1. Introduction

Neural progenitor cells (NPCs) derived from human induced pluripotent stem cells (iPSCs) can be used to study the pathological model and cellular treatment of traumatic brain injury and neurodegenerative diseases [1,2].However, differentiation from iPSCs to NPCs is a complex and highly regulated process [3,4], the underlying molecular mechanisms of which are currently unknown but urgently need to be addressed. Among them, mitochondria are the main organelles in cells that produce energy and play a crucial role in cell differentiation and determine cell fate by regulating energy metabolism and signal transduction [5–8]. Mitochondrial pH is a key factor in mitochondrial metabolism and function, and changes in mitochondrial pH may have a significant impact on cell differentiation processes [9–11].Therefore, studying the fluctuation of mitochondrial pH value during iPSCs differentiation can reveal the acid-base balance information during differentiation, and help to further understand the metabolic mechanism during differentiation at the organelle level [12–14].

Due to the intricate cellular environment and small size, detecting mitochondrial pH is a challenging endeavor. The detection techniques primarily rely on principles such as fluorescence and Surface-enhanced Raman Scattering (SERS) [15,16] . Although fluorescence imaging is widely utilized [17,18], it suffers from issues like easy quenching of fluorescent dyes and the arduous removal of cell autofluorescence [19,20]. These problems result in non-repeatable measurements and background interference. On the other hand, SERS technology operates on the principle of significantly amplifying the Raman signal of substances on the nanostructure of the plasma. This technology exhibits ultra-high sensitivity, enabling even single molecule detection [21]. Moreover, compared to fluorescence technology, SERS technology demonstrates resistance to photobleaching and quenching, displays a narrower spectral peak, and can be excited across a wider wavelength range [22]. Thanks to these unique properties, numerous pH nanosensors based on SERS technology have been developed and extensively applied in the pH detection of cells and tissues [23–26]. However, due to the high complexity and extended differentiation period of cell differentiation, accurately characterizing the fluctuation of mitochondrial pH during long-term cell differentiation remains a challenge. SERS technology can serve as an effective means to address this issue.

Herein, we studied the evolution of mitochondrial microenvironment during directionally induced iPSCs differentiation into NPCs by SERS technique. Mitochondria targeted pH response to mature SERS nanoprobes was used to explore the changes in mitochondrial pH at different stages of the cell-induced differentiation process, and to further compare the intra-mitochondrial pH values between the successful and failed batches of differentiation. The main objective of this paper is not method development, but to apply the method to stem cell research, and its novelty and important findings can be summarized in three aspects: (1) The change of mitochondrial pH during induced differentiation was reported for the first-time using SERS technology. We found that mitochondrial pH value decreased in the early stage of differentiation and increased in the late stage of differentiation, which may be a sign of the transformation of mitochondrial metabolic state during differentiation. (2) By comparative analysis of the change trend of mitochondrial pH value between abnormal differentiation batch and normal batch, the response to embryoid bodies (EB) abnormal growth stage was carried out, thus providing a means to evaluate the differentiation level. (3) NPCs have lower mitochondrial pH than iPSCs, which reflects the acidic regulation of the microenvironment of differentiated cells in the direction of neuroectoderm by iPSCs.

2. Material and methods

2.1 Chemicals and reagents

The iPSCs (code: ACS-1011TM) were introduced by iCell Bioscience Inc (iCell, Shanghai). The iPSCs base medium (with additives) and iPSC digestive solution were purchased from Sebikon (China). AggreWell800 six-well plate, Neural Rosette Selection Reagent and Neural Progenitor Medium were purchased from Stem Cell Technologies (Canada). Nestin was purchased from RI&D systems (USA), SOX2, PAX6, goat anti-Rabbit IgG HI&L (Alexa Fluor 488) and goat anti-mouse IgG HI&L (Alexa Fluor555) were purchased from Abcam (China). Gold nanorods (AuNRs) were purchased from NanoSeedz (China). Mitochondrial localization signal (MLS) was purchased from Apeptide (China), MLS peptide: MLALLGWWWFFSRKKC, 4-mercaptopyridine (4-MPy) was purchased from Beijing Putian Tongchuang (China). N2 Supplement (100X), B27 Supplement (50X), GlutaMAX, DMEM culture medium (1X) and phosphate buffered saline (PBS) pH 7.4 buffer (1X), purchased from Gibco (USA). BCECF-AM was purchased from Beyotime (China). Ultrapure water from the Milli-Q Millipore system was used for all experiments. All chemicals are analytical grade and can be used without further purification.

