At Mika Biologics, we build manufacturability by default. Microbes are our engine. Speed, quality, and scale are the outputs.

Executive take

Plasmid DNA & mRNA manufacturing now sits at the backbone of modern biologics—from genetic vaccines and in vivo gene editing to engineered cell therapies and next-generation immuno-oncology. Demand for fast, GMP-grade supply chains is exploding; capacity, however, remains constrained and expensive. The bottleneck isn’t just steel or square footage—it’s microbial mastery at scale, clean process architectures that prevent endotoxin at the source, and an analytics stack that makes quality visible in real time.

This post lays out Mika’s microbial-first approach to plasmid DNA & mRNA manufacturing: how we design upstream and downstream for speed and reproducibility, how we collapse tech-transfer drag with digital QMS + QbD, and how our endotoxin-control philosophy removes the most stubborn risk in nucleic-acid manufacturing. If you’re scouting partners, here’s what “ready-to-run” looks like—and how a specialized microbial CDMO turns months into weeks without compromising release criteria.

Why microbial-first matters now

Most of the world’s plasmids are born in E. coli, and most mRNA programs rise or fall on the quality of that plasmid template plus a clean IVT (in-vitro transcription) ecosystem. That makes microbial expertise—not mammalian capacity—the decisive variable for schedule, cost, and CQA conformance. A microbial-optimized facility unlocks:

• High space-time yields for pDNA via strain engineering, media design, and growth-phase control.
• Predictable supercoiled fractions through lysis chemistry, shear management, and polishing tuned to plasmid topology.
• Low-endotoxin architectures that start with the organism, not only with the column you buy after the fact.
• Tight IVT performance because your enzymes, buffers, and template quality are designed to work together, not “integrated” via last-minute firefighting.

Microbial-first isn’t a slogan here—it’s the default physics of our plant.

The pDNA manufacturing chain, de-risked end-to-end

1) Host, vector, and chassis strategy

Manufacturability starts at construct architecture. We interrogate: copy number regime (low/med/high), replication origin behavior under fed-batch stress, antibiotic vs. auxotrophy selection, plasmid size class (sub-5 kb, 5–10 kb, >10 kb behaves as a different animal), sequence liabilities (direct repeats, palindromes, high GC plateaus), and the intended unit operation path (is this template for IVT, transfection, or direct use?).

On the host side, we match metabolic burden to chassis: growth kinetics, membrane robustness, recombination propensity, and lysis amenability. For high-burden constructs we pivot to recombination-deficient, endA- and RNase-tamed strains; for low LPS risk we deploy genetically detuned outer-membrane phenotypes plus LPS-sparing media to reduce load upstream of DSP.

Design for release, not for luck. We map QTPP → CQAs → CPPs before seed train:

• CQAs (pDNA): supercoiled content (%SC), residual gDNA and RNA, topology distribution (SC/OC/LIN), A260/280 & A260/230, residual salts/solvents, endotoxin (EU/µg), bioburden/sterility.
• CPP exemplars: μ (specific growth rate), DO control window, feed composition/flux, lysis pH/temperature/exposure time, shear budget (P/V, Np), conductivity at load, gradient shape in AEX, TFF TMP.

Make it visible: Inline/at-line analytics—qPCR for HCD/RNA, microfluidic CE for topology kinetics, kinetic chromogenic LAL (validated) for LPS trendlines, and UV ratio telemetry—are wired into the historian so process drift is caught before release.

Typical pitfalls & mitigations

• Instability >10 kb: lower copy origin, stabilize with cer or par systems; lower temp growth; tune feed C:N:P to reduce replication stress.
• Recombination hotspots: recA-, sbcCD- backgrounds; codon-shave repeats if design latitude exists.
• Selection drag: consider auxotrophy when GMP antibiotic sourcing or clearance is problematic.

