Angiogenin mediates paternal inflammation-induced metabolic disorders in offspring through sperm tsRNAs

Abstract

Paternal environmental inputs can influence various phenotypes in offspring, presenting tremendous implications for basic biology and public health and policy. However, which signals function as a nexus to transmit paternal environmental inputs to offspring remains unclear. Here we show that offspring of fathers with inflammation exhibit metabolic disorders including glucose intolerance and obesity. Deletion of a mouse tRNA RNase, Angiogenin (Ang), abolished paternal inflammation-induced metabolic disorders in offspring. Additionally, Ang deletion prevented the inflammation-induced alteration of 5'-tRNA-derived small RNAs (5'-tsRNAs) expression profile in sperm, which might be essential in composing a sperm RNA 'coding signature' that is needed for paternal epigenetic memory. Microinjection of sperm 30-40 nt RNA fractions (predominantly 5'-tsRNAs) from inflammatory Ang^+/+ males but not Ang^-/- males resulted in metabolic disorders in the resultant offspring. Moreover, zygotic injection with synthetic 5'-tsRNAs which increased in inflammatory mouse sperm and decreased by Ang deletion partially resembled paternal inflammation-induced metabolic disorders in offspring. Together, our findings demonstrate that Ang-mediated biogenesis of 5'-tsRNAs in sperm contributes to paternal inflammation-induced metabolic disorders in offspring.

Fig. 1. Body weight and metabolic parameters of F1 males with paternal inflammation.

Male mice were treated with LPS (Inf) or saline (Con) once every other day for a total of four injections in 7 days to establish an inflammatory model. These mice were then allowed to mate with normal females for two days. a Growth curves for F1 male offspring from Con- and Inf- fathers. Statistical analysis was performed by a MIXED procedure for repeated measures. ^*P < 0.05; ^**P < 0.01. b Body weight of male offspring at 15 weeks of age. c–e Blood glucose by GTT at 13 weeks (c), ITT at 14 weeks (d), and PTT at 15 weeks (e) of age. In (c, d) and (e), n = 7 mice per group. AUC, area under the curve. ^*P < 0.05; ^***P < 0.001; ^****P < 0.0001. f Ratio of fat-to-muscle weight of male offspring at 15 weeks of age. g, h Representative images showing adipocyte area in perigonadal fat of Con F1 (g) and Inf F1 (h) male mice. n = 5 mice per group. Scale bar = 100 μm. i Treadmill test results of male offspring at 12 weeks of age. In (b, f) and (i), n = 14 in Con and n = 15 in Inf. Statistical analysis was performed by two-tailed unpaired Student’s t test (b, f, i), by two-tailed, two-way analysis of variance (ANOVA), uncorrected Fisher’s least significant difference (LSD) (c–e). Data are show as mean ± SEM. j, k Representative images showing cryosections of gastrocnemius muscle (cross sections) from Con F1 (j) and Inf F1 (k) mice stained with antibodies against GLUT4 (red) and DAPI (blue). n = 5 mice per group. Scale bar = 50 μm. l The top ten significant clusters of gene ontology (KEGG) terms enriched in gastrocnemius muscle samples of Con F1 and Inf F1 male offspring determined by GSEA and clustered under parent terms were related to synaptic signaling (n = 4, NES > | 1.5| and FDR < 0.05). Source data are provided as a Source Data file.

Fig. 2. Expression of Ang in inflammatory mice and metabolic parameters of F1 males from inflammatory and Ang-deleted mice.

a The relative expression level of Ang mRNA in mouse testis, caput epididymis, and cauda epididymis under LPS (Inf) or saline (Con) treatment for 24 h. b Western blots of Ang protein in mouse caput epididymis under LPS or saline treatment for 24 h or 7 d. In (a, b), n = 6 mice per group. c The Ang^+/+ and Ang^−/− F0 male mice were injected with LPS or saline once every other day for a total of four injections in 7 d in order to establish an inflammatory model. These mice were then allowed to mate with normal females for two days. Subsequently, offspring mice were exposed to chow diet and water ad libitum without any challenge throughout their lives. d Body weight of F1 males at 16 weeks of age. e Ratio of fat-to-muscle weight of F1 males at 16 weeks of age. f Blood glucose levels as assayed by GTT of F1 male mice at 14 weeks of age. ^** P < 0.01, ^*** P < 0.001 (Ang^+/+ Inf versus Ang^+/+ Con); ^&P < 0.05, ^&&P < 0.01, ^&&&& P < 0.0001 (Ang^+/+ Inf versus Ang^−/−Con); ^# P < 0.05, ^### P < 0.001 (Ang^+/+ Inf versus Ang^–/– Inf). g Area under the curve (AUC) statistics for (f). In (f, g), n = 7 mice per group. h Treadmill test results of F1 males at 8 weeks of age. In (d, e) and h, n = 12 in Ang^+/+ Con group, n = 13 in Ang^+/+ Inf group, n = 10 in Ang^–/– Con group, and n = 13 in Ang^–/– Inf group. Statistical analysis was performed by two-tailed unpaired Student’s t test (a), by two-tailed, one-way ANOVA (d, e, g, h) or two-way ANOVA (f), uncorrected Fisher’s LSD. Data are show as mean ± SEM. Source data are provided as a Source Data file.