2.2 Synthesis and characterization of SERS probes

The mitochondrial localization signal (MLS) peptides and the pH responsive Raman reporter molecule 4-MPy were modified on the surface of AuNRs to obtain the SERS probe with two functions of mitochondrial targeting and pH response. The specific steps are as follows: 112$\mathrm{\mu}$L 0.5nM MLS aqueous solution is added into 1.0mL AuNRs solution with mass fraction of 145.36$\mathrm{\mu}$g/mL, and the reaction is stirred at room temperature for 8 h. The excess MLS molecules were then removed by centrifugation (6000 rpm, 8min). Then 200$\mathrm{\mu}$L 40 $\mathrm{\mu}$M 4-MPy ethanol solution was added to 1ml AuNRs-MLS and stirred at room temperature for 8 h. The excess 4-MPy molecules were removed by centrifugation (6000 rpm, 8min) to obtain AuNRs-MLS-(4-MPy). Ultraviolet-visible spectrophotometer (UV-2700, Shimadzu, Japan), transmission electron microscope (JEM-1400PLUS, JEOL, Japan) and nano particle size potential analyzer (NanoZS-90, Malvern Zetasizer, UK) were used to characterize the preparation of SERS probes.

2.3 pH sensing performance of SERS probes

The pH calibration curve was generated by cell medium (DMEM with 10% fetal bovine serum added) and pH adjusted by NaOH and HCl. SERS spectrum of AuNRs-MLS-(4-MPy) at different pH values were obtained by confocal Raman system (Invia, Renishaw, UK). The confocal Raman system uses He-Ne laser as the excitation source (632.8 nm wavelength), sets the excitation light power to 7.444 mW, and the spectrum acquisition range is 400 cm$^{-1}$ to 1800 cm$^{-1}$, and the exposure time of each spectrum is 10 seconds. The SERS spectra of nanosensors were measured at different pH values, with 30 spectra per group. Pre-processing analysis using Wire 4.3 software, including narrow cosmic ray removal, baseline correction, and noise smoothing. The probes were dispersed in different pH calibration fluids, and SERS spectra of the probes were collected. The relative intensity of the spectrum at 1001cm$^{-1}$ (ring breathing) and 1091 cm$^{-1}$ (X- sensitive/ C-S) from the reporter molecule 4-MPy was plotted as a pH calibration curve based on linear analysis by Origin 2021 software.

2.4 Cell culture

In this study, differentiation was induced by single inhibition of bone morphogenetic protein (BMP) [27]. The iPSCs cultured in Matrigel coated vials or petri plates at 37 $^{\circ }$C and 5% CO$_{2}$ using a specific iPSC medium. iPSCs are treated with a unique iPSC digestive solution to separate it from Matrigel and then inoculated onto a Matrigel-coated 24-well plate. When the cells filled the bottom of the pore plate, they were transferred to the AggreWell800 six-well plate at a density of 6$\times$106 cells per pore. The cells were then cultured with M1 medium (basal nerve medium, 1% GlutaMAX, 1% penicillin/streptomycin, 1% N2, 2% B27, and 1.25 $\mathrm{\mu}$M AMPK inhibitor). EBs of various sizes and shapes are formed in a single day. Replace 50-75% of the medium daily. After EB grew to day 10, the cell debris was filtered out with 40 $\mathrm{\mu}$m cell filter to obtain pure EB and washed with M2 medium (DMEM/F-12 medium, 1% GlutaMAX, 1% N2, 2% B27, and 20 ng/mL FGF2). Then the EBs were resuspended in Matrix Matrigel-coated six-well plates, neural rosettes were visible within 5 days. NPCs were isolated and selected by using Neural Rosette Selection Reagent and repeated the selection process for neural rosettes during subsequent passages for further purification of NPCs. NPCs were cultured with Neural Progenitor Medium.

