Systematic CRISPRi screening reveals genetic modulators of E. coli isoprenoid production

This study employs a systematic CRISPRi screening approach to identify 31 genetic modulators across diverse metabolic pathways in *E. coli* that significantly influence lycopene production, thereby revealing novel targets for rebalancing cellular metabolism to enhance isoprenoid yields.

The Gist

Imagine you are trying to bake the world's most delicious, bright red cake (which, in this scientific story, is a molecule called lycopene). You have a very talented baker (the bacterium E. coli), but the baker is also trying to build a house, fix a car, and write a novel all at the same time. Because the baker is so busy with these other tasks, the cake doesn't turn out as big or as red as you'd like.

This paper is about a team of scientists who decided to figure out exactly which of the baker's other tasks were stealing the ingredients needed for the cake, and how to stop them without firing the baker.

Here is the story of how they did it, broken down into simple steps:

1. The Problem: The Baker is Overworked

Lycopene is a valuable red pigment used in food and cosmetics. Making it inside bacteria is a great idea, but it's "expensive" for the bacteria. It uses up a lot of the sugar and energy the bacteria need just to survive and grow.

• The Analogy: Think of the bacteria's cell as a busy factory. The "lycopene line" is a new, high-priority project. But the factory is also running a "growth line" (making more bacteria) and a "maintenance line" (fixing cell walls). The new project is stealing raw materials from the other lines, causing the whole factory to slow down or produce a poor-quality cake.

2. The Tool: The "Volume Knob" (CRISPRi)

Usually, if you want to stop a machine in a factory, you might smash it with a hammer (knock out the gene). But that's too risky; if you smash the wrong machine, the whole factory shuts down.

• The Innovation: The scientists used a tool called CRISPRi. Think of this not as a hammer, but as a remote control with a volume knob. Instead of destroying a gene, they could turn its volume down to 10%, 50%, or 90%. This allowed them to test exactly how much a specific task needed to be slowed down to help the cake without stopping the factory entirely.

3. The Experiment: The "Taste Test" of 180 Tasks

The scientists created a massive library of these "volume knobs." They picked 180 different jobs the bacteria does—like making fatty acids (oil), amino acids (protein blocks), or dealing with stress—and assigned a specific knob to each one.

• The Setup: They put these bacteria into 96-well plates (like a giant egg carton). They turned the volume down on one job in each little cup.
• The Timing Trick: They realized that when they turned the volume down mattered. If they turned it down too early, the bacteria died. If they waited too long, the cake wasn't made. They found the "sweet spot" was turning the knobs down after the bacteria had been growing for 4 hours.

4. The Discovery: What Happened When They Turned the Knobs?

After 48 hours, they looked at the results. They measured two things:

1. How much red cake (lycopene) was made.
2. How many bacteria were alive (biomass).

They found 31 "winners"—jobs that, when turned down, made the cake redder and bigger.

• The Surprises:
  • The "Oil" Leak: They found that turning down the genes for making fatty acids (oils) helped. It was like plugging a leak in the fuel tank; the energy that was leaking out as oil was now flowing into the cake.
  • The "Stress" Relief: They found that turning down the genes for "stress response" (how the bacteria panics when things go wrong) actually helped. It seems the bacteria were wasting energy worrying about stress instead of baking the cake.
  • The "Protein" Overload: In their specific lab conditions (where the bacteria were eating a rich soup of amino acids), the bacteria didn't need to make their own proteins. Turning down the genes for making amino acids saved energy for the cake.

5. The "Villains": What Not to Touch

They also found jobs that, if turned down, made the cake worse.

• The Core Engine: If they turned down the genes for the "TCA cycle" (the bacteria's main energy generator) or the specific enzymes that make the cake ingredients, the cake production crashed. This makes sense; you can't turn off the engine and expect the car to move.

The Big Picture

This paper is a roadmap. Before, scientists were guessing which genes to tweak to make better bacteria. Now, they have a map showing exactly which "volume knobs" to turn to get the most red cake.

Why does this matter?

• Efficiency: It shows us how to make bacteria work smarter, not harder.
• New Products: This method can be used for any valuable chemical, not just red cake. Whether it's biofuels, medicines, or perfumes, we can now systematically find the best way to rewire the bacteria's factory.
• The Future: The scientists suggest that in the future, we could use computers and AI to look at this map and figure out the perfect combination of volume knobs to turn, creating a "super-baker" that produces massive amounts of valuable products.

In short: They taught bacteria to stop multitasking so poorly and focus on making the red stuff, using a remote control to gently nudge their other habits into the background.

Technical Summary

1. Problem Statement

Isoprenoids are high-value natural products with applications in therapeutics, cosmetics, and biofuels. Microbial biosynthesis in Escherichia coli is a sustainable alternative to extraction, but production is often limited by a substantial metabolic burden. The synthesis of isoprenoids (like lycopene) diverts significant carbon flux and cofactors (NADPH, ATP) from central metabolism, creating competition for precursors (pyruvate, glyceraldehyde-3-phosphate, acetyl-CoA) and causing toxicity from intermediate accumulation.

Current metabolic engineering strategies often rely on rational, localized tuning of specific pathway genes (e.g., the MEP pathway). However, optimizing yields requires a systems-level understanding of how global metabolic networks and regulatory pathways interact with the biosynthetic pathway. There is a lack of high-throughput methods to systematically map the relationship between gene expression perturbations across the entire genome and biosynthetic pathway yield.

