Volume 6, Issue 1 pp. 3-16
REVIEW
Open Access

Fuelling the future of sustainable sugar fermentation across generations

David Lips

Corresponding Author

David Lips

Spiber Inc., Tsuruoka, Japan

Correspondence

David Lips, Spiber Inc. Tsuruoka, Kakuganji Mizukumi 234-1, 997-0052, Japan.

Email: [email protected]

Search for more papers by this author
First published: 10 December 2021
Citations: 1

Abstract

Biomanufacturing in the form of industrial sugar fermentation is moving beyond pharmaceuticals and biofuels into chemicals, materials, and food ingredients. As the production scale of these increasingly consumer-facing applications expands over the next decades, considerations regarding the environmental impact of the renewable biomass feedstocks used to extract fermentable sugars will become more important. Sugars derived from first-generation biomass in the form of, for example, corn and sugarcane are easily accessible and support high-yield fermentation processes, but are associated with the environmental impacts of industrial agriculture, land use, and competition with other applications in food and feed. Fermentable sugars can also be extracted from second- and third-generation feedstocks in the form of lignocellulose and macroalgae, respectively, potentially overcoming some of these concerns. Doing so, however, comes with various challenges, including the need for more extensive pretreatment processes and the fermentation of mixed and unconventional sugars. In this review, we provide a broad overview of these three generations of biomass feedstocks, outlining their challenges and prospects for fuelling the industrial fermentation industry throughout the 21st century.

1 INTRODUCTION

In the last several years, products made via the industrial fermentation of sugars from renewable biomass feedstocks by genetically engineered microbes have moved beyond their traditional role in biofuels and medicine to consumer-facing applications in food and materials (Figure 1) [1-3]. This growth will likely continue as both synthetic biology capacities and consumer demand for fossil-free and animal-free products expand, such that over the next decades, products made via fermentation and engineered biology will increasingly permeate society [4].

Details are in the caption following the image

Overview of the sugar-based biomanufacturing process, showing how sugars from a variety of sources may be fermented by genetically engineered microorganisms to create products for different industries. Examples of the product categories shown include bioethanol, compostable plastics derived from, for example, 1–4 butanediol, textiles made from recombinant proteins, recombinant heme as an ingredient of plant-based meats, and recombinant insulin. Heme and textile images were reproduced with permission from Impossible Foods and Spiber Inc., respectively

Currently, a substantial part of the industrial fermentation industry that makes these products is fuelled by sugars from first-generation feedstocks, that is, the edible parts of sucrose- and starch-containing biomass crops like sugarcane and maize (corn) [1, 5-7]. Products made from these feedstocks often display more favourable environmental impact profiles compared to fossil-based and animal-based alternatives [8], but concerns regarding the impacts of the intensive industrial agriculture practices required to grow these feedstocks at a large scale have been voiced in the context of corn bioethanol and may be echoed in other parts of the industry as production scales increase [9, 10].

Related to these concerns is the fact that the expansion of the bioeconomy does not occur in isolation, but in the context of a growing global population [11] whose per capita demand for crops has historically been a function of per capita real income—largely due to higher meat consumption and the associated need for animal feed [12]. Although biotechnology could potentially offset some of the pressure of this demand by improving crop yields or enabling plant-based and cell-based meat alternatives [13, 14], any business-as-usual growth scenario could see significant increases in biomass demand for food and feed in addition to that of a larger bioeconomy.

To put the potential growth of the bioeconomy and its impact in a practical perspective, consider the size of annual global textile fibre production, estimated at around 100 million tonnes per year in 2017, and projected to grow to 120 million tonnes in 2025 [15]. Several biomanufacturers are producing bio-based fibres via the fermentation of sugars derived from first-generation feedstocks (e.g. protein-based fibres [1] and bio-based nylon [16]). Assuming a hypothetical 10% yield of biomanufactured fibre on corn starch-derived glucose, scaling up to 10% of 2017's global fibre production would require around 22 million acres of corn, which is 1.7 times the number of harvested corn acres in Iowa [17], the world's most productive corn region [18], and on the same order of corn acres that were used for bioethanol production in the United States in 2018/2019 (∼28 million acres) [19]. Although the fermentation-based textile fibre industry is far from reaching this scale and improvements in yield would reduce biomass requirements relative to a given amount of product, various other industries are being targeted by fermentation as well. Together, their demand for biomass may eventually start adding up to magnitudes on the order of the above and beyond.

