Investigation of the effect of verapamil on the regional absorption of sofosbuvir from rabbit intestine in situ

Abstract

Purpose

Sofosbuvir, a nucleotide antiviral drug, is a Biopharmaceutics Classification System (BCS) class III prodrug suffering from limited intestinal absorption due to its high hydrophilicity and low intestinal permeability. This research aims to investigate the luminal stability of Sofosbuvir, the influence of anatomical site on its intestinal absorption and the effects of verapamil on such absorption.

Method

The study utilized in situ rabbit intestinal perfusion technique to examine absorption of Sofosbuvir from duodenum, jejunum, ileum and ascending colon. This was conducted both with and without verapamil.

Results

The luminal stability study showed that Sofosbuvir was subjected to premature degradation with varying fractions degraded from the different intestinal segments. The in situ perfusion data showed incomplete absorption of Sofosbuvir from small and large intestinal segments. The recorded values of the absorptive clearance per unit length (Pe.A/L) of Sofosbuvir were 0.026, 0.0075, 0.0026, & 0.054 ml/min.cm for duodenum, jejunum, ileum, and ascending colon, respectively. The Pe.A/L values were ordered as colon > duodenum > jejunum > ileum. This is the opposite rank of P-gp content in the different intestinal segments. The recorded values of the length required for complete Sofosbuvir absorption (L95%) were 29.58, 128.47, 949.2 and, 13.63 cm for duodenum, jejunum, ileum, and ascending colon, respectively. Co-perfusion with verapamil significantly increased Pe.A/L and reduced the L95% of Sofosbuvir from both jejunum and ileum (P-value < 0.05).

Conclusion

The results indicated that the absorptive clearance of Sofosbuvir was site dependent and associated with the content of P-glycoprotein, in addition to the expected drug interactions that can occur in polymedicated hepatitis C virus (HCV) infected patients.

Graphical abstract

Introduction

Sofosbuvir, a nucleotide antiviral analogue, is a BCS class III prodrug suffering from limited intestinal absorption due to its high aqueous solubility and low intestinal permeability. This antiviral drug is used as a part of combination therapeutic regimen for treatment of liver chronic Hepatitis C that is caused by Hepatitis C Virus (HCV) [1, 2]. Sofosbuvir has an aqueous solubility of more than 2 mg/mL at 37 °C with a molecular weight of 529 gm/mol, its log Po/w is 1.62 and its pKa is 9.3 [3].

Sofosbuvir has a limited oral bioavailability with peak plasma concentration around 0.567 mg/mL and t_max within about 0.5 to 2 h with a terminal half-life of 0.4 h [2, 3]. The drug is mainly metabolized by the liver by carboxylesterase 1 (CES1) that involves hydrolytic cleavage of the carboxyl ester group followed by phosphoramidate moiety cleavage by the histidine triad nucleotide-binding protein1 and the last step is the phosphorylation of the compound by the pyrimidine nucleotide pathway to form the final pharmacologically active triphosphate compound [4, 5]. The main inactive metabolite (GS-331007) resulting from the hydrolytic cleavage is excreted by renal elimination. However, cytochrome P450 isozymes are not likely to affect Sofosbuvir disposition [6].

One of the possible reasons for the limited oral absorption of Sofosbuvir is its efflux by the intestinal P-glycoprotein (P-gp) affecting its oral bioavailability. P-gp efflux transporters are expressed in different regions of the body as intestine, kidneys, liver, and brain. The intestinal P-gp is expressed on the apical membrane of enterocytes bind to substrates recycling the drugs back to the intestinal lumen lowering its intestinal permeability[7]. Several methods have been utilized to study the effect of P-gp on the intestinal absorption of drugs ranging from in vitro to in vivo methods [8, 9]. Although in vitro methods are associated with huge amount of results, these methods failed to define the effect of P-gp transporters in different intestinal segments [10].

Single pass in situ intestinal perfusion technique using animal models was developed as an alternative strategy to measure the intestinal membrane transport mechanism(s) of the drug, quantitative evaluation of factors affecting drug absorption reflecting the extent of the drug absorption and monitor the influence of efflux transporters on the drug absorption [9, 11].

