A conserved role of bam in maintaining metabolic homeostasis via regulating intestinal microbiota in Drosophila

Drosophila strains

Fly stocks were reared with standard culture mediums (6.65% cornmeal, 7.15% dextrose, 5% yeast, 0.66% agar supplemented with 2.2% nipagin and 3.4 ml/l propionic acid). The w¹¹¹⁸ strain was used as the control and the tool strain for isogenization. The P{NP1-Gal4 } (gut driver), P{Nos-Gal4 } (germline driver), P{Ppl-Gal4 } (fat body driver), P{Elav-Gal4 } (neuron driver), bam^Δ86 (bam loss-of-function mutant), P{Uasp-artmiR-bam } (bam KD), and P{Uasp-bam} strains were obtained from the Bloomington Drosophila Stock Center and described previously (Guo et al., 2014; Ji et al., 2017).

Mice

Wild-type (WT, C57BL/6) and Gm114 knock out (Gm114 KO) mice (Tang, Ross & Capel, 2008) were purchased from Nanjing Myris Biotechnology Co. and maintained under pathogen-free conditions in the animal care center of Anhui Agricultural University (22 ± 2 °C, 12/12 h light/dark cycle). Each mouse has been taken as the experimental unit, and the mice used in the experiments were randomly selected. All mice were carried out in accordance with the “Regulations on Animal Management” of the Ministry of Health of the People’s Republic of China and the plan approved by the Laboratory Animal Center of Anhui Agricultural University (SYXK 2020-007). Intraperitoneal injection of sodium pentobarbital was used to euthanize the mice at the end of the experiments. The operating procedures involving animals in animal experiments have been approved by experimental Animal Center of Anhui Agricultural University, and the IACUC approval number is AHAU 210429.

Antibodies

Western blotting assays were performed according to methods described previously (Hua et al., 2022). The following antibodies were utilized for Western blotting: Bam (mouse, 1:3000, DSHB), Actin (mouse, 1:5000, Cat#A4700; Sigma), and Goat anti-Mouse IgG H&L (1:3000, Cat#ab150078; Abcam).

Examination of total lipids via GC–MS

The GC-MS assays were performed as previously described (Papadimitropoulos et al., 2018; Tennessen et al., 2014). Briefly, indicated flies were added with 800 µl prechilled 90% methanol containing 1.25 µg/ml succinic-d4 acid (293075; Sigma) and 6.25 µg/ml U-13C, U-15N amino acid mix (CDNLM-6784; Cambridge Isotope). Samples were then homogenized for 30 s at 6.45 m/s and incubated at −20 °C for 1 h to enhance protein precipitation, followed by centrifugation at 20,000 g for 5 min at 4 °C to remove the resulting precipitate. The supernatant was transferred to a 1.5 ml microfuge tube and the solvent removed with a Speed-Vac (Genevac). The GC-MS detection and analysis software was opened to create the experimental method (6890 Agilent GC with Leco pegasus IV time-of-flight MS instrument). 0.5ul of mixed reagent was drawn with a micro syringe and the sample was injected through the sample inlet of the gas chromatograph. Last, we use the BinBase database software to perform analysis.

Measurements of metabolites

The analyses of metabolites including TAG (triacylglyceride), Gly (glycogen), and Glu (glucose) were performed to examine the metabolic profiles of indicated flies (Tennessen et al., 2014; Fan et al., 2017). Briefly, flies were collected and frozen at −80 °C until use. 5 flies per group were homogenized in 1 × PBST and centrifuged at 1,000 g for 2 min. Proper aliquot of the supernatant was used to measure protein concentration by the Bradford assay (BCA1; Sigma), or TAG level by working reagent (T2449; Sigma), or Gly and Glu levels using HK kit (MAK091; Sigma). The total amounts of TAG, Gly, and Glu were normalized to the total protein level.

Cafe and Fluorescein feeding assays

The cafe assay (Cai et al., 2021) was performed to monitor the levels of food consumption of indicated flies. Briefly, two labeled calibrated glass micropipettes (5 µl, VWR) were filled with liquid medium and were inserted through the cap via truncated 200 µl pipette tips. 10 flies per vial were subjected to the long-term experiments (six independent replicates for each group) under a 12 h light/12 h dark cycle in a room kept at 25 °C and 60% humidity. Each experiment included an identical Cafe chamber without flies to determine evaporative losses (typically <10% of ingested volumes), which were subtracted from experimental readings.

