Where did the glucose come from?

Currently there are two distinct views on the origin of life. The majority of scientists think that life arose in a prebiotic soup of complex organic molecules. Most of them think this "warm little pond" was the ocean (!) and most of them have bought into the stories about asteroids and comets delivering complex organic molecules to create a soup of amino acids and sugars. Presumably, all the earliest forms of life had to do was to join together the amino acids to make proteins and hook up the nucleotides to make RNA. The energy for these reactions was derived from breaking down all the glucose in the sweet ocean.

The other view is the one supported by the majority of experts and people who make a serious study of the origin of life. It proposes a "metabolism first" view where the initial products of non-enzymatic reactions were small molecules like pyruvate and glycine and gradually pathways evolved to make the more complex molecules like glucose, more complex amino acids, and nucleotides. The energy for these reactions came from proton gradients in the pores of hydrothermal vents. In this view, life arose in tiny compartments, where concentrations could be significant, then spread to the ocean. [Changing Ideas About The Origin Of Life] [Was the Origin of Life a Lucky Accident?] [Why Are Cells Powered by Proton Gradients?] [Metabolism First and the Origin of Life].

In biochemistry courses we distinguish between catabolic pathways where something is broken down or degraded (= catabolism) and anabolic pathways where complex molecules are synthesised (= biosynthesis, anabolism). Glycolyis—the breakdown of glucose to pyruvate—is the classic example of catabolism. Gluconeogenesis—the synthesis of glucose from pyruvate—is the classic anabolic (biosynthesis) pathway. Similar contrasting pathways exist for amino acid metabolism and nucleotide metabolism.

Simple organisms, like bacteria, concentrate on biosynthetic pathways. Some of them can grow and thrive in the absence of any external organic molecules. They are called chemoautotrophs. Some bacteria can use external organic molecules as carbon sources and some of them absolutely require them. However, the main metabolic pathways in bacteria are biosynthetic and the enzymes that catalyze these reactions are the most ancient enzymes. This strongly suggests that the most primitive pathways were biosynthetic pathways and not catabolic pathways.

Human biochemisty—as it is taught in most universities—is dominated by catabolic pathways. Students are given the impression that the most important pathways are those that break down complex molecules for energy. According to this human-centric view, glycolysis is the most important pathway and everything is treated as a potential food source.¹ Students can often graduate with a major in biochemistry without ever understanding where the glucose comes from.

I think this explains some of the strange views of those who work on origin of life problems. They begin with the prejudice that glycolysis is the most primitive pathway and not gluconeogenesis because they were taught biochemistry from a human perspective where the synthesis of glucose (glucoenogenesis) is often ignored or downplayed.

In May 2014 I posted a commentary on a paper by Keller et al. (2014) that illustrated this bias [see More primordial soup nonsense]. The authors of that paper examined the possibility that nonenzymatic reactions could have been the first step leading to the evolution of metabolic pathways. They presented evidence that iron could catalyze many of the reactions in the glycolytic pathway and the pentose phosphate pathway. However, both of those pathways require glucose and it's easy to show that the necessary sweet ocean is impossible.

The senior author of that paper, Markus Ralser of the University of Cambridge (UK), participated in the discussion and so did Bill Martin, one of the leading proponents of metabolism first. Ralser quickly abandoned the thread only to pop up again nine months later with ...

Now its quite some time ago so that debate could cooled down a bit; so I would like to comment on this blog. All living cells use a conserved network of biochemical reactions to catalyse their metabolic reactions. This network, called the metabolic network, had an origin in early evolution, but this origin not understood. One of the main questions in this field is about how did the first catalysts about two conserved pathways, called glycolysis and the pentose phosphate pathway, looked like. Were these RNA molecules? Or were these minerals or other molecules? That's what we test in our paper. We joined up with Earth Scientists at the University of Cambridge. They told us what they think was abundantly available in the Archean oceans. And we tested systematically with very advanced mass spectrometry methods whether these molecules can catalyse reactions observed within the most conserved part of metabolism. We got a hit in the metal ions. And that's very remarkable, because it shows that the first catalysts capable to catalyse the reactions as found now in modern cells central metabolism, did not need to have a complex enzyme fold structures for a start. This makes it much easier to explain the origin of the metabolic network. Nobody, really nobody, claims here that the ocean was a soup full of ribose 5-phosphate. But its fun to read Larrys calculation, but I have to admit it would not have hurt him to read a bit about what is known and not known about the origin of the metabolic network structure before starting shouting out loud against the work of others.

