Transmission electron microscopy and scanning transmission electron microscopy study on B-site...

Delivered by Publishing Technology to Korea Institute of Science amp Technology (KIST)IP 1611226190 On Wed 10 Dec 2014 020928

Copyright American Scientific Publishers

Copyright copy 2015 American Scientific PublishersAll rights reservedPrinted in the United States of America

ArticleJournal of

Nanoscience and NanotechnologyVol 15 4229ndash4238 2015

wwwaspbscomjnn

Transmission Electron Microscopy and Scanning

Transmission X-Ray Microscopy Studies on the

Bioaccumulation and Tissue Level Absorption of

TiO2 Nanoparticles in Daphnia magna

Dongwook Kwon Hyun Woo Nho and Tae Hyun Yoonlowast

Laboratory of Nanoscale Characterization and Environmental Chemistry Department of ChemistryCollege of Natural Sciences Hanyang University Seoul 133-791 Korea

In this study bioaccumulation and tissue-level absorption of TiO2 nanoparticles (NPs) in freshwaterinvertebrates were investigated using transmission electron microscopy (TEM) and scanning trans-mission X-ray microscopy (STXM) The TiO2 NPs were used to test impacts of core sizes (ie5plusmn 2 nm and 23plusmn 7 nm for TiOSyn

2 and TiOP252 respectively) and agglomerations (ie well dis-

persed vs highly agglomerated) on the uptake of TiO2 NPs in Daphnia magna (D magna) Highlyagglomerated TiO2 NPs regardless of their core sizes were heavily taken up into the digestivetract of D magna and no detectable penetration of both TiO2 NPs into the gut epithelial cells ofD magna was observed in TEM and STXM images However significant damages involving mor-phological changes in the microvilli and gut epithelial cells (eg irregular shaped microvilli epithelialcell protrusion and dilatation of cytoplasmic inclusion) were observed only with the commercial TiO2

NPs (TiOP252 ) with larger core size and mixed crystalline phase while the laboratory synthesized

TiO2 NPs (TiOSyn2 ) with smaller core size and single crystalline phase showed slight morphological

changes in the gut microvilli and epithelial cells In the case of D magna exposed to the well dis-persed synthetic TiO2 NP (CitTiOSyn

2 only a negligible amount of TiO2 NPs were found within thedigestive tract of the D magna without any significant damages in the gut microvilli and epithelialcells and any detectable penetrations of TiO2 NPs into epithelial cells of D magna gut These TEMand STXM observations confirmed us that uptake of NP into D magna are strongly dependent ontheir agglomeration (ie hydrodynamic sizes) rather than their core sizes while direct penetrationof NPs into tissues of digestive tract seems unlikely without significant morphological changes (egcollapse of the epithelial tissue) caused by high toxicity of NPs or released metal ions

Keywords TiO2 Nanopaticle Uptake Absorption TEM STXM Daphnia magna

1 INTRODUCTIONAs a response to recent concerns on the potential haz-

ards of manufactured nanomaterials (MNs) increasing

number of studies on the toxicities of MNs has been

published12 which have demonstrated that these MNs

may cause acute and chronic toxicity for many bio-

logical systems13 Among these MNs TiO2 nanoparti-

cle (NP) is one of the well-known and widely used

MNs with many industrial applications such as photocat-

alysts biomaterials cosmetics solar cells and environ-

mental catalysts4ndash7 However Zhu et al reported an acute

lowastAuthor to whom correspondence should be addressed

toxicity (48-h LC50 as 143 mgL) of TiO2 NP for Daphniamagna D magna8 while increased mortalities were also

observed in chronic toxicity assays where uptake of TiO2

NPs into D magna have induced enhanced level of reac-

tive oxygen species (ROS) generation910 Similarly Fed-

erici et al have reported enhancements of oxidative stress

at gill and internal organs of rainbow trout (Oncorhynchusmykiss) exposed to TiO2 NPs11

Moreover the observed toxicities of TiO2 NPs are con-

sidered as closely related to their uptake while these uptake

are strongly dependent on their physicochemical properties

such as core and hydrodynamic size distributions surface

chemistry and surface charge12ndash19 For instance Cho et al

J Nanosci Nanotechnol 2015 Vol 15 No 6 1533-48802015154229010 doi101166jnn20159695 4229



TEM and STXM Studies on the Bioaccumulation and Absorption of TiO2 NPs Kwon et al

recently reported that the uptake of Au NPs by mammalian

cells is closely related with the types of surface ligand

rather than the shape of the NPs12 while Chithrani et al

also demonstrated that the uptake of Au NPs by HeLa cells

is heavily dependent upon their size (maximum NP uptake

was observed for the Au NPs with 50 nm core sizes)13

It was also reported that bacterial association with NPs

was mainly determined by the surface charge of NPs pos-

itively charged Ag NPs were more easily associated with

the Bacillus (gram positive) compared with the negatively

charged Ag NPs16 Additionally in the case of a freshwa-

ter invertebrate our recent study has shown that the uptake

of NPs were closely related with their hydrodynamic size

larger sized TiO2 NPs were accumulated than small sized

particles in the D magna gut17 The quantum dot particles

were also reported similar results18 and uptake of 20 nm

polystyrene (PS) particles by D magna was found much

less than of 1000 nm PS particle19

To understand these NP uptake processes in biological

systems various imaging techniques have been applied

For instance transmission electron microscopy combined

with energy dispersive X-ray spectroscopy (TEM-EDS)

