Oxalobacter formigenes–Derived Bioactive Factors Stimulate Oxalate Transport by Intestinal Epithelial Cells

Abstract

Hyperoxaluria is a major risk factor for kidney stones and has no specific therapy, although Oxalobacter formigenes colonization is associated with reduced stone risk. O. formigenes interacts with colonic epithelium and induces colonic oxalate secretion, thereby reducing urinary oxalate excretion, via an unknown secretagogue. The difficulties in sustaining O. formigenes colonization underscore the need to identify the derived factors inducing colonic oxalate secretion. We therefore evaluated the effects of O. formigenes culture conditioned medium (CM) on apical ¹⁴C-oxalate uptake by human intestinal Caco-2-BBE cells. Compared with control medium, O. formigenes CM significantly stimulated oxalate uptake (>2.4-fold), whereas CM from Lactobacillus acidophilus did not. Treating the O. formigenes CM with heat or pepsin completely abolished this bioactivity, and selective ultrafiltration of the CM revealed that the O. formigenes–derived factors have molecular masses of 10–30 kDa. Treatment with the protein kinase A inhibitor H89 or the anion exchange inhibitor 4,4'-diisothiocyano-2,2'-stilbenedisulfonic acid completely blocked the CM-induced oxalate transport. Knockdown of the oxalate transporter SLC26A6 also significantly restricted the induction of oxalate transport by CM. In a mouse model of primary hyperoxaluria type 1, rectal administration of O. formigenes CM significantly reduced (>32.5%) urinary oxalate excretion and stimulated (>42%) distal colonic oxalate secretion. We conclude that O. formigenes–derived bioactive factors stimulate oxalate transport in intestinal cells through mechanisms including PKA activation. The reduction in urinary oxalate excretion in hyperoxaluric mice treated with O. formigenes CM reflects the in vivo retention of biologic activity and the therapeutic potential of these factors.

Nephrolithiasis is the second most prevalent kidney disease in the United States after hypertension, with a rising prevalence and complications including CKD and ESRD.¹ Hyperoxaluria is a major risk factor for kidney stones, and 70%–80% of kidney stones are composed of calcium oxalate.² Urinary oxalate is an important determinant of supersaturation, and the risk of stone formation is affected by small increases in urine oxalate.^3,4 Mild-to-moderate hyperoxaluria is found in patients with diabetes and obesity and contributes to their increased risk of stone formation.^5–9 More severe hyperoxaluria is seen in enteric hyperoxaluria (EH), most commonly observed in inflammatory bowel disease patients and after small-bowel resection and bariatric surgery for obesity.^10–12 Primary hyperoxaluria (PH) is an inherited disease in which there is endogenous oxalate overproduction.¹³ Both EH and PH lead to recurrent calcium oxalate kidney stones (COKS), nephrocalcinosis, and ESRD.

The mammalian intestine plays a crucial role in oxalate homeostasis, by regulating the amount of absorbed dietary oxalate and providing an avenue for enteric oxalate excretion.¹⁴ Anion exchanger SLC26A6 (A6)-mediated intestinal oxalate secretion plays a critical role in preventing hyperoxaluria and COKS.^15,16 Oxalobacter formigenes is an anaerobic bacterium and utilizes oxalate as its exclusive energy source.¹⁷ O. formigenes colonization correlates with reduced risk of COKS formation in a number of studies, presumably by reducing intestinal oxalate absorption and urinary oxalate excretion.^18–21

In addition to degrading dietary oxalate, O. formigenes also interacts with colonic epithelium by inducing distal colonic oxalate secretion, leading to reduced urinary excretion via a potential secretagogue.²² This effect was demonstrated in rodent colon and whether similar regulation is also active in human colon remains to be determined. O. formigenes colonization of PH1 mice significantly reduced serum and urinary oxalate levels due to induction of colonic oxalate secretion.²³ However, all PH1 mice lost colonization within 18 days when switched from a high oxalate/low calcium diet (1.5% oxalate/0.5% calcium) to regular mouse chow (0.25% oxalate/1% calcium).²³ In addition, colonization cannot be maintained without reducing dietary calcium.^22,24 Moreover, it has been suggested from studies in PH patients and PH1 mice that the intraluminal environment in PH is not supportive of sustained O. formigenes colonization.^23,25 Collectively, maintaining O. formigenes colonization in the absence of high exogenous oxalate remains problematic, underscoring the need for identifying O. formigenes–derived bioactive factors exerting effects similar to live O. formigenes.

We therefore examined whether O. formigenes conditioned medium (CM) modulates intestinal oxalate transport by testing its effects on oxalate uptake by Caco2-BBE (C2) cells. We find that O. formigenes CM significantly stimulates oxalate uptake by C2 cells through PKA activation. small interfering RNA (siRNA) knockdown studies show that a significant component of the observed stimulation is A6-mediated. Importantly, O. formigenes CM also significantly reduces urinary oxalate excretion in PH1 mice by stimulating distal colonic oxalate secretion, reflecting its translational relevance. Of note is that A6 (mediates ≥50% of apical oxalate uptake by C2 cells²⁶) operates in the direction of exchanging intracellular oxalate for mucosal Cl during the process of transepithelial intestinal oxalate secretion. However, A6 can operate in either direction,²⁷ and we therefore measured its activity by the more convenient assay of cellular oxalate uptake.

Results

We previously established C2 cells as a model to study regulation of intestinal oxalate transport by extracellular nucleotides.²⁸ We have now used C2 cells to examine whether intestinal oxalate transport is regulated by O. formigenes CM. As shown in Figure 1A, compared with untreated (UT) cells and cells treated with O. formigenes growth medium (OM), the CM significantly stimulated (>2.4-fold) oxalate uptake. Similar effects were also observed with 6-hour and 16-hour incubations, but no effect was seen at 1 hour (data not shown). The pH of the medium is not affected by OM or CM. Importantly, OM or CM has no effect on the transepithelial resistance (TER) (mean TER in Ω cm², before and after treatment, respectively: UT=434.17±19.49 and 443.83±20.09; OM=417.45±7.24 and 430.11±11.74; CM=436.25±11.26 and 438.57±13.10), indicating that OM or CM did not affect the paracellular permeability. These results indicate that potential secreted factors are responsible for the observed stimulation, likely by modulating the activity of the involved transporter(s).

Figure 1.

