A highly reliable and budget-friendly Peltier-cooled camera for biological fluorescence imaging microscopy

A highly reliable and budget-friendly Peltier-cooled camera for biological fluorescence imaging microscopy
of 8
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
   Journal of Microscopy, Vol. 228, Pt 3 2007, pp. 264–271Received 18 January 2007; accepted 16 June 2007 Ahighlyreliableandbudget-friendlyPeltier-cooledcameraforbiologicalfluorescenceimagingmicroscopy KOEN JOLLING † , MARTIN VANDEVEN ‡ , JIMMY VAN DEN EYNDEN ‡ , MARCEL AMELOOT ‡ & EMMY VAN KERKHOVE † Hasselt University,  † Centre for Environmental Sciences and   ‡ Biomedical Research Institute,Laboratory of Physiology, Diepenbeek, Belgium Keywords.  Fluorescence imaging, photomultiplier, SAC8.5 CCD camera,signal-to-noise ratio. Summary The SAC8.5, a low-cost Peltier-cooled black and white8-bit CCD camera for astronomy, was evaluated forits use in imaging microscopy. Two camera–microscopeconfigurations were used: an epifluorescence microscope(Nikon Eclipse TE2000-U) and a bottom port laser scanningconfocal microscope system (Zeiss LSCM 510 META). Mainadvantages of the CCD camera over the currently usedphotomultiplier detection in the scanning setup are fastimage capturing, stable background, an improved signal-to-noise ratio and good linearity. Based on DAPI-labelledChinese Hamster Ovarian cells, the signal-to-noise ratiowas estimated to be 4 times higher with respect to thecurrently used confocal photomultiplier detector. A linearrelationship between the fluorescence signal and the FITC-inulin concentrations ranging from 0.05 to 1.8 mg mL − 1 could be established. With the SAC8.5 CCD camera andusing DAPI, calcein-AM and propidium iodide we couldalso distinguish between viable, apoptotic and necrotic cells:exposure to CdCl 2  caused necrosis in A6 cells. Additionalexamples include the observation of wire-like mitochondrialnetworks in Mito Tracker Green-loaded Madin–Darby caninekidney cells. Furthermore, it is straightforward to interfacethe SAC8.5 with automated shutters to prevent rapidfluorophore photobleaching via easy to use  ASTROVIDEO software. In this study, we demonstrate that the SAC8.5black and white CCD camera is an easy-to-implementand cost-conscious addition to quantitative fluorescencemicrofluorimetry on living tissues and is suitable for teachinglaboratories. Correspondenceto:KoenJolling.Tel:0032(0)11268532;fax:0032(0)11168599;e-mail: koen.jolling@uhasselt.be Introduction Essential requirements for quantitative fluorescence imagingare a fluorescence signal that is linearly proportional tothe concentration of the labelled molecule, low and stabledark noise, acceptable dynamic range and appropriate highimaging speed matching fast changing cellular and tissuephenomena.Fordescannedfluorescencedetection,confocalmicroscopesare standard equipped with photomultiplier tubes (PMTs).However, one of the major drawbacks when using PMT-basedscanningconfocalimagingisthefluorophorephotobleaching(Lansing Taylor & Salmon, 1989; Foldes-Papp  et al ., 2003).This bleaching effect is even more seriously aggravatingwhen a mercury lamp is used to supplement for missinglaser excitation, especially in the ultraviolet (UV) range of the spectrum. In this so-called pseudoconfocal fluorescencemode the sample is fully illuminated while the image iscollectedbyscanningthedetection.Therefore,evenpixelsthatare not directly sampled by the scanning mirrors continueto bleach. Additionally, the phototoxicity has to be takeninto account because of the long overall exposure (Pawley,2006).Whole image spatial as well as temporal information canbe collected much faster by using a CCD or CMOS camera.Since commonly used scientific CCD camera systems areprohibitively expensive we evaluated the SAC8.5, a low-costPeltier-cooled black and white 8-bit CCD camera for fastquantitative fluorescence measurements. The performanceof this camera was tested on a nonscanning epifluorescencemicroscope (Nikon Eclipse TE2000-U) and in non-descannedmode on a laser scanning confocal microscope (LSCM)Zeiss LSCM 510 META. Samples comprise nonconfluent cellmonolayersandfreshlyisolatedkidneyproximaltubulesfrommice. C  2007 The Authors Journal compilation  C  2007 The Royal Microscopical Society  BUDGET-FRIENDLY CAMERA FOR IMAGING MICROSCOPY  265 Table1.  