2.5 iPSCs and NPCs identification with immunofluorescence

Frist, both iPSCs and NPCs were seeded onto clean slides pretreated with Matrix. Then, at 4 $^{\circ }$C, immobilize 4% formaldehyde for 20 minutes. The cells were then cleaned with PBST (PBS 0.1% TritonTM X-100). To allow the cells to penetrate, they were soaked in PBST at 4 $^{\circ }$C for 10 minutes. The cells were then blocked with 10% goat serum (Sigma-Aldrich). When immunostained, key antibodies including Anti-SOX2 antibody (Abcam, 1/200), PAX6 antibodies (Abcam, 1/350), Oct-3/4 antibody (Santa, 1/200) and Nestin antibody (R&D system, 1/40) were diluted in a closed buffer and cultured with cells at 4 $^{\circ }$C for 18 24 hours. The cells were then cleaned with PBST. Secondary antibodies, including goat anti-rabbit IgG HI&L (Abcam) and goat anti-mouse IgG HI&L (Abcam), were diluted in a closed buffer and incubated with cells for 1 hour at room temperature under no light. Finally, the cells were re-stained with Hoechst33342 nuclear staining. Finally, the air-dried fluorescently stained samples were observed and imaged using an Olympus IX83 inverted microscope. ImageJ software was utilized for processing the fluorescent images.

2.6 Cell morphology characterization

Detecting morphological changes induced by pluripotent stem cell differentiation, utilizing the Olympus IX 83 inverted microscope and the digital holographic microscope developed independently by the research team for phase-contrast imaging and quantitative phase imaging to assess cell morphology. Cells before and after differentiation were respectively seeded on glass slides pre-coated with Matrix, incubated at 37 $^{\circ }$C in a 5% CO2 incubator until adherent, remove the culture medium, add 1 mL of paraformaldehyde, place in a 4 $^{\circ }$C environment to fix the sample for 20 min, then wash the sample with PBS, rinse twice with ultrapure water, air dry, and store at 4 $^{\circ }$C for testing. The camera used for digital holography is CMOS (BFS-U3-16S2M-CS), with a pixel size of p = 3.45 $\mathrm{\mu}$m, and a pixel size of 1440 $\times$ 1080 ; using a semiconductor laser with a wavelength of 639.49 nm and a linewidth of 1.229 nm (638 nm - 70 mW - R44768) as the light source. Objective magnification: 20$\times$. Use ImageJ software to add scale bars to the images.

2.7 Measurement of mitochondrial pH

Frist, both iPSCs and NPCs were seeded onto clean slides pretreated with Matrix, and 50 $\mathrm{\mu}$L SERS probes were added to each sample. The SERS probes were incubated with the cells for 24 h. Aspirate the supernatant from the cells, and add Accutase enzyme to detach the cells from the substrate. Then the cells were fixed with tissue fixating solution for 20 min, centrifuged with PBS twice, and washed with ultra-pure water twice to obtain cell suspension. The SERS spectra of intracellular probes was measured by dropping cell suspensions on a clean gold film. The SERS spectrum of intracellular probes was measured after the cell suspension was air-dried by dropping on a clean gold film. For SERS measurements, the experiment employs a 633 nm laser as the excitation light source, with a beam size of 1$\mathrm{\mu}$m and a power of 100% . Dynamic mode is utilized with an exposure time of 10 s and integration performed once, using a Leica microscope with a 50$\times$ objective lens (NA=0.75). The collected spectral wavenumber range is from 400 to 1800 cm $^{-1}$. The intracellular pH level can be understood according to the signal intensity changes of SERS probes. 10 spectra were collected from different positions of each cell, and 10 cells were randomly selected at different stages for SERS spectrum collection. Firstly, the obtained SERS spectral data was pre-processed by using WiRE4.3 software to remove cosmic rays and correction of baselines , and the mitochondrial pH was obtained by the SERS spectral ratio of the intracellular probe and the pH calibration curve of the probe, and the Origin 2021 software was used for mapping.

