PLATFORM TECHNOLOGY

Industrial PHA bioplastic production on Haloferax Mediterranei

An extreme halophilic archaeon that converts waste-derived volatile fatty acids into biodegradable polyesters. Non-sterile continuous fermentation. Cell lysis in water. No endotoxin contamination. A platform designed from first principles for the lowest achievable cost of polymer production.

The organism

The carbon source is the single largest cost driver in PHA production, typically accounting for 30–50% of total production cost. Current industrial PHA platforms are built on sugar-based feedstocks: glucose, fructose, corn starch hydrolysates, and in some cases sucrose or glycerol. These substrates are commodity-priced, food-competitive, and subject to agricultural market volatility.

H. mediterranei can utilise all of these conventional carbon sources. But its structural advantage lies elsewhere: in its ability to efficiently metabolise volatile fatty acids (VFAs) — acetate, propionate, and butyrate — as primary carbon and energy sources for PHA biosynthesis. Published research has demonstrated that VFA mixtures derived from anaerobic fermentation of food waste yield 55% higher PHBV production than glucose on an equivalent carbon basis.

This matters because VFAs are the primary output of anaerobic digestion, an established industrial process applied globally to food waste, agricultural residues, wastewater sludge, and municipal organic waste. The feedstock for our platform is not a commodity to be purchased but rather a waste stream that facility operators currently pay to dispose of. This inverts the cost structure: instead of competing for sugar on global commodity markets, we accept organic waste and convert it to a high-value biopolymer.

Beyond VFAs, H. mediterranei has been demonstrated to produce PHBV from hydrolysed cheese whey and whey permeate, crude glycerol from biodiesel production, olive mill wastewater, hydrolysed rapeseed meal, confectionery industry waste, and pretreated vinasse. This substrate versatility enables regional feedstock adaptation: dairy waste in Northern Europe, olive mill waste in the Mediterranean, food waste digestate in urban centres. This makes it a viable platform for decentralized, regional production.

Feedstock advantage: volatile fatty acids from organic waste

Precursor-free PHBV: In all bacterial PHA producers, producing the copolymer PHBV requires supplementation with valerate or propionate as exogenous precursors — an additional raw material cost and process complexity. H. mediterranei is the only well-studied production organism that synthesises both the 3-hydroxybutyrate and 3-hydroxyvalerate monomers endogenously from simple carbon sources, producing PHBV as its default storage polymer.

Process architecture

Non-sterile continuous fermentation

Hypersaline growth conditions create a thermodynamic contamination barrier that substitutes for aseptic engineering. At 200–250 g/L NaCl, maintaining cellular osmotic balance requires either the salt-in strategy of haloarchaea (intracellular K⁺ at molar concentrations, proteins evolved for function at saturating ionic strength) or compatible-solute accumulation at metabolic cost exceeding 20% of ATP budget in non-adapted organisms. Neither is accessible to environmental contaminants on operational timescales. A narrow set of extremely halophilic bacteria (Salinibacter ruber) and halophilic fungi (Wallemia ichthyophaga, Hortaea werneckii) can persist at these salinities, but none are plausible fermenter invaders outside of solar-saltern geographies, and none grow fast enough to compete with H. mediterranei at production densities.

This reshapes bioreactor capex and opex along three axes. Vessels no longer require ASME Section VIII pressure rating for steam sterilisation, cutting fabricated-vessel cost by roughly 40–60% per litre of working volume at pilot and commercial scale. Steam-in-place utility load is eliminated along with its water, energy, and cycle-time overhead. Sterile-grade HEPA air filtration is replaced by basic particulate filtration, reducing both filter capex and differential-pressure energy cost. Collectively these are the line items that dominate the capex delta between biologics-grade and commodity-chemical fermentation, and the halophilic platform captures most of the delta in the right direction.

