Comparison of effects of aminosalicylic acid, glucocorticoids and immunosuppressive agents on the expression of multidrug-resistant genes in ulcerative colitis

To compare the effects of aminosalicylic acid, glucocorticoids and immunosuppressants on the expression levels of multidrug resistance genes in patients with ulcerative colitis (UC), with the aim of providing a theoretical and therapeutic basis for the diagnosis, treatment, and prevention of UC. Fresh colonic mucosal tissues or postoperative pathological biopsies from 148 UC patients were collected, and the distribution sites and morphology of P-glycoprotein (P-gp) were detected using immunohistochemical staining. RT-PCR was used to quantify the expression levels of multidrug resistance gene (MDR1) mRNA before and after the corresponding treatment, and the effects of aminosalicylic acid, glucocorticoids and immunosuppressive drugs on P-gp were compared. In addition, the effects of the three drugs on MDR1 mRNA were analyzed. Administration of 5-aminosalicylic acid (5-ASA) drugs did not correlate with MDR1 expression in UC, whereas administration of glucocorticoids and immunosuppressive drugs was positively correlated with MDR1 expression profile. The expression levels of MDR1 mRNA and its product P-gp were significantly upregulated in patients who did not respond to glucocorticoids and immunosuppressive drugs. 5-ASA had no effect on the expression levels of MDR1 and its product P-gp in patients with a confirmed diagnosis of UC. However, the use of glucocorticoids and immunosuppressants can increase the expression level of MDR1.

Diagnostic criteria. The diagnosis of UC can be confirmed by clinical manifestations, such as diarrhea, mucopurulent blood, with abdominal pain, defecation and various degrees of systemic symptoms; persistent, recurrent episodes; duration of UC > 1 month, colonoscopic observation of distal colon such as rectum and sigmoid colon with continuous, diffuse distribution, mucosal vascular congestion, hemorrhage and blurred texture, etc. 1 month, colonoscopic observation of rectum, sigmoid colon and other distal colon with continuous Diffuse distribution, mucosal vascular congestion, hemorrhage, edema, blurred, disorganized or absent texture, purulent secretion adhesions, barium enema examination including jagged or burr-like intestinal margins, multiple small filling defects in the intestinal wall, coarse mucosa with/ or granular changes, shortening of the intestine, disappearance of the intestinal pouch into a lead tube, pathological examination. Exclusions were those with bacillary dysentery, amebic dysentery, chronic schistosomiasis, intestinal tuberculosis and other infectious colitis, colonic Crohn's disease (CD), ischemic colitis, and radiation colitis.
Inclusion criteria. Those who did not receive ASA medication, oral or intravenous glucocorticoids, glucocorticoids plus immunosuppressants and herbal medicine 4 weeks before the start of the study; those who received oral ASA medication after treatment: oral or intravenous glucocorticoids; oral or intravenous glucocorticoids plus immunosuppressants, ASA medication and glucocorticoids and other topical treatment medications.
Prognosis and outcome. All patients were divided into effective (complete remission and effective), ineffective and control groups according to their prognosis and efficacy. In the complete remission group, the patients' symptoms disappeared, and the mucosa was normal on colonoscopy. In the effective treatment group, the patients' physical symptoms of colitis largely disappeared, and diarrhea and abdominal pain decreased; colo-  Table 3. Sutherland disease activity index. A total score below 2 was considered symptom relief; between 3 and 5 was mild activity; 6-10 was moderate activity; 11-12 was severe activity. www.nature.com/scientificreports/ noscopy showed mild inflammation of the mucosa or pseudo-polyps formation; pathological sections showed restoration of the mucosal layer of colonic tissue and reduction of neutrophil infiltration.
In the ineffective treatment group, patients had no relief of symptoms after treatment and frequent diarrhea and abdominal pain; colonoscopy showed severe inflammation of the mucosa and breakdown of the mucosal layer; pathological sections showed defects in the mucosal layer of colonic tissue and increased neutrophil infiltration.
Specimen preparation. Fresh colonic mucosal tissue specimens or endoscopic pathological mucosal tissues or postoperative pathological biopsies were immediately stored in a refrigerator at − 80 °C for RNA extraction. The remaining specimens were adequately fixed in 10% formalin, routinely dehydrated, and embedded in paraffin, and 4-μm tissue sections were prepared for immunohistochemical staining (P-gp).
Immunohistochemical staining. Immunohistochemical staining method. Using the SP method, antigen repair was required for the determination of P-gp. The negative control was replaced with phosphate-buffered saline (PBS) buffer, and the rest of the procedure was performed according to the manufacturer's instructions. p-gp positive control was provided by Beijing Zhongshan Biotechnology Co. Positive controls were set up for each antibody. Routine dewaxing was hydrated using xylene twice for 15 min each, hydrated with gradient alcohol, washed three times with PBS (pH = 7.4) for 5 min each, and incubated with 3% hydrogen peroxide solution for 15 min at room temperature to eliminate endogenous peroxidase. Then, the samples were washed with distilled water for 5 min for three times. For antigen repair, 10 ml of ethylenediaminetetraacetic acid (EDTA) antigen repair solution was collected and diluted with 490 ml of triple distilled water for 3-4 min of hot repair. Finally, samples were cooled to room temperature, washed three times with PBS for 5 min each, sealed with normal goat serum working solution, and incubated at 37 °C for 40 min. The primary antibody was added to the wet box overnight at 4 °C, rewarmed at 37 °C for 45 min, and washed three times with PBS for 5 min each. The labeled secondary antibody was added to biotin drops and incubated at 37 °C for 25 min, washed three times with PBS, and horseradish peroxidase-labeled streptomycin was incubated at 37 °C for 20 min, washed three times with PBS for 5 min. The color development reaction was observed under a microscope with a chromogenic solution of 3,3-N-diaminobenzidine tetrahydrochloride (DAB). Samples were rinsed with running water, gently re-stained with hematoxylin for 10-30 s, dehydrated with a gradient of alcohol, clear xylene, and then mounted with a neutral mounting sheet.
Evaluation of immunohistochemical staining results. Using DAB immunohistochemical color development system, positive results were shown as brownish-yellow particles in the corresponding positions. p-gp positive particles were mainly distributed in the lamina propria of colonic mucosa and intestinal epithelium, and a distinct brownish-yellow color in the membrane and cytoplasm was considered positive. Ten pathological cells were randomly counted in the high magnification field. the number of P-gp treated positive cells was < 10% as negative, 10-25% as (+) and 25-75% as (+++). This refers to the criteria provided by Zhongshan Biotechnology Co.

