The significance of the work lies in what it replaces. Virtually all industrial fermentation that yields nitrogen-rich products, from amino acids to nucleotides, depends on ammonium salts or urea as the nitrogen feedstock, and those inputs trace back to Haber–Bosch ammonia synthesis. That process, which combines atmospheric nitrogen with hydrogen under extreme pressures and temperatures, consumes an estimated one to two percent of global energy output and generates substantial carbon dioxide emissions as a byproduct of hydrogen production from natural gas. Biological nitrogen fixation, carried out by the nitrogenase enzyme complex in certain bacteria and archaea, performs the same chemical transformation at ambient temperature and pressure using ATP and electrons, offering a potentially far gentler route to usable nitrogen.

Azotobacter vinelandii has long served as the model organism for aerobic nitrogen fixation, a biologically awkward combination since oxygen both damages nitrogenase and competes for the electrons the enzyme needs. The bacterium survives this paradox through extraordinarily high respiratory rates that consume oxygen before it can reach the enzyme. Previous efforts had engineered A. vinelandii to excrete ammonium, the natural next step after nitrogen fixation, since the fixed nitrogen normally feeds the cell’s own biosynthesis. But the researchers behind the new study found that studies on actually using that excreted ammonium as a nitrogen source for other microbes remained limited, and that ammonium production levels were too low to be industrially interesting.

To push yields higher, the team took aim at NifA, the transcriptional activator that switches on the entire nif regulon encoding the nitrogen fixation machinery. They constructed A. vinelandii strains by integrating a nifA overexpression cassette, driven by the strong tac promoter, into the algU locus of the chromosome. This chromosomal integration strategy matters for stability: rather than relying on a plasmid that could be lost during cultivation, the engineered construct is inherited by every daughter cell, ensuring the nitrogen fixation program stays dialed up throughout a fermentation run. The resulting strain stably produced ammonium at a concentration of 1 gram per liter in simple flask cultures, a level the researchers describe as a solid baseline for the platform.

Flask cultures, however, are a proving ground rather than a production environment. The team then scaled the process into a jar fermenter, where they could control oxygen transfer, pH and feeding with far greater precision. By combining fed-batch cultivation, in which fresh carbon source is supplied incrementally to avoid depletion or overflow metabolism, with antifoam treatment to keep the aerated broth from foaming over and lactic acid addition to manage the culture’s chemistry, they raised ammonium production to 1.5 grams per liter. Each of these process interventions addresses a practical bottleneck: fed-batch keeps the energy supply matched to the nitrogenase’s enormous ATP appetite, antifoam protects oxygen transfer and prevents contamination pathways, and acid addition stabilizes the pH as ammonium accumulates and shifts the broth’s acid-base balance.

With a reliable ammonium source in hand, the researchers turned to the second half of the partnership. Corynebacterium glutamicum is arguably the most successful amino acid production organism in industrial biotechnology, responsible for the bulk of the world’s several-million-ton annual L-glutamate output, the flavor-enhancing component of monosodium glutamate. In the co-culture scheme, the engineered A. vinelandii functions as a living nitrogen fertilizer, continuously fixing atmospheric nitrogen gas and releasing ammonium into the shared medium, while C. glutamicum assimilates that ammonium and channels it through its existing metabolic machinery into L-glutamate. The division of labor elegantly sidesteps the need to purify or concentrate the intermediate: the product of one microbe is the substrate of the other, delivered in situ.

The results demonstrated the concept convincingly. Co-cultivation of the ammonium-producing A. vinelandii strain with C. glutamicum enabled the production of 2 grams per liter of L-glutamate from nitrogen gas. While that titer remains well below the tens of grams per liter achieved in conventional glutamate fermentations fed with commercial ammonium, the demonstration establishes a complete biological pipeline from atmospheric dinitrogen to a finished amino acid in a single vessel. The researchers frame the strategy as a contribution to the development of environment-friendly fermentation processes for producing various nitrogen-containing compounds from nitrogen gas, suggesting the platform could extend well beyond glutamate to other amino acids, nucleotides and nitrogenous chemicals.

