Apolipoprotein A-V deficiency enhances chylomicron production in lymph fistula mice

Abstract

Apolipoprotein A-V (apoA-V), a liver-synthesized apolipoprotein discovered in 2001, strongly modulates fasting plasma triglycerides (TG). Little is reported on the effect of apoA-V on postprandial plasma TG, an independent predictor for atherosclerosis. Overexpressing apoA-V in mice suppresses postprandial TG, but mechanisms focus on increased lipolysis or clearance of remnant particles. Unknown is whether apoA-V suppresses the absorption of dietary lipids by the gut. This study examines how apoA-V deficiency affects the steady-state absorption and lymphatic transport of dietary lipids in chow-fed mice. Using apoA-V knockout (KO, n = 8) and wild-type (WT, n = 8) lymph fistula mice, we analyzed the uptake and lymphatic transport of lipids during a continuous infusion of an emulsion containing [³H]triolein and [¹⁴C]cholesterol. ApoA-V KO mice showed a twofold increase in ³H (P < 0.001) and a threefold increase in ¹⁴C (P < 0.001) transport into the lymph compared with WT. The increased lymphatic transport was accompanied by a twofold reduction (P < 0.05) in mucosal ³H, suggesting that apoA-V KO mice more rapidly secreted [³H]TG out of the mucosa into the lymph. ApoA-V KO mice also produced chylomicrons more rapidly than WT (P < 0.05), as measured by the transit time of [¹⁴C]oleic acid from the intestinal lumen to lymph. Interestingly, apoA-V KO mice produced a steadily increasing number of chylomicron particles over time, as measured by lymphatic apoB output. The data suggest that apoA-V suppresses the production of chylomicrons, playing a previously unknown role in lipid metabolism that may contribute to the postprandial hypertriglyceridemia associated with apoA-V deficiency.

after dietary fat intake, the increase of plasma triglycerides (TG) is a normal metabolic consequence characterized by a transient accumulation of TG-rich lipoproteins comprised of intestinally derived chylomicrons and liver-derived very low density lipoprotein (VLDL). Hydrolysis of their TG cores releases free fatty acids to be taken up by muscle and adipose tissues, generating remnant particles that are efficiently removed from the circulation by the liver. Aberrations in lipoprotein metabolism result in elevated plasma TG and a prolonged accumulation of remnant particles that are atherogenic (16, 76, 24) and constitute a major risk factor for cardiovascular disease (51, 43).

Apolipoproteins play crucial roles in the regulation of lipoprotein metabolism by acting as structural proteins, as cofactors for enzymes involved in lipolysis, or as ligands for receptor-mediated clearance of lipoproteins (35). Apolipoprotein A-V (apoA-V) was discovered in 2001 to be expressed exclusively in the liver and located ∼30 kb proximal to the APOAI/CIII/AIV gene cluster (49, 69), a region well-known for regulating lipid metabolism (19). Early studies using genetically engineered mouse models revealed apoA-V to be a potent modulator of plasma TG. Knocking out apoA-V in mice resulted in a fourfold increase in plasma TG while overexpressing human apoA-V in mice reduced plasma TG by threefold compared with WT mice (49). Structural analysis showed that apoA-V is very hydrophobic and predominately lipid-bound (3, 72), associating with plasma high-density lipoprotein, VLDL, and chylomicrons (47). What potentially made apoA-V elusive in its discovery was its low plasma concentration, a range of 114–258 ng/ml in normolipemic individuals (26, 47, 59, 64). The interesting question is how can an apolipoprotein of such low plasma concentration affect plasma TG so dramatically?

