Detection of titanium particles in human liver and spleen and possible health implications

Titanium dioxide (TiO2) is produced at high volumes and applied in many consumer and food products. Recent toxicokinetic modelling indicated the potential of TiO2 to accumulate in human liver and spleen upon daily oral exposure, which is not routinely investigated in chronic animal studies. A health risk from nanosized TiO2 particle consumption could not be excluded then. Here we show the first quantification of both total titanium (Ti) and TiO2 particles in 15 post-mortem human livers and spleens. These low-level analyses were enabled by the use of fully validated (single particle) inductively coupled plasma high resolution mass spectrometry ((sp)ICP-HRMS) detection methods for total Ti and TiO2 particles. The presence of TiO2 in the particles in tissues was confirmed by Scanning Electron Microscopy with energy dispersive X-ray spectrometry. These results prove that TiO2 particles are present in human liver and spleen, with ≥24% of nanosize (< 100 nm). The levels are below the doses regarded as safe in animals, but half are above the dose that is deemed safe for liver damage in humans when taking into account several commonly applied uncertainty factors. With these new and unique human data, we remain with the conclusion that health risks due to oral exposure to TiO2 cannot be excluded.


Background
Titanium dioxide (TiO 2 ) is produced as titanium white at high production volumes, up to 6 million tons per year [1]. It is incorporated in many products, such as in food (additive E171), toothpaste, supplements and medicines, as well as in applications like paints, plastics, and cosmetics [1]. Food grade TiO 2 contains a fraction of particles in the nanosize range, which is around 10% number-based [2,3]. No acceptable daily intake (ADI) for oral ingestion of TiO 2 has been derived in the past due to the absence of observed toxic effects in the available chronic rodent study [4], the generally assumed negligible uptake of TiO 2 following ingestion [5], and the assumed insolubility and inertness of the material [6,7]. Recent human volunteer studies, however, show elevated blood Ti levels (and indications of TiO 2 particles) 6 h after ingestion of food grade TiO 2 [8], confirming earlier reports of increased blood Ti-levels after ingestion of 160 nm and 380 nm TiO 2 particles [9]. Upon evaluating food grade TiO 2 , the European Food Safety Agency (EFSA) acknowledged that TiO 2 is absorbed after oral application, albeit to a low extent, and transported to various organs [10]. Recently, very low oral (0.02 and 0.6%) absorption of TiO 2 nanoparticles has been shown in rats, with a retention of these particles in mainly the liver and spleen [11,12]. This calls for (nano)particle biokinetic studies in humans [13].
Toxicokinetic modelling of TiO 2 levels in human organs, based on animal studies and accounting for accumulation, has recently led to the conclusion that a human health risk from the oral intake of TiO 2 nanoparticles cannot be excluded [14]. Although most accumulation was seen in spleen, in the final risk assessment, a potential risk was found for the liver. It remained uncertain whether the modelled levels of TiO 2 nanoparticles for human liver and spleen are accurate, which is best verified by measurements. Although total-Ti has been detected before in human tissues like liver and spleen [15,16], there currently are no data on the presence of TiO 2 (nano)particles in human tissues from people without titanium implants [17,18]. Here, we present the first quantitative measurements of particles, both in size and concentration, in post-mortem liver and spleen of 15 human subjects (see Table 1) with a corresponding assessment of the risks that can potentially be associated with the observed total Ti and TiO 2 particle concentrations in these tissues.

Methods
Firstly, we determined the total-Ti content in human liver and spleen samples using a fully validated procedure that included the acid digestion of the formaldehyde-fixed homogenized human tissues and ICP-HRMS detection (Peters RJB, Undas A, Memelink J, van Bemmel G, Munniks S, Bouwmeester H, et al.: Development and validation of a method for the detection of titanium dioxide particles in human tissue, submitted). Next, a new, independent subsample was prepared to quantitatively determine the presence of TiO 2 particles in these tissues.
For this, highly sensitive and selective spICP-HRMS was used [2,[19][20][21]. The enzymatic and gentle chemical sample clean-up and detection method for Ti in tissues and organs was recently fully validated (Peters RJB, Undas A, Memelink J, van Bemmel G, Munniks S, Bouwmeester H, et al.: Development and validation of a method for the detection of titanium dioxide particles in human tissue, submitted). The sample preparation is known not to affect the presence and size of particles [2].

