Background/Aim: Colorectal cancer is still considered a leading cause of death in the United States and worldwide. One potential way to improve survival besides detection is to look to new therapeutic agents that can be taken prophylactically to reduce the risk of tumor formation. For cancer cells to grow and invade, a higher (more alkaline) intracellular pH must occur. We chose to examine a specific nutraceutical agent, which is Vitamin C. The acute effect of Vitamin C exposure on normal colonic crypts has been studied, providing some insight into how Vitamin C achieve its effect. Methods: Distal colon was excised from rats. Following enzymatic digestion single colonic crypts were isolated. Colonic crypts were loaded with pH sensitive dye to measure the intracellular pH changes. Crypts were exposed to solutions +/- Vitamin C. Results: 10 mM Vitamin C decreased Na+-dependent intracellular pH recovery. Vitamin C modulates SVCT leading to changes in proton extrusion. Vitamin C entry occurs via either SVCT2 on the basolateral membrane or by transcellular passive diffusion through tight junctions to the apical membrane and then active transport via SVCT1. Conclusion: Acute addition of Vitamin C to the basolateral membrane maintains low intracellular pH for a longer period which could halt and/or prevent tumor formation.
Colorectal cancer is the third most common cancer worldwide and the third most common cause of death with just over one hundred thousand new cases diagnosed each year equally divided between men and women . The United States has among one of the highest levels for all countries and has approximately 93, 000 diagnosed cases per year, and an estimated death rate of 49, 000 per year .
One of the common pathophysiological pathways for cancer to grow, proliferate and invade neighboring cells, is that the cancer cell must maintain an elevated intracellular pH (pHi) and concurrently a significantly lower pH of the extracellular environment (pHe) [3, 4]. By understanding the intracellular ionic signaling pathways, it may be possible to manipulate the pHi of a cancer cell thereby arresting its growth. Similarly, modulating the pHe may prevent or retard the growth of the tumor. The modulation of these two environments (pHi and pHe) via targeting the specific ionic exchange proteins could thus be a promising novel treatment strategy. Vitamin C (Ascorbic Acid) is known to play a role in the differentiation process of cells , showing that Vitamin C has capabilities to modulate cells proliferation, which suggests a possible inhibitory action also on tumor cells . As an antitumorigenic, Vitamin C has been proven to protect against several types of non-hormonal cancers in lung and colon [7-9]. Maulén et al. showed that the analysis of quantitative PCR of colonic carcinoma cell line, which revealed that the expression of the Na-dependent Vitamin C transporters SVCT 1 mRNA increased by 4 fold without changes in SVCT2 mRNA levels. Moreover, this study also showed that the Vmax increased also by at least 2-fold .
Recent studies have shown that Vitamin C is absorbed throughout the intestine via an energy-dependent Na-dependent process that is saturable. In previous studies, this absorption of Vitamin C was associated with Glucose transport [11, 12]. If this relationship to Na-dependent Glucose transport can be exploited, it may be possible to enhance the uptake of Vitamin C and thus further provide an antitumorigenic effect . Under these conditions, we would expect an enhanced modulation of the pH-dependent processes both intracellularly and extracellularly. Our studies presented below also focused on the uptake of Vitamin C via SVCT1 and SVCT2 [14-16], and glucose Transporters (GLUT) and their effects on acid extrusion or uptake. We further examined the concentration dependence of these effects by using 3 concentrations that were either 10-fold lower or higher than the suggested maximal daily dose . The highest concentration used in this study was comparable to a dose used clinically in attempting to reduce tumor growth . In the present study, we chose to investigate the physiological process(es) that occurs in an acute exposure of Vitamin C on normal colonic crypts and to determine if there are any modulations in sodium dependent hydrogen ion extrusion.
Materials and Methods
Male Sprague-Dawley rats weighing 300-650g were used in this study. All animals were kept in a climate controlled facility. The animals were fed standard chow and had ad lib access to water prior to experimentation. The animals were handled by animal care personnel according to the approved protocol by The Institutional Animal Care and Use Committee at Yale University (protocol number 2015-10253). The animals were fasted for 12 hours before tissue harvest with ad lib access to water during this period .