The engineering choices embedded in the study reveal a careful reading of nitrogenase regulation. NifA sits atop a hierarchy of control mechanisms that bacteria use to avoid wasting energy on nitrogen fixation when fixed nitrogen is already available, a regulatory logic that normally shuts the system down precisely when engineers want it running. By overexpressing NifA from a constitutive tac promoter, the team effectively overrides the ammonium-sensing feedback that would otherwise silence the nif genes as product accumulates. Placing the cassette at the algU locus, which governs stress responses in A. vinelandii, reflects a deliberate choice of a neutral genomic landing site that disrupts native function minimally while granting stable, high-level expression of the activator.

Scaling challenges remain before such co-cultures could challenge conventional plants. Nitrogenase is an enzyme of notorious fragility and metabolic cost, demanding roughly sixteen ATP per molecule of nitrogen reduced, and maintaining two microbial populations with different physiological optima in one fermenter requires balancing oxygen availability, carbon source preference and growth rates. The fed-batch jar fermenter results, with their combination of antifoam and lactic acid management, hint at the kind of process engineering refinement that will determine whether titers can climb toward commercial relevance. The involvement of Kikkoman Corporation, a company with deep roots in fermentation technology, alongside academic groups at The University of Tokyo’s Department of Biotechnology and Collaborative Research Institute for Innovative Microbiology, suggests industrial interest in closing that gap. Several of the authors have filed patent applications on the work, underscoring its perceived commercial potential.

For now, the study stands as a proof of concept with a compelling narrative: a flavor compound that seasons much of the world’s food, assembled in part from the air itself, by two bacteria cooperating in a fermenter. If the platform’s titers can be improved through further strain and process optimization, nitrogen-fixing co-cultures could offer fermentation industries a route to decouple amino acid production from synthetic fertilizer inputs, trimming both energy demand and carbon emissions. The researchers position their work as a step toward fermentation processes that draw their nitrogen directly from the atmosphere, converting a century-old industrial dependency into a biological partnership.

Subject of Research: L-glutamate production from nitrogen gas via co-culture of engineered Azotobacter vinelandii and Corynebacterium glutamicum

Article Title: L-Glutamate production from nitrogen gas by co-culturing Corynebacterium glutamicum with Azotobacter vinelandii

Article References: Ito, Y., Yoshidome, D., Araki, Y., Ito, K., Hidaka, M., Kosono, S., & Nishiyama, M. (2026). L-Glutamate production from nitrogen gas by co-culturing Corynebacterium glutamicum with Azotobacter vinelandii. Applied Microbiology and Biotechnology. https://doi.org/10.1007/s00253-026-14031-5

Image Credits: AI Generated

DOI: 10.1007/s00253-026-14031-5

Keywords: Azotobacter vinelandii, Corynebacterium glutamicum, nitrogen fixation, L-glutamate, co-culture, nifA overexpression, ammonium production, Haber-Bosch, metabolic engineering, fed-batch fermentation, nitrogenase, sustainable biotechnology

The central problem these platforms address is the tumor microenvironment, the dynamic and immunosuppressive niche that governs cancer progression, immune evasion, and therapeutic resistance. Tumors are not merely masses of malignant cells; they are complex ecosystems containing cancer stem cells, stromal cells such as fibroblasts and endothelial cells, angiogenic vessels, extracellular matrix, signaling molecules, and a diverse cast of immune populations including macrophages, dendritic cells, neutrophils, natural killer cells, B cells, and T cells. Within this milieu, regulatory T cells, myeloid-derived suppressor cells, and elevated checkpoint molecules conspire to blunt anti-tumor immunity. The review frames tumor–immune dynamics through the cancer immunoediting framework of elimination, equilibrium, and escape, emphasizing that immune evasion during the escape phase is what allows clinically detectable malignancy to emerge and that the microenvironment is an active driver of tumor evolution and therapeutic resistance rather than a passive backdrop.

