Pets & Animals

Observation of the Dynamics of Live Cardiomyocytes through a Free-Running Scanning Near-Field Optical Microscopy Setup

Observation of the Dynamics of Live Cardiomyocytes through a Free-Running Scanning Near-Field Optical Microscopy Setup
of 5
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.
  Observation of the dynamics of livecardiomyocytes through a free-runningscanning near-field optical microscopy setup Ruggero Micheletto, Morgan Denyer, Martin Scholl, Ken Nakajima, Andreas Offenhauser, Masahiko Hara, and Wolfgang Knoll We report the observation of live-cell dynamics by noncontact scanning near-field optical microscopy  SNOM   modified to work with living biological samples that are fully immersed in liquid. We did notuse the SNOM setup in strictly near-field conditions   we used 1-  m constant-height mode  ; however, wecould examine the dynamics of rhythmically beating cardiac myocytes in culture with extremely high vertical sensitivity below the nanometric range. We could halt scans at any point to record localizedcontraction profiles of the cell membrane. We show that the contractions of the organisms changedshape dramatically within adjacent areas. We believe that the spatial dependency of the contractionsarises because of the measurement system’s ability to resolve the behavior of individual submembraneactin bundles. Our results, combining imaging and real-time recording in localized areas, reveal a new,toourknowledge,noninvasivemethodforusingSNOMsetupsforstudyingthedynamicsoflivebiologicalsamples. © 1999 Optical Society of America OCIS codes:  180.5810, 170.6920, 170.1420, 170.1530. 1. Introduction and Experiment Submicrometer studies of cell ultrastructure havebeen performed with scanning- and transmission-electron microscopes and atomic-force microscopes.However, live cells cannot be examined with the firsttwo instruments because those systems require cellfixation and observation in vacuum, and images of live cells obtained with the last-named instrumentmay be compromised by direct contact between thecantilever and sample-deforming soft tissue.Scanning near-field optical microscope 1  SNOM  systems include a sharp optical fiber used as a prob-ing element to collect the optical field created in prox-imity to a sample. There are a number of opticalconfiguration modes in which a SNOM can be oper-ated. The most common ones are collection, reflec-tion, and illumination modes, as described in detailelsewhere. 2 In this study we aim to detect morpho-logical activity of a rhythmically contracting synci-tium of a cultured cardiac myocyte 3 isolated from1–3-day-old rats. The optical configuration that wechose for this purpose is shown schematically in Fig.1. Light illuminates the sample from the bottom,creatingaspotoflightthatistransmittedthroughoutthe whole sample. To permit working in liquid, wesurrounded the sample by a Teflon ring that containsa culture medium to allow the cardiomycytes to sur- vive. Above the sample a fiber tip is used as a prob-ing element to collect the intensity of the optical fieldinasubmicrometer-sizedlocalizedarea. Ifthereare variations in the myocyte’s morphological status, thecoupled field will change accordingly, allowing us torecord a signal associated with the cell activity. Thesample holder is mounted on a piezoelectric three-dimensional actuator that scans an area of the sam-ple, yielding an image of the specimen underinvestigation. The setup and the control softwarethatwedevelopedallowusbothtoscanandtostopatany point of the image that is forming on the screenof our monitor. When the scan is halted we are ableto start recording the localized light field in that R. Micheletto     is with the Depart-ment of Material Science, Graduate School of Engineering, Uni- versity of Kyoto, Yoshida, Sakyo-ku, Kyoto 606-8501, Japan. M.Denyer, M. Scholl, K. Nakajma, and M. Hara are with FrontierResearch Program, The Institute of Physical and Chemical Re-search   RIKEN  , Wako, Saitama 351-0198, Japan. A. Offen-hauser and W. Knoll are with Max-Planck Institute for PolymerResearch, Ackermannweg 10, D-55128, Mainz, Germany.Received 25 February 1999; revised manuscript received 14June 1999.0003-6935  99  316648-05$15.00  0 © 1999 Optical Society of America 6648 APPLIED OPTICS    Vol. 38, No. 31    1 November 1999  point. Astheopticalprobeweusedplatinum-coatedchemically etched glass fiber tips 4,5 sharpened by useof the difference in etching rates between core andcladdinginaselectiveetchingsolutionofbufferedHFacid. A typical scanning-electron microscope pic-tureoftheapexofoneofourtipsisshownintheinsetof Fig. 1.We cultivated rat muscle cells as explained in Ap-pendix A. Then the culture was transferred to theoptical setup for examination. The area surround-ing the sample was maintained at 33 °C to allow thecells to survive and beat as long as possible during the experiment. The cells were alive for almost 12 hin our setup. 