Mutation of the lbp-5 gene alters metabolic output in Caenorhabditis elegans.

Intracellular lipid-binding proteins (LBPs) impact fatty acid homeostasis in various ways, including fatty acid transport into mitochondria. However, the physiological consequences caused by mutations in genes encoding LBPs remain largely uncharacterized. Here, we explore the metabolic consequences of lbp-5 gene deficiency in terms of energy homeostasis in Caenorhabditis elegans. In addition to increased fat storage, which has previously been reported, deletion of lbp-5 attenuated mitochondrial membrane potential and increased reactive oxygen species levels. Biochemical measurement coupled to proteomic analysis of the lbp-5(tm1618) mutant revealed highly increased rates of glycolysis in this mutant. These differential expression profile data support a novel metabolic adaptation of C. elegans, in which glycolysis is activated to compensate for the energy shortage due to the insufficient mitochondrial β-oxidation of fatty acids in lbp-5 mutant worms. This report marks the first demonstration of a unique metabolic adaptation that is a consequence of LBP-5 deficiency in C. elegans.

INTRODUCTION

Fatty acids play many important roles in various cellular processes and in nutritional balance (1,2). Because fatty acids are hydrophobic molecules, they are solubilized and transported within the cell mainly by specific lipid-binding proteins (LBPs) in C. elegans and by lipid chaperones termed fatty acid-binding proteins (FABPs) in mammals (3,4). These proteins are also involved in targeting fatty acids to specific metabolic regulatory pathways and in controlling gene expression during cell growth (5), influencing energy homeostasis (5,6).

We previously investigated the novel functions of LBPs in energy homeostasis in vivo through RNA interference (RNAi) of C. elegans lbp genes (lbp-1 to -9) and found that fat accumulation was most pronounced in worms with silenced or mutated (7), which was one of the nine genes studied. This was further supported by the severe fat accumulation observed in worms with the lbp-5(tm1618) mutant (7), suggesting insufficient transport of fatty acids to the mitochondria for β-oxidation.

However, little is known about the roles of LBPs in other related cellular metabolic processes, such as glycolysis and the tricarboxylic acid (TCA) cycle. We address this issue by introducing a nonfunctional mutant and analyzing the consequences of losing the function provided by Ibp-5 mutant in related metabolic pathways, such as glycolysis. Furthermore, we hypothesize that improper regulation or disruption of cellular lipid transport by LBPs would alter the metabolic processes in a compensatory mechanism to maintain energy homeostasis in C. elegans.

Here, we present evidence that when the intracellular lipid chaperone gene is disrupted, C. elegans suffers from functional attenuation of the mitochondria-where fatty acids are usually metabolized to provide energy-but metabolically adapts by activating aerobic glycolysis to overcome the energy shortage.

RESULTS

Mutation of lbp-5 in C. elegans causes significant metabolic alterations

Previously, it was found that RNAi-mediated knockdown or mutation of lbp-5 caused more fat accumulation in C. elegans than the disruption of other members of the LBP family did (7). To better understand the role of lbp-5 in energy metabolism, we performed a series of biochemical assays to assess the state of energy homeostasis in lbp-5(tm1618) mutant worms (Supplemental Information). First, to evaluate the efficiency of energy production in mitochondria, ΔΨm was determined using the TMRE uptake in the lbp-5(tm1618) mutant and N2 worms (8). Surprisingly, the ΔΨm in lbp-5(tm1618) mutant worms was much lower than that in N2 worms according to TMRE uptake (Fig. 1A, B). This result suggests the influence of reduction in energy production on other metabolic pathways.

Fig. 1.

Metabolic consequences of lbp-5 deficiency in Caenorhabditis elegans. (A) Deletion of lbp-5 decreases mitochondrial membrane potential. Confocal (upper panel) and fluorescence (lower panel) images of tetramethylrhodamine ethyl ester (TMRE)-stained nematodes. Scale bars, 100 μm. (B) Quantification of the data shown in (A). (C) Levels of lactate production in lbp-5(tm1618) mutant and control N2 worms. (D) Levels of lactate dehydrogenase activity in lbp-5(tm1618) mutant and control N2 worms. (E) The effect of deoxyglucose treatment on lactate production in lbp-5(tm1618) mutant and control N2 worms. (F) Deletion of lbp-5 decreases ATP levels in adult nematodes compared with N2 worms. (G) Effect of lbp-5 mutation on ROS levels. Intracellular ROS levels in mutant and control worms were measured by MitoSOX. (H) Effect of paraquat treatment on the survival of mutant and control worms. The number of living worms was counted 72 h after treatment with 4 mM paraquat. The mev-1(kn1) mutant was used as the positive control. Data represent the mean ± SD of at least three measurements from three independent experiments. *P < 0.05 as calculated by Student’s t-test versus the control sample.