X-ray fluorescence (XRF) microscopy and laser confo-

cal microscopy (LCM) have been used to investigate the

distribution and chemical speciation of NPs in biological

samples19ndash22 Among these approaches due to its excellent

spatial resolution TEM-EDS technique has been widely

used to monitor morphological changes in the tissues and

cells after exposure to NPs and localization of NPs within

those biological matrixes2324 Another promising analyt-

ical technique that can complement TEM approach in

the study NP uptake process is synchrotron-based scan-

ning transmission X-ray microscopy (STXM) which is

a novel spectromicroscopic technique that typically uses

finely focused soft X-rays from a highly brilliant third-

generation synchrotron source Due to recent develop-

ments in X-ray optics and the extended availability of

third-generation synchrotron sources STXM is considered

as one of the most unique and powerful synchrotron-based

analytical tools that can complement TEM approach by

providing nanometer-scale chemical information on com-

plex biological environmental speciemens25

In this study uptake of TiO2 NPs in freshwater inver-

tebrates was investigated using TEM combined with

STXM Three different TiO2 NPs (ie TiOP252 TiO

Syn

2

and CitTiOSyn

2 ) were used to test the impacts of core sizes

(TiOP252 vs TiO

Syn

2 ) and hydrodynamic sizes (TiOSyn

2 vsCitTiO

Syn

2 ) on the uptake of TiO2 NPs in D magna as well

as their absorption of TiO2 NP into the gut epithelial cells

of D magna

2 MATERIALS AND METHODS21 Preparation and Characterization of NPsTwo different TiO2 NPs commercially available aeroxide

P25 TiO2 NPs (ie TiOP252 ) and laboratory synthesized

TiO2 NPs (ie TiOSyn

2 ) were used in this study The TiOP252

was provided by the manufacturer (AEROXIDEreg TIO2

P25 Evonik Degussa Germany) for this study and the

TiOSyn

2 was synthesized in our laboratory using a previ-

ously reported protocol26 Briefly 007 mL TiCl4 (99

Cat No 206-12595 Wako Japan) was added drop wise

in 1913 mL cooled deionized water (DIW) at 1 C under

magnetic stirring followed by 30 min of slow stirring

This colloid solution was dialyzed through dialysis mem-

brane tube (SpectraPor MWCO 3500 Spectrum CA

USA) until reaching pH 25 The TiOP252 stock solution

of 10 gL was prepared by bulk powder in DIW and

sonicated for 30 min using a probe sonicator (420 W

20 kHz Sonosmasher Ulsso Hitech Korea)27 The as-

synthesized TiOSyn

2 NP was well dispersed in DIW how-

ever in moderately hard synthetic freshwater (MHW)

media it was highly agglomerated with hydrodynamic

sizes up to several microns Therefore well dispersed

TiOSyn

2 NP in MHW media (ie CitTiOSyn

2 was also pre-

pared with 020 mM sodium citrate (990 Cat No

S4641 Sigma MO USA) and compared with the highly

agglomerated TiOSyn

2 NP

The surface area of TiO2 NPs was determined by multi-

point Brunauer Emmett Teller (BET) analysis method with

N2 (nano-POROSITY Mirae SI Korea) X-ray diffrac-

tion (XRD) pattern was measured using a Cu-K radia-

tion source (Xrsquopert Pro MPD PANalytical Netherlands)

The hydrodynamic sizes of NPs in aqueous solution were

examined using a dynamic light scattering instrument

(DLS Scatteroscope I Qudix Inc Korea) To measure

the TiO2 concentration samples were digested with mix-

ture of HNO3 HCl and HF and were analyzed by using

inductively coupled plasma atomic emission spectroscopy

(ICP-AES Optima-4300 Perkin Elmer DV MA USA)

22 Exposure of TiO2 NPs to D magnaD magna were cultured according to the US EPA

standard operating procedure26 Briefly five neonates of

D magna were incubated at 20 C in each test well with

100 mL MHW media then exposed to 25 mgL TiO2 NPs

The tests were conducted at 20plusmn 2 C with a 16 h light

8 h dark photoperiod for 48 h Three replicates with five

neonates of D magna for each exposure conditions were

conducted All the exposure tests were performed at 20 Cin MHW media containing NaHCO3 96 mgL CaSO4 middot2H2O 60 mgL MgSO4 60 mgL and KCl 4 mgL

23 Sample Preparation for TEM and STXMTo prepare cross-sectioned TEM and STXM samples

D magna were fixed with 25 glutaraldehyde (25 Cat

No 18427 Ted Pella CA USA) in 01 M phosphate

buffered saline (PBS) buffer for 1 h and the glutaralde-

hyde was removed from D magna by washing with PBS

solution Then the samples were fixed in 10 osmium

tetroxide (Cat No 18456 Ted Pella CA USA) for 1 h

4230 J Nanosci Nanotechnol 15 4229ndash4238 2015



Kwon et al TEM and STXM Studies on the Bioaccumulation and Absorption of TiO2 NPs

which was followed by a washing step with PBS Dehy-

dration was performed at room temperature using a series

of 50 70 80 95 and 100 ethanol (999 Cat No

8006-02 JT Baker Malaysia) and finished by propylene

oxide for 5 min Then the samples were embedded with

propylene oxide (99 Cat No 149620010 Acros Organ-

ics Belgium) and epoxy mixture (21 12 13 (vv)) for

1 h Finally samples were polymerized by epoxy mix-

ture at 48 h at 60 C Sectioning of samples was done

with an ultramicrotome (Ultracut E ReichertLeica Ger-

many) using a diamond knife Ultra-thin sections (about

80 nm) were placed on formvar-coated 300-mesh copper

grids (Cat No 01813-F Ted Pella CA USA)

24 TEM and EDS AnalysisTo obtain high resolution images of TiO2 NPs TEM

(7600S Hitachi Japan) and HVEM (JEM-ARMI1300S

JEOL Japan) was used in this study The TiO2 NP sam-

ples were prepared by dropping TiO2 NP suspensions on

formvar-coated 300-mesh copper grids The grids were

then allowed to air dry overnight The ultra-thin sections

of the D magna gut were stained with uranyl acetate

and lead citrate and observed with TEM (7600S Hitachi

Japan) and Bio-TEM (Technai G2 Spirit FEI OR USA)

A field emission TEM (FE-TEM JEM-2100F JEOL

Japan) equipped with EDS (JSM-7000F JEOL Japan) at

an acceleration voltage of 200 was used to identify NPs in

the thin sections of D magna gut

25 STXM AnalysisScanning transmission X-ray microscopy (STXM) obser-

vations were carried out at the Canadian Light Source

(CLS Saskatoon Canada) on a 10ID-1 (Spectromi-

croscopy SM) beamline with the synchrotron storage ring

operating at 29 GeV and 200ndash300 mA stored current

Energy calibrations were performed using the O 3p Ryd-

berg peak at 5389 eV of gaseous CO2 The STXM sample

chamber was filled with He to reduce attenuation of the

soft X-ray by the air The detector for all measurements

was a photomultiplier tube (PMT) with a phosphor scin-

tillator STXM image stacks with X-ray absorption near

edge structure (XANES) spectra at the O K-edge or Ti

L-edge were obtained with the following method The

sample was raster scanned in x- and y-directions throughthe focused X-ray beam to collect STXM images at a

fixed energy Stacks of STXM images were acquired by

scanning in the xndashy direction of selected sample areas

at each energy increment over the energy range of inter-

est here x direction is horizontal y direction is verti-

cal and the xndashy plane is the plane perpendicular to the

X-ray beam XANES spectra were normalized and cor-

rected for background by dividing each spectrum by a

second spectrum (I0) taken at a location on the sample at

which the element of interest was absent AXis2000 soft-

ware (httpunicornmcmasterca) was used to align image

stacks and extract XANES spectra from the raw data of

the image stack and redraw chemical maps using XANES

spectra

3 RESULTS AND DISCUSSION31 Physicochemical Characteristics of TiO2 NPsThe primary core sizes of TiO2 NPs measured by using

