Week 1 HW: A color-changing wound-status patch made by cell-free synthetic biology

(1) What you would like to biomanufacture and why?

I would like to biomanufacture a smart wound dressing with a visible “traffic-light” readout. The path would collect a small amount of wound fluid and display green, yellow, or red on its outside surface to communicate whether the wound environment appears compatible with healing, needs monitoring, or may be persistently inflamed and at risk of delayed healing. The display turns the no so pretty wound into a geometric pattern that turns something painful into a beautiful path.

The core product would not diagnose infection on its own. Instead, it would report a meaningful and accessible biological condition: whether the wound fluid shows a combination of abnormally alkaline pH and high MMP-9 protease activity. Chronic or non-healing wounds often have persistent protease activity, including MMP-9, and elevated MMP-9 has been associated with poor healing outcomes. Excessive MMP activity can degrade extracellular matrix and pro-healing growth factors, keeping a wound in an inflammatory, non-healing state.

Product: “Wound Traffic Light” The product is a disposable, flexible wound patch with three layers:

Wound-contact layer A sterile, absorbent, non-adherent hydrogel or dressing material that manages wound moisture and collects wound exudate.

Sealed sensing layer A thin compartment containing freeze-dried, cell-free synthetic-biology reagents. Wound fluid reaches this sensing layer through a controlled membrane, but the reagents remain physically contained in the patch.

External display layer A large, easily read color window visible from outside the bandage.

The visual output would be:

Color Biological reading Meaning Green pH is in a healing-compatible range and MMP-9 activity is low The wound environment appears relatively favorable for healing Yellow One indicator is abnormal or changing Monitor the wound and check the dressing more frequently Red Both pH and MMP-9 indicate a persistently inflammatory environment Possible delayed-healing risk; the wound should be assessed by a clinician, especially if symptoms worsen

I am interested in this project because wound care has a communication problem. A normal bandage hides the wound, so a person often has to remove it to inspect it. That can be inconvenient, disruptive, and difficult for people managing chronic wounds, especially diabetic foot ulcers, pressure injuries, or post-surgical wounds. The patch would turn subtle biochemical information into a simple visual language that can be understood without batteries, specialist equipment, or a smartphone:

The target users would be people with chronic wounds, home-care patients, caregivers, wound clinics, and eventually hospitals. The initial use case would be a chronic diabetic foot ulcer, where early recognition of a stalled healing trajectory could be especially valuable. MMP-9 is a strong candidate biomarker in this context because higher MMP-9 activity and elevated MMP-9/TIMP-1 ratios have been linked with poorer healing outcomes in diabetic foot ulcers and other non-healing wounds.

(2) What living system would you use?

Proposed chassis: freeze-dried cell-free biosensor I would use a cell-free synthetic-biology system rather than living engineered bacteria directly in the wound dressing.

A cell-free system uses the molecular machinery of cells—such as enzymes, ribosomes, transcription factors, and metabolic cofactors—outside intact living cells. These components can be freeze-dried into a stable reaction pellet or film, then activated by liquid. In this project, wound exudate would rehydrate the sensing chemistry when it reaches the sealed sensing chamber.

This is a better first chassis than live cells because:

The wound is a sensitive environment, so I would avoid applying engineered microorganisms directly to damaged tissue. The sensing chemistry can be enclosed inside the patch. Freeze-dried cell-free systems are compatible with paper, hydrogel, and membrane-based formats. The final output can be a direct color signal rather than a measurement that requires a laboratory instrument.

High-level biological logic The sensor would use a two-input AND gate:

Red output=alkaline pH signal∧high MMP-9 activity

The system would be designed so that one concerning signal alone produces yellow, or no strong alarm, while the combination produces red.

pH-responsive module (Wound-fluid pH): Detects a shift toward an alkaline wound environment Protease-responsive module (Active MMP-9): Uses an MMP-9-cleavable peptide or related recognition element to activate a reporter pathway Logic module (Both inputs): Creates a red output only when both signals meet the threshold Color reporter (Activated pathway): Produces or reveals a stable visible pigment Saturation strip (Moisture/exudate volume): Indicates when the absorbent dressing should be changed.

What goes in, what happens, and what comes out Inputs:

• Wound exudate entering through a fluid-control membrane.
• The exudate contains water, salts, proteins, enzymes, and potential inflammation-associated biomarkers.
• The sensing layer begins as a dry, sterile cell-free reagent film.

Biological processes:

• Wound fluid hydrates the freeze-dried reaction.
• A pH-sensitive component measures whether the wound fluid has shifted into an alkaline range.
• An MMP-9-responsive element detects high protease activity.
• The two signals are combined through a molecular logic circuit.
• mIf both inputs are present above preset thresholds, the circuit activates a chromogenic output.

Outputs:

• A visible color change viewed from outside the dressing.
• Green/yellow/red is a deliberately simple user interface rather than a quantitative lab test.
• A moisture strip changes color when the dressing absorbs too much exudate.

Why pH and MMP-9? pH is useful because it is fast and easy to measure. Wound pH tends to become more acidic as healing progresses, whereas chronic non-healing wounds are often slightly alkaline. However, pH alone is not specific enough: normal early inflammation, blood, dressing conditions, and local environment can change it.