To investigate this, we assessed the rate of glycolysis using the conventional lactate dehydrogenase (LDH) activity assay. Additionally, we also measured lactate production levels by comparing the accumulation of L-(+)-lactate, a major stereo-isomer of lactate formed during glycolysis, between mutant and control worms. Our results showed that the bodies of lbp-5(tm1618) mutant worms contained approximately 22% (33 ±1.7 vs. 27 ±1.4 nmol/mg protein; n=3) more lactate than those of N2 worms (Fig. 1C). Consistent with this finding, our examination revealed 3-fold higher LDH activity in lbp-5(tm1618) mutants than in N2 worms (Fig. 1D). Measurement of lactate production following treatment with DOG (17), an inhibitor of glycolysis, demonstrated the connection between the increased lactate production and glycolysis (Fig. 1E). Finally, lbp-5 mutant worms exhibited lower ATP levels than N2 worms, implying that energy production were repressed in lbp-5 mutant worms (Fig. 1F). These results demonstrate that lbp-5 deletion causes a reduction in mitochondrial function but induces a substantial increase in the rate of glycolysis in aerobically cultured worms.

Deficiency of lbp-5 leads to increased reactive oxygen species (ROS) production

Based on the observed suppression of mitochondrial function (Fig. 1A and B), we hypothesized that reduced mitochondrial potential may also accompany compromised anti-oxidant activity. To test this hypothesis, we compared the ROS levels (18) between lbp-5(tm1618) mutant and N2 worms by using mev-1(kn1) mutant worms as the positive control (Supplemental Information). The mev-1 gene encodes the cytochrome b large subunit in complex II of the mitochondrial electron transport chain. Thus, mev-1(kn1) worms display increased levels of superoxide in the mitochondria and sensitivity to paraquat, a superoxide generator frequently used to examine ROS resistance. We found that lbp-5 deficiency led to approximately 48% higher ROS levels in the lbp-5(tm1618) worms than those of the control worms (Fig. 1G). These higher ROS levels are quite similar to the levels found in mev-1(kn1) worms (61% higher than the levels in N2 worms). For further confirmation, we measured sensitivity to paraquat by testing the ability of adults to survive in the presence of a high concentration of the compound (4 mM). The lbp-5(tm1618) mutant worms showed decreased survival with paraquat treatment (Fig. 1H). This result suggests a link between the inability to transport fatty acids in the lbp-5 mutant and the attenuated mitochondrial anti-oxidant function (9). These data also indicate that the reduction in fatty acid transport in this mutant may weaken mitochondrial protection against ROS produced by paraquat.

Proteomic analysis of lbp-5 deficiency in C. elegans supports the basis for metabolic adaptation

