Determination of 30 elements in colorectal biopsies by sector field inductively coupled plasma mass spectrometry: method development and preliminary baseline levels

RAPID COMMUNICATIONS IN MASS SPECTROMETRY

Rapid Commun. Mass Spectrom. 2007; 21: 1776–1782

) DOI: 10.1002/rcm.3016
Published online in Wiley InterScience (www.interscience.wiley.com
Determination of 30 elements in colorectal biopsies

by sector field inductively coupled plasma mass

spectrometry: method development and preliminary

baseline levels

Beatrice Bocca1, Antonella Lamazza2, Anna Pino1, Ercole De Masi3, Maura Iacomino3,

Daniela Mattei1, Siavash Rahimi3, Enrico Fiori2, Alberto Schillaci2, Alessandro Alimonti1*

and Giovanni Forte1

1Istituto Superiore di Sanita, Department of Environment and Primary Prevention, Viale Regina Elena 299, 00161 Rome, Italy2Hospital ‘Umberto I’, Viale del Policlinico 155, 00161 Rome, Italy3Hospital ‘San Carlo’, IDI IRCCS, Via Aurelia 275, 00165 Rome, Italy

Received 22 February 2007; Revised 28 March 2007; Accepted 28 March 2007

*CorrespoDepartmeRegina EE-mail: aContract/Tumori (IISS/N3A

An analytical procedure applicable to restricted sample sizes was developed and applied to the

analysis of 30 chemical elements in colorectal biopsies of healthy patients. Acidic microwave

digestion processed £10mg of tissue at 80-C in 15-mL polystyrene liners. The digests were diluted

to a volume of 2mL with high-purity water and directly analyzed by sector field inductively coupled

plasma mass spectrometry without further specimen handling. A careful selection of isotopes and

instrumental resolution permitted the quantification in a single analytical sequence both of the

elements present at parts per trillion and of those at parts permillion. The accuracy calculated on BCR

184 ranged from 93.3–110%, the recoveries of the biopsy material was in the range 95.2–105%, the

precision was <10%, and the blank levels were much below those expected in biopsy samples. The

metal concentrations (on a dry-weight basis) in colorectal normal tissue showed a large range of

variation: Ag, Au, Be, Bi, Co, Li, Sb, Tl, V, W and Zr were below 50ng gS1; As, Ba, Cd, Cr, Cs, Hg, Mo,

Ni, Pb, Se and Sn were distributed from 100 to 500ng gS1; Al, Cu, Fe, Mn, Sr and Zn were from a

few mg gS1 to 100mg gS1; and Ca andMg were at a level of 1000mg gS1. These data represent the first

attempt to achieve an elemental profile in the colorectal mucosa of healthy patients as baseline level

measurements for studies focused on the imbalance of chemical elements in diseased mucosa.

Copyright # 2007 John Wiley & Sons, Ltd.

Assessment of the elemental profile in human tissues is

essential to extending knowledge on the healthy and the

diseased states. From this point of view, colon-rectal mucosa

has been rarely and not fully characterized. In general,

studies on elemental concentration involved only one (e.g.,

Se or Mg)1,2 or a few analytes such as Cu, Cd, Cr, Se, Ni and

Zn.3–9 Information about other elements such as Al, Ba, Ca,

Co, Hg, Mg, Mn, Pb, and Fe is rather sparse.10–12 An

imbalance in the normal homeostasis of essential elements

and tissue accumulation of potential toxic metals might,

however, be associated with the risk of developing diseases

such as colorectal cancer.13–17

This paucity of information on elemental profiles can be

traced back to the difficulties inherent in the multi-elemental

analysis of biological tissues. First, the amount of the

ndence to: A. Alimonti, Istituto Superiore di Sanita,nt of Environment and Primary Prevention, Vialelena 299, 00161 Rome, [email protected] sponsor: LILT, Lega Italiana per la Lotta contro italian League against Cancer); contract/grant number:(2006-08).

biological sample is often limited – for example, only a

few milligrams of sample are available from human tissue

biopsies. Secondly, the analytes of interest often occur at

trace (ng g�1) or ultratrace (from pg g�1 to fg g�1)

concentration level. In addition, tissue matrix contains high

amounts of interferents which can affect the final analytical

results.