Properties of the objectives (A) and filter sets (B) used for experiments with the Zeiss LSCM 510 META setup.A Objective NA WD (mm) Eff. pixel size ( µ m) × 10 plan neofluar 0.3 5.6 1.8 × 1.8 × 20 plan apochromat 0.75 0.66 0.90 × 0.90 × 63 plan apochromat oil immersion 1.4 0.18 0.28 × 0.28B Filter set  λ ex (nm) Beamsplitter  λ em  (nm)1 BP 365/12 FT 395 LP 3979 BP 450–490 FT 510 LP 51515 BP 546/12 FT 580 LP 590NA: numerical aperture; WD: working distance; Eff. pixel size: effective pixel size; λ ex : excitation wavelength; λ em : emission wavelength. Materialandmethods Specifications of the SAC8.5 The SAC8.5 camera (SAC-imaging, Melbourne, FL, U.S.A.)is a Peltier-cooled black and white 8-bit CCD camera systemwith USB-2 interface. It comes standard with power supplyand housing. This budget-friendly, approximately 600 USdollar CCD camera is generally used in astronomy. Home-madefocusadjustableflanges,withapossibilitytoincorporatezoom lenses, connected the camera with the Nikon EclipseTE2000-UmicroscopeandtheZeissLSCM510METAconfocalsetup. The SAC8.5 camera is equipped with a Sony 1/3-inch.EXview HAD TM interline transfer CCD. Pixel size is 9.6  × 7.5  µ  m 2 and the displayed pixel layout is 640  ×  480 (HxVpixilated). Video lux rating amounts to 0.0003 and enablesdetection of low-intensity fluorescence signals. The SAC8.5 issoftware controlled ( ASTROVIDEO , COOA, Portim˜ao, Portugal)and runs preferably on Win2K, XP or MAC OSX operatingsystems. The minimal integration time for capturing shortexposure images is  160  s. Longer arbitrary exposure timesand video mode (30 frames per second) are possible. Flexibleimage export in a number of standard imaging formats, suchas JPEG, BITMAP 16 and 32 bit, FITS (the astronomicalstandard of most major observatories) and AVI digital videois offered in the  ASTROVIDEO  software. FITS images can beopened by the accompanying  FITSX  software. Furthermore,it was straightforward to interface the SAC8.5 CCD camerawith automated shutters (Lambda 10-2, Sutter InstrumentCompany, Novato, CA, U.S.A.) to prevent rapid fluorophorebleaching. To this end the software, running in a Video forWindows(VFW),activatesandcontrolsthecamerashutterviaserial port commands. In both excitation and detection pathssynchronously operating shutters could be placed (Ewen-Smith, 2005). Fluorescence imaging With the scan head occupying the left side port, the SAC8.5CCD camera was attached to the bottom port of a ZeissLSCM (Zeiss, Jena, Germany). Camera image collection tookplace in non-descanned mode using Hg-lamp excitationvia spectral selection cubes with selectable Zeiss filter sets(Table 1). Due to a front surface mirror reflecting the lamplight to the specimen at an angle of 90 ◦ , the reflected light ispartially polarized (Fresnell). For pseudoconfocal descannedfluorescenceimagingthesame100-WHg-lampillumination,but a descanned photomultiplier detector were utilised. Forclarity,indescannedmodethefluorescencesignalspassinsidethe scan head via its scanning mirrors and optical filtersto scan-head PMT detector(s). In non-descanned mode thefluorescence image is directly detected by the CCD cameracompletely bypassing the scan head and its scanning mirrorsand optics. Images were captured via prime focus imagingwithout an additional focusing lens in front of the camerafurther simplifying the setup and reducing costs. Cameramodeimageswerecomparedwithwide-field(detectionpinholeat maximum) descanned mode PMT images. The currentlyimplementedPMTdetectionrequiresabout15sperscanfora512 × 512(HxVpixilated)image.Allimageswerecollectedatroom temperature, 20 ◦ C. The properties of the objectives andfilter sets used in this setup are displayed in Table 1.Prior to imaging, the SAC8.5 CCD camera was properlyoriented and aligned with respect to the PMT image on theZeiss LSCM 510 META. For every experiment the camera wasplaced in exactly the same position