot predict which technology will ultimately prevail more established due to the low technological maturity, uncertain future development, and trade-offs in terms of cost and energy intensity of different technologies involved in DACCU-based synthetic fuel production.
+
+We first derive the installed capacities of each technology from the amounts of synthetic jet fuel required and from calculations of DACCS-based offset, as explained in the previous sections. To calculate their costs and energy consumption, we apply the parameters and assumptions summarized in Supplementary Tables 1–3. To calculate changes in energy efficiency, we polynomially interpolate between current values and future estimates (see Supplementary Table 2). In the case of CAPEX, we apply a learning rate following Eq. 1:
+
+$$CAPEX\left(t\right)=CAPEX\left({t}_{0}\right)*{\left(\frac{{Q}_{t}}{{Q}_{{t}_{0}}} \right)}^{-b}$$
+
+## (1)
+
+Where *Q* is the quantity of installed capacity of a technology and *b* equals ${log}_{2}\left(1-LR\right)$, and *LR* is the learning rate.
+
+To calculate the increase in ticket price per passenger for three representative flights, we first calculate the cost of achieving CO<sub>2</sub> or carbon neutrality per kilometer flown each year. We then assume that the non-fuel cost of tickets will remain constant in the future, while replacing the cost of fuel with the cost of achieving either CO<sub>2</sub> or climate neutrality, including the cost of DACCS and DACCU. The relevant parameters for these calculations are shown in Supplementary Table 4.
+
+Finally, we conduct a local sensitivity analysis on key parameters highlighted in Supplementary Table 1–3, including those associated with alternative policy scenarios outlined in Supplementary Table 5. To ensure comparability, we systematically vary uncertain input parameters by fixed percentages, which are right-skewed towards increases in the input assumption to accommodate the constraint that many parameters cannot be reduced below −100% due to the impracticality of negative values.
+
+# References
+
+1. Klöwer, M. et al. Quantifying aviation’s contribution to global warming. *Environ. Res. Lett.* **16**, 104027 (2021).
+2. Burkhardt, U., Bock, L. & Bier, A. Mitigating the contrail cirrus climate impact by reducing aircraft soot number emissions. *Npj Clim. Atmospheric Sci.* **1**, 1–7 (2018).
+3. Kärcher, B. Formation and radiative forcing of contrail cirrus. *Nat. Commun.* **9**, 1824 (2018).
+4. Lee, D. S. et al. Aviation and global climate change in the 21st century. *Atmos. Environ.* **43**, 3520–3537 (2009).
+5. Lee, D. S. et al. The contribution of global aviation to anthropogenic climate forcing for 2000 to 2018. *Atmos. Environ.* **244**, 117834 (2021).
+6. Airbus. *Global Market Forecast 2018-2037, Global Networks, Global Citizens*. (2018).
+7. Boeing. Commercial Market Outlook 2021-2040. https://www.boeing.com/commercial/market/commercial-market-outlook/ (2021).
+8. Dray, L. et al. Cost and emissions pathways towards net-zero climate impacts in aviation. *Nat. Clim. Change* **12**, 956–962 (2022).
+9. Gössling, S. & Humpe, A. The global scale, distribution and growth of aviation: Implications for climate change. *Glob. Environ. Change* **65**, 102194 (2020).
+10. Grewe, V. et al. Evaluating the climate impact of aviation emission scenarios towards the Paris agreement including COVID-19 effects. *Nat. Commun.* **12**, 3841 (2021).
+11. Terrenoire, E., Hauglustaine, D. A., Gasser, T. & Penanhoat, O. The contribution of carbon dioxide emissions from the aviation sector to future climate change. *Environ. Res. Lett.* **14**, 084019 (2019).
+12. Gössling, S. & Humpe, A. Net-zero aviation: Time for a new business model? *J. Air Transp. Manag.* **107**, 102353 (2023).
+13. Larsson, J., Elofsson, A., Sterner, T. & Åkerman, J. International and national climate policies for aviation: a review. *Clim. Policy* **19**, 787–799 (2019).
+14. Scheelhaase, J., Maertens, S., Grimme, W. & Jung, M. EU ETS versus CORSIA–A critical assessment of two approaches to limit air transport’s CO2 emissions by market-based measures. *J. Air Transp. Manag.* **67**, 55–62 (2018).
+15. Committee on Climate Chnage. *Biomass in a Low-Carbon Economy*. https://www.theccc.org.uk/publication/biomass-in-a-low-carbon-economy/ (2018).
+16. Dooley, K., Christoff, P. & Nicholas, K. A. Co-producing climate policy and negative emissions: trade-offs for sustainable land-use. *Glob. Sustain.* **1**, (2018).
+17. Smith, P. et al. Biophysical and economic limits to negative CO2 emissions. *Nat. Clim. Change* **6**, 42–50 (2016).
+18. Gnadt, A. R., Speth, R. L., Sabnis, J. S. & Barrett, S. R. H. Technical and environmental assessment of all-electric 180-passenger commercial aircraft. *Prog. Aerosp. Sci.* **105**, 1–30 (2019).
+19. Noland, J. K. Hydrogen Electric Airplanes: A Disruptive Technological Path to Clean Up the Aviation Sector. *IEEE Electrification Mag.* **9**, 92–102 (2021).
+20. Peeters, P., Higham, J., Kutzner, D., Cohen, S. & Gössling, S. Are technology myths stalling aviation climate policy? *Transp. Res. Part Transp. Environ.* **44**, 30–42 (2016).
+21. Schäfer, A. W. et al. Technological, economic and environmental prospects of all-electric aircraft. *Nat. Energy* **4**, 160–166 (2019).
+22. Becattini, V., Gabrielli, P. & Mazzotti, M. Role of Carbon Capture, Storage, and Utilization to Enable a Net-Zero-CO2-Emissions Aviation Sector. *Ind. Eng. Chem. Res.* **60**, 6848–6862 (2021).
+23. Becken, S. & Mackey, B. What role for offsetting aviation greenhouse gas emissions in a deep-cut carbon world? *J. Air Transp. Manag.* **63**, 71–83 (2017).
+24. Bergero, C. et al. Pathways to net-zero emissions from aviation. *Nat. Sustain.* **6**, 404–414 (2023).
+25. Brazzola, N., Patt, A. & Wohland, J. Definitions and implications of climate-neutral aviation. *Nat. Clim. Change* **12**, 761–767 (2022).
+26. Sacchi, R. et al. How to make climate-neutral aviation fly. *Nat. Commun.* **14**, 3989 (2023).
+27. Scheelhaase, J., Maertens, S. & Grimme, W. Synthetic fuels in aviation – Current barriers and potential political measures. *Transp. Res. Procedia* **43**, 21–30 (2019).
+28. Terwel, R. & Kerkhoven, J. *Carbon Neutral Aviation with Current Enginge Technology: The Take-off of Synthetic Kerosene Production in the Netherlands*. 62 . https://kalavasta.com/assets/reports/Kalavasta_Carbon_Neutral_Aviation.pdf (2018).
+29. Timmons, D. & Terwel, R. Economics of aviation fuel decarbonization: A preliminary assessment. *J. Clean. Prod.* **369**, 133097 (2022).
+30. Fuhrman, J. et al. Food–energy–water implications of negative emissions technologies in a +1.5 °C future. *Nat. Clim. Change* **10**, 920–927 (2020).
+31. Fuhrman, J. et al. The role of direct air capture and negative emissions technologies in the shared socioeconomic pathways towards $\\mathplus$1.5 °C and $\\mathplus$2 °C futures. *Environ. Res. Lett.* **16**, 114012 (2021).
+32. Terlouw, T., Treyer, K., Bauer, C. & Mazzotti, M. Life Cycle Assessment of Direct Air Carbon Capture and Storage with Low-Carbon Energy Sources. *Environ. Sci. Technol.* **55**, 11397–11411 (2021).
+33. The Royal Society. *Net Zero Aviation Fuels: Resource Requirements and Environmental Impacts*. (London, UK, 2023).
+34. Ali, M. et al. Recent advances in carbon dioxide geological storage, experimental procedures, influencing parameters, and future outlook. *Earth-Sci. Rev.* **225**, 103895 (2022).
+35. Andreoni, P., Emmerling, J. & Tavoni, M. Inequality repercussions of financing negative emissions. *Nat. Clim. Change* 1–7 (2023) doi:10.1038/s41558-023-01870-7.
+36. Küng, L. et al. A roadmap for achieving scalable, safe, and low-cost direct air carbon capture and storage. *Energy Environ. Sci.* (2023) doi:10.1039/D3EE01008B.
+37. Meckling, J. & Biber, E. A policy roadmap for negative emissions using direct air capture. *Nat. Commun.* **12**, 2051 (2021).
+38. Nemet, G. F. *How Solar Energy Became Cheap: A Model for Low-Carbon Innovation*. (Routledge, 2019).
+39. Young, J. et al. The cost of direct air capture and storage can be reduced via strategic deployment but is unlikely to fall below stated cost targets. *One Earth* **0**, (2023).
+40. Owen, A., Burke, J. & Serin, E. Who pays for BECCS and DACCS in the UK: designing equitable climate policy. *Clim. Policy* **22**, 1050–1068 (2022).
+41. Addepalli, S., Pagalday, G., Salonitis, K. & Roy, R. Socio-economic and demographic factors that contribute to the growth of the civil aviation industry. *Procedia Manuf.* **19**, 2–9 (2018).
+42. Nemet, G. F. et al. Near-term deployment of novel carbon removal to facilitate longer-term deployment. *Joule* **0**, (2023).
+43. IPCC. *Global Warming of 1.5°C. An IPCC Special Report on the Impacts of Global Warming of 1.5°C above Pre-Industrial Levels and Related Global Greenhouse Gas Emission Pathways, in the Context of Strengthening the Global Response to the Threat of Climate Change, Sustainable Development, and Efforts to Eradicate Poverty*. https://www.ipcc.ch/site/assets/uploads/sites/2/2019/06/SR15_Full_Report_High_Res.pdf (2018).
+44. IPCC. Summary for Policymakers. in *Climate Change 2022: Mitigation of Climate Change. Contribution of Working Group III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change* (Cambridge University Press, Cambridge, UK and New York, NY, USA, 2022).
+45. Cames, M., Chaudry, S., Göckeler, K., Kasten, P. & Kurth, S. *E-Fuels versus DACCS*. https://www.transportenvironment.org/wp-content/uploads/2021/08/2021_08_TE_study_efuels_DACCS.pdf (2021).
+46. Braun-Unkhoff, M., Riedel, U. & Wahl, C. About the emissions of alternative jet fuels. *CEAS Aeronaut. J.* **8**, 167–180 (2017).
