15.2: Glycogenesis

Search Fundamentals of Biochemistry

Learning Goals (ChatGPT o3-mini)

1. Describe the Enzymatic Steps of Glycogenesis:

   • Outline the six-step process of glycogen synthesis, identifying the roles of hexokinase, phosphoglucomutase, glycogen synthase (GS), UDP-glucose pyrophosphorylase (UGPase), glycogenin (GN), and glycogen branching enzyme (GBE).

2. Explain the Role of Glycogen Synthase in Glycogen Assembly:

   • Discuss how GS extends the glycogen main chain via the formation of alpha 1→4 glycosidic bonds using UDP-glucose and how its activity is regulated by insulin signaling and allosteric effectors such as glucose-6-phosphate.

3. Understand the Function and Mechanism of UDP-Glucose Pyrophosphorylase (UGPase):

   • Describe how UGPase catalyzes the reversible conversion of glucose 1-phosphate and UTP into UDP-glucose and inorganic pyrophosphate, emphasizing the enzyme’s dependency on divalent cations (e.g., Mg²⁺) and the key active site residues and loops involved in catalysis.

4. Explain the Role of Glycogenin in Priming Glycogen Synthesis:

   • Illustrate how glycogenin self-catalyzes its glycosylation to provide the primer required for GS activity, including the role of Mn²⁺ and critical aspartate residues in coordinating UDP-glucose.

5. Analyze Glycogen Branching Enzyme (GBE) Function and Structure:

   • Describe the mechanism by which GBE introduces alpha 1→6 branches into the glycogen molecule, increasing the number of non-reducing ends, and discuss the domain organization and structural features that distinguish eukaryotic GBE from its prokaryotic counterparts.

6. Integrate Enzyme Cooperation in Glycogen Synthesis:

   • Explain how glycogenin and glycogen synthase interact within a multi-enzyme complex, detailing the conformational states (closed/inactive, partially open, and open/active) that regulate their cooperative activity during glycogen assembly.

7. Examine the Regulation of Glycogen Synthase Activity:

   • Discuss the dual regulation of GS via allosteric activation by glucose-6-phosphate and reversible phosphorylation (and dephosphorylation), including the role of GSK-3, phosphatases, and the structural basis for the inactive (T) and active (R) states.

8. Assess the Impact of Mutations and Post-translational Modifications:

   • Analyze how mutations in critical regions (e.g., the 309-loop of UGPase, the arginine cradle in GS) and post-translational modifications influence enzyme activity and overall glycogen synthesis.

9. Utilize Structural Models to Understand Enzyme Function:

   • Interpret interactive structural models (e.g., iCn3D models of human UGPase, glycogenin-GS complexes, and GBE) to correlate structural features with catalytic mechanisms and regulatory interactions.

10. Synthesize the Overall Pathway of Glycogen Synthesis and Its Regulation:

    • Integrate the individual enzymatic steps and regulatory mechanisms into a comprehensive overview of glycogen synthesis, emphasizing how the cooperative and coordinated activities of these enzymes contribute to cellular energy storage and metabolic homeostasis.

These learning goals will guide students through both the mechanistic details and the regulatory complexities of glycogenesis, preparing them to analyze and interpret experimental data and structural information in this critical area of metabolic biochemistry.

Introduction

The process of forming glycogen is called glycogenesis, and it requires the activity of six enzymes, as illustrated in Figure \(\PageIndex{1}\).  We have already discussed several, including hexokinase, which phosphorylates the 6'-OH of glucose, and phosphoglucomutase, which converts glucose-6-phosphate to the glucose-1-phosphate isomer. This section will discuss the remaining four enzymes and their role in glycogen biosynthesis. They are Glycogen Synthase, UDP-Glucose Pyrophosphorylase (preferred name UTP-glucose-1-phosphate uridylyltransferase), glycogenin, and glycogen branching enzyme.

Figure \(\PageIndex{1}\): Enzymes involved in Glycogenesis. Image from Mark Cidade

Given their importance in synthesizing the second main energy storage molecule, we must probe them in detail.