HVEM and TEM were estimated as 5plusmn2 nm for TiOSyn

2

and 23plusmn7 nm for TiOP252 (see Fig 1) The TiO

Syn

2 was con-

sisted of anatase phase while TiOP252 was found as a mix-

ture of anatase (87) and rutile (13) crystalline phases28

Their mean particle sizes estimated from the Scherrerrsquos

line broadening analysis of the XRD data were 4 nm for

TiOSyn

2 and 19 nm for anatase form and 54 nm for rutile

from of TiOP252 which agreed well with the estimation

from the representative TEM images The BET surface

areas were found as 284 m2g for TiOSyn

2 and 57 m2g for

TiOP252 and the average particle diameters estimated from

these BET surface areas were 54 nm and 302 nm for

TiOSyn

2 and TiOP252 respectively which also agreed well

with the previous TEM and XRD analyses results

As shown in Figure 2 the hydrodynamic sizes of the as-

dispersed TiO2 NP stock solutions (pH 25ndash30) were mea-

sured as 9plusmn2 nm for TiOSyn

2 and 202plusmn18 nm for TiOP252 in

DIW Since the point of zero charge of TiO2 is in the range

of 58ndash68 TiO2 NPs in these stock solutions should have

highly positive surface charges under acidic conditions and

resulted in well dispersed TiO2 NP suspensions via electro-

static repulsion2829 However in MHW (pH 76ndash78 ionic

strength 675 mM) these TiO2 NPs were highly agglom-

erated and their hydrodynamic sizes were increased up to

several microns Previously it is well known that the pH

and ionic strength of the suspensions have strong influence

on the degree of TiO2 NP agglomeration29ndash31 The TiO2

NPs used in the study were agglomerated and sedimented

by the increased pH and ionic strength of the MHW media

Therefore to prevent these agglomeration and sedimenta-

tion processes we have treated TiOSyn

2 NPs with citrate

ions and prepared well-dispersed CitTiOSyn

2 NPs in MHW

media Colloidal stability of CitTiOSyn

2 NP dispersion was

confirmed by the temporal changes of the hydrodynamic

sizes which were measured as 11plusmn3 nm and 12plusmn3 nm

(after 1 h and 48 h respectively) even in the MHW media

This is mainly due to the highly negative charge devel-

oped on the TiOSyn

2 NP surface that prevents agglomeration

process283032 In the following study in addition to the

comparison of the primary core size effects (ie TiOP252

vs TiOSyn

2 ) the impact of hydrodynamic size was tested by

comparing well-dispersed TiO2 NPs (ie CitTiOSyn

2 ) with

highly agglomerated TiO2 NP (ie TiOSyn

2 )

32 Uptake of Highly AgglomeratedTiO2 NPs by D magna

TEM images shown in Figure 3 displayed ultrastructure of

unexposed D magna gut epithelial cells with clean shape

J Nanosci Nanotechnol 15 4229ndash4238 2015 4231




Figure 1 (a) and (b) HVEM images of the TiOSyn

2 NPs and (c) and (d) TEM images of the TiOP252 NPs

and regular density no significant artifact were induced

during the sample preparation procedure (ie dehydra-

tion fixation and sectioning process) The gut struc-

tures included microvilli (MV) muscular cell (MC) basal

lamina (BL) cell junctions (CJ) nucleus (N) mitochon-

dria (M) apical membrane (AM) and peritrophic mem-

brane (PTM) The PTM is a secreted cellular layer which

protects the epithelial cell of the gut and regulates the

exchange of nutrients and enzymes33

In Figure 4 TEM images of cross-sectioned guts of

D magna were presented which were exposed to highly

agglomerated TiO2 NPs (25 mgL 48 h) with different

core sizes (ie TiOP252 vs TiO

Syn

2 ) The TiO2 NPs with

high electron density were clearly visible within the cross-

sectioned gut of D magna As shown in the TEM images

both TiOP252 and TiO

Syn

2 NPs were found heavy uptake in

the gut of D magna This observation was not surprising

since D magna was known as filter feeders with fine mesh

sizes ranging from 024 to 064 m34 Through this fine

mesh filters D magna are able to actively filter particles

larger than the gap between their setulae such as bacte-

ria with sizes of 200 nm152735 It is also reported that

particles up to 50 m in diameter were found in the gut

content of large individuals36 Thus as shown in Figure 4

agglomerated TiO2 NPs (both TiOP252 and TiO

Syn

2 ) in MHW

media were easily taken up by the D magna Moreover

TiO2 NPs in the gut were further agglomerated via peri-

staltic contractions of the gut wall which is typically used

for processing and passing food through the gut As shown

in Figure 4 the gut of D magna seems to make TiO2

NPs highly packed form Similar results were previously

observed for the aquatic invertebrate species (eg cope-

pod) exposed to single walled carbon nanotube (SWNT)

in which production of compacted pure SWNT bundles

were reported in pellets of copepod fecal material through

the peristaltic contractions of the gut37 In this study both

the ingested TiOP252 and TiO

Syn

2 NPs were heavily agglom-

erated to several tens of microns and found in the vicinity

of the gut microvilli which may disturb chemical transfor-

mations occurring in the digestive tract24 These accumu-

lated NPs may interfere with food intake and ultimately

affect growth and reproduction of D magna8 Previously

it has been reported that the acute toxicity of TiO2 NPs

is mediated by reactive oxygen species (ROS) generation





Figure 2 Hydrodynamic sizes of TiOP252 TiO

Syn

2 and CitTiOSyn

2 NPs in

(a) deionized water (DIW) for pH 25ndash30 and (b) moderately hard syn-

thetic freshwater (MHW) for pH 76ndash78

via oxidative stress while the increased mortality in the

chronic bioassay was attributed to accumulated TiO2 NPs

in the intestine of D magna28

On the other hand these highly agglomerated TiO2 NPs

with different core materials (eg TiOP252 and TiO

Syn

2 )

seem to make some observable differences in their NP

induced disturbance As demonstrated in Figures 4(a)ndash(c)