Over the years, diverse functions have been proposed for apoA-V. ApoA-V may facilitate lipolysis by interacting with lipoprotein lipase (LPL) and heparin sulfate proteoglycans (38) or with an endothelial cell surface protein, glycosylphosphatidylinositol high-density lipoprotein-binding protein 1 (4). ApoA-V may also interact with members of the low-density lipoprotein-receptor family to enhance receptor-mediated particle uptake (13, 20, 45, 46). However, its low plasma concentration indicates that apoA-V exists on a subfraction of plasma lipoproteins, ∼1 apoA-V molecule for every 24 VLDL particles (38), suggesting that apoA-V must be recycled for efficient lipolysis or clearance of particles to occur. Some studies suggest that apoA-V exerts its effects intracellularly rather than extracellularly, by interacting with lipid droplets in hepatocytes (8, 17, 62, 63). In the hepatocyte, and also in the enterocyte although the mechanisms are nuanced, TG accumulates as cytosolic lipid droplets on the endoplasmic reticulum (ER) membrane before being incorporated into lipoproteins for secretion (see Ref. 41). Hence apoA-V is proposed to facilitate the intrahepatic accumulation of lipid (8, 64), which may reduce the production rate of VLDL (8, 49), although this effect was not seen in other studies (15, 20, 38).

In 2001, apoA-V was also discovered to be a novel factor significantly upregulated during early liver regeneration (70). Originally, the authors thought apoA-V stimulated lipid uptake by the liver for regeneration of cell membranes. However, they observed that apoA-V expression and plasma levels peaked at 6 h after partial hepatectomy, while cell mitosis began 20 h later. ApoA-V was speculated to inhibit lipid flux to protect the reduced-size liver against a possible lipid overload, thereby behaving like a hepatic signal protein. It was apparent that, since its discovery, apoA-V was thought as an atypical apolipoprotein with its low plasma concentration and significant ability to modulate plasma TG through a variety of proposed mechanisms.

While mechanisms of apoA-V have been focused on the liver or plasma, no studies have investigated its physiological role in the gut. Only a few studies have looked at effects of apoA-V on plasma TG after a fatty meal. High levels of apoA-V in transgenic mice protected against postprandial lipemia after an intragastric dose of sunflower oil (15). Adenoviral gene transfer of apoA-V to mice also dose-dependently reduced postprandial TG response (49). Authors of these studies concluded that, since apoA-V was only expressed in the liver, the effects were unlikely a result from reduced absorption but rather increased clearance of lipoproteins. However, what has not been investigated thus far is the critical question: does apoA-V play a role in the absorption of dietary lipids by the gut?

The present study was designed to examine the effects of apoA-V deficiency on the steady-state absorption and lymphatic transport of dietary lipids in chow-fed mice. Based on previous observations of the effects of apoA-V on postprandial TG levels (15, 49), and its association with lipid droplets and potential facilitation of TG accumulation in the ER (8, 62, 63, 64), we hypothesized that apoA-V deficiency would increase the transport of dietary TG across the enterocyte and its subsequent secretion as chylomicrons. To test this, we employed the conscious lymph fistula model to directly examine the transport of dietary lipids from the intestinal lumen to the lymph in apoA-V knockout (KO) and wild-type (WT) mice, providing insight into whether a lack of apoA-V affected uptake, intracellular metabolism, or the secretion of lipid into lymph. Our studies show that apoA-V deficiency in mice significantly increases the absorption and lymphatic transport of dietary lipids into lymph, providing compelling evidence for a role of apoA-V in inhibiting lipid absorption and thereby acting as a hepatic signal to modulate the production of chylomicrons by the gut.

MATERIALS AND METHODS

Animals.

ApoA-V KO mice used in this study were backcrossed to a C57BL6/J background for at least eight generations, and their genotypes were confirmed by PCR. ApoA-V KO and age-matched WT control mice were maintained on standard rodent chow (LM-485 Mouse/Rat Sterilizable Diet; Harlan Laboratories, Madison, WI) under a normal 12:12-h light-dark cycle at the University of Cincinnati Laboratory Animal Medical Services. Only 3- to 4-mo-old males were used in the study. All procedures performed were approved by the University of Cincinnati Internal Animal Care and Use Committee and complied with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Lymph and duodenal cannulation.

