Living cells can already sense their surroundings, but turning that information into complex decisions has remained a major challenge.

Researchers can program human cells to perform the same binary calculations as a simple digital circuit.

By building compact genetic logic systems inside single cells, researchers moved programmable cell therapies a step closer to reality.

Splicing as a switch

The circuits rely on a process that already takes place inside cells every day. When a cell switches on a gene, the cell copies it into a working message.

The cell then trims that message, cutting out filler and joining the useful pieces into a finished instruction. That editing step is called splicing.

Normally the cell joins pieces of a single message. The team instead used trans-splicing, a natural but less common version.

Within this process, two separate messages are fused into one. Each input gene produces half of the finished instruction, and the cell joins the halves only when both inputs are active.

That arrangement behaves like an AND gate, an electronic component that produces an output only when two inputs are switched on at once.

Turning RNA into logic

The researchers wired the output to a red fluorescent protein, so a successful calculation showed up as a red glow.

Under the microscope, cells lit up at least 45 times brighter when both inputs were active than in any other state.

Trans-splicing is naturally less efficient than the everyday kind, so getting a clean signal took careful tuning of the genetic parts.

The design comes from the synthetic biology group of Dr. Lior Nissim at the Hebrew University of Jerusalem (HUJI) with Keren Roas as the study’s first author.

The team borrowed its key part rather than inventing it. Researchers had explored trans-splicing as a way to repair faulty genes.

Before this study, no one had used it to build computing circuits in mammalian cells.

Math in one layer

The group first built a half adder, the basic unit that adds two binary digits and reports a sum along with a carry, the extra digit that moves into the next column when the first exceeds its limit.

Then came the harder target, a full adder, which takes three input digits and again returns a sum and a carry.

The researchers assigned each output a different color, red for the sum and cyan for the carry, and the circuit produced the correct pair across all eight possible input combinations.

Earlier attempts to run a full adder in human cells took a different, unsuccessful route.

One effort built the calculation from nine separate populations of engineered cells that passed chemical messages between them.

Nissim’s circuit runs the whole computation inside one population, with every reaction happening at the same time in a single step.

From error to output

The team’s most elaborate circuit worked as a multiplexer, a switch that picks one of several inputs and passes it to a single output line.

It follows instructions from separate selector signals. The version they built handled three input channels, each reporting through its own color of fluorescent protein.

A switch like this has a quirk. Two selector signals will encode four settings, but only three are needed to choose among three inputs.

This leaves one combination unused. Instead of wasting it, the researchers assigned that state its own job, a dedicated output called a Selector Overload Status.

The unused signal

When both selectors switched on, the circuit lit a fourth color, yellow, that none of the ordinary channels could produce.

The authors suggest that spare output could run a built-in self-test, flag an abnormal cell state such as a tumor, or trip a safety mechanism that shuts the circuit down.

It is the first mammalian circuit of its kind to include this extra operating state. That extra slot gives designers a place to build in monitoring without adding a separate device.

Future research paths

The circuits so far run in a dish, inside human kidney cells. The outputs have mostly been glowing proteins used as convenient readouts.

In one test, however, the team swapped in a working therapeutic signal, a human immune-stimulating protein called IL-15, that the circuit produced only when both inputs were active.

Nissim’s group has worked on RNA-based circuits like these for years. An earlier study used them to single out cancer cells and switch on immune-boosting outputs.

Packing more of that decision-making into fewer genetic parts is what the new design adds. Human cells can run multi-step logic on their own.

The work shows a full adder and a working multiplexer inside a single layer, with no separate cell populations passing signals and no heavy drain on the cell’s resources.

Using fewer parts helps. A circuit that uses less genetic material is easier to deliver and less likely to overload its host.

Smarter cell therapies

The practical target is disease that hides behind combinations of signals rather than one clear marker.

Cancers and autoimmune disorders often reveal themselves through a particular mix of molecular cues, and a treatment keyed to a single cue can strike healthy tissue by mistake.

Cell therapies already build upon this idea. One approach engineers immune cells to attack only when two tumor markers appear together.

This spares cells that carry just one, and circuits that add up several markers at once could sharpen that targeting further.

From proof to practice

Real therapies remain theoretical. Researchers still have to deliver many DNA pieces at once.

Integrating them stably into a cell’s own DNA is an unsolved problem the authors flag for later work. Even so, the central result holds.

Complex, calculator-style computation now fits inside a single human cell, giving smart diagnostics and targeted therapies a building block they have long lacked.

The study is published in Nature Communications.

NOTE – This article was originally published in Earth and can be viewed here