Introduction

Manganese is a micronutrient that is essential for human life in only trace amounts [1, 2], but deficiency is rare [3]. Existing primarily in the Mn(IV) oxidation state in the earth’s crust [4], manganese is the third most abundant transition metal in the crust [4] and is taken up by plants, where it is utilized in green leaves as an oxidizer in photosynthesis [4]. Manganese(II) is by far the most electrochemically stable oxidation state for manganese [4] and the state that is predominantly used as an enzymatic cofactor. Manganese in the blood is found as Mn(III) bound to binding sites for Fe(III) on transferrin [5, 6] or as Mn(II) bound to other serum proteins [7], but the vast majority of manganese leaves the blood quickly and is accumulated in various organs or excreted [8]. The necessity of manganese as a nutrient is sometimes overshadowed by its neurotoxic effects that can result in manganism [2], a condition that was first described in 1837 [9]. In mangansim, early-stage patients often present with psychiatric symptoms [10, 11] that are sometimes referred to as “manganese madness” [12]. Later-stage patients develop symptoms of parkinsonism, including intellectual impairment, tremors, and rigidity similar to idiopathic Parkinson’s disease [9, 12–14], with a few key differences [12].

In addition to studying the neurotoxicity of manganese, an additional motivation for understanding the biodistribution of free manganese is that Mn(II) can be utilized in contrast agents for manganese-enhanced magnetic resonance imaging (MEMRI), which was studied early in the history of MRI by P. Lauterbur and colleagues [15] and later achieved in vivo [16]. Manganese(II) is “high-spin” from its five unpaired valence electrons, which shorten the spin-lattice time constant (T₁) for nearby ¹H nuclei, resulting in brighter signal (when using a T₁-weighted MR pulse sequence) [8, 17–19]. In fact, T₁-weighted MRI may be used as an indicator of manganese-induced parkinsonism [20, 21]. Additionally, chelated Mn(II) could serve as an alternative to Gd(III), which is currently the most commonly used cation core in T₁-shortening contrast agents for MRI, since free Gd(III) is excreted through the kidneys, but free Mn(II) is excreted almost exclusively in feces, primarily from bile from the liver [22–25].

Thus far, the only manganese-based MR contrast agent to reach the market in the United States was Teslascan® (mangafodipir, manganese(II)-dipyridoxyl diphosphate, Mn-DPDP) (Amersham plc, Amersham, Buckinghamshire, United Kingdom), a chelated form of Mn(II) that reduces cardiac toxicity, by slowing the availability of free Mn(II) [17, 26–28] due to dissociation [29]. Mn-DPDP was officially indicated for detecting liver metastases of colorectal cancer [26, 27], but it no longer on the market in the United States. Various other chelated forms of Mn(II) and Mn(III) [30] are being developed, numerous kinds of Mn-doped inorganic and organic nanoparticles are in the in vitro and pre-clinical stages [19], and oral administration of liquids containing Mn(II) has even been tested in humans [31, 32]. However, since MRI is not inherently as quantitative as PET, it is valuable to develop surrogates that are radiolabeled with a positron-emitting radioisotope of manganese [27, 28, 33, 34] for quantitative confirmation of their biodistribution.

Although imaging of large amounts of accumulated Mn(II) is possible by T₁-weighted MRI, true quantification of the biodistribution of non-radioactive manganese in vivo is difficult. However, pre-clinical imaging and ex vivo biodistribution using tracer amounts of radiomanganese can provide a route to quantify the amount of manganese as a percentage of administered dose in animal models. Past studies have examined the biodistribution of manganese in various forms administered by various routes, mostly in rodents, and have generally found that manganese is distributed to many tissues throughout the body. Several of these studies have used the radiotracers ⁵²Mn (t_1/2 = 5.6 d [35]), and ⁵⁴Mn (t_1/2 = 312 d [35]), or by administering stable manganese (⁵⁵Mn: 100% natural abundance [35]) followed by various techniques for quantifying non-radioactive metals in the tissues [17, 36–38]. More recently, researchers at the University of Wisconsin have published results for ex vivo biodistribution of IV ⁵²Mn(II) in rodents: Graves, et al. [39] have investigated biodistribution in tumor-bearing mice at 96 h post-injection (p.i.), as well as by in vivo PET at timepoints 4–96 h p.i. Additionally, Brunnquell, et al. [40] have investigated biodistribution in rats at 4 and 48 h p.i.