Reverse transcription polymerase chain reaction (RT-PCR). Primer. Primers designed by Primer
Premier 5.0 were used as reference 15 . Primers for MDRI and β-actin were synthesized by Beijing Sebring Bioengineering Co.
MDR1-specific primers. Extraction of total tissue RNA. Fresh specimens collected by colonoscopic biopsy or post-surgical tissue biopsy were immediately stored at − 80 °C in a low-temperature refrigerator to detect the expression level of MDR1.
Trizol extraction 16 . Harvest frozen tissue samples and grind to a powder and transfer to pre-cooled 1.5 ml EP tubes before evaporation of liquid nitrogen. Add LML Trizol, insert into a microtissue homogenizer for 2-3 min, then add chloroform and shake for 15 s, allow to stand for 5 min, then centrifuge at 12,000 rpm for 15 min at 4 °C. The aqueous component of the upper layer was carefully collected into another new EP tube. Pre-chilled isopropanol at − 20 °C was added and mixed thoroughly with the supernatant phase. After 10 min of ice bath and centrifugation at 12,000 rpm for 15 min, the supernatant was discarded and dried, 1 ml of 75% ice ethanol was added and mixed well, the precipitate was washed well and centrifuged at 8000 rpm for 5 min at 4 °C, then the supernatant was discarded and the precipitate was dried. If to be used immediately, leave the sample at ambient temperature for 20 min and add 30 μl of DEPC water. If not for immediate use, add 75% ethanol LML and store at − 80 °C.  Fig. 1.

