Abstract

   Dietary restriction (DR) has been a very important discovery in modern
   aging biology research. Its remarkable anti-aging effect has been
   proved in a variety of organisms, including members of Lepidoptera, but
   mechanisms by which DR increases longevity are not fully understood. By
   using the silkworm (Bombyx mori), a model of lepidopteran insect, we
   established a DR model, isolated hemolymph from fifth instar larvae and
   employed LC–MS/MS metabolomics to analyze the effect of DR on the
   endogenous metabolites of silkworm, and tried to clarify the mechanism
   of DR to prolong lifespan. We identified the potential biomarkers by
   analyzing the metabolites of the DR and control groups. Then, we
   constructed relevant metabolic pathways and networks with
   MetaboAnalyst. DR significantly prolonged the lifespan of silkworm. The
   differential metabolites between the DR and control groups were mainly
   organic acids (including amino acid), and amines. These metabolites are
   involved in metabolic pathways such as amino acid metabolism. Further
   analysis showed that, the levels of 17 amino acids were significantly
   changed in the DR group, indicating that the prolonged lifespan was
   mainly due to changes in amino acid metabolism. Furthermore, we
   identified 41 and 28 unique differential metabolites in males and
   females, respectively, demonstrating sex differences in biological
   responses to DR. The DR group showed higher antioxidant capacity and
   lower lipid peroxidation and inflammatory precursors, with differences
   between the sexes. These results provide evidence for various DR
   anti-aging mechanisms at the metabolic level and novel reference for
   the future development of DR-simulating drugs or foods.

   Subject terms: Developmental biology, Physiology, Zoology, Biomarkers

Introduction

   With the increasing problem of global population aging, anti-aging
   research has become a hot research field. DR with adequate nutrition is
   the gold standard for delaying aging and extending health span and
   lifespan in diverse species, including rodents and non-human
   primates^[34]1–[35]4. At the same time, it can effectively prevent the
   deterioration of biological functions and prevent and/or delay the
   occurrence of aging-related diseases^[36]5–[37]7.

   DR increases lifespan in a broad variety of organisms and improves
   health in humans. Although increasing evidence suggests that DR works
   through the key nutrient and stress-responsive metabolic signaling
   pathways including IIS/FOXO, mTOR, AMPK, sirtuins, NRF2, and
   autophagy^[38]8–[39]12, the mechanism by which DR promotes longevity is
   not fully understood, and the interaction between dietary intake, the
   microbiome, and the immune system remains poorly described. In
   addition, the long-term transgenerational consequences of dietary
   interventions are poorly understood. DR commonly reduces reproduction
   and long-term DR can be difficult to sustain in
   humans^[40]4,[41]5,[42]13. Therefore, it is important to develop new
   experimental animals with a small size and short life cycle that are
   more suitable for lifespan research than mice, that are used more
   frequently in the usual lifespan studies. Invertebrates with simple
   structures and relatively complete organ differentiation are very good
   candidates. Currently, only a few invertebrates have been used
   extensively for lifespan research, namely Drosophila and nematodes
   (e.g. C.elegans)^[43]14–[44]16. Therefore, new experimental animals
   suitable for aging lifespan research, especially invertebrate models,
   need to be established.

   The silkworm (Bombyx mori) is an economically important organism. It
   has been domesticated for thousands of years and has unique advantages
   as an experimental invertebrate progeny because it produces a large
   number of progeny, and has a short life cycle, and it has a clear
   genetic background that has been used as a model in scientific research
   for over one century, and a simple living environment. Many studies on
   the silkworm have also proved that it can be used for aging
   research^[45]8,[46]17–[47]20.

   Metabolomics is a relatively new high-throughput sequencing technology
   that has been widely used in various research fields, especially to
   evaluate the impact of the environment on species^[48]21. Metabolomics
   provides information on metabolites in biological cells, tissues,
   organs, and organisms. These metabolites reflect the life activities of
   small molecular compounds downstream of the genome at the level of
   cellular metabolism. They can be measured with various metabolomics
   methods. The application of metabolomics to silkworm research has
   increased in recent years^[49]22–[50]25.