The half-life of ⁵²Mn is long enough for in vivo imaging at timepoints as long as days or even weeks p.i., and ⁵²Mn emits positrons at a sufficient branching ratio (I_β+ = 29.6% [35]) for in vivo medical imaging by positron emission tomography (PET). Manganese-52 provides the additional benefit of low-energy positron emission (E_β+ = 242 keV [35]) (resulting in better spatial resolution in PET). Manganese-52 can be produced efficiently by the ⁵²Cr(p,n) or ^natCr(p,x) reactions (⁵²Cr: 83.8% natural abundance) by bombarding target material with low-energy protons (6<E_p<20 MeV [41, 42]). Thus, ⁵²Mn can be produced at medical centers using protons from a low-energy cyclotron without expensive isotopically enriched target material. So far, ⁵²Mn has been used in several ex vivo and in vivo studies, including the biodistribution of Mn(II) [43], cell tracking [44], antibody imaging [39], and as a positron-emitting surrogate of a MEMRI contrast agent [34].

In this work, we utilized the radioisotope ⁵²Mn to study the biodistribution of free Mn(II) in sixteen different tissues in mice following intravenous injection or inhalation in saline solution.

Materials and methods

Results

For the biodistribution-only study, bombardment of ^natCr metal foils with ~13.4 MeV protons for a total of 27 μA·h (12 μA), which produced 41 MBq (1.1 mCi) (decay-corrected to end-of-bombardment), which was 69.6% of theoretical yield based on our published cross-section results [41]. Fig 1 and Table 1 show the results from ex vivo biodistribution of ⁵²Mn in saline administered via intravenous injection or by inhalation. Manganese-52 was cleared rapidly from the blood, as demonstrated by its rapid decrease from 1 h to 1 d p.i., and the activity in the gastrointestinal tract—stomach and intestines—also decreased rapidly between these timepoints. At 1 h p.i., the highest activity was found, in decreasing order, in the liver, kidney, lung, heart, pancreas, spleen, and salivary glands (Table 1); however, by 1 d p.i., the concentration of ⁵²Mn in all of these organs of high initial uptake had decreased substantially, except for the salivary glands, pancreas, and kidney, as shown in Fig 1. Among the other tissues, which had lower uptake at 1 h p.i., ⁵²Mn was retained to various degrees in the brain, bone, and thyroid.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Plot showing results from ex vivo biodistribution of saline solutions containing ⁵²Mn administered via intravenous injection or inhalation.

The timepoints shown represent the time after administration of the dose. Results for the p-values from Student’s t-test with Welch’s correction at 1 d p.i. only: *: p<5%; **: p<0.1%; ***: p<0.01%; -: p≥5%. Sample sizes for each timepoint were n = 4 mice for injection and n = 3 for inhalation. Uncertainties: one absolute standard deviation in units of %ID/g. Error bars were only drawn in the positive direction for visual clarity. The plotted data is presented in numerical form in Table 1.

https://doi.org/10.1371/journal.pone.0174351.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Results from ex vivo biodistribution of saline solutions containing ⁵²Mn administered via intravenous injection or inhalation.

https://doi.org/10.1371/journal.pone.0174351.t001

Comparing the biodistribution of ⁵²Mn(II) at 1 d after IV injection or inhalation, we considered two sample means to be statistically significant if the p-value from the t-test was <0.05. Applying the t-test and this cutoff value, nearly all tissues that we measured showed a statistically significant difference, with the exception of brain, thyroid, and stomach, all of which demonstrated large relative sample standard deviations (%RSD). Salivary glands and muscles demonstrated especially strong statistical significance (p<0.0001). S1 Fig (Supplementary Information) shows co-registered PET/CT images of two mice (species: C57-Black-6) from a separate but similar study in which larger doses of [⁵²Mn]MnCl₂ were administered for in vivo PET imaging. These images demonstrate a distribution of PET signal in many of the same tissues with high uptake in the current ex vivo biodistribution results, and these images suggest the potential utility of ⁵²Mn for further quantitative PET/CT studies.

Discussion

For the comparison of IV injection to inhalation of ⁵²Mn in mice, the concentration of ⁵²Mn in most tissues at both 1 h and 1 d timepoints is higher than in the inhalation results. The activity in the gastrointestinal tract was much greater in the inhalation results, suggesting that some of the dose may have been swallowed directly from the aerosol or inhaled, cleared by mucociliary clearance, and then swallowed. Also in our studies, we observed uptake in the pancreas and brain, while ⁵²Mn that entered the stomach or intestines did not appear to be retained, although the one hour amounts were much higher in the inhalation groups likely due to putative swallowing of some of the dose. Retention of ⁵²Mn in the brain, thyroid, and thymus was observed in results from both injection and inhalation. We also found significant uptake of ⁵²Mn at 1 d in the bone (2.6%ID/g) resulting from injection, but not from inhalation.