+47. Voigt, C. et al. Cleaner burning aviation fuels can reduce contrail cloudiness. *Commun. Earth Environ.* **2**, 1–10 (2021).
+48. PwC. Green hydrogen economy - predicted development of tomorrow. *PwC* https://www.pwc.com/gx/en/industries/energy-utilities-resources/future-energy/green-hydrogen-cost.html (2023).
+49. World Energy Council. *World Energy Insights: Executive Summary Regional Insights Into Low-Carbon Hydrogen Scale-Up*. https://www.worldenergy.org/assets/downloads/World_Energy_Insights_Executive_Summary_Regional_insights_into_low-carbon_hydrogen_scale_up_April_2022.pdf?v=1680701563 (2022).
+50. Malm, A. & Carton, W. Seize the Means of Carbon Removal: The Political Economy of Direct Air Capture. *Hist. Mater.* **29**, 3–48 (2021).
+51. Arning, K. et al. Same or different? Insights on public perception and acceptance of carbon capture and storage or utilization in Germany. *Energy Policy* **125**, 235–249 (2019).
+52. Arning, K., Linzenich, A., Engelmann, L. & Ziefle, M. More green or less black? How benefit perceptions of CO2 reductions vs. fossil resource savings shape the acceptance of CO2-based fuels and their conversion technology. *Energy Clim. Change* **2**, 100025 (2021).
+53. Markusson, N., McLaren, D. & Tyfield, D. Towards a cultural political economy of mitigation deterrence by negative emissions technologies (NETs). *Glob. Sustain.* **1**, e10 (2018).
+54. Satterfield, T., Nawaz, S. & St-Laurent, G. P. Exploring public acceptability of direct air carbon capture with storage: climate urgency, moral hazards and perceptions of the ‘whole versus the parts’. *Clim. Change* **176**, 14 (2023).
+55. Geels, F. W. Technological transitions as evolutionary reconfiguration processes: a multi-level perspective and a case-study. *Res. Policy* **31**, 1257–1274 (2002).
+56. Kemp, R., Schot, J. & Hoogma, R. Regime shifts to sustainability through processes of niche formation: The approach of strategic niche management. *Technol. Anal. Strateg. Manag.* **10**, 175–198 (1998).
+57. Kemp, R. & Volpi, M. The diffusion of clean technologies: a review with suggestions for future diffusion analysis. *J. Clean. Prod.* **16**, S14–S21 (2008).
+58. Roberts, C. et al. The politics of accelerating low-carbon transitions: Towards a new research agenda. *Energy Res. Soc. Sci.* **44**, 304–311 (2018).
+59. Rogers, E. M. Diffusion of Innovations. *J. Pharm. Sci.* **52**, 612 (1963).
+60. Kayak. Search Flights, Hotels & Rental Cars | KAYAK. https://www.kayak.com/ (2023).
+61. IRENA. Renewable Power Generation Costs in 2022. https://www.irena.org/Publications/2023/Aug/Renewable-Power-Generation-Costs-in-2022 (2023).
+62. Kost, C. et al. *Levelized Cost of Electricity - Renewable Energy Technologies*. https://www.ise.fraunhofer.de/en/publications/studies/cost-of-electricity.html (2021).
+63. Grahn, M. et al. Review of electrofuel feasibility—cost and environmental impact. *Prog. Energy* **4**, 032010 (2022).
+64. IRENA. *Green Hydrogen Cost Reduction: Scaling up Electrolysers to Meet the 1.5C Climate Goal*. (2020).
+65. Fuhrman, J. et al. Diverse carbon dioxide removal approaches could reduce impacts on the energy–water–land system. *Nat. Clim. Change* 1–10 (2023) doi:10.1038/s41558-023-01604-9.
+66. Odenweller, A., Ueckerdt, F., Nemet, G. F., Jensterle, M. & Luderer, G. Probabilistic feasibility space of scaling up green hydrogen supply. *Nat. Energy* **7**, 854–865 (2022).
+67. IEA. *Renewable Electricity – Analysis*. https://www.iea.org/reports/renewable-electricity (2022).
+68. Angliviel de La Beaumelle, N. et al. The Global Technical, Economic, and Feasible Potential of Renewable Electricity. *Annu. Rev. Environ. Resour.* **48**, 419–449 (2023).
+69. Riebl, S., Braun-Unkhoff, M. & Riedel, U. A Study on the Emissions of Alternative Aviation Fuels. *J. Eng. Gas Turbines Power* **139**, (2017).
+70. Chen, C.-C. & Gettelman, A. Simulated 2050 aviation radiative forcing from contrails and aerosols. *Atmospheric Chem. Phys.* **16**, 7317–7333 (2016).
+71. Dray, L. et al. Cost and emissions pathways towards net-zero climate impacts in aviation. *Nat. Clim. Change* **12**, 956–962 (2022).
+72. IATA. Passenger Demand Recovery Continued in 2021 but Omicron Having Impact. https://www.iata.org/en/pressroom/2022-releases/2022-01-25-02/ (2021).
+73. IEA. *World Energy Outlook 2021*. 386 https://www.iea.org/reports/world-energy-outlook-2021 (2021).
+74. Klöwer, M. et al. Quantifying aviation’s contribution to global warming. *Environ. Res. Lett.* **16**, 104027 (2021).
+75. U.S. Energy Information Administration. International Energy Outlook - U.S. Energy Information Administration (EIA). https://www.eia.gov/outlooks/ieo/tables_side_xls.php (2021).
+76. International Civil Aviation Organisation. Long term global aspirational goal (LTAG) for international aviation. https://www.icao.int/environmental-protection/Pages/LTAG.aspx (2022).
+77. Bain & Company. Air Travel Forecast to 2030: The Recovery and the Carbon Challenge. *Bain* https://www.bain.com/insights/air-travel-forecast-interactive/ (2023).
+78. Filippone, A. *Advanced Aircraft Flight Performance*. (Cambridge University Press, 2012).
+79. Anuar, A., Undavalli, V. K., Khandelwal, B. & Blakey, S. Effect of fuels, aromatics and preparation methods on seal swell. *Aeronaut. J.* **125**, 1542–1565 (2021).
+80. Zickfeld, K., Azevedo, D., Mathesius, S. & Matthews, H. D. Asymmetry in the climate–carbon cycle response to positive and negative CO2 emissions. *Nat. Clim. Change* **11**, 613–617 (2021).
+81. Zickfeld, K. et al. Net-zero approaches must consider Earth system impacts to achieve climate goals. *Nat. Clim. Change* **13**, 1298–1305 (2023).
+82. Moretti, C., Moro, A., Edwards, R., Rocco, M. V. & Colombo, E. Analysis of standard and innovative methods for allocating upstream and refinery GHG emissions to oil products. *Appl. Energy* **206**, 372–381 (2017).
+83. Griffiths, S., Sovacool, B. K., Kim, J., Bazilian, M. & Uratani, J. M. Decarbonizing the oil refining industry: A systematic review of sociotechnical systems, technological innovations, and policy options. *Energy Res. Soc. Sci.* **89**, 102542 (2022).
+84. Deutz, S. & Bardow, A. Life-cycle assessment of an industrial direct air capture process based on temperature–vacuum swing adsorption. *Nat. Energy* **6**, 203–213 (2021).
+85. Delpierre, M., Quist, J., Mertens, J., Prieur-Vernat, A. & Cucurachi, S. Assessing the environmental impacts of wind-based hydrogen production in the Netherlands using ex-ante LCA and scenarios analysis. *J. Clean. Prod.* **299**, 126866 (2021).
+86. Adnan, M. A. & Kibria, M. G. Comparative techno-economic and life-cycle assessment of power-to-methanol synthesis pathways. *Appl. Energy* **278**, 115614 (2020).
+87. McQueen, N. et al. Cost Analysis of Direct Air Capture and Sequestration Coupled to Low-Carbon Thermal Energy in the United States. *Environ. Sci. Technol.* **54**, 7542–7551 (2020).
+88. Smith, S. M. et al. *The State of Carbon Dioxide Removal - 1st Edition*. http://dx.doi.org/10.17605/OSF.IO/W3B4Z (2023) doi:10.17605/OSF.IO/W3B4Z.
+89. IRENA. *Making the Breakthrough: Green Hydrogen Policies and Technology Costs*. https://www.irena.org/-/media/Files/IRENA/Agency/Publication/2020/Nov/IRENA_Green_Hydrogen_breakthrough_2021.pdf?la=en&hash=40FA5B8AD7AB1666EECBDE30EF458C45EE5A0AA6 (2021).
+90. Christensen, A. *Assessment of Hydrogen Production Costs from Electrolysis: United States and Europe*. https://theicct.org/publication/assessment-of-hydrogen-production-costs-from-electrolysis-united-states-and-europe/ (2020).
+91. Zang, G. et al. Synthetic Methanol/Fischer–Tropsch Fuel Production Capacity, Cost, and Carbon Intensity Utilizing CO2 from Industrial and Power Plants in the United States. *Environ. Sci. Technol.* **55**, 7595–7604 (2021).
+92. Jouny, M., Luc, W. & Jiao, F. General Techno-Economic Analysis of CO2 Electrolysis Systems. *Ind. Eng. Chem. Res.* **57**, 2165–2177 (2018).
+93. Shin, H., Hansen, K. U. & Jiao, F. Techno-economic assessment of low-temperature carbon dioxide electrolysis. *Nat. Sustain.* **4**, 911–919 (2021).
+94. Moretti, C. Reflecting on the environmental impact of the captured carbon feedstock. *Sci. Total Environ.* **854**, 158694 (2023).
+95. Gabrielli, P., Gazzani, M. & Mazzotti, M. The Role of Carbon Capture and Utilization, Carbon Capture and Storage, and Biomass to Enable a Net-Zero-CO2 Emissions Chemical Industry. *Ind. Eng. Chem. Res.* **59**, 7033–7045 (2020).
+96. Schmidt, P., Weindorf, W., Roth, A., Batteiger, V. & Riegel, F. *Power-to-Liquids: Potentials and Perspectives*. (2016).
+97. Sutter, D., van der Spek, M. & Mazzotti, M. 110th Anniversary: Evaluation of CO2-Based and CO2-Free Synthetic Fuel Systems Using a Net-Zero-CO2-Emission Framework. *Ind. Eng. Chem. Res.* **58**, 19958–19972 (2019).
+98. Yates, J. et al. Techno-economic Analysis of Hydrogen Electrolysis from Off-Grid Stand-Alone Photovoltaics Incorporating Uncertainty Analysis. *Cell Rep. Phys. Sci.* **1**, 100209 (2020).
+99. Kopp, M. et al. Energiepark Mainz: Technical and economic analysis of the worldwide largest Power-to-Gas plant with PEM electrolysis. *Int. J. Hydrog. Energy* **42**, 13311–13320 (2017).
