Mammalian colonocytes possess a carrier-mediated mechanism for uptake of vitamin B3 (niacin): studies utilizing human and mouse colonic preparations

Abstract

Niacin (vitamin B3; nicotinic acid) plays an important role in maintaining redox state of cells and is obtained from endogenous and exogenous sources. The latter source has generally been assumed to be the dietary niacin, but another exogenous source that has been ignored is the niacin that is produced by the normal microflora of the large intestine. For this source of niacin to be bioavailable, it needs to be absorbed, but little is known about the ability of the large intestine to absorb niacin and the mechanism involved. Here we addressed these issues using the nontransformed human colonic epithelial NCM460 cells, native human colonic apical membrane vesicles (AMV) isolated from organ donors, and mouse colonic loops in vivo as models. Uptake of ³H-nicotinic acid by NCM460 cells was: 1) acidic pH (but not Na⁺) dependent; 2) saturable (apparent K_m = 2.5 ± 0.8 μM); 3) inhibited by unlabeled nicotinic acid, nicotinamide, and probenecid; 4) neither affected by other bacterially produced monocarboxylates, monocarboxylate transport inhibitor, or by substrates of the human organic anion transporter-10; 5) affected by modulators of the intracellular protein tyrosine kinase- and Ca²⁺-calmodulin-regulatory pathways; and 6) adaptively regulated by extracellular nicotinate level. Uptake of nicotinic acid by human colonic AMV in vitro and by mouse colonic loops in vivo was also carrier mediated. These findings report, for the first time, that mammalian colonocytes possess a high-affinity carrier-mediated mechanism for nicotinate uptake and show that the process is affected by intracellular and extracellular factors.

the water-soluble vitamin niacin (also known as nicotinic acid, vitamin B3) metabolically functions as part of the coenzymes NAD and NADP. These coenzymes participate in a variety of metabolic reactions that influence cellular redox state (12, 13). Recent studies have also suggested a role for niacin (nicotinamide) in regulating the activity of the mammalian target of rapamycin (mTOR) signaling pathway (which is involved in cell proliferation, protein synthesis, and transcription; 7, 10) and the level of production of intestinal antimicrobial peptides (10); it also appears to be important for maintaining normal intestinal homeostasis and reducing gut inflammation (10, 31). Niacin deficiency leads to pellagra, a disease that is still endemic in many countries due to malnutrition (31). Pellagra is characterized by inflammation of mucous membranes, skin lesions, diarrhea, and dementia; more than 90% of patients with pellagra also develop colitis (29). Deficiency and suboptimal levels of niacin occur in chronic alcoholics, in patients with inflammatory bowel disease (in fact, pellagra has been reported in several patients with Crohn's disease) (1, 9, 23, 26), and in patients with Hartnup's disease (individuals with the latter disease have mutations in the membrane transporter of the amino acid tryptophan, which is the endogenous precursor of niacin) (14). Optimizing niacin body homeostasis, on the other hand, may protect against Alzheimer's disease and against age-related cognitive decline (20). Pharmacological doses of niacin also have lipid-lowering effects (5, 18, 19).

Niacin is obtained from both exogenous and endogenous sources, with the latter source being via the enzymatic conversion of the amino acid tryptophan. As to the exogenous source, obviously diet is a source of niacin, and absorption of this source in the small intestine has been recently characterized in our laboratory and shown to be via a high-affinity and specific carrier-mediated mechanism (21). Another exogenous source for niacin that has been ignored is the bacterially synthesized vitamin by the normal microflora of the large intestine. Although existence of the latter source has been known for quite some time (3, 4, 22, 24), it has been assumed to be of little or no nutritional value due to the belief that the large intestine is incapable of absorbing the vitamin (4). Until recently, the same assumption existed in the case of the other water-soluble vitamins that are also produced by the normal microflora of the large intestine (e.g., biotin, thiamin, folate, riboflavin). Only recently, direct studies using colonic preparations have shown that the latter assumption is incorrect (reviewed in Ref. 27). Several transporters have been suggested as having a potential role in intestinal niacin uptake, but none seem to be involved in the uptake of physiological concentrations of the vitamin as we discussed previously (21). One more system has been recently suggested, i.e., human organic anion transporter-10 (hOAT10), a proton-driven transporter that also transports urate and p-aminohippurate (2). Thus our aim in this investigation was to directly test the ability of mammalian colonocytes to take up niacin and to find out through which mechanism. We used nicotinic acid as a substrate and the well-established and nontransformed human colonic epithelial NCM460 cells, as well as native human colonic apical membrane vesicles obtained from organ donors, and intact mouse colonic loops in vivo in our investigations. The results showed, for the first time, the existence of an efficient, specific, and regulated carrier-mediated mechanism for niacin uptake by human and mouse colonocytes. These findings raise the possibility that the bacterially synthesized niacin in the large intestine may contribute to the overall niacin body homeostasis and especially toward the nutritional needs of the local colonocytes.