   In the previous research, we established a DR model by using the B.
   mori (Bombyx mori) Qiufeng strain. One group was submitted to ad
   libitum feeding (AL) while the other group was submitted to DR. We
   recorded life indexes such as daily food intake, body weight, age
   progression, and reproductive capacity to ensure successful
   establishment of the DR model. Our finding indicates that food
   restriction can significantly prolong lifespan^[51]20,[52]26. Moreover,
   the larvae have sex-limited marks: female larvae have common spots and
   male larvae have plain spots. Hence, we could separate the larvae sex
   to evaluate sex-specific changes. The lifespan of female and male in DR
   model was 144 h and 136 h longer than that in AL group respectively,
   and the life-prolonging effect was even more obvious after
   maturity(including pupal and moth stage)^[53]26. Based on these
   results, we made the same model in this study and collected hemolymph
   from fifth instar larvae and performed non-targeted metabolomics using
   liquid chromatography–tandem mass spectroscopy (LC–MS/MS). Through
   systematic analysis of differential metabolites, we found that
   silkworms in the DR group had better antioxidant capacity and lower
   lipid peroxidation and inflammatory precursors. The results provide
   experimental evidence and reference for elucidating the biochemical
   mechanisms by which DR prolongs lifespan, and also lay a theoretical
   basis to explore lifespan-related biomarkers and to analyze the
   long-term effects of DR on organisms.

Materials and methods

Establishment of the DR silkworm model

   The B. mori Qiufeng strain was bred and preserved by the Sericulture
   Research Institute of the Chinese Academy of Agricultural Sciences. It
   produces diapause eggs when parental eggs are incubated in 25 °C
   Diapause of eggs were terminated by dipping eggs in a 20% (V/V) HCl
   solution in 25 °C (for 60 min and thoroughly washed in clean water. The
   eggs were then incubated in indoor natural light at 25 °C. After
   hatching, silkworm larvae were reared with fresh mulberry leaves at
   25 ± 1 °C. At the second instar stage, the larvae were randomly divided
   into two groups. The AL group was given food twice a day and allowed
   uninterrupted feeding during the entire day. The DR group was given
   mulberry leaves once a day and allowed to feed for 16 h. There was no
   food available for the other 8 h. The DR group feeding time was 66.6%
   of that of the AL group, which meets the DR standard of other animal
   models (60%)^[54]27,[55]28. From the fourth instar, females and males
   of each groups were reared separately according to their body marks,
   with three replicates for each group and 30 larvae for each repetition.

Sample preparation for LC–MS/MS

   Fifth instar male and female larvae from the AL and DR were wiped with
   75% alcohol and hemolymph was collected from the caudal legs. Eighty
   larvae were used for each experimental group. And 300 μL hemolymph was
   collected from each 10 larvae as one sample, totally 8 hemolymph
   samples of each experimental group were used for LC–MS/MS. The
   hemolymph was collected in 1.5 mL centrifuge tubes that contained
   phenylthiourea, then immediately transferred to -80° for freezing.
   Samples were prepared as follows:
   Groups            AL♀      AL♂      DR♀      DR♂
   Quantity * volume 8*300 μL 8*300 μL 8*300 μL 8*300 μL
   [56]Open in a new tab

   All samples were thawed at 4 °C. Then, 200 μL of each sample was
   combined with 800 μL of methanol; the tube was shook for 60 s, and then
   mixed well. The samples were centrifuged at 12,000 rpm for 10 min at
   4 °C. The supernatant was transferred to a new 1.5 mL centrifuge tube,
   then concentrated to dryness in vacuo. Dried samples were reconstituted
   with 300 µL of 80% methanol and filtered with a 0.22 μm membrane. These
   samples were subjected to LC–MS.

   Twenty microliters of each sample was mixed to form a quality control
   (QC) sample, which was used to correct deviations of the analysis
   results and errors caused by the analytical instrument itself. The
   remaining samples were used for LC–MS on-machine detection.

