One-third of ischemic heart disease is associated with high cholesterol, which is also related to increased risk of stroke and heart disease. Moreover, hypercholesterolemia is related to 2.6 million deaths, according to the World Health Organization. In 2008, the prevalence of increased cholesterol among adults was 39%. More than half of United States adults (55%, or 43 million) who could benefit from cholesterol medicine are currently using it [1
Gemfibrozil is a fibrate used for the management of dyslipidemia. It is well-known to decrease low-density lipoprotein (LDL) and increase high-density lipoprotein (HDL) cholesterol, and it can also be used to reduce the development of coronary heart disease [2
]. Gemfibrozil activates peroxisome proliferator-activated receptor alpha (PPAR-α), which leads to changes in lipid metabolism, decreasing serum triglyceride levels by modulating the lipoprotein lipase in adipose and muscle tissues [9
]. Some side effects are associated with the use of fibrates, and anemia is one of them [10
Anemia is characterized by a reduced ability of the blood to carry oxygen, and reduced levels of hemoglobin are observed in the peripheral blood, which usually also reflects the reduced number of red blood cells (RBCs). Moreover, anemia is the most common blood disorder, affecting about one-third of the global population [12
]. It can be caused by blood loss, decreased red-blood-cell production and increased red-blood-cell breakdown [13
]. It increases medical costs and lowers a person’s productivity through decreased ability to work [14
Leukopenia is the term for a low level of white blood cells, or leukocytes, which are responsible for systemic defense against infections and diseases. Leukopenia can be caused by chronic conditions, infections and some drugs that slow bone-marrow function, leading to a reduced white-blood-cell count. The greatest danger of this condition is susceptibility to infections [15
Considering that the prevalence of hyperlipidemia is high and that treatment with fibrates is frequent, we decided to establish a reliable model for studying gemfibrozil-induced anemia and check whether this effect was mediated by PPAR-α.
The prevalence of hyperlipidemia in the modern world is high, thus, the use of hypolipemiant drugs is increasing every day. However, the side effects must be better understood in order to find new tools to avoid them. Gemfibrozil is a fibrate, and it is used to activate PPAR-α [20
]. Here, we developed a reliable model to study gemfibrozil-induced anemia in mice. We observed that gemfibrozil led to anemia seven days after the first administration and that hemoglobin levels diminished 14 days after the first dose and normalized 21 days later. In a study, after four weeks of 100 mg/kg bw/day of fenofibrate, rats developed slight anemia [10
]. In our model, the development of anemia was clear, and we could see a reduction in all parameters analyzed by complete blood count. In order to better understand its mechanism, we analyzed renal tissue, where hypoxia-inducible factor has an important role when low oxygen levels are detected, and HIF-2α is crucial for erythropoietin induction [17
HIF-2α-knockout mice presented anemia, which was not the result of a cell-autonomous defect in erythroid-precursor maturation but of inadequate renal EPO production; moreover, the location of renal HIF-2α-expressing interstitial cells coincides with that of renal EPO-producing cells [22
]. Although hemoglobin, hematocrit and RBCs decreased, indicating an anemic condition in our mice, HIF-2α
mRNA levels increased in the renal tissue, and increased serum EPO levels were observed with gemfibrozil treatment, which suggests that a compensatory mechanism was triggered to increase EPO production. This explains the normalization of hemoglobin levels on Day 21 after gemfibrozil treatment. Additionally, leukopenia, a common side effect of gemfibrozil treatment found in humans, was observed in gemfibrozil-treated mice [11
]. Bone-marrow analyses corroborated blood analyses, in which we found decreased hematopoietic-stem-cell counts in addition to a decreased percentage of erythrocyte markers. Our results suggest that gemfibrozil may present some cytotoxicity. However, it has already been shown that fibrates can alter cell-cycle distributions, increasing G0/G1 phase and reducing G2/M phase, which shows that fibrates reduce cell proliferation with no or very low cytotoxic effects [25
]. Further investigation of the mechanism that reduces the number of hematopoietic progenitor cells and the erythroid population through investigating cell arrest and cell-death mechanisms is necessary. There are no mechanisms described for gemfibrozil that alter hematopoiesis, but the inhibition of Stat-3, an important transcription factor in myelopoiesis, by gemfibrozil was observed in adipose and hepatic tissues [26
The colony-forming-unit assay is widely used for analyses of hematopoietic and progenitor cells. It allows for the measurement of cell differentiation and proliferation. Both total CFU counts and CFU-granulocyte counts were increased in gemfibrozil-treated animals, showing that the bone marrow of gemfibrozil-treated mice increased the production of granulocyte colonies in order to potentially reverse leukopenia.
Here, we presented for the first time that gemfibrozil led to anemia in mice, decreasing erythrocytes and hematopoietic stem cells in the bone marrow and increasing signaling for RBC production in the renal tissue and WBC production in the bone marrow in order to compensate for these adverse effects. Moreover, we also showed that this effect of gemfibrozil was dependent on PPAR-alpha, since its ablation could reverse all these side effects. Additional studies are necessary to further clarify the mechanisms of gemfibrozil-induced anemia, but we showed that gemfibrozil effects were dependent on PPAR-α.
Wild type (WT, C57/BL6J) and PPAR-α knockout (PPARα KO, B6; 129S4-Pparatm1Gonz/J, Jackson laboratory) male mice weighing 23–27 g and aged 9–12 weeks, were used for these experiments. The animals were obtained from the Animal Care Facility of the Federal University of São Paulo (UNIFESP). All animals were housed in individual, standard cages and had free access to water and food. All procedures were previously reviewed and approved by the internal ethics committee of the Federal University of São Paulo (CEUA 6823010319).
