Introduction
Radioactive iodine therapy (RIT) in benign and malignant thyroid diseases is very important thanks to the birth of nuclear medicine [1]. But can it play a role in the future, for example, in the treatment of brain tumours like glioblastoma multiforme (GM)? It should be noted that the sodium iodide symporter (NIS) plays the main role in the latest RIT. This underestimated NIS is nothing more than a protein responsible for the active transport of iodine to the thyroid cell. The NIS protein was discovered in the rat thyroid, and the sequence of the human NIS gene was determined in 1996 [2]. Spitzweg et al. [3] discovered the presence of NIS in extrathyroid tissues. Interestingly, this protein, as well as iodine, is responsible for the transport of many other ions, such as ClO3–, SCN–, SeCN–, NO3–, Br–, BF4–, IO 4–, BrO 4–, SO 4–, F–, HPO4–2, and ReO4– [4].
The function of NIS as a protein located in the membrane of thyroid follicular cells
NIS is a cell membrane glycoprotein located in the basolateral membrane of the thyroid follicular cells, mediating the active transport of iodide to the thyroid, which is an important condition for the biosynthesis of thyroid hormones [5, 6]. NIS-mediated iodide transport can be inhibited by competitive inhibitors, thiocyanate and perchlorate, as well as by the Na+Ka+-ATPase inhibitor ouabain [7]. The functional expression of NIS is the basis for the diagnostic and therapeutic use of radioactive iodide, which has been widely used in the treatment of differentiated thyroid cancer for 80 years [8]. The cytoreductive effect of NIS-mediated targeted radioisotope therapy is related to the so-called “cross-effect”, i.e. the impact through particle decay of the build-up of radioisotope radiation in NIS-expressing cells on neighbouring cells that do not express NIS [9]. In addition to radioiodide, alternative radionuclides, such as the beta emitter 188Re or alpha emitter 211At, are transported by NIS [10–13]. By successfully cloning the NIS cDNA in 1996, Dai et al. [2] produced a new and well-proven dual function tool to enable image-guided selective NIS gene transfer to non-thyroid tumours. Next, therapeutically effective radionuclides were shown to be successful in restoring radioiodide accumulation both in vitro and in vivo after stable ex vivo transfection of rat thyroid cells (FRTL-Tc) transformed with rat NIS cDNA [2, 9, 14].
The role of the NIS protein in the thyroid gland
The NIS polypeptide chain is arranged in a spiral through the cell membrane. There are 13 transmembrane segments and 14 extramembrane segments, including terminal fragments with the NH2 end located extracellularly and the COOH end located intracellularly. There are 3 potential glycosylation sites for the NIS protein: one is in the seventh extramembrane domain and the other 2 are in the 13th extramembrane domain. It has been proven that the level of glycosylation of the NIS protein does not affect its activity, stability, or location in the cell membrane [14]. The NIS protein is found in about 20–30% of normal thyroid follicular cells [7, 15–17] and shows the diversity of its existence both within the thyroid follicles and within the thyrocytes of a given follicle [7, 15]. There is greater symporter expression observed in the small and medium-sized follicles made of cylindrical and cubic cells, considered hormonally active, than in the flat thyrocytes that form large follicles [15–17]. The expression of the NIS protein in the thyroid tissue is subject to complex regulation and depends on many factors, including transcription factors, enhancers in the NIS gene, oncogenes, cytokines, growth factors, hormones, antibodies, and some drugs [18].