O. formigenes CM stimulates oxalate uptake by C2 cells. (A) C2 cells were UT or were treated apically with OM or CM in the culture medium (1:50 dilution × 24 hours), and then ¹⁴C-oxalate uptake was measured as described in the Concise Methods. Values are means±SEM of 12 independent experiments each of which was done in triplicate. The CM significantly stimulated ¹⁴C-oxalate uptake (*P<0.001 for CM compared with UT and OM, by ANOVA). (B) C2 cells were UT or were treated apically with L. acidophilus growth medium (MRS) or CM in the culture medium (1:25 dilution × 24 hours), and then ¹⁴C-oxalate uptake was measured as described in the Concise Methods. Values are means±SEM of three independent experiments each of which was done in triplicate. L. acidophilus CM had no significant effect on ¹⁴C-oxalate uptake by C2 cells.

A crucial question is, how specific is the observed stimulation? To address specificity, we similarly evaluated the effect of Lactobacillus acidophilus (also degrades intraluminal oxalate,²⁹ but it is unknown if it interacts with intestinal cells and modulates oxalate transport) CM. Compared with UT and cells treated with the control medium (MRS), L. acidophilus CM had no effect on oxalate uptake by C2 cells (Figure 1B). These results indicate that O. formigenes CM-mediated stimulation of oxalate transport is specific and is possibly mediated by one or more O. formigenes–derived factor(s).

To ensure that the CM effects on intestinal oxalate transport are not cell-line specific, we evaluated the CM effects on oxalate uptake by the human colonic cell line T84. C2 and T84 cells are considered models for absorptive epithelial cells and secretory crypt cells, respectively. We previously demonstrated that A6 mediates most of oxalate transport in T84 cells.³⁰ Compared with UT and OM, the CM significantly stimulated (>2.6-fold) oxalate uptake by T84 cells (Figure 2). These results indicate that the CM effects on oxalate transport by intestinal epithelial cells are not cell-line specific.

Figure 2.

O. formigenes CM stimulates oxalate uptake by T84 cells. T84 cells were UT or were treated apically with OM or CM in the culture medium (1:50 dilution × 24 hours), and then ¹⁴C-oxalate uptake was measured as described in the Concise Methods. Values are means±SEM of three independent experiments each of which was done in triplicate. The CM significantly stimulated ¹⁴C-oxalate uptake (*P<0.05 and 0.01 for CM compared with UT and OM, respectively, by ANOVA).

To test the possibility that the O. formigenes–derived bioactive factor(s) might be proteinaceous in nature, the CM was subjected to heat treatment. Compared with UT, the CM significantly stimulated (>1.9-fold) oxalate uptake by C2, an effect completely abolished by heat treatment (heated CM) (Figure 3A). These results indicate that the O. formigenes–derived factor(s) is/are likely to be protein(s) and/or peptide(s). To provide more evidence that the O. formigenes–derived bioactive factor(s) is/are protein(s) and/or peptide(s), the CM was treated with the protease pepsin. Compared with UT, the CM significantly stimulated (>4-fold) oxalate uptake by C2, an effect completely abolished by pepsin treatment (CM+Pepsin) (Figure 3B). Treatment of the CM with another protease, trypsin, also completely abolished its bioactivity (data not shown). These results provide further evidence that the O. formigenes–derived factor(s) is/are protein(s) and/or peptide(s).

Figure 3.

Heat and pepsin treatment abolish the CM bioactivity. (A) C2 cells were UT or were treated apically with CM or a CM that was subjected to heat treatment (heated CM) in the culture medium (1:50 dilution × 24 hours), and then ¹⁴C-oxalate uptake was measured as described in the Concise Methods. Values are means±SEM of four independent experiments each of which was done in triplicate. The CM significantly stimulated ¹⁴C-oxalate uptake, an effect completely abolished by heat treatment (*P<0.01 and <0.05 for CM compared with UT and heated CM, respectively, by ANOVA). (B) C2 cells were UT or were treated apically with CM or a CM that was treated with pepsin (CM+Pepsin) in the culture medium (1:50 dilution × 24 hours), and then ¹⁴C-oxalate uptake was measured as described in the Concise Methods. Values are means±SEM of four independent experiments each of which was done in triplicate. The CM significantly stimulated ¹⁴C-oxalate uptake, an effect completely abolished by pepsin treatment (* P<0.01 for CM compared with UT and CM+Pepsin, by ANOVA).

To determine the molecular mass(es) (MM) of the factor(s), the CM was subjected to selective ultrafiltration using 10 and 30 kDa cutoff spin columns. C2 cells were UT or treated with the CM, filtrate (F), retentate (R), or F+R. Using a 10 kDa column, the R and the F+R significantly stimulated oxalate uptake by C2 cells, whereas the F had no effect (Figure 4A). Using a 30 kDa column, the F and the F+R significantly stimulated oxalate uptake by C2 cells, whereas the R had no effect (Figure 4B). Collectively, these results indicate that the MMs of the factors are largely between 10 and 30 kDa. Because F+R had a better stimulatory effect than F, whereas R had no effect, such data suggest the possibility that these factors might exist as a multifunctional complex requiring a bacterial product of >30 kDa for optimal functioning.

Figure 4.

Selective ultrafiltration shows the O. formigenes–derived factors to have molecular masses of 10–30 kDa. (A) Selective ultrafiltration of the CM using 10 kDa cutoff spin column. C2 cells were UT or were treated with CM, F, R, or the combined fractions (F+R) in the culture medium (1:50 dilution × 24 hours), and then ¹⁴C-oxalate uptake was measured as described in the Concise Methods. Values are means±SEM of five independent experiments each of which was done in duplicate or triplicate. The CM, the R, and the F+R significantly stimulated ¹⁴C-oxalate uptake (*P<0.001, <0.001, and <0.01 for CM compared with UT, F, and R; **P<0.01 for R compared with UT and F; and ***P<0.001, <0.001, and <0.01 for F+R compared with UT, F, and R, respectively, by ANOVA). (B) Selective ultrafiltration of the CM using 30 kDa cutoff spin column. C2 cells were UT or were treated with CM, F, R, or F+R in the culture medium (1:50 dilution × 24 hours), and then ¹⁴C-oxalate uptake was measured as described in the Concise Methods. Values are means±SEM of eight independent experiments each of which was done in triplicate. The CM, the F, and the F+R significantly stimulated ¹⁴C-oxalate uptake (*P<0.001 and <0.01 for CM compared with UT and R, respectively; ** P<0.05 for F compared with UT and R; and ***P<0.001 and <0.05 for F+R compared with UT and R, respectively, by ANOVA).