RT-PCR.
We used a two-step RT-PCR for quantification. The first step was cDNA reverse transcription. Vortex and shake the reagent solution from the melted kit for 2 s; transient centrifuge in a micro benchtop centrifuge for 5 s; transfer 4.5 µl of the bootstrap solution to a new 0.5 ml PCR tube; add 8 µl of RNA sample (5 µg of total RNA or 0.25 µg of POLY A + RNA); mix with a gun tip; incubate in a dry thermostat at 70 °C for 3 min; immediately place in an ice bath; add 8 µl of Reverse Solution (containing Invitrogen SuperScript III Reverse Transcriptase); mix well with the tip of the gun; centrifuge instantaneously for 5 s in a micro benchtop centrifuge; incubate for 60 min at 37 °C in a dry thermostat; place immediately in an ice bath; dilute with 80 µl of buffer and mix well; place in an ice bath or place in a − 20 °C refrigerator. Storage. The second step was cDNA amplification. Transfer 30 µl of reaction solution to a new 0.5 ml PCR tube; add primers F and R (350 ng total, respectively) and Taqase; add 2 µl of the above diluted cDNA synthesis; finally add buffer to a total of 50 µl and mix well; place in a 4 °C micro benchtop centrifuge for 5 s instantaneously; add 2 drops of mineral oil using a 1000 µl gun tip. The reaction was carried out in a PCR amplifier and the products were later subjected to agarose electrophoresis for characterization.
Identification of amplification products. A 2% agar gel (containing ethidium bromide 0.5 μl/ml) was prepared with TAE, and 10UL amplification products were sampled by electrophoresis at 100 V for 120 min. The amplification products of MDRl and B-actin were identified under UV light at 167 bp and 301 bp, respectively, which were consistent with the designed amplification fragments of MDR1 and P-actin primers.
Quantitative analysis. Two bands of MDRl and B-actin primers were scanned with a Cs-910 chromatography scanner made by Shimadzu. The length and width were 2.1 mm and 0.1 mm. The wavelength input was 550 nm. The data were entered into a computer for relative quantitative analysis, and MDR1/B-actin was the relative amount of MDR1.
Statistical analysis. The SPSS 19.0 statistical package was used for statistical analysis (SPSS Inc., Chicago).
Percentages were compared using the Chi-squire test, and t-tests were used for comparison of measurements between the two groups. p-values less than 0.05 were considered statistically significant.
Ethics approval and consent to participate. This study was approved by the ethics committee of the first affiliated hospital, and college of clinical medicine of henan university of science and technology. Informed consent for all patients (including the patients of the control group) was obtained prior to therapy and performed in accordance with the Declaration of Helsinki and Good Clinical Practice Guidelines.

Results
Results of immunohistochemical staining for P-gp in colon tissue. The particles positive for P-gp staining were mainly distributed in the lamina propria of colonic mucosa and intestinal epithelial cells, which showed a distinctive brownish-yellow color in the cell membrane and cytoplasm. In the control group, P-gp stained positively in the epithelial cells and lamina propria ( Fig. 2A,B), and in the UC colonic mucosa P-gp stained positively in both epithelial and lamina propria cells (Fig. 2C,D). with ASA, there was a significant difference in the expression rate of P-gp before and after treatment in the effective group of UC compared with the normal control group (P < 0.05). However, in the effective and ineffective groups, there was no significant difference in the positive expression rate of P-gp after treatment compared with that before treatment (all P > 0.05), as shown in Table 4 and Fig. 3.

Effect of glucocorticoids on P-gp expression in colonic tissue of UC patients.
In 53 patients treated with glucocorticoids, the rate of positive P-gp expression was significantly downregulated in the effective group compared with the control group, both before and after treatment (all P < 0.05). The P-gp positive expression rate in the inactive group was not significantly different before and after treatment (all P > 0.05). In the nulliparous group, the positive expression rate of P-gp was significantly higher than that of the control group after treatment (P < 0.05). As shown in Table 5 and Fig. 4, the positive expression rate of P-gp was significantly higher in the ineffective group than in the effective group before and after treatment (all P < 0.05).

Effect of immunosuppression on P-gp expression in UC.
In 37 cases of UC with immunosuppression, the positive expression rate of P-gp was significantly lower in the effective group than in the control group before and after treatment (all P < 0.05). However, there was no significant difference in the positive expression rate of P-gp in the effective group before and after treatment (P > 0.05). The positive expression rate of P-gp before treatment did not differ between the ineffective and control groups (P > 0.05), while it increased significantly after treatment compared with the control group (P < 0.05). In the effective group, the positive expression rate of P-gp was not significantly different before and after treatment (P > 0.05), as shown in Table 6 and Fig. 5.  www.nature.com/scientificreports/

Effect of ASA on MDR1 gene expression in UC.
In 58 UC patients treated with ASA, the expression levels of MDR1 gene before and after treatment in the effective group were 0.590 ± 0.071 and 0.514 ± 0.018, respectively, which were significantly lower (both P < 0.05) compared with the control group (1.374 ± 0.022). However, the expression levels of MDR1 gene in the effective and ineffective groups after treatment were not significantly different from those before treatment (all P > 0.05), as shown in Table 7.