2.8 Measurement of cytosolic pH

2’,7’-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein, acetoxymethyl ester (BCECF-AM, Molecular Probes, Beyotime, China) was used as a measure of the pH change trend of cytoplasm during differentiation of induced pluripotent stem cells. The iPSCs use dedicated iPSC digestive solution and other cells in the differentiation process use Accutase enzyme for detaching the cells from the substrate. After 1 min, fresh medium is added, cell suspension was centrifuge at 1000 rpm for 5 min, then added 1 $\mathrm{\mu}$M 1 mL BCECF-AM staining solution in a cell suspension with a density of 2 $\times$ 10$^{6}$ cells /ml, and incubated in a 5% CO$_{2}$ incubator at 37 $^{\circ }$C for 30 min. The trend of cytoplasmic pH fluctuations was determined by the ratio of the emission intensity (535 nm) at 488 nm (pH-dependent) excitation to that at 440 nm (PH-independent) excitation with a fluorescence spectrophotometer (F-4600, HITACHI, Japan) [28].

3. Result and discussion

3.1 Characterization of SERS probe

The synthesis process of SERS probe is shown in Fig. 1(a). Referring to previous method [26], the mitochondrial localization signal (MLS) peptides and the pH responsive Raman reporter molecule 4-MPy were successively modified on the surface of AuNRs (AuNRs-MLS-(4-MPy) for short) to obtain the SERS probe with two functions of mitochondrial targeting and pH response. The transmission electron microscopy (TEM) image of the final synthesized SERS probe shows that the probe has an average length of 80$\pm$2 nm and width of 40$\pm$1 nm, covered with a layer of MLS and 4-MPy (Fig. 1(b)). Fig. 1(c) shows the absorption spectra of the probe at different synthesis stages. Compared with pure AuNRs, the absorption peaks of modified probes showed a redshift, which could be attributed to the coupling effect of MLS and MPy [29]. Besides, the Zeta potential of the modified probe decreases (Fig. 1(d)), indicating that the chemical characteristics of AuNPs surface have changed after each step of modification, verifying the success of SERS probe synthesis. Further, to prove the mitochondria-targeting of the AuNPs-MLS-(4-MPy) probes, relevant fluorescence localization experiments were performed, and the results are shown in Fig. 1(e-g). The SERS probes were modified with fluorescein isothiocyanate (FITC) and then incubated with iPSCs for 24 h. The mitochondria in iPSCs were stained with Mito-Tracker Red CMXRos. The red fluorescence in Fig. 1(e) represents the location of the mitochondria in iPSCs, while the green fluorescence in Fig. 1(f) represents the distribution of the probes. The merged image shown in Fig. 1(g) presents a high overlap between green and red fluorescence, indicating the successful targeting of the SERS probes to the mitochondria in iPSCs.

Since the interior of mitochondria is a weakly alkaline environment, we investigated the pH response of the probe in a weakly alkaline environment. In the pH range of 6.2-9.0, 6 groups of pH values were set, and SERS spectra of the probes under different pH values were detected, as shown in Fig. 2(a). We observed that the peak intensity ratio of the SERS spectrum at 1091 cm$^{-1}$ (X-sensitive/ C-S) and 1001 cm$^{-1}$ (ring breathing) positions vary with changes in pH . This is due to the pH-dependent ionization behavior and resonant structure of 4-MPy, as illustrated in Fig. 2(b) [30]. Therefore, these two peaks are often used to describe fluctuations in pH [24,31]. Linear equation fitting was performed according to the ratio of pH value in Fig. 2(a) to the corresponding spectrum at 1001 and 1091 cm$^{-1}$ , and the calibration curve obtained was shown in Fig. 2(c). The fitting formula is ${y=-0.286x-0.328}$, $R^{2}$ =0.996,which proves that the mitochondria targeting pH response probe has a good linear response in the weakly alkaline pH range of 6.2-9.0.

 

Fig. 1. (a) The synthesis process of the AuNRs-MLS-(4-MPy) probe. (b) TEM image of AuNRs-MLS-(4-MPy) probes. (c) UV-vis spectrum and (d) Zeta potential of AuNRs, AUNRs-MLS, AUNRs-MLS-(4-MPy). (e-g) Fluorescent images of iPSCs treated with the AuNRs-(MLS-FITC)-(4-MPy) for 24 h; mitochondria of iPSCs were stained red with Mito Tracker red CMXRos (e) and AuNRs-(MLS-FITC)-(4-MPy) probes were stained green with FITC (f). (g) Fluorescence merge image of e and f.