Figure 3 TEM images of the gut of unexposed D magna (control) (a) Holocrine cell (asterisks) in the gut Regular microvilli (MV) muscular cell

(MC) basal lamina (BL) (b) Cellular junctions (CJ) nucleus (N) mitochondria (M) apical membrane (AM) peritrophic membrane (PTM)

the gut microvilli exposed to highly agglomerated TiOP252

NPs were found irregular in their shapes and diameters

while protruded epithelial cells were also observed These

observations were very similar with the cellular damage

reported for the C60 phototoxicity in D magna and fish

(ie fathead minnow)3839 which can be mostly ascribed

to the oxidative stress from lipid peroxidation Also ROS

generation may also damage biomolecules such as lipids

carbohydrates and proteins and may result in dysfunction

of tissues such as morphological changes (eg irregu-

lar microvilli epithelial cell protrusion and dilatation of

cytoplasmic inclusion)40ndash42 Thus morphological changes

of microvilli and epithelial cells observed in this study

might be understood as a result of ROS production from

the TiO2 NPs accumulated in the gut of D magna Com-

pared to the TEM images of control specimen shown in

Figure 3 these gut epithelial cells clearly displayed distinct

morphological changes including dilatation of cytoplas-

mic inclusions (arrows in Figs 4(a) and (c)) which might

be due to the problems in osmotic adjustment as previ-

ously suggested by Bianchini et al43 Similar observations

in gut microvilli and epithelial cells were also reported

for D magna exposed to CuO and C60 NPs2438 In con-

trast cross-sectioned images of the D magna exposed to

TiOSyn

2 displayed no significant morphological changes in

the gut microvilli and epithelial cells although those NPs

have made direct contact with microvilli due to the PTM

damage (Figs 4(d)ndash(f)) The morphological changes found

in the gut microvilli and epithelial cells (eg irregular

microvilli epithelial cell protrusion and dilatation of cyto-

plasmic inclusion) were observed only with TiOP252 NPs

(Figs 4(a)ndash(c)) while the TiOSyn

2 did not cause any mor-

phological changes in the gut microvilli and epithelial cells

(Figs 4(d)ndash(f))

In addition to the different core sizes the observed

differences between TiOP252 and TiO

Syn

2 NPs might be





Figure 4 TEM images of the gut of D magna exposed to (a)ndash(c) TiOP252 and (d)ndash(f) TiO

Syn

2 NPs (a) TiOP252 NPs isolated from the microvilli of

gut Lipid like cytoplasmic inclusions (arrow) and protrusion of gut epithelial cells (asterisk) observed (b) Protrusion of gut epithelial cells (asterisk)

observed (c) Lipid like cytoplasmic inclusions (arrow) observed (d) TiOSyn

2 NPs isolated from the microvilli (MV) of gut Cellular junctions (CJ) (e)

Peritrophic membrane (PTM) in different formation stages (arrow) (f) Cross section images of microvilli (MV) of gut

explained by the differences in their crystal phases The

TiO2 NPs are known to cause toxic effect via ROS gen-

eration resulted from their photoactivity and the photocat-

alytic activity of TiO2 NPs was reported as closely related

to their crystal phases944 Previously it was reported that

the TiO2 NPs containing both anatase and rutile phases

showed better photocatalytic activity than those TiO2 NPs

containing only single phases45 In the current study

since the TiOP252 is known as mixture of anatase and

rutile phases it might have higher photocatalytic activity

and generate more ROS than the TiOSyn

2 containing only

anatase crystal form Thus differences in crystal phases

may have produced different levels of ROS in the gut of

D magna and have caused the observed differences in the

morphologies of the gut microvilli and epithelial cells

Additionally chemical mapping of the cross-sectioned

specimen of D magna was also performed using STXM

at the oxygen K-edge (5320 eV) and titanium L-edge

(4652 eV) The O K-edge STXM images shown in

Figures 5(a) and (d) they were observed epithelial cells

microvilli of the gut and agglomerated TiO2 NPs while

the Ti L-edge STXM images shown in Figures 5(b) and (e)

displayed highly agglomerated TiO2 NPs in the gut of

D magna (Figs 5(b) and (e)) Moreover Ti maps shown

in Figures 5(c) and (f) were obtained from the subtrac-

tion of the STXM image taken at 4500 eV (below Ti

L-edge absorption energy) from the STXM image taken

at 4652 eV (above Ti L-edge absorption energy) which

helps us to clearly observe the absorbed TiO2 NPs As

shown in Figures 5(c) and (f) both TiOP252 and TiO

Syn

2 NPs

were not observed in epithelial cells of the gut and both

particles were present only in the gut lumen of D magnaPrevious studies have reported that Au NPs were not

absorbed by the microvilli although present in the vicin-

ity of gut microvilli of D magna19 A number of stud-

ies on carbon-based NPs also reported that there were

no NP absorptions by epithelial cells of D magna gut46

Our observations agreed well with the results in the lit-

eratures and confirmed the facts that the agglomerated

TiO2 NPs (gt200 nm) could be taken up into the gut of

D magna via filter feeding process but the TiO2 NPs

were unable to enter inside the epithelial cells Based on

the microscopic images it seemed that the agglomerated

TiO2 NPs were blocked by the PTM and the microvilli

of the gut Moreover it has been reported that epithe-

lial cells can only absorb molecules rather than phagocyte

particles36 Thus it is conceivably possible that smaller

sized TiO2 NPs than cross section of microvilli are able to

enter inside the epithelial cells through microvilli How-

ever TiO2 NPs in MHW media were heavily agglomer-

ated and cannot be absorbed by the epithelial cells and

microvilli of the gut Similarly most of the previous stud-

ies on the NPs uptake by D magna were also performed in

freshwater media with high ionic strength which can cause





Figure 5 STXM images of the gut of D magna exposed to (a)ndash(c) TiOP252 and (d)ndash(f) TiO

Syn

2 NPs (a) and (d) Map of oxygen absorbing at 5320 eV

showing the structure of epithelial cells and TiO2 NPs (b) and (e) Map of titanium absorbing at 4652 eV showing the presence of TiO2 NPs (c) and