MATERIALS AND METHODS

[5,6-³H]-Nicotinic acid (specific activity 50 Ci/mmol; radiochemical purity >97%) was obtained from American Radiolabeled Chemicals (St. Louis, MO). Culture medium (F-12 Ham) and all other cell culture constituents were purchased from Invitrogen (Carlsbad, CA). Other chemicals and reagents used in these investigations were all of analytical grade and were purchased from commercial sources.

Cell culture and uptake studies.

The nontransformed human-derived colonic epithelial NCM460 cells were obtained from INCELL (San Antonio, TX) and were maintained and cultured as described before (15). Cells were trypsinized and plated into 12-well plates (∼2 × 10⁵ cells/well) in Ham's F-12 culture medium supplemented with 20% (vol/vol) fetal bovine serum and 1% antibiotics (vol/vol) and used for uptake investigations 2–3 days postconfluence. Uptake of nicotinic acid was examined in monolayers incubated at 37°C in Krebs-Ringer (KR) buffer (in mM: 133 NaCl, 4.93 KCl, 1.23 MgSO₄, 0.85 CaCl₂, 5 glucose, 5 glutamine, 10 HEPES, and 10 MES, pH 5.5, unless otherwise mentioned). Uptake was terminated after 5 min (unless otherwise stated) by the addition of 1 ml of ice-cold KR buffer, which was followed by rinsing and then digestion with NaOH (1 N). Lysates were then neutralized with HCl, and radioactivity was counted in a liquid scintillation counter (Beckman Coulter, Brea, CA). Protein content of cell lysates was measured using Bio-Rad DC protein assay kit (Bio-Rad, Hercules, CA).

The metabolic form of the taken-up substrate following incubation of the NCM460 cells with ³H-nicotinic acid (6 nM; 5 min) was checked by mean of thin-layer chromatography (TLC, silica-precoated plates) and a solvent system of n-butanol:acetic acid:water (4:4:2, vol/vol).

Human colonic apical membrane vesicles and uptake studies.

The human colonic apical membrane vesicles (AMV) used in this study were kindly supplied by Dr. P. K. Dudeja of the University of Illinois at Chicago and were isolated from the proximal and distal colonic mucosa scrape of human organ donors by an established procedure (17). Use of the human colonic tissue was done in accordance with approved institutional protocols (i.e., by the Institutional Review Board of the University of California). Uptake studies (10 s, 37°C) were performed using a rapid-filtration technique (11). The intravesicular buffer was (in mM) 280 mannitol and 20 HEPES-Tris, pH 7.4; the extravesicular (incubation) buffer was (in mM) 140 mannitol, 100 NaCl, 10 HEPES, and 10 MES (pH 5.5) containing labeled and unlabeled nicotinic acid.

Uptake studies utilizing mouse colonic loops in vivo.

Adult C57BL/6 mice (Jackson Laboratories, Bar Harbor, ME) were used and were anesthetized using ketamine and xylazine. Perfusion of colonic loop (∼1 cm) in vivo was performed (6, 25). Briefly, the loop was filled with 100 μl of KR buffer (pH 5.5) containing ³H-nicotinic acid with or without unlabeled nicotinic acid (1 mM). Uptake was measured after 5 min (linear phase of uptake, data not shown) and expressed in fmol/mg protein per 5 min. Mouse studies performed were approved by the Long Beach VA Subcommittee on Animal Studies.

siRNA analysis.