Download Full Size | PDF

 

Fig. 2. (a)SERS spectra of AuNRs-MLS-(4-MPy) probes in different pH environment. (b) Ionization behavior and resonance structure of Raman reporter 4-MPy modified on AuNRs at different pH. (c) Standard curve of SERS spectral intensity ratio $I_{1091}$ / $I_{1001}$ changing with pH under different pH.

Download Full Size | PDF

3.2 Differentiation of iPSCs to NPCs

Fig. 3(a) shows corresponding phase-contrast images of cells at different differentiation stages. In the undifferentiated stage, iPSCs appear in a single-cell state (Fig. 3(a1)) and exhibit stable growth. As iPSCs form stable colonies, a inducer named M1 is added to induce them into the EB stage. On the first day after adding M1 inducer to the cells, the cell state is termed EB1d, and the cells show features of aggregation and overlap (Fig. 3(a2)). On the fourth day after adding M1 inducer, the cell state is termed EB4d, and the cells transition from adherent to suspension growth, changing their state (Fig. 3(a3)). On the seventh day after adding M1 inducer, the cell state is termed EB7d, and the cells continue to maintain a suspension state and grow (Fig. 3(a4)). On the tenth day after adding M1 inducer, the cell state is termed EB10d. It can be observed from Fig. 3(a2-a5) that there is no significant difference in the morphology of cells in suspension state. The first ten days of pluripotent stem cell differentiation are collectively referred to as the EB period, during which cells in EB further mature. Subsequently, cells are treated with a differentiation inducer named M2 to promote the transition of EB to NPC, and the cells transition to an adherent state; After repeated sorting and purification, NPCs can be obtained, with stable and dispersed cell morphology (Fig. 3(a6)). The single-cell phase maps of iPSC and NPC shown in Fig. 3(b1 and b2) better illustrate the significant morphological differences between the two cell types before and after differentiation. The optical height of iPSC can reach up to 4.8 rad, while the differentiated NPC has a maximum optical height of only 3.7 rad. NPCs are flatter than iPSCs, with more protrusions.

Throughout the whole differentiation process, we can see that the morphological changes of cells are complex. However, it is not enough to rely solely on morphological features to characterize differentiation. Therefore, we conducted another immunofluorescence staining experiment, and the results were shown in Fig. 4. SOX2 and Oct4 are the main regulator of iPSCs pluripotency [32]. Nestin, a cytoskeletal protein, is a widely used molecular marker for NPCs. PAX6 is a neuroectodermal marker that differentiates the human central nervous system [33]. Nucleus staining as shown in Fig. 4(a1-e1) plays a role in cell localization. In iPSCs, the strong expression of nuclear pluripotency markers SOX2 and Oct4 is shown in Fig. 4(a2 and a3), and the weak expression of NPCs marker Nestin is shown in Fig. 4(b2 and b3). In NPCs, Nestin and PAX6 are strongly expressed in NPCs as shown in Fig. 4(c2 and c3), SOX2 is weakly expressed as shown in Fig. 4(d2) and d3, and OCT4 is not expressed as shown in Fig. 4(e2 and e3). The fourth column in Fig. 4(a4-e4) are the merged of the staining results from the first three columns. The results of immunofluorescence staining indicated the successful induction of iPSC into NPC.

 

Fig. 3. The phase contrast images of cells at different differentiation stages (a1-a6). Single-cell phase images of iPSC (b1) and NPC (b2).

Download Full Size | PDF

 

Fig. 4. Immunofluorescence staining of iPSCs and NPCs.(a1) (b1) (c1) (d1) (e1) Stained with Hoechst; (a2) (b2) (c2) (d2) (e2) immunostaining with SOX2 or PAX6; (a3) (b3) (c3) (d3) (e3) immunostaining with OCT4 or Nestin and (a4) (b4) (c4) (d4) (e4) merged.