(f) Titanium map obtained from the subtraction of the image taken at 4500 eV from the image taken at 4652 eV

heavy agglomeration of NPs and resulted in no NP absorp-

tions by epithelial cells of D magna However in natural

aquatic system containing natural organic matter (NOM)

the stability of NPs might be significantly improved via

electrosteric stabilization of NOM4748 Therefore micro-

scopic study using highly agglomerated NPs might not be

sufficient to confirm no NP absorptions by epithelial cells

of D magna in natural aquatic system

33 Uptake of Well Dispersed TiO2

NPs by Gut of D magnaThe well dispersed TiO2 NPs (ie CitTiO

Syn

2 ) was also

used in this study to test the impact of hydrodynamic size

on the uptake of NPs via comparison with the highly

agglomerated TiO2 NP (ie TiOSyn

2 ) Similar with the above

comparison study on the effect of primary core sizes (ie

TiOP252 vs TiO

Syn

2 ) TEM and HVEM observations were

conducted on the corss-sectioned specimens D magnaexposed to the well dispersed TiO2 NPs and the results

were presented in Figures 6 and 7 TEM images shown

in Figure 6 display the cross-sectioned gut specimen of

D magna exposed to 25 mgL CitTiOSyn

2 NPs for 48 h Inter-

estingly many electron-dense regions were found in the

vicinity of microvilli and gut epithelial cells of D magna(Figs 6(b) and (c)) However no significant morphological

changes were observed in the microvilli and epithelial cells

of D magna (Fig 6) EDS and HVEM analysis were also

performed for these electron-dense regions to verify if they

are CitTiOSyn

2 NPs present in the microvilli of D magna(Fig 6(d) and Table I) The EDS analysis results shown in

Table I showed no Ti signals at the Ti K (4511 KeV) and

Ti K (4931 KeV) energies and confirmed that there were

no TiO2 NPs present in both 1 and 2 regions of Figure 6(d)

HVEM images shown in Figure 7 also confirmed that those

electron dense regions found in the microvilli of D magnawere having different crystalline forms with TiO

Syn

2 As pre-

viously shown in Figure 1(b) the TiOSyn

2 NPs was found

to have anatase nanocrystal form with core size of sim5 nm

(Fig 1(b)) however those electrons dense NPs found in

the microvilli regions are composed clusters of cores with

diameters ofsim1 nm (Fig 7) These HVEM and EDS obser-

vations have confirmed that those electron dense regions

found in the microvilli of D magna do not correspond

to the TiOSyn

2 NPs These NPs found in microvilli regions

are clusters of metal oxides included during the sample

preparation procedures such as osmium tetra oxides These

results also suggest that even the well-dispersed TiOSyn

2 NPs

with 5 nm core size and 11 nm hydrodynamic sizes can-

not penetrate into the microvilli and gut epithelial cells of

D magnaThere were a few recent studies on the NPs absorption

into the tissues of D magna Allen et al has reported that

they observed a few citrate-coated Ag NPs in the epithe-

lial cells of D magna gut49 However in their case the

gut tissues of the organisms (ie microvilli and epithe-

lial cells) were seriously damaged by the toxicity of the





Figure 6 TEM images of the D magna gut exposed to CitTiOSyn

2 NPs

citrate-coated Ag NPs since the D magna was exposed

to an approximate LC50 concentration (about 11 ugL

AgNPs) of Ag NPs On the other hand there have been

many studies suggesting that NPs do not exist in the tissues

Figure 7 HVEM images of electron dense region in microvilli of D magna gut exposed to CitTiOSyn

2 NPs

of D magna without a serious damage such as collapse

of the cell tissue27384650 These observations also suggest

that NP uptake into D magna tissue will be possible only

with sufficient toxicity such as collapsing of the epithelial





Table I EDS analysis results of electron dense region in Figure 6(d)

image (In the EDS results Cu signal is originated from the TEM grid

while Si signal is originated from sample holder)

1 2

Element Mass () Error () Mass () Error ()

C 7950 0 8683 0

Culowast 930 006 619 010

Os 842 014 301 042

Silowast 240 010 310 009

O 037 051 086 025

Ti ND ndash ND ndash

cells and generally improbable for those NPs with rela-

tively weak toxicity This is a significant contrast with the

toxic chemical (ie deltamethrin) which was known to

permeate the epithelial cell membranes of D magna with-

out damaging tissue51

4 CONCLUSIONSIn this study uptakes of three different TiO2 NPs (ie

TiOP252 TiO

Syn

2 and CitTiOSyn

2 ) in D magna were investi-

gated using TEM and STXM imaging techniques Highly

agglomerated TiO2 NPs regardless of their core sizes

were found heavy uptake within the digestive tract of

D magna and no detectable penetration of both TiO2 NPs

into the gut epithelial cells of D magna was observed in

TEM and STXM images However significant damages

involving morphological changes in the microvilli and gut

epithelial cells (eg irregular shaped microvilli epithe-

lial cell protrusion and dilatation of cytoplasmic inclu-

sion) were observed only with the commercial TiO2 NPs

(ie TiOP252 ) with larger core size (23plusmn7 nm) and mixed

crystalline phase while the laboratory synthesized TiO2

NPs (ie TiOSyn

2 ) with smaller core size (5plusmn 2 nm) and

single crystalline phase showed only slight morphological

changes in the gut microvilli and epithelial cells In con-

trast to above cases no significant morphological damages

as well as bioaccumulations of TiO2 NPs were observed

for the D magna exposed to the well dispersed synthetic

TiO2 NP (CitTiOSyn

2 )

These TEM and STXM observations confirmed us that

uptake of NP in D magna are strongly dependent on their

hydrodynamic sizes rather than their core sizes which

are in good agreements with our recent study of the

impact of agglomeration on the bioaccumulation of TiO2

NPs17 Additionally NP penetration into the gut epithelial

cells of D magna seems to occurs rarely unless signifi-

cant morphological changes involved such as collapse of

the epithelial tissue These observations also implied that

direct penetration of NPs into biological tissues is unlikely

for those NPs with relatively weak toxicity (eg TiO2 and

SiO2)