Before surgery, mice were fasted overnight with free access to water. The following morning, the animals were anesthetized with ketamine (80 mg/kg) and xylazine (20 mg/kg), and their intestinal lymph ducts were cannulated with polyvinylchloride (PVC) tubing (0.20 mm ID; 0.50 mm OD), with slight modifications to the procedures originally described (9). Instead of being secured with a suture string, the cannula was secured with a drop of tissue glue. A surgical incision was made on the fundus of the stomach, and a second PVC tubing (0.5 mm ID, 0.8 mm OD) was introduced through the fundus of the stomach into the duodenum. The tubing was secured by a purse-string and then with a drop of cryanoacrylate glue. After surgery, the animals were allowed to recover for at least 24 h in Bollman restraining cages that were maintained at a warm temperature of 30°C. Despite the restraints, the animals still had considerable freedom to move forward and backward and laterally. During recovery, the animals received a continuous infusion of a 5% saline-glucose solution at a constant rate of 0.3 ml/h to compensate for fluid and electrolyte loss due to lymphatic drainage.

Lipid infusion and lymph collection.

A lipid emulsion was prepared by dissolving triolein {labeled with [9,10-³H(N)]triolein}, cholesterol {labeled with [1,2-¹⁴C(N)]cholesterol}, and egg phosphatidylcholine in chloroform, which was then evaporated under nitrogen gas. The chloroform-free mixture was emulsified with 19 mM sodium taurocholate (NaTC) in phosphate-buffered saline (PBS) at pH 6.4. The emulsion was pulse sonicated until the emulsion appeared homogeneous. Homogeneity was verified by determining radioactivity in samples taken from the top, middle, and bottom of the emulsion. Emulsions were considered homogeneous if the counts did not vary >1%. The morning after surgery, the overnight saline-glucose solution was replaced with the lipid emulsion described above. Each mouse received a continuous intraduodenal infusion of the lipid emulsion for 6 h at a rate of 0.3 ml/h. A continuous infusion enables measurement of transport rates at the steady state (5). The hourly infusate contained 4 μmol triolein labeled with 1 μCi [³H]triolein, 0.78 μmol cholesterol labeled with 0.1 μCi [¹⁴C]cholesterol, 0.78 μmol egg phosphatidylcholine, and 5.7 μmol NaTC in PBS. Lymph samples were collected hourly in microcentrifuge tubes on ice. Because the animals are conscious during lipid infusion, our experiments lacked the complications from anesthesia, which could affect the production of chylomicrons (56).

Collection of intestinal tissues and luminal contents.

At the end of the 6-h study, mice were anesthetized in preparation for tissue and organ removal. The small intestine was carefully removed with its mucosal tissue left intact and cut into four equal-length segments through three separate suture ties. The segments were labeled M1–M4 [M1, (duodenum), M2 and M3 (two equal length segments of the jejunum), and M4 (ileum)]. The content from each segment was emptied into a tube and followed by three washes of 10 mM of NaTC solution. Luminal contents from the stomach, small intestine, and colon were collected from three rinses of 10 mM NaTC. The mucosal tissue of the small intestine was left intact and homogenized using a Polytron homogenizer. Samples from the hourly lymph, luminal washings, and intestinal tissues were taken for determination of radioactivity by scintillation spectroscopy. Lipids recovered from the intestinal lumen and mucosa at the end of the study allowed for analysis of steady-state accumulation of lipids that were not secreted (1).

Absorptive and lymphatic transport index calculations.

To determine the efficiency of the enterocyte in absorbing and transporting lipids into the lymph, absorptive and lymphatic transport indexes were computed as previously described (7). The absorptive index represents the percentage of infused lipid taken up by the enterocyte and is calculated from the following equation: absorptive index = 100% − (%total dose recovered in the lumen). The lymphatic transport index represents the percentage of absorbed lipid secreted into the lymph. It determines the ability of the small intestine to transport the absorbed lipid into the lymph independent of the amount of lipid absorbed. The transport index is calculated from the equation: lymphatic transport index = (%infused dose recovered in lymph)/(absorptive index) × 100.

Chylomicron appearance time.

