*3.1. Surgically Induced Models*

The most well-characterized surgical model is the subtotal nephrectomy model, also called 5/6 nephrectomy. There are different methods for performing this surgery; all involve unilateral nephrectomy. Reduction of the remaining kidney mass is then performed either by cauterization, ligation or slicing off the two poles, or by occlusion of the branches of the main kidney artery. Unilateral nephrectomy and reduction of the remaining kidney mass are either performed on the same day (1-step) or with a 7- to 14-day recovery phase after the unilateral nephrectomy (2-step). Various studies describe a cardiac phenotype resulting from CKD in these mice, which is aggravated with the time elapsed since surgery (see Table 2). The surgically induced model has been reported to mimic most phenotypic changes observed in the human disease. However, the resulting phenotype is highly dependent on the genetic background of the mice.

A study performed in nephrectomized C57BL/6 mice showed impaired renal function, anemia, cardiac hypertrophy, cardiac fibrosis and decreased systolic and diastolic heart function compared to sham-operated mice [75,89]. In addition, there was increased expression of natriuretic peptides, another marker of progressive heart failure, in nephrectomized male 129X1/SvJ mice [88]. Subtotal nephrectomy in male CD1 mice resulted in physiological and morphological changes that also mimicked the cardiac phenotype in patients with CKD [91]. In a study with male BALB/c mice, analyzed for up to 24 weeks after 5/6 nephrectomy, cardiac and arterial structure and function showed signs of fibrosis, oxidative damage and endothelial dysfunction [90,91].

RAAS is a key player in blood-pressure control and has been described as being highly relevant in the cardiorenal syndrome. Subtotal nephrectomy in angiotensin II type 1A receptor (AT1) knockout mice resulted in significantly reduced cardiac hypertrophy, fibrosis and capillary rarefaction compared to their wildtype littermates [87]. Other widely used cardioprotective and antihypertensive treatments such as β1-receptor blockers have also been shown to reduce cardiac hypertrophy in CKD [95].

CKD patients suffer from reduced erythropoietin (EPO) levels, and it has been shown that besides causing anemia, EPO reduction had direct effects on the cardiovascular system of these patients. In line with this, 5/6 nephrectomized mice receiving recombinant human erythropoietin (rhEPO) had a better outcome compared to saline-treated controls in terms of cardiac function and remodeling. These effects were independent of anemia. Thus, a control group that received an EPO derivate (asialo-EPO) still suffered from anemia but showed the same beneficial effects on cardiac remodeling [79,95]. Both compounds similarly attenuated LVH, indicating that EPO receptor signaling protected the hearts of CKD mice through mechanisms independent of erythropoiesis. The production of erythrocytes requires continuous stimulation of EPO receptors, whereas a brief stimulation is described to be sufficient for neuroprotection. The same mechanism might explain the cardioprotective effect in this model [79].

In addition, unilateral ureteral obstruction (UUO) can induce some cardiac hypertrophy and fibrosis [79,92]. However, since UUO does not cause renal failure in the presence of a non-injured contralateral kidney, the model is probably not suitable to induce a severe cardiac phenotype.

By performing a graded ureter obstruction model in combination with a systemic expression of humanized S100A8, S100A12, and S100A9 in C57BL/6 mice, the association between S100/RAGE/FGF23 and cardiac hypertrophy was revealed [85]. Elevated serum concentrations of S100A12 are associated with inflammatory diseases, and thus might accelerate pathological cardiac remodeling in CKD patients. As mice do not express S100A12 [34,96], a humanized model was generated and CKD was induced via reversible unilateral ureteral obstruction. For this method, the right ureter was obstructed using a clip that was relocated every other day to prevent irreversible obstruction. After 7 days, the clip was removed to allow recovery of the right kidney, followed by irreversible ligation of the left ureter. A potential mechanism to explain the results of the study was presented with the finding that in addition to osteocyte expression, in CKD patients, FGF23 is also expressed in the heart by cardiomyocytes, cardiac fibroblasts, vascular smooth muscle cells and endothelial cells in coronary arteries, and by inflammatory macrophages [60]. It was suggested that S100/RAGE-mediated chronic

sustained systemic inflammation caused cardiac fibroblasts to upregulate FGF23 synthesis, and in turn, increased cardiac FGF23 levels were linked to pathological cardiac remodeling [86].
