Comparison of olive leaf, olive oil, palm oil, and omega-3 oil in acute kidney injury induced by sepsis in rats

Induction of sepsis

Rats were intraperitoneally injected with 10 mg/kg Escherichia coli LPS (serotype 055: B5; Sigma, St Louis, MO, USA), dissolved to 0.9% in distilled water. The control (CTL) group received the same volume of distilled water only. To begin the experimental protocol, sepsis was stimulated by LPS treatment after 24 and 48 h. Rats were subsequently untreated or treated with test substances after 24 and 48 h. The treatment protocols are described in detail below.

Treatment protocol: Brazilian olive

The oil and fresh leaves of Brazilian olive trees were provided by Cerro dos Olivais, located in Picada Grande, Caçapava do Sul-RGS, Brazil, associated with the Associação Rio-Grandense de Olivicultores. A lyophilized extract was prepared from the olive leaves by Universidade de Ribeirão Preto (UNAERP/SP) using ethanol as the solvent.

Ethanol has been reported to be the most suitable solvent for preparing O. europaea L. (olive leaf extracts) with a high content of phenolic compounds (Lafka et al., 2013; Pérez-Bonilla et al., 2013).

Olive leaf extracts were analyzed at Phytobios/Centroflora Brazil, using a high-performance liquid chromatography system (Merck-Hitachi LaChrom Elite; Hitachi, Tokyo, Japan), equipped with a UV/vis detector (model L-2400 UV).

The olive leaf extract was resuspended in 10 mg/mL methanol, filtered through membrane with 0.45-μm pores (Fig. S1A), and analyzed by electrospray ionization-mass spectrometry (ESI-MS, Fig. S1B).

The phenolic compounds in Ex and Olv were evaluated using a UV/vis spectrophotometer (model SP-220; Biospectro, Curitiba, Brazil). Aliquots of the sample solution (50 μL) were mixed with 250 μL of Folin-Ciocalteau reagent (Blainski, Lopes & De Mello, 2013) and 750 μL of 20% sodium carbonate and the absorbance at 735 nm was measured after 2 h. A calibration curve was constructed using caffeic acid as the standard compound, and the results are expressed in terms of milligrams of caffeic acid/liter of sample.

The total phenol content of Olv was higher than that of Ex. The fatty acid content of Olv was analyzed by Shimadzu gas chromatography, according to the Hartman & Lago method (Mello & Pinheiro, 2012; Hartman & Lago, 1973), in which a 400-mg aliquot of the sample was converted to methyl esters using an ammonium chloride solution and sulfuric acid in methanol as an esterifying agent. The different types of fatty acids were identified by chromatography through a comparison of retention times of the samples and standards. Quantification was performed using area normalization, and the results were expressed as percentage per 100 grams of sample. The results are presented in Table S1.

Rats received 100 mg/kg of lyophilized Ex, administered by gavage daily throughout the experimental period. Using a pipette, 100 μL of Olv was carefully dispensed by dripping directly into the mouth of rats. Sepsis was induced by LPS treatment after 24 and 48 h and rats were subsequently treated with Ex, Olv, and Ex + Olv (24 and 48 h).

Treatment protocol: omega-3 fish oil

Omega-3 fish oil was directly extracted from 1,000-mg capsules containing 21.3% EPA and 11.4% DHA. Using a pipette, 100 μL of Omg was carefully dispensed by dripping directly into the mouth of rats. Sepsis was stimulated by LPS treatment after 24 and 48 h and rats were subsequently treated with Omg (24 and 48 h).

Treatment protocol: palm oil

The composition of the Pal used in this study was as follows: 42% saturated fatty acids (39–44% palmitic acid); 5–5.8% stearic acid; 43% monounsaturated fatty acids (38–46% oleic acid); 10–14% linoleic acid; 500–700 ppm carotenoids; and 600–1,200 ppm tocopherols, especially tocotrienols. Using a pipette, 100 μL of Pal was carefully dispensed by dripping directly into the mouth of rats. Sepsis was stimulated by LPS treatment after 24 and 48 h and rats were subsequently treated with Pal (24 and 48 h).