NCM460 cells (∼85% confluent) were transiently transfected with siRNA [pool of 3 different siRNA duplexes (duplex-1, sense: GUGAAGAACCACACUUUCAtt, antisense: UGAAAGUGUGGUUCUUCACtt; duplex-2, sense: CAAGACCAGCCUUGCUUAUtt, antisense: AUAAGCAAGGCUGGUCUUGtt; duplex-3, sense: CUUGUCCUCUAACAAUCUUtt, antisense: AAGAUUGUUAGAGGACAAGtt)] specific for hOAT10 and control siRNA (scrambled) using Lipofectamine 2000. After 48 h of transient transfection, cells were used to determine the effect of hOAT10 knockdown on ³H-nicotinic acid uptake. Protein and total RNA were also isolated after 48 h of transfection from these cells to determine hOAT10 protein and mRNA expression levels.

Western blot analysis.

Total protein was isolated from NCM460 cells using RIPA buffer (Sigma). Sixty micrograms of proteins were resolved onto premade 4–12% Bis-Tris minigel (Invitrogen) and subjected to Western blot analysis (32). Proteins were electroblotted onto immobilon PVDF (Fisher Scientific, Waltham, MA) after electrophoresis. The blots were blocked in blocking buffer (LI-COR Bioscience, Lincoln, NE) and incubated overnight with hOAT10 (Santa Cruz Biotechnology, Santa Cruz, CA)-specific polyclonal antibodies along with the β-actin monoclonal antibody (Santa Cruz Biotechnology). Specificity of the hOAT10 band was determined using antigenic peptide (Santa Cruz Biotechnology) (data not shown). The immunoreactive bands were detected and quantified using Odyssey software as described before (32).

Quantitative real-time PCR.

Total RNA (5 μg) isolated from NCM460 cell was treated with DNase I and subjected to reverse transcription (RT) using Superscript II kit (Invitrogen). The products were then used for real-time PCR amplification with hOAT10 (forward 5′-CACTGCAGGTCACCCAGA-3′ and reverse 5′-GCCCAGCCATGAACACTG-3′) and human β-actin (forward: 5′-CATCCTGCGTCTGGACCT-3′ and reverse 5′-TAATGTCACGCACGATTTCC-3′) primers. Data were normalized to human β-actin, and Ct values were calculated using the relative relationship method (16).

Data presentation and statistical analysis.

Values are presented as means ± SE of multiple individual uptake determinations and are expressed in fmol/mg protein per unit of time or as a percentage relative to simultaneously performed controls. The Student's t-test was used for statistical analysis, and a P < 0.05 was considered statistically significant. Kinetic parameters of the saturable nicotinic acid uptake process [i.e.; maximal velocity (V_max) and the Michaelis-Menten constant (K_m)] were calculated as described by Wilkinson (33). Saturable uptake of nicotinic acid was calculated by taking out uptake by simple diffusion [determined from the slope of the uptake line between the point of origin and that of uptake at high concentration of nicotinic acid (1 mM)] from total uptake at any given concentration.

RESULTS

Uptake of nicotinic acid by human colonic NCM460 cells: general observations.

Uptake of nicotinic acid (6 nM) by NCM460 confluent monolayers incubated in buffer pH 5.5 was linear up to 10 min and occurred at the rate of 1,586 ± 79 fmol·mg protein⁻¹·5 min⁻¹. Thus an incubation period of 5 min was chosen as the standard uptake time in all other studies (Fig. 1A). The initial rate of uptake of nicotinic acid (6 nM) by NCM460 cells was also temperature dependent, being significantly (P < 0.01) higher at 37°C than at 25°C and 4°C (864 ± 7, 323 ± 15, and 40 ± 2.2 fmol·mg protein⁻¹·5 min⁻¹, respectively). The effect of varying the pH of the incubation buffer on initial rate of nicotinic acid (6 nM) uptake by NCM460 cells was also investigated with the results (Fig. 1B), showing a dramatic increase in uptake as the pH was reduced from 8.0 to 5.0 (139 ± 9, 270.2 ± 16.3, 372 ± 51, and 600 ± 22.2 fmol·mg protein⁻¹·5 min⁻¹, at pH 8, 7, 6, and 5, respectively). A buffer pH of 5.5 was chosen as the standard incubation buffer pH in all other studies. With the initial rate of uptake of nicotinic acid showing acidic pH dependence, we next examined the effect of pretreating the NCM460 cells (for 30 min) with protonophore carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP; 50 μM) on the initial rate of nicotinic acid uptake. The results showed pretreatment of NCM460 monolayers with FCCP to lead to a significant (P < 0.01) decrease in nicotinic acid uptake (807 ± 12 and 237 ± 19 fmol·mg protein⁻¹·5 min⁻¹ for control and FCCP-pretreated cells, respectively). In yet another study, we examined the role of extracellular Na⁺ in nicotinic acid uptake by NCM460 cells. Here Na⁺ was replaced in the incubation buffer with an isoosmotic amount of K⁺, Li⁺, or mannitol, and the initial rate of nicotinic acid (6 nM) uptake was examined. The results showed a similar nicotinic acid uptake under all these ionic conditions (603 ± 25, 599 ± 31, 574 ± 26, and 575 ± 23 fmol·mg protein⁻¹·5 min⁻¹ in the presence of Na, K⁺, Li⁺, and mannitol, respectively). Pretreatment of NCM460 monolayers with the Na⁺-K⁺-ATPase inhibitor, ouabain (1 mM; 30 min), also did not significantly affect the initial rate of nicotinic acid (6 nM) uptake by these cells (713 ± 13 and 680 ± 22 fmol·mg protein⁻¹·5 min⁻¹ for control and ouabain-pretreated cells, respectively). Finally, the effect of pretreatment (30 min) with the metabolic inhibitor iodoacetate (1 mM) on initial rate of nicotinic acid (6 nM) uptake was examined, and a significant (P < 0.01) inhibition was observed (901 ± 37 and 473 ± 55 fmol·mg protein⁻¹·5 min⁻¹ for control and in cell pretreated with iodoacetate, respectively).