+100. Alfian, M. & Purwanto, W. W. Multi-objective optimization of green urea production. *Energy Sci. Eng.* **7**, 292–304 (2019).
+101. Schmidt, P., Batteiger, V., Roth, A., Weindorf, W. & Raksha, T. Power-to-Liquids as Renewable Fuel Option for Aviation: A Review. *Chem. Ing. Tech.* **90**, 127–140 (2018).
+102. Nosherwani, S. A. & Neto, R. C. Techno-economic assessment of commercial ammonia synthesis methods in coastal areas of Germany. *J. Energy Storage* **34**, 102201 (2021).
+103. Schmidt, O. et al. Future cost and performance of water electrolysis: An expert elicitation study. *Int. J. Hydrog. Energy* **42**, 30470–30492 (2017).
+104. Treyer, K., Sacchi, R. & Bauer, C. *Life Cycle Assessment of Synthetic Hydrocarbons for Use as Jet Fuel: ‘Power-to-Liquid’ and ‘Sun-to-Liquid’ Processes*. (2022).
+105. Liu, C. M., Sandhu, N. K., McCoy, S. T. & Bergerson, J. A. A life cycle assessment of greenhouse gas emissions from direct air capture and Fischer–Tropsch fuel production. *Sustain. Energy Fuels* **4**, 3129–3142 (2020).
+106. Rosental, M., Fröhlich, T. & Liebich, A. Life Cycle Assessment of Carbon Capture and Utilization for the Production of Large Volume Organic Chemicals. *Front. Clim.* **2**, (2020).
+107. Doty, F. D., Doty, G. N., Staab, J. P. & Holte, L. L. Toward Efficient Reduction of CO2 to CO for Renewable Fuels. in 775–784 (American Society of Mechanical Engineers Digital Collection, 2010). doi:10.1115/ES2010-90362.
+108. Matute, G., Yusta, J. M. & Correas, L. C. Techno-economic modelling of water electrolysers in the range of several MW to provide grid services while generating hydrogen for different applications: A case study in Spain applied to mobility with FCEVs. *Int. J. Hydrog. Energy* **44**, 17431–17442 (2019).
+109. Reksten, A. H., Thomassen, M. S., Møller-Holst, S. & Sundseth, K. Projecting the future cost of PEM and alkaline water electrolysers; a CAPEX model including electrolyser plant size and technology development. *Int. J. Hydrog. Energy* **47**, 38106–38113 (2022).
+110. Terlouw, T., Bauer, C., McKenna, R. & Mazzotti, M. Large-scale hydrogen production via water electrolysis: a techno-economic and environmental assessment. *Energy Environ. Sci.* (2022) doi:10.1039/D2EE01023B.
+111. Elsernagawy, O. Y. H. et al. Thermo-economic analysis of reverse water-gas shift process with different temperatures for green methanol production as a hydrogen carrier. *J. CO2 Util.* **41**, 101280 (2020).
+
+# Footnotes
+
+1. With Boeing 787-9 Dreamliner
+2. Average of Boeing 787-9 Dreamliner, Boeing 777-200LR, Boeing 777-3000LR, and Boeing 767−300
+3. With Boeing 737–900
+
+# Supplementary Files
+
+- [SupplementaryInformation.docx](https://assets-eu.researchsquare.com/files/rs-3981416/v1/1311b53361c2a443334c8658.docx)

015d01aca154cd482068a6a8178be082317064ce45885171da176f0dc7e6914b/preprint/images_list.json

@@ -0,0 +1,50 @@
+[
+    {
+        "type": "image",
+        "img_path": "images/Figure_1.jpg",
+        "caption": "\u03b1CD40.RBD vaccine targeting and immunogenicity in hu-mice. (A) Binding to solid-phase attached human CD40 ectodomain protein by anti-CD40 12E12 human monoclonal antibody (Ab, filled pink circles) by the anti-CD40 12E12-RBD vaccine (\u03b1CD40.RBD, filled green triangles) and control IgG4 (bckg, filled orange triangles). (B) Binding of 12E12 antibody (Ab) and \u03b1CD40.RBD vaccine to CD40-expressing PBMCs of three naive cynomolgus macaques measured by flow cytometry. Cell subsets were defined by the gating strategy shown in Extended data Fig. 1B. (C) Fold change of the geometric mean fluorescence intensity (MFI) of activation markers after 18 h of incubating NHP (n = 3) PBMCs with the \u03b1CD40.RBD vaccine for cell subsets targeted by the \u03b1CD40.RBD vaccine and identified in (B). (D) Schematic overview of vaccination strategies in NSG humanized (hu) mice, including three experimental groups, 9 to 10 animals/group. (E) SARS-CoV-2 S protein-specific IgG-switched human B-cell frequencies within the hu-B cells in the blood of hu-mice three weeks after the priming injection. (F) Flow cytometry t-SNE analysis of splenic CD19+ B cells based on five markers (mCD45, hCD45, hCD19, hCD20, hCD38) showing the clustering of PCs, early plasma blasts (PBs), and a population of PBs and immature PCs. (G) Mapping of CD20 and CD38 onto the splenic hu-B-cell clusters obtained following t-SNE analysis. (H) Representative examples of t-SNE of one hu-mouse from each group. (I) SARS-CoV-2 S protein-specific IgG-switched human B-cell frequencies within the hu-B cells in the spleen of hu-mice six weeks after the priming injection. (J) Mapping of the SARS-CoV-2 S protein trimer, CXCR4, and CCR10 onto the splenic hu-B-cell clusters obtained following t-SNE analysis.",
+        "footnote": [],
+        "bbox": [],
+        "page_idx": -1
+    },
+    {
+        "type": "image",
+        "img_path": "images/Figure_2.jpg",
+        "caption": "SARS-CoV-2 specific B- and T-cell responses induced by \u03b1CD40.RBD in convalescent NHP. (A) Study design in cynomolgus macaques. (B) Relative MFI of IgG binding to SARS-CoV-2 S protein, measured using a Luminex-based serology assay, in serum samples (mean \u00b1 SD of 6 animals per group). The red and blue vertical dotted lines indicate viral exposure and vaccination, respectively. (C) SARS-CoV-2 S protein-specific binding before any exposure to SARS-CoV-2 (week -26) and on the week of vaccine injection (week 0) in macaques (n = 12) compared to convalescent humans (n = 7) sampled 24 weeks after the onset of symptoms. The horizontal dotted line represents the background threshold. (D) Quantification of SARS-CoV-2 antibodies against RBD measured in the serum of NHPs using a multiplexed solid-phase chemiluminescence assay. Each plain line indicates the individual values, and the bold dotted lines represent the mean for each experimental group. (E) Quantification of antibody-induced inhibition of ACE-2 binding in NHP serum. Symbols are as for panel D. (F) Frequency of RBD-specific Th1 CD4+ T cells (CD154+ and IFN-\u0263+/- IL-2+/- TNF-\u03b1) in the total CD4+ T cell population for each non-immunized convalescent macaque (blue lines and symbols) and \u03b1CD40.RBD-vaccinated convalescent macaque (green lines and symbols). PBMC were stimulated overnight with SARS-CoV-2 RBD overlapping peptide pools. Time points in each experimental group were compared using the Wilcoxon signed rank test. (G) Frequency of cytokine producing cells in the RBD-specific CD4+ T cells (CD154+) for \u03b1CD40.RBD-vaccinated convalescent macaque. Each bar indicated the mean of the 6 vaccinated convalescent macaques. Distribution of cytokines is indicated within each bar. BL: Baseline approximately 1 week before immunization; \u201cPost imm.\u201d: Two weeks post immunization.",
+        "footnote": [],
+        "bbox": [],
+        "page_idx": -1
+    },
+    {
+        "type": "image",
+        "img_path": "images/Figure_3.jpg",
+        "caption": "Efficacy of \u03b1CD40.RBD in convalescent cynomolgus macaques. (A) Genomic viral RNA (gRNA) quantification in tracheal swabs of naive (left, grey lines), convalescent (middle, blue lines), and \u03b1CD40.RBD-vaccinated convalescent macaques (right, green lines). The bold line represents the mean viral load for each experimental group. (B) Mean of subgenomic (sgRNA) viral loads in tracheal swabs. (C) Percentage of macaques with viral gRNA above the limit of detection (LOD) over time in tracheal swabs. Experimental groups were compared using log Rank tests. (D) Area under the curve (AUC) of gRNA viral loads in tracheal (left panel) and nasopharyngeal swabs (right panel). (E) gRNA viral quantification in BAL three days post-exposure (d.p.expo). Groups were compared using the non-parametric Mann-Whitney test. (F) Quantification of SARS-CoV-2 IgG binding N, S, and RBD after challenge. Each plain line indicates individual values, and the bold dotted lines represent the mean for each experimental group. (G) Quantification of antibody-induced inhibition of ACE-2 binding. Lines as in panel F. (H) Lung CT-scores of macaques before challenge and at 2 and 6 d.p.expo to SARS-CoV-2. The CT score includes lesion type and lesion volume summed for each lobe. (I) Correlation matrix between virological and immune parameters. The heatmap indicates the Spearman r values (Only values between -0.7 and -1, and 0.7 and 1 are colored in the heatmap). (J) Correlation between antibody-induced inhibition of ACE-2 binding at 0 d.p.expo. and tracheal gRNA viral loads (left) or IL-1RA plasma concentration (right) at 2 d.p.expo. ",
+        "footnote": [],
+        "bbox": [],
+        "page_idx": -1
+    },
+    {
+        "type": "image",
+        "img_path": "images/Figure_4.jpg",
+        "caption": "Modeling of viral dynamics. Estimations of the effect of the \u03b1CD40.RBD-vaccinated convalescent  group on the infectivity (\u03b2) and clearance of the infected cells (\u03b4) in the trachea relative to na\u00efve control group. T: uninfected target cells, I1: Infected cells, I2: productively-infected cells, Vs: Virus inoculum, grey hexagon: non-infectious viral particles (VNI), Red hexagon: infectious viral particles (VI).",
+        "footnote": [],
+        "bbox": [],
+        "page_idx": -1
+    },
+    {
+        "type": "image",
+        "img_path": "images/[IMAGE_METHODS_1].png",
+        "caption": "",
+        "footnote": [],
+        "bbox": [],
+        "page_idx": -1
+    },
+    {
+        "type": "image",
+        "img_path": "images/[IMAGE_METHODS_2].png",
+        "caption": "",
+        "footnote": [],
+        "bbox": [],
+        "page_idx": -1
+    }
+]

015d01aca154cd482068a6a8178be082317064ce45885171da176f0dc7e6914b/preprint/preprint.md

@@ -0,0 +1,179 @@
+# Abstract
+
+Controlling the circulation of the recently emerged SARS-CoV-2 in the human populations requires massive vaccination campaigns. Achieving sufficient worldwide vaccination coverage will require additional approaches to first generation of approved viral vector and mRNA vaccines. Subunit vaccines have excellent safety and efficacy records and may have distinct advantages, in particular when immunizing individuals with vulnerabilities or when considering the vaccination of children and pregnant women. We have developed a new generation of subunit vaccines with enhanced immunogenicity by the targeting of viral antigens to CD40-expressing antigen-presenting cells, thus harnessing their intrinsic immune-stimulant properties. Here, we demonstrate that targeting the receptor-binding domain (RBD) of the SARS-CoV-2 spike protein to CD40 (αCD40.RBD) induces significant levels of specific T and B cells, with a long-term memory phenotype, in a humanized mouse model. In addition, we demonstrate that a single dose of the αCD40.RBD vaccine, injected without adjuvant, is sufficient to boost a rapid increase in neutralizing antibodies in convalescent non-human primates (NHPs) exposed six months previously to SARS-CoV-2. Such vaccination thus significantly improved protection against a new high-dose virulent challenge versus that in non-vaccinated convalescent animals. Viral dynamics modelling showed the high efficiency of the vaccine at controlling the viral dissemination.
