Magnetotactic Author(s): Richard Blakemore Source: Science, New Series, Vol. 190, No. 4212 (Oct. 24, 1975), pp. 377-379 Published by: American Association for the Advancement of Science Stable URL: http://www.jstor.org/stable/1741081 . Accessed: 12/07/2011 23:51

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=aaas. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

American Association for the Advancement of Science is collaborating with JSTOR to digitize, preserve and extend access to Science.

http://www.jstor.org cussion. First, the remarkable precision of 8. T. R. Marshall, C. S. Parmenter, M. Seaver, in finite cylinders. Diameters can be found with pre- preparation. cisions better than one-half percent. size determination at resonance is on a rel- 9. The values d = 2.956 Am and m = 1.581 are given 11. Contribution No. 2632 from the Chemical Labora- ative basis only. It presumes that m is by the manufacturer (Dow Chemical Co., Mid- tories of Indiana University. Supported by the Na- land, Michigan). The batch number is unknown. tional Science Foundation. C.S.P. appreciates the fixed, and therefore any uncertainty in in 10. W. A. Farone and M. Kerker [J. Opt. Soc. Am. help of E. C. Hartman in the formative stages of the absolute In trials to 56, 481 (1966)] and D. Cooke and M. Kerker this work degrades precision. [Rev. Sci. Instrum. 39, 320 (1968)] have character- determine both m and d simultaneously at ized nonresonance differential scattering from in- 16 June 1975 resonance, the precision of each seems lim- ited to about 0.5 percent. This is also what we find for nonresonant scattering. Sec- ond, the relationship of "optical" diame- Magnetotactic Bacteria ters determined at resonance to "physical" diameters is not entirely clear, particularly Abstract. Bacteria with directed by the local geomagneticfield have been ob- when relative precisions of a few ang- served in marine sediments. These magnetotactic microorganisms possess flagella and stroms are considered. In polystyrene contain novel structured particles, rich in , within intracytoplasmic membrane vesi- aerosols ranging from about I to 3 gm, di- cles. Conceivably these particles impart to cells a . This could explain ameters determined by our optical method the observed migration of these organisms infields as weak as 0.5 gauss. (away from resonances) match those from physical methods and microscopy Few studies have unequivocally revealed During attempts to isolate Spirochaeta to within a few percent (7, 8). effects of the earth's on liv- plicatilis from marine marsh muds (3), I The sensitivity of scattering to particle ing organisms, although recent work in- observed microorganisms which rapidly parameters at resonance can be either a dicates that birds (1) and elasmobranchs migrated (4) toward one side of drops of useful tool or a liability. It is clearly the (2) detect and may use geomagnetism as a the mud transferred to microscope slides latter if resonances within the range of cue for orientation. I now describe a bacte- (Fig. 1). I presumed this to be a phototactic aerosol parameters are not recognized. rial tactic response to magnetic fields, a response toward light from a northwest Techniques based on differential scattering phenomenon for which the term magneto- laboratory window. It became apparent, in particular may be subject to substantial is appropriate. however, that light was not the stimulus di- errors if one is not careful (8). On the other recting the migration of these organisms as hand, the availability of tunable mono- cells aggregated at the same side of mud chromatic light sources makes these reso- of the distribution of O SEC ? . ,* drops regardless light nances generally accessible for many sys- on the slides, as well as in the dark. The di- tems. Since the effects of resonance occur rection in which these organisms moved for absorbing as well as nonabsorbing immediately changed when small magnets particles, and since they should be seen were moved about in the vicinity of the mi- with cylindrical objects (10) as well as A croscope preparations. This suggested that r N spheres, numerous applications are pos- A--;,,^ geomagnetism was the stimulus for the be- *v.