To explore the mechanism by which lbp-5 deficiency not only alters cellular metabolism by augmenting the rate of glycolysis but also suppresses mitochondrial function, we conducted a proteomic analysis of lbp-5(tm1618) mutant and N2 wild-type worms (Supplemental Information). This proteomic analysis was designed to evaluate changes in protein abundance and measure the expression of representative genes involved in energy metabolism and oxidative stress. First, differences in protein levels between the two worm strains (N2 vs. the lbp-5 mutant) were documented through 2DE and mass spectrometry (MS) analysis (Fig. S1) (19). Because most enzymes involved in energy metabolism are soluble proteins, it was thought that 2DE would facilitate the detection of changes in the expression of proteins involved in energy metabolism in the lbp-5(tm1618) mutant. From this experiment, we were able to identify 199 protein spots by MALDI-TOF MS analysis. Some of those differentially expressed proteins were further verified by qRT-PCR. Table 1 outlines 22 representative genes that were differentially expressed by at least 2-fold in the lbp-5 mutant strain at both the protein and mRNA levels (in three independent experiments). Fig. 2 shows the qRT-PCR data sorted by different functional groups selected from the proteomics data. Changes in the expression levels of genes involved in glycolytic reactions and mitochondrial energy production are consistent with the novel metabolic adaptation of lbp-5-deficient C. elegans (Table 1). For example, most mitochondrial proteins (ATP-2, ATP-5, TSFM-1, and PDHK-2) were decreased by 2.0- to 2.5-fold, whereas ALH-1, which acts in ethanol oxidation,was increased by 4.1-fold. In particular, the expression levels of LDH-1 (4.9-fold) and GPI-1 (3.4-fold), which are involved in glycolysis, were increased substantially, whereas those of GPD-2 and aldolase, which are involved in side reactions of glycolysis, were decreased by 2.0- and 3.6-fold, respectively. These results confirm earlier findings that glycolysis is activated (Fig. 1C-E) when the mitochondrial energy cycle (TCA) is attenuated (Fig. 1A and B). These differential expression profile data elucidate the novel metabolic adaptation of C. elegans, in which glycolysis is activated to compensate for the energy shortage due to the insufficient mitochondrial β-oxidation/TCA cycle in lbp-5(tm1618) mutant worms (Table 1, Fig. 2A and B). For instance, the proteomics and qRT-PCR data show that the mutant had decreased expression levels of acyl-CoA dehydrogenase (acdh), which catalyzes the first step of fatty acid β-oxidation and thus, plays a key role in energy production. The expression levels of fatty acyl-CoA synthetase (acs) and acetyl-CoA acetyl transferase (B0303.3) catalyzing the production of acetyl-CoA, a key product of fatty acid β-oxidation, were also decreased. Decreased levels of these enzymes indicate reduced fatty acid β-oxidation in the mitochondria. Down-regulation of atp-2 and atp-5, which encode subunits of complex V of the mitochondrial respiratory chain, also reflects lower TCA activity. Decreased was PDHK-2, which negatively regulates the pyruvate dehydrogenase complex (10,11). This catalysis blocks the use of the metabolic fuel from the TCA cycle (Table 1) by initiating the conversion of glycolytic products into fatty acids, resulting in the biosynthesis of more fatty acids and triglyceride (fat accumulation) (11). In contrast, the expression levels of several genes related to glycolysis (e.g., ldh-1) were increased (Fig. 2B). LDH, which catalyzes the irreversible last step of aerobic glycolysis, can be used as an indicator of glycolytic activity. These results are consistent with our biochemical results (Fig. 1). Together with the observed decrease in ΔΨm, these data explain the basis of the novel metabolic adaptation caused by lbp-5 deficiency.

Table 1.

Proteins identified to be differentially expressed between N2 and lbp-5(tm1618) nematodes by two-dimensional electrophoresis and matrix-assisted laser desorption/ionization-time of flight analysis

