Technology

A concerted protocol for the analysis of mineral deposits in biopsied tissue using infrared microanalysis

Description
A concerted protocol for the analysis of mineral deposits in biopsied tissue using infrared microanalysis
Categories
Published
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  ORIGINAL PAPER Jennifer Anderson  Æ  Jessica DellomoAndre ´ Sommer  Æ  Andrew Evan  Æ  Sharon Bledsoe  A concerted protocol for the analysis of mineral depositsin biopsied tissue using infrared microanalysis Received: 4 June 2004/ Accepted: 25 October 2004/Published online: 10 February 2005   Springer-Verlag 2005 Abstract  The mechanism(s) by which crystals areretained in the kidney resulting in stone disease remainsunclear. Intratubular aggregation as well as crystal cellbinding, or internalization and translocation, or alter-natively nucleation and growth in the interstitial fluidare possible models. Our group is testing the hypothesisthat calcium phosphate deposits in kidneys of patientswith calcium renal stones arise in unique anatomicalregions of the kidney. Furthermore, we believe thattheir formation is conditioned by specific stone formingpathophysiologies. To test this hypothesis, we per-formed intra-operative renal papillary biopsies duringpercutaneous nephrolithotomy of kidneys from 15idiopathic calcium stone formers as well as kidneytissue from a patient who ingested ethylene glycol, anddeveloped a new protocol to accurately identify thecomposition of the calcium deposits located in the renaltissue. We developed a new histological approach thatincorporated a low-energy (low-E) reflective slide sub-strate that has similar characteristics to a commonmicroscope slide and infrared absorption microspec-troscopy. Infrared absorption microspectroscopyrevealed the crystal deposits in the idiopathic calciumoxalate stone formers to be hydroxyapatite in compo-sition with an occasional region of calcium carbonate,while calcium oxalate was the predominant mineral inthe kidney of the patient who had ingested ethyleneglycol. The results demonstrate that mixed sampletypes containing tissue and mineralized deposits areeasily analyzed while mounted on a low-E slide usingthe attenuated total internal reflectance (ATR) method.Reflection/absorption (R/A) analysis allows one toquickly survey a tissue section and provides qualitativeinformation about its components. Once interestingsites have been identified by R/A analysis, ATR anal-ysis can then be used to collect the best data possible.ATR analysis provides spectra free from many of theartifacts associated with transmission and R/A analysis,and completes the full picture of the componentscontained in the crystal deposits and tissue. We presenta method of analysis for mineralized materialsembedded in kidney tissue that uses readily or easilyobtainable materials and instrumentation. The sensi-tivity of this method allows tissue sections to remainunstained, alleviating the tedious and time-consumingconstraints of earlier methods of visual analysis. Thepresent method will save time and training, whilesimultaneously offering an unbiased analysis of miner-alized components that is more accurate and conduciveto patient treatments than previous methods. Keywords  Infrared microspectroscopy  Æ  Mineraldeposits  Æ  Renal stones  Æ  Tissue biopsy  Æ  Reflection/absorption  Æ  Imaging Introduction For well over 160 years, disease detection in biopsiedtissue has relied on the painstaking preparation of thinsections followed by contrast staining to visibly signalthe presence or absence of disease. For most tissuebiopsies, thin sectioning and contrast staining proce-dures are considered routine. The identification of crystals in renal biopsies has relied on the visualinspection of the morphology of a crystal using a lightmicroscope and/or employing selective staining proto-cols to take advantage of the mineral properties of aparticular crystal. Such approaches are subject toinvestigator bias and skill level.Infrared and x-ray diffraction methodologies havebeen successfully used on isolated renal calculi because J. Anderson  Æ  J. Dellomo  Æ  A. Sommer ( & )Molecular Microspectroscopy Laboratory,Department of Chemistry and Biochemistry,Miami University, Oxford, OH 45056 USAA. Evan  Æ  S. BledsoeDepartment of Anatomy and Cell Biology,Indiana University School of Medicine,Indiana University—Purdue University Indianapolis,Indianapolis, IN 46202 USAUrol Res (2005) 33: 213–219DOI 10.1007/s00240-004-0456-0  of their ability to yield highly accurate spectra of mineralcomposition [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22]. However, the problem with these previous studies is that they analyzed macroscopicsized samples of mineral and not the micron-sizedcrystalline deposits we are interested in identifying in thekidney tissue of stone formers [23]. The integration of amicroscope to the infrared spectrometer allows theinfrared beam to be focused in microscopic crystal do-mains at specific areas of a larger sample. This approachis termed infrared