L-Threonine Production Process
L-Threonine from E. coli fermentation — disc-stack centrifugation, activated carbon, strong cation exchange, evaporation, cooling crystallization, and fluidized bed drying to >98.5% food/feed grade purity
Process Overview
L-Threonine is an essential amino acid produced at ~500,000 tonnes/year globally, primarily as an animal feed supplement to replace costly soy and fishmeal proteins. Metabolically engineered E. coli strains achieve fermentation titers of 80–120 g/L in 30–40 hour fed-batch processes with glucose as carbon source. The downstream process exploits threonine’s zwitterionic nature and relatively low solubility at neutral pH for efficient crystallization purification. Overall yield: 75–85%, purity >98.5% (food/feed grade).
Process Steps
1
Clarification
Disc-Stack Centrifugation — Cell Removal
Remove E. coli cells from the fermentation broth by continuous disc-stack centrifugation (10,000–15,000 × g). The centrate (supernatant) contains threonine and all soluble broth components. The cell sludge is typically processed for single-cell protein or biomass disposal. For high-titer broths (>100 g/L threonine), dilution 1:1 with water before centrifugation may be needed to reduce viscosity. Centrate turbidity: <50 NTU. Recovery of threonine in centrate: >99%.
Yield: >99%
Cell removal: >99.9%
2
Adsorption
Activated Carbon Treatment (0.5% w/v)
Add powdered activated carbon (food-grade, 0.5% w/v) to the centrate and stir at 60°C for 30 minutes. Activated carbon adsorbs color bodies (melanoidins from Maillard reactions during fermentation), pigments, and hydrophobic impurities that would cause discoloration of the final product. Filter the carbon slurry through a plate-and-frame filter press with filter aid. The decolorized centrate should be water-clear (OD420 < 0.1). Threonine loss to carbon: typically <2%.
Yield: >98%
Color removal: >90%
3
Ion Exchange Chromatography
Strong Cation Exchange (H+ Form)
Acidify the decolorized centrate to pH 2.0 with hydrochloric acid or sulfuric acid. Load onto sulfonic acid strong cation resin (H+ form, e.g., Amberlite IR-120, Dowex 50W). At pH 2, threonine is fully protonated (α-NH3+) and binds strongly to the resin. Glucose, organic acids, and inorganic anions pass through (non-binding or weakly binding). Wash with water at pH 2 to remove non-retained impurities. Elute threonine with 2–3 M ammonia solution (NH3) to pH 8–9. This step concentrates and purifies threonine, removing sugars, organic acids, and minerals.
Yield: 88–93%
Purity: 92–96%
4
Concentration
Neutralization + Evaporation (300–400 g/L)
Neutralize the alkaline IEX eluate to pH 5.0–6.0 (near threonine’s pI of 5.87) with dilute hydrochloric acid. Concentrate by vacuum evaporation at 50–60°C to 300–400 g/L threonine. High concentration is required to achieve good crystallization yield. Threonine is stable at these temperatures (no racemization, no significant degradation). The concentrate is a nearly saturated solution at 60°C ready for cooling crystallization.
Concentration: 300–400 g/L
Yield: >99%
5
Crystallization
Cooling Crystallization (4°C, 24 h, >98% purity)
Cool the concentrated threonine solution from 60°C to 4°C over 24 hours with gentle stirring (50–100 rpm paddle agitator). Threonine solubility decreases from ~900 g/L at 60°C to ~170 g/L at 4°C, driving crystallization of >60% of dissolved threonine. The crystals are thin plate-like monoclinic prisms. Seed crystals added at 45°C improve nucleation control and crystal size. After crystallization, harvest by basket centrifuge. Wash with cold water (0.5 cake volumes) to remove mother liquor. Purity after washing: >98%.
Yield: 80–88%
Purity: >98%
6
Drying
Centrifugal Drying + Fluidized Bed Dryer (60°C)
After centrifugal washing, the crystal cake has 5–8% residual moisture. Transfer to a fluidized bed dryer (inlet air 60°C, product temperature <45°C) for 2–3 hours to achieve <0.3% moisture. Fluidized bed provides gentle, uniform drying with minimal crystal attrition. Screen dried product through 0.5–2 mm sieves to separate fines and oversized aggregates. Final product: free-flowing white crystalline L-threonine powder, >98.5% purity by HPLC, <3 ppm heavy metals, suitable for direct addition to animal feed premixes.