The biological matrix can be mineralized by use of

conventional hot plates and furnaces but these procedures

take time to complete the digestion and cause possible

contamination and loss of analytes. Microwave (MW)

digestion in PTFE closed vessels has advantages over

conventional methods because it reduces external contami-

nation and achieves total recovery of elements.18 Never-

theless, treatment by a high-pressure MW system is not ideal

when samples are of small sizes, such as human biopsies,

because of the large volume of the vessels (100mL). Further,

the method does not facilitate routine sample preparation


Table 1. Operating conditions for digestion and SF-ICP-MS

analysis of biopsies

Digestion

Sample weight 2.0–10 mgReagent added 0.5mL HNO3

MW program (step) Temperature (8C) Time (min)

1 25–70 152 70 153 70–80 104 80 90Samples processed up to 80Final volume 2 mL

Analysis

Geometry Double focusing reverseNier-Johnson

RF power 1.2 kWArgon flows (L min�1) Plasma, 14.0; auxiliary, 0.9;

nebulizer, 0.85Nebulization Meinhard-type glass, water-cooled

Scott spray chamberInterface Pt sampler and skimmer conesMeasurements Electric scan, 2 runs and 2 passes

for each resolution

Determination of 30 elements in colorectal biopsies 1777

because the MW rotor can handle only a maximum of

12 vessels; finally, the PTFE bomb method requires tedious

and time-consuming operations such as transfers between

containers and repeated cleaning cycles to avoid memory

effects from the walls.

Different analytical methods have been used for the

elemental analysis of tissues: neutron activation analysis,

electrothermal atomic absorption spectrometry, X-ray fluor-

escence spectrometry, etc.11,19,20 As a rule, these techniques

have high detection limits, require large sample amounts and

have to analyze each element separately in a time-consuming

process. More recently, inductively coupled plasma optical

emission spectrometry (ICP-OES) and ICP-mass spectrom-

etry (ICP-MS) have been used successfully for tissue analysis

because of their multi-elemental capabilities and their good

sensitivity.18,21–24 Sector field (SF) ICP-MS is now the method

of choice because it attains very low detection limits, up to

fg mL�1, and allows us to quantify both essential and toxic

metals in one analytical sequence so reducing sample usage

and, thus, the required sample amount.

The goal of this studywas tomeasure the concentrations of

trace (Ag, As, Au, Bi, Co, Li, Sb, Tl, V, W and Zr), minor (Al,

Ba, Cd, Cr, Cs, Hg, Mo, Mn, Ni, Pb, Se and Sn) and major

(such as Ca, Cu, Fe,Mg, Sr and Zn) elements in small samples

of tissue (�10mg) as colonic biopsies. The entire analytical

procedure was designed, from the sample collection to the

final instrumental measurement, to lower sample consump-

tion, to minimize metal contamination or loss risk, and to

carry out the analyses of all the elements in a single

SF-ICP-MS run. The method was then applied to the

investigation of baseline metal levels in healthy human

biopsies to be used for comparisons with possible abnormal

levels in colorectal cancer tissues or polyps.

EXPERIMENTAL

Reagents and instrumentationHigh-purity water from an EASY-pure apparatus (PBI,

Milan, Italy) and ultrapure HNO3 (Carlo Erba, Milan, Italy)

were used throughout the study. Screw-capped polystyrene

liners, 15mL Falcon1 (Becton Dickinson Labware, Franklin

Lakes, NJ, USA), were used for sample collection, storage,

digestion and analysis. The liners were specifically cleaned

by soaking in 10% (v/v) HNO3 for 24 h and then washed

thoroughly with high-purity water. Single-element standard

solutions (SPEX, Edison, NJ, USA) at concentrations of

1mgmL�1 were used to prepare multi-element calibrants

and internal standards (ISs).

For MW digestions, a Milestone Ethos 900-Mega II oven

(FKV Milestone, Milan, Italy) loading a MultiPREP 80 rotor

(FKV Milestone), with a capacity of 79 liners plus one

containing the temperature sensor, was used. The SF-ICP-MS

system was a ThermoFinnigan Element II model (Bremen,

Germany) equipped with a concentric glass nebulizer,

water-cooled spray chamber, and torch with guard electrode

device. The instrumental settings are summarized in Table 1.