using an external mark asreference. The field of view was calibrated using a Zeiss stagemicrometer.ThecollectedPMTimageswerecroppedtomatchtheCCDimagesizetakingintoaccountamultiplicationfactorof 1.25 × (Fig. 1).AsecondinvertedepifluorescencesetupconsistedofaNikonEclipse TE2000-U microscope (Nikon Instruments EuropeB.V., Badhoevedorp, The Netherlands) equipped with 30-WhalogenilluminationandaNikon × 10CFIachromatobjective(numerical aperture, NA  =  0.25). The SAC8.5 CCD camerawas attached to the optical side port and images werecapturedusingthemicroscopeopticsonly.Excitation(BP450-490 nm) and emission (LP 515 nm) filters were changedmanually.Whennecessaryneutraldensityfilterswereinsertedin the light path to reduce the intensity of the illumination C  2007 The Authors Journal compilation  C  2007 The Royal Microscopical Society,  Journal of Microscopy , 228 , 264–271  266  KOEN JOLLING  ET AL . Fig.1.  FluorescenceofDAPI-labelledCHOcellstakenwiththeSAC8.5CCDcamera(A)andthePMT(B)(ZeissLSCM510METAsetup).Thecirclesindicatethe ROIs that were used to calculate the average SNR. The scale difference of the SAC8.5 and the PMT detection amounts to a multiplication factor of 1.25. source.Allopticalfiltersanddichroicmirrorswerepurchasedfrom AHF analysentechnik AG (T¨ubingen, Germany). Thisparticular camera–microscope configuration was used toevaluate the usefulness of the SAC8.5 CCD camera forfutureluminalperfusionexperimentsonkidneytubulesusingfluorescence − labelled inulin (FITC − inulin). Kidney tubules Anin-housemaleC57BL/6mouseof10weeksofagewasused.Housed under standard vivarium conditions (temperature:20–22 ◦ C, humidity: 50%) it had access to an elementalpellet diet (Carfill, Oud-Turnhout, Belgium) and fresh waterad libitum. The mouse was sacrificed by means of cervicaldislocation. Subsequently the two kidneys were rapidlyremovedandtransferredintoice-coldmicrodissectionsolution(MDS)ofthefollowingcomposition(inmmol L − 1 ):120NaCl,5 KCl, 21 NaHCO 3 , 1.3 CaCl 2 , 1 MgCl 2  and 2 Na 2 HPO 4 and 5 mannitol (pH  =  7.4). The kidneys were decapsulatedusingfineforceps(DumontNo.5)andwerecutlongitudinallythrough the hilus into two identical pieces using a bistoury.The slices were transferred to a Petri dish with fresh ice-coldMDS. Again using a bistoury, the cortex was dissected andcut into small pieces. Proximal tubules were then dissectedmechanically using sharpened forceps (Dumont No. 5). Thenuclei of the isolated proximal tubules were labelled with300 n M  DAPI (Invitrogen, Merelbeke, Belgium), dissolved inPBS (pH = 7.4) for 5 min. After rinsing them three times inPBS (pH  =  7.4) the proximal tubules were transferred ontoglass cover slips using a glass pasteur pipet. The Zeiss  × 10plan neofluar objective and filter set 1, with the Hg lamp atminimalintensity,wereusedtovisualizethekidneytubules. Cell lines A Chinese Hamster Ovary (CHO) cell line, stably expressinghomomeric  α 2 glycine receptors (Mangin  et al ., 2003),was cultured in Dulbecco’s modified Eagle’s medium(DMEM) containing 10% (v/v) fetal calf serum, 2%penicillin/streptomycin and 500  µ g/100 mL Zeocin. Growthmedium was replaced twice a week and the cells weresubcultured weekly. Versene 0.2 g L − 1 (Invitrogen) was usedforcelldetachmentbyincubatingthecellsfor10minatroomtemperature (21 ◦ C). Trypsin-EDTA 0.25% (Invitrogen) for20 min at room temperature was applied if cells still adheredto the culture flask. Cells were plated on poly- L -lysine(100  µ g mL − 1 ) coated cover slips (Menzel, Braunschweig,Germany), and were allowed to adhere for 1 h at 37 ◦ C.They were fixed with 4% paraformaldehyde (Sigma-Aldrich,Bornem, Belgium) for 10 min and permeabilized with Triton-X-100 0.2% (Sigma-Aldrich) for 10 min. Subsequently, theywere labelled with 300-n M  DAPI (Invitrogen) for 5 min.Standard No. 1 cover slips (Menzel), mounted on a slide withProlong Gold antifade reagent (Invitrogen), were allowed todry in the dark for 1 day at room temperature. The next day,theCHOcellswerevisualizedusingaZeiss × 10planneofluarobjective. The fluorescence signal was measured using filterset 1.Madin–Darbycaninekidney(MDCK)cells,acelllineofdistaltubular