Bacteria play a crucial role in the production of biofuels, as they are a highly genetically diverse group of microorganisms capable of converting various substrates into fuels such as bioethanol, biobutanol, biodiesel, and others. The workhorses of GMO bacteria include species such as Zymomonas mobilis, Escherichia coli, and Bacillus subtilis.

Zymomonas mobilis is a well-known producer of bioethanol [21], but it has the drawback of not being able to use pentose sugars, which are abundant in lignocellulosic feedstocks. To address this, Wang et al. [29] modified Z. mobilis to allow pentose utilization. A recombinant strain of Z. mobilis was developed that produced ethanol concentrations of 136 g/L in a solution of 295 g/L glucose without the need for any additions of amino acids or vitamins. The enhanced strain harbored multiple gene modules, i.e., the xylA/xlyB/tktA/talB operon for xylose utilization, the metB/yfdZ operon for lysine and methionine biosynthesis, and the thioesterase gene tesA to enhance free fatty acid biosynthesis for increased ethanol tolerance. This strain is now commercially available under the DuPont brand and is used for several biotechnological developments [27].

E. coli and B. subtilis have been chosen as biological models for ethanol production due to their ability to utilize diverse substrates. E. coli was one of the first microorganisms to be successfully modified through metabolic engineering [30]. Ajit et al. [31] transformed E. coli into a strain capable of producing bioethanol by adding genes from Z. mobilis encoding pyruvate decarboxylase and alcohol dehydrogenase. Further modifications led to the creation of a homoethanoleogenic derivative called the KO11 PPAL strain, which had improved expressions of heterologous pyruvate decarboxylase and alcohol dehydrogenase from Z. mobilis.

An example of the metabolic diversion of carbon is the strain of B. subtilis that was modified for bioethanol production through the disruption of the native lactate dehydrogenase gene and the addition of the Z. mobilis pyruvate decarboxylase and alcohol dehydrogenase II genes. This led to increased rates of cell growth and glucose consumption compared to the wildtype [32].

Saccharomyces cerevisiae is a microorganism with probably one of the longest histories of domestication and usage by humanity. With a genetic background that has been widely explored, manipulated, and tested at industrial scales, it is among the most popular biological platforms for producing bioethanol. Its metabolic routes have been modified using diverse genetic engineering tools, with a focus on gene regulatory systems, increasing stress tolerance, and improving transport mechanisms [33]. Although yeasts cannot directly convert some sugars (e.g., xylose) into alcohols, by using different metabolic alternatives, it has been possible to make yeast capable of co-fermenting non-native carbon sources. However, there are still limitations to overcome, such as low xylose fermentation yields compared to glucose and inhibitions due to toxic compounds in the hydrolysates.

For the production of biodiesel, oleaginous yeasts such as Yarrowia lipolytica can accumulate high levels of lipids and consume a wide range of lignocellulosic feedstocks while tolerating high inhibitor concentrations and operational conditions [34]. Qiao et al. [35] genetically modified strains of Y. lipolytica to produce high amounts of fatty acid methyl ester (FAME). They achieved this by creating synthetic pathways that helped funnel glycolytic NADH into NADPH and carbon into acetyl-CoA. The best strain achieved a productivity improvement of 25% over other engineered yeast strains. This improvement was due to the reduction in oxygen requirements caused by decreased NADH oxidation by aerobic respiration. Targeting the lipase-dependent pathway to the subcellular lipid body compartment resulted in a 10-fold higher fatty acid methyl ester (FAME) titer compared to the cytosolic pathway, and a combined simultaneous targeting of lipase pathways to the lipid body, endoplasmic reticulum, and peroxisome gave rise to the highest FAME titer of 1644.8 mg/L [36]. These strategies aimed at targeting metabolic pathways towards lipid bodies are expected to promote subcellular compartment engineering and offer new options for the biosynthesis of other lipid products.

