The Maillard reaction, a chemical reaction between amino acids and sugars, is exploited to produce flavorful food ubiquitously, from the baking industry to our everyday lives. However, the Maillard reaction also occurs in all cells, from prokaryotes to eukaryotes, forming advanced glycation end-products (AGEs). AGEs are a heterogeneous group of compounds resulting from the irreversible reaction between biomolecules and α-dicarbonyls (α-DCs), including methylglyoxal (MGO), an unavoidable byproduct of anaerobic glycolysis and lipid peroxidation. We previously demonstrated that Caenorhabditis elegans mutants lacking the glod-4 glyoxalase enzyme displayed enhanced accumulation of α-DCs, reduced lifespan, increased neuronal damage, and touch hypersensitivity. Here, we demonstrate that glod-4 mutation increased food intake and identify that MGO-derived hydroimidazolone, MG-H1, is a mediator of the observed increase in food intake. RNAseq analysis in glod-4 knockdown worms identified upregulation of several neurotransmitters and feeding genes. Suppressor screening of the overfeeding phenotype identified the tdc-1-tyramine-tyra-2/ser-2 signaling as an essential pathway mediating AGE (MG-H1)-induced feeding in glod-4 mutants. We also identified the elt-3 GATA transcription factor as an essential upstream regulator for increased feeding upon accumulation of AGEs by partially controlling the expression of tdc-1 gene. Furthermore, the lack of either tdc-1 or tyra-2/ser-2 receptors suppresses the reduced lifespan and rescues neuronal damage observed in glod-4 mutants. Thus, in C. elegans, we identified an elt-3 regulated tyramine-dependent pathway mediating the toxic effects of MG-H1 AGE. Understanding this signaling pathway may help understand hedonistic overfeeding behavior observed due to modern AGE-rich diets.
feedingadvanced glycation end-productsglod-4elt-3tyraminepharyngeal pumpingResearch organismC. eleganshttp://dx.doi.org/10.13039/100000002National Institutes of HealthR01AG061165KapahiPankajhttp://dx.doi.org/10.13039/100000002National Institutes of HealthR01AG068288KapahiPankajhttp://dx.doi.org/10.13039/100001167Larry L. Hillblom Foundation2021-A-007-FELKapahiPankajhttp://dx.doi.org/10.13039/100000002National Institutes of HealthR01DK133196GalliganJames Jhttp://dx.doi.org/10.13039/100000002National Institutes of HealthR35GM137910GalliganJames JThe funders had no role in study design, data collection, and interpretation, or the decision to submit the work for publication.Author impact statementAdvanced glycation end-products (compounds which make the food appetizing and aromatic) intricately modulate organism’s homeostatic and hedonistic signaling pathways thereby inducing preferential consumption of excess nutrients offering preliminary insights into overconsumption of modern-day processed food.Introduction
Processed modern diets enriched with advanced glycation end-products (AGEs), formed by the Maillard reaction, are tempting to eat but at the same time deleterious for health (Chaudhuri et al., 2018; Nowotny et al., 2018; Zhang et al., 2020). In 1912, a French Chemist, L.C. Maillard, reported a reaction between glucose and glycine upon heating, resulting in the formation of brown pigments (Maillard, 1912). Later, the covalent bonds formed between carbohydrates and proteins during heating in a non-enzymatic browning reaction was named the Maillard reaction (Jaeger et al., 2010; Liu et al., 2020). Glycation is a part of the Maillard reaction, or browning of food, during cooking which enhances the taste, color, and aroma of the food to make it more palatable (Maillard, 1912; Lima, 2013). The Maillard reaction has revolutionized the food industry by playing an important role in food chemistry (Machiels and Istasse, 2022); however, this reaction also results in the formation of adverse AGEs as well as toxic byproducts including acrylamide (Luca et al., 2010; Mottram et al., 2022; Stadler et al., 2002).
In addition to food sources, AGEs are also endogenously produced in cells when α-dicarbonyl compounds (α-DCs) (such as glyoxal [GO], methylglyoxal [MGO], and 3-deoxyglucosone [3DG]) non-enzymatically react with biomolecules. α-DCs are unavoidable byproducts of cellular metabolisms, such as glycolysis and lipid peroxidation (Figure 1). AGEs include GO derivatives such as carboxymethyl lysine (CML) and glyoxal lysine dimer (GOLD). AGEs derived from MGO include hydroimidazolone (MG-H1), carboxyethyl lysine (CEL), and methylglyoxal lysine dimer (MOLD), and 3DG derivatives include 3-deoxyglucosone-derived imidazolium cross-link (DOGDIC), pyrraline, etc. (Chaudhuri et al., 2018; Allaman et al., 2015; Chen et al., 2018; Vistoli et al., 2013). The glyoxalase system utilizes enzymes Glo1 and Glo2 and reduced glutathione (GSH) to detoxify α-DCs stress, especially MGO to lactate (Figure 1), in cytosol and nucleus. Differential expression levels of glyoxalases are reported in various disease conditions such as diabetes, hypertension, neurodegenerative disorders, anxiety disorders, infertility, and cancer, suggesting their role in exacerbating their pathogenesis (He et al., 2020). Glo1 has been linked with several behavioral phenotypes, such as anxiety, depression, autism, and pain, among other mental illnesses (Distler and Palmer, 2012). Also, we have previously demonstrated increased neuronal damage in the Caenorhabditis elegans glod-4 glyoxalase mutant model, which is shown to accumulate high levels of α-DCs and AGEs (Figure 1 and Figure 2—figure supplement 1H, I); (Chaudhuri et al., 2016). AGEs accumulate in long-lived proteins, such as collagen (Monnier et al., 1984); furthermore, quantifying the glycated form of hemoglobin (HbA1c) is utilized as a biomarker in diabetes (Rahbar et al., 1969). Increased AGEs are associated with aging, obesity, diabetes, neurodegeneration, inflammation, cardiomyopathy, nephropathy, and other age-related diseases (Chaudhuri et al., 2018; Ramasamy et al., 2005; Prasad et al., 2019). Furthermore, neurodegenerative diseases have also demonstrated a strong correlation between increased levels of AGEs and pathogenesis.
Graphical representation for the formation of α-dicarbonyls and advanced glycation end-products (AGEs).
Dicarbonyls are highly reactive byproducts from metabolic pathways such as lipid peroxidation and glycolysis. In the above example, methylglyoxal (MGO) spontaneously forms from dihydroxyacetone phosphate which interacts with biomolecules resulting in the formation of AGEs. Toxic MGO is detoxified by glyoxalase enzymes to non-toxic lactate. One of the examples of glyoxalase enzyme in C. elegans is glod-4. Lack of glyoxalase enzyme leads to increased levels of MGO resulting in increased accumulation of AGEs.
Overconsumption of food and excessive availability of cheap, highly processed foods have contributed to the obesity pandemic. Obesity is a key risk factor for other diseases including diabetes, hypertension, cancers, cardiovascular, inflammatory, and neurodegenerative disorders, among other non-communicable chronic diseases (Lee and Yau, 2020; Miller and Spencer, 2014; Wolin et al., 2010; Ellulu et al., 2017; Keramat et al., 2021; Uribarri et al., 2015; Ruhm, 2012; Hossain et al., 2007). Thus, identifying signaling pathways that modulate increased feeding behavior is important to understand the underlying causes of obesity and identify novel therapeutics to overcome it. Here, we report that loss of the glyoxalase system or exogenously feeding MGO-derived AGEs increased feeding behavior in C. elegans. We also identified the mechanism for the observed phenotype and found that the MGO-derived AGE, MG-H1, acts via the elt-3 GATA transcription factor (TF), to partially regulate the expression of tdc-1 gene (tyramine decarboxylase – an enzyme that biosynthesis neurotransmitter tyramine), and tyramine receptors (tyra-2 and ser-2) to mediate adverse effects of AGEs such as increased feeding, reduced lifespan, and neuronal damages. This study is the first to identify the signaling pathway mediated by specific AGEs molecules downstream of MGO (such as MG-H1) to enhance feeding and neurodegeneration. Our study emphasizes that AGEs accumulation is deleterious and enhances disease pathology in different conditions, including obesity and neurodegeneration. Hence, limiting AGEs accumulation is relevant to the global increase in obesity and other age-associated diseases.
ResultsAGEs increase food intake and food-seeking behavior in <italic>C. elegans</italic>
Our initial observations revealed that glod-4 glyoxalases enzyme mutants exhibit a significantly enhanced pharyngeal pumping than wildtype N2 animals (Figure 2A). This increase in pharyngeal pumping was consistent from day 1 (young adult, post-65 hr of timed egg laying) till day 3 of adulthood (Figure 2A). We performed a food clearance assay to validate whether increased pharyngeal pumping was accompanied by enhanced food intake (Figure 2B and Figure 2—figure supplement 1A). We found increased bacterial clearance after 72 hr in glod-4 mutants.
The glyoxalase mutant, <italic>glod-4,</italic> and methylglyoxal (MGO)-derived advanced glycation end-product (AGE), MG-H1, increases pharyngeal pumping and feeding in <italic>C. elegans</italic>.
(A) Quantification of pharyngeal pumping (#/30 s) in N2 (wt) and glod-4 (gk189) mutant at different stages of adulthood. (B) Food clearance assay in N2 (wt) and glod-4 (gk189) mutant after 72 hr of feeding. (C) Quantification of pharyngeal pumping (#/30 s) in N2 (wt) after treatment, with either 150 µM of arginine (control) or MG-H1. (D) Food clearance assay in N2 (wt) and glod-4 (gk189) mutant worms after treatment for 72 hr with either 150 µM of arginine (control) or MG-H1. (E) Quantification of pharyngeal pumping with different concentrations of MG-H1. (F) Food racing assay in N2 (wt) and glod-4 (gk189) at different stages of adulthood toward OP50-1. (G) Food racing assay of N2 (wt) toward OP50-1 when combined with either MGO or MG-H1 (100 µM). (H) Food racing assay of glod-4 (gk189) mutants toward OP50-1 when combined with either MGO or MG-H1 (100 µM). Student’s t-test for A, B, C, E, and F. One-way analysis of variance (ANOVA) with Fisher’s LSD (Least Significant Difference) multiple comparison test for D, G, and H. The data points in the graphs represent the sample size (n). Comparison between two specific groups are indicated by lines above the bars; otherwise, the groups are compared with control group. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Error bar ± standard deviation (SD).
The glyoxalase mutant, <italic>glod-4,</italic> and methylglyoxal (MGO)-derived advanced glycation end-product (AGE), MG-H1, increases pharyngeal pumping and feeding in <italic>C. elegans</italic>.
(A) Food clearance assay demonstrating increased food intake with an increasing number of worms. (B, C) Food clearance assay of wildtype N2 (wt) and glod-4 (gk189) mutant with 5 mM treatment of serotonin, respectively. (D) Quantification of pharyngeal pumping in genetic mutants such as tph-1 (tryptophan hydroxylase), glod-4, and tph-1;glod-4 double mutant compared with N2 (wt) wildtype worms. (E) Quick visual quantification of pharyngeal pumping after treatment with different advanced glycation end-products (AGEs) molecules at 100 µM for 12–18 hr. (F) Quantification of pharyngeal pumping with treatment of either N2 (wt) wildtype or glod-4 mutants with either arginine or MG-H1 at 150 µM. (G) Quantification of pharyngeal pumping in N2 (wt) worms after treatment with phosphate-buffered saline or 150 µM of arginine for 24 hr. (H) Liquid chromatography–multiple reaction monitoring (LC-MRM) chromatograms for the exacted ion peaks for TNP-MG-H1 in N2 wildtype (left) and glod-4 mutant (right). (I) Relative quantification of MG-H1 in worm lysates of N2 wildtype and glod-4 mutant background from (H). (J) Pictorial representation of food racing assay. Student’s t-test for B, C, I. One-way analysis of variance (ANOVA) for A, D, E–G. The data points in the graphs represent the sample size (n). *p < 0.05, ***p < 0.001, and ****p < 0.0001. Error bar ± standard deviation (SD).
Serotonin treatment increased bacterial clearance in both wildtype N2 worms and glod-4 mutants (Figure 2—figure supplement 1B, C). Furthermore, worms lacking tph-1 (tryptophan hydroxylase, an enzyme that catalyzes the formation of 5-hydroxy-tryptophan, precursor for serotonin) enzyme as well as tph-1;glod-4 double mutants which lacks serotonin (Dallière et al., 2017; Ji Ying Sze et al., 2000) show decreased pumping compared to wildtype N2 worms (Figure 2—figure supplement 1D). However, tph-1;glod-4 double mutants show significantly increased pumping compared to tph-1 single mutants (Figure 2—figure supplement 1D). These data suggest that glod-4 null mutation mediated increase in pharyngeal pumping is independent of the serotonin signaling (Dallière et al., 2017). These preliminary observations lead to the hypothesis that enhanced feeding in glod-4 mutant worms is mediated by endogenous accumulation of AGEs characterized previously (Chaudhuri et al., 2016; Golegaonkar et al., 2015; Morcos et al., 2008). To this end, we explored AGEs such as MG-H1, CEL, CML, and F-ly as possible mediators of feeding and identified MG-H1 and CEL as potential MGO-derived AGEs to increase feeding in C. elegans (Figure 2—figure supplement 1E). Just feeding MGO was not sufficient to increase the pharyngeal pumping rate (Figure 2—figure supplement 1E). Time course analysis in wildtype N2 worms treated with MG-H1 showed that 24 hr of MG-H1 (150 µM) treatment was enough to increase pharyngeal pumping significantly (Figure 2C). A significant increase in bacterial clearance was observed after 72 hr (Figure 2D). Also, note that treatment of glod-4 null mutants with MG-H1 did not further increase either the bacterial clearance or the pharyngeal pumping (Figure 2D and Figure 2—figure supplement 1F), suggesting that MG-H1 and glod-4 null mutation increases feeding by overlapping mechanism. In addition, we also demonstrated that MG-H1 regulates pharyngeal pumping rate in a dose-dependent manner (Figure 2E). Since MG-H1 is the product of arginine modification by MGO (see Materials and methods), we used arginine as a negative control for our MG-H1 treatment. We did not observe a significant difference between worms treated with arginine versus water versus phosphate-buffered saline (Figure 2—figure supplement 1G). Since MG-H1 induces increased pharyngeal pumping (Figure 2C, D), we validated an increase in MG-H1 in the glod-4 null mutants compared to N2 wildtype using liquid chromatography–multiple reaction monitoring (LC–MRM) mass spectrometry (Figure 2—figure supplement 1H, I). In addition to food consumption, the glod-4 mutant exhibited a significantly increased preference toward food source OP50-1 at days 1 and 3 of adulthood compared to wildtype N2 worms (Figure 2F and Figure 2—figure supplement 1J). Furthermore, we noticed that wildtype N2 worms preferred exogenous MG-H1 compared to MGO when provided with bacterial food source Escherichia coli OP50-1 (Figure 2G). We did not observe this phenotype in the glod-4 mutant background (Figure 2H), suggesting MG-H1 in food makes it more appealing for control worms.