LC and MS conditions

   For LC, a Thermo Ultimate 3000 instrument and ACQUITY UPLC^® HSS T3
   1.8 µm (2.1 × 150 mm) column were used. The autosampler temperature was
   4 °C, the flow rate was 0.25 mL/min, and the column temperature was
   40 °C. Two microliters of the sample was injected for gradient elution.
   The mobile phase for the positive ion mode was 0.1% formic acid water
   (A) and 0.1% formic acid acetonitrile (B). The mobile phase for
   negative ion mode was 5 mM formic acid amine water (C) and acetonitrile
   (D). The gradient elution program was 0–1 min, 2% B/D; 1–9 min, 2–50%
   B/D; 9–14 min, 50–98% B/D; 14–15 min, 98% % B/D; 15–15.5 min, 98–2%
   B/D; 15.5–17 min, 2% B/D.

   For MS, a Thermo Q Exactive Focus instrument with an electrospray ion
   source (ESI) was used in positive and negative ion ionization modes.
   The positive ion spray voltage was 3.80 kV, the negative ion spray
   voltage was 2.50 kV, the sheath gas 45 bar, and auxiliary gas was
   15 bar. The capillary temperature was 325 °C, the full scan was
   performed with a resolution of 70 000, the scan range was 81–1000, the
   secondary fragmentation was performed by high-energy collisions (HCD),
   the collision voltage was 30 eV, and the unnecessary MS/MS information
   was removed by dynamic exclusion.

Data analysis

   The acquired LC–MS/MS data raw files were imported into Proteowizard
   software (v3.0.8789). Then, the acquired raw data were converted into
   the mzXML format (xcms input file format). The XCMS package of R
   (v3.3.2) was used to identify, filter, and align peaks. The main
   parameters were bw = 5, ppm = 15, peakwidth = c (10, 20),
   mzwid = 0.015, mzdiff = 0.01, and method = centWave. The data matrix
   was obtained including the mass-to-charge ratio (m/z), retention time
   (rt), peak area (intensity), and other information. To enable data of
   different magnitudes to be compared, peak areas were batched normalized
   and then imported into SIMCA-P 14.1 (Umetrics, Sweden) for principal
   component analysis (PCA), partial least squares discriminant analysis
   (PLS-DA), and orthogonal partial least squares discriminant analysis
   (OPLS-DA). The differential variables that contributed the most were
   determined by variable importance in projection (VIP) > 1 in the S-plot
   and p < 0.05 on an independent samples t-test. Finally, the
   differential variables were identified through the Human Metabolome
   Database (HMDB) database ([57]https://hmdb.ca/) combined with secondary
   fragment ions. The differential metabolite data obtained from the
   identification was used to carry out pathway enrichment analysis with
   MetaboAnalyst5.0 ([58]https://www.metaboanalyst.ca/).

Results

Effects of DR on the life index of silkworms

   By controlling the feeding time, the feeding amount of DR group was
   65.07% of that in AL group (Fig. [59]1A).The entire lifespan (including
   larval, pupal and adult stage) lasted significantly longer (p < 0.05),
   specifically 143 and 137 h longer for females and males, respectively,
   compared with the AL group, and the extension of survival time was more
   significant in the pupal and adult stage (Fig. [60]1B). Although diet
   restriction caused the decrease of body weight in the later period of
   silkworm life (Fig. [61]1C), and the number of eggs laid by females
   declined (Fig. [62]1D), it had no obvious negative effect on cocoon
   layer rate (Fig. [63]1E). These results indicate that rational diet
   restriction can significantly prolong life expectancy with little
   impact on economic and life indexes.

Figure 1.

   [64]Figure 1
   [65]Open in a new tab

   Effects of DR on the Life Index of Silkworms. (A) The feeding amount in
   DR and AL group; (B) Comparison of the duration of each instar of
   Bombyx mori in DR and AL group; (C) Body weight of silkworm larvae at
   different stages,the chart shows the weight per 50 larvea; (D)
   Comparison of the number of eggs laid by females in DR and AL group;
   (E) Comparison of cocoon layer rate between DR and AL group.