Glycogen synthase (GS) is a key enzyme whose activity is highly regulated. In Chapter 15.1, we have already explored how insulin signaling upregulates the activity of this enzyme by inhibiting phosphorylation by GSK-3. Other effectors include the allosteric binding of glucose-6-phosphate, which also increases the activity of the GS. In a later section, we will also see that the hormone glucagon can also regulate the activity of the GS through protein kinase A (PKA) in a fashion that decreases glycogen synthesis and increases glycogen breakdown.

In the glycogenesis pathway, GS is responsible for building most of the main α-1 → 4 chain glucose acetal linkages. The GS requires a primer of 4-6 glucose residues linked together by α-1 → 4 bonds to initiate synthesis. Since GS can only form α-1 → 4 linkages in the main chain, it CANNOT create the α-1 → 6 branches inherent to the core structure of glycogen.

GS uses the glycogen primer and glucose activated through covalent attachment to uridine diphosphate (UDP) at the 1-position to build the glycogen main chain. Upon completion of one round of synthesis, the 1 position of the incoming UDP-glucose is covalently attached to the 4-position of the nascent glycogen molecule, releasing the UDP as a leaving group.

\[\text { Glycogen }_{(n)}+\text { UDP-glucose } \rightarrow \text { Glycogen }_{(n+1)}+\text { UDP } \nonumber \]

UTP--glucose-1-phosphate uridylyltransferase (or UDP-Glucose Pyrophosphorylase)

The formation of the UDP-glucose required for the synthesis of the main chain of glycogen is mediated by UTP-glucose-1-phosphate uridylyltransferase (preferred name), which is also called UDP-glucose pyrophosphorylase (GalU or UGPase; EC 2.7.7.9). UGPase catalyzes the reversible reaction of glucose 1-phosphate and UTP into UDP-glucose and inorganic pyrophosphate (PP_i) (Figure \(\PageIndex{2}\)). Enzymes of the UGPase family are ubiquitous and can be found in the tree of life.

Figure \(\PageIndex{2}\): Formation of UDP-Glucose

UDP-glucose is an activated form of glucose used to synthesize other glycans, including sucrose, cellulose, starch, and glycogen, and the glycan parts of glycoproteins, glycolipids, and proteoglycans.  Hence, it is a key metabolite that gives more importance to understanding UGPase.

Like many other nucleotidyl transferases, UGPase requires divalent cations to promote the reaction (Figure \(\PageIndex{3}\)). In most cases, magnesium ions are employed. The reaction mechanism follows a sequential bi-bi-mechanism, starting with the binding of UTP to the active site in the presence of a magnesium ion, followed by the binding of glucose 1-phosphate. The octahedral coordination sphere of magnesium positions the substrates correctly and enables the nucleophilic attack of glucose 1-phosphate on UTP. A lysine, an aspartate, and several water molecules within the active site help stabilize the position of the substrates and cofactor for the proper nucleophilic attack of the phosphoryl oxygen of glucose 1-phosphate on the α-phosphorus atom of UTP. Finally, PP_i is released from the UGPase/Mg²⁺/UDP-glucose complex. UDP-Glucose then dissociates from the complex, restoring the enzyme's active site for another round of synthesis.

Figure \(\PageIndex{3}\): Proposed Reaction Mechanism of UTP--glucose-1-phosphate uridylyltransferase (UDP-Glucose Pyrophosphorylase). Image from Kumpf, A., et al. (2019) Int. J. Mol. Sci. 20(22) 5809; https://doi.org/10.3390/ijms20225809

Figure \(\PageIndex{4}\) shows an interactive iCn3D model of the human UDP-glucose pyrophosphorylase tetramer (3R2W).

Figure \(\PageIndex{4}\):  Human UDP-glucose pyrophosphorylase tetramer (3R2W). (Copyright; author via source). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/icn3d/share.html?9yq9STPGvq87Wnbt8

Substrate binding and active site residues are shown as CPK-colored sticks and labeled in one of the subunits.  Three key loops whose correct positioning is required for catalysis are colored as follows:

• Latch Loop:  406-416, yellow, which contains Glu 412 (shown in spacefill)
• SB Loop:  275-282, red
• 309 Loop: 309-311, cyan, which contains Ser 309 (shown in spacefill).