Immunosuppression within the tumor microenvironment operates through multiple mechanistic layers. An immunosuppressive cytokine milieu dominated by TGF-β, IL-10, and IL-6 suppresses T cell activation while promoting the expansion of myeloid-derived suppressor cells and M2-polarized macrophages. Chemokine gradients, including CCL22 and the CXCR4/CXCL12 axis, actively recruit suppressive cell populations to the tumor site while excluding cytotoxic effector cells, creating spatially organized immunosuppressive niches that shield tumor cells from immune destruction. Tumor-associated macrophages, among the most abundant and plastic immune populations in the microenvironment, are typically driven toward a pro-tumorigenic M2 phenotype through IL-4, IL-13, and IL-10 signaling, converting a potentially anti-tumor force into an active promoter of progression and therapy resistance. Tumors are further classified as inflamed or hot, immune-excluded, or immune-desert or cold, with the latter two categories being largely refractory to immunotherapy, a reality that underscores why models preserving spatial and stromal complexity are so valuable.

At the heart of immune evasion lies the exploitation of checkpoint pathways. PD-L1 on tumor cells engages PD-1 on T cells, inducing a state of exhaustion characterized by reduced proliferation, cytokine production, and cytotoxicity, while CTLA-4 dampens T cell priming by competing with CD28 for B7 ligands. Emerging checkpoints including LAG-3, TIM-3, and TIGIT further contribute to intratumoral T cell exhaustion and represent targets for next-generation immunotherapy. Immune checkpoint inhibitors have demonstrated efficacy across a remarkable range of malignancies, including non-small cell lung cancer, urothelial bladder cancer, head and neck squamous cell carcinoma, breast cancer, cutaneous squamous cell cancer, melanoma, renal cell cancer, and Hodgkin’s lymphoma. Yet predicting which patients will benefit remains difficult, which is precisely where co-culture platforms that preserve viable immune cells alongside living tumor tissue promise to change clinical practice.

The review traces the evolution of cancer models from flat two-dimensional monolayers, which advanced early understanding of tumor biology but fail to capture complex tumor–microenvironment interactions, to three-dimensional systems that more accurately recapitulate in vivo tumor structure and behavior. Gene expression profiling of three-dimensional multicellular tumor spheroids has revealed upregulation of hypoxia-responsive genes and downregulation of cell cycle-related genes compared to two-dimensional models, and enhanced mevalonate pathway activity has been observed in quiescent spheroid cells, underscoring the context-dependent nature of anticancer responses in 3D. Spheroids, typically 200 to 500 micrometers in diameter, form through integrin- and cadherin-mediated self-assembly and can be generated by pellet culture, hanging drop, liquid overlay, or spinner techniques, each with distinct trade-offs in scalability, monitoring, and hypoxic core formation. When co-cultured with immune cells such as PBMCs, spheroids enable modeling of tumor–immune interactions and intratumoral heterogeneity, making them a valuable preclinical research tool.

Organoids, often described as mini-organs, go further by self-organizing to recapitulate native tissue architecture and function through lineage commitment and spatial cell sorting guided by extracellular matrix and culture medium cues. Patient-derived tumor organoids have now been generated across colorectal, pancreatic, breast, lung, and brain cancers, among others, retaining the mutational profiles, histopathology, and cellular diversity of the source tumor. Matrigel-based systems remain the most widely used due to accessibility and standardized workflows, despite batch-to-batch variability, while bioengineered synthetic matrices offer tunable stiffness and improved reproducibility at higher cost. Advanced technologies are pushing the field further: microfluidic organ-on-a-chip systems enable perfused, vascularized organoids modeling fluid flow and immune infiltration; air–liquid interface culture preserves epithelial, stromal, and immune components with improved oxygenation; and microwell arrays, droplet encapsulation, and acoustic aggregation enable high-throughput, size-controlled production, though typically limited to short-term culture.

The translational centerpiece of the review is the PDTO–PBMC co-culture system, which the authors describe as enabling reconstitution of autologous immune responses in an antigen-agnostic manner. The landmark study by Dijkstra and colleagues established that co-culturing peripheral blood lymphocytes with tumor organoids expands CD8-positive T cells that kill tumor organoids but spare healthy tissue-derived organoids, with killing efficiency of 20 to 80 percent depending on tumor type, and confirmed antigen-specificity through HLA-blocking experiments. The review is careful to define what constitutes genuine functional immune reconstitution rather than mere physical co-localization: HLA-dependent tumor-selective killing, CD137 upregulation as a marker of tumor-reactive T cell activation, interferon-gamma secretion, and granzyme B-mediated cytotoxicity must all be demonstrated. Checkpoint blockade responsiveness has been validated in immune-enhanced organoid platforms, with anti-PD-1 treatment increasing CD3-positive and CD8-positive T cell recruitment, and air–liquid interface models showing organoid responses to PD-1/PD-L1 blockade correlating with clinical outcomes in approximately 85 percent of cases.