Thyroid-stimulating hormone in NIS regulation
The most important regulator of NIS protein activity in thyroid cells is thyroid stimulating hormone (TSH), whose effect on the sodium-iodine symporter is cAMP-dependent [19]. Thyrotropin stimulates the expression of thyroid transcription factors (i.e. transcription factor 1 and 2 [TTF1, TTF2] and PAX-8), which in turn stimulate expression of the following genes: thyroglobulin, thyroperoxidase, TSH receptor, and sodium-iodine symporter [18, 20, 21]. Thyrotropic hormone not only increases the biosynthesis of the sodium-iodine symporter but is also an essential factor in making sure the NIS protein is correctly localised in the cell membrane [22]. It has been noted that in the absence of TSH, the half-life of NIS protein degradation shortens from 5 to 3 days, with the amount of NIS protein decreasing faster in the cell membrane than in the cytoplasm. This suggests that when the influence of TSH is removed, the NIS protein moves from the cell membrane to intracellular compartments and is unable to transport iodine [22]. TSH in the cAMP-dependent mechanism probably affects the post-transcriptional phosphorylation of the NIS protein [23], which may play a role in how the symporter is distributed intracellularly [22]. Phosphorylation is a known cellular mechanism that modulates the activity, intracellular localisation, and/or degradation of proteins [11]. It may occur in the absence of TSH, but then the phosphorylation sites in the NIS protein are different to those when in the presence of TSH [22].
Iodine, thyroglobulin, and NIS symporter
Iodine and thyroglobulin, which show an autoregulatory effect on iodine uptake in the thyroid, reduce the level of expression of the NIS protein [24–28]. The effect of externally administered iodine depends on the dose. An inadequate supply of iodine increases the sensitivity of thyrocytes to the stimulating effect of TSH, which increases iodine uptake. However, as the amount of iodine administered increases, the process of its organisation is inhibited, which causes the acute Wolff-Chaikoff effect [29]. This effect is short-term, and as the thyroid gland adapts to the continued influx of iodine, thyroid hormone synthesis returns almost to normal, which is called an escape from the acute Wolff-Chaikoff effect [30]. As a result of studies in which iodine administration suppressed the thyroid mRNA concentration of the NIS gene [24] and/or decreased the concentration of the NIS protein [31, 32], it has been suggested this escape from the acute Wolff-Chaikoff effect is caused by a decrease in the degree of expression and NIS protein activity. This reduces the intrathyroid iodine concentration below the critical limit, which allows the maintenance of iodine organisation [32, 33].
The role of cytokines in NIS
Cytokines such as interleukin 1 (IL-1), tumour necrosis factor alpha (TNF-a), and interferon gamma (INF-g) inhibit NIS gene expression and the uptake of iodine in the rat thyroid cell line (FRTL-5, Fischer rat thyroid line-5) as well as in human thyrocytes [34, 35]. Reducing the NIS gene expression, these cytokines, which are produced in inflammatory infiltrates, may play a role in the development of autoimmune hypothyroidism [34]. In addition, by suppressing the expression of the symporter, cytokines may cause attenuation of the effect caused by antibodies stimulating the TSH receptor in Graves’ disease [36]. This could explain, at least partially, the lack of correlation between the concentration of antibodies stimulating the TSH receptor and the clinical severity of hyperthyroidism in this disease [36]. Antithyroid drugs such as methimazole or propylthiouracil are mentioned among the external factors that reduce the expression of the NIS protein [31].
Exogenous factors affecting NIS (drugs and RITs)
Among the external factors that reduce the expression of the NIS protein, antithyroid drugs such as methimazole or propylthiouracil are mentioned [31]. The clinical effect of the negative effect of thyrostatics on the NIS protein is the lower effectiveness of radioactive iodine treatment in patients who previously received thyrostatics [36–40].
The factors influencing the expression of the NIS protein also include those whose activity has been proven only in the studies of the FRTL-5 cell line, and they include, among the inhibitors — cytokine transforming growth factor beta 1 (TGF-b1) and IL-6, ceramides and sphingomyelinase, triiodothyronine, dexamethasone, and oestradiol, and among the inducers — adenosine [31, 33, 41, 42].
Application of NIS in glioblastoma multiforme imaging and therapy
The role of NIS imaging in glioblastoma multiforme
Glioblastoma (GBM) is the most common primary brain tumour with poor prognosis, and it is mainly treated with palliative therapy. Several mechanisms are used by this highly complex tumour to evade treatment, which means new GBM treatment strategies are urgently required [43]. The blood-brain barrier (BBB), which can block radiotracers and gene vectors, is one reason for the limitations in effective treatment and detection of GBM. Because NIS-mediated radionuclide imaging and therapy do not require complicated radiolabelling procedures, the small-size radionuclides used are able to penetrate the BBB and diffuse into the tumour [44].