Metabolites differ significantly between DR and AL groups

   PCA generated new characteristic variables by linear combination of
   metabolite variables according to a certain weight, and categorized
   each group of data through the main new variables (principal
   components) to remove poor repeatability (outlier samples) and abnormal
   samples. This approach intuitively reflects the distribution of each
   sample in the mathematical model space. From the PCA score plot, the
   samples were obviously aggregated into groups, indicating a similar
   composition and concentration of the variables/molecules in the same
   group of samples (Fig. [66]2).

Figure 2.

   Figure 2
   [67]Open in a new tab

   Scatter plot of the principal component analysis. (A) positive ion
   mode; (B) negative ion mode. Note AL, ad libitum feeding; DR, Dietary
   restriction.

   PLS-DA can measure the impact strength and explanatory power of
   expression patterns of metabolites to classify and discriminate
   samples, thereby assisting to screen metabolic markers. The PLS-DA
   model revealed that the DR and AL groups could be distinguished in the
   positive and negative ion modes (Supplementary Fig. [68]1).

   We used OPLS-DA regression to establish the relationship between
   metabolite expression and sample category, to predict the sample group.
   There was good discrimination between the DR and AL groups in the
   positive and negative ion modes (Supplementary Fig. [69]2).

Potential biomarkers for DR-induced lifespan extension in silkworms

   Biomarkers can provide important information regarding the state of
   metabolic pathways. Identifying biomarkers would allow one to analyze
   comprehensively the mechanism by which DR increases lifespan. The
   effect of DR on the hemolymph metabolites of silkworm was obvious.Among
   the detected metabolites, 1042 metabolites were upregulated and 1,692
   were downregulated compared with the AL group (Supplementary
   Fig. [70]3).We identified 80 differential metabolites between the DR
   and AL groups (based on p < 0.05 and VIP ≥ 1 for the first component of
   OPLS-DA). Most of these differential metabolites are organic acids, of
   which amino acids, peptides, and their derivatives represent the
   largest proportion (53.3%, Fig. [71]3, Supplementary Table [72]1). The
   metabolites with the most significant differences are amino acids
   (including peptides and their derivatives, such as citrulline,
   L-glutamine, 5-oxo-L-proline, beta-alanine, N-acetyl-L-phenylalanine,
   L-phenylalanine, and L-histidine),the levels of 17 amino acids were
   significantly changed in the DR group(Table [73]1), followed by other
   organic acids (2-oxoglutaric acid, 3-hydroxy-3-methylglutaric acid,
   5-aminosalicylic acid, and aminohippuric acid), amines (orotidine,
   ribothymidine, guanosine, 2-methylguanosine, uridine, and inosine), and
   biogenic amines (N-omega-acetylhistamine, histamine, putrescine,
   thymine, epsilon-caprolactam, and cyclohexylammonium) (Supplementary
   Table [74]1, Supplementary Fig. [75]4). Organic acids, amines, and
   biogenic amines are metabolized vigorously in silkworm hemolymph,
   indicating that the changes in these metabolites are inseparable from
   DR.

Figure 3.

   [76]Figure 3
   [77]Open in a new tab

   Classification of differential metabolites.

Table 1.