2. Results In Fig. 2 we show two scans of the culture and therelative recordings at two points, P 1  and P 2 . Super-imposed over the enlarged image in Fig. 2  b   arewhite stripes, which are visible especially in thelower central area. These discontinuities are barely visible in the lower-resolution image   Fig. 2  a   be-cause of the different scales of the two figures.These features are the result of the beating cell’smodifying the optical coupling between tip and sam-ple, producing and thus distorting the image of thecell along the scan direction. The central bottomareaoftheimageismoreintenselystriped,indicating that in that region the cardiomyocyte is more active.In Fig. 2  c   we show an image of the same cardiacmyocyte culture as viewed with a conventional opti-cal microscope   Olympus, Model IX70, phase con-trast  . It is possible to distinguish the elongatedcardiac cells. The round white spots are cell bodiesthat did not grow properly on the glass substrate;such cells are inevitably produced in a culture.Comparing the conventional optics image of Fig.2  c   with our two scanned maps in Figs. 2  a   and 2  b  ,we can see that the same structures are visible in allthe images and that the morphologies of the imagesare compatible. We have to stress that we are fo-cusing our efforts not on obtaining good lateral reso-lution but on establishing a noncontact method todetectlocallythedynamicsofasamplewithsubnano-metric sensitivity, as we show below.While we scanned the surface of a cell it was pos-sible to halt the scan at any point so the contractionprofiles of the cardiac cell could be recorded in realtime and in highly localized regions. We found thatcontraction profiles showed marked repeatability.In Fig. 3  a   we can see a beating profile composed of a main peak followed by a much smaller one, slightlydelayed in time. All the peaks in this recording show similar profiles, indicating that we are actuallyobserving a complex dynamic feature of the cell’s con-traction. The recording shown in Fig. 3  b   revealsthat the amplitudes of contractions can change dra-maticallybetweenadjacentrecordingareas. There-cording sites are adjacent, separated by 0.9   m; thetwo profiles show distinct behavior, revealing thatthe dynamic properties of the cell are discriminatedat different points. After many hours of experi-ments the cells were still beating properly withouthaving been removed from the sample holder. 3. Discussion There is substantial evidence in the literature 6,7 thatcontracting and broadening actin bundles may act asthe detectable scattering elements in contractilecells. Actin bundles in cardiac myocytes have diam-eters of approximately 1   m. Near-field microscopyalso can be used for examining unstained submem-brane cytoskeletal structures in fixed hippocampalneurons. 8 Therefore we believe that the spatiallydependent contraction profiles reported in this studyarise as a result of the setup’s ability to resolve thecontractile activity of individual submembrane actinbundles. This possibility has a range of significantimplications for biological studies in which mechan-ical or optical modifications are involved.Oursystemdoesnotworkstrictlyinaconventionalnear-field collection-mode configuration. The tip isfree running over a distance of the order of 1   m  estimated by observation through a long-focus mi-croscope until apparent contact  . In these condi-tions the probe is actually dipped inside and outsidethe optical near field at every contraction of the cell,which can change the shape of the cell by as much asof 500 nm. 3,9  Also, we wanted to enhance the cou-pling by direct illumination of the sample from thebottom without total internal reflection.Becauseoftheconditionsthatwehavedescribedinwhichoursetupworks,wewereunsurewhattoname Fig.1. Schematicviewofthecoreofourhomemadesystem. Themetal-coated optical fiber tip is fabricated by chemical etching. A submicrometer aperture is created at its apex to permit