Several preclinical studies have demonstrated the potential use of NIS in GBM imaging and therapy. A study by Cho et al. [45] used a rat model with F98 intracerebral gliomas that were retrovirally transduced with human NIS. The authors demonstrated the possibility of non-invasive imaging of GBM using by [99mTc]pertechnetate- and [123I]NaI-scintigraphy, and then prolonging the survival time of rats after 131I therapy. Guo et al. [44] published imaging and therapy experiments with 188Re or 131I in mice with tumour xenografts injected with the U87 human glioblastoma cell line that had been transfected with a recombinant lentiviral vector containing human NIS in the right armpit. In vivo imaging results, assessed by gamma camera imaging, showed the 188Re/131I accumulated in the NIS-containing tumours, and effective reduction of the tumour volume was achieved in the 188Re or 131I treated mice compared to untreated control mice. In another study, using one of the most extensively studied oncolytic viruses for NIS gene transfer, Opyrchal et al. [46] demonstrated the effectiveness of the [123I]NaI or [99mTc]pertechnetate gamma camera or micro-single-photon emission tomography/computed tomography (microSPECT/CT) s.c. imaging and orthotopic murine glioblastoma xenografts after intratumoural infection with measles virus encoding NIS (MV-NIS) to induce NIS expression in brain tumour tissue. Combined radiovirotherapy with MV-NIS and 131I resulted in improved antitumour activity and survival compared to viral therapy alone in both cases of GBM.
The advantage of PET, as opposed to scintigraphy or SPECT, is the potentially more accurate detection of small-volume GBM lesions with a relatively low level of NIS expression when systemic gene transfer methods are used [47]. Preclinical imaging studies with [18F]TFB as a PET tracer were performed in athymic mice with C6 human glioma expressing NIS s.c. tumour xenografts producing adventurous tumour uptake of [18F]TFB via NIS [48]. Recently, Kitzberger et al. [49] used the U87 human glioblastoma cell line stably transfected with a plasmid expressing NIS (CMV-NIS-pcDNA3) (U87-NIS) to track tumour expression of NIS s.c. and orthotopic brain tumours. Using this, they made a direct comparison of 124I and [18F]TFB as radiotracers for PET imaging of small animals. The localisation of the NIS protein on the cell membrane of the U87-NIS cells and its active transport of iodide was confirmed by immunocytochemistry and [125I]iodide uptake assays in vitro. Very interestingly, these U87 tumours revealed endogenous NIS-mediated 123I uptake, in addition to the thyroid gland, which was observed in other organs such as the stomach, salivary glands, and in the bladder due to renal excretion [50]. In our opinion, it would be interesting to check the endogenous NIS-mediated 123I uptake when assessing GM tumours.
In a second group of mice, NIS-based radionuclide biodistribution was investigated using 3D preclinical PET scanners after an intravenous injection of [124I]NaI or [18F]TFB. The results showed a high accumulation of NIS PET markers in the U87-NIS tumours [51]. In these animals, the physiological signal of organs endogenously expressing NIS (thyroid, mammary glands, salivary glands, stomach), and tumour uptake was effectively blocked [50]. Kitzberger et al. [49] showed the possibility of PET imaging by monitoring the expression of the NIS gene in these brain tumours. They conducted the study based on a mouse, an appropriate orthotopic model. Nude mice with U87-NIS orthotopic brain tumours received [124I]NaI or [18F]TFB for PET imaging. Both radiopharmaceuticals caused NIS-mediated accumulation of radionuclides in the brain tumours, which was comparable to [124I]NaI- and [18F]TFB-PET [50].
The NIS gene therapy concept for glioblastoma multiforme
Future prospects: non-viral systemic NIS gene delivery to glioblastoma multiforme
The potential of NIS as a theranostic gene and the improvement of new gene delivery systems have broadened the possibilities of using the NIS gene therapy concept for extrathyroidal tumours [9, 51]. Based on the gene therapy approaches summarised above, the preclinical development of the NIS gene therapy approach will be extended to other aggressive non-thyroid cancers in the future, such as GBM, with the primary objective being a phase I/II clinical trial.