The low (LR, 300m/Dm), medium (MR, 4000m/Dm) and

high resolution (HR, 10 000m/Dm) settings were selected

appropriately as a function of the isotope (i.e. abundance

and type of interference) and the amount of analyte expected


in the matrix (see Table 2). Three different ISs were used: in

LR mode, 115In for Ag, Au, Ba, Be, Cd, Cs, Li, Mo, Sb, Sn, Sr,

and Zr, and 195Pt for Bi, Hg, Pb, Tl and W; in the MR mode,45Sc for Al, Co, Cr, Cu, Mn, Ni and V; in HR mode, 115In for

As, Ca, Fe, Mg, Se and Zn. All the ISs were at a concentration

of 1 ng mL�1 in the analytical solutions. For metal

quantification the standard addition approach on at least

five concentration levels was used. Limits of quantification

(LoQs) were calculated as 10 times the standard deviation

(SD) of 10 readings of biopsy digests. The certified reference

material BCR 184 (bovine muscle) purchased from the

Institute for Reference Materials and Measurements (IRMM,

Geel, Belgium) was used for the accuracy control. For

elements not certified in BCR 184, recovery tests were

performed by analyzing 10 biopsy samples fortified with

known amounts of the elements. The precision was

expressed as the relative SD of 10 replicated measurements

of a pooled biopsy sample.

Sample collection and treatmentColorectal biopsy samples were provided by S. Carlo

Hospital (Rome, Italy) and Umberto I Polyclinic (Rome,

Italy) from 10 healthy individuals aged 58.3� 14.9 years

(7 males and 3 females) during a prevention campaign for

colorectal cancer. Informed consent to the study was

obtained, prior to enrolment, from the patients or their

parents. Sample collection was performed with the per-

mission of the ethics committee of the institutions.

Amounts of tissue ranging from 2.0 to 10mgwere sampled

by stainless steel biopsy forceps, previously tested for metal

contamination. Tissues were then weighed in 15-mL

polystyrene liners and stored at� 208C. The tissues were

dried at 1058C overnight, with water content ranging

between 59 and 84%. Then 0.5mL of ultrapure HNO3 was


DOI: 10.1002/rcm

Table 2. Isotopes quantified by using medium or high resolution mode

Element m/z Abundance (%) Resolution (m/Dm) Interferences separated with the selected resolution

Al 27 100 4000 11B16O, 13C14N, 12C14NH, 54Feþþ

As 75 100 10 000 40Ar35Cl, 36Ar39K, 36Ar38Ar1H,Ca 44 2.1 10 000 28Si16O, 27Al16O1H, 12C16O16O, 14N14N16OCo 59 100 4000 40Ar19F, 43Ca16O, 41K18O, 36Ar23Na, 40Ar18O1H, 42Ca16O1HCr 52 83.8 4000 40Ar12C, 36Ar16O, 38Ar14N, 35Cl17O, 37Cl15N, 35Cl16O1HCu 63 69.2 4000 40Ar23Na, 44Ca19FFe 56 91.7 10 000 40Ar16O, 40Ca16OMg 24 79 10 000 12C12C, 48Caþþ

Mn 55 100 4000 37Cl18O, 40Ar15N, 39K16O, 40Ar14N1HNi 60 26.1 4000 44Ca16O, 23Na37Cl, 36Ar24MgSe 82 9.20 10 000 66Zn16O, 40Ar42Ca, 65Cu16O1H, 64Zn17O1HV 51 99.8 4000 35Cl16O, 37Cl14N, 40Ar11B, 36Ar14N1HZn 64 48.6 10 000 40Ar24Mg, 48Ca16O, 36Ar28Si

1778 B. Bocca et al.

added to the biopsies in the liners and the samples were

MW-digested following a temperature ramp from 25 to 808C,for a total digestion time of about 2 h. The digests were

diluted to 2mL with high-purity water, the ISs were added

and the samples were analyzed. Details of the digestion

procedure are given in Table 1.

Aliquots of ca. 20mg of BCR 184 were added with 1.0mL

of ultrapure HNO3, subjected to MW irradiation together

with the biopsies and the samples were adjusted to the final

volume with high-purity water. Blank samples were run

under the same MW conditions as the matrices and were

considered in the final evaluation.