For biomanufacturers relying on first-generation biomass, this means that it becomes increasingly important to evaluate the environmental impacts of their current feedstock strategy, to mitigate negative externalities as much as possible, and to eventually develop the capacity to use sugars from alternative sources if warranted by future trends in biomass supply and demand. Regarding the latter, alternative sugar sources that do not compete with food resources exist in various forms, out of which lignocellulosic biomass (second generation) and brown macroalgae (third generation) have been among the most prominent and promising [20, 21]. Despite the earlier-stage nature of these feedstock and mixed results with early lignocellulosic biofuel facilities [22], progress in the processing of these feedstocks and their fermentation into target compounds has been ongoing [23-25]. In this perspective, we provide a broad overview of these three generations of biomass feedstocks, outlining the challenges and prospects for sustainably and efficiently using them as the bioeconomy expands.

2 FIRST GENERATION: FOOD CROPS

2.1 Overview

Sugars derived from first-generation feedstocks are the current mainstay of the industrial fermentation industry, providing the input for a wide variety of chemicals, ingredients, and materials. The main sources of these sugars are sucrose from sugarcane and sugar beets, as well as glucose derived from corn starch [26] (Figure 2a). The processing of these crops for sugar extraction has been well-established and takes place in specialized sugar mills. Together, these mills and related trading companies form a global market from which biomanufacturers can easily procure fermentable sugars (Figure 2b) [27]. Altogether, these sugars are easy to extract, are delivered at high concentrations and purity, and can be fermented by industrial microbes to support high productivity and yields for a wide variety of target products.

Details are in the caption following the image

First-generation biomass feedstocks. (a) Three common forms of first-generation biomass from which fermentable sugars are extracted, showing from left to right: maize (corn), sugarcane, and sugar beets. (b) Simplified process showing fermentation of sugars extracted from corn and sugarcane. The main sugar constituents are shown with their molecular structure (corn; glucose, sugarcane; sucrose). (c) Sustainability practices that can be implemented by biomanufacturers (together with their suppliers) to improve the environmental footprint of their first-generation biomass feedstock supply chains. Images in (c) showing cover cropping, prairie strips, and Bonsucro's certification mark have been reprinted with permission from Practical Farmers of Iowa, Omar de Kok-Mercado (Iowa State University), and Bonsucro, respectively

2.2 Challenges

Although the processing of first-generation feedstocks into sugars is well-established, sustainability challenges exist with regard to the industrial agriculture practices associated with their cultivation. Both corn and sugarcane, for example, are industrial biomass crops grown at a large scale in various regions worldwide. In line with intensive agriculture practices, their cultivation typically involves lands clearing that can harm biodiversity [28], soil practices that can lead to erosion [29], and fertilizer application that can result in eutrophication as has previously been seen in, for example, the Gulf of Mexico and Great Barrier Reef along Australia's North West coast [30, 31].

Beyond the direct environmental impacts of intensive agriculture, high corn demand from the bioethanol industry has previously been associated with both higher food prices [32] and indirect land use change [33]. Although these indirect effects are difficult to quantify and estimates for their associated magnitude vary widely [34], the impacts involved are likely non-negligible and can contribute significantly to a bio-based product's environmental impact if included in Life Cycle Analysis (LCA) studies [35].

2.3 Outlook

Given their accessibility and compatibility with high-yield fermentation processes, sugars derived from first-generation biomass crops will likely continue to be one of the key feedstocks in industrial fermentation for the foreseeable future. It will, therefore, be important for biomanufacturers to take a proactive stance and mitigate the associated negative environmental impacts of growing this biomass as much as possible. Practically, this translates to the application of sustainable agriculture practices to the cultivation of first-generation biomass crops.

Potential sustainable agriculture practices are as plentiful as they are diverse [36], but the implementation of a relatively small set of practices can likely mitigate a significant portion of the negative impacts related to biodiversity, eutrophication, and soil degradation commonly associated with industrial agriculture (Figure 2c). In the US Midwest's Corn Belt, for example, the integration of strips of native prairie species amid corn and soybean crops (replacing 10% of cropland) increased biodiversity and ecosystem services, reduced total water runoff by 37%, and resulted in the retention of 20 times more soil, though yields decreased by the amount of area taken out of crop production, which could increase overall land use [37].

In places where eutrophication and soil degradation levels are high, the inclusion of cover crops between the harvesting and planting of subsequent main crops has been shown to provide various ecosystem services, reduce nutrient runoff, and help protect against soil erosion [38, 39]. In Australia, various sugarcane programs are being implemented together with growers to reduce the aforementioned eutrophication in the Great Barrier Reef, including the application of soil analysis together with soil-specific nutrient management practices [40]. Further sustainability improvements for first-generation biomass crops may be achieved by enhancing the soil microbiome with nitrogen-fixing microbes, which can reduce the needed amount of synthetic nitrogen fertilizers [41], and which have recently become commercially available for corn farmers [42].