Acknowledgments This work was supported under

the framework of an international cooperation program

managed by the National Research Foundation of

Korea (F01-2009-000-10138-0) The STXM measure-

ments described in this paper were supported by Pohang

Accelerator Laboratory (PAL) through the abroad beam-

time program of Synchrotron Radiation Facility Project

under MEST and have been performed at the Cana-

dian Light Source (CLS) Canadarsquos national synchrotron

research facility We thank Chithra Karunakaran Jian

Wang and Yingshen Lu for providing expert support at the

CLS spectromicroscopy beamline 10ID-1 We would also

like to thank the Korea Basic Science Institute (KBSI) for

the use of HVEM TEM-EDS and XRD We also acknowl-

edge Professor Kyungho Choirsquos Environmental Toxicology

Laboratory of Seoul National University (Seoul Koera) for

providing the D magna for this study

References and Notes1 V L Colvin Nat Biotechnol 21 1166 (2003)2 S J Klaine P J Alvarez G E Batley T F Fernandes R D

Handy D Y Lyon S Mahendra M J McLaughlin and J R Lead

Enviro Toxicol Chem 27 1825 (2007)3 OECD Environment directorate joint meeting of the chemicals com-

mittee and the working party on chemicals pesticides and biotech-

nology Manufactured Nanomaterials Work Programe 2009-2012

ENVJMMONO(2009)22

4 J Schulz H Hohenberg F Pflucker E Gartner T Will S Pfeiffer

V Wendel H Gers-Barlag and K P Wittern Adv Drug DeliverRev 54 S157 (2002)

5 D A Tryk A Fujishima and K Honda Electrochim Acta 45 2363

(2000)6 M Gratzel Nature 414 338 (2001)7 A Salvador M C Pascual-Marti J R Adell A Requeni and J G

March J Pharmaceut Biomed 22 301 (2000)8 X Zhu Y Chang and Y Chen Chemosphere 78 209 (2010)9 K T Kim S J Klaine J Cho S H Kim and S D Kim Sci Total

Environ 408 2268 (2010)10 A Dabrunz L Duester C Prasse F Seitz R Rosenfeldt

C Schilde G E Schaumann and R Schulz PLoS One 6 e20112

(2011)11 G Federici B J Shaw and R D Hand Aquat Toxicol 84 415

(2007)12 E C Cho Y Liu and Y Xia Angew Chem Int Ed 49 1976

(2010)13 B D Chithrani A A Ghazani and W C Chan Nano Lett 6 662

(2006)14 W Zhang B Rittmann and Y Chen Environ Sci Technol 45 2172

(2011)15 P Rosenkranz Q Chaudhry V Stone and T F Fernandes Environ

Toxicol Chem 28 2142 (2009)16 A M El Badawy R G Silva B Morris K G Scheckel M T

Suidan and T M Tolaymat Environ Sci Technol 45 283 (2011)17 D Kwon S K Jeon and T H Yoon Colloid Surf B Biointerfaces

116 277 (2014)18 N A Lewinski H Zhu H J Jo D Pham R R Kamath C R

Ouyang C D Vulpe V L Colvin and R A Drezek Environ SciTechnol 44 1841 (2010)

19 S B Lovern H A Owen and R Klaper Nanotoxicology 2 43

(2008)20 B Wang Z Wang W Feng M Wang Z Hu Z Chai and Y Zhao

Anal Bioanal Chem 398 667 (2010)21 A Elsaesser A Taylor G S De Yanes G McKerr E M Kim

E OrsquoHare and C V Howard Nanomedicine 5 1447 (2010)





22 K Tsuji K Nakano H Hayashi K Hayashi and C U Ro AnalChem 80 4421 (2008)

23 C Muhlfeld B Rothen-Rutishauser D Vanhecke F Blank P Gehr

and M Ochs Part Fibre Toxicol 4 11 (2007)24 M Heinlaan A Kahru K Kasemets B Arbeille G Prensier and

H C Dubourguier Water Res 45 179 (2011)25 T H Yoon Appl Spectrosc Rev 44 91 (2009)26 M Kolar H Mestankova J Jirkovsky M Heyrovsky and J Subrt

Langmuir 22 598 (2006)27 US EPA Methods for measuring the acute toxicity of effluents and

receiving waters to freshwater and marine organisms office of water

US Environmental Protection Agency Washington DC (2002)28 D Kwon S H Lee J Kim and T H Yoon Toxicol Environ

Health Sci 2 78 (2010)29 R A French A R Jacobson B Kim S L Isley R L Penn and

P C Baveye Environl Sci Technol 43 1354 (2009)30 J K Jiang G Oberdorster and P Biswas J Nanopart Res 11 77

(2009)31 K Suttiponparnit J K Jiang M Sahu S Suvachittanont

T Charinpanitkul and P Biswas Nanoscale Res Lett 6 27 (2011)32 N Mandzy E Grulke and T Druffel Powder Technol 160 121

(2005)33 E Gaino M Rebora A R Taddei and M Mazzini

in Ephemeroptera and Plecoptera Biologye Ecologye Systematics

edited by P Landolt and M Sartori Ultrastructural aspects of the

alimentary canal in some mayflies (1997) pp 332ndash33734 W Geller and H Mtiller Oecologia 49 316 (1981)35 M Gophen and W Geller Oecologia 64 408 (1984)36 D Ebert Ecology Epidemiology and Evolution of Parasitism in

Daphnia Bethesda (MD) National Library of Medicine (US)

National Center for Biotechnology (2005)

37 R C Templeton P L Ferguson K M Washburn W A

Scrivens and G T Chandler Environ Sci Technol 40 7387

(2006)38 X Y Yang R E Edelmann and J T Oris Aquat Toxicol 100 202

(2010)39 J E Weinstein J T Oris and D H Taylor Aquat Toxicol 39 1

(1997)40 K Kelly C Havrilla T Brady K A E Levin Environ Health

Perspect 106 375 (1998)41 G Federici B J Shaw and R D Handy Aquat Toxicol 84 415

(2007)42 J Kim S Lee C Kim J Seo Y Park D Kwon S H Lee T H

Yoon and K Choi Ecotox Environ Safe 101 240 (2014)43 A Bianchini S E G Martins and I F Barcarolli Int Congr Ser

1275 189 (2004)44 H Ma A Brennan and S A Diamond Environ Toxicol Chem

31 1621 (2012)45 T Ohno K Sarukawa K Tokieda and M Matsumura J Catal

203 82 (2001)46 K Tervonen G Waissi E J Petersen J Akkanen and J V

Kukkonen Environ Toxicol Chem 29 1072 (2010)47 E Illes and E Tombaacutecz J Colloid Interf Sci 295 115 (2006)48 M Baalousha Sci Total Environ 407 2093 (2009)49 H J Allen C A Impellitteri D A Macke J L Heckman H C