The second economic lever is continuous operation. Industrial PHA production on C. necator runs as fed-batch, with a duty cycle of roughly 60–70%: the balance is consumed by turnaround — drain, clean-in-place, steam-in-place, media charge, inoculation, lag phase. Continuous open halophilic fermentation removes the turnaround requirement. Published work with Halomonas has demonstrated 14-day and, in one reported case, 65-day continuous runs without contamination; continuous operation on H. mediterranei itself is less validated and is one of our engineering priorities. Annual plant throughput under continuous operation scales closer to theoretical maximum, with corresponding reductions in fixed-cost allocation per kilogram of polymer.

Downstream processing typically accounts for 30–50% of total PHA production cost in conventional bacterial platforms, and the cost is structural rather than incidental. Cupriavidus necator and similar freshwater systems require high-pressure homogenisation or bead milling to rupture the cell envelope, followed by chlorinated-solvent extraction (chloroform, dichloromethane) or detergent washing to separate polymer from cellular debris, then solvent recovery, granule washing, and drying. Each step adds capex and opex; high-shear disruption also produces measurable reductions in polymer molecular weight, narrowing the applications for which the recovered polymer is materially suitable.

H. mediterranei removes the two most expensive steps in that sequence. Cells cultivated at 200–250 g/L NaCl (~3.4–4.3 M) maintain intracellular osmolarity against a medium whose osmotic pressure exceeds 17 MPa. Resuspending harvested biomass in fresh water or dilute buffer imposes the full differential across the archaeal monolayer, and lysis is rapid and substantially complete. PHA granules release directly into the aqueous phase and are recovered by centrifugation or microfiltration. The remaining DSP train — biomass harvest, hypotonic lysis, granule recovery and washing, drying — runs on standard unit operations compatible with food- and pharma-grade manufacturing, without chlorinated solvents and without the high-shear step that limits molecular weight in conventional platforms.

Two second-order benefits follow. First, preserving native polymer molecular weight widens the addressable application envelope toward segments where MW matters — oriented fibres, thin films, resorbable biomedical devices. Second, the saline supernatant recovered after lysis can be partially recycled as base medium for subsequent fermentation cycles; published work with H. mediterranei has demonstrated substitution of a substantial fraction of fresh mineral medium with recovered spent broth without yield loss, reducing both salt input cost and saline effluent disposal.

Osmotic lysis downstream processing

The material properties (flexibility, elongation at break, melting temperature, crystallinity) of PHA bioplastics are determined by the monomer composition of the polymer, which is in turn mainly determined by the substrate specificity of the PHA synthase enzyme. Controlling which monomers are incorporated, and at what ratios, is the central challenge in producing application-specific PHA materials.

We are developing proprietary PHA synthase variants with modified substrate specificity using structure-guided computational protein engineering. Our approach combines structural analysis of the enzyme active site and substrate access tunnel with computational design methods to predict and validate amino acid substitutions that alter monomer incorporation profiles.

This work targets both optimisation of existing capabilities such as fine-tuning the 3HV/3HB ratio in PHBV and expansion of the monomer repertoire to include medium-chain-length monomers that confer elastomeric properties to the resulting copolymers. Specific engineered variants, structural insights, and computational methods are covered by provisional patent filings.

Computational enzyme engineering

Multiple provisional patent applications have been filed covering engineered PHA synthase variants with modified substrate tunnel geometry and altered monomer specificity. These filings cover both computationally designed variants and the computational methods used to identify them. We do not disclose specific amino acid substitutions, structural coordinates, or computational pipelines prior to patent publication.

Product portfolio

H. mediterranei natively produces two PHA types from unmodified wild-type or minimally engineered strains:

PHB — poly(3-hydroxybutyrate). The simplest PHA homopolymer. Rigid, highly crystalline, with properties similar to polypropylene. Suitable for rigid packaging, injection-moulded articles, and fibre applications. Produced when the organism is grown on simple sugars under conditions that suppress 3HV incorporation.