RESULTS AND DISCUSSION

Selection of the analytical procedureFrom the analytical point of view, the most critical aspect of

this multi-elemental study was the small quantity of biopsy

material, sometimes as little as 2mg. The ideal analytical

protocol should be able to restrict sample consumption as

much as possible and, at the same time, measure very low

concentrations of several elements. It thus was necessary to

redesign both the sample preparation step (that normally

needs hundreds of milligrams of sample) and the quanti-

fication step – this latter usually requires two different

techniques, one devoted to quantify major elements (e.g.

ICP-AES or atomic absorption spectrometry) and another to

trace elements (ICP-MS).21,24,25

The small tissue portions were digested by modifying a

digestion method previously used for human fluids.26 The

procedure consisted of MW digestion at atmospheric

pressure in plastic liners and offered many advantages

comparedwith traditional pressuredMWdigestions in PTFE

vessels.18 First, the liners used throughout all the work had a

small volume, i.e. they were adequate for small sample sizes.

Secondly, the same liner was used from collection to final

quantification so avoiding transfer of liquids and cleaning

cycles typical when working with PTFE vessels. Thirdly, the

method permitted us to mineralize, in a total time of 2 h, a

large number of samples at the same time instead of a

maximum of 10 samples mineralized in previous systems.

This new procedure was less prone to any contamination or


losses of analytes, and easily applicable when large batches

of samples had to be tested.

For quantification, SF-ICP-MS was the technique of choice

because it had more favourable detection limits than other

techniques, and permitted quantification of the elements in

spite of the high dilution factor applied (from 200 to

1000).27,28 By adjusting the resolution setting it was possible

to determine in a single analytical run all the elements at

concentration ranges from mgmL�1 to pgmL�1. The isotopes

and instrumental resolution were carefully selected with the

aim of eliminating spectral interferences and maintaining, at

the same time, suitable sensitivity. The LR setting was used

for 109Ag, 197Au, 138Ba, 9Be, 209Bi, 111Cd, 133Cs, 202Hg, 7Li,98Mo, 208Pb, 121Sb, 118Sn, 88Sr, 205Tl, 184W and 90Zr because

none of these elements was hampered by any relevant

interference; for each analyte the most abundant isotope was

selected to maintain the maximum instrumental sensitivity.

Exceptions were Ag at mass 107, Cd at mass 114 and Sn at

mass 120, because they were, respectively, interfered with by40Ar67Zn, 114Sn and 40Ar80Se on their most abundant masses.

The MR and the HR settings were used for those analytes

heavily overlapped by interferents resulting from the

plasma, the reagent (nitric acid) and the matrix (high

contents of Ca, K, Na and Mg) (see Table 2). The HR setting,

in addition to its main task of eliminating spectral

interferences, also helped reduce the signal intensity of

elements such as Ca and Mg present at very high

concentration in biopsies, leading to a more accurate

analysis. Only Ca, Ni and Se were not quantified at their

highest abundant masses: Ca at mass 40 was interfered with

by 40Ar and Ni at mass 58 by 58Fe. For Se determination,

among all the available isotopes, those at the 76, 78 and

80 masses could not be selected because of interference from

the huge signal of the Ar dimers. Of the masses 77 and 82, the

latter was chosen because it was completely removed from40Ar37Cl and 38Ar39K that interfered with mass 77.

Figures 1(a)–1(c) display the spectra of Se obtained at

different resolutions. In the LR setting (Fig. 1(a)) only one

peak was observed, indicating that this resolution was not

sufficient to separate the Se peak from that of the interfering

species, i.e. 40Ar42Ca, 44Ar38Ca 65Cu17O, 64Zn18O and66Zn16O. When the MR mode (Fig. 1(b)) was used, again

the Se peak appears not to be completely free from


DOI: 10.1002/rcm

Figure 1. Spectra of Se obtained at (a) low, (b) medium, and

(c) high resolution.



interferences. In the HR mode (Fig. 1(c)) the interferences

were all shifted well away from the Se peak enabling us to

quantify Se unequivocally and accurately. In Fig. 1(c) the

central peak included the signals of the interfering species40Ar42Ca, 44Ar38Ca and 66Zn16O, and the peak on the right

resulted from the sum of the signals of the 65Cu17O and64Zn18O ions.

