A human iPSC line previously established from peripheral mononuclear cells obtained from a healthy subject employing episomal vectors (pCXLE-hOCT3/4-shp53-F, Cat.#. 27077; pCXLE-hSK, Cat.#. 27078; pCXLE-hUL, Cat.#. 27080; and pCXWB-EBNA1, Cat.#. 37624, Addgene, Watertown, MA, United states) was used in this study. Human iPSCs were maintained on iMatrix-511 (Cat.#. 892021, Matrixome, Osaka, Japan)-coated 6-well plates (Cat.#. 3516, Corning, Glendale, Arizona, United States) in StemFit AK02N (Cat.#. AK02N, Ajinomoto, Tokyo, Japan), and were passaged every 7 days.

“pAAVS1_SA_loxP_PGK_Neo_loxP/CAG_IRES_EGFP/AmpR” that we previously generated was used for a donor vector (15). Human SLC5A5 open reading frame (ORF) subcloned into pCMV6-XL5 vector (Cat.#. OHu19270C) was purchased from GenScript (Piscataway, NJ, United States). SLC5A5 ORF was amplified by polymerase chain reaction (PCR), purified, and ligated with an XhoI-digested control donor vector using In-Fusion HD Cloning Kit w/Competent Cells (Cat.#. 639642, Clontech, Mountain View, CA, United States) to obtain “pAAVS1_SA_loxP_PGK_Neo_loxP/CAG_ SLC5A5_IRES_EGFP/AmpR”. The PCR primers are listed in Table 1.

Human iPSCs were harvested using StemPro Accutase Cell Dissociation Reagent (Cat.#. A1110501, Thermo Fisher Scientific, Waltham, MA, United States). Ten micrograms of a non-linearized donor vector, 5 μg of AAVS1 1l TALEN (Cat.#. 35431, Addgene) and 5 μg of AAVS1 1R TALEN (Cat.#. 35432, Addgene) were transduced using Nucleofector Ⅱ/Program B-016 (Cat.#. AAD-1001N, Amaxa Biosystems, Nordrhein-Westfalen, Germany) and Amaxa Human Stem Cell Nucleofector Kit 1 (Cat.#. VAPH-5012, Lonza, Basel, Switzerland). The cells were treated with 50 μg/ml G-418 (Cat.#. 4727878001, Roche Applied Science, Penzberg. Germany) for 10 days. EGFP-positive colonies were picked up and expanded on iMatrix-511-coated plates using StemFit AK02N. The established clones were genotyped by PCR. The PCR primers used in this study are listed in Table 1.

¹⁸F-TFB was synthesized in-house as previously described (16). The radiochemical purity of ¹⁸F-TFB was consistently > 98%, with activity > 1.8 GBq/ml. ¹²⁵I was supplied as Na¹²⁵I (Cat.#. NEZ033l; PerkinElmer, Inc., Boston, MA, United States). The radiochemical purity of the radiolabeled compounds was greater than 99%.

¹⁸F-TFB and ¹²⁵I uptake study using human iPSCs were performed as previously described (17) with minor modifications. In brief, human iPSCs were seeded on iMatrix-511-coated 24-well plates at 1 × 10⁵ cells/well and maintained in StemFit AK02N for 3 days. The cells were then washed twice with Hanks' balanced salt solution (HBSS, Cat.#. 14175079, Thermo Fisher Scientific), and incubated in 450 µl of HBSS with or without the NIS inhibitor NaClO₄ (200 µM). 2.0 MBq/ml ¹⁸F-TFB and 0.8 MBq/ml ¹²⁵I in 50 µl of HBSS were added to each well (n = 3), followed by incubation for 60 min (37°C, 5% CO₂). The incubation buffer was removed, and ¹⁸F-TFB and ¹²⁵I uptake into the cells was immediately terminated by adding 1 ml of ice-cold HBSS.

Human iPSC-CMs were seeded on iMatrix-511-coated 24-well plates at 3 × 10⁵ cells/well and maintained in Dulbecco's modified Eagle's medium (DMEM) with low glucose (Cat.#. 11885092, Thermo Fisher Scientific) supplemented with 5% fetal bovine serum (FBS) (Cat.#. F7524, Merck, Kenilworth, NJ, United States) for 3 days. Cells were washed twice with HBSS and incubated in 450 µl of DMEM with low glucose supplemented with 5% FBS with or without NaClO₄ (200 µM). 0.8 MBq/ml ¹²⁵I in 50 µl of HBSS was added to each well (n = 5), followed by incubation for 60 min (37°C, 5% CO₂). The incubation buffer was removed and ¹²⁵I uptake into the cells was immediately terminated by adding 1 ml of ice-cold HBSS.