The efficacy of non-viral systemic NIS gene delivery systems based on mesenchymal stem cells or synthetic polyplexes to target glioma and using advanced [124I]NaI and [18F]TFB in animals — PET imaging has been reported [52, 53]. Spellerberg et al. [53] addressed tumours of mice that received non-targeted mono DBCO-PEG24/NIS polyplexes, which exhibited an uptake of 1.96 ± 0.52% ID/mL. Vadysirisack et al. [54] studied doxycycline-induced NIS expression that was established in cell lines of various tissue types, including human (FTC133 thyroid cancer cells), HeLa human (cervical cancer cells), and rat (PC12 pheochromocytoma cells).
There is a double benefit to NIS gene transfer: it is therapeutic and enables imaging of transgenic protein expression [45]. Until now, evaluating the expression of a transgenic protein required an invasive biopsy or even the death of the animal undergoing gene therapy. Meanwhile, using NIS gene transfer may enable non-invasive and reproducible visualisation of vector expression, which may be an important tool in preclinical and clinical gene therapy trials [45, 54]. The use of the NIS gene as a vector transfer monitoring gene (imaging reporter gene, Imagene) makes it possible to assess both the location and concentration and duration of expression of the transgenic protein [45, 54]. Gene therapy using the NIS gene, however, requires the resolution of several issues, including whether the post-translational processes, including the distribution of the NIS protein to the cell membrane, affect iodine uptake in cells expressing the exogenous symporter [54]. As many studies have shown, the cellular distribution of the symporter is a disorder-prone process and may, therefore, be a limiting factor for iodine uptake by the exogenous NIS protein. However, the study carried out in the FTC133, HeLa, and PC12 cell lines did not show distribution disorders of the transgenic NIS protein. In addition, the uptake of radioactive iodine by the exogenous NIS protein observed in many other cell types indicates that disturbances in the cellular distribution of the symporter do not play a role in the use of NIS gene transfer [54]. The problem of NIS gene therapy remains the inability of transfected cells of many tumours to retain iodine. Huang et al. [55] noted the limited effectiveness of the radioactive iodine destruction of non-small cell lung cancer cells transfected with the NIS gene due to the rapid elimination of 131I. For comparative purposes, the study used a combined transfer of the NIS gene and the thyroperoxidase gene, which enables the organification and maintenance of iodine in the cell. This increased the effectiveness of the radiotherapy. However, as shown in other studies, the organification of iodine may not be a prerequisite for successful tumour radio ablation using the NIS gene [24, 56]. RIT reduced the tumour volume by more than 75% [57]. Meanwhile, there are no grounds to claim that prostate cells have the ability to organise iodine, if only because they do not initially capture iodine. However, gene therapy using the NIS gene to treat prostate cancer is potentially possible and requires further in vivo research. One study compared the human and rat NIS proteins by transducing retroviral vectors with the genes of these proteins into various human cancer cell lines and rodent cancer cells. It has been shown that the radioactive iodine concentration of the rat NIS protein is up to 5 times higher than that of the human, in both rodent and human cell lines [58]. The difference in the functionality of both symporters may result from the different structure of the rat and human NIS protein or from different regulatory mechanisms. Whether the rat NIS protein has a greater uptake ability for radioactive iodine in vitro only or also in vivo needs to be answered [58].
In addition, this study showed the differentiation of iodine uptake by the transgenic NIS protein in individual cancer cell lines. Therefore, further research is needed to assess which cancers will respond best to RIT after NIS gene transfer. Hence, the most important task for researchers is to transfer the knowledge gained in laboratories to clinical trials.