Ralser eventually agreed that gluconeogenesis was important but he justified his paper's emphasis on glycolysis by saying ...

Agreed, in fact we also write clearly in the paper that gluconeogeneis was probably before glycolysis - the detail is important also here however: Its the same catalysts for most of the reactions that allows both glycolysis and gluconeogenesis. Without glycolytic enzymes (and their precursors), cells couldn’t do gluconeogenesis either. So its chemically and catalytically not two different pathways.

That's not incorrect but it kinda misses the point. If you know that biosynthesis of glucose is required before you can degrade it then why not look at nonenzymatic reactions that could lead to the biosynthesis of glucose instead of reactions that break it down? Why emphasize glycolysis?

Ralser also claims that the concentration of glucose (or glucose-6-phosohate) in the primitive oceans is irrelevant because they were looking at iron-catalyzed reactions. He said,

For the sugar phosphates, we used 7.5uM because that is a good concentration to conduct the experiment: Its lower than its concentration in cells, but can still be perfectly detected on our masspecs. Catalysis is not limited by substrate concentration, Fe(II) can perfectly catalyse the reaction even at much lower concentration. So this result is concentration independent.

He also complained about the references I included in my post to explain the metabolism first scenario. I criticized his paper for not presenting alternative views on the origin of life in the introduction to his paper. He responded by pointing out that this wasn't a review paper and, besides, none of the papers I listed discussed the origin of glycolytic enzymes!

Ralser's group has just published another paper that covers much of the same ground (Keller et al., 2016). Here's the abstract.

Little is known about the evolutionary origins of metabolism. However, key biochemical reactions of glycolysis and the pentose phosphate pathway (PPP), ancient metabolic pathways central to the metabolic network, have non-enzymatic pendants that occur in a prebiotically plausible reaction milieu reconstituted to contain Archean sediment metal components. These non-enzymatic reactions could have given rise to the origin of glycolysis and the PPP during early evolution. Using nuclear magnetic resonance spectroscopy and high-content metabolomics that allowed us to measure several thousand reaction mixtures, we experimentally address the chemical logic of a metabolism-like network constituted from these non-enzymatic reactions. Fe(II), the dominant transition metal component of Archean oceanic sediments, has binding affinity toward metabolic sugar phosphates and drives metabolism-like reactivity acting as both catalyst and cosubstrate. Iron and pH dependencies determine a metabolism-like network topology and comediate reaction rates over several orders of magnitude so that the network adopts conditional activity. Alkaline pH triggered the activity of the non-enzymatic PPP pendant, whereas gentle acidic or neutral conditions favored non-enzymatic glycolytic reactions. Fe(II)-sensitive glycolytic and PPP-like reactions thus form a chemical network mimicking structural features of extant carbon metabolism, including topology, pH dependency, and conditional reactivity. Chemical networks that obtain structure and catalysis on the basis of transition metals found in Archean sediments are hence plausible direct precursors of cellular metabolic networks.

They looked at nonenzymatic reactions—catalyzed by iron—that cause the breakdown of complex sugars and sugar-phosphates to more simple compounds. They tested the reactions under three different pH conditions to produce this figure. The read arrows represent the reactions they observed with the direction indicated by the arrowhead.

The main point of the paper is to show that, "the existence and specificity of these reactions imply that pathways of central carbon metabolism could directly originate from pre-enzymatic metal/sugar phosphate chemistry." It's clear from the figure that there's no pathway leading from a simple molecule, such as pyruvate, to glucose. These pathways are supposed to be examples of how central carbon metabolism evolved. I'll ask the same question I asked before.

Where did the glucose come from?

---

1. These courses end up being courses on fuel metabolism and nutrition rather than courses on fundamental biochemistry. American lecturers justify them on the grounds that they are good preparation for the MCATs and in other countries they cater to what the students want to learn rather than what they should learn.

Keller, M.A., Turchyn, A.V. and Ralser, M. (2014) Non‐enzymatic gycolysis and pentose phosphate pathway‐like reactions in a plausible Archean ocean. Molecular Systems Biology 10:725 [doi: 10.1002/msb.20145228]

Keller, M.A., Zylstra, A., Castro, C., Turchyn, A.V., Griffin, J.L., and Ralser, M. (2016) Conditional iron and pH-dependent activity of a non-enzymatic glycolysis and pentose phosphate pathway. Science Advances, 2(1), e1501235. [doi: 10.1126/sciadv.1501235]