Tyramine regulates MG-H1-mediated feeding behavior via G-protein-coupled receptors TYRA-2 and SER-2
Next, we sought to elucidate how MG-H1 increases the feeding behavior in worms. We performed an unbiased RNA sequencing approach to analyze the global transcriptome profile between control and glod-4 knockdown worms (Supplementary file 1). Our RNAseq analysis identified a total of 20,277 genes, of which 5035 genes were significantly changed (2237 upregulated genes and 2798 downregulated genes) in glod-4 RNAi knockdown worms compared to N2 wildtype. Gene set enrichment analysis showed that the functional category of genes regulating feeding behavior was significantly upregulated in glod-4 knockdown worms (>twofold enrichment score) (Figure 3—figure supplement 1, red * marked GO category). This analysis supports our above observation that glod-4 mutants have an altered feeding rate. Previous studies in C. elegans have documented the role of neurotransmitters in C. elegans feeding behavior (Avery and Horvitz, 1990; Chase and Koelle, 2007; Trojanowski et al., 2016) and we observed differential expression of 66 neurotransmitters and feeding genes (which comprises ~19% of the total feeding and neurotransmission-related genes in C. elegans) (Figure 3A).
Role of <italic>tdc-1</italic> and tyramine receptors in mediating the MG-H1-induced feeding behavior.
(A) Differential expression of 66 neurotransmitters and feeding genes in glod-4 RNAi background. (B) The flowchart shows the pathway of biogenic amine synthesis, which functions as a neurotransmitter. (C) Quantification of pharyngeal pumping in N2 (wt) and tdc-1 (n3419) mutant worms after 24 hr of treatment of MG-H1. (D) Quantification of pharyngeal pumping in N2 (wt), tdc-1 (n3419), glod-4 (gk189), and tdc-1;glod-4 double mutants. (E, F) Quantification of pharyngeal pumping in N2 (wt), tyra-2 (tm1846), ser-2 (ok2103), tyra-2;glod-4, and ser-2;glod-4 mutants. One-way analysis of variance (ANOVA) with Fisher’s LSD multiple comparison test for C–F. The data points in the graphs represent the sample size (n). Comparison between two specific groups are indicated by lines above the bars; otherwise, the groups are compared with control group. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Error bar ± standard deviation (SD).
Gene ontology analysis for upregulated genes in <italic>glod-4</italic> mutant worms.
Role of <italic>tdc-1</italic> and tyramine receptors in mediating the MG-H1-induced feeding behavior.
(A) Quantification of pharyngeal pumping (quick screening by visual counting) on mutants of enzymes involved in the biosynthesis of biogenic amines after MG-H1 treatment (suppressor screen). (B) Quantification of pharyngeal pumping in genetic mutants such as tdc-1 (tyrosine decarboxylase) n3420 allelic mutant, glod-4 mutant, and tdc-1;glod-4 double mutant compared with N2 wildtype worms. (C) Quantification of pharyngeal pumping (quick screening by visual counting) on receptor mutants involved in feeding behavior after MG-H1 treatment. Student’s t-test for A, C. One-way analysis of variance (ANOVA) for B. The data points in the graphs represent the sample size (n). **p < 0.01, ***p < 0.001, and ****p < 0.0001. Error bar ± standard deviation (SD).
We next tested the involvement of these neurotransmitter genes in regulating MG-H1-mediated feeding behavior and systematically analyzed (suppressor screen) MG-H1-induced feeding in the background of genetic mutants limited in producing different biogenic amines and neurotransmitters in C. elegans (Figure 3B and Figure 3—figure supplement 2A). We found that mutation in tdc-1, the gene involved in synthesizing neurotransmitter tyramine, suppressed the enhanced feeding phenotype in MG-H1 treated animals (Figure 3—figure supplement 2A, indicated by a black arrow and Figure 3C). We also confirmed suppression of increased feeding rate in tdc-1;glod-4 double mutant animals (utilizing two different tdc-1 allelic mutants n3419 and n3420) compared to glod-4 single mutants (Figure 3D and Figure 3—figure supplement 2B). Next, we checked putative receptors for tyramine that could potentially mediate downstream signaling. Receptors for tyramine and octopamine are well-studied G-protein-coupled receptors (GPCRs) (Gross et al., 2014; Pirri et al., 2009). We screened seven GPCRs to identify the potential link in regulating tyramine-mediated increased feeding rate exhibited by glod-4 mutant or MG-H 1-treated worms. Observed results showed a mutation in ser-2 and tyra-2 suppresses enhanced feeding in MG-H1 treated animals (Figure 3—figure supplement 2C, indicated by black arrows). A similar reversal of feeding phenotype was observed in tyra-2;glod-4, and ser-2;glod-4 double mutant strains (Figure 3E, F). Our findings support the idea that MG-H1-induced overfeeding is mediated by tyramine signaling.
To check for putative TFs that could regulate the differentially expressed genes in glod-4 knockdown worms (Figure 3), we performed a motif-enrichment analysis (based on available ChIP-Seq data) (Figure 4A, B). We chose the top five TFs (with a threshold of >18.75% target sequence match for TF-binding motif) for further screening. We knocked down each of the five TFs individually and checked for the suppression of MG-H1-induced feeding behavior (Figure 4—figure supplement 1A). Knocking down pha-4 and elt-3 suppressed the increase in pharyngeal pumping induced by MG-H1 treatment (Figure 4—figure supplement 1A, indicated by black arrows). The pha-4 gene is crucial for pharynx development, and loss of pha-4 results in a morphological defect of the pharynx (Mango, 2007; Mango et al., 1994), therefore, we followed the results from elt-3 knockdown in elt-3 mutants (Figure 4C).
Role of <italic>elt-3</italic> transcription factor in regulating MG-H1-induced feeding in <italic>C. elegans</italic>.
(A) List of transcription factors identified by motif analysis. (B) Flowchart demonstrating the method of identification of transcription factors. (C) Quantification of pharyngeal pumping after treatment with either arginine or MG-H1 in elt-3 (gk121) mutants. (D) Quantification of pharyngeal pumping in N2 (wt), glod-4 (gk189), elt-3 (gk121), and double mutant worms. (E) Quantification of tyramine pathway genes in elt-3 (gk121) mutant worms. (F) Quantification of elt-3 and tyramine pathway genes expression in wildtype N2 (wt) worms after MG-H1 treatment. The horizontal dotted line indicates the normalized expression levels of genes in N2 (wt) and untreated control in E and F, respectively. One-way analysis of variance (ANOVA) with Fisher’s LSD multiple comparison test for C, D. Student’s t-test for E, F. The data points in the graphs represent the sample size (n) in C,D and number of biological repeats in E,F. Comparison between two specific groups are indicated by lines above the bars; otherwise, the groups are compared with control group. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Error bars ± standard deviation (SD).
Role of <italic>elt-3</italic> transcription factor in regulating MG-H1-induced feeding in <italic>C. elegans</italic>.
(A) Quantification of pharyngeal pumping (Quick visual counting), suppressor screen for top 5 transcription factors listed in Figure 4A. (B) Quantification of pharyngeal pumping in glod-4 mutant worms along with elt-3 gene knockdown by RNAi feeding comparted with N2 (L4440 represent the control feeding plasmid without dsRNA). (C) List of genes obtained by Hypergeometric Optimization of Motif EnRichment (HOMER) analysis that are potentially regulated by the elt-3 transcription factor. Student’s t-test in A. One-way analysis of variance (ANOVA) for B. The data points in the graphs represent the sample size (n). ****p < 0.0001. Error bar ± standard deviation (SD).
Analysis of pharyngeal pumping in elt-3;glod-4 double mutant showed that elt-3 is essential to increase pharyngeal pumping observed in glod-4 mutant worms (Figure 4D). Also, note that the knockdown of elt-3 using RNAi feeding in glod-4 single mutants suppressed the pumping (Figure 4—figure supplement 1B). To determine the role of elt-3 in the tyramine signaling pathway, we performed a HOMER (Hypergeometric Optimization of Motif EnRichment) analysis and identified the binding site of elt-3 on the tdc-1 promoter, which suggested elt-3 may potentially regulate tdc-1 expression levels (Figure 4—figure supplement 1C). This was further validated by reduced expression of tdc-1 mRNA levels in the elt-3 mutant worms (Figure 4E). Next, to check if the elt-3 expression is changed on exposure to MG-H1, we treated wildtype N2 worms with MG-H1 and quantified mRNA levels of elt-3. We observed a moderate but significant increase in the elt-3 expression (Figure 4F). Although tdc-1 and tyra-2 did not change significantly, expression levels of other receptors, tyra-3 and ser-2, increased significantly after MG-H1 exposure (Figure 4F). Note that ser-2 is necessary to mediate the increased pharyngeal pumping (Figure 3F). Together, these experiments identified a key role for elt-3 in tyramine-induced feeding increase in response to MG-H1.
Tyramine signaling is necessary to increase feeding in <italic>glod-4</italic> mutants
To strengthen the role of tyramine in mediating increased feeding in glod-4 KO worms, we treated mutants lacking tyramine signaling with exogenous tyramine. Tyramine has been demonstrated to decrease pharyngeal pumping in wildtype N2 worms (Dallière et al., 2017). Consistent with this finding, we found that exogenous treatment of tyramine significantly decreased pharyngeal pumping (Figure 5A). However, exogenous treatment of tyramine to rescue tyramine signaling in both elt-3;glod-4 and tdc-1;glod-4 double mutants shows a significant increase in pharyngeal pumping to the levels similar to that of glod-4 single mutants (Figure 5B, C). Exogenous tyramine did not increase pumping in the double mutants lacking tyramine receptors such as tyra-2;glod-4 and ser-2;glod-4 (Figure 5D, E). These results strongly demonstrate that tyramine suppresses pharyngeal pumping in N2 wildtype worms; however, it increased the pumping in the glod-4 mutant background. These data support the notion that tyramine signaling is necessary to mediate glod-4 mutant-dependent increase in feeding behavior.
Exogenous tyramine rescues the suppressed pumping in double mutants.
(A) Quantification of pharyngeal pumping in N2 wildtype worms treated with tyramine at various concentrations. (B) Quantification of pharyngeal pumping in elt-3, glod-4, elt-3;glod-4 untreated and elt-3;glod-4 double mutant treated with tyramine. (C–E) Quantification of pharyngeal pumping after treatment of double mutant worms (tdc-1;glod-4, ser-2;glod-4, tyra-2;glod-4) with exogenous tyramine. One-way analysis of variance (ANOVA) with Fisher’s LSD multiple comparison test for A, B. Student’s t-test for C–E. The data points in the graphs represent the sample size (n). Comparison between two specific groups are indicated by lines above the bars; otherwise, the groups are compared with control group. **p <0 .01, and ****p < 0.0001. Error bars ± standard deviation (SD).
Absence of tyramine rescues α-DCs and AGEs mediated pathogenic phenotypes
Accumulation of α-DCs in glod-4 mutants results in pathogenic phenotypes, including neurodegeneration and shortening of lifespan (Chaudhuri et al., 2016). Here, chronic accumulation of MGO leads to the build-up of AGEs (Figure 2—figure supplement 1I), thereby increasing feeding in glod-4 worms (Figure 2).
Furthermore, accumulation of MG-H1 significantly reduced the lifespan of N2 wildtype worms; however, it did not further exacerbate the damage in glod-4 mutant worms (Figure 6—figure supplement 1). To test whether tyramine signaling is essential for mediating the pathogenic phenotypes such as neuronal damage and reduced lifespan in glod-4 mutants, we compared the lifespan between wildtype N2 and glod-4 worms in the genetic mutants that lack tyramine. The lifespan of glod-4 was significantly increased upon inhibition of tyramine signaling in the tdc-1;glod-4 double mutation (Figure 6A). Next, we tested if the absence of tyra-2 and ser-2 could also rescue the shortened lifespan of glod-4 mutants. Lifespan increased significantly in the absence of either tyra-2 or ser-2 in double mutant animals (Figure 6B, C). In addition to rescuing lifespan and feeding rate, the lack of tyramine also resulted in the partial but significant rescue of neuronal damage in glod-4 animals (Figure 6D, E).
Suppression of <italic>glod-4</italic> phenotypes in <italic>tdc-1;glod-4</italic> double mutant.
(A) Survival assay with N2 (wt), tdc-1, glod-4, and tdc-1;glod-4 double mutants. (B) Survival assay with N2 (wt), tyra-2, glod-4, and tyra-2;glod-4 double mutants. (C) Survival assay with N2 (wt), ser-2, glod-4, and ser-2;glod-4 double mutants. (D) Image of worm neurons showing neuronal damage at day 8 of adulthood. Red arrows indicates damages. (E) Quantification of neuronal damage with pan-neuronal GFP marker in glod-4 versus tdc-1;glod-4 double mutants, 2 biological repeats. Scale bar – 10 µm. Log-rank (Mantel–Cox) test for survival assays. One-way analysis of variance (ANOVA) with Fisher’s LSD multiple comparison test for E. Comparison between two specific groups are indicated by lines above the bars; otherwise, the groups are compared with control group. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Error bar ± standard deviation (SD).
Survival analysis of N2 and <italic>glod-4</italic> mutant worms after MG-H1 treatment at 150 µM.
Log-rant (Mantel–Cox) test for statistical analysis. **p < 0.01, ***p < 0.001, and ****p < 0.0001.
Discussion
Our observation that glod-4 mutants run out of bacterial lawn faster than wildtype N2 animals during routine maintenance led to the elucidation of a novel signaling pathway that mediates AGE-induced feeding behavior in C. elegans. Glyoxalases are enzymes involved in the detoxification of α-DCs (Figure 1), and we have previously characterized glod-4 mutant, which lacks one of the glyoxalase enzymes, to accumulate increased levels of α-DCs (Chaudhuri et al., 2016) and thereby AGEs, especially MG-H1 (Figure 2—figure supplement 1H, I). In this study, using genetic mutants, RNAi knockdown, synthesized AGEs, and functional genomics, we elucidate that AGEs (especially MG-H1) induce increased feeding through tyramine signaling regulated by GATA transcription factor ELT-3. The glod-4 KO worms, with enhance AGEs accumulation, showed increased feeding, which led to the hypothesis that increased accumulation of AGEs is a potential stimulator of binge feeding. Thus, we studied changes in pumping rate by exogenous administration of MGO and AGEs. As previously reported by Ravichandran et al., 2018, MGO treatment did not change the pumping rate; however, MG-H1 and CEL increased the pumping rate in wildtype N2 worms (Figure 2—figure supplement 1E; Ravichandran et al., 2018). Furthermore, a recent study demonstrated that treatment with sugar-derived AGE-modified bovine serum albumin accelerated the pharyngeal pumping rate (Papaevgeniou et al., 2019). Our study demonstrates that either treatment with purified MG-H1 or endogenous production and accumulation of MG-H1 via genetic mutation increases feeding and adversely affects lifespan. We also found that MG-H1-induced hyper-feeding is independent of serotonin-mediated hyper-feeding (Dallière et al., 2017) in C. elegans (Figure 2—figure supplement 1B–D).