A new hypothesis of RIT treatment in GM
Is it possible to use 131I alone without using the NIS gene? After all, the NIS symporter was detected not only in the thyroid gland, but also in different tumours. The administration of RIT is completely harmless, the only complication being hypothyroidism. However, there are contraindications and concerns regarding increased tumourigenesis in individuals undergoing 131I treatment, as well as recommendations limiting procreation. Indeed, it has been shown recently that, for example, in the case of thyroid cancer, the maximum therapy is 1000 mCi. Or is it worth using RIT in an ablative dose of 800 MBq (22 mCi) of 131I in an outpatient setting, with the possibility of repeating it? What do we have to lose? After all, GM patients live very short lives. And yet NIS gene expression was observed in GM, although only in animals.
As early as 1955, Amyes et al. [59] had already succeeded in pinpointing the location of brain tumours using radioactive iodine and phosphorus. A needle probe was used in this procedure for the first time, which proved to be very useful in quickly locating and defining the affected area.
Radioisotopes of various elements are now used with increasing frequency in nuclear medicine. This additionally proves that using RIT does not damage the skull. This is where beta rays, in addition to imaging, may be effective in RIT. It is not always possible to remove the whole GM, especially grade IV. Attention was also paid to the revolution in targeted (individual) therapy in the case of tyrosine kinase inhibitors (imatinib, sunitinib, and sorafenib) [60–64] and in the case of the latest drugs such as crizotinib, entrectinib, or larotrectinib [65, 66]. NanoTherm® therapy is also used in patients with GBM who have exhausted conventional treatment methods [67]. Recently developed individualised multimodal immunotherapy (IMI) is based on cancer vaccines [68–78] and oncolytic viruses [79, 80].
And maybe we could try using non-virology therapy in the form of RIT, even without genetic aspects – just classical therapy. As before, they turned to a completely new therapy. Just over 60 years ago, Amyes et al. [59] showed that some benign and malignant pathological disorders in the brain (e.g. inflammation, vascular disease, and, most importantly, tumours) tend to increase the rate at which certain ions pass through the so-called BBB (especially 131I).
Or maybe we should just revolutionise and try to use RIT in GM. And what is very important, this therapy can be completely free, and even if not, it does not have to be very expensive. Hypothyroidism is just a “complication”. But perhaps we are getting a new life, or perhaps patients already freed from GM.
Side effects of antithyroid drugs in non-standard therapy amiodarone-induced thyrotoxicosis
Similarly, we used and still use RIT in patients with amiodarone-induced thyrotoxicosis (AIT) with very low uptake (RAIU) [81] modelled on the publication of Hermida et al. [82] or of Gursoy et al. [83], whose RAIU was slightly elevated. But a preliminary study was the first in Poland to use RIT in euthyroid patients with a history of hyperthyroidism and permanent atrial fibrillation prior to the administration of amiodarone [84]. However, the authors of the study used very high radioiodine activities (up to 80 mCi 131I), which are not routinely used in the treatment of hyperthyroidism. Based on our experience, RIT is relatively safe and leads to hypothyroidism [81, 84].
And in the case of AIT, when antithyroid drugs (ATDs), including thionamide derivatives of thiouracyil (PTU), i.e. propylthiouracil and imidazole-thiamazole, are followed by agranulocytosis, hepatitis, or vasculitis and lupus-like syndrome [85–87], then RIT is necessary. The authors believe that, as with AIT, the application of RIT to individual therapy in GM may be a useful addition to other therapies. This therapy, in particular, can play a very important role in the case of GM relapse. There is no current evidence, only for Wistar rats [88] and mice [89], but this is a combination of genNIS genetics and RIT. However, the authors believe it will be a completely different perspective for non-GM therapy (such as AIT) or in the use of other therapies.