In addition to working directly with suppliers to implement sustainable agriculture practices, biomanufacturers can also procure sugars based on voluntary sustainability standards (VSS). These are essentially industry guidelines that cover a range of social and environmental impacts and are administered by NGOs, who grant farmers and biomass operators with certification and assess compliance via independent verification systems. Bonsucro, for example, is one of the fastest growing VSS providers with 5.8% of the global sugarcane area certified under its system as of June 2021 [43]. Research has shown that global compliance with all of their standards would reduce irrigation water use by 65%, eutrophication potential by 34%, and GHG emissions from cultivation by 51%, with most of the benefit coming from targeting the 10% of global cane production area furthest from complying with Bonsucro's criteria [44]. Simply by procuring feedstock from suppliers that are certified by credible VSS providers, biomanufacturers could, therefore, improve the performance of their operations across a range of environmental and socio-economic impact metrics. That said, care should be given to the specific coverage of a VSS scheme and its implementation details on a case-by-case basis, such that additional measures can be implemented as needed.

3 SECOND GENERATION: LIGNOCELLULOSE

3.1 Overview

Lignocellulose makes up most of a plant's dry matter and is a complex structure composed of cellulose encapsulated by hemicellulose and lignin. With an estimated 181.5 billion tonnes produced annually, it is the most abundant renewable source of biomass on earth [45]. Lignocellulose can come in a variety of forms, including specialized energy crops like miscanthus that can be grown on marginal land, wood and its derivatives, and the non-edible residues of food crops like bagasse, corn stover, and wheat straw [46] (Figure 3a). The nature of these forms—neither edible nor grown on land that could be easily used for food crops—relate to lignocellulose's main environmental appeal compared to first-generation feedstocks. A large number of LCA case studies have confirmed that using lignocellulosic feedstocks can lead to less land usage and reduce environmental impact across most categories due to the uncoupling with the intensive agriculture practices applied to first-generation biomass cultivation [47, 48].

Details are in the caption following the image

Second-generation biomass feedstocks. (a) Three representative forms of lignocellulosic biomass from which fermentable sugars can be extracted, showing from left to right: corn stover, wood chips, and miscanthus. (c) Simplified process showing the conversion of lignocellulose into fermentable sugars and fermentation. The main sugar constituents are shown with their molecular structures

The three major components of lignocellulose are cellulose (30%–50%), hemicellulose (15%–30%), and lignin (10%–25%) [49]. Cellulose has a linear structure and is a chain of glucose molecules linked by β-1–4 glycosidic bonds. Due to its crystalline form, it is recalcitrant to chemical hydrolysis. Hemicellulose is composed of pentose sugars (mostly xylose) and hexose sugars (mostly glucose). This results in a relatively amorphous structure that is easier to depolymerize. Lignin is an aromatic polymer of lignols that provides rigidity to the plant structure.

Because of its complex structure, the sugars locked up in the cellulosic and hemicellulosic parts of lignocellulosic biomass have to be extracted before they can be used as substrates for fermentation (Figure 3b). This extraction process typically starts with a physical, chemical, and/or biological pretreatment process aimed to reduce the crystallinity of the cellulose to facilitate enzymatic hydrolysis, remove or degrade the lignin, and hydrolyse the hemicellulose. This can then be followed by enzymatic hydrolysis to break down the cellulose into glucose units, orchestrated most commonly through cellulase enzyme mixtures produced by fungi [50]. The final product is typically a fermentable mixture of pentose and hexose sugars, mixed with a variety of other—potentially inhibitory—compounds depending on the nature of the pretreatment process.

3.2 Challenges

Over the years, a variety of different biological, chemical and physical methods for pretreatment have been developed and applied to lignocellulosic biomass. The choice of pretreatment method depends on the type and composition of the lignocellulosic biomass. Among the more commonly used methods are hydrothermal treatment, dilute acid pretreatment, and concentrated acid hydrolysis [20]. Each of these technologies, however, comes with various drawbacks—including the generation of inhibitor compounds at high temperatures, the need for expensive corrosion-resistant metals in plant construction, and the application of expensive chemicals and the subsequent need for their recovery [20].

The challenges surrounding the enzyme hydrolysis step that follows pretreatment has mostly been described in terms of costs [51, 52]. An evaluation from 2016, for example, showed that cellulase enzymes used for lignocellulose hydrolysis can account for up to 48% of the minimum selling price of bioethanol, based on a price of around $23/kg as seen in industrial cellulose markets at the time [53]. Although enzymes can potentially be produced through consolidated bioprocessing in which the microbial cell factory also secretes enzymes that hydrolyse lignocellulose, ideally enzyme costs would also be brought down through improvements in enzymatic efficiencies resulting in lower total enzyme requirements.

The breakdown of both hemicellulose and cellulose results in an output that is a mixture of hexose and pentose sugars, rather than the hexose-only sugars of first-generation feedstocks. Although technically it is possible to separate these sugar streams, this adds processing costs and means fewer sugars are available for fermentation. For this reason, efficient co-fermentation of both the hexose and pentose sugars is a more desirable lignocellulose utilization scenario [54]. For most industrial microbes, however, the uptake and metabolism of xylose are inhibited by glucose, which makes the efficient and complete utilization of lignocellulosic sugar mixtures a challenge [55]. Escherichia coli, for example, does have native transporters and metabolic pathways for xylose, but carbon catabolite repression (CCR) prevents the efficient utilization of glucose-xylose mixtures [56].