Flies were put into vials with meals that contained 50 µM Fluorescein (F6377; Sigma) for the Fluorescein feeding assay. 10 flies were collected and lysed with 10 mM K-phosphate buffer (pH 6.0) 2 h later and 6 independent replicates for each group. Lysates were put to a 96-well plate (Corning) after centrifugation, and the plate reader (Tecan) was used to conduct the analysis. Excitation at 480 nm and emission at 521 nm were used to measure the fluorescein levels. Equal volume of K-phosphate buffer was utilized as the baseline control in this assay.

RT-qPCR assays

Total RNA from dissected gut, head, fat body, reproductive organ, thorax, and the whole body were extracted using Trizol reagent (15596018; Thermo Fisher) according to the manufacturer’s instructions. cDNA was synthesized using a first-strand cDNA synthesis kit (Transgene, AT301-02). Quantitative PCR was performed in triplicate using SYBR Green Master Mix (A25780; Thermo Fisher) on a Light Cycler 480 following previously published methods (Cai et al., 2018). Relative gene expression levels were normalized to the levels of rp49 in each sample. Primers used for RT-qPCR assays are as follows.

Bam-Fw: CGAGGAAAGCCACTTGTGAG;

Bam-Rv: GTTGCAAGCAATCCAAACCG;

Rp49-Fw: CACGATAGCATACAGGCCCAAGATCGG;

Rp49-Rv: GCCATTTGTGCGACAGCTTAG;

TNF α-Fw: TCCCCAAAGGGATGAGAAGTT;

TNF α-Rv: GTTTGCTACGACGTGGGCTAC;

IL6-Fw: TCGGAGGCTTAATTACACATGTTCT;

IL6-Rv: TGCCATTGCACAACTCTTTTCT.

Measurements of gut microbiota

Intestines (from indicated flies or mice) were collected and subjected to genomic DNA extraction according to the manufacturer’s protocols (E.Z.N.A.® Insect DNA Kit). Quantitative PCR assays were further performed to determine the population of the total intestinal microbiota, Bacilli, Gammaproteobacteria, or Alphaproteobacteria using specific primers. The expression levels of actin or GAPDH were used as internal controls for fly or mice samples, respectively. The sequences for the primers are outlined below.

16S-Fw: AGAGTTTGATCCTGGCTCAG;

16S-Rv: CTGCTGCCTYCCGTA;

Ba-Fw: CGACCTGAGAGGGTAATCGGC;

Ba-Rv: GTAGTTAGCCGTGGCTTTCTGG;

Ga-Fw: GGTAGCTAATACCGCATAACG;

Ga-Rv: TCTCAGTTCCAGTGTGGCTGG;

Al-Fw: CCAGGGCTTGAATGTAGAGGC;

Al-Rv: CCTTGCGGTTCGCTCACCGGC;

Actin-Fw: TTGTCTGGGCAAGAGGATCAG;

Actin-Rv: ACCACTCGCACTTGCACTTTC;

GAPDH-Fw: ATGACATCAAGAAGGTGGTG;

GAPDH-Rv: CATACCAGGAAATGAGCTTG.

Rearing Drosophila under axenic conditions

To raise flies under axenic conditions, we prepared Drosophila foods supplemented with antibiotics (500 µg/ml ampicillin, 50 µg/ml tetracycline, and 200 µg/ml rifamycin) as previously described (Koyle et al., 2016). The vials containing foods were autoclaved for 30 min, followed by 12 h irradiation by a radioactive cesium source.

