Metabolic pathways producing adenosine

Adenosine can be produced from AMP by 5’-nucleotidase. AMP increases during metabolic stress, when cells use adenylate kinase to make one ATP and one AMP from two ADP molecules. Increased AMP activates AMP-activated protein kinase (AMPK), which suppresses energy-consuming processes in the cell and activates energy-producing processes, or is converted to adenosine, which can be released from cells via the equilibrative nucleoside transporter (ENT). In the S-adenosylmethionine (SAM) transmethylation pathway, the combination of ATP and methionine produces SAM, the major methyl group donor (red) for the majority of methylations that occur in the cell. During methylation, SAM is converted to S-adenosylhomocysteine (SAH), which is rapidly converted to homocysteine and adenosine by adenosylhomocysteinase (Ahcy), as it would otherwise block further methylation. Ahcy works bidirectionally, and for the “SAM to homocysteine + adenosine” direction to dominate, homocysteine and adenosine must be rapidly metabolized or cleared from the cell. Homocysteine is remethylated back to methionine or further metabolized via the transsulfuration pathway (source of cysteine, H2S, and antioxidants). The catabolism of adenosine to inosine is mediated by adenosine deaminase (ADA) or recycled to AMP by adenosine kinase which becomes ATP through adenylate kinase, glycolysis or oxidative phosphorylation. Another possible source of adenosine is RNA degradation, which is not shown in this scheme.

Analysis of the SAM transmethylation pathway in larval hemocytes by ex vivo stable 13C isotope tracing

(A) Schematic representation of the SAM transmethylation pathway with polyamine synthesis and transsulfuration branches and labeling with L-methionine-13C5 from media (red 13 represents the labeled carbon). Light blue CH3 represents the methyl group used during transmethylation or remethylation. Enzymes and processes are italicized. RNAi in blue rectangle represents adenosylhomocysteinase knockdown. (B) Expression heat map (bulk RNAseq) of methionine transporters and enzymes in circulating hemocytes from uninfected (Uninf) and infected (INF) third instar larvae collected at 0, 9 and 18 hpi (0 hpi = 72 hours after egg laying). Means shown in each cell are transcripts per million (TPM) - data in S1_Data. (C-E,K,M) 13C-labeling of metabolites in hemocytes, which were incubated ex vivo for 20 min in media containing 0,33 µM L-methionine-13C5. The graphs show the fraction of the compound with one (m+1), four (m+4) or five (m+5) 13C-labeled carbons – methionine m+5 (C), S-adenosylmethionine (SAM) m+5 (D), S-adenosylhomocysteine (SAH) m+4 (E), 5-methylthioadenosine (MTA) m+1 (K) and cystathionine (CTH) m+4 (M). (F) Methionine m+4 portion of total labeled methionine (m+4 and m+5) in media representing the labeling impurity of used methionine (white bar setting threshold - dashed line), and in hemocyte samples. (G-I,L,N) Total levels of methionine (G), SAM (H), SAH (I), MTA (L) and CTH (N) in hemocytes shown as the mean metabolite amounts expressed by the normalized peak area. (J) Methylation index calculated as the ratio of SAM:SAH levels (peak areas in H and I). (C-N) Bars represent mean values with 95% CI of uninfected (Uninf, light grey) and infected (INF, dark grey) control and uninfected (Uninf, light blue) and infected (INF, dark blue) Ahcy-RNAi samples; each dot represents one biological replicate (numerical values in S1_Data and S2_Data); asterisks represent significant differences between samples tested by ordinary one-way ANOVA Tukey’s multiple comparison test (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001).

Expression of SAM transmethylation pathway enzymes and Ahcy RNAi

(A) Expression analysis of enzymes in the SAM transmethylation pathway in circulating hemocytes from uninfected (Uninf; light gray) and infected (INF; dark gray) third instar larvae collected at 18 hpi by RT-qPCR. Bars show fold change compared to uninfected Ahcy samples (expression levels were normalized to RpL32 expression in each sample), each dot represents a biological replicate. Asterisks indicate significant differences between samples tested by multiple unpaired t-test. (B) Analysis of Ahcy knockdown efficiency at 0 and 18 hpi by RT-qPCR. RNAi was specifically induced in hemocytes by SrpD-Gal4 tubP-GAL80ts driven expression of UAS-Ahcy-RNAiHMS05799in larvae maintained at 18°C for the first 3 days of development and then transferred to 25°C. RNAi reduces Ahcy expression to 16% of control at the onset of infection (0 h). Bars represent means with 95% CI of uninfected (Uninf, light gray) and infected (INF, dark gray) control (driver crossed to P{CaryP}Msp300attP40 control without RNAi) and uninfected (Uninf, light blue) and infected (INF, dark blue) Ahcy-RNAi samples; each dot represents a biological replicate (numerical values in S1_Data); asterisks indicate significant differences between samples tested by ordinary one-way ANOVA Tukey’s multiple comparison test (****p<0.0001).