2) Upstream: fed-batch tuned for topology, not just titer

Objective: maximize intrinsic supercoiled potential pre-lysis. We run oxygen-cascaded fed-batch with substrate flux control (glucose or glycerol), phosphate and trace metal titration, and temperature profiling (e.g., 30–37 °C ramps) to balance growth with plasmid amplification.

Signals we track and act on

• OTR/kLa and DO dynamics: flagging oxygen limitation that skews plasmid amplification vs. biomass.
• μ profile and acetate trajectory (overflow control): carbon flux mismanagement will bite %SC later.
• Viscosity & capacitance: early warnings for cell envelope stress and eventual lysis aggressiveness.
• pH & base addition signature: proxy for metabolic regime shifts; we correlate to later isoform drift.

Design-space tightening across campaigns
Our historian correlates upstream fingerprints (e.g., late-phase DO sawtooth or base-addition spikes) with downstream QC (%SC, nicked fraction). As lots accrue, CPP bounds narrow; alarms become predictive, not reactive.

Failure modes & countermeasures

• Acetate spikes → lower %SC: adopt mixed feeds (glucose:glycerol), raise DO setpoint, increase P/V.
• Plasmid loss at high cell density: temper μ with temp drop, re-tune feed ratio, re-assess selection pressure.
• Unexpected shear during seed transfers: standardized tubing IDs/lengths, low-shear pumps, validated transfer flow-rates.

3) Harvest & lysis: chemical precision, mechanical restraint

Alkaline lysis is chemistry that deserves SOP rigor. We bound NaOH/SDS concentration, temperature (±0.5 °C), exposure time (±5–10 s), and mixing energy to protect topology and minimize nicking.

Clarification employs staged depth filtration (e.g., 10 µm → 3 µm → 0.65/0.45 µm) with calibrated flow-per-area and ΔP limits to keep shear under budget. Where helpful, we dose endo-aware surfactants and ionic conditions that keep LPS disaggregated (removable) instead of embedding into pDNA.

Key envelopes

• Shear budget: define allowable Np/P/V across impeller, pump, and filter trains; measure, don’t assume.
• Thermal envelope: lysis and neutralization temperature tightly controlled; excursions shift SC→OC.
• Hold-time ceiling: post-lysis holds are the silent topology killer; we design to flow, not to hold.

Common misses

• Vortexing or uncontrolled mixing: replace with validated recirculation/mild agitation geometries.
• Single-stage clarification: two- to three-stage depth trains reduce fouling and protect downstream membranes.
• pH shock on neutralization: inline neutralization with residence-time control prevents localized over-exposure.

4) Capture, polish, and buffer architecture

Capture: anion exchange (AEX) still earns its keep, but resin chemistry and load conductivity are tailored to plasmid size and isoform profile. For high-throughput trains, membrane adsorbers reduce cycle times and enable aggressive sanitization without resin memory effects.

Polish is impurity-fingerprint driven:

• RNA load: RNase-managed upstream + salt/chaotrope choice; polish via HIC or mixed-mode tuned to RNA hydropathy.
• gDNA residuals: AEX gradient shape, Mg2+ modulation, and shear-aware handling reduce co-elution.
• Host proteins: mixed-mode or secondary AEX; detergent traces stripped in UF/DF.
• Endotoxin: targeted flow-through steps, membrane adsorbers with LPS affinity, ionic conditions that discourage re-association.

UF/DF (TFF) finalizes buffer (pH/ionic strength/osmolality) for your next step: IVT, transfection reagent complexation, or drug-product fill. We set TMP ceilings, shear limits, and path sterility to protect topology.

Design details that matter

• Load CV and linear velocity: residence time is your selectivity knob—don’t turn it blindly.
• Conductivity windows: small drifts change selectivity; inline EC control stabilizes capture.
• Sanitization and memory effects: resin lifetime tracking & LPS decontamination protocols prevent batch-to-batch ghosts.