absorption microspectroscopic imag-ing and provides a unique molecular fingerprint of thecrystalline deposits to be studied. This technique hasbeen previously employed on tissue sections with verypositive results [6, 15, 16, 17, 18, 19, 20, 21, 22, 24]. However, unlike the research presented here, the sam-ples analyzed in these studies did not contain mineralsembedded in the tissue. In addition, the previous tissuesamples had been stained and held between two infra-red-transparent windows, while here, they have the op-tion of remaining unstained and merely mounted onto areflective glass substrate.Our laboratory has been interested in determining thepathogenesis of the most common form of human kid-ney stones: idiopathic calcium oxalate stone disease. Inorder to determine the mechanisms responsible for theformation of these stones, we are required to be able toaccurately characterize the mineral composition of verysmall crystalline deposits located within the tissue com-partments of these kidneys. Thus, the goal of this studywas to develop a new histological approach that incor-porated a low-energy (low-E) reflective slide substratethat has similar characteristics to a common microscopeslide, and infrared absorption microspectroscopy toaccurately determine the mineral composition of thesesmall crystals. The initial step in this new protocol al-lows the histologist to prepare tissue sections in thetraditional manner, but then mount the tissue on thereflective slide. This same reflective slide is employed forvisible as well as infrared analysis of a single tissue sec-tion. When combined with infrared analysis, this pro-tocol allows fast and easy examination of mixedsamples, omitting the process of staining tissue sampleswhile allowing the analysis and identification of bothtissue and embedded crystal deposits. Both reflection/absorption (R/A) as well as attenuated total internalreflection (ATR) analysis of the tissue and embeddedmaterial can be performed on the low-E substrate,yielding qualitative, unbiased and accurate informationabout the mineralized stone components and theembedding tissue media. Materials and methods Tissue biopsy samples were obtained from the IndianaUniversity Medical School with the informed consent of the patients, and were prepared by a certified histologist.The patients varied in age, symptoms and backgrounds,and are described in our recent study [23]. The unstainedtissue sections on the low-E substrates were imagedusing the visible CCD camera and frame grabber on theSpectrum Spotlight. A serial section stained with Yasuesilver replacement stain was employed as a control toconfirm the location of the calcium [25]. This serialsection was reviewed using a standard visible microscopein an attempt to visually determine the general area of the mineralized crystals of interest.Tissue sections from 15 patients with idiopathic cal-cium oxalate kidney stone disease and an individual whoingested ethylene glycol, as well as cross-sections of un-embedded calculi were analyzed in the present study.Papillary biopsies were obtained at the time of percu-taneous nephrolithotomy from a group of 15 well-characterized idiopathic calcium oxalate stone formers.In addition, renal biopsies were collected from fourcontrol kidneys at the time of surgical removal of upperurinary tract cancer while large tissue samples wereobtained at the time of autopsy from one patient whodied from ethylene glycol ingestion.All samples were analyzed by a Perkin-Elmer Spec-trum Spotlight 300 infrared imaging microscope thatwas equipped with an array detector for the rapidacquisition of molecular images and a single pointdetector for the acquisition of high signal to noisespectra with a lower wavenumber cutoff of 580 cm  1 .Both detectors were based on the well-established mer-cury cadmium telluride (MCT) technology. The mini-mum sample size that can be analyzed using eitherdetector was approximately two wavelengths (ca. 6  l m).The majority of spectra presented in this report werecollected using the single point detector and representthe average of 64 individual scans collected at a spectralresolution of 4 cm  1 . The microscope can be operated intransmission, reflection or attenuated total internalreflection (ATR) modes. In the latter mode, a drop-down Ge internal reflection element (IRE) was em-ployed. A 50 · 50  l m confocal aperture was employed toisolate the sample region of interest for the transmissionand reflection modes. The same aperture was employedfor the ATR mode, however, the Ge IRE provides anadditional 4 ·  magnification resulting in a sampling areaof    13 · 13  l m.The substrate used in this investigation was a low-Eglass slide (Kevley Technologies, Chesterland, Ohio)[26]. The slides have the same dimensions as those of standard glass microscope slides, making the mountingof tissue samples routine for histologists. Results Early procedures to determine an infrared microspec-troscopic technique for inclusion determination revolvedaround transmission analysis. Figure 1 displays both avisible image