The selection of the IS is also a key factor in obtaining

reliable data in ICP-MS analyses. The IS should have definite

characteristics: not to be a natural component of the sample

matrix; to be free from spectral interferences; and to have

a mass relatively close to that of the analyte under

investigation – i.e. to have plasma/temperature behaviour

similar that of the analyte. In this work the internal

standardization using 115In, 195Pt and 45Sc met those

requirements and was an effective compromise. In a few

cases – namely 7Li and 9Be corrected with 115In in LR, and24Mg, 44Ca and 56Fe corrected with 115In in HR – the selection

of the IS was imposed on us. In fact, in LR mode none of the

potential IS masses in the range from 20 to 80 were free from

interferences, while in HR mode the lower instrumental

sensitivity directed the choice towards an IS with an

abundance of 100% in order to have the maximum signal

intensity. Figure 2 shows the stability pattern of 7Li and24Mg signals in digested biopsies as acquired over a 1-h

period by SF-ICP-MS with and without internal standard-

ization. In normalization with an IS the RSDs were 4.16% (Li)

and 2.99% (Mg) vs. 10.8% (Li) and 13.7% (Mg) when

operating without an IS, confirming that the IS approach was

effective in minimizing possible instrumental drift.

Method performancesTable 3 shows the performances of the SF-ICP-MS method

applied to analysis of mucosa biopsies. The LoQ values were

under 0.1 ng mL�1 for most of the elements; between 0.2 and

1.0 ng mL�1 for As, Cu, Fe, Ni, Se, Sn and Sr; and slightly

higher than 1.0 ng mL�1 for Al, Ca, Mg and Zn. The blank

levels were in most cases under the LoQ values; the blanks of

Al, Ca, Cu, Fe, Mg and Zn, although at a level of a few

ng mL�1, were much below the concentrations found in real

samples. The described method had satisfactory recovery

efficiency, considering that the recoveries were often

calculated on a few ng mL�1 of element in the analytical

solution. The concentrations found were in agreement either

with the BCR expected values, ranging between 93.3% (Pb)

and 110% (Cd), or with the spiked amounts on biopsies,

ranging from 95.2% for As to 105% for Zr.

These data convinced us that there were no serious losses

or contaminations during the matrix drying and mineraliz-

ation procedures and that, despite the large dilution factor,

the sensitivity was still enough to quantify all the elements.

The only exceptionwasHg,which could not be detectedwith

sufficient accuracy because its concentration was below the

LoQ in the analytical solution. For this reason, a recovery test

by adding an adequate amount of this element to biopsies

was carried out, obtaining a recovery of 96% and confirming

that the method was satisfactory for the purpose. Finally, the

method also appeared to be precise with the variation always

being <10% for all the elements.


DOI: 10.1002/rcm

Figure 2. Stability pattern of 7Li and 24Mg signals in digested biopsies as acquired over a 1-h

period by SF-ICP-MS with and without internal standardization.

Table 3. Performances of the SF-ICP-MS analysis of biopsies

LoQ(ng mL�1)

Blank(ng mL�1)

Accuracy/Recovery (ng g�1)