The in situ technique has many advantages over the other methods as it provides input control, conserves tissue viability with complete blood and nerve supply, and excludes the influence of gastric residence time and food on the drug absorption with flexibility of choice a suitable animal model [8, 12, 13]

Literature reviews show that P-gp is expressed unequally through the intestinal segments with the lowest amount in duodenum, increasing in jejunum and expressed with the highest amount in ileum [14–17]. P-gp is present in large intestine mainly in the central part but the total amount is less than that is present in the small intestinal segments [18]. As a result, studying the regional absorption of Sofosbuvir from different regions of the rabbit intestine as well the effect of verapamil as a p-gp inhibitor on the regional absorption can give an idea about the susceptibility of Sofosbuvir for efflux transporters. But this needs to be confirmed and the current study is the first study to examine this supposition.

Accordingly, this research aims to study the luminal stability of Sofosbuvir, investigate the regional absorption of Sofosbuvir from different anatomical sites of rabbit intestine as well as the influence of verapamil on such absorption.

Materials and methods

Materials

Sofosbuvir was obtained from PHARCO Pharmaceuticals Company (Alexandria, Egypt). Sodium hydroxide was purchased from Fisher chemical (England, UK). Normal saline (Sodium chloride solution 0.9% w/w) was obtained from El-Nasr Pharmaceutical Chemicals Company (Cairo, Egypt). Fexofenadine was a gift from Sanofi-Aventis (Cairo, Egypt), Methanol HPLC grade was ordered from Germany and Fisher chemical (England, UK). Monobasic ammonium phosphate was purchased from Oxford laboratory (Mumbai, India). Chlorpromazine HCl injection (25 mg/ml) was a product from Misr Pharmaceutical Company (Cairo, Egypt). Ketamine HCl injection (50 mg/ml) was produced by Sigma-Tec Pharmaceutical Industries (Giza, Egypt).

Methods

Chromatography

Sofosbuvir was quantified using a high-performance liquid chromatography (HPLC) system (System controller, Shimadzu, Japan) equipped with a variable wavelength UV detector (SPD- 10VP, Shimadzu, Japan), and an automatic injector was used in this study (SIL-10AD VP, Shimadzu, Japan), The separation was accomplished with a 150 mm X 4.6 mm, Thermo Scientific Hypersil BDS C18 column with stationary phase average particle size of 5 µm (Walthman, MA, USA). The peak areas were calculated utilizing the data analysis program Class VP 2015. Fexofenadine was used as internal standard at a fixed concentration (227µg/ml) in each sample. The mobile phase was a mixture of methanol with water (50:50%v/v) at pH 6 at a flow rate of 1.5 ml /min and wavelength was monitored at 262 nm.

Preparation of drug solutions for calibration curves

Drug solutions were prepared with normal saline by spiking known amounts of Sofosbuvir and Fexofenadine (internal standard) into a clean set of test tubes to produce Sofosbuvir concentrations of (5, 50, 100, 300, 500, 600 and 700 µg/ml). The internal standard was used at concentration of (227 µg/ml).

Assay validation

Linearity and range

The linearity is calculated from standard curves by plotting the drug / internal standard peak areas ratios versus drug concentration

Accuracy

Accuracy of the assay is defined as the intimacy to the true reference value and expressed as the percent recovery in relation to the nominal values.

Precision of the assay:

The precision of the assay is represented as Relative Standard Deviation (RSD %), it is calculated by dividing the standard deviation by the corresponding standard drug concentration and the resulting value is expressed as percentage.

Limit of detection

It was calculated from the drug calibration curves utilizing the following equation:

$${}^{\prime }\mathrm{LOD}={3.3}^{*}\left({\mathrm{SD}}_{\mathrm{intercept}},/,\mathrm{slope}\right)\cdots {\left(\mathbf{I},\mathbf{C},\mathbf{H},.,1996\right)}^{\prime }$$

Limit of Quantitation (LOQ)

The limit of quantitation is the lowest drug amount in a sample that the assay can quantify precisely & accurately (BP., 2007). And it is determined by signal: noise ratio of 10:1 or from the calibration curves using this equation:

$${}^{\prime }\mathrm{LOQ}={10}^{*}\left(\mathrm{SDintercept},/,\mathrm{slope}\right)\cdots {\left(\mathbf{I},\mathbf{C},\mathbf{H},.,1996\right)}^{\prime }$$

Experimental set up

Eighteen rabbits (with an average weight of 2 kg) were used for this study. The rabbit was designated as the animal model depending on the highly matched characteristics concerning the physiology and the anatomy of the gastrointestinal tract between rabbit and human. Animal handling and the study treatments were performed according to the permission of the Ethical Committee of College of Pharmacy, Tanta University (Approval number, 108,017).

