The sustainable production of petroleum-based compounds such as fine chemicals is a key challenge in the modern society. Biotechnology provides tools for the high-yield production of these compounds. Generally, production is performed through microbial, plant or animal cell cultures, and this strategy is especially suitable for sterile production (Twyman et al., 2003). Moreover, compounds with a high demand but low value are also targets of biotechnological approaches. Here, cell cultures are too expensive to maintain, prone to contamination and inflexible to rapidly changing demands (Twyman et al., 2003). Transgenic plants grown in the field can serve as cost-effective alternatives. Several such products are already available in the market, as reviewed in (Ricroch et al., 2022). Field-grown tobacco exhibits a high yield of bacterial cellulase in the chloroplasts (Schmidt et al., 2019). Moreover, plants can synthesize polyhydroxyalkanoates with properties similar to polyacrylates or other synthetic polymers (such as polyhydroxybutyrate (PHB) (Anderson and Dawes, 1990; Poirier, 2002; Börnke and Broer, 2010) or the polydisperse biopolymer cyanophycin granular polypeptide (CGP) (Borzi, 1887). The transition from fossil to plant-based production can decrease the energy consumption. Such production processes exhibit a decreased cost, and the adverse influences of the use of fossil raw materials can be alleviated (Mooibroek et al., 2007). They can also facilitate the transition toward sustainable food systems, as discussed at the UN Food System Summit in 2021 (Trigo et al., 2021).

However, in all of these cases, the return on investment occurs solely via the market, depending on the value of the transgene–encoded compound and changes in the production costs. The manufacturing of valuable compounds as coproducts for plant-made industrials may be a cost-effective and environmentally friendly alternative. In this case, production does not depend on additional arable land or energy. The compound can be purified from the leftovers after isolation of the primary product. The biopolymer CGP has been produced as such a coproduct for starch potato (Hühns et al., 2008; Schmidt et al., 2017). CGP is a polydisperse (25–125 kDa) polymer consisting of mainly L-aspartic acid (Asp) and L-arginine (Arg), synthesized by a cyanophycin synthetase. Many cyanobacteria and several nonphotosynthetic bacteria produce the enzyme encoded by a cphA gene (Krehenbrink et al., 2002; Ziegler et al., 2002).

CGP, Asp and Arg have been applied as fine chemicals (Mooibroek et al., 2007). In particular, Arg-Asp dipeptides can substitute free Arg and Asp in several food and feed additives since their bioavailability is superior to that of free amino acids. Various commercial products of these forms are already available in the market (Sallam and Steinbüchel, 2010). In addition, various potential applications in the nonfood/feed industry have been suggested. By removing Arg from CGP, poly-asp can be produced, which has applications ranging from water-softening or detergent applications to applications in the paper, building material, petroleum, cosmetics, or leather industries and dispersant-based applications (Schwamborn, 1996; Schwamborn, 1998; Elbahloul et al., 2005; Mooibroek et al., 2007; Börnke and Broer, 2010). Moreover, Asp can be converted to acrylonitrile, which is widely applied in acrylic fibers, nitrile rubbers, and carbon fibers.

Arg can be converted to 1,4-butanediamine and urea. At present, 1,4-butanediamine is used to produce nylon-4,6 (Mooibroek et al., 2007). The material is primarily used in applications that require high thermal stability and tensile strength, such as engineering materials and heavy machinery. The remaining urea can also be applied in industries.

The biopolymers extracted from CGP are highly valued. As reviewed in Sallam and Steinbüchel, (2010), dipeptide L-alanyl-L-glutamine, which is used as an infusion to treat patients with immunosuppression, postoperative complications or malnutrition, is offered by Mediatech Inc. (Manassas, Virginia, United States) for €4,732/kg. Salam and Steinbüchel assumed that CGP dipeptides are at least as effective as these products in several fields of application. Hence, the researchers predicted that CGP dipeptides can reach a market price of over €3,000/kg. Accordingly, plant-based CGP has considerable market potential, especially in Argentina and other Latin American countries with a substantial amount of tobacco production, and can facilitate the development of a sustainable bioeconomy (Trigo et al., 2021). Nevertheless, this product must compete with CGP derived from cyanobacteria (Lippi et al., 2018), or synthesized from Escherichia coli or yeast (Sallam et al., 2011).which all represents active areas of research Certain commercial success has been achieved with synthesis based on Escherichia coli (Kwiatos and Steinbüchel, 2021).

