Any sample extraction for trace metal measurement should include a sample blank so that the difference between the sample and the blank is the metal you want to measure.and not contamination from your sample preparation.As for sample preparation simple dilution with an ammonia, edta, ipa, triton X would be OK.

thanks, and what about the amount of proteins?

Dear Antonio, 

For ICP-MS determination we use a few amount of sample (in case of extraction this amount keep around 100 up to 500 microliter), usually. As mentioned by Trevor, you try sample preparation by cloud point extraction (see articles that discuss cloud point extration for analytical purposes in biologic samples).

Moreover, at the moment of analysis, should consider possible isobaric interferences on the determination of copper due to presence of triton X (if you use). If your ICP-MS have the device "reaction-collision cell" you will input some gases in order to minimize those spectral interferences. I hope it helped you.

Thanks Rodrigo, yes, concerning isobaric interferences, i'm ok, because the device i'm using is equipped with a collision cell. the problem was just the minimum amount of protein i need from an extraction in order to quantify trace metals.

In any case is it ok if i digest my extractions in a HNO3:H2O2 equimolar mix rather performing a CPE? Because my protein samples consist of solution with a unique phase.

You shouldn't just assume that because you are using a collision cell that this eliminates interferences - it may minimizes them. You should check on the ICP MS software for a list of interferences for the isotope or element you want to measure which includes, isobaric interference, polyatomic interferences, presence of divalent species, oxides, halides etc.

As for the amount of protein I would check with the literature to see what levels of trace metals to expect and from the limit of detection I would estimate the amount of protein required for detection. If this looks feasible I would then go on and extract the samples. However if you have ready access to an ICP MS I would extract and analyse the samples. If you want you could perform a multielement scan to see which elements may be present and follow this up with quantitation using the optimized conditions for their detection and quantitation.

thanks so much trevor, unfortunately i haven't very much in literature. I'll just try as you suggested me ;)

Hi Antonio - no apologies for saying this but don't forget to analyse a blank so the difference between the sample and blank reading is solely due to what's in your sample. All the best - Trevor

yes, as usual:D

Trevor's suggestions are correct.  I would first determine your limit of detection for Cu on your ICP.  Run a calibration with low conc stds and then measure your blank buffer a few times. From this you can calculate the LOD eg blk conc + 3x std dev.  You then need to know the stoichiometry of your protein eg  ceruloplasmin has about 7 atoms per molecule of protein.  This means there is a lot more protein than metal

Cu is a common contaminant in HNO3 and H2O2 that you are likely to use for protein digestion so the correct choice of reagents is critical. For example, Fluka Trace Select H2O2 is both relatively inexpensive and very low in Cu (

Respected Stanislav Strekopytov,

I am pleased to know that you are interested in metallomics. Inductively-Coupled Plasma Mass Spectrometer (ICP-MS) is the most suitable equipment to do so. The unique feature of ICP-MS is that its estimation is usually not influenced by the matrix. Argon Plasma Torch burns at as high as 5000-8000C, which can induce pyration, atomization, and ionization of the sample within seconds and you can estimate as many as 32 metals within 20 minutes. However, as mentioned by earlier scientists, this requires a considerable purity of sample, equipment, and environment around you because it can estimate as low as part per Trillian to parts per Zillian. Even a slight puff of air or respiration can make a significant difference, so you will be wearing lab suit, face mask, and powder-free gloves, ultra-pure ion free water which is relatively difficult to obtain.

I used sterilized, ions, and pyrogen-free water to prepare my samples from the mice tissue (10 mg), and cultured gene-manipulated cells (1 mg: pellete containing 10-6 cells; You can can easily estimate metal ions from as low as 10 ug protein. But to be practical, you would prefer to use at least 100 ug protein sample. If you are estimating metals from the biological fluids such as blood, serum, plasma, urine, you will like to estimate protein content from at least 100 ul sample before estimating .

To determine protein bound metals, you can use pure 0.4 M Perchloric acid. Homogenize at 14,000 rpm in the micro-centrifuge and inject as low as 10 ul of the supernatant after passing through the ultramicrofilter for nice BKg. You need not to dilute CSF samples but require ultra-filtration before injection.

You will also like to prepare your standards under strict sterilized, pyrogen-free, and clean micro-environment, preferably in the Laminar Flow Hood.

You would like to repeat your experiment at least thrice for reproducibility, precision, and accuracy.

What, I did not follow from your question, that whether you are interested in only estimating copper. If that is so, then you really do not need to use ICP-MS. Even relatively economical and simple Varian Atomic Absorption Spectrophotometer with Win-Lab computer software and a specific lamp can be used. It will also provide similar sensitivity and accuracy and will be O.K. However, its limitation is that it can estimate one element at a time. At any given time you change the lamp for estimating a metal ion, the calibration is disturbed and old lamp may not provide better sensitivity, accuracy, and reproducibility.

I guess that you have Fisher Thermoscientific ICP-MS in your lab, which is almost 10 times costly, but provides more efficient, accurate, and precise answers for the metallomic analysis: particularly free vs protein-bound metal ions specifically to metal-binding proteins for the personalized diagnosis, treatment, and prognosis of acute and chronic diseases.

I do hope this information will be useful to you. If you have specific question, please let me know. I will be pleased to answer to the best of my knowledge and ability. Thanks and good luck with your experiments. Regards,

Sushil Sharma, Ph. D; D.M.R.I.T

Academic Dean

American International School of Medicine

Georgetown, Guyana, South America