Drug perfusion solutions preparation

Sofosbuvir was accurately weighed and dissolved in normal saline to prepare a perfusion solution containing 50µg/ml of Sofosbuvir in absence and presence of Verapamil (60µg/ml). The selection of the drug concentration was based on the daily dose of Sofosbuvir and the average fluid volume present in the gastrointestinal tract.

Surgery and segment preparation

The surgery and segment preparation were performed according to well-known methodology [19–21]. The rabbits under study were left fasted overnight with free supply of water. In the same day of the experiment, the rabbit was injected with chlorpromazine HCl (2 mg/kg) as a muscle relaxant then 15 min later, the rabbit was anaesthetized by ketamine HCl divided into two sequential doses (50µg/kg) separated by 15 min time interval. Additional smaller dose of the anesthetic agent (25 mg/kg) was given when needed. All the experiment was performed in a thermo-controlled environment. The anaesthetized rabbit was settled in a supine position. After shaving the abdominal region, the abdominal skin was incised with a 6–8 cm midline abdominal cut. The desired intestinal segment was extracted, and the required length was measured then tied by surgical thread followed by cannulation from both sides. This defines the segment and excludes the influence of food while preserving the regional blood supply and tissue viability. The cannulated segment was rinsed with warm normal saline for cleansing from any food debris and intestinal remains. The length was measured according to the desired intestinal segment with length of 15, 30, 30 and 10 cm for duodenum, jejunum, ileum and ascending colon, respectively. In order to examine the luminal stability of the drug, normal saline was perfused alone through the different intestinal segments in order to gently extract the intestinal enzymes of the mucosal surface in the perfusate. This perfusate was collected and used as a dispersion medium for preparing the drug at concentration of 50 µg/ml. Known volumes of the drug containing perfusate were taken at predetermined time intervals at 0, 2, 5, 30, 60, 90 min. The samples were immediately immersed in iced water bath in order to inhibit any enzymatic activity. The samples were filtered using 0.22 ml syringe filter and the drug concentration in each sample was determined by the developed HPLC method. The concentration of the drug remaining in each sample was plotted versus time.

In order to investigate the membrane transport parameters of the drug through different segments of rabbit intestine, the confined ligated segment was perfused with a freshly prepared drug solution at a constant rate of (0.27 ml/min) for 2 hours while collecting perfusate samples at intervals of 10 minutes. The perfusion solution was pumped through a pumping Harvard Apparatus (Millis, MA, USA). Each perfusate sample was weighted and recorded to help in net water flux calculation by subtracting the recorded sample volume from the expected sample volume. The collected perfusate samples were centrifuged and Sofosbuvir concentration was determined using the developed HPLC assay method. This was performed by adding specified volume of the sample into clean test tubes containing definite amount of the internal standard then the samples were vortexed before HPLC assay.

Data analysis

The intestinal in situ technique is a useful tool that helps in defining the absorption mechanisms of drugs. Also, the data generated by the in situ studies are used for water flux determination. Literature reports had documented the equations used to calculate the absorption parameters [13, 19, 21–24].The following section will present the equations utilized in data analysis generated by the in-situ method.

Absorptive clearance

Sofosbuvir concentration was quantified in each perfusate sample and adjusted regarding to the water flux to determine the definite drug amount present in each sample (C_out). The fraction remaining of the drug after perfusion is calculated from the ratio between (C_out) and the drug concentration in the perfusion solution entering the intestinal segment (C_in).

The fraction remaining during the steady state {(C_out/C_in) ss} is calculated by taking the mean of the fractions remaining in the last six perfusate samples. The absorptive clearance (Pe.A) is calculated by using the following equation and expressed as ml/min:-

$${}^{\prime }\mathrm{Pe}.\mathrm{A}=-{\mathrm{Q}}^{*}ln{\left\{{\left({\mathrm{C}}_{\mathrm{out}},/,{\mathrm{C}}_{\mathrm{in}}\right)}_{\mathrm{ss}}\right\}}^{\prime }$$ 1

Pe represents the apparent permeability coefficient (cm/min), A is the effective surface area (cm²), and the average flow rate of the perfusate through the target segment is expressed as Q and its unit is (ml/min).