Spot no.	Accession no.	Protein name	Score	Theoretical	Sequence coverage (%)	MP/TPa	Fold ratiob
				kDa/pI
Mitochondrial metabolism
1	gi|25144756	ATP synthase subunit family (ATP-2)	82	57/5.52	33	15/60	2.2↓
2	gi|17557712	ATP synthase subunit family (ATP-5)	98	22/6.67	52	10/60	2.5↓
3	gi|17561440	TS elongation factor (TSFM-1)	60	34/6.27	34	11/60	2.0↓
4	gi|17556919	Pyruvate dehydrogenase kinase family (PDHK-2)	61	45/6.02	31	12/60	2.5↓
Glycolytic metabolism
5	gi|71996708	Glucose-6-phosphate isomerase family (GPI-1)	165	65/6.08	35	16/60	3.4↑
6	gi|17535107	Lactate dehydrogenase family (LDH-1)	89	36/6.42	37	13/60	4.9↑
7	gi|32566163	Glyceraldehyde 3-phosphate dehydrogenase (GPD-2)	144	36/6.76	52	18/60	2.0↓
8	gi|1703238	Fructose-bisphosphate aldolase 1 (Aldolase CE-1)	114	39/6.26	40	16/60	3.6↓
Ethanol metabolism
9	gi|17562582	Sorbitol dehydrogenase family (SODH-1)	136	38/6.07	48	17/60	4.2↑
10	gi|25144435	Aldehyde dehydrogenase family (ALH-1)	153	55/7.14	28	23/60	4.1↑
11	gi|17538498	S-adenosylmethionine synthetase (SAMS-4)	117	44/5.87	40	14/60	4.5↑
Oxidative stress resistance
12	gi|71981879	Superoxide dismutase family (SOD-1)	68	16/6.14	57	7/60	2.2↓
13	gi|32565758	Glutathione S-transferase family (GST-4)	56	25/5.47	44	7/60	2.1↓
14	gi|32565831	Peroxiredoxin family (PRDX-2)	87	21/5.53	51	11/60	2.1↓
15	gi|17506189	Gamma-glutamyltransferase (C53D5.5)	66	70/5.78	17	12/60	4.0↓
16	gi|25147792	Catalase family (CTL-1)	95	57/6.45	29	14/57	2.0↓
17	gi|25151141	Catalase family (CTL-3)	120	59/6.41	33	16/57	3.7↓
18	gi|17563248	ATPase-like family member (rpt-1)	55	48/6.08	40	13/72	4.2↑
19	gi|17536181	Cytochrome P450 family member (cyp-13A8)	44	58/8.38	21	8/44	5.0↑
Other
20	gi|17506493	Elongation factor family (EFT-2)	57	95/6.10	18	14/60	6.2↑
21	gi|17508687	Ribosomal protein, small subunit family member (rps-6)	58	28/10.26	30	9/60	11.5↓
22	gi|32563802	Neuronal symmetry family member (nsy-1)	67	162/6.07	15	19/56	8.0↓

aNumber of matched peptides and total searched peptides, bn = 3, three independent experiments.

Fig. 2.

Gene expression in lbp-5(tm1618) mutant worms. (A) Quantitative reverse transcription–polymerase chain reaction (qRT-PCR) analysis (20) of genes related to mitochondrial energy metabolism in lbp-5(tm1618) and control N2 worms. The mRNA level of each gene was normalized to actin and presented as the ratio of expression in lbp-5(tm1618) to control worms. Shown here are: acdh (acyl-CoA dehydrogenase)-8, acs-2, and acs-15; atp (ATP synthase)-2 and atp-5; B0303.3 (thiolase), pdhk (pyruvate dehydrogenase kinase)-2, and tsmf (TS elongation factor)-1. The data represent the average of triplicate measurements from three samples per experimental group. *P ＜ 0.05 as calculated by Student’s t-test versus the control sample. (B) Quantitative RT-PCR of genes related to glycolysis in lbp-5(tm1618) and control N2 worms. The mRNA level of each gene was normalized to actin and presented as the ratio of expression in lbp-5(tm1618) to that in control worms. Shown here are: aldolase CE-1 (fructose-bisphosphate aldolase), gpd (glyceraldehyde 3-phosphate dehydrogenase)-2, gpi (glucose-6-phosphate isomerase)-1, and ldh (lactate dehydrogenase)-1. The data represent the average of triplicate measurements from three samples per experimental group. *P ＜ 0.05 as calculated by Student’s t-test versus the control sample. (C) Quantitative RT-PCR analysis of genes related to ethanol metabolism in lbp-5(tm1618) and control N2 worms. The mRNA level of each gene was normalized to actin and presented as the ratio of expression in lbp-5(tm1618) to that in control worms. Shown here are: sodh (sorbitol dehydrogenase)-1, sams (S-adenosylmethionine synthetase)-4, and alh (aldehyde dehydrogenase)-1. (D) Quantitative RT-PCR analysis of genes involved in oxidative stress in lbp-5(tm1618) and control N2 worms. Shown here are: ctl (catalase)-1 and ctl-3, gst (putative glutathione-requiring prostaglandin D synthase)-4, prdx (peroxiredoxin)-2, sod (superoxide dismutase)-1, C53D5.5 (γ-glutamyltransferase), rpt-1 (ATPase-like family member), and cyp-13A8 (cytochrome P450 family member). The mRNA level of each gene was normalized to actin and presented as the ratio of expression in lbp-5(tm1618) to that in control worms. (E) Quantitative RT-PCR analysis of other functional genes in lbp-5(tm1618) and control N2 worms. Shown here are: eft-2 (elongation factor family), nsy-1 (neuronal symmetry family member), and rps-6 (ribosomal protein, small subunit family member). The mRNA level of each gene was normalized to actin and presented as the ratio of expression in lbp-5(tm1618) to that in control worms. The data represent the average of triplicate measurements from three samples per experimental group. *P ＜ 0.05 as calculated by Student’s t-test versus the control sample.