For clarity, only four of the eight subunits are shown.  Note the proximity of the interacting loops positioned between the brown and magenta subunits.  The SB and 309 loops at that location are part of the brown subunit, while the yellow latch loop is part of the magenta subunit.  The latch loop is between the SB and 309 loops. The cyan 309 loop is prevented from interacting with the substrate and the movements of the SB and 309 loops needed for catalysis.  In other species, it has been observed that the SB loop moves down to bind UDP-glucose at the active site.  The structure above represents the apo-enzyme, which is inactive and in its closed form without a bound substrate.

Mutations in the 309 loop (S309N/S311R) still had 84% of normal activity.  Glu412 in the latch loop is highly conserved in vertebrates (but not in yeast).  Mutations in Glu 412 didn't affect the formation of oligomers of the enzyme but did affect activity.  

• replacing E412 with a short aspartate (E412D) significantly increases activity  (176%)
• E412Q, which eliminates the charge while retaining the approximate size of the side chain, showed just a marginal increase in activity (19%)
• E412K, which flips the charge and increases the length of the side chain, decreased activity to 22%.

These mutations generally suggest that steric effects in the region of subunit interaction are most important in activity.

Glycogen Synthase

UDP-glucose is then utilized by glycogen synthase (GS) to extend the main chain of glycogen by one glucose residue. In this reaction, the 4’-OH group of the glycogen main chain attacks the anomeric carbon of UDP-glucose (Figure \(\PageIndex{5}\)). The UDP functional group serves as a good leaving group, allowing for the formation of the alpha 1 --> 4 bond.

Figure \(\PageIndex{4}\): Formation of the Glycogen Mainchain by Glycogen Synthase. Image modified from Mikael Häggström

Glycogen synthase has two isoforms: GS1, which is expressed in tissues where glycogen is present (for example, muscle), and GS2, which is expressed only in the liver. We will explore its mechanism more fully below.

Glycogenin

Previously, we mentioned that GS requires a glycogen primer of 4 – 6 glucose residues to add new residues to the main chain. The small docking protein, Glycogenin (GN or GYG), provides this primer. This protein is a homodimer that self-catalyzes its own glycosylation at amino acid Tyr-194. In this reaction, UDP-glucose is coordinated by a Mn²⁺ metal cofactor and critical aspartate residues (Figure \(\PageIndex{6}\)). The –OH group of Tyr-194 then mediates nucleophilic attack on the anomeric carbon of UDP-glucose. Thus, glycogenin is tethered to the reducing end of the glycogen molecule.

Figure \(\PageIndex{6}\): Coordination of UDP-Glucose by Glycogenin. Critical aspartic acid residues and a manganese ion cofactor are required for the coordination of UDP-glucose by glycogenin. Image from Hedberg Oldfors, C., Glamuzina, E., Ruygrok, P. and Anderson, L. (2016) J. Inhert. Met. Dis. 40(1) DOI: 10.1007/s10545-016-9978-1

As with glycogen synthase, glycogenin (GN or GYG) has two isoforms, with GN2 (or GYG2) mainly expressed in the liver, pancreas, and heart.

Mechanisms for glycogen synthesis by glycogenin and glycogen synthase

How do glycogenin and glycogen synthase cooperate in the synthesis of glycogen?  Structures of the complex of glycogenin-1 (GYG1), which seeds the molecule by starting glycogen synthesis by autoglucosylation, and glycogen synthase-1 (GS1 or GYS1), which extends the molecule, show allosteric transitions between three primary states, the closed/inactive, partially open, and open active complex, as shown in Figure \(\PageIndex{7}\):

Figure \(\PageIndex{7}\): GYS1 chains A, B, C, and D are colored orange, turquoise, purple, and navy, respectively. GYG1 globular domains and the GYG1-tail fragment are colored in gray shades for all chains.  The bottom left structure shows the apo GYS1:GYG1 mobile complex.  The bottom middle structure is the apo GYS1:GYG1 ordered complex.  The bottom right structure is the +G6P GYG1:GYS1 complex.  Fastman et al., 2022, Cell Reports 40, 111041 July 5, 2022, 2022. https://doi.org/10.1016/j.celrep.2022.111041.  Creative Commons Attribution (CC BY 4.0)

It makes sense that both enzymes bind to each other and cooperate in the synthesis of glycogen. In the closed state, GS (GYS) is a tetramer, which, like hemoglobin, can be described as a T or closed/inactive state.  In a slight difference, one of the GS (GYS) subunits appears to display an asymmetric conformation, which leads to close interactions with GN (GYG), allowing the glycogen seed polymer on GN (GYG) to move to GS (GYS) for elongation (a partially open state).  Multiple conformations of the complex have been resolved.  Further conformational changes lead to the open/active state, resulting in a more open binding groove for GN (GYG).   