PHBV — poly(3-hydroxybutyrate-co-3-hydroxyvalerate). The native default product of H. mediterranei. A copolymer with improved flexibility, lower melting point, and reduced crystallinity compared to PHB. The 3HV content (6–15 mol%) is tunable by carbon source selection. Suitable for flexible films, coatings, and biomedical applications. Produced without valerate precursors — a unique advantage of this organism.

Available now (native production)

Our engineering programme targets the expansion of H. mediterranei's monomer repertoire to include:

P(3HB-co-4HB) — A copolymer incorporating 4-hydroxybutyrate monomers, yielding highly elastomeric materials. Requires introduction of heterologous 4HB biosynthesis pathway genes. This copolymer has FDA-approved medical applications in sutures and tissue engineering scaffolds — applications where the LPS-free nature of our archaeal platform provides a decisive advantage.

PHBHHx — poly(3-hydroxybutyrate-co-3-hydroxyhexanoate). A copolymer with elastomer-like properties, excellent biocompatibility, and thermal stability. Requires engineered PHA synthase variants with broadened substrate specificity and medium-chain-length CoA supply pathways. This is the primary target of our computational enzyme engineering programme.

Terpolymers — P(3HB-co-3HV-co-4HB) and other multi-monomer compositions with tailored mechanical properties. Long-term platform targets enabled by the combination of native 3HV biosynthesis with engineered 4HB and MCL-3HA pathways.

Under development (engineered production)

The biomedical PHA market — resorbable sutures, bone fixation hardware, tissue engineering scaffolds, controlled-release drug delivery matrices — is the highest-value application segment in the polymer. Finished-device pricing sits several orders of magnitude above commodity packaging grade; Tepha's P4HB suture line (MonoMax, GalaFLEX, Phasix), produced from engineered E. coli and now part of the BD portfolio, illustrates the economics. At those price points, the constraint on market entry is not polymer cost. It is regulatory-grade purity, and specifically endotoxin burden.

USP <85> and FDA pyrogen and endotoxins guidance set the thresholds: 20 EU per device for most implantables contacting body fluids, 2.15 EU per device for devices contacting cerebrospinal fluid, 0.5 EU/mL for parenteral-contact surfaces. Endotoxin is measured by the Limulus amebocyte lysate (LAL) assay or its recombinant Factor C successor, and both are specific to lipopolysaccharide — the outer-membrane glycolipid of Gram-negative bacteria.

Every established PHA production chassis — Cupriavidus necator, Halomonas bluephagenesis, E. coli, engineered Pseudomonas — is Gram-negative. LPS is a structural component of the outer membrane and co-purifies with intracellular PHA during extraction. Reaching medical-grade thresholds requires a dedicated purification cascade bolted onto standard DSP: sodium hypochlorite or hydrogen peroxide oxidation, Triton X-114 phase separation, repeated solvent washes, ultrafiltration, or combinations of the above, depending on the manufacturer. Each step carries cost and yield penalty; oxidative treatments additionally cleave polymer chains in a material whose mechanical performance is already molecular-weight-sensitive.

H. mediterranei is an archaeon. Archaea synthesise neither LPS nor any structurally analogous outer-membrane glycolipid — their envelope is a monolayer of ether-linked isoprenoid lipids covered by an S-layer glycoprotein. PHA recovered from archaeal biomass tests clean on LAL at the source, not after multi-step purification. Medical-grade purity still requires the universal removal of host-cell protein, residual nucleic acid, and process solvents — those apply to every production organism — but the LPS-specific purification train is not required. The effect is not "low-endotoxin after treatment." It is the structural absence of the contaminant in the source organism, removing the single most failure-prone purification step in the chain and preserving polymer molecular weight through the DSP train.

Biomedical-grade PHA: the LPS-free advantage

Frequently asked questions

What is Haloferax mediterranei and why is it used for PHA production?