O. formigenes was suggested to utilize a potential secretagogue as a survival strategy when dietary oxalate is limited.²² To test the hypothesis that lower growth medium oxalate concentration might lead to more secreted factor(s) and therefore to a CM with higher bioactivity, oxalate concentration was reduced from 37.5 mM to 18.8 and 9.4 mM, and conditioned media under these conditions were simultaneously prepared (CM37.5, CM18.8, and CM9.4, respectively). C2 cells were UT or treated with CM37.5, CM18.4, and CM9.4. Interestingly, the bioactivity of the CM was significantly higher (>2-fold) with CM9.4 compared with CM37.5 (Figure 5). These results strongly indicate that the secretion of these factors is inducible, which will significantly facilitate their characterization.

Figure 5.

Lowering O. formigenes growth medium oxalate concentration leads to a CM with higher bioactivity. C2 cells were UT or treated with CM37.5, CM18.8, and CM9.4 (conditioned media prepared with growth medium oxalate concentration of 37.5 mM, 18.8 mM, and 9.4 mM, respectively) in the culture medium (1:50 dilution × 24 hours), and then ¹⁴C-oxalate uptake was measured as described in the Concise Methods. Values are means±SEM of three independent experiments each of which was done in triplicate. CM9.4 had significantly higher bioactivity compared with CM37.5 (*P<0.01 and <0.05 for CM9.4 compared with UT and CM37.5, respectively, by ANOVA).

Probiotic-derived factors regulate intestinal epithelial homeostasis by modulating specific signaling pathways.^31,32 To elucidate the involved signaling pathway(s), we tested the effects of several pharmacologic inhibitors. Preincubation of C2 cells with the PKA inhibitor H89³³ before incubation with the CM (CM+H89) completely blocked the CM-induced stimulation, whereas it had no effect on baseline transport (H89) (Figure 6). We previously found PKC activation inhibits intestinal oxalate transport through a Gö6983-sensitive pathway^28,30,34; however, this inhibitor had no effect on the observed stimulation (pmol/cm² per minute: UT=3.12±0.32; CM=10.54±0.43; CM+Gö6983=9.64±0.67; Gö6983=3.60±0.77). On the other hand, Gö6976, U0126, SB 202190, LY 294002, genistein, and PP2 (described in the Concise Methods) had no significant effect on the observed stimulation (data not shown). Taken together, these results indicate that the CM stimulates oxalate transport in C2 cells by activating PKA, whereas PKC, MAPKs, and Src kinases are unlikely to be involved.

Figure 6.

The PKA inhibitor H89 completely blocked the CM-induced oxalate transport. C2 cells were UT or were treated apically with CM in the culture medium (1:50 dilution × 24 hours), and then ¹⁴C-oxalate uptake was measured as described in the Concise Methods. C2 cells were also treated apically with H89 (20 µM) for 30 minutes followed by the CM (1:50 dilution × 24 hours) with continued presence of H89 (CM+H89), or with H89 (20 µM) alone for 24.5 hours, and then ¹⁴C-oxalate uptake was similarly measured. Values are means±SEM of seven independent experiments each of which was done in triplicate. The CM significantly stimulated ¹⁴C-oxalate uptake, an effect completely blocked by H89 (*P<0.001 for CM compared with UT, CM+H89, and H89, by ANOVA).

To assess whether the CM-induced stimulation of oxalate uptake by C2 cells is due to an active transcellular transport process, the effects of the anion exchange inhibitor 4,4^׳diisothiocyanostilbene-2,2^׳-disulfonic acid (DIDS) were tested on the observed stimulation. Compared with UT cells, the CM significantly stimulated (>3.4-fold) oxalate uptake by C2, an effect completely abolished by DIDS (100 µM) treatment (CM+DIDS) (Figure 7). These results indicate that the observed stimulation is mediated by DIDS-sensitive transporters such as A6 and/or SLC26A2 (A2).^26,30,35,36

Figure 7.

The CM-induced oxalate transport is DIDS-sensitive. C2 cells were UT or were treated apically with CM in the culture medium (1:50 dilution × 24 hours), and then ¹⁴C-oxalate uptake was measured as described in the Concise Methods. CM-treated cells were treated apically with DIDS (100 µM) for 36 minutes (30 minutes immediately before the flux as well as during the 6 minute flux period) (CM+DIDS). Values are means±SEM of seven independent experiments each of which was done in triplicate. The CM significantly stimulated ¹⁴C-oxalate uptake, an effect completely abolished by DIDS (*P<0.001 for CM compared with UT and CM+DIDS, by ANOVA).

To examine whether the CM stimulates oxalate transport in C2 cells by increasing the mRNA expression of the likely involved transporters (SLC26A1 [A1], A2, and/or A6), C2 cells were UT or similarly treated with the CM and then total RNA was isolated for qPCR analysis. Compared with UT, the CM did not significantly affect A1, 2, and 6 mRNA expression (Figure 8), indicating that the observed stimulation is unlikely to be through modulation of these transporters at a transcriptional level. In addition, the CM did not significantly affect A6 total and surface protein expression assessed by immunoblotting and surface biotinylation (Supplemental Figure 1). However, because of a lack of good antibodies, we did not test the CM effects on A1 or A2 total and surface protein expression.

Figure 8.

The CM has no effect on A1, A2, and A3 mRNA expression. C2 cells were UT or apically treated with CM in the culture medium (1:50 dilution × 24 hours), and then total RNA was isolated for real-time PCR analysis. Values are means±SEM of three independent experiments each of which was done in duplicate or triplicate. Relative A1, A2, and A6 mRNA expression levels were expressed as a percentage of UT normalized to GAPDH. The CM had no significant effect on A1, A2, and A6 mRNA expression levels.

To examine whether the CM affects oxalate transport kinetic characteristics (i.e., the apparent affinity for oxalate [k_m] and maximal velocity [V_max)]), ¹⁴C-oxalate uptake as a function of increasing ¹⁴C-oxalate concentration in the flux medium (0.3, 1, 3, 10, 20, 75, 150, and 225 µM) was assessed. Compared with UT cells, the CM significantly stimulated (>2.3-fold) oxalate uptake (which is saturable in the presence of increasing oxalate concentrations, reflecting a carrier-mediated transport process) at each concentration (Figure 9). Analysis of these results with a nonlinear regression Michaelis–Menten fit yielded a K_m of 67.79±15.47 and 19.67±3.64 µM and a V_max of 12.63±1.04 and 25.78±1.21 pmol/cm² per minute for UT and CM, respectively. These data indicate that the observed stimulation is due to CM-induced increase (>2-fold) in V_max (i.e., greater transport capacity) and reduction (>3.4-fold) in K_m (i.e., greater affinity for oxalate) of the involved transporter(s).