Fig. 1.

A: uptake of ³H-nicotinic acid by NCM460 cells as a function of time. NCM460 monolayers were incubated at 37°C in Krebs Ringer (KR) buffer (pH 5.5) for different time points with ³H-nicotinic acid (6 nM). Each data point represents the mean ± SE of at least 3 separate uptake determinations. B: effect of incubation buffer pH on ³H-nicotinic acid uptake by NCM460 cells. NCM460 monolayers were incubated at 37°C in KR buffer of varying pH, and ³H-nicotinic acid (6 nM) was added to the incubation medium at the onset of 5-min incubation (i.e., initial rate). Data are means ± SE of at least 3 separate uptake determinations.

The metabolic form of the transported substrate after incubating NCM460 monolayers with ³H-nicotinic acid was examined by mean of TLC (see materials and methods). Little metabolism appeared to occur in the transported substrate after 5-min incubation, as more than 96% of the substrate taken up by the cells was found to be in the form of intact nicotinic acid.

Evidence for existence of a specialized carrier-mediated system for nicotinic acid uptake by colonocytes: studies with cultured NCM460 cells, purified human native colonic AMV from organ donors, and intact mouse colon loops in vivo.

The initial rate of nicotinic acid uptake by confluent NCM460 monolayers as a function of concentration was examined and found to include a saturable component (Fig. 2). Kinetic parameters of the saturable component (calculated as described in materials and methods) were 2.5 ± 0.8 μM and 27.4 ± 3.74 pmol·mg protein⁻¹·5 min⁻¹ for the apparent K_m and V_max, respectively.

Fig. 2.

Uptake of ³H-nicotinic acid by NCM460 cells as a function of substrate concentration. NCM460 monolayers were incubated for 5 min at 37°C in KR buffer (pH 5.5) in the presence of different concentrations (0.1–10.0 μM) of labeled and unlabeled nicotinic acid. Data are means ± SE of at least 3 separate uptake determinations.

We also examined the effect of unlabeled nicotinic acid and that of its structural analogs nicotinamide, isonicotinic acid, isonicotinic acid hydrazide, and nicotinyl alcohol, as well as 5-methyl-¹H-pyrazole-3-carboxylic acid (the latter compound is a specific ligand for the nicotinic acid receptor, HM74A; 34) on initial rate of ³H-nicotinic acid uptake (6 nM) by NCM460 monolayers. The results showed a significant (P < 0.01) inhibition in ³H-nicotinic acid uptake by unlabeled nicotinic acid and nicotinamide but not by the other tested compounds (Table 1). The above-described findings of saturability in the uptake of ³H-nicotinic acid as a function of concentration and the inhibition caused by unlabeled nicotinic acid and its close analog nicotinamide clearly indicate the involvement of a carrier-mediated mechanism for nicotinic acid uptake by NCM460 colonic epithelial cells.

Table 1.