Figure \(\PageIndex{8}\) shows an interactive iCn3D model of the  Human glycogenin-1 and glycogen synthase-1 complex in the presence of glucose-6-phosphate (8CVX).  Glucose-6-phosphate is an allosteric activator of glycogen synthase.

Figure \(\PageIndex{8}\):  Human glycogenin-1 and glycogen synthase-1 complex in the presence of glucose-6-phosphate (8CVX). (Copyright; author via source). Click the image for a popup or use this external link:  https://structure.ncbi.nlm.nih.gov/i...ivh4gZCupGcXZ7

The larger glycogen synthase subunits in the tetramer are shown in various colors, while the glycogenin fragments bound to each GS (GYS) monomer are shown in gray.  

Figure \(\PageIndex{9}\) illustrates the conformational changes induced on the binding of glucose-6-phosphate to glycogen synthase. 

Figure \(\PageIndex{9}\): G6P binding induces a conformational change across the GYS1 tetramer, leading to an open conformation for all four active sites.  Fastman et a.l, ibid.

Panel (A) shows a cartoon representation of the +G6P GYS1:GYG1 complex (left) with GYS1 and GYG1 chains colored as in the previous figure. G6P is shown as spheres colored by heteroatom. Blue and yellow lines and boxes (right) indicate relative positions for perpendicular views. A black dotted circle indicates the oligomeric interface between the CTD-loop region and tetramerization core domains. The CTD-loop, residues 484–488, forms cross-protomer interactions on the other side of the active site.  Green dashed semicircles indicate open active sites.

Panel (B) shows G6P-binding site interactions. G6P (white) and proximal interacting residues are shown as sticks and colored by heteroatom. Yellow dotted lines indicate polar interactions.

Panel (C) shows changes at the G6P-binding site across conformations. A G6P-bound protomer is highlighted in orange, with G6P shown as spheres. The adjacent active state protomer (chain C) is shown as a green cartoon. A basal state protomer (chain C) is modeled relative to the orange G6P-bound protomer and shown in yellow. Key changes to the G6P-sensing loop and a regulatory helix are highlighted by outline and non-transparent representation with arrows indicating the relevant motions. The rest of each chain is shown in a transparent cartoon representation.

We will return to the mechanism after exploring the last enzyme in the pathway.

Glycogen Branching Enzyme

The final enzyme, the glycogen branching enzyme (GBE), catalyzes the hydrolytic cleavage of an α(1→4) glycosidic linkage and subsequent inter- or intra-chain transfer of the non-reducing terminal fragment to the C6 hydroxyl position of an α-glucan (Figure \(\PageIndex{10}\)). In this example, an inter-chain transfer is occurring. At the top of the scheme, above the arrow, you can see that the GBE enzyme transiently removes several glucose residues (usually around 7) from one linear glycogen chain and then attaches it as an alpha 1 → 6 branch to the other chain. In this process, an additional non-reducing end is created, which can act as a primer site for Glycogen Phosphorylase (the main enzyme that breaks down glycogen). Thus, glucose residues can be released very quickly when needed.

Figure \(\PageIndex{10}\): Formation of Glycogen alpha 1 --> 6 Branches. Image from van der Vlist, J., et al (2012) Polymers 4(1) 674-690. DOI: 10.3390/polym4010674

Details of the structure and domain organization of the human glycogen branching enzyme are shown in Figure \(\PageIndex{11}\). The four domains include N1, CBM48, the catalytic domain, and the C-terminal domain.

Figure \(\PageIndex{11}\):  Crystal structure of hGBE1.   Froese et al. Human Molecular Genetics, Volume 24, Issue 20, 15 October 2015, Pages 5667–5676, https://doi.org/10.1093/hmg/ddv280.  Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/)

Panels (A and B) show orthogonal (perpendicular) views of hGBE1 showing the N-terminal helical segment (orange), CBM48 (pink), central catalytic domain (green), and C-terminal domain (blue). The catalytic triad Asp357-Glu412-Asp481 is shown as red sticks. Numbers refer to domain boundaries. N- and C-termini are labeled as grey spheres.