One challenge in microbial oil production is the high production and extraction costs. Ledesma-Amaro et al. [37] proposed the possibility of secreting lipids into the culture broth to uncouple production and biomass processing. Two synthetic approaches were considered, one based on increasing fatty acids produced by enhancing flux through neutral lipid formation through the overexpression of diacylglycerol acyltransferase (DGA2) under the control of the pTEF promoter in the free fatty acids (FFAs) accumulating mutant Δfaa1Δmfe1 (disrupting acyl-CoA and B-oxidation). Simultaneously, TGLA (the intracellular lipase of Y. lipolytica) was overexpressed, resulting in a significant increase in FFA. The other strategy mimicked the bacterial system by a complete depletion of neutral lipid formation and by the expression of endogenous, re-localized heterologous acyl-CoA thioesterases in the same mutant (Δfaa1Δmfe1) to produce free fatty acids. The results showed an increased lipid excretion of up to 85%. Remarkably, the engineered strains secreted a titer of 120% of dry cell weight in coupled fermentation.

Microalgae are a diverse and unique group of photosynthetic microorganisms that have the capability to produce a wide range of high-value commercial products, such as vitamins, antioxidants, omega-3 fatty acids, and immunostimulants [38]. Microalgae feature rapid proliferation and photoautrophic efficiency, converting CO₂ and light into biomass and high-energy molecules such as lipids for biodiesel. Moreover, microalgae do not compete with food crops or agricultural land [39].

The production of biofuels from microalgae is a promising source of sustainable and economically viable environmental energy with low carbon emissions and the potential to replace limited fossil fuels. The composition of cells in microalgal cultures is influenced by environmental factors such as light, temperature, nutrient availability, pH, and salinity. In addition to the direct modification of metabolic pathways, genetic engineering involves the manipulation of enzymatic processes, transport, and photosynthetic functions to improve activities [40].

The transfer of the Arabidopsis thaliana WRINKLED 1 transcription factor to Nannochloropsis salina is an example of recombinant ME. This increased the lipid content by 36.5% and the FAME yield by 64%, making it a good choice for making biofuels in industrial microalgae [41]. Another case is the transformation of Chlorella sp. with the gene hydrogenase (HydA) mutated to avoid O₂ enzymatic inhibition and increase H₂ production. HydA was designed to prevent O₂ from reaching the active site by substituting amino acid residues A105I, V256W, G113I, or V273I close to the gas tunnel, resulting in 7 to 30 times (depending on the tested O₂ concentrations) increases in H₂ production [42].

The targets for ME can be achieved through dedicated bioinformatic annotation, in particular with understudied organisms such as algae. In an interesting study, Sahoo et al. [43] examined the genomes of 26 species of microalgae belonging to various phylogenetic lineages and determined the complete sequence and structure of hundreds of hypothetical proteins coding for enzymes involved in lipid metabolism. This annotation included the development of 3D models with a special emphasis on active-site descriptions employing dynamic simulation methods. These findings have the potential to increase the oil content in microalgae, which is used for biofuel production, primarily biodiesel. Studies demonstrating the potential that microalgae have as a sustainable source of high-value commercial products and biofuels are presented in Table 1. Relevant examples include the organisms utilized, the engineering techniques applied, and the resulting improvements in bioproducts and yield. The successful cases of ME show the promise of further enhancing the metabolic landscape of microalgae to convert lignocellulosic residues into high-value products.

Most of the microbial capacities of interest for biomass conversion have their roots in natural metabolic activities that have been detected in ecological niches. Recently, with the advent of economical sequencing services, modern RNA/DNA-seq methodologies and derived databases can be incorporated into screenings [69]. This would help to overcome the historical bias for microorganisms that can be cultured in laboratory conditions and to incorporate into the biotechnological toolbox those uncultured genotypes that also have mechanisms of practical potential. Enzyme-coding genes, operons, and theoretical microbial genomes can be obtained using the large genomic databases currently available. For example, Naas et al. [70] used the well-studied system of cellulolytic organisms, mammal rumen combined with switchgrass adherence, and applied a metagenomic approach. In this way, the sequence of the genetic determinant is more important than the cellular context from which it originates. Metagenomes were assembled, and with the guidance of experimentally determined proteomes, the pathways of these non-culturable microorganisms were proposed. Extracellular carbohydrate-active enzymes (CAZymes), membrane transporters, multicellular membrane complexes, and biochemical pathways for the catabolic use of all carbohydrate-based components of plant cell walls (cellulose, xylose polysacharides, and mannans) were obtained. Finally, the use of these bioinformatically assembled sequences was illustrated with the recombinant expression (in this case from synthetic genes) of the most expressed cellulase and the characterization of its enzymatic activity, showing the feasibility of recovering functional activities from metagenomic studies and their combination with structural overlap with available registers in the Protein Data Bank.