Figure 9.

The CM increases the V_max and reduces the K_m. C2 cells were UT or apically treated with CM in the culture medium (1:50 dilution × 24 hours), and then¹⁴C-oxalate uptake as a function of increasing ¹⁴C-oxalate concentration in the flux medium (0.3, 1, 3, 10, 20, 75, 150, and 225 µM) was assessed. Values are means±SEM of three independent experiments each of which was done in triplicate. The CM significantly stimulated ¹⁴C-oxalate uptake at each oxalate concentration (*P<0.03, 0.003, 0.003, 0.01, 0.002, 0.01, 0.01, and 0.001 for CM compared with UT at the above corresponding ¹⁴C-oxalate concentrations, by unpaired t test).

To investigate the A6 role in the observed stimulation, A6 expression in C2 cells was knocked down using siRNA. C2 cells were untransfected or transfected with a negative control siRNA (NC si) or an A6 specific siRNA (A6 si). A6 si significantly reduced A6 protein expression (approximately 61%) and baseline oxalate uptake (approximately 49%) by C2 cells (Figure 10), results which were similar to those previously reported,²⁶ whereas NC si had no effect, indicating the observed reduction in A6 expression is not due to a general effect of the silencing procedure. Equal loading was verified by observing no difference in GAPDH. Silencing A6 greatly decreased the CM-induced stimulation of oxalate transport by C2 cells (Figure 10C), indicating that a significant component of the observed stimulation is mediated by A6.

Figure 10.

Silencing A6 greatly reduced the CM-induced stimulation of oxalate transport. (A) A representative Western blot analysis of total A6 protein expression. A6 protein expression was evaluated in C2 cell lysate (10 µg protein/lane: untransfected cells; NC si, C2 cells transfected with the NC si; A6 si, C2 cells transfected with the siRNA targeting A6). The lower half of the same blot was probed with an anti-GAPDH antibody to normalize loading of protein in each lane (lower panel). (B) Densitometry of immunoblot results. Western blot band density was quantified using ImageJ software. Values are means±SEM for six independent experiments of relative total A6 abundance to GAPDH and are presented as a percentage of the UT value. A6 siRNA knockdown significantly reduced A6 protein expression (*P<0.001 for A6 si compared with UT and NC si, by ANOVA). (C) Effect of A6 silencing on the CM-induced stimulation of ¹⁴C-oxalate uptake by C2 cells. Untransfected (Control), NC siRNA (transfected with the NC si), and A6 siRNA (transfected with the siRNA targeting A6) C2 cells were UT or were treated apically with the CM in the culture medium (1:50 dilution × 24 hours), and then ¹⁴C-oxalate uptake was measured as described in the Concise Methods. Values are means±SEM of eight independent experiments each of which was done in triplicate. Silencing A6 significantly reduced the CM-induced stimulation of oxalate transport by C2 cells (*P<0.001 for CM compared with UT in the Control and NC siRNA; **P<0.01 for UT in A6 siRNA compared with UT in Control and NC siRNA; ***P<0.001 for CM in A6 siRNA compared with CM in Control and NC siRNA, by ANOVA).

Although C2 cells closely resemble the native epithelium, confirmation of the in vitro findings in vivo is of paramount physiologic and translational interest. To this end, we evaluated the CM effects on the hyperoxaluria observed in PH1 mice.³⁷ We first confirmed that these mice have significant (>3.6-fold) hyperoxaluria compared with their controls (C57BL/6J) (µM/µM creatinine: controls=0.16±0.01; PH1=0.57±0.03; n=6–10). PH1 mice (males; 38–42 weeks old) were given CM or OM (100 µl twice daily as rectal enemas × 21 days) and urine was collected before (Before Rx) and after treatment (After Rx) for oxalate measurements. Importantly, the CM significantly reduced (>32.5%) urinary oxalate excretion, whereas OM had no effect (Figure 11). These results indicate that the O. formigenes–derived factors retain their biologic activity in vivo and effectively lower urinary oxalate excretion. Of note is that distal colonic histology from OM - and CM-treated mice (compared with UT mice) looks normal, with no evidence of injury, irritation, or inflammation (Supplemental Figure 2).

Figure 11.

O. formigenes CM reduces urinary oxalate excretion in PH1 mice. The PH1 mice were given the CM or OM (100 µl twice daily as rectal enemas × 21 days; n=6–8) and urine was collected before (Before Rx) and after treatment (After Rx) for oxalate measurements. The CM significantly reduced urinary oxalate (adjusted for urinary creatinine [Cr]) excretion (*P<0.03 for CM compared with OM, by unpaired t test).

To test the possibility that the observed reduction in urinary oxalate excretion is due to CM-induced colonic oxalate secretion, distal colonic tissues from CM- and OM-treated PH1 mice were isolated and mounted in Ussing chambers at the end of treatment. Although a small net oxalate secretory flux (−4.71±2.44) is observed in distal colonic tissues from OM-treated PH1 mice, a >4.2-fold higher net oxalate secretory flux (−20.01±5.51) is seen in distal colonic tissues from CM-treated PH1 mice, which is due to significantly increased (>42%) secretory flux (J_SM) (Table 1). CM or OM had no significant effect on the absorptive flux. These results indicate that the O. formigenes–derived factors reduce urinary oxalate excretion in PH1 mice through mechanisms including enhanced net distal colonic oxalate secretion. Although the transepithelial conductance is not affected, the CM significantly increased the short-circuit current (Table 1), suggesting that the CM also affects the movement of other ion(s), as observed with O. formigenes colonization of PH1 mice in some series, and rats treated with O. formigenes lysate.^22,23 The latter was explained by concomitant induction of net chloride secretion, which could potentially stimulate transepithelial colonic oxalate secretion by providing a counter-ion gradient for exchanging intracellular oxalate for mucosal Cl.

Table 1.