Effect of glucocorticoids on MDR1 gene expression in UC.
The expression levels of MDR1 gene before and after treatment in the effective group were 0.675 ± 0.103 and 0.509 ± 0.106, respectively, which were significantly lower than that of 1.374 ± 0.022 in the control group (all P < 0.05). In the effective group, the expression levels of MDR1 gene were not significantly different before and after treatment (P > 0.05). The expression level of MDR1 in the inactive group was not significantly different from the control group before treatment (P > 0.05), and the expression level was substantially upregulated after treatment (P < 0.05). The expression levels of MDR1 gene in the inactive group were significantly higher than those in the effective group before and after treatment (all P < 0.05), as shown in Table 8.

Effect of immunosuppression on MDR1 gene expression in UC. The expression levels of MDR1
gene in the effective group before and after treatment were 0.618 ± 0.095 and 0.523 ± 0.0201, respectively, which were significantly different from 1.374 ± 0.022 in the control group (both P < 0.05). The expression levels of MDR1 gene in the effective group were not significantly different before and after treatment (P > 0.05). In the  Table 4, shown as mean values. Significant differences between groups are detailed in Table 4. www.nature.com/scientificreports/ ineffective group, the expression levels were not significantly different before treatment compared with the control group (P > 0.05), while they increased sharply after treatment (P < 0.05). The expression levels of MDR1 gene before and after treatment in the null group were statistically significant compared with the effective group (both P < 0.05), as shown in Table 9.

Discussion
The regulatory relationship between NF-κB and MDR1 has been the subject of several studies [17][18][19] . Ogretmen et al. demonstrated that a protein complex composed of NF-κB /p65 and c-Fos transcription factors interacts with the CAAT promoter region in MCF7 cells to negatively regulate human mdr1 promoter activity. It has also been reported that insulin-induced mdr1 expression is mediated by NF-κB in rat hepatoma cells 20 , NF-κB can protect renal proximal tubular cells from cadmium and oxidative stress by increasing the expression of P-gp.
Recently, it has also been reported that NF-κB is involved in TNF-α-induced mdr1 expression in hepatocytes, in 2-acetylaminofluorene-induced MDR expression in hepatocytes and in constitutive MDR expression in drugresistant cells 21 . Therefore, the effect of drugs on NF-κB during the treatment of inflammatory diseases like UC and whether it subsequently causes upregulation of MDR1 and increased drug resistance are questions that need to be addressed. For the aminosalicylates, our immunohistochemical staining and RT-PCR assays showed that the expression of both P-gp and MDRl genes before and after treatment with aminosalicylates was significantly different from that of normal controls. However, there was no statistically significant difference in the expression level of P-gp after treatment compared with that before treatment in both the effective and ineffective groups. The above data  Table 5, shown as mean values. Significant differences between groups are detailed in Table 5. Table 6. Effect of immunosuppressants on P-gp expression in UC. a P < 0.05 versus normal control. b P > 0.05 versus pre-treatment. c P < 0.05 versus pre-treatment. d P < 0.05 versus effective group pre-treatment. e P < 0.05 versus effective group post-treatment. www.nature.com/scientificreports/ Figure 5. Effect of immunosuppression on P-gp expression in ulcerative colitis. Calculated from the data in Table 6, shown as mean values. Significant differences between groups are detailed in Table 6. Table 7. Effect of ASA drugs on the expression level of MDR1 gene in UC (mean ± SD). a P < 0.05 versus normal control. b P > 0.05 versus pre-treatment. www.nature.com/scientificreports/ suggest that MDR gene expression in colonic tissues of patients with ulcerative colitis has been changed before treatment, and drug treatment does not affect MDR gene expression, i.e., it does not change the increase in MDR gene resistance or decrease its expression. Therefore, aminosalicylates can be used for long-term pharmacotherapy of UC without causing pharmacogenic tolerance. SASP is an NF-κB (P65) inhibitor with potent inhibitory effects on NF-κB (P65) activation. This has been confirmed by several studies. For example, SASP has been shown to inhibit cell proliferation by blocking NF-κB (P65) activation. And the downregulation of mRNA and protein levels of Bcl-2, Cyclin D1, MDRl and NF-κB (P65) during chemotherapy is one of the reasons for and MDR gene expression in pancreatic cancer 22 , suggesting that NF-κB negatively regulates the expression of MDR1.