Experimental groups

The experimental protocol was approved by the Ethics Committee of the Universidade Federal de São Paulo (CEP 0350/12) and was performed in accordance with the Brazilian guidelines for scientific animal care and use (Concea (CNDCDEA/MDCTEI), 2013; Brazilian Ministry of Science, Technology, Innovation and Communication, 2016).

Thirty male Wistar rats (weight: 230–250 g), not subjected to sepsis, were divided into five treated groups; Ex, Olv, Ex + Olv (ExOlv), Omg, and Pal. The baseline analysis values for these groups are presented in Table S2.

A total of 49 male Wistar rats (weight: 230–250 g) were divided into seven groups (n = 7). The CTL group was not submitted to sepsis. The LPS group was submitted to sepsis. Other rats were subjected to sepsis and were treated subsequently with Ex (LPSEx group), Ex + Olv (LPSExOlv group), Olv (LPSOlv group), Omg (LPSOmg group), and Pal (LPSPal group).

Rats were allowed free access to tap water and food during the experimental period. At 0 and 17 h, blood samples were collected from the orbital sinus and rats were maintained in metabolic cages for 24 h for urine collection. In some experiments, the urine was collected after 24 h and the volume was measured. Rats were euthanized, 48 h after the beginning of the experimental protocol, by intraperitoneal injection of a toxic dose of ketamine (90 mg/kg)/xylazine (10 mg/kg; Agribands de Brasil Ltda, Sao Paulo, Brazil). The right and left kidneys were then removed for immunohistochemical analysis. Biochemical parameters were measured in plasma and urine samples.

Survival

A total of 62 male Wistar rats (weight: 230–250 g), were used exclusively to Survival experiment, and were divided into the following groups: CTL (n = 6), LPS (n = 16), LPSEx (n = 10), LPSExOlv (n = 8), LPSOlv (n = 7), LPSOmg (n = 8), and LPSPal (n = 7). Survival was assessed at 0, 24, and 48 h. Results were expressed as the percentage of surviving animals (%).

Measurement of systolic blood pressure

Systolic blood pressure (SBP) was indirectly measured by tail plethysmography. Rats were placed in a warm chamber for 10 min and the cuff and wrist receiver were attached to the tail. Blood pressure was recorded using an electric sphygmomanometer coupled to a two-channel gould model 2200 S polygraph (Record 2200 S; Gould Inc., Cleveland, OH, USA). Measurements were taken at 0, 24, and 48 h, and the results are expressed as means ± SD.

Creatinine clearance

Serum and urine creatinine levels were assayed spectrophotometrically at 0 and 48 h, according to standard procedures, using commercially available diagnostic kits (Labtest Diagnostica, Minas Gerais, Brazil). Creatinine clearance (mL/min) was calculated according to the following formula: (urine creatinine concentration × urine volume)/(serum creatinine concentration × 1,440).

Lipidic peroxidation

To assess lipid peroxidation, levels of the peroxidation product, malondialdehyde, were determined by measuring thiobarbituric acid-reactive substances (TBARS). In this assay, the reactive substances in the test sample combine with thiobarbituric acid to form a red compound (Beuge & Aust, 1978). Urine samples were added to a solution of 0.375% thiobarbituric acid, 15% trichloroacetic acid, and 0.25 N HCl (Sigma, St. Louis, MO, USA). Subsequently, the samples were continuously agitated while heating to 95 °C for 20 min and were then allowed to cool to room temperature. Finally, the absorbance of the solution at 534 nm was determined using a spectrophotometer. Assays were performed at 0, 24, and 48 h. Results are expressed as nmol/mg creatinine.

Serum cytokines

Tumor necrosis factor alpha (TNF-α), interleukin-1 alpha (IL-1α), interleukin-1 beta (IL-1β), interleukin-6 (IL-6), interleukin-10 (IL-10), and granulocyte-macrophage colony-stimulating factor receptor (GM-CSF), were measured in rat serum using a Luminex High Performance Assay (RnD Systems, Minneapolis, MN, USA), according to the manufacturer’s instructions. Assays were performed at 0, 17, and 48 h, and results are expressed as pg/mL.

