2019 年 60 巻 11 号 p. 2456-2463
Cell sheet technology contributes to advances in tissue engineering. Although various approaches to control cell adhesion and detachment to a culture dish have been devised for the purpose of recovering an intact cell sheet so far, they all have disadvantages in application. Therefore, the search for superior cell sheet detachment technology is still ongoing. The present study describes detachment of human mesenchymal stem cells (hMSC) and their cell sheets via decomposition of pH-responsive CaCO3 particles that are precipitated on a culture dish. A concern in this detachment technology is harmful effects of the acidic environment, which is needed to decompose the CaCO3 particles, on cell viability. The time course of detachment behavior showed that the hMSC sheet was detached in a shorter time than individual cells. Shortening of the operation time for detachment suppressed cell death in the acidic environment. Thus, the hMSC sheet was successfully detached without cell death.
Fig. 1 A schematic illustration of cell detachment via decomposition of CaCO3 particles precipitated on a culture dish.
Tissue engineering has significantly progressed for regenerative medicine. For example, reconstruction of damaged tissues with scaffold materials has been applied in medical practice.1,2) However, for reconstruction with scaffold materials, the cell density supplied to the damaged tissue is often low because scaffolds occupy the spaces in tissue, resulting in limited applications in reconstructions. Another promising technology for tissue engineering is cell sheet technology in which a cell sheet is prepared with cells and extracellular matrix without a scaffold, which has a high cell density and can be supplied to damaged tissue.3)
One of issues to be resolved in cell sheet technology is detachment of the cell sheet from the culture dish without damage. Cell sheet detachment originally developed by Okano et al.4) was the one via the thermo-responsive poly(N-isopropylacrylamide) surface which shows fully hydrated and extended conformation at a low temperature of 20°C, but extensively dehydrates and changes to a compact chain conformation over 32°C. In this case, a cell sheet can be successfully detached by temperature control. Since this development, many methods for cell sheet detachment have been proposed, including light-induced detachment,5) detachment via electrochemical desorption of an adhesion layer,6) magnetic force-based detachment,7) detachment by pH control,8,9) detachment via decomposition of hydrogels,10,11) detachment via enzymatic decomposition,12,13) detachment related to a weak boundary layer,14) detachment with infused polymers,15) and ion-induced detachment.16) However, these methods could not solve problems such as high costs, specialized skills and equipment, low reliability, and a long operation time for cell sheet detachment. Therefore, the development of cell detachment technology is ongoing.
Calcium carbonate is not only easily formed without specialized skills or equipment, but also shows excellent cytocompatibility.17) Thus, calcium carbonate is expected to be used as a scaffold for bone grafting18) and nanoparticles for enteric drug delivery.19) In addition, it is easy to decompose calcium carbonate in acidic environment. Hence, cell sheets can be easily detached by decomposing calcium carbonate precipitated on a substrate, as shown in Fig. 1. The present study describes controllable detachment of human mesenchymal stem cells (hMSCs) and their cell sheets via decomposition of calcium carbonate. The detachment behavior for individual hMSCs and hMSC sheets was investigated from the viewpoint of the time course of detachment.
A schematic illustration of cell detachment via decomposition of CaCO3 particles precipitated on a culture dish.
The reagents used for cell culture and assays were as follows: Mesenchymal Stem Cell Growth Medium (MSCGM, PT-3001) and Trypsin/EDTA for MSCs from Lonza (Walkersville, MD, USA), and Trypan Blue Stain (0.4%), Dulbecco’s Phosphate Buffered Saline (DPBS) without Ca and Mg, and Live-dead viability/cytotoxicity kit from Invitrogen (Carlsbad, CA, USA). Materials used as a culture substrate were glass wafers with a diameter of 18 mm and thickness of 130–170 µm from Matsunami Glass industries (Osaka, Japan), calcium chloride and ammonium carbonate from Nacalai Tesque (Kyoto, Japan), and polyacrylic acid (average molecular weight: 250 kDa) from Wako Pure Chemical Industries (Osaka, Japan).