Chylomicron appearance time was determined as previously described (67, 68). Mice were given a continuous intraduodenal infusion of the same lipid emulsion as described above at a rate of 0.3 ml/h. At the 4th h of lipid infusion when steady-state lymphatic TG transport was reached, the infusion pump was stopped, and a bolus dose of [¹⁴C]oleic acid (0.5 μCi in 100 μl oleic acid) was injected intraduodenally. After injection, the infusion was resumed to maintain steady-state lymphatic TG transport, and lymph samples were collected every minute for 45 min in small glass scintillation vials. Radioactivity was determined by scintillation counting. The first appearance of radioactivity (in min) represented the time for radioactive oleic acid to be taken up by the enterocyte, resynthesized as TG, packaged, and secreted as chylomicrons in the lymph. It is very important to assess the cellular packaging and release of chylomicrons at the steady state of lymphatic output, since chylomicron production rates can depend on factors such as lymph flow or interstitial fluid hydration as previously studied (67, 68).

Measurement of lymphatic outputs of apoB and apoA-IV.

The lymphatic outputs of apoB and apoA-IV were measured by Western blotting. Samples equivalent to 3 s of lymph flow were resolved on 4–15% polyacrylamide gradient gels and transferred to polyvinylidene difluoride membranes. Following transfer, membranes were blocked with 5% nonfat milk in 0.1% Tween 20 in Tris-buffered saline (TBS) for 1 h. Membranes were incubated with antibodies against apoB and apoA-IV (1:5,000) overnight. After being rinsed with TBS plus 0.1% Tween 20, membranes were incubated with peroxidase-conjugated anti-goat antibody at 1:10,000 for 30 min, and signal was developed with Immobilon Western Chemiluminescent HRP substrate (EMD Millipore, Billerica, MA). Protein bands were quantified by densitometric analysis using Total Lab Quant Analysis Software (Fotodyne, Hartland, WI).

Statistical analysis.

The data shown are mean values ± SE. To compare groups throughout the 6-h infusion period, two-way repeated-measures ANOVA with Tukey posttest analysis were used. For comparison of data that have two independent variables, two-way ANOVA was used. A t-test was used for comparisons of only two groups. Statistical analyses were performed using GraphPad Prism version 6. Data were considered significant if P < 0.05.

RESULTS

Lymphatic transport of lipids.

Lymph flow was maintained at 0.2–0.3 ml/h by setting the intraduodenal infusion rate at 0.3 ml/h (Fig. 1). There were no significant differences between the two groups except at the 5th and 6th h, where the lymph flow of the WT mice decreased to 0.11–0.12 ml/h. It should be noted that this decrease occurred after steady state has been achieved and does not affect our analyses of lymphatic output of lipid (68).

Fig. 1.

Lymph flow was maintained at 0.2–0.3 ml/h in apolipoprotein A-V (apoA-V) knockout (KO) and wild-type (WT) mice. Mesenteric lymph flow was measured hourly for 6 h. *P < 0.05. Values are means ± SE.

The amount of TG transported into the lymph per hour was measured by ³H and expressed as a percentage of the hourly infused dose (Fig. 2A). Interestingly, apoA-V KO mice showed substantially higher lymphatic ³H transport at the steady state (52.9 ± 3.3%) than WT controls (30.2 ± 1.5%), with differences seen as early as 1 h after the start of lipid infusion. Steady state was achieved within 3–4 h for both groups of animals.

Fig. 2.

ApoA-V KO mice show increased triglyceride (TG) and cholesterol transport into the lymph compared with WT. Lymphatic outputs of ³H (A) and ¹⁴C (B) were expressed as percentages of the hourly dose. Lymph fistula mice were intraduodenally infused with a lipid emulsion containing [³H]TG and [¹⁴C]cholesterol for 6 h. *P < 0.05. Values are means ± SE.

The amount of cholesterol transported into the lymph was measured by ¹⁴C and expressed as a percentage of the hourly infused dose (Fig. 2B). As with ³H, the lymphatic transport of ¹⁴C was also significantly higher in apoA-V KO (29.4 ± 2.9%) than in WT mice (9.9 ± 4.3%) at the 6th h, with differences evident within 2 h after the start of lipid infusion. With lymphatic transports of both ³H and ¹⁴C higher in apoA-V KO mice, the data suggest that chylomicron production is significantly higher in these mice, revealing a novel role for apoA-V in regulating intestinal lipid secretion under physiological conditions.