Immunohistochemistry

Hematoxylin and eosin staining was performed on kidney tissue sections. Paraffin-embedded tissues were cut into four-μm-thick sections on a rotary microtome (Leica Microsystems, Herlev, Denmark). The kidney slices were then deparaffinized, rehydrated, and boiled in a target retrieval solution (one mmol/L Tris, pH 9.0, with 0.5 mM EGTA) for 10 min to facilitate antigen binding. Nonspecific binding was prevented by incubating sections in PBS containing 1% BSA, 0.05% saponin, and 0.2% gelatin. Endogenous peroxidase activity was blocked by incubation in 5% H₂O₂ in absolute methanol for 10 min at 18–21 °C. Sections were then incubated overnight at 4 °C with primary antibodies against proliferating cell nuclear antigen (PCNA; diluted 1:500) and cleaved caspase-3 (diluted 1:100). Sections were washed and incubated with appropriate horseradish peroxidase-conjugated secondary antibodies (Dako, Glostrup, Denmark) for 1 h at room temperature. The sites of antibody-antigen reactions were visualized by staining with 0.5% 3,3′-diaminobenzidine tetrachloride (Dako) in 0.1% H₂O₂. The areas of PCNA were quantified using image analysis software (Leica DFC 310 FX, LAS software, version 3.8) to calculate the mean intensity value and expressed as a stained area. A total of 36 randomized microscope fields from different groups for cleaved caspase-3 were counted, with the same-sized square and the mean calculated and expressed as μm².

Statistical analysis

Results are expressed as the mean ± standard deviation (SD). Data were analyzed by Kaplan–Meier (survival curves), two-way analysis of variance (treated groups vs CTL and LPS groups), or simple linear regression analysis (treated groups versus Omg group), followed by a Bonferroni post hoc test. Values of p < 0.05 were considered statistically significant.

Blood pressure

Figure 1 shows the results of SBP measurements. Mean SBP values (Fig. 1A and 1B) of the LPS group were significantly lower than those of the CTL group after 24 h. Sepsis was characterized by hypotension. Treatments had no significant effect on blood pressure in septic rats. The blood pressure values for all groups are presented in Table S3.

Survival

There were no deaths in animals kept under CTL conditions. A total of 37% (6 out of 16) of rats in the LPS group survived for the entire 48-h experimental period. In septic rats treated with Ex, the overall survival rate was 60% (6 out of 10). The survival rate was 75% (6 out of 8) in the ExOlv and Omg groups and 85% (6 out of 7) in the Olv and Pal groups. All treatments tested resulted in a significant increase in the mean survival rate of septic rats at 48 h in comparison to the LPS group (Fig. 1C). The survival data for all groups are presented in Table S3.

Creatinine clearance

As shown in Fig. 1D, we observed a significant decrease in creatinine clearance in LPS group at 48 h, when compared to the CTL group, the sepsis-induced AKI. All treatments had no significant effect on creatinine clearance in septic rats. The values for all groups are presented in Table S3.

Urinary levels of TBARS

Abnormal production of ROS may result in cellular injury through peroxidation of membrane lipids, protein denaturation, and DNA damage (Srivastava & Kumar, 2015). Figure 1E shows the urinary levels of TBARS in experimental rats and the antioxidant effect of experimental treatments. There was an increase in lipid peroxidation in LPS group, compared to the CTL group at 24 h. Conversely, the levels of TBARS were lower in all treated groups than in the LPS group, within their respective experimental periods, but Ex and Pal showed the lowest antioxidant effects. The lipid peroxidation data for all groups are presented in Table S3.

Anti-inflammatory and proinflammatory mediators

Interleukin-10 is an important anti-inflammatory cytokine (Latorre et al., 2014). IL-10 levels in the LPS group were not significantly different from those in the CTL group. When compared to the LPS group, all other groups showed higher levels of IL-10, within their respective experimental periods (Fig. 2A). The IL-10 levels for all groups are presented in Table S3.