srcin, were cultured in DMEM:Ham’s F-12 (1:1),supplementedwith10%fetalcalfserum,14m M L -glutamine,25 m M  NaHCO3, 100-U mL − 1 penicillin and 100-g mL − 1 streptomycin.Cellsweremaintainedinahumidifiedincubatorat 37 ◦ C in the presence of 5% CO 2 . The medium was renewedevery3–4days.Forallexperiments,1 × 10 5 cellswereseededonto round glass cover slips with a diameter of 24 mm. After5 days of culture, confluent monolayers were used as follows.Briefly,thecellswererinsedthreetimeswithastandardMDCKcellphysiologicalsolution.Thentheywereloadedwith200n M MitoTrackerGreen(MTG;Invitrogen)for30minat37 ◦ C.Afterbeing loaded, the cells were gently washed three times withthe same physiological solution. Fluorescence measurementswereperformedat37 ◦ CwithaZeiss × 63plan-apochromatoil-immersion objective. The mitochondria were visualized usingfilter set 9 with the 100-W Hg lamp (at 90% of maximumintensity). C  2007 The Authors Journal compilation  C  2007 The Royal Microscopical Society,  Journal of Microscopy , 228 , 264–271  BUDGET-FRIENDLY CAMERA FOR IMAGING MICROSCOPY  267 A6 cells, a cell line derived from the distal kidney tubule of thefrog Xenopuslaevis ,wereculturedinstandardplastictissuecultureflasksinahumidifiedincubatorat28 ◦ Cinthepresenceof 1% CO 2 . The cells were cultured in Leibovitz’s L-15:HAM’sF-12 (1:1) supplemented with 10% fetal calf serum, 20%water, 3.8 m M L -glutamine, 8 m M  NaHCO3, 87 IU penicillinand87- µ gmL − 1 streptomycin.Growthmediumwasreplacedtwice a week and the cells were subcultured weekly. Celldetachment was obtained as described above. Subsequently,cellswereseededonglasscoverslips( ≈ 2.8 × 10 3 cellscm − 2 )andincubatedunderthesameconditionsasmentionedbefore.After three days the cells were exposed to culture mediumsupplemented with 10  µ M  and 10 m M  CdCl 2  for 24 h. Onthe last day staining and experiments were conducted asfollows.Briefly,thecellswererinsedthreetimeswithastandardA6 cell physiological solution. Both control and cadmium-treated cells were simultaneously stained with 300-n M  DAPI,1- µ  M  calcein-AM and 1- µ  g mL − 1 propidium iodide (PI) forabout 10 min. After being washed three times with the samephysiological solution cells were visualized with the SAC8.5camera using filter set 1, 9 and 15 for DAPI, calcein-AM andPI,respectively,withthe100-WHglamp(at30%ofmaximumintensity) and a Zeiss  × 20 plan apochromat objective. Thistriple staining allowed distinguishing between viable (calceinpositive), apoptotic and necrotic cells (PI positive). FITC-inulin measurements Fluorescein isothiocyanate-labelled inulin (Sigma-Aldrich)was dissolved in heated MDS. FITC-inulin standard samples,ranging from 0.01 to 2 mg mL − 1 , were transferred to 1  µ Lconstant-bore capillary tubes (Drummond Scientific,Broomall, PA, U.S.A.). These tubes were imaged witha Nikon  × 10 CFI achromat objective (NA  =  0.25) onthe epifluorescence Nikon Eclipse TE2000-U microscope.Excitation was through a BP 450-490 nm using halogenillumination at 90% of maximal intensity, i.e. 30 W.Fluorescence emission was observed through an LP 515 nm. Image processing and analysis Transmission and fluorescence images were analyzed usingthe ASTROVIDEO and FITSX (COOA)commercialsoftwareaswellas IMAGEJ freeware(NIH,Bethesda,MD,U.S.A.)runningunderMicrosoft Windows XP Professional. Fitting of the calibrationdata, fluorescence intensity versus FITC concentration, wascarriedoutviaaroutinewrittenforS-PLUS2000ProfessionalRelease 1 for Windows (Insightful, Seattle, WA, U.S.A.).Tocomparetheaveragesignal-to-noiseratio(SNR)followingDAPI staining, single images of the same CHO cells weretaken sequentially using SAC8.5 camera and PMT detectionimmediately after identical UV exposure on the Zeiss LSCMsetup. The intensity of the 365-nm Hg line on the samples’illuminated area of about 10 mm 2 using a × 10 plan neofluarobjective was estimated at approximately 503  µ  W  ± 3%(Lasermate-Q TM powermeter,CoherentInc.,SantaClara,CA,U.S.A.). The integration time of the camera was  160  s. Scanspeed for PMT detection was reduced to an acceptable level