The anatomical reserve length (ARL)

It is the intestinal length remaining after drug absorption is complete. The next equation is utilized to calculate the (ARL):

$${}^{\prime }\mathrm{ARL}=\left({\mathrm{L}}^{*}\right)-{\left(1^{*}\right)}^{\prime }$$ 2

L* is the intestinal segment's anatomical length (cm) and l* is the intestinal length through which the drug is absorbed completely (cm).

Practically, it is difficult to consider the remaining fraction to be zero due to the logarithmic situation. As a result, 5% fraction of the remaining drug is taken as criteria for approximate complete drug absorption.

The length required for complete drug absorption (L95%) is calculated by the following equation:

$${}^{\prime }0.05={exp}^{-\left(\mathrm{Pe},.,\mathrm{A},.,{\mathrm{l}}^{*},/,\mathrm{Q}\right),}$$ 3

Pe.A is the absorptive clearance normalized to intestinal length and l* is L95% for the drug.

Influence of solvent drag on the intestinal absorptive clearance

It is crucial to examine the water flux effect on the drug absorption when studying the mechanism of the drug transport through the intestinal membrane. It is attained by plotting the absorptive clearance per unit length (Pe.A/L) as a function of net water flux normalized to intestinal length (Jw/L) that is determined from the difference between the expected sample volume within a certain time interval (Q_in) and the perfusate volume obtained from the intestinal segment in the same time interval (Q_out).

The net amount of drug absorbed per unit time relies on the involvement of two pathways of drug transport which are the convective paracellular and the diffusive transcellular processes. The solute flux at the steady state (J_ss) that is determined in (µg/min) is calculated from the following equation [25, 26]:

$${}^{\prime }{\mathrm{J}}_{\mathrm{ss}}=\frac{{\mathrm{DAK}}_{\mathrm{p}}}{\mathrm{\Delta }\mathrm{x}}(Css)+\varnothing \mathrm{s}{\mathrm{J}}_{\mathrm{w}}{(Css)}^{\prime }$$ 4

where J_ss is solute flux at the steady state (µg/min), D is the drug diffusion coefficient, K_p symbolizes the (octanol/water) partition coefficient of the drug, A represents the effective surface area of drug absorption, Δx denotes the length of path and C_ss is the length averaged steady state solute concentration in the gastrointestinal lumen (µg/ml), Ø_s is the sieving coefficient of the given drug and J_w is the net water flux. The diffusive process is represented by the part of equation {DAK_p/ Δx (C_ss)} at which (DAK_p/ Δx) is the drug diffusive permeability coefficient, while the convective process is represented by (Ø_s Jw Css), at which Ø_s is the sieving coefficient of the given drug.

Accordingly, rearrangement of the equation (4) yields

$${}^{\prime }{\mathrm{J}}_{\mathrm{ss}}/{\mathrm{C}}_{\mathrm{ss}}={\mathrm{DAK}}_{\mathrm{p}}/\mathrm{\Delta }\mathrm{x}+{\varnothing }_{\mathrm{s}}{\mathrm{J}}_{\mathrm{w}}^{\prime }$$ 5

The term J_ss/C_ss is the whole Pe.A of the solute (ml/min) that is attained by different transport pathways and it is practically determined as the product of permeability surface area Pe.A.

Statistical analysis

Sofosbuvir membrane transport parameters were statistically analyzed to examine the significance of the recorded data differences. This achieved by Kruskal–Wallis test with multiple pairwise comparisons using the two tailed Conover-Iman procedure for comparison between individual segments and groups. Significance was considered when P-value is less than 0.05.

Results

Chromatographic Analysis

Sofosbuvir was eluted after 7.98 min. The calibration curve was linear within concentration range of 0.5 μg/mL to100 μg/ ml. The standard curve equation was y = 0.0261x -0.00389 with correlation coefficient of 0.9999. Accuracy of the assay was represented by the percent recovery for the studied drug concentrations. The back calculated concentration for intraday quality control concentrations ranged from 98.02% to 101.76%, and the back calculated concentration for inter-day ranged from 97.02% to100.88% of the nominal values. Within day and interday precision of the assay was expressed by the relative standard deviation (RSD). The RSD was in the range of 1.046% to 2.657% for the within day analysis and was in the range of 0.942% to 1.854% for the inter-day results. The limit of detection determined from the drug calibration curves was 0.17µg/ml and the limit of quantitation was 0.5µg/ml.