The diversity of environments where microbes face evolutionary pressure to express active cell wall-degrading proteomes includes compost from agricultural wastes. Composting plant biomass is a common agricultural practice sustained by plant biomass decomposition. These fermentation techniques are ecological niches enriched with cellulolytic enzymes. Meneses et al. [71] built genomic libraries from red rice compost and screened them to find a cellulase. This bioinformatic analysis allowed the discovery of an open reading frame that was cloned and expressed as a recombinant to characterize it. As expected, when this cellulose was tested as an additive against the substrate where it was isolated, it released sugars that were later used as substrates by S. cerevisae to produce ethanol. Although the cell from which the cellulose was obtained is unknown, the logic of using it to lyse the original substrate proved a useful methodology.

Hyperthermophilic cellulolytic bacteria are an interesting subject of study and testing among the diversity of microorganisms that are suitable for the valorization of biomass. For example, Caldicellulosiruptor bescii is capable of producing exo- and endoglucanases, creating an exproteome that sustains growth in complex plant extracts with ethanol production. However, since its genome does not code for extracellular xylosidases, Kim et al. [72] complemented the exoproteome of C. bescii with xylose processing enzymes (xylosidase, endoglucanase, and xylanase) from other thermophilic bacteria and augmented growth and sugar release from model and plant-derived mixtures. The biological deconstruction of biomass is usually carried out with commercial enzyme cocktails. However, engineering the expression of these enzymes from the genome of suitable microorganisms (e.g., thermophiles) would decrease the cost of processing. This recombinant strategy helps to avoid adding costly external enzymes to the fermentation and allows the integral deconstruction of plant biomass into ethanol.

The use of thermostable enzymes in these processes allows for their direct use without the need for costly heat-transfer operations. Takeda et al. [73] applied experimental metagenomic techniques to samples obtained from hot spring sediments. The authors were able to construct sequences and determine the phylogenetics of digitally isolated glycoside hydrolase family members. The use of available protein crystallographic coordinates allowed the prediction of their structures and thermodynamic mechanisms of thermotolerance, as well as their physical production and enzymatic characterization through recombinant techniques.

In the future, the metabolic landscape of genetically modified organisms used in biomass conversion can be expanded by characterizing the molecular mechanisms governing the complex biological relations of natural niches. Kreuzenbeck et al. [74] approached the fungi–termites symbiotic relationship with biochemical, genetic, and analytical tools to analyze four different cellulolytic niches with a high diversity of organic molecules. The authors discovered dozens of chemical moieties related to terpenes of a diverse nature that were products of a rich enzymatic cocktail that could be partially reconstructed and included several cyclases. Characterizing the most frequently expressed sequence, which codes for a terpene cyclase, made it possible to suggest several conversion pathways that lead to terpenes of industrial interest, such as limonene or bisabolene. In addition to being interesting from a biological point of view, the relationship between fungi and insects could be used in the future to make fine chemicals from simple carbon substrates. This enzyme, which yielded multiple products, is a candidate for directed evolution to suppress or enhance substrate preferences or increase the yield of selected compounds.