DSS-induced colitis in Gm114 mice and HE staining

To induce colitis in mice, 2% (w/v) DSS (molecular weight 36,000-50,000; MP Biomedicals, 9011-18-1) was added to the drinking water for 7 d. For the measurement of weight, mice were further treated with normal drinking water for 14 d and measured for daily weight. For HE staining of colon, mice were sacrificed on d 8, and the entire colon was excised. The clean colons opened longitudinally were fixed with 10% formalin. Small pieces of fixed colon tissues were dehydrated and embedded in paraffin (Paraplast Tissue Embedding Medium, LEICA) using a modular tissue embedding system (Leica EG1150 H). The 5 µm sections were cut using a fully automated rotary microtome (Leica RM2255) and mounted on positively charged slides (Adhesion Microscope Slides; CitoGlas; Haimen, Jingsu, China). HE staining was carried out using a kit (Boster Biological Technology Company, Pleasanton, CA, USA).

Statistical analysis

For all statistical analyses, data are shown as means and standard errors. Statistical significances were determined using the two-tailed Student’s t test (in Figs. 1A–1D, 2C–2E, 2G–2H, 3A–3H, 4A, 4C–4F, 5C, Figs. S1A–S1B, Figs. S2A–S2B) or LogRank test in the PASW Statistics 18 software in Fig. 5B. The p value of less than 0.05 was considered statistically significant. * p < 0.05; ** p < 0.01; *** p < 0.001; ns, not significant. Results from Figs. 1A–1D, 2A, 2C–2E, 2G–2H, 3A–3H, Figs. S1A–S1B, Figs. S2A–S2B were obtained from 3 biological replicates. Data from Figs. 4C, Figs. 4E and 4F were obtained from 7 biological replicates. In Figs. 4A and 4D, the numbers of experimental mice were 38 for WT,fig35 for Gm114 KO. In Figs. 5B and 5D, the numbers for DSS-treated WT and Gm114 KO mice were 15 and 17, respectively.

Loss of bam leads to excessive energy storage in Drosophila

To explore the potential involvements of bag-of-marbles (bam) in impacting the metabolic states in Drosophila, we utilized the bam^Δ86 homozygous mutant (bam^−/−), heterozygous mutant (bam^+/−), and w¹¹¹⁸ (bam^+/+, referred as wild-type control), and performed gas chromatography and mass spectrometer (GC-MS) analyses to assess the levels of total lipids. As shown in Fig. 1A, bam loss-of-function mutant flies (at the age of 10-day) exhibited elevated (by ∼20%) levels of total lipids relative to controls. The occurrences of total lipids in bam mutants increased more significantly (by ∼40% and ∼70%) from samples at the ages of 30-day and 50-day, respectively, when compared to those of the indicated controls (Fig. 1A), implying that silencing of bam likely lead to energetic increases in Drosophila.

To further test our hypothesis, we first examined the index of triacylglyceride (TAG), which is the main form of lipid storage and has long been used to define obesity in fruit fly, as in humans (Arrese & Soulages, 2010; Bharucha, Tarr & Zipursky, 2008; Carley & Lewandowski, 2016). Even though we observed overall decreases of TAG in all samples with indicated genotypes during aging, which was consistent with the previous observations (Johnson & Stolzing, 2019), the levels of TAG in bam mutant adults were significantly higher than those in the controls (Fig. 1B). We then sought to quantify the levels of glycogen (Gly), which represents the state of energy storage in the host (Yamada et al., 2018). Consistently, we observed that mutation of bam resulted in marked augmentations (by ∼10% to ∼30%) of Gly abundances throughout the whole adult life (Fig. 1C). We last quantified the levels of the basic metabolite glucose (Glu), since previous studies have shown that Glu is one of the primary forms of circulating carbohydrates in Drosophila, and serves as an essential energy source and substrate for biosynthetic reactions of glycogen (Fan et al., 2017). As illustrated in Fig. 1D, the circulating Glu was approximately equivalent from samples of controls and flies with only one copy of bam mutation at various stages of age. However, this sugar accumulated more in age-paired bam homozygous adults (Fig. 1D). Taken together, our results indicate that bam largely plays a critical role in regulating the metabolic homeostasis in Drosophila.