The cells were washed twice with 1 ml of ice-cold HBSS and lysed with NaOH (0.5 N, 0.5 ml) for 60 min for protein and radioactivity measurements. The protein content of each cell lysate was measured using Pierce BCA Protein Assay Kit (Cat.#. 23225, Thermo Fisher Scientific) and a filter-based microplate photometer (Multiskan FC, Thermo Fisher Scientific). The radioactivity of each lysate was measured using a γ-counter (AccuFLEX γ7001, Nippon RayTech Co., Tokyo, Japan). Following decay correction of the cell lysate, the percentage of cell uptake radioactivity in each lysate was calculated as follows:(UptakeIndex)=100[(Radioactivityincelllysate(cpm))/proteinvolumeincelllysate(mg))]Radioactivityinaddedradiotracer(cpm)

Animal experiments followed the Animal Research: Reporting of in vivo Experiments (ARRIVE) guidelines. All mice and rats were bred under appropriate care at Department of Animal Resources, Advanced Science Research Center, Okayama University. Isoflurane (Isoflurane Inhalation Solution, Pfizer Japan, Tokyo, Japan) was used for inhalation anesthesia. Carbon dioxide inhalation was used for euthanasia. The humane endpoint for discontinuation of the experiment was defined as a decrease in body weight of more than 20%/3 days, a decrease in water and food intake, or a decrease in locomotion, when the animal's distress was deemed intolerable. No corresponding signs were observed in this study. No randomization, no strategy used to minimize potential confounders, or no blinding were done in this study.

To evaluate teratoma formation capacity, 1 × 10⁷ NIS-expressing or NIS non-expressing human iPSCs suspended in 50% Matrigel (Cat.#. 354277, Corning)/50% StemFit AK02N with 10 µmol/L Y-27632 were injected subcutaneously into the back skin of 8-week-old male NSG mice (NOD.Cg-Prkdc^scidIl2rg^tm1Wjl/SzJ, Jackson Laboratory, Yokohama, Japan). Cell injection was performed after an acclimation period of at least one week. Six mice were used in this study. Six weeks later, teratomas were harvested after euthanasia by carbon dioxide inhalation, fixed in 10% formalin for 48 h, embedded in paraffin, and sectioned for tissue evaluation using hematoxylin and eosin staining.

To detect teratomas by SPECT/CT imaging, 1 × 10⁷ human iPSCs suspended in 50% Matrigel/50% StemFit AK02N with 10 µmol/L Y-27632 were injected subcutaneously into the back skin of 8-week-old male severe combined immunodeficiency (SCID) rats (F344-II2rg/Rag2^em1lexas, NBRP-Rat with support in part by the National BioResource Project of the MEXT, Japan). Cell injection was performed after an acclimation period of at least one week. Two rats were used in this study. Six weeks later, ^99mTcO₄⁻ SPECT/CT was performed, and teratomas were harvested and evaluated using hematoxylin and eosin staining and immunostaining.

Human iPSCs were seeded at 1 × 10⁴/cm² on iMatrix-511-coated 24-well or 96-well plates and cultured in Essential 8 Flex Medium (Cat.#. A2858501, Thermo Fisher Scientific) or RPMI 1640 (Cat.#. 11875093, Thermo Fisher Scientific) supplemented with B-27 supplement (Cat.#. 17504044, Thermo Fisher Scientific) and sodium iodide (Cat.#. 194-02272, Fujifilm Wako Pure Chemical Corporation, Osaka, Japan) for 4 days. The cell area fraction relative to the plate surface area was measured using F-actin/nuclear staining to evaluate cell numbers. To measure the cell area fraction using ImageJ (18), the image was converted to 8-bit, “Threshold” was manually set to (20, 255), “Black background” was checked, “Convert to Mask” was clicked, and “Area fraction” was measured. In addition, MTT assay (Cat.#. M009, Dojindo Laboratories, Kumamoto, Japan) was performed. Absorbance was measured using FlexStation 3 (Molecular Devices, Sunnyvale, CA, United States).