Unidirectional (mucosa to serosa [J_MS, absorptive flux], and serosa to mucosa [J_SM, secretory flux]) and net (J_net) transepithelial oxalate fluxes across distal colonic tissues (n=5 tissue pairs) isolated from OM- and CM-treated PH1 mice and mounted ex vivo in modified Ussing chambers

PH1 Mice	JMS (pmol/h per cm2)	JSM (pmol/h per cm2)	Jnet (pmol/h per cm2)
OM	37.84±5.92	42.56±5.48	−4.71±2.44
CM	40.56±3.72	60.57±3.13a	−20.01±5.51b

The CM significantly increased the J_SM as well as the J_net flux. OM or CM did not significantly affect the transepithelial conductance (G_T) (OM = 20.94±1.39 mS/cm²; CM = 24.08±2.37 mS/cm²). The CM significantly increased the short-circuit current (I_sc) (OM = 0.51±0.10 µEq/cm² per hour; CM = 1.07±0.24 µEq/cm² per hour; P<0.05, by unpaired t test). Data are expressed as means±SEM.

^aP<0.03 for CM compared with OM, with regards to J_SM, by unpaired t test.

^bP<0.05 for CM compared with OM, with regards to J_net, by unpaired t test.

Discussion

Probiotic bacteria have several health benefits; however, difficulties in determining intestinal bacterial bioavailability and biosafety concerns when administering live probiotics are potential problems facing current probiotics clinical application. Developing probiotics-derived factors as novel therapeutic agents is an alternative approach to address such concerns. Secreted factors from many probiotics stimulate several intestinal epithelial cell protective responses by activating different protein kinases.^{31,32,38–41} We now find that O. formigenes–derived factors significantly stimulate oxalate transport in C2 cells through a PKA-dependent signaling pathway. These factors retain their biologic activity in vivo and significantly reduce urinary oxalate excretion in PH1 mice by stimulating distal colonic oxalate secretion.

Hatch et al. reported that O. formigenes interacts with colonic epithelium by inducing distal colonic oxalate secretion, leading to reduced urinary excretion via a potential secretagogue.²² They also tested O. formigenes whole cells, cell membranes, and lysates on oxalate transport across rat distal colonic tissues mounted in Ussing chambers, and found no effect. Also, the effects of these products were tested for 45 minutes, which might be too short in view of the reported longer incubations (2–24 hours) for other probiotics,^32,42 and the fact that the CM had no effect with 1 hour incubation. Feeding hyperoxaluric rats encapsulated O. formigenes lysate for 5 days significantly induced distal colonic oxalate secretion and reduced urinary oxalate excretion,²² which supports this assumption. In addition, the whole cells and lysates significantly degraded oxalate in the chamber, necessitating heat treatment of samples to eradicate this degradative effect.²² The authors acknowledged that such heat treatment might have destroyed any potential O. formigenes–derived factor(s), and our finding that heated CM lost its bioactivity supports this possibility.

A rising prevalence of hyperoxaluria is seen in association with many conditions, including inflammatory bowel disease, bariatric surgery, and obesity. Significant hyperoxalemia is seen in PH and ESRD.^13,43 Unfortunately, there is currently no specific therapy that effectively lowers urine and/or plasma oxalate level(s). Dietary oxalate restriction is ineffective in PH, and is only of modest benefit in EH. The risk of recurrent COKS, nephrocalcinosis, oxalate nephropathy, ESRD, and systemic oxalosis remains substantial in the absence of treatment. In those PH patients unresponsive to pyridoxine, liver transplantation is the only effective treatment.¹³ Cardiovascular diseases are the leading cause of morbidity and mortality in ESRD patients,⁴⁴ and a recent report raised the possibility that the ESRD-associated hyperoxalemia might contribute to this increased risk.⁴³ Lowering serum oxalate might improve cardiovascular outcomes in ESRD patients if these findings are confirmed. The difficulties in maintaining O. formigenes colonization and biosafety concerns with administering live probiotics necessitate the identification of the O. formigenes–derived factors. The fact that these factors are active in vivo provides compelling evidence to pursue their characterization, given their significant therapeutic potential. In addition, the observation that the secretion of these factors is inducible by lowering O. formigenes culture medium oxalate concentration will significantly facilitate their characterization. Of interest in this regard is that we identified several interesting O. formigenes secreted proteins by mass spectrometry, and the most promising candidates will be overexpressed in Escherichia coli to obtain the corresponding recombinant proteins in future studies. If this approach fails to identify the O. formigenes–derived bioactive factor(s), then an alternative future strategy will be to look at the O. formigenes transcriptome under low and high oxalate concentrations in the culture medium.

A6 plays a critical role in mouse duodenal and ileal oxalate secretion.^15,16 A6 is also expressed apically in mouse colon,⁴⁵ but its role in colonic oxalate transport remains unknown. Knockdown studies showed a significant component of the observed stimulation is A6-mediated. An increase in V_max indicates greater transport capacity, which could result from more membrane transporters. However, because the CM did not affect A6 surface protein expression, and in view of the reduced K_m (reflecting greater A6 affinity for oxalate), it is likely possible that the observed stimulation is due to mechanisms including CM-induced enhanced A6 transport activity (resulting from an increase in the intrinsic activity of the preexisting A6 membrane transporters). A2 is apically expressed in enterocytes (colon > small intestine).^46–48 It mediates oxalate transport when expressed in heterologous systems and it is DIDS-sensitive³⁵; however, its role in intestinal oxalate transport is unknown. The observed changes in K_m and V_max could also result from CM-mediated increased A2 transport activity. Although A2 mRNA expression is not affected, it is possible that the observed stimulation is due to CM-induced enhanced A2 total and/or surface protein expression, which will be tested in future studies. Because transcellular oxalate secretion requires oxalate influx into the enterocyte from the blood side, where A1 might be involved, and then its efflux from the luminal side, it is possible that CM-induced changes in A1 expression and/or activity also contribute to the observed enhanced distal colonic oxalate secretion.