Srp-Gal4 Gal80 driver expression

SrpD-Gal4 tubP-GAL80ts driven expression of UAS-GFP in 3rd-instar larvae maintained at 18°C for the first 3 days of development and then transferred to 25°C. GFP expression is detected in all lobes of lymph gland, pericardial cells and circulating and sessile hemocytes.

Generation of adenosine in the SAM transmethylation pathway and its systemic effects

(A) Levels of intracellular adenosine in hemocytes shown as the mean metabolite amount expressed by the normalized peak area at 20 hpi. Infection significantly decreases the level in control hemocytes (uninfected - Uninf - light gray vs. infected - INF - dark gray). Hemocyte-specific adenosylhomocysteinase knockdown (Ahcy RNAi; uninfected - Uninf - light blue and infected - INF - dark blue) significantly decreases intracellular adenosine in the uninfected state. (B) Levels of extracellular adenosine (right), released from hemocytes ex vivo after 20 min, and extracellular inosine (left) generated by adenosine deaminase ADGF-A. While infection leads to increased release of adenosine and generated inosine in the control (gray), no such increase is detected with Ahcy RNAi (blue). (C) Pupation is delayed upon infection in control larvae (n = 270, uninfected dashed gray line and n = 240, infected solid black line) but significantly less in hemocyte-specific Ahcy RNAi larvae (n = 260, uninfected dashed blue line and n = 265, infected solid blue line). Lines represent percentages of pupae at hours after egg laying (h AEL); rates were compared using Log-rank survival analysis. (D) The number of lamellocytes is significantly lower in infected Ahcy-RNAi (blue) larvae compared to infected control (gray). (E) Percentage of infected larvae surviving to adulthood is significantly lower in Ahcy-RNAi (blue) compared to control (gray) while the survival of uninfected individuals is not affected. (A,B,E) Bars represent means with SEM of uninfected (Uninf, light gray) and infected (INF, dark gray) control and uninfected (Uninf, light blue) and infected (INF, dark blue) Ahcy-RNAi samples; each dot represents a biological replicate; (D) dots represent the number of lamellocytes in a larva and the line mean; numerical values are in S1_Data and S2_data; asterisks represent significant differences between samples tested and unpaired t test (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001).

Adenosine kinase Adk3 and adenylate kinase Ak1 knockdown efficiency in hemocytes

Analysis of Adk3 (A) and Ak1 (B) knockdown efficiency at 0 and 18 hpi by RT-qPCR. RNAi was specifically induced in hemocytes by SrpD-Gal4 tubP-GAL80ts driven expression of Adk3-RNAiHMC06354 or UAS-Ak1-RNAiGL00177 in larvae maintained at 18°C for the first 3 days of development and then transferred to 25°C. RNAi reduces Adk3 expression to 13% and Ak1 expression to 20% of control during infection (18 hpi). Bars represent means with 95% CI of uninfected (Uninf, light gray) and infected (INF, dark gray) control (driver crossed to P{CaryP}Msp300attP40 or P{CaryP}attP2 control without RNAi) and uninfected (Uninf, light blue) and infected (INF, dark blue) Ahcy-RNAi samples; each dot represents a biological replicate (numerical values in S1_Data); asterisks indicate significant differences between samples tested by ordinary one-way ANOVA Tukey’s multiple comparison test (****p<0.0001). Development and survival to adulthood is not affected either by SrpD-Gal4 tubP-GAL80ts driven expression of Adk3-RNAiHMC06354 (C) or UAS-Ak1-RNAiGL00177(D).

Analysis of adenosine recycling to SAM in larval hemocytes by ex vivo stable 13C isotope tracing

(A) Schematic representation of the SAM transmethylation pathway, de novo purine synthesis and adenosine recycling to ATP and SAM and labeling with adenosine-13C5 from media (red 13 represents the labeled carbon). Enzymes and processes are italicized. (B) Expression heat map (bulk RNAseq) of adenosine metabolizing enzymes and transporters in circulating hemocytes from uninfected (Uninf) and infected (INF) third instar larvae collected at 0, 9 and 18 hpi (0 hpi = 72 hours after egg laying). Means shown in each cell are transcripts per million (TPM) - data in S1_Data. (C, D) Expression analysis of adenosine kinases Adk2, 3 (C) and adenylate kinases Ak1, 2 (D) in circulating hemocytes from uninfected (Uninf) and infected (INF) third instar larvae collected at 0, 9 and 18 hpi by RT-qPCR. Bars show fold change compared to 0 h Adk3 samples (expression levels were normalized to RpL32 expression in each sample), each dot represents a biological replicate. (E-H, L) Total levels of IMP (E), AMP (F), ADP (G), ATP (H) and S-adenosylmethionine - SAM (L) in hemocytes shown as the mean metabolite amounts expressed by the normalized peak area. (I-K) 13C-labeling of metabolites in hemocytes, which were incubated ex vivo for 20 min in media containing 10 µM adenosine-13C5. The graphs show the fraction of the compound with five 13C-labeled carbons – AMP m+5 (I), ATP m+5 (J) and SAM m+5 (K). (C-L) Bars represent mean values with 95% CI of uninfected (Uninf, light gray or pink) and infected (INF, dark gray or red) samples; each dot represents a biological replicate (numerical values in S1_Data and S2_data); asterisks represent significant differences between samples tested by tested by ordinary one-way ANOVA Sidak’s multiple comparison test (C,D) and unpaired t test (E-L) (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001).