5) Release analytics that predict behavior, not just pass/fail

Mika’s QC panels are nucleic-acid-centric and use-case aligned:

Identity & topology

• Agarose or capillary electrophoresis: SC/OC/LIN quantification with method linearity verified.
• Topological integrity score: %SC as headline CQA when IVT is downstream.

Impurities

• gDNA & RNA residuals by qPCR/RT-qPCR.
• Endotoxin by validated LAL (with known interference controls).
• Residual solvent/salt profile; osmolality.

Micro & sterility

• Bioburden/sterility per pharmacopeia; environmental trending piped to the batch pack.

IVT-readiness add-ons

• Nicking/linearization readiness: confirm restriction map and cut efficiency; verify no partial nick-heavy tails that depress IVT yield.
• Template homogeneity: CE traces correlated to IVT transcription rates on your construct, not a demo plasmid.

IVT mRNA that starts clean, transcribes hard, and purifies right

1) Template preparation that respects the reaction physics

IVT excellence is won before T7 ever engages. We linearize with restriction enzymes or enzymatic cassettes designed for clean blunt/defined ends; residual nicked/relaxed plasmid is controlled because it changes transcription kinetics and elevates dsRNA.

We confirm end integrity, template size, and cut completeness by CE; then set Mg2+/NTP/pyrophosphate stoichiometry to your cap strategy and ORF length. Buffer components are screened for LAL interference and enzyme compatibility (detergent ghosts are a real thing).

Typical traps

• Partial linearization: raises abortive transcription and dsRNA; enforce cut checks, not just assume.
• Template carryover salts/solvents: depress IVT velocity; UF/DF with conductivity targets solves it.
• Inhibitory buffers from pDNA DSP: switch buffers upstream—IVT shouldn’t be a scavenger hunt.

2) Enzymes, caps, and poly(A): coherent, not bolted on

We support co-transcriptional capping (e.g., m7G dinucleotides/CleanCap-like systems) or post-transcriptional enzymatic capping when that improves stability or throughput.

Poly(A) strategy: encoded vs. enzymatically appended depends on tail length precision, context, and innate-immune profile you seek. We validate reaction time/temperature/enzyme load against capping %, tail length distribution, and dsRNA formation.

Compatibility matrix we actually maintain

• Cap analogue ↔ polymerase ↔ Mg2+ window ↔ NTP stoichiometry ↔ ORF length ↔ desired innate sensing.
• Enzymatic cap kits ↔ salt sensitivity ↔ residual enzyme clearance path in DSP.

Watchouts

• Capping competition with high NTP: capping % falls off; tune NTP ratios and cap analogue excess.
• Encoded poly(A) instability in certain contexts: consider enzymatic tailing for tight distributions.

3) Purification that removes what biology cares about

We condition viscosity with TFF (not to beat RNA up), remove bulk salts, then run chromatography tuned to mRNA size and impurities: plasmid template, enzymes, short RNA fragments, and especially dsRNA.

dsRNA is the villain. We employ orthogonal assays (e.g., anti-dsRNA dot blot plus enzyme-based confirmation) and unit ops intentionally selected for dsRNA minimization (e.g., AEX/HIC modes with dsRNA de-enrichment). We validate by reduction curves and LOD demonstrations, not just single numbers.

Good practice specifics

• Flow-per-area and TMP ceilings to protect integrity.
• DNase strategy with thorough inactivation/clearance controls (residual enzyme becomes a CQA).
• Hold-time minimization between steps; RNA doesn’t forgive.

4) Potency-correlated analytics

Release covers identity (sequence), integrity (CE or LC-MS fragmentation maps), capping %, poly(A) profile, dsRNA, residual DNA & protein (enzyme), residual salts/solvents, endotoxin, osmolality, pH, bioburden/sterility.

Where appropriate, we add cell-free translation or in-cell reporter correlations so your analytics predict biological behavior, not just pass specs.