Silencing of bam is dispensable for impacting food consumption in Drosophila

We sought to determine whether bam is involved in impacting food consumption of Drosophila, since previous evidence has shown that increased nutrient absorption and storage might be a consequence of elevated intake of foods (Kohyama-Koganeya, Kurosawa & Hirabayashi, 2015; Branch & Shen, 2017). To do this, we first performed the cafe assay, a widely used method for directly and accurately measuring feeding rates by using capillary feeds containing liquid media (Ja et al., 2007). However, we failed to observe any obvious increases in food intake of bam mutant adults, when compared with the age-matched controls (Fig. S1A). To further confirm these findings, we subsequently used the Fluorescein as a food tracer (Wang et al., 2005), and conducted natural feeding experiment as previously described (Danilov et al., 2015). We obtained consistent results, as there were no apparent variances among the levels of consumed-foods from the age-paired bam homozygous and heterozygous mutant flies, and the w¹¹¹⁸ controls (Fig. S1B). Collectively, our results suggest that the increased energy storage caused by silencing of bam is not likely through abnormal food consumption in Drosophila.

Prevention of bam in intestines results in host obesity

In order to investigate how bam contributes to regulating the metabolic states in Drosophila, we first sought to explore whether this regulatory relationship is organ- or tissue-dependent. Numerous pioneering studies have shown that bam is highly expressed in reproductive organs (namely ovary and testis), and both necessary and sufficient for the differentiation of germline stem cells and cystoblasts (Ohlstein & McKearin, 1997; McKearin & Spradling, 1990; McKearin & Ohlstein, 1995). Significantly, bam was reported to exist in intestinal cells, contributing to the maintenance of gut homeostasis and host lifespan (Ji et al., 2019). Therefore, we determined to examine the expression profiles of bam in various tissues. We dissected w¹¹¹⁸ adult flies and separately collected tissues including fat body, head, gut, thorax, and reproductive organ. The reverse transcriptional quantitative polymerase chain reaction (RT-qPCR) and Western blotting assays further displayed that bam was highly abundant in reproductive organs, moderately expressed in intestinal and fat body cells, and relatively low expressed in tissues of head and thorax (Figs. 2A and 2B).

We then utilized the widely-used UAS/Gal4 driver system (Brand & Perrimon, 1993) to down-regulate bam expression in specific organs. As shown in Fig. 2C, the P{Nos-Gal4 } (germline driver), P{Ppl-Gal4 } (fat body driver), P{NP1-Gal4 } (gut driver), and P{Elav-Gal4 } (neuron driver) were crossed with Bam KD (Ji et al., 2017) to generate Nos>Bam KD, Ppl>Bam KD, NP1>Bam KD, and Elav>Bam KD, respectively. When we investigated the levels of total lipids and TAG from those flies throughout the whole adult life, we found that down-regulation of bam in either the reproductive organ, or the fat body, or the nervous system was dispensable for affecting the lipid metabolism in Drosophila (Fig. 2C). In particular, prevention of bam expression in intestinal cells resulted in markedly elevated (by ∼20% to ∼40%) storages of energy relative to age-paired (30-day and 50-day, respectively) controls (Figs. 2C and 2D), even though there were no significant statistical variances from samples collected at the early adult stage (10-day). These results were consistent with our previous observations in bam mutant flies (Fig. 1), indicating that bam functions in the intestinal cells to control the metabolic homeostasis of the host during aging.

Restoring bam in guts rescues energetic increase in bam^Δ86 homozygotes

To further verify the regulatory role and ability of intestinal bam for the lipid storage and metabolism, we next performed rescue experiments. As shown in Figs. 2E and 2F, the expression levels of bam were almost comparable in gut samples from both the control (NP1 > +; bam^+/+) and the rescue groups (NP1>bam; bam^−/−). When we collected these flies and performed total lipid and TAG analyses, we found that over-expression of bam in gut cells nearly fully rescued the excessive energy storages in bam^Δ86 homozygotes (Figs. 2G and 2H). Collectively, our results suggest that bam is responsible for controlling lipid metabolism through its essential role in the intestinal tract in Drosophila.

