Supplementary information Size dependent translocation and fetal accumulation of gold nanoparticles from maternal blood in the rat

The sulfonated triphenylphosphine (S-TPP) surface functionality of the ligand protected AuNP are characterized by the high mobility of the S-TPP molecules on the AuNP surface. As it is known, they can easily be substituted by stronger ligands, following the rules of complex chemistry [1]. In a biological system, numerous stronger ligands such as proteins, sulfur containing molecules etc. are available and will substitute the original phosphines partially or quantitatively. Hence, the administration of initially S-TPP-coated AuNP provides binding dynamics of serum proteins and biomolecules in blood to the naked surface of the injected AuNP after the phosphine ligand has been replaced. Note that the free S-TPP did not cause any detectable toxic response in in vitro tests [2]. Our latest in vivo AuNP inhalation studies support the notion of the rapid replacement of the S-TPP ligand coating in the lungs: When comparing the 24 hours biodistribution after intratracheal instillation of 18 nm AuNP with that of freshly generated, pristine 20 nm AuNP inhaled as an aerosol, the biodistribution in all organs, blood, remaining carcass and in excretion was remarkably similar suggesting rapid removal of the S-TPP coating and similar patterns of lung protein binding after both applications.

: Hydrodynamic size distribution (volume) of original and neutron-irradiated 18 nm AuNP coated with sulfonated triphenylphosphine; Z avg is 21 nm and 22 nm, respectively; PdI is 0.07 and 0.14, respectively. Figure S1b: Hydrodynamic size distribution (volume) of original and neutron-irradiated 80 nm AuNP coated with sulfonated triphenylphosphine; Z avg is 89 nm and 94 nm, respectively; PdI is 0.15 and 0.12, respectively.

AuNP concentrations per mass of organs and tissues
In Figure S2 the retained 198 AuNP fractions of the uterus and blood of pregnant rats versus nonpregnant controls are given as concentrations (fractions per weight of uterus or blood) for all three 198 AuNP sizes. Au-mass concentrations (in µg) are obtained when multiplying these fractions with the IV administered AuNP masses given in Table 2   AuNP size the concentrations are rather similar between pregnant rats and non-pregnant controls.
(n=4). Au-mass concentrations (in µg) are obtained when multiplying these fractions with the IV administered AuNP masses given in Table 1. (n=4; **** p < =0.0001). Statistical analysis of pregnant vs. non-pregnant rats by one-way analysis of variance (ANOVA) followed by post hoc Sidak's multiple comparisons test.

AuNP elimination from blood
Regarding the measurements of AUC after an IV bolus injection of AuNP we have analyzed the biodistribution at two time points (1h, 24h) after IV injection which is published in a previous paper [3]. We now are comparing the total AuNP contents in blood at two time points (1h, 24h) of nonpregnant rats with the data of this MS.  Table S1 from the previous paper [3] shows that about half of the 1.4 nm AuNP were still circulating in blood one hour after IV injection. This high level decreased by a factor of 10 within the next 23 h such that still 7 % were circulating. About 99% of the 18 nm AuNP have been eliminated from blood within 1 h and this level was found also after 24 h. Surprisingly, a five-fold higher concentration of the 80 nm AuNP were left in blood after 1 h which declined to about half during 24 h after injection.
Since the AuNP concentrations in blood were reproduced rather well by non-pregnant rats of the current manuscript ( Figure 3) and, in addition, there is very little difference between pregnant and non-pregnant rats as shown in Figures 3 and S2, the assumption is plausible that similar data after 1 h as those in Table 3 are to be expected in pregnant rats 1 h after injection. So the elimination kinetics from blood to the organs and tissues within 24 h after injection is similar between pregnant and nonpregnant rats for the same sized AuNP but differs largely between the three different AuNP sizes.

Hepato-biliary clearance (HBC) of AuNP
In both IV injection studies, [3] and the current, fecal excretion results predominantly from hepatobiliary AuNP clearance. The kinetics data of hepato-biliary clearance (HBC, i.e. AuNP in GIT + feces) after 1 h and 24 h of the previous paper [3] are shown in the Table S2 below: The fecal excretion of all three AuNP is dominant compared to urinary excretion but differs significantly not only between the three AuNP sizes but also between pregnant and non-pregnant rats.
In Figure S4  The rapidly decreasing HBC with increasing AuNP size had already been shown in our previous paper [3] using an even larger set of these AuNP. As noted there, NP in the liver,-not immediately trapped by Kupffer cells -are translocated through the fenestrated vascular endothelium into the Dissé spaces to be taken up by hepatocytes and processed into biliary canaliculi [4]. From there, they are between pregnant and non-pregnant rats. The reduced HBC occurs simultaneously with a higher AuNP retention in liver during pregnancy ( Figure 2); yet, the underlying mechanisms, particularly, in pregnancy are not yet understood and require further investigations. Since very little is known about HBC of different NP and their physico-chemical properties and holds also for nanostructured drugs, no rational comparisons or extrapolations can be drawn currently. Our data provide a first attempt to shed light on the physiological function of HBC including the differences between pregnancy and non-pregnancy.

Estimate of translocated AuNP rate across the placental barrier by transcytosis
Rather than estimating AuNP translocation rates through an in vitro monolayered, epithelial cell culture [5] or similar, we base this estimate on our previous in vivo translocation data across the alveolar-capillary-barrier of the alveolar region of rat lungs of the same Wistar-Kyoto strain for which we have determined data obtained for the same 1.4 nm and 18 nm 198 AuNP that were used in the present paper [6]. Since in the present paper no 80 nm 198 AuNP were found in the fetuses, transcytosis -if it did occur -must have been below the detection limit. The lung alveolar-capillarybarrier consists of a single layer of epithelial cells, a basal membrane and the single layer of endothelial cells and does not have canaliculi [7]. The placenta also consists of a maternal epithelial layer, a basal membrane and the fetal endothelial layer but the epithelium consists of multiple layers of syncytiotrophoblastic cells through which transcytosis would have to occur to reach the fetal blood circulation. Furthermore, in both the placental and the alveolar-capillary-barrier tight junctions between epithelial cells are reported [7]. Therefore, we suggest that the alveolar-capillary-barrier may be viewed as a model barrier for AuNP translocation kinetics across barriers. Because the single cell layers in the lungs are much thinner than in the placenta the translocation rate through the alveolar-capillary-barrier may be considered an overestimate of AuNP transcytosis across the placenta. For the rat lungs we determined 24-hours translocation rates towards blood circulation of 0.084 and 0.0020 d -1 of the intratracheally instilled 1.4 and 18 nm 198 AuNP, respectively [6]. This implies that the placental transcellular pathway is negligible and suggests a dominant transport through another route such as the transtrophoblastic canaliculi. For the 1.4 nm 198 AuNP the transfer rate across the lung barrier is a factor of 5 higher than for the placenta supporting the notion that according to the lower transfer rate, a small fraction of the 1.4 nm 198 AuNP may have crossed the placental barrier by transcytotic processes. However, additional differences in transfer mechanisms of the 1.4 nm 198 AuNP between alveolar and placental barriers need to be considered: while fast diffusion across the thin alveolar-capillary-barrier may occur, an equally fast translocation through the transtrophoblastic channels of the thicker trophoblast may result in the above mentioned fivetimes smaller placental transfer rate. In fact, fast diffusion and cellular uptake in the alveolar epithelium has indeed been reported by Geiser and co-workers [8].