Precision(%)Expected values Found values

Ag 0.003 <LoQ 25a 24.2�1.1 4.75Al 1.25 3.25 2500a 2568� 104 5.88As 0.50 <LoQ 250a 238� 21 7.46Au 0.01 <LoQ 25a 24.1� 1.5 3.66Ba 0.04 0.09 250a 257� 10 6.08Be 0.03 <LoQ 25a 25.5� 1.7 8.12Bi 0.03 <LoQ 25a 25.8� 1.2 6.38Cab 0.0015 0.010 (150) 149� 9.8 2.08Cd 0.02 0.03 13� 2c 14.3� 1.7 4.33Co 0.01 <LoQ 25a 24.0� 1.9 7.07Cr 0.06 0.10 (76–153) 111� 13 3.58Cs 0.003 <LoQ 250a 260� 18 2.07Cub 0.0008 0.002 2.36� 0.06c 2.42� 0.15 4.26Feb 0.0005 0.010 79.0� 2.0c 80.9� 3.3 4.88Hg 0.10 0.11 250a 240� 13 4.00Li 0.01 <LoQ 25a 24.5� 0.8 4.16Mgb 0.0012 0.002 (1020) 1002� 60 2.99Mn 0.08 <LoQ 334� 28c 349� 25 6.37Mo 0.04 <LoQ 250a 242� 12 4.71Ni 0.20 <LoQ (270) 265� 19 5.68Pb 0.06 0.08 239� 11c 223� 18 2.62Sb 0.02 <LoQ 25a 24.2� 1.6 4.75Se 0.30 <LoQ 183� 12c 175� 12 6.38Sn 0.30 0.45 250a 259� 17 8.44Sr 1.00 <LoQ 2500a 2402� 158 2.53Tl 0.003 <LoQ 25a 24.1� 0.7 3.68V 0.02 <LoQ 25a 24.2� 0.6 8.27W 0.02 <LoQ 25a 25.8� 1.4 6.07Znb 0.0013 0.002 166� 3c 161� 9 2.48Zr 0.03 <LoQ 25a 26.2� 1.2 7.83

aAmount spiked on a biopsies pool for recovery test.b Concentration in mg g�1.c Certified values of the BCR 184, bovine muscle. Values in brackets are indicative only.

Copyright # 2007 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2007; 21: 1776–1782

DOI: 10.1002/rcm


Table 4. Values (ng g�1 dry weight) in healthy colorectal

biopsies

Mean� SD Median Range

Ag 16.1� 13.1 9.79 2.97–37.5Al 2746� 1449 2519 864–5191As <LoQa — —Au 17.0� 8.1 13.7 7.49–28.9Ba 611� 396 489 223–1477Be <LoQb — —Bi 51.9� 30.6 36.3 24.5–97.4Cac 819� 304 756 358–1425Cd 199� 112 191 35.9–422Co 35.5� 28.1 30.2 11.5–89.2Cr 445� 342 343 92.4–1148Cs 287� 166 251 142–708Cuc 26.0� 10.0 24.1 13.3–39.6Fec 121� 75.0 107 48.2–315Hg 252� 179 221 40.5–527Li 12.1� 7.50 8.72 7.00–27.8Mgc 1094� 239 1090 756–1602Mn 3343� 1258 2814 1969–6352Mo 150� 55.0 135 85.5–290Ni 123� 40.0 120 82.2–183Pb 137� 50.0 155 50.1–217Sb 14.3� 2.8 15.1 10.3–16.6Se 737� 359 574 420–1436Sn 201� 35.0 203 163–245Sr 2791� 2227 2051 542–7846Tl 16.6� 15.5 11.5 4.61–55.3V 35.5� 20.9 27.5 19.1–80.4W 24.5� 20.5 13.5 11.9–48.1Znc 89.5� 19.4 86.5 59.0–123Zr 27.8� 25.9 17.2 8.52–88.0

aConcentration <150 ng g�1.b concentration <12 ng g�1.c concentration in mg g�1.

Table 5. Comparison with published data on colorectal tissue (m

Reference no. (year) This work 7 (1970) 12 (1971) 10 (1983) 6

Technique SF-ICP-MS NAA ES XRF

Weight dry dry dry wet

Al 2746� 1449 2900Ba 611� 396 980Cab 820� 304 804Cd 199� 112Co 35.5� 28.1Cr 445� 342 500Cub 25.9� 10.0 27.5� 0.4 8.2 1.34� 0.3 1.Feb 122� 75 109Mgb 1094� 239 589Mn 3343� 1258 1300 72� 67Ni 406� 527 66� 60Pb 137� 508 79� 53Se 737� 359Znb 89.5� 19.4 171� 13 101 15.2� 3.1 18

SF-ICP-MS: sector field inductively coupled plasma mass spectrometry; NX-ray fluorescence; AAS: atomic absorption spectrometry.aNot specified; b concentration in mg g�1; cmedians.