Group Total (n) Relative expression of MDR1
For glucocorticoids, the expression levels of MDR1 gene and P-gp were significantly higher in the ineffective group after treatment compared to the pre-treatment and healthy controls. Not only that, the expression levels of MDR1 gene and P-gp in the ineffective group were significantly higher than those in the effective group. These results may be due to the ability of glucocorticoids to be exocytosed by P-gp.
Clinically, the effectiveness of UC treatment depends mainly on the patient's response to glucocorticoid therapy. For some patients, there is no response even after high doses of the drug, which leads to treatment delays 23 . P-gp can mediate glucocorticoid resistance through several pathways. First, P-gp can induce glucocorticoid resistance by acting on drug transport pumps and protecting cells from apoptosis. In addition, P-gp can also mediate hormone resistance by acting on the entire cellular immune system 24 . The induction of resistance by glucocorticoids during the treatment of UC has also been validated by several studies. In patients with UC without hormone therapy, P-gp expression and activity in peripheral blood, intestinal mucosal lamina propria and intestinal epithelial lymphocytes were significantly lower than in normal controls 25 . For those patients receiving hormone therapy peripheral blood mononuclear cells MDR1 mRNA expression showed a significant positive correlation with the total hormone dose, suggesting that the high expression of MDR1 mRNA in UC patients is a response induced by the application of high dose hormone therapy 26 .
For immunosuppressants, our study demonstrated that immunosuppressants can upregulate P-gp expression as well as glucocorticoids. It may be that lymphocytes act as a protective mechanism to specifically increase P-gp expression and reduce the toxic effects of drugs by external stimuli. fiedler also found that combined interventions such as glucocorticoids and spinosad also increased MDR gene as well as P-gp protein expression in rat liver and intestine. The above evidence suggests that MDR gene and P-gp protein expression are not only related to individual genetics, but that glucocorticoids and immunosuppressants themselves can also increase MDR gene and P-gp protein expression 27 . Therefore, although competitive substrate resistance reversal can improve drug efficacy in a short period of time, the high expression of MDR-induced genes and P-gp protein may be associated with drug resistance or drug dependence after long-term use, which may also be one of the reasons for relapse after drug administration in many diseases.
Other studies have similarly identified resistance induced by immunosuppressive therapy. For example, patients with lupus erythematosus with higher disease activity have increased MDR gene and P-gp protein expression in the presence of immunosuppressive agents such as cyclophosphamide and the glucocorticoid methylprednisolone 28 . Also, the different expression levels of P-gp in lymphocytes of SLE patients indicate a different requirement for immunosuppression 29 , which corroborates that immunosuppression can induce the development of drug resistance. More direct evidence is that intestinal MDR1 expression is also upregulated after tacrolimus administration 30 . Related mechanistic studies suggest that immunosuppression induces phosphorylation of AKT and ERK, leading to activation of PI3K/AKT and MAPK/ERK pathways and increased expression of MDR1/P-gp in target cells 31 . The relationship between MDR1/P-gp and the development of UC has also been continuously validated by mechanistic studies 32 .
In summary, the expression level of MDR1/P-gp must be precisely determined and monitored during the treatment of UC to avoid the increase of drug resistance caused by drug abuse, which reduces the efficacy.

Conclusion
ASA had no significant effect on the expression level of MDR gene in UC patients. The use of glucocorticoids and immunosuppressants significantly increased the expression levels of MDR genes in the colonic tissues of UC patients, and the underlying mechanisms remain to be further elucidated. Table 9. Effect of immunosuppressive agents on the expression of MDR1 gene in UC (mean ± SD). a P < 0.05 versus control group. b P > 0.05 versus pre-treatment. c P < 0.05 versus pre-treatment. d P < 0.05 versus effective group pre-treatment. e P < 0.05 versus effective group post-treatment.

Group
Total (n) Relative expression of MDR1

Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.