Effect of unlabeled nicotinic acid and structural analogs on the initial rate of ³H-nicotinic acid (6 nM) uptake by NCM460 cells

Compound	Concentration, μM	Uptake, % of Control	P Value
Control		100 + 8
Nicotinic Acid	50	30 ± 4	<0.01
Nicotinamide	50	77 ± 8	<0.01
Isonicotinic Acid	50	100 ± 3	NS
Isonicotinic Acid Hydrazide	50	105 ± 5	NS
Nicotinyl Alcohol	50	85 ± 7	NS
5-Methyl-1H-pyrazole-3-carboxylic acid	50	101 ± 3	NS

Data are means ± SE of at least 3 separate uptake determinations. NCM460 monolayers were incubated at 37°C in Krebs Ringer buffer (pH 5.5) in the presence of the structural analogs and ³H-nicotinic acid (6 nM). Uptake was determined after 5 min of incubation. NS, not significant.

To determine whether a carrier-mediated uptake mechanism is also involved in nicotinic acid uptake by human native human colon, we used AMV isolated from the proximal and distal colon of human organ donors and examined the effect of unlabeled nicotinic acid (1 mM) on initial rate of uptake of ³H-nicotinic acid (0.12 μM). The results showed a significant (P < 0.01 for both) inhibition in ³H-nicotinic acid uptake by proximal and distal colonic AMV in the presence of unlabeled nicotinic acid, suggesting the existence of a carrier-mediated process for nicotinic acid uptake in human native colon (Fig. 3).

Fig. 3.

Effect of unlabeled nicotinic acid on the initial rate of ³H-nicotinic acid uptake by purified apical membrane vesicles (AMV) isolated from proximal (A) and distal (B) colon of human organ donors. ³H-nicotinic acid (0.12 μM) uptake was performed at 37°C for 10 s as described in materials and methods. Data are means ± SE of at least 3 separate uptake determinations. *P < 0.01.

Both of the above studies were performed in vitro, and thus we also examined the effect of unlabeled nicotinic acid (1 mM) on ³H-nicotinic acid (0.1 μM) uptake by intact mouse colonic loop in vivo (see materials and methods). The results again showed a significant (P < 0.01) inhibition in ³H-nicotinic acid uptake by unlabeled nicotinic acid (Fig. 4).

Fig. 4.

Effect of unlabeled nicotinic acid on the initial rate of ³H-nicotinic acid uptake by native mouse colonic loop. ³H-nicotinic acid (0.1 μM) uptake was performed for 5 min in intact mouse colonic loop as described in materials and methods. Data are means ± SE of at least 5 different mice. *P < 0.01.

Effect of probenecid, carboxylic acids, and a monocarboxylic acid transporter inhibitor on ³H-nicotinic acid uptake by NCM460 cells.

This study examined the effect of the anion transport inhibitor probenecid (1 mM) on the initial rate of uptake of 6 nM ³H-nicotinic acid (nicotinic acid itself is an anion with a pKa of 4.9 by NCM460 monolayers). The results showed probenecid to cause a significant (P < 0.01) inhibition in nicotinic acid uptake (561 ± 41.2 and 384 ± 33 fmol·mg protein⁻¹·5 min⁻¹ for control and probenecid, respectively). We also examined the effect of selected carboxylic acids (including monocarboxylates that are also produced by the normal microflora of the large intestine) on the initial rate of ³H-nicotinic acid (6 nM) uptake. The selected compounds utilize different monocarboxylic acid transporter (MCT) systems for their own uptake; also a role for MCTs in nicotinic acid uptake has been suggested (8, 30). The results, however, showed that none of the tested compounds significantly affect ³H-nicotinic acid uptake (Table 2). Similarly, the MCT inhibitor α-cyano-4-hydroxycinnamate (50 μM) failed to affect the initial rate of ³H-nicotinic acid uptake (Table 2).

Table 2.

Effect of unrelated carboxylic acids, a monocarboxylic acid transporter inhibitor, urate, and p-aminohippurate on the initial rate of ³H-nicotinic acid uptake by NCM460 cells

Condition	Concentration, μM	Uptake, % of Control	P Value
Control		100 ± 4
Acetate	50	104 ± 19	NS
Butyrate	50	92 ± 12	NS
Propionate	50	91 ± 4	NS
Oxalate	50	110 ± 3	NS
Succinate	50	94 ± 4	NS
Valproate	50	101 ± 6	NS
Pyruvate	50	100 ± 6	NS
Lactate	50	99 ± 8	NS
α-cyano-4-hydroxycinnamate	50	108 ± 11	NS
Urate	1,000	100 ± 9	NS
p-aminohippurate	1,000	100 ± 10	NS

Values are means ± SE of at least 3 separate uptake determinations. NCM460 monolayers were treated as stated in Table 1. Different carboxylic acid compounds were added at the time of incubation.