Our detailed investigation of the time-dependent increase in pumping rate after MG-H1 treatment indicates a more robust and highly significant increase after 24 hr of treatment (Figure 2C). We also observed a MG-H1 dose-dependent increase in feeding rate (Figure 2E), which is based on the stronger significance (lower p-value) caused by reduced dispersion of the data at higher concentrations of MG-H1. Our analysis indicates that higher concentrations of MG-H1 can increase the pharyngeal pumping in almost all the treatment worms, thus predisposing the worms with lower pumping rates in the Gaussian distribution to higher pumping rates. It is well established that AGEs are formed during cooking, browning the food during dry heating, making the food more appetizing (Vistoli et al., 2013). Furthermore, feeding is a multisensorial process regulated by several signaling pathways subjected to evolutionary adaptations (Luca et al., 2010). Thus, we wanted to analyze the changes in sensory behavior of C. elegans induced by either endogenous accumulation of AGEs or by exogenous administration of MG-H1 with the food. Since the glod-4 mutant lacks glyoxalase system to detoxify MGO and leads to the accumulation of AGEs (Figure 1), the MG-H1-mediated signaling pathway can be responsible for the increased chemoattraction of glod-4 mutant worms to food source OP50-1 (Figure 2F). It can be explained that including MG-H1 in bacterial lawn increased the chemoattraction of wildtype N2 worms toward food, resulting in increased attraction to palatable MG-H1-mixed bacterial food OP50-1 (Figure 2G). However, unlike wildtype N2 worms, exogenous MG-H1 treatment had no further increase in the feeding rate or chemoattraction of glod-4 mutant worms (Figure 2D, H and Figure 2—figure supplement 1F), indicating the maximum sensory modulation attained by the endogenous accumulation of MG-H1 in the glod-4 mutant. Although our screening identified CEL, a lysine-derived adduct of MGO, as another AGEs increasing the food intake, a detailed analysis is necessary to conclude the effect of CEL on feeding behavior (Figure 2—figure supplement 1E).
We utilized RNAseq data from glod-4 knockdown worms to identify the novel signaling pathway that mediates AGE-induced feeding in C. elegans. Since glod-4 knockdown data are enriched with several genes regulating the synthesis of neurotransmitters and feeding (Figure 3A and Figure 3—figure supplement 1), we performed suppression screening in mutant worms for genes involved in synthesizing biogenic amine neurotransmitters after MG-H1 treatment (Figure 3B and Figure 3—figure supplement 2A, C). Thus, our screen identified tdc-1, involved in tyramine biosynthesis, and tyramine receptors (tyra-2 and ser-2) to mediate AGE-induced increased pharyngeal pumping (Figure 3C–F and Figure 3—figure supplement 2). Tryptophan and tyrosine are the substrates for synthesizing biogenic amines implicated in modulating various behaviors in C. elegans (Chase and Koelle, 2007; Alkema et al., 2005). Tyrosine to tyramine conversion in the presence of the enzyme tyrosine decarboxylase (TDC-1) followed by tyramine β-hydroxylase (TBH-1), is crucial for the synthesis of neurotransmitters tyramine and octopamine, respectively (Alkema et al., 2005; Greer et al., 2008). Previous studies have shown the role of tyramine and its receptor (ser-2) in regulating feeding and foraging behavior in C. elegans (Dallière et al., 2017; Greer et al., 2008; Rex et al., 2004; Li et al., 2012). Furthermore, the tyra-2 receptor is expressed in MC (Marginal Cell) and NSM (NeuroSecretory Motor) pharyngeal neurons and is discussed to regulate pharyngeal pumping potentially (Dallière et al., 2017; Rex et al., 2005). Especially, tyramine has been shown to reduce pharyngeal pumping when applied exogenously to the worms (Figure 5A; Dallière et al., 2017). Supporting previous findings (Dallière et al., 2017), our observation shows increased pharyngeal pumping in tyra-2 and ser-2 single mutant worms (Figure 3E, F); at the same time, tdc-1 single mutants did not increase pumping (Figure 3D). Converse to our observation of tyra-2 and ser-2 single mutants, Greer et al., 2008 did not find any difference in the pumping rate of tyra-2 and ser-2 single mutants compared to wildtype N2 worms. However, the same study reported no changes in the pumping rate of the tdc-1 single mutant, similar to our results (Greer et al., 2008), which is also demonstrated by Li et al., 2012. Interestingly, double mutants of either tdc-1 or its receptors (tyra-2-partial suppression and ser-2) with glod-4 mutant significantly suppress the increased pharyngeal pumping observed in either glod-4 or tyra-2 or ser-2 single mutants (Figure 3D–F). It is to be noted that only two interneurons, namely RIM (Ring Interneuron M) and RIC (Ring Interneuron C) neurons, uv1 cells near vulva and gonadal sheath cells (Alkema et al., 2005) express the tdc-1 gene, which is involved in the biosynthesis of tyramine; however, receptors of tyramine are expressed in distant tissues explaining an endocrine activity for tyramine neurotransmitter (Dallière et al., 2017) leading to the multi-pathway mode of action to exert differential response which should be elucidated in the future. Since ser-3 mutant worms did not suppress the pumping (Figure 3—figure supplement 2C) and ser-3 has been demonstrated to be a receptor for octopamine (Dallière et al., 2017), we conclude that octopamine is not responsible for mediating MG-H1-induced feeding in C. elegans.
Our suppressor screen for the upstream effector of the tdc-1-tyramine-tyra-2/ser-2 pathway that mediates MG-H1-induced increased feeding identified the elt-3 TF (Figure 4C, D). Thus, we examined whether elt-3 TF regulates the tdc-1, tyra-2, or ser-2. Our analysis revealed that in elt-3 mutant worms, the tdc-1 gene is significantly reduced (Figure 4E), concluding that elt-3 TF regulates tyramine biosynthesis. In favor of the data, HOMER analysis identified that the tdc-1 gene is potentially regulated by elt-3 TF (Figure 4—figure supplement 1C). Although elt-3 TF is predominantly expressed in hypodermal cells, its expression is also reported in the pharyngeal–intestinal valve, intestine, a few neurons (head neurons and mechanosensory PVD (Posterior Ventral process D) neuron), etc. (Wormbase.org). In accordance with elt-3 expression in PVD neurons and head neurons, tyra-2 is also expressed in PVD neurons (Rex et al., 2005) and tdc-1 in RIM and RIC head interneurons, respectively, suggesting a possible direct/partial regulation of tdc-1 expression by elt-3. Also, tyra-2 expression has been reported in pharyngeal MC neurons, which directly regulate pharyngeal pumping (Rex et al., 2005), suggesting direct endocrine action of tyramine. Similarly, ser-2 is expressed in pharyngeal muscle segment cells (Rex et al., 2004; Li et al., 2012; Tsalik et al., 2003). Increased expression of ser-2 in elt-3 mutant worms can be inferred as a compensatory mechanism for reduced tyramine signaling by increasing the expression of the tyramine receptor. Also, ser-2 expression is significantly increased in MG-H1 treated wildtype N2 worms (Figure 4F). Although the mechanism of MG-H1-induced expression of ser-2 is unclear, it is evident that the ser-2 genetic mutant can suppress the increased feeding in the glod-4 mutant (double mutants) (Figure 3F), demonstrating an important role of the SER-2 receptor in mediating the MG-H1-induced feeding via tyramine. Furthermore, elt-3 expression levels significantly increased after MG-H1 treatment. Altogether, our data strongly suggest the role of the elt-3-tdc-1-tyramine-tyra-2/ser-2 pathway in mediating enhanced feeding. Finally, it is essential to note that the ser-2 gene is upregulated in the glod-4 knockdown RNAseq dataset, similar to significant upregulation after MG-H1 treatment, validating that MG-H1 is a critical player in mediating adverse phenotypes observed in glod-4 mutant worms.
Exogenous tyramine suppressed pharyngeal pumping in N2 wildtype worms; however, tyramine rescued the pumping in double mutants (elt-3;glod-4 and tdc-1;glod-4) to that of glod-4 single mutants (Figure 5). Although our data strongly demonstrate the role of tyramine signaling in increasing feeding rate in glod-4 mutant background, our study also identified a paradox in tyramine signaling to regulated pharyngeal pumping. From the literature (Dallière et al., 2017) as well as from our data, it is evident that tyramine suppresses pumping in N2 wildtype genetic background (Figure 5A). The mechanism behind this behavioral switch, that is, from suppressor of pharyngeal pumping in wildtype to a stimulator in glod-4 mutant background, remains elusive. It can be speculated that MG-H1 can modify the tissue-specific expression of tyramine receptors (Figure 4F), resulting in an observed behavioral switch in response to tyramine signaling. This hypothesis can only be addressed by methodologies such as single-cell RNA sequencing and exploration of cellular signaling circuitry in further studies. Thus, our current understanding is that MG-H1 (either exogenous or endogenous accumulation in glod-4 KO) modifies tyramine signaling by modulating genes in the tyramine pathway, leading to an increased feeding rate in C. elegans.
Previously, we have demonstrated reduced lifespan, hyperesthesia, and accelerated neurodegeneration-like phenotypes observed in diabetic conditions caused by excessive accumulation of α-DCs in glod-4 mutant worms (Chaudhuri et al., 2016). Lack of dicarbonyl detoxification by glyoxalases enzyme in glod-4 mutant worms (Figure 1) should result in the accumulation of AGEs (Figure 2—figure supplement 1H, I), which at sufficient concentration act as signaling molecules to modulate the feeding behavior (Figure 2) by causing differential gene expression (Figure 3). The increased amount of dicarbonyl stress, thereby AGEs, is observed in several systemic diseases such as obesity, diabetes, cardiovascular and neurodegenerative diseases, among other age-associated diseases (Chaudhuri et al., 2018). In diabetic patients, three times higher plasma levels of MGO have been reported and is a leading cause of neuropathic pain (Bierhaus et al., 2012; Rabbani and Thornalley, 2014; Kold-Christensen et al., 2019). Earlier reports in the literature show that the dicarbonyl levels correlate with diabetic complications. One of the major risk factors for diabetes is obesity (Ismail et al., 2021), which is caused by overfeeding. Thus, exploring the regulatory pathways of feeding is essential to understand and identify ways to modulate feeding behavior.
Here, we show that AGEs can modulate feeding behavior in evolutionary primitive model organisms, and it will be worth exploring this pathway in mammals. The TF elt-3 belongs to the GATA TF family (Gilleard et al., 1999). Shobatake et al., 2018 report that GATA 2 and 3 TFs induce the expression of appetite regulator genes such as POMC and CART (Shobatake et al., 2018). With the easy availability and unlimited access to modern-day processed food enriched in sugars and AGEs resulting in overeating, a significant cause of the obesity pandemic, it is necessary to explore signals regulating feeding. Importantly, our study shows exogenous treatment with MG-H1 increases feeding in worms (Figure 2C, D), indicating that a high AGEs diet in our day-to-day life can modulate feeding behavior in humans. It is well known that food cooked by grilling, broiling, roasting, searing, and frying accelerates the formation of AGEs in food; thus, methods are explored to cook food with fewer AGEs accumulation (Uribarri et al., 2010). Furthermore, increased caloric intake and changes in eating habits have been reported in a behavioral variant of frontotemporal dementia (Aiello et al., 2016) and medication of antipsychotic drugs (Perez-Gomez et al., 2018).
Finally, we show that a lack of tdc-1-tyramine signaling rescues glod-4 mutant phenotypes (lifespan and neuronal damage) (Figure 6). A strong association between worsening PD phenotypes with increased aggregation of α-synuclein and specific sites of increased glycation has been demonstrated in different genetic models with increased AGEs (Vicente Miranda et al., 2017). Thus, it will be interesting to investigate the role of the tdc-1-tyramine pathway in modulating pathways enhancing neurodegeneration and feeding. Recent research identified neurodegenerative diseases to be influenced by metabolism (Muddapu et al., 2020; Procaccini et al., 2016) and glod-4 mutants demonstrate increased neuronal damage, decreased lifespan, and increased feeding. Thus, it is essential to investigate the balance in energy metabolism to identify critical pathways to modulate the outcome of neurodegenerative diseases.
Materials and methodsStrains
Strains were either obtained from Caenorhabditis Genetic Center (CGC), Minneapolis, USA or National Bioresource Project, Tokyo, Japan and the following strains were used: N2 (wt), VC343 glod-4(gk189), VC143 elt-3(gk121), MT13113 tdc-1(n3419), MT10661 tdc-1(n3420), FX1846 tyra-2(tm1846), RB1690 ser-2(ok2103), MT9455 tbh-1(n3247), CB1112 cat-2(e1112), MT15434 tph-1(mg280), DA1814 ser-1(ok345), RB1631 ser-3(ok2007), RB745 ser-4(ok512), VC125 tyra-3(ok325), and OH438 otls117[unc-33p::gfp + unc-4(+)]. All the mutant strains were outcrossed at least three times or more with N2 wildtype. Mutant strains are crossed to get the double mutants elt-3;glod-4, tdc-1(n3419);glod-4, tdc-1(n3420);glod-4, tph-1;glod-4, tyra-2;glod-4, ser-2;glod-4, glod-4;unc-33p::gfp, and tdc-1;glod-4;unc-33p::gfp. RNAi clones were obtained from Ahringer’s RNAi feeding library and the following were used: pha-4, ces-1, elt-3, lin-39, egl-5, and tdc-1.
Growth and maintenance
Worms were cultured at 20°C for at least two generations on standard NGM (Nematode Growth Media) agar plates seeded with 5× E. coli OP50-1 bacterial strain (Broth culture of OP50-1 was cultured overnight at 37°C at 220 rpm), which was propagated at room temperature for 2 days. For feeding RNAi bacteria, synchronized L1 larvae were transferred to NGM plates containing 3 mM of isopropyl β-D-1-thiogalactopyranoside (IPTG; referred to as RNAi plates) seeded with 20× concentrated HT115 bacteria (cultured overnight at 37°C at 220 rpm), carrying the desired plasmid for RNAi of a specific gene or bacteria carrying empty vector pL4440 as control and allowed to grow on plates for 48 hr at 37°C. For drug assays, synchronized young adult worms (60–65 hr from egg laying) were transferred to NGM plates (with or without IPTG) with 20× HT115 RNAi bacteria or 5× OP50-1 bacteria, respectively, which are freshly overlayed by the desired drug (or vehicle control) that was air dried and diffused. Final drug concentrations were calculated considering the total media volume on the NGM plates.