and a false-color infrared image of a 4  l m 214  thick tissue section containing an interstitial mineraldeposit that has been embedded in paraffin and com-pressed between two BaF 2  windows. The patient wasdiagnosed as an idiopathic calcium oxalate stone former,although currently, infrared analysis classifies the min-eralized deposits within the kidney to be calcium phos-phate, specifically, hydroxyapatite.Transmission spectra were collected from the boxedarea on the visible image in Fig. 1 to produce the false-color infrared image. The false-color image is made upof approximately 4,000 spectra, which were collected inparallel during a 15-min experiment. This image is basedon the peak height of the asymmetric stretching vibra-tion (1,050 cm  1 ) of the orthophosphate group associ-ated with hydroxyapatite. The image exemplifies thebenefit of infrared imaging for the study of tissue. Inaddition to eliminating the staining procedure, inter-pretation of the results is much less subjective thancurrent methods of crystal deposit identification.In the context of the present study, mineral depositspresent a problem in that they absorb strongly andscatter infrared radiation, which ultimately affects theresults and potential quantitative capacity of the meth-od. Figure 2 illustrates the transmission spectra of acrystal located within a kidney section (top) and of thetissue only (bottom) extracted from the infrared image inFig. 1. The spectrum of the tissue is photometricallyaccurate, whereas that of the crystal is not. The feature(1,050 cm  1 ) associated with the hydroxyapatite is to-tally absorbing, exhibiting percent transmission valuesthat approach zero. In addition, the entire spectrum hasa positive sloping baseline going toward lower wave-numbers. These artifacts arise from the strong absor-bance of the deposit and the propensity for scattering,respectively. As the size of the crystal decreases, scat-tering becomes more problematic, which could makeidentification difficult. Strong negative transmissionbands can be noticed in the spectra of Fig. 2 for both the Fig. 1  Visible image ( left ) andan IR transmission false-colorimage ( right ) of a 4  l m-thicktissue section with a renal stoneinclusion between two BaF 2 discs Fig. 2  Transmission spectra of both tissue ( bottom ) andcalculus ( top ) from anembedded sample215  crystal and tissue near 2,900 and 1,450 cm  1 . Thesefeatures are due to uncompensated embedding (paraffin)material. The embedding material is a hydrocarbon,which absorbs near 2,900, 1,450 and 720 cm  1 .Subsequent to transmission analysis, stained samplescontaining inclusions were mounted on low-E slides forR/A analysis. Figure 3 illustrates the visible image of astained tissue sample from the patient who ingestedethylene glycol. Fig. 4 illustrates infrared spectra ob-tained on a stained mineral deposit using R/A combinedwith ATR sampling modes. Based on infrared featuresobserved in the R/A spectrum, it appears as though thedeposit is comprised mostly of tissue. Again, the slopingbaseline indicates the presence of scattering. The mostprominent features in the infrared spectrum are the N-Hasymmetric stretch and the amide I and amide II bandslocated at 3250, 1650 and 1550 cm  1 , respectively. All of these features are characteristic of protein. Upon closerinspection, the spectrum reveals an inverted absorptionlocated near 780 cm  1 . This feature is present in all R/Aspectra of the deposits collected from this particularseries of samples, and is absent in spectra of the tissue.The patient was diagnosed with ethylene glycol poison-ing and oxalate crystals in the kidney tubules; sincecalcium oxalate has a feature near the absorption inquestion, the presence of the inverted band was em-ployed as evidence for the presence of calcium oxalate.The negative absorption (Reststrahlen band) arises fromanomalous dispersion, which is caused by increasedFresnel reflection near strong absorption bands. Thespectrum demonstrates that R/A is a viable method inthis particular cases. Discussion Initial attempts to study the tissue sections involvedthe use of transmission infrared microspectroscopy dueto the fact that this method typically yields the bestresults when the long-term goal includes quantitativeanalysis. However, this mode requires extensive samplepreparation to ensure that the sample thickness is onthe order of 2–6  l m. This thickness yields infraredspectra of proteinaceous materials in which the mini-mum transmittance is no less than 20%, a requirementfor quantitative analysis and photometrically accurateband intensities. Further, the sample must be mountedflat on an infrared transparent window in order toavoid sloping baselines or artifacts in the spectrum.Most transmission analysis of tissue samples involvesmounting a wet or dry section between two bariumfluoride (or similar) windows [6, 15, 17, 18, 19, 20, 22, 24], which are transparent from 50,000 cm  1 down to750 cm  1 [27]. Barium fluoride is hard and non-hygroscopic, unlike alkali halide windows such asNaCl or KBr. Mounting