   J 17 amino acids that level changed significantly in the DR group.
   Amino acid Mean_AL Mean_DR Fold change Log2 (FC) p value FDR
   L-Phenylalanine 19,200,919.35 14,296,314.96 0.74456 − 0.42553 3.01E−06
   0.00030349
   Citrulline 98,769,293.16 206,759,036.3 2.0934 1.0658 3.32E−06
   0.000321484
   L-Glutamine 4,944,020,962 5,896,912,379 1.1927 0.25427 0.000189198
   0.004417884
   5-Oxo-L-Proline 3,318,487,561 3,963,115,740 1.1943 0.25611 0.000237059
   0.004974341
   L-Histidine 10,976,341,701 9,151,456,845 0.83374 − 0.26232 0.000343632
   0.006361452
   beta-Alanine 512,919,211.5 614,938,175.3 1.1989 0.26171 0.000501213
   0.007897442
   Glycine Betaine 5,676,864,188 9,211,875,930 1.6227 0.6984 0.002378735
   0.019590033
   3-Hydroxy-DL-kynurenine 210,599,831.8 118,910,653.1 0.56463 − 0.82463
   0.002863303 0.022173951
   ADMA 2,591,734,286 1,985,470,489 0.76608 − 0.38444 0.009657901
   0.048930094
   L-Asparagine 421,067,355.7 530,654,290.9 1.2603 0.33372 0.011296623
   0.054180258
   L-Kynurenine 69,673,758.07 46,561,638.19 0.66828 − 0.58147 0.02034124
   0.078170374
   L-Isoleucine 11,010,270,065 7,834,365,254 0.71155 − 0.49096 0.022427126
   0.083545406
   N-Acetyl-L-Phenylalanine 676,923.76 1,475,764.267 2.1801 1.1244
   0.000125896 0.003624221
   PS(18:1(9Z)/0:0) 5,570,516.541 3,811,703.205 0.68426 − 0.54738
   0.004071645 0.028608243
   N-Acetylproline 6,140,567.251 4,537,858.364 0.739 − 0.43636 0.005675749
   0.035375506
   L-Leucine 7,961,590.504 5,914,799.329 0.74292 − 0.42873 0.015074813
   0.064331681
   4-Hydroxy-L-Proline 7,724,893.857 5,715,649.62 0.7399 − 0.4346
   0.022334042 0.084458617
   [78]Open in a new tab

   FDR false discovery rate.

Metabolic pathway analysis of potential biomarkers

   We used the MetaboAnalyst database to analyze the differential
   metabolites between the DR and AL groups. We integrated the raw p
   values, the Holm p value, false discovery rate (FDR), and impact value,
   and we evaluated the importance of related metabolic pathways. We
   obtained comprehensive data of metabolic pathways by removing
   metabolites with a raw and Holm p > 0.1 and false positive rate > 10%
   (0.1). In the DR group, the differential metabolites were mainly
   concentrated in amino acid metabolism, carbohydrate metabolism, and
   energy metabolism (Fig. [79]4A), and the differential metabolites
   showed significant differences in multiple metabolic pathways
   (Fig. [80]4B). In addition, the tricarboxylic acid (TCA), pyrimidine
   and purine metabolism, biotin metabolism, and riboflavin metabolism
   showed significant changes. Notably, amino acid metabolism accounted
   for more than 50% of the total metabolic pathways, followed by
   carbohydrate metabolism. Combined with the results of cluster analysis,
   DR altered the ratio of amino acids and increased amino acid
   utilization, which may be the most important reason for the increased
   longevity of silkworms submitted to DR. We confirmed that DR could
   promote fatty acid oxidative metabolism by intervening in mitochondria,
   thereby prolonging lifespan. The TCA cycle was activated after DR, and
   this change might be the result of mitochondrial activity.

Figure 4.

   [81]Figure 4
   [82]Open in a new tab

   Enrichment of differential metabolites. (A) Metabolism pathway
   enrichment of differential metabolites; (B) Pathway impact of
   differential metabolites.

   We also analyzed the potential relationship between differential
   metabolites, genes (Fig. [83]5A, Attachment 1), and diseases
   (Fig. [84]5B, Attachment 1). We found that the differential metabolites
   are involved in a variety of geriatric diseases and multiple
   lifespan-related genes. Hence, DR may modulate lifespan genes and
   promote anti-aging processes.

Figure 5.

   [85]Figure 5
   [86]Open in a new tab

   Analysis of interaction between different metabolites and diseases,
   genes. (A) Gene–metabolite interaction networks. (B) Metabolite–disease
   interaction.