Investigation of the luminal Stability of Sofosbuvir

The luminal stability of Sofosbuvir was examined to assess the possibility of its premature degradation by the intestinal enzymes in the brush border membrane layer. This was performed by monitoring the drug concentration at predetermined time intervals in the perfusate containing extract of the intestinal enzymes of mucosal cells. The drug concentration analysis demonstrated a significant loss of the drug due to its enzymatic degradation by the rabbit intestinal extraction through different intestinal segments.

The drug incubation with the perfusate collected from the intestinal lumen containing the extracted metabolizing enzymes resulted in extensive degradation of the prodrug. After 90 min of the incubation, the amount remaining of the drug in the perfusate collected from duodenum, jejunum, ileum, and ascending colon was 41%, 55%, 55% and 73%, respectively (Fig. 1).

Fig. 1.

Percent Sofosbuvir remaining concentration versus time during incubation in intestinal perfusate fluids extracted from different rabbit intestinal segments (The results are represented as the mean + SD, n = 3)

Membrane transport parameters of Sofosbuvir in the rabbit intestine

The intestinal membrane transport study includes the calculation of the membrane transport parameters of Sofosbuvir during the steady state. The last six samples of the second hour of perfusion represented the steady state. The membrane transport parameters of Sofosbuvir through the different intestinal segments are presented in Table 1.

Table1.

Membrane transport parameters of Sofosbuvir alone and with verapamil

Segment	PeA/L (ml/min.cm)	R(out)/R(in)	PeA (ml/min)	L (95%ab)		ARL (cm)	JW (ml/min)	JW/L (ml/min.cm)
Sofosbuvir alone
Duodenum	0.0264 (0.0085)	0.1315 (0.082)	0.5403 (0.18)	29.58 (9.5)	-9.58 (9.5)		0.067 (0.03)	0.003 (0.001)
Jejunum	0.0075 (0.001)	0.43 (0.06)	0.226 (0.03)	128.478 (32.2)	-8.478 (32.2)		0.070 (0.01)	0.0023 (0.0003)
Ileum	0.0026 (0.0005)	0.803 (0.15)	0.0771 (0.02)	949.2 (122.4)	-889.2 (122.4)		0.106 (0.0017)	0.0035 (0.00005)
Colon	0.054 (0.01)	0.085 (0.03)	0.59 (0.04)	13.63 (2.9)	1.37 (2.9)		0.11 (0.01)	0.0098 (0.0017)
Sofosbuvir with verapamil
Jejunum	0.013 (0.005)	0.216 (0.12)	0.3 (0.15)	60.4 (24)		59.6 (24)	0.091 (0.02)	0.003 (0.00)
Ileum	0.011 (0.003)	0.235 (0.1)	0.342 (0.11)	64.01 (19.75)		-4.04 (19.75)	0.101 (0.00)	0.003 (0.00)

Values between parentheses are SD, n = 3. Pe.A/L is the absorptive clearance normalized to intestinal length, Rout/Rin is the fraction remaining to be absorbed, Pe.A is the overall absorptive clearance, L95% is the length required for 95% absorption, ARL is the anatomical reserve length, JW is the water flux and JW/L is the water flux normalized to intestinal length

The absorptive clearance normalized to intestinal length was plotted versus time as represented in Fig. 2. This plot assures the steady state pattern of the results. The plot of the Pe.A per unit length during the steady state reflects the descending pattern of data through the different intestinal segments which was ordered as colon > duodenum > jejunum > ileum. The results show the dependence of the intestinal absorption of Sofosbuvir on the absorption site throughout the rabbit intestine. As we move distally through the rabbit small intestine, the absorption extent decreases with the highest absorption from the duodenum and lowest in the ileum. However, the absorption increases again from the colon with the highest value throughout the rabbit intestine. According to the statistical analysis of the results, there was significant differences in the absorption of Sofosbuvir among the tested segments (P-value < 0.05). This rank was assured by the reverse order of L95% in the selected intestinal segments (Table 1). The drug was incompletely absorbed from the small intestine as reflected by the negative values of the ARL (Table 1).