2.2 Synthesis and characterization of calcium carbonateA precursor solution was prepared by mixing calcium chloride (10 mM) and polyacrylic acid (10 mM) in distilled water. A circular glass coverslip, which was used as a culture dish, was thoroughly washed with ethanol and distilled water, immersed in the precursor solution, and placed in a desiccator with a wide mouth bottle containing 15 g ammonium carbonate for 1 h. Carbon dioxide generated by decomposition of ammonium carbonate was dissolved in an immersion liquid to lower the pH of the solution and precipitate calcium carbonate on the culture dish. After calcium carbonate precipitation, the culture dish was dried in air for 24 h.
The precipitates on the culture dish were observed by a scanning electron microscope (SEM) equipped with an energy-dispersive X-ray spectroscope (EDXS). EDXS analysis was performed on the dish before and after precipitation. In addition, X-ray diffraction analysis (XRD) was carried out on the precipitated dishes before and after annealing at 450°C for 3 h.
2.3 Decomposition of calcium carbonate by pH controlThe pH of DPBS was adjusted to 2.3, 4.1, and 7.6 by addition of hydrochloric acid. The pH of the adjusted DPBS was measured by a pH tester. The culture dishes with calcium carbonate were immersed in pH-adjusted DPBS for 1 h and then dried in air for 24 h. Subsequently, EDXS analysis was carried out on the dishes.
2.4 Cell cultureNormal bone marrow-derived hMSCs were purchased from Lonza (Walkersville, MD, USA) and cultured in MSCGM. Subconfluent hMSCs on polystyrene dishes were detached with trypsin/EDTA, and then hMSCs were seeded on glass coverslips, on which calcium carbonate was precipitated, and cultured in a humidified incubator (5% CO2, 37°C).
2.5 Individual cell attachment and detachment assayhMSCs were seeded on culture dishes with calcium carbonate precipitates at a cell density of 5 × 103 cells/cm2 and cultured for 24 h. For comparison, the cells were seeded on culture dishes without the precipitates at the same cell density and cultured for 24 h. The culture medium was removed and hMSCs were immersed in DPBS adjusted to pH 2.3, 4.1, and 7.6. Then, the cell behaviors of adhesion to the dish and detachment from the dish were monitored under an inverted optical microscope (CKX41, Olympus, Tokyo, Japan) with a digital camera (C-5060 (with calcium carbonate) or DP74 (without calcium carbonate), Olympus, Tokyo Japan). Attached and detached cells were counted to investigate the attached cell rate (RA) that was calculated as 100 × NA/(NA + NN) (%) (NA: the number of attached cells; NN: the number of detached cells4)). The identical assays were repeated three times under each pH condition.
2.6 Cell viability assayhMSCs were seeded on culture dishes with calcium carbonate precipitates at a cell density of 1 × 104 cells/cm2 and cultured for 24 h. After the culture medium was removed, pH-adjusted DPBS was added and hMSCs were cultured in a humidified incubator for 15 min. Then, the cultured cells (attached cells and detached cells) were collected by trypsin/EDTA treatment. The collected cells were stained with trypan blue, and live and dead cells were counted using a hemocytometer (Erma, Tokyo, Japan) to investigate cell viability calculated as 100 × NL/(NL + ND) (%) (NL: the number of viable cells; ND: the number of dead cells20)). Cell viability assays were repeated three times under each pH condition.
2.7 Formation of hMSC cell sheets and detachment assayTo form intact hMSC sheets, hMSCs were seeded on a culture dish with calcium carbonate precipitates at a cell density of 2 × 105 cells/cm2 and cultured for 24 h. After removal of the culture medium, the cell sheets were detached by immersion in DPBS adjusted to pH 4.1. These operations were carried out promptly and carefully to avoid applying mechanical stimulus to the cell sheets and dishes. The detachment behavior of the cell sheets was monitored using an optical microscope (BX51N-T31KS, Olympus, Tokyo, Japan). Cell sheet detachment experiments were repeated three times under the same conditions.
2.8 Live-dead staining of detached hMSC cell sheetsThe viability of hMSC cell sheets was assessed by the live-dead staining kit. The cell sheets were immersed in DPBS adjusted to pH 4.1 for 15 min and detached from a culture dish via decomposition of calcium carbonate precipitates. Thereafter, the cell sheets were thoroughly washed with DPBS and then incubated with a mixture of 1 µM calcein-AM and 2 µM ethidium homodimer for 30 min. After rinsing three times with DPBS, the cell sheets were observed using a fluorescence microscope (BX53, Olympus, Tokyo, Japan) with a digital camera (DP74, Olympus, Tokyo Japan), and analyzed with cellSens Standard software (Olympus, Tokyo Japan).