For both injection and inhalation, ⁵²Mn was cleared rapidly from the blood, and the injection data showed ⁵²Mn in the liver and kidney in <1 h. The inhalation data also suggest accumulation in the liver and kidneys, but this accumulation took much longer than one hour. Although manganese appears to accumulate in both liver and kidneys, results from the literature [22–25] suggest that very little manganese is excreted in urine. Instead, the vast majority of manganese is excreted in feces, primarily due to release of manganese from the liver into the enterohepatic circulation. In other studies, animals that were administered manganese or radiomanganese intravenously [17, 22, 55–58], high uptake was generally observed in the kidney, pancreas, liver, adrenal glands, and intestine, as well as comparatively small amounts in the brain. For intratracheal instillation, Heilig, et al. [59] and Brain, et al. [60] reported that [⁵⁴Mn]MnCl₂ was retained in the lungs to large extent, but was also found in significant amounts in the intestines, liver, kidneys, and in small amounts in the brain. These intratracheal instillation studies also showed that Mn can enter the bloodstream from the lungs, possibly by ion channels, but not by the divalent metal transporter-1 (DMT-1) in that particular tissue [59, 60]. These general trends in manganese distribution agree with studies that focused more narrowly on the in vitro or in vivo uptake of manganese by cells or tissues that included liver [25, 61], kidney [61], pancreas [61], intestines [60, 62], salivary glands [17, 63], myocardium [58], and bone [64].

Owing to its neurotoxic effects, particular interest has been paid to uptake of manganese in the brain by various routes. Kanayama, et al. [65] administered to mice by eight different routes a solution containing sixteen different radiotracers, mostly transition metals, and, for most administration routes, ⁵⁴Mn(II) had higher brain uptake than most of the other tracers. In several other studies, rodents received radiomanganese by carotid injection or continuous in situ brain perfusion brain. These studies suggest that manganese enters the brain by more than one mechanism, including carrier-mediated uptake of Mn-citrate [66], by store-operated calcium channels as Mn(II) [67], by transferrin-receptor mediated endocytosis as Mn-transferrin [68, 69], or by other unspecified mechanism(s) that are faster than simple diffusion [69, 70]. In a comparison of ⁵⁴Mn species administered by in situ brain perfusion, Mn-citrate was transported across the blood-brain barrier faster than either free Mn(II) or Mn-transferrin, suggesting that Mn-citrate might be an important route for manganese uptake in the brain [66]. Furthermore, it is likely that manganese accumulates in the brain because it effluxes at a rate that is consistent with simple diffusion [71, 72].

Since certain inhaled substances can enter the brain directly from the olfactory bulb—without entering the bloodstream—there have been a considerable number of studies examining the possibility of manganese entering the brain by such mechanisms. Characterization of this mechanism(s) would be particularly relevant to the problem of inhaled manganese in metal workers. Animal studies using radiomanganese [73, 74] or MEMRI [75, 76], suggested that Mn(II) can be transported from the olfactory bulb and into olfactory neurons, transported through those neurons, across synaptic junctions, into secondary olfactory neurons, and then into the diencephalon and cerebrum, and (in rats) into the spinal cord. Thompson, et al. [77] tested uptake in rats with defective divalent metal transporter-1 (DMT-1) proteins and confirmed that this transporter is involved in manganese uptake in the rat olfactory bulb based on higher uptake in healthy control rats. In rats, ninety minute nose-only inhalation of aerosolized [⁵⁴Mn]MnCl₂ [78] or [⁵⁴Mn]MnHPO₄ [79] solution leads to uptake in the olfactory bulb and olfactory tubercle. In both studies, they observed uptake in lungs, liver, kidney, and pancreas, and—at much lower amounts—in the striatum of the brain. In a similar study, Lewis, et al. [38] exposed mice and rats to nebulized, non-radioactive MnCl₂ for 10 nose-only doses of 6 hours each. Sample analysis by proton induced X-ray emission (PIXE) revealed elevated manganese in trigeminal ganglia, suggesting this nerve as a possible pathway for brain entry of manganese. Along with these published examples of brain uptake of Mn cations, the present study has shown uptake of ⁵²Mn(II) in the mouse brain following administration by IV injection or inhalation or aerosol. The magnitude of uptake in the brain that we observed was low compared to some other tissues, with uptake being greater following IV injection than from inhalation. However, the presence of an exogenous substance in the brain in any amount demonstrates an ability to cross the blood-brain barrier, which is an important result in itself.