MMP-9 adds biological specificity. Controlled protease activity is normal during healing, but persistent or excessive MMP-9 is concerning because it can degrade the extracellular matrix and growth factors required for repair. MMP-9 activity has been reported to be several times higher in non-healing than healing wounds, with some reports noting substantially larger differences between chronic and acute wounds.

The important design decision is therefore not to label a single high MMP-9 measurement as “bad.” The patch would interpret persistent, combined signals as concerning. A wound has some inflammation during normal early healing; the problem is an inflammatory state that does not resolve.

(3) How would you use automation?

Prototype automation: Opentrons OT-2 liquid handler For prototype development, I would use an Opentrons OT-2 desktop liquid-handling robot. It is well suited to preparing many small cell-free reactions with controlled variation in reagent concentrations, pH, enzyme inputs, and material formulations. The OT-2 can perform liquid transfers from 1–1,000 µL and supports plate filling, serial dilution, normalization, and sample pooling.

The automation is important because this project has a large design space. For example, I may need to test:

8 pH conditions. 8 MMP-9 activity conditions. 4 reporter concentrations. 3 sensing-layer materials. 3 replicates. That is: 8×8×4×3×3=2,304reactions

This is too repetitive and error-prone to prepare manually. An OT-2 with an 8-channel pipette can distribute reagents across a 96-well plate in seconds to minutes depending on the transfer protocol; published work has also specifically explored semiautomated production of cell-free colorimetric and fluorescent biosensors using the OT-2.

Manufacturing automation at larger scale The OT-2 is a prototyping tool, not the final manufacturing tool. At production scale, I would use:

• Roll-to-roll coating or printing to deposit cell-free reagents into discrete sensing spots on a flexible film.
• Automated freeze-drying or controlled low-humidity drying to stabilize the cell-free reagents.
• Automated lamination to assemble the absorbent layer, membrane, sensor layer, transparent window, and adhesive backing.
• Vision systems for quality control, verifying patch dimensions, reagent spot placement, baseline color, and seal integrity.
• Batch release testing using standardized pH and protease controls.

This separates design automation from manufacturing automation: the OT-2 helps discover the best chemistry, while roll-to-roll manufacturing makes millions of uniform patches.

(4) How much would you need to make, and how long would it take?

Scale and timeline For an initial pilot, I would make 10,000 wound patches for 500 users over two weeks.

Each person would use one patch per day:

500 users×14 days=7,000 patches I would produce an additional 3,000 patches for quality testing, failed units, training, and backup supply:

7,000+3,000=10,000 patches If each patch is 10 cm × 10 cm, it has an area of 100 cm². For 10,000 patches, this equals about 100 m² of dressing material.

Each patch would contain about 50 µL of cell-free sensor mixture:

10,000×50 μL=0.5 L With extra material for manufacturing losses, I would prepare around 0.6 L of sensor mixture.

Time estimate To develop the sensor, I would test about 2,304 combinations of pH, MMP-9 activity, material type, and reporter concentration. Since one 96-well plate holds 96 tests, this would require 24 plates.

An Opentrons OT-2 could prepare around four plates per day, so the first automated screen would take roughly six working days. Including repeats, data analysis, and material testing, development would likely take 2–3 weeks.

For manufacturing, a small automated line could make and package about 1,000 patches per day:

10,000÷1,000=10 days Including sensor preparation, drying, assembly, packaging, and quality control, producing the first 10,000-patch batch would take approximately 3–4 weeks once the design is finalized.

At larger scale, the physical patch-making process could be much faster using roll-to-roll manufacturing. The main bottlenecks would probably be preparing stable sensor reagents, drying them correctly, sterile packaging, and quality testing—not cutting the patches themselves.

(5) What I don’t know yet

1. Can an MMP-9 sensor work reliably in real wound fluid? This is the most important technical uncertainty. Wound fluid is not a clean buffer. It contains salts, blood components, proteins, bacteria, proteases other than MMP-9, inflammatory molecules, and variable amounts of fluid. I would need to determine:

• Whether the recognition element is selective enough for MMP-9 activity.
• Whether other wound proteases cause false-positive signals.
• Whether blood, pus, or common topical wound treatments interfere with the color output.
• Whether the sensitivity matches clinically meaningful MMP-9 activity levels.

2. What thresholds best separate normal healing from stalled healing? A major risk is designing a patch that flags normal early inflammation as dangerous. MMPs and inflammation are necessary parts of ordinary healing; the key variable is their amount, duration, and balance with inhibitors such as TIMPs.

I would need wound-fluid samples or carefully designed simulated wound fluid to establish:

• The pH threshold for green, yellow, and red.
• The MMP-9 activity threshold for concern.
• Whether the patch should integrate a signal over hours or days.
• How thresholds differ between acute surgical wounds, diabetic foot ulcers, burns, and pressure injuries.

My initial hypothesis is that trend detection will be more valuable than a one-time reading. A wound that changes from yellow to green may be improving even if it is not yet fully normal; a wound that stays red over several dressing changes may deserve attention.