SLC26A3 (A3) plays a critical role in transcellular mouse colonic oxalate absorption.⁴⁹ However, its role in oxalate transport in C2 cells remains to be determined. Worthy of mentioning is that in contrast to the observed robust Cl-HCO₃ exchange, human A3 transports oxalate poorly (both uptake and efflux, despite using a very high ¹⁴C-oxalate concentration of 1 mM) when expressed in Xenopus oocytes,⁵⁰ strongly suggesting that Cl-HCO₃ exchange is the main physiologic function of A3. In addition, A3-mediated Cl/OH exchange activity has been assessed as DIDS-sensitive (using 600 µM DIDS) ³⁶Cl uptake by C2 cells,^42,51 reflecting it is relatively DIDS-insensitive. It should be noted that A3 was proposed to serve as a DIDS-sensitive SO₄/HCO₃ exchanger, while it simultaneously operates as the DIDS-resistant Cl/HCO₃ exchanger.⁵² At 0.5 mM, DIDS only weakly inhibited A3 activity (assessed as ³⁶Cl efflux in A3 expressing Xenopus oocytes) in buffers containing >100 mM Cl,^50,53 whereas 1 mM DIDS almost completely inhibited A3-mediated ³⁶Cl uptake by A3 expressing Xenopus oocytes in buffers containing 1 mM Cl.⁵⁴ On the basis of these observations, Whittamore et al. suggested that the observed differential sensitivity of A3-mediated Cl absorption and SO₄ secretion to DIDS (0.5 mM) may be due to the extracellular concentrations of Cl (122 mM) and SO₄ (0.5 mM), and that A3 may be sensitive to DIDS in the absence of Cl.⁵² Of note is that studies with the anion exchanger AE1 suggest that extracellular Cl allosterically lowers DIDS affinity for AE1.⁵⁵ However, ³⁶Cl uptake by A3 expressing HEK cells was found to be insensitive to 1 mM DIDS in Cl-free buffers.⁵⁶ Therefore, the issue of A3 being DIDS-sensitive in Cl-free buffers is not yet resolved. In addition, it remains to be determined whether lower DIDS concentrations (e.g., 100 µM, which completely abolished the CM-induced stimulation by C2 cells) would have a similar effect as that observed with 0.5 mM DIDS (which could still reflect relative DIDS-insensitivity). Moreover, what was observed ex vivo in native mouse epithelium⁵² might not be operative in vitro in a human cell line like C2 cells. Importantly, we also observed that 100 µM DIDS in buffers containing >120 mM Cl also completely abolished the CM-induced stimulation by C2 cells (data not shown). Collectively, in the absence of clear data directly showing that A3 contributes to DIDS-sensitive oxalate transport in C2 cells, the previously reported L. acidophilus CM-induced increase in A3 activity^42,51 should not necessarily result in enhanced oxalate uptake by C2 cells in Cl-free buffers as seen in Figure 1B. Furthermore, CM-mediated stimulation of A3 activity in native epithelium is expected to enhance transcellular colonic oxalate absorption, leading to increased urinary oxalate excretion. However, the CM has no effect on distal colonic oxalate absorption, as well as it significantly reduced urinary oxalate excretion in PH1 mice. Taken together, these observations strongly argue against an important role for A3 in the CM-induced stimulation of oxalate transport in C2 cells (which is very DIDS-sensitive) and native colonic epithelium. Of note is that the CM did not affect A3 mRNA level (data not shown).

Our finding that O. formigenes CM stimulates ¹⁴C-oxalate uptake by C2 cells through a PKA-dependent pathway fit with previous studies demonstrating cAMP stimulates oxalate secretion in rabbit proximal and distal colonic tissues.^57,58 In view of these findings and because C2 cells closely resemble the native epithelium,^59,60 we anticipate that the observed CM-induced stimulation of mouse distal colonic secretion might potentially be PKA-mediated in vivo. Of interest in this regard is that A1 and A2 are predicted to have highly conserved PKA phosphorylation sites (S467 and S693, respectively), raising their functional significance and potential involvement in PKA regulation of these transporters. However, A6 is not predicted to have a PKA phosphorylation site. Indirect PKA regulation of A6 is possible through its interaction with CFTR. CFTR stimulates A3 and A6 functions in cultured pancreatic duct cells.⁶¹ PKA-dependent phosphorylation of CFTR domain promotes its binding to SLC26 exchanger domain, resulting in significant mutual activation of CFTR and the SLC26 exchanger.^62,63 Therefore, A6 might be involved in the observed CM-induced stimulation through mechanisms including PKA-dependent CFTR activation. We don’t know whether the O. formigenes–derived secretagogue directly modulates the activity of the involved transporter(s) or acts indirectly through other regulatory/accessory proteins, which might involve translational, post-translational, and/or other genomic changes. Future mechanistic studies will explore these possibilities, which will help explain why the CM requires >1 hour to stimulate oxalate transport in C2 cells. Worthy of mentioning is that we previously found that PKC activation (including cholinergic and purinergic signaling acting through PKC) negatively regulate A6-mediated oxalate transport in intestinal cells through a Gö6983-sensitive pathway.^28,30,34 However, Gö6983 has no effect on the observed stimulation, which is expected because the CM is stimulating rather than inhibiting oxalate transport.

In summary, we have shown that O. formigenes CM significantly stimulates oxalate transport in C2 cells through mechanisms including PKA activation, with A6 being responsible for a significant component of the observed stimulation. The CM also significantly reduces urinary oxalate excretion in PH1 mice by stimulating distal colonic oxalate secretion. Given the significant potential for clinical application of the O. formigenes–derived bioactive factors as novel therapeutic agents for prevention and/or treatment of hyperoxaluria, hyperoxalemia, and related COKS, characterization of these factors is currently under active investigation.

Concise Methods

Cell Culture

Human intestinal C2 and T84 cells were grown and maintained (including monitoring the TER) as we reported previously.^28,30 The oxalate flux and other studies described above were performed using confluent cells grown for 5–14 days postplating on 0.4 µm collagen-coated polystyrene transwell membrane filters (Corning, Corning, NY) in 12 mm inserts. The cells were switched from a DMEM-containing 8%–10% FBS to a DMEM with 0.1% FBS, and then incubated for 1 hour before treatment with the specified reagents (e.g., OM, CM, etc.) in the same 0.1% FBS medium.