Note: For glod-4 mutant animals, we found that the pathogenic phenotypes discussed in this paper are contingent on strictly maintaining an ad libitum feeding regimen. Hence, care was taken not to allow the animals to starve by maintaining a low worm-to-bacteria ratio and transferring to fresh plates frequently (at least once every 2 days).
Pharyngeal pumping assay
C. elegans pharyngeal pumping was measured using a Leica M165 FC stereomicroscope utilizing a modified previously established method (Raizen et al., 2012) on day 2 young adult worms (unless otherwise specified). Grinder movement in the terminal bulb was used as a read-out for the pumping rate phenotype. Pharyngeal pumping was recorded using a Leica M165 FC microscope; thus, obtained movies were played at ×0.25 times the original speed and a manual counter was used to count the number of pumps for 30 s. For quick pumping screening (pumping data in the figure supplements), the pumping rate was counted in real time for 30 s using a stopwatch and a manual counter focusing the grinder using an Olympus SZ61 stereomicroscope. Ten to thirty animals were counted per biological repeat and two to three repeats were obtained for each experiment. The pumping data from all the repeats were combined for the presentation of data in the figures. At least one biological replicate was counted blind. Since pharyngeal pumping is very dynamic and changes with worms’ development, we decided to record the minimum required treatment groups as possible to reduced variations caused by delayed time and worms’ development. In case, when more treatment/genetic groups need to be compared the video recordings between different groups were staggered (recording of 10 worms per group followed by recording worms from other groups and repeating this cycle until 30 worms per group were recorded) to minimize the variations. To reduce the variations induced by fluctuations in the room temperature, 3 plates of worms were prepared for a single treatment group and only 10 worms were recorded per plate while other treatment plates were incubated at 20°C. Animals that did not pump during the recording time, worms that were stationary for prolonged amount of time and worms that are potentially injured with visible damages were eliminated from the analysis as well a few outliers were identified using the Gaussian distribution curve. Under exogenous drug treatment, animals were incubated in the drug at least 18–24 or until 48 hr before measurement of the pump rate. The drugs were overlaid on the NGM plate containing bacterial lawn and air dried before the addition of worms.
Food clearance assay
Food clearance assay was performed following minor modifications to the established protocol by Wu et al., 2019. In brief, 20–25 age synchronized (L3–L4 stage) worms were washed twice in S basal then once with S complete medium and transferred to a 96-well plate containing 160 µl assay medium (S-complete medium, growth-arrested OP50-1 at final OD 0.8 (at 600 nm), antibiotics, FuDR and either 150 µM arginine or MG-H1 or 5 mM serotonin). Initial bacterial density was measured by obtaining OD at 600 nm. Following the indicated number of hours, bacterial density was measured at OD600 after a brief and gentle mixing using a multichannel pipet. For each experimental data point, at least six wells were measured (at least 120–150 worms in total), with the results shown being representative of at least two to three independent assays. The relative food intake was determined by the change in OD for each well, normalized to the number of worms. Under these conditions, ample OP50-1 was available for feeding throughout the analysis, and worms were maintained in the same wells for the entire duration of the experiment.
Food race assay
The food race assay to evaluate C. elegans choice or attraction for a specific diet, a chemosensory behavior, was performed utilizing a previously established protocol (Nyamsuren et al., 2007). For this assay, synchronized adult worms (50 per race) were spotted on a 60-mm NGM agar plate, freshly seeded with E. coli OP50-1 (with or without drug) approximately 2 cm from the edge of the Petri plate. Adult animals were aliquoted on the plate diametrically opposite to the food source to estimate the percentage of worms that reached the food source within 30 min. An illustration of the food race assay has been provided (Figure 2—figure supplement 1J).
Organic synthesis of AGEs
MG-H1 (3) was synthesized according to the literature procedure with a slight modification as follows: (L)-arginine (1) (6.07 g, 34.8 mmol, 1 equiv) was dissolved in 12 M HCl (50 ml). To this was added methylglycol dimethyl acetal (2) (4.53 g, 38.3 mmol, 1.1 equiv). It was then stirred at room temperature for 11 hr. At this time, the reaction mixture was diluted with water (200 ml) and concentrated in vacuo. The resulting dark-red solution was purified by SiO2-gel column chromatography (4:2:1 ethyl acetate:methanol:acetic acid) to give MG-H1 (3) as a yellow solid (5.23 g, 22.9 mmol, 66%) (Scheme 1). The spectroscopic data obtained are consistent with those previously reported in the literature (Hellwig et al., 2011).
CML and CEL were synthesized according to the reported procedure (Hellwig et al., 2011) with a slight modification. To a 25-ml flask was added Nα-(tert-butoxycarbonyl)-L-lysine (4) (1.0 mmol, 1 equiv), palladium on carbon (10 wt% loading, 100 mg, 0.94 mmol), and distilled H2O (7 ml). To this was added glyoxylic acid (120 mg, 1.3 mmol, 1.3 equiv) for CML synthesis or pyruvic acid (115 mg, 1.3 mmol, 1.3 equiv) for CEL synthesis. 1 N NaOH(aq) was added dropwise to make pH of this solution 9. A balloon filled with hydrogen gas was attached, and the resulting solution was stirred at room temperature for 14 hr. At this time, the reaction mixture was filtered through a celite pad and the filtrate was concentrated in vacuo. Purification by SiO2-gel column chromatography (1:2 ethyl acetate:methanol) yielded (6a) (260 mg, 0.85 mmol) or (6b) (263 mg, 0.83 mmol), respectively. To this was added 1 N HCl(aq) (3 ml), and it was then stirred at room temperature for 3 hr. The resulting solution was concentrated in vacuo to give CML (7a) (164 mg, 0.80 mmol, 80%) or CEL (7b) (172 mg, 0.79 mmol, 79%) (Scheme 2). The spectroscopic data obtained are consistent with those previously reported in the literature (Hellwig et al., 2011).
Nε-carboxymethyl-lysine (CML) (7a) and Nε-(1-carboxyethyl)-lysine (CEL) (7b).
F-ly (12) was synthesized according to the literature procedure with a slight modification as follows: To a 200-ml round-bottomed flask was added Nα-(tert-butoxycarbonyl)-L-lysine (4) (510 mg, 2.8 mmol, 1 equiv), D-(+)-glucose (6.15 g, 30.0 mmol, 10 equiv), and MeOH (90 ml). The condenser was attached, and it was refluxed for 7 hr. After that, it was cooled to room temperature and concentrated in vacuo. The generated solid residue was purified by reversed-phase SiO2-gel chromatography (H2O only) to provide the desired compound (11) in 53% yield (599 mg, 1.47 mmol). This compound was reacted with 1 N HCl(aq) (3.5 ml) at room temperature and stirred overnight. After the concentration in vacuo, F-ly (12) was obtained in 97% yield (438 mg, 1.42 mmol) (Scheme 3). The spectroscopic data obtained are consistent with those previously reported in the literature (Thornalley et al., 1999).
Synthesis of F-ly (12).
Preparation of samples and methodology for RNAseq
RNA preparation for RNAseq was performed using the Qiagen RNeasy Mini kit (Cat. No. 73404). Total RNA extraction was performed from day 1 adult animals (n = 30) picked and collected in 20 µl M9 buffer per condition. Five biological replicates were used for wildtype N2 and mutant animals. RNAseq on the extracted total RNA was executed at the University of Minnesota Genomics Core (UMGC) using their sequencing protocol for HiSeq 2500 High Output (HO) mode and 50 bp paired-end sequencing following Illumina Library Preparation. RNAseq coverage was ~22 million reads per sample to perform downstream bioinformatics analyses.
Bioinformatic analysis
RNAseq global transcriptome data were subjected to Gene Ontology (GO) based functional classification using the Database for Annotation, Visualization and Integrated Discovery (DAVID) v.6.8. We employed the heatmap2 (Galaxy Version 3.0.1) function from R ggplot2 package to visualize the bioinformatics data.
HOMER analysis was used to identify the TFs for the 66 differentially expressed gene from Figure 3A. A threshold of 18% was used to select potential TFs for further screening. Please refer to the flowchart in Figure 4B.
Reverse transcription polymerase chain reaction
Total RNA was extracted from nearly 100 µl of tightly packed age-synchronized adult worm pellet collected in 1 ml TRIzol reagent provided by Qiagen RNeasy Mini kit (Cat. No. 73404) following manufacturer’s protocol. Subsequently, 1 μg total RNA was used as a template for cDNA synthesis. cDNA was synthesized using the iScriptTM cDNA synthesis kit (Bio-Rad, CA) following the manufacturer’s protocol. q-PCR was carried out using the PCR Biosystems Sygreen Blue Mix Separat -ROX (Cat. No. 17-507DB) in a LightCycler 480 Real-Time PCR system (Roche Diagnostics Corp, IN). Quantification was performed using the comparative ΔΔCt method and normalization for internal reference was done using either act-5 or pmp-2. All assays were performed with 3 technical replicates followed by five to six biological replicates. Following are qPCR primers used: (1) act-5 gene primers are ‘TCCAATCTATGAAGGATATGCCCTCCC and AAAGCTTCTCTTTGATGTCCCGGAC’. pmp-2 primer pair is ‘ATCTTTCAAAGCCAATCCTCGAC and GAGATAAGTCAGCCCAACTCC’. glod-4: TGTTCTGAATATGAAAGTTCTTCGCCACG and GATGACGATTGCTCTATAATCATTACCCAACTC. elt-3: GCCGTTCAATATTTTTGAATTGAACCTTTCAAACTT and TTTTTTTCATCGGCTTCGGCTCG. tdc-1: CGACGAGTTGTTCCTGCTATT and CGGATGTTGCCAATGAGTTATTC. tyra-2: GGAAGAGGAGGAAGAAGATAGCGAAAGTAG and ATCTCGCTTTTCATCCGAGTCTTCATC. tyra-3: CATCGATGGCCGCTTGGTC and CTTGTTCTCGGGTATTTGAGCGGT. And ser-2: GGAACAATTACGTACTTGGTAATTATTGCAATGAC and ATATCGCCACCGCCAGATCG.
Lifespan assay
Lifespan assays were performed in Thermo Scientific Precision incubators at 20°C. Timed egg laying was performed to obtain a synchronized animal population, which were either placed onto NGM plates seeded with 5× concentrated E. coli OP50-1. Post-L4 stage or young adult worms (60–65 hr from egg laying) were added to FuDR (5-fluoro-2 deoxyuridine) NGM plates to inhibit the development and growth of progeny. After 3 days, animals were transferred to a new 60 mm NGM seeded with OP50-1 and scored every other day thereafter. Forty-five to eighty animals were considered for each lifespan experiment, and two to three biological replicates were performed. Animal viability was assessed visually and with gentle prodding on the head. Animals were censored in the event of internal hatching of the larvae, body rupture, or crawling of larvae from the plates (Chaudhuri et al., 2016).
Assay for assessing neuronal damage
Neuronal damage was assayed using a pan-neuronal GFP reporter strain under different conditions on day 8 of adulthood. Animals were paralyzed using freshly prepared 5 mM levamisole in M9 buffer and mounted on 2% agar pads under glass coverslips. Neuronal damage was visually inspected under an upright Olympus BX51 compound microscope coupled with a Hamatsu Ocra ER digital camera. Images were acquired under the ×40 objective. Neuronal deterioration was examined and characterized by loss of fluorescent intensity of nerve ring, abnormal branching of axon/dendrite, and thinning and fragmentation of axons and neuronal commissures (Bijwadia et al., 2021). Quantification and imaging of animals harboring damage were performed using the ImageJ software (http://imagej.nih.gov/ij/). To reduce experimental bias, this assay was performed genotype blind with two biological repeats.
Mass spectrometry quantification of MG-H1
MG-H1 in worm homogenates were quantified using LC–MRM following 2,4,6-trinitrobenzene sulfonate derivatization as described in Hashimoto et al., 2013. An Agilent 1260 HPLC connected to a Sciex 5500 QQQ mass spectrometer was used. The LC conditions were modified as the follows: an Acquity UPLC BEH C18 (130 Å, 1.7 µm, 2.1 mm × 30 mm) column was used. The mobile phase (organic: methanol, aqueous: water containing 0.1% FA) was used with a linear gradient of 0–50% of the organic mobile phase over 1.5 min at 0.40 ml/min at 40°C. The gradient was increased to 100% organic at 1.6 min, held for 0.4 min, and equilibrated for an additional 0.4 min. The injections volume was 3 µl. MRM parameters were used as described in Hashimoto et al. with instrument-specific optimization (Hashimoto et al., 2013).
Statistical analysis
All data analyses for lifespan, pharyngeal pumping assays, and gene expression were performed using GraphPad Prism (GraphPad Software, Inc, La Jolla, CA). Survival curves were plotted using the Kaplan–Meier method, and a comparison between the survival curves to measure significance (p values) was performed using log-rank (Mantel–Cox) test. Two groups were compared for significance using an unpaired Student’s t-test. Multiple group comparison was performed by one-way analysis of variance with either Fisher’s LSD or Dunnett’s multiple comparisons test, or Sidak’s multiple comparisons test was used to compare between specific groups. p values from the significance testing were designated as follows: *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.
Additional informationCompeting interests
No competing interests declared
Reviewing editor, eLife
Author contributions
Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing – original draft, Writing – review and editing
Conceptualization, Data curation, Formal analysis, Methodology, Writing – original draft
Conceptualization, Data curation, Formal analysis, Writing – review and editing
Data curation, Formal analysis, Writing – review and editing
Conceptualization, Data curation, Formal analysis, Writing – original draft
Data curation, Formal analysis, Writing – review and editing
Formal analysis
Data curation, Formal analysis, Methodology
Formal analysis, Methodology
Data curation, Formal analysis
Supervision, Funding acquisition
Formal analysis, Methodology
Data curation, Formal analysis
Conceptualization, Resources, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing – original draft, Project administration, Writing – review and editing
Additional files
A list of genes identified in RNAseq between N2 wildtype and <italic>glod-4</italic> knockdown along with the data of fold change and significance.
Data availability
All data generated during this study are included in the manuscript. The RNAseq data is included as supplementary file.
Acknowledgements
This work was supported by grants from NIH (R01AG061165 and R01AG068288) to PK, the Larry L Hillblom Foundation (2021-A-007-FEL) to PK as well as R01DK133196 and R35GM137910 to JJG. We thank Professors Keith Blackwell, Suneil Koliwad, Malene Hansen, and the members of the Kapahi lab for their valuable suggestions. We thank Dr. Feimei Zhu for her guidance in troubleshooting the food clearance assay, Dr. Kiyomi Kaneshiro for her guidance in organizing the RNAseq data and Professor Jingru Sun as well as Phillip Wibisono for material support. We also thank the Buck Institute’s morphology and imaging core for their valuable assistance.