Regarding increased ROS production in the lbp-5(tm1618) mutant, the net effect from the decreased abundance of anti-oxidative enzymes (e.g., GST-4, PRDX-2, CTL-1, and CTL-3) and increased expression levels of oxidation-related enzymes (e.g., SAMS-4, EFT-2, cyp-13A8, and rpt-1) seemingly increased the ROS production and reduced the oxidative stress resistance in lbp-5(tm1618) mutant worms (Table 1, Fig. 1H and 2D). The amounts of ethanol metabolism-related enzymes (SODH-1 and ALH-1) were greatly increased (4.1- to 4.2-fold), producing more oxidative products due to the reduction in mitochondrial function. Thus, up-regulation of both glycolysis and ethanol metabolism may compensate for the attenuation of mitochondrial function (Fig. 2 and Table 1). Both the qRT-PCR and proteomics profile data are consistent with our hypothesis that lbp-5 deficiency causes a unique metabolic adaptation in this nematode (Fig. 3).

Fig. 3.

Summary of metabolic consequences of lbp-5deficiency. Deletion of lbp-5 decreases mitochondrial activity but increases glycolysis and ethanol metabolism. In addition, lbp-5gene deficiency increases ROS production and decreases oxidative stress resistance. Fat accumulation in lbp-5(tm1618) mutant nematodes was most likely be due to the down-regulation of fatty acid β-oxidation.

DISCUSSION

This study was designed to understand the potential effect of lbp-5 gene mutation on metabolic pathway by use of biochemical assays and proteomic analysis. Our data suggest that disruption of the lbp-5 gene not only attenuated mitochondrial function but also enhanced aerobic glycolysis as indicated by lactate accumulation. Core metabolic pathways have been generally well conserved among eukaryotes from yeast to mammals, which use both glycolytic and mitochondrial metabolism depending on extracellular conditions and cues, cellular needs, and the stage of metabolic or circadian cycle (12). For example, fat metabolism in C. elegans occurs through a complex network that includes those genes involved in neural signaling, fatty acid uptake, intracellular transport, storage, and consumption (1).

The evolutionarily conserved intracellular lipid-binding proteins FABP and LBP exhibit different functions in energy homeostasis (5,6,9). Multiple sequence alignment and phylogenetic analysis of the nematode lbp genes indicate that, although the fatty acid-binding domain is highly conserved, the complete sequences and functions are diverse (7). For instance, mutating lbp-5 had metabolic consequences in the worms (Table 1), and the presence of other LBP homologues in C. elegans did not overcome these effects. Our previous work has shown that silencing lbp-5 induces fat accumulation (7). Moreover, we demonstrate that fat accumulation due to lbp-5 deficiency is caused in an nhr-49 dependent manner via effects on some genes involved in mitochondrial and peroxisomal beta-oxidation, fatty acid desaturation/elongation, and gluconeogenesis. In addition, similar to PPAR-α-deficient mice, nhr-49 mutant worms exhibit a decrease in glucose and increase in lactate concentrations (13). These data suggest that LBP-5 and NHR-49 may work together as functional partners in lipid metabolism. However, this relationship should be investigated further before we conclude that the metabolic adaptation is mainly caused by reduced NHR-49 function.

The changes in expression levels of genes associated with various metabolic pathways (e.g., mitochondrial energy production, glycolysis, fat accumulation, ROS generation, ethanol metabolism, and oxidative stress resistance) in lbp-5(tm1618) (Fig. 3) indicate a classical metabolic adaptation in which glycolysis is promoted. Generally, pyruvate has one of three metabolic fates: to become acetyl-CoA for consumption in the TCA cycle, to become ethanol, or to become lactate. Thus, up-regulation of both glycolysis and ethanol metabolism may compensate for the attenuation of mitochondrial function.