4.2. Experimental Protocol
The mice were divided into 4 groups for each experiment: WT vehicle group, PPARKO vehicle group, WT gemfibrozil group and PPAR-α-KO gemfibrozil group. We used n = 5–6 for each experiment and condition.
4.3. Drug Treatment
Gemfibrozil was given by oral gavage (150 mg/kg) diluted in 0.5% methylcellulose [27
]. The mice were treated for 4 days with gemfibrozil and euthanized 1 week after the beginning of treatment or followed for 21 days. Vehicle-group mice were given the same volume with 0.5% methylcellulose.
4.4. Blood Sampling and Tissue Collection
The mice were anesthetized with ketamine (91 mg/kg) and xylazine (9.1 mg/kg) intraperitoneally and blood was collected via cardiac puncture. Blood samples for serum collection were allowed to clot for 2 h at room temperature and then centrifuged for 20 min at 2000× g. The samples were then stored at –20 °C. Kidney tissue was collected, and the renal capsule was removed. Transversal cuts were performed, and the kidneys were immediately frozen in nitrogen and then stored at −80 °C.
4.5. Blood Count Test
Blood samples were collected with EDTA (Merck, Darmstadt, Germany), and blood counts were performed using ABX Micros ABC Vet equipment (Horiba ABX, Montpellier, France). Leukocyte differential and morphological analyses were performed on blood smears stained using the May–Grünwald–Giemsa (Merck, Darmstadt, Germany) technique.
4.6. Hemoglobin Measurement
The HemoCue portable system for hemoglobinometry can measure hemoglobin (Hb) concentration in less than one minute. Briefly, this device measures Hb by spectrophotometry using a small-volume (tailcut) optical-measuring cuvette and short light path. This device allows for the measurement of Hb concentration in blood samples several times and with low sample volumes.
4.7. Real-Time PCR
Kidney samples were frozen at –80 °C immediately after collection. Total RNA was isolated using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA). The RNA integrity was assessed by electrophoresis on an agarose gel. cDNA was synthesized using the “High Capacity cDNA Reverse Transcription Kit” (Applied Biosystems). Standard curves were plotted to determine the amplification efficiency for each primer pair. Real-time PCR was performed with the SYBR Green system (Thermo Scientific, Waltham, MA, USA) using specific primers for β-actin, 18S, EPO, HIF-2a; the primers were designed using Primer3 web, and their specificity was confirmed using NCBI primer-BLAST; their sequences were β-actin forward 5′ CTGGCCTCACTGTCCACCTT 3′, reverse 5′ CGGACTCATCGTACTCCTGCTT 3′; 18s forward 5′ CGCCGCTAGAGGTGAAATTC 3′, reverse 5′ TCTTGGCAAATGCTTTCGC 3′; EPO forward 5′ TCCACTCCGAACACTCACA 3′, reverse 5′ CCTCTCCCGTGTACAGCTT 3′; HIF2a forward 5′ CTGGACAAAGCCTCCATCAT 3′, reverse 5′ TTGCTGATGTTTTCCGACAG 3′. The cycling conditions were as follows: 10 min at 95 °C, followed by 45 cycles of 30 s at 95 °C, 30 s at 60 °C, and 30 s at 72 °C. Target mRNA expression was normalized to β-actin and 18s, and expressed as a relative value using the comparative threshold cycle (Ct) method (2−ΔΔCt). The expression levels of the genes of interest were normalized to the control group and presented as fold change.
4.8. Serum Erythropoietin Assay
Serum samples were frozen and stored at –20 °C immediately after collection. Serum EPO levels (KA1998) were quantified using the ELISA mouse kit specific to this analyte (Novus Biologicals, Littleton, CO, USA), according to the manufacturer’s instructions.
4.9. Colony-Forming-Unit Assay
Bone-marrow cells were collected from the mice’s femurs and suspended in Iscove’s Modified Dulbecco’s Medium (Invitrogen, Waltham, MA, USA). Twenty thousand cells were mixed into 1 mL of MethoCult GF M3434 (StemCell Technologies, Vancouver, BC, Canada) and supplemented according to the manufacturer’s instructions. The CFU assay was performed by placing cells in a 35-mm Petri dish. Cells were cultured in a humidity chamber at 37 °C with 5% CO2 for 14 days. At the end of the incubation period, colonies of more than 50 cells were counted using an inverted microscope at 20× magnification.
Hematopoietic cell populations were labeled with antibodies as previously described [31
]. The following monoclonal antibodies were used to define the different hematopoietic cell populations: Ter-119 (Ly-76) and Mac-1 (M1/70). The cells were incubated with the antibodies for 30 min and then washed. In order to identify HSC, whole bone marrow cells (1 × 106
/sample) were labeled with a mature lineage (Lin) antibody cocktail (Gr1-PE, Mac-1-PE, CD3e-PE, Ter-119-PE and B220-PE), CD90.1 (Thy1.1.)-FITC, FLK-2-PE, Sca-1-PE/Cy7 and c-Kit-APC for 30 min. The cells were then washed. Gr-1+
cells were defined as myeloid cells, and Ter-119+
(Gr-1-) cells were defined as erythroid cells. HSCs were defined as Lin-
]. All antibodies were purchased from Becton Dickinson (Franklin Lakes, NJ, USA). All cytometry analyses in this study were performed with an Accuri C6 (Becton Dickinson, Franklin Lakes, NJ, USA) flow cytometer. Total bone marrow cells count and total labelled cells count are found in supplementary materials (Figure S1)
4.11. Statistical Analysis
All data are presented as mean ± SEM. Intergroup-difference significance was assessed by two-way analysis of variance (ANOVA) with Tukey’s correction for multiple comparisons. The value for statistical significance was established at p < 0.05. All statistical analyses were performed using GraphPad Prism 8 (GraphPad, La Jolla, CA, USA).