Fig. 3. Sperm tsRNAs profile in inflammatory and Ang-deleted mice.

Male Ang^+/+ and Ang^–/– mice were treated with LPS (Inf) or saline (Con) once every other day for a total of four injections in 7 d to establish an inflammatory model. Small RNA deep sequencing was performed on cauda sperm isolated from Ang^+/+ Con, Ang^+/+ Inf, Ang^–/– Con, and Ang^–/– Inf mice. Sperm total RNAs extracted from two mice were mixed together for Small RNA deep sequencing. a–d Subcellular fractionation of tsRNAs (5′-tsRNAs, 3′-tsRNAs, CCA-tsRNAs, and other tsRNAs) in Ang^+/+ Con (a), Ang ^+/+ Inf (b), Ang^–/– Con (c), and Ang^–/– Inf (d) sperm. e–h Length distribution and pattern changes of different tsRNAs in Ang^+/+ Con (e), Ang^+/+ Inf (f), Ang^–/– Con (g), and Ang^–/– Inf (h) sperm. i Heat map of differentially expressed 30–35 nt 5′-tsRNAs in sperm. j Venn diagram showing numbers of 5′-tsRNAs with significant upregulated in sperm from inflammatory mice and downregulated in sperm from Ang-deleted mice. k–n Expression levels of select 5′-tRNAs that are altered with inflammation and Ang deletion. Statistical analysis was performed by two-tailed, one-way ANOVA, uncorrected Fisher’s LSD. n = 6 mice per group. All data are plotted as mean ± SEM. Values with different superscripts are significantly different from each other (P < 0.05). o Northern blot analyses of 5′tsRNA-Gly (shown by arrow heads) in Ang^+/+ Con, Ang ^+/+ Inf, Ang^–/– Con, and Ang^–/– Inf cauda sperm and caput epididymis. n = 6 mice per group. Sperm total RNAs extracted from two mice were mixed together for each lane in the experiment. Sperm total RNAs were run on a 15% denatured PAGE gel as shown as a loading control. Blots are shown as representatives of three independent experiments with similar results. Source data are provided as a Source Data file.

Fig. 4. Body weight and metabolic parameters of F1 males generated by zygotic injection of sperm 30–40 nt RNAs or synthetic tsRNAs.

a Illustration of zygotic injection of sperm 30–40 nt RNAs or a pool of synthetic tsRNAs to generate F1 male offspring for phenotypic examination. b Blood glucose during GTT of F1 males generated from sperm 30–40 nt RNAs injection at 15 weeks of age. ^**** P < 0.0001 (Ang^+/+ Inf versus Ang^+/+ Con); ^&&& P < 0.001, ^&&&& P < 0.0001 (Ang^+/+ Inf versus Ang^–/– Con); ^## P < 0.01, ^#### P < 0.0001 (Ang^+/+ Inf versus Ang^–/– Inf); ^$$$ P < 0.001, ^$$$$ P < 0.0001 (Ang^+/+ Inf versus control injection). c Area under the curve (AUC) statistics for (b). d Body weight of F1 males at 17 weeks of age. e Ratio of fat-to-muscle weight of F1 males at 17 weeks of age. In (b–e), n = 14 in Ang^+/+ Con group, n = 12 in Ang^+/+ Inf group, n = 14 in Ang^–/– Con group, n = 14 in Ang^–/– Inf group, and n = 14 in control group. f Blood glucose during GTT of F1 males generated from scrambled RNA, synthetic tsRNAs (synthetic tsRNAs), and sperm 30–40 nt RNAs from Ang^+/+ Inf mice (sperm RNAs) injection and normal male mice (control) at 12 weeks of age. ^* P < 0.05 (scrambled RNA versus synthetic tsRNAs); ^&&&& P < 0.0001 (scrambled RNA versus sperm RNAs); ^# P < 0.05, ^### P < 0.001 (synthetic tsRNAs versus sperm RNAs); ^$$ P < 0.01, ^$$$ P < 0.001 (sperm RNAs versus control). g AUC statistics for (f). h Body weight of F1 males at 15 weeks of age. i Ratio of fat-to-muscle weight of F1 males at 15 weeks of age. In (f–i), n = 17 in scrambled RNA group, n = 10 in synthetic tsRNAs group, n = 17 in sperm RNAs, and n = 7 in control. Statistical analysis was performed by two-tailed, two-way ANOVA (b, f) or one-way ANOVA (c-e, g-i), uncorrected Fisher’s LSD. Data are show as mean ± SEM. Source data are provided as a Source Data file.