The expression of the gene cphATe from the cyanobacterium Thermosynechococcus elongatus BP-1 under the control of the cauliflower mosaic virus (CaMV) 35S promoter and terminator in transgenic tobacco Petit Havana SR1 plants resulted in up to 1.0% of CGP dry weight (dw) (Neumann et al., 2005). Phenotypic damage caused by the production of CGP in the cytosol was eliminated via the plastid location of the synthetase. This modification led to CGP contents of up to 1.7 and 6.8% dw in the T0 and T2 descendants, respectively, without detectable stress symptoms (Hühns et al., 2008). Moreover, CGP-producing potato plants were analyzed in 6 years of field trials (Schmidt et al., 2017). Without additional fertilization or other inputs, the plants produced up to 4 μg/mg dw (0.4% dw) of CGP in the tubers without a significant reduction in the starch content. However, the tuber size was significantly decreased. Although the number of tubers increased, the overall yield was less compared to near isogenic control (NIC) (Schmidt et al., 2017).

The production of CGP in the commercial tobacco cultivars Badischer Geudertheimer (BG) and Virginia Golta (VG) in the greenhouse was noted to be a highly effective process. In the F1 hybrids (max. 9.4% CGP dw) and T0 transformants (max. 8.8% CGP dw), the CGP content was significantly higher than that achieved through Petit Havana SR1 transformants. In the greenhouse, no significant yield penalty was observed (Nausch et al., 2016). In addition, the cytosolic expression of a cyanophycinase, encoded by the gene cphEPa, in CGP-producing tobacco led to the formation of stable Arg-Asp dipeptides after cell disruption (Nausch and Broer, 2017; Ponndorf et al., 2017; Nausch et al., 2020). Hence, commercial tobacco varieties may be a suitable production platform for CGP.

Moreover, tobacco can be used to produce CGP as a coproduct of a number of other products in addition to extracted nicotine. Recently, tobacco has received renewed attention as a bioenergy crop (Gowtham Rajan et al., 2021). Tobacco plants can provide substantial amounts of oil stored in the seeds (Grisan et al., 2016) and yield fermentable sugars for bioethanol or biogas (Schneider et al., 2021). Tobacco oil has been successfully converted to biodiesel that can satisfy the European standard (Usta et al., 2011). The biomass remaining after extracting CGP from the plant can be applied in the biobased economy.

The possibility of storing the harvest is key to cost-effective processing because the harvest can be processed over a prolonged period in a facility with a smaller maximum capacity and lower investment cost. CGP-containing tobacco leaves can be stored in a dried form or as silage without loss of CGP (Nausch et al., 2016). Isolation of plant-made CGP on a small scale has been reported for tobacco and potato, achieving purities of 58–90%, respectively (Neubauer et al., 2012). This method is based on acid extraction on lyophilized starting material. Freeze-drying is an efficient technique to expand plant cells for extraction; however, the process is highly energy intensive to be applied at a large scale (Guldhe et al., 2014). In addition, the common extraction protocol applies 0.1 M HCl, which is corrosive and requires specially adapted machinery. In this study, we evaluate the possibility of macerating dried and ensiled tobacco leaf material.at pH 5, omitting low pH conditions in the expensive equipment, followed by extraction at pH 1, which can be performed in a simple tank. The findings can facilitate the processing of farm-scale quantities of CGP-containing tobacco.

The approval to cultivate GM crops in Europe is difficult to obtain (Sparrow et al., 2013; Smart et al., 2015). Many companies attempt to set bases in other countries, such as Argentina or the United States, where the approval process to cultivate GM crops is considered to be transparent, less complicated, and less expensive than that in Europe (Whelan and Lema, 2015; Eriksson et al., 2019). Nonfood/feed products derived from transgenic plants do not require any additional GMO labeling in the EU (Regulation (EC) No. 1829/2003, (Wesseler and Kalaitzandonakes, 2019)). Nevertheless, the import and processing of CGP from transgenic tobacco require approval. These procedures and costs can be avoided by processing CGP into the final product in the home country and exporting the final product to the EU, similar to the export of textiles derived from GM cotton. Nevertheless, a market for GM-free labeled textiles has emerged for cotton. Similar voluntary labeling has been observed for a number of food and nonfood products in the EU, with substantial implications for the marketability of GM food/feed and nonfood/feed products owing to consumer responses (Castellari et al., 2018). Hence, any product isolated from genetically modified plants (GMP) to be offered in a consumer market requires consumer acceptance. Therefore, we conducted choice experiments for two hypothetical applications mimicking consumers’ shopping and consumption patterns: food wrapping and cosmetics.

To clarify whether the production of CGP can be a sustainable alternative, we conducted field trials using several independent transgenic events from the tobacco cultivar BG in Argentina, tested the isolation from storable materials and evaluated the production and potential commercial value of CGP products along with the consumer acceptance for these products in four European countries.