+
+[Virology](/browse?subjectArea=Virology) | [Infectious Diseases](/browse?subjectArea=Infectious%20Diseases) | [Immunology](/browse?subjectArea=Immunology) | [Vaccine Development](/browse?subjectArea=Vaccine%20Development) | Vaccine | SARS-CoV-2 | DC-targeting | convalescent | Non-Human Primate
+
+# Main Text
+
+Coronavirus-induced disease 2019 (COVID-19) is caused by a zoonotic virus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which has rapidly spread during the last year and a half, infecting over 100 million humans and causing more than two million deaths worldwide. Durable control of the pandemic requires mass vaccination strategies, for which the first vaccine candidates became available at the end of 2020. Although there are a limited number of previously licensed vector-based vaccines for human use, recombinant DNA vector and synthetic mRNA vaccines have nevertheless become the most advanced in the fight against COVID-19 because of the many possibilities offered for genetic engineering and rapid scalability<sup>1-4</sup>. Given that the benefits outweigh the risks for their use in humans, an Emergency Use Authorization (EUA) was favorably evaluated by the US Food and Drug Administration (FDA) for the first two mRNA vaccines encoding a pre-fusion stabilized SARS-CoV-2 spike glycoprotein<sup>3,5-7</sup>. The estimated efficacy after phase III clinical trial interim analysis was approximately 94% to 95% in preventing COVID-19 in the short term following the second immunization. Continued evaluation of the effectiveness of the vaccines following EUA issuance is needed to confirm these initial promising results. Long-term efficacy data will be critical for estimating their impact on progression of the pandemic. Initial reports on adverse events may not limit their deployment, but safety assessments require extended follow-up. Further evaluations will also be needed to assess the efficacy of the vaccines in preventing asymptomatic infections and reducing viral shedding to the level required to prevent secondary transmission. If not efficiently prevented, asymptomatic infections in combination with reduced mask wearing and social distancing could result in significant continuing circulation of the virus<sup>5</sup>.
+
+A new generation of COVID-19 vaccines is needed to counteract the development of the pandemic. Providing the necessary billions of doses to achieve sufficient global coverage will not be possible with any single product. In addition, there are uncertainties about the long-term efficacy and safety of these first-in-class vector or mRNA vaccine platforms, with a limited history of use, particularly in vulnerable individuals, including frail, older individuals, people with co-morbidities, and immunosuppressed patients. Importantly, the use of vector-based vaccines will require cautious and long-term safety assessment before considering their use in children and pregnant women. Although younger individuals are less prone to develop severe disease, they are susceptible to mild COVID-19 or asymptomatic infection and may facilitate circulation of the virus and the potential for further mutation. Control of the pandemic will also require the mass immunization of children.
+
+The constraints of antigen design and engineering and the time required for the production of large numbers of doses make subunit vaccines difficult to develop as first countermeasures for suddenly emerging and non-anticipated epidemics. However, licensed subunit vaccines have proven tolerability and safety in diverse population classes<sup>8</sup>. Several adjuvanted SARS-CoV-2 spike protein vaccines are able to elicit neutralizing antibodies to protective levels in relevant animal models, including non-human primate (NHP) challenge studies<sup>9-11</sup>. These advantages may be decisive in the development of the next-generation vaccines aimed at controlling the long-term circulation of SARS-CoV-2, in particular if the virus continues provoking seasonal epidemic waves of COVID-19.
+
+Dendritic cells (DCs) are immune system controllers that can deliver differential signals to other immune cells through intercellular interactions and soluble factors, resulting in a variety of host immune responses of varying quality. Targeting vaccine antigens to DCs via surface receptors represents an appealing strategy to improve subunit-vaccine efficacy while reducing the amount of required antigen. Direct delivery of the antigen, which can additionally activate cell receptors, may also evoke a danger signal, stimulating an immune response without the need of additional immune stimulants, such as adjuvants. Among the various DC receptors tested, including lectins and scavenger receptors, we reported the superiority of vaccines targeting diverse viral antigens to CD40 expressing antigen-presenting cells in evoking strong antigen-specific T- and B-cell responses<sup>12-16</sup>. Drawing from this knowledge, we developed a vaccine that targets the receptor-binding domain (RBD) of the SARS-CoV-2 spike antigen to the CD40 receptor (αCD40.RBD). We proved its immunogenicity in two different animal models. A single dose of the αCD40.RBD administered without adjuvant boosted the protective response in COVID-19 convalescent NHPs.
+
+## Humanized anti-CD40 monoclonal antibody fused to the receptor-binding domain of the spike antigen targets and activates human and macaque antigen-presenting cells
+
+The human ACE2 receptor is the crucial target for the receptor-binding domain (RBD) of the spike (S) protein of SARS-CoV2, for which this strong interactive synapse assists viral entry into host cells<sup>17</sup>. The RBD is a logical target for the development of neutralizing antibodies, as well as serving as a potential source of T-cell epitopes to elicit cellular immune responses. Thus, we engineered vectors expressing SARS-CoV-2 RBD (residues 173-591 of sequence ID: QIC50514.1) fused to the C-termini of the anti-human CD40 humanized 12E12 IgG4 antibody to generate the αCD40.RBD vaccine<sup>16,18,19</sup>.
+
+As evaluated by a solid-phase direct-binding assay<sup>16</sup>, there was no significant difference in CD40 binding affinity (EC50 30 pM) between the 12E12 anti-CD40 monoclonal antibody (mAb) and 12E12 anti-CD40 fused to RBD (EC50 35 pM) (Fig. 1A and Extended data Fig. 1A). We have previously shown that 12E12 anti-CD40 fused to viral antigens enhances CD40-mediated internalization and antigen-presentation by mononuclear cells and *ex vivo* generated monocyte-derived DCs<sup>12,18</sup>. Similarly, we show here that the αCD40-RBD vaccine binds (Fig. 1B; Extended data Fig. 1B-C) and activates (Fig. 1C; Extended data Fig. 1D) macaque monocytes, DCs, and B cells obtained from peripheral blood mononuclear cells (PBMCs).
+
+## The αCD40-RBD vaccine induces robust human B- and T-cell responses in humanized mice
+
+We first assessed the immunogenicity of the αCD40-RBD vaccine in NSG (NOD/SCID gc<sup>-/-</sup>) mice with a human immune system (hu-mice) generated by reconstituting newborns with human fetal liver hematopoietic stem cells (Fig. 1D). A single injection of αCD40-RBD (10 μg) adjuvanted with polyinosinic-polycytidylic acid (Poly-IC, 50 μg) by the intraperitoneal route was sufficient to elicit SARS-CoV-2 S protein-specific IgG-switched human B cells in the blood of 50% of immunized mice (Fig. 1E). At week 6, one week after the last αCD40.RBD boost, unbiased t-SNE analysis of the splenic human CD19<sup>+</sup> B cells revealed cell clusters corresponding to well-described subsets of terminally differentiated plasma cells (PCs), early plasma blasts (PBs), and a contingent of PBs and immature PCs in the vaccine groups but not controls (Fig. 1F-I). At the same timepoint, splenic SARS-CoV-2 S protein-specific IgG-switched human B cells were detected in all vaccinated hu-mice (Fig. 1H), mainly of the PB and immature PC phenotype (Fig. 1H). All spike protein-specific IgG-switched human B cells expressed CXCR4 and a discrete cell island was observed in the t-SNE analysis driven by high expression of CCR10 (Fig. 1J), which was confirmed using manual back gating (Extended data Fig. 2A). We next evaluated the capacity of the vaccines to induce specific and functional CD4<sup>+</sup> and CD8<sup>+</sup> memory T cells. Th1 (IFN-ɣ<sup>+</sup> IL-2<sup>+</sup> TNF-α) type CD4<sup>+</sup> T-cell responses and IFNg-secreting CD8<sup>+</sup> T-cells were observed for the vaccinated hu-mice following *ex vivo* stimulation of splenocytes with RBD peptide pools (Extended data Fig. 2B-C). We confirmed the presence of human CD8<sup>+</sup> T cells specific for the predicted optimal epitopes from SARS-CoV-2 RBD protein in the spleens of vaccinated hu-mice using HLA-I tetramers (Extended data Fig. 2D-E).
+
+Subunit vaccines could also be considered as boosters for other type of vaccines in human vaccination campaigns. Thus, in addition to a homologous prime-boost regime, we tested the capacity of αCD40.RBD to boost heterologous priming with a vector-based vaccine. The DNA-launched self-amplifying RNA replicon vector encoding the SARS-CoV-2 spike glycoprotein (DREP)-S is a previously described platform<sup>20</sup> based on the alphavirus genome encoding the genes for the viral RNA replicase but lacking those encoding the structural proteins of the virus<sup>21</sup>. We demonstrated that αCD40.RBD efficiently boosted (DREP)-S primed B- and T-cell SARS-CoV-2 specific responses (Fig. 1E, 1I; Extended data Fig. 2).