Bacterial Cultivation and Preparation of Conditioned Media

O. formigenes was obtained from the American Type Culture Collection (ATCC) (35274; Manassas, VA) and grown under a 95% CO₂, 5% H₂ atmosphere at 37°C in Oxalobacter medium⁶⁴ with varying concentrations of sodium oxalate (1 g/L=9.4 mM; 2.5 g/L=18.8 mM; and the standard 5 g/L=37.5 mM). Cultivation medium was dispensed in 50 ml aliquots per 250 ml anaerobic bottles in an anaerobic chamber (Coy Laboratories, Grass Lake, MI). After autoclaving, a solution of filter-sterilized sodium oxalate (using 0.22 µm filter) was then aseptically added to each bottle to achieve the desired final concentration of sodium oxalate. Batches of media were prepared in 50 ml volumes for each sodium oxalate concentration in duplicate and inoculated with 1:10 volume of actively growing culture. Cultures were then harvested after a 12 hour incubation for one set and after an 18 hour incubation for the second set (OD at 600 nm of 0.03, 0.075, and 0.150, respective to sodium oxalate concentration). O. formigenes cultures were centrifuged (3000 × g at 4°C for 10 minutes) and the supernatant (CM) was filtered through a 0.22 µm filter to sterilize and remove all bacterial cells and stored at −80°C. Of note is that we found CM9.4 (harvested at 12 hours from cultures containing 1g/L of sodium oxalate) had the highest bioactivity compared with CM18.8 or CM37.5 harvested at 12 or 18 hours (Figure 5). Therefore, most of the studies were done using CM9.4. Earlier in the study, we had used CM37.5 harvested at 12 hours before comparing the three different oxalate concentrations described above.

For the studies with heated CM, the CM was subjected to heat treatment (boiling at 100°C for 20–30 minutes) in a thermoregulatory water bath and then allowed to cool to room temperature before use.

For the studies with protease-treated CM, the CM was digested with pepsin. Because pepsin is active at acidic pH, the CM was first made slightly acidic by titrating with concentrated HCl to a pH of 4 before adding pepsin. The CM was incubated with pepsin (0.4% final concentration) at 37°C for 60 minutes, and then pepsin was irreversibly inactivated by altering the pH to 8.0 with concentrated NaOH and incubation for 10 minutes. Alternatively, after the 60 minute incubation, pepsin (MM=34.7 kDa) was removed by subjecting the pepsin-treated CM to sizing filtration using a 30 kDa cutoff spin column. The pH was adjusted to 7.4 before experimental use. Of note is that pepsin treatment completely abolished the CM bioactivity under both conditions (i.e., pepsin inactivation by pH 8.0 or pepsin removal by sizing filtration).

To determine the molecular weight(s) of the O. formigenes–derived secreted bioactive factor(s) present in the CM, the CM was subjected to selective ultrafiltration using 10 and 30 kDa molecular mass cutoff spin columns (Ultracel YM-10 and YM-30 membranes; EMD Millipore, Billerica, MA), per manufacturer’s protocols. The CM samples were spun until 10%–15% of the overall volume was retained upon concentration. These concentrated fractions were then used in the experiments and referred to as the filtrate (F; containing molecules <10 or 30 kDa) and the retentate (R; containing molecules >10 or 30 kDa).

L. acidophilus was obtained from ATCC (4357) and grown as previously reported.⁵¹ Briefly, L. acidophilus was grown overnight in MRS broth at 37°C in a 5% CO₂, 95% O₂ incubator. L. acidophilus CM was similarly prepared as described with O. formigenes CM.

Radioactive Flux Studies

Apical ¹⁴C-oxalate flux studies in C2 and T84 cells were performed following our previously published methods.^28,30 Briefly, an outward Cl gradient is imposed by removing extracellular Cl (Cl_i>Cl_o, by incubating in a Cl-free solution containing 20 μM ¹⁴C-oxalate for 6 minutes) and the influx of ¹⁴C-oxalate in exchange for intracellular Cl (i.e., apical Cl-oxalate exchange activity, which we previously found to be inhibited [>91%] by 100 µM luminal DIDS [anion exchange inhibitor]²⁸) is measured. The influx of ¹⁴C-oxalate was terminated by two to three rapid washes of the cell monolayers with ice-cold Cl-free solution and the transwells were then placed upside down and were allowed to dry for several minutes. Membrane filters containing the cells were cut from the support and placed into vials with scintillation fluid (Opti-Fluor, Packard), and the radioactivity was measured by scintillation spectrometry after overnight solubilization.

Animals

A mouse model of PH type 1 (called PH1 mice in this study), due to deficiency in liver alanine-glyoxylate aminotransferase,³⁷ was kindly provided by Dr. Roy-Chowdhury (Albert Einstein College of Medicine). PH1 mice exhibit hyperoxalemia and hyperoxaluria.³⁷ Mice were bred and maintained in an specific pathogen free facility, with free access to water and standard mouse chow. All animal studies were approved by the University of Chicago Institutional Animal Care and Use Committee and were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Rectal Administration of O. formigenes OM or CM

O. formigenes OM or CM was administered to PH1 mice as rectal enemas (100 µl twice daily × 21 days) using disposable animal feeding needles (catalog no. 01–208–87; Fisher Scientific, Waltham, MA). The needles were introduced approximately 2 cm from the anus, and the mice were then held from their tails with the heads down and kept in that position for approximately 1 minute. This protocol was developed in pilot studies, where CM or OM was given to PH1 mice for up to 28 days, and the CM was found to significantly reduce urine oxalate excretion, whereas OM had no effect.

Urine Collection and Oxalate Measurements

Mice were placed individually in well ventilated boxes containing sterile 96-well plates (in which the urine get collected) and baseline urine samples were collected over 1 hour. This method was approved by our Institutional Animal Care and Use Committee. Urine samples from the OM- and CM-treated PH1 mice were directly collected from the bladders at the end of treatment. The urine samples were acidified with 4N HCL (50 µl per ml of urine) and stored at −80°C. Urine oxalate was measured enzymatically using oxalate oxidase (Trinity Biotech, Bray, Ireland), employing sodium nitrite⁶⁵ to prevent interference by ascorbic acid. Oxalate measurements in OM and CM were performed by ion chromatography. Urine creatinine was measured by a modified Jaffe alkaline picrate method.