Using the data pertaining to T0 and T1, we can calculate the expected CGP production per hectare. Since we selected elite events, the CGP content of the best performer (531) with a maximum CGP production of 4 g per plant was considered. The planting density of the trial was 22,000 plants per hectare, which is a standard value for high biomass tobacco production (Berbeć and Matyka, 2020). Hence, an extrapolated CGP production of 88 kg CGP per hectare was obtained. Since the extraction and purification yield and CGP prices involved uncertainties, we performed a Monte Carlo simulation. The input data for the simulation are presented in Table 2. Sallam and Steinbüchel (2010) reported that the industrial production cost of CGP from E. coli fermentation was between 250 and 500 EUR per kg CGP. We considered these values as references for the minimum CGP sales prices as crude material. Thus, these values were set as the min and max values of the CGP price distributions (P). The distribution results are summarized in Table 3, with the mean CGP production per hectare of 60 kg and a CGP revenue mean of 23.065 USD per hectare. Supplementary Figure S5 shows the revenue distribution.

Moreover, Sallam and Steinbüchel (2010) predicted that a CGP dipeptide product would reach a market price of over 3,000 EUR per kg. If 3,300 USD per kg is considered the fine chemical CGP application price, a revenue of 198,000 USD per hectare can be expected in the fine chemical market for a CGP production per hectare of 60 kg. These calculated revenues show the margins for cultivation, isolation, purification and formulation costs. In other words, investments in extracting and processing CGP from tobacco plants for crude and fine chemical products can yield positive returns per hectare.

In 2018, biopolymer production reached 7.5 million tonnes (2% of the petrochemical polymer volume), and the compound annual growth rate of the biopolymer production is expected to be 4% until 2023 (Chinthapalli et al., 2019). The Coca Cola Company introduced the PlantBottle™, which contains 30% bio-PET, and 35 billion ¹ plant bottle packages were distributed between 2009 and 2015. Danone GmbH switched to a polylactic acid (PLA) version of the dairy cup to decrease the thickness of the cup wall for its Activia yogurt brand in Germany and Switzerland by 2011 (Schut, 2016). The car company Renault saved 0.40 EUR per part by switching to a biobased high heat ABS in the dashboard of Clio line (Roma, 2016). Farmers can lower their labor costs by switching to biodegradable plastic mulch film (van den Oever et al., 2017). These examples show that industries are seeking new biopolymers for different applications, and CGP is a promising candidate.

The sustainable production of polymers is a relevant goal at present. Plants grown in the field provide an excellent platform for important compounds, especially when they are coproducts of traditional plant compounds such as oil, sugar or starch. Tobacco seems to be especially suited as a production platform for the following reasons: First, tobacco plants can be easily modified to produce specific compounds (Horsch et al., 1985), and second, the decline in smoking globally has imposed pressure on tobacco farmers (Jha and Chaloupka, 2000). Therefore, CGP production in leaves can provide an alternative income source for tobacco farmers. Nevertheless, it remains to be shown whether CGP production in tobacco is truly sustainable from ecological, economic and socioeconomic viewpoints. Several of these aspects were addressed in this paper.

One of the main factors influencing the extent of sustainability in production is the CGP yield/hectare. The average biomass of the CGP-producing plants in the field was not significantly decreased compared to that of the NIC. This finding is in agreement with (Schmidt et al., 2019), in which the production of a bacterial cellulase at 20% of the total soluble protein did not lead to a yield penalty. In contrast, CGP-producing potatoes were influenced in terms of the tuber size and yield in several field trials, even with considerably lower CGP contents (Schmidt et al., 2017). Hence, tobacco may be assumed to be appropriate for producing novel compounds and enabling higher production levels for transgene-encoded proteins compared to potato. This assumption was supported by the data obtained for the production of PHB in plants. Although the amount of PHB in the chloroplast of transgenic tobacco (0.4 mg/g dw) was lower than that in Arabidopsis thaliana (132 mg/g dw), it was considerably lower in potato (0,09 mg/g dw) (Bohmert et al., 2002).

The decrease in the seed yield observed in the greenhouse indicates that the production of CGP is an additional burden that, although insignificant for the leaf yield in optimal circumstances, may lead to decreased stress resilience of the transgenic plants. This aspect is also indicated by the decrease in the CGP content in the T1 generation in the field and greenhouse and might be caused by a selection of low CGP-producing cells to form seeds. This result is supported by the fact that the CGP production level was similar for all events in the bulk samples regardless of whether the T0 parent was a high or low producer (Figure 4D). Even the ranking of the expression levels between the six events was different among generations and growth conditions (field and greenhouse), indicating that the integration locus of the transgene is not the main limiting factor.