* *> s * v- 1- + sible. The phenomenon might be used to havior of the cells. It was experimentally detect extremely small changes in the di- confirmed that the migration of the bac- mensions of a scatterer due to externally teria was, indeed, directed by the earth's applied stress or modifications to its envi- magnetic field (5). ronment, or to changes in its composition. Magnetotactic organisms were present The effect might also be used as a contin- in surface sediments collected from salt . uous monitor in the manufacturing of SEC marshes of Cape Cod, Massachusetts, and wires or fibers optical requiring a precise IB in surface layers of sedimentary cores col- maintenance of the diameter or refractive lected from a depth of 15 m in Buzzards index. It is also possible that resonant scat- Bay. Population densities in these environ- tering may be a useful tool for studying ments ranged from 200 to 1000 cells per regularly shaped nonspherical particles in milliliter. Mud samples placed under sev- biological systems. eral centimeters of seawater in glass jars THOMASR. MARSHALL and kept in dim light in the laboratory un- CHARLESS. PARMENTER derwent an ecological succession (3). Pop- MARK SEAVER ulations of magnetotactic organisms in- Department of Chemistry, Indiana creased to hundreds of thousands of cells 1. frames recorded University, Bloomington 47401 Fig.quential Cinematographic sequenceon Kodakof Tri-X bacteria re- per milliliter in many such mud samples displayingmagnetotaxis. Portions of three se- after several months to several years of References and Notes quentialframes recordedon Kodak Tri-X re- versal 16-mm movie film at 18 frames storage. Magnetotactic bacteria were 1. M. The per sec- Kerker, Scattering of Light and Other with a Zeiss RA 38 distributed in surface of the Electromagnetic Radiation (Academic Press, New ond, microscope. The im- evenly layers York, 1969), pp. 50, 94-96. ages shown were photographically reversed and stored muds even though the jars were po- 2. H. C. van de Hulst, Light Scattering by Small enlarged. (A) Freely swimming magnetotactic Particles New York, 151 sitioned for long periods in the same geo- (Wiley, 1957), pp. 127, bacteria aggregated at the northern extremity of 158. graphic orientation. Apparently factors 3. E. J. Meehan, J. Colloid Interface Sci. 49, 146 a water drop by responding to geomagnetism. (1974). At the time of recording, a small permanent additional to also determine 4. For example, see pp. 97--127 in (1). magnet was used to reverse the field. the distribution of the in 5. J. Tima and F. T. the magnetic organisms natural Gucker, Proceedings of 2nd The cells then in the International Clean Air Congress (Academic migrated opposite direction environments, since larger populations of New as recorded in frames Press, York, 1971), p. 463. (B) (0.5 second or 9 bacteria were not detected 6. F. T. Gucker, J. Tuma, H.-M. Lin, C.-M. Huang, frames and 1 second or 18 magnetotactic S. later) (C) (recorded C. Ems, T. R. Marshall, J. Aerosol Sci. 4, 389 frames The arrow in the northern areas of a marsh as com- (1973). later). indicates the direction 7. T. R. Marshall, C. S. Parmenter, M. Seaver, Natl. of the earth's north geomagnetic pole (bar, 100 pared to other locations in the marsh. Bur. Stand. (U.S.) Spec. Publ. 412 (1974), p. 41. gm). The organisms have not yet been iso- 24 OCTOBER 1975 377 lated and grown in pure culture. I obtained nisms which rapidly migrated to the sea- and B). Each bundle consisted of approxi- purified cell suspensions for electron mi- water drop were collected with micropipets mately seven flagella which were inserted croscopic examination by taking advan- and prepared for electron microscopy. in a disk-shaped structure (Fig. 2B, arrows) tage of the cells' magnetotactic behavior. I Magnetotactic cells were roughly spher- similar to those present in Ectothiorhodos- placed drops of seawater in contact with ical and averaged 1 um in diameter (Fig. pira mobilis (6). Two chains, each con- the northern edge of drops of mud enrich- 2, A to C). Two bundles of flagella were sisting of approximately five to ten elec- ment material on microscope slides. Orga- observed at one side of the cell (Fig. 2, A tron-opaque crystal-like particles, charac- terized magnetotactic cells (Fig. 2, A to D). In chemically fixed, thin-sectioned mag- netotactic bacteria, these chains of parti- cles were found to be internal cell com- 'k i~~~~~~~~~~54j~ ~ ~ ~ ~ ~ ~~C &M ponents (Fig. 2C, white arrowheads). The Wj.r`' crystal-like