Elemental profile in the mucosa biopsiesThe developed method was tested on 10 colorectal biopsies

from healthy subjects. The concentration distribution of

30 elements in 10 colorectal biopsies from healthy subjects is

illustrated in Table 4. The amounts of the element in tissue

are expressed on a dry-weight basis. Elements could be

roughly divided into four groups ordered by increasing

concentration; the first group consisted of Ag, Au, Be, Bi, Co,

Li, Sb, Tl, V, W and Zr for whichmedian concentrations were

below 50 ng g�1, with Be as the lowest element (<12 ng g�1)

and Bi the highest (36.8 ng g�1). The second group was

represented by As, Ba, Cd, Cr, Cs, Hg, Mo, Ni, Pb, Se and Sn

which were distributed from 100 to 500 ng g�1, with the

lowest median level for Mo (135 ng g�1) and the highest for

Se (574 ng g�1). The third group comprised Al, Cu, Fe, Mn, Sr

and Zn with concentrations from a few mg g�1 to 100mg g�1;

and the last included Ca andMg at level of about 1000mg g�1.

Good agreement between means and median values was

observed for the majority of the elements.

The elemental concentrations found in the mucosa of

healthy subjects were tentatively compared with published

data (see Table 5). Previous results on metals in colorectal

tissue are very scarce, however, being limited to some

elements only and, in general, concern the ‘healthy portion’

of colonic tissue obtained from cancerous patients. Only the

data obtained by Martin Mateo and Martın4 were collected

from subjects not affected by tumours. However, Al, Cd, Fe

and Se biopsy concentrations agreed with those previously

reported,5,11,12 and Ba, Co and Pb were of the same order of

magnitude as previously reported data.11,12 In contrast, the

levels of Cu, Mg, Mn and Zn were generally higher than in

earlier studies.3,4,6,10,11 For Ca and Ni, the colon concen-

trations overlapped those observed in two previous

studies,5,12 but were much higher than those reported

elsewhere.8,10 The Cr concentrations were the same as those

of Arriola et al.11 andMulay et al.,12 but differed considerably

from those of Martin Mateo and Martin4 and Raithel et al.8

ean�SD, in ng g�1)

(1983) 4 (1988) 8 (1989) 5 (1990) 3 (1993) 11 (1999)

AAS AAS AAS AAS AAS NAA

wet a wet a dry dry

100� 1711.2� 0.93

27510� 1

8840� 1628 79.4c 460� 8053� 0.3 9.5� 4.5 1.79� 0.6 0.40� 0.06

131.5� 4.219.76� 0.7

40.1c 500

1310� 200.3� 3.8 18.6� 11.1 18.98� 6.4 7.53� 0.5

AA: neutron activation analysis; ES: emission spectrography; XRF:


DOI: 10.1002/rcm


CONCLUSIONS

The ideal protocol for the quantification of elements in

human biopsies should be applicable to very limited sample

sizes, process many samples in the shortest possible time,

and be almost contamination free.

The analytical procedure developed in this study has the

advantages of greatly reducing sample consumption

(2–10mg of tissue) and processing time (80 samples in

2 h), and maintaining very low blank levels. The SF-ICP-MS

quantification allowed us to accurately and precisely

quantify micro (at parts per trillion) and macro (at several

parts per million) concentrations of elements in human

biopsies in a single sequence by utilizing three different

resolution settings. The elemental distribution obtained in

healthy colorectal biopsies was as follows: Ag, Au, Be, Bi, Co,

Li, Sb, Tl, V,W andZr below 50ng g�1; As, Ba, Cd, Cr, Cs, Hg,

Mo, Ni, Pb, Se and Sn from 100 to 500 ng g�1; Al, Cu, Fe, Mn,

Sr and Zn from a few mg g�1 to 100mg g�1; and Ca andMg at

a level of 1000mg g�1. These preliminary results provide

unique information on a wide elemental profile in healthy

colonic biopsies, and they could provide a very useful

baseline for comparisons with possible abnormal levels of

metals in colorectal cancer tissues or polyps.

AcknowledgementsThis work was financially supported by the LILT, Lega

Italiana per la Lotta contro i Tumori (Italian League against

Cancer), as a part of project no. ISS/N3A (2006-08).

REFERENCES

1. Marchante-Gayon JM, Sanchez-Urıa JE, Sanz-Medel A.J. Trace Elements Med. Biol. 1996; 10: 229.

2. Pasternak K, Przyszlak W. Magnesium Res. 1999; 12: 139.3. Gupta AK, Shukla VK, Vaidya MP, Roy SK, Gupta S. J. Surg.

Oncol. 1993; 52: 172.4. MartinMateoMC,Martın G.Clin. Physiol. Biochem. 1988; 6: 321.5. Martin Mateo MC. Clin. Physiol. Biochem. 1990; 8: 261.


