Glycan processing in the Golgi as optimal information coding that constrains cisternal number and enzyme specificity

Alkesh Yadav, Quentin Vagne, Pierre Sens, Garud Iyengar, Madan Rao

TL;DR

This paper models the complex process of glycan synthesis in the Golgi apparatus, revealing that the "glycan code" on cell surfaces is a sophisticated information system. The study shows that achieving complex glycan patterns with high fidelity requires a delicate balance between the number of Golgi cisternae, enzyme specificity, and their precise arrangement, offering quantitative insights into cellular evolution and design principles.

What Problem Does This Paper Address?

Cells use glycans (sugar chains attached to proteins and lipids) on their surface as a form of molecular "barcode" for identity and communication. The precise arrangement of these glycans, known as the glycan profile, is crucial for accurate cell recognition. However, the cellular machinery responsible for creating these complex glycan patterns – the Golgi apparatus – is a dynamic and noisy environment. This paper tackles the question of how cells achieve high fidelity in generating these crucial glycan codes, especially for complex patterns, and what constraints this imposes on the structure and function of the Golgi. Specifically, it investigates the trade-offs between the number of Golgi compartments (cisternae), the number and specificity of the enzymes involved, and the rates of transport between compartments, all in the pursuit of accurately replicating a "target" glycan distribution.

Methods

The authors developed a simplified mathematical model of the N-glycosylation process in the Golgi. This model captures the sequential enzymatic reactions within each cisterna and the transport of molecules between them. Key aspects of their approach include:

  • Chemical Synthesis Model: A mean-field model that treats enzyme-substrate interactions using Michaelis-Menten kinetics.
  • Enzyme Specificity Model: A biophysical model based on "induced fit" to describe how enzyme specificity affects binding probabilities. This model allows for a spectrum from highly specific to promiscuous enzymes.
  • Optimization Framework: They define "fidelity" as the normalized Kullback-Leibler (KL) divergence between the synthesized and target glycan distributions. An optimization problem is formulated to minimize this fidelity metric by adjusting parameters like cisternal number ($N_c$), enzyme number ($N_E$), enzyme specificity ($\sigma$), and transport rates ($\mu$).
  • Complexity Measure: Glycan complexity is quantified using Gaussian Mixture Models (GMMs) fitted to mass spectrometry data, where the number of GMM components needed for a good fit serves as a complexity metric.
  • Fidelity Landscape Analysis: The authors analyze the "fidelity landscape" by examining the eigenvalues and eigenvectors of the Hessian matrix of the optimization problem. This helps identify "stiff" (sensitive) and "sloppy" (robust) directions in the parameter space, revealing how the system responds to perturbations.

Datasets

The study utilizes mass spectrometry (MS) data from human T-cells to define target glycan distributions. This data is pre-processed and approximated using Gaussian Mixture Models (GMMs) to represent varying levels of complexity (e.g., 3-GMM for less complex, 20-GMM for more complex). The authors note that no new data was generated for this manuscript, and the code and data are available on GitHub.

Key Insights and Ideas

  • Glycans as an Information Code: The paper frames glycan profiles as a molecular code essential for cell identity, emphasizing the importance of fidelity in its generation.
  • Complexity Demands Compartmentalization: Synthesizing complex glycan patterns requires multiple Golgi cisternae. A single compartment, even with many enzymes, would be less efficient and potentially more costly.
  • Optimal Enzyme Specificity: For a given number of enzymes and cisternae, there's an optimal level of enzyme specificity that balances fidelity and diversity. Too specific or too promiscuous enzymes can hinder accurate pattern generation.
  • Fidelity Landscape Geometry: The multidimensional "fidelity landscape" reveals stiff and sloppy directions, indicating which parameters the cell needs to control tightly (e.g., enzyme specificity) and which offer more flexibility.
  • Evolutionary Argument for Golgi Structure: The model provides a quantitative basis for why the Golgi evolved multiple compartments, as this structure inherently supports higher fidelity and robustness in glycan synthesis.

Key Results

  • Trade-offs: Achieving high fidelity for complex glycan distributions necessitates a significant increase in both the number of cisternae ($N_c$) and enzymes ($N_E$). The impact of increasing $N_c$ and $N_E$ on fidelity is roughly symmetrical.
  • Optimal Specificity: The optimal enzyme specificity ($\sigma_{min}$) required for high fidelity increases with the complexity of the target glycan distribution.
  • Diversity vs. Complexity: There's a tension between achieving high glycan diversity and high complexity. Low-specificity enzymes can lead to diversity but at the cost of complexity. This tension can be relieved by increasing $N_c$ and $N_E$.
  • Robustness: The synthesis model exhibits degeneracies, meaning multiple parameter combinations can yield similar glycan profiles. This, along with sloppy directions in the parameter space, contributes to robustness against noise and parameter fluctuations.
  • Stiff vs. Sloppy Directions: Enzyme specificity ($\sigma$) and transport rate ($\mu$) are identified as "stiff" directions, requiring tighter cellular control, while reaction rates ($R$) are "sloppy."

Limitations and Future Work

  • Simplified Model: The authors acknowledge that their model is a simplification of the actual cellular machinery, particularly in how it treats enzyme partitioning and ignores some stochastic effects.
  • Apparent Symmetry: The initial model showed an apparent symmetry between $N_E$ and $N_c$, which was later attributed to a simplification in the enzyme distortion model. A more refined model lifts this symmetry.
  • Future Directions: The authors suggest exploring more complex chemical models, investigating optimal enzyme partitioning in detail, understanding the role of intracellular transport in proofreading mechanisms, and analyzing the dynamics of non-equilibrium self-assembly of Golgi cisternae in conjunction with chemical processing. They also propose that the insights gained could inform glycoengineering efforts.

Summary updated

Abstract

Many proteins that undergo sequential enzymatic modification in the Golgi cisternae are displayed at the plasma membrane as cell identity markers. The modified proteins, called glycans, represent a molecular code. The fidelity of this <i>glycan code</i> is measured by how accurately the glycan synthesis machinery realizes the desired target glycan distribution for a particular cell type and niche. In this article, we construct a simplified chemical synthesis model to quantitatively analyse the trade-offs between the number of cisternae, and the number and specificity of enzymes, required to synthesize a prescribed target glycan distribution of a certain complexity to within a given fidelity. We find that to synthesize complex distributions, such as those observed in real cells, one needs to have multiple cisternae and precise enzyme partitioning in the Golgi. Additionally, for a fixed number of enzymes and cisternae, there is an optimal level of specificity (promiscuity) of enzymes that achieves the target distribution with high fidelity. The geometry of the <i>fidelity landscape</i> in the multidimensional space of the number and specificity of enzymes, inter-cisternal transfer rates, and number of cisternae provides a measure for robustness and identifies stiff and sloppy directions. Our results show how the complexity of the target glycan distribution and number of glycosylation enzymes places functional constraints on the Golgi cisternal number and enzyme specificity.
BibTeX
@inproceedings{yadav2022glycan,
  author = {Alkesh Yadav and Quentin Vagne and Pierre Sens and Garud Iyengar and Madan Rao},
  title = {{Glycan processing in the Golgi as optimal information coding that constrains cisternal number and enzyme specificity}},
  booktitle = {eLife},
  year = {2022},
  doi = {10.7554/elife.76757},
}
Figure