Measurement of Oxalate Fluxes across Distal Colon Tissues

Immediately after euthanasia, distal colonic tissues were removed, opened longitudinally along the mesenteric border, thoroughly cleansed by flushing with ice-cold saline, and then mounted (one pair of tissues per mouse) as intact sheets in modified Ussing chambers (Physiologic Instruments, San Diego, CA) (exposed tissue surface area of 0.30 cm²). Two unidirectional oxalate fluxes (mucosa to serosa [J_MS]; and serosa to mucosa [J_SM]) were assessed and the magnitude and direction of the net flux (J_net) across conductance-matched tissues was determined by calculating the difference between the two measured unidirectional fluxes. The mucosal and serosal surfaces of the tissue were bathed with 4 ml of warmed (37°C) Ringer buffer (140 mM Na⁺, 119.8 mM Cl⁻, 5.2 mM K⁺, 1.2 mM Mg²⁺, 1.2 mM Ca²⁺, 25 mM HCO₃⁻, 2.4 mM HPO₄⁻, 0.4 mM H₂PO₄²⁻, pH 7.4, gassed with 95% O₂-5% CO₂), containing 10 mM mannitol or glucose on the mucosal or serosal side, respectively. Two micromolar ¹⁴C-oxalate was added either to the mucosal or the serosal bath, and 2 µM unlabeled oxalate was added to the opposite bath. All buffers contained 5 µM indomethacin to inhibit endogenous prostanoid production. After a 15 minute equilibration period, unidirectional ¹⁴C-oxalate fluxes were measured for 60 minutes at 15 minute intervals under short-circuit conditions. The electrical parameters of the tissues were continuously recorded (at 10–20 second intervals) using Acquire and Analyze Software (Physiologic Instruments) attached to the chamber. The viability of the tissue was ensured by treating the tissue with carbachol (100 µM, serosally) at the end of the experiment and observing an immediate increase in short-circuit current.⁶⁶

A6 Knockdown in C2 Cells

A6 expression in C2 cells was knocked down as previously reported,²⁶ using the same A6 and negative control siRNAs, with minor modifications as described below. C2 cells were untransfected or transfected with the NC si or the A6 si. We observed approximately 61% reduction in A6 total protein expression 2–3 days after transfection, and most of the studies were performed 3 days after transfection.

SDS-PAGE, Western Blotting, and Surface Biotinylation

C2 cells were scraped and lysed in lysis buffer (in mM: NaCl=150, TRIS-HCl=50 [pH 7.4], EDTA=2, Sodium Fluoride=50, NP-40=1%, Sodium Deoxycholate=0.5%, SDS=0.1%) supplemented with Complete Protease Inhibitor Cocktail (Roche Diagnostics, Indianapolis, IN). The lysate was then centrifuged (14,000 × g, 4°C, 10 minutes), and the supernatant was saved for gel electrophoresis. Total protein levels were determined and equal amounts of protein lysates were separated by SDS-PAGE using 7.5% polyacrylamide gels, with subsequent electro-transfer to polyvinylidene difluoride (Immobilon-P; EMD Millipore). All immunoblots were stained with Ponceau S Solution (0.1% Ponceau S [w/v] in 5% acetic acid [v/v]) after transfer. For Western blotting and to block nonspecific binding, polyvinylidene difluoride strips were first incubated in blotto (5% nonfat dry milk and 0.2% Tween 20 in tris buffered saline) for 1 hour. Immunoblots were then incubated overnight at 4°C with anti-A6 antibody (1:100 dilution) (catalog no. sc-26728; Santa Cruz Biotechnology, Santa Cruz, CA) and anti-GAPDH antibody (1:5000 dilution) (catalog no. sc-32233; Santa Cruz Biotechnology). The strips then were washed in blotto and incubated for 1 hour with horseradish peroxidase–conjugated secondary antibodies (donkey anti-mouse or mouse anti-goat IgG, 1:10,000 dilution). Antibody reactivity was detected with the enhanced chemiluminescence system (SuperSignal West; Thermo Fisher Scientific, Vernon Hills, IL) according to the manufacturer’s protocol.

For the surface biotinylation studies, C2 cells were grown in 0.4 µm collagen-coated polystyrene transwell membrane filters in 24 mm inserts and treated with the CM as described above. Surface biotinylation studies were performed as previously described.²⁸

Real-Time PCR

RNA was extracted from UT and CM-treated C2 cells using E.Z.N.A. Total RNA Kit (Omega Bio-tek, Norcross, GA). RNA was reverse transcribed and amplified with Power SYBR Green PCR Master Mix kit (Applied Biosystems, Foster City, CA). Human A1 was amplified with gene-specific primers (forward primer, 5=-GGACTCCCTCAGCGTGCAAACA-3=; reverse primer, 5=-GCGTTAGGGAGGCGGTGCTTAG-3=). Human A2 was amplified with gene-specific primers (forward primer, 5=-CGCGGGCGTTTACACTGGCT-3=; reverse primer, 5=-GTCCCTTTCCCCGCCCATCG-3=). Human A6 was amplified with gene-specific primers (forward primer, 5=-GCAACACAAGCAATGGACCTG-3=; reverse primer, 5=-CCGGGGTAACCAGACCAAAAC-3=). Human GAPDH was amplified as an internal control using gene-specific primers (forward primer, 5=-TCCCTGAGCTGAACGGGAAG-3=; reverse primer, 5=-GGAGGAGTGGGTGTCGCTGT-3=). Relative A1, A2, and A6 mRNA levels were expressed as a percentage of UT normalized to GAPDH.

Histologic Examination

Distal colonic tissues were immediately placed in 10% neutral buffered formalin and were fixed at room temperature for 24 hours. After fixation, the tissues were transferred to 70% reagent alcohol and then submitted to our histology facility for processing and embedding. Hematoxylin and eosin-stained sections (4 μm thick) were blindly examined under a light microscope by a gastrointestinal pathologist.

Materials

¹⁴C-oxalate was purchased from Vitrax (specific activity: 54 mCi/mmol). H89, PP2, SB 202190, Gö6976, Gö6983, carbachol, genistein, and LY 294002 were purchased from Calbiochem (San Diego, CA). U0126 was purchased from Promega (Madison, WI). DIDS was purchased from Sigma-Aldrich (St. Louis, MO). H89, PP2, carbachol, Gö6983, SB 202190, LY 294002, U0126, and genistein were dissolved in DMSO and stored at −20°C. Gö6976 was dissolved in DMSO and stored at 4°C. Equivalent volumes of DMSO (0.1%–0.2%) were added to control media. Of note is that the PKC inhibitors Gö6983 and Gö6976, the ERK1/2 inhibitor U0126, the p38 inhibitor SB 202190, the PI3K inhibitor LY 294002, the general tyrosine kinase inhibitor genistein, and the specific Src family kinase inhibitor PP2 were used at concentrations known to effectively block the relevant signaling pathways in C2 and T84 intestinal cells.^28,67–70

Statistical Analyses

Experimental data are presented as means±SEM. Data were analyzed by one-way ANOVA followed by Bonferroni or Student–Newman–Keuls post hoc test, or by t test for paired or unpaired samples when comparing two groups. P values <0.05 were considered statistically significant.

Disclosures

I.G. and J.A. are employed by LabCorp (Chicago, IL).