The TEM assessment showed the shape and morphology of the QDs used in this study (). The analysis based on the nanoparticle size analyzer suggested that the diameter of QDs was about 4 nm. And the evaluation of the fluorescence spectrum indicated that the maximum emission wavelength was 590±10 nm, and maximum half width was ≤32 nm ().
The TEM image and the fluorescent curve of QDs used in the present study.
To evaluate the biological influences of QD accumulation in tissues from mice with the acute and chronic exposure to QDs, we first assessed QD content in various tissues. Regarding the results from the acute exposure (), QDs reflected by the cadmium amount were predominantly spread into livers, spleens and kidneys, especially livers, from mice with exposure to CdSe QDs for both doses at 20 nM and 200 nM, consistent with previous studies 
. The cadmium concentration in above tissues from the 200 nM QD-treated mice was significantly higher than that from the 20 nM QD-treated mice (P
0.017 for liver, P<0.001 for kidney and P
0.04 for spleen, ). In contrast, only a little cadmium was found in blood and bone marrow, and importantly, little cadmium was detected in the lavages from the abdominal cavities, where CdSe solution was administered (). This finding suggested CdSe particles were quickly taken away and distributed into tissues within 48 hrs post the intraperitoneal administration. To better understand the pattern of tissue distribution and accumulation of QD particles in mice, we also performed a chronic exposure at lower concentrations, 5 nM and 10 nM QDs and 20 nM CdCl2
for 6 wks. The pattern of QD accumulation in tissues was similar to that from the acute exposure as described above, and the cadmium concentration in liver from the 20 nM QD-treated mice for 6 wks was comparable to that from the 200 nM QD-treated mice for 48 hrs (). Importantly, kidney tended to retain more QD particles over 6 wks under the chronic treatment, as the cadmium concentration in kidney from the 20 nM QD-treated mice for 6 wks was similar to that in liver, and was much higher than that from the 200 nM QD-treated mice for 48 hrs ().
The distribution of QDs in various tissues from mice with acute exposure (A) and chronic exposure (B).
Substantial QD acquisition in livers, spleens and kidneys might result in injuries to these organs. We therefore evaluated tissue changes from the acute and chronic exposure at the microscopic level via histological examination. No noticeable alternation was detected in the spleens and kidneys based on histological examination for both the acute and chronic exposure (data not shown). In contrast, there were dramatic morphological alternations to the hepatic lobules in livers from mice treated with 20 nM or 200 nM QDs for 48 hrs, as indicated by disordered hepatic cords and enlarged central veins (), compared to those from the vehicle control mice. Hepatic impairments observed in livers from mice treated with 5 nM and 10 nM QDs for 6 wks were similar to those in livers from mice treated with 20 nM and 200 nM for 48 hrs (data not shown). No injury was detected in the livers of mice treated with 200 nM CdCl2 for 48 hrs (), and in the livers of mice treated with 20 nM CdCl2 for 6 wks (data not shown), compared to the vehicle control.
To further study the cytotoxicity caused by QDs at the cellular level, we delineated morphological alternations upon QD exposure by employing two types of in vitro cultured cells, murine hepatoma cells, Hepa 1–6, and monocyte-macrophage, J774A.1 cells. For this purpose, Hepa 1–6 might be able to represent hepatocytes from liver, and J774A.1 could represent Kuppfer cells from liver. Hepa 1–6 and J774A.1 cells were treated with 20 nM CdCl2, 5 nM, 10 nM and 20 nM QDs for 24 hrs and 48 hrs, and then cellular morphologies were assessed under a phase-contrast microscope. As shown in , Hepa 1–6 cells treated with 20 nM QDs for 24 hrs were condensed compared to the control cells, and the average size of the cells became smaller than that of the control cells. Similar morphological alternations were observed in cells treated with 20 nM QDs for 48 hrs (data not shown); however, no significant alternations to cellular morphologies were recognized in cells treated with 20 nM CdCl2, 5 nM and 10 nM QDs for 24 hrs and 48 hrs (data not shown). As shown by phase-contrast microscopy, the control J774A.1 cells formed typical macrophages with outward protrusions at the peripheral membrane after 24 hr culture (). In contrast, J774A.1 cells upon 20 nM QD exposure were round and condensed with fewer outward protrusions, suggesting macrophagic differentiation was impaired by QDs (). Similar observations to J774A.1 cells were recognized in response to 20 nM QDs for 48 hrs (data not shown). No significant morphological differences were noted in cells treated with 20 nM CdCl2, 5 nM and 10 nM QDs for 24 hrs and 48 hrs, compared to the control cells (data not shown). Importantly, FACS analysis of apoptosis did not suggest cell death at 24 and 48 hrs upon exposure to CdCl2 and QDs at various concentrations described above using FITC-Annexin V and PI stains (data not shown). These findings collectively demonstrated that QD particles robustly induced cytotoxicity to hepatocytes and macropahges, and potently attenuated cell differentiation without causing cell death, in parallel to the observations of the in vivo hepatoxicity ().
The cytotoxicity of QDs to in vitro cultured murine monocyte-macrophage J774A.1 cells.