Finally, the presence of minor reaction products such as furfural, phenolic compounds and weak organic acids that can arise from the application of heat and chemicals during pretreatment can hinder efficient fermentation [57]. Although additional processing can be implemented to reduce inhibitor compound levels, the inclusion of extra steps increases processing complexity and costs [20].

3.3 Outlook

The ideal pretreatment process would use low cost or easily recovered and recycled chemicals, limit sugar losses prior to fermentation, minimize the formation of inhibitors, have low energy consumption, enable high solids loading to increase sugar concentrations, and come with low waste treatment costs. Although no pretreatment process is currently available that meets all of these conditions, efforts to improve existing methods are still underway. The use of ionic liquids, for example, has long been considered promising because it can efficiently dissolve all lignocellulose components and results in fewer degradation products [23]. However, it currently still faces difficulties in both sugar and ionic liquid recovery and the inhibition of subsequently applied enzymes, which requires further process improvements [58, 59]. Similarly, the process of hydrothermal treatment—theoretically ideal in its simplicity and lack of required chemicals—would benefit from progress improvements to overcome the large quantities of water needed that result in dilute sugar streams and high energy costs [60]. Such improvements, however, would likely require significant breakthroughs in process design and equipment, and may not be commercially viable for all feedstock types.

For enzyme performance, general improvements in efficiency, robustness, and cost (closely related due to the impact of robustness and efficiency on the amount of enzyme needed) are desirable to expand the range of products that could be made from lignocellulose in a commercially feasible fashion. The discovery and development of more efficient enzymes is an active research area, with recent discoveries of new enzymes [25, 61, 62] offering promising avenues for improvement that will need to be tested in a range of enzyme mixtures.

In the absence of a pretreatment process that delivers highly pure sugar streams with no inhibitors, biomanufacturers will have to bioengineer their own solutions to the challenges inherent to current pretreatment. Industrial microbes, for example, can be engineered to display greater tolerance to the inhibitors that arise during pretreatment. There is precedent in the literature for this, with reports of E. coli and yeast strains that were engineered to be tolerant to furfural, 5-hydroxymethylfurfural, and acetate [63-65]. If inhibitors are, indeed, present in the fermentation broth, such traits will have to be engineered into existing producer strains while maintaining high yields and productivity.

The other bioengineering challenge relates to the efficient utilization of mixed pentose and hexose sugar streams. This can be accomplished either through the adaptation of existing production hosts to perform well on mixed sugars or by adapting microbes with the native capacity to do so (e.g. specific clostridia strains [66]) to make a target product at a high yield. Although both approaches may be feasible, there is still a general lack of genetic tools for non-conventional strains [67], and industrial microbes like E. coli have already been engineered to overcome CCR and do well on glucose-xylose mixtures [68]. This suggests that adaptation of existing industrial microbes is likely the fastest path to lignocellulosic success if high yields and productivity can be maintained. Notably, undigested sugars, impurities, and inhibitors may also require adaptation of downstream purification and wastewater treatment, depending on the contents of the lignocellulosic sugar mixture [69].

From the biomanufacturer's perspective, there is of course still the matter of obtaining the lignocellulosic sugar inputs in the first place. Compared to the well-established global markets for sugar derived from first-generation feedstocks, the market for lignocellulosic sugars is small and fragmented. As a result, biofuel manufacturers have often resorted to establishing their own lignocellulose processing facilities in the past [70-72], adding processing complexity, capital burden, and resulting in siloed innovation. It would be helpful for biomanufacturers if there were more specialized commercial providers that could aggregate lignocellulosic biomass and leverage economies of scale to drive down cost and improve sugar quality. Such providers could benefit from innovations in, for example, lignin valorization to create additional revenue streams [73] while providing biomanufacturers with the sugars needed for commercial fermentation. Although several of such providers have offered their services in the past, in most cases, the scale and scope were too small for commercial production [74]. That said, the recent emergence of a Thailand-based demonstration facility that processes sugarcane using new membrane-based separation technologies and offers the resulting cellulosic sugars to biomanufacturers is a promising example of continued innovation and emerging opportunities in this space [75]. Sustained demand from the biomanufacturing industry will be required for these and other lignocellulosic sugar suppliers to operate successfully.