2. Methodology

The authors developed a high-throughput, next-generation sequencing (NGS)-coupled CRISPR interference (CRISPRi) screening platform to identify genetic modulators of lycopene production.

• Chassis Strain Engineering:

  • Started with E. coli K-12 MG1655 containing a genomically integrated, tetracycline-inducible dCas9.
  • Introduced the lycopene biosynthesis operon (crtEBI and idi) from Pantoea agglomerans on a plasmid.
  • Optimization: Used Oligonucleotide Recombineering followed by Bxb-1 Integrase Targeting (ORBIT) to replace native ribosome binding sites (RBS) of the MEP pathway genes dxs and dxr with high-efficiency sequences. This created strain DD2, which served as a high-yield chassis comparable to previously evolved strains.

• CRISPRi Library Design:

  • Designed a library targeting 180 genes spanning central carbon metabolism, amino acid biosynthesis, fatty acid metabolism, isoprenoid pathways, and stress responses.
  • Included both fully complementary sgRNAs (strong repression) and mismatch variants (7 mismatches) to allow for graded repression levels.
  • Total library size: 359 guides (plus non-targeting controls).

• Screening Protocol:

  • Induction Timing Optimization: Determined that inducing CRISPRi 4 hours after inoculation (during early exponential growth) yielded the strongest phenotypic effects without causing immediate lethality or selecting for "escaper" mutants.
  • High-Throughput Assay: Cultures were grown in 96-deep-well plates. Lycopene yield was quantified via acetone extraction and absorbance at 475 nm, normalized to optical density (OD600) to account for biomass differences.
  • NGS Integration: DNA was extracted from each well, and sgRNA cassettes were amplified with unique barcodes (well-specific and plate-specific). Pooled amplicons were sequenced on an Illumina MiSeq to map sgRNA abundance to specific wells and phenotypes.

3. Key Contributions

• Methodological Innovation: Established a scalable, 96-well plate-based CRISPRi screening workflow coupled with NGS (CRISPRi-seq) specifically for metabolic engineering. This allows for the simultaneous assessment of gene repression effects on both growth and metabolite yield.
• Strain Development: Created a robust, high-yield E. coli chassis (DD2) with optimized translation of MEP pathway genes, providing a sensitive background for detecting subtle metabolic changes.
• Systematic Mapping: Performed the first broad-scale screen (180 genes) linking global metabolic perturbations to isoprenoid yield, moving beyond the traditional focus on the MEP pathway alone.

4. Key Results

The screen identified 31 genes whose repression significantly altered lycopene yield (18 increased yield, 13 decreased yield).

• Genes Increasing Lycopene Yield (Positive Modulators):

  • Fatty Acid Biosynthesis: Repression of accA and accD increased yield but severely reduced biomass, suggesting a trade-off.
  • Central Carbon Metabolism: Repression of TCA cycle genes (aceE, aceF, acnA, sdhD) and fermentation genes (pflB) increased yield. This suggests redirecting flux away from respiration/overflow metabolism toward lycopene precursors.
  • Amino Acid Biosynthesis: Repression of branched-chain amino acid genes (ilvD, leuC) and chorismate pathway genes (aroK, aroF) increased yield. This is likely because LB media provides excess amino acids, making endogenous synthesis a wasteful drain on resources.
  • Phospholipid Biosynthesis: Repression of dgkA and pgsA increased yield.
  • Stress Response: Repression of SOS response genes (umuC, uvrB, uvrD) and the stringent response protease clpP increased yield.

• Genes Decreasing Lycopene Yield (Negative Modulators):

  • Isoprenoid Pathway: As expected, repression of core MEP genes (dxs, dxr, ispA) drastically reduced yield.
  • TCA Cycle: Repression of sdhB, sdhC, sucB, and acnB decreased yield, likely due to reduced ATP/NADPH availability.
  • Stress Response: Repression of rpoS (sigma factor) and dksA decreased yield, likely due to increased oxidative stress degrading the oxidation-sensitive lycopene molecule.

• Operon Complexity: The study highlighted that CRISPRi effects can vary within an operon. For example, repressing sdhC (upstream) decreased yield, while repressing sdhD (downstream) increased yield, likely due to polar effects and differential regulation by CRP and ArcB.

5. Significance

• Metabolic Rebalancing: The study demonstrates that optimizing isoprenoid production requires "metabolic rebalancing" across the entire cell, not just the target pathway. It reveals that under rich media (LB) conditions, endogenous pathways for amino acid and lipid synthesis can become metabolic sinks that compete with isoprenoid production.
• Novel Targets: It identifies previously uncharacterized targets (e.g., ilvD, leuC, aroK, aroF, pgsA) for carotenoid engineering, offering new strategies for strain improvement.
• Generalizable Platform: The developed CRISPRi-NGS pipeline is a generalizable tool for mapping genotype-to-phenotype relationships in metabolic engineering. It enables the discovery of complex genetic interactions (epistasis) and provides a foundation for machine learning models to predict optimal gene expression landscapes for complex biosynthetic pathways.
• Context Dependence: The results underscore that metabolic engineering strategies are highly dependent on culture conditions (e.g., LB vs. minimal media), emphasizing the need for context-specific screening.