Oxidative stress is currently believed to be the main modulator of toxicity upon exposure to nanomaterials including both ultrafine sphere-like nanoparticles (e.g. QDs and nanosliver) and long fiber-like carbon nanotubes 
. Thus, to shed light on the mechanism responsible for hepatic damage induced by QDs, we evaluated oxidative stress stimulated by QDs. The glutathione peroxidase (GSH-Px, as a reactive oxygen species scavenger) activity in the livers was assayed, as shown in . There was an increase in the GSH-Px activity in the livers from mice treated with 20 nM and 200 nM QDs for 48 hrs, especially for 200 nM QDs (P
0.02), compared to the vehicle control. Although not so robust as QDs, CdCl2
exposure also increased the hepatic GSH-Px activity (P>0.05, ), in agreement with a previous study 
. The enhancement in the activity of the antioxidation indicated that the oxidative stress elevated the antioxidant capability of hepatocytes. Lipid peroxidation is considered as an important index for the identification of oxidative stress. Decomposition of lipid peroxides generates a lot of products including malondialdehyde (MDA). MDA is widely used as a marker of lipid peroxidase. We observed a significant increase of MDA content in the livers from the acute QD-treated mice compared to that in the control mice (P<0.05, ). The MDA level in the livers of acute CdCl2
–treated mice was also increased compared to that in the control mice (P<0.05, ), similar to the previous studies 
. Similarly in the chronic exposure, MDA content in the livers from the mice treated with 5 nM and 10 nM QDs (especially 10 nM QDs) for 6 wks was significantly increased compared to that in the control mice (P<0.05, ). The MDA level in the livers from mice exposed to 20 nM CdCl2
for 6 wks was also increased, but not statistically significant, compared to that in the control mice (P>0.05, ). It has been documented that cadmium causes hepatic oxidative stress in mice, thus leading to liver damage characterized by increased lipid peroxidation and altered antioxidant system 
. Therefore, oxidative stress stands for a major mechanism of acute and chronic cadmium toxicity. These data together suggested that the accumulation of QDs in livers resulted in hepatic injury mainly via oxidative stress, and the hepatic toxicity from QDs appeared greater than that from an equal amount of CdCl2
. Thus, QDs might cause more severe hepatic damage than CdCl2
in that QDs could readily spread and deposit in liver in comparison to cadmium ions (as shown in ). Moreover, owing to the superfine size, CdSe particles might have the capability to readily enter hepatocytes, where they could trigger severe intracellular impairments 
QD-induced oxidative stress in vivo.
To confirm the novel hepatic toxicity induced by QDs, we performed similar assays in vitro
by treating murine hepatic Hepa 1–6 cells with CdCl2
and QDs. 24 hrs post treatment, increased GSH-Px activity was observed in QD-treated cells in a dose-dependent manner compared to the vehicle control (P<0.001, reflected by the one-way ANOVA test), particularly in the 20 nM QD-treated cells (P<0.001) (). However, no significant increase in the GSH-Px activity was detected in the CdCl2
-treated cells, even though the concentration of CdCl2
(200 nM) was 10 times higher than the highest concentration for QDs (20 nM) used in this study (). This finding indicated that the intracellular antioxidant system was considerably stimulated by QD-triggered oxidative stress to cells, but not by CdCl2
treatment. Regarding the induction of MDA, QDs enhanced the MDA level in a dose-dependent manner compared to the vehicle control (P
0.059, reflected by the one-way ANOVA test), particularly in the 20 nM QD-treated cells (P
0.016) (). CdCl2
at 200 nM also increased the MDA production compared to the vehicle control (P
0.059); however, the increase was less than that in the 20 nM QD treatment ().
QD-induced oxidative stress in vitro.
To illustrate the mechanism by which QDs promoted oxidative stress in vivo
and in vitro
, we investigated intracellular reactive oxygen species (ROS) production in response to QDs. As presented in , QDs largely induced the production of intracellular ROS in Hepa 1–6 cells at a low concentration of 5 nM for 6 hrs, similar to that upon 200 nM CdCl2
(27.54% VS 28.32%). At 10 nM, QDs generated more ROS compared to 5 nM (36.66% VS 27.54%) (). To demonstrate that ROS played a critical role in mediating oxidative stress upon QD exposure, we pre-treated cells with a potent ROS scavenger, beta-mercaptoethanol (β-ME) 
, to quench intracellular ROS. As a result, the cytotoxicity to Hepa 1–6 cells characterized by the intracellular MDA level was dramatically reduced compared to that in cells treated with 20 nM QDs only (P<0.05, ). However, the MDA level in cells upon 20 nM QD exposure with pre-treatment of β-ME was still higher than that in the control cells (P<0.05, ). These data together suggested ROS played a crucial role in mediating cytotoxicity caused by QDs; however, other unidentified mediators derived from QDs might also be contributive to the cellular impairments.
ROS and MDA generation in Hepa 1–6 cells upon QD treatment.
Taken together, CdSe particles could readily distribute into various organs upon the in vivo
administration, and liver appeared to be the predominant site for the QD accumulation. QDs induced dramatic hepatic toxicity in vivo
and in vitro
, which was much greater than that induced by cadmium ions at a similar or even a higher dose. The mechanism responsible for QD-triggered hepatoxicity might derive from the toxicity from QD particles themselves and cadmium-stimulated oxidative stress as well. The toxic effect of QDs might be partially due to the liberation of cadmium ions from the QD core 
, and the toxicity of free cadmium ions (such as cadmium-stimulated ROS) is presumably an important contribution to the overall toxicity of QDs 
. Additionally, QDs per se
as fine nanoparticles represent distinct toxic characteristics from cadmium, such as size/shape-dependent effects, and aggregation- and surface composition-associated influences. The active QD cores are involved in free radical formation (such as ROS), and free radical-mediated oxidative stress is considered as another crucial contribution to the QD toxicity 
. To sum up, the influences from both cadmium and QD cores together build up novel QD toxicities, including the hepatoxicity as discussed in the current study.