4 THIRD GENERATION: MACROALGAE

4.1 Overview

Macroalgae, commonly referred to as seaweeds, are considered a promising feedstock candidate for the industrial fermentation industry (Figure 4a). In contrast to first and second-generation biomass, the production of macroalgae takes place in coastal waters and does therefore not involve arable land, freshwater, or fertilizers [76]. Their photosynthetic efficiencies (PE) and biomass densities are also substantially higher than terrestrial crops (6%–8% vs. 1.8%–2.2% for PE and 565 ton/ha vs. 180 ton/ha for biomass density) [77]. Macroalgae are furthermore high in carbohydrates while being low in lignin, thereby enabling enzymatic hydrolysis with less intensive or no pretreatment [78].

Details are in the caption following the image

Third-generation biomass feedstocks in the form of macroalgae. (a) A member of the Laminaria genus of brown seaweed in the order Laminariales (kelp). (b) A coastal macroalgae farm in Xiapu, Fujian province, China. (c) Simplified process showing the conversion of brown macroalgae into fermentable sugars. The main sugar constituents are shown with their molecular structures

The large-scale cultivation of macroalgae is practised in several countries, yielding over 16 million metric tonnes per year in 2019 [76]. Macroalgae are categorized as red, green or brown algae, each of which has a particular composition of carbohydrates that could be converted into fermentable sugars. Red and brown macroalgae constitute 76% of the global commercial production of macroalgae, with the majority of production being based on species from several genera of red macroalgae (Eucheuma and Gracilaria) and brown macroalgae (Laminaria, Undaria, and Porphyra) [76]. Although natural populations of macroalgae have been collected for food purposes for centuries, at present about 90% of macroalgae harvested worldwide is derived from cultivated sources [21]. In China, Japan, and Chile—the largest producers of macroalgae—cultivation is mostly practised in nearshore coast farms (Figure 4b), although it can also be practised in offshore farms and land-based ponds. Harvesting is performed either manually or mechanically with the help of boats or ships for the operation of, for example, mowing with rotating blades or dredging with cutters, depending on the type of macroalgae and its cultivation method [21]. Since brown macroalgae are typically larger and have higher areal productivities than green or red macroalgae [21], they have received the greatest attention as a promising feedstock alternative for the fermentation industry and will be used as a representative example in the rest of this section.

The carbohydrate content of macroalgae constitutes 40%–77% of dry matter (DM) [79]. The most abundant of these carbohydrates in brown macroalgae are alginate, mannitol, and laminarin. Mannitol is a sugar alcohol formed by the reduction of mannose and can make up to 30% of the dry weight of brown macroalgae [80], where it functions as a major storage carbohydrate and osmoprotectant or local osmolyte [81]. Laminarin is a β-1,3 glucan chain and one of the main storage carbohydrates in brown macroalgae. Levels vary widely depending on the species but can reach up to 35% of total dry weight in the Laminariaceae family of brown macroalgae (many genera of which are known as kelp) [82]. Alginate is found in the cell walls of brown macroalgae, where it contributes to the organism's flexibility [83]. It composes 30%–60% of the total sugars in brown macroalgae [84] and is a linear block copolymer of two uronic acids, b-D-mannuronate and a-L-guluronate, arranged in varying sequences. Before fermentable sugars can be extracted from macroalgae, the biomass will have to be processed to remove debris, crushed into particle sizes more amenable to processing, and dried to reduce water content, increase shelf life, and reduce transportation costs [21] (Figure 4c).

4.2 Challenges

Among the main carbohydrates in brown macroalgae, mannitol is easily accessible and fermented without pretreatment. Laminarin and alginate, however, are polysaccharides that require hydrolysis into monomers before they can be used as a fermentation feedstock. Most macroalgae-to-bioethanol studies to date use a combination of dilute acid pretreatment before enzymatic hydrolysis in order to increase the reaction surface area of the macroalgal biomass available for enzymatic breakdown and maximize sugar yield [85-90]. Including dilute acid treatment, however, comes with the drawbacks of inhibitor formation and the requirement for a neutralization step prior to fermentation [85]. A study using enzymatic hydrolysis alone without pretreatment has been reported as well [91], though the differences in macroalgae and the used enzymes do not allow for a direct comparison.

Enzymatic hydrolysis of macroalgal polysaccharides has mostly been performed using cellulase enzyme cocktails like Celluclasat [88] and Viscozyme [92]. These were originally developed for hydrolysis of lignocellulosic biomass but can perform well on polysaccharides like alginate and laminarin when combined with pretreatment methods such as acid and hot water treatment [89, 92]. The reported sugar conversion efficiencies are in the range of 50%–70% of the total carbohydrate content [79].