+
+## The αCD40.RBD vaccine recalls specific immune responses in convalescent macaques
+
+The results in the hu-mice model are consistent with those of our previous CD40-targeted influenza and HIV vaccine studies<sup>13,14,18,19</sup> and demonstrate that αCD40.RBD is a potent prime or boost vaccine for eliciting RBD-specific T- and B-cell responses similar in magnitude to previously reported protective responses<sup>11</sup>. In addition, we previously showed that nanomolar amounts of αCD40 vaccines can elicit *in vitro* recall responses in PBMCs collected from individuals primed by the natural viral infection<sup>18,22</sup>. Thus, we tested the hypothesis that the αCD40.RBD vaccine can efficiently elicit recall responses *in vivo* in SARS-CoV-2 convalescent individuals. The improved immunogenicity obtained by CD40 targeting and the stimulating capacity of the αCD40.RBD vaccine also suggested that adjuvant may not be necessary to elicit a protective recall response. We thus subcutaneously injected six convalescent cynomolgus macaques with 200 µg of the vaccine without adjuvant. An additional 12 animals (six convalescent and six naive) were injected with PBS as controls (Fig. 2A). All the convalescent macaques, randomly distributed between the vaccine and control groups, had been infected approximately six months before this study (range = 26-24 weeks) with SARS-CoV-2 in a study to evaluate pre-exposure or post-exposure prophylaxis with hydroxychloroquine (HCQ). No evidence of antiviral efficacy<sup>23</sup> of HCQ was observed and after this first exposure to the virus, all animals developed similar profiles of viral load (Extended data Fig. 3A and 3B) and suffered from transient and moderate disease, resulting in increased levels of anti-S IgG antibodies detected in the serum (Fig. 2B). At the time of the αCD40.RBD-vaccine injection, anti-S IgG levels in the two groups of convalescent macaques were comparable and in the average range of specific responses detected in the sera of convalescent patients (Fig. 2C). Before vaccination, the infection of macaques with SARS-CoV-2 generated both anti-RBD antibodies (Fig. 2D) and low but detectable levels of antibodies inhibiting the binding of the spike protein to the ACE2 receptor (Fig. 2E). Before vaccination, low Th1 (IFN-ɣ<sup>+</sup> IL-2<sup>+</sup> TNF-α) type CD4<sup>+</sup> T-cell responses were observed for both groups of convalescent macaques following *ex vivo* stimulation of PBMCs with RBD and N-peptide pools (Fig. 2F; Extended data Fig. 2E). None of the convalescent animals had detectable anti-RBD or anti-N CD8<sup>+</sup> T cells (Extended data Fig. 2F).
+
+Two weeks after αCD40.RBD vaccine injection, all six vaccinated macaques exhibited significantly increased levels of anti-S (Fig. 2B) and anti-RBD IgG (Fig. 2D) in the serum, which correlated with an increased capacity of inhibition of RBD binding to the ACE2 receptor (p = 0.022, Fig. 2E), as they remained elevated four weeks after vaccination. None of these parameters increased in PBS-injected convalescent controls (Fig. 2D and 2E). In addition, anti-S IgG levels in the vaccinated macaques were higher (p = 0.0018) than those typically observed in humans 1 to 3 months after symptomatic SARS-CoV-2 infection (Extended data Fig. 3C). The immunization also elicited a significant increase in the anti-RBD Th1 response in all six immunized animals (p = 0.031; Fig. 2F-G), whereas no changes in the magnitude of anti-N CD4<sup>+</sup> T cells (Extended data Fig. 4E) or SARS-CoV-2 specific CD8<sup>+</sup> T cells was observed (Extended data Fig. 4F).
+
+## The αCD40.RBD vaccine improves the protection of convalescent macaques against SARS-CoV-2 reinfection
+
+Four weeks following vaccine or placebo injection, the 12 convalescent macaques were exposed a second time to a high dose (1x10<sup>6</sup> pfu) of SARS-CoV-2 administered via the combined intra-nasal and intra-tracheal route using a previously reported challenge procedure<sup>23</sup>. Six SARS-CoV-2 naive animals were also challenged as controls.
+
+All naive animals became infected, as shown by the detection of viral genomic (gRNA) and sub-genomic (sgRNA) RNA in tracheal (Fig. 3A-D; Extended data Fig. 5A and 5B) and nasopharyngeal (Fig. 3D; Extended data Fig. 5C-G) swabs and broncho-alveolar lavages (BAL, Fig. 3E and Extended data Fig. 5H). Of note, the dynamics of viral replication in these animals was comparable to that observed during the first infection six months earlier in the two groups of convalescent macaques (Extended data Fig. 3A-B). The non-vaccinated convalescent animals were not protected against the second SARS-CoV-2 challenge, but significantly lower viral RNA levels were detected in the upper respiratory tract than in the naive animals (Fig. 3A-E and Extended data Fig. 5). The αCD40.RBD vaccine remarkably improved the partial protection observed in the convalescent macaques. All vaccinated animals exhibited significantly lower viral gRNA levels (p = 0.015, Fig. 3D) than the non-vaccinated convalescent animals. The levels of sgRNA remained below the limit of detection in upper respiratory tract samples for 5 of 6 vaccinated animals, whereas sgRNA was detected in 4 of 6 non-vaccinated convalescent and all naive control animals (Extended data Fig. 5B and 5G). Moreover, the time post-exposure (p.expo.) to reach undetectable gRNA levels was significantly lower in vaccinated convalescent than non-vaccinated and control animals (Fig. 3C and Extended data Fig. 5E, log rank, p < 0.0001). The efficacy of vaccination was also higher in the lower respiratory tract, as only 3 of 6 vaccinated macaques were above the limit of detection for gRNA in BAL at day 3 p.expo. versus day 6 for the six non-vaccinated convalescent animals (Fig. 3E). Complete protection from shedding of the virus from the gastrointestinal tract was noted in the immunized macaques (Extended data Fig. 5I), which is probably an important factor to prevent secondary viral transmission<sup>24</sup>.
+
+The reduction of viral load in vaccinated and non-vaccinated convalescent macaques relative to naive infected animals was associated with a limited impact on leukocyte numbers (Extended data Fig. 6) and reduced cytokine concentrations in the plasma, in particular those of IL-1RA and CCL2 (Extended data Fig. 7B). Such viral loads and cytokine profiles were also associated with a reduction in lung lesions (Fig. 3H and Extended data Fig. 8), as scored by X-ray computerized tomography (CT).
+
+We then analyzed the immune responses of all animals following SARS-CoV-2 viral challenge. The naive controls showed the slowest development of anti-S, anti-RBD, and anti-N IgG (Fig. 3F), of which the levels remained significantly lower than for the other two groups at day 20 p.expo. (p = 0.022). The non-vaccinated convalescent animals raised a rapid and robust anamnestic antibody response (Fig. 3F), which was associated with a significant increase (p = 0.031) in the serum capacity to neutralize ACE2 binding to RBD (Fig. 3G) by p.expo. day 9. The anti-S- and anti-RBD-specific antibody responses and neutralization activity of the serum was maintained in the vaccinated macaques at the high levels already achieved at the time of challenge and remained superior to that of the control macaques (Fig. 3F and 3G). The anti-RBD Th1 CD4<sup>+</sup> response increased post challenge for most of the control (convalescent and naive) animals, with higher levels for some of the naive controls as early as p.expo. day 9 (Extended data Fig. 4G). On the contrary, all 18 animals showed comparable antibody and CD4<sup>+</sup> T cell responses to the N-peptide pool (Extended data Fig. 4G), probably reflecting a predominance of the response against non-structural antigens in infected individuals. The IFN-ʏ-mediated CD8<sup>+</sup> T-cell response was also mainly directed against the N peptides (Extended data Fig. 4D), but with a significantly reduced intensity in all convalescent macaques than in the naive controls (Extended data Fig. 4H), probably reflecting the lower exposure to viral antigens as a result of better control of viral replication.
+
+Spearman analysis between all recorded parameters revealed that the induction of anti-RBD- and ACE2-inhibiting antibodies was the strongest parameter to correlate with the reduction of viral load and disease markers, as were the plasma levels of the inflammatory cytokines IL-1RA and CCL2 (Fig. 3I and J).
+
+## Modeling of the dynamics of viral replication supports the capacity of αCD40.RBD to induce the blockade of initial viral entry into host cells and then limit secondary transmission
+
+We developed a mathematical model to better characterize the impact of the immune response on viral gRNA and sgRNA dynamics. We wished to compare the differences in immunity, in particular in the reduction of the cell-infection rate and the increase in the clearance of infected cells generated by vaccination versus immunity developed after infection. The model was adapted from previously published studies<sup>25,26</sup> and includes uninfected target cells (T) that can be infected (I<sub>1</sub>) and then produce virus after an eclipse phase (I<sub>2</sub>). The virus generated can be infectious (V<sub>I</sub>) or non-infectious (V<sub>NI</sub>). We completed the model with a compartment for the inoculum to distinguish between the injected virus (V<sub>s</sub>) and the virus produced *de novo* by the host (V<sub>I</sub> and V<sub>NI</sub>). We estimated viral infectivity (β) and the loss rate of infected cells (δ) using the sRNA and sgRNA viral loads, measuring V<sub>S</sub>, V<sub>I</sub>, and V<sub>NI</sub>.
+
+In this model, the αCD40.RBD vaccine reduced the infection of target cells in the trachea by an estimated 99% (Fig. 4, Table 1) relative to the two other groups, suggesting that the levels of anti-RBD antibodies induced by the vaccine are highly efficient for the neutralization of new infections *in vivo*. In addition, both specific antibodies and specific CD8<sup>+</sup> T cells are mechanisms commonly considered to be important for killing infected cells and thus reducing dissemination of the virus. According to our model, the estimated clearance of infected cells was 0.94 /day (95% CI 0.87; 1.02) in naive macaques, which was increased by 2.18-fold (118%) in the non-vaccinated convalescent and 2.86 fold in the αCD40.RBD-vaccinated convalescent animals (149% relative to naive controls and 31% relative to convalescent controls). Hence, the model predicts that the target cell levels (all infected and non-infected cells expressing ACE2) would be decreased by the previous infection in the naive and convalescent groups, whereas it would be preserved in vaccinated animals due to the blockade of new infections and increased clearance of infected cells. Similar effects were predicted for the nasopharyngeal compartment (Extended data Fig. 9).