Cardiac differentiation was performed according to the protocol reported by Lian et al. (19). Human iPSCs were dissociated into single cells using StemPro Accutase Cell Dissociation Reagent and seeded on 0.5 µg/cm² iMatrix-511 silk-coated 12-well plates (Cat#. 3513, Corning) at 2.5 × 10⁴/cm² in StemFit AK02N supplemented with 10 μmol/L Y-27632. The medium was replaced daily with a medium without Y-27632. Four days after plating, on day 0 of differentiation, cells were treated with 12 μmol/L CHIR99021 (Cat#. S2924, Selleck chemicals, Houston, TX, United States) in RPMI 1640 (Cat#. 11875093, Thermo Fisher Scientific) supplemented with B-27 supplement, minus insulin (Cat#. A1895601, Thermo Fisher Scientific). After 24 h, on day 1, the medium was changed to RPMI 1640 supplemented with B-27 supplement, minus insulin and 100 μmol/L ascorbic acid 2-glucoside (Cat#. AG121, Hayashibara, Okayama, Japan). On day 3, the medium was changed to RPMI 1640 supplemented with B-27 supplement, minus insulin and 5 μmol/L IWP2 (Cat#. S7085, Selleck chemicals). On day 5, the medium was changed to RPMI 1640 supplemented with B-27 supplement, minus insulin. From day 7, the medium was changed every 2–3 days. On day 12, the medium was changed to DMEM without glucose (Cat#. 11966025, Thermo Fisher Scientific) supplemented with 8 mmol/L L-(+)-Lactic acid (Cat#. L1750, Merck) to remove non-cardiomyocytes (20). After day 14, the cells were maintained in low-glucose DMEM supplemented with 5% FBS. Purified cardiomyocytes were cryopreserved with Bambanker hRM (GC Lymphotec, Tokyo, Japan) on day 17 for transplantation.

Human iPSC-CMs were seeded at 2 × 10⁴/well on iMatrix-511-coated 96-well plates and cultured with DMEM, low glucose supplemented with 5% FBS and sodium iodide for 4 days. Cell Counting Kit-8 assay (Cat.#. CK04, Dojindo Laboratories) was used to evaluate cell numbers. Absorbance was measured using a FlexStation 3 (Molecular Devices, Sunnyvale, CA, United States). Cell area fraction was measured by ImageJ (18).

After an acclimatization period of at least one week, the inguinal skin of 8-week-old male SCID rats (F344-II2rg/Rag2^em1lexa) was incised under isoflurane (Isoflurane Inhalation Solution, Pfizer Japan, Tokyo, Japan) inhalation anesthesia. Human iPSC-CMs at 1 × 10⁷ were suspended in MatriMix (511) with 10 µmol/L Y-27632 (Cat. #. 899001, Nippi, Tokyo, Japan) and injected into the inguinal subcutaneous adipose tissue. ^99mTcO₄⁻ SPECT/CT was performed 1, 6, and 10 weeks after cell injection.

After an acclimatization period of at least one week, 10-week-old male SCID rats were intubated and ventilated, and the left fourth intercostal space was opened under isoflurane inhalation anesthesia. A 7 mm-diameter aluminum rod cooled in liquid nitrogen for 1 min was placed once on the left ventricle to create cryoinjury, and 1 × 10⁷ human iPSC-CMs suspended in MatriMix (511) with 10 µmol/L Y-27632 (total volume 100 µl) were then injected into three locations around the injury. ^99mTcO₄⁻ SPECT/CT was performed 12 and 16 weeks after cell injection.

SPECT/CT scans were performed using a dedicated small-animal SPECT/CT system (NanoSPECT/CT, Mediso Medical Imaging Systems, Budapest, Hungary). General anesthesia was induced by using 5% isoflurane and maintained with 2% isoflurane during the experiment. The rats were injected 140–150 MBq of ^99mTcO₄⁻ via tail vein. The acquisition started at 30 min post-injection with a total scan time of 40 min. Following the SPECT imaging session, CT was conducted for 10 min to confirm the precise position of the injected sites. Fusion images of CT and SPECT were created, and the obtained images were analyzed using the public domain tool such as AMIDE imaging software (A Medical Imaging Data Examiner) (21). ^99mTcO₄⁻ uptake was calculated by multiplying the volume of regions of interest with increased ^99mTcO₄⁻ uptake (mm³) and its mean uptake value and normalizing with both the injected dose (MBq) and body weight (g) of the rat. The uptake of radioactive tracer was quantitatively evaluated in four rats for which SPECT/CT could be performed over time. Eight rats (inguinal injection, 1; heart injection, 7) that died before SPECT/CT imaging, 5 rats (inguinal injection, 1; heart injection, 4) that died during SPECT/CT imaging, and 1 rat (inguinal injection) that died after the first SPECT/CT imaging were excluded from the analysis.