With the advent of mass sequencing technologies, it is now possible to remotely screen ecological niches. Conspicuous niches to look after are hot springs. Different research groups have taken this prospective approach, and the raw sequences have been deposited in different databases [69]. Reichart et al. [75] bioinformatically explored the microbial diversity of hot springs for the discovery of extremozymes, both in pH and temperature. The authors found representatives of almost every CAZyme family in the dozens of samples screened. All the discovered CAZymes can be physically retrieved by standard PCR reactions or by artificial gene synthesis, a commercial methodology that has seen costs abate recently. A consequence of these types of research is the urgent global conservation of these unique ecosystems, which act as a living library of catalytic activities. Although celluloytic environments showed more diversity than hot springs, the CAZymes found in the latter are more likely to be thermoactive enzymes, rendering several candidates for biomass conversion catalysts. The source of novel CAZymes can also be non-traditional niches, for example, microbial eukaryotes. Chang and Lai [76] bioinformatically explored the family of glycoside hydrolases that encompass many members of biomass depolymerizing enzymes. These CAZymes present in metazoan genomes had representatives from every family of glycoside hydrolases. However, as the metazoan lineage advances toward multicellularity, these representatives decrease. Besides describing the evolutionary stages at which the loss of members of this family occurred and possible horizontal gene transfer events, they were also able to identify crustaceans as a potential source of such enzymes. The potential of these virtual isolates would be in conditions similar to the habitat of these organisms, such as high salinity and tolerance to denaturation.

Computer phylogenetic comparisons are the basis of decades-long research on enzyme and pathway discovery in microorganisms. The discovery of the initial sequence and its characterization are extended to new organisms by nucleotide or amino acid similarities. This knowledge has led to a new application of next-generation sequencing, which allows researchers to reconstruct genomes in a historically unprecedented manner and provides new contexts for discovering new enzymes. Piao et al. [77] highlighted the importance of bacterial genome organization in regulons. They also thought that uncharacterized genes next to genes that were already known to be CAZymes might also work on carbohydrates. The authors observed that a quarter of bacterial open reading frames still lack functional annotations and labeled this uncharacterized molecular information as “genomic dark matter”. After setting up a bioinformatic pipeline to digitally isolate cellulase-like candidates, dozens of predicted CAZymes were expressed in E. coli and tested on Miscanthus grass biomass to confirm their biochemical activity as sugar-releasing enzymes. This interesting research design allows for obtaining insight into nominal biochemical activity, investigating if they are part of operons, as well as their possible roles in their ecological niche. Some of these undiscovered sequences showed little or no activity in the tested conditions; however, interesting candidates for further improvement were obtained. These enzymes with low phylogenetic identity represent previously uncharacterized groups that may allow for the discovery of novel structures and mechanisms.

Sugar-based biomass has received the most attention in biological conversions to chemical products. However, as part of the plant cell wall, lignin is an abundant polymer built from aromatic rings derived from amino acids that can also provide carbon and electrons for cell metabolism. In nature, there are microbes with the genetic capacity to catabolize this complex polymer, which can be enriched and isolated as strains or as metagenomes in both prokaryotes and eukaryotes. That is the case in tropical forests, where the release of tons of lignified matter is continuously mobilized. Therefore, they must possess a microbial biome with the enzymatic ability to perform this catalysis. DeAngelis et al. [78] explored, through metagenomic techniques. the diversity of this niche and found that the transition of populations depends on the lignin load of the samples. Although the trend showed a loss of diversity in lignin-amended samples, the specialized metagenomes can be used in the future to retrieve the specific sequences of the measured enriched activities (e.g., peroxidases, oxidases, and cellobiohydrolases).

Once there is experimental evidence providing sufficient contextual information about a microorganism’s value as a testing chassis, its advancement is enriched by a genomic project. Through recombination or other mutagenesis methods, it is possible to improve strains capable of using lignin derivates, which can help make integral use of plant biomass. Morya et al. [79] isolated microbes from forest soil by challenging them to having several lignin-derivates such as vanillic, galic, and benzoic acid, among others, as the only carbon source. A Burkholderia sp. strain was isolated due to its distinct enhanced growth in these non-conventional substrates, assimilating them into oxalate and malonate. The genomic sequencing of the isolate showed the existence of genes coding for laccases and peroxidases that might become future targets for genetic improvement. The use of molecules representing polyaromatic structural diversity to screen the microorganisms led to three different potential catabolic pathways. Additionally, in this reference, there is a collection of quick testing protocols for activity using spectrophotometric scans.