Samples and sample preparation
The inertness of TiO 2 allowed the use of livers (15) and spleens (15) obtained from bodies that were donated to the Department of Anatomy of the University Medical Centre Utrecht for educational and research purposes ( Table 1). All ethical regulations concerning the use of these organs were followed, and approval for this specific scientific use was obtained from the board of University Medical Center Utrecht. The bodies, 6 men and 9 women who died at the age of 56 to 104 years, had been fixed in 4% formaldehyde. From these persons written informed consent was obtained during life that allowed the use of their entire bodies for educational and research purposes. While there is no information about their diets, it is known that all persons involved are of Caucasian ethnicity except one who was of Asian ethnicity. All have lived in the Netherlands their entire life and it is therefore assumed that most followed a Dutch diet [22]. Of the 15 persons involved, 2 received titanium implants during their lifetime. For sample preparation, each organ was cut into small pieces and grinded to a size of 0.5-1 mm diameter. To investigate potential sample contamination, all materials that had been in contact with the organs were collected. The total-Ti concentrations in these materials or released by these materials were determined. The average of the analytical results of those blank materials were calculated and subtracted from the sample results if they were above the limit of detection (LOD).

Determination of total-Ti content
An analytical sample of 1 g was collected from each grinded and homogenized sample and brought into a perfluoroalkoxy (PFA) microwave digestion tube to which 6 mL of nitric acid (70% HNO 3 ) and 2 mL of hydrofluoric acid (40% HF), were added. All subsamples were digested for 55 min in a MARS microwave system (CEM Corporation, Matthews, NC, USA). The temperature program was as follows: at 1600 W power from 20 to 120°C in 15 min, then to 160°C in 10 min, and then to 210°C in 30 min and hold for 1 min. Following digestion and cooling to room temperature, ultra-pure water was added to a total volume of 50 mL. The extracts were shaken manually, diluted 2 times, and analysed with ICP-HRMS. Quantification was based on ionic titanium standards diluted in the same acidic matrix as the samples. Method blanks were determined by performing the complete procedure, however, without the addition of a sample. The total-Ti content in the blanks was below the method LOD.

Determination of TiO 2 particles
For the determination of particle-TiO 2 , a digestion procedure is followed to liberate the particles. This digestion procedure consists of two steps. In the first step, the tissue in the formaldehyde-fixed sample is depolymerized, while in the second step, a standard enzymatic digestion is performed. An analytical sample of 200 mg was collected from the grinded subsamples and brought into a 12-mL PE tube. In the first step, 4 mL of the digestion buffer was added and the sample was vigorously vortexed for 30 s. The digestion buffer was prepared by dissolving 300 mg of Tris buffer, 92.5 mg EDTA, 5 mg SDS and 3 g NaCl in 100 mL of Milli-Q water. Next, 4 g of glycine are added to the solution and mixed with a magnetic stirrer until complete dissolution. This solution was diluted with Milli-Q water to a final volume of 250 mL. The tube was heated for 3 h. at 100°C to depolymerize the formaldehyde-fixed tissue. In the second step, and after cooling to room temperature, 910 μL of proteinase K (2.5 mg/ml) was added. The tube was incubated for 16 h. at 37°C in a shaking water bath. After cooling to room temperature, the digest was diluted with ultra-pure water and analyzed using spICP-HRMS.