Panel (C) shows the superposition of branching enzyme structures from human (hGBE1, this study), O. sativa SBE1, and M. tuberculosis GBE, highlighting the conserved domain architecture and three regions of structural variation.

Panel (D) shows the domain organization of hGBE1, O. sativa SBE1, and M. tuberculosis GBE, revealing differences in the N-terminus between prokaryotic and eukaryotic polypeptides. Prokaryotic GBEs contain two N-terminal carbohydrate-binding domains (N1, N2,) whereas eukaryotes contain only one (CBM48) and replace the prokaryotic N1 domain with a helical extension.

Figure \(\PageIndex{12}\) shows an interactive iCn3D model of the human glycogen branching enzyme (GBE1)

Figure \(\PageIndex{12}\): human glycogen branching enzyme (GBE1). (Copyright; author via source). Click the image for a popup or use this external link:  https://structure.ncbi.nlm.nih.gov/i...47aqw5qto8hkU7

The domains in the iCn3D module are colored similarly to the previous figures.

The enzyme core is similar to amylase with the conserved active site.  Diseases of glycogen storage often result from mutations in the amylase core domain. For example, late-onset adult polyglucosan body disease (APBD) arises from a common mutation, Y329S.  The effect of this mutation may arise from misfolding.  A tetrapeptide, Leu-Thr-Lys-Glu, administered to patients, increased activity twofold, likely by acting as a chaperone to facilitate proper protein folding.

Putting it all together: Glycogen Synthesis and its Regulation

Figure \(\PageIndex{13}\) shows the steps involved in the addition of glucose from the donor UDP-glucose to glycogenin and through glycogen synthase to the growing glycogen polymer, which becomes branched through the activity of the glycogen branching enzyme.

Figure \(\PageIndex{13}\):  Summary of glycogen synthesis. Marr, L., Biswas, D., Daly, L.A. et al. Nat Commun 13, 3372 (2022). https://doi.org/10.1038/s41467-022-31109-6.  Creative Commons Attribution 4.0 International License.  http://creativecommons.org/licenses/by/4.0/

The top reaction shows the step catalyzed by glycogenin (GN), while the bottom reactions are those catalyzed by glycogen synthase and the glycogen branching enzyme. 

The regulation of glycogen synthesis occurs in part through the phosphorylation of key residues in both GS (GYS) and GN (GYG).  Those sites are shown in Figure \(\PageIndex{14}\).

Figure \(\PageIndex{14}\): Domain and phosphorylation sites for glycogen synthase 1 (GS1) and glycogenin 1 (GN1).  Marr et al., ibid.

Panel c shows the domain architecture of human GS (top) and GN (bottom). Known in vivo phosphorylation sites of GS are shown in red and labeled with the corresponding residue number and classical nomenclature (in bold). GN tyrosine 195, which becomes auto-glucosylated and was mutated to phenylalanine (Y195F) in this study, is indicated. Not to scale. Panel e shows the cartoon representation of GN WT and Y195F.

As we showed above, glycogen synthase is allosterically activated on the binding of glucose-6-phosphate, which can be thought to activate the enzyme by a T to R state change.  The enzyme is inactivated by phosphorylation at multiple sites, as shown above. Activation can also occur through dephosphorylation.  Phosphorylation of a site can create a binding site that facilitates additional phosphorylations.  This can lead to a flexible  "spike" of hyperphosphorylated residues forming from two monomers. Particularly important is pSer641 (site 3A), which interacts with a series of arginine residues in a regulatory helix of glycogen synthase.  This arginine cluster has been referred to as the arginine cradle.  The interaction sites are illustrated in Figure \(\PageIndex{15}\).

 Figure \(\PageIndex{15}\): The phosphoregulatory region of human GS. Marr et al, ibid

Panel a shows the human (Hs)GS-GN³⁴ structure in ribbons (top left). The N- and C-terminal tails of one GS protomer (chain A) lie next to one another and move towards the adjacent protomer, meeting the N- and C-terminal tails from chain B. Arrows indicate a continuation of cryo-EM density (top right). Electron density (C1 symmetry) for phosphorylated S641 (pS641) interacting with R588 and R591 on the regulatory helices α22 (bottom left). Residues interacting with the N- and C-terminal tails that are mutated in this study are shown (bottom right). 

