The rat cardiomyoblast-derived H9C2 cell line was purchased from the Korean Cell Line Bank. The H9C2 cells were maintained in Dulbecco's modified Eagle's medium supplemented with fetal bovine serum(FBS; 10%), penicillin (100 µg/mL), and streptomycin (100 µg/mL) (DMEM) in 5% CO₂ humidified air at 37℃.

The added concentration and maintenance time for NaCl was selected using the forward side scatter pulse height (FSC-H) values from flow cytometry (Table 1, Supplement 1). Based on this, 30mM NaCl was added to the DMEM. The media were filtered through 0.4 µm filters. Cells were maintained for 48 h to produce the in vitro hypertrophic models.

Suspended H9C2 cells in PBS (phosphate-buffered saline) were used. To exclude osmotic effects, the cells were incubated in normal osmotic PBS for 30min. After incubation, 10,000 cells were automatically analyzed by FSCH to measure the cell size and side scatter pulse height (SSC-H) in order to study the intracellular granules (FACS Calibur™).

Cells were fixed in −20℃ methanol for 5 min at room temperature (RT). The cells were washed 3 times with ice-cold PBS and then blocked with 10% FBS in PBS for 30 min at RT. After blocking, the cell nuclei were stained with 1 µg/mL of 2-(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI) in 3% FBS and PBS for 30 min at RT. After staining with DAPI, the images were visualized with a confocal microscope and a coupled image analysis system(Nikon).

A search of the amino acid sequences of a conserved domain of HSP22 and various LEA3 proteins was conducted using BLAST15). Utilizing this program, the amino acid sequences in the conserved domains were compared between HSP22 and LEA316).

The trypan blue exclusion test was performed17). The results were calculated using the following equation: (the total number of viable cells/the total number of cells)×100.

Total RNA was extracted. The RNA quantity and quality were measured using the 260/280 ratio obtained from a Nano-drop spectrophotometer. Using RT-PCR, cDNA was synthesized from RNA. For the analysis of mRNA expression, forward and reverse primers were used (Table 2). One µL of each set of primers was used with TOPreal™ qPCR premix (10 µL), nuclease-free water (20 µL), and a cDNA sample (1 µL). The Tm value of the primers was 58℃. After an initial 10 min at 95℃, 45 cycles of the following protocol were completed: 95℃ for 10s, 60℃ for 15s, and 72℃ for 30s. The fold changes of HSP22 were compared with those of a house keeping gene (18s ribosomal RNA(rRNA)). The data were analyzed using the ΔC_t method18). To confirm the primer activities, an ethidium bromide-stained agarose gel (2%) was used for DNA gel electrophoresis. The size of the HSP22 product was 129 base pairs (bp), while that of the 18s rRNA product was 135 bp.

The cells were loaded with 2mM of the fluorescence probe Calcium Green-1™ in DMEM for 2 h. Cells were washed twice with PBS, and 10,000 cells were analyzed using flow cytometry. Single cells were observed under a confocal microscope after fixation with ice-cold methanol.

All data were expressed as the mean±S.E., and n indicated the number of cells used. For all tests, cells were obtained from a minimum of 3 different culture dishes for each treatment and experiment. Analysis of the data was conducted using Student's t-test and one-way ANOVA. p<0.05 was considered to be significantly different in the statistical analyses.

To compare the characteristics of the in vitro hypernatremic model, an Ang II-induced hypertrophic model (Ang II model) was used, because treatment with Ang II induces cellular hypertrophy in H9C2 cells, which mimics the hypertrophic responses of primary cardiomyocytes3). First, the FSC-H values significantly increased following supplementation with 30 mM NaCl (p<0.05) and 1 µM Ang II for 48 h (p<0.01) (Figs. 1 and 2), indicating an increase in the size of the cells. The mean of the FSC-H was 391.21±6.63 for the control group. The mean of the FSC-H for the 24 h group with 30mM NaCl did not change. However, the mean of the FSC-H for 48 h was 416.49±17.08. The mean FSC-H value was 356.77±9.26 for the control group. The mean of the FSC-H for the 24 h group with 1 µM Ang II did not exhibit a statistically significant change. However, the mean of the FSCH value for the cells maintained for 48 h was 375.09±7.61. There were several factors that could have affected the control samples. For this reason, different controls were used in each experiment to ensure an accurate comparison. The difference in values was more evident for the supplementation with NaCl. Second, SSC-H was constant among the controls after treatment for 24 h and 48 h with each hypertrophy-inducing agent (Fig. 3). Third, the nuclear and cell morphologies were similar among the controls and the 2 models under confocal microscopy with DAPI staining (Supplement 2). Our hypertrophic model, induced by the addition of NaCl, contained less granular content in the cytoplasm than other models. The results indicated that the addition of 30mM NaCl did not strongly affect the cells. Finally, the cell viability was examined (Fig. 4). For all of the groups, there were no changes in the cell viability. Therefore, this model did not exhibit abnormal morphological changes, and the cellular hypertrophy induced by these conditions was termed the in vitro hypernatremic model.

RT-PCR was conducted to compare the mRNA expression patterns after exposure to Ang II or NaCl for 24 or 48 h. HSP22 was compared with a house-keeping gene. The mRNA expression of HSP22 did not increase after treatment with Ang II for 24 h, but the expression of HSP22 increased after 48 h. In contrast to the Ang II model, the expression of HSP22 in the in vitro hypernatremic model gradually decreased in a time-dependent manner (Fig. 5). After these experiments, DNA gel electrophoresis was used to confirm the product of each primer set used for RT-PCR (Supplement 3). The results indicated that the design of the primers and PCR analysis were conducted appropriately.

A gradual decreasing pattern, in response to high NaCl, has been shown for LEA3 in Artemia19). Based on the similarity of the decreasing patterns for LEA3 in Artemia and HSP22, a comparison of the amino acid sequences was conducted. LEA3 proteins have several types of conserved domains (Table 3)20). The sequences of these conserved domains were also compared. The consensus sequence of HSP22 was found at residues 108-128, which corresponds to residues 117-137 of LEA3 from Artemia. The identities were 33%(7/21); 57%(12/21) were positive, there were 0 gaps, and the expectation was 3.9. The similarity was meaningful, because part of the α-crystalline domain of HSP22 was similar to 3^a motif sequence of LEA3. The similar expression patterns and amino acid sequences for HSP22 and LEA3 indicated that HSP22 might have an LEA-like function20).

To determine the roles of HSP22 in hypertrophy, the intracellular Ca²⁺ level was compared using a Ca²⁺-binding dye for the control; the FSC-H increased for the 2 hypertrophic models (p<0.001). Unbound Ca²⁺ increased in the Ang II model (p<0.05). However, compared with the Ang II model, Ca²⁺ significantly decreased in the in vitro hypernatremic model (p<0.01). To compensate for the size of the cells using the equation: unbound Ca²⁺/FSC-H×100, the corrected Ca²⁺ decreased in the Ang II model (p<0.05), and the decrease in Ca²⁺ was more evident with NaCl (p<0.001) (Fig. 6). The corrected result showed that size of the 2 hypernatremic models increased as the intracellular free Ca²⁺ decreased. Moreover, Ca²⁺ was confirmed in a single cell (Fig. 7). The results indicated that the role of HSP22 in cellular hypertrophy is related to impaired intracellular Ca²⁺ homeostasis.