CGP production in the field was as high as that in the greenhouse in terms of the CGP percentage in the dry weight. Since tobacco plants produce significantly more biomass in the field, the CGP yield/plant was considerably higher in the field. Further field trials with seeds from homozygous best performers in the field must be performed to examine whether the CGP production can be increased. In addition, Pergamino (Argentina) is not a typical tobacco cultivation area. In conditions, for instance, those in Salta (Northern Argentina), in which tobacco is grown commercially, the stress may be decreased, and a higher biomass and decreased yield penalty may be achieved.

Other important factors for success are storage and extraction at a large scale. To date, CGP has only been extracted from lyophilized leaf material. The reported experiments show that CGP can be extracted from dried leaf and ensiled tobacco leaves. This observation proves that the storage of harvest is a feasible option. In this study, a blender was used to simulate the milling of dried leaves and maceration of silage. Since cell destruction, which requires the use of a refiner with steel components, can occur at pH = 5, corrosion of the steel must be drastically alleviated. The separation of CGP from the remaining plant material, which must occur at pH = 1, can occur in acid-resistant plastic containers. The difference in the CGP yield between dried leaf (55%) and silage extraction (97%) was considerable. This result may be attributed to the available CGP for extraction. Ensiling keeps the leaves wet, and microbial processes result in cell wall degradation. Moreover, ensiling decreases the pH through lactic acid production and increases the accessibility of cell components for extraction (Rooke and Hatfield, 2003). In contrast, CGP in dried leaves is less available for extraction, resulting in the lower extraction yield observed in this study. The availability of CGP in dried leaves can be enhanced by soaking dried leaves prior to extraction, milling the dried leaf to a powder form or increasing the extraction duration. The modifications introduced to the laboratory isolation protocol can also decrease the cost and facilitate the upscaling of this process.

The potential market for CGP-derived products remains to be validated. In addition to its potential for producing several different products, plant-made CGP is biobased and biodegradable and is thus more sustainable than most petroleum-based compounds. The advantages of biobased, biodegradable compounds have already been recognized by several companies.

Nevertheless, several technical and economic challenges remain in terms of decreasing the cultivation cost and efficiently upscaling the extraction while maintaining high extraction yields. At present, the extraction costs and yield at the hectare scale are unknown. This information can only be generated by establishing a pilot plant that can operate at the commercial scale. Such establishment requires upfront investment and political will to take risks. Farmers must modify their tobacco production strategies and may generate lower revenues for a certain period. Business models that moderate risk can help incentivize farmers and other parties to invest (Carraresi and Bröring, 2021). Nevertheless, in relation to the investments made in the EU for establishing biorefineries, the amount of investment required will be rather moderate and the risk will be relatively low (Clomburg et al., 2017). Policymakers can support development not only by supporting investments via subsidies but also by providing a stimulating policy environment (Smith et al., 2021). Cultivating transgenic tobacco plants and extracting and processing the CGP in Argentina appears to be a more cost-efficient strategy than that in which the tobacco is cultivated in Europe or Argentina and the crude CGP is exported to the EU, as indicated by recent research on the costs for approvals (Wesseler, et al., 2022 in press).

In addition, the decrease in the seed yield in the greenhouse may also occur in the field. This decrease may reduce the revenue of the primary product oil and influence the final revenue from the plants. This aspect must be analyzed in further field trials.

According to consumer choice experiments, on average, consumers reject petroleum-based products and favor natural or GM-based alternatives. These results indicate that consumer responses to GMO-derived products are context-sensitive. Environmental pollution has emerged as a notable concern of European citizens (Eurobarometer 19.1, 2019), and fossil fuel-based products and technologies are key causal factors. In situations in which consumers attempt to balance fossil fuel-based products and other options, GM-based alternatives gain acceptance. This observation might be suggestive of acceptance for products including CGP and its derivatives in Europe, which is important even if the product is not labeled.

This first field trial with CGP-producing tobacco showed that CGP production, storage of the leaf material and CGP extraction can be realized in the field on a larger scale without yield penalty. Further trials must be performed to examine whether the seed yield decreases in the field as in the greenhouse, and whether this decrease influences the usage of CGP as a coproduct. A purified CGP yield of 35–79 kg/ha can be obtained, which may be sufficient for commercial production; nevertheless, this yield depends on the final production and isolation costs. In addition, the CGP yield may change with the environment, planting date and year. Sallam and Steinbüchel mentioned the presence of a small market for CGP produced in microorganisms, and it remains to be seen whether possible new applications will further expand the market. Nevertheless, owing the high consumer acceptance for CGP products in even food-related applications, such as wrapping paper, CGP may emerge as a desirable raw material for companies.