ReferencesAielloMSilaniVRumiatiRI2016You stole my food! Eating alterations in frontotemporal dementiaNeurocase2240040910.1080/13554794.2016.119795227327171AlkemaMJHunter-EnsorMRingstadNHorvitzHR2005Tyramine Functions independently of octopamine in the Caenorhabditis elegans nervous systemNeuron4624726010.1016/j.neuron.2005.02.02415848803AllamanIBélangerMMagistrettiPJ2015Methylglyoxal, the dark side of glycolysisFrontiers in Neuroscience92310.3389/fnins.2015.0002325709564AveryLHorvitzHR1990Effects of starvation and neuroactive drugs on feeding in Caenorhabditis elegansThe Journal of Experimental Zoology25326327010.1002/jez.14025303052181052BierhausAFlemingTStoyanovSLefflerABabesANeacsuCSauerSKEberhardtMSchnölzerMLasitschkaFNeuhuberWLKichkoTIKonradeIElvertRMierWPiragsVLukicIKMorcosMDehmerTRabbaniNThornalleyPJEdelsteinDNauCForbesJHumpertPMSchwaningerMZieglerDSternDMCooperMEHaberkornUBrownleeMReehPWNawrothPP2012Methylglyoxal modification of Nav1.8 facilitates nociceptive neuron firing and causes hyperalgesia in diabetic neuropathyNature Medicine1892693310.1038/nm.275022581285BijwadiaSRMortonKMeyerJN2021Quantifying levels of dopaminergic neuron morphological alteration and degeneration in Caenorhabditis elegansJournal of Visualized Experiments206289410.3791/6289434866619ChaseDLKoelleMR2007Biogenic amine neurotransmitters in C. elegansWormBookWormBook11510.1895/wormbook.1.132.118050501ChaudhuriJBoseNGongJHallDRifkindABhaumikDPeirisTHChamoliMLeCHLiuJLithgowGJRamanathanAXuXZSKapahiP2016A Caenorhabditis elegans model Elucidates a conserved role for TRPA1-Nrf Signaling in reactive α-Dicarbonyl DetoxificationCurrent Biology263014302510.1016/j.cub.2016.09.02427773573ChaudhuriJBainsYGuhaSKahnAHallDBoseNGugliucciAKapahiP2018The role of advanced glycation end products in aging and metabolic diseases: Bridging association and causalityCell Metabolism2833735210.1016/j.cmet.2018.08.01430184484ChenJHLinXBuCZhangX2018Role of advanced glycation end products in mobility and considerations in possible dietary and nutritional intervention strategiesNutrition & Metabolism157210.1186/s12986-018-0306-730337945DallièreNHolden-DyeLDillonJO’ConnorVWalkerRJ2017Caenorhabditis elegans feeding behaviorsOxford Research Encyclopedia of NeuroscienceNeuroscienceDistlerMGPalmerAA2012Role of Glyoxalase 1 (Glo1) and methylglyoxal (MG) in behavior: recent advances and mechanistic insightsFrontiers in Genetics325010.3389/fgene.2012.0025023181072ElluluMSPatimahIKhaza’aiHRahmatAAbedY2017Obesity and inflammation: the linking mechanism and the complicationsArchives of Medical Science1385186310.5114/aoms.2016.5892828721154GilleardJSShafiYBarryJDMcGheeJD1999ELT-3: A Caenorhabditis elegans GATA factor expressed in the embryonic epidermis during morphogenesisDevelopmental Biology20826528010.1006/dbio.1999.920210191044GolegaonkarSTabrezSSPanditASethurathinamSJagadeeshaprasadMGBansodeSSampathkumarS-GKulkarniMJMukhopadhyayA2015Rifampicin reduces advanced glycation end products and activates DAF-16 to increase lifespan in Caenorhabditis elegansAging Cell1446347310.1111/acel.1232725720500GreerERPérezCLVan GilstMRLeeBHAshrafiK2008Neural and molecular dissection of a C. elegans sensory circuit that regulates fat and feedingCell Metabolism811813110.1016/j.cmet.2008.06.00518680713GrossADCoatsJRDukeSOSeiberJN2014Biopesticides: state of the art and future opportunitiesG-Protein-Coupled Receptors (GPCRs) as Biopesticide Targets: A Focus on Octopamine and Tyramine ReceptorsWashington, DCACS Poblications455610.1021/bk-2014-1172HashimotoCIwaiharaYChenSJTanakaMWatanabeTMatsuiT2013Highly-Sensitive detection of free advanced Glycation End-products by Liquid Chromatography-Electrospray Ionization-Tandem Mass Spectrometry with 2,4,6-Trinitrobenzene Sulfonate derivatizationAnalytical Chemistry854289429510.1021/ac400294qHeYZhouCHuangMTangCLiuXYueYDiaoQZhengZLiuD2020Glyoxalase system: A systematic review of its biological activity, related-diseases, screening methods and small molecule regulatorsBiomedicine & Pharmacotherapy = Biomedecine & Pharmacotherapie13111066310.1016/j.biopha.2020.11066332858501HellwigMGeisslerSMatthesRPetoASilowCBrandschMHenleT2011Transport of free and peptide-bound glycated amino acids: synthesis, transepithelial flux at Caco-2 cell monolayers, and interaction with apical membrane transport proteinsChembiochem121270127910.1002/cbic.20100075921538757HossainPKawarBEl NahasM2007Obesity and diabetes in the developing world--a growing challengeThe New England Journal of Medicine35621321510.1056/NEJMp06817717229948IsmailLMaterwalaHAl KaabiJ2021Association of risk factors with type 2 diabetes: A systematic reviewComputational and Structural Biotechnology Journal191759178510.1016/j.csbj.2021.03.00333897980JaegerHJanositzAKnorrD2010The Maillard reaction and its control during food processing: the potential of emerging technologiesPathologie-Biologie5820721310.1016/j.patbio.2009.09.01619896291Ji Ying SzeMVLoerCShiYRuvkunG2000Food and metabolic signalling defects in a Caenorhabditis elegans serotonin-synthesis mutantNature40356056410.1038/35000609KeramatSAAlamKRanaRHChowdhuryRFarjanaFHashmiRGowJBiddleSJH2021Obesity and the risk of developing chronic diseases in middle-aged and older adults: Findings from an Australian longitudinal population survey, 2009-2017PLOS ONE16e026015810.1371/journal.pone.026015834784404Kold-ChristensenRJensenKKSmedegård-HolmquistESørensenLKHansenJJørgensenKAKristensenPJohannsenM2019ReactELISA method for quantifying methylglyoxal levels in plasma and cell culturesRedox Biology2610125210.1016/j.redox.2019.10125231254735LeeTHYYauSY2020From obesity to hippocampal neurodegeneration: pathogenesis and non-pharmacological interventionsInternational Journal of Molecular Sciences2220110.3390/ijms2201020133379163LiZLiYYiYHuangWYangSNiuWZhangLXuZQuAWuZXuT2012Dissecting a central flip-flop circuit that integrates contradictory sensory cues in C. elegans feeding regulationNature Communications377610.1038/ncomms178022491324LimaM2013Encyclopedia of Biological ChemistryElsevierLiuXXiaBHuLNiZThakurKWeiZ2020Maillard conjugates and their potential in food and nutritional industries: A reviewFood Frontiers138239710.1002/fft2.43LucaFPerryGHDi RienzoA2010Evolutionary adaptations to dietary changesAnnual Review of Nutrition3029131410.1146/annurev-nutr-080508-14104820420525MachielsDIstasseL2022Maillard reaction: importance and applications in food chemistryAnnales de Médecine Vétérinaire146347352MaillardLC1912Action des acides amines sur les sucres: formation des melanoidines par voie methodiqueComptes Rendus de l’Académie Des Sciences1546668MangoSELambieEJKimbleJ1994The pha-4 gene is required to generate the pharyngeal primordium of Caenorhabditis elegansDevelopment12030193031MangoSE2007The C. elegans pharynx: a model for organogenesisWormBookWormBook12610.1895/wormbook.1.129.118050503MillerAASpencerSJ2014Obesity and neuroinflammation: a pathway to cognitive impairmentBrain, Behavior, and Immunity42102110.1016/j.bbi.2014.04.00124727365MonnierVMKohnRRCeramiA1984Accelerated age-related browning of human collagen in diabetes mellitusPNAS8158358710.1073/pnas.81.2.5836582514MorcosMDuXPfistererFHutterHSayedAARThornalleyPAhmedNBaynesJThorpeSKukudovGSchlottererABozorgmehrFEl BakiRASternDMoehrlenFIbrahimYOikonomouDHamannABeckerCZeierMSchwengerVMiftariNHumpertPHammesHPBuechlerMBierhausABrownleeMNawrothPP2008Glyoxalase-1 prevents mitochondrial protein modification and enhances lifespan in Caenorhabditis elegansAging Cell726026910.1111/j.1474-9726.2008.00371.x18221415MottramDSWedzichaBLDodsonAT2022Acrylamide is formed in the Maillard reactionNature41944844910.1038/419448aMuddapuVRDharshiniSAPChakravarthyVSGromihaMM2020Neurodegenerative diseases - is metabolic deficiency the root cause?Frontiers in Neuroscience1421310.3389/fnins.2020.0021332296300NowotnyKSchröterDSchreinerMGruneT2018Dietary advanced glycation end products and their relevance for human healthAgeing Research Reviews47556610.1016/j.arr.2018.06.00529969676NyamsurenOFaggionatoDLochWSchulzeEBaumeisterR2007A mutation in CHN-1/CHIP suppresses muscle degeneration in Caenorhabditis elegansDevelopmental Biology31219320210.1016/j.ydbio.2007.09.03317961535PapaevgeniouNHoehnATurJAKlotzL-OGruneTChondrogianniN2019Sugar-derived AGEs accelerate pharyngeal pumping rate and increase the lifespan of Caenorhabditis elegansFree Radical Research531056106710.1080/10715762.2019.166140331514551Perez-GomezACarreteroMWeberNPeterkaVToATitovaVSolisGOsbornOPetrascheckM2018A phenotypic Caenorhabditis elegans screen identifies A selective suppressor of antipsychotic-induced hyperphagiaNature Communications9527210.1038/s41467-018-07684-y30532051PirriJKMcPhersonADDonnellyJLFrancisMMAlkemaMJ2009A tyramine-gated chloride channel coordinates distinct motor programs of A Caenorhabditis elegans escape responseNeuron6252653810.1016/j.neuron.2009.04.01319477154PrasadCDavisKEImrhanVJumaSVijayagopalP2019Advanced glycation end products and risks for chronic diseases: Intervening through lifestyle modificationAmerican Journal of Lifestyle Medicine1338440410.1177/155982761770899131285723ProcacciniCSantopaoloMFaicchiaDColamatteoAFormisanoLde CandiaPGalganiMDe RosaVMatareseG2016Role of metabolism in neurodegenerative disordersMetabolism651376139010.1016/j.metabol.2016.05.01827506744RabbaniNThornalleyPJ2014Measurement of methylglyoxal by stable isotopic dilution analysis LC-MS/MS with corroborative prediction in physiological samplesNature Protocols91969197910.1038/nprot.2014.12925058644RahbarSBlumenfeldORanneyHM1969Studies of an unusual hemoglobin in patients with diabetes mellitusBiochemical and Biophysical Research Communications3683884310.1016/0006-291x(69)90685-85808299RaizenDSongBMTrojanowskiNYouYJ2012Methods for measuring pharyngeal behaviorsWormBook1811310.1895/wormbook.1.154.123255345RamasamyRVannucciSJYanSSDHeroldKYanSFSchmidtAM2005Advanced glycation end products and RAGE: a common thread in aging, diabetes, neurodegeneration, and inflammationGlycobiology1516R28R10.1093/glycob/cwi05315764591RavichandranMPriebeSGrigolonGRozanovLGrothMLaubeBGuthkeRPlatzerMZarseKRistowM2018Impairing L-threonine catabolism promotes healthspan through methylglyoxal-mediated proteohormesisCell Metabolism2791492510.1016/j.cmet.2018.02.00429551589RexEMolitorSCHapiakVXiaoHHendersonMKomunieckiR2004Tyramine receptor (SER-2) isoforms are involved in the regulation of pharyngeal pumping and foraging behavior in Caenorhabditis elegansJournal of Neurochemistry911104111510.1111/j.1471-4159.2004.02787.x15569254RexEHapiakVHobsonRSmithKXiaoHKomunieckiR2005TYRA-2 (F01E11.5): a Caenorhabditis elegans tyramine receptor expressed in the MC and NSM pharyngeal neuronsJournal of Neurochemistry9418119110.1111/j.1471-4159.2005.03180.x15953361RuhmCJ2012Understanding overeating and obesityJournal of Health Economics3178179610.1016/j.jhealeco.2012.07.00422935355ShobatakeRTakasawaKOtaHItaya-HironakaAYamauchiASakuramoto-TsuchidaSUchiyamaTMakinoMSugieKTakasawaSUenoS2018Up-regulation of POMC and CART mRNAs by intermittent hypoxia via GATA transcription factors in human neuronal cellsThe International Journal of Biochemistry & Cell Biology9510010710.1016/j.biocel.2017.12.01229275211StadlerRHBlankIVargaNRobertFHauJGuyPARobertMCRiedikerS2002Acrylamide from Maillard reaction productsNature41944945010.1038/419449a12368845ThornalleyPJLangborgAMinhasHS1999Formation of glyoxal, methylglyoxal and 3-deoxyglucosone in the glycation of proteins by glucoseThe Biochemical Journal344 Pt 110911610548540TrojanowskiNFRaizenDMFang-YenC2016Pharyngeal pumping in Caenorhabditis elegans depends on tonic and phasic signaling from the nervous systemScientific Reports62294010.1038/srep2294026976078TsalikELNiacarisTWenickASPauKAveryLHobertO2003LIM homeobox gene-dependent expression of biogenic amine receptors in restricted regions of the C. elegans nervous systemDevelopmental Biology2638110210.1016/s0012-1606(03)00447-014568548UribarriJWoodruffSGoodmanSCaiWChenXPyzikRYongAStrikerGEVlassaraH2010Advanced glycation end products in foods and a practical guide to their reduction in the dietJournal of the American Dietetic Association11091191610.1016/j.jada.2010.03.01820497781UribarriJCaiWWoodwardMTrippEGoldbergLPyzikRYeeKTansmanLChenXManiVFayadZAVlassaraH2015Elevated serum advanced glycation endproducts in obese indicate risk for the metabolic syndrome: a link between healthy and unhealthy obesity?The Journal of Clinical Endocrinology and Metabolism1001957196610.1210/jc.2014-392525695886Vicente MirandaHSzegoÉMOliveiraLMABredaCDarendeliogluEde OliveiraRMFerreiraDGGomesMARottROliveiraMMunariFEnguitaFJSimõesTRodriguesEFHeinrichMMartinsICZamoloIRiessOCordeiroCPonces-FreireALashuelHASantosNCLopesLVXiangWJovinTMPenqueDEngelenderSZweckstetterMKluckenJGiorginiFQuintasAOuteiroTF2017Glycation potentiates α-synuclein-associated neurodegeneration in synucleinopathiesBrain1401399141910.1093/brain/awx05628398476VistoliGDe MaddisDCipakAZarkovicNCariniMAldiniG2013Advanced glycoxidation and lipoxidation end products (AGEs and ALEs): an overview of their mechanisms of formationFree Radical Research47 Suppl 132710.3109/10715762.2013.81534823767955WolinKYCarsonKColditzGA2010Obesity and cancerThe Oncologist1555656510.1634/theoncologist.2009-028520507889WuZIsikMMorozNSteinbaughMJZhangPBlackwellTK2019Dietary restriction extends lifespan through metabolic regulation of innate immunityCell Metabolism291192120510.1016/j.cmet.2019.02.01330905669ZhangQWangYFuL2020Dietary advanced glycation end-products: Perspectives linking food processing with health implicationsComprehensive Reviews in Food Science and Food Safety192559258710.1111/1541-4337.125933333697210.7554/eLife.82446.sa0Editor's evaluationLeiserScott FReviewing Editorhttps://ror.org/00jmfr291University of MichiganUnited States
This work, examining how Advanced Glycation End-products (AGEs), commonly found in processed and other cooked foods, affect eating behavior and signaling in the nematode C. elegans, is in a fundamentally important area of research with clear translational potential for humans. The authors present a combination of solid and compelling evidence to mechanistically study how AGEs affect eating behavior. The objectives of this study are not only to provide basic information relevant to phenomena that are likely to be conserved throughout the animal kingdom, but also to provide information that could be important in human health for the understanding of disorders caused by the consumption of processed foods.