tomaximizetheSNR.ThecurrentlyimplementedPMTdetectionrequired about 15 s per scan for a 512 × 512 (HxV pixilated)image. The offset slider (black level) and the photomultipliergain (detector gain slider) were set carefully to ensure a lowbackground and to prevent saturation. This was confirmedfor each measurement by using the ‘palette’ feature of theLSCM software. FITS images taken by the SAC8.5 camerawere opened via the  FITSX  software and exported as bmp files.These images together with the laser-scanning images wereanalyzed using  IMAGEJ . Average signal and standard deviationwererecordedforfivedifferentcells(circularregionsofinterest;Ø ≈ 7  µ m) as well as for five arbitrarily chosen backgroundareas. The average SNR for each region of interest (ROI) wascalculated by the following equation:SNR = ¯ x cell − ¯ x background   σ  2cell + σ  2background , wherefortheROI ¯ x cell = averageofcellularsignals, ¯ x background = average from background signals,  σ  2cell  =  variance fromcellular signals and  σ  2background  =  variance of backgroundsignals. Results Dark signal level and stability The SAC8.5 dark signal and its stability over time weremeasured on the LSCM setup starting after a 10-minstabilization period of the Peltier cooling according tomanufacturer’sinstructions.Darksignalswerecollectedevery30 s for 20 min. Dark signal levels remained constant overtimeanddidnotfurtherdecreasewithintheexperimentaltimeframe (linear regression: slope =  –0.0002,  p  >  0.05). Roomtemperature did not vary appreciably during the collectionperiod. Signal-to-noise ratio DAPI-labelled CHO cells (Fig. 1) were used to compare theSNR between the SAC8.5 CCD camera and the PMT detectionsystem.CirclesindicatetheROIsthatwereusedtocalculatetheSNR.TheaverageSNRoftheSAC8.5camera(SNR = 37.2)wasestimated to be approximately 4 times higher in comparisonwith the PMTs (SNR = 9.6). DAPI staining and photodegradation DAPI is a well-known and frequently used DNA-intercalatingfluorophore for staining the cell nucleus and detecting C  2007 The Authors Journal compilation  C  2007 The Royal Microscopical Society,  Journal of Microscopy , 228 , 264–271  268  KOEN JOLLING  ET AL . Fig.2.  Transmission (left) and DAPI fluorescence (right) image of isolated mouse proximal tubules taken with the SAC8.5 CCD camera (Zeiss LSCM 510META setup). The arrow indicates one of many DAPI stained nuclei. stress-induced nuclear fragmentation (Somji  et al ., 2004).DAPIstaininginisolatednonperfusedmouseproximaltubuleswasreadilydetectedasthetransmissionandDAPIfluorescenceimagesshowninFig.2.Thebrightfluorescentspotsrepresentthe proximal tubule cell nuclei.DAPI-labelled isolated mouse proximal tubules were usedto investigate photodegradation as a result of continuousexposure through low UV illumination. Figure 3 displays theDAPIfluorescencesignalasafunctionoftime.Thelogarithmicfluorescence values represent the mean of four DAPI stainednuclei and are expressed as a percentage of the maximumsignal. On a semilogarithmic scale, the DAPI signal decreasedlinearlyovertimeandwasundertheseilluminationconditionsno longer detectable after approximately 2.5 min (linearregression:  R 2 = 0.95,  p < 0.01). Visualization of mitochondria in MDCK cells MTG is a fluorophore that selectively stains the mitochondriaand can be used in colocalization studies (Smets  et al .,2004). MTG staining in MDCK cells was readily detectedand an intact wire-like mitochondrial network was observed(Fig. 4). Transmission and fluorescence images weredeconvoluted using Huygens Essential deconvolutionmicroscopy software (Scientific Volume Imaging B.V.,Hilversum, The Netherlands). Cadmium-induced cell death of A6 cells Cell death occurred when A6 cells were cultured in growthmedium supplemented with 10 m M  CdCl 2  (Fig. 5). To Fig.3.  PhotodegradationovertimeofDAPI-labelledisolatedmouseproximaltubulenuclei( n = 4).TheDAPIfluorescencewasmeasuredusingtheSAC8.5CCD camera (Zeiss LSCM 510 META setup). C  2007 The Authors Journal compilation  C  2007 The Royal Microscopical Society,  Journal of Microscopy , 228 , 264–271
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