The cells were plated on iMatrix-511-coated plates and fixed in 4% paraformaldehyde (Cat.#. 09154-85, Nacalai, Kyoto, Japan) or 100% cold methanol (Cat.#. 137-01823, Fujifilm Wako Pure Chemical Corporation) for 15 min. The cells were permeabilized with 0.1% Triton X-100 (Cat.#. T8787, Sigma-Aldrich, St. Louis, MO, United States) in PBS and blocked with 10% goat serum (Cat.#. 16210064, Thermo Fisher Scientific). Primary and secondary antibodies were diluted in 0.1% Triton X-100 in PBS containing 5% goat serum. The cells were stained with Hoechst 33,342 (Cat.#. H3570, Thermo Fisher Scientific) at 1 µg/ml, rabbit anti-OCT4 IgG (Cat.#. sc-9081, Santa Cruz Biotechnology, Dallas, TX, United States) diluted at 1:500, mouse anti-SSEA-4 IgG (MAB4304, Sigma-Aldrich) diluted at 1:200, mouse anti-TRA-1-60 IgM (MAB4360, Sigma-Aldrich) diluted at 1:200, mouse anti-TRA-1-81 IgM (MAB4381, Sigma-Aldrich) diluted at 1:200, mouse anti-cardiac troponin T monoclonal IgG1 (Cat.#. GTX28295, GeneTex, Irvine, CA, United States) diluted at 1:10,000, mouse anti-α-actinin monoclonal IgG1 (Cat.#. A7811, Sigma-Aldrich) diluted at 1:1,000, rabbit anti-MLC2v monoclonal Ig (Cat.#. ab92721, Abcam, Cambridge, United Kingdom) diluted at 1:500, rabbit anti-Connexin 43 IgG (Cat.#. 83649S, Cell Signaling Technology, Danvers, MA, United States) diluted at 1:1,000, rabbit anti-NKX2-5 monoclonal IgG (Cat.#. 8972S, Cell Signaling Technology) diluted at 1:2,500, mouse anti-NIS monoclonal IgG₁ (Cat.#. MAB3564, Sigma Aldrich) diluted at 1:2,000, mouse anti-Ki67 IgG₁ (Cat.#. 550609, BD Pharmingen, Franklin Lakes, NJ, United States) diluted at 1:1,000, and cardiac troponin I (Cat.#. sc-15368, Santa Cruz Biotechnology) diluted at 1:500. The cells were incubated with a secondary antibody, goat anti-mouse or rabbit polyclonal IgG conjugated with Alexa Fluor 488, 555, 568 or 647 (Thermo Fisher Scientific) diluted at 1:2,000, for 1 h at room temperature. The cells were observed with IX71 (Olympus, Tokyo, Japan) or BZ-X700 (Keyence, Osaka, Japan). Images were processed using ImageJ (22).

The tissues harvested after SPECT/CT were fixed with 10% formalin (Cat.#. 061-00416, Fujifilm Wako Pure Chemical Corporation) for 48 h, embedded in paraffin, and sectioned into 5-µm- thick slices using Microm HM400R Sliding Microtome (Midwest Lab Equipment, Jupiter, FL, United states). Sections were deparaffinized, immersed in citrate buffer at pH 6 or Tris-EDTA buffer at pH 9, and autoclaved at 105°C for 20 min for antigen retrieval. Mouse anti-cardiac troponin T monoclonal IgG₁ (Cat.#. GTX28295, GeneTex) diluted at 1:4,000, rabbit anti-MLC2v monoclonal IgG (Cat.#. ab92721, Abcam) diluted at 1:2,000, mouse anti-NIS monoclonal IgG₁ (Cat.#. MAB3564, Merck) diluted at 1:2,500, and rabbit anti-Ku80 monoclonal IgG (Cat.#. 2180S, Cell Signaling Technology) diluted at 1:800 were used for staining.

All data are expressed as means ± standard deviation (SD). Statistical analysis was performed by Student's t-test using SPSS Statistics 24 (IBM, Armonk, NY, United States). Statistical significance was set at p < 0.05.

To prevent transgene silencing during iPSC differentiation, the SLC5A5 gene encoding NIS was transduced into the AAVS1 locus known as a safe harbor (22–24) using TALEN (Figure 1A). EGFP and SLC5A5 were transduced together to determine the presence or absence of NIS expression using fluorescence microscopy. Gene knock-in to the AAVS1 locus was confirmed using PCR, and human iPSC lines with transgenes in both alleles were used in subsequent experiments: NIS-expressing human iPSC lines, NIS (+) #4, #6 and #8; NIS non-expressing human iPSC lines, NIS (−) #1 and #2 (Figures 1B,C). Fluorescence microscopy confirmed EGFP expression in the transgenic human iPSC lines (Figure 1D).