Poynton J M Lazorchak S Govindaswamy D L Roose and

M N Nadagouda Environ Toxicol Chem 29 2742 (2010)50 A Alves de Matos M S Diniz E Mendonca I Peres L Silva

J B Correia and A Picado Abstract of the Joint Congress of theSpanish and Portuguese Microscopy Societies 139 (2009)

51 T Eybe T Bohn J N Audinot T Udelhoven H M Cauchie H N

Migeon and L Hoffmann Chemosphere 76 134 (2009)

Received 20 November 2013 Accepted 20 February 2014





recently reported that the uptake of Au NPs by mammalian

cells is closely related with the types of surface ligand

rather than the shape of the NPs12 while Chithrani et al

also demonstrated that the uptake of Au NPs by HeLa cells

is heavily dependent upon their size (maximum NP uptake

was observed for the Au NPs with 50 nm core sizes)13

It was also reported that bacterial association with NPs

was mainly determined by the surface charge of NPs pos-

itively charged Ag NPs were more easily associated with

the Bacillus (gram positive) compared with the negatively

charged Ag NPs16 Additionally in the case of a freshwa-

ter invertebrate our recent study has shown that the uptake

of NPs were closely related with their hydrodynamic size

larger sized TiO2 NPs were accumulated than small sized

particles in the D magna gut17 The quantum dot particles

were also reported similar results18 and uptake of 20 nm

polystyrene (PS) particles by D magna was found much

less than of 1000 nm PS particle19

To understand these NP uptake processes in biological

systems various imaging techniques have been applied

For instance transmission electron microscopy combined

with energy dispersive X-ray spectroscopy (TEM-EDS)

X-ray fluorescence (XRF) microscopy and laser confo-

cal microscopy (LCM) have been used to investigate the

distribution and chemical speciation of NPs in biological

samples19ndash22 Among these approaches due to its excellent

spatial resolution TEM-EDS technique has been widely

used to monitor morphological changes in the tissues and

cells after exposure to NPs and localization of NPs within

those biological matrixes2324 Another promising analyt-

ical technique that can complement TEM approach in

the study NP uptake process is synchrotron-based scan-

ning transmission X-ray microscopy (STXM) which is

a novel spectromicroscopic technique that typically uses

finely focused soft X-rays from a highly brilliant third-

generation synchrotron source Due to recent develop-

ments in X-ray optics and the extended availability of

third-generation synchrotron sources STXM is considered

as one of the most unique and powerful synchrotron-based

analytical tools that can complement TEM approach by

providing nanometer-scale chemical information on com-

plex biological environmental speciemens25

In this study uptake of TiO2 NPs in freshwater inver-

tebrates was investigated using TEM combined with

STXM Three different TiO2 NPs (ie TiOP252 TiO

Syn

2

and CitTiOSyn

2 ) were used to test the impacts of core sizes

(TiOP252 vs TiO

Syn

2 ) and hydrodynamic sizes (TiOSyn

2 vsCitTiO

Syn

2 ) on the uptake of TiO2 NPs in D magna as well

as their absorption of TiO2 NP into the gut epithelial cells

of D magna

2 MATERIALS AND METHODS21 Preparation and Characterization of NPsTwo different TiO2 NPs commercially available aeroxide

P25 TiO2 NPs (ie TiOP252 ) and laboratory synthesized

TiO2 NPs (ie TiOSyn

2 ) were used in this study The TiOP252

was provided by the manufacturer (AEROXIDEreg TIO2

P25 Evonik Degussa Germany) for this study and the

TiOSyn

2 was synthesized in our laboratory using a previ-

ously reported protocol26 Briefly 007 mL TiCl4 (99

Cat No 206-12595 Wako Japan) was added drop wise

in 1913 mL cooled deionized water (DIW) at 1 C under

magnetic stirring followed by 30 min of slow stirring

This colloid solution was dialyzed through dialysis mem-

brane tube (SpectraPor MWCO 3500 Spectrum CA

USA) until reaching pH 25 The TiOP252 stock solution

of 10 gL was prepared by bulk powder in DIW and

sonicated for 30 min using a probe sonicator (420 W

20 kHz Sonosmasher Ulsso Hitech Korea)27 The as-

synthesized TiOSyn

2 NP was well dispersed in DIW how-

ever in moderately hard synthetic freshwater (MHW)

media it was highly agglomerated with hydrodynamic

sizes up to several microns Therefore well dispersed

TiOSyn

2 NP in MHW media (ie CitTiOSyn

2 was also pre-

pared with 020 mM sodium citrate (990 Cat No

S4641 Sigma MO USA) and compared with the highly

agglomerated TiOSyn

2 NP

The surface area of TiO2 NPs was determined by multi-

point Brunauer Emmett Teller (BET) analysis method with

N2 (nano-POROSITY Mirae SI Korea) X-ray diffrac-

tion (XRD) pattern was measured using a Cu-K radia-

tion source (Xrsquopert Pro MPD PANalytical Netherlands)

The hydrodynamic sizes of NPs in aqueous solution were

examined using a dynamic light scattering instrument

(DLS Scatteroscope I Qudix Inc Korea) To measure

the TiO2 concentration samples were digested with mix-

ture of HNO3 HCl and HF and were analyzed by using

inductively coupled plasma atomic emission spectroscopy

(ICP-AES Optima-4300 Perkin Elmer DV MA USA)

22 Exposure of TiO2 NPs to D magnaD magna were cultured according to the US EPA

standard operating procedure26 Briefly five neonates of

D magna were incubated at 20 C in each test well with

100 mL MHW media then exposed to 25 mgL TiO2 NPs

The tests were conducted at 20plusmn 2 C with a 16 h light

8 h dark photoperiod for 48 h Three replicates with five

neonates of D magna for each exposure conditions were

conducted All the exposure tests were performed at 20 Cin MHW media containing NaHCO3 96 mgL CaSO4 middot2H2O 60 mgL MgSO4 60 mgL and KCl 4 mgL