Colonic Crypt Isolation and Collection
All animals used in the study were exposed to an overdose of the inhaled anesthetic Isoflurane. We then performed a laparotomy to expose the colon and excise it. We identified the anal junction and made a transverse dissection. We then removed ∼10 cm proximal to the anal colon junctional interface. The colon was then removed and flushed with 4°C cold HEPES Ringer solution (See Table 1) to remove the residual fecal material from the interior of the colon as previously described . Briefly, we cut the clean distal colon into two longitudinal strips [21, 22]. The two strips were then immersed into a digestion solution that was composed of a low Calcium High EDTA/HEPES buffer. Following digestion, the colon was removed and discarded, and the remaining solution was centrifuged at 500 rpm for 45 seconds. Using our previously described technique, we then removed the digestion buffer and rehydrated the remaining pallet with a cold HEPES pH 7.4 at 4°C (see Table 1) .
Dye preparation and cell loading
A coverslip (Fisherbrand®, Fisher Scientific measuring 22 x 50 mm) pretreated with 0.1 µL of Corning® Cell-TakTM (Cell and Tissue Adhesive, Polyphenolic proteins secreted by Mytilus edulis). We then transfer a 200 µL aliquot of the crypts HEPES solution to the chamber, then incubated the crypts with a 1 µL of BCECF/AM dye (Santa Cruz Biotechnology) a pH sensitive dye as previously described [22, 24].
For each BCECF loaded crypt, 5 regions (1 control region and 4 experimental regions) were selected to monitor pHi changes for a duration of a single experiment. The crypts were treated with standard HEPES buffer solution, alkalinized it with NH4Cl, then acidified it by depleting the crypt from Na+ through ØNa perfusion, a reperfusion of the crypts with standard HEPES before lastly calibrated it with High-K/Nigericin. The pHi monitored through measuring the average rate of Hydrogen ion extrusion through experiment cycle. The BCECF loaded crypts were excited at 490 and 440 nm±20nm respectively while collecting the emission at 530 nm±20nm every 15s. The pH was calculated using the High-K/Nigericin calibration technique  previously used in our laboratory.
Addition of Vitamin C and other drugs
L-Ascorbic acid (C6H8O6) (CAS Number: 50-81-7) and Phlorizin Hydrate: CAS Number：60-81-1 C21H24O10·xH2O were purchased from Sigma-Aldrich Co. (St Louis, MO USA).
All data were analyzed with Prism 6 program from GraphPad using Students T- test, when comparing 1 condition to 1 other condition and ANOVA when all columns were compared together. Significance was recorded for each series under all conditions.
Vitamin C Concentration Curve
To develop a concentration curve for Vitamin C effect (see Fig. 1 and 3), different concentrations have been chosen according to previous studies. The pharmacokinetics of Vitamin C show that the body has a tight control keeping the level of Vitamin C ∼100 µM . The maximum tolerable concentration of Vitamin C has been reported to be around 1mM, while it is proven that a concentration of 10mM Vitamin C could kill and fight cancer in vitro and in vivo . First, we compared the recovery rate of pHi among control experiments (ΔpHi/min 0.0160) and experiments containing 100 µM of vitamin C (ΔpHi/min 0.0284); the recovery rates were faster when we added 100 µM of Vitamin C with a p value of <0.0001. On another set of experiments, we increased the concentration of Vitamin C to 1mM which is the same as the daily maximum dose ; we found that the recovery rate of pHi (ΔpHi/min 0.0203) still faster in comparison to control experiments with p value of <0.0001 but slower than that of 100 µM with a p value of <0.0001. However, when we further increase the concentration of Vitamin C to 10mM, a concentration proposed to kill cancer cells, we found that the pHi recovery rate slowed down (ΔpHi/min 0.0080) in comparison to control with a p value of <0.0001. If we compare the pHi recovery rate between 100 µM of Vitamin C and 1mM and 10mM, we observed that 100 µM has the fastest recovery with p value of <0.0001. Conversely, the recovery rate of 10 mM of Vitamin C is slower than 100 µM and 1mM with p value of <0.0001. Tracings of pHi recovery rates are shown in Fig. 2 and the summary data shown in Fig. 3.