A new hope of RIT therapy using NIS with or without the gene other than in glioblastoma multiforme
Attempts to use the anti-cancer effect of radioactive iodine by means of gene therapy using the sodium-iodine symporter are including more and more cancers, even for organs in which the NIS protein is not physiologically detected. This increases the scientific importance and, above all, the potential clinical use of the symporter in the future. Human cancer cell lines transfected with the NIS gene that have been successfully treated with radioactive iodine in animal models include the following: prostate cancer cells [57, 58, 90], multiple myeloma [55], non-small cell lung cancer, neuroendocrine tumours [91], malignant melanoma [58], adenocarcinoma of the breast and ovary [89], cervical cancer [92], renal cancer, GBM [58, 90], and primary liver cancer [88]. There is a dual advantage to performing NIS gene transfer: it is therapeutic, and it enables imaging of transgenic protein expression. The evaluation of transgenic protein expression requires an invasive biopsy or even the death of the animal subjected to gene therapy [93]. Meanwhile, the use of NIS gene transfer may enable non-invasive and reproducible visualisation of vector expression, which may be an important tool in preclinical and clinical gene therapy trials [54]. The use of the NIS gene as an imaging reporter gene (Imagene) allows the assessment of the location and the concentration and duration of expression of the transgenic protein.
NIS, which is found in extrathyroid tissues, does not differ in primary structure — the cDNA of the NIS protein found in the parotid gland, mammary gland, and gastric mucosa has the same nucleotide sequence as the symporter gene found in the thyroid gland [8].
Gene therapy with the use of the NIS gene, however, requires the resolution of several issues, such as whether post-translational processes, including the distribution of the NIS protein to the cell membrane, affect iodine uptake in cells expressing the exogenous symporter [54]. As many studies have shown, the cellular distribution of the symporter is a disorder-prone process and may, therefore, be a limiting factor for iodine uptake by the exogenous NIS protein. This highlights the role of RIT in a highly significant way, and we must especially consider patients with hyperthyroidism. However, in our opinion, it is worth trying RIT even in patients with euthyroidism [81, 84]. But what do we have to lose with a very heavy GM therapy?
Discussion
Is it worth delivering the NIS gene to non-thyroid tumours? Kitzberger et al. [50] introduced an essential, future-proof therapy in the clinical translation of NIS gene therapy (non-virology) for extrathyroidal tumours. This therapy is an effective and safe development of gene delivery vehicles that enable sufficient and tumour-selective NIS expression levels. These authors have additionally proven that this occurs best when there is systemic application of the vector. In addition to options for monitoring and targeting primary tumours, some of these approaches provide metastasis treatment options by enhanced targeted NIS transgene delivery. Synthetic polyplexes and mesenchymal stem cells can deliver anti-cancer therapies after systemic administration through different targeting strategies. Both systems are promising platforms with the potential for clinical success [49]. As previously shown, NIS has also been demonstrated in other organs [8, 17, 94–101].
The sodium-iodine symporter, which is found in extrathyroid tissues, does not differ in primary structure — the cDNA of the NIS protein found in the parotid gland, mammary gland, and gastric mucosa has the same nucleotide sequence as the symporter gene located in the thyroid gland [8]. In extrathyroid tissues, expression of the NIS protein is regulated differently and is generally weaker than in the thyroid gland [49]. Therefore, the question should be asked whether it is worth treating these other tumours (and not just thyroid glands) with RIT alone, but in significantly increased doses. And what is the risk, since GM patients live very short lives, and could this be a “revolution”? The only “complication” is hypothyroidism. Or maybe it is worth trying, especially since the patient does not have anything to risk anymore. The still unresolved question is whether the use of RIT also fulfils its role in the treatment of breast cancer, lung cancer, or glioblastoma multiforme. Although, we have not studied the expression of this protein, and there is still no conclusive evidence for the effect of RIT therapy alone in the case of recurrent glioblastoma, maybe it is worth the risk. Because the most important thing is that life is worth living, and can RIT help this, and perhaps contribute to healing?
Authors’ contributions
Conception: A.Cz. Design: A.Cz., P.G., M.B., M.R. and A.F. Writing manuscript: A.Cz., M.B., P.C., N.S.G., B.K.-K. Writing — review and editing: A.Cz., P.G., M.B., P.C., N.S.G., B.K.-K., A.F. The manuscript was drafted by A.Cz., P.G., B.K.-K., M.R., and A.F. and edited by all authors. All authors have carefully read and improved the manuscript.
Conflict of interest
The authors have no conflicts of interest to declare.
Funding
This manuscript has received no funding.