Distribution of lipid along segments of the small intestine.

With increased lymphatic transport of lipid in the apoA-V KO mice, we wondered if the mucosa accumulation and distribution of lipids were different in these animals. Figure 3 illustrates the distribution of ³H and ¹⁴C accumulated along four equal-length segments of the small intestine, M1–M4 (proximal to distal) at the end of the 6-h infusion period. The pattern of accumulation of ³H and ¹⁴C for both groups of animals showed a decrease from M1 to M4, confirming that the primary sites of lipid absorption were mainly located at the proximal half of the intestine (44). ApoA-V KO mice had significantly less ³H in the whole mucosa than WT (P = 0.0004), with the difference being significant in M1 (P = 0.0025) and M2 (P = 0.0003) (Fig. 3A). The data are consistent with our notion that A-V KO mice are more efficient in packaging the absorbed TG into chylomicrons for secretion.

Fig. 3.

Distribution of ³H (A) and ¹⁴C (B) along the small intestine divided into 4 segments of equal length, from proximal to distal: M1 (duodenum), M2 and M3 (jejunum), and M4 (ileum). *P < 0.05. Values are means ± SE.

This trend was not seen with cholesterol recovery in the mucosa. Although apoA-V KO mice had slightly more ¹⁴C recovered from M1 (P = 0.0359), they had a similar distribution of ¹⁴C recovered from the rest of the segments compared with WT (Fig. 3B). Taken together, the data suggests that apoA-V affects the handling of dietary TG and cholesterol in the intestinal mucosa differently.

Recovery of radiolabeled lipid at the end of 6 h continuous infusion.

By accounting for the total amount of lipid from the lymph and from the intestinal mucosa and lumen at the end of the study, we were able to analyze how a lack of apoA-V affected steady-state transport of lipids from lumen to lymph. We also collected luminal contents from the stomach and colon to account for lipids that may have refluxed or have been excreted. Figure 4 illustrates the total amount of dietary TG and cholesterol remaining in the lymph, intestinal mucosa, intestinal lumen, stomach, and colon after 6 h of infusion. ApoA-V KO mice showed significantly increased ³H in lymph (P = 0.0001) with a concomitant reduction in ³H recovered from the intestinal mucosa (P = 0.0004) and lumen (although insignificant, P = 0.0584) compared with WT (Fig. 4A). apoA-V KO mice also had significantly increased ³H recovered from the lymph (P = 0.0001), with a possible slight reduction in recovery from the lumen (P = 0.08) compared with WT (Fig. 4B). Total ³H recovery was no different between the groups (WT 68.72 ± 3.01 vs. KO 70.25 ± 3.56%). Although total recovery was <100%, the value obtained is consistent with a number of previous studies showing incomplete absorption of long-chain TG into the lymph (10, 75), possibly due to a small percentage of transport to the portal vein (25, 37) or oxidation.

Fig. 4.

Total recovery of ³H (A) and ¹⁴C (B) at the end of the 6-h lipid infusion. Lymph refers to the total radiolabeled lipid collected from the lymph from the 6-h infusion period. Mucosa refers to radioactivity recovered after homogenization of the small intestine. Lumen refers to luminal contents from the small intestine that were washed and collected at the end of infusion. Very low amounts of TG were recovered from the colon and stomach, indicating little excretion and reflux, respectively. *P < 0.05. Values are means ± SE.

Unlike ³H recovery, there was no difference in the amount of ¹⁴C recovered from the mucosa of the two groups of mice, probably as a result of the mixing of labeled cholesterol with unlabeled cholesterol in the various pools within the mucosa. Very little amounts of ³H (Fig. 4A) and ¹⁴C (Fig. 4B) were recovered from the stomach, indicating that the infused lipids refluxed very minimally back into the stomach. Also, very little ³H and ¹⁴C were recovered from the cecum, showing that the infused lipids were minimally excreted during our 6-h study. Overall, the data suggest that the apoA-V KO mice were more efficient in taking up the TG and cholesterol presented in the intestinal lumen and secreting them in the lymph.