Instrumental analysis with ICP-HRMS
A Thermo Finnigan Element 2 (Thermo Fisher Scientific GmbH, Bremen, Germany), a sector-field based high resolution ICP-MS, was used to measure total-Ti in acidic extracts in standard mode and TiO 2 particles in single-particle mode (also called time resolved analysis mode). Single-particle ICP-HRMS is a method for the detection and characterization of (nano-)particles [19,21]. The Thermo Finnigan Element 2 was operated at a forward power of 1300 W and the argon gas flows were at the following settings; plasma, 15.4 L/min; nebulizer, 1.063 L/min; auxiliary, 1.2 L/min. The sample flow rate to the nebulizer was set at 0.5 mL/min. Data acquisition was done in standard mode and in time resolved analysis mode with titanium measured at m/z 46.95 in medium resolution mode to avoid interferences from 36 Ar 12 C, 32 S 16 O and 48 Ca. In standard and time resolved mode the dwell time was 250 and 2 ms respectively, with a total acquisition time of 60 s. The transport efficiency was determined by the analyses of a 50 ng/L diluted aqueous RM8013 (60 nm gold nanoparticle) suspension under the same instrumental conditions as the samples but monitoring m/z 197 for gold. Finally, single-particle data were exported as csv file and processed in a dedicated spreadsheet for the calculation of particle sizes, particle size distributions, and particle number and mass concentrations. Details about this spreadsheet and the calculation of the parameters can be found elsewhere [19]. Method blanks were determined by performing the complete procedure, however, without the addition of a sample. The mass-based TiO 2 particle concentrations in the blanks were below the method LOD. Since the blanks of the sampling materials were below the LOD of the total-Ti method they were not involved in the particle analysis.

LOD
For the total-Ti determination the LOD is calculated as 3 times the standard deviation in the results of a blank sample or a sample with a total-Ti content close to the expected LOD. This sample is analysed on each of the validation days. The LOD is calculated as follows: where k is the number of samples, y i is the result of a single sample and m is average result of the single samples.
For particle-TiO 2 there are two LOD values, one for the number-and mass-based concentration (LOD C ), and one for particle size (LOD S ). LOD C equals the minimum number of particle peaks in the time scan that differentiates a sample from a blank. A way to determine LOD C is by the IUPAC recommended approximation (Poisson) described as [23], where N is the number of particle peaks observed in the time scan of a blank. The particle number LOD C can be converted into mass units if the size and density of the particle are known. The determination of the LOD S is described by Lee et al. and can be estimated as follows [24], where σ m is the standard deviation in the background noise in the time scan, R is the ICP-MS response (cps/ μg), f a is the mass fraction of analysed element in the nanoparticle and ρ is the density of the nanoparticle material (g/cm 3 ). The upper size limit of detection is estimated to be around 1500 nm.

Confirmation of TiO 2 particles wit SEM-EDX
Two subsamples of the grinded sample material of both the livers and the spleens were studied using scanning electron microscopy with energy dispersive X-ray detection (SEM-EDX) to confirm the presence of TiO 2 particles in human liver and spleen. The samples with the highest TiO 2 concentrations (as determined with ICP-HRMS) were selected for confirmation with SEM-EDX. Typically, subsamples of > 100 tissue grains were collected on a sampling stub and dried to remove water. These subsamples were analysed with a high resolution field emission gun scanning electron microscopy in combination with energy dispersive X-ray analysis (FEG-SEM/EDX). Approximately 500 images for each sample were viewed at different magnifications (5.000-100.000 X) to identify TiO 2 particles. For each sample, approximately 10 TiO 2 particles (single nanoparticles as well as aggregates/agglomerates) were detected. The surface of the grains was systematically analysed for TiO 2 particles using the backscattered electron imaging mode. After detection of particles in a field of view, X-ray spectra from the detected particle and surrounding matrix were acquired to determine the identity. Subsequently, plasma-ashing was applied to remove the lipid fraction and obtain a sharper image of the TiO 2 particle.

Results
All tissue levels are given as wet organ weights as obtained after fixation in formaldehyde.

Total Ti measurements
As shown in Table 2, the total-Ti content in the liver ranged from 0.02 to 0.09 mg Ti/kg tissue with an average value of 0.04 ± 0.02 mg Ti/kg tissue 1 . For spleen, the total-Ti content ranged from 0.02 to 0.4 mg Ti/kg tissue with an average value of 0.08 ± 0.1 mg Ti/kg tissue 1 .
In the sparsely available literature on human data, liver and spleen concentrations ranging between 0.2 and 1.9 mg Ti/kg tissue have been detected. These concentrations have been measured using X-ray fluorescence and neutron activation analysis, while we used HR-ICPS [25,26]. The blank-corrected limit of detection (LOD) total Ti was 0.01 mg/kg tissue, while the analytical recovery for total-Ti was 112 ± 34%, which is in the range of accepted analytical standards [20]. None of the specific steps in the tissue and sample preparation contributed to the blank total-Ti. Two human subjects carried a titanium implant, the total-Ti content in liver and spleen of these subjects was comparable to those observed in the liver and spleen in other subjects.

