Cardiac copper deficiency activates a systemic signaling mechanism that communicates with the copper acquisition and storage organs

Abstract

Copper (Cu) is an essential cofactor for a variety of metabolic functions, and the regulation of systemic Cu metabolism is critical to human health. Dietary Cu is absorbed through the intestine, stored in the liver, and mobilized into the circulation; however, systemic Cu homeostasis is poorly understood. We generated mice with a cardiac-specific knockout of the Ctr1 Cu transporter (Ctr1(hrt/hrt)), resulting in cardiac Cu deficiency and severe cardiomyopathy. Unexpectedly, Ctr1(hrt/hrt) mice exhibited increased serum Cu levels and a concomitant decrease in hepatic Cu stores. Expression of the ATP7A Cu exporter, thought to function predominantly in intestinal Cu acquisition, was strongly increased in liver and intestine of Ctr1(hrt/hrt) mice. These studies identify ATP7A as a candidate for hepatic Cu mobilization in response to peripheral tissue demand, and illuminate a systemic regulation in which the Cu status of the heart is signaled to organs that take up and store Cu.

Figure 1. Cardiac-specific knockdown of the Drosophila Ctr1A Cu ion channel

(A) Knock down of Ctr1A in the dorsal vessel. Immunoblot of crudely dissected Drosophila dorsal vessel from wild type (WT) and TinCGal4/UAS-Ctr1A^RNAi flies (Ctr1^RNAi). Total tissue protein extracts were probed with anti-Ctr1A peptide antibody or anti-ERK as a loading control. (B) Representative optical coherence tomography (OCT) images for wild type and TinCGal4/UAS-Ctr1A^RNAi flies. (C) End systolic dimension (ESD) measurements for wild type and TinCGal4/UAS-Ctr1A^RNAi flies. Error bars, s.d., *P = 0.00073 versus wild type. (D) Fractional shortening measurements for wild type and TinCGal4/UAS-Ctr1A^RNAi flies. Error bars, s.d., *P < 0.001 versus wild type.

Figure 2. Generation of a cardiac-specific knock-out of the mouse Ctr1 gene

(A) Growth of Ctr1^hrt/hrt mice. The body mass in grams (g) for Ctr1^hrt/hrt mice (red) or control littermates (blue) as a function of age after birth in days is shown. (B) Ctr1^hrt/hrt and Ctr1^flox/flox littermates at P10. A ruler is included as a size reference (in cm). (C) Kaplan-Meier plot showing postnatal survival of control (blue, n=69; Ctr1^flox/flox, Ctr1^flox/+ and Ctr^hrt/+) and Ctr1^hrt/hrt mice (red, n=25). (D) SDS-PAGE analysis of total Triton X-100 solubilized heart extracts of two representative Ctr1^flox/flox and Ctr1^hrt/hrt mice. Tissues were probed for Ctr1, CCS (copper chaperone for Cu, Zn superoxide dismutase), COX IV (mitochondrial cytochrome oxidase subunit V) and Tubulin as a loading control. The arrows indicate the positions of monomeric (at approximately 17kD) and heterogeneous glycosylated form (at approximately 35 kD) of Ctr1, both of which are dramatically reduced in the Ctr1^hrt/hrt mice.

Figure 3. Ctr1^hrt/hrt mice exhibit cardiac hypertrophy

(A) Gross morphology of P10 hearts from Ctr1^flox/flox and Ctr1^hrt/hrt mice. (B) Quantitation of heart weight/body weight ratio from Ctr1^flox/flox (blue, n=7) and Ctr1^hrt/hrt (red, n=7) mice. Error bars, s.d., P < 0.0001 versus Ctr1^flox/flox. (C) Markers of cardiac hypertrophy. Heart-specific mRNA from male and female Ctr1^flox/flox and Ctr1^hrt/hrt mice was analyzed by RNA blotting using ³²P-labeled DNA probes to atrial natriuretic peptide (ANP), b-type natriuretic peptide (BNP) and skeletal actin (SA). Ribosomal RNA (rRNA) is shown as a loading control. (D) Heart sections from P10 Ctr1^flox/flox and Ctr1^hrt/hrt were stained with hematoxylin and eosin. Left panel: Ctr1^flox/flox; right panel: Ctr1^hrt/hrt (40X magnification; Scale Bar=50 μm). (E) Representative cardiac sections stained with Masson Trichome demonstrates extensive fibrosis (gray staining) in the ventricular endocardium region of tissue from Ctr1^hrt/hrt mice, indicated with an arrowhead, as compared to Ctr1^flox/flox controls (40X magnification; Scale Bar=50 μm). (F) Electron microscopy of cardiac tissue (1800X magnification of cardiac thin sections; upper panels, 3000X magnification of cardiac thin sections; lower panels, Scale Bar=1μm). Left panels, Ctr1^flox/flox; right panels, Ctr1^hrt/hrt. Thin section electron micrographs of cardiac tissues from Ctr1^hrt/hrt mice show abnormally enlarged mitochondria with disrupted cristae structures. Myofilaments and mitochondria are indicated with arrowheads.