Endotoxin governance: prevent the spark, don’t buy bigger hoses

Endotoxin (LPS) quietly wrecks nucleic-acid programs—especially with Gram-negative hosts. You cannot certificate your way out of an LPS-heavy train. Mika’s prevent-don’t-chase doctrine:

Upstream origins

• Strain & media picks that stabilize membranes and lower LPS shedding.
• Temperature & shear envelopes to preserve outer membranes through harvest.
• Feed design that avoids stress-induced LPS release.

Chemistry & buffers

• Surfactant and ionic choices that keep LPS disaggregated and separable, not bound to DNA/RNA.
• Resin/membrane selections with proven LPS clearance and validated sanitization.

Plant behaviors

• Segregation discipline: hose whips, gaskets, totes—classic LPS vectors—are controlled/dedicated.
• Single-use flow paths where risk/throughput math supports it.
• Routine at-line LAL trending—not just at CoA—so we correct in process, not post-mortem.

If your first LPS visibility is at the certificate, the timeline is already lost. Endotoxin control is a habit, not a test.

5) Release analytics that actually predict performance

Digital QMS + QbD: speed without gambling

“Fast” becomes “fragile” if documentation and change control can’t keep up. Mika’s unified digital QMS (ALCOA+ principles baked in) and QbD playbook make speed repeatable:

• Design space definition at kickoff: we pre-specify CPP ranges for growth rate, lysis time/temperature, flow rates, column loads/conductivity, chromatographic pH/ionic strength, and UF/DF TMP—each mapped to the relevant CQA.
• In-process trending with alarms on the variables that matter (e.g., rise in viscosity during IVT hinting at off-stoichiometry, or a DO profile that predicts a topology shift in pDNA).
• Change control that doesn’t stall science: controlled experiments in a sandboxed branch of the eBR (electronic batch record), then formal adoption once the data proves out.
• Right-first-time tech transfer because your method and our method live in the same digital thread from day one.

The result: fewer surprises, shorter deviations, cleaner PPQ.

Scale without the usual bruises

We support microbial campaigns from 2 L process-development through 200–2,000 L GMP, with engineered paths to larger footprints via partner networks when needed. The scale-up philosophy:

• Geometric similarity where it matters (impeller, baffles, sparger) and functional similarity where geometry can’t match (kLa, P/V, mixing time).
• Shear budgeting across the entire flow path—hoses, pumps, valves—so your plasmid’s topology and your mRNA’s integrity survive success.
• Consumable harmonization (membrane areas, resin volumes, flow-per-area) to keep residence times predictable.

Scale should amplify your design, not expose its cracks.

Program modes: where you are, what you need

• Template-only mode (pDNA-for-IVT): You own the IVT; we deliver linearized or linearization-ready, high-supercoiled pDNA with an endotoxin profile that won’t dominate your next step.
• End-to-end mRNA mode: We deliver formulated or bulk mRNA with analytics aligned to your clinical path (e.g., capping %, dsRNA) and fit-for-purpose stability data.
• Hybrid mode: You bring a favorite enzyme or proprietary cap; we integrate without re-learning the same lessons on your clock.

Every mode starts with a manufacturability brief—a rapid mapping of your QTPP to a practical process envelope so everyone sees the same mountain.

Regulatory & quality posture (without the buzzword salad)

We operate to GMP with a documentation framework aligning to ICH Q7/Q9, QbD, data integrity (ALCOA+), and nucleic-acid-relevant USP chapters (e.g., endotoxin testing, sterility). For mRNA programs we partner on control strategy narratives that make capping efficiency and dsRNA first-class citizens in your CMC. Batch records are designed for audit-readiness—clear line-of-sight from raw materials through release, with validated cleaning and cross-contamination controls that withstand diligence.