Particle measurements
The presence of TiO 2 particles in the tissues is evidenced by the characteristic spikes in the time scans of the spICP-HRMS analysis of liver and spleen samples (Fig. 1a, b). TiO 2 particles were detected in 7/15 liver and 13/15 spleen samples ( Table 2). The smallest TiO 2 particle that can be detected with this method (LOD size ) in these tissues is 85 nm. The number-based TiO 2 particle size distributions in liver and spleen were comparable and had a size range of 85-550 and 85-720, respectively ( Table 2 and Fig. 1c). SpICP-HRMS does not allow a further characterisation of the particles being The signal height of the peaks is directly proportional to the particle's mass from which the equivalent spherical particle size is calculated [22,23]. c The calculated number-based particle size distribution (left axis and dark colours) and the calculated mass-based particle size distribution (right axis and light colours). Since the particle size distribution in liver (red bars) and spleen (blue bars) are very similar, they are stacked present as agglomerates, aggregates or primary particles. In the tissues, 24% of the TiO 2 particles in the numberbased size distribution was < 100 nm, but this fraction may be underestimated considering the LOD size of 85 nm.
The TiO 2 particle mass concentration in liver ranged from 0.01 to 0.3 mg Ti/kg tissue (1.0 × 10 9 to 21 × 10 9 TiO 2 particles/kg tissue) ( Table 2). In spleen, this concentration ranged from 0.01 to 0.4 mg Ti/kg tissue (1.2 × 10 9 to 56 × 10 9 TiO 2 particles/kg tissue). The LOD particle-number in the tissue matrix is 0.8 × 10 9 particles per kg tissue. The analytical recovery of TiO 2 particles by enzymatic digestion of the matrix is 32 ± 7%. This low analytical recovery is in accordance with best international practices for this sample preparation and detection technique [27]. Because of the low analytical recovery, the Ti concentration of the particles is presented both as a minimum (not corrected for analytical recovery), and a maximum (corrected for analytical recovery) (see Table 2).
The total-Ti values are in general within this Ti concentration range in the particles, Furthermore, tissues with high total-Ti concentrations also contained high TiO 2 particle concentrations, and vice versa. Based on the maximum Ti concentration values in particles, on average minimally 51% (liver) and 67% (spleen) of total Ti is present in these tissues as particle. Taking into account the analytical recovery (32 ± 7%) and the LOD size (85 nm) for the particles measurements, we assume all total Ti is present as particles.
We did not observe a correlation in the abundance of the TiO 2 particles in liver and spleen from the same subjects, while this would be expected based on the shared external exposure. The reason for this lack of correlation may be related to inter-individual differences in the various involved biodistribution processes.
Lastly, small tissue grains of liver and spleen from two subjects were analysed using SEM-EDX to visualize the TiO 2 particles. As shown in Fig. 2, the observed particles are composed of Ti and oxygen and are present as an aggregate or agglomerate, consisting of smaller primary particles of 75-150 nm. Presence of Ti was also confirmed semi-quantitatively by EDX analysis in dry-ashed liver and spleen samples (Fig. 2d).
Together, these analyses show that approximately all TiO 2 is present as particles in human liver and spleen, with sizes ranging 85-550 and 85-720, respectively (upper size limit of detection was > 1.5 μm). Probably also smaller particles are present, however these cannot be detected with the current methods. The SEM analysis of the particles suggests that the larger particles consist of smaller primary particles. Therefore, for the purpose of risk assessment, we assume that all Ti is ultimately present as TiO 2 nanoparticles.

Risk assessment
In a next step, the total TiO 2 levels in liver and spleen were compared to the toxicologically safe tissue levels for TiO 2 (0.14 mg/kg for spleen and a b c d 0.008 mg/kg for liver), as reported earlier [14]. For liver, the measured TiO 2 concentrations are all below the level where effects occurred in animals, which were the occurrence of liver edema and liver enzyme level changes. However, the seven measurements >LOD are above the level at which effects may occur in humans (Fig. 3). For the estimation of the safe level in humans, interspecies differences were considered and sensitive subpopulations were accounted for, which would include children, elderly, and diseased people (see also Additional file 1). It can therefore not be excluded that the observed liver levels lead to adverse effects in humans, such as a liver functioning less well, leading to e.g. less detoxification of substances in the blood, and less albumin production. For spleen, it is unlikely that adverse effects will occur in humans as the measured levels are distributed around the estimated safe levels (please note that in the key toxicological study [28], no adverse effects were reported, thus the highest tested dose was used here [14] (see also Additional file 2).