Haloferax mediterranei (ATCC 33500) is an extreme halophilic archaeon originally isolated from solar salterns in the Mediterranean. It naturally produces the bioplastic copolymer PHBV (poly-3-hydroxybutyrate-co-3-hydroxyvalerate) without requiring valerate precursors — a capability unique among well-studied PHA producers. Its extreme salt tolerance (200+ g/L NaCl) enables non-sterile, contamination-free fermentation, and the large osmotic differential allows polymer extraction through simple osmotic lysis in fresh water, eliminating the need for mechanical cell disruption or organic solvents.

Can H. mediterranei produce PHBV without valerate precursors?

Yes. Haloferax mediterranei is unique among well-characterised PHA-producing microorganisms in that it natively synthesises PHBV — incorporating 3-hydroxyvalerate (3HV) monomers at 6–15 mol% — without requiring exogenous valerate or propionate as precursors. This is a major cost advantage, as other organisms (including Cupriavidus necator and Halomonas bluephagenesis) require supplementation with expensive precursor compounds to produce PHBV.

How does osmotic lysis work for PHA extraction from halophilic archaea?

Extreme halophilic archaea like H. mediterranei maintain intracellular osmolarity balanced against their hypersaline growth medium (200+ g/L NaCl). When harvested cells are transferred to low-salt or pure water, the massive osmotic differential (exceeding 200 g/L) causes rapid water influx and cell lysis. The intracellular PHA granules are released without mechanical disruption, solvent extraction, or enzymatic digestion. This eliminates the most expensive downstream processing step in conventional PHA production.

How does H. mediterranei compare to H. bluephagenesis for PHA production?

H. mediterranei and H. bluephagenesis occupy different biological domains (Archaea vs Bacteria) and have different structural advantages. H. mediterranei offers stronger contamination resistance (200+ g/L vs 10–60 g/L NaCl), cheaper downstream processing (osmotic lysis in water vs SDS washing), broader feedstock flexibility (VFAs, whey, food waste vs primarily glucose), native PHBV production without precursors, and LPS-free polymer output suitable for biomedical use. H. bluephagenesis has advantages in cell density (132 g/L vs ~20 g/L CDW), growth rate, genetic tool maturity, copolymer diversity, and industrial validation at 10,000+ tonnes per annum scale. See the comparison table above for a detailed breakdown.

What waste feedstocks can H. mediterranei convert to bioplastics?

H. mediterranei has been demonstrated to produce PHBV from a wide range of waste-derived feedstocks: volatile fatty acids from anaerobic digestion of food waste (yielding 55% higher PHBV than glucose), hydrolysed cheese whey and whey permeate, crude glycerol from biodiesel production, olive mill wastewater, hydrolysed rapeseed meal, molasses wastewater, and candy industry waste. Its ability to efficiently metabolise VFA mixtures — acetate, propionate, and butyrate — makes it uniquely suited for integration with existing anaerobic digestion infrastructure.

Why is LPS-free PHA important for biomedical applications?

Lipopolysaccharide (LPS) endotoxins are components of the outer membrane of Gram-negative bacteria, including all Halomonas and Cupriavidus species used for PHA production. LPS co-purifies with PHA during extraction and triggers severe immune responses in humans, making Gram-negative-derived PHA unsuitable for biomedical implants, drug delivery systems, and tissue engineering scaffolds without costly additional purification. H. mediterranei is an archaeon and does not produce LPS, yielding inherently endotoxin-free PHA suitable for medical applications without extra processing.

Is Haloferax mediterranei genetically engineerable?

Yes. H. mediterranei possesses an endogenous Type I-B CRISPR-Cas system that has been repurposed for both gene repression (CRISPRi, demonstrated in 2021 with up to 98% repression and 76% improvement in PHBV production) and gene knockout (CRISPR-KO, established in 2025 with 27% efficiency). Additional genetic tools include pop-in/pop-out homologous recombination, shuttle vectors, and established transformation protocols. The genetic toolbox is advancing rapidly.