SEM images of precipitates on a culture dish are shown in Figs. 2(a) and (b). The precipitate sizes were in the order of several micrometers. EDXS analyses showed that the mapping areas of Ca, C, and O corresponded to the precipitate distribution (Fig. A1). However, mapping areas of other elements (Si, Au, Na, Al, K, Ti, and Zn) were not changed by the precipitation, indicating that the precipitates consisted of only Ca, C, and O.
Characterization of CaCO3 particles precipitated on a culture dish. (a, b) SEM images with low magnification (a) and high magnification (b) of CaCO3 particles precipitated on a culture dish. The size of precipitates was in the order of several micrometers. Scale bars indicates 10 µm (a) and 1 µm (b). (c) XRD results of the precipitates before and after annealing at 450°C for 3 h. The XRD analysis suggested that the precipitates were amorphous CaCO3.
XRD results of the precipitates before and after annealing are shown in Fig. 2(c). Peaks for CaCO3 were found after annealing, although the peaks were not found before annealing. This result suggests that the precipitates on the culture dish were amorphous CaCO3.
3.2 Decomposition of precipitates by pH controlThe EDXS results of Ca and K peaks for dishes immersed in DPBS adjusted to pH 2.3, 4.1, and 7.6 are shown in Fig. 3. K peaks were found in all investigated dishes, suggesting that the K peaks were related to the compositions on dishes. However, a Ca peak was found only on the dish immersed in DPBS adjusted to pH 7.6. This indicated that the precipitated CaCO3 particles were decomposed in DPBS adjusted to pH 2.3 and 4.1.
EDXS results of Ca and K peaks for dishes immersed in DPBS adjusted to pH 2.3, 4.1, and 7.6. The control indicates EDXS results of the dish before immersing in DPBS. The EDXS analysis suggested that the precipitated CaCO3 particles were decomposed in DPBS adjusted to pH 2.3 and 4.1, but not pH 7.6.
We used individual hMSCs that were attached on the culture dish and cultivated in an incubator for 24 h to investigate the detachment behavior via decomposition of calcium carbonate. Brightfield images of individual cells detached from dishes with CaCO3 precipitates (pH+CaCO3) and without CaCO3 precipitates (pH only) immersed in DPBS adjusted to pH 2.3, 4.1, and 7.6 for 0, 15, and 20 min are shown in Figs. 4(a) and (b), where the immersion time of 0 min is the time just after immersion. Clearly, the cells were detached by immersion in acidic media at pH 2.3 and 4.1, as shown in Fig. 4(a). On the other hand, Cell detachment hardly occurred in in the case of no precipitation of CaCO3 (Fig. 4(b)). Variations in the ratio of attached cells on the dishes with CaCO3 precipitates (pH+CaCO3) and without CaCO3 precipitates (pH only) as a function of the immersion time are shown in Figs. 4(c) and (d). As shown in Fig. 4(c), about 20% of the cells were detached in 0 min. This is probably because a small amount of calcium carbonate was decomposed in DPBS. Hence, about 20% of cell detachment was not related to decomposition of calcium carbonate in the acidic environment. The strong acidic environment at pH 2.3 led to detachment via decomposition of calcium carbonate immediately after immersion, whereas immediate detachment via decomposition of calcium carbonate did not occur in the weak acidic environment at pH 4.1. However, the cells were detached by successively immersing in the weak acidic medium at pH 4.1, and finally attached cell rate (RA (pH+CaCO3)) was decreased to about 20% by immersion for 20 min in the weak acidic medium as well as in the strong acidic medium. However, the cells on the dishes without CaCO3 precipitates still spread well, and attached cell rate (RA (pH only)) remained high even after immersion for 20 min in the strong and weak acidic medium (Fig. 4(d)). The results of quantitative comparison between RA (pH+CaCO3) and RA (pH only) is shown in Fig. 4(e). Clearly, there is a significant difference between with and without calcium carbonate on the dishes after 15 min of immersion in the acidic environment at pH 2.3 and 4.1. Therefore, cell detachment is related to the decomposition of calcium carbonate.