After hydrolysis and the generation of monomers, the obtained mixture is to be fermented into a target product by industrial microbes. For mannitol and the glucose subunits derived from laminarin, their fermentation is relatively straightforward and feasible with most industrial microbes. Alginate, however, has been less commonly used as a fermentation feedstock and will require metabolic engineering if current production hosts are to co-ferment its monomers. In microbes that can natively metabolize alginate, alginate lyases catalyze the depolymerization of alginate into unsaturated monomers that spontaneously rearrange into 4-deoxy-L-erythro-5-hexoseulose uronic acid (DEHU) [93-95]. DEHU is then enzymatically reduced into 2-keto-3-deoxygluconate (KDG), a common metabolite that feeds into the Entner-Doudoroff (ED) pathway where it eventually yields pyruvate and glyceraldehyde 3-phosphate [96]. Since industrial microbes will not have the membrane transporters for DEHU nor the reductase to convert DEHU to KDG, these will have to be engineered into the production host of choice if alginate is to be utilized [97]. Alternatively, natively alginate-metabolizing microbes can be engineered to produce a target compound. The latter feat has previously been demonstrated with Sphingomonas sp. for bioethanol production [98], but such strains typically still lack both robustness under industrial fermentation conditions as well as tools for genetic manipulation, making further optimization difficult [99].

From an environmental impact perspective, an LCA study has shown that the impacts related to the usage of brown macroalgae in the form Laminar sp. as a fermentation feedstock can compare favourably to first- and second-generation feedstock systems, but only if renewable energy is used for the energy intensive drying step [35]. Expanding the coastal areas in which brown macroalgae are farmed may also come with unknown environmental trade-offs related to biodiversity [35], as well as climate change mitigation benefits [100], but it is currently difficult to evaluate these accurately.

4.3 Outlook

Given the high proportion of alginate in the carbohydrate fraction of brown macroalgae, the utilization of alginate in the fermentation process is important for maximizing product yields on macroalgal biomass. Although most studies are limited to the fermentation of mannitol and glucose, a small handful of literature reports have shown that existing industrial microbes can be engineered to co-ferment alginate as well. Wargacki et al., for example, engineered E. coli to express alginate lyases to break down alginate into monomers, creating a consolidated bioprocess for alginate breakdown [101]. Genes from Vibrio splendidus for both alginate monomer transport and the NADH-dependent reduction of DEHU to KDH catalyzed by DehR were also added. Their E. coli platform enabled bioethanol production with a yield equivalent to ∼80% of the maximum theoretical yield from macroalgae's sugar composition [100]. This result was similar to the yield achieved by Enquist-Newman et al., who engineered Saccharomyces cerevisiae to co-ferment alginate monomers together with mannitol and glucose to produce ethanol [102]. A later study emphasized the importance of achieving redox balance to optimize ethanol production, predicting ideal rates of mannitol dehydrogenase and DehR in the range 1–1.6 to support optimal growth rates and ethanol yields in S. cerevisiae [103]. To expand the range of products that can be made from brown macroalgae feedstocks beyond bioethanol, similar traits of alginate monomer uptake and DEH reduction to KDG will have to be integrated into other production strains and optimized for high-yield production while ensuring careful redox balance.

Before fermentation takes place though, an enzymatic hydrolysis step with or without chemical pretreatment will be required to depolymerize both alginate and laminarate. As was demonstrated by Wargacki et al., alginate could potentially be degraded by alginate lyases expressed by the industrial host in a consolidated bioprocess [101]. The additional expression of alginate lyase enzymes on the yield of the final target product is not clear, however, and will likely require significant optimization or the application of microbial co-cultures to become economically viable. So far, most literature studies demonstrating macroalgae sugar fermentation utilize external enzymes in the form cellulase enzyme mixtures developed for lignocellulose breakdown [88, 92]. More efficient hydrolysis can likely be achieved with enzyme mixtures specifically developed for third-generation macroalgae resources, using, for example, a combination of alginate lyases and laminarinases targeted to the specific carbohydrate composition of the macroalgae source. Whether or not the trade-offs of including a chemical pretreatment step like dilute acid treatment are worth the improvements in sugar yield compared to their extra cost and/or inhibitor formation should be carefully evaluated.

As with the situation with lignocellulosic feedstocks, how to procure fermentable sugars derived from macroalgae will be a key question for the biomanufacturer. Unlike second-generation feedstocks, however, no pilot facilities for the fermentation of macroalgae-derived sugars have been demonstrated so far, and no supply chain exists to support the procurement of macroalgae-derived sugars in an industrial fermentation context [21]. It is also an open question whether macroalgae-producing countries can expand nearshore coastal production to the scale needed for a mature bioeconomy without incurring environmental impact trade-offs and affecting macroalgae-based food production. Besides coastal production, other options for cultivating macroalgae include offshore farms and land-based ponds [21], and the merits of each form of cultivation should be carefully evaluated when establishing early pilot facilities for the fermentation industry. Considerations to take into account include factors such as the scale of farms required to meet production needs, the availability and cost of space and nutrients, projected environmental impacts, and competition with other uses.