To numerically represent the efficiency of the intestine to take up lipid and to transport the absorbed lipid into the lymph, two indexes were calculated and shown in Table 1. Both the absorptive index and lymph transport index for ³H were higher in apoA-V KO mice than in WT, suggesting more efficient uptake and secretion of dietary TG. The lymph transport index for ¹⁴C was significantly higher in apoA-V KO mice, indicating more efficient secretion of dietary cholesterol into the lymph. These indexes illustrate apoA-V deficiency results in more efficient chylomicron secretion, independent of lipid uptake.

Table 1.

Absorptive and lymphatic transport index

	TG		Cholesterol
	Absorptive Index	Lymphatic Transport Index	Absorptive Index	Lymphatic Transport Index
WT	91.75 ± 1.54	25.81 ± 3.56	74.80 ± 2.82	7.62 ± 1.51
apoA-V KO	95.36 ± 0.32*	49.08 ± 3.08*	81.23 ± 1.99	22.00 ± 1.30*

Values are means ± SE.

TG, triglyceride, WT, wild type; apoA-V, apolipoprotein A-V; KO, knockout.

^*P < 0.05.

Chylomicron appearance time.

As an alternative measure of chylomicron secretion efficiency, we measured the chylomicron appearance time, which is the time for [¹⁴C]oleic acid placed in the intestinal lumen to appear in the lymph as ¹⁴C chylomicrons during steady-state transport. It represents the total time required for lipid uptake, resynthesis into TG, assembly into chylomicrons, and secretion into the lymphatics. Figure 5 shows the chylomicron appearance time is significantly (P < 0.05) shorter in apoA-V KO mice (21.5 ± 3.3 min) than in WT mice (32.8 ± 2.1 min), further demonstrating that apoA-V KO mice are faster at generating chylomicrons.

Fig. 5.

ApoA-V KO mice have a shorter chylomicron appearance time (min) after [¹⁴C]oleic acid injection in the intestinal lumen. apoA-V KO and WT mice were infused intraduodenally with a lipid emulsion containing TG. At the 4th h of lipid infusion when steady-state lymphatic TG transport was reached, [¹⁴C]oleic acid was introduced into the lumen, and lymph samples were collected every minute afterward for 60 min. *P < 0.05. Values are means ± SE.

Lymphatic output of apoB and -A-IV.

Because the data thus far suggest that chylomicron production is enhanced in apoA-V KO mice, we measured the lymphatic output of apoB, which represented the number of chylomicrons produced over time, since only one apoB protein is assumed to be one chylomicron particle (2). While WT mice had a constant lymphatic output of apoB as expected, apoA-V KO mice showed a steady increase in the lymphatic output of apoB (Fig. 6, A and B), suggesting a greater number of chylomicron particles produced over time. Because apoA-IV is presumed to facilitate the assembly of chylomicrons (73), we also measured the lymphatic output of apoA-IV. There was no difference between the two groups in the lymphatic output of apoA-IV relative to fasting (Fig. 6C) although both groups displayed an ∼1.5-fold increase over 6 h that is comparable to previous studies (22, 28). This suggests that apoA-V does not affect the lipid-stimulated secretion of apoA-IV by the intestine. Overall, these data suggest that the increased lymphatic output of dietary TG in apoA-V KO mice is a result of increased production of chylomicron particles.

Fig. 6.

ApoA-V KO mice have increased intestinal apoB secretion into the lymph. A: lymph samples were analyzed for apoB by Western blot. Protein concentration was quantified by densitometric analysis and normalized to hour 0 (fasting). B: a representative blot is shown for lymphatic apoB output. C: lymph samples were analyzed for apoA-IV by Western blot.