What good looks like: signals of a strong CDMO fit

If you’re screening partners, ask for these proofs:

1. Topology trend vs. lysis envelope over multiple lots (not just one pretty gel).
2. dsRNA reduction curve before and after the chosen chromatography step—plus the assay orthogonality.
3. Template readiness metrics (nicked %, linearization completeness, residual enzyme) tied to IVT yield on your construct, not a demo sequence.
4. Endotoxin load-in vs. load-out per unit op, with hypotheses for any excursions.
5. Digital lineage: show me a change, show me the data, show me the approval—on screen, in minutes.

You’re not buying claims; you’re buying behaviors you can audit.

Partnering with Mika: onboarding in 30 days

Day 0–5: Manufacturing brief (QTPP→CQA→CPP), gap list, draft control strategy.
Day 6–15: Lab characterization runs; template/IVT scouting, endotoxin stress tests.
Day 16–30: Process lock for non-GMP or engineering run; draft eBR, raw-material alignment, PPQ plan outline.
Beyond: GMP slotting, PPQ, and clinical supply with a digital data room your team actually uses.

We call it Design → Data → Decision. It’s how we turn intent into inventory.

The market context: why this creates real ROI

The nucleic-acid economy is no longer a spike; it’s a platform—vaccines beyond COVID, in vivo editing, ex vivo engineered cells, cardiac and neuro targets, and protein replacement via mRNA. Across this stack, plasmid DNA & mRNA manufacturing is the throughput governor: the most expensive slips now happen between pDNA and IVT and inside endotoxin-heavy process trains. A microbial-first CDMO that ships clean pDNA and reproducible IVT is the difference between timelines that compress and programs that stall.

For sponsors, that translates directly to enterprise value: earlier INDs, earlier clinical signal, and fewer credibility dents during diligence. The budget saved on delays dwarfs line items for resin or caps. Critically, when plasmid DNA & mRNA manufacturing is architected to prevent LPS at the source—and analytics are tied to CQAs like supercoiled %, capping efficiency, and dsRNA—teams avoid “late learnings” that trigger rework, re-qualification, or worst-case reformulation.

From a portfolio view, plasmid DNA & mRNA manufacturing also unlocks strategic option value: pDNA template lots that are topology-stable feed multiple IVT campaigns; IVT methods with validated dsRNA reduction curves scale across new ORFs with minimal re-engineering. That’s how process knowledge becomes a compounding asset instead of a one-off expense.

Top 20 Plasmid DNA CDMO FAQ

1) What plasmid formats do you support?

Circular plasmids (2–20+ kb), minicircles, and linearized pDNA for IVT. We tailor host/vector, copy number, and selection (antibiotic or auxotrophy) to manufacturability and release needs.

2) Which host strains do you use?

Primarily E. coli chassis selected for low recombination (recA-), reduced endonuclease (endA-), and robust outer membranes. For endotoxin-sensitive programs we deploy strains + media that minimize LPS shedding upstream.

3) Can you run antibiotic-free selection?

Yes. Options include auxotrophy, operator/repressor systems, or post-segregational killing modules. We’ll propose antibiotic-free if GMP clearance or regional regs make antibiotics awkward.

4) What scales are available?

PD: 2–10 L
Engineering/GMP: ~50–2,000 L microbial. Larger via partner network with harmonized unit ops (AEX, MF/DF, membranes).

5) What yields should we expect?

Construct-dependent. We state yields as mg pDNA/L culture and % supercoiled (SC) post-DSP. During scoping we share historical ranges for similar size/origin/GC content and the levers (feed, temp, stress) that move them.

6) How do you protect supercoiled topology?

By design, not luck: upstream μ/DO control to build SC potential; bounded alkaline lysis (time/temp/NaOH/SDS, ±5–10 s); shear-budgeted pumps/filters; residence-time control in clarification; topology-aware polishing.

7) What are your standard CQAs and release tests?

Identity and topology (SC/OC/LIN) by gel or capillary electrophoresis; gDNA/RNA residuals by qPCR/RT-qPCR; endotoxin (validated LAL with interference controls); residual protein/solvent/salt; A260/280, A260/230; osmolality; bioburden/sterility. IVT-bound lots add linearization readiness and cut efficiency.