6. Margalioth EJ, Schenker JG, ChevionM. Cancer 1983; 52: 868.DOI: 10.1002/1097-0142(19830901)52:5<868::AID-CNCR2820520521>3.0.CO;2-K.

7. Danielsen A, Steinnes E. J. Nucl. Med 1970; 11: 260.8. Raithel HJ, Henning F, Schaller KH. J. Environ. Pathol. Tox-

icol. Oncol. 1989; 9: 115.9. Kucharzewski M, Braziewicz J, Majewska U, Gozdz S. Biol.

Trace Elem. Res. 2003; 92: 1.10. Gregoriadis GC, Apostolidis NS, Romanos AN, Paradellis

TP. Cancer 1983; 52: 508. DOI: 10.1002/1097-0142(19830801)52:3<508::AID-CNCR2820520322>3.0.CO;2-8.

11. Arriola H, Longoria L, Quintero A, Guzman D. Biol. TraceElem. Res. 1999; 71–72: 563.

12. Mulay IL, Roy R, Knox BE, Suhr NH, Delaney WE. J. Nat.Cancer Inst. 1971; 47: 1.

13. Senesse P, Meance S, Cottet V, Faivre J, Boutron-Ruault MC.Nutr. Cancer 2004; 49: 66. DOI: 10.1207/s15327914nc4901_9.

14. Weinberg ED. Biometals 1994; 7: 211. DOI: 10.1007/BF00149550.

15. Nelson RL. Nutr. Rev. 2001; 59: 140.16. Yang CY, Chiu HF. Int. J. Cancer 1998; 77: 528. DOI: 10.1002/

(SICI)1097-0215(19980812)77:4<528::AID-IJC9>3.0.CO;2-W.17. Connelly-Frost A, Poole C, Satia JA, Kupper LL,Millikan RC,

Sandler RS. Cancer Epidemiol. Biomarkers Prev. 2006; 15: 486.DOI: 10.1158/1055-9965.EPI-05-0759.

18. Krachler M, Radner H, Irgolic KJ. Fresenius J. Anal. Chem.1996; 355: 120. DOI: 10.1007/s0021663550120.

19. Nicholson JRP, Savory MG, Savory J, Wills MR. Clin. Chem.1989; 35: 488.

20. Varga I, Szebeni A, Szoboszlai N, Kovacs B. Anal. Bioanal.Chem. 2005; 383: 476. DOI: 10.1007/s00216-005-0010-0.

21. Carrilho ENVM, Gonzalez MH, Nogueira ARA, Cruz GN,Nobrega JA. J. Agric. Food Chem. 2002; 50: 4164. DOI: 10.1021/jf011605x.

22. Rahil-Khazen R, Bolann BJ, Ulvik RJ. Biometals 2002; 15: 87.DOI: 10.1023/A:1013197120350.

23. Rahil-Khazen R, Bolann BJ, Myking , Ulvik RJ. J. Trace Elem.Med. Biol. 2002; 16: 15. DOI:10.1016/S0946-672X(02)80004-9.

24. Vanhoe H. J. Trace Elem. Electrolytes Health Dis. 1993; 7:131.

25. Niazi SB, Littlejohn D, Halls DJ. Analyst 1993; 118: 821. DOI:10.1039/AN9931800821.

26. Bocca B, Alimonti A, Petrucci F, Violante N, Sancesario G,Forte G, Senofonte O. Spectrochim. Acta B 2004; 59: 559. DOI:10.1016/j.sab.2004.02.007.

27. Rodushkin I, Odman F, Olofsson R, AxelssonMD. J. Anal. At.Spectrom. 2000; 15: 937. DOI: 10.1039/b003561k.

28. Begerow J, Turfeld M, Dunemann L. J. Anal. At. Spectrom.2000; 15: 347. DOI: 10.1039/a909056h.


DOI: 10.1002/rcm

Determination of 30 elements in colorectal biopsies by sector field inductively coupled plasma mass spectrometry: method development and preliminary baseline levels

Documents