DISCUSSION

While apoA-V is well-demonstrated to profoundly regulate fasting plasma TG in both mice (20, 49, 15, 49, 61, 70, 64) and humans (1, 14, 23, 32, 21), little is known about the contribution of apoA-V to lipid absorption and postprandial TG metabolism. Results from apoA-V transgenic mice (15) and adenovirus-mediated gene transfer experiments to mice (49) suggest that apoA-V suppresses the surge in plasma TG after a lipid meal. However, because apoA-V is only synthesized in the liver (49, 69), mechanistic studies have been focused on enhanced lipolysis (18, 38, 64) or clearance of plasma lipoproteins (20, 38) and not on actions in the gut. The present studies demonstrate a novel role for apoA-V in the absorption of dietary lipids and secretion of chylomicrons by the gut. We show that apoA-V deficiency in mice enhances the transport rate of dietary lipids across the enterocyte, which is accompanied by their rapid secretion from the mucosa into the lymph. This is the first report of apoA-V mediating physiological actions in the intestine.

Postprandial hypertriglyceridemia is an independent risk factor for coronary artery disease (see Ref. 30). The surge in postprandial TG-rich lipoproteins, both intestinally derived chylomicrons and hepatic-derived VLDL, are thought to have deleterious effects on arterial vessel walls (53, 36) and thus contribute to atherosclerosis (34, 71, 76). Increased postprandial plasma TG is also an inherent feature of diabetic dyslipidemia (12, 33), but the underlying mechanisms remain unclear. Since nearly all dietary lipids enter the bloodstream through the intestine, understanding the contribution of intestine to postprandial hypertriglyceridemia is of high clinical and therapeutic importance.

The lymph cannulation method allows us to examine chylomicrons before entering the circulation, which dissociates chylomicron production from clearance. A previous study by Merkel et al. (38) reported that the intestinal secretion of dietary TG was not different in apoA-V transgenic mice compared with WT, although the authors elegantly demonstrated that apoA-V enhanced LPL-mediated hydrolysis of lipoproteins in the plasma. The lipids were administered by intragastric gavage, which is commonly used to measure total absorption but cannot determine its absorptive ability due to variability of stomach emptying (1, 5). Also, studies performed with a bolus dose of lipid cannot give information on the absorption rate of lipid at the steady state, which is an important question when evaluating the contribution of chylomicron delivery to the postprandial TG surge. Furthermore, in the study by Merkel et al., it is possible that a superphysiological level of apoA-V cannot alter the absorption of TG, hence the lack of difference seen in apoA-V transgenic mice. Therefore, to directly address the hypothesis that apoA-V plays an important role in regulating the production rate of chylomicrons, we employed the conscious lymph fistula model under a continuous infusion setting in apoA-V-deficient mice and WT controls.

The results of this study demonstrate that apoA-V deficiency enhances the efficiency of the small intestine to absorb and secrete dietary lipid into the lymph. We show that the transport of [³H]TG from lumen to lymph is accelerated in apoA-V KO mice, as reflected by an increase in the lymphatic transport rate of ³H (Fig. 2A), a decrease in the amount of [³H]TG retained in the intestinal mucosa, and a trend for decrease in the intestinal lumen (Fig. 4A). Further quantitative analyses show that apoA-V KO mice have a greater ability to take up [³H]TG and secrete TG into the lymph (Table 1). This effect is seen specifically in the proximal sections of the small intestine (Fig. 3A). Because more TG is transported as chylomicrons in the apoA-V KO mice, more cholesterol is presumed to be transported out of the small intestine. We find that the transport of ¹⁴C from lumen to lymph is indeed enhanced in apoA-V KO mice (Figs. 2B and 4B), indicating the effect of apoA-V deficiency was not specific to TG. However, there is no difference in ¹⁴C retained in the intestinal mucosa (Fig. 3B), which could be explained by the large intracellular cholesterol pool diluting the exogenous labeled cholesterol.