Silencing of adenosine kinase and adenylate kinase in hemocytes affects the SAM transmethylation pathway and immune response

(A) Total levels of AMP in hemocytes shown as the mean metabolite amounts expressed by the normalized peak area. Infection significantly increases the AMP level in control hemocytes (uninfected - Uninf - light gray vs. infected - INF - dark gray). Hemocyte-specific adenosine kinase Adk3 knockdown (Adk3 RNAi; uninfected - Uninf - light blue and infected - INF - dark blue) significantly decreases AMP during infection compared to control. (B) Methylation index, calculated as the ratio of SAM:SAH levels (peak areas; numerical values in S1_Data and S2_data), in control (uninfected - Uninf - light gray and infected - INF - dark gray) and in hemocyte-specific Adk3-RNAi and adenylate kinase Ak1-RNAi (uninfected - Uninf - light blue and infected - INF - dark blue). (C) Pupation is delayed (10 h) upon infection in control larvae (n = 270, uninfected dashed gray line and n = 240, infected solid black line) but significantly less (7 h) in hemocyte-specific Adk3 RNAi larvae (n = 275, uninfected dashed blue line and n = 275, infected solid blue line). (D) The number of lamellocytes is significantly lower in infected Adk3-RNAi (blue) larvae compared to infected control (gray). (E) Percentage of infected larvae surviving to adulthood is significantly lower in Adk3-RNAi (blue) compared to control (gray). (F) Pupation is delayed (9 h) upon infection in control larvae (n = 225, uninfected dashed gray line and n = 270, infected solid black line) but significantly less (6 h) in hemocyte-specific Ak1 RNAi larvae (n = 195, uninfected dashed blue line and n = 225, infected solid blue line). (G) The number of lamellocytes is significantly lower in infected Ak1-RNAi (blue) larvae compared to infected control (gray). (H) Percentage of infected larvae surviving to adulthood is significantly lower in Ak1-RNAi (blue) compared to control (gray). (A,B,D,E,G,H) Bars/lines represent mean values with 95% CI of uninfected (Uninf, light grey) and infected (INF, dark grey) control and uninfected (Uninf, light blue) and infected (INF, dark blue) Adk3-RNAi or Ak1-RNAi samples; each dot represents one biological replicate (numerical values in S1_Data and S2_data); asterisks represent significant differences between samples tested by unpaired t-test or ordinary one-way ANOVA Tukey’s multiple comparison test (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001). (C,F) Lines represent percentages of pupae at hours after egg laying (h AEL); rates were compared using Log-rank survival analysis.

Hypothetical role of adenosine as a sensor of the balance between cell activity and nutrient supply

The hypothetical scheme shows an immune cell with either sufficient (left) or insufficient (right) nutrient supply for the given cellular activity. For each individual transmethylation event using S-adenosylmethionine (SAM) as a methyl group donor and producing S-adenosylhomocysteine (SAH), S-adenosylhomocysteinase (Ahcy) generates one molecule of adenosine (and homocysteine). Adenosine thus reflects the sum of all SAM transmethylations in the cell and may be considered as a proxy for cell activity (thick gray arrow on the right). If there are enough nutrients (cell on the left) to maintain high levels of ATP, adenosine can be recycled first to AMP by adenosine kinase, which requires the first ATP, then to ADP by adenylate kinase, which requires the second ATP, and finally to ATP by glycolysis (or oxidative phosphorylation) using glucose (the gray arrow on the left as thick as the one on the right expresses the balance). Recycled ATP can then enter the next round of transmethylation with methionine to form SAM. If there are not enough nutrients (cell on the right, thin gray arrow on the left) to regenerate ATP from ADP, ATP is produced from two ADP molecules by adenylate kinase, simultaneously generating AMP. Accumulated AMP prevents adenosine recycling by adenosine kinase (thin gray arrow on the left), and on the contrary, more adenosine can be produced from AMP by 5’nucleotidase. In this case, adenosine is pushed out of the cell (thick gray arrow down) via the equilibrative nucleoside transporter (ENT) and becomes an extracellular signaling molecule. Extracellular adenosine can, for example, suppress the development of Drosophila larvae or increase blood flow in mammals to provide more nutrients to immune cells.