Although there were some differences in the magnitude of the changes in the expression levels of certain proteins involved in glycolysis (e.g., GPD-2 and aldolase) between the proteomic analysis and qRT-PCR data, the overall transcription profile data indicate how cellular metabolism is redirected from the mitochondrial β-oxidation/TCA cycle to glycolysis in lbp-5(tm1618) mutant worms (Fig. 2). In fact, differences between mRNA and protein levels of a particular gene are not uncommon (14). The genes listed in Table 1 were analyzed at both the mRNA and protein levels to ensure that the pattern of changes in a certain group of genes and their corresponding proteins were consistent in response to lbp-5 gene mutation.

A major readout of cell status is the level of ROS. Down-regulations of atp-2 and atp-5 also indicate attenuated mitochondrial function, which can decrease the production of ROS in the lbp-5(tm1618) mutant (Fig. 2A). However, ROS levels in the lbp-5 mutant were paradoxically higher than those of the control, supporting the likelihood that some sort of mitochondrial uncoupling had occurred in the lbp-5 mutant. Yet, higher levels of ROS may be more likely caused by the decrease in anti-oxidative genes rather than increased mitochondrial activity, resulting in reduced oxidative stress resistance in lbp-5(tm1618) mutant worms (Table 1 and Fig. 2D). For instance, the expression levels of catalases (ctl-1, ctl-3) and sod-1 were decreased in the lbp-5 mutant, which were also seen in the 2D data (Table 1). However, the exact cause of these changes requires further investigation.

The most notable effect of lbp-5 deficiency is fat accumulation (7), which can be due to storage signals or reduced lipolysis. When cells encounter insufficient amounts of mobilized fatty acids due to lbp-5 deficiency, a certain energy-sensing system yet to be identified responds to the diminished mitochondrial metabolism and initiates the compensatory activation of glycolysis. This activation may divert glycolytic products into biosynthetic pathways, such as the pentose phosphate pathway, leading to fatty acid biosynthesis.

LBP-5 protein is expressed widely in many different cell types. In particular, LBP-5 expression is high in the hypodermis and intestine, which are major sites of fat accumulation. LBP-GFP fusion protein expression was observed during embryogenesis and in all subsequent larval and adult stages (7), suggesting that LBP-5 most likely plays important roles throughout development. Thus, the broad cellular distribution of LBP-5 expression may provide valuable information about how LBP-5 reduces glycolysis in the life cycle of a normal worm.

In conclusion, our data reveal that LBP-5 plays a regulatory role in the energy homeostasis of C. elegans. The direct consequences of lbp-5 deficiency are fat accumulation and ROS production and reduced ΔΨm and anti-oxidation activity. More importantly, lbp-5 mutation leads to a classical metabolic adaptation in which glycolysis is activated and the mitochondrial TCA cycle is suppressed, but this finding needs to be confirmed through additional experiments. Thus, cellular lipid transport proteins, particularly LBP-5, are important subjects of research on obesity and metabolic diseases (15).

MATERIALS AND METHODS

Deoxyglucose (DOG) treatment

Details on the maintenance of strains, mitochondrial assays, and LDH activity analysis are described in Supplemental Information. For DOG treatment, synchronized L1 stage worms were grown to L4 stage, as described previously in (16). Next, nematodes were transferred to plates containing 25 μM 5-fluoro-2’-deoxyuridine (Sigma-Aldrich) to prevent progeny formation. After incubation for 16 h at 20℃, the worms were transferred to plates containing 5 mM DOG and cultured for another 48 h at 20℃. NGM agar plates without DOG were used as control. Worms were harvested by washing three times with a ice-cold M9 buffer to separate the nematodes from bacteria. Nematodes were frozen in liquid nitrogen and stored at −80℃ until further processing. Worm extracts were prepared as described (17,21).

Paraquat resistance

Synchronized L1-stage worms were grown on normal NGM plates until their progeny reached L4 stage, as previously described (22). A total of 50 worms were transferred to NGM plates with 4 mM paraquat and incubated for 72 h at 20℃, after which the number of surviving worms was counted. The data represent the mean ± standard deviation from triplicate measurements obtained from three independent experiments.

Statistical analyses

Statistical significance was determined using Student's t-test (SPSS 16.0, SPSS, Inc., Chicago, IL). Results with P values less than 0.05 were deemed statistically significant.