Expression of Endoplasmic Reticulum Chaperones in Cardiac Development

Animals

Friend leukemia virus B-sensitive mice (FVB) mice were used. Timed pregnancies were performed by randomly breeding FVB mice and checking vaginal plug formation each morning. Observed vaginal plugs were designated as gestational day 0 (denoted as E0). At an experimentally specified gestational day (day 13 to day 18, i.e., E13 to E18), pregnant FVB mice were sacrificed by cervical dislocation, embryos removed and hearts were dissected out. The tissues were then used for Western blot analysis. Minimum 15 embryonic hearts were used for atrial preparations and 10 for ventricular. As it was impossible to separate atria from ventricles at E13, minimum 10 whole embryonic hearts were used for each preparation. For adult atrial and ventricular preparations, non-pregnant animals approximately 6 months old were used.

Cell Culture

To provide a positive control for experiments with western blotting of ATF6α, mouse embryonic fibroblasts were grown on high-glucose Dulbecco’s Modified Eagle’s Medium supplemented with 10% fetal bovine serum and 2 mM L-glutamine. The cells were grown to 80% confluence in tissue culture dishes followed by overnight treatment with thapsigargin (500 nM). Cells were collected the next day and were subjected to Western blot analysis.

Western Blotting

Cells and tissues were homogenized in lysis buffer (1 M Tris-Cl PH 8.0, 1.2 M NaCl, 0.5% v/v Nonidet P-40) with 10 μl/ml of protease inhibitor cocktail freshly added (Sigma) for 15 minutes on ice, followed by centrifugation at 14000 X g for 15 minutes at 4°C. Supernatants were then measured for protein content according to the manufacturer’s instructions (Bio-Rad). Samples were boiled for 5 minutes in reducing sample buffer with 2-mercaptoethanol (5 μl/100ml) and then subjected to 10% SDS-PAGE. The separated sample proteins were transferred to nitrocellulose membranes, which were then blocked in 5% milk in PBS with 0.01% Tween-20 (PBST) overnight at 4°C. After blocking, membranes were incubated in diluted primary antibodies for 1 hour at room temperature, followed by washing in PBST for 15 minutes (3 times). Membranes were then incubated in diluted (1:10000) secondary horseradish peroxidase (HRP) conjugated antibodies for another 1 hour, followed by washing in PBST for 15 minutes (3 times). ECL Plus Western blotting detection reagents (Amersham Biosciences) and Kodak x-ray film were utilized to visualize the protein bands. Primary antibodies diluted in PBST were as follows: goat anti-calreticulin, 1:300; rabbit anti-Grp94 1:1000; rabbit anti-PDI 1: 1000; rabbit anti-ERp57 1:1000; rabbit anti-ATF6α polyclonal antibody 1: 200 (Santa Cruz); rabbit anti-calnexin polyclonal antibody (spa-860) 1: 2000 (Stressgen Biotechnologies); rabbit anti-caspase-12 polyclonal antibody 1:100 (Cell Signaling Technology); rabbit anti-caspase-7 polyclonal antibody 1:1000 (Cell signaling Technology); rabbit anti-Grp78 polyclonal antibody 1:2000 (Stressgen Biotechnologies); rabbit anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) 1:2000 (Labfrontier, Korea). The protein bands in each blot were normalized by using GAPDH antibody and relative protein levels were quantified using Image J software.

Differential Expression of Calreticulin and Calnexin During Heart Development

There are several families of ER chaperones, which work in co-ordinated pairs and/or groups. The calreticulin/calnexin cycle of chaperoning has been well established [23-25]. We first investigated the expression pattern of this pair of chaperones during heart development. We have previously shown that calnexin is not induced by ER stress, but that calreticulin is indeed inducible by ER stress [26]. Fig. (1A) shows that calreticulin expression was higher in embryonic heart tissues compared to their adult counterparts. In particular, calreticulin expression diminished dramatically in ventricular tissue postnatally, and remained low into adulthood (Fig. 1, Fig. 4). Earlier studies have demonstrated a role for ER resident chaperones in embryonic development using knockout mice for specific chaperones [10, 27, 28]. Two chaperones, such as calreticulin and Grp94, were found to be critical in embryonic development, as their deficiency induced embryonic death. When calreticulin was ablated, calreticulin-deficient mouse embryos died at gestational day 14.5 as a result of faulty cardiac organogenesis. The cardiac structural defects included very thin ventricular walls with deep intertrabecular recesses [10, 11]. In line with this gene knockout study, our current data showed higher expression of calreticulin in both mouse embryonic atrial and ventricular tissue than in the adult. Calreticulin expression began to subside 17 days postnatally in both atria and ventricles, and in adult ventricles, calreticulin was barely detectable.

Calnexin expression, on the other hand, remained largely unchanged during cardiac development, as well as into adulthood (Fig. 1B). Calnexin deficiency is not embryonic lethal. Calnexin-null mice are viable but exhibit ataxia [27] (and unpublished data from Dr. Michalak). Our data showed that calnexin expression did not change during mouse heart development [26], which may indicate different roles for calnexin and calreticulin during embryonic development. This is not surprising, as calreticulin and calnexin, which often work in tandem as chaperones, also have unique substrates [23, 29]. In addition, calreticulin is a soluble ER lumenal protein, whereas calnexin is a type I ER membrane protein, which may also lead to their differences in function [15]. We have previously demonstrated that ER protein distribution contributes to the heterogeneous function of different ER domains [30].