particles may exist as relatively thin plates. Apparent variation in the elec- tron opacity of these structures (Fig. 2, A and D) may be due either to variation in thickness or to electrical charging encoun- tered when they were placed in the electron beam of the microscope. In addition to a typical gram-negative type of wall, cells possessed intracytoplasmic unit mem- branes arranged as vesicles about their pe- riphery (Fig. 2C, black arrows). The crys- tallike particles were often found within these membrane vesicles (Fig. 2D, arrow). The intravesicular location of the particles suggests that they may be synthesized by the membranes. Clumps of these particles, outside of cells, were also observed by transmission electron microscopy of sedi- ments underlying older enrichments. These Fig. 2. Transmission electron micrographs recorded on Kodak EM estar thick base film with a Hi- observations that intracellular tachi HU-12 electron at 75 kv. Electron of a suggest microscope operating (A) micrograph magnetotactic the bacterium. Two bundles of flagella insert at one side of the cell. Two chains of iron-rich particles are particles synthesized by magnetotactic present (arrows). Deposits believed to be polyphosphate are abundant. The outer cell layers bacteria are released after death and lysis appear disrupted. Cell stained with phosphotungstic acid (bar, 0.5 gm). (B) Electron micrograph of a of cells, and accumulate in sediments. portion of a magnetotactic bacterium showing the two disks (arrows) into which flagella insert. Os- Mishustina (7) isolated similar particles motically lysed cell stained with phosphotungstic acid (bar, 0.25 um). (C) Electron micrograph of a from sediments of the Barents Sea but did chemically fixed and thin-sectioned magnetotactic bacterium stained with lead and uranyl salts. The cell has a gram-negative type wall. The iron-rich particles (white arrowheads) have been unintention- not realize their probable origin or proper- ally displaced during thin-sectioning revealing intracytoplasmic membranes arranged as vesicles ties. The tendency of these particles to (black arrows) adjacent to the cell plasma membrane. Much of the central portion of the cell ap- clump when outside of cells is consistent 0.25 Electron pears to have been extracted during chemical preparation (bar, um). (D) micrograph with the that they be a per- iron in a cell as described for The possibility may of a single chain of particles containing present prepared (C). substance such as particles are enclosed within vesicles consisting of triple-layered membranes (arrow) (bar, 0.07 ,tm). manently magnetic mag- netite. Unstained whole cells and their crystal- like inclusions were to energy Fig. 3. A portionof an energypro- subjected file of x-rayscollected after electron dispersive x-ray microanalysis (8). An en- excitationof part of an unstained ergy profile of x-rays collected after elec- magnetotacticbacterium. The area tron excitation of chains of the particles analyzed containedtwo chains of within a cell in 3. Iron the similarto those in single appears Fig. particles Fig. was the detectable element 2D. X-rayswith energies of 6.4 and predominant 7.0 kev correspondto those for Ka within the particles. Elemental ratio analy- and K: emission lines of iron, re- sis (8) of the signals obtained from each of spectively. X-rays from elements several chains of particles indicated that presentin the sample,but not in a the other elements detected were not controlarea from whichcells were in a ratio to iron. These absent,are indicatedby peakscon- present constant sistingof verticalbars toppedwith elements were also far less abundant than white dots. Silicon (solid white iron in the crystal-like particles found out- peak, 1.7 kev) was more abundant side the cells. Thus, these elements were in the control than in the sample. not constituents of the novel The unstainedspecimen was ana- apparently lyzed for 120 seconds on a copper cell inclusions and probably represent con- gridin a berylliumholder positioned tamination from areas adjacent to the 360 incident to the 0.25-gum-diam- particles. eter electronbeam. Analysis was performedin the transmissionmode at x 43,000and 60 kv. The The relation between the chains of iron- electronbeam was to accommodatethe elongatedshape of the chains of particles. astigmatized is Largeamounts of phosphorussuggest contamination from adjacent granules (see Fig. 2A) believed rich particles and magnetotaxis probably to be polyphosphate. not coincidental. Recently, I observed in 378 SCIENCE, VOL. 190 marsh muds a or of References and Notes lated area free of detectable geomagnetic dis- larger type species tortion. bacterium than that de- I. W. T. Keeton, in Animal Orientation and Naviga- 6. C. C. Remsen et J. Bacteriol. 