Finally, the effect of urate and p-aminohippurate (both at high concentration of 1 mM), substrates of hOAT10, on the initial rate of ³H-nicotinic acid (6 nM) uptake by NCM460 monolayers was examined. This was done, as a role for hOAT10 in the pH-dependent nicotinic acid uptake process has been suggested recently (2). The results, however, did not show any effect by either compound on nicotinic acid uptake (Table 2). We also used the molecular approach of siRNA knockdown to examine a possible role of hOAT10 in colonic uptake of nicotinic acid. As can be seen in Fig. 5A, hOAT10 siRNA did not affect the initial rate of carrier-mediated ³H- nicotinic acid (6 nM) uptake, despite significant reduction in hOAT10 protein and mRNA levels of expression (P < 0.01 and P < 0.02, respectively) in hOAT10 siRNA-treated NCM460 cells compared with scrambled siRNA (control)-treated NCM460 cells (Fig. 5, B and C).

Fig. 5.

Effect of human organic anion transporter-10 (hOAT10) siRNA on the initial rate of ³H-nicotinic acid uptake by NCM460 cells. A: NCM460 cells were transfected with hOAT10-specific siRNA or with the negative (scrambled) control siRNA. 48 h later, the initial rate of ³H-nicotinic acid (6 nM) uptake was then measured for 5 min at 37°C. Data are means ± SE of at least 4 separate uptake determinations. B: total proteins (60 μg) isolated from NCM460 cells were separated on a 4–12% mini gel and electroblotted to PVDF membrane and probed with specific anti-hOAT10. With the use of an LI-COR detection system, the immunoreactive bands were detected, and we determined the specific band intensity. Data are means ± SE from at least 3 independent sample preparations from different batches of cells transfected with siRNA. *P < 0.01. C: quantitative real-time PCR was performed using total RNA isolated from NCM460 cells transfected with hOAT10 siRNA, scrambled siRNA, and gene-specific primers. Data are from 3 different experiments and expressed relative to β-actin as means ± SE. **P < 0.02.

Regulation of nicotinic acid uptake process by colonocytes.

We examined (using specific modulators) the possible roles of selective intracellular regulatory pathways in the regulation of colonic epithelial nicotinic acid uptake using NCM460 cells as model. Thus the effect of pretreating (1 h) the cells with inhibitors of the protein-tyrosine kinase (PTK)-mediated pathway, i.e., genistin and tyrophostin A25, on initial rate of nicotinic acid (6 nM) uptake was examined, and the results were compared with the effect of their negative controls, i.e., genistin and tyrophostin A1, respectively (all compounds were used at 100 μM). The results showed that genistin and tyrophostin A25 pretreatment led to a significant (P < 0.01 for both) inhibition in the initial rate of nicotinic acid uptake, whereas their negative controls were without an effect (631 ± 11, 178 ± 7, 267 ± 27, 603 ± 31, and 599 ± 19 fmol·mg protein⁻¹·5 min⁻¹ for control and cells pretreated with genistin, tyrophostin A25, genistin, and tyrophostin A1, respectively). Similarly, pretreating NCM460 cells (1 h) with modulators for the Ca²⁺/calmodulin-mediated pathway, i.e., trifluoperazine and calmidazolium (both at 50 μM), led to a significant (P < 0.01 for both) inhibition in initial rate of nicotinic acid (6 nM) uptake (744 ± 27, 561 ± 19.5 and 135 ± 3.3 fmol·mg protein⁻¹·5 min⁻¹ for control and cells pretreated with trifluoperazine and calmidazolium, respectively). In contrast, pretreating (1 h) of NCM460 cells with modulators of the PKA-mediated pathway [dibutyryl cAMP (1 mM) and the PKA inhibitor H-89 (50 μM)] was without an effect on the initial rate of uptake of nicotinic acid (6 nM) uptake (653 ± 26.7, 688 ± 17.92, 621 ± 8.2 fmol·mg protein⁻¹·5 min⁻¹ for control, dibutyryl cAMP, and H-89-pretreated cells, respectively). Similarly, pretreatment (1 h) of NCM460 cells with modulators of the PKC-mediated pathway [the PKC activator phorbol 12-myristate 13-acetate (PMA; 10 μM), and the PKC inhibitors staurosporine and chelerythrine (both at 1 μM)] showed no effect (724 ± 24.4, 743 ± 39.8, 700 ± 47.3, 734 ± 32.6 fmol·mg protein⁻¹·5 min⁻¹ for control and cells pretreated with PMA, staurosporine, and chelerythrine, respectively).