+
+# Conclusions
+
+In humans, the durability of protection induced by natural SARS-CoV-2 infection and the first vaccine candidates is unknown. In convalescent humans, the virus neutralizing-antibody response wanes and re-infections have been reported within months following previous exposure<sup>24,27</sup>. The decrease in neutralizing-antibody levels observed in most patients within three months post-infection may suggest that vaccine boosters will be required to provide long-lasting protection<sup>28</sup>. In contrast to previous NHP re-challenge studies performed shortly after a first infection<sup>29</sup>, we demonstrate that SARS-CoV-2 reinfection is not fully prevented in convalescent macaques six months after initial exposure to the virus, confirming that protective immunity wanes over time. In addition, the vaccines currently used in humans are aimed at preventing severe disease and information is still lacking as to their capacity to prevent infection and reduce initial viral replication. Vaccinated individuals who develop an asymptomatic or mild symptomatic infection may continue transmitting the virus and actively contribute to circulation of the virus. The αCD40.RBD vaccine we developed significantly improved immunity when administered to convalescent macaques, resulting in a reduction of viral load following re-exposure to the virus down to levels that may avoid such secondary transmission. This vaccine may therefore represent an excellent booster of pre-existing immunity, either induced by natural infection or previous priming with vector-based vaccines. Indeed, we confirmed the efficacy of αCD40.RBD as a booster in (DREP)-S primed hu-mice. This new-generation subunit vaccine, improved by targeting of the antigen to CD40-expressing cells, may have decisive advantages for the rapid provision of a safe and efficient boosting strategy. The capacity to induce protective immunity without requiring an adjuvant would accelerate the development of a vaccine with improved tolerability for people with specific vulnerabilities and children, an important part of the population to consider in the control of circulation of the virus.
+
+# Methods
+
+## Ethics and biosafety statement animal studies
+
+The NSG (NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ) humanized mice (hu-mice) were supplied by the Jackson Laboratories (Bar Harbor, ME, USA) under MTA #1720. Five donors whose HLA typing is recapitulated in the supplemental table S1 provided hematopoietic stem cells for human immune system reconstitution of the mice. The level of human immune cells reconstitution reached an average of 70%. The hu-mice were housed in Mondor Institute of Biomedical Research infrastructure facilities (U955 INSERM-Paris East Creteil University, Ile-de-France, France). The protocols were approved by the institutional ethical committee “Comité d’Ethique Anses/ENVA/UPEC (CEEA-016)” under statement number 20-043 #25329. The study was authorized by the “Research, Innovation and Education Ministry” under registration number 25329-2020051119073072 v4.
+
+Cynomolgus macaques (*Macaca fascicularis*), aged 37-58 months (8 females and 13 males) and originating from Mauritian AAALAC certified breeding centers were used in this study. All animals were housed in IDMIT facilities (CEA, Fontenay-aux-roses), under BSL-3 containment (Animal facility authorization #D92-032-02, Préfecture des Hauts de Seine, France) and in compliance with European Directive 2010/63/EU, the French regulations and the Standards for Human Care and Use of Laboratory Animals, of the Office for Laboratory Animal Welfare (OLAW, assurance number #A5826-01, US). The protocols were approved by the institutional ethical committee “Comité d’Ethique en Expérimentation Animale du Commissariat à l’Energie Atomique et aux Energies Alternatives” (CEtEA #44) under statement number A20-011. The study was authorized by the “Research, Innovation and Education Ministry” under registration number APAFIS#24434-2020030216532863v1.
+
+## αCD40.RBD vaccine
+
+Production and quality assurance of the αCD40.RBD vaccine Vectors and sequences for humanized anti-human CD40 12E12 IgG4 and control human IgG4 antibodies have been described previously<sup>1,2,3</sup>. GenBank sequences HQ738666.1 and KP684037 describe the human IgG4 chimeric forms of the 12E12, anti-human CD40 H and L chains. Methods for expression vectors and protein production and purification, via transient or stable CHO-S (Chinese Hamster Ovary cells) transfection and quality assurance including CD40 binding specificity were as are described. CHO-optimized codons encoding SARS-CoV-2 RBD residues 173-591 of sequence ID: QIC50514.1 with appended residues encoding a C-tag (EPEA) and a stop codon were inserted between the vector *Nhe*I and *Not*I sites positioned distal to the H chain C-terminal codon. Expression plasmids encoding the antibody H chain RBD fusion and the L chain were transiently transfected into Expi-CHO cells with TransIT-PRO Pro reagent (Mirus Bio) using the manufacturers protocol. The product was purified by protein A affinity capture of the culture medium followed by elution with a gradient of 1M L-Arginine monohydrochloride in H2O, from pH 8.0 and pH 1.8. Product was formulated in phosphate buffered saline (pH 7.4) with 125 mM cyclodextrin (average MW 1420). The LPS value was .037 ng/mg. Using a solid phase assay direct binding assay previously described<sup>3</sup> these was no significant difference in the CD40 binding affinity of anti-CD40 12E12 (EC50 30 pM) versus anti-CD40 12E12-RBD (EC50 35 pM).
+
+## DREP-S vaccine
+
+DREP-S vaccine constructs were made by cloning the sequences encoding S of SARS-CoV-2 spike protein into the Semliki Forest Virus (SFV) DREP plasmid vector backbone 3 using BamHI and SpeI restriction sites<sup>4</sup>. The S construct encodes the surface glycoprotein of SARS-CoV-2 (Wuhan-Hu-1) with an 18-aa deletion in the cytoplasmic tail (D18). The synthesis of the construct with the appropriate restriction sites was ordered from Twist bioscience. The spike variant was codon optimized for human expression and the construct’s sequence was confirmed by sequencing. Plasmid DNA of the DREP-S vaccine candidate was purified from bacterial cultures using the EndoFree Plasmid Maxi or Giga Kit (QIAGEN) and the concentration and purity was measured on a NanoDrop One (ThermoFisher).
+
+## Binding of αCD40.RBD vaccine to non-human primate cells and activation PBMC assays
+
+PBMC from 3 naïve macaques were isolated and stained for 15 min with anti-CD11b-V450 (ICRF44, BD), anti-CD3-V500 (SP34-2, BD), anti-CD11c-BV605 (3.9, BioLegend), anti-CD8-BV650 (BW135/80, Miltenyi Biotec), anti-CD20-SB702 (2H7, Invitrogen), anti-CD163-APC (GHI/61, BioLegend), anti-CD14-A700 (M5E2, BioLegend), anti-HLA-DR-APC-H7 (L243, BD), anti-CD4-FITC (L200, BD), anti-CD45-PerCP (D058-1283, BD) and anti-CD40-AF594 (12E12) or αCD40.RBD-AF594. Next, cells were washed twice with PBS and acquired on the ZE5 flow cytometer (Biorad). Moreover, a part of these PMBCs were also incubated 18 hours with culture medium (RPMI 1640 media with L-Glutamax supplemented with Penicillin / Streptomycin and 10% of fetal calf serum (FBS)) and stimulated with αCD40.RBD (10 µg/mL) or LPS (100 ng/mL, Invivogen). Next, cells were washed in PBS and incubated 15min with LIVE/DEAD fixable Blue Dead Cell marker (Life Technologies), anti-CD11b-V450 (ICRF44, BD), anti-CD3-V500 (SP34-2, BD), anti-CD86-BV605 (2331, BD), anti-CD11c-APC (3.9, BioLegend), anti-CD20-SB702 (2H7, Invitrogen), anti-CD80-BV786 (L307.4, BD), anti-CD8-BV650 (BW135/80, Miltenyi Biotec), anti-CD14-A700 (M5E2, BioLegend), anti-HLA-DR-APC-H7 (L243, BD), anti-CD4-FITC (L200, BD), anti-CD45-PerCP (D058-1283, BD), anti-CD69-PE-Cy7 (FN50, BD) and anti-CD40-AF594 (42G5). Next, cells were washed twice with PBS and acquired on the ZE5 flow cytometer (Biorad). Analysis was performed on FlowJo v.10 software. For activation markers, results were expressed as fold change geometric MFI, obtained by dividing the geometric MFI measure in αCD40.RBD or LPS stimulation by the background geometric MFI measure in control stimulation (incubation with medium only).
+
+## Vaccination of humanized mice
+
+The hu-mice received immunizations at week 0, 3, and 5. The priming injection was an intraperitoneal administration of 10 μg of αCD40-RDB adjuvanted with 50 μg of polyinosinic-polycytidylic acid (Poly-IC; Invivogen) combined or not with an intramuscular injection of DREP-S (10 μg). Then hu-mice received booster i.p injections of αCD40-RDB (10 μg) plus Poly-IC (50 μg). Blood was collected at weeks 0 (before immunization), 3, and 6. Hu-mice were euthanized at week 6.
+
+## T-cells response in hu-mice
+
+To analyze the SARS-CoV-2 RBD protein-specific T cell using functional recall assay, we used fifteen-mer peptides (n = 70) overlapping by 11 amino acids (aa) and covering the vaccine RBD sequence (aa281-571 from Spike) synthesized by JPT Peptide Technologies (Berlin, Germany) and used at a final concentration of 1 µg/mL. We also used HLA class I PE labelled tetramers purchased from ProImmune Ltd (Oxford, UK). We used the following two specificities: SARS-CoV-2 A*0201 KIA (KIADYNYKL), SARS-CoV-2 A*0301 KCY (KCYGVSPTK).
+
+Cryopreserved hu-mice spleen cells from 6 weeks after the priming immunization (one week after final immunization) were thawed and counted. Cells were rested overnight in RPMI 1640 media with L-Glutamax supplemented with Penicillin / Streptomycin and 10% of human serum. Subsequently, cells from HLA-A*0201 and HLA-A*0301 donors were pooled together for the mock group and group 2 plus 3 vaccinated hu-mice, then cultured at 0.6x10<sup>6</sup> cells per condition with 1µg/mL of 15-mers peptides JPT Peptide Technologies (Berlin, Germany). As a negative control no stimulant was added, and as a positive control 1 µL of DynabeadsTM CD3/CD28 (ThermoFischer Scientific) were used. IL-2 (100 IU/mL, R&D System) was added on day 2, half of the volume of each culture well was refreshed with fresh media containing IL-2 (10 U/mL) at day 5 and with fresh media without IL-2 at day 7. On day 8, cells were re-stimulated: no stimulant was added in the negative control, 100 ng/mL Staphylococcal enterotoxin B (LL-122, Cliniscience) was added in the positive control and 15-mers peptides in the condition of interest. BD GolgiPlug (Becton Dickinson France) was added in all conditions and the culture was continued for additional 18 hours. Next, spleen cells were washed using FACS buffer (PBS, supplemented with 1% FBS) and incubated with tetramer-PE (ProImmune Ltd, Oxford, UK), Live dead fixable Aqua Dead marker (Life Technologies) and the following antibodies: anti-hCD3-A700 (UCHT1, Sony), anti-h-CD4-BV605 (RPA-T4, Sony), anti-hCD8-APC-Cy7 (SK1, Sony) for 30 minutes. Following fixation and permeabilization, spleen cells were stained with intracellular antibodies: anti-hIFNγ-PerCPCy5.5 (B27, Sony), anti-hIL-2-BV421 (17H12, Sony), anti-hTNFα-PC7 (Mab11, Sony) for 30 minutes. Stained cells were acquired on the LSRII flow cytometer (BD Biosciences). FlowJo v.10.7.1 software was used for data analysis (TreeStar, Inc., Ashland, OR).