Undifferentiated markers of NIS-expressing human iPSCs and NIS non-expressing human iPSCs were confirmed by immunostaining. NIS protein expression was confirmed by immunostaining, and the signal was detected only in NIS-expressing human iPSCs (Figure 2A). In vitro radioactive tracer uptake experiments were performed to reveal the functionality of the expressed NIS. NIS-expressing human iPSCs acquired the ability to take up ¹²⁵I and ¹⁸F-TFB. This uptake was completely inhibited by NaClO₄, an inhibitor of NIS, confirming that the uptake of these radioactive tracers was due to the transduced NIS (Figures 2B,C).

NIS-expressing human iPSCs and NIS non-expressing human iPSCs were transplanted subcutaneously into the back of immunodeficient mice and observed 6 weeks later. NIS non-expressing human iPSC-injected mice showed grossly visible mass formation. In contrast, mice injected with NIS-expressing human iPSCs showed no obvious mass formation on the external surface (Figure 3A). When the skin at the injection site was cut open, a small mass was observed in NIS-expressing human iPSC-injected mice, and histological analysis confirmed that it was a teratoma (Figure 3B).

Next, NIS-expressing human iPSCs were injected into the right side of the back and NIS non-expressing human iPSCs into the left side of the back of immunodeficient rats, respectively, and ^99mTcO₄⁻ SPECT/CT was performed 6 weeks later. As with transplantation into mice, NIS-expressing human iPSC-derived teratomas were smaller than NIS non-expressing human iPSC-derived teratomas, but uptake of radioactive tracer was observed only on the right side of the back (Figure 3C). Histological analysis confirmed that NIS-expressing human cells formed teratomas (Figure 3D).

The reason why overexpression of NIS reduces the size of teratomas was investigated in vitro. NIS-expressing human iPSCs grew well in Essential 8 Flex Medium, the maintenance iPSC medium. However, addition of 30 µmol/L sodium iodide (NaI), the substrate of NIS, for 4 days significantly suppressed cell proliferation (Figures 4A,B). This concentration of iodine is approximately 30 times the physiological blood concentration and thus cannot explain the phenomenon seen in teratomas. In contrast, when iPSCs were cultured for 4 days with RPMI 1640/B-27 supplement medium, a commonly used medium for differentiation, NIS-expressing iPSC-derived differentiated cells showed a marked decrease in proliferation. In addition, the addition of NaI rescued the inhibition of their cell proliferation (Figures 4C–E).

Cardiomyocytes were induced from NIS-expressing human iPSCs and NIS non-expressing human iPSCs. After purification with no glucose DMEM supplemented with lactic acid, most cells expressed NIS (Figure 5A), working cardiomyocyte markers (NKX2-5 and Connexin 43), pan-cardiomyocyte markers (cardiac Troponin T, and α-actinin), and ventricular cardiomyocyte marker (MLC2v) (Figures 5B–D). NIS-expressing human iPSC-CMs were capable of ¹²⁵I uptake, which was inhibited by the NIS inhibitor NaClO₄ (Figure 5E). The effect of iodine on cell proliferation was examined. The addition of 0.3 and 3 µmol/L NaI for 10 days significantly increased the number of NIS-expressing human iPSC-CMs compared to NIS non-expressing human iPSC-CMs (Figure 5F). In contrast, no effect of NaI on cell number was observed in NIS non-expressing human iPSC-CMs (Supplementary Figures S1A,B).

NIS-expressing human iPSC-CMs were injected into the right inguinal subcutaneous adipose tissue and NIS non-expressing iPSC-CMs into the left inguinal subcutaneous adipose tissue of an immunodeficient rat, and ^99mTcO₄⁻ SPECT/CT was performed one week later. Radioactive tracer uptake was detected only in the right inguinal subcutaneous adipose tissue (Figure 6A). Histologically, human cardiomyocytes were detected in the subcutaneous adipose tissue of the bilateral inguinal regions (Figure 6B).

NIS-expressing human iPSC-CMs were injected into the right inguinal subcutaneous adipose tissue of immunodeficient rats, and ^99mTcO₄⁻ SPECT/CT was performed 6 and 10 weeks after transplantation, showing uptake of radioactive tracer at the transplant site over time (Figures 7A,B). NIS-expressing human iPSC-CMs were injected into the heart, and ^99mTcO₄⁻ SPECT/CT showed uptake of radioactive tracer in the heart over time (Figure 7C). The ^99mTc signal was measured and the uptake value was assessed over time (Figure 7D). The same number of cardiomyocytes were administered, but the variation in ^99mTcO₄⁻ uptake is likely to be large. In the histological evaluation, almost all of the transplanted cells were NIS-positive and cardiac troponin T-positive. Therefore, this variation appears to be more a matter of dosing technique than a possible reduction in uptake due to decreased NIS expression (Figures 7E,F).