Lastly, the drying step to reduce macroalgae's high water content will be the largest environmental impact factor due to its energy-intensive nature [35]. Drying can potentially be done via solar means in the open air, but the feasibility of doing so at scale is unknown and depends on geography and climate. Ideally, the energy required for intensive macroalgae drying at scale would, therefore, come from renewable resources to reduce the environmental impact associated with fossil energy usage. From an environmental impact perspective, this would likely make macroalgae feedstocks the best-in-class option compared to sugars derived from either corn starch or lignocellulosic sugars derived from corn stover [35]. However, the availability of renewable energy is also contingent on location, emphasizing the importance of considering the geography of future production facilities. Importantly, all of the above considerations will need to be tested empirically through pilot facilities that develop and demonstrate cost-effective sugar extraction processes and subsequent fermentation into value-added products. This outlook, as well as the key challenges and prospects for the other biomass approaches discussed in this review, is summarized in Table 1.

TABLE 1. Overview of key challenges and prospects of the three biomass feedstock generations described in the text
First-generation biomass Second-generation biomass Third-generation biomass
Commercial compatibility Glucose and sucrose from first-generation biomass crops support productive fermentation processes with most industrial microbes Mixed sugars streams from lignocellulose require specialized microbial cell factories and processes Mixed sugars streams and alginate content from macroalgae will require specialized microbial cell factories/fermentation
Global markets and infrastructure for sugar procurement are in place and easily accessed Small number of examples of commercial success, mostly with in-house biomass processing Commercial feasibility not demonstrated; no pilot facilities for macroalgae-to-sugar processing have been operational at the time of writing
Sustainability considerations Industrial agriculture impacts Reduced land use and no competition with food No terrestrial land use, no fertilizer and freshwater requirements
Direct and indirect land use If lignocellulose was originally used as a fuel source (bagasse) or soil conditioner (corn stover), fossil fuels or fertilizer consumption may increase to compensate Energy requirement of macroalgae drying
Competition with food and feed applications Unclear impacts on marine ecosystems if scaled up
Key challenges Sustainable expansion to meet growing bioeconomy demand Recalcitrance and required pre-treatment processes Early-stage nature
Cost-effective production while reducing the formation of inhibitor compounds Lack up pilot facilities to demonstrate the feasibility and enable optimization towards process development goals
Proposed solutions Sourcing based on voluntary sustainability standards Process optimization of selected approaches Funding for pilot scale development to enable process development and optimization
Implementation of sustainable agriculture practices Lignin recovery and valorization to support commercial lignocellulosic sugar suppliers Strain optimization for alginate co-fermentation and high-yield production of target compound on mixed macroalgal sugar streams
Strain optimization for high-yield production of target compound on mixed lignocellulosic sugar streams

5 DISCUSSION

Although biomass-derived sugar fermentation may be the dominant paradigm in industrial biomanufacturing at present, it is not the only one. Specifically, single carbon (C1) substrates—including methane, methanol, CO2, and formate—are increasingly acknowledged as promising feedstocks for the biomanufacturing industry, largely due to their natural abundance, availability as industrial by-products, and capacity to be generated electrochemically with excess electricity [104-106]. As with the xylose or alginate content of second- and third-generation biomass feedstocks, biomanufacturers that wish to use C1 feedstocks face the choice of either developing natural C1-utilizing organisms for industrial production or engineering the effective use of C1 compounds into existing industrial microbes. Several companies are taking the former approach [107, 108], but doing so for a wider range of products will require extensive development of new hosts, an effort that is currently constrained by a lack of industrial familiarity and synthetic biology tools applicable to such hosts. As an alternative approach, recent advances in genome engineering and laboratory evolution have enabled the creation of synthetic C1-utilizing microbes, illustrated by the first eukaryote (Pichia pastoris) and prokaryote (E. coli) able to synthetically fix CO2 in 2019 [109, 110], as well as the first E. coli cells able to grow on methanol and formate in 2020 [111, 112]. At present, the growth rate of these synthetic C1-utilizing microorganisms is still well below their growth rate on sugar; for example, the doubling time of wild-type E. coli on sugar feedstocks is typically around 20 min, whereas methylotrophic E. coli growing on methanol and autotrophic E. coli growing on CO2 as the sole carbon sources have doubling times of 8.5 h and 18 ± 4 h, respectively [106]. Although there is thus much work to be done to make C1-based production for a wider range of products a reality, throughout the 21st century, these efforts could create opportunities to increasingly shift biomanufacturing feedstocks from biomass-derived sugars to single carbons.

In the meantime, the extent to which the biomanufacturing industry might transition from first-generation biomass feedstocks to second- and third-generation ones depends on both basic economic and technical considerations as well as global trends in biomass supply and demand. As an illustration of the latter, the Nova Institute evaluated a number of scenarios for global biomass supply and demand projections until 2050 [113]. In their ‘bio-based’ demand scenario, biomass demand for the bioeconomy would rise sharply (including an increase in the share of biomass needed to cover the demand of the chemical and plastic industry from 10% to 40% by 2050), and would only be met by supply if both extensive intensification as well as an expansion of agricultural cropland area with 360 million additional ha would take place. In a context of accelerating climate change, biodiversity preservation goals, and the sensitivity of the public when it comes to using biomass for non-food purposes, such growth—if realized—may stimulate a transition to second- and third-generation feedstocks to minimize these trade-offs.