Discussion
The TiO 2 particles observed in the human liver and spleens may have entered the body through dermal, inhalatory or oral exposure. No data on exposure, and thus neither on exposure route, of the subjects included in this study during life is available. However, dermal uptake of TiO 2 particles is unlikely, as TiO 2 particles do not penetrate the (intact) human skin [17,29]. It is likely that inhalatory uptake can be neglected as the chance is small that these people (all) had occupations with TiO 2 exposure through air. In addition, the estimated maximal non-occupational exposure through this route is 4.5 μg Ti/day (with an average of 0.75 μg TiO 2 /day; based on the Ti concentrations in non-occupational settings of 0.01-0.1 μg/m 3 ) [30]. Furthermore, most of these inhaled TiO 2 particles are eliminated from the lungs by mucociliary clearance in the ciliated part of the lungs, and subsequently swallowed, as seen in some studies [31,32]. Probably, most human subjects followed a West European diet and used toothpaste, which may result in a mean oral intake of 0.06-5.5 mg TiO 2 /kg body weight/ day [3,10,33,34]. Recent human volunteer studies indicate the systemic uptake following ingestion of TiO 2 particles [8,9]. Strikingly, the size range of the TiO 2 particles in the human livers and spleens (i.e. 86-421 and 88-445 nm, respectively) falls within that of the TiO 2 particles in food products (30-600 nm diameter [2]). In conclusion, intestinal exposure, e.g. from food, toothpaste and supplements, but also from any inhaled and swallowed particles, is the most likely source of the Ti and TiO 2 particles as found in the liver and spleens of these 15 subjects. This justifies our comparison with safe tissue levels derived from oral toxicity studies. The current study shows that both the element Ti and TiO 2 particles are present in post mortem fixed human liver and spleen and that health risks from liver damage due to oral exposure to TiO 2 still cannot be excluded, especially in elderly people. Clearly, some issues as Fig. 3 Observed liver and spleen concentrations compared to toxicological effect levels. Measured total Ti (expressed as TiO 2 to enable comparison, red diamonds) and TiO 2 particle concentrations (blue ranges) in human liver and spleen are plotted against age, together with liver or spleen concentrations that are relevant for risk assessment (black lines). Open diamonds and arrows represent the possible levels in the samples where the Ti level was below the limit of detection (LOD). Measured particle concentrations are given as a range between the minimum possible level (no correction for the analytical recovery) and the maximum possible level (corrected for the analytical recovery). The solid black line represents the organ level at the No Observed Adverse Effect Level (NOAEL) in the animal study, i.e. the highest level at which no adverse effect was observed. The dashed black line represents the organ level below which no effects are expected in humans, considering several uncertainties in the animal data addressed in Heringa et al. [14] remain unresolved, like the limitations in the toxicological data set and the impact that different forms of TiO 2 , with different size [35], surface properties or crystalline structure can have on the observed toxicity. In addition, the available organs for this study were, understandably, limited to relatively older people and their TiO 2 exposure and health condition is not known. Recently, concern has been raised on the potential contribution of TiO 2 on tumor formation in the intestine [36,37]. More information on the adverse effects of TiO 2 particles, including potential effects on liver as well as on potential carcinogenic induction and promotion in the gastrointestinal tract, would reduce the uncertainties in the current risk assessment.

Conclusion
Using two independent particle characterization techniques, we unequivocally show the presence of TiO 2 particles in (post mortem) human liver and spleen and provide quantitative data on the total human organ burden of TiO 2 particles for the first time. Particles with a size between 85 and 720 nm were found in tissue, of which at least 24% was smaller than 100 nm. This unique study thereby adds another critical piece to the risk assessment puzzle for TiO 2 (nano)particles, showing that health risks related to liver damage (i.e. liver edema and liver enzyme changes) due to TiO 2 particles still cannot be excluded.