A kitchen sink has an overflow outlet to prevent water from spilling onto the floor if the tap is left on. A toilet has a valve that shuts off input when a certain level is reached. A bacterium has both. It can maintain a steady supply of materials by controlling both inputs and outputs, with automated systems that mimic what engineers create.

To regulate its supply of pyrimidines (building blocks of DNA), bacteria employ both negative feedback to quench oversupply (like the toilet valve), and overflow to excrete excess input (like the kitchen sink). This redundancy provides homeostasis under varying conditions — when raw materials are abundant, and when they are scarce.

Five Princeton scientists publishing in Nature unveiled this phenomenon in E. coli bacteria. “These upstream and downstream regulatory mechanisms work in concert to balance speed, efficiency and robustness,” the authors write (emphasis added).

We designate the general mechanism, whereby feedback inhibition of a downstream pathway step leads to excretion of a pathway intermediate or by-product, directed overflow metabolism (Fig. 4a). Such overflow is triggered by excessive biosynthetic pathway flux and is carried out by a degradation pathway sensitive to levels of an accumulating biosynthetic intermediate. It is analogous to overflow in central carbon metabolism …. it forms a choke point that instead directs the enzyme’s normal substrate towards by-product formation and excretion (Fig. 4b). In the case of pyrimidine biosynthesis, the cooperative inhibition of UMP kinase by UTP renders the directed overflow mechanism exquisitely precise in controlling UTP and CTP levels.

They didn’t know about the overflow mechanism till they mutated bacteria to turn off the feedback mechanism. To their surprise, homeostasis of primidines was maintained, even though the mutants were unable to grow as fast. Looking deeper, they found the “directed overflow” pathway. The “input valve” didn’t stop intermediates from accumulating, so the “overflow outlet” came into play. How was it executed? An “evolutionary conserved” phosphatase enzyme of previously unknown function converts the excess to uracil, which can be excreted. So another “exquisitely precise” mechanism was found — in bacteria, the simplest of organisms.

And that’s just one of thousands of mechanisms in bacteria. The paper begins,

The metabolic network of E. coli consists of approximately 1,000 metabolites connected by around 2,000 enzyme-catalysed reactions. Control of metabolite concentrations and fluxes occurs through the regulation of enzyme concentrations, activities and substrate occupancies. Metabolic control analysis provides a systematic framework for investigating the impact of particular enzymes on cellular metabolic activities. Studies modulating the concentrations of enzymes suggest that control of metabolic flux is frequently distributed across multiple enzymes, with demand for end product often having a key role in controlling biosynthetic fluxes.

Feedback inhibition (negative feedback) in bacteria was known before. What’s new was the directed overflow mechanism. It takes more energy to convert the excess supply to uracil when the feedback inhibition is broken, but it gives the bacterium a backup plan to prevent runaway supply, which would be toxic to the cell. So, there’s regulation both at the front end and the back end — supply and demand.

Thus, pyrimidine homeostasis involves two strategies for regulation. The canonical feedback architecture contributes to metabolic efficiency by decreasing unnecessary de novo flux. Directed overflow metabolism provides end-product homeostasis by diverting excess flux to uracil, thereby ensuring end-product homeostasis in response to altered availability of the full range of pathway substrates and intermediates. These upstream and downstream regulatory mechanisms work in concert to balance speed, efficiency and robustness.

Commenting on this discovery, Athel Cornish-Bowden wrote in Nature that it’s like an engineered system:

In bacterial cells, the overproduction of metabolites is normally avoided by mechanisms that are similar in principle to control systems in engineering. In this issue, Reaves et al.1 (page 237) report what happens in mutant bacteria that lack a supposedly essential control mechanism to prevent excessive production of pyrimidine nucleotides — metabolites that act as building blocks for the synthesis of genes, but which are potentially toxic if allowed to accumulate. Instead of observing the expected accumulation, the authors discovered a mechanism in which excess nucleotides are eliminated. In so doing, they identified a plausible role for an enzyme whose physiological function had hitherto been unknown.

The sink and toilet analogy is given by Cornish-Bowden in a paragraph that twice compares the mechanisms to “engineered systems”:

We all know that a kitchen sink is liable to overflow if the tap is left on with the plughole blocked. In most domestic sinks this danger is averted, at least partially, by having an overflow outlet near the top. But in more elaborate engineered systems, such as the domestic toilet, an overflow is avoided by means of negative feedback: as soon as the water in the tank reaches a certain level, the inflow is switched off. Bacterial metabolism is in many ways similar, in that feedback mechanisms prevent the potentially harmful build-up of metabolites. The great explosion of interest in biological regulatory mechanisms in the 1960s followed the realization that negative feedback in metabolism operates in the same way as in engineered systems, by allowing the output of an end product to match demand.

The bacterium’s mechanisms, though, are worthy of more elegant analogies. The “supply and demand” comparison to economics is better:

Metabolic regulation is most easily analysed in economic terms of supply and demand, especially given that the primary function of feedback inhibition is to regulate metabolite concentrations, rather than fluxes. Biological responses to the demand for metabolic end products are common in many systems, thus explaining the occurrence of cooperative feedback inhibition, such as that by CTP in pyrimidine biosynthesis. But the degradation of CTP to uracil observed by Reaves et al. is not a response to demand for uracil, because there is no particular demand for it. Instead, this degradation is a response to the excessive supply of nucleotides, and allows the concentrations of CTP and its biosynthetic precursors to be kept essentially constant when demand for CTP is low….

Could Evolution Produce Backup Regulatory Mechanisms?

The Reaves paper does not mention evolution at all. Cornish-Bowden tries to sneak it in twice. In the first instance, it’s hard to see what evolution has to do with it:

From an evolutionary perspective, what matters in metabolism is not so much the ‘overflow’ that occurs in the absence of biological feedback controls, but the build-up of compounds that arises when the requirement for a metabolite is too low to be handled adequately by feedback inhibition….

OK, so? If evolution only cares about oversupply, it already had feedback inhibition (how that arose is not indicated). Cornish-Bowden merely assumes that evolution would take care of the situation when it doesn’t work, because demand is too low. Here, he grants magical powers to evolution:

The CTP-production pathway is therefore an example of a system in which regulation by demand for end product occurs side-by-side with regulation by degradation and excretion of excess end products. Evolution cannot have generated and conserved the excretion mechanism purely to compensate for the artificial deletion of feedback in experiments, so this pathway must represent a back-up strategy for physiological states in which demand falls below the finite range within which feedback is effective. If so, then we should expect to find examples of the same sort of behaviour in other pathways.

This is just one more example of evolution being treated as a magic wand that generates “exquisitely precise” mechanisms on demand. As we have explained many times, natural selection has no foresight. It cannot see a demand, and generate a mechanism to supply it through blind, mindless processes. It is hopeless to imagine mutation and selection coming up with either of these mechanisms. Cornish-Bowden uses evolution as a crutch.

Whenever we see exquisitely precise mechanisms that regulate other mechanisms, we know from uniform experience with “engineered systems” that they were designed. It’s because of this uniformity of experience that “we should expect to find examples of the same sort of behaviour in other pathways.” If “the realization that negative feedback in metabolism operates in the same way as in engineered systems” led to a flurry of discoveries since the 1960s, history shows that a de facto intelligent design perspective (thinking like an engineer) will motivate many more such discoveries without Darwin’s help.

Photo credit: Fazimoto/Flickr.