In a separate study, we examined possible regulation of the nicotinic acid uptake process of human colonic NCM460 cells by extracellular nicotinate levels. For this, cells were maintained (for 7 days) in a regular growth medium (which contains about 1.2 μM nicotinate) and in the same growth medium but in the presence of 10 mM nicotinate. Initial rate of ³H-nicotinic acid (6 nM) uptake was then measured with the results showing a significantly (P < 0.01) lower nicotinic acid uptake in cells maintained in the presence of a high level of nicotinate in the growth medium compared with controls (Fig. 6). This effect appears to be specific for nicotinic acid, as uptake of the unrelated biotin (5 nM) was similar under the two conditions (18.9 ± 0.8 and 18.8 ± 1.1 fmol·mg protein⁻¹·5 min⁻¹, for control and 10 mM nicotinic acid, respectively). In these studies, we also determined possible changes in the levels of expression of the hOAT10 protein and RNA (hOAT10 has been recently suggested as being involved in pH-dependent uptake of nicotinic acid; 2) but found no change in expression level (Fig. 6).

Fig. 6.

Effect of maintaining NCM460 cells in growth media in the presence and absence of different levels of nicotinic acid (low and high) on initial rate of ³H-nicotinic acid uptake. A: NCM460 cells were maintained for 7 days in a regular growth medium (which contains around 1.2 μM nicotinate) and in the same medium but in the presence of 10 mM nicotinate. Initial rate of ³H-nicotinic acid (6 nM) uptake was then performed for 5 min. Data are means ± SE of at least 5 separate uptake determinations. *P < 0.01. B: total proteins (60 μg) isolated from NCM460 cells were separated on a 4–12% mini gel, and the detection was performed as mentioned in Fig. legend 5. Data are means ± SE from at least 4 independent sample preparations from different batches of cells. C: quantitative real-time PCR was performed using total RNA isolated from NCM460 cells and gene-specific primers. Data are from 3 different experiments and expressed relative to β-actin as means ± SE.

DISCUSSION

The aim of this study was to determine whether mammalian colonic epithelial cells are capable of absorbing luminally presented nicotinic acid and, if so, through which mechanism. Nicotinic acid, like many other water-soluble vitamins (e.g., folate, biotin, thiamin, riboflavin), is produced by the normal microflora of the large intestine (3, 4, 22, 24). This source of niacin, however, has been considered to be of little or no nutritional value, as it is assumed to be lost in the feces due to the inability of the large intestine to take it in. Because the same situation existed in the case of other water-soluble vitamins that are produced by the normal microflora, only to be corrected recently by demonstrating the existence of highly efficient uptake systems for these micronutrients in colonocytes (reviewed in Ref. 27), we hypothesized that a similar situation may exist in the case of niacin. To test this possibility, we used a well-established and characterized human colonic epithelial NCM460 cell, native human colonic AMV preparations isolated from human organ donors, and intact mouse colonic loops in vivo. The results obtained provided clear evidence for the existence of a specific and high-affinity carrier-mediated system for uptake of luminal nicotinic acid by all of these colonic preparations.

Our studies with the human colonic NCM460 cells showed the initial rate of physiological concentrations of nicotinic acid uptake to be acidic pH (but not Na) dependent and energy/temperature dependent. This event appears to involve a carrier-mediated system as indicated by the saturation in the initial rate of uptake of nicotinic acid as a function of concentration and by the significant cis inhibition caused by unlabeled nicotinic acid and nicotinamide. Similarly, uptake of the vitamin by native human colonic preparations appeared to be carrier mediated, as indicated by the significant inhibition in ³H-nicotinic acid uptake by unlabeled nicotinic acid in colonic AMV of organ donors. The latter inhibition was observed in AMV prepared from both the proximal and the distal portions of the human colon, suggesting distribution of the nicotinic acid carrier-mediated uptake process along the colon. Finally, in vivo uptake investigations utilizing intact mouse colon segments showed the involvement of a carrier-mediated mechanism for nicotinic acid uptake. Collectively, these in vitro and in vivo findings clearly demonstrate the existence of a functional uptake mechanism for nicotinic acid in the mammalian colon. This raises the possibility that the bacterially synthesized nicotinic acid in the large intestine is bioavailable and may contribute to overall nicotinic acid nutrition of the host, and especially toward the cellular nutrition and health of the local colonocytes. In support of the latter suggestion is the previous reporting of colitis in more than 90% of patients with pellagra (29), a disease that is caused by niacin deficiency. Furthermore, a recent study has shown that niacin supplementation can alleviate the severe colitis induced in mice deficient in the angiotensin I converting enzyme 2 (10).