2374 magnetotactic S. R. K. G. J. Ja- al., 95, (1968). tion, Galler, Schmidt-Koenig, 7. I. E. Mishustina, Ecol. Res. Comm. Bull. 17, 143 scribed. This organism measures 1.2 by 3.2 cobs, R. E. Belleville, Eds. (Government Printing W. (1973). gm and differs from the aforementioned in Office, Washington, D.C., 1972), p. 579; 8. Dr. J. Russ, W. MacMillan (of EDAX Interna- Wiltschko, in ibid., p. 569. tional Prairie and I a number of details when 2. A. J. in Handbook Laboratories, View, 111.), per- morphological Kalmijn, of Sensory Physi- formed these analyses on unstained thin-sectioned ology, A. Fessard, Ed. (Springer-Verlag, Berlin, cells and on unstained whole A viewed by phase contrast microscopy. Both vol. section organisms. Phillips 1974), 3, 3,p. 147. EM 301G scanning-transmission electron micro- types of magnetotactic bacteria contain 3. R. P. Blakemore and E. Canale-Parola, Arch. with an EDAX detector and Mikrobiol. 273 scope equipped x-ray chains of iron-rich At least five 89, (1973). and multichannel pulse height analyzer coupled to particles. 4. In the absence of obvious artificial magnetic fields an solid state was the downward and northward EDIT/NOVA minicomputer morphologically distinct types of magneto- organisms migrated used. Conditions have been described [ J. Russ, J. at about 100 um/sec. Submicrosc. Cvtol. 6, 55 tactic bacteria are in 5. Dr. A. J. and I tested these ori- (1974)1. recognizable samples Kalmijn organisms' 9. Supported by grant AI-08248 to Dr. E. collected from marshes in the of entation in the undisturbed earth's magnetic field Canale-Parola at the of Massachusetts, vicinity and in fields where either the horizontal or the ver- University Amherst, and by a grant from the Sarah Mellon Woods Hole. This raises the interesting tical component was reversed with Helmholtz Scaife Foundation to the Woods Hole Oceano- coils. The field was uniform applied within 0.2 per- Institution. I thank Drs. V. T. Bowen, E. possibility that several (perhaps previously cent in a 10 cm in diameter. The coils were graphic region Canale-Parola, H. W. Jannasch, and A. J. Kalmijn unrecognized) species of bacteria may be adjusted and oriented appropriately for the geo- for conversations and criticism under consideration. stimulating helpful those which can be magnetic component Mag- of the manuscript. I thank L. Surprenant for her among organisms sepa- netotactic bacteria in sufficient numbers to be vis- contribution to the work here and Dr. R. ible as a mass at the of a seawater reported rated from their environment by their edge drop (see Turner for the use of her cinematographic equip- Fig. 1) were sealed beneath a glass cover slip on a ment. Contribution 3581 of the Woods Hole magnetotactic responses. slide. The bacteria in the microscope migrated op- Oceanographic Institution. could not result from a posite direction when current was applied to the Magnetotaxis coils. The experiments were conducted in an iso- 9 June 1975; revised 30 July 1975 magnetic force tending to "pull" these bac- teria northward since, in uniform magnetic fields such as those used to demonstrate cell responses to geomagnetism (5), a di- Mass Measurement: A Study of Anomalies pole would not translocate in any direction as a consequence of magnetism; it would Abstract. It has always been assumed that the measurement of the difference in mass merely rotate to a preferred orientation between two objects would be the same in all laboratories. Recent National Bureau of within the magnetic field. Cells suspended Standards measurements involving dissimilar objects (effective density ranging from 2.7 in seawater did not migrate when killed to 16.6 grams per cubic centimeter) at a wide variety of pressures (0.5 to 2 atmospheres) with vapors of osmium tetroxide. Those have been made with sufficient precision to test this assumption. The results show unsus- not stuck to glass surfaces rotated so as to pected discrepancies which