**4. Discussion**

In this study, we focused on the determination of USPION-originated iron content and its distinction from the naturally present iron in the tissues. For that purpose, we developed the method for determination of the presence of iron originated from USPIONs in the tissues of rats after i.v. administration of a low dose of NPs. We showed that the SQUID magnetometry is able to determine and to distinguish USPION-originated iron even if it is in low amount and not clearly detectable by Perl's histochemical method.

We have dealt with the application of USPIONs, and we investigated their effects on various organs of the WKY rats. As bare iron oxide nanoparticles administered in higher doses were previously shown to be toxic as they produce the development of oxidative stress [9,11], we used a low dose of PEG-coated magnetite USPIONs. PEG is a neutral, hydrophilic and biocompatible polymer which improves the dispersion of NPs in water, it improves their bio-distribution and increases blood circulation time [30]. However, despite the PEG-coating and administration of a low dose of USPIONs, we found elevated production of the superoxide, which was increased in all tissues and organs investigated in this study, similarly as it was found using various types of NPs [31]. On the other hand, we were unable to unequivocally confirm elevated iron content in the tissues using the histochemical method. Thus, we used SQUID magnetometry for determination of the distribution and content of USPION-originated iron in the samples of the organs as well as in the blood of rats and for differentiation of USPIONs from the biogenic iron and determination of it content.

However, due to extremely weak magnetic signals (with changing polarity), it was necessary to analyse and adjust the measurement conditions in the MPMS measuring system. Special attention was paid to preparing and mounting of the samples and to the design of the sample holders. As the magnetic moment of the samples was extremely weak, close to the value of standard holders/capsules, it was necessary to design such a configuration, where the material in vicinity of the measured sample is homogeneous in the whole operating space. In such a way, it is possible to subtract the background from the total measured signal.

We found that the proper mounting of the sample into the MPMS system is extremely important. The frequently used configuration for measurements of biological samples is to use standard plastic or gelatine capsules which are inserted into the drinking-straw and fixed by a cotton stopper [19,21,23,27,32]. However, this configuration leads to several problems. Firstly, there is a need to subtract the background signal generated by the capsule and the cotton. Secondly, when the value of the magnetic moment of the sample is close to that of the capsule, but in the opposite direction, this often leads to instability of scan through gradiometer coils, the usually centred peak is shifted, and the system is not able to properly fit the measured values (as shown in Figure 3). Similar problems were observed in previous studies [33,34], but the authors ascribed them mistakenly as a sort of hardware problem. In some studies, the authors have similar issues with the proper sample mounting. The problem of sample geometry and proper mounting for inorganic samples has been discussed previously in the article of Sawicki et al. [29]. The authors recommended to glue the sample to the long silicone rod attached at the end of the measuring rod for MPMS instead of using a plastic straw. In the study of [25], freeze-drying of the samples and mounting to a straw without capsule was used. In another study [18], the authors described the mounting of cancer cell samples treated with nanoparticles to the long strip of filtration paper.

Another problem which we have to solve is the attachment of the small samples into the measurement straw. We have solved this problem by the development of the cylindrical shaped instrument to obtain repeatedly the samples of the same shape, because we observed differences in the results if the shapes of the samples were different. Our method with the cutting of the biological samples of the same shape and their attachment during the vacuum drying worked well for samples from the muscles, kidneys or heart, which retain its shape after cutting. The rings of the aorta can also be used when the wire is embedded via the lumen of the artery. However, the samples of the liver or brain were more difficult to be prepared as a cylindrical sample due to their soft structure. In this study, we used a higher mass of the liver to ensure a suitable shape of the sample. With specific caution to the preparation of the sample (cut from the bigger mass of the "crude" sample), our method can also be used for the liver. However, we were unable to perform the measurements successfully in the brain tissues due to their soft structure. We assume that for samples of the brain, their homogenisation in proper buffer and manipulation similar to that for the liquid samples would be more suitable than the procedure for solid tissue samples.

For the liquid samples, such as blood or its derivatives, which need to be prepared differently from solid tissues, we developed a procedure which was inspired with the work of Hashimoto et al. [18]. Liquid samples (10 μL) were pipetted on the suitable type of paper. The type of paper used for the measurement of liquid samples may differ. The laboratory filtration paper can be used for the cell cultures measurement. For blood samples measurement, we used the standard white V-shaped strip of the office (80 g/m2) paper, which we found more suitable than laboratory filtration paper. When filtration paper was used, the sample was soaked into it and spread on a relatively big area, which produced inconsistent results. Thus, to keep a small area of the dried sample is essential. Such a preparing of the samples has several advantages: (a) fine localised sample; (b) the sample is time-stable; (c) sample can be stored at room temperature and (d) it is ready for repeated measurements. In addition, as the paper is longer than the base length of gradiometer, the paper has a negligible additional signal to the magnetic moment of the sample. We also assume that this procedure can be used for determination of the presence of the USPIONs in homogenates of the soft tissues that cannot be measured in the solid state.

In the conditions used in our study, we set the detection limit of SQUID magnetometry to 0.6 μg Fe/g of the sample dry weight using PEGylated NPs. Another way of iron NPs determination in rodent tissues has recently been described by Poller at al., who used magnetic particle spectroscopy [35] to analyse fate and metabolic processing of very small iron oxide NPs in atherosclerotic mice after injection of 300 μmol of iron NPs.

It is worthy to note that we observed a relatively high variability of the results during experiment, mainly in blood. This may result from individual metabolic differences between rats. We also found that the amount of USPIONs in blood and tissues can be affected by stress, which significantly alters blood circulation and can also potentially affect the stability of USPIONs [36]. Thus, our experience has shown that caution is needed when manipulating experimental subjects and ensuring the best possible stress-free conditions is needed to reduce the variability of results.

In conclusion, we have improved the SQUID magnetometric method for determination of the low content of nanoparticle-originated iron in the solid and liquid samples of the animal tissues at 300 K, which allows distinguishing the low dose of USPION-originated iron from the biogenic iron naturally present in the tissues and blood. This method is more cost-effective and less time-consuming then measurements at 2 K, which requires liquid He for cooling of the system. Furthermore, the once-prepared sample is stable, and it can be stored at room temperature and repeatedly measured. This method also allows for the determination of the USPIONs in the tissues after administration of the low doses of NPs when standard methods are unable to determine USPION-originated iron in the tissues.

**Author Contributions:** Conceptualization, M.Š. and I.B.; Data curation, M.Š.; Formal analysis, M.Š., A.D., A.M., J.M. and I.B.; Funding acquisition, J.M. and I.B.; Investigation, M.Š., A.D., M.K., A.B., P.B., A.M., A.E.-A. and J.M.; Methodology, M.Š., M.K., A.B., P.B., A.C., J.M. and I.B.; Project administration, J.M. and I.B.; Visualization, M.Š., A.B., A.E.-A. and I.B.; Writing—original draft, M.Š. and I.B.; Writing—review & editing, M.Š., A.D., A.C., J.M. and I.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Slovak Research and Development Agency under Contract No. APVV-16-0263 and by the grants VEGA 2/0160/17 and VEGA 2/164/17.

**Acknowledgments:** The authors thank R. Kurjakova, J. Petova, B. Bolgacova and R. Tanglmajer for their excellent technical assistance.

**Conflicts of Interest:** The authors declare no conflict of interest.