Among the impacts of first-generation biomass utilization for bio-based products, much has been published on land use change in particular. Associated impact estimates vary widely, ranging from large increases in indirect land usage and related biodiversity and carbon losses as a result of diverting corn to ethanol production [33, 114, 115], to a study showing no evidence of ‘lands converting to agriculture because of biofuel’ [116]. Since 1) it seems a priori unlikely that a large increase in demand would not result in a significant additional land requirement to meet this demand unless compensated by yield increases and/or demand decreases elsewhere, and 2) future biomass demand for food and feed is set to grow in accordance with increases in population and prosperity, we suspect the actual land use change impact of a larger bio-economy will be non-negligible. In this context, the yield of a target product on sugar will be a key metric for biomanufacturers to optimize towards its theoretical maximum. A doubling of a compound's yield on sugar, for example, translates to a halving of the amount of feedstock required to produce a similar amount of product. Given how yields on sugar for bio-based products are initially typically much lower than the theoretical maximum yield [117], there is likely much to be gained here by applying the tools of synthetic biology to iteratively improve on this important metric.

It should be noted that although second- and third-generation feedstocks may largely avoid the land change issue, they do come with their own inherent trade-offs. Agricultural waste residues like sugarcane bagasse or corn stover, for example, currently fulfil respective roles as fuel in sugar mills and as fertilizer [118, 119]. Using this biomass instead to generate fermentable sugars may, therefore, result in an increase of synthetic fertilizer application and fossil fuel usage, offsetting lignocellulose's environmental benefits for various impact metrics [35]. For third-generation feedstock processing, the trade-offs are currently less known and will depend on cultivation location and harvesting methods, but may affect marine ecosystems in various ways.

Beyond environmental concerns, the practical feasibility of using any particular feedstock is largely an economic matter. Although the procurement of second- and third-generation biomass is relatively cheap compared to purchasing first-generation sugars ― $24 to $121 per ton for lignocellulosic biomass depending upon the crop, yield, region, and method of analysis [120] and $112 per ton for brown macroalgae in the form of Laminaria sp. [21] versus $440 per ton for sucrose [121]―the efforts required for feedstock aggregation, processing into fermentable sugars, and reductions in yield due to the often more dilute and impure nature of the resulting sugars, all add up to a final cost that will likely be significantly higher than first-generation sugars costs.

The practical reality of this has been shown before in the biofuels industry, in which numerous commercial efforts at lignocellulosic biofuels were attempted, but many have subsequently been shelved due to economic intractabilities [122]. Ethanol, however, averaged a price of less than $2 per gallon from 2016 to 2021 [123], which means that feedstock costs can quickly make up a substantial fraction of the final biofuel price. The fact that some efforts do in fact appear successful [124] suggests that the economics can be feasible with the right feedstock aggregation strategy, process design, and fermentation process. For higher-value products like speciality chemicals, food ingredients, and fermentation-derived alternatives to animal materials, the feedstock price component will be a smaller fraction of the price, suggesting more flexibility and robustness in terms of the maximum affordable feedstock price. Techno-economic evaluations can quantify such economic considerations for various feedstock choices and do so in the context of environmental impact trade-offs when combined with an LCA [125]. Such evaluations will likely be an important first step for biomanufacturers making a stepwise transition from first- to second- and third-generation biomass feedstocks when the incentives to do so emerge as a result of, for example, consumer and investor demand.

Ultimately, sustainable biomanufacturing involves a complex set of trade-offs between different social and environmental impact metrics and their magnitude. At present, the trade-offs inherent in making bio-based products via the fermentation of first-generation feedstocks are often (though not always) favourable compared to fossil- or animal-based alternatives, especially from a climate change perspective [8]. As the bio-economy expands throughout the next decades though, trade-offs related to land usage and the conflicting use-cases for first-generation biomass may become more dominant, potentially spurring at least a partial transition to the second or third-generation feedstocks outlined here. Practically speaking, we, therefore, believe it not only prudent to put into place sustainable agriculture practices for first-generation biomass cultivation, but also to make strategic investments in the technologies needed to efficiently procure and ferment sugars from alternative biomass sources such as the ones described here.

ACKNOWLEDGEMENT

No specific funding has been received with regard to the creation of this manuscript.

    CONFLICT OF INTEREST

    The author is an employee of Spiber Inc., a biomanufacturing company based in Tsuruoka, Japan.

    DATA AVAILABILITY STATEMENT

    Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.