Download Full Size | PDF

3.3 Detection of mitochondrial pH during iPSCs differentiation to NPCs

The corresponding mitochondrial pH was obtained from the probe’s SERS spectra at different differentiation stages (Fig.S3), as shown by the black data in Fig. 5(a). From iPSCs to EB4d, the mitochondrial pH value showed a downward trend: decreased from 8.0$\pm$0.1 to 7.6$\pm$0.1. It can be explained as follows, at the initial stage, the cell is in an adherent state (Fig. 3(a1 and a2)), the cell proliferates and the cell requires a lot of energy to support the cell division and growth, so the mitochondria increase significantly during the initial differentiation to meet the cell’s energy needs [34,35]. Increased mitochondrial respiration, initiation of glycolysis, and increased rate of glycolysis producing metabolic acid resulted in decreased mitochondrial pH [36,37]. From EB4d to NPCs, mitochondrial pH increased and NPCs mitochondrial pH increased to 7.9 $\pm$ 0.1. The reason for this was that EB changed from an adherent state to a suspension state on day 4, which restricted cell proliferation and increased cell differentiation [38], as shown in Fig. 3(a3). The metabolic behavior in this stage is different from that in the previous proliferative stage, mitochondrial metabolism tends to be oxidative phosphorylation (OxPhos), accompanied by an increase in the ratio of NAD+/NADH, resulting in an upward trend in mitochondrial pH at the late stage of differentiation [39,40]. Trends in cytoplasmic pH during differentiation were detected by BCECF-AM fluorescence probes, as shown by Fig.S5. The change trend of cell intracellular pH value is roughly the same as that of mitochondrial pH value, confirming the accuracy of SERS in detecting mitochondrial pH changes. It is worth noting that at the stage of EB4d to EB10d, the change trend of intracellular and mitochondrial changes is somewhat different, showing a trend of first rising and then decreasing, due to the difference in the number of cells detected and the location of the cells. Furthermore, due to the easy quenching of fluorescent dyes, fluorescence spectra can only be measured a limited number of times, while SERS spectroscopy measurements do not affect the signal of SERS probes, demonstrating the stability of SERS technology in measuring intracellular pH values.

 

Fig. 5. (a) Mitochondrial pH value measured by SERS spectra. (b) Phase contrast images of two cell differentiation EB7d and EB10d, where the first column corresponds to normal differentiation batch and the second column corresponds to abnormal differentiation batch.

Download Full Size | PDF

To investigate whether this particular change in mitochondrial pH occurs only during successful differentiation, we also examined mitochondrial pH in cells during abnormal differentiation, as shown in the red data in Fig. 5(a). Dysplasia of EB prevented this population of cells from differentiating into NPCs stages and differed from the mitochondrial pH of the successful batch. The mitochondrial pH of aberrant cells decreased from day 4 to day 10 of EB growth to 7.3 $\pm$0.1 on day 10 compared to successfully differentiated cells. Fig. 5(b) provides phase contrast images of the two batches of differentiated EB, showing that stunted EB volumes were smaller than normal at day 7 and showed signs of fragmentation at day 10. Mitochondrial pH imbalance is a key feature of cellular abnormalities and will result in reduced ATP synthesis [41]. Inadequate mitochondrial energy supply will lead to EB dysplasia and failure of cell differentiation.

Therein, we detected and quantified the mitochondrial pH during the differentiation of iPSCs to NPCs, and found that there was a large difference in the change of differentiation pH between the failed batch of differentiation and the successful batch of differentiation. By comparing the mitochondrial pH of these two differentiation batches, we could infer that the growth of EBs started to be abnormal at 4 to 7 days, and therefore, by the trend of mitochondrial pH change, we could make a preliminary judgment of the growth level of EBs.

4. Conclusions

In this study, we employed SERS technology to longitudinally monitor the changes in mitochondrial pH during the differentiation of iPSCs into NPCs induced by a single BMP inhibition method. During normal differentiation, mitochondrial pH decreased from 8.0 to 7.6 in the early stage and increased to 7.9 in the later stage, as shown by our results. Moreover, it was verified by fluorescent probe that iPSC regulated the acidic microenvironment of NPC-differentiated cells. The pH value of mitochondria of abnormally differentiated cells was different, and decreased to 7.3 in the late stage of differentiation, reflecting EB dysplasia and abnormal cell differentiation. Our results help to reveal the changes and regulatory mechanisms of mitochondrial metabolism during cell differentiation, and lay a foundation for further understanding of the physiological processes and regulatory networks of cell differentiation. From another perspective, our study also reveals potential targets for future research aimed at improving iPSC differentiation protocols for regenerative medicine applications. In particular, mitochondrial pH levels were modified by manipulating mitochondrial ion channels to improve the efficiency of iPSC differentiation.