Detachment behavior of individual hMSCs. Brightfield images of individual cells detached from dishes (a) with CaCO3 precipitates and (b) without CaCO3 precipitates immersed in DPBS adjusted to pH 2.3, 4.1, and 7.6 for 0, 15, and 20 min. The immersion time of 0 min is the time just after immersion. All scale bars are 200 µm. Variations in the ratio of attached cells on the dishes (c) with CaCO3 precipitates and (d) without CaCO3 precipitates as a function of the immersion time. Results are expressed as ± standard deviation. * denotes significant difference between pH 7.6 and pH 2.3 or 4.1 with p < 0.05. (e) Difference in the ratio of attached cells between the dishes with CaCO3 precipitates and without CaCO3 precipitates as a function of the immersion time.
The cell viability after immersion in DPBS for 15 min was investigated by trypan blue exclusion (Fig. 5), where the control was cell viability at seeding. Most cells died in the strong acidic medium at pH 2.3, whereas many cells were alive in the weak acidic medium at pH 4.1. It was noted that about half of the cells had detached after immersion in the weak acidic medium for 15 min, as shown in Fig. 4(c), and many of the detached cells were alive.
Cell viability after immersion in DPBS for 15 min. The control is cell viability at cell seeding. Most cells died in the strong acidic medium at pH 2.3. However, many cells were alive in the weak acidic medium at pH 4.1. Results are expressed as ± standard deviation.
We investigated the detachment behavior of hMSC sheets immersed in DPBS adjusted to pH 4.1. Spontaneous detachment of the cell sheet was initiated after about 5 min. Brightfield images of the immersed cell sheet at 8 and 9 min are shown in Figs. 6(a) and (b). The cell sheet was smoothly detached, as shown in Figs. 6(a) and (b). Finally, the detachment was completed by immersion for 12 min without extrinsic stimuli such as vibration. The cell sheet detachment experiments were repeated three times, and the detachment was completed by immersion for 10–15 min in all experiments. Hence, cell sheet detachment via decomposition of calcium carbonate has high reliability. The live/dead fluorescent staining in the detached cell sheet showed that 97.8% of the cells remained alive after detachment (Fig. 6(c)). Cell-cell bonding was sustained in the detached cell sheet and the cell sheet was successfully detached without breaking the sheet.
Detachment of a cell sheer in the weak acidic medium at pH 4.1. (a) Brightfield image of a cell sheet immersed for 8 min. (b) Brightfield image of a cell sheet immersed for 9 min. The cell sheet was smoothly detached. Finally, the detachment was completed by immersion for 12 min without extrinsic stimuli such as vibration. (c) Live/dead fluorescent staining in the detached cell sheet. Most of the cells were alive after detachment.
The CaCO3 particles precipitated on the culture dish were decomposed in acidic media adjusted to pH 2.3 and 4.1. Decomposition of the precipitates on the culture dish was observed visually (data not shown). However, careful observation of the culture dishes revealed the presence of crystalline residues on the dishes after immersion in acidic media adjusted to pH 2.3 and 4.1. EDXS analyses showed that the residues consisted of Na and Cl, but not Ca (Fig. A2). Therefore, the residues may be precipitates of substances such as NaCl, which were dissolved in DPBS or the culture medium. Clearly, calcium carbonate was decomposed by immersion in acidic medium. Figure 4(a) showed that the cells were detached from the dish by immersion in the acidic medium. Hence, it is obvious that the detachment of cells is related to the decomposition of calcium carbonate. Detachment of cell sheets via decomposition of hydrogels has been reported previously,10,11) where the detachment was completed in a short time of 5 min.11) However, it should be noted that the area of the cell sheet recovered by the hydrogel method is much smaller than that by dissolution of calcium and other methods. In the present study, the detachment via decomposition of calcium carbonate was also completed in a short time of 10–15 min. Thus, detachment via decomposition is likely to be promising to shorten the operation time.