bam contributes to metabolic homeostasis via regulating gut microbiota

Numerous pioneering studies have shown that dysregulation of gut microbiota leads to metabolic disorders both in invertebrates and vertebrates (Vijay-Kumar et al., 2010; Wong, Vanhove & Watnick, 2016; Ji et al., 2019; Guo et al., 2014; Lee & Hase, 2014; Clark et al., 2015), suggesting a pivotal role of the commensal intestinal bacteria in affecting the host metabolism. To further address whether bam is involved in controlling gut microbiota of Drosophila, we first examined the total bacterial load in fly guts at various ages. As shown in Fig. 3A, bam^Δ86 homozygotes displayed higher (more than two-fold) populations of intestinal bacteria compared to those of bam^Δ86 heterozygotes or w¹¹¹⁸ controls. We next performed qPCR assays using different primers determining the occurrences of bacteria including Bacilli, Gammaproteobacteria, and Alphaproteobacteria, which have been shown to be the main causes of intestinal dysfunction in Drosophila (Clark et al., 2015). Indeed, we observed marked elevations (by ∼60% to ∼220%) of the populations of both Bacilli and Gammaproteobacteria but not the Alphaproteobacteria in bam^Δ86 homozygous mutants, when compared to those in the controls (Figs. 3B–3D). These results imply that loss of bam contributes to the dysbiosis of microbiota in Drosophila guts, thus may leading to increased lipid storage. Then, we raised bam mutants and the control flies under axenic condition (Koyle et al., 2016) to eliminate the influence of gut microbiota. As shown in Figs. 3E–3H, bam mutants and the control flies displayed similar metabolic profiles in the context of total lipids and various metabolites, when the gut microbiota was abolished. Collectively, our results suggest that loss of bam leads to metabolic disorder, which may be related to the imbalance of intestinal microbiota in Drosophila.

Gm114 regulates lipid metabolism in mice

Previous study has shown that Gm114 displays similar expression profile in mouse testis to that of its Drosophila homologue bam (Tang, Ross & Capel, 2008). However, silencing of Gm114 hardly affected the development of germ cells and the fertility of mice to a large extent, which indicates that there are no functional parallels between Gm114 and bam in regulating spermatogonial differentiation (Tang, Ross & Capel, 2008). When we raised Gm114 knockout (Gm114 KO) mice under normal condition, we observed that they gained more weights than the wild-type (WT) controls and this phenomenon was more significant in 10- and 20-week animals (Fig. 4A). We also found that Gm114 KO mice displayed obvious fat accumulation (mostly the white adipose tissue) in the abdominal cavity covering most of the internal organs (Figs. 4B and 4C).

We further performed a series of examinations detecting the physiological profiles of these animals. As shown in Fig. 4D, higher blood pressures were observed in Gm114 KO mice, when compared to those of the age-paired WT controls. Moreover, we noted marked increases in the host serum levels of TAG and Glu when Gm114 was knocked out (Figs. 4E and 4F), suggesting mammalian Gm114 plays a similar regulatory role on lipid metabolism to that of its Drosophila homologue bam.

Loss of Gm114 leads to intestinal dysfunction in mice

We then sought to explore whether Gm114 is also involved in modulating intestinal homeostasis. To do this, we first performed qPCR assays to determine whether the total population of gut microbiota was affected by silencing of Gm114. As shown in Fig. 5A, the level of commensal bacterium in Gm114 KO guts was markedly increased (by ∼70%) when compared to that of the WT. We next observed that the mRNA levels of inflammation markers such as TNF α and IL6 were elevated by knocking out of Gm114 (Figs. S2A and S2B). Then, we administered Gm114 KO mice and WT controls with 2% DSS in drinking water for 7 days to induce a colitis model. We indeed observed an apparent loss of body weight in the WT mice and a faster leveling off in the weight of Gm114 KO mice (Fig. 5B). Of note, when we dissected the intestines from Gm114 KO and WT mice, we found that the DSS treatment-induced reduction of colon length was more evident in Gm114 KO mice than in WT controls, though the colon lengths were comparable among mock-treated animals (Figs. 5C and 5D). To further analyze the histology of the colons, we carried out hematoxylin-eosin (HE) staining assays. Our results showed that the intestinal injury of Gm114 KO mice was more severe than that of the WT mice: the intestinal cell morphology was completely lost and tissue was atrophied after being treated with 2% DSS (Fig. 5E). Most of our observations on DSS-treated Gm114 KO mice are consistent with what were found in a recent study (Sun et al., 2022). Taken together, these results strongly suggest that Gm114 is essential for maintaining the intestinal homeostasis in mice.