Fig. 2.

The absorptive clearance of Sofosbuvir per unit length (PeA/L) from different intestinal segments during the steady state. (Data are represented as mean + SD, n = 3)

The water flux per unit length was calculated and the recorded data revealed that there were no statistically significant differences between the three segments of the rabbit small intestine; on the other hand, the water flux per unit length of the colon was significantly higher than that of the small intestinal segments (P-value < 0.05) (Table 1).

The effect of water flux on the absorptive clearance of Sofosbuvir in each segment was studied by plotting the absorptive clearance per unit length versus the net water flux normalized to intestinal length. Table 2 shows the regression data of these plots. The intercept of the plots was used to determine the relative participation of the transcellular pathway to the overall absorptive clearance. The paracellular contribution was calculated from the difference between the overall absorptive clearance and the percentage of Sofosbuvir absorbed by the transcellular pathway. A positive contribution is indicated if the slope of the regression line fitted to the data of the absorptive clearance as a function of water flux was significantly different from zero.

Table 2.

Influence of water flux (ml/min.cm) on overall absorptive clearance (ml/min.cm) both normalized to intestinal length of Sofosbuvir alone and with verapamil. Regression parameters are determined by fitting data to Eq. (5)

Regression parameters	Intercept (DAKp/Δx)	Slope (∅)	% Transcellular	% Paracellular
Control
Duodenum	0.01617** (0.004)	3.153** (0.99)	61%	39%
Jejunum	0.0066** (0.003)	0.389* (1.07)	88%	12%
Ileum	0.0022* (0.001)	0.117* (0.39)	86%	14%
Ascending colon	0.0166* (0.014)	3.75** (1.38)	31%	69%
Verapamil
Jejunum	0.011** (0.003)	0.55* (0.88)	86%	14%
Ileum	0.011** (0.002)	0.178* (0.63)	88%	12%

Values between parentheses are SE, n = 3. (DAKp/ x) represents the permeability coefficient and (∅s) is the sieving coefficient of the given drug. ** Significantly different from zero (P < 0.05). * Not significantly different from zero (P > 0.05)

The intercept obtained from the linear regression equation was significantly different from zero in case of duodenum and jejunum; however, the intercept was not significantly different from zero in case of ileum and ascending colon. 61% of Sofosbuvir was transported by the transcellular pathway in the duodenum (P-value < 0.05). For the jejunum the relative involvement of the transcellular route increased to 88%. The absorption of Sofosbuvir was insignificant by either pathway from the ileum (P-value > 0.05). The drug was absorbed predominantly by the paracellular pathway in the ascending colon which contributed to 69% of the total amount absorbed of the drug (Table 2, Fig. 3).

Fig. 3.

The relative contribution of transcellular and paracellular pathways in Sofosbuvir absorption from rabbit small and large intestinal segments

Effect of verapamil on membrane transport parameters of Sofosbuvir from Jejunum and Ileum of the rabbit in situ

The effect of co-perfusion of verapamil as a p-glycoprotein inhibitor on the membrane transport parameters of Sofosbuvir was studied in jejunum and ileum of rabbit intestine as shown in Table1.

Co perfusion of verapamil with Sofosbuvir modified the membrane transport parameters of the drug. This was evidenced by the significant enhancement of the Pe.A/L values in both segments in comparison to the results obtained when Sofosbuvir was perfused alone (P-value < 0.05) (Table 1, Fig. 4). This effect was further verified by the significant decrease in the L95% values in both jejunum and ileum (P < 0.05) when verapamil was co-perfused with Sofosbuvir as compared to the values obtained after perfusion of Sofosbuvir alone in the tested segments (Table 1).

Fig. 4.

The effect of Verapamil on the absorptive clearance per unit length of Sofosbuvir in jejunum and ileum. (Data are presented at steady state as mean + SD, n = 3)

Regarding the transport pathway of Sofosbuvir from jejunum and ileum, co-perfusion of verapamil with Sofosbuvir changed the intestinal transport of the drug to become significantly absorbed by the transcellular route (P-value < 0.05) in the ileum. However, there was non-significant change in the comparative contribution of the absorption pathways in the jejunum after co-perfusion of verapamil (Table 2).