+
+## SARS-CoV-2 S protein-specific B cell analysis
+
+Hu-mice PBMC from 3 weeks after the priming immunization and hu-mice PBMC and spleen cells from 6 weeks (one week after the last recall injection) were incubated first with the biotinylated SARS-CoV-2 S protein for 30 min at 4°C. After a washing step, cells were stained for 30 min at 4°C with streptavine-AF700 (ThermoFisher Scientific), anti-human (h) CD45-PeCy7 (HI30, Sony), anti-mouse (m) CD45-BV711 (30F11, Sony), anti-hCXCR4-Pe-Dazzle (12G5, eBiosciences), anti-hCCR10-PE (314305; R&D System), anti-CD3-FITC (SK7, Biolegend), anti-CD14-FITC (M5E2, Sony), anti-IgM-FITC (MHM-88, Biolegend) antibodies and the following B cell-specific antibodies: anti-hCD19-PacBlue (HIB19, Sony), anti-hCD20-APC (2H7, Sony), anti-hIgG-BV786 (G18-145, BD Biosciences), anti-hCD38-APC-H7 (HIT2, Sony). Staining on spleen cells also included a viability marker (LiveDead aqua or yellow stain ThermoFisher Scientific). Cells were washed twice with FACS buffer (PBS 1% FCS) and acquired on the LSRII flow cytometer (BD Biosciences). Analyses were performed on FlowJo v.10.7.1.
+
+## Non-human primate study design
+
+Convalescent cynomolgus macaques previously exposed to SARS-CoV-2 and used to assess hydroxychloroquine (HCQ) and azithromycin (AZTH) antiviral efficacy. None of the AZTH neither HCQ nor the combination of HCQ and AZTH showed a significant effect on viral replication<sup>5</sup>. Six months (24-26 weeks) post infection (p.i.), twelve of these animals were randomly assigned in two experimental groups. The convalescent vaccinated group (n=6) received 200 ug of αCD40.RBD vaccine by subcutaneous (SC) route diluted in PBS and without any adjuvant. The other six convalescent animals were used as controls and received the equivalent volume of PBS by SC. The two groups of convalescent animals were sampled at week 2 and 4 following vaccine or PBS injection for anti-SARS-CoV-2 immune response evaluation. Additional six age matched (43.7 months +/-6.76) cynomolgus macaques from same origin were included in the study as controls naïve from any exposure to SARS-CoV-2.
+
+## Evaluation of anti-Spike, anti-RBD and neutralizing IgG antibodies
+
+Anti-Spike IgG from human and NHP sera were titrated by multiplex bead assay. Briefly, Luminex beads were coupled to the Spike protein as previously described<sup>6</sup> and added to a Bio-Plex plate (BioRad). Beads were washed with PBS 0.05% tween using a magnetic plate washer (MAG2x program) and incubated for 1h with serial diluted individual serum. Beads were then washed and anti-NHP IgG-PE secondary antibody (Southern Biotech, clone SB108a) was added at a 1:500 dilution for 45 min at room temperature. After washing, beads were resuspended in a reading buffer 5 min under agitation (800 rpm) on the plate shaker then read directly on a Luminex Bioplex 200 plate reader (Biorad). Average MFI from the baseline samples were used as reference value for the negative control. Amount of anti-Spike IgG was reported as the MFI signal divided by the mean signal for the negative controls. Human sera from convalescent patients who were hospitalized with virologically confirmed COVID-19 were collected three months after symptoms recovery and used as controls for the titration of anti-Spike antibodies.
+
+Anti-RBD and anti-Nucleocapside (N) IgG were titrated using a commercially available multiplexed immunoassay developed by Mesoscale Discovery (MSD, Rockville, MD) as previously described<sup>7</sup>. Briefly, antigens were spotted at 200−400 μg/mL in a proprietary buffer, washed, dried and packaged for further use (MSD® Coronavirus Plate 2). Then, plates were blocked with MSD Blocker A following which reference standard, controls and samples diluted 1:500 and 1:5000 in diluent buffer were added. After incubation, detection antibody was added (MSD SULFO-TAG<sup>TM</sup> Anti-Human IgG Antibody) and then MSD GOLD<sub>TM</sub> Read Buffer B was added and plates read using a MESO QuickPlex SQ 120MM Reader. Results were expressed as arbitrary unit (AU)/mL.
+
+The MSD pseudo-neutralization assay was used to measure antibodies neutralizing the binding of the spike protein to the ACE2 receptor. Plates were blocked and washed as above, assay calibrator (COVID- 19 neutralizing antibody; monoclonal antibody against S protein; 200 μg/mL), control sera and test sera samples diluted 1:10 and 1:100 in assay diluent were added to the plates. Following incubation of the plates, an 0.25 μg/mL solution of MSD SULFO-TAG<sup>TM</sup> conjugated ACE-2 was added after which plates were read as above. Electro-chemioluminescence (ECL) signal was recorded and results expressed as 1/ECL.
+
+## Antigen specific T cell assays using non-human primate cells
+
+To analyze the SARS-CoV-2 protein-specific T cell using functional assay, 15-mer peptides (n = 70) overlapping by 11 amino acids (aa) and covering the vaccine RBD sequence (n=70, aa 281-571 from Spike) and the SARS-CoV-2 Nucleoprotein sequence (n=102, aa 1-419 from N) synthesized by JPT Peptide Technologies (Berlin, Germany) and used at a final concentration of 2 µg/mL.
+
+IFNy ELISpot assay of PBMC was performed using the Monkey IFNy ELISpot PRO kit (Mabtech Monkey IFNy ELISPOT pro, #3421M-2APT) according to the manufacturer’s instructions. PBMC were stimulated with RBD or N sequence overlapping peptide pools at a final concentration of 2 μg/mL. Plates were incubated for 18 h at 37°C in an atmosphere containing 5% CO2, then washed 5 times with PBS and incubated for 2 h at 37°C with a biotinylated anti-IFNy antibody. After 5 washes, spots were developed by adding 0.45 μm-filtered ready-to-use BCIP/NBT-plus substrate solution and counted with an automated ELISpot reader ELRIFL04 (Autoimmun Diagnostika GmbH, Strassberg, Germany). Spot forming units (SFU) per 1.0×10<sup>6</sup> PBMC are means of duplicates for each animal.
+
+T-cell responses were also characterized by measurement of the frequency of PBMC expressing IL-2 (PerCP5.5, MQ1-17H12, BD), IL-17a (Alexa700, N49-653, BD), IFN-γ (V450, B27, BD), TNF-α (BV605, Mab11, BioLegend), IL-13 (BV711, JES10-5A2, BD), CD137 (APC, 4B4, BD) and CD154 (FITC, TRAP1, BD) upon stimulation with the two peptide pools. CD3 (APC-Cy7, SP34-2, BD), CD4 (BV510, L200, BD) and CD8 (PE-Vio770, BW135/80, Miltenyi Biotec) antibodies was used as lineage markers. One million of PBMC were cultured in complete medium (RPMI1640 Glutamax+, Gibco; supplemented with 10 % FBS), supplemented with co-stimulatory antibodies (FastImmune CD28/CD49d, Becton Dickinson). Then cells were stimulated with S or N sequence overlapping peptide pools at a final concentration of 2 μg/mL. Brefeldin A was added to each well at a final concentration of 10µg/mL and the plate was incubated at 37°C, 5% CO2 during 18 h. Next, cells were washed, stained with a viability dye (LIVE/DEAD fixable Blue dead cell stain kit, ThermoFisher), and then fixed and permeabilized with the BD Cytofix/Cytoperm reagent. Permeabilized cell samples will be stored at -80 °C before the staining procedure. Antibody staining was performed in a single step following permeabilization. After 30 min of incubation at 4°C, in the dark, cells were washed in BD Perm/Wash buffer then acquired on the ZE5 flow cytometer (Biorad). Analysis was performed on FlowJo v.10 software.
+
+## Experimental infection of macaques with SARS-CoV-2
+
+Four weeks after immunization, all animals were exposed to a total dose of 10<sup>6</sup> pfu of SARS-CoV-2 virus (hCoV-19/France/ lDF0372/2020 strain; GISAID EpiCoV platform under accession number EPI_ISL_406596) via the combination of intranasal and intra-tracheal routes (0.25 mL in each nostril and 4.5 mL in the trachea, i.e. a total of 5 mL; day 0), using atropine (0.04 mg/kg) for pre-medication and ketamine (5 mg/kg) with medetomidine (0.05 mg/kg) for anesthesia. Nasopharyngeal, tracheal and rectal swabs, were collected at 1, 2, 3, 4, 6, 9, 14 and 20 days post exposure (d.p.exp.) while blood was taken at 2, 4, 6, 9, 14 and 20 d.p.exp. Bronchoalveolar lavages (BAL) were performed using 50 mL sterile saline at 3 d.p.exp in order to be close to the peak of viral replication and to be able to observe a difference between the vaccinated and control groups. In our earlier study<sup>5</sup>, we found that at later time-points, viral loads in the BAL were very low or negative. Chest CT was performed at baseline and at 2 and 6 d.p.exp. on anesthetized animals using tiletamine (4 mg/kg) and zolazepam (4 mg/kg). Lesions were scored as we previously described<sup>5</sup>. Blood cell counts, haemoglobin and haematocrit were determined from EDTA blood using a DXH800 analyzer (Beckman Coulter).
+
+## Virus quantification in cynomolgus macaque samples
+
+Upper respiratory (nasopharyngeal and tracheal) and rectal specimens were collected with swabs (Viral Transport Medium, CDC, DSR-052-01). Tracheal swabs were performed by insertion of the swab above the tip of the epiglottis into the upper trachea at approximately 1.5 cm of the epiglottis. All specimens were stored between 2°C and 8°C until analysis by RT-qPCR with a plasmid standard concentration range containing an RdRp gene fragment including the RdRp-IP4 RT-PCR target sequence. The limit of detection was estimated to be 2.67 log<sub>10</sub> copies of SARS-CoV-2 gRNA per mL and the limit of quantification was estimated to be 3.67 log<sub>10</sub> copies per mL. SARS-CoV-2 E gene subgenomic mRNA (sgRNA) levels were assessed by RT-qPCR using primers and probes previously described (Corman et al., 2020; Wölfel et al., 2020): leader-specific primer sgLeadSARSCoV2-F CGATCTCTTGTAGATCTGTTCTC, E-Sarbeco-R primer ATATTGCAGCAGTACGCACACA and E-Sarbeco probe HEX-ACACTAGCCATCCTTACTGCGCTTCG-BHQ1. The protocol describing the procedure for the detection of SARS-CoV-2 is available on the WHO website (https://www.who.int/docs/default-source/coronaviruse/real-time-rt-pcr-assays-for-the-detection-of-sars-cov-2-institut-pasteur-paris.pdf?sfvrsn=3662fcb6_2). The limit of detection was estimated to be 2.87 log<sub>10</sub> copies of SARS-CoV-2 sgRNA per mL and the limit of quantification was estimated to be 3.87 log<sub>10</sub> copies per mL.