DR Effects on metabolism in each sex

   We found that metabolite differences in silkworm hemolymph were not
   only affected by DR, but were also closely related to silkworm sex. The
   lifespan extension induced by DR was more obvious in female silkworms
   than that in male indicating that sex is inseparable from the effects
   of DR. To analyze the metabolic differences between female and male
   silkworms, we performed a multivariate analysis based on metabolomics
   data, and we separated the two groups of differential metabolites based
   on sex. We found 41 unique differential metabolites in the male group
   and 28 unique differential metabolites in female silkworms (Fig. [87]6,
   Supplementary Tables [88]2 and [89]3), further indicating that sex has
   a great effect on metabolites in the hemolymph. Thirty-three
   differential metabolites appeared simultaneously in the two
   experimental groups, indicating that these metabolites are closely
   related to aging. Of these metabolites, 25 showed similar trends
   (Supplementary Fig. [90]5A, Supplementary Table [91]4) and the other 8
   showed the opposite trend (Supplementary Fig. [92]5B, Supplementary
   Table [93]4). The differential metabolites with the identical trend are
   predominantly amino acids (Supplementary Fig. [94]6A). The unique
   differential metabolites are predominantly amino acids and
   glycerophospholipids in females (Supplementary Fig. [95]6B) and
   predominantly pyrimidine, amino acids, pantothenic acid, and coenzyme A
   in males (Supplementary Fig. [96]6C). These findings make sense:
   females lay eggs, so they have higher amino acid and energy needs than
   males.

Figure 6.

   Figure 6
   [97]Open in a new tab

   Venn diagram of differential metabolites in the male and female groups.
   AL, ad libitum feeding; DR, dietary restriction.

Discussion

   As the global population continues to age, there has been increased
   attention of health problems caused by aging and aging-related
   diseases. Over the past decade, research on aging and its accompanying
   diseases has accelerated markedly. DR is a powerful way to prolong
   lifespan and is one of the most recognized and successful anti-aging
   methods. However, how it induces metabolic changes associated with
   aging has been poorly studied, and the focus has mainly been on
   vertebrates^[98]29–[99]32. Studies have shown that organisms respond to
   DR stimuli in complex and unpredictable ways. At this stage, the
   available research is still insufficient to determine how DR prolongs
   lifespan and allows organisms to resist aging-related problems. Several
   pathways known to be involved in anti-aging mechanisms (e.g., free
   radical generation, formation of advanced glucose oxidation end
   products, lipid peroxidation, and inflammatory responses) are closely
   related to metabolism. Therefore, metabolite analysis can be used to
   discover potential aging biomarkers and to test various hypotheses of
   the mechanisms by which DR prolongs lifespan by determining the
   metabolic changes induced by DR. By measuring and mathematically
   modeling changes in the levels of hemolymph metabolites in the
   invertebrate silkworm, we have provided new insights into the
   mechanisms by which DR prolongs life. At the same time, our
   non-targeted metabolic analysis covers the whole set of metabolic
   pathways, and our results provide detailed metabolic characteristics of
   biological systems responding to DR.

DR induced marked changes in amino acids

   Amino acids are essential nutrients for silkworms to maintain body
   functions and growth. They compose most enzymes and some hormones and
   other active substances that regulate the physiological activities of
   silkworms^[100]33. In the DR group, the total amount of amino acids in
   the silkworm diet decreased, and there were specific reductions in
   isoleucine and arginine but increases in alanine, asparagine, and other
   amino acids. Some specific amino acids have established effects on
   lifespan—for example, methionine restriction increases lifespan in
   Drosophila by downregulating TOR signaling in the state of low amino
   acid concentrations^[101]34. Consistently, tryptophan restriction
   increases lifespan in mice and Evans rats, and dietary methionine
   restriction in rats and mice reduces mitochondrial oxygen free
   radicals, blood glucose, insulin, and IGF-1 levels, increasing average
   and maximum lifespan^[102]35,[103]36. These studies indicate a key role
   for proteins in the regulation of lifespan in Drosophila and mammals.