highlylocalized light coupling. The laser light wavelength is 633 nm at15 mW. To enhance the transmission we do not use total internalreflection. The incident angle is 0°; the spot diameter is 2–3 mm.The light reaching the fiber probe is real transmitted light.Sample–probe separation is not controlled by feedback and it is of the order of 1   m   see text  . Cardiac myocyte cells are preparedon a standard cover glass immersed in a HEPES-buffered HamsF10 culture medium containing 0.5% insulin–transferrin–selenite  ITS  , antibiotics, and 3% fetal calf serum   FCS   feeding medium  .The atmosphere surrounding the sample area is maintained atconstant temperature and humidity   33 °C and   80%   by circula-tion of warm humid air under the plastic curtain that surroundsthe optical system. 1 November 1999    Vol. 38, No. 31    APPLIED OPTICS 6649  our setup. Can it still be considered a SNOM sys-tem? In other systems that have been proposed inthe literature, for example, illumination-modeSNOM setups, 8,10 the light comes from the probe andilluminates a complex structured sample at a con-stant height. In those systems the presence of anevanescent field is not certain; nevertheless, the sys-tems are still generally called SNOM’s because theprobe is located near the sample and thus the nameincludes the term “near field.”In our case the sample–probe separation is esti-mated to be roughly 1   m. The sample that we areinvestigating is alive and contracting vertically by asmuch as 500 nm. We thus believe that the tip isinserted in and out of the near-field range by these vertical movements of the sample. Also, as in anyexperiment of this kind, we have to consider thatdistances are only a rough estimation of the real,unknown ones. So we can be either be closer to thenear-field region or farther away, depending on theexperiment. For these reasons we call our system aSNOM setup, and we clearly describe the working conditions to clarify its limits and advantages.We are aware that the optical signal that werecord, as with other optical microscopes, does notcontain pure morphological information. Differ-ences in optical coupling could arise also from otherphenomena in the cell body, such as local modifica-tionsintheindexofrefractionandmovementofinnercellular structures. In the hypothesis that the mor-phological displacement plays the major role in ourmeasurements, we can estimate the vertical sensitiv-ity of our recordings. As we stated above, contrac-tions of cardiac myocytes are known to extend vertically approximately 500 nm. We could observethe beating by using a noncalibrated voltage signal.The strongest peaks observed were   5 V from thebase to the tip of the peak. Considering a minimumdiscrimination of 2 mV, from electronic analog-to-digital conversion the minimal cell displacement de-tected was well below 1 nm.These numbers are intended to be only estimates, Fig. 2.   a   Scan of 40   m    40   m over live cardiac myocytes; 128    128 pixels.   b   Close-up of the region outlined by the rectangle in  a  . The image is 8   m    15   m wide; 90    90 pixels. The horizontal stripes are due to the beating of the cells during the scanning.White points indicate recording sites where we stopped the scan to monitor the cell activity in real time.   c   Conventional phase-contrastoptical microscope picture of the same cell specimen, shown for comparison with the SNOM maps. The area shown is not the same asin   a   and   b  . 6650 APPLIED OPTICS    Vol. 38, No. 31    1 November 1999  particularly because they assume linearity betweensignal and contraction displacement. This linearityis not guaranteed, because of the complexity of thephenomena involved. However, we believe that it isareasonableapproximationoftheorderofmagnitudeof the vertical sensitivity of our instrument.Withrespecttothelateralresolution,inspiteofthesubnanometric vertical resolution mentioned abovewe are currently working over a range of 1   m or less  see Fig. 2  b   for an estimate  . We have to considerthat, in regular SNOM experiments performed withsolid samples, the resolution is usually higher, lim-ited only by the size of the aperture tip. Instead, weare working with the tip fully immersed in liquid andwith a live, soft, and conctracting sample. The tip iskept relatively far from the sample to avoid harshcontact that could harm the cells and disturb thequality of the probe. We have chosen to work in thesocalled  free-running mode . Thesample–probesep-aration is not modified by any feedback system and isconsidered to be   1   m. We believe that this is themain cause of a decreased lateral resolution; how-ever,ourchoicesimplifiedtheopticalsetupextremelyand preserved the quality of the cells for the entiretime of the experiment. We were able to make allthe recordings that we required, and we believe thatfurther improvement of this method is possible. A confocal scanning microscope 11 may reach the resolv-ingpoweroftheimagescurrentlypresentedhere,butimprovement beyond that power is limited because of thediffractionlimitsintrinsicintheconfocalmethod.In a number of papers it was speculated that me-chanicalresponsesexistinbiologicallivesamples,forexample, in the retina of a squid or in the process of axonal transport. 12–14  Application of the methodol-ogy presented here could be extremely useful in pro- viding new insight into these studies, especiallyconsideringitsgoodverticalsensitivity. Oneareaof our intense interest is understanding the mecha-nisms of cell guidance to engineer tissue regenera-tion. In the nervous system, for example, theextending nerve processes navigate by means of thesensory growth cone. 15 This navigation is probablyaided by secondary messenger-mediated cytoskeletalresponses to environmental guidance cues 16 ; how-ever, the exact mechanisms of cytoskeletal reorgani-zation and repolymerization are complex and are notfully understood. 17  A system similar to the one thatwe have described here can easily be modified to hostfluorescent applications. 18 Thus cells under investi-gation could be marked with dyes to specifically op-tically resolve submembrane structures in live cellsandfacilitatethestudyoftheirdynamicsinthegrow-ing process.More generally speaking, the setup described is anew method thatshouldprovidefurther insightin allstudies in which biological live samples show mor-phological activity of any kind. Light can be ana-lyzed by its spectrum, giving further information onthe nature of the sample. The high-resolution glassfiber presented could be applied not only as localprobes, as shown here, but also as highly localizedlight sources to be used as stimuli in experimentswith light-sensitive biological specimens, giving riseto a wide variety of interesting applications.  Appendix A  Heart-cell cultures were prepared by a method usedby Courtois  et al . 19 Briefly, hearts were removedfrom four neonatal Sprague Dawley rats   aged 1–2days  . Each heart on removal was placed in a Petridishcontaining10mLofCa  Mg-freeHanksbalancedsalt solution   HBSS; GIBCO, BRL  , massagedquickly to remove excess blood, and chopped into1-mm 2 pieces. Thechoppedheartswerethenpooledina20-mlconicalcentrifugetubeandwashedatleastthree times in ice cold Ca  Mg-free HBSS. Afterwashing, the HBSS was replaced with 8 ml of 0.05%crude trypsin   Sigma-Aldrich   in versene buffer  GIBCO  , and after 8 min of incubation at 37 °C thesupernatant was discarded. The chopped heartswere then enzymatically dissociated by the additionof 100   L of DNAase Type II solution   10,000  ml;Sigma   followed 1–2 min later by the addition of 2.5ml of 0.05% crude trypsin–versene and stirred for 8min at 37 °C. We then collected the supernatant,leaving the undissociated tissue in the centrifugetube, and added it to 4 ml of HEPES-buffered Hams Fig. 3. Recordings taken at the two points, P1 and P2, in Fig. 2.Numbers in parentheses are the coordinates   srcin at the lower-right-hand corner   of the actual pixel. The distance between thetwo points is   900 nm. Recording time resolution is 0.03 s.Eachrecordinglasted  10s;thenscanningwasrestartedfromthesame point. The two scans were taken 30 s apart. The tworecordings have a common vertical scale to permit comparisonbetween them. The time scales are independent, and the the twoimages are not synchronized. 