Interleukin-6 is a reliable marker for the early diagnosis and follow-up of sepsis (Gabay, Lamacchia & Palmer, 2010). IL-6 levels increased in the LPS group at 17 and 48 h, compared to the CTL group. However, when compared to the LPS group, the LPSExOlv, LPSOlv, and LPSOmg groups showed significantly lower IL-6 levels at 17 and 48 h. IL-6 levels in the LPSEx and LPSPal groups were not significantly different from IL-6 levels in the LPS group, within their respective experimental periods (Fig. 2B). The IL-6 levels for all groups are presented in Table S3.

The secretion of proinflammatory cytokines, such as TNF-α and IL-1 (including IL-1α and IL-1β), in sepsis has been demonstrated in numerous studies (Cannon et al., 1990; Hesse et al., 1988). TNF-α levels increased in the LPS group when compared to the CTL group at 17 h. However, when compared to the LPS group, there was a significant decrease in TNF-α levels in all other treatment groups at 17 h. No significant differences in TNF-α levels were observed between experimental groups at 48 h, indicating a short half-life of TNF-α (Fig. 2C). The TNF-α levels for all groups are presented in Table S3.

The inhibition of IL-1β protects against LPS-induced sepsis (Kang et al., 2004). We observed higher levels of IL-1β in the LPS group after 17 h when compared to the CTL group. Inversely, IL-1β levels were lower at 17 and 48 h in other treatment groups than in the LPS group (Fig. 2D), suggesting anti-inflammatory effects of the tested oils and extracts.

Compared to the CTL group, IL-1α levels increased at 17 and 48 h in the LPS group, which is suggestive of systemic inflammation (Fig. 2E). There was no significant effect of extract or oil treatment on IL-1α levels, with the exception of the LPSExOlv group, in which a significant decrease was seen at 17 h, compared to the LPS group. The IL-1α levels for all groups are presented in Table S3.

Once released, TNF-α and IL-1 act on different target cells to promote the proliferation, activation, differentiation, and survival of macrophages (Witsell & Schook, 1992; Fahlman et al., 1994; Conte et al., 2006) and all these effects enhance proinflammatory responses during sepsis, we confirmed an increase in GM-CSF levels at 17 and 48 h in the LPS group, when compared to the CTL group. GM-CSF levels only decreased at 17 h in the LPSOmg group, but were not statistically different from the LPS group at 17 h in the other treatment groups. However, at 48 h, GM-CSF levels were lower in all treatment groups, except the LPSEx group (Fig. 2F). The GM-CSF levels for all groups are presented in Table S3.

Cell proliferation and apoptosis in the kidney

The number of apoptotic cells per field was also analyzed by immunohistochemical detection of cleaved caspase-3 (Fig. 3). We observed an increase in apoptosis at 48 h in the LPS group when compared to the CTL group. The number of apoptotic cells in the LPSEx, LPSExOlv, LPSOlv, LPSOmg, and LPSPal groups was significantly lower the number of apoptotic cells in the LPS group (Fig. 3). The cleaved caspase-3 immunohistochemistry data for all groups are presented in Table S3.

Light microscopic images of kidney tissue sections are presented in Fig. 4. We observed a decrease in PCNA expression in the LPS group compared to the CTL group. However, a high level of PCNA expression was detected at 48 h in kidneys of the LPSEx, LPSExOlv, LPSOlv, LPSOmg, and LPSPal groups, when compared to the LPS group, indicating an increase in cell proliferation. The PCNA expression data for all groups are presented in Table S3.

Comparative effects of olive and palm products and omega-3 fish oil

Table 1 compares the results of LPSEx, LPSOlv, LPSExOlv, and LPSPal treatments with the results of LPSOmg treatment. In our experimental conditions, only Olv showed similar results to Omg, since there was no significant difference in lipid peroxidation, IL-1β, and IL-6 data between these two treatments.