LAUSR

Dilated cardiomyopathy (DCM) is a progressive disease characterized by ventricular chamber enlargement and contractile dysfunction. Natriuretic peptides (NPs) are state of the art for diagnosis and prognosis.1 Their secretion appears to depend on cardiac location,2 and, despite their known role in heart failure (HF), no studies have been conducted to elucidate their secretory mechanism. The Golgi apparatus represents the central processing and sorting station of the secretory pathway, and several studies in canine models have demonstrated that the size and complexity of the Golgi apparatus in cardiomyocytes increase during HF.3 In previous studies, we observed elevated NP mRNA levels in DCM LV tissue.4,5 However, NP protein levels did not change. Here we studied the Golgi apparatus and its relationship with NP levels and the LV remodelling process. Patient characteristics, homogenization of samples, protein determination, polyacrylamide gel electrophoresis, and western blot analysis are described in Tarazón et al.5 The primary detection antibodies used were anti-BNP (1:500) from HyTest Ltd, anti-atrial natriuretic peptide (ANP; 1:50) and anti-C-type natriuretic peptide (CNP; 1:50) from Santa Cruz biotechnology, and anti-glycerldehyde phosphate dehydrogenase (GAPDH; 1:1000) from Abcam. Tissue samples were obtained from the left ventricle of 17 patients with non-ischaemic DCM and 8 control (CNT) samples from non-diseased donor hearts. All patients gave written informed consent to participate in the study. The project was approved by the Biomedical Investigation Ethics Committee of our Hospital. The investigation was conducted in accordance with the guidelines of the Declaration of Helsinki. Isolation of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . human Golgi vesicles (hGVs) was performed in 14 of the 25 samples (DCM, n= 7; CNT, n= 7) using a modified version of a method previously developed for a rat model.6 Then, they were processed and observed with a JEOL JEM-1010 electron microscope. We analysed the numerical density and the morphology of vesicles (iTEM FEI program, v. 5.0, 2008). All statistical analyses were performed using SPSS software v. 20, 2012 for Windows (IBM SPSS Inc., Chicago, IL, USA). Significance was accepted at the P< 0.05 level. We analysed the NP protein levels in total LV tissue, and we found no differences from the CNT group (BNP, 107± 36 vs. 100± 23, NS; ANP, 80± 18 vs. 100± 19, NS; CNP, 73± 64 vs. 100± 40, NS). We also isolated for the first time the hGVs from the LV tissue, and analysed the NP levels. All three peptides were increased in DCM: BNP (132± 18 vs. 100±10, P< 0.05), ANP (207± 69 vs. 100± 44, P< 0.05), and CNP (154± 18 vs. 100± 29, P< 0.01). This shows a new adapted secretory mechanism that maintains the elevated levels of ANP, BNP, and CNP in the plasma. In fact, our hypothesis concerning the new secretory regime fits with the variations found in the morphology of hGVs. We assessed the shape and size of the hGVs (CNT, n= 4138; DCM, n= 7641), finding that the vesicular density of the isolates (vesicles per μm2) was higher (10.52± 4.24 vs. 7.98± 3.24, P< 0.0001), the area was smaller [8452 (4226–16 466) nm2 vs. 8981 (4402–21 132) nm2, P< 0.0001] (Figure 1A), and the shape was more ellipsoidal [0.94 (0.88–0.98) vs. 0.95 (0.91–0.99), P< 0.0001] in the DCM group, which corresponds to an immediate secretion to produce elevated NP plasma levels. Moreover, we also confirmed the desired quality of the isolated hGVs for both groups (Figure 1B). Thus, the rate of secretion could influence the shape and size of the vesicles responsible for the secretion of all proteins. In this sense, we found an inverse correlation between hGV area and both LV end-diastolic (r=−0.828) and end-systolic (r=−0.844) diameters (P< 0.05); and a direct correlation with LVEF (r= 0.873) (P< 0.01) (Figure 1C); smaller and more numerous vesicles indicated a worse functional situation. This is . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . in accordance with what occurs in healthy CNT; vesicles were fewer in number, larger, and more spherical. In HF, NP demand increases to regulate plasma volume and pressure homeostasis. This demand results in an augmentation of the production of transcription factors that regulate their expression (e.g. GATA4 mRNA), and in a faster NP synthesis that has been attributed to endoplasmic reticulum stress. It has been published that in the presence of large amounts of proteins, the Golgi activates a stress response pathway to induce the transcription of genes involved in its function.7 Consequently, we categorize the Golgi apparatus as the organelle that plays an important role in the NP transport to plasma, with its morphological NP-linked changes in area and number being related to LV new dimensions and function.