DISCUSSION

Sofosbuvir, a parent nucleotide analogue inhibitor, is used combined with other antiviral drugs in the treatment of liver diseases caused by Hepatitis C Virus (HCV). The antiviral drug suffers from poor intestinal absorption due to its poor permeability with subsequent poor oral bioavailability [1, 2]. This research aims to study the luminal stability of Sofosbuvir. Another objective of the study was to examine the membrane transport parameters of Sofosbuvir from different anatomical sites in the rabbit utilizing, through and through, intestinal perfusion technique to assess the permeation data of Sofosbuvir and investigate the effect of verapamil, a published P-gp inhibitor, on its membrane transport parameters. The rabbit was chosen as the animal model because of the large physiological resemblance of the animal intestine to that of the human regarding the average lipid to protein ratio along the gastrointestinal tract and change of surface area from one site to another [27, 28].

In order to study the luminal stability of Sofosbuvir, the drug was incubated in the fluid extracted from the rabbit small and large intestinal segments. The stability study showed that Sofosbuvir was subjected to premature degradation with varying fractions degraded from the different intestinal segments (Fig. 1). The degradation of the drug in the different segments could be attributed to either chemical or enzymatic instability. Varieties of metabolizing enzymes are secreted in the intestinal lumen. Among the enzymes responsible for the premature degradation of prodrugs are Carboxyl esterases (CES1 and CES2) that are responsible for the enzymatic hydrolysis of prodrugs. Since Sofosbuvir is an ester prodrug, it can be subjected to premature hydrolysis by CES. These results are in accordance with previous studies which demonstrated that Sofosbuvir was found to be CES1 substrate and CES2 inhibitor as well that could be used to enhance the oral bioavailability of other substrate prodrugs [29–32]. The premature degradation of the prodrug results in liberation of the parent drug that is less lipophilic than its corresponding prodrug. Also, the existence of CES1, 2 in the rabbit intestine was reported in previous studies [33–36]. In addition to the presence of CES in the colon [37–39]. Accordingly, this can explain the resulted hydrolysis of Sofosbuvir after incubation of the prodrug in the fluid extracted from the different segments.

In order to examine the regional absorption of Sofosbuvir, the drug was perfused in the rabbit small and large intestine at concentration of 5µg/ml. This concentration was determined depending on the oral drug daily dose and the mean fluid volume present in the gastrointestinal lumen; this method was reported in previous in situ studies [40–42].

Sofosbuvir was incompletely absorbed from the rabbit small intestinal segments as reflected from the negative values of the ARL. This is in accordance with the documented limited oral bioavailability of the drug since Sofosbuvir is a BCS class III prodrug suffering from poor membrane permeability [1, 2]. The fraction remaining to be absorbed (Rout/Rin) was used to calculate the absorptive clearance (Pe.A) of Sofosbuvir from the rabbit small and large intestinal segments. The calculated absorptive clearance per unit length (Pe.A/L) of Sofosbuvir showed that the drug absorption was reliant on the anatomical site throughout the rabbit intestine. The Pe.A/L values from the different segments were ranked as ascending colon > duodenum > jejunum > ileum with significant differences among them (P-value < 0.05) (Table 1). Many factors can account for this order of the absorptive clearance. One of the factors affecting the intestinal absorption is the available surface area for drug absorption from different segments. Previous studies demonstrated that the order of surface area of the different segments was jejunum > duodenum > ileum > ascending colon [43, 44]. However, the reported findings of the Pe.A/L were not in accordance with the rank of the intestinal surface area order. As a result, the surface area change is not the main factor influencing the regional absorption of Sofosbuvir especially with the recorded value of Pe.A/L in the ascending colon being the highest among the segments under study.

The relative involvement of transcellular and paracellular pathways in transporting Sofosbuvir can be taken into consideration in the explanation of the regional absorption of the drug from the different anatomical sites. The reported findings demonstrated that Sofosbuvir was predominantly absorbed by the transcellular pathway from duodenum and jejunum (Table 2, Fig. 3). Consequently, any factor influencing the transcellular transport can affect the absorptive clearance of Sofosbuvir from the small intestine. Regarding the transport pathway of Sofosbuvir from the ascending colon, the drug was mostly absorbed by the paracellular pathway as the recorded water flux per unit length (Jw/L) in the colon was significantly higher than that of the other small intestinal segments. The possible reason for this is the physiological function of the colon as it absorbs mainly water and sodium and secrets potassium and bicarbonates [45–48].