10.7554/eLife.82446.sa1Decision letterLeiserScott FReviewing Editorhttps://ror.org/00jmfr291University of MichiganUnited States
Our editorial process produces two outputs: (i) public reviews designed to be posted alongside the preprint for the benefit of readers; (ii) feedback on the manuscript for the authors, including requests for revisions, shown below. We also include an acceptance summary that explains what the editors found interesting or important about the work.
Decision letter after peer review:
Thank you for submitting your article "Methylglyoxal-derived hydroimidazolone, MG-H1, increases food intake by altering tyramine signaling via the GATA transcription factor ELT-3 in Caenorhabditis elegans" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Jessica Tyler as the Senior Editor. The reviewers have opted to remain anonymous.
The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.
Essential revisions:
1) As you will see from the reviews attached below, all reviewers found this manuscript interesting and were enthusiastic about the central idea of the work. However, all three reviewers also agreed that there were mechanistic deficiencies in the manuscript that make it unsuitable for publication in eLife without additional data. In particular the reviewers agreed that mechanistic data about how MGH1 acts on the tyraminergic pathway to potentiate food intake were lacking in several places and weak in others (e.g., the elt-3 and tdc-1 data), and require strengthening to better support the proposed model. Specific rescues (elt-3 and tdc-1), functional experiments, and manipulation of tyramine-producing cell activity are all important approaches to obtain mechanistic information that will help identify how some of these players interact with each other (e.g., the cells and circuits that govern the processes).
2) Additional experiments, such as bacterial clearance (reviewer 1, comment 4), and something to rectify reviewer 3, comment 2, should also be completed.
3) Additionally, please note that both reviewer 1 and reviewer 3 point out several instances in the writing and/or figures that require clarification/revision, although these revisions may or may not require data depending on the response (e.g., pumping controls, backcrossing of worms). All reviewer comments are included below for individual responses.
Reviewer #1 (Recommendations for the authors):
These are the points that I feel need to be addressed:
1) The authors claim that MGH-1 modulates (positively) the tyraminergic signal through the activation of the TF elt-3. This is based on the fact that mutations in both tdc-1 and elt-3 suppress the exacerbated pharyngeal pumping of glod-4 mutants (Figure 2E and 3D), that elt-3 expression is increased in MGH1-treated worms (Figure 3F) and that in elt-3-deficient mutants tdc-1 expression is low (Figure 3 E).
First, only two determinations are shown in Figure 3 E (where the mRNA of various elt-3 targets is measured). I think it is too premature to draw conclusions with n=2. It is important to increase the n, especially considering that this is a relevant figure for the whole conclusion of the paper.
Assuming that the conclusions obtained in Figure 1 E are valid, the reasoning that MGH-1 induces the tyraminergic signal through the activation of the TF elt-3 sounds logical. However, this hypothesis does not agree with other data. For instance, MGH-1 treatment does not significantly modulate tdc-1 expression (Figure 1 F). Ser-2 expression is induced upon MGH-1 treatment (Figure 1F) but does not respond to elt-3 (Figure 1E). This raises questions as to how all these players interrelate to modulate the increase in feeding due to AGEs.
It may be possible that the increase in elt-3 upon MGH1 treatment is not sufficient to positively modulate tdc-1 Expression? In this context, one can think that a basal level of elt-3 is needed because otherwise, tyramine release would be low (and this explains the glod-4, tdc-1 and glod-4; elt-3 phenotypes?).
tdc-1 rescue assays with promoters lacking the elt-3 TF binding sequence could demonstrate how important the modulation of tdc-1 expression by elt-3 is during MGH-1 accumulation.
2) Results page 6 (line 117): As exogenous serotonin increases bacterial clearance in glod-4 animals, the authors claim that the increase in pharyngeal pumping in these animals is independent of serotonin. This is not concluded from this experiment. It may in fact be the case that even though it depends on endogenous serotonin release, the exogenous addition of serotonin at a high concentration may further induce pharyngeal pumping. The experiment in the manuscript that actually shows that the induction of pharyngeal pumping by MGH-1 does not depend on serotonin is shown in Figure suppl. 2 A, where MGH-1 treatment induces pumping even in tph-1 worms. To assume that the increase in pumping in glod-4 is independent of 5ht, one would have to compare the pharyngeal pumping (and bacterial clearance) of tph-1 and glod-4;tph-1 worms.
3) Results Page 7 "In addition, we also demonstrated that MG-H1 regulates pharyngeal pumping rate in a dose-dependent manner (Figure 1E)". To assert this, one would have to compare the pharyngeal pumping between the MGH-1 conditions themselves, rather than doing a student test against the naive condition. In addition, it would be advisable to do the experiments on the same day (given the variation in pharyngeal pumping in the naive condition. For example, the first naive shows a lot of dispersion and the third naive condition seems to have a lower average than the first one).
4) Results Page 7 (line 130). "In addition to food consumption, glod-4 mutant exhibited a significantly increased preference towards food source OP50-1 at day 1 and day 3 of adulthood compared to wild-type N2 worms (Figure 1F + Figure Suppl. 1F). Furthermore, we noticed that wild-type N2 worms preferred exogenous MG-H1 compared to MGO when provided with bacterial food source E. coli OP50-1. We did not observe this phenotype in the glod-4 mutant background (Figure 1G), suggesting that MG-H1 and glod-4 null mutation increases feeding by overlapping mechanism."
While the food race experiments may be interesting, the connection with increased food intake is, at the very least, very indirect. The experiment that would actually demonstrate that MGH1 and glod-4 null mutation increases feeding by overlapping mechanism would be the bacterial clearance assay shown in Figure 1D, where the addition of MGH1 does not increase bacterial clearance in a glod-4 mutant background. Something similar could be done with pharyngeal pumping, to confirm the overlapping mechanisms.
With respect to figure 1G, what would be the value corresponding to the attraction to MGO? It is not indicated on the axis.
Beyond that, these food race experiments should not be included in the context of enhanced feeding rate.
5) There is a very large variation in wild-type pharyngeal pumping between experiments. For example, in some experiments it is as high as (160 pp/30sec. Figure 3c) and in others it can be as low as (110 pp/30sec. Figure 2Suppl.) While I understand that for the supplementary figures the authors performed a quick count the difference is still large. Even within the main figures there is a high variation in the controls (see, for example, Figures 2 E, F and G). While I understand that experiments such as pharyngeal pumping can be variable between experiments, I am particularly struck by such variation. Are all the experiments done with animals of the same age and raised under exactly the same conditions?
6) It is not clear, and the authors do not present any model of the mechanism by which MGH1 excess modulates the tyraminergic signal to increase food intake. It is only shown that elt-3 may be important and that tdc-1, ser-2 and tyra-2 appear to be relevant. There are no rescue or functional experiments showing where these players would be playing their role.
For example, would tyramine release be affected at the level of RIM, RIC or UV cells? Perhaps it would be good to specifically silence these neurons (e.g. with the expression of histamine-triggered anion channels, see Pokala et al., PNAS, 2014 or De Rosa et al., Nature, 2019) and see if MGH-1 is able to induce pharyngeal pumping.
It is also unclear where elt-3 is relevant. It would be important to express this TF specifically in tdc-1 expressing cells and analyze whether the response is similar to wild-type.
This is one of my main concerns regarding the manuscript. The lack of mechanistic details to understand how AGEs accumulation increases food intake argues against the significance of the work.
7) Exogenous TA has been shown to reduce pharyngeal pumping (Greer et al., 2008). Furthermore, the absence of SER-2 suppresses the tyraminergic inhibition of exacerbated pharyngeal pumping in serotonin-treated worms (Rex et a, 2004. https://doi.org/10.1111/j.1471-4159.2004.02787.x).
Since the authors' results point more towards a positive modulation of pharyngeal pumping by tyramine and ser-2 , it would be important to discuss this paradox further.
8) Many of the strains listed have not been backcrossed to clean up the background. For example: RB1690 ser-2(ok2103); VC343 glod-4(gk189); VC125 tyra-3(ok325); RB745 ser-4(ok512), etc. For tyra-2 it is not indicated which strain is used.
If backcrosses have not been performed, it is essential to perform the experiments with the clean strains. If they have done them, the strain they used for the experiments is not called as they named it. Please clarify this.
At the same time, it would be advisable that in the absence of rescue experiments, they use other alleles to confirm the observed phenotypes.
Reviewer #2 (Recommendations for the authors):
The study is generally well done, but the strains don't appear to be outcrossed, and none of the effects are rescued. At least for the main results, tdc-1 and elt-3, there are multiple mutants available at CGC. The authors should rescue the original mutant or show the same results using a different allele. It would also be helpful for the reader to show the alleles in the figure legend.
Reviewer #3 (Recommendations for the authors):
There are several points that need to be substantially improved.
1. The writing of the manuscript needs to be improved substantially. The introduction is very confusing with too many abbreviations. I would suggest a diagram to show the reaction of α-dicarbonyl (α-DG), AGEs, glyoxalases. In each of these terms, you can list the examples in that category, for example: α-DG (MGO, GO, 3DG etc.), AGEs (MG-H1, CEL, CML, GOLD, MOLD), glyoxalases (glod-4, glod-1, etc.). The introduction is not organized with a focus in each paragraph. Transitions between sentences need to be improved substantially too.
2. It is not explained or addressed in the manuscript why glod-4 KO in which we would expect less MG-H1, and MG-H1 supplementation both increase pumping. It will be helpful to measure MG-H1 level in glod-4 KO. This will also help to understand the result that the pumping rate increases are MG-H1 dose-dependent, however, glod-4 KO and MG-H1 supplementation is not additive in the increase of pumping.
3. Figure 2B shows all 66 genes are upregulated in glod-4 compared to WT. A concern is whether the gene expression is properly normalized. Are there downregulated genes other than the 66 genes? How many genes are up- and down- regulated in glod-4 vs. WT. All these necessary descriptions are not mentioned in the results part. The labeling of both A and B panels are too small for recognizing the gene and pathway names. Among the 66 "neurotransmitters and feeding genes", how many of them are neurotransmitter genes and how were they defined? Are tph-1, tdc-1, tbh-1, cat-2 genes in this list? Similar question, how were the "feeding genes" defined?
4. Figure 2B used glod-4 RNAi while other results pumping data were collected on glod-4KO. Why glod-4 KO was not used in transcriptomics data. And please label on panel B indicating glod-4 RNAi. And it is described in the method part that RNA-seq was done on "N2 and mutant animals" but not RNAi.
5. tdc-1 and the two tyramine receptors knockouts abrogate pumping rate by MG-H1 and glod-4. This would indicate that tyramine release is necessary for the pumping rate increase by MG-H1 and glod-4. Did the author try rescue the abrogation by tyramine supplementation? This is a valid control to strengthen the hypothesis.
6. The authors previously showed that glod-4 KO shortens lifespan, and supplementation of glod-4 substrates MGO also shortens lifespan. The results in Figure 4 of lifespan rescue by tdc-1 is not sufficient to support that increased feeding shortens glod-4 KO lifespan suggested in the manuscript. The valid conclusion from this data is that the lifespan decreases by glod-4 KO requires tyramine synthesis and probably tyramine signaling since the two receptors are also required for the shorter lifespan. Does MG-H1 addition rescue or further shorten glod-4 KO lifespan since MG-H1 also increases pumping rate not additively to glod-4 KO? Figure 4D labeling is too small.
10.7554/eLife.82446.sa2Author response
Essential revisions:
1) As you will see from the reviews attached below, all reviewers found this manuscript interesting and were enthusiastic about the central idea of the work. However, all three reviewers also agreed that there were mechanistic deficiencies in the manuscript that make it unsuitable for publication in eLife without additional data. In particular the reviewers agreed that mechanistic data about how MGH1 acts on the tyraminergic pathway to potentiate food intake were lacking in several places and weak in others (e.g., the elt-3 and tdc-1 data), and require strengthening to better support the proposed model. Specific rescues (elt-3 and tdc-1), functional experiments, and manipulation of tyramine-producing cell activity are all important approaches to obtain mechanistic information that will help identify how some of these players interact with each other (e.g., the cells and circuits that govern the processes).
We thank the editors and reviewers for their time in reviewing our manuscript. We also appreciate the strengths pointed out by the reviewers and their constructive feedback. We address these concerns below.
To strongly demonstrate the necessity of tyramine signaling in glod-4 mutants to increase pharyngeal pumping, we supplemented tyramine exogenously (instead of laser ablation/manipulation of tyramine-producing cells) to either elt-3;glod-4 or tdc-1;glod-4 double mutants and analyzed the pumping behavior. Further, we chose to supplement tyramine as a rescue strategy in the mutants that lack tyramine signaling. Our new results in Figure B+C demonstrate that exogenous tyramine treatment of double mutants increased their pumping levels to glod-4 single mutant worms. Thus, supporting the hypothesis that lack of elt-3 reduces tyramine signaling (note the reduced expression of tdc-1 in elt-3 mutants, Figure 4E) to suppress the pumping in elt-3;glod-4 double mutants. This data also demonstrates the importance of elt-3 in mediating increased pumping in glod-4 mutants via tyramine. Similarly, exogenous tyramine supplementation in tdc-1;glod-4 double mutant also increased pharyngeal pumping to glod-4 single mutant levels. However, the pumping is not increased in double mutants such as ser-2;glod-4, and tyra-2;glod-4 which lacks tyramine receptors (Figure D+E).