   In this study, many amino acid metabolic pathways changed significantly
   after dietary restriction (Fig. [104]4). For example, the metabolic
   changes of the three branched-chain amino acids—valine, leucine, and
   isoleucine—are closely related to longevity and aging^[105]37
   suggesting that lifespan extension in DR may also be the result of
   changes in branched-chain amino acids. In addition, compared with the
   AL group, the content of citrulline increased by 2.09 times. As an
   important part of the liver, citrulline can help the immune system
   resist infection and increase energy, indicating that DR may prolong
   lifespan by improving immunity. The types and quantities of various
   amino acids in the silkworm follow a certain strict ratio. Too little
   or too much intake of amino acids will hinder protein synthesis, thus
   affecting their lifespan^[106]33. The essential amino acids required by
   the silkworm are the same as those required by other insects or higher
   animals^[107]38, among which tryptophan and methionine are the main
   limiting amino acids^[108]33. We found no differences in the contents
   of these two limiting amino acids in the hemolymph, suggesting that
   although the AL group ingested more amino acids, they did not utilize
   amino acids better than the DR group. Therefore, changes in the ratio
   of amino acids and utilization by DR may also be an important reason
   for prolonging life^[109]39. Metabolomics has advanced our
   understanding of the complex relationship between amino acid metabolism
   and lifespan to identify the role of specific amino acids in
   maintaining health and the impact on lifespan. This technique also
   facilitates the study of amino acid functions, metabolic regulation,
   and safety, and serves as a reference for the development and design of
   dietary regimens restricting specific amino acids.

DR altered carbohydrate metabolites

   The TCA cycle is a crucial metabolic pathway in silkworms. It contains
   and affects a variety of metabolites in the body, and has a wide and
   profound connection with aging^[110]40. TCA metabolites such as citric
   acid were increased in the DR group, indicating an increase in TCA
   cycle activity in this group. Studies have shown that a decrease in the
   levels of circulating TCA metabolites reduces the number of cells, and
   NAD + concentrations decrease during aging; these changes increase the
   body’s oxidative stress response. Accompanied by the functional decline
   of these cellular signals, age-related oxidative stress and
   inflammation increase gradually, but restoring NAD + levels in the body
   can extend lifespan to a certain extent^[111]41. In other words, DR may
   have facilitated a better energy supply and antioxidant capacity by
   activating the TCA cycle, indicating that silkworms submitted to DR
   have a better health state and thus prolonged lifespan. In addition,
   the TCA cycle intermediate citrate may play a role in amino acid and
   fatty acid metabolism^[112]42. Interestingly, elevated citrate, which
   inhibits phosphofructokinase, and glycolysis inhibitors such as
   2-deoxyglucose have been proposed as viable DR mimics^[113]43.

   Organisms respond when energy supply levels are low and try to delay
   aging and enhance resistance to aging pathologies. Determining the
   mechanisms behind this response could provide a way to treat a variety
   of diseases and prolong life in one fell swoop. Mitochondria are
   cellular energy factories. The enzymes and proteins distributed on the
   inner and outer membranes of mitochondria are important components of
   the oxidative respiratory chain and participate directly in the energy
   production process^[114]44. Metabolomics revealed that the silkworm TCA
   cycle was activated after DR, and this change was apparently the result
   of mitochondrial activity.

DR altered biogenic amines

   We detected changes in biogenic amines such as spermine, putrescine,
   spermidine, histamine, and tyramine. Except for histamine, the rest
   showed a downward trend. Appropriate amounts of biogenic amines can
   promote the normal physiological activities of organisms, while
   excessive intake can cause adverse reactions^[115]45. Biogenic amines
   are mainly formed by microbial strains producing amino acid
   decarboxylase catalyzing the decarboxylation of free amino acids.
   Therefore, microbial strains producing amino acid decarboxylase are a
   key factor that limits the content of biogenic amines^[116]46. The
   intestinal flora metabolize the ingested amino acids into biogenic
   amines, which enter the hemolymph through the intestinal wall^[117]33.
   Therefore, the reduced biogenic amine content suggests that DR can
   reduce the accumulation of biogenic amines by inhibiting microbial
   growth and metabolic activity.

   The gut microbiota has a marked impact on lifespan. Studies have shown
   that the gut microbiota induces symbiosis in the host after a problem
   with the immune system, thereby promoting the occurrence of
   aging-related diseases^[118]29. In short, the gut microbiota indirectly
   promotes aging-related diseases. Diet has a huge impact on the
   intestinal microbes of silkworm, which indicates that DR induces
   changes in the intestinal flora and changes the intestinal microecology
   of silkworms, thereby reducing the occurrence of inflammation,
   stabilizing the immune system, and prolonging silkworm lifespan.