A concern in cell detachment via decomposition of calcium carbonate is that the acidic environment may lead to cell death. Gabi et al.21) reported that although C2C12 myoblasts died in acidic media of pH 1.2 and 2.0, most cells were alive in weak acidic medium with a pH of more than 3.0. This finding corresponds to the results obtained in the present study in which cell viability at pH 4.1 was higher than that at pH 2.3. Thus, the appropriate pH for cell detachment is suggested to be around 4. On the other hand, the acidic environment may have effect on other cell behaviors such as differentiation. In addition to acidic environment, calcium ion concentration in the medium may have effect on hMSCs. Lee et al.22) reported that when calcium ion concentration in the medium reaches 30 mM, it shows toxicity to cells. And, they also reported to promote hMSCs proliferation and calcification, even at calcium concentrations below 20 mM. Under the conditions for cell sheet detachment in this paper, the maximum concentration of calcium ion in solution is estimated to be 1.4 mM.23) This suggests that there is no toxicity, and rather an advantage for cell proliferation. It is important to investigate the effects for applications. The research is in progress. Anyway, it is important to shorten the operation time because a long-time operation may give harmful effects to cells even in a weak acidic environment.
An interesting result in the present study is that the operation time for detachment was shorter for the cell sheet than for individual cells. Specifically, the cell sheet was completely detached by immersion for 10–15 min, whereas about half of individual cells were attached to the dish after immersion at pH 4.1 for 15 min. The strength of cell adhesion depends on the adhesion energy which is the sum of three contributors: specific interfacial interactions, non-specific interfacial interactions, and the elastic energy stored in the cell membrane.24) For a cell sheet, a detached cell draws neighboring cells attached on the dish due to cell-cell binding because the detached cell tends to shrink to a spherical shape (Fig. 7(a)). Actually, it was observed that neighboring cells were drawn by the cell sheet, as shown in Fig. 7(b) where the drawn cell is shown by a red arrow. In the case of individual cells, however, such a drawing force is not generated because of no cell-cell binding. The drawing force affects the elastic energy stored in the cell membrane of the cell sheet, resulting in the reduced adhesion strength. The cell-cell binding depends on interactions via cadherins. The interactions via cadherins are enhanced by Ca2+.25) Because the Ca2+ concentration is increased by the decomposition of calcium carbonate, the cell-cell binding may be strengthened by immersion in acidic medium at pH 4.1, resulting in the reduced adhesion strength. The time course of detachment behavior showed that the hMSC sheet was detached in a shorter time than the individual cells, which may be related to the increased concentration of Ca2+ by the decomposition of calcium carbonate.
Cell-cell bindings induce rapid cells detachment. (a) Schematic illustrations of the reduced adhesion strength of a cell sheet. For a cell sheet, a detached cell draws neighboring cells because of cell-cell binding. However, such a drawing force is not generated because of no cell-cell binding in the case of individual cells. (b) Brightfield images of a cell sheet during detachment from a dish. The cell shown by a red arrow is drawn by the cell sheet. The scale bar indicates 100 µm.
In conclusion, individual hMSCs and a hMSC sheet were detached by decomposition of calcium carbonate precipitated on culture dishes. In particular, the hMSC sheet was successfully detached with suppressing cell death. The hMSC sheet was detached in a shorter time than individual cells. The shortening of operation time for detachment is suggested to suppress cell death in the cell sheet.
Cell sheet detachment via decomposition of calcium carbonate is promising because it is a controlled detachment method with low cost. However, further study is needed because it remains unknown whether this detachment technology can be applied to other cell types. Also, it is important to investigate effects of acidic environment on other cell behaviors such as differentiation.
EDXS results of precipitates on a culture dish. (a) SEM image, (b) Ca mapping, (c) C mapping, (d) O mapping, (e) Si mapping, (f) Au mapping, (g) Na mapping, (h) Al mapping, (i) K mapping, (j) Ti mapping, and (k) Zn mapping. The results indicated that the precipitates consisted of only Ca, C, and O.
EDXS results of crystalline residues on a dish after immersion in DPBS adjusted to pH 4.1. (a) SEM image, (b) Ca mapping, (c) Na mapping, and (d) Cl mapping. The EDXS analyses showed that the residues consisted of Na and Cl, but not Ca.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