Further, we have shown in Figure 4F that MG-H1 treatment of N2 wildtype worms modulates the expression levels of the gene in the tyramine signaling pathway, i.e., increases the expression of elt-3 as well as ser-2 which are essential for increased feeding in glod-4 mutants (Figure 3+4). Thus, either exogenous treatment of MG-H1 or glod-4 mutants (with increased accumulation of MG-H1, Figure2—figure supplement 1H+1I) increased tyramine signaling by modulating genes in the tyramine pathway. The mechanism is explained in the discussion of the revised manuscript on pages 21 and 22.
However, treatment of N2 wildtype worms with tyramine resulted in a significant decrease in the pharyngeal pumping (Figure 5A). Also, tyramine is shown to suppress pharyngeal pumping in N2 wildtype worms (1). These findings demonstrate that tyramine increases pharyngeal pumping only in the glod-4 mutant background but decreases pumping in the N2 wildtype background. The mechanism behind this behavioral switch upon activation of the tyramine signaling pathway (i.e., from a suppressor of pharyngeal pumping in wildtype into a stimulator in glod-4 mutant background) remains elusive. Although tdc-1 is expressed only in RIC, RIM, UV2 cells and gonadal sheath cells, tyramine receptors are located at distinct tissues (1) (2); thus, it can be speculated that AGEs, especially MG-H1, can modify the tissue-specific expression of tyramine receptors resulting in observed behavioral switch in response to tyramine signaling. This hypothesis can only be addressed by methodologies such as single-cell RNA sequencing which is beyond the scope of the current study.
2) Additional experiments, such as bacterial clearance (reviewer 1, comment 4), and something to rectify reviewer 3, comment 2, should also be completed.
We have addressed reviewer 1-comment 4 and reviewer 3-comment 2 with appropriate experiments and explanation, please refer below.
3) Additionally, please note that both reviewer 1 and reviewer 3 point out several instances in the writing and/or figures that require clarification/revision, although these revisions may or may not require data depending on the response (e.g., pumping controls, backcrossing of worms). All reviewer comments are included below for individual responses.
All the comments that do not require additional experiments but require appropriate explanation are revised as suggested by the reviewers in the new revised manuscript. Please see the rebuttal below.
Reviewer #1 (Recommendations for the authors):
These are the points that I feel need to be addressed:
1) The authors claim that MGH-1 modulates (positively) the tyraminergic signal through the activation of the TF elt-3. This is based on the fact that mutations in both tdc-1 and elt-3 suppress the exacerbated pharyngeal pumping of glod-4 mutants (Figure 2E and 3D), that elt-3 expression is increased in MGH1-treated worms (Figure 3F) and that in elt-3-deficient mutants tdc-1 expression is low (Figure 3 E).
First, only two determinations are shown in Figure 3 E (where the mRNA of various elt-3 targets is measured). I think it is too premature to draw conclusions with n=2. It is important to increase the n, especially considering that this is a relevant figure for the whole conclusion of the paper.
As recommended by the reviewer, we have now increased the sample size to 5 biological replicates (Figure 4E). Our data shows that tdc-1 expression levels are significantly reduced in elt-3 mutant worms. Further, we now have data demonstrating that exogenous tyramine treatment on elt-3;glod-4 double mutant increased the pharyngeal pumping similar to glod-4 single mutant worms (Figure 5B) indicating that elt-3 is necessary for tyramine signaling.
However, it can be observed that ser-2 expression significantly increased in elt-3 mutant worms which can be inferred as a compensatory mechanism for lack of tyramine signaling by increasing the expression of tyramine receptor.
Assuming that the conclusions obtained in Figure 1 E are valid, the reasoning that MGH-1 induces the tyraminergic signal through the activation of the TF elt-3 sounds logical. However, this hypothesis does not agree with other data. For instance, MGH-1 treatment does not significantly modulate tdc-1 expression (Figure 1 F). Ser-2 expression is induced upon MGH-1 treatment (Figure 1F) but does not respond to elt-3 (Figure 1E). This raises questions as to how all these players interrelate to modulate the increase in feeding due to AGEs.
It may be possible that the increase in elt-3 upon MGH1 treatment is not sufficient to positively modulate tdc-1 Expression?. In this context, one can think that a basal level of elt-3 is needed because otherwise, tyramine release would be low (and this explains the glod-4, tdc-1 and glod-4; elt-3 phenotypes?).
tdc-1 rescue assays with promoters lacking the elt-3 TF binding sequence could demonstrate how important the modulation of tdc-1 expression by elt-3 is during MGH-1 accumulation.
We show that MG-H1 treatment modulates factors in the elt-3-tdc-1/tyramine pathways (Figure 4F, especially increased expression of elt-3 and ser-2 genes) to increase pharyngeal pumping. In order to strengthen the involvement of tyramine signaling, we supplemented tyramine (tyramine rescue) to either elt-3;glod-4, or tdc-1;glod-4 double mutant instead of tdc-1 rescue suggested by the reviewer. Our results in Figure B+C demonstrate that exogenous tyramine treatment of double mutants increased the pumping levels to glod-4 single mutant worms. Thus, we can understand that lack of elt-3 modulates tyramine levels to suppress the pumping in elt-3;glod-4 double mutants. Similarly, exogenous tyramine supplementation in tdc-1;glod-4 double mutant also increased pharyngeal pumping to glod-4 single mutant levels. However, the pumping is not increased in double mutants such as ser-2;glod-4, and tyra-2;glod-4 which lack tyramine receptors.
Further, from the literature we can find that tyramine suppresses the pharyngeal pumping in C. elegans (1). Thus, treatment of N2 wildtype worms with tyramine resulted in a significant decrease in the pharyngeal pumping (Figure 5A). These findings demonstrate that tyramine increases pharyngeal pumping only in glod-4 mutant background but decreases pumping in N2 wildtype background. And further analysis is required to understand the mechanistic shift in the tyramine signaling, however it is beyond the scope of this research article.
2) Results page 6 (line 117): As exogenous serotonin increases bacterial clearance in glod-4 animals, the authors claim that the increase in pharyngeal pumping in these animals is independent of serotonin. This is not concluded from this experiment. It may in fact be the case that even though it depends on endogenous serotonin release, the exogenous addition of serotonin at a high concentration may further induce pharyngeal pumping. The experiment in the manuscript that actually shows that the induction of pharyngeal pumping by MGH-1 does not depend on serotonin is shown in Figure suppl. 2 A, where MGH-1 treatment induces pumping even in tph-1 worms. To assume that the increase in pumping in glod-4 is independent of 5ht, one would have to compare the pharyngeal pumping (and bacterial clearance) of tph-1 and glod-4;tph-1 worms.
As recommended by the reviewer, we compared the pharyngeal pumping in tph-1 (mg280) and tph-1;glod-4 mutants background (Figure 2—figure supplement 1D). Worms lacking tph-1 (tryptophan hydroxylase) enzyme, as well as tph-1;glod-4 double mutants, which lack serotonin (1) (3) show decreased pumping compared to N2 wildtype worms; however, tph-1;glod-4 double mutant worms show significantly increased pumping compared to tph-1 single mutants demonstrating that glod-4 mutation dependent increase in pharyngeal pumping is independent of serotonin. Also, note that MG-H1 treatment significantly increased pharyngeal pumping in tph-1 mutants in Figure 3—figure supplement 2A which supports our claim that glod-4 mutant induced pumping is independent of serotonin.
3) Results Page 7 "In addition, we also demonstrated that MG-H1 regulates pharyngeal pumping rate in a dose-dependent manner (Figure 1E)". To assert this, one would have to compare the pharyngeal pumping between the MGH-1 conditions themselves, rather than doing a student test against the naive condition. In addition, it would be advisable to do the experiments on the same day (given the variation in pharyngeal pumping in the naive condition. For example, the first naive shows a lot of dispersion and the third naive condition seems to have a lower average than the first one).
Our conclusion is based on the stronger significance (lower p value) caused by reduced dispersion of the data at higher concentration of MG-H1. This indicates that higher concentrations of MG-H1 can increase the pharyngeal pumping in almost all the treatment worms thus predisposing the worms with lower pumping rates in the Gaussian distribution to higher pumping levels. Now, it is made clear in the manuscript on page 18.
It is important to explain that there were a few technical difficulties in recording pumping videos from several treatment groups in a narrow window of time. From Figure 2A it is very evident that pharyngeal pumping is very dynamic which changes with the worm’s development. Further, it takes about a minute to record a 40 sec pumping video which includes finding a worm and focusing to start the recording. Approximately 26 to 30 worms were recorded per treatment group which requires 1 hour (including personal fatigue experienced by the recorder). Dynamic pumping behavior can induce variations between the worms recorded in the first treatment group and the last treatment group especially when there are 5 treatment groups to record pumping videos which takes around 5 hours. Thus, we decided to record the minimum required treatment groups as possible to reduce variations caused by delayed time and worms’ development. In case, when more treatment/genetic groups need to be compared the video recording between different groups are staggered (recording of 10 worms per group followed by recording worms from other groups and repeating this cycle until 30 worms per group were recorded) to minimize the various caused.
Further, variations are also induced by fluctuations in the lab temperature, which is overcome by preparing 3 plates of worms for a single treatment group and only 10 worms are recorded per plate while the other plates are incubated at 20 ̊C.
An insight into the above-mentioned strategies has been updated on the “Materials and methods” section on page 25.
4) Results Page 7 (line 130). "In addition to food consumption, glod-4 mutant exhibited a significantly increased preference towards food source OP50-1 at day 1 and day 3 of adulthood compared to wild-type N2 worms (Figure 1F + Figure Suppl. 1F). Furthermore, we noticed that wild-type N2 worms preferred exogenous MG-H1 compared to MGO when provided with bacterial food source E. coli OP50-1. We did not observe this phenotype in the glod-4 mutant background (Figure 1G), suggesting that MG-H1 and glod-4 null mutation increases feeding by overlapping mechanism."
While the food race experiments may be interesting, the connection with increased food intake is, at the very least, very indirect. The experiment that would actually demonstrate that MGH1 and glod-4 null mutation increases feeding by overlapping mechanism would be the bacterial clearance assay shown in Figure 1D, where the addition of MGH1 does not increase bacterial clearance in a glod-4 mutant background. Something similar could be done with pharyngeal pumping, to confirm the overlapping mechanisms.
With respect to figure 1G, what would be the value corresponding to the attraction to MGO? It is not indicated on the axis.
Beyond that, these food race experiments should not be included in the context of enhanced feeding rate.
As recommended by the reviewer, we performed pharyngeal pumping quantification on glod-4 mutant worms treated with 150 µM MG-H1 and compared with N2 and glod-4 either treated with arginine or MG-H1 (Author response image 1). Our data demonstrates that glod-4 ARG does not differ significantly with glod-4 MG-H1 suggesting both glod-4 mutants and MG-H1 treatment modulates similar pathway to increase feeding. From Figure 4F, it is also evident that MG-H1 treatment induces gene expression changes in elt-3-tyramine signaling pathways which also validates that MG-H1 acts through elt-3-tyramine signaling pathway.
Quantification of pharyngeal pumping.
One way ANOVA with Fisher’s LSD test. **** p<0.0001. Error bar ± SD.
We thank the reviewer for pointing out the lack of description in old Figure 1G. We apologize that a small proofreading error resulted in a wrong graph which is updated now with two graphs as shown in Figure 2G and 2H. Figure 2G is a comparison between N2 wildtype worm treated with either MGO or MG-H1 with untreated control, showing that mixing of MGO to OP50-1 did not significantly increase the attraction of worms towards food, unlike MG-H1. Figure 2H demonstrates that glod-4 mutant’s attraction towards food is unaffected by MG-H1 when mixed with OP50-1.
5) There is a very large variation in wild-type pharyngeal pumping between experiments. For example, in some experiments it is as high as (160 pp/30sec. Figure 3c) and in others it can be as low as (110 pp/30sec. Figure 2Suppl.) While I understand that for the supplementary figures the authors performed a quick count the difference is still large. Even within the main figures there is a high variation in the controls (see, for example, Figures 2 E, F and G). While I understand that experiments such as pharyngeal pumping can be variable between experiments, I am particularly struck by such variation. Are all the experiments done with animals of the same age and raised under exactly the same conditions?
Yes, the worms were grown in the same conditions and the videos were recorded at similar ages, ± few minutes difference, throughout the project. As pointed out by the reviewer pharyngeal pumping is variable and dynamic. Thus, we realized the importance of validating our findings with alternative methodologies, for example – very important pumping data was supported by food clearance assay (Figure 2B+2D in the manuscript). Further, a few important pharyngeal pumping counts were performed double-blinded to minimize the bias.
It is important to explain that there were a few technical difficulties in recording pumping videos from several treatment groups in a narrow window of time. From Figure 2A it is very evident that pharyngeal pumping is very dynamic which changes with the worm’s development. Further, it takes about a minute to record a 40 sec pumping video which includes finding a worm and focusing to start the recording. Approximately 26 to 30 worms were recorded per treatment group per biological replicate which requires 1 hour (including personal fatigue experienced by the recorder). Dynamic pumping behavior can induce variations between the worms recorded in the first treatment group and the last treatment group especially when there are 5 treatment groups to record pumping videos which takes around 5 hours. Thus, we decided to record the minimum required treatment groups as possible to reduce variations caused by delayed time and worms’ development. In case, when more treatment/genetic groups need to be compared the video recordings between different groups are staggered (recording of 10 worms per group followed by recording worms from other groups and repeating this cycle until 30 worms per group were recorded) to minimize the variations.
Further, variations are also induced by fluctuations in the lab temperature, which is overcome by preparing 3 plates of worms for a single treatment group and only 10 worms are recorded per plate while the other plates are incubated at 20 ̊C.
To further average out the variations, the biological repeats were pooled together to represent the final data which is already explained in the “Materials and methods” section on Page 25.
Also, some odd-behaving worms were not recorded such as worms that are stationary for prolonged amounts of time, worms that did not show any pharyngeal pumping, and worms that are potentially injured with any visible damages, etc.
An insight on the above-mentioned strategies has been updated on the “Materials and methods” section on page 25.
6) It is not clear, and the authors do not present any model of the mechanism by which MGH1 excess modulates the tyraminergic signal to increase food intake. It is only shown that elt-3 may be important and that tdc-1, ser-2 and tyra-2 appear to be relevant. There are no rescue or functional experiments showing where these players would be playing their role.