+
+## Viral dynamics modeling
+
+For the structure of the model, we started from previously published models<sup>8,9</sup> where we added a compartment for the inoculum to be able to distinguish between the injected virus (V<sub>s</sub>) and the virus produced de novo (V<sub>I</sub> and V<sub>NI</sub>). The model included uninfected target cells (T) that can be infected (I<sub>1</sub>) and produce virus after an eclipse phase (I<sub>2</sub>). The virus generated can be infectious (V<sub>I</sub>) or non infectious (V<sub>NI</sub>). The model can be written as a set of differential equations as follows:
+
+[IMAGE_METHODS_1]
+
+Using the concentration of viral load, measuring V<sub>S</sub>, V<sub>I</sub> and V<sub>NI</sub>, we estimated the viral infectivity (β) and the loss rate of infected cells (δ). The effect of each intervention group (convalescent macaques vaccinated or not and previously uninfected macaques) was tested on the viral infectivity and the loss rate of infected cells. Furthermore, individual variation of β, ρ and δ was estimated through random effects. Maximum likelihood estimation was performed using a stochastic approximation EM algorithm implemented in the software Monolix (www.lixoft.com). The duration of the eclipse phase (1/κ) and the clearance of the virus (c) were estimated by profile likelihood. The production of viral particles by infected cells (ρ) has been fixed to 19,000 in trachea and 36,000 in nasopharynx copies per productively infected cell per day according to previous estimations<sup>5</sup>. The proportion of infectious virus (m) has been fixed to 1/1000 according to previous work<sup>5</sup>. The initial concentration of target cells, that are the epithelial cells expressing the ACE2 receptor, was assumed to be 1.33x10<sup>5</sup> cells/mL in the nasopharynx and 2.25x10<sup>4</sup> cells/mL in trachea (Gonçalves et al., 2020). The initial concentration of the inoculum was assumed to be 2.3x10<sup>9</sup> copies/mL corresponding to 10<sup>6</sup> pfu (Table 1).
+
+[IMAGE_METHODS_2]
+
+## Statistical analysis
+
+Differences between unmatched groups were compared using an unpaired t-test or the Mann-Whitney U test (Graphpad Prism 8.0), and differences between matched groups were compared using a paired t-test or the Wilcoxon signed-rank test (Graphpad Prism 8.0). Viral kinetic parameter was compared using log-rank tests (Graphpad Prism 8.0). Correlation between viral and immune parameter was determined using nonparametric Spearman correlation (Graphpad Prism 8.0).
+
+# References
+
+1. M. Voysey et al., Safety and efficacy of the ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim analysis of four randomised controlled trials in Brazil, South Africa, and the UK. *Lancet (London, England)*, (2020).
+2. K. S. Corbett et al., Evaluation of the mRNA-1273 Vaccine against SARS-CoV-2 in Nonhuman Primates. *N Engl J Med* **383**, 1544-1555 (2020).
+3. F. P. Polack et al., Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine. *N Engl J Med*, (2020).
+4. N. B. Mercado et al., Single-shot Ad26 vaccine protects against SARS-CoV-2 in rhesus macaques. *Nature* **586**, 583-588 (2020).
+5. S. E. Oliver et al., The Advisory Committee on Immunization Practices' Interim Recommendation for Use of Pfizer-BioNTech COVID-19 Vaccine - United States, December 2020. *MMWR Morb Mortal Wkly Rep* **69**, 1922-1924 (2020).
+6. A. T. Widge et al., Durability of Responses after SARS-CoV-2 mRNA-1273 Vaccination. *N Engl J Med*, (2020).
+7. E. J. Anderson et al., Safety and Immunogenicity of SARS-CoV-2 mRNA-1273 Vaccine in Older Adults. *N Engl J Med* **383**, 2427-2438 (2020).
+8. J. T. Schiller, D. R. Lowy, Raising expectations for subunit vaccine. *J Infect Dis* **211**, 1373-1375 (2015).
+9. X. Ma et al., Nanoparticle Vaccines Based on the Receptor Binding Domain (RBD) and Heptad Repeat (HR) of SARS-CoV-2 Elicit Robust Protective Immune Responses. *Immunity* **53**, 1315-1330.e1319 (2020).
+10. P. J. M. Brouwer et al., Two-component spike nanoparticle vaccine protects macaques from SARS-CoV-2 infection. *bioRxiv*, 2020.2011.2007.365726 (2020).
+11. K. McMahan et al., Correlates of protection against SARS-CoV-2 in rhesus macaques. *Nature*, (2020).
+12. W. Yin et al., Functional Specialty of CD40 and Dendritic Cell Surface Lectins for Exogenous Antigen Presentation to CD8(+) and CD4(+) T Cells. *EBioMedicine* **5**, 46-58 (2016).
+13. V. Godot et al., TLR-9 agonist and CD40-targeting vaccination induces HIV-1 envelope-specific B cells with a diversified immunoglobulin repertoire in humanized mice. *PLoS pathogens* **16**, e1009025 (2020).
+14. L. Cheng et al., TLR3 agonist and CD40-targeting vaccination induces immune responses and reduces HIV-1 reservoirs. *J Clin Invest* **128**, 4387-4396 (2018).
+15. A. Bouteau et al., DC Subsets Regulate Humoral Immune Responses by Supporting the Differentiation of Distinct Tfh Cells. *Frontiers in immunology* **10**, 1134 (2019).
+16. G. Zurawski et al., Superiority in Rhesus Macaques of Targeting HIV-1 Env gp140 to CD40 versus LOX-1 in Combination with Replication-Competent NYVAC-KC for Induction of Env-Specific Antibody and T Cell Responses. *Journal of virology* **91**, (2017).
+17. J. Shang et al., Cell entry mechanisms of SARS-CoV-2. *Proc Natl Acad Sci U S A* **117**, 11727-11734 (2020).
+18. A. L. Flamar et al., Targeting concatenated HIV antigens to human CD40 expands a broad repertoire of multifunctional CD4+ and CD8+ T cells. *Aids* **27**, 2041-2051 (2013).
+19. A. L. Flamar et al., Noncovalent assembly of anti-dendritic cell antibodies and antigens for evoking immune responses in vitro and in vivo. *Journal of immunology (Baltimore, Md. : 1950)* **189**, 2645-2655 (2012).
+20. I. Szurgot et al., DNA-launched RNA replicon vaccines induce potent anti-SARS-CoV-2 immune responses in mice. *Scientific Reports*, (2021).
+21. K. Ljungberg, P. Liljeström, Self-replicating alphavirus RNA vaccines. *Expert Rev Vaccines* **14**, 177-194 (2015).
+22. E. Klechevsky et al., Cross-priming CD8+ T cells by targeting antigens to human dendritic cells through DCIR. *Blood* **116**, 1685-1697 (2010).
+23. P. Maisonnasse et al., Hydroxychloroquine use against SARS-CoV-2 infection in non-human primates. *Nature*, (2020).
+24. C. Gaebler et al., Evolution of Antibody Immunity to SARS-CoV-2. *bioRxiv*, (2020).
+25. P. Baccam, C. Beauchemin, C. A. Macken, F. G. Hayden, A. S. Perelson, Kinetics of influenza A virus infection in humans. *Journal of virology* **80**, 7590-7599 (2006).
+26. A. Gonçalves et al., Timing of Antiviral Treatment Initiation is Critical to Reduce SARS-CoV-2 Viral Load. *CPT Pharmacometrics Syst Pharmacol* **9**, 509-514 (2020).
+27. J. I. Cohen, P. D. Burbelo, Reinfection with SARS-CoV-2: Implications for Vaccines. *Clin Infect Dis*, (2020).
+28. J. Seow et al., Longitudinal observation and decline of neutralizing antibody responses in the three months following SARS-CoV-2 infection in humans. *Nat Microbiol* **5**, 1598-1607 (2020).
+29. A. Chandrashekar et al., SARS-CoV-2 infection protects against rechallenge in rhesus macaques. *Science* **369**, 812-817 (2020).
+
+## Methods references
+
+1. Flamar, A. L. et al., Targeting concatenated HIV antigens to human CD40 expands a broad repertoire of multifunctional CD4+ and CD8+ T cells. *Aids* **27**, 2041-2051, doi:10.1097/QAD.0b013e3283624305 (2013).
+2. Li, D. et al., Targeting self- and foreign antigens to dendritic cells via DC-ASGPR generates IL-10-producing suppressive CD4+ T cells. *J Exp Med* **209**, 109-121, doi:10.1084/jem.20110399 (2012).
+3. Zurawski, G. et al., Superiority in Rhesus Macaques of Targeting HIV-1 Env gp140 to CD40 versus LOX-1 in Combination with Replication-Competent NYVAC-KC for Induction of Env-Specific Antibody and T Cell Responses. *Journal of virology* **91**, doi:10.1128/jvi.01596-16 (2017).
+4. Szurgot, I. et al., DNA-launched RNA replicon vaccines induce potent anti-SARS-CoV-2 immune responses in mice. *Scientific Reports* (2021).
+5. Maisonnasse, P. et al., Hydroxychloroquine use against SARS-CoV-2 infection in non-human primates. *Nature*, doi:10.1038/s41586-020-2558-4 (2020).
+6. Fenwick, C. et al., Changes in SARS-CoV-2 Spike versus Nucleoprotein Antibody Responses Impact the Estimates of Infections in Population-Based Seroprevalence Studies. *Journal of virology* **95**, doi:10.1128/jvi.01828-20 (2021).
+7. Johnson, M. et al., Evaluation of a novel multiplexed assay for determining IgG levels and functional activity to SARS-CoV-2. *J Clin Virol* **130**, 104572, doi:10.1016/j.jcv.2020.104572 (2020).
+8. Baccam, P., Beauchemin, C., Macken, C. A., Hayden, F. G. & Perelson, A. S. Kinetics of influenza A virus infection in humans. *Journal of virology* **80**, 7590-7599, doi:10.1128/jvi.01623-05 (2006).
+9. Gonçalves, A. et al., Timing of Antiviral Treatment Initiation is Critical to Reduce SARS-CoV-2 Viral Load. *CPT Pharmacometrics Syst Pharmacol* **9**, 509-514, doi:10.1002/psp4.12543 (2020)
+
+# Supplementary Files
+
+- [ExtendedDataFigures.pdf](https://assets-eu.researchsquare.com/files/rs-244682/v1/84c31b5b43d62df52bb8f08c.pdf)