Figure 4. Cardiac dysfunction resulting from loss of the Ctr1 Cu ion channel

(A) Representative echocardiogram M-mode images from Ctr1^flox/flox and Ctr1^hrt/hrt at postnatal 10 days. (B) Quantitation of left ventricular diastolic dimension (LVDd), left ventricular systolic dimension (LVDs), fractional shortening and left ventricular mass (LVm) for Ctr1^flox/flox (n=10) and Ctr1^hrt/hrt mice (n=9). Error bars, s.d. *P < 0.0001 versus Ctr1^flox/flox. (C) Mouse heart rate measurements demonstrate a ~ 50% decrease in heart rate in Ctr1^hrt/hrt mice (n=10) relative to control mice (n=9). Error bars, s.d., *P < 0.0001 versus Ctr1^flox/flox. (D) Electrocardiographic analysis of Ctr1^flox/flox and Ctr1^hrt/hrt mice from P10 reveal a bradycardia due to cardiac-specific loss of Ctr1. (E) SDS-PAGE analysis of total heart extracts of two independent Ctr1^flox/flox (flox) and two independent Ctr1^hrt/hrt mice (hrt). Immunoblots were probed for SERCA-2, P-PLB (phospho-phospholamban, Ser-16), PLB (phospholamban) and Tubulin as a loading control. (F) Immunoblot of cardiac tissues for Ctr1, CCS and SOD1 (included as loading control) from two representative 2 month-old Ctr1^flox/flox mice and two Ctr1^flox/flox; MHC-Cre-ER mice after tamoxifen (OHT) treatments. 1 and 2 indicate samples from two independent mice of each genotype. (G) Heart sections from 2 month and 7 month-old Ctr1^flox/flox and Ctr1^flox/flox; MHC-Cre-ER mice injected with OHT were stained with hematoxylin and eosin. Left panel: Ctr1^flox/flox, right panel: Ctr1^flox/flox; MHC-Cre-ER (40X magnification, Scale Bar=50 μm).

Figure 5. ATP7A expression is increased in the liver and intestine of Ctr1^hrt/hrt mice

(A) Tissue Cu levels in Ctr1^hrt/hrt mice are shown as a percentage of Cu levels from Ctr1^flox/flox mice (100%). Error bars, s.d., *P < 0.005 versus Cu levels in Ctr1^flox/flox. H, heart; B, brain; M, muscle; L, liver; S, serum. (B) Mitochondrial cytochrome oxidase activity in cardiac tissue from Ctr1^flox/flox (blue, n=4) and Ctr1^hrt/hrt (red, n=5) mice. Error bars, s.d., *P < 0.005 versus Ctr1^flox/flox. (C) Immunoblot of representative Ctr1^flox/flox (flox) and Ctr1^hrt/hrt (hrt) mouse tissues for ATP7A and Actin (as loading control). H, heart; M, muscle; B, brain; L, liver; K, kidney. (D) Semi-quantitative RT-PCR for ATP7A and ATP7B. GAPDH was assayed as an input control. (E) Immunohistochemistry of Ctr1^flox/flox and Ctr1^hrt/hrt mouse liver with anti-ATP7A antibody. Enriched ATP7A expression toward the sinusoidal space is indicated with arrowheads. The central vein (CV) is indicated. Scale Bar=50μm. (F) Immunoblot analysis of total Triton X-100 solubilized liver extracts from two representative Ctr1^flox/flox and Ctr1^hrt/hrt mouse tissues for ATP7A, ATP7B and ceruloplasmin (Cp) and Tubulin as a loading control). (G) Immunoblot analysis of ATP7A and CCS in the intestinal epithelial cells of two representative Ctr1^flox/flox and Ctr1^hrt/hrt mice. Actin levels were assayed as a loading control. For panels F and G, 1 and 2 indicate samples from two independent mice of each genotype.

Figure 6. ATP7A expression correlates with hepatic Cu mobilization

(A) Immunoblot analysis of ATP7A, Ceruloplasmin (Cp), COX IV and SOD1 in the liver of wild type C57BL mice during maturation. P5; postnatal day 5, P10; postnatal day 10, 4W; 4 week-old mice, 6M; 6 month-old mice. (B) Liver Cu levels from wild type mice. P5; postnatal day 5, P10; postnatal day 10, 4W; 4 week-old mice, 6M; 6 month-old mice. Error bars, s.d., n=5 for each group. (C) Comparison of hepatic ATP7A levels between a representative 10 day old Ctr1^hrt/hrt (hrt) mouse and wild type mice at P5; postnatal day 5, P10; postnatal day 10, 4W; 4 week-old mice, 6M; 6 month-old mice. (D) ATP7A expression levels (short exposure as well as longer exposure shown) in the intestinal epithelial cells (IEC), liver (LIV) and heart (HRT) in response to a Cu-limited diet in wild type C57BL mice. CCS and COX IV were assayed as cellular Cu level indicators and SOD1 is shown as a loading control. A indicates a copper adequate diet and D indicates a copper deficient diet.

Figure 7. in vitro activation of ATP7A protein in serum treated human cells

(A) Immunoblot analysis of ATP7A and CCS levels in HUVEC cells treated with serum from control Ctr1^flox/flox and Ctr1^hrt/hrt mice. Actin levels were assayed as a loading control. (B) Immunoblot analysis of ATP7A and CCS levels in Caco-2 cells treated with serum from control Ctr1^flox/flox and Ctr1^hrt/hrt mice. Tubulin levels were assayed as a loading control. The source of the serum is indicated by the mouse genotype.