For example, would tyramine release be affected at the level of RIM, RIC or UV cells? Perhaps it would be good to specifically silence these neurons (e.g. with the expression of histamine-triggered anion channels, see Pokala et al., PNAS, 2014 or De Rosa et al., Nature, 2019) and see if MGH-1 is able to induce pharyngeal pumping.
It is also unclear where elt-3 is relevant. It would be important to express this TF specifically in tdc-1 expressing cells and analyze whether the response is similar to wild-type.
This is one of my main concerns regarding the manuscript. The lack of mechanistic details to understand how AGEs accumulation increases food intake argues against the significance of the work.
To further demonstrate the role of tyramine signaling in mediating MG-H1-induced increased pharyngeal pumping, as well as in glod-4 mutants, we supplemented tyramine to either elt-3;glod-4 or tdc-1;glod-4 double mutant and analyzed the pumping behavior. Our results in Figure 5B-E demonstrate that exogenous tyramine treatment of double mutants increased the pumping levels to glod-4 single mutant worms. Thus, we show that lack of elt-3 modulates tyramine signaling to suppress the pumping in elt-3;glod-4 double mutants. This demonstrates the importance of elt-3 in mediating increased pumping in glod-4 mutants via tyramine. Similarly, exogenous tyramine supplementation in tdc-1;glod-4 double mutant also increased pharyngeal pumping to glod-4 single mutant levels. However, the pumping is not increased in double mutants such as ser-2;glod-4, and tyra-2;glod-4 which lacks tyramine receptors.
Silencing the tyramine synthesis neurons to dissect the neuronal circuitry is an interesting experiment, as suggested by the reviewers; however, it is very time-consuming and beyond the scope of this study. We have demonstrated that lack of either tyramine or tyramine receptor suppressed the pumping in the glod-4 mutant background which was rescued by exogenous tyramine in elt-3;glod-4 and tdc-1;glod-4 double mutant.
7) Exogenous TA has been shown to reduce pharyngeal pumping (Greer et al., 2008). Furthermore, the absence of SER-2 suppresses the tyraminergic inhibition of exacerbated pharyngeal pumping in serotonin-treated worms (Rex et a, 2004. https://doi.org/10.1111/j.1471-4159.2004.02787.x).
Since the authors' results point more towards a positive modulation of pharyngeal pumping by tyramine and ser-2 , it would be important to discuss this paradox further.
Treatment of N2 wildtype worms with tyramine resulted in a significant decrease in the pharyngeal pumping (Figure 5). Tyramine suppressing pharyngeal pumping has already been described in the literature (1), as well as indicated by the reviewer. However, our data from elt-3;glod-4 and tdc-1;glod-4 double mutants show that tyramine treatment increased the pumping to glod-4 single mutant levels (Figure B+C). Also, double mutants lacking tyramine receptors were not rescued by tyramine (Figure D+E). These findings demonstrate that tyramine increases pharyngeal pumping only in glod-4 mutant background but decreases pumping in N2 wildtype background.
The mechanism behind this behavioral switch upon activation of the tyramine signaling pathway (i.e., from a suppressor of pharyngeal pumping in wildtype into a stimulator in glod-4 mutant background) remains elusive. Although tdc-1 is expressed only in RIC, RIM, UV2 pair of cells and gonadal sheath cells, tyramine receptors are located at distinct tissues (1) (2); thus, it can be speculated that AGEs, especially MG-H1, can modify the tissue-specific expression of tyramine receptors resulting in observed behavioral switch in response to tyramine signaling. This hypothesis can only be addressed by methodologies such as single-cell RNA sequencing which is beyond the scope of the current study. Currenty, we discussed this paradox in the ‘Discussion’ section on pages 21 and 22.
8) Many of the strains listed have not been backcrossed to clean up the background. For example: RB1690 ser-2(ok2103); VC343 glod-4(gk189); VC125 tyra-3(ok325); RB745 ser-4(ok512), etc. For tyra-2 it is not indicated which strain is used.
If backcrosses have not been performed, it is essential to perform the experiments with the clean strains. If they have done them, the strain they used for the experiments is not called as they named it. Please clarify this.
At the same time, it would be advisable that in the absence of rescue experiments, they use other alleles to confirm the observed phenotypes.
All the strains are outcrossed at least 3 times or more with N2 wildtype strain obtained from CGC, we have updated the “Materials and methods” now.
The strain used for tyra-2 is now mentioned in the “Materials and methods” as FX1846 tyra-2 (tm1846).
As recommended by the reviewer, we have used another allelic mutant of tdc-1 (n3420) and evaluated its pumping in n3420 as well as in tdc-1(n3420);glod-4 double mutant (Author response image 2). Our results demonstrate that glod-4 single mutant shows significantly increased pumping compared to N2 wildtype worms; however, tdc-1 (n3420);glod-4 (gk189) double mutant worms do not show any significant difference indicating tdc-1 enzyme is essential for increased feeding in glod-4 single mutants.
Similarly, we also evaluated the role of elt-3 in mediating the pharyngeal pumping in glod-4 single mutants by knockdown elt-3 using RNAi (Author response image 2). Again, our results demonstrate that elt-3 is essential for increased feeding in glod-4 single mutants.
Quantification of pharyngeal pumping in N2 wildtype, <italic>glod-4 (gk189)</italic>, <italic>tdc-1 (n3420)</italic> as well as in double mutants (A) and with <italic>elt-3</italic> knockdown (B).
One way ANOVA with Fisher’s LSD test. **** p<0.0001. Error bar ± SD.
Reviewer #2 (Recommendations for the authors):
The study is generally well done, but the strains don't appear to be outcrossed, and none of the effects are rescued. At least for the main results, tdc-1 and elt-3, there are multiple mutants available at CGC. The authors should rescue the original mutant or show the same results using a different allele. It would also be helpful for the reader to show the alleles in the figure legend.
We thank the reviewer for pointing out the importance of using either multiple allelic mutants or rescue of the mutant phenotype to validate, as well as strengthen, the role of tdc-1 and elt-3 genes in modulating the feeding behavior.
The strains used in this study are outcrossed at least 3 times, which is now indicated in the ‘Materials and methods’ section on Page 24.
As requested by the reviewer, we now included pharyngeal pumping data from another allelic mutant of MT10661 tdc-1(n3420) to demonstrate that lack of tdc-1 suppresses pharyngeal pumping in glod-4;tdc-1 double mutant. Also, we knocked down the elt-3 in glod-4 mutants worms using RNAi feeding approach and demonstrated that elt-3 is essential of increased feeding in glod-4 mutants (Author response image 2) .
Finally, we rescued increased pharyngeal pumping (glyoxalase glod-4 mutant phenotype) in either tdc-1;glod-4 or elt-3;glod-4 double mutants with exogenous administration of tyramine. This result shows that tyramine signaling is essential for increased feeding in glyoxalase glod-4 mutants which is suppressed with the lack of either tdc-1 or elt-3 genes. However, exogenous tyramine did not increase pharyngeal pumping in tyramine receptor-glod-4 double mutants.
Reviewer #3 (Recommendations for the authors):
There are several points that need to be substantially improved.
1. The writing of the manuscript needs to be improved substantially. The introduction is very confusing with too many abbreviations. I would suggest a diagram to show the reaction of α-dicarbonyl (α-DG), AGEs, glyoxalases. In each of these terms, you can list the examples in that category, for example: α-DG (MGO, GO, 3DG etc.), AGEs (MG-H1, CEL, CML, GOLD, MOLD), glyoxalases (glod-4, glod-1, etc.). The introduction is not organized with a focus in each paragraph. Transitions between sentences need to be improved substantially too.
We thank the reviewer for pointing out an unclear Introduction. As per reviewer’s request, we now introduced a pictorial diagram in the Figure 1 to explain the interactions between dicarbonyls, AGEs and glyoxalases enzymes as well as explained it in the pages 3 and 4. Also, a detailed understanding about dicarbonyls, AGEs, glyoxalases, and their impact on the health as well as disease can be obtained from a review from our lab, Chaudhuri et la. 2018 (4).
2. It is not explained or addressed in the manuscript why glod-4 KO in which we would expect less MG-H1, and MG-H1 supplementation both increase pumping. It will be helpful to measure MG-H1 level in glod-4 KO. This will also help to understand the result that the pumping rate increases are MG-H1 dose-dependent, however, glod-4 KO and MG-H1 supplementation is not additive in the increase of pumping.
It has been previously shown that glod-4 KO worms show more MGO (5). We now further clarify that glyoxalase enzymes detoxify dicarbonyls (such as MGO, GO, etc.) and the lack of glyoxalase enzymes results in accumulation of dicarbonyls which interacts with biomolecules to increase the formation and accumulation of AGEs in pages 3 and 4 (Figure 1). Thus, glod-4 glyoxalase mutant worms accumulate more MG-H1 which signals to increase the feeding which is comparable to increased feeding after exogenous treatment of MG-H1. Now, we also quantified the levels of MG-H1 in glod-4 mutant worms using mass spectrometry (LC-MRM) and included the data in Figure 2—figure supplement 1H+1I, which shows significantly increased MG-H1 levels.
3. Figure 2B shows all 66 genes are upregulated in glod-4 compared to WT. A concern is whether the gene expression is properly normalized. Are there downregulated genes other than the 66 genes? How many genes are up- and down- regulated in glod-4 vs. WT. All these necessary descriptions are not mentioned in the results part. The labeling of both A and B panels are too small for recognizing the gene and pathway names. Among the 66 "neurotransmitters and feeding genes", how many of them are neurotransmitter genes and how were they defined? Are tph-1, tdc-1, tbh-1, cat-2 genes in this list? Similar question, how were the "feeding genes" defined?
The RNAseq is normalized by FPKM (fragments per kilobase of exon per million fragments mapped) method.
The total number of genes identified in the RNAseq analysis is 20,277. The total genes that changed significantly are 5035 with upregulated genes amounting to 2237 and downregulated amounts to 2798 genes. This information is now included on page 10 of the manuscript.
Gene ontology was performed in DAVID v6.8 which identified neurotransmitter and feeding genes which is mentioned in the “Materials and methods” section. Because neurotransmitters also regulate feeding behavior, it is essential to understand that a clear separation between neurotransmitter genes and feeding genes is difficult.
Gene tph-1 was identified in RNAseq but there was no significant difference between N2 and glod-4 RNAi. However, RNAseq did not identify the following genes tdc-1, tbh-1, and cat-2.
4. Figure 2B used glod-4 RNAi while other results pumping data were collected on glod-4KO. Why glod-4 KO was not used in transcriptomics data. And please label on panel B indicating glod-4 RNAi. And it is described in the method part that RNA-seq was done on "N2 and mutant animals" but not RNAi.
As requested by the reviewer, we labeled the RNAseq heatmap appropriately (Figure 3A). We prepared samples for RNAseq from both glod-4 KO worms and by glod-4 RNAi knockdown. During further processing, samples from glod-4 RNAi worms demonstrated the best quality for RNAseq analysis. However, samples from glod-4 KO required further standardization to pass the quality control. Thus, we utilized samples from glod-4 RNAi instead of samples from glod-4 null mutants for RNAseq analysis.
5. tdc-1 and the two tyramine receptors knockouts abrogate pumping rate by MG-H1 and glod-4. This would indicate that tyramine release is necessary for the pumping rate increase by MG-H1 and glod-4. Did the author try rescue the abrogation by tyramine supplementation? This is a valid control to strengthen the hypothesis.
As recommended by the reviewer, we exogenously supplemented tyramine in different genetic backgrounds and demonstrated its role in modulating pharyngeal pumping. Tyramine supplementation in N2(wt) worms significantly reduced the pumping at various concentrations (Figure 5A). However, tyramine supplementation in double mutants such as elt-3;glod-4 and tdc-1;glod-4 resulted in a significant increase in the pumping comparable to glod-4 single mutants. Further, tyramine supplementation in tyramine receptor mutants leads to no changes in the pumping. This result demonstrates that in wildtype background tyramine suppresses pumping consistent with the existing literature (1); however, in the glod-4 mutant background tyramine increases pumping. Thus far, the mechanisms by which tyramine increased the pumping (signaling switch from a suppressor in wildtype worms into pharyngeal pumping activator in glod-4 mutant background) in glod-4 mutants remains elusive and needs further studies in the future.
6. The authors previously showed that glod-4 KO shortens lifespan, and supplementation of glod-4 substrates MGO also shortens lifespan. The results in Figure 4 of lifespan rescue by tdc-1 is not sufficient to support that increased feeding shortens glod-4 KO lifespan suggested in the manuscript. The valid conclusion from this data is that the lifespan decreases by glod-4 KO requires tyramine synthesis and probably tyramine signaling since the two receptors are also required for the shorter lifespan. Does MG-H1 addition rescue or further shorten glod-4 KO lifespan since MG-H1 also increases pumping rate not additively to glod-4 KO? Figure 4D labeling is too small.
We thank the reviewer for suggesting appropriate inference of the data. The conclusion was updated in the manuscript on pages 16 and 17, as suggested by the reviewer. Further, we also performed lifespan assay in N2 and glod-4 mutant worms with 150 µM MG-H1 treatment (Figure 6—figure supplement 1). MG-H1 treatment significantly reduced the lifespan in N2 wildtype worms. Glyoxalase mutant worms showed significantly shorter lifespans than N2 worms; however, MG-H1 treatment did not further affect the lifespan of glod-4 mutant worms.
We enlarged figure 6D (old figure 4D).
References:
1. Dallière N, Holden-Dye L, Dillon J, O'Connor V, Walker RJ. Caenorhabditis elegans Feeding Behaviors. Oxford Research Encyclopedia of Neuroscience2017.
2. Rex E, Hapiak V, Hobson R, Smith K, Xiao H, Komuniecki R. TYRA-2 (F01E11.5): a Caenorhabditis elegans tyramine receptor expressed in the MC and NSM pharyngeal neurons. J Neurochem. 2005;94(1):181-91.
3. Ji Ying Sze MV, Curtis Loer, Yang Shi & Gary Ruvkun. Food and metabolic signalling defects in a Caenorhabditis elegans serotonin-synthesis mutant. Nature. 2000;403:560-4.
4. Chaudhuri J, Bains Y, Guha S, Kahn A, Hall D, Bose N, et al. The Role of Advanced Glycation End Products in Aging and Metabolic Diseases: Bridging Association and Causality. Cell Metab. 2018;28(3):337-52.
5. Chaudhuri J, Bose N, Gong J, Hall D, Rifkind A, Bhaumik D, et al. A Caenorhabditis elegans Model Elucidates a Conserved Role for TRPA1-Nrf Signaling in Reactive α-Dicarbonyl Detoxification. Curr Biol. 2016;26(22):3014-25.