Yield: >99%
Moisture: <0.3%
Target Molecule: L-Threonine
| Molecular Weight | 119.12 Da |
| Isoelectric Point (pI) | 5.87 |
| pKa values | 2.09 (α-COOH), 9.10 (α-NH3+) |
| Solubility (20°C) | ~90 g/L (increases strongly with temperature) |
| Crystal form | Monoclinic platelets (anhydrous) |
| Application | Limiting amino acid in wheat/maize-based animal feeds; human pharmaceutical (rare) |
Cost Considerations
| Step | Key Cost Driver | Relative Cost |
|---|
| Disc Centrifugation | Centrifuge capital, energy | Medium |
| Activated Carbon | Carbon cost (not recyclable), filter aid, press | Low |
| Strong Cation Exchange | Resin, HCl/NH3 regeneration, wastewater | High |
| Evaporation | Steam consumption, evaporator capital | Medium |
| Crystallization | Refrigeration, crystallizer vessel, agitation | Low |
| Fluidized Bed Drying | Energy, inlet air heating, bag filter | Low |
Ion exchange is the dominant cost — some producers skip it. For feed-grade threonine, the IEX step is sometimes replaced by direct evaporation-crystallization of the clarified broth. This sacrifices some purity (96–97% vs 98.5%) but significantly reduces capital and operating cost. The IEX step is retained for food/pharmaceutical grade. Mother liquor from crystallization is recycled to reduce overall yield loss. Use
untangle.bio to compare process routes with and without IEX.
Frequently Asked Questions
Why is L-threonine important as an animal feed supplement?
L-Threonine is the second or third limiting amino acid in monogastric animal diets based on corn and wheat (lysine is first). Supplementing feeds with crystalline L-threonine allows reduction of total dietary protein (crude protein) by 2–3 percentage points, saving on expensive soybean meal. This reduces nitrogen excretion by 10–20%, lowering environmental impact. The economic value is driven by the cost differential: crystalline threonine ($1–2/kg) vs soybean meal ($0.4–0.6/kg protein), justified by the much higher bioavailability of the crystalline amino acid.
How does E. coli produce threonine in such high titers?
Wild-type E. coli produces threonine at only trace levels due to strict feedback inhibition of aspartate kinase by threonine and lysine. Metabolic engineering removes these regulatory controls: overexpression of feedback-resistant aspartate kinase (thrA*), amplification of the threonine biosynthesis operon (thrABC), deletion of threonine degradation genes (tdh, kbl), and deletion of competing pathways (lysine, isoleucine). These modifications channel aspartate carbon flux toward threonine, achieving titers of 80–120 g/L with yields of 0.4–0.5 g threonine/g glucose.
Can the ion exchange step be replaced by direct crystallization?
Yes, for feed-grade threonine (>98% purity specification). Direct evaporation of clarified broth to 300–400 g/L followed by cooling crystallization yields threonine at 96–97% purity. The IEX step adds significant cost but improves purity to 98.5%+ by removing glucose, organic acids, and mineral salts that would otherwise contaminate the crystal lattice or mother liquor. Major Chinese producers at very large scale (>100,000 tonnes/year) often omit IEX to reduce capital investment and operating cost.
What analytical methods verify food/feed grade threonine quality?
Key quality tests for food/feed-grade L-threonine include: (1) HPLC amino acid analysis for purity and D-threonine content (<0.1%), (2) optical rotation (specific rotation [α]D +28° for pure L-form), (3) heavy metals by ICP-MS (Pb <2 ppm, As <1 ppm, Hg <0.1 ppm), (4) residual solvents (negligible for aqueous process), (5) microbial tests (total aerobic count <10,000 CFU/g for feed grade), and (6) moisture by Karl Fischer or loss on drying. Pharmacopoeia grades (USP, EP) require additional tests for clarity, color, and specific amino acid impurities.
Design Your L-Threonine Production Process
Build the full threonine downstream train from centrifugation through crystallization and drying, simulate mass balance at your fermentation titer, and compare process routes with and without ion exchange.
Open untangle.bio