Beyond T cells, the platforms extend across the immune repertoire. Cancer-associated fibroblasts co-cultured with organoids enhance tumor growth, promote epithelial–mesenchymal transition, and confer therapy resistance across colorectal, pancreatic, hepatocellular, and esophageal cancers, while endothelial co-culture illuminates tumor-induced angiogenesis and vascular niche-dependent drug resistance. Macrophage co-cultures reveal that sirtuin-1 promotes M2 polarization and suppresses CD8-positive T cell activity in colorectal cancer, and that the CCL5–Sp1–AREG axis mediates tumor–macrophage crosstalk in pancreatic cancer. Dendritic cell co-cultures expose tolerogenic shifts in tumor microenvironments, and natural killer cell studies in breast and pancreatic cancer models have demonstrated both therapeutic potential and tumor-induced immune impairment. CAR T cell evaluation has been particularly transformative: patient-derived bladder cancer organoids have been validated as reliable preclinical platforms for CAR T testing, neuroblastoma organoids have modeled CAR T infiltration and antigen loss, and glioblastoma organoids now serve as real-time avatars for assessing responses to clinical CAR T cell therapy.

The review does not shy away from sobering limitations. Most published work remains at the proof-of-concept stage, with prospective clinical validation data still sparse and no defined response thresholds for what magnitude of in vitro cytotoxicity reliably predicts clinical benefit. Organoid establishment success rates vary dramatically, from 22 percent rising to 75 percent with optimized protocols in metastatic colorectal cancer, 58 percent in pancreatic cancer, and as low as 17 percent for conventional lung cancer protocols versus over 90 percent with optimized free-floating platforms. PBMC viability after cryopreservation, delays exceeding 24 hours between blood collection and processing, and inter-patient immune repertoire variation all confound outcomes. Turnaround from biopsy to functional readout typically spans three to six weeks, though accelerated platforms have demonstrated feasibility within 7 to 14 days. Clonal evolution poses a further threat, as culture-adapted subclones may displace clinically relevant immune-evasive populations, and therapy-induced changes can alter neoantigen repertoires and checkpoint expression. The authors call for the newly proposed Minimum Information about Organoid Research reporting standard, reference organoid lines, external quality control programs, and a structured evidentiary roadmap toward companion diagnostic status under FDA and EU IVDR frameworks.

Looking forward, the convergence of microfluidics, spatial transcriptomics, single-cell multi-omics, and artificial intelligence promises to expand both the biological fidelity and translational utility of these platforms. Three-dimensional bioprinting allows spatially controlled deposition of tumor, immune, and stromal components that better recapitulate immune exclusion zones and stromal barriers, while machine learning applied to high-dimensional co-culture datasets holds promise for predicting patient-specific immunotherapy responses. The authors conclude that clinical validation of tumor–immune co-culture systems will be fundamental to achieving truly personalized cancer immunotherapy, in which each patient’s tumor is prospectively tested against their own immune cells to guide individualized treatment decisions, transforming organoid co-culture from a research curiosity into a functional companion diagnostic for precision oncology.

Subject of Research: Patient-derived tumor organoid co-culture systems for modeling tumor–immune interactions and evaluating cancer immunotherapy and precision oncology

Article Title: Patient-derived tumor organoid co-culture systems as translational platforms for cancer immunotherapy and precision oncology

Article References: Gadelkarim, M., Elsayed, A., Abaza, T., Bahr, A. R., Elsayed, Y., & Iqbal, O. (2026). Patient-derived tumor organoid co-culture systems as translational platforms for cancer immunotherapy and precision oncology. Clinical Cancer Bulletin, 5(1), Article 18. https://doi.org/10.1007/s44272-026-00067-1

Image Credits: AI Generated

DOI: 10.1007/s44272-026-00067-1

Keywords: tumor organoids, co-culture, tumor microenvironment, cancer immunotherapy, immune checkpoint inhibitors, CAR T cells, PBMC, precision oncology, patient-derived models, adoptive cell therapy, organ-on-a-chip, companion diagnostics