In our studies, relatively high uptake was observed in the lung and thyroid following either route of administration. Although several papers confirmed thyroid uptake of manganese from different administration routes in animal models [80–83], the biology of uptake of manganese in the thyroid does not seem clear [83]. Interestingly, electron spin resonance (ESR) has shown that only a small fraction of the manganese naturally found in the rat thyroid is in the 2+ oxidation state [37], perhaps suggesting that it is bound to serum protein(s) as Mn(III) and not present as a free Mn(II). Additionally, following subcutaneous injection of Mn(II)Cl_2, in guinea pigs [81], manganese rapidly accumulated in thyroid by 1 d p.i., but then 70% of the manganese in the thyroid at 1 d had cleared by 2 d p.i.; and then 85%, by 96 hours p.i. Inhaled manganese uptake in salivary glands was in agreement with observations in welders exposed to manganese [84]. Interestingly, our results showed that injection of [⁵²Mn]MnCl₂ exhibited higher and faster uptake in salivary gland than in the inhalation study. Our results demonstrate uptake of ⁵²Mn in the brain, thyroid, and pancreas in all trials. Interestingly, uptake from IV administration was higher in brain and pancreas, but lower in thyroid, compared to inhalation results. Importantly, Graves, et al. [85] have shown that anesthesia by isoflurane can significantly decrease uptake of Mn(II) in the pancreas in fasted mice. This study concluded that isoflurane initiates a sequence in pancreatic β-cells that reduces opening of voltage-dependent calcium channels (VDCCs)—the channels through which Ca(II), Mn(II), and potentially other divalent cations can enter the pancreatic beta cells. In one part of this work, ⁵²Mn(II) in aqueous sodium acetate was administered IV with and without anesthesia by isoflurane, and the uptake in the pancreas at 1 h p.i. under anesthesia was similar to our result (in %ID/g). However, the uptake in the pancreas without anesthesia was ~3x greater than with anesthesia. Since our study only used anesthesia during IV administration of ⁵²Mn(II), and not during inhalation, we expect that in the absence of anesthesia in both routes of administration the difference would be even greater in uptake in the pancreas between IV injection and inhalation.

Along with recently renewed interest in manganese for signal in both MRI and PET, Brunnquell, et al. [40] have published results for brain uptake of manganese in female rats following tail vein infusion at 2 mL/h of non-carrier-added ⁵²Mn(II) in radiotracer concentrations. In addition to monitoring brain uptake of ⁵²Mn(II), this study included results for biodistribution from several tissues at 4 and 48 h p.i. Compared to Brunnquell, et al., our biodistribution results were roughly 10x greater for many tissues (in %ID/g), likely due to a roughly 10x lower body mass of the mice in our study compared to rats, along with different timepoints and gender. However, the relative changes from 4 to 48 h p.i. in this study was similar to our results at 1 h and 1 d, including brain, spleen, liver, intestine, heart, lung, and muscle, while results for blood, pancreas, thymus, kidney, and bone demonstrated results that were less similar to our results, based on tissue concentrations relative to other tissues and/or relative changes from 4 to 48 h. Manganese-52 might someday be used as a PET agent for niche clinical research applications, but our biodistribution results in mice can be used, in conjunction with other published results, to inform regulations for safe levels of environmental and occupational exposure to manganese. The primary objective of this work was to contribute to basic research with regards to the in vivo behavior of manganese cations, which are increasingly being studied preclinically for PET agents and MRI contrast agents—with only the possibility for future translation to the clinic.

Conclusion

In this work, we have produced ⁵²Mn via the ^natCr(p,x) reaction, chemically isolated this radioisotope of manganese from the chromium metal target material, and reported new data for the biodistribution of manganese in mice following administration by IV injection or inhalation. Manganese-52 was produced by cyclotron bombardment with low-energy protons on chromium metal, and subsequent chemical separation of ⁵²Mn(II) was performed by cation-exchange chromatography and re-dissolved in saline solution for ex vivo biodistribution studies in mice. The doses were administered by IV injection or inhalation, always resulting in uptake of ⁵²Mn in the brain, thyroid, and pancreas. Interestingly, uptake from IV administration was higher in brain and pancreas, but lower in thyroid, compared to inhalation results. While ⁵²Mn might find use as a PET agent in preclinical studies, our biodistribution results in mice may be used to inform regulations for safe levels of environmental and occupational exposure to Mn(II) and the predicted biodistribution of Mn(II) following exposure by IV injection or inhalation.

Disclaimer

“This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or limited, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.”

Supporting information

Acknowledgments

The authors gratefully acknowledge: Four Star Finishing (Saint Louis) for suggesting the method for fabricating batches of thin chromium metal foils and for performing the electroplating of chromium; the Instrument Machine Shop at WUSM for machine shop work and related advice; the Electronics Shop at WUSM for performing soldering; the Cyclotron Facility at WUSM/MIR for operation and maintenance of the CS-15 cyclotron, for other services, and for helpful advice; and the Pre-Clinical Imaging Facility at WUSM/MIR for conducting injections, dissections, gamma-ray counting of tissues, and data analysis.