These results suggest a role for apoA-V in the lipidation of primordial chylomicron particles, which is supported by studies in the hepatocytes where apoA-V is postulated to stabilize lipid droplets (8, 62, 63). It is possible that apoA-V may gain access to enterocyte lipid droplets and, in doing so, serve to modulate TG secretion, and that apoA-V deficiency could cause more rapid movement of intracellular TG into primordial chylomicrons and subsequently enhances chylomicron production. Indeed, when chylomicron appearance time was measured, we found that apoA-V KO mice took less time to produce chylomicrons (Fig. 5) than WT mice, which further strengthens the argument that apoA-V plays an important role in suppressing chylomicron production.

While some reports (15, 20, 38), plus studies in our lab (data not shown), have found no effect of apoA-V on hepatic VLDL secretion, there is a body of evidence that shows that apoA-V KO mice have impaired removal of lipoprotein remnants (15, 20). Our finding that apoA-V deficiency resulted in greater chylomicron particle secretion (Fig. 6, A and B) provides an additional mechanism by which a lack of apoA-V contributes to postprandial lipemia. An increased rate of chylomicron production may result in a greater flux of chylomicron particles to the remnant pool in the plasma. Because apoA-V-deficient mice were previously shown to have reduced LPL activity (20), the rapid influx of chylomicrons might tax the capacity for hydrolysis of TG-rich lipoproteins (chylomicrons plus VLDL) and thereby reduce their removal rate.

Applied to the human population, only a few studies have examined the association of apoA-V polymorphisms with postprandial plasma TG although a clear association of apoA-V polymorphisms and elevated fasting plasma TG has been documented in a wide range of ethnic groups (29, 32, 42, 50, 57). These studies observed an inverse relationship between plasma apoA-V and postprandial TG in patients after a lipid meal. For example, the C allele of the T-1131C polymorphism was found to be associated with hypertriglyceridemia and higher postprandial TG in nonobese Korean (27) and Caucasian (39) populations. Additionally, healthy young males with the APOA5*2 haplotype (−1131C>T) or APOA5*3 haplotype (c.56C>G) also showed higher postprandial response after a fat load test (40). However, the mechanisms underlying these effects were not further explored and were thought to be caused by slower TG hydrolysis and removal of remnant lipoproteins based on evidence in mouse models. Our studies direct attention to a major contribution by the gut in the delivery of chylomicron particles in the postprandial state, signifying the importance of dietary considerations or gut-targeted therapeutics in patients with apoA-V polymorphisms. Our findings are particularly important because they predict higher levels of postprandial apoB48, and significant evidence links increased apoB48 with adverse cardiovascular events (see Ref. 66) and the presence of apoB48 in atherosclerotic plagues (48, 54).

In summary, we have demonstrated for the first time that apoA-V deficiency in mice significantly increases the production rate of chylomicrons by the gut and that this effect seems to come from more efficient shuttling of dietary TG out of the mucosa and into the lymph. Our findings suggest that apoA-V does not behave as a typical classical apolipoprotein but rather a signal protein, communicating between the liver and the gut. This suggestion was first put forth by Van der Vliet and colleagues in 2001, who speculated that apoA-V may act as a signal released by the liver during regeneration to protect itself against a lipid overload (70). In a normal physiological condition, apoA-V may be released by the liver to reduce lipid flux from the gut. Interestingly, cases of apoA-V deficiency in humans do demonstrate a higher postprandial TG excursion (27, 31, 40). By extension, the postprandial hypertriglyceridemia seen in these patient populations may be attributed to an increased rate of delivery of dietary lipids to the bloodstream.

GRANTS

This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants DK-092138 and DK-059630 (University of Cincinnati MMPC) to P. Tso and American Heart Association Grant 13PRE17140013 to L. Zhang.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Author contributions: L.S.Z. and P.T. conception and design of research; L.S.Z., M.X., and Q.Y. performed experiments; L.S.Z. analyzed data; L.S.Z. interpreted results of experiments; L.S.Z. prepared figures; L.S.Z. drafted manuscript; L.S.Z., R.O.R., P.N.H., and P.T. edited and revised manuscript; L.S.Z., M.X., Q.Y., R.O.R., P.N.H., and P.T. approved final version of manuscript.