1 November 1999    Vol. 38, No. 31    APPLIED OPTICS 6651  F10   GIBCO   containing a 0.5% ITS solution  GIBCO   and 36% FCS   GIBCO   to block trypsiniza-tion. We centrifuged the sample for 5 min at 1500U  min, resuspended it in 0.5–1.5 ml of ice coldHEPES-buffered Hams F10 containing 0.5% ITS and10% FCS, and stored it at 0–4 °C. Meanwhile, theundissociated chopped heart tissue remaining in the20-ml conical centrifuge tube was once more enzy-matically dissociated. This cyclical procedure wasrepeated until all the heart tissue had been dis-persed. The collected cell suspensions were pooledin a 25-cm 2 tissue culture flask and incubated for 1 hat 37 °C. This incubation allowed the majority of the cell debris and fibroblasts to adhere to the flask,leaving an increased proportion of myocytes in sus-pension. Finally, the heart cells were plated ontoglass slides coated with 2   g   cm 2 of fibronectin   Sig-ma   at densities of 10 6 cells  mL and incubated at37 °C for 24 h in HEPES-buffered Hams F10 contain-ing 0.5% ITS 2.5 ml  100 ml of an antibiotic solution  consisting of 200 mM glutamine   Sigma  ; 5000 U  mlof penicillin–streptomycin   GIBCO  ; and 250   g   mlof Fungizone   GIBCO  , and 10% FCS. The cellswere then fed daily with prewarmed HEPES-buffered Hams F10 containing 0.5% ITS, antibiotics,and 3% FCS   feeding media  .We are in debt to Motoichi Ohtsu and Angelo DeMarco for partially inspiring this research and forkeen discussions and suggestions. References 1. E. Betzig, J. K. Trautman, T. D. Harris, J. S. Weiner, and R. L.Kostelak, “Breaking the diffraction barrier: optical micros-copy on a nanometric scale,” Science  251,  1468–1470   1991  .2. M. Ohtsu, “Photon STM: from imaging to fabrication,” Opto-electron. Devices Technol.  10,  147–166   1995  .3. A. Shrier and J. R. Clay, “Pacemaker currents in chick embry-onic heart cells change with development,” Nature  283,  670–671   1980  .4. S. Mononobe, M. Naya, T. Saiki, and M. Ohtsu, “Reproduciblefabrication of a fiber probe with a nanometric protrusion fornear-field optics,” Appl. Opt.  36,  1496–1500   1997  .5. T. Saiki, S. Mononobe, M. Ohtsu, N. Saito, and K. Kusano,“Tailoring a high-transmission fiber probe for photon scanning tunnelingmicroscope,”Appl.Phys.Lett. 68, 2612–2614  1996  .6. J. H. Rose, M. R. Kaufmann, S. A. Wickline, C. S. Hall, andJ. G. Miller, “A proposed microscopic elastic wave theory forultrasonic backscatter from myocardial tissue,” J. Acoust. Soc. Am.  97,  656–668   1995  .7. J. Meunier and M. Bertrand, “Echographic image mean graylevel changes with tissue dynamics: a system-based modelstudy,” IEEE Trans. Biomed. Eng.  42,  403–410   1995  .8. R.U.Maheswari,S.Mononobe,H.Tatsumi,Y.Katayama,andM.Ohtsu,“Observationofsubcellularstructuresofneuronsbyan illumination mode near-field optical microscope under anoptical feedback control,” Opt. Rev.  3,  463–467   1996  .9. M. F. Arnsdorf and R. Lal, “Recent progress with atomic forcemicroscopy in biology: molecular resolution imaging of cellmembranes, constituent biomolecules, and microcrystals,” in  Imaging Technologies and Applications , R. J. Heaston, ed.,Proc. SPIE  1778,  112–16   1992  .10. R. U. Maheswari, H. Kadono, and M. Ohtsu, “Power spectralanalysis for evaluating optical near-field images of 20 nm goldparticles,” Opt. Commun.  131,  133–142   1996  .11. B. Bailey, D. L. Farkas, D. L. Taylor, and F. Lanni, “Enhance-ment of axial resolution in fluorescence microscopy bystanding-wave excitation,” Nature  366,  44–48   1993  .12. I. Tasaki and T. Nakaye, “Rapid mechanical responses of thedark-adapted squid retina to light pulses,” Science  223,  411–413   1984  .13. R. D. Allen, J. Metuzals, I. Tasaki, S. T. Brady, and S. P.Gilbert, “Fast axonal transport in squid giant axon,” Science 218,  1127–1129   1982  .14. G. C. Bonazzola, A. DeMarco, M. Maringelli, S. Rinaudo, and A.Trabucco,“Possibleapplicationoftheopticaltunneleffecttomembrane biophysics,” Eur. Biophys. J.  12,  51–55   1985  .15. P. R. Gordon-Weeks, “The Ultrastructure of the neuronalgrowthcone: newinsightsfromsubcellularfractionationandrapid freezing studies,” Electron Microsci. Rev.  1,  201–219  1988  .16. P. R. Gordon-Weeks, “Growth cones—the mechanism of neu-rite advance,” Bioessays  13,  235–239   1991  .17. J. Q. Zheng, J. J. Wan, and M. M. Poo, “Essential role of filopodia in chemotropic turning of nerve growth cone in-duced by a glutamate gradient,” J. Neurosci.  16,  1140–1149  1994  .18. J.Hwang,L.K.Tamm,C.Bohm,T.S.Ramalingam,E.Betzig,and M. Edidin, “Nanoscale complexity of phospholipid mono-layers investigated by near-field scanning optical microscopy,”Science  270,  610–614   1995  .19. M. Courtois, S. Khatami, E. Fantini, P. Athias, P. Mielle, and A. Grynberg, “Polyunsaturated fatty acids in cultured cardio-myocytes: effect on physiology and  b -adrenoceptor function,” Am. J. Physiol.  262,  H451–H456   1992  . 6652 APPLIED OPTICS    Vol. 38, No. 31    1 November 1999
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

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!