The nicotinic acid uptake process of NCM460 cells appears to be specific for nicotinic acid and nicotinamide, as none of the other structural analogs tested affected the initial rate of ³H-nicotinic acid uptake. Similarly, neither of the prototypical ligands of the MCTs (e.g., lactate, pyruvate, butyrate, propionate) nor the MCT inhibitor α-cyano-4-hydroxycinnamate affected nicotinic acid uptake. The latter findings suggest that the MCT systems are not involved in the uptake of physiologically relevant concentrations of nicotinic acid by colonic epithelial cells. Recently, a role for hOAT10 in the acidic pH-dependent nicotinic acid uptake has been suggested (2). This transporter is found to be mainly expressed in the kidneys, but some expression was also reported in the colon (2). However, our results do not seem to support a role for this transporter in colonic uptake of nicotinic acid. First, two substrates that utilize hOAT10, namely urate and p-aminohippurate, failed to affect nicotinic acid uptake (even when used at a high concentration of 1 mM). Second, whereas colonic uptake of nicotinic acid appears to be adaptively regulated by substrate level in the growth medium (with a significantly higher uptake by cells maintained in the presence of low nicotinic acid compared with those maintained in the presence of high concentration), no changes in the level of expression of hOAT10 protein and mRNA was observed under these conditions. Finally, selective knockdown of the hOAT10 with the use of gene-specific siRNA (verified by a clear reduction in the level of hOAT10 expression) failed to significantly affect nicotinic acid uptake by NCM460 cells. Further studies are needed to establish the molecular identity of the nicotinic acid uptake carrier of colonocytes.

We also obtained some evidence that the colonic nicotinic acid uptake process is under the influence of intracellular and extracellular factors/conditions. Thus, although no role for the intracellular PKA- and PKC-mediated pathways were evident, a role for the intracellular PTK- and Ca/calmodulin-mediated pathways was observed. Although specific modulators of these pathways were used (and when possible negative controls employed), further studies at the molecular level (when identity the colonic nicotinic uptake system is determined) are needed. The colonic nicotinic acid uptake process was also found to be adaptively regulated by the extracellular substrate level with higher uptake by cells maintained in the presence of low compared with high nicotinic acid concentrations. Similar adaptive regulation has been seen in the case of absorption of other water-soluble vitamins (reviewed in Refs. 27 and 28) and is attributed to the essentiality of these micronutrients for cell function and survival. Again, at which level (i.e., transcriptional and/or posttranscriptional) the adaptive response in colonic nicotinic acid uptake process by extracellular vitamin level is occurring is not clear and in need of further investigations when molecular identity of the system is determined.

In summary, our studies demonstrate for the first time that the human and mouse colonic epithelial cells possess an efficient, specific, and regulated nicotinic acid uptake mechanism. This mechanism may be involved in the absorption of the vitamin that is synthesized by the normal microflora of the large intestine and thus may contribute toward host nutrition and especially that of the local colonocytes.

GRANTS

The study was supported by grants from the Department of Veterans Affairs and NIH (DK-56061, DK-58057).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

Author contributions: J.S.K., V.S.S., M.L.K., and H.M.S. conception and design of research; J.S.K., V.S.S., and R.K. performed experiments; J.S.K., V.S.S., R.K., M.L.K., and H.M.S. analyzed data; J.S.K., V.S.S., R.K., M.L.K., and H.M.S. interpreted results of experiments; J.S.K. and V.S.S. prepared figures; J.S.K., V.S.S., and H.M.S. drafted manuscript; J.S.K., V.S.S., M.L.K., and H.M.S. edited and revised manuscript; J.S.K., V.S.S., R.K., M.L.K., and H.M.S. approved final version of manuscript.