may approach 1 milligram in a kilogram in the assignment of remain aligned with an applied magnetic mass values when dissimilar materials are involved. These discrepancies have not been field when the direction of the field was noted in the past because precision comparisons of both like and unlike materials have changed. Thus, cell motility is required for nearly always been made in a relatively restricted range of environmental conditions. The magnetotaxis, and cell orientation is the worldwide mass measurement system is therefore consistent, because similar materials primary response to magnetism. Freely have been used in the construction of weight sets, but possibly offset with respect to the suspended killed cells also frequently mass unit as embodied in the platinum-iridium defining artifact. formed chains of up to ten cells. Each chain behaved as a single dipole in re- The unit of mass of the Systeme Inter- able with standard deviations of a single sponse to changes in the position of a near- national d'Unites (SI) system of measure- weighing on the order of 0.03 mg, and re- by permanent magnet. Perhaps the iron- ment units is defined by an artifact, the In- search balances in national laboratories rich cell inclusions serve as magnetic di- ternational Prototype Kilogram preserved have standard deviations on the order of poles which convey a magnetic moment at the Bureau International des Poids et 0.005 mg. Careful attention must be given upon the cell, thus orienting the cells in Mesures near Paris. The transfer of a mass to the manner in which the buoyant force is magnetic fields. Magnetotaxis would result value from such a defining artifact to an- accounted for if this precision is to be uti- if, within each cell, a fixed spatial relation- other cannot be done directly in a normal lized in the transfer of the mass unit from ship existed between the orienting mecha- laboratory environment. The mass value one object to another. nism and cell propulsion. Studies of the be- assigned to another object must be inferred In such a transfer the observed differ- havior of living and dead cells in uniform from the results of a comparison of the ar- ence, as determined by the balance, is ad- and nonuniform fields should contribute to tifact and the "unknown" by means of a justed or corrected by the product of the an understanding of the magnetotactic balance. The balance, in turn, is a force computed air density, D, at the time of the mechanism. It may also be possible to de- comparator, responding to both the grav- measurement and the difference in the pre- tect a preferred orientation of cells or their itational force and the buoyant force of the viously measured displacement volumes, flagella and their iron-rich inclusions in atmosphere acting on the object or objects V1 and V2, of the objects being compared. magnetic fields. Results of survival experi- on the balance pan or pans. In a given loca- In essence, the observed difference, Y, is ments indicate that these organisms are ei- tion, the gravitational force acting on an the difference between the resultant forces ther or strict anaerobes. object is a constant and proportional to the acting on the two objects (the gravitational Because the vertical component of the mass of the object. The buoyant force, force minus the buoyant force). To deter- earth's magnetic field has a greater magni- however, is proportional to the dis- mine the gravitational force, one must add tude than the horizontal in the locations placement volume of the object and the a correction of Y for the difference D(V, - where they have been found, magnetotaxis density of the surrounding air. Although V2) in buoyant force; that is, the value for might serve to direct these organisms the buoyant force is not large with respect the mass difference, M, -- M2, is given by downward toward sediments and anaerob- to the gravitational force, being only on the M, -M2 = - ic areas favorable to their growth. order of 0.015 percent for materials of nor- Y+ D(VI V2) RICHARD BLAKEMORE mal density, it is large with respect to the One expects the mass difference for any Woods Hole Oceanographic Institution, precision of currently available balances. given pair of objects to be constant regard- Woods Hole, Massachusetts 02543 Commercial kilogram balances are avail- less of location. 24 OCTOBER 1975 379