SVCT (Sodium Vitamin C Transporter) Role in Vitamin C-Dependent pH modification
To observe and establish a role of SVCT in Vitamin C-dependent pH modification and how it could modulate the efficacy of Vitamin C at different concentrations, we designed experiments with the same components of the solutions as found in Table 1 except for reducing the Na+ concentration to 77mM (see Fig. 5). Comparing control experiments of standard Na+ concentration (117 mM) with control experiments of 77 mM Na+ concentration, there was no significance difference in the recovery rate of pHi, the ΔpHi /min for 77 mM Na+ was 0.0169 and 0.0160 for the 117 mM Na+ respectively. The recovery rate of pHi with 100µM of Vitamin C dissolved in 77 mM solution (ΔpHi /min 0.0098) is slower than the control experiments of the same Na+ concentration(77mM) with a p value of <0.0001. However, when we increased the concentration of Vitamin C to 1mM in the same 77 mM Na+ concentration condition we observed a slower rate of pHi recovery (ΔpHi /min 0.0128) compared to 77mM Na+ control rates with a p value of 0.0021. Interestingly, when we increased the concentration of Vitamin C that is dissolved in 77mM Na+ to 10 mM; there was no significant difference in the recovery rate of pHi (ΔpHi /min 0.0190) from that of the control solution of 77 mM Na+. Tracings of pHi recovery rates are shown in Fig. 4, and the summary data shown in Fig. 5.
GLUT Effect On Vitamin C-Dependent pH Modification
To study the effect of GLUT on Vitamin C-dependent pHi modification, in the first set of this series of experiments we used the same solutions for the first set of experiments in Table 1, but excluding Glucose (117 NaCl) (see Fig. 7). The recovery rate of pHi to resting physiological pH values with 10 mM Vitamin C in glucose-free solutions (ΔpHi /min 0.0080) was slower with a p value of < 0.0001in comparison to glucose-free control solutions (ΔpHi /min 0.0133). However, there was no significant difference between 10 mM of Vitamin C dissolved in Glucose solutions and 10 mM Vitamin C dissolved in Glucose-free solutions (See Fig. 7).
In the second set of experiments we dissolved Phlorizin a known inhibitor of glucose transport (SGLT1 and SGLT2), and Vitamin C in Glucose-free solutions to blunt the activity of GLUT transport. The recovery rate of pHi with 10 mM Vitamin C plus Phlorizin dissolved in Glucose-free solutions (ΔpHi/min 0.0081) was slower in comparison with control Glucose-free solutions (ΔpHi/min 0.0133) with a p value of <0.0001. Moreover, there was no significance difference in pHi recovery rate when we compared 10 mM Vitamin C +/- Glucose versus 10 mM Vitamin C plus Phlorizin dissolved in glucose-free solutions. Tracings of pHi recovery rates are shown in Fig. 6 with summary data presented in Fig. 7.
The difference between GLUT and SVCT Effects On Vitamin C-Dependent Na-Dependent pH recovery
In this set of studies, we examined the difference in recovery rates between GLUT and SVCT on Vitamin C-dependent pH modification (See Fig. 9). We compared the rate of pHi recovery rates between control experiments that either had no Glucose (ΔpHi /min 0.0133) or had a Na+ concentration of 77 mM plus 10 mM Glucose (ΔpHi /min 0.0169). We observed that the control experiments that were Glucose free had a slower pHi recovery rate than those experiments with 77 mM Na+ concentration plus 10 mM Glucose concentration with a p value of 0.0002. When we added 10 mM Vitamin C to the various groups in a separate series of experiments, we found that the pHi recovery rate in the absence of Glucose (ΔpHi /min 0.0080) became dramatically slower than that of the 177 mM Na+ Glucose-free (ΔpHi /min 0.0133) with a p value of <0.0001. Original tracings of pHi recovery rates please refer to Fig. 8 with summary data presented in Fig. 9.
Vitamin C is believed to be an anti-tumorigenic agent in chronic exposure to the colon, but the effect of Vitamin C in the acute condition is controversial and appears to only be antitumorigenic when infused directly into the bloodstream at high concentrations . In retrospective studies published about the benefits of using Vitamin C in cancer treatment there appeared to be some effects [25, 26]. Later, double-blind trials showed no effect of Vitamin C as a tumoricidal agent . However, in 2008, Chen et al. considered Vitamin C as a cancer therapy from a different perspective from the previous studies mentioned above, notably it was infused directly at high concentrations and was tumor cytotoxic at these high levels .