The different levels of expression of P-gp are supposed to have a great influence on the drug absorption. The documented site dependent distribution of the P-gp amount is inversely correlated with the measured overall Pe.A of Sofosbuvir in each segment. It is documented that the amount of P-gp expression in the different intestinal segments is ranked as duodenum < jejunum < ileum [14–16, 49]. This rank is the opposite to the order of the absorptive clearance of Sofosbuvir from the small intestinal segments as the drug absorption was higher in duodenum than jejunum. However, the least absorption of Sofosbuvir was from the ileum. Taking these findings, P-gp efflux transporters can have a significant influence in determining the regional absorption of Sofosbuvir from the different intestinal segments. Regarding the presence of P-gp in the large intestine, it is reported that P-gp is mainly present in the central part. However, the overall amount is less than that expressed in the small intestinal segments [18]. Regarding Sofosbuvir absorption from the ascending colon, it was mainly by the paracellular convective pathway with the influence of solvent drag dominating.

The influence of P-gp efflux on Sofosbuvir absorption through the different intestinal segments was further confirmed by studying the effect of co perfusion of verapamil which is a known P-gp inhibitor on Sofosbuvir absorption from jejunum and ileum where P-gp is expressed with abundant amount in both segments reaching its highest magnitude in the ileum [14–16, 49]. Verapamil, a calcium channel blocker [50], is a commonly taken medication in poly medicated HCV patient with arrhythmia and hypertension. Verapamil significantly modulated the perfusion data of Sofosbuvir in jejunum and ileum in comparison with the results obtained when Sofosbuvir was perfused without verapamil (Table 1, Fig. 4). The significant enhancement in the absorptive clearance per unit length of Sofosbuvir is accompanied by a significant decrease in the L95% values when co-perfused with verapamil suggests the inhibitory influence of verapamil on the efflux transporters where P-gp is highly expressed in jejunum and ileum [14–16, 49]. This hypothesis is supported by the reported inhibitory influence of verapamil on P-gp transporters [30, 51, 52]. The reported results of the obtained rank of Sofosbuvir absorption from the small and large intestinal segments together with the enhancing influence of verapamil on Sofosbuvir absorptive clearance from jejunum and ileum suggests the important role of P-gp on determining the permeation parameters of Sofosbuvir. Accordingly, P-gp inhibitors can be used as absorption enhancers to increase the intestinal absorption of Sofosbuvir which suffers from P-gp efflux. Optimization of the intestinal absorption can improve the oral bioavailability of the drug offering an opportunity for dose reduction with subsequent minimization of adverse effects.

Conclusion

The intestinal absorption of Sofosbuvir was site dependent and associated with the content of P-glycoprotein with reduced absorption in segments having higher P-gp concentration. This was confirmed by studying the influence of verapamil as an inhibitor for the P-gp on Sofosbuvir absorption which led to significant enhancement in the absorption of the drug from regions having high P-gp content. This study demonstrated the main factors controlling the regional absorption of Sofosbuvir and the possible drug interactions that can occur in polymedicated HCV infected patients receiving Sofosbuvir.

Author contributions

Nada M. Mohsen: Investigation, Data curation and Writing drafted manuscript. Esmat E. Zein El-Din: Visualization, Supervision, Writing, Reviewing and Editing. Mohamed A. Osman: Conceptualization, Methodology, Data curation, Visualization and Supervision. Shimaa M. Ashmawy: Data Curation, Visualization, Writing, Reviewing and Editing.

Funding

No funding was received for conducting this study.

Data availability

All authors are sure that all data and materials support our published claims and comply with field standards. And all listed authors have approved the manuscript before submission, including the names and order of authors.

Ethical approval.

Animal handling and the study treatments were performed according to the permission of the Ethical Committee of College of Pharmacy, Tanta University (Approval number, 108017).

Declarations

Conflicts of interest/Competing interests

The authors have no relevant financial or non-financial interests to disclose.

Consent to participate

Not applicable.

Consent to publish

Not applicable.