23 Sample Preparation for TEM and STXMTo prepare cross-sectioned TEM and STXM samples

D magna were fixed with 25 glutaraldehyde (25 Cat

No 18427 Ted Pella CA USA) in 01 M phosphate

buffered saline (PBS) buffer for 1 h and the glutaralde-

hyde was removed from D magna by washing with PBS

solution Then the samples were fixed in 10 osmium

tetroxide (Cat No 18456 Ted Pella CA USA) for 1 h

























































spectra



2


Syn

2 was con-





TiOSyn





2 and 57 m2g for



TiOSyn





















2 NPs with citrate


2 NPs in MHW


2 NP dispersion was





oped on the TiOSyn




vs TiOSyn



2 ) with


2 )























Syn





Syn











Syn

2 ) in MHW













Syn














Syn

2 and CitTiOSyn

2 NPs in








Syn

2 )





























TiOSyn











(Figs 4(d)ndash(f))



Syn

2 NPs might be






Syn

















2 containing only




















Syn

2 NPs





























Syn














Syn

2 ) was also






TiOP252 vs TiO

Syn












are CitTiOSyn







Syn

2 As pre-


2 NPs was found







to the TiOSyn





2 NPs














2 NPs






2 NPs












1 2


C 7950 0 8683 0

Culowast 930 006 619 010

Os 842 014 301 042

Silowast 240 010 310 009

O 037 051 086 025








TiOP252 TiO

Syn

2 and CitTiOSyn














NPs (ie TiOSyn







TiO2 NP (CitTiOSyn

2 )












SiO2)























ENVJMMONO(2009)22




















































































































spectra



2


Syn

2 was con-





TiOSyn





2 and 57 m2g for



TiOSyn





















2 NPs with citrate


2 NPs in MHW


2 NP dispersion was





oped on the TiOSyn




vs TiOSyn



2 ) with


2 )























Syn





Syn











Syn

2 ) in MHW













Syn














Syn

2 and CitTiOSyn

2 NPs in








Syn

2 )





























TiOSyn











(Figs 4(d)ndash(f))



Syn

2 NPs might be






Syn

















2 containing only




















Syn

2 NPs





























Syn














Syn

2 ) was also






TiOP252 vs TiO

Syn












are CitTiOSyn







Syn

2 As pre-


2 NPs was found







to the TiOSyn





2 NPs














2 NPs






2 NPs












1 2


C 7950 0 8683 0

Culowast 930 006 619 010

Os 842 014 301 042

Silowast 240 010 310 009

O 037 051 086 025








TiOP252 TiO

Syn

2 and CitTiOSyn














NPs (ie TiOSyn







TiO2 NP (CitTiOSyn

2 )












SiO2)























ENVJMMONO(2009)22















































































Syn





Syn











Syn

2 ) in MHW













Syn














Syn

2 and CitTiOSyn

2 NPs in








Syn

2 )





























TiOSyn











(Figs 4(d)ndash(f))



Syn

2 NPs might be






Syn

















2 containing only




















Syn

2 NPs





























Syn














Syn

2 ) was also






TiOP252 vs TiO

Syn












are CitTiOSyn







Syn

2 As pre-


2 NPs was found







to the TiOSyn





2 NPs














2 NPs






2 NPs












1 2


C 7950 0 8683 0

Culowast 930 006 619 010

Os 842 014 301 042

Silowast 240 010 310 009

O 037 051 086 025








TiOP252 TiO

Syn

2 and CitTiOSyn














NPs (ie TiOSyn







TiO2 NP (CitTiOSyn

2 )












SiO2)























ENVJMMONO(2009)22

































































Syn

2 and CitTiOSyn

2 NPs in








Syn

2 )





























TiOSyn











(Figs 4(d)ndash(f))



Syn

2 NPs might be






Syn

















2 containing only




















Syn

2 NPs





























Syn














Syn

2 ) was also






TiOP252 vs TiO

Syn












are CitTiOSyn







Syn

2 As pre-


2 NPs was found







to the TiOSyn





2 NPs














2 NPs






2 NPs












1 2


C 7950 0 8683 0

Culowast 930 006 619 010

Os 842 014 301 042

Silowast 240 010 310 009

O 037 051 086 025








TiOP252 TiO

Syn

2 and CitTiOSyn














NPs (ie TiOSyn







TiO2 NP (CitTiOSyn

2 )












SiO2)























ENVJMMONO(2009)22

































































Syn

















2 containing only




















Syn

2 NPs





























Syn














Syn

2 ) was also






TiOP252 vs TiO

Syn












are CitTiOSyn







Syn

2 As pre-


2 NPs was found







to the TiOSyn





2 NPs














2 NPs






2 NPs












1 2


C 7950 0 8683 0

Culowast 930 006 619 010

Os 842 014 301 042

Silowast 240 010 310 009

O 037 051 086 025








TiOP252 TiO

Syn

2 and CitTiOSyn














NPs (ie TiOSyn







TiO2 NP (CitTiOSyn

2 )












SiO2)























ENVJMMONO(2009)22

































































Syn














Syn

2 ) was also






TiOP252 vs TiO

Syn












are CitTiOSyn







Syn

2 As pre-


2 NPs was found







to the TiOSyn





2 NPs














2 NPs






2 NPs












1 2


C 7950 0 8683 0

Culowast 930 006 619 010

Os 842 014 301 042

Silowast 240 010 310 009

O 037 051 086 025








TiOP252 TiO

Syn

2 and CitTiOSyn














NPs (ie TiOSyn







TiO2 NP (CitTiOSyn

2 )












SiO2)























ENVJMMONO(2009)22

































































2 NPs






2 NPs












1 2


C 7950 0 8683 0

Culowast 930 006 619 010

Os 842 014 301 042

Silowast 240 010 310 009

O 037 051 086 025








TiOP252 TiO

Syn

2 and CitTiOSyn














NPs (ie TiOSyn







TiO2 NP (CitTiOSyn

2 )












SiO2)























ENVJMMONO(2009)22



































































1 2


C 7950 0 8683 0

Culowast 930 006 619 010

Os 842 014 301 042

Silowast 240 010 310 009

O 037 051 086 025








TiOP252 TiO

Syn

2 and CitTiOSyn














NPs (ie TiOSyn







TiO2 NP (CitTiOSyn

2 )












SiO2)























ENVJMMONO(2009)22





























































Transmission electron microscopy and scanning transmission electron microscopy study on B-site...

Documents

Transcript of Transmission electron microscopy and scanning transmission electron microscopy study on B-site...