Pandey et al. . established the risk of all cancers in the body being reduced significantly (39%) when >113 mg/d Vitamin C was consumed, compared to the control group who consumed <82 mg/d. However, as in all previous studies there were no published data about the effects of Vitamin C when taken acutely. We, therefore, decided in this paper to concentrate our focus on the acute effect(s) of Vitamin C on colonic crypts and observe any changes in cellular pH regulation. These potential acute changes in pHi could have a positive effect in regards to carcinogenesis/chemoprevention in the colon according to Vaughan-Jones proposed theory of cancer development .
To do so, we noted the changes of intracellular pH before and after Vitamin C administration. Our results were obtained using the isolated colonic crypt model from Sprague-Dawley rats. Vitamin C shows a marked effect on the acidity of the crypt cells from intact isolated crypts. Maintaining the pH acidic intracellularly would inhibit the growth of tumor cells, as tumor cells require the intracellular milieu to be alkaline in order to proliferate.
Our first finding that Vitamin C appears to affect pHi is supported by our experiments on the colonic crypts acidity when superfusing the crypts with solutions mixed with Vitamin C. Adding 10 mM Vitamin C induced a slower rates of pHi recovery (ΔpHi/min 0.0080) in comparison with the control experiment (ΔpHi/min 0.0160) (Fig. 3). With intracellular pH becoming more acidic, an intracellular milieu has been created that is optimized to prevent tumor cell proliferation.
Furthermore, In the setting of acute Vitamin C administration, we observed in Fig. 3 that exposure to low concentrations (100 μM and 1 mM) of Vitamin C resulted in accelerated recovery rates of pHi (ΔpHi/min 0.0284 for 100 µM Vitamin C and ΔpHi/min 0.0203 for Vitamin C 1 mM) following an acid load which could mean a favorable environment for stimulation of tumor growth, this was reversed (tumor inhibitor effect, slow Na-dependent acid extrusion) with high acute Vitamin C exposure (10 mM). In other words, the effect of Vitamin C on the intracellular pH is concentration-dependent, as observed in Fig. 3 (Vitamin C Concentration Curve), when the colonic crypts were superfused with different concentrations resulting in a biphasic concentration dependent response, low concentrations of Vitamin C (100µM and 1 mM) maintains high (alkali) pHi during the recovery period while high concentration of Vitamin C (10 mM) maintains low (acidic) pHi during recovery period. This observation can be explained by dose dependent modulation of the Na+ dependent Vitamin C transporter (SVCT). This increased activity leads to both increases in H+ concentration within the cell as well as increases in intracellular Na+ concentration. The higher intracellular Na+ concentration prevents NHE from working efficiently due to the increased Na+ gradient within the cell. On the contrary, with lower concentrations of Vitamin C, the SVCT is less active which leads to a slower build up in the intracellular Na+ concentration. This results in a faster Na-dependent recovery rate.
As we have observed that Vitamin C transport is dependent on Na+ and since we use a 77 mM Na+ concentration for our NH4Cl prepulse technique different from that of HEPES (177 mM) (Table 1), we wanted to eliminate any influence on the recovery rate that may result from this Na+ concertation difference between solutions. We performed a new set experiments with HEPES solution that made of 77mM Na+ concentration (Table 1), contain or devoid of different Vitamin C concentrations to observe its ability to maintain an acidic intracellular environment of the crypts. There was no significant difference when we compare the recovery rate of pHi between control experiments of 77 mM Na+ concentration (ΔpHi/min 0.0169) with control experiments of standard 117mM Na+ concentration (ΔpHi/min 0.0160). Comparing the results of the control experiments of 77mM Na+ concentration (ΔpHi/min 0.0169), to those of the 100 µM Vitamin C dissolved in 77 mM Na+ solutions (ΔpHi/min 0.0098) we observed a slowing of the rate of pHi recovery to the normal resting physiological rate (control conditions) with a p value of <0.0001. A similar slow recovery rate was observed when we increased the concentration of Vitamin C from 100µM to 1mM (ΔpHi/min 0.0128) with a p value of 0.0021. However, when we increased the concentration to 10mM (ΔpHi/min 0.0190) (a concentration that is used clinically to kill cancer cells), there was no significant difference in the recovery rate of pHi from that of the control solution (0mM Vitamin C). We established that the difference in Na+ concentration effects the Vitamin C dependent intracellular pH changes, by the rate of Na dependent proton extrusion, at a (77 mM) Na+ concentration the opposite effects of Vitamin C dependent proton extrusion are observed, as seen in Fig. 5 (SVCT Role in Vitamin C Dependent pH modification). This reversal in the recovery rates (Fig. 5) in comparison to 117mM Na+ concentration (Fig. 3) can be explained by the difference in the Na+ concentration and the effects on the Vitamin C modulation of SVCT and subsequently on NHE activity, resulting in a modulation in the extrusion rate of H+.