Grp94 and Grp78 Expression in Embryonic Heart Development

An important class of ER chaperones are the glucose-regulated proteins (Grps), which belong to the heat shock family of stress proteins. Grp78, also known as BiP, and Grp94 function as molecular chaperones and can bind to either misfolded proteins or unassembled polypeptides. They are induced in response to stress, but are then post-transcriptionally modified into inactive forms upon removal of the stress [31]. In order to detect the expression of Grp94 and Grp78 during embryonic development, Western blot analyses were performed on mouse heart tissues at embryonic and adult stages. Both Grp94 and Grp78 exhibited similar expression levels as that observed for calreticulin during mouse heart development (Fig. 2A, 2B, Fig. 4). In all tissues examined, Grp94 and Grp78 expression was higher in embryonic versus the adult tissues. Barnes and Smoak [12, 13], showed that Grp94 and Grp78 levels in the heart were highest when the myocardial cells were proliferating rapidly, indicating a high demand for the synthesis of these two proteins at certain stages of heart development. Our present data shows that Grp78 and Grp94 expression is high in both atria and ventricles until gestational day 18. At postnatal day 15, Grp78 expression level in the atria and ventricles is very low, and not detectable in the adult, indicating that there may also be a peak in the regulation of Grp78 during mammalian heart development. These data indicate that Grp78 and Grp94 play an important role during cardiomyocyte differentiation and heart development.

PDI and ERp57 Expression During Heart Development

Another class of chaperones we examined were the protein disulfide isomerase (PDI) family, which assist in disulfide bond formation. We specifically examined PDI and ER protein 57 (ERp57), which is an ortholog of PDI that acts on glycosylated proteins [32]. To examine the expression of PDI and ERp57 during mouse embryo development, we performed Western blot analyses of PDI and ERp57 in mouse embryonic and adult heart tissues. Fig. (3A, 3B) and Fig. (4)show that the higher expression levels of ERp57 and PDI were during embryonic development, while these proteins are barely detectable in adult tissues. This expression pattern is similar to that of calreticulin, Grp94 and Grp78.

In summary, three groups of ER resident stress-related chaperones have been found to have a similar pattern of expression during mouse embryonic development, and are all greatly downregulated in adult tissues. We thus conclude that there is a form of ER stress during normal cardiac development. We next sought to investigate the degree of this ER stress in embryos compared to the degree of stress in adulthood, in order to understand the mechanism behind this upregulation in the embryonic stages. Is it a large synthetic load to which these chaperones are responding or is embryonic development also accompanied by a form of ER stress?

Detection of UPR factors in Heart Development

In order to investigate whether the UPR, an ER stress initiated signaling network, is also detectable in embryonic development, we examined several factors in each UPR pathway.

We could not detect eIF-2α protein,phospho-eIF-2α protein or cleaved XBP-1 mRNA (a cleavage product of IRE1) either during embryonic heart development or during adulthood (data not shown). We also performed Western blot analysis of ATF6α expression in mouse embryonic heart tissues. The expression of ATF6α remained unchanged during embryonic heart development and was slightly higher in adult tissues, especially in the adult ventricles (Fig. 5). It seems then, that there is not a great degree of ER stress within the developing mouse heart. We could not detect UPR activation in developing mouse hearts, but there seemed to be some degree of ATF6 induction in the adult heart. It can only be speculated on that such an induction may be due to cardiac remodelling. Further examinations need to be undertaken to determine the degree of such possible remodelling and to determine if it is normal or pathologic.

ER-Stress Activated Caspases During Heart Development

To further determine the degree of ER stress in developing and adult mouse hearts, Western blot analysis was performed for caspase-7 and caspase-12, which have been shown to be activated during ER stress [33, 34]. The antibodies used here were able to detect both pro-caspases and cleaved caspases 7 and 12. Fig. (6) shows that both pro-caspases and cleaved caspases were much more abundant in embryonic ventricles than in atria. More importantly, these caspases were more abundant in adult hearts compared to embryonic hearts, suggesting a greater degree of apoptosis in adult hearts, which may be related to cardiac remodelling.

In summary, in terms of markers of UPR activation and markers of apoptosis, there seems to be a greater degree of ER stress in adult hearts than embryonic hearts. This is contrary to the high degree of ER stress found in embryonic eye and brain [26]. In the embryonic heart, although ER-resident stress-related chaperones are more abundant than in the adult, ATF6α is decreased and eIF2α and XBP-1, components of the UPR, are not detectable. Thus, during embryonic heart development, the driving force for the extensive upregulation of ER stress chaperones seems to be the great synthetic load and not ER stress. On the other hand, in adult hearts, ER-resident stress-related chaperones are less abundant but there are increased levels of ATF6α and the ER stress-related caspases 7 and 12. It seems, then, that protein synthesis (synthetic load) during embryonic development is not as intense a stress as is required for apoptosis that is found during postnatal remodelling. Thus, during embryonic heart development, ER stress-related chaperones are activated but the UPR is not activated and neither is apoptosis induced to the levels found in adult hearts. The ER chaperones are an important part of a “quality” control” machinery of the ER [35], and may participate in UPR and ER stress, which may in turn lead to apoptosis [16]. ER stress (and UPR) [17-21] as well as apoptosis itself [22], have been associated with cardiomyopathies. Postnatal remodelling of the heart normally occurs but it may also be related to a variety of cardiac pathologies, such as cardiac hypertrophy and dilated cardiomyopathies.