DR altered other metabolites

   DR also enriched nucleic acid and vitamin metabolism in silkworms.
   Elevated purine/pyrimidine metabolites (1.63-fold increase for thymine
   and 1.39-fold increase for hypoxanthine) support RNA synthesis and
   recombination, which also confirms that DR may prolong lifespan by
   epigenetic mechanisms^[119]47. Vitamins increase the lifespan of
   various organisms primarily by acting as antioxidants^[120]42. However,
   vitamins such as biotin, riboflavin, ascorbic acid, and choline all
   showed a downward trend. While vitamins are generally considered to
   have health benefits, there is growing evidence that they also shorten
   lifespan. The antioxidant function of ascorbic acid reduces the long
   lifespan conferred by mildly increased reactive oxygen species levels
   in C. elegans^[121]48. Moreover, vitamin A supplementation for 6 months
   increased C-reactive protein, an inflammatory factor^[122]49. These
   studies suggest that the conventional wisdom that vitamins promote
   health benefits and slow aging should be applied with caution. In
   addition, we must consider the possibility that different species have
   different physiological responses or dietary signal responses, and the
   mechanism of action of many vitamins in silkworm remains to be proved
   by further studies.

DR promoted sex-specific metabolite changes

   The impact of sex on life expectancy is an issue that cannot be
   ignored^[123]50. In this study, the lifespan extension caused by DR was
   more obvious in female silkworms than that in male silkworms,
   indicating that the effect of DR on lifespan extension was
   sex-specific, and the effect was better in female. Among the 33 common
   differential metabolites, 8 showed opposite trends in the sexes, namely
   L-cystathionine, L-tyrosine, phenethylamine, L-arginine, adipic acid,
   dulcitol, L-lysine, and L-piperic acid (Supplementary Fig. [124]5).
   Most of these metabolites are amino acids and their intermediates.
   Differences in cystathionine are related to egg formation, and female
   silkworms require more of it than males. Cystathionine decreased 3.23
   times in males and increased 3.24 times in females after DR treatment,
   indicating that the utilization of specific amino acids was changed in
   a sex-specific manner.

Conclusion

   We employed LC–MS/MS-based metabolomics to evaluate DR-induced changes
   in hemolymph metabolites. DR altered organic acids and amines,
   especially the ratio and utilization of amino acids, indicating that
   the DR group has a higher amino acid utilization, a healthier gut
   flora, and a more active energy response than the AL group. DR promotes
   better antioxidant capacity and lower lipid peroxidation and
   inflammatory precursors, which may be the reason for the lifespan
   extension. Restricting the content of certain amino acids and
   maintaining adequate vitamin levels will be key considerations in the
   future development of food and medicine for aging. We also explored
   some lifespan-related biomarkers and genes based on the relationship
   between the metabolome and genes and diseases. Our future research will
   explore these biomarkers and genes and evaluate the long-term effects
   of food restriction on the body.

Supplementary Information

   [125]Supplementary Information.^ (18.8KB, csv)
   [126]Supplementary Figures.^ (1.2MB, docx)
   [127]Supplementary Tables.^ (30.8KB, docx)

Author contributions

   X.S. designed the research project. Z.T., B.F. and A.Y. conducted the
   experiments. M.W. and Y.S. analyzed the data. M.W. prepared the
   manuscript. The product names or products mentioned in this article is
   for specific information only and does not indicate a recommendation by
   the authors. All authors reviewed the manuscript.

Funding

   This work was supported by the National Natural Science Foundation of
   China (Grant No. 32072791), and Postgraduate Research & Practice
   Innovation Program of Jiangsu Province (No. KYCX22-3761; SJCX22-1983).

Data availability

   Most of the analytical data is provided in the article and
   supplementary data. More original datasets used and/or analysed during
   the current study available from the corresponding author on reasonable
   request.

Competing interests

   The authors declare no competing interests.

Footnotes

   Publisher's note

   Springer Nature remains neutral with regard to jurisdictional claims in
   published maps and institutional affiliations.

   These authors contributed equally: Meixian Wang, Yichen Shen and
   Zhicheng Tan.

Supplementary Information

   The online version contains supplementary material available at
   10.1038/s41598-023-34132-9.

References