In a recent study published by Maulén et. al. using CaCo-2 cells they determined that these cells express SVCT 1 selectively on the apical side. Moreover, in the same study they found that SCVT 1 and SVCT 2 have different Km values, SVCT 1 has a lower affinity for Vitamin C (Km 125 µM) while SVCT 2 has a higher affinity (Km 8 µM) . Their study also finds that there was no transport of Vitamin C across the basolateral surface and that all of the transported Vitamin C is being transported on the apical membrane through SVCT1 . However, in our study, we superfused the intact colonic crypts (isolated from rat colon) in vitro so that all effects were predominantly via changes in basolateral transport. With that being said, and since we did observe a basolateral Vitamin C effect as measured by changes in pHi, these results show that there could be a direct difference in Vitamin C transport from cultured CaCo-2 cells and intact freshly isolated crypts. In our study Vitamin C could potentially be transported via SVCT2, or it could diffuse through the tight junctions to reach the apical side and interact with SVCT1, or alternatively there could be another protein transporter that has yet to be identified that allows Vitamin C to enter the cells.
The third outcome of our study is that GLUT transporters have a role to play in Vitamin C transport inside the cell and thereby could manipulate the effect(s) of Vitamin C. As shown in Fig. 7 10 mM Vitamin C still maintained an acidic pH (ΔpHi/min 0.0080) in the intracellular environment after removal of Glucose from the solutions.
When the drug Phlorizin was added to the glucose-free solutions to specifically bind and block SGLT transporters, Vitamin C still was able to change intracellular pH (ΔpHi/min 0.0081) in a similar pattern as previously seen. These results suggest that attenuating GLUT through glucose removal and blocking SGLT with Phlorizin did not modulate the Vitamin C effect on the recovery rate of pH to the normal physiological value. These results could mean that using Vitamin C as a therapeutic agent for colon cancer in diabetic patients would still lead to an elevation in intracellular acidity and potentially lead to reduced tumor burden.
Finally, we compared the influences of Glucose and Na+ on Vitamin C-dependent proton extrusion rates. We observed that the pHi recovery rate of 117mM Na is faster when we compared to control experiments of 77mM Na+ concentration that contain 10 mM Glucose (ΔpHi/min 0.0169) with control experiments that have the standard concentration Na+ and contain no Glucose (ΔpHi/min 0.0133) p value 0.0002. Remarkably, the rate of recovery become more aberrant when we add 10mM Vitamin C to the Glucose-free solution (ΔpHi/min 0.0080) with a p value of <0.0001 while there was no significant difference between 77 mM Na+ control and 10mM Vitamin C with the same 77 mM Na+ concentration in the extracellular environment. We believe that Na+ concentrations have the major influences on Vitamin C-dependent proton extrusion (Fig. 9) through modulations in Na-Glucose exchange and Na-H Exchange.
In summary, our data for the first time demonstrate the acute effects of Vitamin C on modulating intracellular pH regulation in isolated colonic crypts. This data is highly suggestive that a positive effect of Vitamin C as an antitumorigenic agent during chronic exposure can also be achieved in the acute exposure setting.
There are no competing interests or sponsors that affected this work. All studies are done in compliance with the Yale school of Medicine conflict of interest committee.
This work was supported by the Charles Ohse Grant program from the department of surgery.