8) How do you control endotoxin (LPS)?

Prevention first: strain/media, temperature & shear envelopes, LPS-aware buffers/surfactants, segregated flow paths. Removal via membrane adsorbers/flow-through steps tuned to conductivity/pH. We trend LAL in-process, not only at CoA.

9) Do you offer linearization and template prep for IVT?

Yes. Restriction or enzymatic cassettes with end-verification by CE. We also condition conductivity/osmolality for IVT kits and document cut completeness and nicking profile.

10) Which purification strategies do you use?

Capture by anion exchange (AEX)—resin or membranes—then polish based on impurity fingerprint (RNA, gDNA, HCP, LPS) using mixed-mode or HIC. UF/DF (TFF) sets final buffer for IVT, transfection, or DP fill.

11) What documentation and QMS do you run?

Unified digital QMS (ALCOA+), electronic batch records, controlled specs/methods, and change control tied to QTPP→CQA→CPP. We supply a complete CMC data pack for tech transfer and audits.

12) How do you approach tech transfer?

Day-0 manufacturability brief; method harmonization; risk register; DoE or bracketing runs to lock CPPs; reference runs with predefined success criteria; live dashboards for your reviewers.

13) Do you support antibiotic removal and residual testing?

Yes. We avoid antibiotics where possible; when used, we define clearance strategy and quantify residuals per spec with validated methods.

14) What about host-cell DNA/protein residuals?

gDNA quantified by qPCR with construct-specific primers; HCP via ELISA or orthogonal LC methods. We show stepwise reduction curves (load-in vs. load-out) so you see where clearance actually happens.

15) How are deviations handled?

Risk-ranked impact assessment (CQA/CPP link), immediate containment, documented root-cause (5-Why/FMEA), CAPA, and transparent customer communication. Batch impact on %SC, LPS, residuals is explicitly analyzed.

16) How do you ship and store pDNA?

Cold chain per buffer and spec (commonly 2–8 °C; frozen for long holds). We validate container-closure, perform in-use stability, and include ship-backups and temp logs.

17) Can you provide non-GMP material first?

Yes—Tox/Research lots mirror the intended GMP train so the translational jump is minimal. Great for IVT optimization and formulation screening.

18) What regulatory frameworks do you align to?

GMP with ICH Q7/Q9/Q10 principles, USP/EP microbiology/endotoxin chapters, and nucleic-acid-relevant guidances. We fold your control strategy narrative into CMC sections.

19) How is pricing structured?

Typically per campaign and scale, with adders for linearization, expedited slots, extra analytics, or special raw materials. We quote unit cost vs. yield bands so your finance model can breathe.

20) What makes microbial CDMO work fail most often—and how do you avoid it?

Top causes: uncontrolled lysis exposure, shear in clarification, hidden LAL interference, weak EC control at AEX load, and missing topology analytics. We prevent with tight envelopes, in-process trending, and design choices that make supercoiled % and LPS predictable instead of aspirational.

Conclusion: Mika is Ready when you are

Whether you need pDNA templates that make IVT easy or end-to-end mRNA with dsRNA and capping locked down, Mika’s microbial systems are tuned for modern nucleic-acid programs. If your constraints are timeline, topology targets, or a dsRNA ceiling, we’ll convert them into a manufacturable plan—and we’ll ground the plan in plasmid DNA & mRNA manufacturing fundamentals so your analytics predict behavior, not just pass specs.

Mika Biologics — Microbial Systems. Beyond the cell. Built for GMP.

Want to learn about Biologic Nanoparticles? Visit our page -> Biologic Nanoparticles

When you’re ready to operationalize plasmid DNA & mRNA manufacturing with audit-ready documentation and measurable endotoxin prevention, we’re ready to start.

Email our team at info@mikabiologics.com