Panel b shows a comparison of distances between regulatory helices of adjacent monomers of HsGS (reported here), low activity inhibited mimic (PDB ID 5SUL), basal state (PDB ID 3NAZ), and G6P-activated (PDB ID 5SUK) yeast GS (yGS) crystal structures. Quoted distances were measured from Cα of Arg591 (chain A) and -Cα of Arg580 (chain B) of HsGS and corresponding yeast residues.

The strong electrostatic arginine-pSer interactions lock the tetramer into the inactive T state.

A cartoon model illustrating the regulation of glycogen synthase by phosphorylation/dephosphorylation and interconversion between T and R state is shown in Figure \(\PageIndex{16}\).

Figure \(\PageIndex{16}\): GS and GN cooperate to synthesize glycogen. Marr et al, ibid

 The inhibition by phosphorylation can be relieved by binding the allosteric effector glucose-6-phosphate and does not require phosphatases,

Summary

This chapter provides a comprehensive overview of glycogenesis—the process by which cells synthesize glycogen, the primary storage form of glucose. It builds on earlier discussions of hexokinase and phosphoglucomutase, focusing on the four key enzymes that complete the pathway: Glycogen Synthase (GS), UDP-Glucose Pyrophosphorylase (UGPase), Glycogenin (GN), and Glycogen Branching Enzyme (GBE).

Glycogen Synthase (GS):
GS is the central enzyme responsible for elongating the glycogen main chain by forming α(1→4) glycosidic bonds using UDP-glucose as the activated donor. Its activity is intricately regulated—upregulated by insulin signaling (which inhibits phosphorylation by GSK-3) and further enhanced allosterically by glucose-6-phosphate. Despite its pivotal role, GS requires a preformed glycogen primer (4–6 glucose residues) provided by glycogenin, and it cannot generate the branch points in glycogen.

UDP-Glucose Pyrophosphorylase (UGPase):
UGPase catalyzes the reversible conversion of glucose-1-phosphate and UTP into UDP-glucose and inorganic pyrophosphate. This reaction, which requires divalent cations (typically Mg²⁺) and involves key active site loops (the latch, SB, and 309 loops), is crucial for generating the activated glucose needed for glycogen synthesis. Structural studies reveal how substrate positioning and specific residue interactions, such as those involving Glu412 and Ser309, influence enzyme activity and regulation.

Glycogenin (GN):
GN acts as the primer for glycogen synthesis by self-catalyzing the addition of a short glucose chain via autoglucosylation at a specific tyrosine residue. Coordinated by a Mn²⁺ cofactor and essential aspartate residues, glycogenin provides the initial oligosaccharide that glycogen synthase can then extend, ensuring that glycogen synthesis can be efficiently initiated.

Glycogen Branching Enzyme (GBE):
GBE introduces α(1→6) branches into the glycogen molecule by cleaving α(1→4) bonds and transferring a short glucan fragment to a C6 hydroxyl group on a neighboring chain. This branching increases the number of non-reducing ends, which is critical for rapid mobilization of glucose during energy demand. Structural analyses of human GBE (hGBE1) illustrate its domain organization and active site features, and highlight the differences between prokaryotic and eukaryotic enzymes.

Integration and Regulation:
The chapter concludes by integrating these enzymes into the broader context of glycogen synthesis. It explains how glycogenin and glycogen synthase cooperate through conformational transitions—shifting from closed/inactive to open/active states—to extend the glycogen molecule, while the branching enzyme remodels the structure to optimize storage and mobilization. Additionally, the regulation of glycogen synthesis is underscored by reversible phosphorylation events and allosteric mechanisms that fine-tune enzyme activities in response to cellular energy status.

Overall, this chapter elucidates the molecular mechanisms and regulatory networks underlying glycogenesis, providing students with a detailed understanding of how glucose is stored and mobilized in cells—a fundamental aspect of energy metabolism and homeostasis.

References

1. Kumpf, A., Partzsch, A., Pollender, A., Bento, I., and Tischler, D. (2019) Two Homologous Enzymes of the GalU Family in Rhodococcus opacus 1CP-RoGalU1 and RoGalU2. Int. J. Mol. Sci. 20(22), 5809. https://doi.org/10.3390/ijms20225809