the tissue between twowindows in a low compression cell ensures that thesample remains relatively flat over an area of approximately 1 cm 2 .The major drawback to the transmission mode of analysis is the extensive sample preparation and theintroduction of substrate materials that are consideredforeign to the histologist. In order for a technique to bequickly accepted by the medical community, it shouldemploy methods that are commonly practiced in thatcommunity. In addition, barium fluoride windows areexpensive, somewhat brittle, and are commonly fur-nished as round disks with a diameter of 13 mm and athickness of 2 mm. Although micro-arrays of tissuesections deposited on barium fluoride have been devel-oped [28, 29, 30, 31, 32, 33], they can be very costly and time consuming to prepare. Finally, barium fluoride isonly transparent down to 750 cm  1 . The region below750 cm  1 is sometimes useful for further differentiationof mineralized deposit components.Following the analysis of tissue sections using atransmission process, reflection/absorption measure-ments were conducted using the low-E substrate [26]. These slides have similar physical characteristics toconventional glass microscope slides with the exceptionof a thin three-layer reflective coating on the surface of one side. Tissue analysis using low-E slides has beenperformed previously with much success [34].The homogeneity of the low-E substrate was testedby spin coating a thin film (approximately 0.6  l m) of poly(methylmethacrylate) on the surface of the sub-strate. Six slides, taken from two different batches pro-duced by the manufacturer, were coated with theresultant films exhibiting a faint purple hue. The peakarea of the carbonyl (C=O) absorption from 125infrared spectra in each map collected over a  4.0 · 4.0 mm 2 area for each slide was determined. Theaverage relative standard deviation of the peak area wasless than 10.7% over the mapped area. The thickness of spin cast films is known to vary by   10% independentof area. The results demonstrate that the reflectionproperties of the low-E slide are homogeneous over an Fig. 3  Visible image of a Yasue-stained tissue section mountedupon a low-E slide from an individual who ingested ethylene glycol.Many oxalate crystals were embedded in the tissue216  area of 4.0 · 4.0 mm 2 , which should be sufficient for mosthistological analyses.Although the R/A spectrum is useful from a quali-tative perspective, quantitative considerations exemplifytwo shortcomings. First, in the reflectance mode, theoptical path length is approximately double that of atransmission mode (see Fig. 5). It is approximatelydouble since the radiation must pass through the sample,reflect from the substrate, and pass back through thesample again before reaching the detector. As such, thethickness of the section should be one-half that requiredfor transmission, should any quantitative analysis beattempted. Second, scattering and reflection artifacts arestill present, which could significantly affect the inter-pretation of results. Finally, the identification of calciumoxalate should be based on more than one feature, evenif it is a well-behaved and reproducible one as in the casepresented here.In contrast to the R/A spectrum, Fig. 4 also presentsa spectrum obtained on the same mineralized depositusing the ATR method. In this method, a shaped ger-manium IRE is placed in contact with the sample. Thetip of the IRE is  100  l m, however, the sampled area isdependent upon the refractive index of the IRE and thesize of the confocal aperture [35]. For the same aperturesize as used in transmission or reflection/absorptionanalysis, the sampled area is four times smaller in theATR analysis due to the 4 ·  magnification associatedwith the refractive index of the germanium IRE. Thespectrum obtained on the deposit clearly showsthe asymmetric and symmetric stretching modes of theoxalate anion located at 1,620 and 1,318 cm  1 , respec-tively. A reference ATR spectrum of calcium oxalate isprovided for comparison. An added benefit of the ATRmethod is that the penetration of the infrared radiationinto the sample is less than 1  l m. Thus, for samplesthicker than this penetration depth, the optical paththrough the sample is independent of sample thickness.As a result, photometrically accurate spectra can beobtained on all the materials associated with renal stoneswithout having to worry about the thickness of thesection. The reduced path length also reduces scatteringthat is not only particle size dependent, but path lengthdependent as well. The only drawback to the ATRmethod is that the IRE contacts the sample, which couldpotentially damage it. However, the quality of thespectrum is, by far, the best of all the methods presentedwhen taking all things into consideration. Fig. 5  Transmission, R/A, andATR diagrams showing thepath of light through a sample Fig. 4  Spectra of calciumoxalate in human biopsy. R/A( top ), ATR ( middle ), andcalcium oxalate ATR standard( bottom )217
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks